Compare commits

..
Author SHA1 Message Date
Sydney RunkleandClaude Opus 4.7 4f01115fa8 refactor(add_messages): clearer step-by-step structure
Restructure add_messages so fast vs. slow path, REMOVE_ALL handling,
and format application each live in a single numbered section with a
short lead-in comment. No behavior changes; the previous commits'
optimizations are preserved.

- fold the two path branches into one `if pure_append else slow_path`
  so format handling happens at a single exit instead of being
  duplicated between fast and slow paths
- drop the now-redundant `left_seq` local; index `left` directly after
  the coerce step (with a single `cast(list, left)` for the type
  checker)
- use `set.isdisjoint` for the overlap check

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-04-24 07:31:03 -04:00
Sydney RunkleandClaude Opus 4.7 e601988ab3 test(add_messages): cover fast-path guards and format handling
Nine new tests pin the behavioral boundaries introduced by the
optimization: chunk / dict / tuple / missing-id left inputs must fall
through to full conversion, duplicate right ids and None right ids
must still be handled correctly, format="langchain-openai" and invalid
format must round-trip through the fast path, and the fast path must
return a fresh list rather than aliasing left.

Also picks up a ruff-format reflow in test_time_travel.py.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-04-24 07:31:03 -04:00
Sydney RunkleandClaude Sonnet 4.6 c606652958 fix(add_messages): guard fast path against None IDs and intra-right duplicates
Two bugs in the fast-path optimisation:

1. The left-side type guard checked isinstance(BaseMessage) but not
   id is not None. Messages without IDs (e.g. HumanMessage(content="hi"))
   would skip ID assignment and return None IDs.

2. The pure-append short-circuit only checked for overlaps between
   right and left, not duplicates within right itself. A right list
   containing two messages with the same ID would bypass the slow-path
   deduplication and return both.

Fixes:
- Add left_seq[0].id is not None to the type-guard condition.
- Replace the any() overlap check with a set-intersection check that
  also verifies len(right_id_set) == len(right_msgs) (no intra-right
  duplicates) before taking the fast return.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-04-24 07:31:03 -04:00
Sydney RunkleandClaude Sonnet 4.6 6c35d4e97b perf(add_messages): skip left-side conversion and fast-path pure appends
Two optimizations for the hot path in add_messages, which is called on
every write to a messages channel:

1. Skip conversion of left: when left is already list[BaseMessage] with
   IDs assigned (true for every call after the first), skip
   convert_to_messages + message_chunk_to_message + the ID-None loop.
   These are O(n) no-ops on already-resolved messages that allocate two
   intermediate lists.

2. Pure-append short-circuit: when right contains no RemoveMessage and
   no ID overlaps with left, return left + right directly. Replaces the
   O(n) copy + dict build + filter with a single set-membership check.

Benchmarks (median of 2000 iterations, pure-append scenario):
  10-msg thread:   2.9x faster
  100-msg thread:  6.6x faster
  1000-msg thread: 7.3x faster
  200-step simulation (2 msgs/step): 3.4x faster end-to-end

Also adds tests/test_add_messages_benchmark.py with correctness tests
for all scenarios (append, update, remove) and a runnable benchmark.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-04-24 07:31:03 -04:00
Sydney Runkle afec98f369 internal for now 2026-04-24 07:29:52 -04:00
Sydney RunkleandClaude Opus 4.7 f25d1935ef fix(postgres): handle missing checkpoint_id in _get_channel_writes_history; update test signatures
Two fixes exposed by running the postgres test suite against a local
postgres instance:

1. `PostgresSaver._get_channel_writes_history` /
   `AsyncPostgresSaver._aget_channel_writes_history` required
   `checkpoint_id` in the passed config, raising `KeyError` when called
   with just `thread_id` (e.g. `graph.aget_state({"thread_id": "..."})`).
   Now resolves to the latest checkpoint via `get_tuple`/`aget_tuple`
   when the id is missing.

2. `test_get_checkpoint_no_channel_values` (sync + async) monkeypatched
   `_load_checkpoint_tuple` with the old `(value, cur)` signature. Method
   now takes `(value)` only since delta reconstruction moved out of the
   tuple-load path — updated both tests.

Local postgres (`brew install pgvector postgresql@16`, running on port
5441) now exercises all 40 non-vector postgres tests green.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-04-23 14:47:22 -04:00
Sydney RunkleandClaude Opus 4.7 3a7ed5b454 refactor(delta-channel): honest data model, private experimental API
Restructure DeltaChannel reconstruction so the hydration path matches
pregel's storage axes (blobs + writes) without leaking internal DTOs
into the public checkpoint contract.

Key changes:

* Deleted `DeltaChannelWrites` dataclass and `SEED_UNSET` sentinel.
  Reconstruction data no longer flows through `Checkpoint.channel_values`
  as a wrapped DTO — that field now carries a value or `DELTA_SENTINEL`,
  never a reconstruction shape.
* Added private `_ChannelWritesHistory(seed: Any, writes: list[PendingWrite])`
  NamedTuple as the return type for the new storage-level query.
* Added private, experimental `_get_channel_writes_history` /
  `_aget_channel_writes_history` on `BaseCheckpointSaver` — reference
  impl via `get_tuple` + `parent_config` walk, overridden on
  `InMemorySaver` / `PostgresSaver` / `AsyncPostgresSaver` for perf.
  Fixes a latent migration bug in the base fallback (now inspects
  ancestor `channel_values` for pre-delta seed).
* `DeltaChannel.from_checkpoint(seed)` simplified to two cases
  (sentinel/MISSING → empty, else → seed). New `replay_writes` method
  folds `list[PendingWrite]` through the reducer.
* Delta hydration consolidated inside `channels_from_checkpoint` via
  optional `saver` + `config` kwargs (+ async mirror
  `achannels_from_checkpoint`). All six pregel call sites updated.
  `get_tuple` no longer patches `channel_values` — removed
  `_resolve_delta_channels` (memory) and per-tuple reconstruction from
  `_load_checkpoint_tuple` (postgres sync + async).
* Hydration short-circuits on the target's own blob: if
  `channel_values[k]` is a real value (pre-migration tip, `update_state`
  result), use it directly. Only walks ancestors when the target holds
  sentinel or is missing. Fixes a correctness bug where migration-tip
  and `update_state` values would be lost.
* New test_delta_channel_migration.py: 10 scenarios covering
  BinaryOperatorAggregate → DeltaChannel migration (basic + async,
  time-travel, fork, `update_state`, tip-of-pre-migration, base-saver
  fallback parity, cross-thread isolation).

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-04-23 14:15:02 -04:00
Sydney Runkle 31ef0e942a refactor(delta-channel): drop snapshot_every and saver Overwrite terminator
snapshot_every was a knob for bounding reconstruction cost on deep threads.
Benchmarks (notes/add_messages_replay_problem.md + scratch work on
sr/add-messages-replay-bench) showed the add_messages fast-path
(optimize/add-messages-fast-path) closes the quadratic replay cost for
threads under ~1000 turns, where the crossover to snapshots makes sense.
For deeper threads we'll ship a first-class compaction primitive instead.

Removals:

* DeltaChannel: snapshot_every ctor param, _writes_since_snapshot counter,
  should_snapshot() / snapshot_write() methods, counter threading through
  _apply_write / update / from_checkpoint / copy.
* Pregel loop: post-checkpoint snapshot-injection block and
  SNAPSHOT_TASK_ID import + constant.
* Checkpoint base: _overwrite_types() helper and the ancestor-walk
  short-circuit on user-emitted Overwrite in sync + async
  get_channel_writes.
* InMemory + Postgres savers: same walk-terminator shortcut. The
  pre-delta blob terminator (seed-from-ancestor-blob) stays — it's
  required for migration correctness, not a snapshot optimization.
* Tests for all of the above.

Preserved:

* Channel-level Overwrite semantics in DeltaChannel / BinOpAggregate:
  Overwrite still resets the value at reducer level; same-super-step
  dedup and InvalidUpdateError on multiple Overwrites still enforced.
* Pre-delta migration seeding.
2026-04-23 09:54:12 -04:00
Sydney Runkle d120f127ca refactor(delta-channel): plain SELECT WHERE replaces recursive CTE
The recursive CTE was bottlenecked by a JSON-expression join
(`bl.version = checkpoint->'channel_versions'->>bl.channel`) that the
planner could not index, producing an O(ancestors x blobs) nested-loop.
At depth 1000 it ran ~275 ms and removed ~2M filter rows; the recursion
itself was 2.4 ms.

Switch to three plain indexed SELECTs per delta channel
(checkpoints, checkpoint_writes, checkpoint_blobs); a pure helper on
BasePostgresSaver walks the parent chain and assembles
DeltaChannelWrites. Sync (__init__.py) and async (aio.py) each own
their three-roundtrip I/O wrappers.

Bench numbers (notes/delta_channel_query_bench.md): 3x at depth 50,
15x at depth 200, ~100x at depth 1000. Plain over-fetches sibling rows
when the thread branches but still wins at every realistic depth on
both local and remote postgres.

Multi-channel coalescing dropped — reconstruction is per-channel now.
Same shape as InMemorySaver. Can come back as a SQL-level
optimization later if needed.
2026-04-23 08:21:00 -04:00
Sydney Runkle 9e330c96dc contextvar 2026-04-23 07:42:51 -04:00
Sydney RunkleandClaude Opus 4.7 cd8fad5905 fix(delta-channel): target-exclusion, pre-delta seed, one-query postgres walk
Four fixes from an independent review of the reconstruction pipeline, plus
a structural cleanup:

1. Ancestor walk excludes the target checkpoint itself (matches pregel:
   writes stored under checkpoint_id=T are pending for the NEXT step and
   applied separately via apply_writes). Memory saver previously included
   them, diverging from Postgres and causing pending writes to be folded
   into the reconstructed snapshot — visible via get_state during
   interrupts and time-travel into a non-leaf checkpoint.

2. Pre-delta blob terminator. When the walk hits an ancestor whose blob
   for the channel is a real value (not DELTA_SENTINEL), bind that blob
   as DeltaChannelWrites.seed and stop. Without this, threads migrated
   from pre-delta storage would replay ancestor writes to the root
   forever AND lose any value that lived only in the old blob
   (e.g. from update_state). Per-ancestor, the blob is checked BEFORE
   its writes — a pre-delta blob subsumes writes at the same checkpoint,
   so including them would double-count.

3. Base-fallback get_channel_writes follows parent_checkpoint_id instead
   of list(before=...). The previous form returned every tuple with
   id<target, including sibling branches on forked threads.

4. seed replaces the Overwrite-wrapping hack for pre-delta values.
   DeltaChannelWrites(writes, seed=SEED_UNSET) makes the saver's
   reconstruction terminator semantically explicit; drops the lazy
   _make_overwrite import dance. User-emitted Overwrite still reset the
   chain via _apply_write as before.

Postgres: recursive CTE enumerates on-path ancestors and joins once
against checkpoint_writes and once against checkpoint_blobs for every
delta channel in the get_tuple — one roundtrip instead of the previous
3 queries × N channels.

Tests added:
- Pre-delta blob seeding (seed binding, no double-counting of ancestor
  writes at the terminator, pending-at-target excluded).
- Root checkpoint returns empty writes.
- Seed-based from_checkpoint replay (three scenarios: with writes,
  seed-only, seed=None distinct from SEED_UNSET).

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-04-22 22:09:08 -04:00
Sydney Runkle acc7eda8c5 optimizations i sure hope 2026-04-22 20:47:51 -04:00
Sydney Runkle 5b7fdf5655 eh 2026-04-22 18:41:06 -04:00
Sydney RunkleandClaude Opus 4.7 4cad68f767 fix(delta-channel): unwrap NotRequired[X] for dict/set reducers
Annotated[NotRequired[dict[...]], DeltaChannel(reducer)] (the shape used
by deepagents' filesystem middleware) fell through type inference to
`list`, so the first operator call blew up with
"'list' object is not a mapping". `_is_field_channel` now unwraps a
parameterized Required[X]/NotRequired[X] before stripping extras, which
lets dict/set/mapping outer types reach the abc normalization block.

Also type-annotates the `new` locals in DeltaChannel.copy() and
from_checkpoint() so mypy can infer them through the abstract return
type.

Adds tests covering: dict Overwrite in update and in writes replay,
snapshot_write with a dict reducer, dict backwards-compat checkpoints,
NotRequired type inference, and a filesystem-shaped end-to-end graph.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-04-22 17:58:36 -04:00
Sydney Runkle 9d8c0be068 arbitrary 2026-04-22 17:54:12 -04:00
Sydney Runkle 51154be4ab lint 2026-04-22 17:47:53 -04:00
Sydney Runkle 2e7edb2b60 lint 2026-04-22 17:30:22 -04:00
Sydney Runkle 325cb42f19 lint and snapshot every 2026-04-22 17:06:53 -04:00
Sydney Runkle b9fad696ec cleanup 2026-04-22 16:45:32 -04:00
Sydney Runkle 96760e6267 chore(delta-channel): add PR description 2026-04-22 14:35:48 -04:00
Sydney RunkleandClaude Sonnet 4.6 9342ae215a chore(delta-channel): remove snapshot_every — simpler design, better storage savings
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-04-22 14:12:57 -04:00
Sydney Runkle ee5b3c2639 refactor(postgres): replace recursive CTE with two-query ancestor walk for DeltaChannel
Instead of a recursive SQL CTE, collect the ancestor checkpoint ID chain in
Python by fetching all (checkpoint_id, parent_checkpoint_id) for the thread
in one query, then fetch writes with a plain WHERE checkpoint_id = ANY(...).

Simpler, avoids recursive query planner overhead, and uses well-indexed lookups.
2026-04-22 14:03:37 -04:00
Sydney Runkle d7c3616620 feat(delta-channel): store sentinel in blobs, reconstruct from checkpoint_writes
DeltaChannel.checkpoint() now returns a zero-byte DeltaChannelSentinel
instead of duplicating delta data in checkpoint_blobs. Reconstruction
walks the parent checkpoint chain via checkpoint_writes (which already
holds per-step writes) and replays them through the operator.

In-memory benchmark (100 turns, ~20K tokens):
  storage: 10.2 MB → 40.5 KB (251x reduction)
  read:    0.6ms → 7.9ms (reconstruction cost, amortized by storage savings)

InMemorySaver and PostgresSaver override get_channel_writes() with
efficient implementations (Python dict walk and recursive CTE respectively).
The base class fallback uses self.list() with a thread-local recursion guard.
2026-04-22 14:03:37 -04:00
Sydney Runkle 06e302bbff refactor(delta-channel): infer typ from Annotated outer type instead of constructor arg
Remove the positional `typ` parameter from `DeltaChannel.__init__`. The type is
now injected automatically from the `Annotated` outer type in `_is_field_channel`
(matching how `BinaryOperatorAggregate` receives its type). `copy()` and
`from_checkpoint()` propagate `self.typ` explicitly. Test helpers updated to
use `_get_channel` with the proper `Annotated` path.
2026-04-22 14:03:37 -04:00
Sydney Runkle 10da2326a9 chore(delta-channel): remove supports_delta_channels flag
Rely on the runtime raise in DeltaChannel.from_checkpoint() instead of
a compile-time boolean flag. Savers that assemble DeltaChainValue inside
_load_blobs work transparently; savers that don't will pass through a raw
DeltaValue and hit a clear ValueError on first reload.

Removes: BaseCheckpointSaver.supports_delta_channels, the attribute on
InMemorySaver / PostgresSaver / AsyncPostgresSaver, the compile-time
UserWarning in StateGraph.compile(), and the associated test.
2026-04-22 14:03:37 -04:00
Sydney RunkleandClaude Sonnet 4.6 84f00c51eb chore: remove docs/ from PR
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-04-22 14:03:37 -04:00
ccurmeandSydney Runkle 3cab106bdf fix(prebuilt): handle injected NotRequired keys (#7392)
Resolves https://github.com/langchain-ai/langchain/issues/35585

This would previously raise KeyError:
```python
from typing import Annotated

from langchain_core.tools import tool
from langchain.agents import create_agent
from typing_extensions import NotRequired
from langgraph.prebuilt import InjectedState
from langchain.agents import AgentState


class CustomAgentState(AgentState):
    city: NotRequired[str]


@tool
def get_weather(city: Annotated[str | None, InjectedState("city")] = None) -> str:
    """Get weather for a given city."""
    if city is None:
        city = "Boston"
    return f"It's always sunny in {city}!"


agent = create_agent(
    model="claude-sonnet-4-6",
    tools=[get_weather],
    system_prompt="You are a helpful assistant",
    state_schema=CustomAgentState,
)

input_message = {
    "role": "user",
    "content": "What's the weather?",
}

result = agent.invoke({"messages": [input_message]})
for m in result["messages"]:
    m.pretty_print()
```

---------

Co-authored-by: Sydney Runkle <sydneymarierunkle@gmail.com>
2026-04-22 14:03:37 -04:00
Sydney Runkle 77bb349309 lint 2026-04-22 14:03:37 -04:00
Sydney Runkle 1d8364a749 latest 2026-04-22 14:03:37 -04:00
Sydney Runkle 06e97b0fd2 fix(delta-channel): support non-list reducers (dict) and fix MISSING handling
Use typ() instead of [] throughout DeltaChannel so reducers over dict
(and other non-list types) work correctly. fromCheckpoint(MISSING) now
leaves value as typ() from __init__ instead of overwriting with MISSING.
copy() uses value.copy() to handle dicts. update() initialises base from
typ() when value is MISSING. Add four tests covering the deepagents-style
dict-merge / file-deletion reducer pattern.
2026-04-22 14:03:37 -04:00
Sydney RunkleandClaude Sonnet 4.6 e256a31d00 chore(delta): rename _steps_since_rehydrate → _steps_since_snapshot; add audit tests
- Rename `_steps_since_rehydrate` → `_steps_since_snapshot` in DeltaChannel
  for clarity (counts steps since the last snapshot, not since rehydration)
- Pre-seed cycle-detection `visited` set with current checkpoint ID in both
  sync and async `_assemble_delta_channels` to prevent self-referential chains
- Add 4 new unit tests:
  - `test_delta_channel_snapshot_every_emits_plain_list`: verifies counter
    semantics and snapshot/delta transitions
  - `test_delta_channel_snapshot_every_end_to_end`: graph-level smoke test
  - `test_delta_channel_assembly_fast_path_returns_delta_value`: exercises
    chain traversal via get_channel_blob returning DeltaValue then plain list
  - `test_delta_channel_assembly_broken_chain_logs_warning`: partial chain
    when get_tuple returns None

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-04-22 14:03:37 -04:00
Sydney RunkleandClaude Sonnet 4.6 04b3ae7cd0 chore: rename serde type tag "diff" → "delta" for DeltaValue
Consistent with channel/type naming (DeltaChannel, DeltaValue).

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-04-22 14:03:37 -04:00
Sydney RunkleandClaude Sonnet 4.6 ec52520389 chore: apply format/lint fixes across checkpoint, checkpoint-postgres, prebuilt
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-04-22 14:03:37 -04:00
Sydney RunkleandClaude Sonnet 4.6 82fea763c5 fix: register DeltaValue in SAFE_MSGPACK_TYPES; rename _is_diff_delta; cross-saver benchmark
- Add DeltaValue to SAFE_MSGPACK_TYPES so SQLite and other msgpack-based
  savers don't emit "Deserializing unregistered type" warnings.
- Rename _is_diff_delta → _is_delta_value (leftover from DiffChannel rename).
- Parametrize benchmark by checkpointer: runs InMemory (fast-path) and
  SQLite (get_tuple fallback) in the same table, sharing the _run_turns helper.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-04-22 14:03:37 -04:00
Sydney RunkleandClaude Sonnet 4.6 6cfcad18f3 fix(delta-channel): fix chain assembly and get_state paths
- Fix InMemorySaver.get_channel_blob: use correct storage[thread_id][ns]
  nesting and deserialize the checkpoint before extracting channel_versions.
- Pass checkpoint_id to after_checkpoint() in channels_from_checkpoint so
  DeltaChannel seeds _last_checkpoint_id correctly on load; without this
  every turn broke the chain at its boundary.
- Wire _assemble_delta_channels into _prepare_state_snapshot and
  _aprepare_state_snapshot (get_state / get_state_history paths) and into
  perform_superstep / aperform_superstep (update_state paths) — previously
  only the loop __enter__ path did assembly.
- Fix test_get_channel_blob to use the correct storage structure.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-04-22 14:03:37 -04:00
Sydney Runkle db6b9ec995 test(channels): replace unsupported-saver raise test with fallback assembly test 2026-04-22 14:03:37 -04:00
Sydney Runkle 55fdc7aec6 feat(postgres): remove _load_diff_chains; add get_channel_blob / aget_channel_blob 2026-04-22 14:03:37 -04:00
Sydney Runkle 9969fb9737 feat(memory): implement get_channel_blob; remove diff handling from _load_blobs 2026-04-22 14:03:37 -04:00
Sydney Runkle 40981fdac7 feat(pregel): wire DeltaChannel assembly into loop; pass checkpoint_id to after_checkpoint 2026-04-22 14:03:37 -04:00
Sydney Runkle 5d1b3c4190 feat(pregel): add _assemble_delta_channels helpers for universal DeltaChannel support 2026-04-22 14:03:37 -04:00
Sydney RunkleandClaude Sonnet 4.6 fa83d64eff feat(channels): DeltaChannel tracks checkpoint_id; emits prev_checkpoint_id in DeltaValue
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-04-22 14:03:37 -04:00
Sydney RunkleandClaude Sonnet 4.6 4608af9615 feat(serde): diff type encodes prev_checkpoint_id; loads_typed returns DeltaValue
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-04-22 14:03:37 -04:00
Sydney Runkle 9beda5d3fb docs(checkpoint): expand aget_channel_blob docstring for parity 2026-04-22 14:03:37 -04:00
Sydney Runkle 9a6d7e08fb chore: add .worktrees/ to .gitignore 2026-04-22 14:03:37 -04:00
Sydney Runkle bbeb2759ba feat(checkpoint): DeltaValue uses prev_checkpoint_id; add get_channel_blob stubs 2026-04-22 14:03:37 -04:00
Sydney Runkle b799b95138 chore: rename DiffChannel/DiffDelta/DiffChainValue to Delta* across libs
Renames the diff-channel types to DeltaChannel, DeltaValue, and DeltaChainValue
for consistency with the settled naming convention.
2026-04-22 14:03:37 -04:00
Sydney Runkle ed9711fd33 more tests 2026-04-22 14:03:37 -04:00
Sydney RunkleandClaude Sonnet 4.6 ca883fe4eb chore: format/lint fixes for rehydrate_every benchmark
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-04-22 14:03:37 -04:00
Sydney RunkleandClaude Sonnet 4.6 4b9e4d25ca feat(channels): add rehydrate_every to DiffChannel for bounded chain traversal
Periodic full-snapshot checkpoints cap chain depth, trading a small
amount of extra storage for bounded reconstruction time.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-04-22 14:03:37 -04:00
Sydney RunkleandClaude Sonnet 4.6 ec8fd85ea2 test(channels): add DiffChannel vs BinaryOperatorAggregate storage/time benchmark
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-04-22 14:03:36 -04:00
Sydney RunkleandClaude Sonnet 4.6 ebd98f2e27 chore: format and lint fixes for DiffChannel implementation
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-04-22 14:03:36 -04:00
Sydney RunkleandClaude Sonnet 4.6 318fee9fc6 fix(checkpoint/postgres): pass cursor to avoid deadlock in diff chain traversal
Fixes a critical deadlock that occurs when _load_diff_chains calls self._cursor()
from within _load_blobs while the outer _load_checkpoint_tuple already holds
self._cursor(). On bare (non-pool) connections, the threading.Lock is not
reentrant, causing a deadlock.

Solution: Pass the cursor as a parameter to _load_diff_chains and _load_blobs
instead of acquiring a new cursor within those methods. Updated _load_checkpoint_tuple
to acquire a cursor once at the top level and pass it through the call chain.

Changes:
- Updated _load_blobs signature to accept optional cur parameter
- Updated _load_diff_chains signature (base and implementations) to accept optional cur parameter
- Modified _load_checkpoint_tuple in PostgresSaver to acquire cursor and pass it
- Modified _load_checkpoint_tuple_async to acquire cursor only when diff_payloads exist
- Removed nested self._cursor() calls in _load_diff_chains and _load_diff_chains_async

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-04-22 14:03:36 -04:00
Sydney RunkleandClaude Sonnet 4.6 e37299af87 feat(checkpoint/postgres): diff chain reconstruction in async saver
Add `_load_diff_chains_async` to `AsyncPostgresSaver` and override
`_load_checkpoint_tuple` to inline blob-parsing and diff-chain
resolution via async point-lookup traversal, mirroring the sync
`PostgresSaver._load_diff_chains` implementation. Add integration test
`test_diff_channel_chain_reconstruction` that skips gracefully when
`langgraph` core is not installed.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-04-22 14:03:36 -04:00
Sydney RunkleandClaude Sonnet 4.6 65d6ab2609 feat(checkpoint/postgres): diff chain reconstruction in _load_blobs (sync)
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-04-22 14:03:36 -04:00
Sydney RunkleandClaude Sonnet 4.6 888a308814 test(pregel): strengthen DiffChannel time-travel and reply assertions
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-04-22 14:03:36 -04:00
Sydney RunkleandClaude Sonnet 4.6 c7086ed7e2 feat(pregel): call after_checkpoint hook when loading and saving channels
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-04-22 14:03:36 -04:00
Sydney RunkleandClaude Sonnet 4.6 e645c2a085 fix(checkpoint/memory): warn on broken diff chain, guard against cycles
- Add logger.warning when a mid-chain blob is missing (fixes silent truncation bug)
- Add cycle guard to prevent infinite loops on corrupt blob stores
- Fix type annotation on diff_channels from dict[str, Any] to dict[str, str]

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-04-22 14:03:36 -04:00
Sydney RunkleandClaude Sonnet 4.6 e52b7b2d54 feat(checkpoint/memory): chain-traverse diff blobs in _load_blobs
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-04-22 14:03:36 -04:00
Sydney Runkle 6581fdd0a5 fix(channels): align DiffChannel.is_available with BinaryOperatorAggregate 2026-04-22 14:03:36 -04:00
Sydney RunkleandClaude Sonnet 4.6 fe2bc286fc feat(channels): implement DiffChannel for incremental checkpoint storage
Adds DiffChannel, a new channel type that stores only per-step write
deltas in checkpoints and reconstructs the full list by replaying the
chain through the operator at load time.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-04-22 14:03:36 -04:00
Sydney Runkle c345d337bb feat(channels): add no-op after_checkpoint hook to BaseChannel 2026-04-22 14:03:36 -04:00
Sydney RunkleandClaude Sonnet 4.6 d0af83b746 fix(checkpoint/serde): use lazy isinstance check for DiffDelta
Replace duck-typing check with lazy import inside _is_diff_delta helper
function to avoid module-level circular dependency while using proper
isinstance semantics.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-04-22 14:03:36 -04:00
Sydney RunkleandClaude Sonnet 4.5 eabf926a4f feat(checkpoint/serde): serialize DiffDelta as 'diff' type tag
Add serde support for DiffDelta by implementing dump/load for the "diff" type tag.
This allows the checkpoint system to efficiently store delta objects by serializing
them as msgpack-encoded dicts with {"d": delta, "p": prev_version} structure.

The implementation uses runtime type checking to avoid circular imports and
leverages the existing msgpack ext hooks for proper deserialization of complex
types like LangChain messages.

Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>
2026-04-22 14:03:36 -04:00
Sydney RunkleandClaude Sonnet 4.6 6852e0478e feat(checkpoint): add DiffDelta and DiffChainValue protocol types
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-04-22 14:03:36 -04:00
Sydney RunkleandClaude Sonnet 4.6 68b2f7bfac docs: add DiffChannel implementation plan
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-04-22 14:03:36 -04:00
Sydney RunkleandClaude Sonnet 4.6 95e0fe060c docs: add DiffChannel incremental checkpoint storage design spec
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-04-22 14:03:36 -04:00
Sydney RunkleandGitHub a529b9bede chore(langgraph): bump version 1.1.8 -> 1.1.9 (#7563)
## Summary

- Bumps `langgraph` patch version from `1.1.8` to `1.1.9` in
`libs/langgraph/pyproject.toml`

## Test plan

- [ ] Verify version string is correct in `pyproject.toml`
- [ ] Confirm release workflow triggers on merge
2026-04-21 09:38:58 -04:00
0a26b471d3 fix(langgraph): don't propagate ReplayState to subgraphs on plain resume (#7561)
**Description:**
When clients resume an interrupted subgraph with `Command(resume=...)`
plus an explicit `checkpoint_id` in the config (the pattern LangGraph
Studio and the API server emit on every resume), the subgraph restarts
from its first node instead of continuing at the interrupted node.

Fix: gate `ReplayState` propagation on `is_time_traveling` rather than
`is_replaying`, so a resume that happens to carry a head checkpoint_id
behaves the same as one with just a thread_id.

**Verification:** added regression test
`test_subgraph_interrupt_resume_with_explicit_head_checkpoint_id` (fails
on main, passes with fix, across memory/sqlite/sqlite_aes).
Full `test_time_travel.py`, `test_time_travel_async.py`,
`test_interruption.py`, and all pregel subgraph/interrupt/resume/replay
tests still pass.

Co-authored-by: Jessie Ibarra <jessie.ibarra@langgraph.dev>
2026-04-20 21:29:18 -04:00
Eugene YurtsevandGitHub b674dd4622 feat(prebuilt): expose available tools on ToolRuntime (#7512) 2026-04-17 16:54:16 -04:00
Eugene YurtsevandGitHub 8df0a377d0 chore(langgraph): undo unnecessary changes in stream handler (#7536)
This change is no longer necessary as langchain-core will continue
copying checkpoint_ns for backwards compatibility
2026-04-17 16:22:34 -04:00
dependabot[bot]GitHubdependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
216cf33a54 chore(deps): bump the pip group across 3 directories with 1 update (#7537)
Bumps the pip group with 1 update in the
/libs/cli/examples/graph_prerelease_reqs/deps/zuper_deps directory:
[langchain-openai](https://github.com/langchain-ai/langchain).
Bumps the pip group with 1 update in the
/libs/cli/examples/graph_prerelease_reqs_fail directory:
[langchain-openai](https://github.com/langchain-ai/langchain).
Bumps the pip group with 1 update in the
/libs/cli/examples/graph_prerelease_reqs directory:
[langchain-openai](https://github.com/langchain-ai/langchain).

Updates `langchain-openai` from 1.0.1 to 1.1.14
<details>
<summary>Release notes</summary>
<p><em>Sourced from <a
href="https://github.com/langchain-ai/langchain/releases">langchain-openai's
releases</a>.</em></p>
<blockquote>
<h2>langchain-openai==1.1.14</h2>
<p>Changes since langchain-openai==1.1.13</p>
<p>release(openai): 1.1.14 (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/36820">#36820</a>)
fix(openai): use SSRF-safe transport for image token counting (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/36819">#36819</a>)
chore(deps): bump pytest to <code>9.0.3</code> (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/36801">#36801</a>)
chore: bump langsmith from 0.6.3 to 0.7.31 in /libs/partners/openai (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/36795">#36795</a>)
chore: bump pillow from 12.1.1 to 12.2.0 in /libs/partners/openai (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/36777">#36777</a>)</p>
<h2>langchain-openai==1.1.13</h2>
<p>Changes since langchain-openai==1.1.12</p>
<p>release(openai): 1.1.13 (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/36729">#36729</a>)
fix(openai): handle content blocks without type key in responses api
conversion (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/36725">#36725</a>)
chore(model-profiles): refresh model profile data (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/36539">#36539</a>)
chore(openai): fix broken vcr cassette playback and add ci guard (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/36502">#36502</a>)
fix(openai,groq,openrouter): use is-not-None checks in usage metadata
token extraction (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/36500">#36500</a>)
fix(core): fixed typos in the documentation (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/36459">#36459</a>)
chore(model-profiles): refresh model profile data (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/36455">#36455</a>)
feat(core): impute placeholder filenames for OpenAI file inputs (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/36433">#36433</a>)
chore: pygments&gt;=2.20.0 across all packages (CVE-2026-4539) (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/36385">#36385</a>)
chore(model-profiles): refresh model profile data (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/36368">#36368</a>)
fix(openai): update computer call test (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/36352">#36352</a>)
fix(openai): let user-provided User-Agent override the Azure default (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/35523">#35523</a>)
chore: bump requests from 2.32.5 to 2.33.0 in /libs/partners/openai (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/36248">#36248</a>)</p>
<h2>langchain-openai==1.1.12</h2>
<p>Changes since langchain-openai==1.1.11</p>
<p>fix(openai): bump min core version (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/36180">#36180</a>)
release(openai): 1.1.12 (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/36178">#36178</a>)
fix(core,model-profiles): add missing <code>ModelProfile</code> fields,
warn on schema drift (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/36129">#36129</a>)
fix(openai): support phase parameter (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/36161">#36161</a>)
fix(openai): preserve namespace field in streaming function_call chunks
(<a
href="https://redirect.github.com/langchain-ai/langchain/issues/36108">#36108</a>)
ci: suppress pytest streaming output in CI (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/36092">#36092</a>)
ci: avoid unnecessary dep installs in lint targets (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/36046">#36046</a>)
chore(model-profiles): refresh model profile data (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/36039">#36039</a>)
chore: bump orjson from 3.11.5 to 3.11.6 in /libs/partners/openai (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/35860">#35860</a>)
fix(openai): add type: message to Responses API input items (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/35693">#35693</a>)
perf(.github): set a timeout on get min versions HTTP calls (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/35851">#35851</a>)
feat(model-profiles): new fields + <code>Makefile</code> target (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/35788">#35788</a>)
fix(openai): close PIL Image handles in token counting to prevent fd
leak (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/35742">#35742</a>)
fix(openai): typo (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/35763">#35763</a>)
chore(model-profiles): refresh model profile data (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/35754">#35754</a>)</p>
<h2>langchain-openai==1.1.11</h2>
<p>Changes since langchain-openai==1.1.10</p>
<p>fix(openai): bump min core version (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/35705">#35705</a>)
release(openai): 1.1.11 (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/35703">#35703</a>)</p>
<!-- raw HTML omitted -->
</blockquote>
<p>... (truncated)</p>
</details>
<details>
<summary>Commits</summary>
<ul>
<li><a
href="https://github.com/langchain-ai/langchain/commit/b7447c6969fc928ec3f29c200e2e56c0a46c4c77"><code>b7447c6</code></a>
fix(infra): skip serdes tests in min-version release step (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/36818">#36818</a>)</li>
<li><a
href="https://github.com/langchain-ai/langchain/commit/41c0cc58b0dac82000d24715f7a4b44dc8b01fd3"><code>41c0cc5</code></a>
release(openai): 1.1.14 (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/36820">#36820</a>)</li>
<li><a
href="https://github.com/langchain-ai/langchain/commit/0516156ef98f5001129f6d47bc8682d6536d58fb"><code>0516156</code></a>
fix(openai): use SSRF-safe transport for image token counting (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/36819">#36819</a>)</li>
<li><a
href="https://github.com/langchain-ai/langchain/commit/338aa8131a8124e7aa1e042616ccd2366ff9f699"><code>338aa81</code></a>
fix(core): restore cloud metadata IPs and link-local range in SSRF
policy (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/3">#3</a>...</li>
<li><a
href="https://github.com/langchain-ai/langchain/commit/51e954877efd2d2c3c5bf09364dcfec8794eadb0"><code>51e9548</code></a>
chore: bump langsmith from 0.6.3 to 0.7.31 in /libs/text-splitters (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/36797">#36797</a>)</li>
<li><a
href="https://github.com/langchain-ai/langchain/commit/e85c418cfa559d4a794ddc6db92c6febab44651c"><code>e85c418</code></a>
chore: bump langsmith from 0.6.3 to 0.7.31 in /libs/model-profiles (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/36798">#36798</a>)</li>
<li><a
href="https://github.com/langchain-ai/langchain/commit/789126e6c78ad74664bea26228dda6e72e135dce"><code>789126e</code></a>
chore: bump langsmith from 0.6.3 to 0.7.31 in /libs/standard-tests (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/36799">#36799</a>)</li>
<li><a
href="https://github.com/langchain-ai/langchain/commit/937b3eb3827551d17ee4736f9acc4aa57e88c716"><code>937b3eb</code></a>
chore: bump langsmith from 0.6.3 to 0.7.31 in /libs/langchain_v1 (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/36800">#36800</a>)</li>
<li><a
href="https://github.com/langchain-ai/langchain/commit/a06c205738cf5953e28c37287ddb1559d67c01f6"><code>a06c205</code></a>
ci(infra): validate issue checkboxes by section (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/36811">#36811</a>)</li>
<li><a
href="https://github.com/langchain-ai/langchain/commit/aa33b06deb0d65489ce254b48a8aaf8a86304c18"><code>aa33b06</code></a>
fix(langchain-classic): suppress mypy errors in compat code (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/36806">#36806</a>)</li>
<li>Additional commits viewable in <a
href="https://github.com/langchain-ai/langchain/compare/langchain-openai==1.0.1...langchain-openai==1.1.14">compare
view</a></li>
</ul>
</details>
<br />

Updates `langchain-openai` from 1.0.0a2 to 1.1.14
<details>
<summary>Release notes</summary>
<p><em>Sourced from <a
href="https://github.com/langchain-ai/langchain/releases">langchain-openai's
releases</a>.</em></p>
<blockquote>
<h2>langchain-openai==1.1.14</h2>
<p>Changes since langchain-openai==1.1.13</p>
<p>release(openai): 1.1.14 (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/36820">#36820</a>)
fix(openai): use SSRF-safe transport for image token counting (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/36819">#36819</a>)
chore(deps): bump pytest to <code>9.0.3</code> (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/36801">#36801</a>)
chore: bump langsmith from 0.6.3 to 0.7.31 in /libs/partners/openai (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/36795">#36795</a>)
chore: bump pillow from 12.1.1 to 12.2.0 in /libs/partners/openai (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/36777">#36777</a>)</p>
<h2>langchain-openai==1.1.13</h2>
<p>Changes since langchain-openai==1.1.12</p>
<p>release(openai): 1.1.13 (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/36729">#36729</a>)
fix(openai): handle content blocks without type key in responses api
conversion (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/36725">#36725</a>)
chore(model-profiles): refresh model profile data (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/36539">#36539</a>)
chore(openai): fix broken vcr cassette playback and add ci guard (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/36502">#36502</a>)
fix(openai,groq,openrouter): use is-not-None checks in usage metadata
token extraction (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/36500">#36500</a>)
fix(core): fixed typos in the documentation (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/36459">#36459</a>)
chore(model-profiles): refresh model profile data (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/36455">#36455</a>)
feat(core): impute placeholder filenames for OpenAI file inputs (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/36433">#36433</a>)
chore: pygments&gt;=2.20.0 across all packages (CVE-2026-4539) (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/36385">#36385</a>)
chore(model-profiles): refresh model profile data (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/36368">#36368</a>)
fix(openai): update computer call test (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/36352">#36352</a>)
fix(openai): let user-provided User-Agent override the Azure default (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/35523">#35523</a>)
chore: bump requests from 2.32.5 to 2.33.0 in /libs/partners/openai (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/36248">#36248</a>)</p>
<h2>langchain-openai==1.1.12</h2>
<p>Changes since langchain-openai==1.1.11</p>
<p>fix(openai): bump min core version (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/36180">#36180</a>)
release(openai): 1.1.12 (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/36178">#36178</a>)
fix(core,model-profiles): add missing <code>ModelProfile</code> fields,
warn on schema drift (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/36129">#36129</a>)
fix(openai): support phase parameter (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/36161">#36161</a>)
fix(openai): preserve namespace field in streaming function_call chunks
(<a
href="https://redirect.github.com/langchain-ai/langchain/issues/36108">#36108</a>)
ci: suppress pytest streaming output in CI (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/36092">#36092</a>)
ci: avoid unnecessary dep installs in lint targets (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/36046">#36046</a>)
chore(model-profiles): refresh model profile data (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/36039">#36039</a>)
chore: bump orjson from 3.11.5 to 3.11.6 in /libs/partners/openai (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/35860">#35860</a>)
fix(openai): add type: message to Responses API input items (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/35693">#35693</a>)
perf(.github): set a timeout on get min versions HTTP calls (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/35851">#35851</a>)
feat(model-profiles): new fields + <code>Makefile</code> target (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/35788">#35788</a>)
fix(openai): close PIL Image handles in token counting to prevent fd
leak (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/35742">#35742</a>)
fix(openai): typo (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/35763">#35763</a>)
chore(model-profiles): refresh model profile data (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/35754">#35754</a>)</p>
<h2>langchain-openai==1.1.11</h2>
<p>Changes since langchain-openai==1.1.10</p>
<p>fix(openai): bump min core version (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/35705">#35705</a>)
release(openai): 1.1.11 (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/35703">#35703</a>)</p>
<!-- raw HTML omitted -->
</blockquote>
<p>... (truncated)</p>
</details>
<details>
<summary>Commits</summary>
<ul>
<li><a
href="https://github.com/langchain-ai/langchain/commit/b7447c6969fc928ec3f29c200e2e56c0a46c4c77"><code>b7447c6</code></a>
fix(infra): skip serdes tests in min-version release step (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/36818">#36818</a>)</li>
<li><a
href="https://github.com/langchain-ai/langchain/commit/41c0cc58b0dac82000d24715f7a4b44dc8b01fd3"><code>41c0cc5</code></a>
release(openai): 1.1.14 (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/36820">#36820</a>)</li>
<li><a
href="https://github.com/langchain-ai/langchain/commit/0516156ef98f5001129f6d47bc8682d6536d58fb"><code>0516156</code></a>
fix(openai): use SSRF-safe transport for image token counting (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/36819">#36819</a>)</li>
<li><a
href="https://github.com/langchain-ai/langchain/commit/338aa8131a8124e7aa1e042616ccd2366ff9f699"><code>338aa81</code></a>
fix(core): restore cloud metadata IPs and link-local range in SSRF
policy (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/3">#3</a>...</li>
<li><a
href="https://github.com/langchain-ai/langchain/commit/51e954877efd2d2c3c5bf09364dcfec8794eadb0"><code>51e9548</code></a>
chore: bump langsmith from 0.6.3 to 0.7.31 in /libs/text-splitters (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/36797">#36797</a>)</li>
<li><a
href="https://github.com/langchain-ai/langchain/commit/e85c418cfa559d4a794ddc6db92c6febab44651c"><code>e85c418</code></a>
chore: bump langsmith from 0.6.3 to 0.7.31 in /libs/model-profiles (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/36798">#36798</a>)</li>
<li><a
href="https://github.com/langchain-ai/langchain/commit/789126e6c78ad74664bea26228dda6e72e135dce"><code>789126e</code></a>
chore: bump langsmith from 0.6.3 to 0.7.31 in /libs/standard-tests (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/36799">#36799</a>)</li>
<li><a
href="https://github.com/langchain-ai/langchain/commit/937b3eb3827551d17ee4736f9acc4aa57e88c716"><code>937b3eb</code></a>
chore: bump langsmith from 0.6.3 to 0.7.31 in /libs/langchain_v1 (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/36800">#36800</a>)</li>
<li><a
href="https://github.com/langchain-ai/langchain/commit/a06c205738cf5953e28c37287ddb1559d67c01f6"><code>a06c205</code></a>
ci(infra): validate issue checkboxes by section (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/36811">#36811</a>)</li>
<li><a
href="https://github.com/langchain-ai/langchain/commit/aa33b06deb0d65489ce254b48a8aaf8a86304c18"><code>aa33b06</code></a>
fix(langchain-classic): suppress mypy errors in compat code (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/36806">#36806</a>)</li>
<li>Additional commits viewable in <a
href="https://github.com/langchain-ai/langchain/compare/langchain-openai==1.0.1...langchain-openai==1.1.14">compare
view</a></li>
</ul>
</details>
<br />

Updates `langchain-openai` from 1.0.0a2 to 1.1.14
<details>
<summary>Release notes</summary>
<p><em>Sourced from <a
href="https://github.com/langchain-ai/langchain/releases">langchain-openai's
releases</a>.</em></p>
<blockquote>
<h2>langchain-openai==1.1.14</h2>
<p>Changes since langchain-openai==1.1.13</p>
<p>release(openai): 1.1.14 (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/36820">#36820</a>)
fix(openai): use SSRF-safe transport for image token counting (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/36819">#36819</a>)
chore(deps): bump pytest to <code>9.0.3</code> (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/36801">#36801</a>)
chore: bump langsmith from 0.6.3 to 0.7.31 in /libs/partners/openai (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/36795">#36795</a>)
chore: bump pillow from 12.1.1 to 12.2.0 in /libs/partners/openai (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/36777">#36777</a>)</p>
<h2>langchain-openai==1.1.13</h2>
<p>Changes since langchain-openai==1.1.12</p>
<p>release(openai): 1.1.13 (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/36729">#36729</a>)
fix(openai): handle content blocks without type key in responses api
conversion (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/36725">#36725</a>)
chore(model-profiles): refresh model profile data (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/36539">#36539</a>)
chore(openai): fix broken vcr cassette playback and add ci guard (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/36502">#36502</a>)
fix(openai,groq,openrouter): use is-not-None checks in usage metadata
token extraction (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/36500">#36500</a>)
fix(core): fixed typos in the documentation (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/36459">#36459</a>)
chore(model-profiles): refresh model profile data (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/36455">#36455</a>)
feat(core): impute placeholder filenames for OpenAI file inputs (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/36433">#36433</a>)
chore: pygments&gt;=2.20.0 across all packages (CVE-2026-4539) (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/36385">#36385</a>)
chore(model-profiles): refresh model profile data (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/36368">#36368</a>)
fix(openai): update computer call test (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/36352">#36352</a>)
fix(openai): let user-provided User-Agent override the Azure default (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/35523">#35523</a>)
chore: bump requests from 2.32.5 to 2.33.0 in /libs/partners/openai (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/36248">#36248</a>)</p>
<h2>langchain-openai==1.1.12</h2>
<p>Changes since langchain-openai==1.1.11</p>
<p>fix(openai): bump min core version (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/36180">#36180</a>)
release(openai): 1.1.12 (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/36178">#36178</a>)
fix(core,model-profiles): add missing <code>ModelProfile</code> fields,
warn on schema drift (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/36129">#36129</a>)
fix(openai): support phase parameter (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/36161">#36161</a>)
fix(openai): preserve namespace field in streaming function_call chunks
(<a
href="https://redirect.github.com/langchain-ai/langchain/issues/36108">#36108</a>)
ci: suppress pytest streaming output in CI (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/36092">#36092</a>)
ci: avoid unnecessary dep installs in lint targets (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/36046">#36046</a>)
chore(model-profiles): refresh model profile data (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/36039">#36039</a>)
chore: bump orjson from 3.11.5 to 3.11.6 in /libs/partners/openai (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/35860">#35860</a>)
fix(openai): add type: message to Responses API input items (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/35693">#35693</a>)
perf(.github): set a timeout on get min versions HTTP calls (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/35851">#35851</a>)
feat(model-profiles): new fields + <code>Makefile</code> target (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/35788">#35788</a>)
fix(openai): close PIL Image handles in token counting to prevent fd
leak (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/35742">#35742</a>)
fix(openai): typo (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/35763">#35763</a>)
chore(model-profiles): refresh model profile data (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/35754">#35754</a>)</p>
<h2>langchain-openai==1.1.11</h2>
<p>Changes since langchain-openai==1.1.10</p>
<p>fix(openai): bump min core version (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/35705">#35705</a>)
release(openai): 1.1.11 (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/35703">#35703</a>)</p>
<!-- raw HTML omitted -->
</blockquote>
<p>... (truncated)</p>
</details>
<details>
<summary>Commits</summary>
<ul>
<li><a
href="https://github.com/langchain-ai/langchain/commit/b7447c6969fc928ec3f29c200e2e56c0a46c4c77"><code>b7447c6</code></a>
fix(infra): skip serdes tests in min-version release step (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/36818">#36818</a>)</li>
<li><a
href="https://github.com/langchain-ai/langchain/commit/41c0cc58b0dac82000d24715f7a4b44dc8b01fd3"><code>41c0cc5</code></a>
release(openai): 1.1.14 (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/36820">#36820</a>)</li>
<li><a
href="https://github.com/langchain-ai/langchain/commit/0516156ef98f5001129f6d47bc8682d6536d58fb"><code>0516156</code></a>
fix(openai): use SSRF-safe transport for image token counting (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/36819">#36819</a>)</li>
<li><a
href="https://github.com/langchain-ai/langchain/commit/338aa8131a8124e7aa1e042616ccd2366ff9f699"><code>338aa81</code></a>
fix(core): restore cloud metadata IPs and link-local range in SSRF
policy (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/3">#3</a>...</li>
<li><a
href="https://github.com/langchain-ai/langchain/commit/51e954877efd2d2c3c5bf09364dcfec8794eadb0"><code>51e9548</code></a>
chore: bump langsmith from 0.6.3 to 0.7.31 in /libs/text-splitters (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/36797">#36797</a>)</li>
<li><a
href="https://github.com/langchain-ai/langchain/commit/e85c418cfa559d4a794ddc6db92c6febab44651c"><code>e85c418</code></a>
chore: bump langsmith from 0.6.3 to 0.7.31 in /libs/model-profiles (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/36798">#36798</a>)</li>
<li><a
href="https://github.com/langchain-ai/langchain/commit/789126e6c78ad74664bea26228dda6e72e135dce"><code>789126e</code></a>
chore: bump langsmith from 0.6.3 to 0.7.31 in /libs/standard-tests (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/36799">#36799</a>)</li>
<li><a
href="https://github.com/langchain-ai/langchain/commit/937b3eb3827551d17ee4736f9acc4aa57e88c716"><code>937b3eb</code></a>
chore: bump langsmith from 0.6.3 to 0.7.31 in /libs/langchain_v1 (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/36800">#36800</a>)</li>
<li><a
href="https://github.com/langchain-ai/langchain/commit/a06c205738cf5953e28c37287ddb1559d67c01f6"><code>a06c205</code></a>
ci(infra): validate issue checkboxes by section (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/36811">#36811</a>)</li>
<li><a
href="https://github.com/langchain-ai/langchain/commit/aa33b06deb0d65489ce254b48a8aaf8a86304c18"><code>aa33b06</code></a>
fix(langchain-classic): suppress mypy errors in compat code (<a
href="https://redirect.github.com/langchain-ai/langchain/issues/36806">#36806</a>)</li>
<li>Additional commits viewable in <a
href="https://github.com/langchain-ai/langchain/compare/langchain-openai==1.0.1...langchain-openai==1.1.14">compare
view</a></li>
</ul>
</details>
<br />


Dependabot will resolve any conflicts with this PR as long as you don't
alter it yourself. You can also trigger a rebase manually by commenting
`@dependabot rebase`.

[//]: # (dependabot-automerge-start)
[//]: # (dependabot-automerge-end)

---

<details>
<summary>Dependabot commands and options</summary>
<br />

You can trigger Dependabot actions by commenting on this PR:
- `@dependabot rebase` will rebase this PR
- `@dependabot recreate` will recreate this PR, overwriting any edits
that have been made to it
- `@dependabot show <dependency name> ignore conditions` will show all
of the ignore conditions of the specified dependency
- `@dependabot ignore <dependency name> major version` will close this
group update PR and stop Dependabot creating any more for the specific
dependency's major version (unless you unignore this specific
dependency's major version or upgrade to it yourself)
- `@dependabot ignore <dependency name> minor version` will close this
group update PR and stop Dependabot creating any more for the specific
dependency's minor version (unless you unignore this specific
dependency's minor version or upgrade to it yourself)
- `@dependabot ignore <dependency name>` will close this group update PR
and stop Dependabot creating any more for the specific dependency
(unless you unignore this specific dependency or upgrade to it yourself)
- `@dependabot unignore <dependency name>` will remove all of the ignore
conditions of the specified dependency
- `@dependabot unignore <dependency name> <ignore condition>` will
remove the ignore condition of the specified dependency and ignore
conditions
You can disable automated security fix PRs for this repo from the
[Security Alerts
page](https://github.com/langchain-ai/langgraph/network/alerts).

</details>

Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
2026-04-17 12:42:55 -07:00
Eugene YurtsevandGitHub 4956134a37 release(langgraph): 1.1.8 (#7545)
releasing 1.1.8
2026-04-17 19:41:13 +00:00
aa94790f36 fix(langgraph): remove strict add_handler type check that breaks OTel instrumentation (#7544)
## Summary

Removes the `add_handler()` overrides on `_GraphCallbackManager` and
`_AsyncGraphCallbackManager` that reject handlers not inheriting from
`GraphCallbackHandler`. This fixes a regression in 1.1.7 where
`opentelemetry-instrumentation-langchain` (and likely other libraries
that patch `BaseCallbackManager.__init__`) crash with `TypeError:
handlers must inherit GraphCallbackHandler` at invocation time.

## Why this is safe

The strict type check is redundant — `_configure_graph_callbacks` and
`_filter_graph_handlers` already filter handlers to
`GraphCallbackHandler` instances at construction time. Non-graph
handlers that enter via external patches (like OTel's monkey-patch) are
harmless because `handle_event("on_interrupt", ...)` /
`handle_event("on_resume", ...)` will simply no-op on handlers that
don't implement those methods.

## What changed

- Deleted `add_handler()` override from `_GraphCallbackManager` (was
lines 248-255)
- Deleted `add_handler()` override from `_AsyncGraphCallbackManager`
(was lines 324-331)
- No other changes — 18 lines removed, 0 added

## Test plan

- [x] All 8 existing `test_graph_callbacks.py` tests pass (`make test
TEST=tests/test_graph_callbacks.py`)
- [x] `make lint` passes
- [x] `make format` passes (no changes needed)
- [x] Verified fix locally: `LangchainInstrumentor().instrument()` +
`create_react_agent()` + `graph.ainvoke()` no longer raises `TypeError`
- [x] Verified the graph lifecycle callbacks (`on_interrupt`,
`on_resume`) still work correctly

Closes #7543

---------

Co-authored-by: Eugene Yurtsev <eyurtsev@gmail.com>
2026-04-17 15:33:10 -04:00
ccurmeandGitHub e002711ede release(prebuilt): 1.0.10 (#7541) 2026-04-17 13:52:20 -04:00
f44b49b33d chore: dedup warnings (#7257)
Co-authored-by: Will Fu-Hinthorn <will@langchain.dev>
2026-04-17 10:13:03 -07:00
a0a95df2ac release(cli): 0.4.23 (#7542)
Release Note: Increase the max bound for langgraph-api

Co-authored-by: Will Fu-Hinthorn <will@langchain.dev>
2026-04-17 16:38:14 +00:00
Eugene YurtsevandGitHub d194c18c06 release(langgraph): 1.1.7 (#7540)
release langgraph 1.1.7
2026-04-17 15:19:18 +00:00
Sydney Runkle eae916719f CLI bump 2026-04-16 13:48:02 -04:00
f093702e4e fix(prebuilt): handle injected NotRequired keys (#7392)
Resolves https://github.com/langchain-ai/langchain/issues/35585

This would previously raise KeyError:
```python
from typing import Annotated

from langchain_core.tools import tool
from langchain.agents import create_agent
from typing_extensions import NotRequired
from langgraph.prebuilt import InjectedState
from langchain.agents import AgentState


class CustomAgentState(AgentState):
    city: NotRequired[str]


@tool
def get_weather(city: Annotated[str | None, InjectedState("city")] = None) -> str:
    """Get weather for a given city."""
    if city is None:
        city = "Boston"
    return f"It's always sunny in {city}!"


agent = create_agent(
    model="claude-sonnet-4-6",
    tools=[get_weather],
    system_prompt="You are a helpful assistant",
    state_schema=CustomAgentState,
)

input_message = {
    "role": "user",
    "content": "What's the weather?",
}

result = agent.invoke({"messages": [input_message]})
for m in result["messages"]:
    m.pretty_print()
```

---------

Co-authored-by: Sydney Runkle <sydneymarierunkle@gmail.com>
2026-04-16 13:06:33 +00:00
Sydney RunkleandGitHub 51cbdbd5cd fix: time travel when going back to interrupt node (#7498)
# Fix: Create fork checkpoint on subgraph time travel

## Problem

When time-traveling to a subgraph checkpoint that has an interrupt, and
then resuming, the resume would load the **wrong state** — it would pick
up the original execution's latest checkpoint instead of the
time-traveled one.

This happened because replaying from a subgraph checkpoint never created
a new parent checkpoint. If the replay hit an interrupt before
`after_tick()` ran, no checkpoint was written at all, so the parent's
"latest" checkpoint was still the old one from the original execution.

## Fix

When the loop detects a time-travel replay (not an `update_state` fork),
it now **eagerly writes a fork checkpoint** at the start of the tick.
This ensures:

1. The parent thread's latest checkpoint points to the replayed state
2. Subsequent `Command(resume=...)` calls find the correct checkpoint
3. Stale `INTERRUPT` pending writes from the old checkpoint are cleared
(they reference old task IDs)

Additionally, the subgraph replay logic now uses the **parent checkpoint
ID** (from `prev_checkpoint_config`) when resolving subgraph checkpoints
during time-travel, matching the existing behavior for `update_state`
forks.

## Checkpoint flow diagrams

### Before fix: time travel leaves no fork

```
Original execution:

  C0 (start) --> C1 (step_a) --> C2 (ask_1 interrupt) --> C3 (resume) --> C4 (ask_2 interrupt) --> C5 (done)

Time travel to C2 (subgraph config):

  Replay runs... hits interrupt... no new checkpoint written.
  Parent "latest" is still C5.

  Command(resume="new_answer"):
    Loads C5 (wrong!) instead of the replayed C2 state.
```

### After fix: time travel creates a fork

```
Original execution:

  C0 --> C1 --> C2 --> C3 --> C4 --> C5 (done)

Time travel to C2 (subgraph config):

  C0 --> C1 --> C2 --> C3 --> C4 --> C5
                  \
                   F1 (fork, source="fork")  <-- new latest

  Command(resume="new_answer"):
    Loads F1 (correct!) --> resumes from the right state.

  After full resume:

  C0 --> C1 --> C2 --> C3 --> C4 --> C5
                  \
                   F1 --> F2 (ask_1 result) --> F3 (ask_2 interrupt) --> F4 (done)
```

### Manual fork via `update_state` (unchanged)

```
  C0 --> C1 --> C2 --> C3
                  \
                   U1 (source="update")  <-- created by update_state()

  This path already worked. The fix skips update/fork sources
  so existing behavior is preserved.
```

## Changes

- **`libs/langgraph/langgraph/pregel/_loop.py`**:
- Extract `is_time_traveling` flag from the existing replay detection
logic for reuse
- Write a fork checkpoint (`source="fork"`) eagerly at the start of a
time-travel tick, before execution begins
- Clear stale `INTERRUPT` pending writes when creating the fork (they
reference old task IDs that won't match the new checkpoint)
- Unify subgraph replay ID resolution: check `source in ("update",
"fork")` instead of a separate `is_time_traveling` condition, since the
new fork checkpoint now has `source="fork"`
- **`libs/langgraph/tests/test_time_travel.py`** and
**`test_time_travel_async.py`**: Added 4 new test cases (sync + async):
- `test_replay_from_before_interrupt_then_resume` — replays from a
checkpoint before an interrupt, resumes with a new answer, and verifies
the full checkpoint history (source, next, values) at each stage
- `test_subgraph_time_travel_resume_from_first_interrupt` — time-travels
to a subgraph's first interrupt, resumes both interrupts with new
answers, and verifies the fork creates a new branch while preserving the
original
- `test_subgraph_time_travel_resume_from_second_interrupt` —
time-travels to a subgraph's second interrupt, resumes with a new
answer, and verifies the first interrupt's original answer is preserved
- `test_subgraph_time_travel_checkpoint_pattern` — verifies the fork
checkpoint branches from the correct replay point and that the full
checkpoint tree is correct after resume
- **`libs/langgraph/tests/test_pregel.py`** /
**`test_pregel_async.py`**: Updated existing
`test_weather_subgraph_state` to account for the new fork checkpoint
appearing in history (history length increases by 1)
2026-04-16 08:29:48 -04:00
hari-dhanushkodiandGitHub 4d64227c13 chore: start tracking cli deploy source (#7520)
Fixes #

<!-- Replace everything above this line with a 1-2 sentence description
of your change. Keep the "Fixes #xx" keyword and update the issue
number. -->

Read the full contributing guidelines:
https://docs.langchain.com/oss/python/contributing/overview

> **All contributions must be in English.** See the [language
policy](https://docs.langchain.com/oss/python/contributing/overview#language-policy).

If you paste a large clearly AI generated description here your PR may
be IGNORED or CLOSED!

Thank you for contributing to LangGraph! Follow these steps to have your
pull request considered as ready for review.

1. PR title: Should follow the format: TYPE(SCOPE): DESCRIPTION

    - feat(langgraph): add multi-tenant support
- Allowed TYPE and SCOPE values:
https://github.com/langchain-ai/langgraph/blob/main/.github/workflows/pr_lint.yml#L19-L43

2. PR description:

  - Write 1-2 sentences summarizing the change.
- The `Fixes #xx` line at the top is **required** for external
contributions — update the issue number and keep the keyword. This links
your PR to the approved issue and auto-closes it on merge.
  - If there are any breaking changes, please clearly describe them.
- If this PR depends on another PR being merged first, please include
"Depends on #PR_NUMBER" in the description.

3. Run `make format`, `make lint` and `make test` from the root of the
package(s) you've modified.

  - We will not consider a PR unless these three are passing in CI.

4. How did you verify your code works?

Additional guidelines:

- All external PRs must link to an issue or discussion where a solution
has been approved by a maintainer, and you must be assigned to that
issue. PRs without prior approval will be closed.
- PRs should not touch more than one package unless absolutely
necessary.
- Do not update the `uv.lock` files or add dependencies to
`pyproject.toml` files (even optional ones) unless you have explicit
permission to do so by a maintainer.

## Social handles (optional)
<!-- If you'd like a shoutout on release, add your socials below -->
Twitter: @
LinkedIn: https://linkedin.com/in/
2026-04-16 08:24:47 -04:00
dependabot[bot]GitHubdependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
6bcac5d72e chore(deps): bump langsmith from 0.6.4 to 0.7.31 in /libs/prebuilt (#7530)
Bumps [langsmith](https://github.com/langchain-ai/langsmith-sdk) from
0.6.4 to 0.7.31.
<details>
<summary>Release notes</summary>
<p><em>Sourced from <a
href="https://github.com/langchain-ai/langsmith-sdk/releases">langsmith's
releases</a>.</em></p>
<blockquote>
<h2>v0.7.31</h2>
<h2>What's Changed</h2>
<ul>
<li>chore(deps-dev): bump langchain-core from 1.2.23 to 1.2.28 in
/python by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2692">langchain-ai/langsmith-sdk#2692</a></li>
<li>chore(deps-dev): bump <code>@​anthropic-ai/sdk</code> from 0.82.0 to
0.84.0 in /js by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2684">langchain-ai/langsmith-sdk#2684</a></li>
<li>chore(deps): bump cryptography from 46.0.6 to 46.0.7 in /python by
<a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2693">langchain-ai/langsmith-sdk#2693</a></li>
<li>chore(deps-dev): bump <code>@​anthropic-ai/sdk</code> from 0.84.0 to
0.85.0 in /js by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2700">langchain-ai/langsmith-sdk#2700</a></li>
<li>feat(py): Tag OpenAI Agent Python SDK runs with ls_agent_type by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2699">langchain-ai/langsmith-sdk#2699</a></li>
<li>feat(js): Adds ls_agent_type metadata to AI SDK runs by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2701">langchain-ai/langsmith-sdk#2701</a></li>
<li>chore(deps-dev): bump types-tqdm from 4.67.3.20260303 to
4.67.3.20260408 in /python by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2710">langchain-ai/langsmith-sdk#2710</a></li>
<li>chore(deps): bump pnpm/action-setup from 5 to 6 by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2705">langchain-ai/langsmith-sdk#2705</a></li>
<li>chore(deps): bump the py-minor-and-patch group across 1 directory
with 10 updates by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2711">langchain-ai/langsmith-sdk#2711</a></li>
<li>chore(deps-dev): bump <code>@​anthropic-ai/sdk</code> from 0.85.0 to
0.86.0 in /js by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2702">langchain-ai/langsmith-sdk#2702</a></li>
<li>chore(deps): bump actions/github-script from 8 to 9 by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2706">langchain-ai/langsmith-sdk#2706</a></li>
<li>chore(deps-dev): bump the js-minor-and-patch group across 1
directory with 7 updates by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2712">langchain-ai/langsmith-sdk#2712</a></li>
<li>chore(deps-dev): bump types-psutil from 7.2.2.20260130 to
7.2.2.20260408 in /python by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2709">langchain-ai/langsmith-sdk#2709</a></li>
<li>chore(deps-dev): bump rich from 14.3.3 to 15.0.0 in /python by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2708">langchain-ai/langsmith-sdk#2708</a></li>
<li>feat: Filter kwargs from new token events by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2714">langchain-ai/langsmith-sdk#2714</a></li>
<li>release(py): 0.7.31 by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2716">langchain-ai/langsmith-sdk#2716</a></li>
</ul>
<p><strong>Full Changelog</strong>: <a
href="https://github.com/langchain-ai/langsmith-sdk/compare/v0.7.30...v0.7.31">https://github.com/langchain-ai/langsmith-sdk/compare/v0.7.30...v0.7.31</a></p>
<h2>v0.7.30</h2>
<h2>What's Changed</h2>
<ul>
<li>feat(python): add service feature to sandbox by <a
href="https://github.com/DanielKneipp"><code>@​DanielKneipp</code></a>
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2665">langchain-ai/langsmith-sdk#2665</a></li>
<li>fix(js): Fix prototype pollution bug in anonymizers by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2690">langchain-ai/langsmith-sdk#2690</a></li>
<li>release(js): 0.5.18 by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2691">langchain-ai/langsmith-sdk#2691</a></li>
<li>chore(js/sandbox): suppress warning log by <a
href="https://github.com/hntrl"><code>@​hntrl</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2694">langchain-ai/langsmith-sdk#2694</a></li>
<li>feat(js): Add metadata to Claude Agent SDK JS tracing by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2695">langchain-ai/langsmith-sdk#2695</a></li>
<li>fix(py): Fix run tree memory leak by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2696">langchain-ai/langsmith-sdk#2696</a></li>
<li>release(py): 0.7.30 by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2698">langchain-ai/langsmith-sdk#2698</a></li>
</ul>
<p><strong>Full Changelog</strong>: <a
href="https://github.com/langchain-ai/langsmith-sdk/compare/v0.7.29...v0.7.30">https://github.com/langchain-ai/langsmith-sdk/compare/v0.7.29...v0.7.30</a></p>
<h2>v0.7.29</h2>
<h2>What's Changed</h2>
<ul>
<li>release(js): 0.5.17 by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2681">langchain-ai/langsmith-sdk#2681</a></li>
<li>feat(py): Fix race condition around Claude Agent SDK instrumentation
by <a href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a>
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2685">langchain-ai/langsmith-sdk#2685</a></li>
<li>release(py): 0.7.29 by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2686">langchain-ai/langsmith-sdk#2686</a></li>
</ul>
<p><strong>Full Changelog</strong>: <a
href="https://github.com/langchain-ai/langsmith-sdk/compare/v0.7.28...v0.7.29">https://github.com/langchain-ai/langsmith-sdk/compare/v0.7.28...v0.7.29</a></p>
<h2>v0.7.28</h2>
<h2>What's Changed</h2>
<ul>
<li>feat(py): Support subagent tracing in Claude Agents SDK, fix usage
and duplicate messages by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2670">langchain-ai/langsmith-sdk#2670</a></li>
<li>chore(deps-dev): bump the py-minor-and-patch group across 1
directory with 11 updates by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2677">langchain-ai/langsmith-sdk#2677</a></li>
<li>chore(deps-dev): bump the js-minor-and-patch group across 1
directory with 8 updates by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2667">langchain-ai/langsmith-sdk#2667</a></li>
<li>chore(deps): bump pnpm/action-setup from 4 to 5 by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2658">langchain-ai/langsmith-sdk#2658</a></li>
</ul>
<!-- raw HTML omitted -->
</blockquote>
<p>... (truncated)</p>
</details>
<details>
<summary>Commits</summary>
<ul>
<li><a
href="https://github.com/langchain-ai/langsmith-sdk/commit/c434999d05c00334efeba88b8bbd2de9f3afbef6"><code>c434999</code></a>
release(py): 0.7.31 (<a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/issues/2716">#2716</a>)</li>
<li><a
href="https://github.com/langchain-ai/langsmith-sdk/commit/47d7c4a783333e716395d802e7632f1f1b4744d3"><code>47d7c4a</code></a>
feat: Filter kwargs from new token events (<a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/issues/2714">#2714</a>)</li>
<li><a
href="https://github.com/langchain-ai/langsmith-sdk/commit/3c57445b543c9a2f86db52024ea2c998bfc2ffab"><code>3c57445</code></a>
chore(deps-dev): bump rich from 14.3.3 to 15.0.0 in /python (<a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/issues/2708">#2708</a>)</li>
<li><a
href="https://github.com/langchain-ai/langsmith-sdk/commit/2be6cd01a2b6e35e811488d3561e7b0b57b06f63"><code>2be6cd0</code></a>
chore(deps-dev): bump types-psutil from 7.2.2.20260130 to 7.2.2.20260408
in /...</li>
<li><a
href="https://github.com/langchain-ai/langsmith-sdk/commit/b8b6ca32d43c919c07a4e13c99a83bcaab8accb0"><code>b8b6ca3</code></a>
chore(deps-dev): bump the js-minor-and-patch group across 1 directory
with 7 ...</li>
<li><a
href="https://github.com/langchain-ai/langsmith-sdk/commit/9897cb33da7698291637f268edd833ca3e1adde6"><code>9897cb3</code></a>
chore(deps): bump actions/github-script from 8 to 9 (<a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/issues/2706">#2706</a>)</li>
<li><a
href="https://github.com/langchain-ai/langsmith-sdk/commit/572c0184285747e027a796e03ea6c9ba171e09a6"><code>572c018</code></a>
chore(deps-dev): bump <code>@​anthropic-ai/sdk</code> from 0.85.0 to
0.86.0 in /js (<a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/issues/2702">#2702</a>)</li>
<li><a
href="https://github.com/langchain-ai/langsmith-sdk/commit/57447524c88b6bba2775161aa449da32fb8e5c42"><code>5744752</code></a>
chore(deps): bump the py-minor-and-patch group across 1 directory with
10 upd...</li>
<li><a
href="https://github.com/langchain-ai/langsmith-sdk/commit/960cae7f490e9ccbe428e6b56c8047bdb7b942a5"><code>960cae7</code></a>
chore(deps): bump pnpm/action-setup from 5 to 6 (<a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/issues/2705">#2705</a>)</li>
<li><a
href="https://github.com/langchain-ai/langsmith-sdk/commit/9370e7670abf7f8f9a36fbb72250bcfd2f91e7c6"><code>9370e76</code></a>
chore(deps-dev): bump types-tqdm from 4.67.3.20260303 to 4.67.3.20260408
in /...</li>
<li>Additional commits viewable in <a
href="https://github.com/langchain-ai/langsmith-sdk/compare/v0.6.4...v0.7.31">compare
view</a></li>
</ul>
</details>
<br />


[![Dependabot compatibility
score](https://dependabot-badges.githubapp.com/badges/compatibility_score?dependency-name=langsmith&package-manager=uv&previous-version=0.6.4&new-version=0.7.31)](https://docs.github.com/en/github/managing-security-vulnerabilities/about-dependabot-security-updates#about-compatibility-scores)

Dependabot will resolve any conflicts with this PR as long as you don't
alter it yourself. You can also trigger a rebase manually by commenting
`@dependabot rebase`.

[//]: # (dependabot-automerge-start)
[//]: # (dependabot-automerge-end)

---

<details>
<summary>Dependabot commands and options</summary>
<br />

You can trigger Dependabot actions by commenting on this PR:
- `@dependabot rebase` will rebase this PR
- `@dependabot recreate` will recreate this PR, overwriting any edits
that have been made to it
- `@dependabot show <dependency name> ignore conditions` will show all
of the ignore conditions of the specified dependency
- `@dependabot ignore this major version` will close this PR and stop
Dependabot creating any more for this major version (unless you reopen
the PR or upgrade to it yourself)
- `@dependabot ignore this minor version` will close this PR and stop
Dependabot creating any more for this minor version (unless you reopen
the PR or upgrade to it yourself)
- `@dependabot ignore this dependency` will close this PR and stop
Dependabot creating any more for this dependency (unless you reopen the
PR or upgrade to it yourself)
You can disable automated security fix PRs for this repo from the
[Security Alerts
page](https://github.com/langchain-ai/langgraph/network/alerts).

</details>

Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
2026-04-15 19:37:47 -07:00
dependabot[bot]GitHubdependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
6177c4311b chore(deps): bump langsmith from 0.6.4 to 0.7.31 in /libs/checkpoint (#7525)
Bumps [langsmith](https://github.com/langchain-ai/langsmith-sdk) from
0.6.4 to 0.7.31.
<details>
<summary>Release notes</summary>
<p><em>Sourced from <a
href="https://github.com/langchain-ai/langsmith-sdk/releases">langsmith's
releases</a>.</em></p>
<blockquote>
<h2>v0.7.31</h2>
<h2>What's Changed</h2>
<ul>
<li>chore(deps-dev): bump langchain-core from 1.2.23 to 1.2.28 in
/python by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2692">langchain-ai/langsmith-sdk#2692</a></li>
<li>chore(deps-dev): bump <code>@​anthropic-ai/sdk</code> from 0.82.0 to
0.84.0 in /js by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2684">langchain-ai/langsmith-sdk#2684</a></li>
<li>chore(deps): bump cryptography from 46.0.6 to 46.0.7 in /python by
<a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2693">langchain-ai/langsmith-sdk#2693</a></li>
<li>chore(deps-dev): bump <code>@​anthropic-ai/sdk</code> from 0.84.0 to
0.85.0 in /js by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2700">langchain-ai/langsmith-sdk#2700</a></li>
<li>feat(py): Tag OpenAI Agent Python SDK runs with ls_agent_type by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2699">langchain-ai/langsmith-sdk#2699</a></li>
<li>feat(js): Adds ls_agent_type metadata to AI SDK runs by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2701">langchain-ai/langsmith-sdk#2701</a></li>
<li>chore(deps-dev): bump types-tqdm from 4.67.3.20260303 to
4.67.3.20260408 in /python by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2710">langchain-ai/langsmith-sdk#2710</a></li>
<li>chore(deps): bump pnpm/action-setup from 5 to 6 by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2705">langchain-ai/langsmith-sdk#2705</a></li>
<li>chore(deps): bump the py-minor-and-patch group across 1 directory
with 10 updates by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2711">langchain-ai/langsmith-sdk#2711</a></li>
<li>chore(deps-dev): bump <code>@​anthropic-ai/sdk</code> from 0.85.0 to
0.86.0 in /js by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2702">langchain-ai/langsmith-sdk#2702</a></li>
<li>chore(deps): bump actions/github-script from 8 to 9 by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2706">langchain-ai/langsmith-sdk#2706</a></li>
<li>chore(deps-dev): bump the js-minor-and-patch group across 1
directory with 7 updates by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2712">langchain-ai/langsmith-sdk#2712</a></li>
<li>chore(deps-dev): bump types-psutil from 7.2.2.20260130 to
7.2.2.20260408 in /python by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2709">langchain-ai/langsmith-sdk#2709</a></li>
<li>chore(deps-dev): bump rich from 14.3.3 to 15.0.0 in /python by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2708">langchain-ai/langsmith-sdk#2708</a></li>
<li>feat: Filter kwargs from new token events by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2714">langchain-ai/langsmith-sdk#2714</a></li>
<li>release(py): 0.7.31 by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2716">langchain-ai/langsmith-sdk#2716</a></li>
</ul>
<p><strong>Full Changelog</strong>: <a
href="https://github.com/langchain-ai/langsmith-sdk/compare/v0.7.30...v0.7.31">https://github.com/langchain-ai/langsmith-sdk/compare/v0.7.30...v0.7.31</a></p>
<h2>v0.7.30</h2>
<h2>What's Changed</h2>
<ul>
<li>feat(python): add service feature to sandbox by <a
href="https://github.com/DanielKneipp"><code>@​DanielKneipp</code></a>
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2665">langchain-ai/langsmith-sdk#2665</a></li>
<li>fix(js): Fix prototype pollution bug in anonymizers by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2690">langchain-ai/langsmith-sdk#2690</a></li>
<li>release(js): 0.5.18 by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2691">langchain-ai/langsmith-sdk#2691</a></li>
<li>chore(js/sandbox): suppress warning log by <a
href="https://github.com/hntrl"><code>@​hntrl</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2694">langchain-ai/langsmith-sdk#2694</a></li>
<li>feat(js): Add metadata to Claude Agent SDK JS tracing by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2695">langchain-ai/langsmith-sdk#2695</a></li>
<li>fix(py): Fix run tree memory leak by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2696">langchain-ai/langsmith-sdk#2696</a></li>
<li>release(py): 0.7.30 by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2698">langchain-ai/langsmith-sdk#2698</a></li>
</ul>
<p><strong>Full Changelog</strong>: <a
href="https://github.com/langchain-ai/langsmith-sdk/compare/v0.7.29...v0.7.30">https://github.com/langchain-ai/langsmith-sdk/compare/v0.7.29...v0.7.30</a></p>
<h2>v0.7.29</h2>
<h2>What's Changed</h2>
<ul>
<li>release(js): 0.5.17 by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2681">langchain-ai/langsmith-sdk#2681</a></li>
<li>feat(py): Fix race condition around Claude Agent SDK instrumentation
by <a href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a>
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2685">langchain-ai/langsmith-sdk#2685</a></li>
<li>release(py): 0.7.29 by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2686">langchain-ai/langsmith-sdk#2686</a></li>
</ul>
<p><strong>Full Changelog</strong>: <a
href="https://github.com/langchain-ai/langsmith-sdk/compare/v0.7.28...v0.7.29">https://github.com/langchain-ai/langsmith-sdk/compare/v0.7.28...v0.7.29</a></p>
<h2>v0.7.28</h2>
<h2>What's Changed</h2>
<ul>
<li>feat(py): Support subagent tracing in Claude Agents SDK, fix usage
and duplicate messages by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2670">langchain-ai/langsmith-sdk#2670</a></li>
<li>chore(deps-dev): bump the py-minor-and-patch group across 1
directory with 11 updates by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2677">langchain-ai/langsmith-sdk#2677</a></li>
<li>chore(deps-dev): bump the js-minor-and-patch group across 1
directory with 8 updates by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2667">langchain-ai/langsmith-sdk#2667</a></li>
<li>chore(deps): bump pnpm/action-setup from 4 to 5 by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2658">langchain-ai/langsmith-sdk#2658</a></li>
</ul>
<!-- raw HTML omitted -->
</blockquote>
<p>... (truncated)</p>
</details>
<details>
<summary>Commits</summary>
<ul>
<li><a
href="https://github.com/langchain-ai/langsmith-sdk/commit/c434999d05c00334efeba88b8bbd2de9f3afbef6"><code>c434999</code></a>
release(py): 0.7.31 (<a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/issues/2716">#2716</a>)</li>
<li><a
href="https://github.com/langchain-ai/langsmith-sdk/commit/47d7c4a783333e716395d802e7632f1f1b4744d3"><code>47d7c4a</code></a>
feat: Filter kwargs from new token events (<a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/issues/2714">#2714</a>)</li>
<li><a
href="https://github.com/langchain-ai/langsmith-sdk/commit/3c57445b543c9a2f86db52024ea2c998bfc2ffab"><code>3c57445</code></a>
chore(deps-dev): bump rich from 14.3.3 to 15.0.0 in /python (<a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/issues/2708">#2708</a>)</li>
<li><a
href="https://github.com/langchain-ai/langsmith-sdk/commit/2be6cd01a2b6e35e811488d3561e7b0b57b06f63"><code>2be6cd0</code></a>
chore(deps-dev): bump types-psutil from 7.2.2.20260130 to 7.2.2.20260408
in /...</li>
<li><a
href="https://github.com/langchain-ai/langsmith-sdk/commit/b8b6ca32d43c919c07a4e13c99a83bcaab8accb0"><code>b8b6ca3</code></a>
chore(deps-dev): bump the js-minor-and-patch group across 1 directory
with 7 ...</li>
<li><a
href="https://github.com/langchain-ai/langsmith-sdk/commit/9897cb33da7698291637f268edd833ca3e1adde6"><code>9897cb3</code></a>
chore(deps): bump actions/github-script from 8 to 9 (<a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/issues/2706">#2706</a>)</li>
<li><a
href="https://github.com/langchain-ai/langsmith-sdk/commit/572c0184285747e027a796e03ea6c9ba171e09a6"><code>572c018</code></a>
chore(deps-dev): bump <code>@​anthropic-ai/sdk</code> from 0.85.0 to
0.86.0 in /js (<a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/issues/2702">#2702</a>)</li>
<li><a
href="https://github.com/langchain-ai/langsmith-sdk/commit/57447524c88b6bba2775161aa449da32fb8e5c42"><code>5744752</code></a>
chore(deps): bump the py-minor-and-patch group across 1 directory with
10 upd...</li>
<li><a
href="https://github.com/langchain-ai/langsmith-sdk/commit/960cae7f490e9ccbe428e6b56c8047bdb7b942a5"><code>960cae7</code></a>
chore(deps): bump pnpm/action-setup from 5 to 6 (<a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/issues/2705">#2705</a>)</li>
<li><a
href="https://github.com/langchain-ai/langsmith-sdk/commit/9370e7670abf7f8f9a36fbb72250bcfd2f91e7c6"><code>9370e76</code></a>
chore(deps-dev): bump types-tqdm from 4.67.3.20260303 to 4.67.3.20260408
in /...</li>
<li>Additional commits viewable in <a
href="https://github.com/langchain-ai/langsmith-sdk/compare/v0.6.4...v0.7.31">compare
view</a></li>
</ul>
</details>
<br />


[![Dependabot compatibility
score](https://dependabot-badges.githubapp.com/badges/compatibility_score?dependency-name=langsmith&package-manager=uv&previous-version=0.6.4&new-version=0.7.31)](https://docs.github.com/en/github/managing-security-vulnerabilities/about-dependabot-security-updates#about-compatibility-scores)

Dependabot will resolve any conflicts with this PR as long as you don't
alter it yourself. You can also trigger a rebase manually by commenting
`@dependabot rebase`.

[//]: # (dependabot-automerge-start)
[//]: # (dependabot-automerge-end)

---

<details>
<summary>Dependabot commands and options</summary>
<br />

You can trigger Dependabot actions by commenting on this PR:
- `@dependabot rebase` will rebase this PR
- `@dependabot recreate` will recreate this PR, overwriting any edits
that have been made to it
- `@dependabot show <dependency name> ignore conditions` will show all
of the ignore conditions of the specified dependency
- `@dependabot ignore this major version` will close this PR and stop
Dependabot creating any more for this major version (unless you reopen
the PR or upgrade to it yourself)
- `@dependabot ignore this minor version` will close this PR and stop
Dependabot creating any more for this minor version (unless you reopen
the PR or upgrade to it yourself)
- `@dependabot ignore this dependency` will close this PR and stop
Dependabot creating any more for this dependency (unless you reopen the
PR or upgrade to it yourself)
You can disable automated security fix PRs for this repo from the
[Security Alerts
page](https://github.com/langchain-ai/langgraph/network/alerts).

</details>

Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
2026-04-15 19:37:35 -07:00
dependabot[bot]GitHubdependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
303769904b chore(deps): bump langsmith from 0.7.20 to 0.7.31 in /libs/sdk-py (#7528)
Bumps [langsmith](https://github.com/langchain-ai/langsmith-sdk) from
0.7.20 to 0.7.31.
<details>
<summary>Release notes</summary>
<p><em>Sourced from <a
href="https://github.com/langchain-ai/langsmith-sdk/releases">langsmith's
releases</a>.</em></p>
<blockquote>
<h2>v0.7.31</h2>
<h2>What's Changed</h2>
<ul>
<li>chore(deps-dev): bump langchain-core from 1.2.23 to 1.2.28 in
/python by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2692">langchain-ai/langsmith-sdk#2692</a></li>
<li>chore(deps-dev): bump <code>@​anthropic-ai/sdk</code> from 0.82.0 to
0.84.0 in /js by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2684">langchain-ai/langsmith-sdk#2684</a></li>
<li>chore(deps): bump cryptography from 46.0.6 to 46.0.7 in /python by
<a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2693">langchain-ai/langsmith-sdk#2693</a></li>
<li>chore(deps-dev): bump <code>@​anthropic-ai/sdk</code> from 0.84.0 to
0.85.0 in /js by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2700">langchain-ai/langsmith-sdk#2700</a></li>
<li>feat(py): Tag OpenAI Agent Python SDK runs with ls_agent_type by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2699">langchain-ai/langsmith-sdk#2699</a></li>
<li>feat(js): Adds ls_agent_type metadata to AI SDK runs by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2701">langchain-ai/langsmith-sdk#2701</a></li>
<li>chore(deps-dev): bump types-tqdm from 4.67.3.20260303 to
4.67.3.20260408 in /python by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2710">langchain-ai/langsmith-sdk#2710</a></li>
<li>chore(deps): bump pnpm/action-setup from 5 to 6 by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2705">langchain-ai/langsmith-sdk#2705</a></li>
<li>chore(deps): bump the py-minor-and-patch group across 1 directory
with 10 updates by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2711">langchain-ai/langsmith-sdk#2711</a></li>
<li>chore(deps-dev): bump <code>@​anthropic-ai/sdk</code> from 0.85.0 to
0.86.0 in /js by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2702">langchain-ai/langsmith-sdk#2702</a></li>
<li>chore(deps): bump actions/github-script from 8 to 9 by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2706">langchain-ai/langsmith-sdk#2706</a></li>
<li>chore(deps-dev): bump the js-minor-and-patch group across 1
directory with 7 updates by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2712">langchain-ai/langsmith-sdk#2712</a></li>
<li>chore(deps-dev): bump types-psutil from 7.2.2.20260130 to
7.2.2.20260408 in /python by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2709">langchain-ai/langsmith-sdk#2709</a></li>
<li>chore(deps-dev): bump rich from 14.3.3 to 15.0.0 in /python by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2708">langchain-ai/langsmith-sdk#2708</a></li>
<li>feat: Filter kwargs from new token events by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2714">langchain-ai/langsmith-sdk#2714</a></li>
<li>release(py): 0.7.31 by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2716">langchain-ai/langsmith-sdk#2716</a></li>
</ul>
<p><strong>Full Changelog</strong>: <a
href="https://github.com/langchain-ai/langsmith-sdk/compare/v0.7.30...v0.7.31">https://github.com/langchain-ai/langsmith-sdk/compare/v0.7.30...v0.7.31</a></p>
<h2>v0.7.30</h2>
<h2>What's Changed</h2>
<ul>
<li>feat(python): add service feature to sandbox by <a
href="https://github.com/DanielKneipp"><code>@​DanielKneipp</code></a>
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2665">langchain-ai/langsmith-sdk#2665</a></li>
<li>fix(js): Fix prototype pollution bug in anonymizers by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2690">langchain-ai/langsmith-sdk#2690</a></li>
<li>release(js): 0.5.18 by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2691">langchain-ai/langsmith-sdk#2691</a></li>
<li>chore(js/sandbox): suppress warning log by <a
href="https://github.com/hntrl"><code>@​hntrl</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2694">langchain-ai/langsmith-sdk#2694</a></li>
<li>feat(js): Add metadata to Claude Agent SDK JS tracing by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2695">langchain-ai/langsmith-sdk#2695</a></li>
<li>fix(py): Fix run tree memory leak by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2696">langchain-ai/langsmith-sdk#2696</a></li>
<li>release(py): 0.7.30 by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2698">langchain-ai/langsmith-sdk#2698</a></li>
</ul>
<p><strong>Full Changelog</strong>: <a
href="https://github.com/langchain-ai/langsmith-sdk/compare/v0.7.29...v0.7.30">https://github.com/langchain-ai/langsmith-sdk/compare/v0.7.29...v0.7.30</a></p>
<h2>v0.7.29</h2>
<h2>What's Changed</h2>
<ul>
<li>release(js): 0.5.17 by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2681">langchain-ai/langsmith-sdk#2681</a></li>
<li>feat(py): Fix race condition around Claude Agent SDK instrumentation
by <a href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a>
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2685">langchain-ai/langsmith-sdk#2685</a></li>
<li>release(py): 0.7.29 by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2686">langchain-ai/langsmith-sdk#2686</a></li>
</ul>
<p><strong>Full Changelog</strong>: <a
href="https://github.com/langchain-ai/langsmith-sdk/compare/v0.7.28...v0.7.29">https://github.com/langchain-ai/langsmith-sdk/compare/v0.7.28...v0.7.29</a></p>
<h2>v0.7.28</h2>
<h2>What's Changed</h2>
<ul>
<li>feat(py): Support subagent tracing in Claude Agents SDK, fix usage
and duplicate messages by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2670">langchain-ai/langsmith-sdk#2670</a></li>
<li>chore(deps-dev): bump the py-minor-and-patch group across 1
directory with 11 updates by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2677">langchain-ai/langsmith-sdk#2677</a></li>
<li>chore(deps-dev): bump the js-minor-and-patch group across 1
directory with 8 updates by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2667">langchain-ai/langsmith-sdk#2667</a></li>
<li>chore(deps): bump pnpm/action-setup from 4 to 5 by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2658">langchain-ai/langsmith-sdk#2658</a></li>
</ul>
<!-- raw HTML omitted -->
</blockquote>
<p>... (truncated)</p>
</details>
<details>
<summary>Commits</summary>
<ul>
<li><a
href="https://github.com/langchain-ai/langsmith-sdk/commit/c434999d05c00334efeba88b8bbd2de9f3afbef6"><code>c434999</code></a>
release(py): 0.7.31 (<a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/issues/2716">#2716</a>)</li>
<li><a
href="https://github.com/langchain-ai/langsmith-sdk/commit/47d7c4a783333e716395d802e7632f1f1b4744d3"><code>47d7c4a</code></a>
feat: Filter kwargs from new token events (<a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/issues/2714">#2714</a>)</li>
<li><a
href="https://github.com/langchain-ai/langsmith-sdk/commit/3c57445b543c9a2f86db52024ea2c998bfc2ffab"><code>3c57445</code></a>
chore(deps-dev): bump rich from 14.3.3 to 15.0.0 in /python (<a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/issues/2708">#2708</a>)</li>
<li><a
href="https://github.com/langchain-ai/langsmith-sdk/commit/2be6cd01a2b6e35e811488d3561e7b0b57b06f63"><code>2be6cd0</code></a>
chore(deps-dev): bump types-psutil from 7.2.2.20260130 to 7.2.2.20260408
in /...</li>
<li><a
href="https://github.com/langchain-ai/langsmith-sdk/commit/b8b6ca32d43c919c07a4e13c99a83bcaab8accb0"><code>b8b6ca3</code></a>
chore(deps-dev): bump the js-minor-and-patch group across 1 directory
with 7 ...</li>
<li><a
href="https://github.com/langchain-ai/langsmith-sdk/commit/9897cb33da7698291637f268edd833ca3e1adde6"><code>9897cb3</code></a>
chore(deps): bump actions/github-script from 8 to 9 (<a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/issues/2706">#2706</a>)</li>
<li><a
href="https://github.com/langchain-ai/langsmith-sdk/commit/572c0184285747e027a796e03ea6c9ba171e09a6"><code>572c018</code></a>
chore(deps-dev): bump <code>@​anthropic-ai/sdk</code> from 0.85.0 to
0.86.0 in /js (<a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/issues/2702">#2702</a>)</li>
<li><a
href="https://github.com/langchain-ai/langsmith-sdk/commit/57447524c88b6bba2775161aa449da32fb8e5c42"><code>5744752</code></a>
chore(deps): bump the py-minor-and-patch group across 1 directory with
10 upd...</li>
<li><a
href="https://github.com/langchain-ai/langsmith-sdk/commit/960cae7f490e9ccbe428e6b56c8047bdb7b942a5"><code>960cae7</code></a>
chore(deps): bump pnpm/action-setup from 5 to 6 (<a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/issues/2705">#2705</a>)</li>
<li><a
href="https://github.com/langchain-ai/langsmith-sdk/commit/9370e7670abf7f8f9a36fbb72250bcfd2f91e7c6"><code>9370e76</code></a>
chore(deps-dev): bump types-tqdm from 4.67.3.20260303 to 4.67.3.20260408
in /...</li>
<li>Additional commits viewable in <a
href="https://github.com/langchain-ai/langsmith-sdk/compare/v0.7.20...v0.7.31">compare
view</a></li>
</ul>
</details>
<br />


[![Dependabot compatibility
score](https://dependabot-badges.githubapp.com/badges/compatibility_score?dependency-name=langsmith&package-manager=uv&previous-version=0.7.20&new-version=0.7.31)](https://docs.github.com/en/github/managing-security-vulnerabilities/about-dependabot-security-updates#about-compatibility-scores)

Dependabot will resolve any conflicts with this PR as long as you don't
alter it yourself. You can also trigger a rebase manually by commenting
`@dependabot rebase`.

[//]: # (dependabot-automerge-start)
[//]: # (dependabot-automerge-end)

---

<details>
<summary>Dependabot commands and options</summary>
<br />

You can trigger Dependabot actions by commenting on this PR:
- `@dependabot rebase` will rebase this PR
- `@dependabot recreate` will recreate this PR, overwriting any edits
that have been made to it
- `@dependabot show <dependency name> ignore conditions` will show all
of the ignore conditions of the specified dependency
- `@dependabot ignore this major version` will close this PR and stop
Dependabot creating any more for this major version (unless you reopen
the PR or upgrade to it yourself)
- `@dependabot ignore this minor version` will close this PR and stop
Dependabot creating any more for this minor version (unless you reopen
the PR or upgrade to it yourself)
- `@dependabot ignore this dependency` will close this PR and stop
Dependabot creating any more for this dependency (unless you reopen the
PR or upgrade to it yourself)
You can disable automated security fix PRs for this repo from the
[Security Alerts
page](https://github.com/langchain-ai/langgraph/network/alerts).

</details>

Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
2026-04-15 19:37:27 -07:00
dependabot[bot]GitHubdependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
cee7dcd523 chore(deps): bump langsmith from 0.7.26 to 0.7.31 in /libs/cli (#7529)
Bumps [langsmith](https://github.com/langchain-ai/langsmith-sdk) from
0.7.26 to 0.7.31.
<details>
<summary>Release notes</summary>
<p><em>Sourced from <a
href="https://github.com/langchain-ai/langsmith-sdk/releases">langsmith's
releases</a>.</em></p>
<blockquote>
<h2>v0.7.31</h2>
<h2>What's Changed</h2>
<ul>
<li>chore(deps-dev): bump langchain-core from 1.2.23 to 1.2.28 in
/python by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2692">langchain-ai/langsmith-sdk#2692</a></li>
<li>chore(deps-dev): bump <code>@​anthropic-ai/sdk</code> from 0.82.0 to
0.84.0 in /js by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2684">langchain-ai/langsmith-sdk#2684</a></li>
<li>chore(deps): bump cryptography from 46.0.6 to 46.0.7 in /python by
<a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2693">langchain-ai/langsmith-sdk#2693</a></li>
<li>chore(deps-dev): bump <code>@​anthropic-ai/sdk</code> from 0.84.0 to
0.85.0 in /js by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2700">langchain-ai/langsmith-sdk#2700</a></li>
<li>feat(py): Tag OpenAI Agent Python SDK runs with ls_agent_type by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2699">langchain-ai/langsmith-sdk#2699</a></li>
<li>feat(js): Adds ls_agent_type metadata to AI SDK runs by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2701">langchain-ai/langsmith-sdk#2701</a></li>
<li>chore(deps-dev): bump types-tqdm from 4.67.3.20260303 to
4.67.3.20260408 in /python by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2710">langchain-ai/langsmith-sdk#2710</a></li>
<li>chore(deps): bump pnpm/action-setup from 5 to 6 by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2705">langchain-ai/langsmith-sdk#2705</a></li>
<li>chore(deps): bump the py-minor-and-patch group across 1 directory
with 10 updates by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2711">langchain-ai/langsmith-sdk#2711</a></li>
<li>chore(deps-dev): bump <code>@​anthropic-ai/sdk</code> from 0.85.0 to
0.86.0 in /js by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2702">langchain-ai/langsmith-sdk#2702</a></li>
<li>chore(deps): bump actions/github-script from 8 to 9 by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2706">langchain-ai/langsmith-sdk#2706</a></li>
<li>chore(deps-dev): bump the js-minor-and-patch group across 1
directory with 7 updates by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2712">langchain-ai/langsmith-sdk#2712</a></li>
<li>chore(deps-dev): bump types-psutil from 7.2.2.20260130 to
7.2.2.20260408 in /python by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2709">langchain-ai/langsmith-sdk#2709</a></li>
<li>chore(deps-dev): bump rich from 14.3.3 to 15.0.0 in /python by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2708">langchain-ai/langsmith-sdk#2708</a></li>
<li>feat: Filter kwargs from new token events by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2714">langchain-ai/langsmith-sdk#2714</a></li>
<li>release(py): 0.7.31 by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2716">langchain-ai/langsmith-sdk#2716</a></li>
</ul>
<p><strong>Full Changelog</strong>: <a
href="https://github.com/langchain-ai/langsmith-sdk/compare/v0.7.30...v0.7.31">https://github.com/langchain-ai/langsmith-sdk/compare/v0.7.30...v0.7.31</a></p>
<h2>v0.7.30</h2>
<h2>What's Changed</h2>
<ul>
<li>feat(python): add service feature to sandbox by <a
href="https://github.com/DanielKneipp"><code>@​DanielKneipp</code></a>
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2665">langchain-ai/langsmith-sdk#2665</a></li>
<li>fix(js): Fix prototype pollution bug in anonymizers by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2690">langchain-ai/langsmith-sdk#2690</a></li>
<li>release(js): 0.5.18 by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2691">langchain-ai/langsmith-sdk#2691</a></li>
<li>chore(js/sandbox): suppress warning log by <a
href="https://github.com/hntrl"><code>@​hntrl</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2694">langchain-ai/langsmith-sdk#2694</a></li>
<li>feat(js): Add metadata to Claude Agent SDK JS tracing by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2695">langchain-ai/langsmith-sdk#2695</a></li>
<li>fix(py): Fix run tree memory leak by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2696">langchain-ai/langsmith-sdk#2696</a></li>
<li>release(py): 0.7.30 by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2698">langchain-ai/langsmith-sdk#2698</a></li>
</ul>
<p><strong>Full Changelog</strong>: <a
href="https://github.com/langchain-ai/langsmith-sdk/compare/v0.7.29...v0.7.30">https://github.com/langchain-ai/langsmith-sdk/compare/v0.7.29...v0.7.30</a></p>
<h2>v0.7.29</h2>
<h2>What's Changed</h2>
<ul>
<li>release(js): 0.5.17 by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2681">langchain-ai/langsmith-sdk#2681</a></li>
<li>feat(py): Fix race condition around Claude Agent SDK instrumentation
by <a href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a>
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2685">langchain-ai/langsmith-sdk#2685</a></li>
<li>release(py): 0.7.29 by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2686">langchain-ai/langsmith-sdk#2686</a></li>
</ul>
<p><strong>Full Changelog</strong>: <a
href="https://github.com/langchain-ai/langsmith-sdk/compare/v0.7.28...v0.7.29">https://github.com/langchain-ai/langsmith-sdk/compare/v0.7.28...v0.7.29</a></p>
<h2>v0.7.28</h2>
<h2>What's Changed</h2>
<ul>
<li>feat(py): Support subagent tracing in Claude Agents SDK, fix usage
and duplicate messages by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2670">langchain-ai/langsmith-sdk#2670</a></li>
<li>chore(deps-dev): bump the py-minor-and-patch group across 1
directory with 11 updates by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2677">langchain-ai/langsmith-sdk#2677</a></li>
<li>chore(deps-dev): bump the js-minor-and-patch group across 1
directory with 8 updates by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2667">langchain-ai/langsmith-sdk#2667</a></li>
<li>chore(deps): bump pnpm/action-setup from 4 to 5 by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2658">langchain-ai/langsmith-sdk#2658</a></li>
</ul>
<!-- raw HTML omitted -->
</blockquote>
<p>... (truncated)</p>
</details>
<details>
<summary>Commits</summary>
<ul>
<li><a
href="https://github.com/langchain-ai/langsmith-sdk/commit/c434999d05c00334efeba88b8bbd2de9f3afbef6"><code>c434999</code></a>
release(py): 0.7.31 (<a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/issues/2716">#2716</a>)</li>
<li><a
href="https://github.com/langchain-ai/langsmith-sdk/commit/47d7c4a783333e716395d802e7632f1f1b4744d3"><code>47d7c4a</code></a>
feat: Filter kwargs from new token events (<a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/issues/2714">#2714</a>)</li>
<li><a
href="https://github.com/langchain-ai/langsmith-sdk/commit/3c57445b543c9a2f86db52024ea2c998bfc2ffab"><code>3c57445</code></a>
chore(deps-dev): bump rich from 14.3.3 to 15.0.0 in /python (<a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/issues/2708">#2708</a>)</li>
<li><a
href="https://github.com/langchain-ai/langsmith-sdk/commit/2be6cd01a2b6e35e811488d3561e7b0b57b06f63"><code>2be6cd0</code></a>
chore(deps-dev): bump types-psutil from 7.2.2.20260130 to 7.2.2.20260408
in /...</li>
<li><a
href="https://github.com/langchain-ai/langsmith-sdk/commit/b8b6ca32d43c919c07a4e13c99a83bcaab8accb0"><code>b8b6ca3</code></a>
chore(deps-dev): bump the js-minor-and-patch group across 1 directory
with 7 ...</li>
<li><a
href="https://github.com/langchain-ai/langsmith-sdk/commit/9897cb33da7698291637f268edd833ca3e1adde6"><code>9897cb3</code></a>
chore(deps): bump actions/github-script from 8 to 9 (<a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/issues/2706">#2706</a>)</li>
<li><a
href="https://github.com/langchain-ai/langsmith-sdk/commit/572c0184285747e027a796e03ea6c9ba171e09a6"><code>572c018</code></a>
chore(deps-dev): bump <code>@​anthropic-ai/sdk</code> from 0.85.0 to
0.86.0 in /js (<a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/issues/2702">#2702</a>)</li>
<li><a
href="https://github.com/langchain-ai/langsmith-sdk/commit/57447524c88b6bba2775161aa449da32fb8e5c42"><code>5744752</code></a>
chore(deps): bump the py-minor-and-patch group across 1 directory with
10 upd...</li>
<li><a
href="https://github.com/langchain-ai/langsmith-sdk/commit/960cae7f490e9ccbe428e6b56c8047bdb7b942a5"><code>960cae7</code></a>
chore(deps): bump pnpm/action-setup from 5 to 6 (<a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/issues/2705">#2705</a>)</li>
<li><a
href="https://github.com/langchain-ai/langsmith-sdk/commit/9370e7670abf7f8f9a36fbb72250bcfd2f91e7c6"><code>9370e76</code></a>
chore(deps-dev): bump types-tqdm from 4.67.3.20260303 to 4.67.3.20260408
in /...</li>
<li>Additional commits viewable in <a
href="https://github.com/langchain-ai/langsmith-sdk/compare/v0.7.26...v0.7.31">compare
view</a></li>
</ul>
</details>
<br />


[![Dependabot compatibility
score](https://dependabot-badges.githubapp.com/badges/compatibility_score?dependency-name=langsmith&package-manager=uv&previous-version=0.7.26&new-version=0.7.31)](https://docs.github.com/en/github/managing-security-vulnerabilities/about-dependabot-security-updates#about-compatibility-scores)

Dependabot will resolve any conflicts with this PR as long as you don't
alter it yourself. You can also trigger a rebase manually by commenting
`@dependabot rebase`.

[//]: # (dependabot-automerge-start)
[//]: # (dependabot-automerge-end)

---

<details>
<summary>Dependabot commands and options</summary>
<br />

You can trigger Dependabot actions by commenting on this PR:
- `@dependabot rebase` will rebase this PR
- `@dependabot recreate` will recreate this PR, overwriting any edits
that have been made to it
- `@dependabot show <dependency name> ignore conditions` will show all
of the ignore conditions of the specified dependency
- `@dependabot ignore this major version` will close this PR and stop
Dependabot creating any more for this major version (unless you reopen
the PR or upgrade to it yourself)
- `@dependabot ignore this minor version` will close this PR and stop
Dependabot creating any more for this minor version (unless you reopen
the PR or upgrade to it yourself)
- `@dependabot ignore this dependency` will close this PR and stop
Dependabot creating any more for this dependency (unless you reopen the
PR or upgrade to it yourself)
You can disable automated security fix PRs for this repo from the
[Security Alerts
page](https://github.com/langchain-ai/langgraph/network/alerts).

</details>

Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
2026-04-15 19:37:13 -07:00
dependabot[bot]GitHubdependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
3413723e5a chore(deps): bump langsmith from 0.6.4 to 0.7.31 in /libs/checkpoint-postgres (#7527)
Bumps [langsmith](https://github.com/langchain-ai/langsmith-sdk) from
0.6.4 to 0.7.31.
<details>
<summary>Release notes</summary>
<p><em>Sourced from <a
href="https://github.com/langchain-ai/langsmith-sdk/releases">langsmith's
releases</a>.</em></p>
<blockquote>
<h2>v0.7.31</h2>
<h2>What's Changed</h2>
<ul>
<li>chore(deps-dev): bump langchain-core from 1.2.23 to 1.2.28 in
/python by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2692">langchain-ai/langsmith-sdk#2692</a></li>
<li>chore(deps-dev): bump <code>@​anthropic-ai/sdk</code> from 0.82.0 to
0.84.0 in /js by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2684">langchain-ai/langsmith-sdk#2684</a></li>
<li>chore(deps): bump cryptography from 46.0.6 to 46.0.7 in /python by
<a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2693">langchain-ai/langsmith-sdk#2693</a></li>
<li>chore(deps-dev): bump <code>@​anthropic-ai/sdk</code> from 0.84.0 to
0.85.0 in /js by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2700">langchain-ai/langsmith-sdk#2700</a></li>
<li>feat(py): Tag OpenAI Agent Python SDK runs with ls_agent_type by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2699">langchain-ai/langsmith-sdk#2699</a></li>
<li>feat(js): Adds ls_agent_type metadata to AI SDK runs by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2701">langchain-ai/langsmith-sdk#2701</a></li>
<li>chore(deps-dev): bump types-tqdm from 4.67.3.20260303 to
4.67.3.20260408 in /python by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2710">langchain-ai/langsmith-sdk#2710</a></li>
<li>chore(deps): bump pnpm/action-setup from 5 to 6 by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2705">langchain-ai/langsmith-sdk#2705</a></li>
<li>chore(deps): bump the py-minor-and-patch group across 1 directory
with 10 updates by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2711">langchain-ai/langsmith-sdk#2711</a></li>
<li>chore(deps-dev): bump <code>@​anthropic-ai/sdk</code> from 0.85.0 to
0.86.0 in /js by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2702">langchain-ai/langsmith-sdk#2702</a></li>
<li>chore(deps): bump actions/github-script from 8 to 9 by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2706">langchain-ai/langsmith-sdk#2706</a></li>
<li>chore(deps-dev): bump the js-minor-and-patch group across 1
directory with 7 updates by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2712">langchain-ai/langsmith-sdk#2712</a></li>
<li>chore(deps-dev): bump types-psutil from 7.2.2.20260130 to
7.2.2.20260408 in /python by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2709">langchain-ai/langsmith-sdk#2709</a></li>
<li>chore(deps-dev): bump rich from 14.3.3 to 15.0.0 in /python by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2708">langchain-ai/langsmith-sdk#2708</a></li>
<li>feat: Filter kwargs from new token events by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2714">langchain-ai/langsmith-sdk#2714</a></li>
<li>release(py): 0.7.31 by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2716">langchain-ai/langsmith-sdk#2716</a></li>
</ul>
<p><strong>Full Changelog</strong>: <a
href="https://github.com/langchain-ai/langsmith-sdk/compare/v0.7.30...v0.7.31">https://github.com/langchain-ai/langsmith-sdk/compare/v0.7.30...v0.7.31</a></p>
<h2>v0.7.30</h2>
<h2>What's Changed</h2>
<ul>
<li>feat(python): add service feature to sandbox by <a
href="https://github.com/DanielKneipp"><code>@​DanielKneipp</code></a>
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2665">langchain-ai/langsmith-sdk#2665</a></li>
<li>fix(js): Fix prototype pollution bug in anonymizers by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2690">langchain-ai/langsmith-sdk#2690</a></li>
<li>release(js): 0.5.18 by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2691">langchain-ai/langsmith-sdk#2691</a></li>
<li>chore(js/sandbox): suppress warning log by <a
href="https://github.com/hntrl"><code>@​hntrl</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2694">langchain-ai/langsmith-sdk#2694</a></li>
<li>feat(js): Add metadata to Claude Agent SDK JS tracing by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2695">langchain-ai/langsmith-sdk#2695</a></li>
<li>fix(py): Fix run tree memory leak by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2696">langchain-ai/langsmith-sdk#2696</a></li>
<li>release(py): 0.7.30 by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2698">langchain-ai/langsmith-sdk#2698</a></li>
</ul>
<p><strong>Full Changelog</strong>: <a
href="https://github.com/langchain-ai/langsmith-sdk/compare/v0.7.29...v0.7.30">https://github.com/langchain-ai/langsmith-sdk/compare/v0.7.29...v0.7.30</a></p>
<h2>v0.7.29</h2>
<h2>What's Changed</h2>
<ul>
<li>release(js): 0.5.17 by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2681">langchain-ai/langsmith-sdk#2681</a></li>
<li>feat(py): Fix race condition around Claude Agent SDK instrumentation
by <a href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a>
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2685">langchain-ai/langsmith-sdk#2685</a></li>
<li>release(py): 0.7.29 by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2686">langchain-ai/langsmith-sdk#2686</a></li>
</ul>
<p><strong>Full Changelog</strong>: <a
href="https://github.com/langchain-ai/langsmith-sdk/compare/v0.7.28...v0.7.29">https://github.com/langchain-ai/langsmith-sdk/compare/v0.7.28...v0.7.29</a></p>
<h2>v0.7.28</h2>
<h2>What's Changed</h2>
<ul>
<li>feat(py): Support subagent tracing in Claude Agents SDK, fix usage
and duplicate messages by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2670">langchain-ai/langsmith-sdk#2670</a></li>
<li>chore(deps-dev): bump the py-minor-and-patch group across 1
directory with 11 updates by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2677">langchain-ai/langsmith-sdk#2677</a></li>
<li>chore(deps-dev): bump the js-minor-and-patch group across 1
directory with 8 updates by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2667">langchain-ai/langsmith-sdk#2667</a></li>
<li>chore(deps): bump pnpm/action-setup from 4 to 5 by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2658">langchain-ai/langsmith-sdk#2658</a></li>
</ul>
<!-- raw HTML omitted -->
</blockquote>
<p>... (truncated)</p>
</details>
<details>
<summary>Commits</summary>
<ul>
<li><a
href="https://github.com/langchain-ai/langsmith-sdk/commit/c434999d05c00334efeba88b8bbd2de9f3afbef6"><code>c434999</code></a>
release(py): 0.7.31 (<a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/issues/2716">#2716</a>)</li>
<li><a
href="https://github.com/langchain-ai/langsmith-sdk/commit/47d7c4a783333e716395d802e7632f1f1b4744d3"><code>47d7c4a</code></a>
feat: Filter kwargs from new token events (<a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/issues/2714">#2714</a>)</li>
<li><a
href="https://github.com/langchain-ai/langsmith-sdk/commit/3c57445b543c9a2f86db52024ea2c998bfc2ffab"><code>3c57445</code></a>
chore(deps-dev): bump rich from 14.3.3 to 15.0.0 in /python (<a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/issues/2708">#2708</a>)</li>
<li><a
href="https://github.com/langchain-ai/langsmith-sdk/commit/2be6cd01a2b6e35e811488d3561e7b0b57b06f63"><code>2be6cd0</code></a>
chore(deps-dev): bump types-psutil from 7.2.2.20260130 to 7.2.2.20260408
in /...</li>
<li><a
href="https://github.com/langchain-ai/langsmith-sdk/commit/b8b6ca32d43c919c07a4e13c99a83bcaab8accb0"><code>b8b6ca3</code></a>
chore(deps-dev): bump the js-minor-and-patch group across 1 directory
with 7 ...</li>
<li><a
href="https://github.com/langchain-ai/langsmith-sdk/commit/9897cb33da7698291637f268edd833ca3e1adde6"><code>9897cb3</code></a>
chore(deps): bump actions/github-script from 8 to 9 (<a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/issues/2706">#2706</a>)</li>
<li><a
href="https://github.com/langchain-ai/langsmith-sdk/commit/572c0184285747e027a796e03ea6c9ba171e09a6"><code>572c018</code></a>
chore(deps-dev): bump <code>@​anthropic-ai/sdk</code> from 0.85.0 to
0.86.0 in /js (<a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/issues/2702">#2702</a>)</li>
<li><a
href="https://github.com/langchain-ai/langsmith-sdk/commit/57447524c88b6bba2775161aa449da32fb8e5c42"><code>5744752</code></a>
chore(deps): bump the py-minor-and-patch group across 1 directory with
10 upd...</li>
<li><a
href="https://github.com/langchain-ai/langsmith-sdk/commit/960cae7f490e9ccbe428e6b56c8047bdb7b942a5"><code>960cae7</code></a>
chore(deps): bump pnpm/action-setup from 5 to 6 (<a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/issues/2705">#2705</a>)</li>
<li><a
href="https://github.com/langchain-ai/langsmith-sdk/commit/9370e7670abf7f8f9a36fbb72250bcfd2f91e7c6"><code>9370e76</code></a>
chore(deps-dev): bump types-tqdm from 4.67.3.20260303 to 4.67.3.20260408
in /...</li>
<li>Additional commits viewable in <a
href="https://github.com/langchain-ai/langsmith-sdk/compare/v0.6.4...v0.7.31">compare
view</a></li>
</ul>
</details>
<br />


[![Dependabot compatibility
score](https://dependabot-badges.githubapp.com/badges/compatibility_score?dependency-name=langsmith&package-manager=uv&previous-version=0.6.4&new-version=0.7.31)](https://docs.github.com/en/github/managing-security-vulnerabilities/about-dependabot-security-updates#about-compatibility-scores)

Dependabot will resolve any conflicts with this PR as long as you don't
alter it yourself. You can also trigger a rebase manually by commenting
`@dependabot rebase`.

[//]: # (dependabot-automerge-start)
[//]: # (dependabot-automerge-end)

---

<details>
<summary>Dependabot commands and options</summary>
<br />

You can trigger Dependabot actions by commenting on this PR:
- `@dependabot rebase` will rebase this PR
- `@dependabot recreate` will recreate this PR, overwriting any edits
that have been made to it
- `@dependabot show <dependency name> ignore conditions` will show all
of the ignore conditions of the specified dependency
- `@dependabot ignore this major version` will close this PR and stop
Dependabot creating any more for this major version (unless you reopen
the PR or upgrade to it yourself)
- `@dependabot ignore this minor version` will close this PR and stop
Dependabot creating any more for this minor version (unless you reopen
the PR or upgrade to it yourself)
- `@dependabot ignore this dependency` will close this PR and stop
Dependabot creating any more for this dependency (unless you reopen the
PR or upgrade to it yourself)
You can disable automated security fix PRs for this repo from the
[Security Alerts
page](https://github.com/langchain-ai/langgraph/network/alerts).

</details>

Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
2026-04-15 19:37:03 -07:00
dependabot[bot]GitHubdependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
07252d2cda chore(deps): bump langsmith from 0.6.4 to 0.7.31 in /libs/langgraph (#7526)
Bumps [langsmith](https://github.com/langchain-ai/langsmith-sdk) from
0.6.4 to 0.7.31.
<details>
<summary>Release notes</summary>
<p><em>Sourced from <a
href="https://github.com/langchain-ai/langsmith-sdk/releases">langsmith's
releases</a>.</em></p>
<blockquote>
<h2>v0.7.31</h2>
<h2>What's Changed</h2>
<ul>
<li>chore(deps-dev): bump langchain-core from 1.2.23 to 1.2.28 in
/python by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2692">langchain-ai/langsmith-sdk#2692</a></li>
<li>chore(deps-dev): bump <code>@​anthropic-ai/sdk</code> from 0.82.0 to
0.84.0 in /js by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2684">langchain-ai/langsmith-sdk#2684</a></li>
<li>chore(deps): bump cryptography from 46.0.6 to 46.0.7 in /python by
<a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2693">langchain-ai/langsmith-sdk#2693</a></li>
<li>chore(deps-dev): bump <code>@​anthropic-ai/sdk</code> from 0.84.0 to
0.85.0 in /js by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2700">langchain-ai/langsmith-sdk#2700</a></li>
<li>feat(py): Tag OpenAI Agent Python SDK runs with ls_agent_type by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2699">langchain-ai/langsmith-sdk#2699</a></li>
<li>feat(js): Adds ls_agent_type metadata to AI SDK runs by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2701">langchain-ai/langsmith-sdk#2701</a></li>
<li>chore(deps-dev): bump types-tqdm from 4.67.3.20260303 to
4.67.3.20260408 in /python by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2710">langchain-ai/langsmith-sdk#2710</a></li>
<li>chore(deps): bump pnpm/action-setup from 5 to 6 by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2705">langchain-ai/langsmith-sdk#2705</a></li>
<li>chore(deps): bump the py-minor-and-patch group across 1 directory
with 10 updates by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2711">langchain-ai/langsmith-sdk#2711</a></li>
<li>chore(deps-dev): bump <code>@​anthropic-ai/sdk</code> from 0.85.0 to
0.86.0 in /js by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2702">langchain-ai/langsmith-sdk#2702</a></li>
<li>chore(deps): bump actions/github-script from 8 to 9 by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2706">langchain-ai/langsmith-sdk#2706</a></li>
<li>chore(deps-dev): bump the js-minor-and-patch group across 1
directory with 7 updates by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2712">langchain-ai/langsmith-sdk#2712</a></li>
<li>chore(deps-dev): bump types-psutil from 7.2.2.20260130 to
7.2.2.20260408 in /python by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2709">langchain-ai/langsmith-sdk#2709</a></li>
<li>chore(deps-dev): bump rich from 14.3.3 to 15.0.0 in /python by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2708">langchain-ai/langsmith-sdk#2708</a></li>
<li>feat: Filter kwargs from new token events by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2714">langchain-ai/langsmith-sdk#2714</a></li>
<li>release(py): 0.7.31 by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2716">langchain-ai/langsmith-sdk#2716</a></li>
</ul>
<p><strong>Full Changelog</strong>: <a
href="https://github.com/langchain-ai/langsmith-sdk/compare/v0.7.30...v0.7.31">https://github.com/langchain-ai/langsmith-sdk/compare/v0.7.30...v0.7.31</a></p>
<h2>v0.7.30</h2>
<h2>What's Changed</h2>
<ul>
<li>feat(python): add service feature to sandbox by <a
href="https://github.com/DanielKneipp"><code>@​DanielKneipp</code></a>
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2665">langchain-ai/langsmith-sdk#2665</a></li>
<li>fix(js): Fix prototype pollution bug in anonymizers by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2690">langchain-ai/langsmith-sdk#2690</a></li>
<li>release(js): 0.5.18 by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2691">langchain-ai/langsmith-sdk#2691</a></li>
<li>chore(js/sandbox): suppress warning log by <a
href="https://github.com/hntrl"><code>@​hntrl</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2694">langchain-ai/langsmith-sdk#2694</a></li>
<li>feat(js): Add metadata to Claude Agent SDK JS tracing by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2695">langchain-ai/langsmith-sdk#2695</a></li>
<li>fix(py): Fix run tree memory leak by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2696">langchain-ai/langsmith-sdk#2696</a></li>
<li>release(py): 0.7.30 by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2698">langchain-ai/langsmith-sdk#2698</a></li>
</ul>
<p><strong>Full Changelog</strong>: <a
href="https://github.com/langchain-ai/langsmith-sdk/compare/v0.7.29...v0.7.30">https://github.com/langchain-ai/langsmith-sdk/compare/v0.7.29...v0.7.30</a></p>
<h2>v0.7.29</h2>
<h2>What's Changed</h2>
<ul>
<li>release(js): 0.5.17 by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2681">langchain-ai/langsmith-sdk#2681</a></li>
<li>feat(py): Fix race condition around Claude Agent SDK instrumentation
by <a href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a>
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2685">langchain-ai/langsmith-sdk#2685</a></li>
<li>release(py): 0.7.29 by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2686">langchain-ai/langsmith-sdk#2686</a></li>
</ul>
<p><strong>Full Changelog</strong>: <a
href="https://github.com/langchain-ai/langsmith-sdk/compare/v0.7.28...v0.7.29">https://github.com/langchain-ai/langsmith-sdk/compare/v0.7.28...v0.7.29</a></p>
<h2>v0.7.28</h2>
<h2>What's Changed</h2>
<ul>
<li>feat(py): Support subagent tracing in Claude Agents SDK, fix usage
and duplicate messages by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2670">langchain-ai/langsmith-sdk#2670</a></li>
<li>chore(deps-dev): bump the py-minor-and-patch group across 1
directory with 11 updates by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2677">langchain-ai/langsmith-sdk#2677</a></li>
<li>chore(deps-dev): bump the js-minor-and-patch group across 1
directory with 8 updates by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2667">langchain-ai/langsmith-sdk#2667</a></li>
<li>chore(deps): bump pnpm/action-setup from 4 to 5 by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2658">langchain-ai/langsmith-sdk#2658</a></li>
</ul>
<!-- raw HTML omitted -->
</blockquote>
<p>... (truncated)</p>
</details>
<details>
<summary>Commits</summary>
<ul>
<li><a
href="https://github.com/langchain-ai/langsmith-sdk/commit/c434999d05c00334efeba88b8bbd2de9f3afbef6"><code>c434999</code></a>
release(py): 0.7.31 (<a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/issues/2716">#2716</a>)</li>
<li><a
href="https://github.com/langchain-ai/langsmith-sdk/commit/47d7c4a783333e716395d802e7632f1f1b4744d3"><code>47d7c4a</code></a>
feat: Filter kwargs from new token events (<a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/issues/2714">#2714</a>)</li>
<li><a
href="https://github.com/langchain-ai/langsmith-sdk/commit/3c57445b543c9a2f86db52024ea2c998bfc2ffab"><code>3c57445</code></a>
chore(deps-dev): bump rich from 14.3.3 to 15.0.0 in /python (<a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/issues/2708">#2708</a>)</li>
<li><a
href="https://github.com/langchain-ai/langsmith-sdk/commit/2be6cd01a2b6e35e811488d3561e7b0b57b06f63"><code>2be6cd0</code></a>
chore(deps-dev): bump types-psutil from 7.2.2.20260130 to 7.2.2.20260408
in /...</li>
<li><a
href="https://github.com/langchain-ai/langsmith-sdk/commit/b8b6ca32d43c919c07a4e13c99a83bcaab8accb0"><code>b8b6ca3</code></a>
chore(deps-dev): bump the js-minor-and-patch group across 1 directory
with 7 ...</li>
<li><a
href="https://github.com/langchain-ai/langsmith-sdk/commit/9897cb33da7698291637f268edd833ca3e1adde6"><code>9897cb3</code></a>
chore(deps): bump actions/github-script from 8 to 9 (<a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/issues/2706">#2706</a>)</li>
<li><a
href="https://github.com/langchain-ai/langsmith-sdk/commit/572c0184285747e027a796e03ea6c9ba171e09a6"><code>572c018</code></a>
chore(deps-dev): bump <code>@​anthropic-ai/sdk</code> from 0.85.0 to
0.86.0 in /js (<a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/issues/2702">#2702</a>)</li>
<li><a
href="https://github.com/langchain-ai/langsmith-sdk/commit/57447524c88b6bba2775161aa449da32fb8e5c42"><code>5744752</code></a>
chore(deps): bump the py-minor-and-patch group across 1 directory with
10 upd...</li>
<li><a
href="https://github.com/langchain-ai/langsmith-sdk/commit/960cae7f490e9ccbe428e6b56c8047bdb7b942a5"><code>960cae7</code></a>
chore(deps): bump pnpm/action-setup from 5 to 6 (<a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/issues/2705">#2705</a>)</li>
<li><a
href="https://github.com/langchain-ai/langsmith-sdk/commit/9370e7670abf7f8f9a36fbb72250bcfd2f91e7c6"><code>9370e76</code></a>
chore(deps-dev): bump types-tqdm from 4.67.3.20260303 to 4.67.3.20260408
in /...</li>
<li>Additional commits viewable in <a
href="https://github.com/langchain-ai/langsmith-sdk/compare/v0.6.4...v0.7.31">compare
view</a></li>
</ul>
</details>
<br />


[![Dependabot compatibility
score](https://dependabot-badges.githubapp.com/badges/compatibility_score?dependency-name=langsmith&package-manager=uv&previous-version=0.6.4&new-version=0.7.31)](https://docs.github.com/en/github/managing-security-vulnerabilities/about-dependabot-security-updates#about-compatibility-scores)

Dependabot will resolve any conflicts with this PR as long as you don't
alter it yourself. You can also trigger a rebase manually by commenting
`@dependabot rebase`.

[//]: # (dependabot-automerge-start)
[//]: # (dependabot-automerge-end)

---

<details>
<summary>Dependabot commands and options</summary>
<br />

You can trigger Dependabot actions by commenting on this PR:
- `@dependabot rebase` will rebase this PR
- `@dependabot recreate` will recreate this PR, overwriting any edits
that have been made to it
- `@dependabot show <dependency name> ignore conditions` will show all
of the ignore conditions of the specified dependency
- `@dependabot ignore this major version` will close this PR and stop
Dependabot creating any more for this major version (unless you reopen
the PR or upgrade to it yourself)
- `@dependabot ignore this minor version` will close this PR and stop
Dependabot creating any more for this minor version (unless you reopen
the PR or upgrade to it yourself)
- `@dependabot ignore this dependency` will close this PR and stop
Dependabot creating any more for this dependency (unless you reopen the
PR or upgrade to it yourself)
You can disable automated security fix PRs for this repo from the
[Security Alerts
page](https://github.com/langchain-ai/langgraph/network/alerts).

</details>

Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
2026-04-15 19:36:52 -07:00
dependabot[bot]GitHubdependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
47fd42abb2 chore(deps): bump langsmith from 0.6.4 to 0.7.31 in /libs/checkpoint-sqlite (#7524)
Bumps [langsmith](https://github.com/langchain-ai/langsmith-sdk) from
0.6.4 to 0.7.31.
<details>
<summary>Release notes</summary>
<p><em>Sourced from <a
href="https://github.com/langchain-ai/langsmith-sdk/releases">langsmith's
releases</a>.</em></p>
<blockquote>
<h2>v0.7.31</h2>
<h2>What's Changed</h2>
<ul>
<li>chore(deps-dev): bump langchain-core from 1.2.23 to 1.2.28 in
/python by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2692">langchain-ai/langsmith-sdk#2692</a></li>
<li>chore(deps-dev): bump <code>@​anthropic-ai/sdk</code> from 0.82.0 to
0.84.0 in /js by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2684">langchain-ai/langsmith-sdk#2684</a></li>
<li>chore(deps): bump cryptography from 46.0.6 to 46.0.7 in /python by
<a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2693">langchain-ai/langsmith-sdk#2693</a></li>
<li>chore(deps-dev): bump <code>@​anthropic-ai/sdk</code> from 0.84.0 to
0.85.0 in /js by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2700">langchain-ai/langsmith-sdk#2700</a></li>
<li>feat(py): Tag OpenAI Agent Python SDK runs with ls_agent_type by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2699">langchain-ai/langsmith-sdk#2699</a></li>
<li>feat(js): Adds ls_agent_type metadata to AI SDK runs by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2701">langchain-ai/langsmith-sdk#2701</a></li>
<li>chore(deps-dev): bump types-tqdm from 4.67.3.20260303 to
4.67.3.20260408 in /python by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2710">langchain-ai/langsmith-sdk#2710</a></li>
<li>chore(deps): bump pnpm/action-setup from 5 to 6 by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2705">langchain-ai/langsmith-sdk#2705</a></li>
<li>chore(deps): bump the py-minor-and-patch group across 1 directory
with 10 updates by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2711">langchain-ai/langsmith-sdk#2711</a></li>
<li>chore(deps-dev): bump <code>@​anthropic-ai/sdk</code> from 0.85.0 to
0.86.0 in /js by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2702">langchain-ai/langsmith-sdk#2702</a></li>
<li>chore(deps): bump actions/github-script from 8 to 9 by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2706">langchain-ai/langsmith-sdk#2706</a></li>
<li>chore(deps-dev): bump the js-minor-and-patch group across 1
directory with 7 updates by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2712">langchain-ai/langsmith-sdk#2712</a></li>
<li>chore(deps-dev): bump types-psutil from 7.2.2.20260130 to
7.2.2.20260408 in /python by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2709">langchain-ai/langsmith-sdk#2709</a></li>
<li>chore(deps-dev): bump rich from 14.3.3 to 15.0.0 in /python by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2708">langchain-ai/langsmith-sdk#2708</a></li>
<li>feat: Filter kwargs from new token events by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2714">langchain-ai/langsmith-sdk#2714</a></li>
<li>release(py): 0.7.31 by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2716">langchain-ai/langsmith-sdk#2716</a></li>
</ul>
<p><strong>Full Changelog</strong>: <a
href="https://github.com/langchain-ai/langsmith-sdk/compare/v0.7.30...v0.7.31">https://github.com/langchain-ai/langsmith-sdk/compare/v0.7.30...v0.7.31</a></p>
<h2>v0.7.30</h2>
<h2>What's Changed</h2>
<ul>
<li>feat(python): add service feature to sandbox by <a
href="https://github.com/DanielKneipp"><code>@​DanielKneipp</code></a>
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2665">langchain-ai/langsmith-sdk#2665</a></li>
<li>fix(js): Fix prototype pollution bug in anonymizers by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2690">langchain-ai/langsmith-sdk#2690</a></li>
<li>release(js): 0.5.18 by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2691">langchain-ai/langsmith-sdk#2691</a></li>
<li>chore(js/sandbox): suppress warning log by <a
href="https://github.com/hntrl"><code>@​hntrl</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2694">langchain-ai/langsmith-sdk#2694</a></li>
<li>feat(js): Add metadata to Claude Agent SDK JS tracing by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2695">langchain-ai/langsmith-sdk#2695</a></li>
<li>fix(py): Fix run tree memory leak by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2696">langchain-ai/langsmith-sdk#2696</a></li>
<li>release(py): 0.7.30 by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2698">langchain-ai/langsmith-sdk#2698</a></li>
</ul>
<p><strong>Full Changelog</strong>: <a
href="https://github.com/langchain-ai/langsmith-sdk/compare/v0.7.29...v0.7.30">https://github.com/langchain-ai/langsmith-sdk/compare/v0.7.29...v0.7.30</a></p>
<h2>v0.7.29</h2>
<h2>What's Changed</h2>
<ul>
<li>release(js): 0.5.17 by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2681">langchain-ai/langsmith-sdk#2681</a></li>
<li>feat(py): Fix race condition around Claude Agent SDK instrumentation
by <a href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a>
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2685">langchain-ai/langsmith-sdk#2685</a></li>
<li>release(py): 0.7.29 by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2686">langchain-ai/langsmith-sdk#2686</a></li>
</ul>
<p><strong>Full Changelog</strong>: <a
href="https://github.com/langchain-ai/langsmith-sdk/compare/v0.7.28...v0.7.29">https://github.com/langchain-ai/langsmith-sdk/compare/v0.7.28...v0.7.29</a></p>
<h2>v0.7.28</h2>
<h2>What's Changed</h2>
<ul>
<li>feat(py): Support subagent tracing in Claude Agents SDK, fix usage
and duplicate messages by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2670">langchain-ai/langsmith-sdk#2670</a></li>
<li>chore(deps-dev): bump the py-minor-and-patch group across 1
directory with 11 updates by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2677">langchain-ai/langsmith-sdk#2677</a></li>
<li>chore(deps-dev): bump the js-minor-and-patch group across 1
directory with 8 updates by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2667">langchain-ai/langsmith-sdk#2667</a></li>
<li>chore(deps): bump pnpm/action-setup from 4 to 5 by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2658">langchain-ai/langsmith-sdk#2658</a></li>
</ul>
<!-- raw HTML omitted -->
</blockquote>
<p>... (truncated)</p>
</details>
<details>
<summary>Commits</summary>
<ul>
<li><a
href="https://github.com/langchain-ai/langsmith-sdk/commit/c434999d05c00334efeba88b8bbd2de9f3afbef6"><code>c434999</code></a>
release(py): 0.7.31 (<a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/issues/2716">#2716</a>)</li>
<li><a
href="https://github.com/langchain-ai/langsmith-sdk/commit/47d7c4a783333e716395d802e7632f1f1b4744d3"><code>47d7c4a</code></a>
feat: Filter kwargs from new token events (<a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/issues/2714">#2714</a>)</li>
<li><a
href="https://github.com/langchain-ai/langsmith-sdk/commit/3c57445b543c9a2f86db52024ea2c998bfc2ffab"><code>3c57445</code></a>
chore(deps-dev): bump rich from 14.3.3 to 15.0.0 in /python (<a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/issues/2708">#2708</a>)</li>
<li><a
href="https://github.com/langchain-ai/langsmith-sdk/commit/2be6cd01a2b6e35e811488d3561e7b0b57b06f63"><code>2be6cd0</code></a>
chore(deps-dev): bump types-psutil from 7.2.2.20260130 to 7.2.2.20260408
in /...</li>
<li><a
href="https://github.com/langchain-ai/langsmith-sdk/commit/b8b6ca32d43c919c07a4e13c99a83bcaab8accb0"><code>b8b6ca3</code></a>
chore(deps-dev): bump the js-minor-and-patch group across 1 directory
with 7 ...</li>
<li><a
href="https://github.com/langchain-ai/langsmith-sdk/commit/9897cb33da7698291637f268edd833ca3e1adde6"><code>9897cb3</code></a>
chore(deps): bump actions/github-script from 8 to 9 (<a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/issues/2706">#2706</a>)</li>
<li><a
href="https://github.com/langchain-ai/langsmith-sdk/commit/572c0184285747e027a796e03ea6c9ba171e09a6"><code>572c018</code></a>
chore(deps-dev): bump <code>@​anthropic-ai/sdk</code> from 0.85.0 to
0.86.0 in /js (<a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/issues/2702">#2702</a>)</li>
<li><a
href="https://github.com/langchain-ai/langsmith-sdk/commit/57447524c88b6bba2775161aa449da32fb8e5c42"><code>5744752</code></a>
chore(deps): bump the py-minor-and-patch group across 1 directory with
10 upd...</li>
<li><a
href="https://github.com/langchain-ai/langsmith-sdk/commit/960cae7f490e9ccbe428e6b56c8047bdb7b942a5"><code>960cae7</code></a>
chore(deps): bump pnpm/action-setup from 5 to 6 (<a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/issues/2705">#2705</a>)</li>
<li><a
href="https://github.com/langchain-ai/langsmith-sdk/commit/9370e7670abf7f8f9a36fbb72250bcfd2f91e7c6"><code>9370e76</code></a>
chore(deps-dev): bump types-tqdm from 4.67.3.20260303 to 4.67.3.20260408
in /...</li>
<li>Additional commits viewable in <a
href="https://github.com/langchain-ai/langsmith-sdk/compare/v0.6.4...v0.7.31">compare
view</a></li>
</ul>
</details>
<br />


[![Dependabot compatibility
score](https://dependabot-badges.githubapp.com/badges/compatibility_score?dependency-name=langsmith&package-manager=uv&previous-version=0.6.4&new-version=0.7.31)](https://docs.github.com/en/github/managing-security-vulnerabilities/about-dependabot-security-updates#about-compatibility-scores)

Dependabot will resolve any conflicts with this PR as long as you don't
alter it yourself. You can also trigger a rebase manually by commenting
`@dependabot rebase`.

[//]: # (dependabot-automerge-start)
[//]: # (dependabot-automerge-end)

---

<details>
<summary>Dependabot commands and options</summary>
<br />

You can trigger Dependabot actions by commenting on this PR:
- `@dependabot rebase` will rebase this PR
- `@dependabot recreate` will recreate this PR, overwriting any edits
that have been made to it
- `@dependabot show <dependency name> ignore conditions` will show all
of the ignore conditions of the specified dependency
- `@dependabot ignore this major version` will close this PR and stop
Dependabot creating any more for this major version (unless you reopen
the PR or upgrade to it yourself)
- `@dependabot ignore this minor version` will close this PR and stop
Dependabot creating any more for this minor version (unless you reopen
the PR or upgrade to it yourself)
- `@dependabot ignore this dependency` will close this PR and stop
Dependabot creating any more for this dependency (unless you reopen the
PR or upgrade to it yourself)
You can disable automated security fix PRs for this repo from the
[Security Alerts
page](https://github.com/langchain-ai/langgraph/network/alerts).

</details>

Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
2026-04-15 19:36:42 -07:00
dependabot[bot]GitHubdependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
554b2db1f8 chore(deps): bump langsmith from 0.7.3 to 0.7.31 in /libs/checkpoint-conformance (#7523)
Bumps [langsmith](https://github.com/langchain-ai/langsmith-sdk) from
0.7.3 to 0.7.31.
<details>
<summary>Release notes</summary>
<p><em>Sourced from <a
href="https://github.com/langchain-ai/langsmith-sdk/releases">langsmith's
releases</a>.</em></p>
<blockquote>
<h2>v0.7.31</h2>
<h2>What's Changed</h2>
<ul>
<li>chore(deps-dev): bump langchain-core from 1.2.23 to 1.2.28 in
/python by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2692">langchain-ai/langsmith-sdk#2692</a></li>
<li>chore(deps-dev): bump <code>@​anthropic-ai/sdk</code> from 0.82.0 to
0.84.0 in /js by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2684">langchain-ai/langsmith-sdk#2684</a></li>
<li>chore(deps): bump cryptography from 46.0.6 to 46.0.7 in /python by
<a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2693">langchain-ai/langsmith-sdk#2693</a></li>
<li>chore(deps-dev): bump <code>@​anthropic-ai/sdk</code> from 0.84.0 to
0.85.0 in /js by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2700">langchain-ai/langsmith-sdk#2700</a></li>
<li>feat(py): Tag OpenAI Agent Python SDK runs with ls_agent_type by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2699">langchain-ai/langsmith-sdk#2699</a></li>
<li>feat(js): Adds ls_agent_type metadata to AI SDK runs by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2701">langchain-ai/langsmith-sdk#2701</a></li>
<li>chore(deps-dev): bump types-tqdm from 4.67.3.20260303 to
4.67.3.20260408 in /python by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2710">langchain-ai/langsmith-sdk#2710</a></li>
<li>chore(deps): bump pnpm/action-setup from 5 to 6 by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2705">langchain-ai/langsmith-sdk#2705</a></li>
<li>chore(deps): bump the py-minor-and-patch group across 1 directory
with 10 updates by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2711">langchain-ai/langsmith-sdk#2711</a></li>
<li>chore(deps-dev): bump <code>@​anthropic-ai/sdk</code> from 0.85.0 to
0.86.0 in /js by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2702">langchain-ai/langsmith-sdk#2702</a></li>
<li>chore(deps): bump actions/github-script from 8 to 9 by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2706">langchain-ai/langsmith-sdk#2706</a></li>
<li>chore(deps-dev): bump the js-minor-and-patch group across 1
directory with 7 updates by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2712">langchain-ai/langsmith-sdk#2712</a></li>
<li>chore(deps-dev): bump types-psutil from 7.2.2.20260130 to
7.2.2.20260408 in /python by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2709">langchain-ai/langsmith-sdk#2709</a></li>
<li>chore(deps-dev): bump rich from 14.3.3 to 15.0.0 in /python by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2708">langchain-ai/langsmith-sdk#2708</a></li>
<li>feat: Filter kwargs from new token events by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2714">langchain-ai/langsmith-sdk#2714</a></li>
<li>release(py): 0.7.31 by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2716">langchain-ai/langsmith-sdk#2716</a></li>
</ul>
<p><strong>Full Changelog</strong>: <a
href="https://github.com/langchain-ai/langsmith-sdk/compare/v0.7.30...v0.7.31">https://github.com/langchain-ai/langsmith-sdk/compare/v0.7.30...v0.7.31</a></p>
<h2>v0.7.30</h2>
<h2>What's Changed</h2>
<ul>
<li>feat(python): add service feature to sandbox by <a
href="https://github.com/DanielKneipp"><code>@​DanielKneipp</code></a>
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2665">langchain-ai/langsmith-sdk#2665</a></li>
<li>fix(js): Fix prototype pollution bug in anonymizers by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2690">langchain-ai/langsmith-sdk#2690</a></li>
<li>release(js): 0.5.18 by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2691">langchain-ai/langsmith-sdk#2691</a></li>
<li>chore(js/sandbox): suppress warning log by <a
href="https://github.com/hntrl"><code>@​hntrl</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2694">langchain-ai/langsmith-sdk#2694</a></li>
<li>feat(js): Add metadata to Claude Agent SDK JS tracing by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2695">langchain-ai/langsmith-sdk#2695</a></li>
<li>fix(py): Fix run tree memory leak by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2696">langchain-ai/langsmith-sdk#2696</a></li>
<li>release(py): 0.7.30 by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2698">langchain-ai/langsmith-sdk#2698</a></li>
</ul>
<p><strong>Full Changelog</strong>: <a
href="https://github.com/langchain-ai/langsmith-sdk/compare/v0.7.29...v0.7.30">https://github.com/langchain-ai/langsmith-sdk/compare/v0.7.29...v0.7.30</a></p>
<h2>v0.7.29</h2>
<h2>What's Changed</h2>
<ul>
<li>release(js): 0.5.17 by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2681">langchain-ai/langsmith-sdk#2681</a></li>
<li>feat(py): Fix race condition around Claude Agent SDK instrumentation
by <a href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a>
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2685">langchain-ai/langsmith-sdk#2685</a></li>
<li>release(py): 0.7.29 by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2686">langchain-ai/langsmith-sdk#2686</a></li>
</ul>
<p><strong>Full Changelog</strong>: <a
href="https://github.com/langchain-ai/langsmith-sdk/compare/v0.7.28...v0.7.29">https://github.com/langchain-ai/langsmith-sdk/compare/v0.7.28...v0.7.29</a></p>
<h2>v0.7.28</h2>
<h2>What's Changed</h2>
<ul>
<li>feat(py): Support subagent tracing in Claude Agents SDK, fix usage
and duplicate messages by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2670">langchain-ai/langsmith-sdk#2670</a></li>
<li>chore(deps-dev): bump the py-minor-and-patch group across 1
directory with 11 updates by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2677">langchain-ai/langsmith-sdk#2677</a></li>
<li>chore(deps-dev): bump the js-minor-and-patch group across 1
directory with 8 updates by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2667">langchain-ai/langsmith-sdk#2667</a></li>
<li>chore(deps): bump pnpm/action-setup from 4 to 5 by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2658">langchain-ai/langsmith-sdk#2658</a></li>
</ul>
<!-- raw HTML omitted -->
</blockquote>
<p>... (truncated)</p>
</details>
<details>
<summary>Commits</summary>
<ul>
<li><a
href="https://github.com/langchain-ai/langsmith-sdk/commit/c434999d05c00334efeba88b8bbd2de9f3afbef6"><code>c434999</code></a>
release(py): 0.7.31 (<a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/issues/2716">#2716</a>)</li>
<li><a
href="https://github.com/langchain-ai/langsmith-sdk/commit/47d7c4a783333e716395d802e7632f1f1b4744d3"><code>47d7c4a</code></a>
feat: Filter kwargs from new token events (<a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/issues/2714">#2714</a>)</li>
<li><a
href="https://github.com/langchain-ai/langsmith-sdk/commit/3c57445b543c9a2f86db52024ea2c998bfc2ffab"><code>3c57445</code></a>
chore(deps-dev): bump rich from 14.3.3 to 15.0.0 in /python (<a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/issues/2708">#2708</a>)</li>
<li><a
href="https://github.com/langchain-ai/langsmith-sdk/commit/2be6cd01a2b6e35e811488d3561e7b0b57b06f63"><code>2be6cd0</code></a>
chore(deps-dev): bump types-psutil from 7.2.2.20260130 to 7.2.2.20260408
in /...</li>
<li><a
href="https://github.com/langchain-ai/langsmith-sdk/commit/b8b6ca32d43c919c07a4e13c99a83bcaab8accb0"><code>b8b6ca3</code></a>
chore(deps-dev): bump the js-minor-and-patch group across 1 directory
with 7 ...</li>
<li><a
href="https://github.com/langchain-ai/langsmith-sdk/commit/9897cb33da7698291637f268edd833ca3e1adde6"><code>9897cb3</code></a>
chore(deps): bump actions/github-script from 8 to 9 (<a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/issues/2706">#2706</a>)</li>
<li><a
href="https://github.com/langchain-ai/langsmith-sdk/commit/572c0184285747e027a796e03ea6c9ba171e09a6"><code>572c018</code></a>
chore(deps-dev): bump <code>@​anthropic-ai/sdk</code> from 0.85.0 to
0.86.0 in /js (<a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/issues/2702">#2702</a>)</li>
<li><a
href="https://github.com/langchain-ai/langsmith-sdk/commit/57447524c88b6bba2775161aa449da32fb8e5c42"><code>5744752</code></a>
chore(deps): bump the py-minor-and-patch group across 1 directory with
10 upd...</li>
<li><a
href="https://github.com/langchain-ai/langsmith-sdk/commit/960cae7f490e9ccbe428e6b56c8047bdb7b942a5"><code>960cae7</code></a>
chore(deps): bump pnpm/action-setup from 5 to 6 (<a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/issues/2705">#2705</a>)</li>
<li><a
href="https://github.com/langchain-ai/langsmith-sdk/commit/9370e7670abf7f8f9a36fbb72250bcfd2f91e7c6"><code>9370e76</code></a>
chore(deps-dev): bump types-tqdm from 4.67.3.20260303 to 4.67.3.20260408
in /...</li>
<li>Additional commits viewable in <a
href="https://github.com/langchain-ai/langsmith-sdk/compare/v0.7.3...v0.7.31">compare
view</a></li>
</ul>
</details>
<br />


[![Dependabot compatibility
score](https://dependabot-badges.githubapp.com/badges/compatibility_score?dependency-name=langsmith&package-manager=uv&previous-version=0.7.3&new-version=0.7.31)](https://docs.github.com/en/github/managing-security-vulnerabilities/about-dependabot-security-updates#about-compatibility-scores)

Dependabot will resolve any conflicts with this PR as long as you don't
alter it yourself. You can also trigger a rebase manually by commenting
`@dependabot rebase`.

[//]: # (dependabot-automerge-start)
[//]: # (dependabot-automerge-end)

---

<details>
<summary>Dependabot commands and options</summary>
<br />

You can trigger Dependabot actions by commenting on this PR:
- `@dependabot rebase` will rebase this PR
- `@dependabot recreate` will recreate this PR, overwriting any edits
that have been made to it
- `@dependabot show <dependency name> ignore conditions` will show all
of the ignore conditions of the specified dependency
- `@dependabot ignore this major version` will close this PR and stop
Dependabot creating any more for this major version (unless you reopen
the PR or upgrade to it yourself)
- `@dependabot ignore this minor version` will close this PR and stop
Dependabot creating any more for this minor version (unless you reopen
the PR or upgrade to it yourself)
- `@dependabot ignore this dependency` will close this PR and stop
Dependabot creating any more for this dependency (unless you reopen the
PR or upgrade to it yourself)
You can disable automated security fix PRs for this repo from the
[Security Alerts
page](https://github.com/langchain-ai/langgraph/network/alerts).

</details>

Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
2026-04-15 19:36:27 -07:00
dependabot[bot]GitHubdependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
93a144b404 chore(deps): bump the uv group across 2 directories with 1 update (#7531)
Bumps the uv group with 1 update in the /libs/cli/uv-examples/monorepo
directory: [langsmith](https://github.com/langchain-ai/langsmith-sdk).
Bumps the uv group with 1 update in the /libs/cli/uv-examples/simple
directory: [langsmith](https://github.com/langchain-ai/langsmith-sdk).

Updates `langsmith` from 0.7.26 to 0.7.31
<details>
<summary>Release notes</summary>
<p><em>Sourced from <a
href="https://github.com/langchain-ai/langsmith-sdk/releases">langsmith's
releases</a>.</em></p>
<blockquote>
<h2>v0.7.31</h2>
<h2>What's Changed</h2>
<ul>
<li>chore(deps-dev): bump langchain-core from 1.2.23 to 1.2.28 in
/python by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2692">langchain-ai/langsmith-sdk#2692</a></li>
<li>chore(deps-dev): bump <code>@​anthropic-ai/sdk</code> from 0.82.0 to
0.84.0 in /js by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2684">langchain-ai/langsmith-sdk#2684</a></li>
<li>chore(deps): bump cryptography from 46.0.6 to 46.0.7 in /python by
<a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2693">langchain-ai/langsmith-sdk#2693</a></li>
<li>chore(deps-dev): bump <code>@​anthropic-ai/sdk</code> from 0.84.0 to
0.85.0 in /js by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2700">langchain-ai/langsmith-sdk#2700</a></li>
<li>feat(py): Tag OpenAI Agent Python SDK runs with ls_agent_type by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2699">langchain-ai/langsmith-sdk#2699</a></li>
<li>feat(js): Adds ls_agent_type metadata to AI SDK runs by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2701">langchain-ai/langsmith-sdk#2701</a></li>
<li>chore(deps-dev): bump types-tqdm from 4.67.3.20260303 to
4.67.3.20260408 in /python by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2710">langchain-ai/langsmith-sdk#2710</a></li>
<li>chore(deps): bump pnpm/action-setup from 5 to 6 by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2705">langchain-ai/langsmith-sdk#2705</a></li>
<li>chore(deps): bump the py-minor-and-patch group across 1 directory
with 10 updates by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2711">langchain-ai/langsmith-sdk#2711</a></li>
<li>chore(deps-dev): bump <code>@​anthropic-ai/sdk</code> from 0.85.0 to
0.86.0 in /js by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2702">langchain-ai/langsmith-sdk#2702</a></li>
<li>chore(deps): bump actions/github-script from 8 to 9 by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2706">langchain-ai/langsmith-sdk#2706</a></li>
<li>chore(deps-dev): bump the js-minor-and-patch group across 1
directory with 7 updates by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2712">langchain-ai/langsmith-sdk#2712</a></li>
<li>chore(deps-dev): bump types-psutil from 7.2.2.20260130 to
7.2.2.20260408 in /python by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2709">langchain-ai/langsmith-sdk#2709</a></li>
<li>chore(deps-dev): bump rich from 14.3.3 to 15.0.0 in /python by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2708">langchain-ai/langsmith-sdk#2708</a></li>
<li>feat: Filter kwargs from new token events by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2714">langchain-ai/langsmith-sdk#2714</a></li>
<li>release(py): 0.7.31 by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2716">langchain-ai/langsmith-sdk#2716</a></li>
</ul>
<p><strong>Full Changelog</strong>: <a
href="https://github.com/langchain-ai/langsmith-sdk/compare/v0.7.30...v0.7.31">https://github.com/langchain-ai/langsmith-sdk/compare/v0.7.30...v0.7.31</a></p>
<h2>v0.7.30</h2>
<h2>What's Changed</h2>
<ul>
<li>feat(python): add service feature to sandbox by <a
href="https://github.com/DanielKneipp"><code>@​DanielKneipp</code></a>
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2665">langchain-ai/langsmith-sdk#2665</a></li>
<li>fix(js): Fix prototype pollution bug in anonymizers by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2690">langchain-ai/langsmith-sdk#2690</a></li>
<li>release(js): 0.5.18 by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2691">langchain-ai/langsmith-sdk#2691</a></li>
<li>chore(js/sandbox): suppress warning log by <a
href="https://github.com/hntrl"><code>@​hntrl</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2694">langchain-ai/langsmith-sdk#2694</a></li>
<li>feat(js): Add metadata to Claude Agent SDK JS tracing by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2695">langchain-ai/langsmith-sdk#2695</a></li>
<li>fix(py): Fix run tree memory leak by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2696">langchain-ai/langsmith-sdk#2696</a></li>
<li>release(py): 0.7.30 by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2698">langchain-ai/langsmith-sdk#2698</a></li>
</ul>
<p><strong>Full Changelog</strong>: <a
href="https://github.com/langchain-ai/langsmith-sdk/compare/v0.7.29...v0.7.30">https://github.com/langchain-ai/langsmith-sdk/compare/v0.7.29...v0.7.30</a></p>
<h2>v0.7.29</h2>
<h2>What's Changed</h2>
<ul>
<li>release(js): 0.5.17 by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2681">langchain-ai/langsmith-sdk#2681</a></li>
<li>feat(py): Fix race condition around Claude Agent SDK instrumentation
by <a href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a>
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2685">langchain-ai/langsmith-sdk#2685</a></li>
<li>release(py): 0.7.29 by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2686">langchain-ai/langsmith-sdk#2686</a></li>
</ul>
<p><strong>Full Changelog</strong>: <a
href="https://github.com/langchain-ai/langsmith-sdk/compare/v0.7.28...v0.7.29">https://github.com/langchain-ai/langsmith-sdk/compare/v0.7.28...v0.7.29</a></p>
<h2>v0.7.28</h2>
<h2>What's Changed</h2>
<ul>
<li>feat(py): Support subagent tracing in Claude Agents SDK, fix usage
and duplicate messages by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2670">langchain-ai/langsmith-sdk#2670</a></li>
<li>chore(deps-dev): bump the py-minor-and-patch group across 1
directory with 11 updates by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2677">langchain-ai/langsmith-sdk#2677</a></li>
<li>chore(deps-dev): bump the js-minor-and-patch group across 1
directory with 8 updates by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2667">langchain-ai/langsmith-sdk#2667</a></li>
<li>chore(deps): bump pnpm/action-setup from 4 to 5 by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2658">langchain-ai/langsmith-sdk#2658</a></li>
</ul>
<!-- raw HTML omitted -->
</blockquote>
<p>... (truncated)</p>
</details>
<details>
<summary>Commits</summary>
<ul>
<li><a
href="https://github.com/langchain-ai/langsmith-sdk/commit/c434999d05c00334efeba88b8bbd2de9f3afbef6"><code>c434999</code></a>
release(py): 0.7.31 (<a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/issues/2716">#2716</a>)</li>
<li><a
href="https://github.com/langchain-ai/langsmith-sdk/commit/47d7c4a783333e716395d802e7632f1f1b4744d3"><code>47d7c4a</code></a>
feat: Filter kwargs from new token events (<a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/issues/2714">#2714</a>)</li>
<li><a
href="https://github.com/langchain-ai/langsmith-sdk/commit/3c57445b543c9a2f86db52024ea2c998bfc2ffab"><code>3c57445</code></a>
chore(deps-dev): bump rich from 14.3.3 to 15.0.0 in /python (<a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/issues/2708">#2708</a>)</li>
<li><a
href="https://github.com/langchain-ai/langsmith-sdk/commit/2be6cd01a2b6e35e811488d3561e7b0b57b06f63"><code>2be6cd0</code></a>
chore(deps-dev): bump types-psutil from 7.2.2.20260130 to 7.2.2.20260408
in /...</li>
<li><a
href="https://github.com/langchain-ai/langsmith-sdk/commit/b8b6ca32d43c919c07a4e13c99a83bcaab8accb0"><code>b8b6ca3</code></a>
chore(deps-dev): bump the js-minor-and-patch group across 1 directory
with 7 ...</li>
<li><a
href="https://github.com/langchain-ai/langsmith-sdk/commit/9897cb33da7698291637f268edd833ca3e1adde6"><code>9897cb3</code></a>
chore(deps): bump actions/github-script from 8 to 9 (<a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/issues/2706">#2706</a>)</li>
<li><a
href="https://github.com/langchain-ai/langsmith-sdk/commit/572c0184285747e027a796e03ea6c9ba171e09a6"><code>572c018</code></a>
chore(deps-dev): bump <code>@​anthropic-ai/sdk</code> from 0.85.0 to
0.86.0 in /js (<a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/issues/2702">#2702</a>)</li>
<li><a
href="https://github.com/langchain-ai/langsmith-sdk/commit/57447524c88b6bba2775161aa449da32fb8e5c42"><code>5744752</code></a>
chore(deps): bump the py-minor-and-patch group across 1 directory with
10 upd...</li>
<li><a
href="https://github.com/langchain-ai/langsmith-sdk/commit/960cae7f490e9ccbe428e6b56c8047bdb7b942a5"><code>960cae7</code></a>
chore(deps): bump pnpm/action-setup from 5 to 6 (<a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/issues/2705">#2705</a>)</li>
<li><a
href="https://github.com/langchain-ai/langsmith-sdk/commit/9370e7670abf7f8f9a36fbb72250bcfd2f91e7c6"><code>9370e76</code></a>
chore(deps-dev): bump types-tqdm from 4.67.3.20260303 to 4.67.3.20260408
in /...</li>
<li>Additional commits viewable in <a
href="https://github.com/langchain-ai/langsmith-sdk/compare/v0.7.26...v0.7.31">compare
view</a></li>
</ul>
</details>
<br />

Updates `langsmith` from 0.7.26 to 0.7.31
<details>
<summary>Release notes</summary>
<p><em>Sourced from <a
href="https://github.com/langchain-ai/langsmith-sdk/releases">langsmith's
releases</a>.</em></p>
<blockquote>
<h2>v0.7.31</h2>
<h2>What's Changed</h2>
<ul>
<li>chore(deps-dev): bump langchain-core from 1.2.23 to 1.2.28 in
/python by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2692">langchain-ai/langsmith-sdk#2692</a></li>
<li>chore(deps-dev): bump <code>@​anthropic-ai/sdk</code> from 0.82.0 to
0.84.0 in /js by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2684">langchain-ai/langsmith-sdk#2684</a></li>
<li>chore(deps): bump cryptography from 46.0.6 to 46.0.7 in /python by
<a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2693">langchain-ai/langsmith-sdk#2693</a></li>
<li>chore(deps-dev): bump <code>@​anthropic-ai/sdk</code> from 0.84.0 to
0.85.0 in /js by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2700">langchain-ai/langsmith-sdk#2700</a></li>
<li>feat(py): Tag OpenAI Agent Python SDK runs with ls_agent_type by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2699">langchain-ai/langsmith-sdk#2699</a></li>
<li>feat(js): Adds ls_agent_type metadata to AI SDK runs by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2701">langchain-ai/langsmith-sdk#2701</a></li>
<li>chore(deps-dev): bump types-tqdm from 4.67.3.20260303 to
4.67.3.20260408 in /python by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2710">langchain-ai/langsmith-sdk#2710</a></li>
<li>chore(deps): bump pnpm/action-setup from 5 to 6 by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2705">langchain-ai/langsmith-sdk#2705</a></li>
<li>chore(deps): bump the py-minor-and-patch group across 1 directory
with 10 updates by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2711">langchain-ai/langsmith-sdk#2711</a></li>
<li>chore(deps-dev): bump <code>@​anthropic-ai/sdk</code> from 0.85.0 to
0.86.0 in /js by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2702">langchain-ai/langsmith-sdk#2702</a></li>
<li>chore(deps): bump actions/github-script from 8 to 9 by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2706">langchain-ai/langsmith-sdk#2706</a></li>
<li>chore(deps-dev): bump the js-minor-and-patch group across 1
directory with 7 updates by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2712">langchain-ai/langsmith-sdk#2712</a></li>
<li>chore(deps-dev): bump types-psutil from 7.2.2.20260130 to
7.2.2.20260408 in /python by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2709">langchain-ai/langsmith-sdk#2709</a></li>
<li>chore(deps-dev): bump rich from 14.3.3 to 15.0.0 in /python by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2708">langchain-ai/langsmith-sdk#2708</a></li>
<li>feat: Filter kwargs from new token events by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2714">langchain-ai/langsmith-sdk#2714</a></li>
<li>release(py): 0.7.31 by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2716">langchain-ai/langsmith-sdk#2716</a></li>
</ul>
<p><strong>Full Changelog</strong>: <a
href="https://github.com/langchain-ai/langsmith-sdk/compare/v0.7.30...v0.7.31">https://github.com/langchain-ai/langsmith-sdk/compare/v0.7.30...v0.7.31</a></p>
<h2>v0.7.30</h2>
<h2>What's Changed</h2>
<ul>
<li>feat(python): add service feature to sandbox by <a
href="https://github.com/DanielKneipp"><code>@​DanielKneipp</code></a>
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2665">langchain-ai/langsmith-sdk#2665</a></li>
<li>fix(js): Fix prototype pollution bug in anonymizers by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2690">langchain-ai/langsmith-sdk#2690</a></li>
<li>release(js): 0.5.18 by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2691">langchain-ai/langsmith-sdk#2691</a></li>
<li>chore(js/sandbox): suppress warning log by <a
href="https://github.com/hntrl"><code>@​hntrl</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2694">langchain-ai/langsmith-sdk#2694</a></li>
<li>feat(js): Add metadata to Claude Agent SDK JS tracing by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2695">langchain-ai/langsmith-sdk#2695</a></li>
<li>fix(py): Fix run tree memory leak by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2696">langchain-ai/langsmith-sdk#2696</a></li>
<li>release(py): 0.7.30 by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2698">langchain-ai/langsmith-sdk#2698</a></li>
</ul>
<p><strong>Full Changelog</strong>: <a
href="https://github.com/langchain-ai/langsmith-sdk/compare/v0.7.29...v0.7.30">https://github.com/langchain-ai/langsmith-sdk/compare/v0.7.29...v0.7.30</a></p>
<h2>v0.7.29</h2>
<h2>What's Changed</h2>
<ul>
<li>release(js): 0.5.17 by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2681">langchain-ai/langsmith-sdk#2681</a></li>
<li>feat(py): Fix race condition around Claude Agent SDK instrumentation
by <a href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a>
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2685">langchain-ai/langsmith-sdk#2685</a></li>
<li>release(py): 0.7.29 by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2686">langchain-ai/langsmith-sdk#2686</a></li>
</ul>
<p><strong>Full Changelog</strong>: <a
href="https://github.com/langchain-ai/langsmith-sdk/compare/v0.7.28...v0.7.29">https://github.com/langchain-ai/langsmith-sdk/compare/v0.7.28...v0.7.29</a></p>
<h2>v0.7.28</h2>
<h2>What's Changed</h2>
<ul>
<li>feat(py): Support subagent tracing in Claude Agents SDK, fix usage
and duplicate messages by <a
href="https://github.com/jacoblee93"><code>@​jacoblee93</code></a> in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2670">langchain-ai/langsmith-sdk#2670</a></li>
<li>chore(deps-dev): bump the py-minor-and-patch group across 1
directory with 11 updates by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2677">langchain-ai/langsmith-sdk#2677</a></li>
<li>chore(deps-dev): bump the js-minor-and-patch group across 1
directory with 8 updates by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2667">langchain-ai/langsmith-sdk#2667</a></li>
<li>chore(deps): bump pnpm/action-setup from 4 to 5 by <a
href="https://github.com/dependabot"><code>@​dependabot</code></a>[bot]
in <a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/pull/2658">langchain-ai/langsmith-sdk#2658</a></li>
</ul>
<!-- raw HTML omitted -->
</blockquote>
<p>... (truncated)</p>
</details>
<details>
<summary>Commits</summary>
<ul>
<li><a
href="https://github.com/langchain-ai/langsmith-sdk/commit/c434999d05c00334efeba88b8bbd2de9f3afbef6"><code>c434999</code></a>
release(py): 0.7.31 (<a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/issues/2716">#2716</a>)</li>
<li><a
href="https://github.com/langchain-ai/langsmith-sdk/commit/47d7c4a783333e716395d802e7632f1f1b4744d3"><code>47d7c4a</code></a>
feat: Filter kwargs from new token events (<a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/issues/2714">#2714</a>)</li>
<li><a
href="https://github.com/langchain-ai/langsmith-sdk/commit/3c57445b543c9a2f86db52024ea2c998bfc2ffab"><code>3c57445</code></a>
chore(deps-dev): bump rich from 14.3.3 to 15.0.0 in /python (<a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/issues/2708">#2708</a>)</li>
<li><a
href="https://github.com/langchain-ai/langsmith-sdk/commit/2be6cd01a2b6e35e811488d3561e7b0b57b06f63"><code>2be6cd0</code></a>
chore(deps-dev): bump types-psutil from 7.2.2.20260130 to 7.2.2.20260408
in /...</li>
<li><a
href="https://github.com/langchain-ai/langsmith-sdk/commit/b8b6ca32d43c919c07a4e13c99a83bcaab8accb0"><code>b8b6ca3</code></a>
chore(deps-dev): bump the js-minor-and-patch group across 1 directory
with 7 ...</li>
<li><a
href="https://github.com/langchain-ai/langsmith-sdk/commit/9897cb33da7698291637f268edd833ca3e1adde6"><code>9897cb3</code></a>
chore(deps): bump actions/github-script from 8 to 9 (<a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/issues/2706">#2706</a>)</li>
<li><a
href="https://github.com/langchain-ai/langsmith-sdk/commit/572c0184285747e027a796e03ea6c9ba171e09a6"><code>572c018</code></a>
chore(deps-dev): bump <code>@​anthropic-ai/sdk</code> from 0.85.0 to
0.86.0 in /js (<a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/issues/2702">#2702</a>)</li>
<li><a
href="https://github.com/langchain-ai/langsmith-sdk/commit/57447524c88b6bba2775161aa449da32fb8e5c42"><code>5744752</code></a>
chore(deps): bump the py-minor-and-patch group across 1 directory with
10 upd...</li>
<li><a
href="https://github.com/langchain-ai/langsmith-sdk/commit/960cae7f490e9ccbe428e6b56c8047bdb7b942a5"><code>960cae7</code></a>
chore(deps): bump pnpm/action-setup from 5 to 6 (<a
href="https://redirect.github.com/langchain-ai/langsmith-sdk/issues/2705">#2705</a>)</li>
<li><a
href="https://github.com/langchain-ai/langsmith-sdk/commit/9370e7670abf7f8f9a36fbb72250bcfd2f91e7c6"><code>9370e76</code></a>
chore(deps-dev): bump types-tqdm from 4.67.3.20260303 to 4.67.3.20260408
in /...</li>
<li>Additional commits viewable in <a
href="https://github.com/langchain-ai/langsmith-sdk/compare/v0.7.26...v0.7.31">compare
view</a></li>
</ul>
</details>
<br />


Dependabot will resolve any conflicts with this PR as long as you don't
alter it yourself. You can also trigger a rebase manually by commenting
`@dependabot rebase`.

[//]: # (dependabot-automerge-start)
[//]: # (dependabot-automerge-end)

---

<details>
<summary>Dependabot commands and options</summary>
<br />

You can trigger Dependabot actions by commenting on this PR:
- `@dependabot rebase` will rebase this PR
- `@dependabot recreate` will recreate this PR, overwriting any edits
that have been made to it
- `@dependabot show <dependency name> ignore conditions` will show all
of the ignore conditions of the specified dependency
- `@dependabot ignore <dependency name> major version` will close this
group update PR and stop Dependabot creating any more for the specific
dependency's major version (unless you unignore this specific
dependency's major version or upgrade to it yourself)
- `@dependabot ignore <dependency name> minor version` will close this
group update PR and stop Dependabot creating any more for the specific
dependency's minor version (unless you unignore this specific
dependency's minor version or upgrade to it yourself)
- `@dependabot ignore <dependency name>` will close this group update PR
and stop Dependabot creating any more for the specific dependency
(unless you unignore this specific dependency or upgrade to it yourself)
- `@dependabot unignore <dependency name>` will remove all of the ignore
conditions of the specified dependency
- `@dependabot unignore <dependency name> <ignore condition>` will
remove the ignore condition of the specified dependency and ignore
conditions
You can disable automated security fix PRs for this repo from the
[Security Alerts
page](https://github.com/langchain-ai/langgraph/network/alerts).

</details>

Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
2026-04-15 19:36:05 -07:00
dependabot[bot]GitHubdependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
2f4611db8f chore(deps): bump langsmith from 0.5.18 to 0.5.20 in /libs/cli/js-examples (#7522)
Bumps [langsmith](https://github.com/langchain-ai/langsmith-sdk) from
0.5.18 to 0.5.20.
<details>
<summary>Commits</summary>
<ul>
<li>See full diff in <a
href="https://github.com/langchain-ai/langsmith-sdk/commits">compare
view</a></li>
</ul>
</details>
<br />


[![Dependabot compatibility
score](https://dependabot-badges.githubapp.com/badges/compatibility_score?dependency-name=langsmith&package-manager=npm_and_yarn&previous-version=0.5.18&new-version=0.5.20)](https://docs.github.com/en/github/managing-security-vulnerabilities/about-dependabot-security-updates#about-compatibility-scores)

Dependabot will resolve any conflicts with this PR as long as you don't
alter it yourself. You can also trigger a rebase manually by commenting
`@dependabot rebase`.

[//]: # (dependabot-automerge-start)
[//]: # (dependabot-automerge-end)

---

<details>
<summary>Dependabot commands and options</summary>
<br />

You can trigger Dependabot actions by commenting on this PR:
- `@dependabot rebase` will rebase this PR
- `@dependabot recreate` will recreate this PR, overwriting any edits
that have been made to it
- `@dependabot show <dependency name> ignore conditions` will show all
of the ignore conditions of the specified dependency
- `@dependabot ignore this major version` will close this PR and stop
Dependabot creating any more for this major version (unless you reopen
the PR or upgrade to it yourself)
- `@dependabot ignore this minor version` will close this PR and stop
Dependabot creating any more for this minor version (unless you reopen
the PR or upgrade to it yourself)
- `@dependabot ignore this dependency` will close this PR and stop
Dependabot creating any more for this dependency (unless you reopen the
PR or upgrade to it yourself)
You can disable automated security fix PRs for this repo from the
[Security Alerts
page](https://github.com/langchain-ai/langgraph/network/alerts).

</details>

Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
2026-04-15 19:21:10 -07:00
dependabot[bot]GitHubdependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
20c5d36efe chore(deps): bump langsmith from 0.5.18 to 0.5.20 in /libs/cli/js-monorepo-example (#7521)
Bumps [langsmith](https://github.com/langchain-ai/langsmith-sdk) from
0.5.18 to 0.5.20.
<details>
<summary>Commits</summary>
<ul>
<li>See full diff in <a
href="https://github.com/langchain-ai/langsmith-sdk/commits">compare
view</a></li>
</ul>
</details>
<br />


[![Dependabot compatibility
score](https://dependabot-badges.githubapp.com/badges/compatibility_score?dependency-name=langsmith&package-manager=npm_and_yarn&previous-version=0.5.18&new-version=0.5.20)](https://docs.github.com/en/github/managing-security-vulnerabilities/about-dependabot-security-updates#about-compatibility-scores)

Dependabot will resolve any conflicts with this PR as long as you don't
alter it yourself. You can also trigger a rebase manually by commenting
`@dependabot rebase`.

[//]: # (dependabot-automerge-start)
[//]: # (dependabot-automerge-end)

---

<details>
<summary>Dependabot commands and options</summary>
<br />

You can trigger Dependabot actions by commenting on this PR:
- `@dependabot rebase` will rebase this PR
- `@dependabot recreate` will recreate this PR, overwriting any edits
that have been made to it
- `@dependabot show <dependency name> ignore conditions` will show all
of the ignore conditions of the specified dependency
- `@dependabot ignore this major version` will close this PR and stop
Dependabot creating any more for this major version (unless you reopen
the PR or upgrade to it yourself)
- `@dependabot ignore this minor version` will close this PR and stop
Dependabot creating any more for this minor version (unless you reopen
the PR or upgrade to it yourself)
- `@dependabot ignore this dependency` will close this PR and stop
Dependabot creating any more for this dependency (unless you reopen the
PR or upgrade to it yourself)
You can disable automated security fix PRs for this repo from the
[Security Alerts
page](https://github.com/langchain-ai/langgraph/network/alerts).

</details>

Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
2026-04-15 19:21:00 -07:00
25470ea435 release(checkpoint): 4.0.2 (#7518)
Co-authored-by: Will Fu-Hinthorn <will@langchain.dev>
2026-04-15 20:52:57 +00:00
70 changed files with 6055 additions and 6607 deletions
+1
View File
@@ -100,3 +100,4 @@ dmypy.json
.turbo
.editorconfig
.scratch
.worktrees/
+3 -3
View File
@@ -306,7 +306,7 @@ test = [
[[package]]
name = "langsmith"
version = "0.7.3"
version = "0.7.31"
source = { registry = "https://pypi.org/simple" }
dependencies = [
{ name = "httpx" },
@@ -319,9 +319,9 @@ dependencies = [
{ name = "xxhash" },
{ name = "zstandard" },
]
sdist = { url = "https://files.pythonhosted.org/packages/8d/bc/8172fefad4f2da888a6d564a27d1fb7d4dbf3c640899c2b40c46235cbe98/langsmith-0.7.3.tar.gz", hash = "sha256:0223b97021af62d2cf53c8a378a27bd22e90a7327e45b353e0069ae60d5d6f9e", size = 988575, upload-time = "2026-02-13T23:25:32.916Z" }
sdist = { url = "https://files.pythonhosted.org/packages/e6/11/696019490992db5c87774dc20515529ef42a01e1d770fb754ed6d9b12fb0/langsmith-0.7.31.tar.gz", hash = "sha256:331ee4f7c26bb5be4022b9859b7d7b122cbf8c9d01d9f530114c1914b0349ffb", size = 1178480, upload-time = "2026-04-14T17:55:41.242Z" }
wheels = [
{ url = "https://files.pythonhosted.org/packages/f4/9d/5a68b6b5e313ffabbb9725d18a71edb48177fd6d3ad329c07801d2a8e862/langsmith-0.7.3-py3-none-any.whl", hash = "sha256:03659bf9274e6efcead361c9c31a7849ea565ae0d6c0d73e1d8b239029eff3be", size = 325718, upload-time = "2026-02-13T23:25:31.52Z" },
{ url = "https://files.pythonhosted.org/packages/1d/a1/a013cf458c301cda86a213dd153ce0a01c93f1ab5833f951e6a44c9763ce/langsmith-0.7.31-py3-none-any.whl", hash = "sha256:0291d49203f6e80dda011af1afda61eb0595a4d697adb684590a8805e1d61fb6", size = 373276, upload-time = "2026-04-14T17:55:39.677Z" },
]
[[package]]
@@ -8,11 +8,13 @@ from typing import Any
from langchain_core.runnables import RunnableConfig
from langgraph.checkpoint.base import (
DELTA_SENTINEL,
WRITES_IDX_MAP,
ChannelVersions,
Checkpoint,
CheckpointMetadata,
CheckpointTuple,
_ChannelWritesHistory,
get_checkpoint_id,
get_serializable_checkpoint_metadata,
)
@@ -23,7 +25,12 @@ from psycopg.types.json import Jsonb
from psycopg_pool import ConnectionPool
from langgraph.checkpoint.postgres import _internal
from langgraph.checkpoint.postgres.base import BasePostgresSaver
from langgraph.checkpoint.postgres.base import (
SELECT_DELTA_BLOBS_SQL,
SELECT_DELTA_PARENTS_SQL,
SELECT_DELTA_WRITES_SQL,
BasePostgresSaver,
)
from langgraph.checkpoint.postgres.shallow import ShallowPostgresSaver
Conn = _internal.Conn # For backward compatibility
@@ -430,6 +437,42 @@ class PostgresSaver(BasePostgresSaver):
with conn.cursor(binary=True, row_factory=dict_row) as cur:
yield cur
def _get_channel_writes_history(
self, config: RunnableConfig, channel: str
) -> _ChannelWritesHistory:
"""Fast-path override of `BaseCheckpointSaver._get_channel_writes_history`.
Three indexed roundtrips (`checkpoints`, `checkpoint_writes`,
`checkpoint_blobs`) each filtered by `(thread_id, checkpoint_ns)` and
the per-table key. Plain SELECTs let the planner pick straight index
scans; rationale + benchmark in `notes/delta_channel_query_bench.md`.
"""
thread_id = config["configurable"]["thread_id"]
checkpoint_ns = config["configurable"].get("checkpoint_ns", "")
checkpoint_id = get_checkpoint_id(config)
if checkpoint_id is None:
# Caller didn't specify a target — resolve to the latest
# checkpoint on the thread. `get_tuple` without `checkpoint_id`
# returns the newest; its config carries the resolved id.
target = self.get_tuple(config)
if target is None:
return _ChannelWritesHistory(seed=DELTA_SENTINEL, writes=[])
checkpoint_id = target.config["configurable"]["checkpoint_id"]
with self._cursor() as cur:
cur.execute(SELECT_DELTA_PARENTS_SQL, (channel, thread_id, checkpoint_ns))
parents_rows = cur.fetchall()
cur.execute(SELECT_DELTA_WRITES_SQL, (thread_id, checkpoint_ns, channel))
writes_rows = cur.fetchall()
cur.execute(SELECT_DELTA_BLOBS_SQL, (thread_id, checkpoint_ns, channel))
blobs_rows = cur.fetchall()
return self._build_delta_channel_writes_history(
channel=channel,
target_id=checkpoint_id,
parents_rows=parents_rows,
writes_rows=writes_rows,
blobs_rows=blobs_rows,
)
def _load_checkpoint_tuple(self, value: DictRow) -> CheckpointTuple:
"""
Convert a database row into a CheckpointTuple object.
@@ -442,6 +485,7 @@ class PostgresSaver(BasePostgresSaver):
including its configuration, metadata, parent checkpoint (if any),
and pending writes.
"""
channel_values = self._load_blobs(value["channel_values"])
return CheckpointTuple(
{
"configurable": {
@@ -454,7 +498,7 @@ class PostgresSaver(BasePostgresSaver):
**value["checkpoint"],
"channel_values": {
**(value["checkpoint"].get("channel_values") or {}),
**self._load_blobs(value["channel_values"]),
**channel_values,
},
},
value["metadata"],
@@ -8,11 +8,13 @@ from typing import Any
from langchain_core.runnables import RunnableConfig
from langgraph.checkpoint.base import (
DELTA_SENTINEL,
WRITES_IDX_MAP,
ChannelVersions,
Checkpoint,
CheckpointMetadata,
CheckpointTuple,
_ChannelWritesHistory,
get_checkpoint_id,
get_serializable_checkpoint_metadata,
)
@@ -23,7 +25,12 @@ from psycopg.types.json import Jsonb
from psycopg_pool import AsyncConnectionPool
from langgraph.checkpoint.postgres import _ainternal
from langgraph.checkpoint.postgres.base import BasePostgresSaver
from langgraph.checkpoint.postgres.base import (
SELECT_DELTA_BLOBS_SQL,
SELECT_DELTA_PARENTS_SQL,
SELECT_DELTA_WRITES_SQL,
BasePostgresSaver,
)
from langgraph.checkpoint.postgres.shallow import AsyncShallowPostgresSaver
Conn = _ainternal.Conn # For backward compatibility
@@ -391,6 +398,45 @@ class AsyncPostgresSaver(BasePostgresSaver):
async with conn.cursor(binary=True, row_factory=dict_row) as cur:
yield cur
async def _aget_channel_writes_history(
self, config: RunnableConfig, channel: str
) -> _ChannelWritesHistory:
"""Fast-path override of `BaseCheckpointSaver._aget_channel_writes_history`.
Three indexed roundtrips (`checkpoints`, `checkpoint_writes`,
`checkpoint_blobs`); rows assembled by the shared pure helper on
`BasePostgresSaver`. Rationale + benchmark in
`notes/delta_channel_query_bench.md`.
"""
thread_id = config["configurable"]["thread_id"]
checkpoint_ns = config["configurable"].get("checkpoint_ns", "")
checkpoint_id = get_checkpoint_id(config)
if checkpoint_id is None:
target = await self.aget_tuple(config)
if target is None:
return _ChannelWritesHistory(seed=DELTA_SENTINEL, writes=[])
checkpoint_id = target.config["configurable"]["checkpoint_id"]
async with self._cursor() as cur:
await cur.execute(
SELECT_DELTA_PARENTS_SQL, (channel, thread_id, checkpoint_ns)
)
parents_rows = await cur.fetchall()
await cur.execute(
SELECT_DELTA_WRITES_SQL, (thread_id, checkpoint_ns, channel)
)
writes_rows = await cur.fetchall()
await cur.execute(
SELECT_DELTA_BLOBS_SQL, (thread_id, checkpoint_ns, channel)
)
blobs_rows = await cur.fetchall()
return self._build_delta_channel_writes_history(
channel=channel,
target_id=checkpoint_id,
parents_rows=parents_rows,
writes_rows=writes_rows,
blobs_rows=blobs_rows,
)
async def _load_checkpoint_tuple(self, value: DictRow) -> CheckpointTuple:
"""
Convert a database row into a CheckpointTuple object.
@@ -403,11 +449,18 @@ class AsyncPostgresSaver(BasePostgresSaver):
including its configuration, metadata, parent checkpoint (if any),
and pending writes.
"""
thread_id = value["thread_id"]
checkpoint_ns = value["checkpoint_ns"]
blob_values = value["channel_values"]
channel_values: dict[str, Any] = {}
if blob_values:
channel_values = self._load_blobs(blob_values)
return CheckpointTuple(
{
"configurable": {
"thread_id": value["thread_id"],
"checkpoint_ns": value["checkpoint_ns"],
"thread_id": thread_id,
"checkpoint_ns": checkpoint_ns,
"checkpoint_id": value["checkpoint_id"],
}
},
@@ -415,15 +468,15 @@ class AsyncPostgresSaver(BasePostgresSaver):
**value["checkpoint"],
"channel_values": {
**(value["checkpoint"].get("channel_values") or {}),
**self._load_blobs(value["channel_values"]),
**channel_values,
},
},
value["metadata"],
(
{
"configurable": {
"thread_id": value["thread_id"],
"checkpoint_ns": value["checkpoint_ns"],
"thread_id": thread_id,
"checkpoint_ns": checkpoint_ns,
"checkpoint_id": value["parent_checkpoint_id"],
}
}
@@ -8,9 +8,12 @@ from typing import Any, cast
from langchain_core.runnables import RunnableConfig
from langgraph.checkpoint.base import (
DELTA_SENTINEL,
WRITES_IDX_MAP,
BaseCheckpointSaver,
ChannelVersions,
PendingWrite,
_ChannelWritesHistory,
get_checkpoint_id,
)
from langgraph.checkpoint.serde.types import TASKS
@@ -152,6 +155,30 @@ INSERT_CHECKPOINT_WRITES_SQL = """
ON CONFLICT (thread_id, checkpoint_ns, checkpoint_id, task_id, idx) DO NOTHING
"""
# DeltaChannel reconstruction: three plain indexed SELECTs per channel.
# Bench (notes/delta_channel_query_bench.md) showed the prior recursive CTE
# carried a hidden O(ancestors x blobs_in_thread) join; plain SELECTs are
# 3x-100x faster in the realistic depth range and the Python walk is O(n).
SELECT_DELTA_PARENTS_SQL = """
SELECT checkpoint_id,
parent_checkpoint_id,
checkpoint -> 'channel_versions' ->> %s AS ver
FROM checkpoints
WHERE thread_id = %s AND checkpoint_ns = %s
"""
SELECT_DELTA_WRITES_SQL = """
SELECT checkpoint_id, type, blob, task_id, idx
FROM checkpoint_writes
WHERE thread_id = %s AND checkpoint_ns = %s AND channel = %s
"""
SELECT_DELTA_BLOBS_SQL = """
SELECT version, type, blob
FROM checkpoint_blobs
WHERE thread_id = %s AND checkpoint_ns = %s AND channel = %s
"""
class BasePostgresSaver(BaseCheckpointSaver[str]):
SELECT_SQL = SELECT_SQL
@@ -185,16 +212,91 @@ class BasePostgresSaver(BaseCheckpointSaver[str]):
)
def _load_blobs(
self, blob_values: list[tuple[bytes, bytes, bytes]]
self,
blob_values: Any,
) -> dict[str, Any]:
if not blob_values:
return {}
return {
k.decode(): self.serde.loads_typed((t.decode(), v))
for k, t, v in blob_values
if t.decode() != "empty"
result: dict[str, Any] = {}
for k, t, v in blob_values:
type_tag = t.decode()
if type_tag != "empty":
result[k.decode()] = self.serde.loads_typed((type_tag, v))
return result
def _build_delta_channel_writes_history(
self,
*,
channel: str,
target_id: str,
parents_rows: Sequence[Any],
writes_rows: Sequence[Any],
blobs_rows: Sequence[Any],
) -> _ChannelWritesHistory:
"""Reconstruct one delta channel's history from rows of the three SELECTs.
Pure data transform shared by sync (`PostgresSaver`) and async
(`AsyncPostgresSaver`); both paths run the queries themselves and
feed the rows here.
Walk is newest → oldest from the target's parent. A non-sentinel
blob in `checkpoint_blobs` (a pre-delta snapshot) terminates the
walk and is returned as the seed so replay starts from it.
Writes stored at `target_id` itself are pending writes for the next
step and are excluded — the walk begins at the target's parent.
"""
parent_of: dict[str, str | None] = {}
ver_of: dict[str, str | None] = {}
for r in parents_rows:
cid = r["checkpoint_id"]
parent_of[cid] = r["parent_checkpoint_id"]
ver_of[cid] = r["ver"]
ancestors: list[str] = []
cid = parent_of.get(target_id)
while cid is not None:
ancestors.append(cid)
cid = parent_of.get(cid)
if not ancestors:
return _ChannelWritesHistory(seed=DELTA_SENTINEL, writes=[])
ancestor_set = set(ancestors)
# Group writes by ancestor cid; sort within (task_id DESC, idx DESC)
# to match the prior CTE ordering — newest write first per ancestor.
writes_by_cid: dict[str, list[tuple[str, bytes, str, int]]] = {}
for r in writes_rows:
cid = r["checkpoint_id"]
if cid not in ancestor_set:
continue
writes_by_cid.setdefault(cid, []).append(
(r["type"], r["blob"], r["task_id"], r["idx"])
)
for ws in writes_by_cid.values():
ws.sort(key=lambda w: (w[2], w[3]), reverse=True)
blob_by_ver: dict[str, tuple[str, bytes]] = {
r["version"]: (r["type"], r["blob"]) for r in blobs_rows
}
collected: list[PendingWrite] = [] # newest first; reversed at the end
for cid in ancestors:
# Pre-delta blob terminator: subsumes any writes at this ancestor.
ver = ver_of.get(cid)
if ver is not None:
seed_blob = blob_by_ver.get(ver)
if seed_blob is not None and seed_blob[0] != "empty":
blob_value = self.serde.loads_typed(seed_blob)
if blob_value is not DELTA_SENTINEL:
collected.reverse()
return _ChannelWritesHistory(seed=blob_value, writes=collected)
for type_tag, write_blob, task_id, _idx in writes_by_cid.get(cid, []):
val = self.serde.loads_typed((type_tag, write_blob))
collected.append((task_id, channel, val))
collected.reverse() # oldest → newest
return _ChannelWritesHistory(seed=DELTA_SENTINEL, writes=collected)
def _dump_blobs(
self,
thread_id: str,
+46 -2
View File
@@ -361,9 +361,9 @@ async def test_get_checkpoint_no_channel_values(
load_checkpoint_tuple = saver._load_checkpoint_tuple
def patched_load_checkpoint_tuple(value):
async def patched_load_checkpoint_tuple(value):
value["checkpoint"].pop("channel_values", None)
return load_checkpoint_tuple(value)
return await load_checkpoint_tuple(value)
monkeypatch.setattr(
saver, "_load_checkpoint_tuple", patched_load_checkpoint_tuple
@@ -371,3 +371,47 @@ async def test_get_checkpoint_no_channel_values(
checkpoint = await saver.aget_tuple(config)
assert checkpoint.checkpoint["channel_values"] == {}
@pytest.mark.parametrize("saver_name", ["base", "pool", "pipe"])
async def test_delta_channel_chain_reconstruction(saver_name: str) -> None:
"""AsyncPostgresSaver reconstructs DeltaChannel chain via point-lookup traversal."""
pytest.importorskip(
"langgraph.channels._delta", reason="langgraph core not installed"
)
from typing import Annotated
from langchain_core.messages import AIMessage, HumanMessage
from langgraph.channels._delta import DeltaChannel
from langgraph.graph import START, StateGraph
from langgraph.graph.message import add_messages
from typing_extensions import TypedDict
class State(TypedDict):
messages: Annotated[list, DeltaChannel(add_messages)]
def respond(state: State) -> dict:
n = len(state["messages"])
return {"messages": [AIMessage(content=f"reply-{n}", id=f"ai-{n}")]}
builder = StateGraph(State)
builder.add_node("respond", respond)
builder.add_edge(START, "respond")
async with _saver(saver_name) as saver:
graph = builder.compile(checkpointer=saver)
config = {"configurable": {"thread_id": "diff-channel-test-1"}}
await graph.ainvoke({"messages": [HumanMessage(content="hi", id="h1")]}, config)
await graph.ainvoke(
{"messages": [HumanMessage(content="there", id="h2")]}, config
)
state = await graph.aget_state(config)
msgs = state.values["messages"]
assert len(msgs) == 4, f"expected 4, got {len(msgs)}: {msgs}"
assert msgs[0].content == "hi"
assert msgs[1].content == "reply-1"
assert msgs[2].content == "there"
assert msgs[3].content == "reply-3"
+123 -4
View File
@@ -259,7 +259,7 @@ wheels = [
[[package]]
name = "langgraph-checkpoint"
version = "4.0.1"
version = "4.0.2"
source = { editable = "../checkpoint" }
dependencies = [
{ name = "langchain-core" },
@@ -382,7 +382,7 @@ test = [
[[package]]
name = "langsmith"
version = "0.6.4"
version = "0.7.31"
source = { registry = "https://pypi.org/simple" }
dependencies = [
{ name = "httpx" },
@@ -392,11 +392,12 @@ dependencies = [
{ name = "requests" },
{ name = "requests-toolbelt" },
{ name = "uuid-utils" },
{ name = "xxhash" },
{ name = "zstandard" },
]
sdist = { url = "https://files.pythonhosted.org/packages/e7/85/9c7933052a997da1b85bc5c774f3865e9b1da1c8d71541ea133178b13229/langsmith-0.6.4.tar.gz", hash = "sha256:36f7223a01c218079fbb17da5e536ebbaf5c1468c028abe070aa3ae59bc99ec8", size = 919964, upload-time = "2026-01-15T20:02:28.873Z" }
sdist = { url = "https://files.pythonhosted.org/packages/e6/11/696019490992db5c87774dc20515529ef42a01e1d770fb754ed6d9b12fb0/langsmith-0.7.31.tar.gz", hash = "sha256:331ee4f7c26bb5be4022b9859b7d7b122cbf8c9d01d9f530114c1914b0349ffb", size = 1178480, upload-time = "2026-04-14T17:55:41.242Z" }
wheels = [
{ url = "https://files.pythonhosted.org/packages/66/0f/09a6637a7ba777eb307b7c80852d9ee26438e2bdafbad6fcc849ff9d9192/langsmith-0.6.4-py3-none-any.whl", hash = "sha256:ac4835860160be371042c7adbba3cb267bcf8d96a5ea976c33a8a4acad6c5486", size = 283503, upload-time = "2026-01-15T20:02:26.662Z" },
{ url = "https://files.pythonhosted.org/packages/1d/a1/a013cf458c301cda86a213dd153ce0a01c93f1ab5833f951e6a44c9763ce/langsmith-0.7.31-py3-none-any.whl", hash = "sha256:0291d49203f6e80dda011af1afda61eb0595a4d697adb684590a8805e1d61fb6", size = 373276, upload-time = "2026-04-14T17:55:39.677Z" },
]
[[package]]
@@ -1284,6 +1285,124 @@ wheels = [
{ url = "https://files.pythonhosted.org/packages/33/e8/e40370e6d74ddba47f002a32919d91310d6074130fe4e17dabcafc15cbf1/watchdog-6.0.0-py3-none-win_ia64.whl", hash = "sha256:a1914259fa9e1454315171103c6a30961236f508b9b623eae470268bbcc6a22f", size = 79067, upload-time = "2024-11-01T14:07:11.845Z" },
]
[[package]]
name = "xxhash"
version = "3.6.0"
source = { registry = "https://pypi.org/simple" }
sdist = { url = "https://files.pythonhosted.org/packages/02/84/30869e01909fb37a6cc7e18688ee8bf1e42d57e7e0777636bd47524c43c7/xxhash-3.6.0.tar.gz", hash = "sha256:f0162a78b13a0d7617b2845b90c763339d1f1d82bb04a4b07f4ab535cc5e05d6", size = 85160, upload-time = "2025-10-02T14:37:08.097Z" }
wheels = [
{ url = "https://files.pythonhosted.org/packages/34/ee/f9f1d656ad168681bb0f6b092372c1e533c4416b8069b1896a175c46e484/xxhash-3.6.0-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:87ff03d7e35c61435976554477a7f4cd1704c3596a89a8300d5ce7fc83874a71", size = 32845, upload-time = "2025-10-02T14:33:51.573Z" },
{ url = "https://files.pythonhosted.org/packages/a3/b1/93508d9460b292c74a09b83d16750c52a0ead89c51eea9951cb97a60d959/xxhash-3.6.0-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:f572dfd3d0e2eb1a57511831cf6341242f5a9f8298a45862d085f5b93394a27d", size = 30807, upload-time = "2025-10-02T14:33:52.964Z" },
{ url = "https://files.pythonhosted.org/packages/07/55/28c93a3662f2d200c70704efe74aab9640e824f8ce330d8d3943bf7c9b3c/xxhash-3.6.0-cp310-cp310-manylinux1_i686.manylinux_2_28_i686.manylinux_2_5_i686.whl", hash = "sha256:89952ea539566b9fed2bbd94e589672794b4286f342254fad28b149f9615fef8", size = 193786, upload-time = "2025-10-02T14:33:54.272Z" },
{ url = "https://files.pythonhosted.org/packages/c1/96/fec0be9bb4b8f5d9c57d76380a366f31a1781fb802f76fc7cda6c84893c7/xxhash-3.6.0-cp310-cp310-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:48e6f2ffb07a50b52465a1032c3cf1f4a5683f944acaca8a134a2f23674c2058", size = 212830, upload-time = "2025-10-02T14:33:55.706Z" },
{ url = "https://files.pythonhosted.org/packages/c4/a0/c706845ba77b9611f81fd2e93fad9859346b026e8445e76f8c6fd057cc6d/xxhash-3.6.0-cp310-cp310-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:b5b848ad6c16d308c3ac7ad4ba6bede80ed5df2ba8ed382f8932df63158dd4b2", size = 211606, upload-time = "2025-10-02T14:33:57.133Z" },
{ url = "https://files.pythonhosted.org/packages/67/1e/164126a2999e5045f04a69257eea946c0dc3e86541b400d4385d646b53d7/xxhash-3.6.0-cp310-cp310-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:a034590a727b44dd8ac5914236a7b8504144447a9682586c3327e935f33ec8cc", size = 444872, upload-time = "2025-10-02T14:33:58.446Z" },
{ url = "https://files.pythonhosted.org/packages/2d/4b/55ab404c56cd70a2cf5ecfe484838865d0fea5627365c6c8ca156bd09c8f/xxhash-3.6.0-cp310-cp310-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:8a8f1972e75ebdd161d7896743122834fe87378160c20e97f8b09166213bf8cc", size = 193217, upload-time = "2025-10-02T14:33:59.724Z" },
{ url = "https://files.pythonhosted.org/packages/45/e6/52abf06bac316db33aa269091ae7311bd53cfc6f4b120ae77bac1b348091/xxhash-3.6.0-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:ee34327b187f002a596d7b167ebc59a1b729e963ce645964bbc050d2f1b73d07", size = 210139, upload-time = "2025-10-02T14:34:02.041Z" },
{ url = "https://files.pythonhosted.org/packages/34/37/db94d490b8691236d356bc249c08819cbcef9273a1a30acf1254ff9ce157/xxhash-3.6.0-cp310-cp310-musllinux_1_2_i686.whl", hash = "sha256:339f518c3c7a850dd033ab416ea25a692759dc7478a71131fe8869010d2b75e4", size = 197669, upload-time = "2025-10-02T14:34:03.664Z" },
{ url = "https://files.pythonhosted.org/packages/b7/36/c4f219ef4a17a4f7a64ed3569bc2b5a9c8311abdb22249ac96093625b1a4/xxhash-3.6.0-cp310-cp310-musllinux_1_2_ppc64le.whl", hash = "sha256:bf48889c9630542d4709192578aebbd836177c9f7a4a2778a7d6340107c65f06", size = 210018, upload-time = "2025-10-02T14:34:05.325Z" },
{ url = "https://files.pythonhosted.org/packages/fd/06/bfac889a374fc2fc439a69223d1750eed2e18a7db8514737ab630534fa08/xxhash-3.6.0-cp310-cp310-musllinux_1_2_s390x.whl", hash = "sha256:5576b002a56207f640636056b4160a378fe36a58db73ae5c27a7ec8db35f71d4", size = 413058, upload-time = "2025-10-02T14:34:06.925Z" },
{ url = "https://files.pythonhosted.org/packages/c9/d1/555d8447e0dd32ad0930a249a522bb2e289f0d08b6b16204cfa42c1f5a0c/xxhash-3.6.0-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:af1f3278bd02814d6dedc5dec397993b549d6f16c19379721e5a1d31e132c49b", size = 190628, upload-time = "2025-10-02T14:34:08.669Z" },
{ url = "https://files.pythonhosted.org/packages/d1/15/8751330b5186cedc4ed4b597989882ea05e0408b53fa47bcb46a6125bfc6/xxhash-3.6.0-cp310-cp310-win32.whl", hash = "sha256:aed058764db109dc9052720da65fafe84873b05eb8b07e5e653597951af57c3b", size = 30577, upload-time = "2025-10-02T14:34:10.234Z" },
{ url = "https://files.pythonhosted.org/packages/bb/cc/53f87e8b5871a6eb2ff7e89c48c66093bda2be52315a8161ddc54ea550c4/xxhash-3.6.0-cp310-cp310-win_amd64.whl", hash = "sha256:e82da5670f2d0d98950317f82a0e4a0197150ff19a6df2ba40399c2a3b9ae5fb", size = 31487, upload-time = "2025-10-02T14:34:11.618Z" },
{ url = "https://files.pythonhosted.org/packages/9f/00/60f9ea3bb697667a14314d7269956f58bf56bb73864f8f8d52a3c2535e9a/xxhash-3.6.0-cp310-cp310-win_arm64.whl", hash = "sha256:4a082ffff8c6ac07707fb6b671caf7c6e020c75226c561830b73d862060f281d", size = 27863, upload-time = "2025-10-02T14:34:12.619Z" },
{ url = "https://files.pythonhosted.org/packages/17/d4/cc2f0400e9154df4b9964249da78ebd72f318e35ccc425e9f403c392f22a/xxhash-3.6.0-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:b47bbd8cf2d72797f3c2772eaaac0ded3d3af26481a26d7d7d41dc2d3c46b04a", size = 32844, upload-time = "2025-10-02T14:34:14.037Z" },
{ url = "https://files.pythonhosted.org/packages/5e/ec/1cc11cd13e26ea8bc3cb4af4eaadd8d46d5014aebb67be3f71fb0b68802a/xxhash-3.6.0-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:2b6821e94346f96db75abaa6e255706fb06ebd530899ed76d32cd99f20dc52fa", size = 30809, upload-time = "2025-10-02T14:34:15.484Z" },
{ url = "https://files.pythonhosted.org/packages/04/5f/19fe357ea348d98ca22f456f75a30ac0916b51c753e1f8b2e0e6fb884cce/xxhash-3.6.0-cp311-cp311-manylinux1_i686.manylinux_2_28_i686.manylinux_2_5_i686.whl", hash = "sha256:d0a9751f71a1a65ce3584e9cae4467651c7e70c9d31017fa57574583a4540248", size = 194665, upload-time = "2025-10-02T14:34:16.541Z" },
{ url = "https://files.pythonhosted.org/packages/90/3b/d1f1a8f5442a5fd8beedae110c5af7604dc37349a8e16519c13c19a9a2de/xxhash-3.6.0-cp311-cp311-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:8b29ee68625ab37b04c0b40c3fafdf24d2f75ccd778333cfb698f65f6c463f62", size = 213550, upload-time = "2025-10-02T14:34:17.878Z" },
{ url = "https://files.pythonhosted.org/packages/c4/ef/3a9b05eb527457d5db13a135a2ae1a26c80fecd624d20f3e8dcc4cb170f3/xxhash-3.6.0-cp311-cp311-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:6812c25fe0d6c36a46ccb002f40f27ac903bf18af9f6dd8f9669cb4d176ab18f", size = 212384, upload-time = "2025-10-02T14:34:19.182Z" },
{ url = "https://files.pythonhosted.org/packages/0f/18/ccc194ee698c6c623acbf0f8c2969811a8a4b6185af5e824cd27b9e4fd3e/xxhash-3.6.0-cp311-cp311-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:4ccbff013972390b51a18ef1255ef5ac125c92dc9143b2d1909f59abc765540e", size = 445749, upload-time = "2025-10-02T14:34:20.659Z" },
{ url = "https://files.pythonhosted.org/packages/a5/86/cf2c0321dc3940a7aa73076f4fd677a0fb3e405cb297ead7d864fd90847e/xxhash-3.6.0-cp311-cp311-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:297b7fbf86c82c550e12e8fb71968b3f033d27b874276ba3624ea868c11165a8", size = 193880, upload-time = "2025-10-02T14:34:22.431Z" },
{ url = "https://files.pythonhosted.org/packages/82/fb/96213c8560e6f948a1ecc9a7613f8032b19ee45f747f4fca4eb31bb6d6ed/xxhash-3.6.0-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:dea26ae1eb293db089798d3973a5fc928a18fdd97cc8801226fae705b02b14b0", size = 210912, upload-time = "2025-10-02T14:34:23.937Z" },
{ url = "https://files.pythonhosted.org/packages/40/aa/4395e669b0606a096d6788f40dbdf2b819d6773aa290c19e6e83cbfc312f/xxhash-3.6.0-cp311-cp311-musllinux_1_2_i686.whl", hash = "sha256:7a0b169aafb98f4284f73635a8e93f0735f9cbde17bd5ec332480484241aaa77", size = 198654, upload-time = "2025-10-02T14:34:25.644Z" },
{ url = "https://files.pythonhosted.org/packages/67/74/b044fcd6b3d89e9b1b665924d85d3f400636c23590226feb1eb09e1176ce/xxhash-3.6.0-cp311-cp311-musllinux_1_2_ppc64le.whl", hash = "sha256:08d45aef063a4531b785cd72de4887766d01dc8f362a515693df349fdb825e0c", size = 210867, upload-time = "2025-10-02T14:34:27.203Z" },
{ url = "https://files.pythonhosted.org/packages/bc/fd/3ce73bf753b08cb19daee1eb14aa0d7fe331f8da9c02dd95316ddfe5275e/xxhash-3.6.0-cp311-cp311-musllinux_1_2_s390x.whl", hash = "sha256:929142361a48ee07f09121fe9e96a84950e8d4df3bb298ca5d88061969f34d7b", size = 414012, upload-time = "2025-10-02T14:34:28.409Z" },
{ url = "https://files.pythonhosted.org/packages/ba/b3/5a4241309217c5c876f156b10778f3ab3af7ba7e3259e6d5f5c7d0129eb2/xxhash-3.6.0-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:51312c768403d8540487dbbfb557454cfc55589bbde6424456951f7fcd4facb3", size = 191409, upload-time = "2025-10-02T14:34:29.696Z" },
{ url = "https://files.pythonhosted.org/packages/c0/01/99bfbc15fb9abb9a72b088c1d95219fc4782b7d01fc835bd5744d66dd0b8/xxhash-3.6.0-cp311-cp311-win32.whl", hash = "sha256:d1927a69feddc24c987b337ce81ac15c4720955b667fe9b588e02254b80446fd", size = 30574, upload-time = "2025-10-02T14:34:31.028Z" },
{ url = "https://files.pythonhosted.org/packages/65/79/9d24d7f53819fe301b231044ea362ce64e86c74f6e8c8e51320de248b3e5/xxhash-3.6.0-cp311-cp311-win_amd64.whl", hash = "sha256:26734cdc2d4ffe449b41d186bbeac416f704a482ed835d375a5c0cb02bc63fef", size = 31481, upload-time = "2025-10-02T14:34:32.062Z" },
{ url = "https://files.pythonhosted.org/packages/30/4e/15cd0e3e8772071344eab2961ce83f6e485111fed8beb491a3f1ce100270/xxhash-3.6.0-cp311-cp311-win_arm64.whl", hash = "sha256:d72f67ef8bf36e05f5b6c65e8524f265bd61071471cd4cf1d36743ebeeeb06b7", size = 27861, upload-time = "2025-10-02T14:34:33.555Z" },
{ url = "https://files.pythonhosted.org/packages/9a/07/d9412f3d7d462347e4511181dea65e47e0d0e16e26fbee2ea86a2aefb657/xxhash-3.6.0-cp312-cp312-macosx_10_13_x86_64.whl", hash = "sha256:01362c4331775398e7bb34e3ab403bc9ee9f7c497bc7dee6272114055277dd3c", size = 32744, upload-time = "2025-10-02T14:34:34.622Z" },
{ url = "https://files.pythonhosted.org/packages/79/35/0429ee11d035fc33abe32dca1b2b69e8c18d236547b9a9b72c1929189b9a/xxhash-3.6.0-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:b7b2df81a23f8cb99656378e72501b2cb41b1827c0f5a86f87d6b06b69f9f204", size = 30816, upload-time = "2025-10-02T14:34:36.043Z" },
{ url = "https://files.pythonhosted.org/packages/b7/f2/57eb99aa0f7d98624c0932c5b9a170e1806406cdbcdb510546634a1359e0/xxhash-3.6.0-cp312-cp312-manylinux1_i686.manylinux_2_28_i686.manylinux_2_5_i686.whl", hash = "sha256:dc94790144e66b14f67b10ac8ed75b39ca47536bf8800eb7c24b50271ea0c490", size = 194035, upload-time = "2025-10-02T14:34:37.354Z" },
{ url = "https://files.pythonhosted.org/packages/4c/ed/6224ba353690d73af7a3f1c7cdb1fc1b002e38f783cb991ae338e1eb3d79/xxhash-3.6.0-cp312-cp312-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:93f107c673bccf0d592cdba077dedaf52fe7f42dcd7676eba1f6d6f0c3efffd2", size = 212914, upload-time = "2025-10-02T14:34:38.6Z" },
{ url = "https://files.pythonhosted.org/packages/38/86/fb6b6130d8dd6b8942cc17ab4d90e223653a89aa32ad2776f8af7064ed13/xxhash-3.6.0-cp312-cp312-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:2aa5ee3444c25b69813663c9f8067dcfaa2e126dc55e8dddf40f4d1c25d7effa", size = 212163, upload-time = "2025-10-02T14:34:39.872Z" },
{ url = "https://files.pythonhosted.org/packages/ee/dc/e84875682b0593e884ad73b2d40767b5790d417bde603cceb6878901d647/xxhash-3.6.0-cp312-cp312-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:f7f99123f0e1194fa59cc69ad46dbae2e07becec5df50a0509a808f90a0f03f0", size = 445411, upload-time = "2025-10-02T14:34:41.569Z" },
{ url = "https://files.pythonhosted.org/packages/11/4f/426f91b96701ec2f37bb2b8cec664eff4f658a11f3fa9d94f0a887ea6d2b/xxhash-3.6.0-cp312-cp312-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:49e03e6fe2cac4a1bc64952dd250cf0dbc5ef4ebb7b8d96bce82e2de163c82a2", size = 193883, upload-time = "2025-10-02T14:34:43.249Z" },
{ url = "https://files.pythonhosted.org/packages/53/5a/ddbb83eee8e28b778eacfc5a85c969673e4023cdeedcfcef61f36731610b/xxhash-3.6.0-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:bd17fede52a17a4f9a7bc4472a5867cb0b160deeb431795c0e4abe158bc784e9", size = 210392, upload-time = "2025-10-02T14:34:45.042Z" },
{ url = "https://files.pythonhosted.org/packages/1e/c2/ff69efd07c8c074ccdf0a4f36fcdd3d27363665bcdf4ba399abebe643465/xxhash-3.6.0-cp312-cp312-musllinux_1_2_i686.whl", hash = "sha256:6fb5f5476bef678f69db04f2bd1efbed3030d2aba305b0fc1773645f187d6a4e", size = 197898, upload-time = "2025-10-02T14:34:46.302Z" },
{ url = "https://files.pythonhosted.org/packages/58/ca/faa05ac19b3b622c7c9317ac3e23954187516298a091eb02c976d0d3dd45/xxhash-3.6.0-cp312-cp312-musllinux_1_2_ppc64le.whl", hash = "sha256:843b52f6d88071f87eba1631b684fcb4b2068cd2180a0224122fe4ef011a9374", size = 210655, upload-time = "2025-10-02T14:34:47.571Z" },
{ url = "https://files.pythonhosted.org/packages/d4/7a/06aa7482345480cc0cb597f5c875b11a82c3953f534394f620b0be2f700c/xxhash-3.6.0-cp312-cp312-musllinux_1_2_s390x.whl", hash = "sha256:7d14a6cfaf03b1b6f5f9790f76880601ccc7896aff7ab9cd8978a939c1eb7e0d", size = 414001, upload-time = "2025-10-02T14:34:49.273Z" },
{ url = "https://files.pythonhosted.org/packages/23/07/63ffb386cd47029aa2916b3d2f454e6cc5b9f5c5ada3790377d5430084e7/xxhash-3.6.0-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:418daf3db71e1413cfe211c2f9a528456936645c17f46b5204705581a45390ae", size = 191431, upload-time = "2025-10-02T14:34:50.798Z" },
{ url = "https://files.pythonhosted.org/packages/0f/93/14fde614cadb4ddf5e7cebf8918b7e8fac5ae7861c1875964f17e678205c/xxhash-3.6.0-cp312-cp312-win32.whl", hash = "sha256:50fc255f39428a27299c20e280d6193d8b63b8ef8028995323bf834a026b4fbb", size = 30617, upload-time = "2025-10-02T14:34:51.954Z" },
{ url = "https://files.pythonhosted.org/packages/13/5d/0d125536cbe7565a83d06e43783389ecae0c0f2ed037b48ede185de477c0/xxhash-3.6.0-cp312-cp312-win_amd64.whl", hash = "sha256:c0f2ab8c715630565ab8991b536ecded9416d615538be8ecddce43ccf26cbc7c", size = 31534, upload-time = "2025-10-02T14:34:53.276Z" },
{ url = "https://files.pythonhosted.org/packages/54/85/6ec269b0952ec7e36ba019125982cf11d91256a778c7c3f98a4c5043d283/xxhash-3.6.0-cp312-cp312-win_arm64.whl", hash = "sha256:eae5c13f3bc455a3bbb68bdc513912dc7356de7e2280363ea235f71f54064829", size = 27876, upload-time = "2025-10-02T14:34:54.371Z" },
{ url = "https://files.pythonhosted.org/packages/33/76/35d05267ac82f53ae9b0e554da7c5e281ee61f3cad44c743f0fcd354f211/xxhash-3.6.0-cp313-cp313-macosx_10_13_x86_64.whl", hash = "sha256:599e64ba7f67472481ceb6ee80fa3bd828fd61ba59fb11475572cc5ee52b89ec", size = 32738, upload-time = "2025-10-02T14:34:55.839Z" },
{ url = "https://files.pythonhosted.org/packages/31/a8/3fbce1cd96534a95e35d5120637bf29b0d7f5d8fa2f6374e31b4156dd419/xxhash-3.6.0-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:7d8b8aaa30fca4f16f0c84a5c8d7ddee0e25250ec2796c973775373257dde8f1", size = 30821, upload-time = "2025-10-02T14:34:57.219Z" },
{ url = "https://files.pythonhosted.org/packages/0c/ea/d387530ca7ecfa183cb358027f1833297c6ac6098223fd14f9782cd0015c/xxhash-3.6.0-cp313-cp313-manylinux1_i686.manylinux_2_28_i686.manylinux_2_5_i686.whl", hash = "sha256:d597acf8506d6e7101a4a44a5e428977a51c0fadbbfd3c39650cca9253f6e5a6", size = 194127, upload-time = "2025-10-02T14:34:59.21Z" },
{ url = "https://files.pythonhosted.org/packages/ba/0c/71435dcb99874b09a43b8d7c54071e600a7481e42b3e3ce1eb5226a5711a/xxhash-3.6.0-cp313-cp313-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:858dc935963a33bc33490128edc1c12b0c14d9c7ebaa4e387a7869ecc4f3e263", size = 212975, upload-time = "2025-10-02T14:35:00.816Z" },
{ url = "https://files.pythonhosted.org/packages/84/7a/c2b3d071e4bb4a90b7057228a99b10d51744878f4a8a6dd643c8bd897620/xxhash-3.6.0-cp313-cp313-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:ba284920194615cb8edf73bf52236ce2e1664ccd4a38fdb543506413529cc546", size = 212241, upload-time = "2025-10-02T14:35:02.207Z" },
{ url = "https://files.pythonhosted.org/packages/81/5f/640b6eac0128e215f177df99eadcd0f1b7c42c274ab6a394a05059694c5a/xxhash-3.6.0-cp313-cp313-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:4b54219177f6c6674d5378bd862c6aedf64725f70dd29c472eaae154df1a2e89", size = 445471, upload-time = "2025-10-02T14:35:03.61Z" },
{ url = "https://files.pythonhosted.org/packages/5e/1e/3c3d3ef071b051cc3abbe3721ffb8365033a172613c04af2da89d5548a87/xxhash-3.6.0-cp313-cp313-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:42c36dd7dbad2f5238950c377fcbf6811b1cdb1c444fab447960030cea60504d", size = 193936, upload-time = "2025-10-02T14:35:05.013Z" },
{ url = "https://files.pythonhosted.org/packages/2c/bd/4a5f68381939219abfe1c22a9e3a5854a4f6f6f3c4983a87d255f21f2e5d/xxhash-3.6.0-cp313-cp313-musllinux_1_2_aarch64.whl", hash = "sha256:f22927652cba98c44639ffdc7aaf35828dccf679b10b31c4ad72a5b530a18eb7", size = 210440, upload-time = "2025-10-02T14:35:06.239Z" },
{ url = "https://files.pythonhosted.org/packages/eb/37/b80fe3d5cfb9faff01a02121a0f4d565eb7237e9e5fc66e73017e74dcd36/xxhash-3.6.0-cp313-cp313-musllinux_1_2_i686.whl", hash = "sha256:b45fad44d9c5c119e9c6fbf2e1c656a46dc68e280275007bbfd3d572b21426db", size = 197990, upload-time = "2025-10-02T14:35:07.735Z" },
{ url = "https://files.pythonhosted.org/packages/d7/fd/2c0a00c97b9e18f72e1f240ad4e8f8a90fd9d408289ba9c7c495ed7dc05c/xxhash-3.6.0-cp313-cp313-musllinux_1_2_ppc64le.whl", hash = "sha256:6f2580ffab1a8b68ef2b901cde7e55fa8da5e4be0977c68f78fc80f3c143de42", size = 210689, upload-time = "2025-10-02T14:35:09.438Z" },
{ url = "https://files.pythonhosted.org/packages/93/86/5dd8076a926b9a95db3206aba20d89a7fc14dd5aac16e5c4de4b56033140/xxhash-3.6.0-cp313-cp313-musllinux_1_2_s390x.whl", hash = "sha256:40c391dd3cd041ebc3ffe6f2c862f402e306eb571422e0aa918d8070ba31da11", size = 414068, upload-time = "2025-10-02T14:35:11.162Z" },
{ url = "https://files.pythonhosted.org/packages/af/3c/0bb129170ee8f3650f08e993baee550a09593462a5cddd8e44d0011102b1/xxhash-3.6.0-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:f205badabde7aafd1a31e8ca2a3e5a763107a71c397c4481d6a804eb5063d8bd", size = 191495, upload-time = "2025-10-02T14:35:12.971Z" },
{ url = "https://files.pythonhosted.org/packages/e9/3a/6797e0114c21d1725e2577508e24006fd7ff1d8c0c502d3b52e45c1771d8/xxhash-3.6.0-cp313-cp313-win32.whl", hash = "sha256:2577b276e060b73b73a53042ea5bd5203d3e6347ce0d09f98500f418a9fcf799", size = 30620, upload-time = "2025-10-02T14:35:14.129Z" },
{ url = "https://files.pythonhosted.org/packages/86/15/9bc32671e9a38b413a76d24722a2bf8784a132c043063a8f5152d390b0f9/xxhash-3.6.0-cp313-cp313-win_amd64.whl", hash = "sha256:757320d45d2fbcce8f30c42a6b2f47862967aea7bf458b9625b4bbe7ee390392", size = 31542, upload-time = "2025-10-02T14:35:15.21Z" },
{ url = "https://files.pythonhosted.org/packages/39/c5/cc01e4f6188656e56112d6a8e0dfe298a16934b8c47a247236549a3f7695/xxhash-3.6.0-cp313-cp313-win_arm64.whl", hash = "sha256:457b8f85dec5825eed7b69c11ae86834a018b8e3df5e77783c999663da2f96d6", size = 27880, upload-time = "2025-10-02T14:35:16.315Z" },
{ url = "https://files.pythonhosted.org/packages/f3/30/25e5321c8732759e930c555176d37e24ab84365482d257c3b16362235212/xxhash-3.6.0-cp313-cp313t-macosx_10_13_x86_64.whl", hash = "sha256:a42e633d75cdad6d625434e3468126c73f13f7584545a9cf34e883aa1710e702", size = 32956, upload-time = "2025-10-02T14:35:17.413Z" },
{ url = "https://files.pythonhosted.org/packages/9f/3c/0573299560d7d9f8ab1838f1efc021a280b5ae5ae2e849034ef3dee18810/xxhash-3.6.0-cp313-cp313t-macosx_11_0_arm64.whl", hash = "sha256:568a6d743219e717b07b4e03b0a828ce593833e498c3b64752e0f5df6bfe84db", size = 31072, upload-time = "2025-10-02T14:35:18.844Z" },
{ url = "https://files.pythonhosted.org/packages/7a/1c/52d83a06e417cd9d4137722693424885cc9878249beb3a7c829e74bf7ce9/xxhash-3.6.0-cp313-cp313t-manylinux1_i686.manylinux_2_28_i686.manylinux_2_5_i686.whl", hash = "sha256:bec91b562d8012dae276af8025a55811b875baace6af510412a5e58e3121bc54", size = 196409, upload-time = "2025-10-02T14:35:20.31Z" },
{ url = "https://files.pythonhosted.org/packages/e3/8e/c6d158d12a79bbd0b878f8355432075fc82759e356ab5a111463422a239b/xxhash-3.6.0-cp313-cp313t-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:78e7f2f4c521c30ad5e786fdd6bae89d47a32672a80195467b5de0480aa97b1f", size = 215736, upload-time = "2025-10-02T14:35:21.616Z" },
{ url = "https://files.pythonhosted.org/packages/bc/68/c4c80614716345d55071a396cf03d06e34b5f4917a467faf43083c995155/xxhash-3.6.0-cp313-cp313t-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:3ed0df1b11a79856df5ffcab572cbd6b9627034c1c748c5566fa79df9048a7c5", size = 214833, upload-time = "2025-10-02T14:35:23.32Z" },
{ url = "https://files.pythonhosted.org/packages/7e/e9/ae27c8ffec8b953efa84c7c4a6c6802c263d587b9fc0d6e7cea64e08c3af/xxhash-3.6.0-cp313-cp313t-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:0e4edbfc7d420925b0dd5e792478ed393d6e75ff8fc219a6546fb446b6a417b1", size = 448348, upload-time = "2025-10-02T14:35:25.111Z" },
{ url = "https://files.pythonhosted.org/packages/d7/6b/33e21afb1b5b3f46b74b6bd1913639066af218d704cc0941404ca717fc57/xxhash-3.6.0-cp313-cp313t-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:fba27a198363a7ef87f8c0f6b171ec36b674fe9053742c58dd7e3201c1ab30ee", size = 196070, upload-time = "2025-10-02T14:35:26.586Z" },
{ url = "https://files.pythonhosted.org/packages/96/b6/fcabd337bc5fa624e7203aa0fa7d0c49eed22f72e93229431752bddc83d9/xxhash-3.6.0-cp313-cp313t-musllinux_1_2_aarch64.whl", hash = "sha256:794fe9145fe60191c6532fa95063765529770edcdd67b3d537793e8004cabbfd", size = 212907, upload-time = "2025-10-02T14:35:28.087Z" },
{ url = "https://files.pythonhosted.org/packages/4b/d3/9ee6160e644d660fcf176c5825e61411c7f62648728f69c79ba237250143/xxhash-3.6.0-cp313-cp313t-musllinux_1_2_i686.whl", hash = "sha256:6105ef7e62b5ac73a837778efc331a591d8442f8ef5c7e102376506cb4ae2729", size = 200839, upload-time = "2025-10-02T14:35:29.857Z" },
{ url = "https://files.pythonhosted.org/packages/0d/98/e8de5baa5109394baf5118f5e72ab21a86387c4f89b0e77ef3e2f6b0327b/xxhash-3.6.0-cp313-cp313t-musllinux_1_2_ppc64le.whl", hash = "sha256:f01375c0e55395b814a679b3eea205db7919ac2af213f4a6682e01220e5fe292", size = 213304, upload-time = "2025-10-02T14:35:31.222Z" },
{ url = "https://files.pythonhosted.org/packages/7b/1d/71056535dec5c3177eeb53e38e3d367dd1d16e024e63b1cee208d572a033/xxhash-3.6.0-cp313-cp313t-musllinux_1_2_s390x.whl", hash = "sha256:d706dca2d24d834a4661619dcacf51a75c16d65985718d6a7d73c1eeeb903ddf", size = 416930, upload-time = "2025-10-02T14:35:32.517Z" },
{ url = "https://files.pythonhosted.org/packages/dc/6c/5cbde9de2cd967c322e651c65c543700b19e7ae3e0aae8ece3469bf9683d/xxhash-3.6.0-cp313-cp313t-musllinux_1_2_x86_64.whl", hash = "sha256:5f059d9faeacd49c0215d66f4056e1326c80503f51a1532ca336a385edadd033", size = 193787, upload-time = "2025-10-02T14:35:33.827Z" },
{ url = "https://files.pythonhosted.org/packages/19/fa/0172e350361d61febcea941b0cc541d6e6c8d65d153e85f850a7b256ff8a/xxhash-3.6.0-cp313-cp313t-win32.whl", hash = "sha256:1244460adc3a9be84731d72b8e80625788e5815b68da3da8b83f78115a40a7ec", size = 30916, upload-time = "2025-10-02T14:35:35.107Z" },
{ url = "https://files.pythonhosted.org/packages/ad/e6/e8cf858a2b19d6d45820f072eff1bea413910592ff17157cabc5f1227a16/xxhash-3.6.0-cp313-cp313t-win_amd64.whl", hash = "sha256:b1e420ef35c503869c4064f4a2f2b08ad6431ab7b229a05cce39d74268bca6b8", size = 31799, upload-time = "2025-10-02T14:35:36.165Z" },
{ url = "https://files.pythonhosted.org/packages/56/15/064b197e855bfb7b343210e82490ae672f8bc7cdf3ddb02e92f64304ee8a/xxhash-3.6.0-cp313-cp313t-win_arm64.whl", hash = "sha256:ec44b73a4220623235f67a996c862049f375df3b1052d9899f40a6382c32d746", size = 28044, upload-time = "2025-10-02T14:35:37.195Z" },
{ url = "https://files.pythonhosted.org/packages/7e/5e/0138bc4484ea9b897864d59fce9be9086030825bc778b76cb5a33a906d37/xxhash-3.6.0-cp314-cp314-macosx_10_13_x86_64.whl", hash = "sha256:a40a3d35b204b7cc7643cbcf8c9976d818cb47befcfac8bbefec8038ac363f3e", size = 32754, upload-time = "2025-10-02T14:35:38.245Z" },
{ url = "https://files.pythonhosted.org/packages/18/d7/5dac2eb2ec75fd771957a13e5dda560efb2176d5203f39502a5fc571f899/xxhash-3.6.0-cp314-cp314-macosx_11_0_arm64.whl", hash = "sha256:a54844be970d3fc22630b32d515e79a90d0a3ddb2644d8d7402e3c4c8da61405", size = 30846, upload-time = "2025-10-02T14:35:39.6Z" },
{ url = "https://files.pythonhosted.org/packages/fe/71/8bc5be2bb00deb5682e92e8da955ebe5fa982da13a69da5a40a4c8db12fb/xxhash-3.6.0-cp314-cp314-manylinux1_i686.manylinux_2_28_i686.manylinux_2_5_i686.whl", hash = "sha256:016e9190af8f0a4e3741343777710e3d5717427f175adfdc3e72508f59e2a7f3", size = 194343, upload-time = "2025-10-02T14:35:40.69Z" },
{ url = "https://files.pythonhosted.org/packages/e7/3b/52badfb2aecec2c377ddf1ae75f55db3ba2d321c5e164f14461c90837ef3/xxhash-3.6.0-cp314-cp314-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:4f6f72232f849eb9d0141e2ebe2677ece15adfd0fa599bc058aad83c714bb2c6", size = 213074, upload-time = "2025-10-02T14:35:42.29Z" },
{ url = "https://files.pythonhosted.org/packages/a2/2b/ae46b4e9b92e537fa30d03dbc19cdae57ed407e9c26d163895e968e3de85/xxhash-3.6.0-cp314-cp314-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:63275a8aba7865e44b1813d2177e0f5ea7eadad3dd063a21f7cf9afdc7054063", size = 212388, upload-time = "2025-10-02T14:35:43.929Z" },
{ url = "https://files.pythonhosted.org/packages/f5/80/49f88d3afc724b4ac7fbd664c8452d6db51b49915be48c6982659e0e7942/xxhash-3.6.0-cp314-cp314-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:3cd01fa2aa00d8b017c97eb46b9a794fbdca53fc14f845f5a328c71254b0abb7", size = 445614, upload-time = "2025-10-02T14:35:45.216Z" },
{ url = "https://files.pythonhosted.org/packages/ed/ba/603ce3961e339413543d8cd44f21f2c80e2a7c5cfe692a7b1f2cccf58f3c/xxhash-3.6.0-cp314-cp314-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:0226aa89035b62b6a86d3c68df4d7c1f47a342b8683da2b60cedcddb46c4d95b", size = 194024, upload-time = "2025-10-02T14:35:46.959Z" },
{ url = "https://files.pythonhosted.org/packages/78/d1/8e225ff7113bf81545cfdcd79eef124a7b7064a0bba53605ff39590b95c2/xxhash-3.6.0-cp314-cp314-musllinux_1_2_aarch64.whl", hash = "sha256:c6e193e9f56e4ca4923c61238cdaced324f0feac782544eb4c6d55ad5cc99ddd", size = 210541, upload-time = "2025-10-02T14:35:48.301Z" },
{ url = "https://files.pythonhosted.org/packages/6f/58/0f89d149f0bad89def1a8dd38feb50ccdeb643d9797ec84707091d4cb494/xxhash-3.6.0-cp314-cp314-musllinux_1_2_i686.whl", hash = "sha256:9176dcaddf4ca963d4deb93866d739a343c01c969231dbe21680e13a5d1a5bf0", size = 198305, upload-time = "2025-10-02T14:35:49.584Z" },
{ url = "https://files.pythonhosted.org/packages/11/38/5eab81580703c4df93feb5f32ff8fa7fe1e2c51c1f183ee4e48d4bb9d3d7/xxhash-3.6.0-cp314-cp314-musllinux_1_2_ppc64le.whl", hash = "sha256:c1ce4009c97a752e682b897aa99aef84191077a9433eb237774689f14f8ec152", size = 210848, upload-time = "2025-10-02T14:35:50.877Z" },
{ url = "https://files.pythonhosted.org/packages/5e/6b/953dc4b05c3ce678abca756416e4c130d2382f877a9c30a20d08ee6a77c0/xxhash-3.6.0-cp314-cp314-musllinux_1_2_s390x.whl", hash = "sha256:8cb2f4f679b01513b7adbb9b1b2f0f9cdc31b70007eaf9d59d0878809f385b11", size = 414142, upload-time = "2025-10-02T14:35:52.15Z" },
{ url = "https://files.pythonhosted.org/packages/08/a9/238ec0d4e81a10eb5026d4a6972677cbc898ba6c8b9dbaec12ae001b1b35/xxhash-3.6.0-cp314-cp314-musllinux_1_2_x86_64.whl", hash = "sha256:653a91d7c2ab54a92c19ccf43508b6a555440b9be1bc8be553376778be7f20b5", size = 191547, upload-time = "2025-10-02T14:35:53.547Z" },
{ url = "https://files.pythonhosted.org/packages/f1/ee/3cf8589e06c2164ac77c3bf0aa127012801128f1feebf2a079272da5737c/xxhash-3.6.0-cp314-cp314-win32.whl", hash = "sha256:a756fe893389483ee8c394d06b5ab765d96e68fbbfe6fde7aa17e11f5720559f", size = 31214, upload-time = "2025-10-02T14:35:54.746Z" },
{ url = "https://files.pythonhosted.org/packages/02/5d/a19552fbc6ad4cb54ff953c3908bbc095f4a921bc569433d791f755186f1/xxhash-3.6.0-cp314-cp314-win_amd64.whl", hash = "sha256:39be8e4e142550ef69629c9cd71b88c90e9a5db703fecbcf265546d9536ca4ad", size = 32290, upload-time = "2025-10-02T14:35:55.791Z" },
{ url = "https://files.pythonhosted.org/packages/b1/11/dafa0643bc30442c887b55baf8e73353a344ee89c1901b5a5c54a6c17d39/xxhash-3.6.0-cp314-cp314-win_arm64.whl", hash = "sha256:25915e6000338999236f1eb68a02a32c3275ac338628a7eaa5a269c401995679", size = 28795, upload-time = "2025-10-02T14:35:57.162Z" },
{ url = "https://files.pythonhosted.org/packages/2c/db/0e99732ed7f64182aef4a6fb145e1a295558deec2a746265dcdec12d191e/xxhash-3.6.0-cp314-cp314t-macosx_10_13_x86_64.whl", hash = "sha256:c5294f596a9017ca5a3e3f8884c00b91ab2ad2933cf288f4923c3fd4346cf3d4", size = 32955, upload-time = "2025-10-02T14:35:58.267Z" },
{ url = "https://files.pythonhosted.org/packages/55/f4/2a7c3c68e564a099becfa44bb3d398810cc0ff6749b0d3cb8ccb93f23c14/xxhash-3.6.0-cp314-cp314t-macosx_11_0_arm64.whl", hash = "sha256:1cf9dcc4ab9cff01dfbba78544297a3a01dafd60f3bde4e2bfd016cf7e4ddc67", size = 31072, upload-time = "2025-10-02T14:35:59.382Z" },
{ url = "https://files.pythonhosted.org/packages/c6/d9/72a29cddc7250e8a5819dad5d466facb5dc4c802ce120645630149127e73/xxhash-3.6.0-cp314-cp314t-manylinux1_i686.manylinux_2_28_i686.manylinux_2_5_i686.whl", hash = "sha256:01262da8798422d0685f7cef03b2bd3f4f46511b02830861df548d7def4402ad", size = 196579, upload-time = "2025-10-02T14:36:00.838Z" },
{ url = "https://files.pythonhosted.org/packages/63/93/b21590e1e381040e2ca305a884d89e1c345b347404f7780f07f2cdd47ef4/xxhash-3.6.0-cp314-cp314t-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:51a73fb7cb3a3ead9f7a8b583ffd9b8038e277cdb8cb87cf890e88b3456afa0b", size = 215854, upload-time = "2025-10-02T14:36:02.207Z" },
{ url = "https://files.pythonhosted.org/packages/ce/b8/edab8a7d4fa14e924b29be877d54155dcbd8b80be85ea00d2be3413a9ed4/xxhash-3.6.0-cp314-cp314t-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:b9c6df83594f7df8f7f708ce5ebeacfc69f72c9fbaaababf6cf4758eaada0c9b", size = 214965, upload-time = "2025-10-02T14:36:03.507Z" },
{ url = "https://files.pythonhosted.org/packages/27/67/dfa980ac7f0d509d54ea0d5a486d2bb4b80c3f1bb22b66e6a05d3efaf6c0/xxhash-3.6.0-cp314-cp314t-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:627f0af069b0ea56f312fd5189001c24578868643203bca1abbc2c52d3a6f3ca", size = 448484, upload-time = "2025-10-02T14:36:04.828Z" },
{ url = "https://files.pythonhosted.org/packages/8c/63/8ffc2cc97e811c0ca5d00ab36604b3ea6f4254f20b7bc658ca825ce6c954/xxhash-3.6.0-cp314-cp314t-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:aa912c62f842dfd013c5f21a642c9c10cd9f4c4e943e0af83618b4a404d9091a", size = 196162, upload-time = "2025-10-02T14:36:06.182Z" },
{ url = "https://files.pythonhosted.org/packages/4b/77/07f0e7a3edd11a6097e990f6e5b815b6592459cb16dae990d967693e6ea9/xxhash-3.6.0-cp314-cp314t-musllinux_1_2_aarch64.whl", hash = "sha256:b465afd7909db30168ab62afe40b2fcf79eedc0b89a6c0ab3123515dc0df8b99", size = 213007, upload-time = "2025-10-02T14:36:07.733Z" },
{ url = "https://files.pythonhosted.org/packages/ae/d8/bc5fa0d152837117eb0bef6f83f956c509332ce133c91c63ce07ee7c4873/xxhash-3.6.0-cp314-cp314t-musllinux_1_2_i686.whl", hash = "sha256:a881851cf38b0a70e7c4d3ce81fc7afd86fbc2a024f4cfb2a97cf49ce04b75d3", size = 200956, upload-time = "2025-10-02T14:36:09.106Z" },
{ url = "https://files.pythonhosted.org/packages/26/a5/d749334130de9411783873e9b98ecc46688dad5db64ca6e04b02acc8b473/xxhash-3.6.0-cp314-cp314t-musllinux_1_2_ppc64le.whl", hash = "sha256:9b3222c686a919a0f3253cfc12bb118b8b103506612253b5baeaac10d8027cf6", size = 213401, upload-time = "2025-10-02T14:36:10.585Z" },
{ url = "https://files.pythonhosted.org/packages/89/72/abed959c956a4bfc72b58c0384bb7940663c678127538634d896b1195c10/xxhash-3.6.0-cp314-cp314t-musllinux_1_2_s390x.whl", hash = "sha256:c5aa639bc113e9286137cec8fadc20e9cd732b2cc385c0b7fa673b84fc1f2a93", size = 417083, upload-time = "2025-10-02T14:36:12.276Z" },
{ url = "https://files.pythonhosted.org/packages/0c/b3/62fd2b586283b7d7d665fb98e266decadf31f058f1cf6c478741f68af0cb/xxhash-3.6.0-cp314-cp314t-musllinux_1_2_x86_64.whl", hash = "sha256:5c1343d49ac102799905e115aee590183c3921d475356cb24b4de29a4bc56518", size = 193913, upload-time = "2025-10-02T14:36:14.025Z" },
{ url = "https://files.pythonhosted.org/packages/9a/9a/c19c42c5b3f5a4aad748a6d5b4f23df3bed7ee5445accc65a0fb3ff03953/xxhash-3.6.0-cp314-cp314t-win32.whl", hash = "sha256:5851f033c3030dd95c086b4a36a2683c2ff4a799b23af60977188b057e467119", size = 31586, upload-time = "2025-10-02T14:36:15.603Z" },
{ url = "https://files.pythonhosted.org/packages/03/d6/4cc450345be9924fd5dc8c590ceda1db5b43a0a889587b0ae81a95511360/xxhash-3.6.0-cp314-cp314t-win_amd64.whl", hash = "sha256:0444e7967dac37569052d2409b00a8860c2135cff05502df4da80267d384849f", size = 32526, upload-time = "2025-10-02T14:36:16.708Z" },
{ url = "https://files.pythonhosted.org/packages/0f/c9/7243eb3f9eaabd1a88a5a5acadf06df2d83b100c62684b7425c6a11bcaa8/xxhash-3.6.0-cp314-cp314t-win_arm64.whl", hash = "sha256:bb79b1e63f6fd84ec778a4b1916dfe0a7c3fdb986c06addd5db3a0d413819d95", size = 28898, upload-time = "2025-10-02T14:36:17.843Z" },
{ url = "https://files.pythonhosted.org/packages/93/1e/8aec23647a34a249f62e2398c42955acd9b4c6ed5cf08cbea94dc46f78d2/xxhash-3.6.0-pp311-pypy311_pp73-macosx_10_15_x86_64.whl", hash = "sha256:0f7b7e2ec26c1666ad5fc9dbfa426a6a3367ceaf79db5dd76264659d509d73b0", size = 30662, upload-time = "2025-10-02T14:37:01.743Z" },
{ url = "https://files.pythonhosted.org/packages/b8/0b/b14510b38ba91caf43006209db846a696ceea6a847a0c9ba0a5b1adc53d6/xxhash-3.6.0-pp311-pypy311_pp73-manylinux1_i686.manylinux_2_28_i686.manylinux_2_5_i686.whl", hash = "sha256:5dc1e14d14fa0f5789ec29a7062004b5933964bb9b02aae6622b8f530dc40296", size = 41056, upload-time = "2025-10-02T14:37:02.879Z" },
{ url = "https://files.pythonhosted.org/packages/50/55/15a7b8a56590e66ccd374bbfa3f9ffc45b810886c8c3b614e3f90bd2367c/xxhash-3.6.0-pp311-pypy311_pp73-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:881b47fc47e051b37d94d13e7455131054b56749b91b508b0907eb07900d1c13", size = 36251, upload-time = "2025-10-02T14:37:04.44Z" },
{ url = "https://files.pythonhosted.org/packages/62/b2/5ac99a041a29e58e95f907876b04f7067a0242cb85b5f39e726153981503/xxhash-3.6.0-pp311-pypy311_pp73-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:c6dc31591899f5e5666f04cc2e529e69b4072827085c1ef15294d91a004bc1bd", size = 32481, upload-time = "2025-10-02T14:37:05.869Z" },
{ url = "https://files.pythonhosted.org/packages/7b/d9/8d95e906764a386a3d3b596f3c68bb63687dfca806373509f51ce8eea81f/xxhash-3.6.0-pp311-pypy311_pp73-win_amd64.whl", hash = "sha256:15e0dac10eb9309508bfc41f7f9deaa7755c69e35af835db9cb10751adebc35d", size = 31565, upload-time = "2025-10-02T14:37:06.966Z" },
]
[[package]]
name = "zstandard"
version = "0.25.0"
+123 -4
View File
@@ -268,7 +268,7 @@ wheels = [
[[package]]
name = "langgraph-checkpoint"
version = "4.0.1"
version = "4.0.2"
source = { editable = "../checkpoint" }
dependencies = [
{ name = "langchain-core" },
@@ -385,7 +385,7 @@ test = [
[[package]]
name = "langsmith"
version = "0.6.4"
version = "0.7.31"
source = { registry = "https://pypi.org/simple" }
dependencies = [
{ name = "httpx" },
@@ -395,11 +395,12 @@ dependencies = [
{ name = "requests" },
{ name = "requests-toolbelt" },
{ name = "uuid-utils" },
{ name = "xxhash" },
{ name = "zstandard" },
]
sdist = { url = "https://files.pythonhosted.org/packages/e7/85/9c7933052a997da1b85bc5c774f3865e9b1da1c8d71541ea133178b13229/langsmith-0.6.4.tar.gz", hash = "sha256:36f7223a01c218079fbb17da5e536ebbaf5c1468c028abe070aa3ae59bc99ec8", size = 919964, upload-time = "2026-01-15T20:02:28.873Z" }
sdist = { url = "https://files.pythonhosted.org/packages/e6/11/696019490992db5c87774dc20515529ef42a01e1d770fb754ed6d9b12fb0/langsmith-0.7.31.tar.gz", hash = "sha256:331ee4f7c26bb5be4022b9859b7d7b122cbf8c9d01d9f530114c1914b0349ffb", size = 1178480, upload-time = "2026-04-14T17:55:41.242Z" }
wheels = [
{ url = "https://files.pythonhosted.org/packages/66/0f/09a6637a7ba777eb307b7c80852d9ee26438e2bdafbad6fcc849ff9d9192/langsmith-0.6.4-py3-none-any.whl", hash = "sha256:ac4835860160be371042c7adbba3cb267bcf8d96a5ea976c33a8a4acad6c5486", size = 283503, upload-time = "2026-01-15T20:02:26.662Z" },
{ url = "https://files.pythonhosted.org/packages/1d/a1/a013cf458c301cda86a213dd153ce0a01c93f1ab5833f951e6a44c9763ce/langsmith-0.7.31-py3-none-any.whl", hash = "sha256:0291d49203f6e80dda011af1afda61eb0595a4d697adb684590a8805e1d61fb6", size = 373276, upload-time = "2026-04-14T17:55:39.677Z" },
]
[[package]]
@@ -1211,6 +1212,124 @@ wheels = [
{ url = "https://files.pythonhosted.org/packages/33/e8/e40370e6d74ddba47f002a32919d91310d6074130fe4e17dabcafc15cbf1/watchdog-6.0.0-py3-none-win_ia64.whl", hash = "sha256:a1914259fa9e1454315171103c6a30961236f508b9b623eae470268bbcc6a22f", size = 79067, upload-time = "2024-11-01T14:07:11.845Z" },
]
[[package]]
name = "xxhash"
version = "3.6.0"
source = { registry = "https://pypi.org/simple" }
sdist = { url = "https://files.pythonhosted.org/packages/02/84/30869e01909fb37a6cc7e18688ee8bf1e42d57e7e0777636bd47524c43c7/xxhash-3.6.0.tar.gz", hash = "sha256:f0162a78b13a0d7617b2845b90c763339d1f1d82bb04a4b07f4ab535cc5e05d6", size = 85160, upload-time = "2025-10-02T14:37:08.097Z" }
wheels = [
{ url = "https://files.pythonhosted.org/packages/34/ee/f9f1d656ad168681bb0f6b092372c1e533c4416b8069b1896a175c46e484/xxhash-3.6.0-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:87ff03d7e35c61435976554477a7f4cd1704c3596a89a8300d5ce7fc83874a71", size = 32845, upload-time = "2025-10-02T14:33:51.573Z" },
{ url = "https://files.pythonhosted.org/packages/a3/b1/93508d9460b292c74a09b83d16750c52a0ead89c51eea9951cb97a60d959/xxhash-3.6.0-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:f572dfd3d0e2eb1a57511831cf6341242f5a9f8298a45862d085f5b93394a27d", size = 30807, upload-time = "2025-10-02T14:33:52.964Z" },
{ url = "https://files.pythonhosted.org/packages/07/55/28c93a3662f2d200c70704efe74aab9640e824f8ce330d8d3943bf7c9b3c/xxhash-3.6.0-cp310-cp310-manylinux1_i686.manylinux_2_28_i686.manylinux_2_5_i686.whl", hash = "sha256:89952ea539566b9fed2bbd94e589672794b4286f342254fad28b149f9615fef8", size = 193786, upload-time = "2025-10-02T14:33:54.272Z" },
{ url = "https://files.pythonhosted.org/packages/c1/96/fec0be9bb4b8f5d9c57d76380a366f31a1781fb802f76fc7cda6c84893c7/xxhash-3.6.0-cp310-cp310-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:48e6f2ffb07a50b52465a1032c3cf1f4a5683f944acaca8a134a2f23674c2058", size = 212830, upload-time = "2025-10-02T14:33:55.706Z" },
{ url = "https://files.pythonhosted.org/packages/c4/a0/c706845ba77b9611f81fd2e93fad9859346b026e8445e76f8c6fd057cc6d/xxhash-3.6.0-cp310-cp310-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:b5b848ad6c16d308c3ac7ad4ba6bede80ed5df2ba8ed382f8932df63158dd4b2", size = 211606, upload-time = "2025-10-02T14:33:57.133Z" },
{ url = "https://files.pythonhosted.org/packages/67/1e/164126a2999e5045f04a69257eea946c0dc3e86541b400d4385d646b53d7/xxhash-3.6.0-cp310-cp310-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:a034590a727b44dd8ac5914236a7b8504144447a9682586c3327e935f33ec8cc", size = 444872, upload-time = "2025-10-02T14:33:58.446Z" },
{ url = "https://files.pythonhosted.org/packages/2d/4b/55ab404c56cd70a2cf5ecfe484838865d0fea5627365c6c8ca156bd09c8f/xxhash-3.6.0-cp310-cp310-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:8a8f1972e75ebdd161d7896743122834fe87378160c20e97f8b09166213bf8cc", size = 193217, upload-time = "2025-10-02T14:33:59.724Z" },
{ url = "https://files.pythonhosted.org/packages/45/e6/52abf06bac316db33aa269091ae7311bd53cfc6f4b120ae77bac1b348091/xxhash-3.6.0-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:ee34327b187f002a596d7b167ebc59a1b729e963ce645964bbc050d2f1b73d07", size = 210139, upload-time = "2025-10-02T14:34:02.041Z" },
{ url = "https://files.pythonhosted.org/packages/34/37/db94d490b8691236d356bc249c08819cbcef9273a1a30acf1254ff9ce157/xxhash-3.6.0-cp310-cp310-musllinux_1_2_i686.whl", hash = "sha256:339f518c3c7a850dd033ab416ea25a692759dc7478a71131fe8869010d2b75e4", size = 197669, upload-time = "2025-10-02T14:34:03.664Z" },
{ url = "https://files.pythonhosted.org/packages/b7/36/c4f219ef4a17a4f7a64ed3569bc2b5a9c8311abdb22249ac96093625b1a4/xxhash-3.6.0-cp310-cp310-musllinux_1_2_ppc64le.whl", hash = "sha256:bf48889c9630542d4709192578aebbd836177c9f7a4a2778a7d6340107c65f06", size = 210018, upload-time = "2025-10-02T14:34:05.325Z" },
{ url = "https://files.pythonhosted.org/packages/fd/06/bfac889a374fc2fc439a69223d1750eed2e18a7db8514737ab630534fa08/xxhash-3.6.0-cp310-cp310-musllinux_1_2_s390x.whl", hash = "sha256:5576b002a56207f640636056b4160a378fe36a58db73ae5c27a7ec8db35f71d4", size = 413058, upload-time = "2025-10-02T14:34:06.925Z" },
{ url = "https://files.pythonhosted.org/packages/c9/d1/555d8447e0dd32ad0930a249a522bb2e289f0d08b6b16204cfa42c1f5a0c/xxhash-3.6.0-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:af1f3278bd02814d6dedc5dec397993b549d6f16c19379721e5a1d31e132c49b", size = 190628, upload-time = "2025-10-02T14:34:08.669Z" },
{ url = "https://files.pythonhosted.org/packages/d1/15/8751330b5186cedc4ed4b597989882ea05e0408b53fa47bcb46a6125bfc6/xxhash-3.6.0-cp310-cp310-win32.whl", hash = "sha256:aed058764db109dc9052720da65fafe84873b05eb8b07e5e653597951af57c3b", size = 30577, upload-time = "2025-10-02T14:34:10.234Z" },
{ url = "https://files.pythonhosted.org/packages/bb/cc/53f87e8b5871a6eb2ff7e89c48c66093bda2be52315a8161ddc54ea550c4/xxhash-3.6.0-cp310-cp310-win_amd64.whl", hash = "sha256:e82da5670f2d0d98950317f82a0e4a0197150ff19a6df2ba40399c2a3b9ae5fb", size = 31487, upload-time = "2025-10-02T14:34:11.618Z" },
{ url = "https://files.pythonhosted.org/packages/9f/00/60f9ea3bb697667a14314d7269956f58bf56bb73864f8f8d52a3c2535e9a/xxhash-3.6.0-cp310-cp310-win_arm64.whl", hash = "sha256:4a082ffff8c6ac07707fb6b671caf7c6e020c75226c561830b73d862060f281d", size = 27863, upload-time = "2025-10-02T14:34:12.619Z" },
{ url = "https://files.pythonhosted.org/packages/17/d4/cc2f0400e9154df4b9964249da78ebd72f318e35ccc425e9f403c392f22a/xxhash-3.6.0-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:b47bbd8cf2d72797f3c2772eaaac0ded3d3af26481a26d7d7d41dc2d3c46b04a", size = 32844, upload-time = "2025-10-02T14:34:14.037Z" },
{ url = "https://files.pythonhosted.org/packages/5e/ec/1cc11cd13e26ea8bc3cb4af4eaadd8d46d5014aebb67be3f71fb0b68802a/xxhash-3.6.0-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:2b6821e94346f96db75abaa6e255706fb06ebd530899ed76d32cd99f20dc52fa", size = 30809, upload-time = "2025-10-02T14:34:15.484Z" },
{ url = "https://files.pythonhosted.org/packages/04/5f/19fe357ea348d98ca22f456f75a30ac0916b51c753e1f8b2e0e6fb884cce/xxhash-3.6.0-cp311-cp311-manylinux1_i686.manylinux_2_28_i686.manylinux_2_5_i686.whl", hash = "sha256:d0a9751f71a1a65ce3584e9cae4467651c7e70c9d31017fa57574583a4540248", size = 194665, upload-time = "2025-10-02T14:34:16.541Z" },
{ url = "https://files.pythonhosted.org/packages/90/3b/d1f1a8f5442a5fd8beedae110c5af7604dc37349a8e16519c13c19a9a2de/xxhash-3.6.0-cp311-cp311-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:8b29ee68625ab37b04c0b40c3fafdf24d2f75ccd778333cfb698f65f6c463f62", size = 213550, upload-time = "2025-10-02T14:34:17.878Z" },
{ url = "https://files.pythonhosted.org/packages/c4/ef/3a9b05eb527457d5db13a135a2ae1a26c80fecd624d20f3e8dcc4cb170f3/xxhash-3.6.0-cp311-cp311-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:6812c25fe0d6c36a46ccb002f40f27ac903bf18af9f6dd8f9669cb4d176ab18f", size = 212384, upload-time = "2025-10-02T14:34:19.182Z" },
{ url = "https://files.pythonhosted.org/packages/0f/18/ccc194ee698c6c623acbf0f8c2969811a8a4b6185af5e824cd27b9e4fd3e/xxhash-3.6.0-cp311-cp311-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:4ccbff013972390b51a18ef1255ef5ac125c92dc9143b2d1909f59abc765540e", size = 445749, upload-time = "2025-10-02T14:34:20.659Z" },
{ url = "https://files.pythonhosted.org/packages/a5/86/cf2c0321dc3940a7aa73076f4fd677a0fb3e405cb297ead7d864fd90847e/xxhash-3.6.0-cp311-cp311-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:297b7fbf86c82c550e12e8fb71968b3f033d27b874276ba3624ea868c11165a8", size = 193880, upload-time = "2025-10-02T14:34:22.431Z" },
{ url = "https://files.pythonhosted.org/packages/82/fb/96213c8560e6f948a1ecc9a7613f8032b19ee45f747f4fca4eb31bb6d6ed/xxhash-3.6.0-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:dea26ae1eb293db089798d3973a5fc928a18fdd97cc8801226fae705b02b14b0", size = 210912, upload-time = "2025-10-02T14:34:23.937Z" },
{ url = "https://files.pythonhosted.org/packages/40/aa/4395e669b0606a096d6788f40dbdf2b819d6773aa290c19e6e83cbfc312f/xxhash-3.6.0-cp311-cp311-musllinux_1_2_i686.whl", hash = "sha256:7a0b169aafb98f4284f73635a8e93f0735f9cbde17bd5ec332480484241aaa77", size = 198654, upload-time = "2025-10-02T14:34:25.644Z" },
{ url = "https://files.pythonhosted.org/packages/67/74/b044fcd6b3d89e9b1b665924d85d3f400636c23590226feb1eb09e1176ce/xxhash-3.6.0-cp311-cp311-musllinux_1_2_ppc64le.whl", hash = "sha256:08d45aef063a4531b785cd72de4887766d01dc8f362a515693df349fdb825e0c", size = 210867, upload-time = "2025-10-02T14:34:27.203Z" },
{ url = "https://files.pythonhosted.org/packages/bc/fd/3ce73bf753b08cb19daee1eb14aa0d7fe331f8da9c02dd95316ddfe5275e/xxhash-3.6.0-cp311-cp311-musllinux_1_2_s390x.whl", hash = "sha256:929142361a48ee07f09121fe9e96a84950e8d4df3bb298ca5d88061969f34d7b", size = 414012, upload-time = "2025-10-02T14:34:28.409Z" },
{ url = "https://files.pythonhosted.org/packages/ba/b3/5a4241309217c5c876f156b10778f3ab3af7ba7e3259e6d5f5c7d0129eb2/xxhash-3.6.0-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:51312c768403d8540487dbbfb557454cfc55589bbde6424456951f7fcd4facb3", size = 191409, upload-time = "2025-10-02T14:34:29.696Z" },
{ url = "https://files.pythonhosted.org/packages/c0/01/99bfbc15fb9abb9a72b088c1d95219fc4782b7d01fc835bd5744d66dd0b8/xxhash-3.6.0-cp311-cp311-win32.whl", hash = "sha256:d1927a69feddc24c987b337ce81ac15c4720955b667fe9b588e02254b80446fd", size = 30574, upload-time = "2025-10-02T14:34:31.028Z" },
{ url = "https://files.pythonhosted.org/packages/65/79/9d24d7f53819fe301b231044ea362ce64e86c74f6e8c8e51320de248b3e5/xxhash-3.6.0-cp311-cp311-win_amd64.whl", hash = "sha256:26734cdc2d4ffe449b41d186bbeac416f704a482ed835d375a5c0cb02bc63fef", size = 31481, upload-time = "2025-10-02T14:34:32.062Z" },
{ url = "https://files.pythonhosted.org/packages/30/4e/15cd0e3e8772071344eab2961ce83f6e485111fed8beb491a3f1ce100270/xxhash-3.6.0-cp311-cp311-win_arm64.whl", hash = "sha256:d72f67ef8bf36e05f5b6c65e8524f265bd61071471cd4cf1d36743ebeeeb06b7", size = 27861, upload-time = "2025-10-02T14:34:33.555Z" },
{ url = "https://files.pythonhosted.org/packages/9a/07/d9412f3d7d462347e4511181dea65e47e0d0e16e26fbee2ea86a2aefb657/xxhash-3.6.0-cp312-cp312-macosx_10_13_x86_64.whl", hash = "sha256:01362c4331775398e7bb34e3ab403bc9ee9f7c497bc7dee6272114055277dd3c", size = 32744, upload-time = "2025-10-02T14:34:34.622Z" },
{ url = "https://files.pythonhosted.org/packages/79/35/0429ee11d035fc33abe32dca1b2b69e8c18d236547b9a9b72c1929189b9a/xxhash-3.6.0-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:b7b2df81a23f8cb99656378e72501b2cb41b1827c0f5a86f87d6b06b69f9f204", size = 30816, upload-time = "2025-10-02T14:34:36.043Z" },
{ url = "https://files.pythonhosted.org/packages/b7/f2/57eb99aa0f7d98624c0932c5b9a170e1806406cdbcdb510546634a1359e0/xxhash-3.6.0-cp312-cp312-manylinux1_i686.manylinux_2_28_i686.manylinux_2_5_i686.whl", hash = "sha256:dc94790144e66b14f67b10ac8ed75b39ca47536bf8800eb7c24b50271ea0c490", size = 194035, upload-time = "2025-10-02T14:34:37.354Z" },
{ url = "https://files.pythonhosted.org/packages/4c/ed/6224ba353690d73af7a3f1c7cdb1fc1b002e38f783cb991ae338e1eb3d79/xxhash-3.6.0-cp312-cp312-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:93f107c673bccf0d592cdba077dedaf52fe7f42dcd7676eba1f6d6f0c3efffd2", size = 212914, upload-time = "2025-10-02T14:34:38.6Z" },
{ url = "https://files.pythonhosted.org/packages/38/86/fb6b6130d8dd6b8942cc17ab4d90e223653a89aa32ad2776f8af7064ed13/xxhash-3.6.0-cp312-cp312-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:2aa5ee3444c25b69813663c9f8067dcfaa2e126dc55e8dddf40f4d1c25d7effa", size = 212163, upload-time = "2025-10-02T14:34:39.872Z" },
{ url = "https://files.pythonhosted.org/packages/ee/dc/e84875682b0593e884ad73b2d40767b5790d417bde603cceb6878901d647/xxhash-3.6.0-cp312-cp312-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:f7f99123f0e1194fa59cc69ad46dbae2e07becec5df50a0509a808f90a0f03f0", size = 445411, upload-time = "2025-10-02T14:34:41.569Z" },
{ url = "https://files.pythonhosted.org/packages/11/4f/426f91b96701ec2f37bb2b8cec664eff4f658a11f3fa9d94f0a887ea6d2b/xxhash-3.6.0-cp312-cp312-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:49e03e6fe2cac4a1bc64952dd250cf0dbc5ef4ebb7b8d96bce82e2de163c82a2", size = 193883, upload-time = "2025-10-02T14:34:43.249Z" },
{ url = "https://files.pythonhosted.org/packages/53/5a/ddbb83eee8e28b778eacfc5a85c969673e4023cdeedcfcef61f36731610b/xxhash-3.6.0-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:bd17fede52a17a4f9a7bc4472a5867cb0b160deeb431795c0e4abe158bc784e9", size = 210392, upload-time = "2025-10-02T14:34:45.042Z" },
{ url = "https://files.pythonhosted.org/packages/1e/c2/ff69efd07c8c074ccdf0a4f36fcdd3d27363665bcdf4ba399abebe643465/xxhash-3.6.0-cp312-cp312-musllinux_1_2_i686.whl", hash = "sha256:6fb5f5476bef678f69db04f2bd1efbed3030d2aba305b0fc1773645f187d6a4e", size = 197898, upload-time = "2025-10-02T14:34:46.302Z" },
{ url = "https://files.pythonhosted.org/packages/58/ca/faa05ac19b3b622c7c9317ac3e23954187516298a091eb02c976d0d3dd45/xxhash-3.6.0-cp312-cp312-musllinux_1_2_ppc64le.whl", hash = "sha256:843b52f6d88071f87eba1631b684fcb4b2068cd2180a0224122fe4ef011a9374", size = 210655, upload-time = "2025-10-02T14:34:47.571Z" },
{ url = "https://files.pythonhosted.org/packages/d4/7a/06aa7482345480cc0cb597f5c875b11a82c3953f534394f620b0be2f700c/xxhash-3.6.0-cp312-cp312-musllinux_1_2_s390x.whl", hash = "sha256:7d14a6cfaf03b1b6f5f9790f76880601ccc7896aff7ab9cd8978a939c1eb7e0d", size = 414001, upload-time = "2025-10-02T14:34:49.273Z" },
{ url = "https://files.pythonhosted.org/packages/23/07/63ffb386cd47029aa2916b3d2f454e6cc5b9f5c5ada3790377d5430084e7/xxhash-3.6.0-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:418daf3db71e1413cfe211c2f9a528456936645c17f46b5204705581a45390ae", size = 191431, upload-time = "2025-10-02T14:34:50.798Z" },
{ url = "https://files.pythonhosted.org/packages/0f/93/14fde614cadb4ddf5e7cebf8918b7e8fac5ae7861c1875964f17e678205c/xxhash-3.6.0-cp312-cp312-win32.whl", hash = "sha256:50fc255f39428a27299c20e280d6193d8b63b8ef8028995323bf834a026b4fbb", size = 30617, upload-time = "2025-10-02T14:34:51.954Z" },
{ url = "https://files.pythonhosted.org/packages/13/5d/0d125536cbe7565a83d06e43783389ecae0c0f2ed037b48ede185de477c0/xxhash-3.6.0-cp312-cp312-win_amd64.whl", hash = "sha256:c0f2ab8c715630565ab8991b536ecded9416d615538be8ecddce43ccf26cbc7c", size = 31534, upload-time = "2025-10-02T14:34:53.276Z" },
{ url = "https://files.pythonhosted.org/packages/54/85/6ec269b0952ec7e36ba019125982cf11d91256a778c7c3f98a4c5043d283/xxhash-3.6.0-cp312-cp312-win_arm64.whl", hash = "sha256:eae5c13f3bc455a3bbb68bdc513912dc7356de7e2280363ea235f71f54064829", size = 27876, upload-time = "2025-10-02T14:34:54.371Z" },
{ url = "https://files.pythonhosted.org/packages/33/76/35d05267ac82f53ae9b0e554da7c5e281ee61f3cad44c743f0fcd354f211/xxhash-3.6.0-cp313-cp313-macosx_10_13_x86_64.whl", hash = "sha256:599e64ba7f67472481ceb6ee80fa3bd828fd61ba59fb11475572cc5ee52b89ec", size = 32738, upload-time = "2025-10-02T14:34:55.839Z" },
{ url = "https://files.pythonhosted.org/packages/31/a8/3fbce1cd96534a95e35d5120637bf29b0d7f5d8fa2f6374e31b4156dd419/xxhash-3.6.0-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:7d8b8aaa30fca4f16f0c84a5c8d7ddee0e25250ec2796c973775373257dde8f1", size = 30821, upload-time = "2025-10-02T14:34:57.219Z" },
{ url = "https://files.pythonhosted.org/packages/0c/ea/d387530ca7ecfa183cb358027f1833297c6ac6098223fd14f9782cd0015c/xxhash-3.6.0-cp313-cp313-manylinux1_i686.manylinux_2_28_i686.manylinux_2_5_i686.whl", hash = "sha256:d597acf8506d6e7101a4a44a5e428977a51c0fadbbfd3c39650cca9253f6e5a6", size = 194127, upload-time = "2025-10-02T14:34:59.21Z" },
{ url = "https://files.pythonhosted.org/packages/ba/0c/71435dcb99874b09a43b8d7c54071e600a7481e42b3e3ce1eb5226a5711a/xxhash-3.6.0-cp313-cp313-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:858dc935963a33bc33490128edc1c12b0c14d9c7ebaa4e387a7869ecc4f3e263", size = 212975, upload-time = "2025-10-02T14:35:00.816Z" },
{ url = "https://files.pythonhosted.org/packages/84/7a/c2b3d071e4bb4a90b7057228a99b10d51744878f4a8a6dd643c8bd897620/xxhash-3.6.0-cp313-cp313-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:ba284920194615cb8edf73bf52236ce2e1664ccd4a38fdb543506413529cc546", size = 212241, upload-time = "2025-10-02T14:35:02.207Z" },
{ url = "https://files.pythonhosted.org/packages/81/5f/640b6eac0128e215f177df99eadcd0f1b7c42c274ab6a394a05059694c5a/xxhash-3.6.0-cp313-cp313-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:4b54219177f6c6674d5378bd862c6aedf64725f70dd29c472eaae154df1a2e89", size = 445471, upload-time = "2025-10-02T14:35:03.61Z" },
{ url = "https://files.pythonhosted.org/packages/5e/1e/3c3d3ef071b051cc3abbe3721ffb8365033a172613c04af2da89d5548a87/xxhash-3.6.0-cp313-cp313-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:42c36dd7dbad2f5238950c377fcbf6811b1cdb1c444fab447960030cea60504d", size = 193936, upload-time = "2025-10-02T14:35:05.013Z" },
{ url = "https://files.pythonhosted.org/packages/2c/bd/4a5f68381939219abfe1c22a9e3a5854a4f6f6f3c4983a87d255f21f2e5d/xxhash-3.6.0-cp313-cp313-musllinux_1_2_aarch64.whl", hash = "sha256:f22927652cba98c44639ffdc7aaf35828dccf679b10b31c4ad72a5b530a18eb7", size = 210440, upload-time = "2025-10-02T14:35:06.239Z" },
{ url = "https://files.pythonhosted.org/packages/eb/37/b80fe3d5cfb9faff01a02121a0f4d565eb7237e9e5fc66e73017e74dcd36/xxhash-3.6.0-cp313-cp313-musllinux_1_2_i686.whl", hash = "sha256:b45fad44d9c5c119e9c6fbf2e1c656a46dc68e280275007bbfd3d572b21426db", size = 197990, upload-time = "2025-10-02T14:35:07.735Z" },
{ url = "https://files.pythonhosted.org/packages/d7/fd/2c0a00c97b9e18f72e1f240ad4e8f8a90fd9d408289ba9c7c495ed7dc05c/xxhash-3.6.0-cp313-cp313-musllinux_1_2_ppc64le.whl", hash = "sha256:6f2580ffab1a8b68ef2b901cde7e55fa8da5e4be0977c68f78fc80f3c143de42", size = 210689, upload-time = "2025-10-02T14:35:09.438Z" },
{ url = "https://files.pythonhosted.org/packages/93/86/5dd8076a926b9a95db3206aba20d89a7fc14dd5aac16e5c4de4b56033140/xxhash-3.6.0-cp313-cp313-musllinux_1_2_s390x.whl", hash = "sha256:40c391dd3cd041ebc3ffe6f2c862f402e306eb571422e0aa918d8070ba31da11", size = 414068, upload-time = "2025-10-02T14:35:11.162Z" },
{ url = "https://files.pythonhosted.org/packages/af/3c/0bb129170ee8f3650f08e993baee550a09593462a5cddd8e44d0011102b1/xxhash-3.6.0-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:f205badabde7aafd1a31e8ca2a3e5a763107a71c397c4481d6a804eb5063d8bd", size = 191495, upload-time = "2025-10-02T14:35:12.971Z" },
{ url = "https://files.pythonhosted.org/packages/e9/3a/6797e0114c21d1725e2577508e24006fd7ff1d8c0c502d3b52e45c1771d8/xxhash-3.6.0-cp313-cp313-win32.whl", hash = "sha256:2577b276e060b73b73a53042ea5bd5203d3e6347ce0d09f98500f418a9fcf799", size = 30620, upload-time = "2025-10-02T14:35:14.129Z" },
{ url = "https://files.pythonhosted.org/packages/86/15/9bc32671e9a38b413a76d24722a2bf8784a132c043063a8f5152d390b0f9/xxhash-3.6.0-cp313-cp313-win_amd64.whl", hash = "sha256:757320d45d2fbcce8f30c42a6b2f47862967aea7bf458b9625b4bbe7ee390392", size = 31542, upload-time = "2025-10-02T14:35:15.21Z" },
{ url = "https://files.pythonhosted.org/packages/39/c5/cc01e4f6188656e56112d6a8e0dfe298a16934b8c47a247236549a3f7695/xxhash-3.6.0-cp313-cp313-win_arm64.whl", hash = "sha256:457b8f85dec5825eed7b69c11ae86834a018b8e3df5e77783c999663da2f96d6", size = 27880, upload-time = "2025-10-02T14:35:16.315Z" },
{ url = "https://files.pythonhosted.org/packages/f3/30/25e5321c8732759e930c555176d37e24ab84365482d257c3b16362235212/xxhash-3.6.0-cp313-cp313t-macosx_10_13_x86_64.whl", hash = "sha256:a42e633d75cdad6d625434e3468126c73f13f7584545a9cf34e883aa1710e702", size = 32956, upload-time = "2025-10-02T14:35:17.413Z" },
{ url = "https://files.pythonhosted.org/packages/9f/3c/0573299560d7d9f8ab1838f1efc021a280b5ae5ae2e849034ef3dee18810/xxhash-3.6.0-cp313-cp313t-macosx_11_0_arm64.whl", hash = "sha256:568a6d743219e717b07b4e03b0a828ce593833e498c3b64752e0f5df6bfe84db", size = 31072, upload-time = "2025-10-02T14:35:18.844Z" },
{ url = "https://files.pythonhosted.org/packages/7a/1c/52d83a06e417cd9d4137722693424885cc9878249beb3a7c829e74bf7ce9/xxhash-3.6.0-cp313-cp313t-manylinux1_i686.manylinux_2_28_i686.manylinux_2_5_i686.whl", hash = "sha256:bec91b562d8012dae276af8025a55811b875baace6af510412a5e58e3121bc54", size = 196409, upload-time = "2025-10-02T14:35:20.31Z" },
{ url = "https://files.pythonhosted.org/packages/e3/8e/c6d158d12a79bbd0b878f8355432075fc82759e356ab5a111463422a239b/xxhash-3.6.0-cp313-cp313t-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:78e7f2f4c521c30ad5e786fdd6bae89d47a32672a80195467b5de0480aa97b1f", size = 215736, upload-time = "2025-10-02T14:35:21.616Z" },
{ url = "https://files.pythonhosted.org/packages/bc/68/c4c80614716345d55071a396cf03d06e34b5f4917a467faf43083c995155/xxhash-3.6.0-cp313-cp313t-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:3ed0df1b11a79856df5ffcab572cbd6b9627034c1c748c5566fa79df9048a7c5", size = 214833, upload-time = "2025-10-02T14:35:23.32Z" },
{ url = "https://files.pythonhosted.org/packages/7e/e9/ae27c8ffec8b953efa84c7c4a6c6802c263d587b9fc0d6e7cea64e08c3af/xxhash-3.6.0-cp313-cp313t-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:0e4edbfc7d420925b0dd5e792478ed393d6e75ff8fc219a6546fb446b6a417b1", size = 448348, upload-time = "2025-10-02T14:35:25.111Z" },
{ url = "https://files.pythonhosted.org/packages/d7/6b/33e21afb1b5b3f46b74b6bd1913639066af218d704cc0941404ca717fc57/xxhash-3.6.0-cp313-cp313t-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:fba27a198363a7ef87f8c0f6b171ec36b674fe9053742c58dd7e3201c1ab30ee", size = 196070, upload-time = "2025-10-02T14:35:26.586Z" },
{ url = "https://files.pythonhosted.org/packages/96/b6/fcabd337bc5fa624e7203aa0fa7d0c49eed22f72e93229431752bddc83d9/xxhash-3.6.0-cp313-cp313t-musllinux_1_2_aarch64.whl", hash = "sha256:794fe9145fe60191c6532fa95063765529770edcdd67b3d537793e8004cabbfd", size = 212907, upload-time = "2025-10-02T14:35:28.087Z" },
{ url = "https://files.pythonhosted.org/packages/4b/d3/9ee6160e644d660fcf176c5825e61411c7f62648728f69c79ba237250143/xxhash-3.6.0-cp313-cp313t-musllinux_1_2_i686.whl", hash = "sha256:6105ef7e62b5ac73a837778efc331a591d8442f8ef5c7e102376506cb4ae2729", size = 200839, upload-time = "2025-10-02T14:35:29.857Z" },
{ url = "https://files.pythonhosted.org/packages/0d/98/e8de5baa5109394baf5118f5e72ab21a86387c4f89b0e77ef3e2f6b0327b/xxhash-3.6.0-cp313-cp313t-musllinux_1_2_ppc64le.whl", hash = "sha256:f01375c0e55395b814a679b3eea205db7919ac2af213f4a6682e01220e5fe292", size = 213304, upload-time = "2025-10-02T14:35:31.222Z" },
{ url = "https://files.pythonhosted.org/packages/7b/1d/71056535dec5c3177eeb53e38e3d367dd1d16e024e63b1cee208d572a033/xxhash-3.6.0-cp313-cp313t-musllinux_1_2_s390x.whl", hash = "sha256:d706dca2d24d834a4661619dcacf51a75c16d65985718d6a7d73c1eeeb903ddf", size = 416930, upload-time = "2025-10-02T14:35:32.517Z" },
{ url = "https://files.pythonhosted.org/packages/dc/6c/5cbde9de2cd967c322e651c65c543700b19e7ae3e0aae8ece3469bf9683d/xxhash-3.6.0-cp313-cp313t-musllinux_1_2_x86_64.whl", hash = "sha256:5f059d9faeacd49c0215d66f4056e1326c80503f51a1532ca336a385edadd033", size = 193787, upload-time = "2025-10-02T14:35:33.827Z" },
{ url = "https://files.pythonhosted.org/packages/19/fa/0172e350361d61febcea941b0cc541d6e6c8d65d153e85f850a7b256ff8a/xxhash-3.6.0-cp313-cp313t-win32.whl", hash = "sha256:1244460adc3a9be84731d72b8e80625788e5815b68da3da8b83f78115a40a7ec", size = 30916, upload-time = "2025-10-02T14:35:35.107Z" },
{ url = "https://files.pythonhosted.org/packages/ad/e6/e8cf858a2b19d6d45820f072eff1bea413910592ff17157cabc5f1227a16/xxhash-3.6.0-cp313-cp313t-win_amd64.whl", hash = "sha256:b1e420ef35c503869c4064f4a2f2b08ad6431ab7b229a05cce39d74268bca6b8", size = 31799, upload-time = "2025-10-02T14:35:36.165Z" },
{ url = "https://files.pythonhosted.org/packages/56/15/064b197e855bfb7b343210e82490ae672f8bc7cdf3ddb02e92f64304ee8a/xxhash-3.6.0-cp313-cp313t-win_arm64.whl", hash = "sha256:ec44b73a4220623235f67a996c862049f375df3b1052d9899f40a6382c32d746", size = 28044, upload-time = "2025-10-02T14:35:37.195Z" },
{ url = "https://files.pythonhosted.org/packages/7e/5e/0138bc4484ea9b897864d59fce9be9086030825bc778b76cb5a33a906d37/xxhash-3.6.0-cp314-cp314-macosx_10_13_x86_64.whl", hash = "sha256:a40a3d35b204b7cc7643cbcf8c9976d818cb47befcfac8bbefec8038ac363f3e", size = 32754, upload-time = "2025-10-02T14:35:38.245Z" },
{ url = "https://files.pythonhosted.org/packages/18/d7/5dac2eb2ec75fd771957a13e5dda560efb2176d5203f39502a5fc571f899/xxhash-3.6.0-cp314-cp314-macosx_11_0_arm64.whl", hash = "sha256:a54844be970d3fc22630b32d515e79a90d0a3ddb2644d8d7402e3c4c8da61405", size = 30846, upload-time = "2025-10-02T14:35:39.6Z" },
{ url = "https://files.pythonhosted.org/packages/fe/71/8bc5be2bb00deb5682e92e8da955ebe5fa982da13a69da5a40a4c8db12fb/xxhash-3.6.0-cp314-cp314-manylinux1_i686.manylinux_2_28_i686.manylinux_2_5_i686.whl", hash = "sha256:016e9190af8f0a4e3741343777710e3d5717427f175adfdc3e72508f59e2a7f3", size = 194343, upload-time = "2025-10-02T14:35:40.69Z" },
{ url = "https://files.pythonhosted.org/packages/e7/3b/52badfb2aecec2c377ddf1ae75f55db3ba2d321c5e164f14461c90837ef3/xxhash-3.6.0-cp314-cp314-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:4f6f72232f849eb9d0141e2ebe2677ece15adfd0fa599bc058aad83c714bb2c6", size = 213074, upload-time = "2025-10-02T14:35:42.29Z" },
{ url = "https://files.pythonhosted.org/packages/a2/2b/ae46b4e9b92e537fa30d03dbc19cdae57ed407e9c26d163895e968e3de85/xxhash-3.6.0-cp314-cp314-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:63275a8aba7865e44b1813d2177e0f5ea7eadad3dd063a21f7cf9afdc7054063", size = 212388, upload-time = "2025-10-02T14:35:43.929Z" },
{ url = "https://files.pythonhosted.org/packages/f5/80/49f88d3afc724b4ac7fbd664c8452d6db51b49915be48c6982659e0e7942/xxhash-3.6.0-cp314-cp314-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:3cd01fa2aa00d8b017c97eb46b9a794fbdca53fc14f845f5a328c71254b0abb7", size = 445614, upload-time = "2025-10-02T14:35:45.216Z" },
{ url = "https://files.pythonhosted.org/packages/ed/ba/603ce3961e339413543d8cd44f21f2c80e2a7c5cfe692a7b1f2cccf58f3c/xxhash-3.6.0-cp314-cp314-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:0226aa89035b62b6a86d3c68df4d7c1f47a342b8683da2b60cedcddb46c4d95b", size = 194024, upload-time = "2025-10-02T14:35:46.959Z" },
{ url = "https://files.pythonhosted.org/packages/78/d1/8e225ff7113bf81545cfdcd79eef124a7b7064a0bba53605ff39590b95c2/xxhash-3.6.0-cp314-cp314-musllinux_1_2_aarch64.whl", hash = "sha256:c6e193e9f56e4ca4923c61238cdaced324f0feac782544eb4c6d55ad5cc99ddd", size = 210541, upload-time = "2025-10-02T14:35:48.301Z" },
{ url = "https://files.pythonhosted.org/packages/6f/58/0f89d149f0bad89def1a8dd38feb50ccdeb643d9797ec84707091d4cb494/xxhash-3.6.0-cp314-cp314-musllinux_1_2_i686.whl", hash = "sha256:9176dcaddf4ca963d4deb93866d739a343c01c969231dbe21680e13a5d1a5bf0", size = 198305, upload-time = "2025-10-02T14:35:49.584Z" },
{ url = "https://files.pythonhosted.org/packages/11/38/5eab81580703c4df93feb5f32ff8fa7fe1e2c51c1f183ee4e48d4bb9d3d7/xxhash-3.6.0-cp314-cp314-musllinux_1_2_ppc64le.whl", hash = "sha256:c1ce4009c97a752e682b897aa99aef84191077a9433eb237774689f14f8ec152", size = 210848, upload-time = "2025-10-02T14:35:50.877Z" },
{ url = "https://files.pythonhosted.org/packages/5e/6b/953dc4b05c3ce678abca756416e4c130d2382f877a9c30a20d08ee6a77c0/xxhash-3.6.0-cp314-cp314-musllinux_1_2_s390x.whl", hash = "sha256:8cb2f4f679b01513b7adbb9b1b2f0f9cdc31b70007eaf9d59d0878809f385b11", size = 414142, upload-time = "2025-10-02T14:35:52.15Z" },
{ url = "https://files.pythonhosted.org/packages/08/a9/238ec0d4e81a10eb5026d4a6972677cbc898ba6c8b9dbaec12ae001b1b35/xxhash-3.6.0-cp314-cp314-musllinux_1_2_x86_64.whl", hash = "sha256:653a91d7c2ab54a92c19ccf43508b6a555440b9be1bc8be553376778be7f20b5", size = 191547, upload-time = "2025-10-02T14:35:53.547Z" },
{ url = "https://files.pythonhosted.org/packages/f1/ee/3cf8589e06c2164ac77c3bf0aa127012801128f1feebf2a079272da5737c/xxhash-3.6.0-cp314-cp314-win32.whl", hash = "sha256:a756fe893389483ee8c394d06b5ab765d96e68fbbfe6fde7aa17e11f5720559f", size = 31214, upload-time = "2025-10-02T14:35:54.746Z" },
{ url = "https://files.pythonhosted.org/packages/02/5d/a19552fbc6ad4cb54ff953c3908bbc095f4a921bc569433d791f755186f1/xxhash-3.6.0-cp314-cp314-win_amd64.whl", hash = "sha256:39be8e4e142550ef69629c9cd71b88c90e9a5db703fecbcf265546d9536ca4ad", size = 32290, upload-time = "2025-10-02T14:35:55.791Z" },
{ url = "https://files.pythonhosted.org/packages/b1/11/dafa0643bc30442c887b55baf8e73353a344ee89c1901b5a5c54a6c17d39/xxhash-3.6.0-cp314-cp314-win_arm64.whl", hash = "sha256:25915e6000338999236f1eb68a02a32c3275ac338628a7eaa5a269c401995679", size = 28795, upload-time = "2025-10-02T14:35:57.162Z" },
{ url = "https://files.pythonhosted.org/packages/2c/db/0e99732ed7f64182aef4a6fb145e1a295558deec2a746265dcdec12d191e/xxhash-3.6.0-cp314-cp314t-macosx_10_13_x86_64.whl", hash = "sha256:c5294f596a9017ca5a3e3f8884c00b91ab2ad2933cf288f4923c3fd4346cf3d4", size = 32955, upload-time = "2025-10-02T14:35:58.267Z" },
{ url = "https://files.pythonhosted.org/packages/55/f4/2a7c3c68e564a099becfa44bb3d398810cc0ff6749b0d3cb8ccb93f23c14/xxhash-3.6.0-cp314-cp314t-macosx_11_0_arm64.whl", hash = "sha256:1cf9dcc4ab9cff01dfbba78544297a3a01dafd60f3bde4e2bfd016cf7e4ddc67", size = 31072, upload-time = "2025-10-02T14:35:59.382Z" },
{ url = "https://files.pythonhosted.org/packages/c6/d9/72a29cddc7250e8a5819dad5d466facb5dc4c802ce120645630149127e73/xxhash-3.6.0-cp314-cp314t-manylinux1_i686.manylinux_2_28_i686.manylinux_2_5_i686.whl", hash = "sha256:01262da8798422d0685f7cef03b2bd3f4f46511b02830861df548d7def4402ad", size = 196579, upload-time = "2025-10-02T14:36:00.838Z" },
{ url = "https://files.pythonhosted.org/packages/63/93/b21590e1e381040e2ca305a884d89e1c345b347404f7780f07f2cdd47ef4/xxhash-3.6.0-cp314-cp314t-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:51a73fb7cb3a3ead9f7a8b583ffd9b8038e277cdb8cb87cf890e88b3456afa0b", size = 215854, upload-time = "2025-10-02T14:36:02.207Z" },
{ url = "https://files.pythonhosted.org/packages/ce/b8/edab8a7d4fa14e924b29be877d54155dcbd8b80be85ea00d2be3413a9ed4/xxhash-3.6.0-cp314-cp314t-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:b9c6df83594f7df8f7f708ce5ebeacfc69f72c9fbaaababf6cf4758eaada0c9b", size = 214965, upload-time = "2025-10-02T14:36:03.507Z" },
{ url = "https://files.pythonhosted.org/packages/27/67/dfa980ac7f0d509d54ea0d5a486d2bb4b80c3f1bb22b66e6a05d3efaf6c0/xxhash-3.6.0-cp314-cp314t-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:627f0af069b0ea56f312fd5189001c24578868643203bca1abbc2c52d3a6f3ca", size = 448484, upload-time = "2025-10-02T14:36:04.828Z" },
{ url = "https://files.pythonhosted.org/packages/8c/63/8ffc2cc97e811c0ca5d00ab36604b3ea6f4254f20b7bc658ca825ce6c954/xxhash-3.6.0-cp314-cp314t-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:aa912c62f842dfd013c5f21a642c9c10cd9f4c4e943e0af83618b4a404d9091a", size = 196162, upload-time = "2025-10-02T14:36:06.182Z" },
{ url = "https://files.pythonhosted.org/packages/4b/77/07f0e7a3edd11a6097e990f6e5b815b6592459cb16dae990d967693e6ea9/xxhash-3.6.0-cp314-cp314t-musllinux_1_2_aarch64.whl", hash = "sha256:b465afd7909db30168ab62afe40b2fcf79eedc0b89a6c0ab3123515dc0df8b99", size = 213007, upload-time = "2025-10-02T14:36:07.733Z" },
{ url = "https://files.pythonhosted.org/packages/ae/d8/bc5fa0d152837117eb0bef6f83f956c509332ce133c91c63ce07ee7c4873/xxhash-3.6.0-cp314-cp314t-musllinux_1_2_i686.whl", hash = "sha256:a881851cf38b0a70e7c4d3ce81fc7afd86fbc2a024f4cfb2a97cf49ce04b75d3", size = 200956, upload-time = "2025-10-02T14:36:09.106Z" },
{ url = "https://files.pythonhosted.org/packages/26/a5/d749334130de9411783873e9b98ecc46688dad5db64ca6e04b02acc8b473/xxhash-3.6.0-cp314-cp314t-musllinux_1_2_ppc64le.whl", hash = "sha256:9b3222c686a919a0f3253cfc12bb118b8b103506612253b5baeaac10d8027cf6", size = 213401, upload-time = "2025-10-02T14:36:10.585Z" },
{ url = "https://files.pythonhosted.org/packages/89/72/abed959c956a4bfc72b58c0384bb7940663c678127538634d896b1195c10/xxhash-3.6.0-cp314-cp314t-musllinux_1_2_s390x.whl", hash = "sha256:c5aa639bc113e9286137cec8fadc20e9cd732b2cc385c0b7fa673b84fc1f2a93", size = 417083, upload-time = "2025-10-02T14:36:12.276Z" },
{ url = "https://files.pythonhosted.org/packages/0c/b3/62fd2b586283b7d7d665fb98e266decadf31f058f1cf6c478741f68af0cb/xxhash-3.6.0-cp314-cp314t-musllinux_1_2_x86_64.whl", hash = "sha256:5c1343d49ac102799905e115aee590183c3921d475356cb24b4de29a4bc56518", size = 193913, upload-time = "2025-10-02T14:36:14.025Z" },
{ url = "https://files.pythonhosted.org/packages/9a/9a/c19c42c5b3f5a4aad748a6d5b4f23df3bed7ee5445accc65a0fb3ff03953/xxhash-3.6.0-cp314-cp314t-win32.whl", hash = "sha256:5851f033c3030dd95c086b4a36a2683c2ff4a799b23af60977188b057e467119", size = 31586, upload-time = "2025-10-02T14:36:15.603Z" },
{ url = "https://files.pythonhosted.org/packages/03/d6/4cc450345be9924fd5dc8c590ceda1db5b43a0a889587b0ae81a95511360/xxhash-3.6.0-cp314-cp314t-win_amd64.whl", hash = "sha256:0444e7967dac37569052d2409b00a8860c2135cff05502df4da80267d384849f", size = 32526, upload-time = "2025-10-02T14:36:16.708Z" },
{ url = "https://files.pythonhosted.org/packages/0f/c9/7243eb3f9eaabd1a88a5a5acadf06df2d83b100c62684b7425c6a11bcaa8/xxhash-3.6.0-cp314-cp314t-win_arm64.whl", hash = "sha256:bb79b1e63f6fd84ec778a4b1916dfe0a7c3fdb986c06addd5db3a0d413819d95", size = 28898, upload-time = "2025-10-02T14:36:17.843Z" },
{ url = "https://files.pythonhosted.org/packages/93/1e/8aec23647a34a249f62e2398c42955acd9b4c6ed5cf08cbea94dc46f78d2/xxhash-3.6.0-pp311-pypy311_pp73-macosx_10_15_x86_64.whl", hash = "sha256:0f7b7e2ec26c1666ad5fc9dbfa426a6a3367ceaf79db5dd76264659d509d73b0", size = 30662, upload-time = "2025-10-02T14:37:01.743Z" },
{ url = "https://files.pythonhosted.org/packages/b8/0b/b14510b38ba91caf43006209db846a696ceea6a847a0c9ba0a5b1adc53d6/xxhash-3.6.0-pp311-pypy311_pp73-manylinux1_i686.manylinux_2_28_i686.manylinux_2_5_i686.whl", hash = "sha256:5dc1e14d14fa0f5789ec29a7062004b5933964bb9b02aae6622b8f530dc40296", size = 41056, upload-time = "2025-10-02T14:37:02.879Z" },
{ url = "https://files.pythonhosted.org/packages/50/55/15a7b8a56590e66ccd374bbfa3f9ffc45b810886c8c3b614e3f90bd2367c/xxhash-3.6.0-pp311-pypy311_pp73-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:881b47fc47e051b37d94d13e7455131054b56749b91b508b0907eb07900d1c13", size = 36251, upload-time = "2025-10-02T14:37:04.44Z" },
{ url = "https://files.pythonhosted.org/packages/62/b2/5ac99a041a29e58e95f907876b04f7067a0242cb85b5f39e726153981503/xxhash-3.6.0-pp311-pypy311_pp73-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:c6dc31591899f5e5666f04cc2e529e69b4072827085c1ef15294d91a004bc1bd", size = 32481, upload-time = "2025-10-02T14:37:05.869Z" },
{ url = "https://files.pythonhosted.org/packages/7b/d9/8d95e906764a386a3d3b596f3c68bb63687dfca806373509f51ce8eea81f/xxhash-3.6.0-pp311-pypy311_pp73-win_amd64.whl", hash = "sha256:15e0dac10eb9309508bfc41f7f9deaa7755c69e35af835db9cb10751adebc35d", size = 31565, upload-time = "2025-10-02T14:37:06.966Z" },
]
[[package]]
name = "zstandard"
version = "0.25.0"
@@ -1,9 +1,10 @@
from __future__ import annotations
import contextvars
import copy
import logging
from collections.abc import AsyncIterator, Collection, Iterator, Mapping, Sequence
from typing import ( # noqa: UP035
from typing import (
Any,
Generic,
Literal,
@@ -18,6 +19,9 @@ from langgraph.checkpoint.base.id import uuid6
from langgraph.checkpoint.serde.base import SerializerProtocol, maybe_add_typed_methods
from langgraph.checkpoint.serde.encrypted import EncryptedSerializer
from langgraph.checkpoint.serde.jsonplus import JsonPlusSerializer
from langgraph.checkpoint.serde.types import (
DELTA_SENTINEL as DELTA_SENTINEL,
)
from langgraph.checkpoint.serde.types import (
ERROR,
INTERRUPT,
@@ -28,6 +32,16 @@ from langgraph.checkpoint.serde.types import (
V = TypeVar("V", int, float, str)
PendingWrite = tuple[str, str, Any]
# Task-local guard: ContextVar is copied per asyncio Task, so concurrent
# requests on the same event-loop thread do not share this flag. A plain
# `threading.local()` would leak across tasks and let one in-flight
# reconstruction silently short-circuit another.
_DELTA_RECONSTRUCTION: contextvars.ContextVar[bool] = contextvars.ContextVar(
"_DELTA_RECONSTRUCTION", default=False
)
logger = logging.getLogger(__name__)
@@ -119,6 +133,30 @@ class CheckpointTuple(NamedTuple):
pending_writes: list[PendingWrite] | None = None
class _ChannelWritesHistory(NamedTuple):
"""Result of `BaseCheckpointSaver._get_channel_writes_history`.
Storage-level view of what one channel wrote across the ancestor chain
of a target checkpoint:
* `seed` — the nearest ancestor's stored blob value for this channel,
or `DELTA_SENTINEL` if the walk reached the root without finding a
stored value. A non-sentinel seed typically indicates a pre-delta
snapshot preserved across a channel-type migration (e.g.
`BinaryOperatorAggregate` storage extended under `DeltaChannel`).
* `writes` — on-path deltas oldest→newest, one `PendingWrite` per
step that wrote to this channel. Writes stored at the target
checkpoint itself are pending for the next super-step and are
excluded.
Experimental: method surface may change; the NamedTuple shape is the
contract.
"""
seed: Any
writes: list[PendingWrite]
class BaseCheckpointSaver(Generic[V]):
"""Base class for creating a graph checkpointer.
@@ -457,6 +495,104 @@ class BaseCheckpointSaver(Generic[V]):
"""
raise NotImplementedError
def _get_channel_writes_history(
self, config: RunnableConfig, channel: str
) -> _ChannelWritesHistory:
"""**Experimental.** Query one channel's writes along the parent chain.
Storage-level query, not channel semantics: returns `(seed, writes)`
reflecting what storage knows about a single channel across the
ancestor chain of the target checkpoint identified by `config`.
* `writes` — on-path deltas oldest→newest as `PendingWrite` tuples.
Writes stored at the target `checkpoint_id` itself are pending
for the next super-step and are excluded.
* `seed` — the nearest ancestor's stored blob value for this
channel; `DELTA_SENTINEL` if the walk reached the root without
finding a stored value. A non-sentinel seed typically indicates
a pre-delta snapshot preserved across a channel-type migration.
Walks the **parent chain** (not `list(before=...)`): for forked
threads, only on-path ancestors contribute.
Reference implementation walks `get_tuple` + `parent_config`,
inspecting each ancestor's `channel_values[channel]` for the seed
terminator. Savers with direct storage access (`InMemorySaver`,
`PostgresSaver`) override for performance; the return contract is
fixed here.
Underscore-prefixed because the method surface is experimental.
"""
# Guard against re-entrant calls: when get_tuple() triggers
# reconstruction which calls get_tuple() again, the inner call
# short-circuits here.
if _DELTA_RECONSTRUCTION.get():
return _ChannelWritesHistory(seed=DELTA_SENTINEL, writes=[])
token = _DELTA_RECONSTRUCTION.set(True)
try:
collected: list[PendingWrite] = [] # newest first; reversed at the end
target_tuple = self.get_tuple(config)
cursor_config: RunnableConfig | None = (
target_tuple.parent_config if target_tuple else None
)
while cursor_config is not None:
tup = self.get_tuple(cursor_config)
if tup is None:
break
# Pre-delta seed terminator: if the ancestor has a stored
# (non-sentinel) value for this channel, that snapshot
# subsumes any earlier writes on the chain. Stop here.
ancestor_value = tup.checkpoint["channel_values"].get(channel)
if ancestor_value is not None and ancestor_value is not DELTA_SENTINEL:
collected.reverse()
return _ChannelWritesHistory(seed=ancestor_value, writes=collected)
if tup.pending_writes:
# Within a superstep, pending_writes are oldest→newest;
# reverse to scan newest-first.
for write in reversed(tup.pending_writes):
if write[1] != channel:
continue
collected.append(write)
cursor_config = tup.parent_config
collected.reverse()
return _ChannelWritesHistory(seed=DELTA_SENTINEL, writes=collected)
finally:
_DELTA_RECONSTRUCTION.reset(token)
async def _aget_channel_writes_history(
self, config: RunnableConfig, channel: str
) -> _ChannelWritesHistory:
"""Async version of `_get_channel_writes_history`. See docstring there."""
if _DELTA_RECONSTRUCTION.get():
return _ChannelWritesHistory(seed=DELTA_SENTINEL, writes=[])
token = _DELTA_RECONSTRUCTION.set(True)
try:
collected: list[PendingWrite] = []
target_tuple = await self.aget_tuple(config)
cursor_config: RunnableConfig | None = (
target_tuple.parent_config if target_tuple else None
)
while cursor_config is not None:
tup = await self.aget_tuple(cursor_config)
if tup is None:
break
ancestor_value = tup.checkpoint["channel_values"].get(channel)
if ancestor_value is not None and ancestor_value is not DELTA_SENTINEL:
collected.reverse()
return _ChannelWritesHistory(seed=ancestor_value, writes=collected)
if tup.pending_writes:
for write in reversed(tup.pending_writes):
if write[1] != channel:
continue
collected.append(write)
cursor_config = tup.parent_config
collected.reverse()
return _ChannelWritesHistory(seed=DELTA_SENTINEL, writes=collected)
finally:
_DELTA_RECONSTRUCTION.reset(token)
def get_next_version(self, current: V | None, channel: None) -> V:
"""Generate the next version ID for a channel.
@@ -9,18 +9,21 @@ from collections import defaultdict
from collections.abc import AsyncIterator, Iterator, Sequence
from contextlib import AbstractAsyncContextManager, AbstractContextManager, ExitStack
from types import TracebackType
from typing import Any
from typing import Any, cast
from langchain_core.runnables import RunnableConfig
from langgraph.checkpoint.base import (
DELTA_SENTINEL,
WRITES_IDX_MAP,
BaseCheckpointSaver,
ChannelVersions,
Checkpoint,
CheckpointMetadata,
CheckpointTuple,
PendingWrite,
SerializerProtocol,
_ChannelWritesHistory,
get_checkpoint_id,
get_checkpoint_metadata,
)
@@ -121,16 +124,91 @@ class InMemorySaver(
return self.stack.__exit__(__exc_type, __exc_value, __traceback)
def _load_blobs(
self, thread_id: str, checkpoint_ns: str, versions: ChannelVersions
self,
thread_id: str,
checkpoint_ns: str,
versions: ChannelVersions,
) -> dict[str, Any]:
channel_values: dict[str, Any] = {}
for k, v in versions.items():
kk = (thread_id, checkpoint_ns, k, v)
if kk in self.blobs:
vv = self.blobs[kk]
if vv[0] != "empty":
channel_values[k] = self.serde.loads_typed(vv)
return channel_values
result: dict[str, Any] = {}
for k, ver in versions.items():
kk = (thread_id, checkpoint_ns, k, ver)
if kk not in self.blobs:
continue
vv = self.blobs[kk]
if vv[0] == "empty":
continue
result[k] = self.serde.loads_typed(vv)
return result
def _get_channel_writes_history(
self, config: RunnableConfig, channel: str
) -> _ChannelWritesHistory:
thread_id = config["configurable"]["thread_id"]
checkpoint_ns = config["configurable"].get("checkpoint_ns", "")
checkpoint_id = config["configurable"].get("checkpoint_id", "")
ns_storage = self.storage.get(thread_id, {}).get(checkpoint_ns, {})
# Walk the parent chain newest→oldest. Skip the target itself —
# writes stored AT `checkpoint_id` are pending for the next step
# (pregel applies them via `apply_writes`; they aren't part of the
# snapshot value AT `checkpoint_id`).
chain: list[str] = []
target_entry = ns_storage.get(checkpoint_id)
current: str | None = target_entry[2] if target_entry is not None else None
while current is not None:
entry = ns_storage.get(current)
if entry is None:
break
chain.append(current)
_, _, parent = entry
current = parent
# Scan newest→oldest. A pre-delta blob on an ancestor terminates the
# walk and is bound as `seed`; without this, a thread migrated from
# pre-delta storage would replay ancestor writes all the way to the
# root AND miss any value that lived only in the old blob (e.g. from
# `update_state`).
#
# At each ancestor, check the blob BEFORE processing its pending
# writes: a pre-delta blob represents the state AT that ancestor,
# which already subsumes any writes stored under it. Processing
# those writes first would fold them into the reconstructed value
# twice (once via the blob, once via replay).
collected: list[PendingWrite] = [] # newest first
for cp_id in chain: # newest → oldest
entry = ns_storage.get(cp_id)
if entry is not None:
ckpt = self.serde.loads_typed(entry[0])
ver = ckpt.get("channel_versions", {}).get(channel)
if ver is not None:
blob_entry = self.blobs.get(
(thread_id, checkpoint_ns, channel, ver)
)
if blob_entry is not None and blob_entry[0] != "empty":
blob_value = self.serde.loads_typed(blob_entry)
if blob_value is not DELTA_SENTINEL:
# Pre-delta snapshot terminator. Skip this
# ancestor's writes — the blob subsumes them.
collected.reverse()
return _ChannelWritesHistory(
seed=blob_value, writes=collected
)
step_writes = self.writes.get((thread_id, checkpoint_ns, cp_id), {})
# Within a superstep, sorted by (task_id, idx) = oldest → newest;
# reverse for newest-first scan.
for (_task_id, _idx), (tid, ch, serialized, _) in sorted(
step_writes.items(), reverse=True
):
if ch != channel:
continue
val = self.serde.loads_typed(serialized)
collected.append((tid, ch, val))
collected.reverse()
return _ChannelWritesHistory(seed=DELTA_SENTINEL, writes=collected)
async def _aget_channel_writes_history(
self, config: RunnableConfig, channel: str
) -> _ChannelWritesHistory:
return self._get_channel_writes_history(config, channel)
def get_tuple(self, config: RunnableConfig) -> CheckpointTuple | None:
"""Get a checkpoint tuple from the in-memory storage.
@@ -153,13 +231,16 @@ class InMemorySaver(
checkpoint, metadata, parent_checkpoint_id = saved
writes = self.writes[(thread_id, checkpoint_ns, checkpoint_id)].values()
checkpoint_: Checkpoint = self.serde.loads_typed(checkpoint)
channel_values = self._load_blobs(
thread_id,
checkpoint_ns,
checkpoint_["channel_versions"],
)
return CheckpointTuple(
config=config,
checkpoint={
**checkpoint_,
"channel_values": self._load_blobs(
thread_id, checkpoint_ns, checkpoint_["channel_versions"]
),
"channel_values": channel_values,
},
metadata=self.serde.loads_typed(metadata),
pending_writes=[
@@ -183,19 +264,26 @@ class InMemorySaver(
checkpoint, metadata, parent_checkpoint_id = checkpoints[checkpoint_id]
writes = self.writes[(thread_id, checkpoint_ns, checkpoint_id)].values()
checkpoint_ = self.serde.loads_typed(checkpoint)
return CheckpointTuple(
config={
resolved_config = cast(
RunnableConfig,
{
"configurable": {
"thread_id": thread_id,
"checkpoint_ns": checkpoint_ns,
"checkpoint_id": checkpoint_id,
}
},
)
channel_values = self._load_blobs(
thread_id,
checkpoint_ns,
checkpoint_["channel_versions"],
)
return CheckpointTuple(
config=resolved_config,
checkpoint={
**checkpoint_,
"channel_values": self._load_blobs(
thread_id, checkpoint_ns, checkpoint_["channel_versions"]
),
"channel_values": channel_values,
},
metadata=self.serde.loads_typed(metadata),
pending_writes=[
@@ -290,21 +378,27 @@ class InMemorySaver(
checkpoint_: Checkpoint = self.serde.loads_typed(checkpoint)
yield CheckpointTuple(
config={
list_config = cast(
RunnableConfig,
{
"configurable": {
"thread_id": thread_id,
"checkpoint_ns": checkpoint_ns,
"checkpoint_id": checkpoint_id,
}
},
)
channel_values = self._load_blobs(
thread_id,
checkpoint_ns,
checkpoint_["channel_versions"],
)
yield CheckpointTuple(
config=list_config,
checkpoint={
**checkpoint_,
"channel_values": self._load_blobs(
thread_id,
checkpoint_ns,
checkpoint_["channel_versions"],
),
"channel_values": channel_values,
},
metadata=metadata,
parent_config=(
@@ -33,19 +33,35 @@ from langchain_core.load.load import Reviver
from langgraph.checkpoint.serde import _msgpack as _lg_msgpack
from langgraph.checkpoint.serde.base import SerializerProtocol
from langgraph.checkpoint.serde.event_hooks import emit_serde_event
from langgraph.checkpoint.serde.types import SendProtocol
from langgraph.checkpoint.serde.types import DELTA_SENTINEL, SendProtocol
from langgraph.store.base import Item
if TYPE_CHECKING:
from langgraph.checkpoint.serde._msgpack import (
AllowedMsgpackModules,
)
from langgraph.checkpoint.serde.types import SendProtocol
LC_REVIVER = Reviver()
EMPTY_BYTES = b""
logger = logging.getLogger(__name__)
# Dedup log warnings across process lifetime; cap bounds state if types are
# dynamically generated (also acts as a circuit breaker on warning volume).
# Dedup is best-effort: racing threads may each emit once for the same key,
# and warnings are silently dropped once _MAX_WARNED_TYPES is reached.
_MAX_WARNED_TYPES = 1000
_warned_unregistered_types: set[tuple[str, str]] = set()
_warned_blocked_types: set[tuple[str, str]] = set()
def _warn_once(
seen: set[tuple[str, str]], key: tuple[str, str], msg: str, *args: object
) -> None:
if key in seen or len(seen) >= _MAX_WARNED_TYPES:
return
seen.add(key)
logger.warning(msg, *args)
class JsonPlusSerializer(SerializerProtocol):
"""Serializer that uses ormsgpack, with optional fallbacks.
@@ -235,6 +251,8 @@ class JsonPlusSerializer(SerializerProtocol):
def dumps_typed(self, obj: Any) -> tuple[str, bytes]:
if obj is None:
return "null", EMPTY_BYTES
elif obj is DELTA_SENTINEL:
return "delta", EMPTY_BYTES
elif isinstance(obj, bytes):
return "bytes", obj
elif isinstance(obj, bytearray):
@@ -261,6 +279,8 @@ class JsonPlusSerializer(SerializerProtocol):
return ormsgpack.unpackb(
data_, ext_hook=self._unpack_ext_hook, option=ormsgpack.OPT_NON_STR_KEYS
)
elif type_ == "delta":
return DELTA_SENTINEL
elif self.pickle_fallback and type_ == "pickle":
return pickle.loads(data_)
else:
@@ -534,7 +554,9 @@ def _create_msgpack_ext_hook(
"name": name,
}
)
logger.warning(
_warn_once(
_warned_unregistered_types,
key,
"Deserializing unregistered type %s.%s from checkpoint. "
"This will be blocked in a future version. "
"Set LANGGRAPH_STRICT_MSGPACK=true to block now, or add "
@@ -556,7 +578,9 @@ def _create_msgpack_ext_hook(
"name": name,
}
)
logger.warning(
_warn_once(
_warned_blocked_types,
key,
"Blocked deserialization of %s.%s - not in allowed_msgpack_modules. "
"Add to allowed_msgpack_modules to allow: [(%r, %r)]",
module,
@@ -14,6 +14,25 @@ INTERRUPT = "__interrupt__"
RESUME = "__resume__"
TASKS = "__pregel_tasks"
class _DeltaSentinel:
"""Singleton marker stored (as zero bytes) in checkpoint_blobs for a
DeltaChannel field. The actual per-step writes live in checkpoint_writes
and are replayed through the reducer at load time.
Compare with `is DELTA_SENTINEL` — `loads_typed` always returns the same
module-level instance.
"""
__slots__ = ()
def __repr__(self) -> str:
return "DELTA_SENTINEL"
DELTA_SENTINEL = _DeltaSentinel()
Value = TypeVar("Value", covariant=True)
Update = TypeVar("Update", contravariant=True)
C = TypeVar("C")
+1 -1
View File
@@ -4,7 +4,7 @@ build-backend = "hatchling.build"
[project]
name = "langgraph-checkpoint"
version = "4.0.1"
version = "4.0.2"
description = "Library with base interfaces for LangGraph checkpoint savers."
authors = []
requires-python = ">=3.10"
+9
View File
@@ -29,6 +29,8 @@ from langgraph.checkpoint.serde.jsonplus import (
EXT_METHOD_SINGLE_ARG,
JsonPlusSerializer,
_msgpack_enc,
_warned_blocked_types,
_warned_unregistered_types,
)
@@ -102,6 +104,13 @@ def test_msgpack_method_pathlib_blocked_encrypted_strict(
class TestEncryptedSerializerMsgpackAllowlist:
"""Test msgpack allowlist behavior through EncryptedSerializer."""
@pytest.fixture(autouse=True)
def _reset_warned_types(self) -> None:
# Warning dedup state is process-global; reset per-test so each case
# sees a fresh slate and assertions about warning emission are stable.
_warned_unregistered_types.clear()
_warned_blocked_types.clear()
def test_safe_types_no_warning(self, caplog: pytest.LogCaptureFixture) -> None:
"""Test safe types deserialize without warnings through encryption."""
serde = _make_encrypted_serde()
+29 -2
View File
@@ -35,6 +35,8 @@ from langgraph.checkpoint.serde.jsonplus import (
JsonPlusSerializer,
_msgpack_enc,
_msgpack_ext_hook_to_json,
_warned_blocked_types,
_warned_unregistered_types,
)
from langgraph.store.base import Item
@@ -580,6 +582,14 @@ def test_msgpack_safe_types_no_warning(caplog: pytest.LogCaptureFixture) -> None
assert result is not None
@pytest.fixture(autouse=True)
def _reset_warned_types() -> None:
# Warning dedup state is process-global; reset per-test so each case sees
# a fresh slate and assertions about warning emission are stable.
_warned_unregistered_types.clear()
_warned_blocked_types.clear()
def test_msgpack_pydantic_warns_by_default(caplog: pytest.LogCaptureFixture) -> None:
"""Pydantic models not in allowlist should log warning but still deserialize."""
current = _lg_msgpack.STRICT_MSGPACK_ENABLED
@@ -595,6 +605,12 @@ def test_msgpack_pydantic_warns_by_default(caplog: pytest.LogCaptureFixture) ->
assert "unregistered type" in caplog.text.lower()
assert "allowed_msgpack_modules" in caplog.text
assert result == obj
# Second deserialization of the same type should NOT produce another warning
caplog.clear()
result2 = serde.loads_typed(dumped)
assert "unregistered type" not in caplog.text.lower()
assert result2 == obj
_lg_msgpack.STRICT_MSGPACK_ENABLED = current
@@ -639,7 +655,6 @@ def test_msgpack_allowlist_silences_warning(caplog: pytest.LogCaptureFixture) ->
def test_msgpack_none_blocks_unregistered(caplog: pytest.LogCaptureFixture) -> None:
"""allowed_msgpack_modules=None should block unregistered types."""
serde = JsonPlusSerializer(allowed_msgpack_modules=None)
obj = MyPydantic(foo="test", bar=42, inner=InnerPydantic(hello="world"))
@@ -657,7 +672,6 @@ def test_msgpack_allowlist_blocks_non_listed(
caplog: pytest.LogCaptureFixture,
) -> None:
"""Allowlists should block unregistered types even if msgpack is enabled."""
serde = JsonPlusSerializer(
allowed_msgpack_modules=[("tests.test_jsonplus", "MyPydantic")]
)
@@ -983,3 +997,16 @@ def test_msgpack_nested_pydantic_serializes_as_dict(
# No blocking should occur - inner is serialized as dict, not ext
assert "blocked" not in caplog.text.lower()
assert result == obj
def test_delta_sentinel_serde_round_trip() -> None:
from langgraph.checkpoint.base import DELTA_SENTINEL
from langgraph.checkpoint.serde.jsonplus import JsonPlusSerializer
serde = JsonPlusSerializer()
type_tag, blob = serde.dumps_typed(DELTA_SENTINEL)
# Zero-byte "delta" tag — no allowlist change needed.
assert type_tag == "delta"
assert blob == b""
loaded = serde.loads_typed((type_tag, blob))
assert loaded is DELTA_SENTINEL
+358 -3
View File
@@ -6,19 +6,32 @@ from langchain_core.runnables import RunnableConfig
from pydantic import BaseModel
from langgraph.checkpoint.base import (
DELTA_SENTINEL,
Checkpoint,
CheckpointMetadata,
create_checkpoint,
empty_checkpoint,
)
from langgraph.checkpoint.memory import InMemorySaver
from langgraph.checkpoint.serde.jsonplus import JsonPlusSerializer
from langgraph.checkpoint.serde.jsonplus import (
JsonPlusSerializer,
_warned_blocked_types,
_warned_unregistered_types,
)
class MemoryPydantic(BaseModel):
foo: str
@pytest.fixture(autouse=True)
def _reset_warned_types() -> None:
# Warning dedup state is process-global; reset per-test so each case sees
# a fresh slate and assertions about warning emission are stable.
_warned_unregistered_types.clear()
_warned_blocked_types.clear()
class TestMemorySaver:
@pytest.fixture(autouse=True)
def setup(self) -> None:
@@ -196,8 +209,6 @@ class TestMemorySaver:
async def test_memory_saver() -> None:
from langgraph.checkpoint.memory import InMemorySaver
memory_saver = InMemorySaver()
assert isinstance(memory_saver, InMemorySaver)
@@ -308,3 +319,347 @@ def test_memory_saver_with_allowlist_proxy_isolated() -> None:
assert direct is not None
expected = obj.model_dump() if hasattr(obj, "model_dump") else obj.dict()
assert direct.checkpoint["channel_values"]["foo"] == expected
class TestInMemorySaverDeltaChannel:
def test_load_blobs_returns_sentinel_for_delta_channel(self) -> None:
"""_load_blobs returns DELTA_SENTINEL for delta channels (reconstruction deferred)."""
saver = InMemorySaver()
serde = JsonPlusSerializer()
thread_id, ns, channel = "t1", "", "messages"
v1 = "00000000000000000000000000000001.0000000000000000"
saver.blobs[(thread_id, ns, channel, v1)] = serde.dumps_typed(DELTA_SENTINEL)
cp1 = empty_checkpoint()
cp1["id"] = "cp1"
cp1["channel_versions"][channel] = v1
saver.storage[thread_id][ns] = {
"cp1": (serde.dumps_typed(cp1), serde.dumps_typed({}), None),
}
result = saver._load_blobs(thread_id, ns, {channel: v1})
assert channel in result
assert result[channel] is DELTA_SENTINEL
def test_get_channel_writes_collects_ancestor_writes_only(self) -> None:
"""_get_channel_writes_history collects ancestor writes oldest→newest,
and excludes writes stored at the target checkpoint itself (those are
pending writes for the next step, applied separately by pregel)."""
saver = InMemorySaver()
serde = JsonPlusSerializer()
thread_id, ns, channel = "t1", "", "messages"
cp1 = empty_checkpoint()
cp1["id"] = "cp1"
cp2 = empty_checkpoint()
cp2["id"] = "cp2"
saver.storage[thread_id][ns] = {
"cp1": (serde.dumps_typed(cp1), serde.dumps_typed({}), None),
"cp2": (serde.dumps_typed(cp2), serde.dumps_typed({}), "cp1"),
}
# Writes stored at cp1 produced the cp1 snapshot; part of history.
saver.writes[(thread_id, ns, "cp1")][("task1", 0)] = (
"task1",
channel,
serde.dumps_typed({"content": "hi"}),
"",
)
# Writes stored at cp2 are pending — they will produce cp3 when the
# step that loaded cp2 completes. They MUST NOT appear in the
# reconstructed snapshot value at cp2.
saver.writes[(thread_id, ns, "cp2")][("task2", 0)] = (
"task2",
channel,
serde.dumps_typed({"content": "pending"}),
"",
)
config: RunnableConfig = {
"configurable": {
"thread_id": thread_id,
"checkpoint_ns": ns,
"checkpoint_id": "cp2",
}
}
result = saver._get_channel_writes_history(config, channel)
assert result.seed is DELTA_SENTINEL
values = [v for _, _, v in result.writes]
assert values == [{"content": "hi"}]
def test_get_channel_writes_at_root_returns_empty(self) -> None:
"""Reconstructing the root checkpoint's state: no ancestors → []."""
saver = InMemorySaver()
serde = JsonPlusSerializer()
thread_id, ns, channel = "t1", "", "messages"
cp1 = empty_checkpoint()
cp1["id"] = "cp1"
saver.storage[thread_id][ns] = {
"cp1": (serde.dumps_typed(cp1), serde.dumps_typed({}), None),
}
saver.writes[(thread_id, ns, "cp1")][("task1", 0)] = (
"task1",
channel,
serde.dumps_typed({"content": "pending"}),
"",
)
config: RunnableConfig = {
"configurable": {
"thread_id": thread_id,
"checkpoint_ns": ns,
"checkpoint_id": "cp1",
}
}
result = saver._get_channel_writes_history(config, channel)
assert result.seed is DELTA_SENTINEL
assert result.writes == []
class TestBaseFallbackGetChannelWrites:
"""Exercises the `BaseCheckpointSaver._get_channel_writes_history` default
implementation — the path third-party savers inherit when they don't
override `_get_channel_writes_history` themselves.
Regression guard for a bug where the fallback passed the caller's config
(with `checkpoint_id`) straight to `self.list()`, which most savers
collapse to a single row — causing the fallback to return `[]`.
"""
def _build_saver_with_chain(self) -> tuple[InMemorySaver, str, str]:
"""Build an InMemorySaver with a 3-checkpoint chain and per-step writes
for a `messages` channel.
Returns `(saver, thread_id, namespace)`. The saver subclass deletes the
InMemorySaver override so the base class fallback is exercised.
"""
class _ThirdPartyStyleSaver(InMemorySaver):
_get_channel_writes_history = (
InMemorySaver.__mro__[1]._get_channel_writes_history # type: ignore[attr-defined]
)
_aget_channel_writes_history = (
InMemorySaver.__mro__[1]._aget_channel_writes_history # type: ignore[attr-defined]
)
saver = _ThirdPartyStyleSaver()
serde = JsonPlusSerializer()
thread_id, ns, channel = "t1", "", "messages"
cp0 = empty_checkpoint()
cp0["id"] = "00000000000000000000000000000001.0000000000000000"
cp1 = empty_checkpoint()
cp1["id"] = "00000000000000000000000000000002.0000000000000000"
cp2 = empty_checkpoint()
cp2["id"] = "00000000000000000000000000000003.0000000000000000"
saver.storage[thread_id][ns] = {
cp0["id"]: (serde.dumps_typed(cp0), serde.dumps_typed({}), None),
cp1["id"]: (serde.dumps_typed(cp1), serde.dumps_typed({}), cp0["id"]),
cp2["id"]: (serde.dumps_typed(cp2), serde.dumps_typed({}), cp1["id"]),
}
# Writes under cp0 produced cp1's state; writes under cp1 produced cp2's.
saver.writes[(thread_id, ns, cp0["id"])][("task1", 0)] = (
"task1",
channel,
serde.dumps_typed({"content": "first"}),
"",
)
saver.writes[(thread_id, ns, cp1["id"])][("task2", 0)] = (
"task2",
channel,
serde.dumps_typed({"content": "second"}),
"",
)
return saver, thread_id, ns
def test_fallback_returns_ancestor_writes_oldest_first(self) -> None:
saver, thread_id, ns = self._build_saver_with_chain()
target_id = "00000000000000000000000000000003.0000000000000000"
config: RunnableConfig = {
"configurable": {
"thread_id": thread_id,
"checkpoint_ns": ns,
"checkpoint_id": target_id,
}
}
result = saver._get_channel_writes_history(config, "messages")
assert result.seed is DELTA_SENTINEL
values = [v for _, _, v in result.writes]
assert values == [{"content": "first"}, {"content": "second"}]
async def test_async_fallback_returns_ancestor_writes_oldest_first(self) -> None:
saver, thread_id, ns = self._build_saver_with_chain()
target_id = "00000000000000000000000000000003.0000000000000000"
config: RunnableConfig = {
"configurable": {
"thread_id": thread_id,
"checkpoint_ns": ns,
"checkpoint_id": target_id,
}
}
result = await saver._aget_channel_writes_history(config, "messages")
assert result.seed is DELTA_SENTINEL
values = [v for _, _, v in result.writes]
assert values == [{"content": "first"}, {"content": "second"}]
async def test_async_fallback_concurrent_tasks_do_not_interfere(self) -> None:
"""Regression: the re-entrancy guard must be task-local, not thread-local.
Two concurrent `_aget_channel_writes_history` calls on the same
event-loop thread must each see their full reconstructed writes. A
`threading.local()` guard would let whichever task set it first
short-circuit the other to `writes=[]`.
"""
import asyncio
saver, thread_id, ns = self._build_saver_with_chain()
# Force the two tasks to interleave across the `set(True)` boundary:
# each `aget_tuple` yields control, so if the guard were thread-local
# the second task would observe `active=True` set by the first.
orig_aget_tuple = saver.aget_tuple
async def slow_aget_tuple(config: RunnableConfig) -> Any:
await asyncio.sleep(0)
return await orig_aget_tuple(config)
saver.aget_tuple = slow_aget_tuple # type: ignore[method-assign]
target_id = "00000000000000000000000000000003.0000000000000000"
config: RunnableConfig = {
"configurable": {
"thread_id": thread_id,
"checkpoint_ns": ns,
"checkpoint_id": target_id,
}
}
results = await asyncio.gather(
saver._aget_channel_writes_history(config, "messages"),
saver._aget_channel_writes_history(config, "messages"),
)
expected_values = [{"content": "first"}, {"content": "second"}]
for result in results:
assert result.seed is DELTA_SENTINEL
values = [v for _, _, v in result.writes]
assert values == expected_values
class TestPreDeltaBlobTerminator:
"""Verify the pre-delta blob terminator: when the ancestor walk hits a
checkpoint whose blob for the channel is a real value (not
DELTA_SENTINEL), reconstruction seeds from it and stops. This guards
* back-compat: a thread written by pre-delta code, then extended under
delta — reconstruction must return the correct value without walking
past the last pre-delta ancestor;
* perf: without the terminator, every reconstruct-after-migration would
walk all the way to the thread root.
"""
def _build_mixed_thread(self) -> tuple[InMemorySaver, str, str, str, str]:
"""Three-checkpoint chain: cp1 (pre-delta, blob=[A]), cp2 (delta,
write=B), cp3 (delta, write=C). Reconstructing at cp3 must yield
seed=[A] + writes=[B, C].
Returns `(saver, thread_id, ns, channel, cp3_id)`.
"""
saver = InMemorySaver()
serde = JsonPlusSerializer()
thread_id, ns, channel = "t1", "", "messages"
v1 = "00000000000000000000000000000001.0"
v2 = "00000000000000000000000000000002.0"
v3 = "00000000000000000000000000000003.0"
# Pre-delta: cp1 stored a real blob for the channel.
saver.blobs[(thread_id, ns, channel, v1)] = serde.dumps_typed(["A"])
# Delta-era: cp2 and cp3 store sentinels; real writes in checkpoint_writes.
saver.blobs[(thread_id, ns, channel, v2)] = serde.dumps_typed(DELTA_SENTINEL)
saver.blobs[(thread_id, ns, channel, v3)] = serde.dumps_typed(DELTA_SENTINEL)
cp1 = empty_checkpoint()
cp1["id"] = "cp1"
cp1["channel_versions"][channel] = v1
cp2 = empty_checkpoint()
cp2["id"] = "cp2"
cp2["channel_versions"][channel] = v2
cp3 = empty_checkpoint()
cp3["id"] = "cp3"
cp3["channel_versions"][channel] = v3
saver.storage[thread_id][ns] = {
"cp1": (serde.dumps_typed(cp1), serde.dumps_typed({}), None),
"cp2": (serde.dumps_typed(cp2), serde.dumps_typed({}), "cp1"),
"cp3": (serde.dumps_typed(cp3), serde.dumps_typed({}), "cp2"),
}
# Write under cp1 would be from the pre-delta era and MUST be ignored
# (the blob already captures it). We add one and assert it is not
# folded into the reconstructed result.
saver.writes[(thread_id, ns, "cp1")][("task0", 0)] = (
"task0",
channel,
serde.dumps_typed("PRE-DELTA-WRITE"),
"",
)
saver.writes[(thread_id, ns, "cp2")][("task2", 0)] = (
"task2",
channel,
serde.dumps_typed("B"),
"",
)
saver.writes[(thread_id, ns, "cp3")][("task3", 0)] = (
"task3",
channel,
serde.dumps_typed("PENDING-AT-TARGET"),
"",
)
return saver, thread_id, ns, channel, "cp3"
def test_seed_from_pre_delta_ancestor_blob(self) -> None:
saver, thread_id, ns, channel, target = self._build_mixed_thread()
config: RunnableConfig = {
"configurable": {
"thread_id": thread_id,
"checkpoint_ns": ns,
"checkpoint_id": target,
}
}
result = saver._get_channel_writes_history(config, channel)
# Seed came from the pre-delta blob at cp1.
assert result.seed == ["A"]
# Delta-era writes from cp2 replay through the reducer on top of seed.
# cp3 is the target — its own write is pending for the NEXT step and
# must be excluded.
values = [v for _, _, v in result.writes]
assert values == ["B"]
def test_pre_delta_blob_terminates_walk_before_older_writes(self) -> None:
"""Writes stored at the pre-delta ancestor itself must not be replayed
(the blob subsumes them)."""
saver, thread_id, ns, channel, target = self._build_mixed_thread()
config: RunnableConfig = {
"configurable": {
"thread_id": thread_id,
"checkpoint_ns": ns,
"checkpoint_id": target,
}
}
result = saver._get_channel_writes_history(config, channel)
values = [v for _, _, v in result.writes]
# The pre-delta write under cp1 must not appear (the blob subsumes it).
assert "PRE-DELTA-WRITE" not in values
# And the pending write at the target is never folded in.
assert "PENDING-AT-TARGET" not in values
+123 -4
View File
@@ -286,7 +286,7 @@ wheels = [
[[package]]
name = "langgraph-checkpoint"
version = "4.0.1"
version = "4.0.2"
source = { editable = "." }
dependencies = [
{ name = "langchain-core" },
@@ -369,7 +369,7 @@ test = [
[[package]]
name = "langsmith"
version = "0.6.4"
version = "0.7.31"
source = { registry = "https://pypi.org/simple" }
dependencies = [
{ name = "httpx" },
@@ -379,11 +379,12 @@ dependencies = [
{ name = "requests" },
{ name = "requests-toolbelt" },
{ name = "uuid-utils" },
{ name = "xxhash" },
{ name = "zstandard" },
]
sdist = { url = "https://files.pythonhosted.org/packages/e7/85/9c7933052a997da1b85bc5c774f3865e9b1da1c8d71541ea133178b13229/langsmith-0.6.4.tar.gz", hash = "sha256:36f7223a01c218079fbb17da5e536ebbaf5c1468c028abe070aa3ae59bc99ec8", size = 919964, upload-time = "2026-01-15T20:02:28.873Z" }
sdist = { url = "https://files.pythonhosted.org/packages/e6/11/696019490992db5c87774dc20515529ef42a01e1d770fb754ed6d9b12fb0/langsmith-0.7.31.tar.gz", hash = "sha256:331ee4f7c26bb5be4022b9859b7d7b122cbf8c9d01d9f530114c1914b0349ffb", size = 1178480, upload-time = "2026-04-14T17:55:41.242Z" }
wheels = [
{ url = "https://files.pythonhosted.org/packages/66/0f/09a6637a7ba777eb307b7c80852d9ee26438e2bdafbad6fcc849ff9d9192/langsmith-0.6.4-py3-none-any.whl", hash = "sha256:ac4835860160be371042c7adbba3cb267bcf8d96a5ea976c33a8a4acad6c5486", size = 283503, upload-time = "2026-01-15T20:02:26.662Z" },
{ url = "https://files.pythonhosted.org/packages/1d/a1/a013cf458c301cda86a213dd153ce0a01c93f1ab5833f951e6a44c9763ce/langsmith-0.7.31-py3-none-any.whl", hash = "sha256:0291d49203f6e80dda011af1afda61eb0595a4d697adb684590a8805e1d61fb6", size = 373276, upload-time = "2026-04-14T17:55:39.677Z" },
]
[[package]]
@@ -1515,6 +1516,124 @@ wheels = [
{ url = "https://files.pythonhosted.org/packages/33/e8/e40370e6d74ddba47f002a32919d91310d6074130fe4e17dabcafc15cbf1/watchdog-6.0.0-py3-none-win_ia64.whl", hash = "sha256:a1914259fa9e1454315171103c6a30961236f508b9b623eae470268bbcc6a22f", size = 79067, upload-time = "2024-11-01T14:07:11.845Z" },
]
[[package]]
name = "xxhash"
version = "3.6.0"
source = { registry = "https://pypi.org/simple" }
sdist = { url = "https://files.pythonhosted.org/packages/02/84/30869e01909fb37a6cc7e18688ee8bf1e42d57e7e0777636bd47524c43c7/xxhash-3.6.0.tar.gz", hash = "sha256:f0162a78b13a0d7617b2845b90c763339d1f1d82bb04a4b07f4ab535cc5e05d6", size = 85160, upload-time = "2025-10-02T14:37:08.097Z" }
wheels = [
{ url = "https://files.pythonhosted.org/packages/34/ee/f9f1d656ad168681bb0f6b092372c1e533c4416b8069b1896a175c46e484/xxhash-3.6.0-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:87ff03d7e35c61435976554477a7f4cd1704c3596a89a8300d5ce7fc83874a71", size = 32845, upload-time = "2025-10-02T14:33:51.573Z" },
{ url = "https://files.pythonhosted.org/packages/a3/b1/93508d9460b292c74a09b83d16750c52a0ead89c51eea9951cb97a60d959/xxhash-3.6.0-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:f572dfd3d0e2eb1a57511831cf6341242f5a9f8298a45862d085f5b93394a27d", size = 30807, upload-time = "2025-10-02T14:33:52.964Z" },
{ url = "https://files.pythonhosted.org/packages/07/55/28c93a3662f2d200c70704efe74aab9640e824f8ce330d8d3943bf7c9b3c/xxhash-3.6.0-cp310-cp310-manylinux1_i686.manylinux_2_28_i686.manylinux_2_5_i686.whl", hash = "sha256:89952ea539566b9fed2bbd94e589672794b4286f342254fad28b149f9615fef8", size = 193786, upload-time = "2025-10-02T14:33:54.272Z" },
{ url = "https://files.pythonhosted.org/packages/c1/96/fec0be9bb4b8f5d9c57d76380a366f31a1781fb802f76fc7cda6c84893c7/xxhash-3.6.0-cp310-cp310-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:48e6f2ffb07a50b52465a1032c3cf1f4a5683f944acaca8a134a2f23674c2058", size = 212830, upload-time = "2025-10-02T14:33:55.706Z" },
{ url = "https://files.pythonhosted.org/packages/c4/a0/c706845ba77b9611f81fd2e93fad9859346b026e8445e76f8c6fd057cc6d/xxhash-3.6.0-cp310-cp310-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:b5b848ad6c16d308c3ac7ad4ba6bede80ed5df2ba8ed382f8932df63158dd4b2", size = 211606, upload-time = "2025-10-02T14:33:57.133Z" },
{ url = "https://files.pythonhosted.org/packages/67/1e/164126a2999e5045f04a69257eea946c0dc3e86541b400d4385d646b53d7/xxhash-3.6.0-cp310-cp310-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:a034590a727b44dd8ac5914236a7b8504144447a9682586c3327e935f33ec8cc", size = 444872, upload-time = "2025-10-02T14:33:58.446Z" },
{ url = "https://files.pythonhosted.org/packages/2d/4b/55ab404c56cd70a2cf5ecfe484838865d0fea5627365c6c8ca156bd09c8f/xxhash-3.6.0-cp310-cp310-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:8a8f1972e75ebdd161d7896743122834fe87378160c20e97f8b09166213bf8cc", size = 193217, upload-time = "2025-10-02T14:33:59.724Z" },
{ url = "https://files.pythonhosted.org/packages/45/e6/52abf06bac316db33aa269091ae7311bd53cfc6f4b120ae77bac1b348091/xxhash-3.6.0-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:ee34327b187f002a596d7b167ebc59a1b729e963ce645964bbc050d2f1b73d07", size = 210139, upload-time = "2025-10-02T14:34:02.041Z" },
{ url = "https://files.pythonhosted.org/packages/34/37/db94d490b8691236d356bc249c08819cbcef9273a1a30acf1254ff9ce157/xxhash-3.6.0-cp310-cp310-musllinux_1_2_i686.whl", hash = "sha256:339f518c3c7a850dd033ab416ea25a692759dc7478a71131fe8869010d2b75e4", size = 197669, upload-time = "2025-10-02T14:34:03.664Z" },
{ url = "https://files.pythonhosted.org/packages/b7/36/c4f219ef4a17a4f7a64ed3569bc2b5a9c8311abdb22249ac96093625b1a4/xxhash-3.6.0-cp310-cp310-musllinux_1_2_ppc64le.whl", hash = "sha256:bf48889c9630542d4709192578aebbd836177c9f7a4a2778a7d6340107c65f06", size = 210018, upload-time = "2025-10-02T14:34:05.325Z" },
{ url = "https://files.pythonhosted.org/packages/fd/06/bfac889a374fc2fc439a69223d1750eed2e18a7db8514737ab630534fa08/xxhash-3.6.0-cp310-cp310-musllinux_1_2_s390x.whl", hash = "sha256:5576b002a56207f640636056b4160a378fe36a58db73ae5c27a7ec8db35f71d4", size = 413058, upload-time = "2025-10-02T14:34:06.925Z" },
{ url = "https://files.pythonhosted.org/packages/c9/d1/555d8447e0dd32ad0930a249a522bb2e289f0d08b6b16204cfa42c1f5a0c/xxhash-3.6.0-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:af1f3278bd02814d6dedc5dec397993b549d6f16c19379721e5a1d31e132c49b", size = 190628, upload-time = "2025-10-02T14:34:08.669Z" },
{ url = "https://files.pythonhosted.org/packages/d1/15/8751330b5186cedc4ed4b597989882ea05e0408b53fa47bcb46a6125bfc6/xxhash-3.6.0-cp310-cp310-win32.whl", hash = "sha256:aed058764db109dc9052720da65fafe84873b05eb8b07e5e653597951af57c3b", size = 30577, upload-time = "2025-10-02T14:34:10.234Z" },
{ url = "https://files.pythonhosted.org/packages/bb/cc/53f87e8b5871a6eb2ff7e89c48c66093bda2be52315a8161ddc54ea550c4/xxhash-3.6.0-cp310-cp310-win_amd64.whl", hash = "sha256:e82da5670f2d0d98950317f82a0e4a0197150ff19a6df2ba40399c2a3b9ae5fb", size = 31487, upload-time = "2025-10-02T14:34:11.618Z" },
{ url = "https://files.pythonhosted.org/packages/9f/00/60f9ea3bb697667a14314d7269956f58bf56bb73864f8f8d52a3c2535e9a/xxhash-3.6.0-cp310-cp310-win_arm64.whl", hash = "sha256:4a082ffff8c6ac07707fb6b671caf7c6e020c75226c561830b73d862060f281d", size = 27863, upload-time = "2025-10-02T14:34:12.619Z" },
{ url = "https://files.pythonhosted.org/packages/17/d4/cc2f0400e9154df4b9964249da78ebd72f318e35ccc425e9f403c392f22a/xxhash-3.6.0-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:b47bbd8cf2d72797f3c2772eaaac0ded3d3af26481a26d7d7d41dc2d3c46b04a", size = 32844, upload-time = "2025-10-02T14:34:14.037Z" },
{ url = "https://files.pythonhosted.org/packages/5e/ec/1cc11cd13e26ea8bc3cb4af4eaadd8d46d5014aebb67be3f71fb0b68802a/xxhash-3.6.0-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:2b6821e94346f96db75abaa6e255706fb06ebd530899ed76d32cd99f20dc52fa", size = 30809, upload-time = "2025-10-02T14:34:15.484Z" },
{ url = "https://files.pythonhosted.org/packages/04/5f/19fe357ea348d98ca22f456f75a30ac0916b51c753e1f8b2e0e6fb884cce/xxhash-3.6.0-cp311-cp311-manylinux1_i686.manylinux_2_28_i686.manylinux_2_5_i686.whl", hash = "sha256:d0a9751f71a1a65ce3584e9cae4467651c7e70c9d31017fa57574583a4540248", size = 194665, upload-time = "2025-10-02T14:34:16.541Z" },
{ url = "https://files.pythonhosted.org/packages/90/3b/d1f1a8f5442a5fd8beedae110c5af7604dc37349a8e16519c13c19a9a2de/xxhash-3.6.0-cp311-cp311-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:8b29ee68625ab37b04c0b40c3fafdf24d2f75ccd778333cfb698f65f6c463f62", size = 213550, upload-time = "2025-10-02T14:34:17.878Z" },
{ url = "https://files.pythonhosted.org/packages/c4/ef/3a9b05eb527457d5db13a135a2ae1a26c80fecd624d20f3e8dcc4cb170f3/xxhash-3.6.0-cp311-cp311-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:6812c25fe0d6c36a46ccb002f40f27ac903bf18af9f6dd8f9669cb4d176ab18f", size = 212384, upload-time = "2025-10-02T14:34:19.182Z" },
{ url = "https://files.pythonhosted.org/packages/0f/18/ccc194ee698c6c623acbf0f8c2969811a8a4b6185af5e824cd27b9e4fd3e/xxhash-3.6.0-cp311-cp311-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:4ccbff013972390b51a18ef1255ef5ac125c92dc9143b2d1909f59abc765540e", size = 445749, upload-time = "2025-10-02T14:34:20.659Z" },
{ url = "https://files.pythonhosted.org/packages/a5/86/cf2c0321dc3940a7aa73076f4fd677a0fb3e405cb297ead7d864fd90847e/xxhash-3.6.0-cp311-cp311-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:297b7fbf86c82c550e12e8fb71968b3f033d27b874276ba3624ea868c11165a8", size = 193880, upload-time = "2025-10-02T14:34:22.431Z" },
{ url = "https://files.pythonhosted.org/packages/82/fb/96213c8560e6f948a1ecc9a7613f8032b19ee45f747f4fca4eb31bb6d6ed/xxhash-3.6.0-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:dea26ae1eb293db089798d3973a5fc928a18fdd97cc8801226fae705b02b14b0", size = 210912, upload-time = "2025-10-02T14:34:23.937Z" },
{ url = "https://files.pythonhosted.org/packages/40/aa/4395e669b0606a096d6788f40dbdf2b819d6773aa290c19e6e83cbfc312f/xxhash-3.6.0-cp311-cp311-musllinux_1_2_i686.whl", hash = "sha256:7a0b169aafb98f4284f73635a8e93f0735f9cbde17bd5ec332480484241aaa77", size = 198654, upload-time = "2025-10-02T14:34:25.644Z" },
{ url = "https://files.pythonhosted.org/packages/67/74/b044fcd6b3d89e9b1b665924d85d3f400636c23590226feb1eb09e1176ce/xxhash-3.6.0-cp311-cp311-musllinux_1_2_ppc64le.whl", hash = "sha256:08d45aef063a4531b785cd72de4887766d01dc8f362a515693df349fdb825e0c", size = 210867, upload-time = "2025-10-02T14:34:27.203Z" },
{ url = "https://files.pythonhosted.org/packages/bc/fd/3ce73bf753b08cb19daee1eb14aa0d7fe331f8da9c02dd95316ddfe5275e/xxhash-3.6.0-cp311-cp311-musllinux_1_2_s390x.whl", hash = "sha256:929142361a48ee07f09121fe9e96a84950e8d4df3bb298ca5d88061969f34d7b", size = 414012, upload-time = "2025-10-02T14:34:28.409Z" },
{ url = "https://files.pythonhosted.org/packages/ba/b3/5a4241309217c5c876f156b10778f3ab3af7ba7e3259e6d5f5c7d0129eb2/xxhash-3.6.0-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:51312c768403d8540487dbbfb557454cfc55589bbde6424456951f7fcd4facb3", size = 191409, upload-time = "2025-10-02T14:34:29.696Z" },
{ url = "https://files.pythonhosted.org/packages/c0/01/99bfbc15fb9abb9a72b088c1d95219fc4782b7d01fc835bd5744d66dd0b8/xxhash-3.6.0-cp311-cp311-win32.whl", hash = "sha256:d1927a69feddc24c987b337ce81ac15c4720955b667fe9b588e02254b80446fd", size = 30574, upload-time = "2025-10-02T14:34:31.028Z" },
{ url = "https://files.pythonhosted.org/packages/65/79/9d24d7f53819fe301b231044ea362ce64e86c74f6e8c8e51320de248b3e5/xxhash-3.6.0-cp311-cp311-win_amd64.whl", hash = "sha256:26734cdc2d4ffe449b41d186bbeac416f704a482ed835d375a5c0cb02bc63fef", size = 31481, upload-time = "2025-10-02T14:34:32.062Z" },
{ url = "https://files.pythonhosted.org/packages/30/4e/15cd0e3e8772071344eab2961ce83f6e485111fed8beb491a3f1ce100270/xxhash-3.6.0-cp311-cp311-win_arm64.whl", hash = "sha256:d72f67ef8bf36e05f5b6c65e8524f265bd61071471cd4cf1d36743ebeeeb06b7", size = 27861, upload-time = "2025-10-02T14:34:33.555Z" },
{ url = "https://files.pythonhosted.org/packages/9a/07/d9412f3d7d462347e4511181dea65e47e0d0e16e26fbee2ea86a2aefb657/xxhash-3.6.0-cp312-cp312-macosx_10_13_x86_64.whl", hash = "sha256:01362c4331775398e7bb34e3ab403bc9ee9f7c497bc7dee6272114055277dd3c", size = 32744, upload-time = "2025-10-02T14:34:34.622Z" },
{ url = "https://files.pythonhosted.org/packages/79/35/0429ee11d035fc33abe32dca1b2b69e8c18d236547b9a9b72c1929189b9a/xxhash-3.6.0-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:b7b2df81a23f8cb99656378e72501b2cb41b1827c0f5a86f87d6b06b69f9f204", size = 30816, upload-time = "2025-10-02T14:34:36.043Z" },
{ url = "https://files.pythonhosted.org/packages/b7/f2/57eb99aa0f7d98624c0932c5b9a170e1806406cdbcdb510546634a1359e0/xxhash-3.6.0-cp312-cp312-manylinux1_i686.manylinux_2_28_i686.manylinux_2_5_i686.whl", hash = "sha256:dc94790144e66b14f67b10ac8ed75b39ca47536bf8800eb7c24b50271ea0c490", size = 194035, upload-time = "2025-10-02T14:34:37.354Z" },
{ url = "https://files.pythonhosted.org/packages/4c/ed/6224ba353690d73af7a3f1c7cdb1fc1b002e38f783cb991ae338e1eb3d79/xxhash-3.6.0-cp312-cp312-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:93f107c673bccf0d592cdba077dedaf52fe7f42dcd7676eba1f6d6f0c3efffd2", size = 212914, upload-time = "2025-10-02T14:34:38.6Z" },
{ url = "https://files.pythonhosted.org/packages/38/86/fb6b6130d8dd6b8942cc17ab4d90e223653a89aa32ad2776f8af7064ed13/xxhash-3.6.0-cp312-cp312-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:2aa5ee3444c25b69813663c9f8067dcfaa2e126dc55e8dddf40f4d1c25d7effa", size = 212163, upload-time = "2025-10-02T14:34:39.872Z" },
{ url = "https://files.pythonhosted.org/packages/ee/dc/e84875682b0593e884ad73b2d40767b5790d417bde603cceb6878901d647/xxhash-3.6.0-cp312-cp312-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:f7f99123f0e1194fa59cc69ad46dbae2e07becec5df50a0509a808f90a0f03f0", size = 445411, upload-time = "2025-10-02T14:34:41.569Z" },
{ url = "https://files.pythonhosted.org/packages/11/4f/426f91b96701ec2f37bb2b8cec664eff4f658a11f3fa9d94f0a887ea6d2b/xxhash-3.6.0-cp312-cp312-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:49e03e6fe2cac4a1bc64952dd250cf0dbc5ef4ebb7b8d96bce82e2de163c82a2", size = 193883, upload-time = "2025-10-02T14:34:43.249Z" },
{ url = "https://files.pythonhosted.org/packages/53/5a/ddbb83eee8e28b778eacfc5a85c969673e4023cdeedcfcef61f36731610b/xxhash-3.6.0-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:bd17fede52a17a4f9a7bc4472a5867cb0b160deeb431795c0e4abe158bc784e9", size = 210392, upload-time = "2025-10-02T14:34:45.042Z" },
{ url = "https://files.pythonhosted.org/packages/1e/c2/ff69efd07c8c074ccdf0a4f36fcdd3d27363665bcdf4ba399abebe643465/xxhash-3.6.0-cp312-cp312-musllinux_1_2_i686.whl", hash = "sha256:6fb5f5476bef678f69db04f2bd1efbed3030d2aba305b0fc1773645f187d6a4e", size = 197898, upload-time = "2025-10-02T14:34:46.302Z" },
{ url = "https://files.pythonhosted.org/packages/58/ca/faa05ac19b3b622c7c9317ac3e23954187516298a091eb02c976d0d3dd45/xxhash-3.6.0-cp312-cp312-musllinux_1_2_ppc64le.whl", hash = "sha256:843b52f6d88071f87eba1631b684fcb4b2068cd2180a0224122fe4ef011a9374", size = 210655, upload-time = "2025-10-02T14:34:47.571Z" },
{ url = "https://files.pythonhosted.org/packages/d4/7a/06aa7482345480cc0cb597f5c875b11a82c3953f534394f620b0be2f700c/xxhash-3.6.0-cp312-cp312-musllinux_1_2_s390x.whl", hash = "sha256:7d14a6cfaf03b1b6f5f9790f76880601ccc7896aff7ab9cd8978a939c1eb7e0d", size = 414001, upload-time = "2025-10-02T14:34:49.273Z" },
{ url = "https://files.pythonhosted.org/packages/23/07/63ffb386cd47029aa2916b3d2f454e6cc5b9f5c5ada3790377d5430084e7/xxhash-3.6.0-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:418daf3db71e1413cfe211c2f9a528456936645c17f46b5204705581a45390ae", size = 191431, upload-time = "2025-10-02T14:34:50.798Z" },
{ url = "https://files.pythonhosted.org/packages/0f/93/14fde614cadb4ddf5e7cebf8918b7e8fac5ae7861c1875964f17e678205c/xxhash-3.6.0-cp312-cp312-win32.whl", hash = "sha256:50fc255f39428a27299c20e280d6193d8b63b8ef8028995323bf834a026b4fbb", size = 30617, upload-time = "2025-10-02T14:34:51.954Z" },
{ url = "https://files.pythonhosted.org/packages/13/5d/0d125536cbe7565a83d06e43783389ecae0c0f2ed037b48ede185de477c0/xxhash-3.6.0-cp312-cp312-win_amd64.whl", hash = "sha256:c0f2ab8c715630565ab8991b536ecded9416d615538be8ecddce43ccf26cbc7c", size = 31534, upload-time = "2025-10-02T14:34:53.276Z" },
{ url = "https://files.pythonhosted.org/packages/54/85/6ec269b0952ec7e36ba019125982cf11d91256a778c7c3f98a4c5043d283/xxhash-3.6.0-cp312-cp312-win_arm64.whl", hash = "sha256:eae5c13f3bc455a3bbb68bdc513912dc7356de7e2280363ea235f71f54064829", size = 27876, upload-time = "2025-10-02T14:34:54.371Z" },
{ url = "https://files.pythonhosted.org/packages/33/76/35d05267ac82f53ae9b0e554da7c5e281ee61f3cad44c743f0fcd354f211/xxhash-3.6.0-cp313-cp313-macosx_10_13_x86_64.whl", hash = "sha256:599e64ba7f67472481ceb6ee80fa3bd828fd61ba59fb11475572cc5ee52b89ec", size = 32738, upload-time = "2025-10-02T14:34:55.839Z" },
{ url = "https://files.pythonhosted.org/packages/31/a8/3fbce1cd96534a95e35d5120637bf29b0d7f5d8fa2f6374e31b4156dd419/xxhash-3.6.0-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:7d8b8aaa30fca4f16f0c84a5c8d7ddee0e25250ec2796c973775373257dde8f1", size = 30821, upload-time = "2025-10-02T14:34:57.219Z" },
{ url = "https://files.pythonhosted.org/packages/0c/ea/d387530ca7ecfa183cb358027f1833297c6ac6098223fd14f9782cd0015c/xxhash-3.6.0-cp313-cp313-manylinux1_i686.manylinux_2_28_i686.manylinux_2_5_i686.whl", hash = "sha256:d597acf8506d6e7101a4a44a5e428977a51c0fadbbfd3c39650cca9253f6e5a6", size = 194127, upload-time = "2025-10-02T14:34:59.21Z" },
{ url = "https://files.pythonhosted.org/packages/ba/0c/71435dcb99874b09a43b8d7c54071e600a7481e42b3e3ce1eb5226a5711a/xxhash-3.6.0-cp313-cp313-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:858dc935963a33bc33490128edc1c12b0c14d9c7ebaa4e387a7869ecc4f3e263", size = 212975, upload-time = "2025-10-02T14:35:00.816Z" },
{ url = "https://files.pythonhosted.org/packages/84/7a/c2b3d071e4bb4a90b7057228a99b10d51744878f4a8a6dd643c8bd897620/xxhash-3.6.0-cp313-cp313-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:ba284920194615cb8edf73bf52236ce2e1664ccd4a38fdb543506413529cc546", size = 212241, upload-time = "2025-10-02T14:35:02.207Z" },
{ url = "https://files.pythonhosted.org/packages/81/5f/640b6eac0128e215f177df99eadcd0f1b7c42c274ab6a394a05059694c5a/xxhash-3.6.0-cp313-cp313-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:4b54219177f6c6674d5378bd862c6aedf64725f70dd29c472eaae154df1a2e89", size = 445471, upload-time = "2025-10-02T14:35:03.61Z" },
{ url = "https://files.pythonhosted.org/packages/5e/1e/3c3d3ef071b051cc3abbe3721ffb8365033a172613c04af2da89d5548a87/xxhash-3.6.0-cp313-cp313-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:42c36dd7dbad2f5238950c377fcbf6811b1cdb1c444fab447960030cea60504d", size = 193936, upload-time = "2025-10-02T14:35:05.013Z" },
{ url = "https://files.pythonhosted.org/packages/2c/bd/4a5f68381939219abfe1c22a9e3a5854a4f6f6f3c4983a87d255f21f2e5d/xxhash-3.6.0-cp313-cp313-musllinux_1_2_aarch64.whl", hash = "sha256:f22927652cba98c44639ffdc7aaf35828dccf679b10b31c4ad72a5b530a18eb7", size = 210440, upload-time = "2025-10-02T14:35:06.239Z" },
{ url = "https://files.pythonhosted.org/packages/eb/37/b80fe3d5cfb9faff01a02121a0f4d565eb7237e9e5fc66e73017e74dcd36/xxhash-3.6.0-cp313-cp313-musllinux_1_2_i686.whl", hash = "sha256:b45fad44d9c5c119e9c6fbf2e1c656a46dc68e280275007bbfd3d572b21426db", size = 197990, upload-time = "2025-10-02T14:35:07.735Z" },
{ url = "https://files.pythonhosted.org/packages/d7/fd/2c0a00c97b9e18f72e1f240ad4e8f8a90fd9d408289ba9c7c495ed7dc05c/xxhash-3.6.0-cp313-cp313-musllinux_1_2_ppc64le.whl", hash = "sha256:6f2580ffab1a8b68ef2b901cde7e55fa8da5e4be0977c68f78fc80f3c143de42", size = 210689, upload-time = "2025-10-02T14:35:09.438Z" },
{ url = "https://files.pythonhosted.org/packages/93/86/5dd8076a926b9a95db3206aba20d89a7fc14dd5aac16e5c4de4b56033140/xxhash-3.6.0-cp313-cp313-musllinux_1_2_s390x.whl", hash = "sha256:40c391dd3cd041ebc3ffe6f2c862f402e306eb571422e0aa918d8070ba31da11", size = 414068, upload-time = "2025-10-02T14:35:11.162Z" },
{ url = "https://files.pythonhosted.org/packages/af/3c/0bb129170ee8f3650f08e993baee550a09593462a5cddd8e44d0011102b1/xxhash-3.6.0-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:f205badabde7aafd1a31e8ca2a3e5a763107a71c397c4481d6a804eb5063d8bd", size = 191495, upload-time = "2025-10-02T14:35:12.971Z" },
{ url = "https://files.pythonhosted.org/packages/e9/3a/6797e0114c21d1725e2577508e24006fd7ff1d8c0c502d3b52e45c1771d8/xxhash-3.6.0-cp313-cp313-win32.whl", hash = "sha256:2577b276e060b73b73a53042ea5bd5203d3e6347ce0d09f98500f418a9fcf799", size = 30620, upload-time = "2025-10-02T14:35:14.129Z" },
{ url = "https://files.pythonhosted.org/packages/86/15/9bc32671e9a38b413a76d24722a2bf8784a132c043063a8f5152d390b0f9/xxhash-3.6.0-cp313-cp313-win_amd64.whl", hash = "sha256:757320d45d2fbcce8f30c42a6b2f47862967aea7bf458b9625b4bbe7ee390392", size = 31542, upload-time = "2025-10-02T14:35:15.21Z" },
{ url = "https://files.pythonhosted.org/packages/39/c5/cc01e4f6188656e56112d6a8e0dfe298a16934b8c47a247236549a3f7695/xxhash-3.6.0-cp313-cp313-win_arm64.whl", hash = "sha256:457b8f85dec5825eed7b69c11ae86834a018b8e3df5e77783c999663da2f96d6", size = 27880, upload-time = "2025-10-02T14:35:16.315Z" },
{ url = "https://files.pythonhosted.org/packages/f3/30/25e5321c8732759e930c555176d37e24ab84365482d257c3b16362235212/xxhash-3.6.0-cp313-cp313t-macosx_10_13_x86_64.whl", hash = "sha256:a42e633d75cdad6d625434e3468126c73f13f7584545a9cf34e883aa1710e702", size = 32956, upload-time = "2025-10-02T14:35:17.413Z" },
{ url = "https://files.pythonhosted.org/packages/9f/3c/0573299560d7d9f8ab1838f1efc021a280b5ae5ae2e849034ef3dee18810/xxhash-3.6.0-cp313-cp313t-macosx_11_0_arm64.whl", hash = "sha256:568a6d743219e717b07b4e03b0a828ce593833e498c3b64752e0f5df6bfe84db", size = 31072, upload-time = "2025-10-02T14:35:18.844Z" },
{ url = "https://files.pythonhosted.org/packages/7a/1c/52d83a06e417cd9d4137722693424885cc9878249beb3a7c829e74bf7ce9/xxhash-3.6.0-cp313-cp313t-manylinux1_i686.manylinux_2_28_i686.manylinux_2_5_i686.whl", hash = "sha256:bec91b562d8012dae276af8025a55811b875baace6af510412a5e58e3121bc54", size = 196409, upload-time = "2025-10-02T14:35:20.31Z" },
{ url = "https://files.pythonhosted.org/packages/e3/8e/c6d158d12a79bbd0b878f8355432075fc82759e356ab5a111463422a239b/xxhash-3.6.0-cp313-cp313t-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:78e7f2f4c521c30ad5e786fdd6bae89d47a32672a80195467b5de0480aa97b1f", size = 215736, upload-time = "2025-10-02T14:35:21.616Z" },
{ url = "https://files.pythonhosted.org/packages/bc/68/c4c80614716345d55071a396cf03d06e34b5f4917a467faf43083c995155/xxhash-3.6.0-cp313-cp313t-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:3ed0df1b11a79856df5ffcab572cbd6b9627034c1c748c5566fa79df9048a7c5", size = 214833, upload-time = "2025-10-02T14:35:23.32Z" },
{ url = "https://files.pythonhosted.org/packages/7e/e9/ae27c8ffec8b953efa84c7c4a6c6802c263d587b9fc0d6e7cea64e08c3af/xxhash-3.6.0-cp313-cp313t-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:0e4edbfc7d420925b0dd5e792478ed393d6e75ff8fc219a6546fb446b6a417b1", size = 448348, upload-time = "2025-10-02T14:35:25.111Z" },
{ url = "https://files.pythonhosted.org/packages/d7/6b/33e21afb1b5b3f46b74b6bd1913639066af218d704cc0941404ca717fc57/xxhash-3.6.0-cp313-cp313t-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:fba27a198363a7ef87f8c0f6b171ec36b674fe9053742c58dd7e3201c1ab30ee", size = 196070, upload-time = "2025-10-02T14:35:26.586Z" },
{ url = "https://files.pythonhosted.org/packages/96/b6/fcabd337bc5fa624e7203aa0fa7d0c49eed22f72e93229431752bddc83d9/xxhash-3.6.0-cp313-cp313t-musllinux_1_2_aarch64.whl", hash = "sha256:794fe9145fe60191c6532fa95063765529770edcdd67b3d537793e8004cabbfd", size = 212907, upload-time = "2025-10-02T14:35:28.087Z" },
{ url = "https://files.pythonhosted.org/packages/4b/d3/9ee6160e644d660fcf176c5825e61411c7f62648728f69c79ba237250143/xxhash-3.6.0-cp313-cp313t-musllinux_1_2_i686.whl", hash = "sha256:6105ef7e62b5ac73a837778efc331a591d8442f8ef5c7e102376506cb4ae2729", size = 200839, upload-time = "2025-10-02T14:35:29.857Z" },
{ url = "https://files.pythonhosted.org/packages/0d/98/e8de5baa5109394baf5118f5e72ab21a86387c4f89b0e77ef3e2f6b0327b/xxhash-3.6.0-cp313-cp313t-musllinux_1_2_ppc64le.whl", hash = "sha256:f01375c0e55395b814a679b3eea205db7919ac2af213f4a6682e01220e5fe292", size = 213304, upload-time = "2025-10-02T14:35:31.222Z" },
{ url = "https://files.pythonhosted.org/packages/7b/1d/71056535dec5c3177eeb53e38e3d367dd1d16e024e63b1cee208d572a033/xxhash-3.6.0-cp313-cp313t-musllinux_1_2_s390x.whl", hash = "sha256:d706dca2d24d834a4661619dcacf51a75c16d65985718d6a7d73c1eeeb903ddf", size = 416930, upload-time = "2025-10-02T14:35:32.517Z" },
{ url = "https://files.pythonhosted.org/packages/dc/6c/5cbde9de2cd967c322e651c65c543700b19e7ae3e0aae8ece3469bf9683d/xxhash-3.6.0-cp313-cp313t-musllinux_1_2_x86_64.whl", hash = "sha256:5f059d9faeacd49c0215d66f4056e1326c80503f51a1532ca336a385edadd033", size = 193787, upload-time = "2025-10-02T14:35:33.827Z" },
{ url = "https://files.pythonhosted.org/packages/19/fa/0172e350361d61febcea941b0cc541d6e6c8d65d153e85f850a7b256ff8a/xxhash-3.6.0-cp313-cp313t-win32.whl", hash = "sha256:1244460adc3a9be84731d72b8e80625788e5815b68da3da8b83f78115a40a7ec", size = 30916, upload-time = "2025-10-02T14:35:35.107Z" },
{ url = "https://files.pythonhosted.org/packages/ad/e6/e8cf858a2b19d6d45820f072eff1bea413910592ff17157cabc5f1227a16/xxhash-3.6.0-cp313-cp313t-win_amd64.whl", hash = "sha256:b1e420ef35c503869c4064f4a2f2b08ad6431ab7b229a05cce39d74268bca6b8", size = 31799, upload-time = "2025-10-02T14:35:36.165Z" },
{ url = "https://files.pythonhosted.org/packages/56/15/064b197e855bfb7b343210e82490ae672f8bc7cdf3ddb02e92f64304ee8a/xxhash-3.6.0-cp313-cp313t-win_arm64.whl", hash = "sha256:ec44b73a4220623235f67a996c862049f375df3b1052d9899f40a6382c32d746", size = 28044, upload-time = "2025-10-02T14:35:37.195Z" },
{ url = "https://files.pythonhosted.org/packages/7e/5e/0138bc4484ea9b897864d59fce9be9086030825bc778b76cb5a33a906d37/xxhash-3.6.0-cp314-cp314-macosx_10_13_x86_64.whl", hash = "sha256:a40a3d35b204b7cc7643cbcf8c9976d818cb47befcfac8bbefec8038ac363f3e", size = 32754, upload-time = "2025-10-02T14:35:38.245Z" },
{ url = "https://files.pythonhosted.org/packages/18/d7/5dac2eb2ec75fd771957a13e5dda560efb2176d5203f39502a5fc571f899/xxhash-3.6.0-cp314-cp314-macosx_11_0_arm64.whl", hash = "sha256:a54844be970d3fc22630b32d515e79a90d0a3ddb2644d8d7402e3c4c8da61405", size = 30846, upload-time = "2025-10-02T14:35:39.6Z" },
{ url = "https://files.pythonhosted.org/packages/fe/71/8bc5be2bb00deb5682e92e8da955ebe5fa982da13a69da5a40a4c8db12fb/xxhash-3.6.0-cp314-cp314-manylinux1_i686.manylinux_2_28_i686.manylinux_2_5_i686.whl", hash = "sha256:016e9190af8f0a4e3741343777710e3d5717427f175adfdc3e72508f59e2a7f3", size = 194343, upload-time = "2025-10-02T14:35:40.69Z" },
{ url = "https://files.pythonhosted.org/packages/e7/3b/52badfb2aecec2c377ddf1ae75f55db3ba2d321c5e164f14461c90837ef3/xxhash-3.6.0-cp314-cp314-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:4f6f72232f849eb9d0141e2ebe2677ece15adfd0fa599bc058aad83c714bb2c6", size = 213074, upload-time = "2025-10-02T14:35:42.29Z" },
{ url = "https://files.pythonhosted.org/packages/a2/2b/ae46b4e9b92e537fa30d03dbc19cdae57ed407e9c26d163895e968e3de85/xxhash-3.6.0-cp314-cp314-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:63275a8aba7865e44b1813d2177e0f5ea7eadad3dd063a21f7cf9afdc7054063", size = 212388, upload-time = "2025-10-02T14:35:43.929Z" },
{ url = "https://files.pythonhosted.org/packages/f5/80/49f88d3afc724b4ac7fbd664c8452d6db51b49915be48c6982659e0e7942/xxhash-3.6.0-cp314-cp314-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:3cd01fa2aa00d8b017c97eb46b9a794fbdca53fc14f845f5a328c71254b0abb7", size = 445614, upload-time = "2025-10-02T14:35:45.216Z" },
{ url = "https://files.pythonhosted.org/packages/ed/ba/603ce3961e339413543d8cd44f21f2c80e2a7c5cfe692a7b1f2cccf58f3c/xxhash-3.6.0-cp314-cp314-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:0226aa89035b62b6a86d3c68df4d7c1f47a342b8683da2b60cedcddb46c4d95b", size = 194024, upload-time = "2025-10-02T14:35:46.959Z" },
{ url = "https://files.pythonhosted.org/packages/78/d1/8e225ff7113bf81545cfdcd79eef124a7b7064a0bba53605ff39590b95c2/xxhash-3.6.0-cp314-cp314-musllinux_1_2_aarch64.whl", hash = "sha256:c6e193e9f56e4ca4923c61238cdaced324f0feac782544eb4c6d55ad5cc99ddd", size = 210541, upload-time = "2025-10-02T14:35:48.301Z" },
{ url = "https://files.pythonhosted.org/packages/6f/58/0f89d149f0bad89def1a8dd38feb50ccdeb643d9797ec84707091d4cb494/xxhash-3.6.0-cp314-cp314-musllinux_1_2_i686.whl", hash = "sha256:9176dcaddf4ca963d4deb93866d739a343c01c969231dbe21680e13a5d1a5bf0", size = 198305, upload-time = "2025-10-02T14:35:49.584Z" },
{ url = "https://files.pythonhosted.org/packages/11/38/5eab81580703c4df93feb5f32ff8fa7fe1e2c51c1f183ee4e48d4bb9d3d7/xxhash-3.6.0-cp314-cp314-musllinux_1_2_ppc64le.whl", hash = "sha256:c1ce4009c97a752e682b897aa99aef84191077a9433eb237774689f14f8ec152", size = 210848, upload-time = "2025-10-02T14:35:50.877Z" },
{ url = "https://files.pythonhosted.org/packages/5e/6b/953dc4b05c3ce678abca756416e4c130d2382f877a9c30a20d08ee6a77c0/xxhash-3.6.0-cp314-cp314-musllinux_1_2_s390x.whl", hash = "sha256:8cb2f4f679b01513b7adbb9b1b2f0f9cdc31b70007eaf9d59d0878809f385b11", size = 414142, upload-time = "2025-10-02T14:35:52.15Z" },
{ url = "https://files.pythonhosted.org/packages/08/a9/238ec0d4e81a10eb5026d4a6972677cbc898ba6c8b9dbaec12ae001b1b35/xxhash-3.6.0-cp314-cp314-musllinux_1_2_x86_64.whl", hash = "sha256:653a91d7c2ab54a92c19ccf43508b6a555440b9be1bc8be553376778be7f20b5", size = 191547, upload-time = "2025-10-02T14:35:53.547Z" },
{ url = "https://files.pythonhosted.org/packages/f1/ee/3cf8589e06c2164ac77c3bf0aa127012801128f1feebf2a079272da5737c/xxhash-3.6.0-cp314-cp314-win32.whl", hash = "sha256:a756fe893389483ee8c394d06b5ab765d96e68fbbfe6fde7aa17e11f5720559f", size = 31214, upload-time = "2025-10-02T14:35:54.746Z" },
{ url = "https://files.pythonhosted.org/packages/02/5d/a19552fbc6ad4cb54ff953c3908bbc095f4a921bc569433d791f755186f1/xxhash-3.6.0-cp314-cp314-win_amd64.whl", hash = "sha256:39be8e4e142550ef69629c9cd71b88c90e9a5db703fecbcf265546d9536ca4ad", size = 32290, upload-time = "2025-10-02T14:35:55.791Z" },
{ url = "https://files.pythonhosted.org/packages/b1/11/dafa0643bc30442c887b55baf8e73353a344ee89c1901b5a5c54a6c17d39/xxhash-3.6.0-cp314-cp314-win_arm64.whl", hash = "sha256:25915e6000338999236f1eb68a02a32c3275ac338628a7eaa5a269c401995679", size = 28795, upload-time = "2025-10-02T14:35:57.162Z" },
{ url = "https://files.pythonhosted.org/packages/2c/db/0e99732ed7f64182aef4a6fb145e1a295558deec2a746265dcdec12d191e/xxhash-3.6.0-cp314-cp314t-macosx_10_13_x86_64.whl", hash = "sha256:c5294f596a9017ca5a3e3f8884c00b91ab2ad2933cf288f4923c3fd4346cf3d4", size = 32955, upload-time = "2025-10-02T14:35:58.267Z" },
{ url = "https://files.pythonhosted.org/packages/55/f4/2a7c3c68e564a099becfa44bb3d398810cc0ff6749b0d3cb8ccb93f23c14/xxhash-3.6.0-cp314-cp314t-macosx_11_0_arm64.whl", hash = "sha256:1cf9dcc4ab9cff01dfbba78544297a3a01dafd60f3bde4e2bfd016cf7e4ddc67", size = 31072, upload-time = "2025-10-02T14:35:59.382Z" },
{ url = "https://files.pythonhosted.org/packages/c6/d9/72a29cddc7250e8a5819dad5d466facb5dc4c802ce120645630149127e73/xxhash-3.6.0-cp314-cp314t-manylinux1_i686.manylinux_2_28_i686.manylinux_2_5_i686.whl", hash = "sha256:01262da8798422d0685f7cef03b2bd3f4f46511b02830861df548d7def4402ad", size = 196579, upload-time = "2025-10-02T14:36:00.838Z" },
{ url = "https://files.pythonhosted.org/packages/63/93/b21590e1e381040e2ca305a884d89e1c345b347404f7780f07f2cdd47ef4/xxhash-3.6.0-cp314-cp314t-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:51a73fb7cb3a3ead9f7a8b583ffd9b8038e277cdb8cb87cf890e88b3456afa0b", size = 215854, upload-time = "2025-10-02T14:36:02.207Z" },
{ url = "https://files.pythonhosted.org/packages/ce/b8/edab8a7d4fa14e924b29be877d54155dcbd8b80be85ea00d2be3413a9ed4/xxhash-3.6.0-cp314-cp314t-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:b9c6df83594f7df8f7f708ce5ebeacfc69f72c9fbaaababf6cf4758eaada0c9b", size = 214965, upload-time = "2025-10-02T14:36:03.507Z" },
{ url = "https://files.pythonhosted.org/packages/27/67/dfa980ac7f0d509d54ea0d5a486d2bb4b80c3f1bb22b66e6a05d3efaf6c0/xxhash-3.6.0-cp314-cp314t-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:627f0af069b0ea56f312fd5189001c24578868643203bca1abbc2c52d3a6f3ca", size = 448484, upload-time = "2025-10-02T14:36:04.828Z" },
{ url = "https://files.pythonhosted.org/packages/8c/63/8ffc2cc97e811c0ca5d00ab36604b3ea6f4254f20b7bc658ca825ce6c954/xxhash-3.6.0-cp314-cp314t-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:aa912c62f842dfd013c5f21a642c9c10cd9f4c4e943e0af83618b4a404d9091a", size = 196162, upload-time = "2025-10-02T14:36:06.182Z" },
{ url = "https://files.pythonhosted.org/packages/4b/77/07f0e7a3edd11a6097e990f6e5b815b6592459cb16dae990d967693e6ea9/xxhash-3.6.0-cp314-cp314t-musllinux_1_2_aarch64.whl", hash = "sha256:b465afd7909db30168ab62afe40b2fcf79eedc0b89a6c0ab3123515dc0df8b99", size = 213007, upload-time = "2025-10-02T14:36:07.733Z" },
{ url = "https://files.pythonhosted.org/packages/ae/d8/bc5fa0d152837117eb0bef6f83f956c509332ce133c91c63ce07ee7c4873/xxhash-3.6.0-cp314-cp314t-musllinux_1_2_i686.whl", hash = "sha256:a881851cf38b0a70e7c4d3ce81fc7afd86fbc2a024f4cfb2a97cf49ce04b75d3", size = 200956, upload-time = "2025-10-02T14:36:09.106Z" },
{ url = "https://files.pythonhosted.org/packages/26/a5/d749334130de9411783873e9b98ecc46688dad5db64ca6e04b02acc8b473/xxhash-3.6.0-cp314-cp314t-musllinux_1_2_ppc64le.whl", hash = "sha256:9b3222c686a919a0f3253cfc12bb118b8b103506612253b5baeaac10d8027cf6", size = 213401, upload-time = "2025-10-02T14:36:10.585Z" },
{ url = "https://files.pythonhosted.org/packages/89/72/abed959c956a4bfc72b58c0384bb7940663c678127538634d896b1195c10/xxhash-3.6.0-cp314-cp314t-musllinux_1_2_s390x.whl", hash = "sha256:c5aa639bc113e9286137cec8fadc20e9cd732b2cc385c0b7fa673b84fc1f2a93", size = 417083, upload-time = "2025-10-02T14:36:12.276Z" },
{ url = "https://files.pythonhosted.org/packages/0c/b3/62fd2b586283b7d7d665fb98e266decadf31f058f1cf6c478741f68af0cb/xxhash-3.6.0-cp314-cp314t-musllinux_1_2_x86_64.whl", hash = "sha256:5c1343d49ac102799905e115aee590183c3921d475356cb24b4de29a4bc56518", size = 193913, upload-time = "2025-10-02T14:36:14.025Z" },
{ url = "https://files.pythonhosted.org/packages/9a/9a/c19c42c5b3f5a4aad748a6d5b4f23df3bed7ee5445accc65a0fb3ff03953/xxhash-3.6.0-cp314-cp314t-win32.whl", hash = "sha256:5851f033c3030dd95c086b4a36a2683c2ff4a799b23af60977188b057e467119", size = 31586, upload-time = "2025-10-02T14:36:15.603Z" },
{ url = "https://files.pythonhosted.org/packages/03/d6/4cc450345be9924fd5dc8c590ceda1db5b43a0a889587b0ae81a95511360/xxhash-3.6.0-cp314-cp314t-win_amd64.whl", hash = "sha256:0444e7967dac37569052d2409b00a8860c2135cff05502df4da80267d384849f", size = 32526, upload-time = "2025-10-02T14:36:16.708Z" },
{ url = "https://files.pythonhosted.org/packages/0f/c9/7243eb3f9eaabd1a88a5a5acadf06df2d83b100c62684b7425c6a11bcaa8/xxhash-3.6.0-cp314-cp314t-win_arm64.whl", hash = "sha256:bb79b1e63f6fd84ec778a4b1916dfe0a7c3fdb986c06addd5db3a0d413819d95", size = 28898, upload-time = "2025-10-02T14:36:17.843Z" },
{ url = "https://files.pythonhosted.org/packages/93/1e/8aec23647a34a249f62e2398c42955acd9b4c6ed5cf08cbea94dc46f78d2/xxhash-3.6.0-pp311-pypy311_pp73-macosx_10_15_x86_64.whl", hash = "sha256:0f7b7e2ec26c1666ad5fc9dbfa426a6a3367ceaf79db5dd76264659d509d73b0", size = 30662, upload-time = "2025-10-02T14:37:01.743Z" },
{ url = "https://files.pythonhosted.org/packages/b8/0b/b14510b38ba91caf43006209db846a696ceea6a847a0c9ba0a5b1adc53d6/xxhash-3.6.0-pp311-pypy311_pp73-manylinux1_i686.manylinux_2_28_i686.manylinux_2_5_i686.whl", hash = "sha256:5dc1e14d14fa0f5789ec29a7062004b5933964bb9b02aae6622b8f530dc40296", size = 41056, upload-time = "2025-10-02T14:37:02.879Z" },
{ url = "https://files.pythonhosted.org/packages/50/55/15a7b8a56590e66ccd374bbfa3f9ffc45b810886c8c3b614e3f90bd2367c/xxhash-3.6.0-pp311-pypy311_pp73-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:881b47fc47e051b37d94d13e7455131054b56749b91b508b0907eb07900d1c13", size = 36251, upload-time = "2025-10-02T14:37:04.44Z" },
{ url = "https://files.pythonhosted.org/packages/62/b2/5ac99a041a29e58e95f907876b04f7067a0242cb85b5f39e726153981503/xxhash-3.6.0-pp311-pypy311_pp73-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:c6dc31591899f5e5666f04cc2e529e69b4072827085c1ef15294d91a004bc1bd", size = 32481, upload-time = "2025-10-02T14:37:05.869Z" },
{ url = "https://files.pythonhosted.org/packages/7b/d9/8d95e906764a386a3d3b596f3c68bb63687dfca806373509f51ce8eea81f/xxhash-3.6.0-pp311-pypy311_pp73-win_amd64.whl", hash = "sha256:15e0dac10eb9309508bfc41f7f9deaa7755c69e35af835db9cb10751adebc35d", size = 31565, upload-time = "2025-10-02T14:37:06.966Z" },
]
[[package]]
name = "zstandard"
version = "0.25.0"
@@ -5,5 +5,5 @@ description = "Test for prerelease stuff"
readme = "README.md"
requires-python = ">=3.10"
dependencies = [
"langchain-openai==1.0.1"
"langchain-openai==1.1.14"
]
@@ -5,7 +5,7 @@ description = "Test for prerelease stuff"
readme = "README.md"
requires-python = ">=3.10"
dependencies = [
"langchain-openai==1.0.0a2",
"langchain-openai==1.1.14",
"langchain-anthropic==1.0.0a5",
"langgraph==1.1.5"
]
@@ -5,7 +5,7 @@ description = "Test for prerelease stuff"
readme = "README.md"
requires-python = ">=3.10"
dependencies = [
"langchain-openai==1.0.0a2",
"langchain-openai==1.1.14",
"langgraph==1.1.2",
"langchain_community>=0.3.0",
]
+3 -3
View File
@@ -3676,9 +3676,9 @@ keyv@^4.5.4:
json-buffer "3.0.1"
"langsmith@>=0.5.0 <1.0.0":
version "0.5.18"
resolved "https://registry.yarnpkg.com/langsmith/-/langsmith-0.5.18.tgz#c691ad23614f0b46eaf07d982e0ac988e1f43880"
integrity sha512-3zuZUWffTHQ+73EAwnodADtf534VNEZUpXr9jC12qyG8/IQuJET7PRsCpTb9wX2lmBspakwLUpqpj3tNm/0bVA==
version "0.5.20"
resolved "https://registry.yarnpkg.com/langsmith/-/langsmith-0.5.20.tgz#4021847d2ccd5a86c5eb96060f9bb5f19f80eca5"
integrity sha512-ULhLM8RswvQDXufLtNtvclHrWCBx8Cb5UPI6lAZC+8Dq59iHsVPz/3Ac9khWNm1VIvChRsuykixD/WrmzuuA3Q==
dependencies:
p-queue "6.6.2"
uuid "10.0.0"
+3 -3
View File
@@ -1328,9 +1328,9 @@ keyv@^4.5.4:
json-buffer "3.0.1"
"langsmith@>=0.5.0 <1.0.0":
version "0.5.18"
resolved "https://registry.yarnpkg.com/langsmith/-/langsmith-0.5.18.tgz#c691ad23614f0b46eaf07d982e0ac988e1f43880"
integrity sha512-3zuZUWffTHQ+73EAwnodADtf534VNEZUpXr9jC12qyG8/IQuJET7PRsCpTb9wX2lmBspakwLUpqpj3tNm/0bVA==
version "0.5.20"
resolved "https://registry.yarnpkg.com/langsmith/-/langsmith-0.5.20.tgz#4021847d2ccd5a86c5eb96060f9bb5f19f80eca5"
integrity sha512-ULhLM8RswvQDXufLtNtvclHrWCBx8Cb5UPI6lAZC+8Dq59iHsVPz/3Ac9khWNm1VIvChRsuykixD/WrmzuuA3Q==
dependencies:
p-queue "6.6.2"
uuid "10.0.0"
+1 -1
View File
@@ -1 +1 @@
__version__ = "0.4.21"
__version__ = "0.4.23"
+10 -3
View File
@@ -26,8 +26,15 @@ class LogData(TypedDict):
params: dict[str, Any]
def get_anonymized_params(kwargs: dict[str, Any]) -> dict[str, bool]:
params = {}
def get_anonymized_params(
kwargs: dict[str, Any], *, cli_command: str
) -> dict[str, bool | str]:
params: dict[str, bool | str] = {}
if cli_command == "deploy" and (
analytics_source := os.getenv("LANGGRAPH_CLI_ANALYTICS_SOURCE")
):
params["source"] = analytics_source
# anonymize params with values
if config := kwargs.get("config"):
@@ -88,7 +95,7 @@ def log_command(func):
"python_version": platform.python_version(),
"cli_version": __version__,
"cli_command": func.__name__,
"params": get_anonymized_params(kwargs),
"params": get_anonymized_params(kwargs, cli_command=func.__name__),
}
background_thread = threading.Thread(target=log_data, args=(data,))
+1 -1
View File
@@ -23,7 +23,7 @@ dependencies = [
path = "langgraph_cli/__init__.py"
[project.optional-dependencies]
inmem = [
"langgraph-api>=0.5.35,<0.8.0 ; python_version >= '3.11'",
"langgraph-api>=0.5.35,<0.9.0 ; python_version >= '3.11'",
"langgraph-runtime-inmem>=0.7 ; python_version >= '3.11'",
]
+3 -3
View File
@@ -290,7 +290,7 @@ wheels = [
[[package]]
name = "langsmith"
version = "0.7.26"
version = "0.7.31"
source = { registry = "https://pypi.org/simple" }
dependencies = [
{ name = "httpx" },
@@ -303,9 +303,9 @@ dependencies = [
{ name = "xxhash" },
{ name = "zstandard" },
]
sdist = { url = "https://files.pythonhosted.org/packages/76/86/6de4f6f0451a9658f26f633e0bb090552a4dafd7df3f1ae7f0d40558e67e/langsmith-0.7.26.tar.gz", hash = "sha256:a3e06f3d689ce7195717aa6b8f91082319819ec7ea9b9a62cdcd3d9dc25bfc7b", size = 1146118, upload-time = "2026-04-06T15:01:03.336Z" }
sdist = { url = "https://files.pythonhosted.org/packages/e6/11/696019490992db5c87774dc20515529ef42a01e1d770fb754ed6d9b12fb0/langsmith-0.7.31.tar.gz", hash = "sha256:331ee4f7c26bb5be4022b9859b7d7b122cbf8c9d01d9f530114c1914b0349ffb", size = 1178480, upload-time = "2026-04-14T17:55:41.242Z" }
wheels = [
{ url = "https://files.pythonhosted.org/packages/81/8e/7eb7d65ce62e98e74b9f18f193ea7ac3996d4fbd71fffcc67d0f7ba3103e/langsmith-0.7.26-py3-none-any.whl", hash = "sha256:fe5c877972cea450c1c48251c8fae0f18543c8d19dfdb9ff9a9c4263763dde4e", size = 360160, upload-time = "2026-04-06T15:01:01.516Z" },
{ url = "https://files.pythonhosted.org/packages/1d/a1/a013cf458c301cda86a213dd153ce0a01c93f1ab5833f951e6a44c9763ce/langsmith-0.7.31-py3-none-any.whl", hash = "sha256:0291d49203f6e80dda011af1afda61eb0595a4d697adb684590a8805e1d61fb6", size = 373276, upload-time = "2026-04-14T17:55:39.677Z" },
]
[[package]]
+3 -3
View File
@@ -266,7 +266,7 @@ wheels = [
[[package]]
name = "langsmith"
version = "0.7.26"
version = "0.7.31"
source = { registry = "https://pypi.org/simple" }
dependencies = [
{ name = "httpx" },
@@ -279,9 +279,9 @@ dependencies = [
{ name = "xxhash" },
{ name = "zstandard" },
]
sdist = { url = "https://files.pythonhosted.org/packages/76/86/6de4f6f0451a9658f26f633e0bb090552a4dafd7df3f1ae7f0d40558e67e/langsmith-0.7.26.tar.gz", hash = "sha256:a3e06f3d689ce7195717aa6b8f91082319819ec7ea9b9a62cdcd3d9dc25bfc7b", size = 1146118, upload-time = "2026-04-06T15:01:03.336Z" }
sdist = { url = "https://files.pythonhosted.org/packages/e6/11/696019490992db5c87774dc20515529ef42a01e1d770fb754ed6d9b12fb0/langsmith-0.7.31.tar.gz", hash = "sha256:331ee4f7c26bb5be4022b9859b7d7b122cbf8c9d01d9f530114c1914b0349ffb", size = 1178480, upload-time = "2026-04-14T17:55:41.242Z" }
wheels = [
{ url = "https://files.pythonhosted.org/packages/81/8e/7eb7d65ce62e98e74b9f18f193ea7ac3996d4fbd71fffcc67d0f7ba3103e/langsmith-0.7.26-py3-none-any.whl", hash = "sha256:fe5c877972cea450c1c48251c8fae0f18543c8d19dfdb9ff9a9c4263763dde4e", size = 360160, upload-time = "2026-04-06T15:01:01.516Z" },
{ url = "https://files.pythonhosted.org/packages/1d/a1/a013cf458c301cda86a213dd153ce0a01c93f1ab5833f951e6a44c9763ce/langsmith-0.7.31-py3-none-any.whl", hash = "sha256:0291d49203f6e80dda011af1afda61eb0595a4d697adb684590a8805e1d61fb6", size = 373276, upload-time = "2026-04-14T17:55:39.677Z" },
]
[[package]]
+465 -383
View File
File diff suppressed because it is too large Load Diff
@@ -66,9 +66,6 @@ CONFIG_KEY_RUNTIME = sys.intern("__pregel_runtime")
# holds a `Runtime` instance with context, store, stream writer, etc.
CONFIG_KEY_RESUME_MAP = sys.intern("__pregel_resume_map")
# holds a mapping of task ns -> resume value for resuming tasks
CONFIG_KEY_STREAM_MESSAGES_V2 = sys.intern("__pregel_stream_messages_v2")
# when True, attach StreamMessagesHandlerV2 so content-block (v2) events
# flow through stream_mode="messages"; set by StreamingHandler only.
# --- Other constants ---
PUSH = sys.intern("__pregel_push")
@@ -110,7 +107,6 @@ RESERVED = {
CONFIG_KEY_CHECKPOINT_ID,
CONFIG_KEY_CHECKPOINT_NS,
CONFIG_KEY_RESUME_MAP,
CONFIG_KEY_STREAM_MESSAGES_V2,
# other constants
PUSH,
PULL,
-18
View File
@@ -245,15 +245,6 @@ class _GraphCallbackManager(BaseCallbackManager):
run_id=run_id,
)
def add_handler(
self,
handler: BaseCallbackHandler,
inherit: bool = True, # noqa: FBT001,FBT002
) -> None:
if not isinstance(handler, GraphCallbackHandler):
raise TypeError("handlers must inherit GraphCallbackHandler")
super().add_handler(handler, inherit=inherit)
def copy(
self,
*,
@@ -321,15 +312,6 @@ class _AsyncGraphCallbackManager(BaseCallbackManager):
run_id=run_id,
)
def add_handler(
self,
handler: BaseCallbackHandler,
inherit: bool = True, # noqa: FBT001,FBT002
) -> None:
if not isinstance(handler, GraphCallbackHandler):
raise TypeError("handlers must inherit GraphCallbackHandler")
super().add_handler(handler, inherit=inherit)
def copy(
self,
*,
+177
View File
@@ -0,0 +1,177 @@
from __future__ import annotations
import copy as _copy
from collections.abc import Callable, Sequence
from typing import Any, Generic
from langgraph.checkpoint.base import DELTA_SENTINEL, PendingWrite
from typing_extensions import Self
from langgraph._internal._typing import MISSING
from langgraph.channels.base import BaseChannel, Value
from langgraph.channels.binop import _get_overwrite
from langgraph.errors import EmptyChannelError
__all__ = ("DeltaChannel",)
def _empty(typ: Any) -> Any:
try:
return typ()
except Exception:
return []
class DeltaChannel(Generic[Value], BaseChannel[Any, Any, Any]):
"""Experimental — private API, subject to change or removal without notice.
Imported from the underscored module `langgraph.channels._delta` on purpose;
not re-exported from `langgraph.channels`. Intended for internal use only
while we validate the design on real workloads.
A channel that stores only a sentinel in checkpoints; per-step writes are
stored in checkpoint_writes and replayed through the operator at load time.
Use with append-style reducers (e.g. `add_messages`) on long-running threads
to eliminate O() blob growth storage is O(N) using the writes table that
every checkpointer already maintains.
Reconstruction replays every ancestor write through the operator, so
per-get cost scales with thread depth. Compaction for deep threads is
a follow-up today, use this on threads of a few hundred turns.
Usage::
from langgraph.channels._delta import DeltaChannel
class State(TypedDict):
messages: Annotated[list[AnyMessage], DeltaChannel(add_messages)]
"""
__slots__ = (
"value",
"operator",
)
def __init__(
self,
operator: Callable[[Any, Any], Any],
) -> None:
super().__init__(list)
self.operator = operator
self.value: Any = []
def __eq__(self, other: object) -> bool:
if not isinstance(other, DeltaChannel):
return False
if (
self.operator.__name__ != "<lambda>"
and other.operator.__name__ != "<lambda>"
):
return self.operator is other.operator
return True
@property
def ValueType(self) -> Any:
return self.typ
@property
def UpdateType(self) -> Any:
return self.typ
def copy(self) -> Self:
new: DeltaChannel[Value] = DeltaChannel(self.operator)
new.typ = self.typ
new.key = self.key
new.value = self.value if self.value is MISSING else _copy.copy(self.value)
return new
def _apply_write(self, value: Any, write: Any) -> Any:
"""Apply one write to `value` and return the new value.
An `Overwrite` replaces the value; any other write is folded through
the operator. Centralizes the Overwrite/reducer branching used by both
`update` (live super-step) and `from_checkpoint` (ancestor replay).
"""
is_overwrite, overwrite_value = _get_overwrite(write)
if is_overwrite:
return (
_copy.copy(overwrite_value)
if overwrite_value is not None
else _empty(self.typ)
)
base = _empty(self.typ) if value is MISSING else value
return self.operator(base, write)
def from_checkpoint(self, checkpoint: Any) -> Self:
"""Initialize from a seed value.
Pregel's hydration path calls this with the `seed` returned by
`saver.get_channel_history`:
* `MISSING` / `DELTA_SENTINEL` channel starts empty. The walk
either reached the root (fresh delta thread) or found nothing
to seed from.
* any other value use as the base value. Typically a pre-delta
blob preserved across a channel-type migration; `replay_writes`
folds subsequent deltas on top.
"""
new: DeltaChannel[Value] = DeltaChannel(self.operator)
new.typ = self.typ
new.key = self.key
if checkpoint is MISSING or checkpoint is DELTA_SENTINEL:
new.value = _empty(new.typ)
else:
new.value = checkpoint
return new
def replay_writes(self, writes: Sequence[PendingWrite]) -> None:
"""Fold a sequence of `PendingWrite` tuples into the current value.
Called after `from_checkpoint` during pregel hydration to replay
per-step deltas from on-path ancestors through the reducer. Writes
are oldestnewest. `Overwrite` values inside the stream reset the
reducer state at that point, same as during a live super-step.
The `task_id` and `channel` fields of each `PendingWrite` are
ignored `_get_channel_writes_history` has already filtered to
this channel.
"""
for _, _, value in writes:
self.value = self._apply_write(self.value, value)
def update(self, values: Sequence[Any]) -> bool:
if not values:
return False
seen_overwrite = False
for value in values:
is_overwrite, _ = _get_overwrite(value)
if is_overwrite:
if seen_overwrite:
from langgraph.errors import (
ErrorCode,
InvalidUpdateError,
create_error_message,
)
msg = create_error_message(
message="Can receive only one Overwrite value per super-step.",
error_code=ErrorCode.INVALID_CONCURRENT_GRAPH_UPDATE,
)
raise InvalidUpdateError(msg)
seen_overwrite = True
elif seen_overwrite:
# Post-Overwrite writes within the same super-step are dropped.
continue
self.value = self._apply_write(self.value, value)
return True
def get(self) -> Any:
if self.value is MISSING:
raise EmptyChannelError()
return self.value
def is_available(self) -> bool:
return self.value is not MISSING
def checkpoint(self) -> Any:
return DELTA_SENTINEL
+74 -40
View File
@@ -184,63 +184,97 @@ def add_messages(
```
"""
remove_all_idx = None
# coerce to list
# 1. Coerce scalars to lists.
if not isinstance(left, list):
left = [left] # type: ignore[assignment]
if not isinstance(right, list):
right = [right] # type: ignore[assignment]
# coerce to message
left = [
message_chunk_to_message(cast(BaseMessageChunk, m))
for m in convert_to_messages(left)
]
right = [
left = cast(list, left)
# 2. Normalize `left`. After the first call, `left` is the previous return
# value of `add_messages` — a list of fully-resolved BaseMessages with
# IDs — and needs no work. Fresh user input (dicts, tuples, message
# chunks, BaseMessages without IDs) falls through to full conversion.
left_msgs: list[BaseMessage]
if (
left
and isinstance(left[0], BaseMessage)
and not isinstance(left[0], BaseMessageChunk)
and left[0].id is not None
):
left_msgs = left
else:
left_msgs = [
message_chunk_to_message(cast(BaseMessageChunk, m))
for m in convert_to_messages(left)
]
for m in left_msgs:
if m.id is None:
m.id = str(uuid.uuid4())
# 3. Normalize `right` — always fresh input. Assign missing IDs and detect
# any RemoveMessage sentinels in a single pass.
right_msgs: list[BaseMessage] = [
message_chunk_to_message(cast(BaseMessageChunk, m))
for m in convert_to_messages(right)
]
# assign missing ids
for m in left:
remove_all_idx: int | None = None
has_remove = False
for idx, m in enumerate(right_msgs):
if m.id is None:
m.id = str(uuid.uuid4())
for idx, m in enumerate(right):
if m.id is None:
m.id = str(uuid.uuid4())
if isinstance(m, RemoveMessage) and m.id == REMOVE_ALL_MESSAGES:
remove_all_idx = idx
if isinstance(m, RemoveMessage):
has_remove = True
if m.id == REMOVE_ALL_MESSAGES:
remove_all_idx = idx
# 4. REMOVE_ALL_MESSAGES: discard everything up to and including the sentinel.
if remove_all_idx is not None:
return right[remove_all_idx + 1 :]
return right_msgs[remove_all_idx + 1 :]
# merge
merged = left.copy()
merged_by_id = {m.id: i for i, m in enumerate(merged)}
ids_to_remove = set()
for m in right:
if (existing_idx := merged_by_id.get(m.id)) is not None:
if isinstance(m, RemoveMessage):
ids_to_remove.add(m.id)
# 5. Decide fast vs. slow path. The fast path (pure append) is only valid
# when `right` has no removals, no intra-right duplicate IDs, and no IDs
# that overlap with `left` — any of those would force the indexed merge
# below to update or dedup.
pure_append = False
if not has_remove:
left_ids = {m.id for m in left_msgs}
right_ids = {m.id for m in right_msgs}
pure_append = len(right_ids) == len(right_msgs) and right_ids.isdisjoint(
left_ids
)
if pure_append:
merged = left_msgs + right_msgs
else:
# 6. Slow path: build id→index map over `left`, then replay `right`.
# In-place replacement for matching IDs, append for new IDs, and a
# deferred removal pass so RemoveMessages can target either side.
merged = left_msgs.copy()
merged_by_id = {m.id: i for i, m in enumerate(merged)}
ids_to_remove = set()
for m in right_msgs:
if (existing_idx := merged_by_id.get(m.id)) is not None:
if isinstance(m, RemoveMessage):
ids_to_remove.add(m.id)
else:
ids_to_remove.discard(m.id)
merged[existing_idx] = m
else:
ids_to_remove.discard(m.id)
merged[existing_idx] = m
else:
if isinstance(m, RemoveMessage):
raise ValueError(
f"Attempting to delete a message with an ID that doesn't exist ('{m.id}')"
)
merged_by_id[m.id] = len(merged)
merged.append(m)
merged = [m for m in merged if m.id not in ids_to_remove]
if isinstance(m, RemoveMessage):
raise ValueError(
f"Attempting to delete a message with an ID that doesn't exist ('{m.id}')"
)
merged_by_id[m.id] = len(merged)
merged.append(m)
merged = [m for m in merged if m.id not in ids_to_remove]
# 7. Apply optional output format.
if format == "langchain-openai":
merged = _format_messages(merged)
elif format:
return _format_messages(merged)
if format:
msg = f"Unrecognized {format=}. Expected one of 'langchain-openai', None."
raise ValueError(msg)
else:
pass
return merged
+34 -8
View File
@@ -1,5 +1,6 @@
from __future__ import annotations
import collections.abc
import inspect
import logging
import typing
@@ -47,7 +48,8 @@ from langgraph._internal._pydantic import create_model
from langgraph._internal._runnable import coerce_to_runnable
from langgraph._internal._typing import EMPTY_SEQ, MISSING, DeprecatedKwargs
from langgraph.channels.base import BaseChannel
from langgraph.channels.binop import BinaryOperatorAggregate
from langgraph.channels.binop import BinaryOperatorAggregate, _strip_extras
from langgraph.channels._delta import DeltaChannel
from langgraph.channels.ephemeral_value import EphemeralValue
from langgraph.channels.last_value import LastValue, LastValueAfterFinish
from langgraph.channels.named_barrier_value import (
@@ -1045,7 +1047,6 @@ class StateGraph(Generic[StateT, ContextT, InputT, OutputT]):
interrupt_after: All | list[str] | None = None,
debug: bool = False,
name: str | None = None,
transformers: Sequence[Callable[[], Any]] | None = None,
) -> CompiledStateGraph[StateT, ContextT, InputT, OutputT]:
"""Compiles the `StateGraph` into a `CompiledStateGraph` object.
@@ -1078,16 +1079,12 @@ class StateGraph(Generic[StateT, ContextT, InputT, OutputT]):
interrupt_after: An optional list of node names to interrupt after.
debug: A flag indicating whether to enable debug mode.
name: The name to use for the compiled graph.
transformers: Optional sequence of zero-arg factories returning
`StreamTransformer` instances. Registered on the compiled
graph and instantiated per-run whenever `stream_v2` /
`astream_v2` is called. Appended after the built-in
`ValuesTransformer` and `MessagesTransformer`.
Returns:
CompiledStateGraph: The compiled `StateGraph`.
"""
checkpointer = ensure_valid_checkpointer(checkpointer)
serde_allowlist: set[tuple[str, ...]] | None = None
if _serde.STRICT_MSGPACK_ENABLED:
schema_types: list[type[Any]] = [
@@ -1165,7 +1162,6 @@ class StateGraph(Generic[StateT, ContextT, InputT, OutputT]):
store=store,
cache=cache,
name=name or "LangGraph",
stream_transformers=transformers,
)
compiled._serde_allowlist = serde_allowlist
@@ -1674,6 +1670,36 @@ def _is_field_channel(typ: type[Any]) -> BaseChannel | None:
# Search through all annotated medata to find channel annotations
for item in meta:
if isinstance(item, BaseChannel):
if isinstance(item, DeltaChannel) and hasattr(typ, "__origin__"):
origin = typ.__origin__
# Unwrap parameterized Required[X]/NotRequired[X] to X
# (e.g. Annotated[NotRequired[dict[...]], ...]).
if hasattr(origin, "__origin__") and origin.__origin__ in (
Required,
NotRequired,
):
origin = origin.__args__[0]
outer = _strip_extras(origin)
if outer in (
collections.abc.Sequence,
collections.abc.MutableSequence,
):
outer = list
elif outer in (
collections.abc.Mapping,
collections.abc.MutableMapping,
):
outer = dict
elif outer in (
collections.abc.Set,
collections.abc.MutableSet,
):
outer = set
item.typ = outer
try:
item.value = outer()
except Exception:
item.value = []
return item
elif isclass(item) and issubclass(item, BaseChannel):
# ex, Annotated[int, EphemeralValue, SomeOtherAnnotation]
+79 -9
View File
@@ -3,11 +3,13 @@ from __future__ import annotations
from collections.abc import Mapping
from datetime import datetime, timezone
from langgraph.checkpoint.base import Checkpoint
from langchain_core.runnables import RunnableConfig
from langgraph.checkpoint.base import DELTA_SENTINEL, BaseCheckpointSaver, Checkpoint
from langgraph.checkpoint.base.id import uuid6
from langgraph._internal._typing import MISSING
from langgraph.channels.base import BaseChannel
from langgraph.channels._delta import DeltaChannel
from langgraph.managed.base import ManagedValueMapping, ManagedValueSpec
LATEST_VERSION = 4
@@ -58,8 +60,22 @@ def create_checkpoint(
def channels_from_checkpoint(
specs: Mapping[str, BaseChannel | ManagedValueSpec],
checkpoint: Checkpoint,
*,
saver: BaseCheckpointSaver | None = None,
config: RunnableConfig | None = None,
) -> tuple[Mapping[str, BaseChannel], ManagedValueMapping]:
"""Get channels from a checkpoint."""
"""Hydrate channels from a checkpoint.
For most channels, `spec.from_checkpoint(checkpoint["channel_values"][k])`
is sufficient the stored value IS the reconstructed state.
`DeltaChannel` is the exception: its stored value is a sentinel; the
full state is spread across `checkpoint_writes` along the ancestor
chain. When `saver` and `config` are provided, this function fetches
that history via `saver._get_channel_writes_history` and folds it
through the channel's reducer. Without them (static contexts — graph
drawing, unit tests), delta channels fall back to empty.
"""
channel_specs: dict[str, BaseChannel] = {}
managed_specs: dict[str, ManagedValueSpec] = {}
for k, v in specs.items():
@@ -67,13 +83,67 @@ def channels_from_checkpoint(
channel_specs[k] = v
else:
managed_specs[k] = v
return (
{
k: v.from_checkpoint(checkpoint["channel_values"].get(k, MISSING))
for k, v in channel_specs.items()
},
managed_specs,
)
channels: dict[str, BaseChannel] = {}
for k, spec in channel_specs.items():
ch: BaseChannel
stored = checkpoint["channel_values"].get(k, MISSING)
if (
isinstance(spec, DeltaChannel)
and saver is not None
and config is not None
and (stored is MISSING or stored is DELTA_SENTINEL)
):
# Target's own blob is empty/sentinel — walk ancestors for
# seed + writes. Skipping this when `stored` is a real value
# preserves state written via `update_state` or sitting at the
# tip of a pre-migration thread: the saver's ancestor walk
# intentionally excludes the target's own blob, so without
# this short-circuit we'd lose it.
history = saver._get_channel_writes_history(config, k)
delta_ch = spec.from_checkpoint(history.seed)
delta_ch.replay_writes(history.writes)
ch = delta_ch
else:
ch = spec.from_checkpoint(stored)
channels[k] = ch
return channels, managed_specs
async def achannels_from_checkpoint(
specs: Mapping[str, BaseChannel | ManagedValueSpec],
checkpoint: Checkpoint,
*,
saver: BaseCheckpointSaver | None = None,
config: RunnableConfig | None = None,
) -> tuple[Mapping[str, BaseChannel], ManagedValueMapping]:
"""Async version of `channels_from_checkpoint`. See docstring there."""
channel_specs: dict[str, BaseChannel] = {}
managed_specs: dict[str, ManagedValueSpec] = {}
for k, v in specs.items():
if isinstance(v, BaseChannel):
channel_specs[k] = v
else:
managed_specs[k] = v
channels: dict[str, BaseChannel] = {}
for k, spec in channel_specs.items():
ch: BaseChannel
stored = checkpoint["channel_values"].get(k, MISSING)
if (
isinstance(spec, DeltaChannel)
and saver is not None
and config is not None
and (stored is MISSING or stored is DELTA_SENTINEL)
):
history = await saver._aget_channel_writes_history(config, k)
delta_ch = spec.from_checkpoint(history.seed)
delta_ch.replay_writes(history.writes)
ch = delta_ch
else:
ch = spec.from_checkpoint(stored)
channels[k] = ch
return channels, managed_specs
def copy_checkpoint(checkpoint: Checkpoint) -> Checkpoint:
+50 -8
View File
@@ -92,6 +92,7 @@ from langgraph.pregel._algo import (
task_path_str,
)
from langgraph.pregel._checkpoint import (
achannels_from_checkpoint,
channels_from_checkpoint,
copy_checkpoint,
create_checkpoint,
@@ -692,7 +693,7 @@ class PregelLoop:
# writes so that interrupt() calls re-fire instead of returning
# stale values. But if we're actively resuming, keep them —
# multi-interrupt scenarios need previously resolved values preserved.
if self.is_replaying and (
is_time_traveling = self.is_replaying and (
# Time-travel to a subgraph checkpoint: the parent sets
# RESUMING=True (it can't distinguish time-travel from resume),
# so we check if this subgraph's own ns is in checkpoint_map.
@@ -710,7 +711,8 @@ class PregelLoop:
# (subgraph input is a Send arg, not a Command)
or configurable.get(CONFIG_KEY_RESUMING, False)
)
):
)
if is_time_traveling:
self.checkpoint_pending_writes = [
w for w in self.checkpoint_pending_writes if w[1] != RESUME
]
@@ -765,6 +767,26 @@ class PregelLoop:
if k in self.checkpoint["channel_versions"]:
version = self.checkpoint["channel_versions"][k]
self.checkpoint["versions_seen"][INTERRUPT][k] = version
# When time-traveling (replaying from a specific checkpoint),
# save a fork checkpoint so the replayed execution creates a
# new branch. Without this, if the execution hits an interrupt
# before after_tick() runs, no new checkpoint is created —
# the parent's latest checkpoint remains the old one and
# subsequent resumes load the wrong state.
# Skip for update_state forks (source=update/fork) since they
# already have their own fork checkpoint.
if is_time_traveling and self.checkpoint_metadata.get("source") not in (
"update",
"fork",
):
# Clear old INTERRUPT writes from the loaded checkpoint.
# The fork will have a new checkpoint_id which changes
# task IDs — stale interrupt writes would accumulate and
# confuse the multiple-interrupt check in future resumes.
self.checkpoint_pending_writes = [
w for w in self.checkpoint_pending_writes if w[1] != INTERRUPT
]
self._put_checkpoint({"source": "fork"})
# produce values output
self._emit(
"values", map_output_values, self.output_keys, True, self.channels
@@ -807,14 +829,28 @@ class PregelLoop:
if not self.is_nested:
# Pass the resolved before-bound checkpoint ID so subgraphs can
# find their corresponding checkpoint without re-fetching the
# parent. For forks (source=update), use the fork's parent
# parent. For forks (source=update/fork), use the fork's parent
# checkpoint ID since the fork was created after the subgraph's
# checkpoints from the original execution.
#
# Only gate on is_time_traveling (not is_replaying). When the
# client resumes with an explicit checkpoint_id that happens to
# point at the current head (e.g. LangGraph Studio sending
# `checkpoint: {checkpoint_id}` alongside Command(resume=...)),
# is_replaying is True but is_time_traveling is False. In that
# case subgraphs should load their latest checkpoint normally,
# not go through ReplayState's before-bound lookup which would
# miss subgraph checkpoints created during processing of the
# current parent step.
replay_state: ReplayState | None = None
if self.is_replaying:
if is_time_traveling:
replay_checkpoint_id = self.checkpoint["id"]
if (
self.checkpoint_metadata.get("source") == "update"
self.checkpoint_metadata.get("source")
in (
"update",
"fork",
)
and self.prev_checkpoint_config
):
replay_checkpoint_id = self.prev_checkpoint_config[CONF].get(
@@ -1238,7 +1274,10 @@ class SyncPregelLoop(PregelLoop, AbstractContextManager):
)
self.submit = self.stack.enter_context(BackgroundExecutor(self.config))
self.channels, self.managed = channels_from_checkpoint(
self.specs, self.checkpoint
self.specs,
self.checkpoint,
saver=self.checkpointer,
config=self.checkpoint_config,
)
self.stack.push(self._suppress_interrupt)
self.status = "input"
@@ -1441,8 +1480,11 @@ class AsyncPregelLoop(PregelLoop, AbstractAsyncContextManager):
self.submit = await self.stack.enter_async_context(
AsyncBackgroundExecutor(self.config)
)
self.channels, self.managed = channels_from_checkpoint(
self.specs, self.checkpoint
self.channels, self.managed = await achannels_from_checkpoint(
self.specs,
self.checkpoint,
saver=self.checkpointer,
config=self.checkpoint_config,
)
self.stack.push(self._suppress_interrupt)
self.status = "input"
+6 -94
View File
@@ -14,7 +14,7 @@ from langchain_core.messages import BaseMessage
from langchain_core.outputs import ChatGeneration, ChatGenerationChunk, LLMResult
from pydantic import BaseModel
from langgraph._internal._constants import NS_END, NS_SEP
from langgraph._internal._constants import NS_SEP
from langgraph.constants import TAG_HIDDEN, TAG_NOSTREAM
from langgraph.pregel.protocol import StreamChunk
from langgraph.types import Command
@@ -24,11 +24,6 @@ try:
except ImportError:
_StreamingCallbackHandler = object # type: ignore
try:
from langchain_core.tracers._streaming import _V2StreamingCallbackHandler
except ImportError:
_V2StreamingCallbackHandler = object # type: ignore
T = TypeVar("T")
Meta = tuple[tuple[str, ...], dict[str, Any]]
@@ -137,23 +132,15 @@ class StreamMessagesHandler(BaseCallbackHandler, _StreamingCallbackHandler):
**kwargs: Any,
) -> Any:
if metadata and (not tags or (TAG_NOSTREAM not in tags)):
task_checkpoint_ns = cast(str, metadata["langgraph_checkpoint_ns"])
checkpoint_ns = (
f"{task_checkpoint_ns.rsplit(NS_END, 1)[0]}{NS_END}"
if NS_END in task_checkpoint_ns
else task_checkpoint_ns
)
ns = tuple(task_checkpoint_ns.split(NS_SEP))[:-1]
ns = tuple(cast(str, metadata["langgraph_checkpoint_ns"]).split(NS_SEP))[
:-1
]
if not self.subgraphs and len(ns) > 0 and ns != self.parent_ns:
return
stream_metadata = dict(metadata)
stream_metadata["langgraph_checkpoint_ns"] = checkpoint_ns
# Preserve backwards-compatible streamed checkpoint metadata shape.
stream_metadata["checkpoint_ns"] = checkpoint_ns
if tags:
if filtered_tags := [t for t in tags if not t.startswith("seq:step")]:
stream_metadata["tags"] = filtered_tags
self.metadata[run_id] = (ns, stream_metadata)
metadata["tags"] = filtered_tags
self.metadata[run_id] = (ns, metadata)
def on_llm_new_token(
self,
@@ -261,78 +248,3 @@ class StreamMessagesHandler(BaseCallbackHandler, _StreamingCallbackHandler):
**kwargs: Any,
) -> Any:
self.metadata.pop(run_id, None)
class StreamMessagesHandlerV2(StreamMessagesHandler, _V2StreamingCallbackHandler):
"""v2 variant of `StreamMessagesHandler`.
Declaring `_V2StreamingCallbackHandler` as a base flips
`BaseChatModel.invoke` to route through `_stream_chat_model_events`
(firing `on_stream_event`) instead of `_stream` (firing
`on_llm_new_token`). Inherits `on_stream_event` from the parent,
which forwards protocol events onto the messages stream channel.
Pregel attaches this class instead of the v1 handler only when
`StreamingHandler` opts in via the internal
`CONFIG_KEY_STREAM_MESSAGES_V2` config key; direct
`graph.stream(stream_mode="messages")` callers keep the v1
AIMessageChunk shape.
"""
def on_llm_new_token(
self,
token: str,
*,
chunk: ChatGenerationChunk | None = None,
run_id: UUID,
parent_run_id: UUID | None = None,
tags: list[str] | None = None,
**kwargs: Any,
) -> Any:
"""Intentional no-op — v1 chunks are not used on v2-flagged runs.
The v2 marker already steers `invoke` to the event generator, so
`on_llm_new_token` should not fire under normal routing. This
override stays a pass-through (no call to `super()`) to make
the intent explicit and to guard against any caller (e.g. a
node that calls `model.stream()` directly, which still fires
the v1 callback) leaking AIMessageChunks onto a v2-flagged
messages stream.
"""
# Intentionally empty: v2 handler does not forward v1 chunks.
def on_stream_event(
self,
event: dict[str, Any],
*,
run_id: UUID,
parent_run_id: UUID | None = None,
tags: list[str] | None = None,
**kwargs: Any,
) -> Any:
"""Forward a protocol event from `stream_v2` as a messages stream part.
Fires once per `MessagesData` event (`message-start`, per-block
`content-block-*`, `message-finish`). The transformer layer
correlates events back to a single `ChatModelStream` via
`metadata["run_id"]` attached here so the v1
`stream_mode="messages"` output (which emits
`(AIMessageChunk, metadata)` via `on_llm_new_token`) keeps its
original metadata shape.
Lives on the v2 handler rather than the v1 base: content-block
events are a v2-only concept, and forwarding them only when the
v2 handler is attached keeps the message channel's shape
predictable for v1 callers.
"""
if meta := self.metadata.get(run_id):
# Record message_id on message-start so on_chain_end's
# dedupe skips the finalized AIMessage the node returns
# (otherwise the messages projection double-counts: once
# from streaming, once from the chain output).
if event.get("event") == "message-start":
msg_id = event.get("message_id")
if msg_id:
self.seen.add(msg_id)
v2_meta = {**meta[1], "run_id": str(run_id)}
self.stream((meta[0], "messages", (event, v2_meta)))
+24 -178
View File
@@ -73,7 +73,6 @@ from langgraph._internal._constants import (
CONFIG_KEY_RUNTIME,
CONFIG_KEY_SEND,
CONFIG_KEY_STREAM,
CONFIG_KEY_STREAM_MESSAGES_V2,
CONFIG_KEY_TASK_ID,
CONFIG_KEY_THREAD_ID,
ERROR,
@@ -123,6 +122,7 @@ from langgraph.pregel._algo import (
)
from langgraph.pregel._call import identifier
from langgraph.pregel._checkpoint import (
achannels_from_checkpoint,
channels_from_checkpoint,
copy_checkpoint,
create_checkpoint,
@@ -134,10 +134,7 @@ from langgraph.pregel._loop import (
AsyncPregelLoop,
SyncPregelLoop,
)
from langgraph.pregel._messages import (
StreamMessagesHandler,
StreamMessagesHandlerV2,
)
from langgraph.pregel._messages import StreamMessagesHandler
from langgraph.pregel._read import DEFAULT_BOUND, PregelNode
from langgraph.pregel._retry import RetryPolicy
from langgraph.pregel._runner import PregelRunner
@@ -344,69 +341,6 @@ class NodeBuilder:
)
_STREAM_V2_MODES: list[StreamMode] = [
"values",
"updates",
"messages",
"custom",
"checkpoints",
"tasks",
"debug",
]
def _build_stream_factories(
compile_time: Sequence[Callable[..., Any]],
call_site: Sequence[Any] | None,
) -> list[Callable[..., Any]]:
"""Assemble the factory list handed to `StreamMux(factories=...)`.
Prepends the auto-registered built-ins `ValuesTransformer`
(state snapshots backing `run.output` / `run.interrupted`),
`MessagesTransformer` (LLM token streaming), and
`SubgraphTransformer` (in-process subgraph handle discovery)
then appends the graph's compile-time `stream_transformers`
followed by any call-site additions. Factories flow down into
subgraph mini-muxes, so per-scope instances propagate
automatically.
`LifecycleTransformer` is opt-in: add it via compile-time
`stream_transformers=[...]` or the per-call `transformers=[...]`
kwarg on `stream_v2()` / `astream_v2()`. Without it, no
`lifecycle` wire events are emitted and `run.lifecycle` is
absent.
"""
from langgraph.stream.transformers import (
MessagesTransformer,
SubgraphTransformer,
ValuesTransformer,
)
builtins: list[Callable[..., Any]] = [
ValuesTransformer,
MessagesTransformer,
SubgraphTransformer,
]
return [*builtins, *compile_time, *(call_site or ())]
def _merge_v2_messages_flag(
config: RunnableConfig | None,
) -> RunnableConfig:
"""Return a config with the v2 messages flag set in `configurable`.
Signals to pregel that `stream_mode="messages"` should attach
`StreamMessagesHandlerV2` for this call so invoke-time model runs
route through the v2 event generator and their protocol events
reach the messages channel.
"""
merged: RunnableConfig = dict(config or {}) # type: ignore[assignment]
configurable = dict(merged.get(CONF) or {})
configurable[CONFIG_KEY_STREAM_MESSAGES_V2] = True
merged[CONF] = configurable
return merged
class Pregel(
PregelProtocol[StateT, ContextT, InputT, OutputT],
Generic[StateT, ContextT, InputT, OutputT],
@@ -738,7 +672,6 @@ class Pregel(
config: RunnableConfig | None = None,
trigger_to_nodes: Mapping[str, Sequence[str]] | None = None,
name: str = "LangGraph",
stream_transformers: Sequence[Callable[..., Any]] | None = None,
**deprecated_kwargs: Unpack[DeprecatedKwargs],
) -> None:
if (
@@ -785,9 +718,6 @@ class Pregel(
self.config = config
self.trigger_to_nodes = trigger_to_nodes or {}
self.name = name
self._stream_transformers: tuple[Callable[..., Any], ...] = tuple(
stream_transformers or ()
)
self._serde_allowlist: set[tuple[str, ...]] | None = None
if auto_validate:
self.validate()
@@ -1123,6 +1053,10 @@ class Pregel(
channels, managed = channels_from_checkpoint(
self.channels,
saved.checkpoint,
saver=self.checkpointer
if isinstance(self.checkpointer, BaseCheckpointSaver)
else None,
config=saved.config,
)
# tasks for this checkpoint
next_tasks = prepare_next_tasks(
@@ -1239,9 +1173,13 @@ class Pregel(
step = saved.metadata.get("step", -1) + 1
stop = step + 2
channels, managed = channels_from_checkpoint(
channels, managed = await achannels_from_checkpoint(
self.channels,
saved.checkpoint,
saver=self.checkpointer
if isinstance(self.checkpointer, BaseCheckpointSaver)
else None,
config=saved.config,
)
# tasks for this checkpoint
next_tasks = prepare_next_tasks(
@@ -1612,6 +1550,11 @@ class Pregel(
channels, managed = channels_from_checkpoint(
self.channels,
checkpoint,
saver=self.checkpointer
if saved is not None
and isinstance(self.checkpointer, BaseCheckpointSaver)
else None,
config=saved.config if saved is not None else None,
)
values, as_node = updates[0][:2]
@@ -2055,9 +1998,14 @@ class Pregel(
)
if saved:
checkpoint_config = patch_configurable(config, saved.config[CONF])
channels, managed = channels_from_checkpoint(
channels, managed = await achannels_from_checkpoint(
self.channels,
checkpoint,
saver=self.checkpointer
if saved is not None
and isinstance(self.checkpointer, BaseCheckpointSaver)
else None,
config=saved.config if saved is not None else None,
)
values, as_node = updates[0][:2]
# no values, just clear all tasks
@@ -2697,13 +2645,8 @@ class Pregel(
# set up messages stream mode
if "messages" in stream_modes:
ns_ = cast(str | None, config[CONF].get(CONFIG_KEY_CHECKPOINT_NS))
messages_handler_cls = (
StreamMessagesHandlerV2
if config[CONF].get(CONFIG_KEY_STREAM_MESSAGES_V2)
else StreamMessagesHandler
)
run_manager.inheritable_handlers.append(
messages_handler_cls(
StreamMessagesHandler(
stream.put,
subgraphs,
parent_ns=tuple(ns_.split(NS_SEP)) if ns_ else None,
@@ -3079,13 +3022,8 @@ class Pregel(
if "messages" in stream_modes:
# namespace can be None in a root level graph?
ns_ = cast(str | None, config[CONF].get(CONFIG_KEY_CHECKPOINT_NS))
messages_handler_cls = (
StreamMessagesHandlerV2
if config[CONF].get(CONFIG_KEY_STREAM_MESSAGES_V2)
else StreamMessagesHandler
)
run_manager.inheritable_handlers.append(
messages_handler_cls(
StreamMessagesHandler(
stream_put,
subgraphs,
parent_ns=tuple(ns_.split(NS_SEP)) if ns_ else None,
@@ -3318,98 +3256,6 @@ class Pregel(
await asyncio.shield(run_manager.on_chain_error(e))
raise
def stream_v2(
self,
input: InputT | Command | None,
config: RunnableConfig | None = None,
*,
interrupt_before: All | Sequence[str] | None = None,
interrupt_after: All | Sequence[str] | None = None,
transformers: Sequence[Any] | None = None,
) -> Any:
"""Start a sync v2 streaming run driven by transformer projections.
Builds a `StreamMux` from the auto-registered built-ins
(`ValuesTransformer`, `MessagesTransformer`,
`SubgraphTransformer`), this graph's compile-time
`stream_transformers`, and any additional `transformers=`
supplied at the call site. Returns a `GraphRunStream` that
the caller drives by iterating any projection no background
thread.
`LifecycleTransformer` (emits `lifecycle` wire events,
exposes `run.lifecycle`) is opt-in add it via
`stream_transformers` at compile time or via the
`transformers=` kwarg here.
Args:
input: Graph input.
config: Optional runnable config forwarded to the graph.
interrupt_before: Nodes to interrupt before, if any.
interrupt_after: Nodes to interrupt after, if any.
transformers: Extra transformer instances appended after
compile-time `stream_transformers`.
Returns:
A `GraphRunStream` the caller iterates to drive the run.
"""
from langgraph.stream._mux import StreamMux
from langgraph.stream.run_stream import GraphRunStream
factories = _build_stream_factories(self._stream_transformers, transformers)
mux = StreamMux(factories=factories, is_async=False)
graph_iter = iter(
self.stream(
input,
_merge_v2_messages_flag(config),
stream_mode=_STREAM_V2_MODES,
subgraphs=True,
version="v2",
interrupt_before=interrupt_before,
interrupt_after=interrupt_after,
)
)
return GraphRunStream(graph_iter, mux)
async def astream_v2(
self,
input: InputT | Command | None,
config: RunnableConfig | None = None,
*,
interrupt_before: All | Sequence[str] | None = None,
interrupt_after: All | Sequence[str] | None = None,
transformers: Sequence[Any] | None = None,
) -> Any:
"""Async counterpart to `stream_v2`.
Returns an `AsyncGraphRunStream` whose projections can be awaited
concurrently; each subscribed cursor drives the pump when its
buffer is empty.
Args:
input: Graph input.
config: Optional runnable config forwarded to the graph.
interrupt_before: Nodes to interrupt before, if any.
interrupt_after: Nodes to interrupt after, if any.
transformers: Extra transformer instances appended after
compile-time `stream_transformers`.
"""
from langgraph.stream._mux import StreamMux
from langgraph.stream.run_stream import AsyncGraphRunStream
factories = _build_stream_factories(self._stream_transformers, transformers)
mux = StreamMux(factories=factories, is_async=True)
graph_aiter = self.astream(
input,
_merge_v2_messages_flag(config),
stream_mode=_STREAM_V2_MODES,
subgraphs=True,
version="v2",
interrupt_before=interrupt_before,
interrupt_after=interrupt_after,
).__aiter__()
return AsyncGraphRunStream(graph_aiter, mux)
@overload
def invoke(
self,
+3 -5
View File
@@ -653,11 +653,10 @@ class RemoteGraph(PregelProtocol):
# coerce to list, or add default stream mode
if stream_mode:
if isinstance(stream_mode, str):
updated_stream_modes.append(cast(StreamModeSDK, stream_mode))
updated_stream_modes.append(stream_mode)
else:
req_single = False
for m in stream_mode:
updated_stream_modes.append(cast(StreamModeSDK, m))
updated_stream_modes.extend(stream_mode)
else:
updated_stream_modes.append(default)
requested_stream_modes = updated_stream_modes.copy()
@@ -666,8 +665,7 @@ class RemoteGraph(PregelProtocol):
(config or {}).get(CONF, {}).get(CONFIG_KEY_STREAM)
)
if stream:
for m in stream.modes:
updated_stream_modes.append(cast(StreamModeSDK, m))
updated_stream_modes.extend(stream.modes)
# map "messages" to "messages-tuple"
if "messages" in updated_stream_modes:
updated_stream_modes.remove("messages")
@@ -1,20 +0,0 @@
"""Streaming infrastructure for LangGraph.
Compile a graph with `transformers=[...]` and call `graph.stream_v2()` /
`graph.astream_v2()` to drive a transformer pipeline that projects the
graph's raw events into ergonomic per-channel streams.
"""
from langgraph.stream._event_log import EventLog
from langgraph.stream._types import ProtocolEvent, StreamTransformer
from langgraph.stream.run_stream import AsyncGraphRunStream, GraphRunStream
from langgraph.stream.stream_channel import StreamChannel
__all__ = [
"AsyncGraphRunStream",
"EventLog",
"GraphRunStream",
"ProtocolEvent",
"StreamChannel",
"StreamTransformer",
]
@@ -1,32 +0,0 @@
from __future__ import annotations
import time
from typing import Any, cast
from langgraph.stream._types import ProtocolEvent, _ProtocolEventParams
from langgraph.types import StreamPart
def convert_to_protocol_event(part: StreamPart) -> ProtocolEvent:
"""Convert a v2 StreamPart to a ProtocolEvent.
Args:
part: A stream part with keys `type`, `ns`, `data`, and
optionally `interrupts` (present on values events).
Returns:
The equivalent ProtocolEvent.
"""
part_dict = cast(dict[str, Any], part)
params: _ProtocolEventParams = {
"namespace": list(part_dict["ns"]),
"timestamp": int(time.time() * 1000),
"data": part_dict["data"],
}
if "interrupts" in part_dict:
params["interrupts"] = part_dict["interrupts"]
return {
"type": "event",
"method": part_dict["type"],
"params": params,
}
@@ -1,318 +0,0 @@
from __future__ import annotations
import asyncio
from collections import deque
from collections.abc import AsyncIterator, Awaitable, Callable, Iterator
from typing import Generic, TypeVar
T = TypeVar("T")
class EventLog(Generic[T]):
"""Single-consumer drainable queue for streaming events.
Items are popped off the front as the consumer advances there is
no retention beyond what's currently queued. A log accepts exactly
one subscriber; a second `__iter__` / `__aiter__` call raises. Use
`tee(n)` / `atee(n)` for fan-out.
Starts unbound neither `__iter__` nor `__aiter__` is available
until the StreamMux calls `_bind(is_async)`. After binding, only
the matching iteration protocol works; the other raises `TypeError`.
Pump wiring (set by the run stream, not by `_bind`):
- `_request_more`: sync pump callable, returns True if a new
event was produced.
- `_arequest_more`: async pump coroutine factory, same contract.
Memory is bounded by caller pace: both sync and async use caller-
driven pumps, so each cursor advance produces at most one event.
The only shape where a log can accumulate meaningfully is
concurrent async consumers at unequal rates a slow consumer's
log grows while fast consumers drive the shared pump. That's the
documented tradeoff for concurrent consumption; consume at similar
rates or use a single consumer if memory matters.
Lazy-subscribe: `push` is a no-op when no subscriber has registered.
Transformers still execute `process()` (so scalar state like
`ValuesTransformer._latest` stays current); only the log append is
skipped.
"""
def __init__(self, maxlen: int | None = None, *, retain: bool = False) -> None:
"""Initialize an empty, unbound log.
Args:
maxlen: Accepted for forward compatibility; currently unused.
The caller-driven pump bounds memory naturally for
single-consumer use.
retain: If True, `push()` retains items regardless of whether
a consumer has subscribed yet. Used for projections
whose consumer only becomes visible after events have
already flowed (e.g. mini-mux logs inside dynamically
discovered subgraph handles, or the `lifecycle` channel
iterated after draining `values`). Subscription
exclusivity on `__iter__` is unchanged.
Raises:
ValueError: If `maxlen` is not a positive integer or `None`.
"""
if maxlen is not None and maxlen <= 0:
raise ValueError("EventLog maxlen must be a positive int or None")
self._items: deque[T] = deque()
self._maxlen: int | None = maxlen
self._closed = False
self._error: BaseException | None = None
# Binding state — None means unbound.
self._is_async: bool | None = None
# Flipped on first __iter__ / __aiter__. Pre-subscription
# pushes are silent no-ops unless `_retain` is True.
self._subscribed = False
self._retain = retain
# Pump wiring set by the run stream after bind.
self._request_more: Callable[[], bool] | None = None
self._arequest_more: Callable[[], Awaitable[bool]] | None = None
# ------------------------------------------------------------------
# Binding
# ------------------------------------------------------------------
def _bind(self, *, is_async: bool) -> None:
"""Bind this log to sync or async mode.
Called by the StreamMux after transformer registration. Must be
called exactly once before any iteration.
Args:
is_async: True to enable async iteration, False for sync.
Raises:
RuntimeError: If the log has already been bound.
"""
if self._is_async is not None:
raise RuntimeError("EventLog is already bound")
self._is_async = is_async
# ------------------------------------------------------------------
# Producer API
# ------------------------------------------------------------------
def push(self, item: T) -> None:
"""Append an item. No-op when no subscriber is registered.
Non-blocking in both sync and async matches v1's
`put_nowait` producer shape. Memory is bounded by caller pace
via the caller-driven pump.
When `retain=True` was set at construction, items are appended
regardless of subscription for projections whose consumer
only reaches them after events have already flowed.
Raises:
RuntimeError: If the log is closed (and subscribed).
"""
if not self._subscribed and not self._retain:
return
if self._closed:
raise RuntimeError("Cannot push to a closed EventLog")
self._items.append(item)
def close(self) -> None:
"""Mark the log as complete."""
self._closed = True
def fail(self, err: BaseException) -> None:
"""Mark the log as errored.
Args:
err: The exception to surface to the subscriber.
"""
self._error = err
self._closed = True
# ------------------------------------------------------------------
# Sync iteration (caller-driven pump)
# ------------------------------------------------------------------
def __iter__(self) -> Iterator[T]:
"""Subscribe and return a sync cursor. Can be called only once.
Raises:
TypeError: If the log is unbound or bound to async mode.
RuntimeError: If the log already has a subscriber.
"""
if self._is_async is None:
raise TypeError(
"EventLog has not been bound yet. "
"Register the transformer with a StreamMux first."
)
if self._is_async:
raise TypeError(
"This EventLog is bound to async mode — use 'async for' instead."
)
if self._subscribed:
raise RuntimeError(
"EventLog already has a subscriber; use .tee(n) for fan-out."
)
self._subscribed = True
return self._sync_cursor()
def _sync_cursor(self) -> Iterator[T]:
while True:
if self._items:
yield self._items.popleft()
elif self._closed:
if self._error is not None:
raise self._error
return
elif self._request_more is not None:
if not self._request_more():
if not self._items and not self._closed:
return
else:
return
# ------------------------------------------------------------------
# Async iteration (caller-driven pump)
# ------------------------------------------------------------------
def __aiter__(self) -> AsyncIterator[T]:
"""Subscribe and return an async cursor. Can be called only once.
Raises:
TypeError: If the log is unbound or bound to sync mode.
RuntimeError: If the log already has a subscriber.
"""
if self._is_async is None:
raise TypeError(
"EventLog has not been bound yet. "
"Register the transformer with a StreamMux first."
)
if not self._is_async:
raise TypeError("This EventLog is bound to sync mode — use 'for' instead.")
if self._subscribed:
raise RuntimeError(
"EventLog already has a subscriber; use .atee(n) for fan-out."
)
self._subscribed = True
return self._async_cursor()
async def _async_cursor(self) -> AsyncIterator[T]:
while True:
if self._items:
yield self._items.popleft()
elif self._closed:
if self._error is not None:
raise self._error
return
elif self._arequest_more is not None:
if not await self._arequest_more():
if not self._items and not self._closed:
return
else:
return
# ------------------------------------------------------------------
# Fan-out via tee
# ------------------------------------------------------------------
def tee(self, n: int = 2) -> tuple[Iterator[T], ...]:
"""Subscribe and return `n` independent sync iterators.
Each branch has its own buffer; items pulled from the
underlying cursor are copied into every branch. Branches are
naturally bounded by caller pace since the sync pump is
caller-driven.
Args:
n: Number of branches to create. Must be >= 1.
Returns:
A tuple of `n` iterators over the same underlying stream.
Raises:
TypeError: If the log is unbound or bound to async mode.
RuntimeError: If the log already has a subscriber.
ValueError: If `n` < 1.
"""
if n < 1:
raise ValueError("tee() requires n >= 1")
source = self.__iter__()
buffers: list[deque[T]] = [deque() for _ in range(n)]
exhausted = [False]
def branch(i: int) -> Iterator[T]:
buf = buffers[i]
while True:
if buf:
yield buf.popleft()
elif exhausted[0]:
return
else:
try:
item = next(source)
except StopIteration:
exhausted[0] = True
return
for b in buffers:
b.append(item)
return tuple(branch(i) for i in range(n))
def atee(self, n: int = 2) -> tuple[AsyncIterator[T], ...]:
"""Subscribe and return `n` independent async iterators.
Caller-driven fan-out: each branch's `__anext__` either pops
from its own buffer or, under a shared `asyncio.Lock`, pulls
one item from the underlying cursor and distributes it to
every branch's buffer.
Args:
n: Number of branches to create. Must be >= 1.
Returns:
A tuple of `n` async iterators over the same underlying
stream.
Raises:
TypeError: If the log is unbound or bound to sync mode.
RuntimeError: If the log already has a subscriber.
ValueError: If `n` < 1.
"""
if n < 1:
raise ValueError("atee() requires n >= 1")
source = self.__aiter__()
buffers: list[deque[T]] = [deque() for _ in range(n)]
exhausted = [False]
error: list[BaseException | None] = [None]
lock = asyncio.Lock()
async def branch(i: int) -> AsyncIterator[T]:
buf = buffers[i]
while True:
if buf:
yield buf.popleft()
continue
if exhausted[0]:
if error[0] is not None:
raise error[0]
return
async with lock:
if buf or exhausted[0]:
continue
try:
item = await source.__anext__()
except StopAsyncIteration:
exhausted[0] = True
continue
except Exception as e:
error[0] = e
exhausted[0] = True
continue
for b in buffers:
b.append(item)
return tuple(branch(i) for i in range(n))
-514
View File
@@ -1,514 +0,0 @@
from __future__ import annotations
import asyncio
import time
from collections.abc import Awaitable, Callable
from typing import Any
from langgraph.stream._event_log import EventLog
from langgraph.stream._types import (
ProtocolEvent,
StreamTransformer,
transformer_requires_async,
)
from langgraph.stream.stream_channel import StreamChannel
TransformerFactory = Callable[["tuple[str, ...]"], StreamTransformer]
"""Factory that builds a scoped transformer for a mux.
Called once per `StreamMux` (root or mini-mux) with the mux's scope
typically a subgraph's namespace or `()` for the root. Standard
transformer classes (`ValuesTransformer`, `MessagesTransformer`,
`SubgraphTransformer`) accept a single positional scope argument, so
the class itself is a valid factory. User transformers can close over
their config: `lambda scope: MyTransformer(scope, foo=...)`.
"""
class StreamMux:
"""Central event dispatcher for the streaming infrastructure.
Owns the main event log and routes events through a transformer
pipeline. StreamChannels discovered in transformer projections are
auto-wired so that every `push()` also injects a `ProtocolEvent`
into the main log.
Pass `is_async=True` when the mux will be consumed via async
iteration (`handler.astream()`). All EventLog and StreamChannel
instances discovered during registration are automatically bound
to the matching mode.
Attributes:
extensions: Merged projection dict across all registered
transformers. Treat as read-only mutations won't be
reflected back in individual transformers' state.
native_keys: Projection keys contributed by transformers with
`_native = True`.
"""
def __init__(
self,
transformers: list[StreamTransformer] | None = None,
*,
is_async: bool = False,
factories: list[TransformerFactory] | None = None,
scope: tuple[str, ...] = (),
) -> None:
"""Initialize the mux and register transformers in order.
Callers pass either `transformers` (pre-built instances) or
`factories` (callables producing fresh instances per mux). A
factory list is preferred mini-muxes built by `make_child()`
inherit the factory list, so transformers propagate naturally
into every subgraph's scope. `transformers` is kept for
back-compat tests that exercise the mux directly.
Each transformer's `init()` is called once during registration,
projections are merged into `extensions`, `_native` keys are
recorded in `native_keys`, and any EventLog / StreamChannel
instances are bound and wired.
Args:
transformers: Already-built transformer instances. Mutually
exclusive with `factories`.
is_async: True for async dispatch (`apush` / `aclose` /
`afail`), False for the sync path.
factories: Zero-or-one-argument callables producing
transformers. Called with this mux's `scope`.
scope: The namespace the mux operates within. The root mux
is `()`; mini-muxes for subgraphs use the subgraph's
namespace tuple.
Raises:
RuntimeError: If any transformer requires an async run but
the mux is in sync mode.
TypeError: If a transformer's `init()` doesn't return a dict.
ValueError: If transformers' projection keys collide, or if
both `transformers` and `factories` are supplied.
"""
if transformers is not None and factories is not None:
raise ValueError("Pass either `transformers` or `factories`, not both.")
self._is_async = is_async
self._factories: list[TransformerFactory] = list(factories or ())
self.scope: tuple[str, ...] = scope
self._pump_fn: Callable[[], bool] | None = None
self._apump_fn: Callable[[], Awaitable[bool]] | None = None
self._events: EventLog[ProtocolEvent] = EventLog()
self._events._bind(is_async=is_async)
self._transformers: list[StreamTransformer] = []
self._channels: list[StreamChannel[Any]] = []
self._logs: list[EventLog[Any]] = []
self._seq = 0
self.extensions: dict[str, Any] = {}
self.native_keys: set[str] = set()
self._projection_owners: dict[str, str] = {}
self._transformer_by_key: dict[str, StreamTransformer] = {}
if factories is not None:
for factory in factories:
self._register(factory(scope))
else:
for transformer in transformers or ():
self._register(transformer)
def make_child(self, scope: tuple[str, ...]) -> StreamMux:
"""Build a mini-mux with the same factories scoped to `scope`.
Used by `SubgraphTransformer` to attach a fresh transformer
pipeline to each discovered subgraph handle. The child mux
inherits the current pump binding (so cursors on its projection
logs drive the root pump) and carries the same factory list
forward to any grandchild subgraphs.
Raises:
RuntimeError: If the mux was not built from a factory list
(i.e., constructed with `transformers=`). Mini-muxes
require factories so each scope gets its own fresh
transformer instances.
"""
if not self._factories:
raise RuntimeError(
"StreamMux.make_child requires the mux to be constructed "
"with factories; pre-built transformers can't be cloned "
"to a new scope."
)
child = StreamMux(
factories=self._factories,
is_async=self._is_async,
scope=scope,
)
# Mini-muxes are created during the pump, after the first
# event at the child's scope has already been dispatched.
# Consumers reach child projections via the parent's
# `subgraphs` handle — necessarily after that first event.
# Flip retain on every log and channel so pushes are buffered
# until the consumer subscribes.
child._events._retain = True
for value in child.extensions.values():
if isinstance(value, EventLog):
value._retain = True
elif isinstance(value, StreamChannel):
value._log._retain = True
if self._pump_fn is not None:
child.bind_pump(self._pump_fn)
if self._apump_fn is not None:
child.bind_apump(self._apump_fn)
return child
def bind_pump(self, fn: Callable[[], bool]) -> None:
"""Wire the sync pull callback onto every EventLog in the mux.
Also propagates to transformers that expose `_bind_pump` so
nested handles (e.g., `ChatModelStream` instances produced by
`MessagesTransformer`) can drive the graph pump from their
projection cursors.
"""
self._pump_fn = fn
self._events._request_more = fn
for value in self.extensions.values():
if isinstance(value, EventLog):
value._request_more = fn
elif isinstance(value, StreamChannel):
value._log._request_more = fn
for transformer in self._transformers:
bind = getattr(transformer, "_bind_pump", None)
if bind is not None:
bind(fn)
def bind_apump(self, fn: Callable[[], Awaitable[bool]]) -> None:
"""Async counterpart to `bind_pump`."""
self._apump_fn = fn
self._events._arequest_more = fn
for value in self.extensions.values():
if isinstance(value, EventLog):
value._arequest_more = fn
elif isinstance(value, StreamChannel):
value._log._arequest_more = fn
for transformer in self._transformers:
abind = getattr(transformer, "_bind_apump", None)
if abind is not None:
abind(fn)
def _register(self, transformer: StreamTransformer) -> None:
"""Register a single transformer.
Calls `transformer.init()`, stores the transformer for event
processing, binds any EventLog or StreamChannel instances in
the projection, and merges the projection into `extensions`.
"""
if transformer_requires_async(transformer) and not self._is_async:
raise RuntimeError(
f"{type(transformer).__name__} requires an async run — "
"it overrides aprocess/afinalize/afail or sets "
"requires_async=True. Use astream(), not stream()."
)
projection = transformer.init()
if not isinstance(projection, dict):
raise TypeError(
f"StreamTransformer.init() must return a dict, "
f"got {type(projection).__name__}"
)
conflicts = set(projection) & set(self.extensions)
if conflicts:
attributions = ", ".join(
f"{key!r} (owned by {self._projection_owners[key]})"
for key in sorted(conflicts)
)
raise ValueError(
f"Transformer {type(transformer).__name__} returned "
f"projection keys that conflict with already-registered "
f"keys: {attributions}"
)
self._transformers.append(transformer)
is_native = bool(getattr(transformer, "_native", False))
self._bind_and_wire(projection, is_native=is_native)
self.extensions.update(projection)
owner_name = type(transformer).__name__
for key in projection:
self._projection_owners[key] = owner_name
self._transformer_by_key[key] = transformer
if is_native:
self.native_keys.update(projection.keys())
on_register = getattr(transformer, "_on_register", None)
if on_register is not None:
on_register(self)
def transformer_by_key(self, key: str) -> StreamTransformer | None:
"""Return the transformer that owns the projection at `key`, if any."""
return self._transformer_by_key.get(key)
def push(self, event: ProtocolEvent) -> None:
"""Route an event through all transformers, then append to the main log.
Each transformer's `process()` is called in registration order
except when the transformer has `scope_exact = True` (the
default) and the event's namespace differs from the mux's
`scope`, in which case the transformer is skipped. Transformers
that need to see cross-scope events opt out by setting
`scope_exact = False` (e.g. `SubgraphTransformer`).
If any transformer returns False, the event is suppressed from
the main log, but transformers that already saw it keep their
side effects.
Seq is assigned right before an event enters the main log, not
before the transformer pipeline runs. This ensures that events
auto-forwarded from StreamChannels during `process()` get
earlier seq numbers than the original event, preserving
monotonic ordering in the log.
Args:
event: The protocol event to dispatch.
"""
ns = tuple(event["params"]["namespace"])
in_scope = ns == self.scope
keep = True
for transformer in self._transformers:
if transformer.scope_exact and not in_scope:
continue
if not transformer.process(event):
keep = False
if keep:
self._seq += 1
event["seq"] = self._seq
self._events.push(event)
def close(self) -> None:
"""Finalize all transformers, close all projections and the main log.
EventLogs and StreamChannels discovered in transformer
projections are auto-closed after `finalize()` runs
transformers don't need to close them manually. If any
transformer's `finalize()` raises, the remaining transformers,
projections, and the main log are still closed; the first error
is re-raised after cleanup completes.
Raises:
BaseException: The first error raised by a transformer's
`finalize()`, re-raised after cleanup finishes.
"""
first_error: BaseException | None = None
for transformer in self._transformers:
try:
transformer.finalize()
except BaseException as e:
if first_error is None:
first_error = e
for log in self._logs:
if not log._closed:
log.close()
for ch in self._channels:
if not ch._log._closed:
ch._close()
self._events.close()
if first_error is not None:
raise first_error
def fail(self, err: BaseException) -> None:
"""Fail all transformers, projections, and the main log.
EventLogs and StreamChannels discovered in transformer
projections are auto-failed transformers don't need to fail
them manually. If any transformer's `fail()` raises, the
remaining transformers, projections, and the main log are still
failed.
Args:
err: The exception that ended the run.
"""
for transformer in self._transformers:
try:
transformer.fail(err)
except BaseException:
pass
for log in self._logs:
if not log._closed:
log.fail(err)
for ch in self._channels:
if not ch._log._closed:
ch._fail(err)
self._events.fail(err)
# ------------------------------------------------------------------
# Async dispatch
# ------------------------------------------------------------------
async def apush(self, event: ProtocolEvent) -> None:
"""Dispatch an event on the async lane.
Awaits each transformer's `aprocess` in registration order
before appending to the main log except when the transformer
has `scope_exact = True` and the event's namespace differs from
`self.scope`, in which case it is skipped. A slow `aprocess`
serializes the pipeline by design that's the guarantee that
lets a later transformer (or a synchronous consumer) see the
result of the async work. For decoupled work, use `schedule()`
from inside `process` / `aprocess` instead.
The main log append is a non-blocking `push` matching v1's
`put_nowait` shape. Memory is bounded by caller pace via the
caller-driven pump; see `EventLog` for the full tradeoff story.
Args:
event: The protocol event to dispatch.
"""
ns = tuple(event["params"]["namespace"])
in_scope = ns == self.scope
keep = True
for transformer in self._transformers:
if transformer.scope_exact and not in_scope:
continue
if not await transformer.aprocess(event):
keep = False
if keep:
self._seq += 1
event["seq"] = self._seq
self._events.push(event)
async def aclose(self) -> None:
"""Finalize on the async lane.
Awaits every task started via `StreamTransformer.schedule()`
across all transformers, then calls `afinalize()` on each,
then auto-closes logs, channels, and the main event log.
If any scheduled task raised under `on_error="raise"`, or any
transformer's `afinalize` raises, the exception propagates.
The caller (the pump) handles it by routing into `afail`.
Raises:
BaseException: The first scheduled-task or `afinalize`
error, re-raised after cleanup.
"""
pending = self._collect_scheduled_tasks()
if pending:
results = await asyncio.gather(*pending, return_exceptions=True)
first_err = next(
(
r
for r in results
if isinstance(r, BaseException)
and not isinstance(r, asyncio.CancelledError)
),
None,
)
if first_err is not None:
raise first_err
first_error: BaseException | None = None
for transformer in self._transformers:
try:
await transformer.afinalize()
except BaseException as e:
if first_error is None:
first_error = e
for log in self._logs:
if not log._closed:
log.close()
for ch in self._channels:
if not ch._log._closed:
ch._close()
self._events.close()
if first_error is not None:
raise first_error
async def afail(self, err: BaseException) -> None:
"""Fail on the async lane.
Cancels every scheduled task across all transformers, awaits
them to completion, then runs each transformer's `afail` hook
and auto-fails logs, channels, and the main event log.
Args:
err: The exception that ended the run.
"""
pending = self._collect_scheduled_tasks()
for task in pending:
task.cancel()
if pending:
await asyncio.gather(*pending, return_exceptions=True)
for transformer in self._transformers:
try:
await transformer.afail(err)
except BaseException:
pass
for log in self._logs:
if not log._closed:
log.fail(err)
for ch in self._channels:
if not ch._log._closed:
ch._fail(err)
if not self._events._closed:
self._events.fail(err)
def _collect_scheduled_tasks(self) -> list[asyncio.Task[Any]]:
"""Return a snapshot of in-flight tasks scheduled via transformers."""
return [
task
for transformer in self._transformers
for task in getattr(transformer, "_stream_scheduled_tasks", ())
if not task.done()
]
# ------------------------------------------------------------------
# Binding and StreamChannel auto-wiring
# ------------------------------------------------------------------
def _bind_and_wire(
self, projection: dict[str, Any], *, is_native: bool = False
) -> None:
"""Bind and wire EventLog / StreamChannel instances in a projection.
`is_native` controls wire naming: native transformer channels
emit events with `method` equal to the channel name, while
non-native channels get a `custom:` prefix to keep user-defined
projections from colliding with built-in method names.
"""
for value in projection.values():
if isinstance(value, StreamChannel):
value._bind(is_async=self._is_async)
self._channels.append(value)
channel_name = value.name
def _make_forward(name: str, native: bool) -> Callable[[Any], None]:
def _forward(item: Any) -> None:
self._forward(name, item, native=native)
return _forward
value._wire(_make_forward(channel_name, is_native))
elif isinstance(value, EventLog):
value._bind(is_async=self._is_async)
self._logs.append(value)
def _forward(self, channel_name: str, item: Any, *, native: bool) -> None:
"""Inject a ProtocolEvent for a StreamChannel push.
Forwarded events bypass the transformer pipeline to avoid
infinite recursion (a transformer that pushes to a channel
during `process()` would re-trigger itself). These events are
visible in the main event log but are not passed through
transformers' `process()` methods.
Native transformers emit with `method` equal to the channel
name (e.g. `"lifecycle"`); non-native transformers get a
`custom:` prefix so user-defined projections can't collide
with built-in method names.
"""
self._seq += 1
method = channel_name if native else f"custom:{channel_name}"
event: ProtocolEvent = {
"type": "event",
"seq": self._seq,
"method": method,
"params": {
"namespace": [],
"timestamp": int(time.time() * 1000),
"data": item,
},
}
self._events.push(event)
-303
View File
@@ -1,303 +0,0 @@
from __future__ import annotations
import asyncio
import logging
from abc import ABC, abstractmethod
from collections.abc import Coroutine
from typing import Any, ClassVar, Literal
from typing_extensions import NotRequired, TypedDict
_logger = logging.getLogger(__name__)
class _ProtocolEventParams(TypedDict):
"""Parameters for a protocol event.
`timestamp` is wall-clock milliseconds since the epoch and can go
backwards across NTP adjustments use `ProtocolEvent.seq` for
ordering.
"""
namespace: list[str]
timestamp: int
data: Any
interrupts: NotRequired[tuple[Any, ...]]
class ProtocolEvent(TypedDict):
"""A protocol event emitted by the streaming infrastructure.
Wraps a raw stream part (values, messages, custom, etc.) in a uniform
envelope with a monotonic sequence number assigned by the StreamMux.
Consumers that need a total order across events should use `seq`, not
`params.timestamp` (which is wall-clock and not monotonic).
"""
type: Literal["event"]
eventId: NotRequired[str]
seq: NotRequired[int]
method: str # StreamMode value: "values", "messages", "custom", etc.
params: _ProtocolEventParams
class StreamTransformer(ABC):
"""Extension point for custom stream projections.
Transformers observe protocol events flowing through the StreamMux and
build typed derived projections (EventLogs, StreamChannels, promises,
etc.).
Set `_native = True` on a transformer to have its projection keys
exposed as direct attributes on the run stream (in addition to
appearing in `run.extensions`).
Subclasses must implement `init` and override at least one of
`process` / `aprocess`. The `finalize` / `afinalize` and `fail` /
`afail` hooks are optional the default implementations are no-ops.
EventLog and StreamChannel instances in the projection dict are
auto-closed / auto-failed by the mux, so most transformers don't
need `finalize` or `fail` at all.
Transformers that need async work pick the async lane by:
1. Overriding `aprocess` (and optionally `afinalize` / `afail`), or
2. Calling `self.schedule(coro)` from inside a sync `process`, or
3. Setting `requires_async = True` explicitly.
The mux detects these cases at registration and raises if they're
used under sync `stream()` they only work under `astream()`.
Use `aprocess` when the pump must wait for async work before the
next transformer sees the event (e.g. PII redaction that mutates
`event` in place). Use `schedule()` for decoupled async work whose
result lands on an independent projection (e.g. async moderation
scoring, cost lookup, external tracing).
Attributes:
scope: Namespace the transformer operates within `()` for the
root mux, a subgraph's namespace tuple inside a mini-mux.
Set at construction from the mux's scope (each factory is
called as `factory(scope)`). Transformers that only care
about events at their own namespace compare against
`self.scope`; subgraph-aware transformers can treat it as
a parent path.
scope_exact: If True (the default), the mux only calls
`process` / `aprocess` for events whose namespace equals
`self.scope` user transformers get scope-scoped events
for free with no boilerplate. Set False for transformers
that need to see events across scopes (e.g.
`SubgraphTransformer` forwards deeper events into child
mini-muxes).
requires_async: Explicit opt-in for transformers that need a
running event loop but don't override any async method (for
example, transformers that call `schedule()` from a sync
`process`). The mux also auto-detects the async lane when
`aprocess`, `afinalize`, or `afail` is overridden.
"""
requires_async: ClassVar[bool] = False
scope_exact: ClassVar[bool] = True
def __init__(self, scope: tuple[str, ...] = ()) -> None:
"""Initialize the transformer with its mux's scope.
Args:
scope: The namespace tuple the owning mux is scoped to.
`()` for the root, the subgraph's namespace inside a
mini-mux. Factories receive this at construction time
(`factory(scope)` in `StreamMux`).
"""
self.scope: tuple[str, ...] = scope
@abstractmethod
def init(self) -> dict[str, Any]:
"""Return the projection dict.
Keys become entries in `run.extensions`. If the transformer has
`_native = True`, keys are also set as direct attributes on the
run stream.
StreamChannel instances in the return value are automatically
wired by the StreamMux for protocol event auto-forwarding.
"""
...
def process(self, event: ProtocolEvent) -> bool:
"""Handle an event on the sync lane.
Called for every event before it is appended to the main event
log. Subclasses must override either `process` or `aprocess`.
The default raises so a missing override fails loudly rather
than silently passing every event through.
Args:
event: The protocol event to observe.
Returns:
True to keep the event in the main log, False to suppress it.
"""
raise NotImplementedError(
f"{type(self).__name__} must override process() or aprocess()"
)
async def aprocess(self, event: ProtocolEvent) -> bool:
"""Handle an event on the async lane.
The mux awaits this before dispatching to the next transformer,
so a slow `aprocess` serializes the pipeline. Use it only when
a later transformer or a consumer reading the event
synchronously must see the result of the async work (e.g.
PII redaction that mutates `event` in place).
The default delegates to `process`, so purely-sync transformers
run unchanged under `astream()`.
Args:
event: The protocol event to observe.
Returns:
True to keep the event in the main log, False to suppress it.
"""
return self.process(event)
def finalize(self) -> None:
"""Called when the run ends normally (sync lane).
Override to close EventLogs, resolve promises, or perform other
teardown. StreamChannel instances are auto-closed by the mux.
"""
async def afinalize(self) -> None:
"""Called when the run ends normally (async lane).
By the time this runs, the mux has already awaited every task
started via `schedule()`, so EventLogs can be closed here
without a last-task-wins race.
The default delegates to `finalize`.
"""
self.finalize()
def fail(self, err: BaseException) -> None:
"""Called when the run ends with an error (sync lane).
Override to fail EventLogs, reject promises, or perform other
teardown. StreamChannel instances are auto-failed by the mux.
Args:
err: The exception that ended the run.
"""
async def afail(self, err: BaseException) -> None:
"""Called when the run ends with an error (async lane).
The mux cancels and awaits every task started via `schedule()`
before calling this, so cleanup doesn't race with in-flight work.
The default delegates to `fail`.
Args:
err: The exception that ended the run.
"""
self.fail(err)
# ------------------------------------------------------------------
# Scheduled async work
# ------------------------------------------------------------------
def schedule(
self,
coro: Coroutine[Any, Any, Any],
*,
on_error: Literal["log", "raise"] = "log",
) -> asyncio.Task[Any]:
"""Schedule a coroutine tied to this transformer's lifecycle.
The mux holds the task reference, awaits all scheduled tasks
during `aclose()` before calling `afinalize()`, and cancels
them on `afail()`. Authors don't need to track tasks or
implement the last-task-closes-the-log dance.
Requires a running event loop call only under `astream()`.
Set `requires_async = True` on the class so registration under
sync `stream()` fails fast with a clear message.
Args:
coro: The coroutine to run. Its lifecycle is owned by the
mux from this point on.
on_error: `"log"` (default) catches and logs any exception
the coroutine raises, so a single failure doesn't tear
down the run. `"raise"` lets the exception propagate
when the mux joins pendings, converting the close path
into the fail path.
Returns:
The asyncio Task. Authors rarely need to await it directly
consumers read results from whatever projection the
coroutine pushes into.
Raises:
RuntimeError: If called without a running event loop (i.e.
under sync `stream()` rather than `astream()`).
"""
try:
asyncio.get_running_loop()
except RuntimeError:
raise RuntimeError(
f"{type(self).__name__}.schedule() requires a running "
"event loop; this transformer must run under astream(), "
"not stream(). Set requires_async=True on the class so "
"this fails at registration rather than at first event."
) from None
wrapped = self._wrap_scheduled(coro) if on_error == "log" else coro
task = asyncio.create_task(wrapped)
tasks = self._scheduled_task_set()
tasks.add(task)
task.add_done_callback(tasks.discard)
return task
@staticmethod
async def _wrap_scheduled(coro: Coroutine[Any, Any, Any]) -> Any:
try:
return await coro
except asyncio.CancelledError:
raise
except BaseException:
_logger.exception("Scheduled StreamTransformer task failed")
def _scheduled_task_set(self) -> set[asyncio.Task[Any]]:
"""Return the lazily-allocated task set.
Avoids requiring subclasses to call `super().__init__()`.
"""
tasks: set[asyncio.Task[Any]] | None = getattr(
self, "_stream_scheduled_tasks", None
)
if tasks is None:
tasks = set()
self._stream_scheduled_tasks = tasks
return tasks
def transformer_requires_async(transformer: StreamTransformer) -> bool:
"""Return True if the transformer needs a running event loop.
A transformer requires async if it explicitly opts in
(`requires_async = True`) or overrides any of the async-lane methods
(`aprocess`, `afinalize`, `afail`).
Args:
transformer: The transformer to inspect.
Returns:
True if the transformer cannot run under sync `stream()`.
"""
if transformer.requires_async:
return True
cls = type(transformer)
for name in ("aprocess", "afinalize", "afail"):
if getattr(cls, name) is not getattr(StreamTransformer, name):
return True
return False
@@ -1,408 +0,0 @@
from __future__ import annotations
import asyncio
from collections.abc import AsyncIterator, Awaitable, Callable, Iterator, Mapping
from types import MappingProxyType, TracebackType
from typing import TYPE_CHECKING, Any
from langgraph.stream._convert import convert_to_protocol_event
from langgraph.stream._mux import StreamMux
from langgraph.stream._types import ProtocolEvent
if TYPE_CHECKING:
from langgraph.stream.transformers import ValuesTransformer
def _drive_until_done(pump: Callable[[], bool]) -> None:
"""Call the sync pump until it returns False."""
while pump():
pass
async def _adrive_until_done(pump: Callable[[], Awaitable[bool]]) -> None:
"""Call the async pump until it returns False."""
while await pump():
pass
class BaseRunStream:
"""Shared shape for any object that wraps a `StreamMux`.
Root (`GraphRunStream` / `AsyncGraphRunStream`) and scoped
(`SubgraphRunStream`) streams both compose a `StreamMux`. The mux
owns the projections `values`, `messages`, `subgraphs`, and any
user-registered keys all exposed via `extensions`. Native
projections (`_native = True`) are also bound as direct attributes
(`run.values`, `run.messages`, ) for ergonomics.
Raw iteration (`for event in run` / `async for event in run`) and
the `interleave(...)` helper both live here so every subclass
behaves consistently. Subclasses only add pump ownership, scope
metadata, or sync/async flavor.
"""
def __init__(self, mux: StreamMux) -> None:
self._mux = mux
self.extensions: Mapping[str, Any] = MappingProxyType(mux.extensions)
for key in mux.native_keys:
setattr(self, key, mux.extensions[key])
@property
def _values_transformer(self) -> ValuesTransformer:
"""Look up the `ValuesTransformer` registered on this mux.
`output` / `interrupted` / `interrupts` need scalar state from
the `ValuesTransformer` without threading it through the
constructor. Raises if none is registered `stream_v2` /
`astream_v2` always register one, so hitting this path means
the caller assembled the mux themselves and forgot.
"""
from langgraph.stream.transformers import ValuesTransformer
for t in self._mux._transformers:
if isinstance(t, ValuesTransformer):
return t
raise RuntimeError(
"No ValuesTransformer is registered on this mux — "
"`output` / `interrupted` / `interrupts` are unavailable."
)
def __iter__(self) -> Iterator[ProtocolEvent]:
"""Sync iteration of protocol events on this mux's main log.
Raises at the EventLog level if the mux is async-bound.
"""
return iter(self._mux._events)
def __aiter__(self) -> AsyncIterator[ProtocolEvent]:
"""Async iteration of protocol events on this mux's main log.
Raises at the EventLog level if the mux is sync-bound.
"""
return self._mux._events.__aiter__()
def interleave(self, *names: str) -> Iterator[tuple[str, Any]]:
"""Iterate multiple projections round-robin, yielding ``(name, item)``.
Each turn advances one projection's cursor; when a cursor's
buffer is empty, pulling from it drives the pump once, which
fans out to every subscribed projection log. Projections whose
items aren't consumed on this turn sit in their own buffers
only until the next turn reaches them, bounding memory by the
skew between projection rates rather than letting any single
log grow to the full run length.
Projections are exhausted independently; a projection that
finishes early drops out of the rotation while others
continue. The overall iterator ends once all named projections
are done.
Args:
*names: Projection keys to interleave. Must match keys in
`extensions`.
Yields:
`(name, item)` tuples in round-robin order across the named
projections.
Raises:
KeyError: If a name doesn't match a registered projection.
Example:
```python
for name, item in run.interleave("messages", "values"):
if name == "messages":
print("msg:", item)
else:
print("val:", item)
```
"""
cursors: dict[str, Iterator[Any]] = {
name: iter(self.extensions[name]) for name in names
}
done: set[str] = set()
while len(done) < len(cursors):
for name, cursor in cursors.items():
if name in done:
continue
try:
item = next(cursor)
except StopIteration:
done.add(name)
continue
yield (name, item)
class GraphRunStream(BaseRunStream):
"""Sync run stream with caller-driven pumping.
The caller's iteration on any projection (`values`, `messages`,
raw events, or `output`) drives the graph forward. No background
thread is used the caller's `for` loop is the pump.
Projections are single-consumer iterating `run.values` twice
raises. Use `projection.tee(n)` if you genuinely need fan-out.
"""
def __init__(
self,
graph_iter: Iterator[Any],
mux: StreamMux,
) -> None:
"""Initialize the run stream.
Args:
graph_iter: Pull-based iterator over the graph's stream.
mux: The StreamMux owning projections and the main log.
Must have a `ValuesTransformer` registered for
`output` / `interrupted` / `interrupts` to work.
"""
super().__init__(mux)
self._graph_iter = graph_iter
self._exhausted = False
mux.bind_pump(self._pump_next)
def _pump_next(self) -> bool:
"""Pull one event from the graph and push it through the mux.
Returns:
True if an event was pulled, False if the graph is
exhausted or has raised.
"""
if self._exhausted:
return False
try:
part = next(self._graph_iter)
except StopIteration:
self._mux.close()
self._exhausted = True
return False
except Exception as e:
self._mux.fail(e)
self._exhausted = True
return False
self._mux.push(convert_to_protocol_event(part))
return True
def abort(self) -> None:
"""Stop the run early.
Closes the mux and marks the stream exhausted. The graph
iterator is dropped; any in-flight nodes see the closure on
their next yield point. Idempotent.
"""
if self._exhausted:
return
self._exhausted = True
try:
self._mux.close()
except Exception:
pass
def __enter__(self) -> GraphRunStream:
return self
def __exit__(
self,
exc_type: type[BaseException] | None,
exc: BaseException | None,
tb: TracebackType | None,
) -> None:
self.abort()
@property
def output(self) -> dict[str, Any] | None:
"""Drive the run to completion and return the final state."""
_drive_until_done(self._pump_next)
err = self._values_transformer.error
if err is not None:
raise err
return self._values_transformer._latest
@property
def interrupted(self) -> bool:
"""Drive the run to completion, then return whether it was interrupted.
Raises:
BaseException: If the run ended with an error.
"""
_drive_until_done(self._pump_next)
err = self._values_transformer.error
if err is not None:
raise err
return self._values_transformer._interrupted
@property
def interrupts(self) -> list[Any]:
"""Drive the run to completion, then return interrupt payloads.
Raises:
BaseException: If the run ended with an error.
"""
_drive_until_done(self._pump_next)
err = self._values_transformer.error
if err is not None:
raise err
return self._values_transformer._interrupts
class AsyncGraphRunStream(BaseRunStream):
"""Async run stream with caller-driven pumping.
Async iteration on any projection drives the graph forward there
is no background task. Concurrent consumers share a single-flight
pump via an `asyncio.Lock`, so each awaiting cursor contributes
one event per acquisition. Backpressure comes from the logs: when
a subscribed log's buffer reaches `maxlen`, `apush` awaits the
subscriber to drain, which holds back the pump and paces the
graph.
Projections are single-consumer a second `aiter(run.values)`
raises. Use `projection.tee(n)` for fan-out.
Use as an async context manager to guarantee clean shutdown on
early exit:
```python
async with await handler.astream(input) as run:
async for msg in run.messages:
...
```
"""
def __init__(
self,
graph_aiter: AsyncIterator[Any],
mux: StreamMux,
) -> None:
"""Initialize the async run stream.
Args:
graph_aiter: Async iterator over the graph's stream.
mux: The StreamMux owning projections and the main log.
Must have a `ValuesTransformer` registered for
`output` / `interrupted` / `interrupts` to work.
"""
super().__init__(mux)
self._graph_aiter = graph_aiter
self._exhausted = False
self._pump_cond = asyncio.Condition()
self._pumping = False
mux.bind_apump(self._apump_next)
async def _apump_next(self) -> bool:
"""Drive one pump step, or wait for the active pumper to drive one.
"Take-a-number" semantics: at most one task at a time calls
`graph_aiter.__anext__()` (asyncio iterators can't be advanced
concurrently). Other callers wait on a Condition that the
active pumper notifies after each step. This lets a "passive"
consumer one whose projection's buffer is being filled by the
active pumper's push — wake up as soon as its data lands,
instead of queueing on the pump and only observing its data one
graph event late.
`except Exception` is intentional `CancelledError` and other
`BaseException` subclasses propagate, matching asyncio's
cancellation contract.
Returns:
True if a pump step completed (by this task or another),
False if the graph is exhausted.
"""
async with self._pump_cond:
if self._exhausted:
return False
if self._pumping:
# Another task is pumping; wait for its progress signal.
await self._pump_cond.wait()
return not self._exhausted
self._pumping = True
try:
try:
part = await self._graph_aiter.__anext__()
except StopAsyncIteration:
self._exhausted = True
await self._mux.aclose()
return False
except Exception as e:
self._exhausted = True
await self._mux.afail(e)
return False
await self._mux.apush(convert_to_protocol_event(part))
return True
finally:
async with self._pump_cond:
self._pumping = False
self._pump_cond.notify_all()
async def abort(self) -> None:
"""Stop the run early.
Marks the stream exhausted, wakes any pump-waiters, and closes
the mux. Any `apush` blocked on backpressure wakes and returns
without appending. Idempotent.
"""
async with self._pump_cond:
if self._exhausted:
return
self._exhausted = True
self._pump_cond.notify_all()
try:
await self._mux.aclose()
except Exception:
pass
async def __aenter__(self) -> AsyncGraphRunStream:
return self
async def __aexit__(
self,
exc_type: type[BaseException] | None,
exc: BaseException | None,
tb: TracebackType | None,
) -> None:
await self.abort()
async def output(self) -> dict[str, Any] | None:
"""Drive the run to completion and return the final state.
Methods (not properties) on the async lane so `run.output`
without `await` raises at type-check time instead of silently
yielding a coroutine object.
Example:
```python
output = await run.output()
```
Raises:
BaseException: If the run ended with an error.
"""
await _adrive_until_done(self._apump_next)
if (err := self._values_transformer.error) is not None:
raise err
return self._values_transformer._latest
async def interrupted(self) -> bool:
"""Drive the run to completion and return whether it was interrupted.
Raises:
BaseException: If the run ended with an error.
"""
await _adrive_until_done(self._apump_next)
if (err := self._values_transformer.error) is not None:
raise err
return self._values_transformer._interrupted
async def interrupts(self) -> list[Any]:
"""Drive the run to completion and return interrupt payloads.
Raises:
BaseException: If the run ended with an error.
"""
await _adrive_until_done(self._apump_next)
if (err := self._values_transformer.error) is not None:
raise err
return self._values_transformer._interrupts
@@ -1,115 +0,0 @@
from __future__ import annotations
from collections.abc import AsyncIterator, Callable, Iterator
from typing import Generic, TypeVar
from langgraph.stream._event_log import EventLog
T = TypeVar("T")
class StreamChannel(Generic[T]):
"""A named projection channel with optional protocol auto-forwarding.
Wraps an event log and declares a protocol channel name. When the
StreamMux detects a StreamChannel in a transformer's `init()`
return value, it automatically wires every `push()` to inject a
`ProtocolEvent` into the main event stream using the channel's
name as the method.
Auto-forwarded events bypass the transformer pipeline other
transformers' `process()` / `aprocess()` methods do not see
`custom:<name>` events produced by a channel push. This prevents a
transformer that pushes to its own channel during `process()` from
re-triggering itself, but it also means filter- or tap-style
transformers cannot observe channel output from peer transformers.
Consumers that need that should iterate the main event stream.
In-process consumers iterate the channel directly (`for item in ch`
or `async for item in ch`). Remote SDK clients subscribe via
`session.subscribe("custom:<channelName>")`.
Like EventLog, a StreamChannel starts unbound. The mux calls
`_bind(is_async)` during registration so the correct iteration
protocol is available by the time user code sees it.
Lifecycle (`_close` / `_fail`) is managed by the mux transformers
using only StreamChannels don't need `finalize` or `fail` hooks.
"""
def __init__(
self, name: str, *, maxlen: int | None = None, retain: bool = False
) -> None:
"""Initialize the channel with an empty inner log.
Args:
name: The protocol channel name used for auto-forwarded
events (`custom:<name>` on the wire).
maxlen: Optional retention cap on the inner EventLog. See
`EventLog.__init__` for semantics.
retain: If True, the inner log retains pushes before any
consumer subscribes needed for channels whose
consumer iterates after events have already flowed
(e.g. `lifecycle` inspected after draining `values`).
"""
self.name = name
self._log: EventLog[T] = EventLog(maxlen=maxlen, retain=retain)
self._wire_fn: Callable[[T], None] | None = None
def _bind(self, *, is_async: bool) -> None:
"""Bind the underlying event log to sync or async mode.
Args:
is_async: True for async iteration, False for sync.
"""
self._log._bind(is_async=is_async)
def push(self, item: T) -> None:
"""Append an item to the log and auto-forward if wired.
Args:
item: The item to push.
"""
self._log.push(item)
if self._wire_fn is not None:
self._wire_fn(item)
# ------------------------------------------------------------------
# Mux lifecycle hooks (not called by transformers directly)
# ------------------------------------------------------------------
def _wire(self, fn: Callable[[T], None]) -> None:
"""Install the auto-forward callback (called by StreamMux)."""
self._wire_fn = fn
def _close(self) -> None:
"""Close the underlying log (called by StreamMux on run end)."""
self._log.close()
def _fail(self, err: BaseException) -> None:
"""Fail the underlying log (called by StreamMux on run error)."""
self._log.fail(err)
# ------------------------------------------------------------------
# Iteration — delegates to the inner event log (multi-cursor)
# ------------------------------------------------------------------
def __iter__(self) -> Iterator[T]:
return iter(self._log)
def __aiter__(self) -> AsyncIterator[T]:
return self._log.__aiter__()
def tee(self, n: int = 2) -> tuple[Iterator[T], ...]:
"""Fan out the channel into `n` independent sync iterators.
Delegates to the underlying EventLog's `tee()`.
"""
return self._log.tee(n)
def atee(self, n: int = 2) -> tuple[AsyncIterator[T], ...]:
"""Fan out the channel into `n` independent async iterators.
Delegates to the underlying EventLog's `atee()`.
"""
return self._log.atee(n)
@@ -1,585 +0,0 @@
from __future__ import annotations
import logging
from typing import TYPE_CHECKING, Any, Literal, cast
from langchain_core.language_models._compat_bridge import message_to_events
from langchain_core.language_models.chat_model_stream import (
AsyncChatModelStream,
ChatModelStream,
)
from langchain_core.messages import AIMessageChunk, BaseMessage
from langchain_protocol.protocol import CheckpointRef, MessagesData
from typing_extensions import TypedDict
from langgraph.errors import GraphInterrupt
from langgraph.stream._event_log import EventLog
from langgraph.stream._types import ProtocolEvent, StreamTransformer
from langgraph.stream.run_stream import BaseRunStream
from langgraph.stream.stream_channel import StreamChannel
if TYPE_CHECKING:
from collections.abc import Awaitable, Callable
from langgraph.stream._mux import StreamMux
logger = logging.getLogger(__name__)
SubgraphStatus = Literal["started", "completed", "failed", "interrupted"]
_TERMINAL_STATUSES: frozenset[SubgraphStatus] = frozenset(
{"completed", "failed", "interrupted"}
)
def _is_new_direct_child(
ns: tuple[str, ...],
scope: tuple[str, ...],
seen: set[tuple[str, ...]] | dict[tuple[str, ...], Any],
) -> bool:
"""Return True iff `ns` is a direct child of `scope` not yet seen.
Shared by `SubgraphTransformer` (in-process handle discovery) and
`LifecycleTransformer` (wire event emission) so the two can't
disagree on what counts as a new subgraph.
"""
return len(ns) == len(scope) + 1 and ns[:-1] == scope and ns not in seen
def _parse_ns_segment(segment: str) -> tuple[str, str | None]:
"""Split `node_name:task_id` into (node_name, task_id).
Task ids are present when Pregel spawned the subgraph as a task;
absent on synthesized namespaces (tests, hand-crafted events).
"""
node_name, sep, task_id = segment.partition(":")
if not sep:
return segment, None
return node_name, task_id or None
class ValuesTransformer(StreamTransformer):
"""Capture values events as a drainable stream of state snapshots.
Keeps `_latest` / `_interrupted` / `_interrupts` as scalar state
regardless of whether the log has a subscriber so `run.output()`
and `run.interrupted` work without forcing the caller to iterate
`run.values`. Log pushes are silent no-ops when unsubscribed.
Native transformer projection keys are exposed as direct
attributes on the run stream (e.g. `run.values`).
`scope` (inherited from `StreamTransformer`) is the namespace the
transformer captures values for. `()` matches the root graph;
subgraph mini-muxes pass their subgraph's namespace, so each
instance sees only its own level.
"""
_native = True
def __init__(self, scope: tuple[str, ...] = ()) -> None:
super().__init__(scope)
self._log: EventLog[dict[str, Any]] = EventLog()
self._latest: dict[str, Any] | None = None
self._interrupted = False
self._interrupts: list[Any] = []
def init(self) -> dict[str, Any]:
return {"values": self._log}
@property
def error(self) -> BaseException | None:
"""The error that ended the run, or `None` if it succeeded.
Set by the mux when it auto-fails the projection log.
"""
return self._log._error
def process(self, event: ProtocolEvent) -> bool:
# Namespace filtering is handled by the mux via `scope_exact`.
if event["method"] != "values":
return True
params = event["params"]
self._latest = params["data"]
interrupts = params.get("interrupts", ())
if interrupts:
self._interrupted = True
self._interrupts.extend(interrupts)
self._log.push(params["data"])
return True
class MessagesTransformer(StreamTransformer):
"""Capture messages events as ChatModelStream objects.
The messages projection yields one `ChatModelStream` (or
`AsyncChatModelStream`) per LLM call. Consumers iterate
`run.messages` to get stream handles, then use each handle's typed
projections (`.text`, `.reasoning`, `.tool_calls`, `.usage`,
`.output`) for per-message content.
Two input shapes are handled (via `params["data"] = (payload,
metadata)` from `StreamMessagesHandler`):
1. Protocol event (dict with `"event"` key) emitted by
`stream_v2()` / `astream_v2()` via the `on_stream_event`
callback. Routed to an existing `ChatModelStream` by
`metadata["run_id"]`. A `message-start` event creates a new
stream; `message-finish` closes it.
2. Whole `AIMessage` emitted from `on_chain_end` when a node
returns a finalized message. Replayed as a synthetic protocol
event lifecycle via `message_to_events`, then the
already-complete stream is pushed to the log.
V1 `AIMessageChunk` tuples (from `on_llm_new_token`) are not
streamed into this projection: chat models that want to populate
`run.messages` with content-block streaming must use
`stream_v2()` / `astream_v2()`. Models called via the legacy
`stream()` method still surface their final `AIMessage` via
`on_chain_end` when a node returns it as state.
`scope` (inherited from `StreamTransformer`) is the namespace the
transformer captures messages for. `()` matches the root graph;
subgraph mini-muxes pass their subgraph's namespace, so each
instance sees only its own level.
Native transformer the `messages` projection is exposed as a
direct attribute on the run stream.
"""
_native = True
def __init__(self, scope: tuple[str, ...] = ()) -> None:
super().__init__(scope)
self._log: EventLog[ChatModelStream] = EventLog()
# Correlate protocol events back to a ChatModelStream by run_id
# (attached to the event's metadata by StreamMessagesHandler).
self._by_run: dict[str, ChatModelStream] = {}
self._pump_fn: Callable[[], bool] | None = None
self._apump_fn: Callable[[], Awaitable[bool]] | None = None
def init(self) -> dict[str, Any]:
return {"messages": self._log}
def _bind_pump(self, fn: Callable[[], bool]) -> None:
"""Wire the sync pull callback. Called by GraphRunStream._wire_request_more."""
self._pump_fn = fn
def _bind_apump(self, fn: Callable[[], Awaitable[bool]]) -> None:
"""Wire the async pull callback.
Called by `AsyncGraphRunStream._wire_arequest_more` so each
`AsyncChatModelStream` this transformer creates can drive the
shared graph pump from its projection cursors.
"""
self._apump_fn = fn
def _make_stream(
self,
*,
namespace: list[str],
node: str | None,
message_id: str | None,
) -> ChatModelStream:
"""Create a ChatModelStream (sync) or AsyncChatModelStream (async).
Wires whichever pump is bound. Prefers the async pump so nested
iteration under `AsyncGraphRunStream` drives the graph forward
without a background task. The unwired fallback (no pump bound)
is used by unit tests that dispatch events manually.
"""
if self._apump_fn is not None:
astream = AsyncChatModelStream(
namespace=namespace,
node=node,
message_id=message_id,
)
astream.set_arequest_more(self._apump_fn)
return astream
if self._pump_fn is not None:
stream: ChatModelStream = ChatModelStream(
namespace=namespace,
node=node,
message_id=message_id,
)
stream.set_request_more(self._pump_fn)
return stream
return AsyncChatModelStream(
namespace=namespace,
node=node,
message_id=message_id,
)
def process(self, event: ProtocolEvent) -> bool:
# Namespace filtering is handled by the mux via `scope_exact`.
if event["method"] != "messages":
return True
params = event["params"]
payload, metadata = params["data"]
node: str | None = metadata.get("langgraph_node")
run_id = str(metadata.get("run_id", "")) if metadata else ""
if isinstance(payload, dict) and "event" in payload:
self._route_protocol_event(
cast("MessagesData", payload), run_id=run_id, node=node
)
elif isinstance(payload, BaseMessage) and not isinstance(
payload, AIMessageChunk
):
self._route_whole_message(payload, node=node)
# Legacy AIMessageChunk tuples (from on_llm_new_token) are ignored;
# v1 streaming callers must switch to stream_v2() to populate this
# projection.
return True
def _route_protocol_event(
self,
event: MessagesData,
*,
run_id: str,
node: str | None,
) -> None:
event_type = event.get("event")
if event_type == "message-start":
message_id = event.get("message_id")
stream = self._make_stream(
namespace=list(self.scope),
node=node,
message_id=str(message_id) if message_id is not None else None,
)
self._by_run[run_id] = stream
self._log.push(stream)
stream.dispatch(event)
elif run_id in self._by_run:
stream = self._by_run[run_id]
stream.dispatch(event)
if event_type == "message-finish":
del self._by_run[run_id]
def _route_whole_message(self, message: BaseMessage, *, node: str | None) -> None:
stream = self._make_stream(
namespace=list(self.scope),
node=node,
message_id=message.id,
)
for evt in message_to_events(message, message_id=message.id):
stream.dispatch(evt)
self._log.push(stream)
def finalize(self) -> None:
"""Clear any routing state — streams close themselves via `message-finish`."""
self._by_run.clear()
def fail(self, err: BaseException) -> None:
"""Propagate run error to any streams still open when the graph fails."""
for stream in list(self._by_run.values()):
stream.fail(err)
self._by_run.clear()
class SubgraphRunStream(BaseRunStream):
"""Scoped view of a single nested subgraph execution.
Yielded on `run.subgraphs` (or `parent.subgraphs` for grandchildren)
when a nested `Pregel` spawns. Wraps a mini-`StreamMux` built with
the same transformer factories as the root mux, so `.values`,
`.messages`, `.subgraphs` are populated by the standard
transformers scoped to this handle's namespace — no duplicated
routing logic. The mini-mux borrows the root's pump via
`make_child`'s pump inheritance, so any cursor on a subagent
projection drives the whole run forward.
Handle fields:
- `path`: the namespace tuple stable for the life of the handle.
- `graph_name` / `trigger_call_id`: parsed from the namespace
segment at discovery (`node_name:task_id`).
- `status`: `started` on discovery; advances to `completed` when
the parent mux closes, or `failed` / `interrupted` when it
errors.
- `error`: set on terminal error.
- `checkpoint`: unused by the current discovery path kept for
compatibility with consumers that inspect it.
`.output` is a snapshot of the latest values seen at this
namespace it doesn't drive the pump (unlike root's
`GraphRunStream.output`), because advancing a subgraph to
completion is only meaningful as part of advancing the whole run.
"""
def __init__(
self,
path: tuple[str, ...],
mux: StreamMux,
*,
graph_name: str | None = None,
trigger_call_id: str | None = None,
) -> None:
super().__init__(mux)
self.path: tuple[str, ...] = path
self.graph_name: str | None = graph_name
self.trigger_call_id: str | None = trigger_call_id
self.status: SubgraphStatus = "started"
self.error: str | None = None
self.checkpoint: CheckpointRef | None = None
@property
def output(self) -> dict[str, Any] | None:
"""Latest values snapshot at this namespace, or `None`.
Snapshot-only iterating other projections or the root's
`.output` is what drives the pump.
"""
values_t = self._mux.transformer_by_key("values")
if isinstance(values_t, ValuesTransformer):
return values_t._latest
return None
class SubgraphTransformer(StreamTransformer):
"""Discover subgraphs and route events into per-subgraph mini-muxes.
Thin dispatcher. At its own `scope` (inherited from
`StreamTransformer`, determined by the enclosing mux), it watches
for the first event at exactly one namespace level deeper to
discover a direct child. Each discovered child gets its own
`SubgraphRunStream` backed by a mini-`StreamMux` built via
`parent_mux.make_child(path)`, so the same factory list produces
fresh transformer instances at the child's scope.
Every incoming event that falls under one of the direct children
(ns starts with a child's `path`) is forwarded into that child's
mini-mux via `push`. The standard transformers in that mini-mux
(`ValuesTransformer`, `MessagesTransformer`, and another
`SubgraphTransformer` for grandchildren) handle the rest. No
duplicated routing or assembly logic.
Discovery is method-agnostic: the first event of any mode whose
namespace places it directly below `scope` spawns the handle.
`graph_name` and `trigger_call_id` are parsed from the namespace
segment, which encodes `node_name:task_id`.
Terminal status for each handle is set by the parent mux's
`close` / `fail` path. `finalize` transitions still-open handles
to `completed`; `fail` transitions them to `failed` or
`interrupted` depending on the error.
Native transformer `subgraphs` exposes the direct-children log.
`scope_exact = False`: this transformer sees events at any
namespace, because it forwards out-of-scope events to the matching
direct-child mini-mux.
"""
_native = True
scope_exact = False
def __init__(self, scope: tuple[str, ...] = ()) -> None:
super().__init__(scope)
self._root_log: EventLog[SubgraphRunStream] = EventLog()
# Direct children only (namespace = scope + one segment).
self._by_ns: dict[tuple[str, ...], SubgraphRunStream] = {}
self._mux: StreamMux | None = None
def init(self) -> dict[str, Any]:
return {"subgraphs": self._root_log}
def _on_register(self, mux: StreamMux) -> None:
"""Capture the enclosing mux so we can build child mini-muxes."""
self._mux = mux
def process(self, event: ProtocolEvent) -> bool:
ns = tuple(event["params"]["namespace"])
depth = len(self.scope)
# 1. Discover: first-seen direct-child namespace registers a
# handle. Any event method triggers discovery — no dedicated
# channel.
if _is_new_direct_child(ns, self.scope, self._by_ns):
self._on_started(ns)
# 2. Forward the event to the matching direct-child mini-mux.
# Prefix-match: ns must start with some child's path.
direct_child_ns = ns[: depth + 1] if len(ns) > depth else None
if direct_child_ns is not None and direct_child_ns in self._by_ns:
self._by_ns[direct_child_ns]._mux.push(event)
return True
def _on_started(self, ns: tuple[str, ...]) -> None:
# `_on_register` is called by the mux during registration, which
# happens before any event can be dispatched — so this should
# always be set by the time we process an event.
assert self._mux is not None, (
"SubgraphTransformer processed an event before _on_register; "
"transformer registration ordering is broken."
)
graph_name, trigger_call_id = _parse_ns_segment(ns[-1])
child_mux = self._mux.make_child(ns)
handle = SubgraphRunStream(
path=ns,
mux=child_mux,
graph_name=graph_name,
trigger_call_id=trigger_call_id,
)
self._by_ns[ns] = handle
self._root_log.push(handle)
@staticmethod
def _close_handle_mux(handle: SubgraphRunStream) -> None:
# Idempotent close — mux.close() runs finalize on its transformers
# (which cascades through grandchildren) and closes projection logs.
if not handle._mux._events._closed:
try:
handle._mux.close()
except Exception:
logger.warning(
"Error closing subgraph mini-mux at %s; subscribers "
"may not see a clean close.",
handle.path,
exc_info=True,
)
def finalize(self) -> None:
"""Transition any still-open direct children to `completed`.
Subgraph interrupts surface as a values event with a populated
`interrupts` field rather than as an exception at the graph
boundary the parent pump exhausts normally and `finalize`
runs the close path. Inspect each child's `ValuesTransformer`
to distinguish "completed cleanly" from "interrupted".
"""
for handle in self._by_ns.values():
if handle.status not in _TERMINAL_STATUSES:
values_t = handle._mux.transformer_by_key("values")
if isinstance(values_t, ValuesTransformer) and values_t._interrupted:
handle.status = "interrupted"
else:
handle.status = "completed"
self._close_handle_mux(handle)
def fail(self, err: BaseException) -> None:
"""Transition any still-open direct children to `failed` / `interrupted`."""
is_interrupt = isinstance(err, GraphInterrupt)
terminal: SubgraphStatus = "interrupted" if is_interrupt else "failed"
error_str = None if is_interrupt else str(err)
for handle in self._by_ns.values():
if handle.status not in _TERMINAL_STATUSES:
handle.status = terminal
if error_str is not None and handle.error is None:
handle.error = error_str
if not handle._mux._events._closed:
try:
handle._mux.fail(err)
except Exception:
logger.warning(
"Error failing subgraph mini-mux at %s; subscribers "
"may not see the terminal error.",
handle.path,
exc_info=True,
)
class LifecyclePayload(TypedDict, total=False):
"""Payload of a lifecycle event emitted by `LifecycleTransformer`."""
event: SubgraphStatus
namespace: list[str]
graph_name: str | None
trigger_call_id: str | None
error: str | None
class LifecycleTransformer(StreamTransformer):
"""Synthesize subgraph lifecycle events from observed namespaces.
Observes the same namespace signal `SubgraphTransformer` uses for
in-process discovery and emits `started` / `completed` / `failed`
/ `interrupted` payloads onto its `lifecycle` channel. Consumers
subscribed to that channel see the events in-process; wire
consumers receive them as protocol events with `method:
"lifecycle"` (unprefixed because this transformer is `_native`).
No `running` event: the ns-discovery signal only fires once a
subgraph has emitted output, so `started` already implies
execution. Consumers needing finer-grained task-start visibility
should read the `tasks` stream mode alongside.
No root `started`: the run object itself signals run start.
Terminal events are synthesized `finalize` emits `completed` for
still-open handles; `fail` emits `failed` or `interrupted`
depending on whether the error is a `GraphInterrupt`.
`scope_exact = False` so the transformer sees events at any
namespace (needed for discovery of direct children).
"""
_native = True
scope_exact = False
def __init__(self, scope: tuple[str, ...] = ()) -> None:
super().__init__(scope)
# retain=True: lifecycle events are low-volume and consumers
# commonly inspect them after draining `values`; without
# retention those pushes would be dropped.
self._channel: StreamChannel[LifecyclePayload] = StreamChannel(
"lifecycle", retain=True
)
self._seen: set[tuple[str, ...]] = set()
self._open: set[tuple[str, ...]] = set()
def init(self) -> dict[str, Any]:
return {"lifecycle": self._channel}
def process(self, event: ProtocolEvent) -> bool:
ns = tuple(event["params"]["namespace"])
if _is_new_direct_child(ns, self.scope, self._seen):
self._emit_started(ns)
return True
def _emit_started(self, ns: tuple[str, ...]) -> None:
graph_name, trigger_call_id = _parse_ns_segment(ns[-1])
self._seen.add(ns)
self._open.add(ns)
payload: LifecyclePayload = {
"event": "started",
"namespace": list(ns),
}
if graph_name:
payload["graph_name"] = graph_name
if trigger_call_id is not None:
payload["trigger_call_id"] = trigger_call_id
self._channel.push(payload)
def finalize(self) -> None:
"""Emit `completed` for every still-open direct child."""
for ns in list(self._open):
self._channel.push({"event": "completed", "namespace": list(ns)})
self._open.clear()
def fail(self, err: BaseException) -> None:
"""Emit `failed` / `interrupted` for every still-open direct child.
Closes the channel after emitting rather than letting the mux
auto-fail it the "failed" payload is the signal to
consumers, so they should be able to iterate it. A failed
channel would raise on iteration and hide the events that just
got pushed.
"""
is_interrupt = isinstance(err, GraphInterrupt)
event_type: SubgraphStatus = "interrupted" if is_interrupt else "failed"
error_str = None if is_interrupt else str(err)
for ns in list(self._open):
payload: LifecyclePayload = {
"event": event_type,
"namespace": list(ns),
}
if error_str is not None:
payload["error"] = error_str
self._channel.push(payload)
self._open.clear()
self._channel._close()
+1 -7
View File
@@ -116,13 +116,7 @@ def ensure_valid_checkpointer(checkpointer: Checkpointer) -> Checkpointer:
StreamMode = Literal[
"values",
"updates",
"checkpoints",
"tasks",
"debug",
"messages",
"custom",
"values", "updates", "checkpoints", "tasks", "debug", "messages", "custom"
]
"""How the stream method should emit outputs.
+2 -2
View File
@@ -4,7 +4,7 @@ build-backend = "hatchling.build"
[project]
name = "langgraph"
version = "1.1.7a2"
version = "1.1.9"
description = "Building stateful, multi-actor applications with LLMs"
authors = []
requires-python = ">=3.10"
@@ -24,7 +24,7 @@ classifiers = [
'Programming Language :: Python :: 3.13',
]
dependencies = [
"langchain-core==1.3.2",
"langchain-core>=1.3.0,<2",
"langgraph-checkpoint>=2.1.0,<5.0.0",
"langgraph-sdk>=0.3.0,<0.4.0",
"langgraph-prebuilt>=1.0.9,<1.1.0",
@@ -0,0 +1,290 @@
"""Benchmark: add_messages fast-path optimizations.
Both implementations are inlined so the benchmark is self-contained and
immune to import-cache or installed-vs-local confusion.
Run directly:
python tests/test_add_messages_benchmark.py
Or via pytest (correctness only, numbers printed to stdout):
pytest tests/test_add_messages_benchmark.py -s -v
"""
import statistics
import time
import tracemalloc
import uuid
from typing import cast
from langchain_core.messages import (
AIMessage,
BaseMessage,
BaseMessageChunk,
HumanMessage,
RemoveMessage,
convert_to_messages,
message_chunk_to_message,
)
from langgraph.graph.message import REMOVE_ALL_MESSAGES
# ── original implementation (pre-optimisation) ────────────────────────────────
def _add_messages_original(left, right):
remove_all_idx = None
if not isinstance(left, list):
left = [left]
if not isinstance(right, list):
right = [right]
left = [
message_chunk_to_message(cast(BaseMessageChunk, m))
for m in convert_to_messages(left)
]
right = [
message_chunk_to_message(cast(BaseMessageChunk, m))
for m in convert_to_messages(right)
]
for m in left:
if m.id is None:
m.id = str(uuid.uuid4())
for idx, m in enumerate(right):
if m.id is None:
m.id = str(uuid.uuid4())
if isinstance(m, RemoveMessage) and m.id == REMOVE_ALL_MESSAGES:
remove_all_idx = idx
if remove_all_idx is not None:
return right[remove_all_idx + 1 :]
merged = left.copy()
merged_by_id = {m.id: i for i, m in enumerate(merged)}
ids_to_remove = set()
for m in right:
if (existing_idx := merged_by_id.get(m.id)) is not None:
if isinstance(m, RemoveMessage):
ids_to_remove.add(m.id)
else:
ids_to_remove.discard(m.id)
merged[existing_idx] = m
else:
if isinstance(m, RemoveMessage):
raise ValueError(
f"Attempting to delete a message with an ID that doesn't exist ('{m.id}')"
)
merged_by_id[m.id] = len(merged)
merged.append(m)
return [m for m in merged if m.id not in ids_to_remove]
# ── optimised implementation ──────────────────────────────────────────────────
def _add_messages_optimized(left, right):
if not isinstance(left, list):
left = [left]
if not isinstance(right, list):
right = [right]
# Optimisation 1: skip conversion + ID assignment on left when it already
# contains fully-resolved BaseMessage objects (the common case after the
# first call, since add_messages always returns list[BaseMessage] with IDs).
if (
left
and isinstance(left[0], BaseMessage)
and not isinstance(left[0], BaseMessageChunk)
):
left = cast(list[BaseMessage], left)
else:
left = [
message_chunk_to_message(cast(BaseMessageChunk, m))
for m in convert_to_messages(left)
]
for m in left:
if m.id is None:
m.id = str(uuid.uuid4())
# always normalise right — it's fresh external input
right = [
message_chunk_to_message(cast(BaseMessageChunk, m))
for m in convert_to_messages(right)
]
remove_all_idx = None
has_remove = False
for idx, m in enumerate(right):
if m.id is None:
m.id = str(uuid.uuid4())
if isinstance(m, RemoveMessage):
has_remove = True
if m.id == REMOVE_ALL_MESSAGES:
remove_all_idx = idx
if remove_all_idx is not None:
return right[remove_all_idx + 1 :]
# Optimisation 2: pure-append fast path — no removals and no ID overlaps.
# Builds one set over left instead of copying left + building a full dict.
if not has_remove:
left_ids = {m.id for m in left}
if not any(m.id in left_ids for m in right):
return left + right
# slow path: updates or removals present — full indexed merge
merged = left.copy()
merged_by_id = {m.id: i for i, m in enumerate(merged)}
ids_to_remove = set()
for m in right:
if (existing_idx := merged_by_id.get(m.id)) is not None:
if isinstance(m, RemoveMessage):
ids_to_remove.add(m.id)
else:
ids_to_remove.discard(m.id)
merged[existing_idx] = m
else:
if isinstance(m, RemoveMessage):
raise ValueError(
f"Attempting to delete a message with an ID that doesn't exist ('{m.id}')"
)
merged_by_id[m.id] = len(merged)
merged.append(m)
return [m for m in merged if m.id not in ids_to_remove]
# ── helpers ───────────────────────────────────────────────────────────────────
def _make_messages(n: int) -> list[BaseMessage]:
return [
(HumanMessage if i % 2 == 0 else AIMessage)(
content=f"message {i}", id=str(uuid.uuid4())
)
for i in range(n)
]
def _bench_time(fn, left, right, *, iters: int = 2_000) -> float:
"""Return median latency in microseconds."""
for _ in range(100):
fn(list(left), list(right))
times = []
for _ in range(iters):
left_copy, right_copy = list(left), list(right)
t0 = time.perf_counter()
fn(left_copy, right_copy)
times.append(time.perf_counter() - t0)
return statistics.median(times) * 1e6
def _bench_memory(fn, left, right) -> int:
"""Return peak memory allocated during a single call (bytes)."""
# one warm-up so any lazy init is excluded
fn(list(left), list(right))
left_copy, right_copy = list(left), list(right)
tracemalloc.start()
tracemalloc.clear_traces()
fn(left_copy, right_copy)
_, peak = tracemalloc.get_traced_memory()
tracemalloc.stop()
return peak
# ── scenarios ─────────────────────────────────────────────────────────────────
SCENARIOS = [
("pure append 1 → 1 msg", 1, 1, "append"),
("pure append 10 → 1 msg", 10, 1, "append"),
("pure append 100 → 1 msg", 100, 1, "append"),
("pure append 1000 → 1 msg", 1000, 1, "append"),
("pure append 1000 → 5 msgs", 1000, 5, "append"),
("update existing 100 → 1 msg", 100, 1, "update"),
("remove message 100 → 1 msg", 100, 1, "remove"),
]
def _make_inputs(n_left, n_right, mode):
left = _make_messages(n_left)
right = _make_messages(n_right)
if mode == "update":
right[0] = AIMessage(content="updated", id=left[0].id)
elif mode == "remove":
right = [RemoveMessage(id=left[0].id)]
return left, right
# ── main output ───────────────────────────────────────────────────────────────
COL = 36
def run_benchmarks() -> None:
print()
print("=" * 88)
print("add_messages benchmark — time (µs, median of 2 000 iterations)")
print("=" * 88)
print(f"{'Scenario':<{COL}} {'Original':>10} {'Optimized':>11} {'Speedup':>8}")
print("-" * 88)
for label, n_left, n_right, mode in SCENARIOS:
left, right = _make_inputs(n_left, n_right, mode)
t_orig = _bench_time(_add_messages_original, left, right)
t_opt = _bench_time(_add_messages_optimized, left, right)
print(f"{label:<{COL}} {t_orig:>10.2f} {t_opt:>11.2f} {t_orig / t_opt:>7.2f}x")
print()
print("=" * 88)
print("add_messages benchmark — peak memory allocated per call (bytes)")
print("=" * 88)
print(f"{'Scenario':<{COL}} {'Original':>10} {'Optimized':>11} {'Reduction':>10}")
print("-" * 88)
for label, n_left, n_right, mode in SCENARIOS:
left, right = _make_inputs(n_left, n_right, mode)
m_orig = _bench_memory(_add_messages_original, left, right)
m_opt = _bench_memory(_add_messages_optimized, left, right)
reduction = (1 - m_opt / m_orig) * 100 if m_orig else 0.0
print(f"{label:<{COL}} {m_orig:>10,} {m_opt:>11,} {reduction:>9.1f}%")
print()
print("=" * 88)
print("Simulated long thread — 200 steps × 2 msgs appended per step")
print("=" * 88)
for name, fn in [
("original", _add_messages_original),
("optimized", _add_messages_optimized),
]:
state: list = []
t0 = time.perf_counter()
for step in range(200):
new_msgs = [
HumanMessage(content=f"step {step} human", id=str(uuid.uuid4())),
AIMessage(content=f"step {step} ai", id=str(uuid.uuid4())),
]
state = fn(state, new_msgs)
elapsed = (time.perf_counter() - t0) * 1_000
print(f" {name:<12} {elapsed:.2f} ms ({len(state)} messages)")
print()
# ── pytest entry-points ───────────────────────────────────────────────────────
def test_add_messages_correctness():
"""Optimised implementation must match original output for every scenario."""
for label, n_left, n_right, mode in SCENARIOS:
left, right = _make_inputs(n_left, n_right, mode)
expected = _add_messages_original(list(left), list(right))
actual = _add_messages_optimized(list(left), list(right))
assert len(actual) == len(expected), f"[{label}] length mismatch"
for a, b in zip(actual, expected):
assert type(a) is type(b), f"[{label}] type mismatch"
assert a.id == b.id, f"[{label}] id mismatch"
assert a.content == b.content, f"[{label}] content mismatch"
def test_add_messages_benchmark(capsys):
run_benchmarks()
out = capsys.readouterr().out
assert "Speedup" in out
assert "Optimized" in out
if __name__ == "__main__":
run_benchmarks()
+504
View File
@@ -2,13 +2,17 @@ import operator
from collections.abc import Sequence
import pytest
from langchain_core.messages import AIMessage, HumanMessage
from langgraph.checkpoint.base import DELTA_SENTINEL
from langgraph._internal._typing import MISSING
from langgraph.channels.binop import BinaryOperatorAggregate
from langgraph.channels._delta import DeltaChannel
from langgraph.channels.last_value import LastValue
from langgraph.channels.topic import Topic
from langgraph.channels.untracked_value import UntrackedValue
from langgraph.errors import EmptyChannelError, InvalidUpdateError
from langgraph.graph.message import add_messages
pytestmark = pytest.mark.anyio
@@ -117,3 +121,503 @@ def test_untracked_value() -> None:
new_channel = UntrackedValue(dict).from_checkpoint(checkpoint)
with pytest.raises(EmptyChannelError):
new_channel.get()
def test_delta_channel_basic_two_steps() -> None:
from langchain_core.messages import AIMessage, HumanMessage
from langgraph.checkpoint.base import DELTA_SENTINEL
from langgraph.channels._delta import DeltaChannel
from langgraph.graph.message import add_messages
ch = DeltaChannel(add_messages).from_checkpoint(MISSING)
# Step 1: one message added
ch.update([HumanMessage(content="hi", id="h1")])
d1 = ch.checkpoint()
assert d1 is DELTA_SENTINEL
# Step 2: another message
ch.update([AIMessage(content="hello", id="a1")])
d2 = ch.checkpoint()
assert d2 is DELTA_SENTINEL
# Full accumulated value is preserved in memory
assert len(ch.get()) == 2
assert ch.get()[0].content == "hi"
assert ch.get()[1].content == "hello"
def test_delta_channel_from_checkpoint_writes_list() -> None:
"""replay_writes on a fresh channel replays through the operator."""
from langchain_core.messages import AIMessage, HumanMessage
from langgraph.channels._delta import DeltaChannel
from langgraph.graph.message import add_messages
spec = DeltaChannel(add_messages)
ch = spec.from_checkpoint(DELTA_SENTINEL)
ch.replay_writes(
[
("t0", "messages", HumanMessage(content="hi", id="h1")),
("t1", "messages", AIMessage(content="hello", id="a1")),
("t2", "messages", HumanMessage(content="bye", id="h2")),
]
)
msgs = ch.get()
assert len(msgs) == 3
assert msgs[0].content == "hi"
assert msgs[1].content == "hello"
assert msgs[2].content == "bye"
def test_delta_channel_from_checkpoint_backwards_compat() -> None:
from langchain_core.messages import HumanMessage
from langgraph.channels._delta import DeltaChannel
from langgraph.graph.message import add_messages
# Old BinaryOperatorAggregate checkpoint: plain list treated as backward compat
spec = DeltaChannel(add_messages)
old_value = [HumanMessage(content="old", id="h1")]
ch = spec.from_checkpoint(old_value)
assert ch.get() == old_value
def test_delta_channel_overwrite() -> None:
from langchain_core.messages import HumanMessage
from langgraph.checkpoint.base import DELTA_SENTINEL
from langgraph.channels._delta import DeltaChannel
from langgraph.graph.message import add_messages
from langgraph.types import Overwrite
ch = DeltaChannel(add_messages).from_checkpoint(MISSING)
ch.update([HumanMessage(content="old", id="h1")])
ch.update([Overwrite([HumanMessage(content="new", id="h2")])])
d = ch.checkpoint()
assert d is DELTA_SENTINEL
# After overwrite, value is reset to only the new message
assert len(ch.get()) == 1
assert ch.get()[0].content == "new"
def test_delta_channel_remove_message_and_replay() -> None:
"""RemoveMessage must round-trip correctly when writes are replayed."""
from langchain_core.messages import AIMessage, HumanMessage, RemoveMessage
from langgraph.channels._delta import DeltaChannel
from langgraph.graph.message import add_messages
spec = DeltaChannel(add_messages)
ch = spec.from_checkpoint(MISSING)
# Step 1: add two messages
ch.update([HumanMessage(content="hi", id="h1")])
ch.update([AIMessage(content="hello", id="a1")])
assert ch.get() == [
HumanMessage(content="hi", id="h1"),
AIMessage(content="hello", id="a1"),
]
# Step 2: remove the AI message
ch.update([RemoveMessage(id="a1")])
assert ch.get() == [HumanMessage(content="hi", id="h1")]
# Replay the writes list from scratch — must reproduce the post-remove state
ch2 = spec.from_checkpoint(DELTA_SENTINEL)
ch2.replay_writes(
[
("t0", "messages", HumanMessage(content="hi", id="h1")),
("t1", "messages", AIMessage(content="hello", id="a1")),
("t2", "messages", RemoveMessage(id="a1")),
]
)
assert ch2.get() == [HumanMessage(content="hi", id="h1")]
def test_delta_channel_update_by_id_and_replay() -> None:
"""Updating a message by ID must round-trip correctly through writes replay."""
from langchain_core.messages import HumanMessage
from langgraph.channels._delta import DeltaChannel
from langgraph.graph.message import add_messages
spec = DeltaChannel(add_messages)
ch = spec.from_checkpoint(MISSING)
# Step 1: add a message
ch.update([HumanMessage(content="original", id="h1")])
# Step 2: update the same message by ID
ch.update([HumanMessage(content="updated", id="h1")])
assert ch.get() == [HumanMessage(content="updated", id="h1")]
# Replay writes — must produce the updated message, not the original
ch2 = spec.from_checkpoint(DELTA_SENTINEL)
ch2.replay_writes(
[
("t0", "messages", HumanMessage(content="original", id="h1")),
("t1", "messages", HumanMessage(content="updated", id="h1")),
]
)
assert len(ch2.get()) == 1
assert ch2.get()[0].content == "updated"
def test_delta_channel_checkpoint_returns_sentinel() -> None:
"""checkpoint() always returns DELTA_SENTINEL regardless of state."""
from langgraph.checkpoint.base import DELTA_SENTINEL
from langgraph.channels._delta import DeltaChannel
from langgraph.graph.message import add_messages
ch = DeltaChannel(add_messages).from_checkpoint(MISSING)
assert ch.checkpoint() is DELTA_SENTINEL
from langchain_core.messages import HumanMessage
ch.update([HumanMessage(content="hi", id="h1")])
assert ch.checkpoint() is DELTA_SENTINEL
def test_delta_channel_inmemory_saver_assembles_writes() -> None:
"""InMemorySaver assembles writes from checkpoint_writes inside get_tuple."""
from typing import Annotated
from langchain_core.messages import AIMessage, HumanMessage
from langgraph.checkpoint.memory import InMemorySaver
from typing_extensions import TypedDict
from langgraph.channels._delta import DeltaChannel
from langgraph.graph import START, StateGraph
from langgraph.graph.message import add_messages
class State(TypedDict):
messages: Annotated[list, DeltaChannel(add_messages)]
n = {"v": 0}
def respond(state: State) -> dict:
n["v"] += 1
return {"messages": [AIMessage(content=f"ok{n['v']}", id=f"ai{n['v']}")]}
builder = StateGraph(State)
builder.add_node("respond", respond)
builder.add_edge(START, "respond")
saver = InMemorySaver()
graph = builder.compile(checkpointer=saver)
config = {"configurable": {"thread_id": "t1"}}
graph.invoke({"messages": [HumanMessage(content="hi", id="h1")]}, config)
graph.invoke({"messages": [HumanMessage(content="bye", id="h2")]}, config)
# get_tuple returns raw storage shape — channel_values stores DELTA_SENTINEL
# for delta channels; the reconstructed writes flow separately via
# saver._get_channel_writes_history.
saved = saver.get_tuple(config)
assert saved is not None
assert "messages" in saved.checkpoint["channel_values"]
assert saved.checkpoint["channel_values"]["messages"] is DELTA_SENTINEL
state = graph.get_state(config)
assert len(state.values["messages"]) == 4 # 2 human + 2 AI
def _delta_channel_with_type(operator, typ):
"""Build a DeltaChannel with an explicit type via the Annotated injection path."""
from typing import Annotated
from langgraph.channels._delta import DeltaChannel
from langgraph.graph.state import _get_channel
return _get_channel("_test", Annotated[typ, DeltaChannel(operator)])
def test_delta_channel_dict_reducer_fresh_channel() -> None:
"""DeltaChannel with a dict reducer starts as empty dict on MISSING checkpoint."""
def merge_dicts(left: dict, right: dict) -> dict:
return {**left, **right}
ch = _delta_channel_with_type(merge_dicts, dict).from_checkpoint(MISSING)
# Should be available (not raise EmptyChannelError) and start empty
assert ch.is_available()
assert ch.get() == {}
def test_delta_channel_dict_reducer_basic_updates() -> None:
"""DeltaChannel with a dict reducer accumulates key/value pairs across steps."""
from langgraph.checkpoint.base import DELTA_SENTINEL
def merge_dicts(left: dict, right: dict) -> dict:
return {**left, **right}
ch = _delta_channel_with_type(merge_dicts, dict).from_checkpoint(MISSING)
ch.update([{"a": 1}])
d1 = ch.checkpoint()
assert d1 is DELTA_SENTINEL
ch.update([{"b": 2}])
d2 = ch.checkpoint()
assert d2 is DELTA_SENTINEL
assert ch.get() == {"a": 1, "b": 2}
def test_delta_channel_dict_reducer_writes_reconstruction() -> None:
"""replay_writes on a fresh channel replays through a dict merge reducer."""
def merge_dicts(left: dict, right: dict) -> dict:
return {**left, **right}
spec = _delta_channel_with_type(merge_dicts, dict)
ch = spec.from_checkpoint(DELTA_SENTINEL)
ch.replay_writes(
[
("t0", "files", {"a": 1}),
("t1", "files", {"b": 2}),
("t2", "files", {"c": 3}),
]
)
assert ch.get() == {"a": 1, "b": 2, "c": 3}
def test_delta_channel_dict_reducer_with_deletions() -> None:
"""Dict reducer that treats None values as deletions works end-to-end (deepagents pattern)."""
def merge_files(left: dict | None, right: dict) -> dict:
if left is None:
return {k: v for k, v in right.items() if v is not None}
result = {**left}
for k, v in right.items():
if v is None:
result.pop(k, None)
else:
result[k] = v
return result
ch = _delta_channel_with_type(merge_files, dict).from_checkpoint(MISSING)
ch.update([{"file1.py": "content1", "file2.py": "content2"}])
# Delete file1, add file3
ch.update([{"file1.py": None, "file3.py": "content3"}])
assert ch.get() == {"file2.py": "content2", "file3.py": "content3"}
# Confirm writes reconstruction produces the same result
spec = _delta_channel_with_type(merge_files, dict)
ch2 = spec.from_checkpoint(DELTA_SENTINEL)
ch2.replay_writes(
[
("t0", "files", {"file1.py": "content1", "file2.py": "content2"}),
("t1", "files", {"file1.py": None, "file3.py": "content3"}),
]
)
assert ch2.get() == {"file2.py": "content2", "file3.py": "content3"}
def test_delta_channel_dict_reducer_overwrite_in_update() -> None:
"""Overwrite(dict) in update() must preserve dict shape, not coerce to list."""
from langgraph.types import Overwrite
def merge_dicts(left: dict, right: dict) -> dict:
return {**left, **right}
ch = _delta_channel_with_type(merge_dicts, dict).from_checkpoint(MISSING)
ch.update([{"a": 1}])
ch.update([Overwrite({"b": 2, "c": 3})])
assert ch.get() == {"b": 2, "c": 3}
def test_delta_channel_dict_reducer_overwrite_in_writes_replay() -> None:
"""Overwrite(dict) embedded in replayed writes must reconstruct as dict."""
from langgraph.types import Overwrite
def merge_dicts(left: dict, right: dict) -> dict:
return {**left, **right}
spec = _delta_channel_with_type(merge_dicts, dict)
ch = spec.from_checkpoint(DELTA_SENTINEL)
ch.replay_writes(
[
("t0", "files", {"a": 1}),
("t1", "files", Overwrite({"x": 10, "y": 20})),
("t2", "files", {"z": 30}),
]
)
assert ch.get() == {"x": 10, "y": 20, "z": 30}
def test_delta_channel_dict_reducer_with_notrequired_annotation() -> None:
"""DeltaChannel infers dict type through `Annotated[NotRequired[dict[...]], ch]`.
This is the shape the deepagents filesystem middleware uses for its
`files` field; without unwrapping NotRequired we'd fall through to `list`
and blow up on the first dict operator call.
"""
from typing import Annotated
from typing_extensions import NotRequired
from langgraph.channels._delta import DeltaChannel
from langgraph.graph.state import _get_channel
def merge_dicts(left: dict | None, right: dict) -> dict:
if left is None:
return dict(right)
return {**left, **right}
annotation = Annotated[
NotRequired[dict[str, int]],
DeltaChannel(merge_dicts),
]
ch = _get_channel("files", annotation).from_checkpoint(MISSING)
assert ch.get() == {}
ch.update([{"a": 1}])
ch.update([{"b": 2}])
assert ch.get() == {"a": 1, "b": 2}
def test_delta_channel_dict_reducer_end_to_end_filesystem() -> None:
"""End-to-end: graph with dict-reducer (filesystem-style) channel wrapped in DeltaChannel.
Mirrors the deepagents filesystem pattern: `files: Annotated[dict, reducer]`
where the reducer merges dicts and treats None values as deletions.
"""
from typing import Annotated
from langgraph.checkpoint.memory import InMemorySaver
from typing_extensions import TypedDict
from langgraph.channels._delta import DeltaChannel
from langgraph.graph import START, StateGraph
def merge_files(left: dict | None, right: dict) -> dict:
if left is None:
return {k: v for k, v in right.items() if v is not None}
result = {**left}
for k, v in right.items():
if v is None:
result.pop(k, None)
else:
result[k] = v
return result
class State(TypedDict):
files: Annotated[dict[str, str], DeltaChannel(merge_files)]
turn = {"v": 0}
def write_file(state: State) -> dict:
turn["v"] += 1
n = turn["v"]
return {"files": {f"/doc_{n}.txt": f"content for turn {n}"}}
builder = StateGraph(State)
builder.add_node("write_file", write_file)
builder.add_edge(START, "write_file")
saver = InMemorySaver()
graph = builder.compile(checkpointer=saver)
config = {"configurable": {"thread_id": "fs"}}
for _ in range(3):
graph.invoke({"files": {}}, config)
# Checkpoint stores only the sentinel — per-step writes live in checkpoint_writes.
saved = saver.get_tuple(config)
assert saved is not None
cv = saved.checkpoint["channel_values"]["files"]
assert cv is DELTA_SENTINEL
state = graph.get_state(config)
assert state.values["files"] == {
"/doc_1.txt": "content for turn 1",
"/doc_2.txt": "content for turn 2",
"/doc_3.txt": "content for turn 3",
}
# Deletion path must round-trip through writes replay.
def delete_file(state: State) -> dict:
return {"files": {"/doc_1.txt": None}}
builder2 = StateGraph(State)
builder2.add_node("write_file", write_file)
builder2.add_node("delete_file", delete_file)
builder2.add_edge(START, "write_file")
builder2.add_edge("write_file", "delete_file")
turn["v"] = 0
saver2 = InMemorySaver()
graph2 = builder2.compile(checkpointer=saver2)
config2 = {"configurable": {"thread_id": "fs2"}}
graph2.invoke({"files": {}}, config2)
state2 = graph2.get_state(config2)
assert state2.values["files"] == {}
def test_delta_channel_dict_reducer_backwards_compat() -> None:
"""A pre-DeltaChannel dict checkpoint must load as a dict, not be listified."""
def merge_dicts(left: dict, right: dict) -> dict:
return {**left, **right}
spec = _delta_channel_with_type(merge_dicts, dict)
old_value = {"a": 1, "b": 2}
ch = spec.from_checkpoint(old_value)
assert ch.get() == {"a": 1, "b": 2}
# ---------------------------------------------------------------------------
# seed / pre-delta migration
# ---------------------------------------------------------------------------
def test_delta_channel_from_checkpoint_honors_seed() -> None:
"""A non-sentinel value to from_checkpoint is used as the pre-delta seed.
Guards the pre-delta migration path: when the saver's ancestor walk hits
a pre-DeltaChannel blob it passes it as `seed` so replay reconstructs
the post-migration state correctly rather than replaying from empty.
"""
spec = DeltaChannel(add_messages)
seed = [HumanMessage(content="pre-delta", id="p1")]
ch = spec.from_checkpoint(seed)
ch.replay_writes(
[
("t0", "messages", AIMessage(content="delta-1", id="d1")),
("t1", "messages", HumanMessage(content="delta-2", id="d2")),
]
)
msgs = ch.get()
assert [m.content for m in msgs] == ["pre-delta", "delta-1", "delta-2"]
def test_delta_channel_from_checkpoint_seed_without_writes() -> None:
"""Reconstruction at a pre-delta ancestor with no newer deltas returns
just the seed the saver's terminator fired immediately."""
spec = DeltaChannel(add_messages)
seed = [HumanMessage(content="only-snap", id="s1")]
ch = spec.from_checkpoint(seed)
ch.replay_writes([])
assert ch.get() == seed
def test_delta_channel_from_checkpoint_seed_none_is_distinct_from_sentinel() -> None:
"""`seed=None` must start replay from None, not from an empty channel.
The DELTA_SENTINEL / MISSING sentinels mean 'no seed'; passing `None`
explicitly should feed None to the reducer as the left operand.
"""
def replace(left, right):
return right
spec = DeltaChannel(replace)
ch = spec.from_checkpoint(None)
ch.replay_writes([("t0", "x", "after")])
# Reducer replaces; seed=None → first write produces "after".
assert ch.get() == "after"
@@ -0,0 +1,398 @@
"""Benchmark: DeltaChannel vs BinaryOperatorAggregate storage and time.
Run directly: python tests/test_delta_channel_benchmark.py
Run via pytest: pytest tests/test_delta_channel_benchmark.py -s
Simulates realistic multi-turn conversations with paragraph-length messages
(~100 tokens each) scaling up to 1M-token-equivalent histories.
Token estimates: 1 token 4 chars; each turn 200 tokens (human + AI).
A 1M-token conversation 5,000 turns of realistic messages.
DeltaChannel stores only a zero-byte sentinel in checkpoint_blobs; the actual
write data lives in checkpoint_writes (already stored there). Reconstruction
walks the parent chain and replays writes through the operator O(N) total
storage vs O() for plain add_messages.
"""
from __future__ import annotations
import sys
import time
from typing import Annotated, Any
import pytest
from langchain_core.messages import AIMessage, HumanMessage
from langgraph.checkpoint.memory import MemorySaver
from typing_extensions import TypedDict
from langgraph.channels._delta import DeltaChannel
from langgraph.graph import END, StateGraph
from langgraph.graph.message import add_messages
try:
from langgraph.checkpoint.sqlite import SqliteSaver
_SQLITE_AVAILABLE = True
except ImportError:
_SQLITE_AVAILABLE = False
try:
from langgraph.checkpoint.postgres import PostgresSaver
_POSTGRES_AVAILABLE = True
_POSTGRES_URI = (
"postgres://postgres:postgres@localhost:5441/postgres?sslmode=disable"
)
except ImportError:
_POSTGRES_AVAILABLE = False
# ---------------------------------------------------------------------------
# Realistic message payload (~100 tokens / ~400 chars each)
# ---------------------------------------------------------------------------
_HUMAN_TEMPLATE = (
"I need help understanding the implications of {topic} on our system architecture. "
"Specifically, I'm concerned about how this interacts with our existing {concern} "
"and whether we need to refactor the {component} layer before proceeding. "
"We've had prior incidents in this area and want to be deliberate. "
"What should we prioritize first, and are there known failure modes we should design around from the start?"
)
_AI_TEMPLATE = (
"Great question about {topic}. The key insight here is that {concern} introduces "
"a subtle ordering dependency that most teams overlook until they hit it in production. "
"For your {component} layer specifically, I'd recommend starting with a careful audit "
"of the interface boundaries before making any structural changes. This will give you "
"a clear picture of the blast radius and let you sequence the migration safely."
)
_TOPICS = [
"distributed tracing",
"eventual consistency",
"schema migration",
"backpressure handling",
"idempotency guarantees",
"cache invalidation",
"connection pooling",
"rate limiting",
"circuit breaking",
"observability pipelines",
]
_CONCERNS = [
"concurrency model",
"retry semantics",
"state management",
"error propagation",
"latency budget",
]
_COMPONENTS = [
"persistence",
"routing",
"ingestion",
"aggregation",
"serialization",
]
def _human_content(i: int) -> str:
return _HUMAN_TEMPLATE.format(
topic=_TOPICS[i % len(_TOPICS)],
concern=_CONCERNS[i % len(_CONCERNS)],
component=_COMPONENTS[i % len(_COMPONENTS)],
)
def _ai_content(i: int) -> str:
return _AI_TEMPLATE.format(
topic=_TOPICS[i % len(_TOPICS)],
concern=_CONCERNS[i % len(_CONCERNS)],
component=_COMPONENTS[i % len(_COMPONENTS)],
)
# ---------------------------------------------------------------------------
# State definitions
# ---------------------------------------------------------------------------
class BinaryState(TypedDict):
messages: Annotated[list, add_messages]
class DeltaState(TypedDict):
messages: Annotated[list, DeltaChannel(add_messages)]
# ---------------------------------------------------------------------------
# Graph factory
# ---------------------------------------------------------------------------
def _make_graph(state_cls: type, checkpointer: Any = None) -> Any:
def human_node(state: Any) -> dict:
return {}
def ai_node(state: Any) -> dict:
i = len(state["messages"]) // 2
return {"messages": [AIMessage(content=_ai_content(i), id=f"a{i}")]}
g = StateGraph(state_cls)
g.add_node("human", human_node)
g.add_node("ai", ai_node)
g.add_edge("human", "ai")
g.add_edge("ai", END)
g.set_entry_point("human")
return g.compile(checkpointer=checkpointer or MemorySaver())
# ---------------------------------------------------------------------------
# Measurement helpers
# ---------------------------------------------------------------------------
def _total_blob_bytes(saver: MemorySaver) -> int:
total = 0
for (_, _, _, _), (type_tag, blob) in saver.blobs.items():
if blob is not None:
total += len(blob)
return total
def _run_turns(
n_turns: int,
state_cls: type,
checkpointer: Any = None,
) -> tuple[float, float, int]:
"""Run n_turns conversation turns.
Returns (write_elapsed_s, read_elapsed_s, total_blob_bytes).
blob_bytes is -1 for savers without in-memory blob stores (e.g. SQLite).
Read latency is measured as the time to invoke the graph with no new
messages after the full history is built this forces state rehydration.
"""
graph = _make_graph(state_cls, checkpointer)
config = {"configurable": {"thread_id": "bench"}}
t0 = time.perf_counter()
for i in range(n_turns):
graph.invoke(
{"messages": [HumanMessage(content=_human_content(i), id=f"h{i}")]},
config,
)
write_elapsed = time.perf_counter() - t0
# Measure read/rehydration: get_state forces the channel to rebuild
t1 = time.perf_counter()
for _ in range(5):
graph.get_state(config)
read_elapsed = (time.perf_counter() - t1) / 5
if isinstance(graph.checkpointer, MemorySaver):
blob_bytes = _total_blob_bytes(graph.checkpointer)
else:
blob_bytes = -1
return write_elapsed, read_elapsed, blob_bytes
def _fmt_bytes(n: int) -> str:
if n >= 1_000_000:
return f"{n / 1_000_000:.1f} MB"
if n >= 1_000:
return f"{n / 1_000:.1f} KB"
return f"{n} B"
def _approx_tokens(n_turns: int) -> str:
# ~100 tokens human + ~100 tokens AI per turn
tokens = n_turns * 200
if tokens >= 1_000_000:
return f"~{tokens / 1_000_000:.1f}M tok"
if tokens >= 1_000:
return f"~{tokens / 1_000:.0f}K tok"
return f"~{tokens} tok"
# ---------------------------------------------------------------------------
# Benchmark matrix
# ---------------------------------------------------------------------------
# Turn counts chosen to demonstrate O(N²) vs O(N) storage growth without running too long.
# Extrapolation: 5,000 turns × ~200 tokens/turn ≈ 1M tokens (Claude's full context window).
TURN_COUNTS = [10, 25, 50, 100, 500]
# Deep-thread counts where add_messages blob storage would exceed 1 GB;
# only DeltaChannel runs here.
DELTA_ONLY_TURN_COUNTS = [1000]
def _checkpointer_factories() -> list[tuple[str, Any]]:
"""Return (label, context_manager_or_none) pairs for available checkpointers."""
return [("InMemory", None)]
def run_benchmark() -> None:
print()
print(
"DeltaChannel vs add_messages (BinaryOperatorAggregate) — checkpoint storage & latency"
)
print("Simulating realistic multi-turn conversations up to ~1M-token histories")
print("(5,000 turns × ~200 tokens/turn ≈ 1M tokens — Claude's full context window)")
print()
checkpointers: list[tuple[str, Any]] = [("InMemory", None)]
if _POSTGRES_AVAILABLE:
try:
import psycopg
psycopg.connect(_POSTGRES_URI).close()
checkpointers.append(("Postgres (plain SELECT)", "postgres"))
except Exception:
pass
for cp_label, cp_hint in checkpointers:
print(f"--- Checkpointer: {cp_label} ---")
_run_benchmark_for_checkpointer(cp_hint)
def _run_benchmark_for_checkpointer(cp_hint: Any) -> None:
import contextlib
import tempfile
@contextlib.contextmanager
def _make_saver():
if cp_hint is None:
yield None
elif cp_hint == "postgres":
with PostgresSaver.from_conn_string(_POSTGRES_URI) as saver:
saver.setup()
with saver._cursor() as cur:
cur.execute("DELETE FROM checkpoints WHERE thread_id = 'bench'")
cur.execute(
"DELETE FROM checkpoint_blobs WHERE thread_id = 'bench'"
)
cur.execute(
"DELETE FROM checkpoint_writes WHERE thread_id = 'bench'"
)
yield saver
else:
with tempfile.NamedTemporaryFile(suffix=".db") as f:
with SqliteSaver.from_conn_string(f.name) as saver:
yield saver
rows: list[tuple[int, Any, Any, Any, Any, Any, Any]] = []
for turns in TURN_COUNTS:
with _make_saver() as saver:
b_wt, b_rt, b_bytes = _run_turns(turns, BinaryState, saver)
with _make_saver() as saver:
d_wt, d_rt, d_bytes = _run_turns(turns, DeltaState, saver)
rows.append((turns, b_bytes, d_bytes, b_rt, d_rt, b_wt, d_wt))
for turns in DELTA_ONLY_TURN_COUNTS:
with _make_saver() as saver:
d_wt, d_rt, d_bytes = _run_turns(turns, DeltaState, saver)
rows.append((turns, None, d_bytes, None, d_rt, None, d_wt))
# ── Table 1: Storage ─────────────────────────────────────────────────────
W = 64
print("Storage (checkpoint blob bytes)")
print("=" * W)
print(
f"{'turns':>6} {'ctx size':>10} {'add_msgs':>12} {'delta':>12} "
f"{'savings':>8}"
)
print("-" * W)
def _bytes_or_na(v: Any) -> str:
if v is None:
return "n/a"
if v < 0:
return "n/a"
return _fmt_bytes(v)
def _ms_or_na(v: Any) -> str:
return "n/a" if v is None else f"{v * 1000:.1f}ms"
for turns, b_bytes, d_bytes, b_rt, d_rt, b_wt, d_wt in rows:
if b_bytes is None or b_bytes < 0 or d_bytes is None or d_bytes < 0:
ratio_str = "n/a"
else:
ratio = b_bytes / d_bytes if d_bytes else float("inf")
ratio_str = f"{ratio:.0f}x"
print(
f"{turns:>6} {_approx_tokens(turns):>10} "
f"{_bytes_or_na(b_bytes):>12} {_bytes_or_na(d_bytes):>12} "
f"{ratio_str:>8}"
)
print("=" * W)
print()
# ── Table 2: Read latency ─────────────────────────────────────────────────
print("Read latency (avg of 5 get_state calls)")
print("=" * W)
print(f"{'turns':>6} {'ctx size':>10} {'add_msgs':>12} {'delta':>12}")
print("-" * W)
for turns, b_bytes, d_bytes, b_rt, d_rt, b_wt, d_wt in rows:
print(
f"{turns:>6} {_approx_tokens(turns):>10} "
f"{_ms_or_na(b_rt):>12} {_ms_or_na(d_rt):>12}"
)
print("=" * W)
print()
# ── Table 3: Per-invoke latency (total write_elapsed / turns) ─────────────
print("Per-invoke latency (total graph.invoke time / turns)")
print("=" * W)
print(f"{'turns':>6} {'ctx size':>10} {'add_msgs':>12} {'delta':>12}")
print("-" * W)
for turns, b_bytes, d_bytes, b_rt, d_rt, b_wt, d_wt in rows:
def _per(wt: Any) -> str:
if wt is None:
return "n/a"
return f"{(wt / turns) * 1000:.1f}ms"
print(
f"{turns:>6} {_approx_tokens(turns):>10} "
f"{_per(b_wt):>12} {_per(d_wt):>12}"
)
print("=" * W)
print()
print("Legend:")
print(" add_msgs = Annotated[list, add_messages] — O(N²) storage")
print(" delta = Annotated[list, DeltaChannel(add_messages)] — O(N) storage")
print()
# ---------------------------------------------------------------------------
# Pytest entry point
# ---------------------------------------------------------------------------
@pytest.mark.skip(
reason="slow benchmark — run manually with: python tests/test_delta_channel_benchmark.py"
)
def test_delta_channel_benchmark(capsys: Any) -> None:
"""Storage grows O(N²) for add_messages, O(N) for DeltaChannel."""
with capsys.disabled():
run_benchmark()
# Correctness assertion: DeltaChannel must use less storage at scale.
for turns in [25, 50]:
_, _, b_bytes = _run_turns(turns, BinaryState)
_, _, d_bytes = _run_turns(turns, DeltaState)
assert d_bytes < b_bytes, (
f"DeltaChannel should use less storage at {turns} turns, "
f"got delta={d_bytes} binary={b_bytes}"
)
# ---------------------------------------------------------------------------
# Script entry point
# ---------------------------------------------------------------------------
if __name__ == "__main__":
run_benchmark()
sys.exit(0)
@@ -0,0 +1,504 @@
"""Tests for the BinaryOperatorAggregate -> DeltaChannel migration path.
A thread written under `BinaryOperatorAggregate(...)` must keep working
after its annotation is swapped to `DeltaChannel(...)` on the same
checkpointer pre-migration state visible at each *settled* ancestor
checkpoint is preserved, and post-migration writes fold on top through
the reducer.
Mechanism under test: the saver's `_get_channel_writes_history(config,
channel)` walks the parent chain; when it encounters an ancestor whose
`channel_values[channel]` is a real value (not `DELTA_SENTINEL`), it
returns that as the `seed`. `DeltaChannel.from_checkpoint(seed)` uses
it as the base value, and `replay_writes(writes)` folds on-path deltas.
Scenarios covered:
1. **Basic migration (sync + async)**: build pre-migration state with
`BinaryOperatorAggregate`, swap the annotation to `DeltaChannel` on
the same checkpointer, and verify that every settled pre-migration
super-step boundary (`next=('__start__',)`) round-trips exactly
under the delta-channel view.
2. **Time travel into a pre-migration checkpoint** after migration
`graph.get_state(pre_migration_config)` at a settled ancestor
returns the same state as under the binop channel.
3. **Continuing a migrated thread**: driving one more super-step after
migration produces a state that includes the pre-migration settled
prefix plus the new delta write proving `from_checkpoint(seed)` +
`replay_writes` correctly fold post-migration deltas onto the
pre-migration seed.
4. **Base-saver fallback path**: a third-party-style subclass that
removes the optimized `InMemorySaver` override and falls back to
`BaseCheckpointSaver._get_channel_writes_history` must produce the
same result as the optimized path.
5. **Channel-type isolation across threads**: two threads on the same
checkpointer under the delta-channel graph one freshly-started,
one migrated from pre-migration state don't cross-contaminate.
The parent-chain walk is scoped to the thread.
TODO: add postgres variants in the existing `libs/checkpoint-postgres`
test files (different fixture setup; not this file).
"""
from __future__ import annotations
import operator
from typing import Annotated, Any
import pytest
from langgraph.checkpoint.memory import InMemorySaver
from typing_extensions import TypedDict
from langgraph.channels.binop import BinaryOperatorAggregate
from langgraph.channels._delta import DeltaChannel
from langgraph.graph import END, START, StateGraph
pytestmark = pytest.mark.anyio
# ---------------------------------------------------------------------------
# Graph factories
#
# A minimal reducer (`operator.add` on lists of str) with a noop node keeps
# state change localized to the HumanMessage-like payload passed through
# `invoke`. That isolates the pre/post-migration parity assertions to
# channel-hydration semantics.
# ---------------------------------------------------------------------------
def _noop(_state: Any) -> dict:
return {}
def _binop_graph(checkpointer: Any) -> Any:
class BinopState(TypedDict):
items: Annotated[list, BinaryOperatorAggregate(list, operator.add)]
return (
StateGraph(BinopState)
.add_node("noop", _noop)
.add_edge(START, "noop")
.add_edge("noop", END)
.compile(checkpointer=checkpointer)
)
def _delta_graph(checkpointer: Any) -> Any:
class DeltaState(TypedDict):
items: Annotated[list, DeltaChannel(operator.add)]
return (
StateGraph(DeltaState)
.add_node("noop", _noop)
.add_edge(START, "noop")
.add_edge("noop", END)
.compile(checkpointer=checkpointer)
)
def _drive(graph: Any, config: dict, tag: str, n: int) -> None:
for i in range(n):
graph.invoke({"items": [f"{tag}{i}"]}, config)
async def _adrive(graph: Any, config: dict, tag: str, n: int) -> None:
for i in range(n):
await graph.ainvoke({"items": [f"{tag}{i}"]}, config)
def _settled_boundaries(history: list) -> list[tuple[dict, list]]:
"""Return `[(config, items), ...]` for every checkpoint in `history`
whose `next == ('__start__',)` the stable boundaries between invokes.
"""
return [
(s.config, list(s.values.get("items", [])))
for s in history
if s.next == ("__start__",)
]
# ---------------------------------------------------------------------------
# 1. Basic migration (sync + async)
# ---------------------------------------------------------------------------
def test_basic_migration_preserves_pre_migration_state() -> None:
"""Build state under `BinaryOperatorAggregate`, migrate to
`DeltaChannel` on the same checkpointer, and verify that every
settled pre-migration super-step boundary round-trips exactly.
Settled boundaries (`next=('__start__',)`) are the stable hydration
targets for the migration path: writes that produced the NEXT
super-step are kept as `pending_writes` on the ancestor, so walking
from a descendant finds the ancestor's blob as the seed and
reconstructs the correct state.
"""
checkpointer = InMemorySaver()
config = {"configurable": {"thread_id": "basic-sync"}}
# Pre-migration: accumulate items across 3 invokes.
binop = _binop_graph(checkpointer)
_drive(binop, config, "u", 3)
pre_boundaries = _settled_boundaries(list(binop.get_state_history(config)))
assert len(pre_boundaries) >= 2, "expected multiple settled boundaries"
# Migrate: swap the annotation on the same checkpointer.
delta = _delta_graph(checkpointer)
for cfg, items in pre_boundaries:
snap = delta.get_state(cfg)
assert list(snap.values.get("items", [])) == items, (
f"snapshot mismatch at {cfg['configurable']['checkpoint_id']}: "
f"expected {items}, got {snap.values.get('items', [])}"
)
async def test_basic_migration_preserves_pre_migration_state_async() -> None:
"""Async variant of the basic migration scenario."""
checkpointer = InMemorySaver()
config = {"configurable": {"thread_id": "basic-async"}}
binop = _binop_graph(checkpointer)
await _adrive(binop, config, "u", 3)
pre_history = [s async for s in binop.aget_state_history(config)]
pre_boundaries = _settled_boundaries(pre_history)
assert len(pre_boundaries) >= 2
delta = _delta_graph(checkpointer)
for cfg, items in pre_boundaries:
snap = await delta.aget_state(cfg)
assert list(snap.values.get("items", [])) == items, (
f"async snapshot mismatch at {cfg['configurable']['checkpoint_id']}"
)
# ---------------------------------------------------------------------------
# 2. Time travel into a pre-migration checkpoint after migration
# ---------------------------------------------------------------------------
def test_time_travel_into_pre_migration_checkpoint() -> None:
"""After migration, `graph.get_state(pre_migration_config)` at a
settled ancestor returns the state as stored at that point."""
checkpointer = InMemorySaver()
config = {"configurable": {"thread_id": "time-travel"}}
binop = _binop_graph(checkpointer)
_drive(binop, config, "u", 3)
pre_boundaries = _settled_boundaries(list(binop.get_state_history(config)))
assert pre_boundaries, "no settled ancestors to time-travel to"
delta = _delta_graph(checkpointer)
# Pick the oldest non-empty boundary — a long distance to walk back.
non_empty = [(cfg, items) for cfg, items in pre_boundaries if items]
assert non_empty, "expected at least one non-empty boundary"
target_cfg, expected_items = non_empty[-1]
snap = delta.get_state(target_cfg)
assert list(snap.values.get("items", [])) == expected_items
# ---------------------------------------------------------------------------
# 3. Continuing a migrated thread: deltas fold onto pre-migration seed
# ---------------------------------------------------------------------------
def test_continuing_migrated_thread_folds_deltas_on_seed() -> None:
"""Resume a pre-migration settled ancestor via `invoke(None, cfg)`
under the delta-channel graph. Since the pre-migration checkpoint
has an existing `pending_writes` entry (the input for the NEXT
super-step), re-running from that ancestor reproduces the same
post-ancestor state as the original binop run.
This proves the seed-terminator + write-replay pipeline works
end-to-end across the migration boundary.
"""
checkpointer = InMemorySaver()
config = {"configurable": {"thread_id": "continue"}}
binop = _binop_graph(checkpointer)
_drive(binop, config, "u", 2)
# Pick the oldest settled boundary with non-empty state.
pre_boundaries = _settled_boundaries(list(binop.get_state_history(config)))
target_cfg, seed_items = next(
(cfg, items) for cfg, items in reversed(pre_boundaries) if items
)
assert seed_items, "need a non-empty seed boundary"
# Migrate and resume from the pre-migration ancestor. `invoke(None,
# cfg)` replays the pending writes staged at `cfg` under the new
# channel; the reducer folds those deltas onto the seed.
delta = _delta_graph(checkpointer)
result = delta.invoke(None, target_cfg)
# The resumed state must include the pre-migration seed items in order.
result_items = list(result.get("items", []))
for idx, prefix_item in enumerate(seed_items):
assert result_items[idx] == prefix_item, (
f"pre-migration seed item at {idx} not preserved: "
f"got {result_items[: idx + 1]}, expected {seed_items}"
)
# ---------------------------------------------------------------------------
# 4. Base-saver fallback path
# ---------------------------------------------------------------------------
class _ThirdPartyStyleSaver(InMemorySaver):
"""Simulates a third-party saver that inherits the reference
`_get_channel_writes_history` implementation from
`BaseCheckpointSaver` rather than overriding it.
We rebind the two methods to the base-class versions (via MRO) so
the fallback path is exercised even though the storage layer is
still the in-memory one.
"""
# MRO: [_ThirdPartyStyleSaver, InMemorySaver, BaseCheckpointSaver, ...]
_get_channel_writes_history = ( # type: ignore[assignment]
InMemorySaver.__mro__[1]._get_channel_writes_history # type: ignore[attr-defined]
)
_aget_channel_writes_history = ( # type: ignore[assignment]
InMemorySaver.__mro__[1]._aget_channel_writes_history # type: ignore[attr-defined]
)
def test_base_saver_fallback_matches_optimized_override() -> None:
"""The reference `BaseCheckpointSaver` implementation must produce
the same migration behavior as the optimized `InMemorySaver`
override. We drive the same migration scenario through both savers
and assert per-snapshot parity in the delta-channel view."""
# Fast path: optimized InMemorySaver override.
fast_saver = InMemorySaver()
fast_config = {"configurable": {"thread_id": "fast"}}
fast_binop = _binop_graph(fast_saver)
_drive(fast_binop, fast_config, "u", 3)
fast_delta = _delta_graph(fast_saver)
fast_history = [
(s.next, list(s.values.get("items", [])))
for s in fast_delta.get_state_history(fast_config)
]
# Slow path: base-class fallback.
slow_saver = _ThirdPartyStyleSaver()
slow_config = {"configurable": {"thread_id": "slow"}}
slow_binop = _binop_graph(slow_saver)
_drive(slow_binop, slow_config, "u", 3)
slow_delta = _delta_graph(slow_saver)
slow_history = [
(s.next, list(s.values.get("items", [])))
for s in slow_delta.get_state_history(slow_config)
]
assert slow_history == fast_history, (
"base-saver fallback should match optimized-override behavior; "
f"fast={fast_history}, slow={slow_history}"
)
# ---------------------------------------------------------------------------
# 5. Thread isolation under mixed-generation storage
# ---------------------------------------------------------------------------
def test_delta_and_migrated_threads_do_not_cross_contaminate() -> None:
"""Two threads sharing a checkpointer — one migrated from
pre-migration state, one freshly-started under DeltaChannel must
maintain independent state. The parent-chain walk in
`_get_channel_writes_history` must be scoped to the target thread.
"""
checkpointer = InMemorySaver()
migrated_cfg = {"configurable": {"thread_id": "migrated"}}
fresh_cfg = {"configurable": {"thread_id": "fresh"}}
# Thread A: pre-migration build-up.
binop = _binop_graph(checkpointer)
_drive(binop, migrated_cfg, "m", 2)
# Thread B: fresh delta-channel run.
delta = _delta_graph(checkpointer)
_drive(delta, fresh_cfg, "f", 2)
# Thread A: migrate and confirm its state is anchored in its own
# thread's pre-migration history (tag 'm'), never mixing in tag 'f'.
migrated_boundaries = _settled_boundaries(
list(delta.get_state_history(migrated_cfg))
)
assert migrated_boundaries, "migrated thread has no settled boundaries"
for _, items in migrated_boundaries:
for it in items:
assert it.startswith("m"), (
f"migrated thread leaked item from other thread: {it}"
)
# Thread B: settled boundaries must only contain 'f' tags.
fresh_boundaries = _settled_boundaries(list(delta.get_state_history(fresh_cfg)))
assert fresh_boundaries, "fresh thread has no settled boundaries"
for _, items in fresh_boundaries:
for it in items:
assert it.startswith("f"), (
f"fresh thread leaked item from migrated thread: {it}"
)
# ---------------------------------------------------------------------------
# 6. Tip-of-pre-migration hydration: the latest checkpoint from a binop-run
# thread has a real accumulated value in its own `channel_values["items"]`.
# When hydrated under the delta-channel graph via `get_state(config)` with no
# `checkpoint_id`, the short-circuit must use that value directly instead of
# walking ancestors (which would skip the tip's own blob).
# ---------------------------------------------------------------------------
def test_tip_of_pre_migration_hydrates_directly() -> None:
"""`graph.get_state(config)` at the latest (pre-migration) checkpoint
returns the full accumulated list stored in that checkpoint's own
`channel_values`. The hydration must not walk ancestors past it."""
checkpointer = InMemorySaver()
config = {"configurable": {"thread_id": "tip-sync"}}
binop = _binop_graph(checkpointer)
_drive(binop, config, "u", 3)
binop_tip = binop.get_state(config)
expected_items = list(binop_tip.values.get("items", []))
assert expected_items == ["u0", "u1", "u2"], (
f"sanity: pre-migration tip should accumulate all 3 items, got {expected_items}"
)
delta = _delta_graph(checkpointer)
snap = delta.get_state(config)
assert list(snap.values.get("items", [])) == expected_items, (
f"tip hydration mismatch: expected {expected_items}, "
f"got {snap.values.get('items', [])}"
)
async def test_tip_of_pre_migration_hydrates_directly_async() -> None:
"""Async variant of the tip-of-pre-migration hydration scenario."""
checkpointer = InMemorySaver()
config = {"configurable": {"thread_id": "tip-async"}}
binop = _binop_graph(checkpointer)
await _adrive(binop, config, "u", 3)
binop_tip = await binop.aget_state(config)
expected_items = list(binop_tip.values.get("items", []))
assert expected_items == ["u0", "u1", "u2"]
delta = _delta_graph(checkpointer)
snap = await delta.aget_state(config)
assert list(snap.values.get("items", [])) == expected_items, (
f"async tip hydration mismatch: expected {expected_items}, "
f"got {snap.values.get('items', [])}"
)
# ---------------------------------------------------------------------------
# 7. `update_state` after migration writes a real value to the new
# checkpoint's `channel_values` (not a sentinel). Hydration must use it
# directly — the ancestor walk would skip this blob and return stale state.
# ---------------------------------------------------------------------------
def test_update_state_after_migration_uses_written_value() -> None:
"""After migrating and running at least one post-migration super-step
(so the thread's tip has a `DELTA_SENTINEL`), `update_state` writes a
concrete value to a new checkpoint's `channel_values`. `get_state`
must reflect that concrete value."""
checkpointer = InMemorySaver()
config = {"configurable": {"thread_id": "update-state"}}
# Pre-migration: accumulate a little state.
binop = _binop_graph(checkpointer)
_drive(binop, config, "u", 2)
# Migrate and run one more super-step so the tip is a post-migration
# checkpoint with `DELTA_SENTINEL` in its own `channel_values`.
delta = _delta_graph(checkpointer)
delta.invoke({"items": ["post"]}, config)
# `update_state` writes a concrete value into a new checkpoint's blob
# via the reducer against the hydrated prior state.
delta.update_state(config, {"items": ["x", "y"]})
snap = delta.get_state(config)
updated_items = list(snap.values.get("items", []))
# Must include the "x","y" update; without the hydration fix, the
# update_state-written blob would be skipped in favor of an ancestor
# walk, and the update values would disappear.
assert "x" in updated_items and "y" in updated_items, (
f"update_state values missing from snapshot: {updated_items}"
)
# The "x","y" items should be folded onto the prior accumulated state,
# not stand alone. This verifies the update-written blob is used
# directly by `get_state` (no ancestor walk past it).
assert len(updated_items) >= 4, (
f"update_state snapshot should preserve pre-update state, got {updated_items}"
)
assert updated_items[-2:] == ["x", "y"], (
f"update_state deltas should be at the tail, got {updated_items}"
)
# ---------------------------------------------------------------------------
# 8. Fork from an `update_state` checkpoint: a new run branched off the
# update_state-produced checkpoint must see that checkpoint's concrete
# `channel_values` as its base, with new deltas folded on top.
# ---------------------------------------------------------------------------
def test_fork_from_update_state_checkpoint() -> None:
"""Branching a new run from the checkpoint produced by `update_state`
must use that checkpoint's concrete blob as the base. Additional
deltas from the forked run fold onto it through the reducer."""
checkpointer = InMemorySaver()
config = {"configurable": {"thread_id": "fork"}}
# Pre-migration build-up, then migrate and add one post-migration step.
binop = _binop_graph(checkpointer)
_drive(binop, config, "u", 2)
delta = _delta_graph(checkpointer)
delta.invoke({"items": ["post"]}, config)
# Apply `update_state` and capture the returned config (references
# the new checkpoint produced by the update).
update_cfg = delta.update_state(config, {"items": ["x", "y"]})
update_snap = delta.get_state(update_cfg)
base_items = list(update_snap.values.get("items", []))
assert "x" in base_items and "y" in base_items, (
f"update_state values missing from snapshot: {base_items}"
)
assert base_items[-2:] == ["x", "y"], (
f"sanity: update_state deltas should be at the tail, got {base_items}"
)
# Fork: invoke from the update_state checkpoint with a new delta.
forked = delta.invoke({"items": ["fork0"]}, update_cfg)
forked_items = list(forked.get("items", []))
# The fork must see the update_state-written blob as its base (not
# walk past it), and the new delta must fold on top of it.
assert forked_items[: len(base_items)] == base_items, (
f"fork lost update_state base: base={base_items}, forked={forked_items}"
)
assert forked_items[-1] == "fork0", f"fork delta not appended: {forked_items}"
@@ -275,3 +275,70 @@ def test_graph_callbacks_accept_base_callback_manager() -> None:
assert "__interrupt__" in first
assert len(graph_handler.interrupt_events) == 1
def test_non_graph_handler_via_add_handler_does_not_crash() -> None:
"""Non-GraphCallbackHandler added via add_handler should not raise.
Libraries like opentelemetry-instrumentation-langchain monkey-patch
BaseCallbackManager.__init__ and inject handlers via add_handler().
These handlers inherit from BaseCallbackHandler, not
GraphCallbackHandler. They must be silently accepted graph lifecycle
events will simply not be dispatched to them.
"""
from langgraph.callbacks import _GraphCallbackManager
manager = _GraphCallbackManager()
plain_handler = _LangChainCustomEventHandler()
manager.add_handler(plain_handler, inherit=True)
assert plain_handler in manager.handlers
def test_non_graph_handler_does_not_receive_lifecycle_events() -> None:
"""Non-GraphCallbackHandler added alongside a GraphCallbackHandler
should not interfere with lifecycle event dispatch."""
graph = _build_interrupt_graph()
graph_handler = _GraphEventHandler()
plain_handler = _LangChainCustomEventHandler()
config = {
"configurable": {"thread_id": "graph-callback-mixed-handlers"},
"callbacks": [plain_handler, graph_handler],
}
first = graph.invoke({"answer": None}, config)
assert "__interrupt__" in first
assert len(graph_handler.interrupt_events) == 1
assert plain_handler.events == []
resumed = graph.invoke(Command(resume="done"), config)
assert resumed == {"answer": "done"}
assert len(graph_handler.resume_events) == 1
assert plain_handler.events == []
@pytest.mark.anyio
@NEEDS_CONTEXTVARS
async def test_non_graph_handler_does_not_receive_lifecycle_events_async() -> None:
"""Async variant: non-GraphCallbackHandler should not interfere."""
graph = _build_interrupt_graph()
graph_handler = _GraphEventHandler()
plain_handler = _LangChainCustomEventHandler()
config = {
"configurable": {"thread_id": "graph-callback-mixed-handlers-async"},
"callbacks": [plain_handler, graph_handler],
}
first = await graph.ainvoke({"answer": None}, config)
assert "__interrupt__" in first
assert len(graph_handler.interrupt_events) == 1
assert plain_handler.events == []
resumed = await graph.ainvoke(Command(resume="done"), config)
assert resumed == {"answer": "done"}
assert len(graph_handler.resume_events) == 1
assert plain_handler.events == []
+118
View File
@@ -5,6 +5,7 @@ import langchain_core
import pytest
from langchain_core.messages import (
AIMessage,
AIMessageChunk,
AnyMessage,
HumanMessage,
RemoveMessage,
@@ -338,6 +339,123 @@ def test_remove_all_messages():
]
def test_fast_path_preserves_format_openai():
"""Pure-append fast path must still apply the `langchain-openai` formatter."""
left = [HumanMessage(content="prior", id="1")]
right = [
AIMessage(
content=[
{
"type": "tool_use",
"name": "foo",
"input": {"bar": "baz"},
"id": "t1",
}
],
id="2",
)
]
result = add_messages(left, right, format="langchain-openai")
assert isinstance(result[0], HumanMessage)
assert result[0].content == "prior"
assert isinstance(result[1], AIMessage)
# formatter collapses the tool_use content block into `tool_calls`
assert result[1].content == ""
assert len(result[1].tool_calls) == 1
assert result[1].tool_calls[0]["name"] == "foo"
assert result[1].tool_calls[0]["args"] == {"bar": "baz"}
assert result[1].tool_calls[0]["id"] == "t1"
def test_fast_path_rejects_invalid_format():
"""Pure-append fast path must validate the `format` arg like the slow path."""
left = [HumanMessage(content="prior", id="1")]
right = [AIMessage(content="new", id="2")]
with pytest.raises(ValueError, match="Unrecognized format="):
add_messages(left, right, format="bogus") # type: ignore[arg-type]
def test_left_starting_with_chunk_is_normalized():
"""Opt-1 guard: a `BaseMessageChunk` at left[0] must trigger full conversion."""
chunk = AIMessageChunk(content="chunk", id="c1")
result = add_messages([chunk], [HumanMessage(content="h", id="h1")])
assert len(result) == 2
# chunk must be converted to a non-chunk message
assert type(result[0]).__name__ == "AIMessage"
assert result[0].id == "c1"
assert result[1].id == "h1"
def test_left_as_dicts_is_normalized():
"""Opt-1 guard: dicts at left[0] must trigger full conversion."""
left = [{"role": "user", "content": "hi", "id": "d1"}]
right = [AIMessage(content="reply", id="a1")]
result = add_messages(left, right)
assert len(result) == 2
assert isinstance(result[0], HumanMessage)
assert result[0].id == "d1"
assert result[0].content == "hi"
def test_left_as_tuples_is_normalized():
"""Opt-1 guard: tuple-form messages must trigger full conversion."""
left = [("user", "hi")]
right = [AIMessage(content="reply", id="a1")]
result = add_messages(left, right)
assert len(result) == 2
assert isinstance(result[0], HumanMessage)
# id is auto-assigned
assert isinstance(result[0].id, str) and UUID(result[0].id, version=4)
def test_left_first_msg_missing_id_is_normalized():
"""Opt-1 guard: a BaseMessage without an id at left[0] falls to the else branch."""
left = [HumanMessage(content="hi")] # no id
right = [AIMessage(content="reply", id="a1")]
result = add_messages(left, right)
assert len(result) == 2
# left's id must have been auto-assigned
assert isinstance(result[0].id, str) and UUID(result[0].id, version=4)
def test_duplicate_ids_in_right_with_nonempty_left():
"""Opt-2 guard: intra-right duplicate ids must take slow path (dedup kept)."""
left = [HumanMessage(content="prior", id="1")]
right = [
AIMessage(content="first", id="2"),
AIMessage(content="second", id="2"),
]
result = add_messages(left, right)
assert len(result) == 2
assert result[0].id == "1"
assert result[1].id == "2"
assert result[1].content == "second"
def test_right_with_none_ids_pure_append():
"""Fast path still correct when right entries start with id=None (fresh uuids assigned)."""
left = [HumanMessage(content="prior", id="1")]
right = [AIMessage(content="a"), AIMessage(content="b")]
result = add_messages(left, right)
assert len(result) == 3
assert result[0].id == "1"
for m in result[1:]:
assert isinstance(m.id, str) and UUID(m.id, version=4)
# fresh uuids must be distinct
assert result[1].id != result[2].id
def test_fast_path_returns_fresh_list():
"""Fast path must return a new list object (not mutate or alias left)."""
left = [HumanMessage(content="prior", id="1")]
right = [AIMessage(content="new", id="2")]
result = add_messages(left, right)
assert result is not left
# left must be untouched
assert len(left) == 1
assert left[0].id == "1"
def test_push_messages_in_graph():
class MessagesState(TypedDict):
messages: Annotated[list[AnyMessage], add_messages]
+193 -3
View File
@@ -615,8 +615,11 @@ def test_run_from_checkpoint_id_retains_previous_writes(
)
]
assert len(new_history) == len(history) + 1
for original, new in zip(history, new_history[1:]):
# +2: one fork checkpoint from time travel, one from the new execution
assert len(new_history) == len(history) + 2
# new_history[0] is the new execution result, new_history[1] is the fork
assert new_history[1].metadata["source"] == "fork"
for original, new in zip(history, new_history[2:]):
assert original.values == new.values
assert original.next == new.next
assert original.metadata["step"] == new.metadata["step"]
@@ -624,7 +627,7 @@ def test_run_from_checkpoint_id_retains_previous_writes(
def _get_tasks(hist: list, start: int):
return [h.tasks for h in hist[start:]]
assert _get_tasks(new_history, 1) == _get_tasks(history, 0)
assert _get_tasks(new_history, 2) == _get_tasks(history, 0)
def test_batch_two_processes_in_out() -> None:
@@ -9397,3 +9400,190 @@ def test_fork_does_not_apply_pending_writes(
# Should be: 1 (input) + 20 (forked node_a) + 100 (node_b) = 121
assert result == {"value": 121}
async def test_delta_channel_end_to_end_inmemory() -> None:
"""Full graph run: DeltaChannel accumulates correctly across multiple turns."""
from langchain_core.messages import AIMessage, HumanMessage
from langgraph.checkpoint.memory import InMemorySaver
from langgraph.channels._delta import DeltaChannel
from langgraph.graph import START, StateGraph
from langgraph.graph.message import add_messages
class State(TypedDict):
messages: Annotated[list, DeltaChannel(add_messages)]
def respond(state: State) -> dict:
n = len(state["messages"])
return {"messages": [AIMessage(content=f"reply-{n}", id=f"ai-{n}")]}
builder = StateGraph(State)
builder.add_node("respond", respond)
builder.add_edge(START, "respond")
graph = builder.compile(checkpointer=InMemorySaver())
config = {"configurable": {"thread_id": "diff-test-1"}}
# Turn 1
graph.invoke({"messages": [HumanMessage(content="hello", id="h1")]}, config)
# Turn 2
graph.invoke({"messages": [HumanMessage(content="world", id="h2")]}, config)
# Turn 3
graph.invoke({"messages": [HumanMessage(content="bye", id="h3")]}, config)
state = graph.get_state(config)
msgs = state.values["messages"]
# 3 human + 3 AI = 6 total
assert len(msgs) == 6, f"expected 6 messages, got {len(msgs)}: {msgs}"
assert msgs[0].content == "hello"
assert msgs[2].content == "world"
assert msgs[4].content == "bye"
assert msgs[1].content == "reply-1"
assert msgs[3].content == "reply-3"
assert msgs[5].content == "reply-5"
async def test_delta_channel_time_travel() -> None:
"""Time-travel back to turn-1 checkpoint and resume; continuation must not include turn-2 deltas."""
from langchain_core.messages import AIMessage, HumanMessage
from langgraph.checkpoint.memory import InMemorySaver
from langgraph.channels._delta import DeltaChannel
from langgraph.graph import START, StateGraph
from langgraph.graph.message import add_messages
class State(TypedDict):
messages: Annotated[list, DeltaChannel(add_messages)]
counter = {"n": 0}
def respond(state: State) -> dict:
counter["n"] += 1
return {
"messages": [
AIMessage(content=f"ai-{counter['n']}", id=f"ai-{counter['n']}")
]
}
builder = StateGraph(State)
builder.add_node("respond", respond)
builder.add_edge(START, "respond")
saver = InMemorySaver()
graph = builder.compile(checkpointer=saver)
config = {"configurable": {"thread_id": "diff-time-travel"}}
# Run 2 turns: h1→ai-1, h2→ai-2
graph.invoke({"messages": [HumanMessage(content="h1", id="h1")]}, config)
graph.invoke({"messages": [HumanMessage(content="h2", id="h2")]}, config)
# Find the checkpoint after turn 1 (2 messages: h1 + ai-1)
history = list(graph.get_state_history(config))
after_turn1 = next(h for h in history if len(h.values.get("messages", [])) == 2)
assert len(after_turn1.values["messages"]) == 2
assert after_turn1.values["messages"][0].content == "h1"
assert after_turn1.values["messages"][1].content == "ai-1"
# Resume from turn-1 checkpoint: inject h3, expect 3 messages total (h1, ai-1, ai-N)
# NOT 5 messages (turn-2 deltas must not bleed into the resumed run)
result = graph.invoke(
{"messages": [HumanMessage(content="h3", id="h3")]},
after_turn1.config,
)
msgs = result["messages"]
# Should be: h1, ai-1, h3, ai-N — 4 messages total
assert len(msgs) == 4, (
f"expected 4 messages after time-travel resume, got {len(msgs)}: {msgs}"
)
assert msgs[0].content == "h1"
assert msgs[1].content == "ai-1"
assert msgs[2].content == "h3"
async def test_delta_channel_remove_message_end_to_end() -> None:
"""RemoveMessage inside a DeltaChannel graph must persist and reload correctly."""
from langchain_core.messages import AIMessage, HumanMessage, RemoveMessage
from langgraph.checkpoint.memory import InMemorySaver
from langgraph.channels._delta import DeltaChannel
from langgraph.graph import START, StateGraph
from langgraph.graph.message import add_messages
class State(TypedDict):
messages: Annotated[list, DeltaChannel(add_messages)]
def respond(state: State) -> dict:
return {"messages": [AIMessage(content="reply", id="ai-1")]}
def delete_first(state: State) -> dict:
# removes the first message
return {"messages": [RemoveMessage(id=state["messages"][0].id)]}
builder = StateGraph(State)
builder.add_node("respond", respond)
builder.add_node("delete_first", delete_first)
builder.add_edge(START, "respond")
builder.add_edge("respond", "delete_first")
graph = builder.compile(checkpointer=InMemorySaver())
config = {"configurable": {"thread_id": "diff-remove-test"}}
graph.invoke({"messages": [HumanMessage(content="hello", id="h1")]}, config)
state = graph.get_state(config)
msgs = state.values["messages"]
# h1 was removed, only ai-1 should remain
assert len(msgs) == 1, f"expected 1 message, got {len(msgs)}: {msgs}"
assert msgs[0].id == "ai-1"
# A subsequent turn must reconstruct from the checkpoint correctly
graph.invoke({"messages": [HumanMessage(content="again", id="h2")]}, config)
state = graph.get_state(config)
msgs = state.values["messages"]
# ai-1 + h2 + ai-1(second reply, same id overwrites) + h2 removed
# more simply: after second run we expect ai-1 updated + h2 remaining minus deleted h2
# just assert h1 is still gone
assert all(m.id != "h1" for m in msgs), (
"h1 should still be absent after second turn"
)
async def test_delta_channel_update_by_id_end_to_end() -> None:
"""Updating a message by ID via DeltaChannel must persist and reload correctly."""
from langchain_core.messages import HumanMessage
from langgraph.checkpoint.memory import InMemorySaver
from langgraph.channels._delta import DeltaChannel
from langgraph.graph import START, StateGraph
from langgraph.graph.message import add_messages
class State(TypedDict):
messages: Annotated[list, DeltaChannel(add_messages)]
def update_msg(state: State) -> dict:
# re-send h1 with updated content
return {"messages": [HumanMessage(content="updated", id="h1")]}
builder = StateGraph(State)
builder.add_node("update_msg", update_msg)
builder.add_edge(START, "update_msg")
graph = builder.compile(checkpointer=InMemorySaver())
config = {"configurable": {"thread_id": "diff-update-id-test"}}
graph.invoke({"messages": [HumanMessage(content="original", id="h1")]}, config)
state = graph.get_state(config)
msgs = state.values["messages"]
assert len(msgs) == 1, f"expected 1 message, got {len(msgs)}: {msgs}"
assert msgs[0].content == "updated"
assert msgs[0].id == "h1"
# Second turn: verify the updated state is the base for further accumulation
graph.invoke({"messages": [HumanMessage(content="new", id="h2")]}, config)
state = graph.get_state(config)
msgs = state.values["messages"]
ids = [m.id for m in msgs]
assert "h1" in ids # h1 persists (updated, not duplicated)
assert "h2" in ids
assert ids.count("h1") == 1, "h1 must not be duplicated"
+6 -3
View File
@@ -2086,8 +2086,11 @@ async def test_run_from_checkpoint_id_retains_previous_writes(
)
]
assert len(new_history) == len(history) + 1
for original, new in zip(history, new_history[1:]):
# +2: one fork checkpoint from time travel, one from the new execution
assert len(new_history) == len(history) + 2
# new_history[0] is the new execution result, new_history[1] is the fork
assert new_history[1].metadata["source"] == "fork"
for original, new in zip(history, new_history[2:]):
assert original.values == new.values
assert original.next == new.next
assert original.metadata["step"] == new.metadata["step"]
@@ -2095,7 +2098,7 @@ async def test_run_from_checkpoint_id_retains_previous_writes(
def _get_tasks(hist: list, start: int):
return [h.tasks for h in hist[start:]]
assert _get_tasks(new_history, 1) == _get_tasks(history, 0)
assert _get_tasks(new_history, 2) == _get_tasks(history, 0)
async def test_cond_edge_after_send() -> None:
File diff suppressed because it is too large Load Diff
@@ -1,258 +0,0 @@
"""Tests for LifecycleTransformer — derives subgraph lifecycle from ns discovery."""
from __future__ import annotations
import operator
import time
from typing import Annotated, Any
import pytest
from typing_extensions import TypedDict
from langgraph.constants import END, START
from langgraph.errors import GraphInterrupt
from langgraph.graph import StateGraph
from langgraph.stream._mux import StreamMux
from langgraph.stream._types import ProtocolEvent
from langgraph.stream.transformers import (
LifecycleTransformer,
ValuesTransformer,
)
TS = int(time.time() * 1000)
def _event(method: str, data: Any, *, namespace: list[str]) -> ProtocolEvent:
return {
"type": "event",
"method": method,
"params": {
"namespace": namespace,
"timestamp": TS,
"data": data,
},
}
def _drain_channel(mux: StreamMux) -> list[dict[str, Any]]:
ch = mux.extensions["lifecycle"]
return list(ch._log._items) # type: ignore[attr-defined]
def _drain_main_events(mux: StreamMux) -> list[ProtocolEvent]:
return list(mux._events._items) # type: ignore[attr-defined]
def _subscribe(mux: StreamMux) -> None:
ch = mux.extensions["lifecycle"]
ch._log._subscribed = True # type: ignore[attr-defined]
mux._events._subscribed = True # type: ignore[attr-defined]
class TestLifecycleTransformerUnit:
def _mux(self) -> StreamMux:
mux = StreamMux(
factories=[ValuesTransformer, LifecycleTransformer], is_async=False
)
_subscribe(mux)
return mux
def test_first_event_at_child_ns_emits_started(self) -> None:
mux = self._mux()
mux.push(_event("values", {"v": 1}, namespace=["child:task_a"]))
events = _drain_channel(mux)
assert events == [
{
"event": "started",
"namespace": ["child:task_a"],
"graph_name": "child",
"trigger_call_id": "task_a",
}
]
def test_no_started_at_root_ns(self) -> None:
mux = self._mux()
mux.push(_event("values", {"v": 1}, namespace=[]))
assert _drain_channel(mux) == []
def test_method_agnostic_discovery(self) -> None:
mux = self._mux()
mux.push(_event("messages", "x", namespace=["c:t"]))
(started,) = _drain_channel(mux)
assert started["event"] == "started"
assert started["namespace"] == ["c:t"]
def test_repeated_events_single_started(self) -> None:
mux = self._mux()
mux.push(_event("values", {"v": 1}, namespace=["c:t"]))
mux.push(_event("values", {"v": 2}, namespace=["c:t"]))
mux.push(_event("updates", {"n": "x"}, namespace=["c:t"]))
events = _drain_channel(mux)
assert len(events) == 1
assert events[0]["event"] == "started"
def test_ns_without_task_id(self) -> None:
mux = self._mux()
mux.push(_event("values", {"v": 1}, namespace=["child"]))
(started,) = _drain_channel(mux)
assert started["graph_name"] == "child"
assert "trigger_call_id" not in started
def test_finalize_emits_completed(self) -> None:
mux = self._mux()
mux.push(_event("values", {"v": 1}, namespace=["c:t"]))
mux.close()
events = _drain_channel(mux)
assert events == [
{
"event": "started",
"namespace": ["c:t"],
"graph_name": "c",
"trigger_call_id": "t",
},
{"event": "completed", "namespace": ["c:t"]},
]
def test_fail_with_graph_interrupt(self) -> None:
mux = self._mux()
mux.push(_event("values", {"v": 1}, namespace=["c:t"]))
mux.fail(GraphInterrupt())
events = _drain_channel(mux)
assert events[-1] == {"event": "interrupted", "namespace": ["c:t"]}
def test_fail_with_generic_error(self) -> None:
mux = self._mux()
mux.push(_event("values", {"v": 1}, namespace=["c:t"]))
mux.fail(RuntimeError("boom"))
events = _drain_channel(mux)
assert events[-1] == {
"event": "failed",
"namespace": ["c:t"],
"error": "boom",
}
class TestLifecycleWireFormat:
"""Native transformer: method on the wire is `"lifecycle"`, no `custom:` prefix."""
def test_wire_method_is_lifecycle_unprefixed(self) -> None:
mux = StreamMux(factories=[LifecycleTransformer], is_async=False)
_subscribe(mux)
mux.push(_event("values", {"v": 1}, namespace=["c:t"]))
# Find the lifecycle event in the main log.
lifecycle_events = [
ev for ev in _drain_main_events(mux) if ev["method"] == "lifecycle"
]
assert len(lifecycle_events) == 1
assert lifecycle_events[0]["params"]["data"]["event"] == "started"
# Also verify no `custom:lifecycle` leaks through.
assert not any(
ev["method"].startswith("custom:") for ev in _drain_main_events(mux)
)
def test_started_precedes_originating_event_on_wire(self) -> None:
"""Seq ordering: synthesized lifecycle event lands before the event that triggered it."""
mux = StreamMux(
factories=[ValuesTransformer, LifecycleTransformer], is_async=False
)
_subscribe(mux)
mux.push(_event("values", {"v": 1}, namespace=["c:t"]))
wire = _drain_main_events(mux)
methods = [ev["method"] for ev in wire]
# Lifecycle's synthetic event is forwarded during process() and
# gets an earlier seq than the originating values event.
assert methods.index("lifecycle") < methods.index("values")
# ---------------------------------------------------------------------------
# End-to-end via stream_v2
# ---------------------------------------------------------------------------
class SimpleState(TypedDict):
value: str
items: Annotated[list[str], operator.add]
def _build_nested_graph():
def inner_node(state: SimpleState) -> dict:
return {"value": state["value"] + "X", "items": ["x"]}
inner_builder = StateGraph(SimpleState)
inner_builder.add_node("inner_node", inner_node)
inner_builder.add_edge(START, "inner_node")
inner_builder.add_edge("inner_node", END)
inner = inner_builder.compile()
outer_builder = StateGraph(SimpleState)
outer_builder.add_node("sub", inner)
outer_builder.add_edge(START, "sub")
outer_builder.add_edge("sub", END)
return outer_builder.compile()
class TestLifecycleEndToEnd:
def test_real_run_emits_started_and_completed(self) -> None:
graph = _build_nested_graph()
run = graph.stream_v2(
{"value": "", "items": []}, transformers=[LifecycleTransformer]
)
# Drain the run so finalize fires.
list(run.values)
lifecycle = list(run.lifecycle) # type: ignore[attr-defined]
events = [e["event"] for e in lifecycle]
assert "started" in events
assert "completed" in events
def test_real_run_error_emits_failed(self) -> None:
def boom(state: SimpleState) -> dict:
raise RuntimeError("kaboom")
inner_builder = StateGraph(SimpleState)
inner_builder.add_node("inner", boom)
inner_builder.add_edge(START, "inner")
inner_builder.add_edge("inner", END)
inner = inner_builder.compile()
outer_builder = StateGraph(SimpleState)
outer_builder.add_node("sub", inner)
outer_builder.add_edge(START, "sub")
outer_builder.add_edge("sub", END)
graph = outer_builder.compile()
run = graph.stream_v2(
{"value": "", "items": []}, transformers=[LifecycleTransformer]
)
with pytest.raises(RuntimeError):
list(run.values)
lifecycle = list(run.lifecycle) # type: ignore[attr-defined]
events = [e["event"] for e in lifecycle]
assert "failed" in events
def test_lifecycle_and_subgraphs_agree(self) -> None:
"""SubgraphTransformer and LifecycleTransformer share the discovery predicate."""
graph = _build_nested_graph()
run = graph.stream_v2(
{"value": "", "items": []}, transformers=[LifecycleTransformer]
)
subs = list(run.subgraphs)
lifecycle = list(run.lifecycle) # type: ignore[attr-defined]
sub_paths = {tuple(s.path) for s in subs}
started_paths = {
tuple(e["namespace"]) for e in lifecycle if e["event"] == "started"
}
assert sub_paths == started_paths
@@ -1,907 +0,0 @@
"""Tests for the MessagesTransformer content-block upgrade (B2).
Verifies that `MessagesTransformer` routes protocol events (emitted by
`stream_v2` via `on_stream_event`) to `ChatModelStream` objects keyed by
run_id, and replays whole `AIMessage` payloads via `message_to_events`.
Legacy v1 `AIMessageChunk` tuples (from `on_llm_new_token`) are ignored.
"""
from __future__ import annotations
import time
from typing import Any
import pytest
from langchain_core.language_models import GenericFakeChatModel
from langchain_core.language_models.chat_model_stream import (
AsyncChatModelStream,
ChatModelStream,
)
from langchain_core.messages import AIMessage, AIMessageChunk
from langgraph.constants import END, START
from langgraph.graph import MessagesState, StateGraph
from langgraph.stream._event_log import EventLog
from langgraph.stream._mux import StreamMux
from langgraph.stream.run_stream import GraphRunStream
from langgraph.stream.transformers import MessagesTransformer, ValuesTransformer
TS = int(time.time() * 1000)
# ---------------------------------------------------------------------------
# Helpers
# ---------------------------------------------------------------------------
def _proto_event(
event: dict[str, Any],
*,
run_id: str = "run-1",
node: str = "llm",
) -> dict[str, Any]:
"""Build a messages ProtocolEvent carrying a protocol event dict (v2 path)."""
metadata: dict[str, Any] = {"langgraph_node": node, "run_id": run_id}
return {
"type": "event",
"method": "messages",
"params": {
"namespace": [],
"timestamp": TS,
"data": (event, metadata),
},
}
def _v1_chunk(
text: str,
msg_id: str = "msg-1",
*,
finish: bool = False,
node: str = "llm",
) -> dict[str, Any]:
"""Build a messages ProtocolEvent carrying a v1 AIMessageChunk tuple."""
rm: dict[str, Any] = {}
if finish:
rm["finish_reason"] = "stop"
message = AIMessageChunk(content=text, id=msg_id, response_metadata=rm)
metadata: dict[str, Any] = {"langgraph_node": node}
return {
"type": "event",
"method": "messages",
"params": {
"namespace": [],
"timestamp": TS,
"data": (message, metadata),
},
}
def _whole_msg(
text: str,
msg_id: str = "msg-10",
*,
node: str = "node",
) -> dict[str, Any]:
"""Build a messages ProtocolEvent carrying a completed AIMessage."""
message = AIMessage(content=text, id=msg_id)
metadata: dict[str, Any] = {"langgraph_node": node}
return {
"type": "event",
"method": "messages",
"params": {
"namespace": [],
"timestamp": TS,
"data": (message, metadata),
},
}
def _make_sync_transformer() -> tuple[MessagesTransformer, EventLog[ChatModelStream]]:
t = MessagesTransformer()
proj = t.init()
log: EventLog[ChatModelStream] = proj["messages"]
log._bind(is_async=False)
# Production subscribes via `iter(log)` from the graph consumer — do that
# up front so `push` during `process` isn't a no-op. Tests read buffered
# items via `log._items` directly rather than re-iterating.
log._subscribed = True
t._bind_pump(lambda: False)
return t, log
def _make_async_transformer() -> tuple[MessagesTransformer, EventLog[ChatModelStream]]:
t = MessagesTransformer()
proj = t.init()
log: EventLog[ChatModelStream] = proj["messages"]
log._bind(is_async=True)
log._subscribed = True
return t, log
# Standard lifecycle events for one streaming LLM call.
def _lifecycle(
*,
text: str = "hello world",
message_id: str = "run-1",
) -> list[dict[str, Any]]:
"""Produce a valid protocol event lifecycle: start, delta, finish, end."""
# Split text into two deltas to exercise delta accumulation.
half = len(text) // 2
first, second = text[:half], text[half:]
return [
{"event": "message-start", "role": "ai", "message_id": message_id},
{
"event": "content-block-start",
"index": 0,
"content_block": {"type": "text", "text": ""},
},
{
"event": "content-block-delta",
"index": 0,
"content_block": {"type": "text", "text": first},
},
{
"event": "content-block-delta",
"index": 0,
"content_block": {"type": "text", "text": second},
},
{
"event": "content-block-finish",
"index": 0,
"content_block": {"type": "text", "text": text},
},
{"event": "message-finish", "reason": "stop"},
]
# ---------------------------------------------------------------------------
# Primary path: protocol event routing
# ---------------------------------------------------------------------------
class TestProtocolEventRouting:
def test_message_start_creates_stream(self) -> None:
t, log = _make_sync_transformer()
t.process(
_proto_event(
{"event": "message-start", "role": "ai", "message_id": "run-1"},
run_id="run-1",
)
)
# Stream is in the log immediately.
log.close()
streams = list(log._items)
assert len(streams) == 1
assert isinstance(streams[0], ChatModelStream)
assert streams[0].message_id == "run-1"
def test_full_lifecycle_yields_done_stream(self) -> None:
t, log = _make_sync_transformer()
for evt in _lifecycle(text="hello world"):
t.process(_proto_event(evt, run_id="run-1"))
log.close()
(stream,) = list(log._items)
assert stream.done
assert stream.output.text == "hello world"
def test_message_finish_cleans_up_routing(self) -> None:
t, log = _make_sync_transformer()
for evt in _lifecycle():
t.process(_proto_event(evt, run_id="run-1"))
assert t._by_run == {}
def test_events_without_prior_start_are_ignored(self) -> None:
"""Orphan delta events (no preceding message-start) are dropped silently."""
t, log = _make_sync_transformer()
t.process(
_proto_event(
{
"event": "content-block-delta",
"index": 0,
"content_block": {"type": "text", "text": "orphan"},
},
run_id="unknown",
)
)
log.close()
assert list(log._items) == []
def test_concurrent_streams_routed_by_run_id(self) -> None:
"""Two interleaved LLM calls each produce their own stream."""
t, log = _make_sync_transformer()
# Interleave events from two different run_ids.
life_a = _lifecycle(text="aaaa", message_id="run-a")
life_b = _lifecycle(text="bbbb", message_id="run-b")
for a, b in zip(life_a, life_b):
t.process(_proto_event(a, run_id="run-a"))
t.process(_proto_event(b, run_id="run-b"))
log.close()
streams = list(log._items)
assert len(streams) == 2
by_id = {s.message_id: s for s in streams}
assert by_id["run-a"].output.text == "aaaa"
assert by_id["run-b"].output.text == "bbbb"
def test_text_deltas_accumulated_on_stream(self) -> None:
t, log = _make_sync_transformer()
for evt in _lifecycle(text="abcdef"):
t.process(_proto_event(evt))
log.close()
(stream,) = list(log._items)
deltas = list(stream._text_proj._deltas)
assert "".join(deltas) == "abcdef"
def test_stream_pushed_on_message_start_not_finish(self) -> None:
"""Consumer can see the stream before it finishes."""
t, log = _make_sync_transformer()
t.process(
_proto_event(
{"event": "message-start", "role": "ai", "message_id": "run-1"},
run_id="run-1",
)
)
# The log has the stream immediately — even though message-finish
# hasn't arrived yet.
assert len(log._items) == 1
def test_node_metadata_set_on_stream(self) -> None:
t, log = _make_sync_transformer()
t.process(
_proto_event(
{"event": "message-start", "role": "ai", "message_id": "run-1"},
run_id="run-1",
node="my_llm",
)
)
(stream,) = [*log._items]
assert stream.node == "my_llm"
# ---------------------------------------------------------------------------
# Non-streaming (whole AIMessage) fallback
# ---------------------------------------------------------------------------
class TestWholeMessageFallback:
def test_whole_ai_message_produces_complete_stream(self) -> None:
t, log = _make_sync_transformer()
t.process(_whole_msg("the full answer"))
log.close()
(stream,) = list(log._items)
assert stream.done
assert stream.output.text == "the full answer"
def test_whole_message_has_full_lifecycle(self) -> None:
t, log = _make_sync_transformer()
t.process(_whole_msg("full"))
log.close()
(stream,) = list(log._items)
event_types = [e["event"] for e in stream._events]
assert event_types == [
"message-start",
"content-block-start",
"content-block-delta",
"content-block-finish",
"message-finish",
]
# ---------------------------------------------------------------------------
# Legacy v1 chunks are ignored (users must migrate to stream_v2)
# ---------------------------------------------------------------------------
class TestLegacyChunksIgnored:
def test_aimessage_chunk_tuple_is_dropped(self) -> None:
t, log = _make_sync_transformer()
t.process(_v1_chunk("hello"))
t.process(_v1_chunk(" world", finish=True))
log.close()
assert list(log._items) == []
# ---------------------------------------------------------------------------
# Filtering behaviors
# ---------------------------------------------------------------------------
class TestFiltering:
def test_non_messages_events_pass_through(self) -> None:
t, _ = _make_sync_transformer()
values_event = {
"type": "event",
"method": "values",
"params": {"namespace": [], "timestamp": TS, "data": {"x": 1}},
}
assert t.process(values_event) is True
def test_subgraph_namespace_dropped(self) -> None:
"""Root MessagesTransformer (via the mux) ignores non-root events."""
from langgraph.stream._mux import StreamMux
mux = StreamMux([MessagesTransformer()], is_async=False)
t = mux.transformer_by_key("messages")
assert isinstance(t, MessagesTransformer)
t._log._subscribed = True
t._bind_pump(lambda: False)
mux.push(
{
"type": "event",
"method": "messages",
"params": {
"namespace": ["subgraph"],
"timestamp": TS,
"data": (
{"event": "message-start", "message_id": "run-x"},
{"run_id": "run-x"},
),
},
}
)
t._log.close()
assert list(t._log._items) == []
# ---------------------------------------------------------------------------
# Lifecycle: finalize / fail
# ---------------------------------------------------------------------------
class TestLifecycle:
def test_fail_propagates_to_open_streams(self) -> None:
t, log = _make_sync_transformer()
t.process(
_proto_event(
{"event": "message-start", "message_id": "run-1"},
run_id="run-1",
)
)
streams = list(log._items)
err = RuntimeError("graph died")
t.fail(err)
assert t._by_run == {}
assert streams[0]._error is err
def test_finalize_clears_routing_state(self) -> None:
t, _ = _make_sync_transformer()
t.process(
_proto_event(
{"event": "message-start", "message_id": "run-1"},
run_id="run-1",
)
)
assert "run-1" in t._by_run
t.finalize()
assert t._by_run == {}
# ---------------------------------------------------------------------------
# Async mode (AsyncChatModelStream)
# ---------------------------------------------------------------------------
class TestAsyncMode:
def test_async_mode_creates_async_stream(self) -> None:
t, log = _make_async_transformer()
for evt in _lifecycle(text="async stream"):
t.process(_proto_event(evt))
streams = list(log._items)
assert len(streams) == 1
assert isinstance(streams[0], AsyncChatModelStream)
@pytest.mark.anyio
async def test_async_text_projection_yields_deltas(self) -> None:
t, log = _make_async_transformer()
for evt in _lifecycle(text="hello world"):
t.process(_proto_event(evt))
(stream,) = list(log._items)
assert isinstance(stream, AsyncChatModelStream)
collected = []
async for delta in stream.text:
collected.append(delta)
assert "".join(collected) == "hello world"
@pytest.mark.anyio
async def test_async_output_awaitable(self) -> None:
t, log = _make_async_transformer()
for evt in _lifecycle(text="async"):
t.process(_proto_event(evt))
(stream,) = list(log._items)
msg = await stream.output
assert msg.text == "async"
# ---------------------------------------------------------------------------
# GraphRunStream integration
# ---------------------------------------------------------------------------
class TestWireRequestMore:
def test_bind_pump_called_on_wire(self) -> None:
values_t = ValuesTransformer()
messages_t = MessagesTransformer()
mux = StreamMux([values_t, messages_t], is_async=False)
assert messages_t._pump_fn is None
run = GraphRunStream(iter([]), mux)
# After wire, the transformer's pump callback is set.
assert messages_t._pump_fn is not None
# And calling it invokes GraphRunStream._pump_next (drains an empty
# graph_iter, returns False).
assert messages_t._pump_fn() is False
assert run._exhausted
def test_created_streams_have_request_more(self) -> None:
values_t = ValuesTransformer()
messages_t = MessagesTransformer()
mux = StreamMux([values_t, messages_t], is_async=False)
GraphRunStream(iter([]), mux)
log: EventLog[ChatModelStream] = mux.extensions["messages"]
log._subscribed = True
for evt in _lifecycle():
messages_t.process(_proto_event(evt))
(stream,) = list(log._items)
# Pump was threaded through: the stream's _request_more points at
# the same callable the transformer was bound with.
assert stream._request_more is messages_t._pump_fn
# ---------------------------------------------------------------------------
# End-to-end via StreamMux
# ---------------------------------------------------------------------------
class TestViaMux:
def test_streaming_via_mux(self) -> None:
t = MessagesTransformer()
v = ValuesTransformer()
mux = StreamMux([v, t], is_async=False)
t._bind_pump(lambda: False)
log: EventLog[ChatModelStream] = mux.extensions["messages"]
# Simulate a consumer subscribing (as `run.messages` iteration would).
log._subscribed = True
for evt in _lifecycle(text="mux stream"):
mux.push(_proto_event(evt))
mux.close()
(stream,) = list(log._items)
assert stream.output.text == "mux stream"
def test_whole_message_via_mux(self) -> None:
t = MessagesTransformer()
v = ValuesTransformer()
mux = StreamMux([v, t], is_async=False)
t._bind_pump(lambda: False)
log: EventLog[ChatModelStream] = mux.extensions["messages"]
log._subscribed = True
mux.push(_whole_msg("result"))
mux.close()
(stream,) = list(log._items)
assert stream.output.text == "result"
@pytest.mark.anyio
async def test_async_streaming_via_mux(self) -> None:
t = MessagesTransformer()
v = ValuesTransformer()
mux = StreamMux([v, t], is_async=True)
log: EventLog[ChatModelStream] = mux.extensions["messages"]
log._subscribed = True
for evt in _lifecycle(text="async mux"):
await mux.apush(_proto_event(evt))
streams = list(log._items)
assert len(streams) == 1
msg = await streams[0].output
assert msg.text == "async mux"
await mux.aclose()
# ---------------------------------------------------------------------------
# End-to-end: full graph → stream_v2 → run.messages
# ---------------------------------------------------------------------------
class TestEndToEnd:
"""Prove the full pipeline works when a node calls `model.stream_v2()`.
These tests exercise the path that the new messages projection is
designed for: a user node invokes `stream_v2` on a chat model,
`on_stream_event` fires on `StreamMessagesHandler`, the handler
forwards to the mux, and the transformer routes events into a
`ChatModelStream` exposed on `run.messages`.
Nothing in Pregel calls `stream_v2` automatically yet; the planned
`graph.stream_v2()` API (B4) and the `create_react_agent`
integration (C2) will wire that up. Until then, populating the
messages projection is opt-in at the node level.
"""
def test_node_calling_stream_v2_populates_messages(self) -> None:
model = GenericFakeChatModel(messages=iter(["hello world"]))
def call_model(state: MessagesState) -> dict[str, Any]:
stream = model.stream_v2(state["messages"])
return {"messages": stream.output}
graph = (
StateGraph(MessagesState)
.add_node("call_model", call_model)
.add_edge(START, "call_model")
.add_edge("call_model", END)
.compile()
)
run = graph.stream_v2({"messages": "hi"})
streams = list(run.messages)
assert len(streams) == 1
assert isinstance(streams[0], ChatModelStream)
assert streams[0].output.text == "hello world"
def test_node_stream_v2_text_deltas_iterate(self) -> None:
"""Consumer can iterate `.text` on the streamed message in real time."""
model = GenericFakeChatModel(messages=iter(["streamed answer"]))
def call_model(state: MessagesState) -> dict[str, Any]:
stream = model.stream_v2(state["messages"])
return {"messages": stream.output}
graph = (
StateGraph(MessagesState)
.add_node("call_model", call_model)
.add_edge(START, "call_model")
.add_edge("call_model", END)
.compile()
)
run = graph.stream_v2({"messages": "go"})
# Pull the stream handle out, then iterate its text deltas.
(stream,) = list(run.messages)
text = "".join(stream.text)
assert text == "streamed answer"
def test_non_llm_message_returned_from_node(self) -> None:
"""Node returns a finalized AIMessage directly — whole-message fallback."""
def return_message(state: MessagesState) -> dict[str, Any]:
return {"messages": AIMessage(content="hardcoded", id="msg-abc")}
graph = (
StateGraph(MessagesState)
.add_node("return_message", return_message)
.add_edge(START, "return_message")
.add_edge("return_message", END)
.compile()
)
run = graph.stream_v2({"messages": "hi"})
streams = list(run.messages)
assert len(streams) == 1
assert streams[0].output.text == "hardcoded"
@pytest.mark.anyio
async def test_async_node_calling_astream_v2(self) -> None:
model = GenericFakeChatModel(messages=iter(["async answer"]))
async def call_model(state: MessagesState) -> dict[str, Any]:
stream = await model.astream_v2(state["messages"])
msg = await stream
return {"messages": msg}
graph = (
StateGraph(MessagesState)
.add_node("call_model", call_model)
.add_edge(START, "call_model")
.add_edge("call_model", END)
.compile()
)
run = await graph.astream_v2({"messages": "hi"})
streams = []
async for stream in run.messages:
streams.append(stream)
assert len(streams) == 1
assert isinstance(streams[0], AsyncChatModelStream)
msg = await streams[0].output
assert msg.text == "async answer"
@pytest.mark.anyio
async def test_nested_async_iteration_yields_text_deltas(self) -> None:
"""Iterate `stream.text` inside `async for stream in run.messages`.
The inner `stream.text` cursor drives the shared graph pump via
`AsyncProjection._arequest_more`, wired by
`MessagesTransformer._bind_apump` and
`AsyncGraphRunStream._wire_arequest_more`.
"""
import asyncio
model = GenericFakeChatModel(messages=iter(["hello world"]))
async def call_model(state: MessagesState) -> dict[str, Any]:
stream = await model.astream_v2(state["messages"])
msg = await stream
return {"messages": msg}
graph = (
StateGraph(MessagesState)
.add_node("call_model", call_model)
.add_edge(START, "call_model")
.add_edge("call_model", END)
.compile()
)
run = await graph.astream_v2({"messages": "hi"})
async def consume_nested() -> list[str]:
collected: list[str] = []
async for stream in run.messages:
async for delta in stream.text:
collected.append(delta)
return collected
deltas = await asyncio.wait_for(consume_nested(), timeout=2.0)
assert "".join(deltas) == "hello world"
class TestEndToEndV2Invoke:
"""Nodes call `model.invoke()`; `stream_v2` routes through v2.
Exercises the auto-routing path added in
`feat(core): route invoke through v2 event path for
_V2StreamingCallbackHandler`: `stream_v2` injects
`CONFIG_KEY_STREAM_MESSAGES_V2` into the config, pregel attaches
`StreamMessagesHandlerV2`, `BaseChatModel._should_stream_v2` sees the
v2 marker and drives the protocol event generator, and
`on_stream_event` forwards each event onto the messages channel.
"""
def test_invoke_with_v2_marker_populates_messages(self) -> None:
"""Node calling `model.invoke()` produces one ChatModelStream with v2 events."""
model = GenericFakeChatModel(messages=iter(["hello world"]))
def call_model(state: MessagesState) -> dict[str, Any]:
return {"messages": model.invoke(state["messages"])}
graph = (
StateGraph(MessagesState)
.add_node("call_model", call_model)
.add_edge(START, "call_model")
.add_edge("call_model", END)
.compile()
)
run = graph.stream_v2({"messages": "hi"})
streams = list(run.messages)
assert len(streams) == 1, (
"Expected exactly one ChatModelStream — the streamed invoke and "
"the node's return of the same AIMessage must dedupe."
)
stream = streams[0]
assert isinstance(stream, ChatModelStream)
assert stream.output.text == "hello world"
def test_invoke_v2_emits_protocol_events(self) -> None:
"""Iterating the stream yields the full v2 lifecycle (not v1 chunks)."""
model = GenericFakeChatModel(messages=iter(["streamed answer"]))
def call_model(state: MessagesState) -> dict[str, Any]:
return {"messages": model.invoke(state["messages"])}
graph = (
StateGraph(MessagesState)
.add_node("call_model", call_model)
.add_edge(START, "call_model")
.add_edge("call_model", END)
.compile()
)
run = graph.stream_v2({"messages": "go"})
(stream,) = list(run.messages)
events = list(stream)
event_types = [e.get("event") for e in events]
assert "message-start" in event_types
assert "content-block-start" in event_types
assert "content-block-delta" in event_types
assert "content-block-finish" in event_types
assert "message-finish" in event_types
# Sanity: every event is a dict carrying an "event" key — not an
# AIMessageChunk tuple from the v1 path.
for event in events:
assert isinstance(event, dict)
assert "event" in event
# Typed projection still assembles the final text.
assert stream.output.text == "streamed answer"
def test_invoke_text_deltas_iterate_live(self) -> None:
"""`.text` projection yields deltas in order."""
model = GenericFakeChatModel(messages=iter(["delta streaming works"]))
def call_model(state: MessagesState) -> dict[str, Any]:
return {"messages": model.invoke(state["messages"])}
graph = (
StateGraph(MessagesState)
.add_node("call_model", call_model)
.add_edge(START, "call_model")
.add_edge("call_model", END)
.compile()
)
run = graph.stream_v2({"messages": "hi"})
(stream,) = list(run.messages)
assembled = "".join(stream.text)
assert assembled == "delta streaming works"
def test_invoke_dedupe_survives_multi_node_graph(self) -> None:
"""Two model-invoking nodes produce exactly two streams, each once."""
model_a = GenericFakeChatModel(messages=iter(["alpha"]))
model_b = GenericFakeChatModel(messages=iter(["beta"]))
def node_a(state: MessagesState) -> dict[str, Any]:
return {"messages": model_a.invoke(state["messages"])}
def node_b(state: MessagesState) -> dict[str, Any]:
return {"messages": model_b.invoke(state["messages"])}
graph = (
StateGraph(MessagesState)
.add_node("node_a", node_a)
.add_node("node_b", node_b)
.add_edge(START, "node_a")
.add_edge("node_a", "node_b")
.add_edge("node_b", END)
.compile()
)
run = graph.stream_v2({"messages": "hi"})
streams = list(run.messages)
assert len(streams) == 2
contents = {s.output.text for s in streams}
assert contents == {"alpha", "beta"}
def test_invoke_plus_constructed_message_two_streams(self) -> None:
"""A v2-streamed node + a node that returns a constructed AIMessage
produces two ChatModelStreams one from the live event lifecycle,
one synthesized from the constructed message via `message_to_events`.
"""
model = GenericFakeChatModel(messages=iter(["live stream"]))
def streaming_node(state: MessagesState) -> dict[str, Any]:
return {"messages": model.invoke(state["messages"])}
def constructed_node(state: MessagesState) -> dict[str, Any]:
return {"messages": [AIMessage(content="hardcoded", id="constructed-1")]}
graph = (
StateGraph(MessagesState)
.add_node("streaming_node", streaming_node)
.add_node("constructed_node", constructed_node)
.add_edge(START, "streaming_node")
.add_edge("streaming_node", "constructed_node")
.add_edge("constructed_node", END)
.compile()
)
run = graph.stream_v2({"messages": "hi"})
streams = list(run.messages)
assert len(streams) == 2
assert streams[0].node == "streaming_node"
assert streams[0].output.text == "live stream"
assert streams[1].node == "constructed_node"
assert streams[1].output.text == "hardcoded"
assert streams[1].message_id == "constructed-1"
@pytest.mark.anyio
async def test_ainvoke_with_v2_marker_populates_messages(self) -> None:
"""Async mirror: `model.ainvoke()` + `astream_v2`."""
model = GenericFakeChatModel(messages=iter(["async invoke"]))
async def call_model(state: MessagesState) -> dict[str, Any]:
return {"messages": await model.ainvoke(state["messages"])}
graph = (
StateGraph(MessagesState)
.add_node("call_model", call_model)
.add_edge(START, "call_model")
.add_edge("call_model", END)
.compile()
)
run = await graph.astream_v2({"messages": "hi"})
streams = []
async for stream in run.messages:
streams.append(stream)
assert len(streams) == 1
assert isinstance(streams[0], AsyncChatModelStream)
msg = await streams[0].output
assert msg.text == "async invoke"
class TestDirectMessagesModeStaysV1:
"""Regression guard: direct `graph.stream(stream_mode="messages")`
(no `stream_v2`) must keep the v1 `(AIMessageChunk, metadata)`
tuple shape. The v2 flag is only injected by `stream_v2` / `astream_v2`.
"""
def test_direct_graph_stream_messages_yields_ai_message_chunks(self) -> None:
model = GenericFakeChatModel(messages=iter(["legacy path"]))
def call_model(state: MessagesState) -> dict[str, Any]:
return {"messages": model.invoke(state["messages"])}
graph = (
StateGraph(MessagesState)
.add_node("call_model", call_model)
.add_edge(START, "call_model")
.add_edge("call_model", END)
.compile()
)
parts = list(graph.stream({"messages": "hi"}, stream_mode="messages"))
# Should have at least one streamed chunk; each part is
# (AIMessageChunk, metadata) — not a v2 event dict.
assert parts, "expected stream_mode='messages' to emit tuples"
for part in parts:
payload, _metadata = part
assert isinstance(payload, AIMessageChunk), (
"direct graph.stream(stream_mode='messages') leaked v2 "
"event dicts — stream_v2 flag bled through."
)
assembled = "".join(
p[0].content for p in parts if isinstance(p[0].content, str)
)
assert assembled == "legacy path"
class TestStreamMessagesHandlerV2Unit:
"""Unit tests on the handler class itself."""
def test_on_llm_new_token_is_noop(self) -> None:
"""v2 handler must not emit v1 chunks even if `on_llm_new_token` fires
(e.g. from a node calling `model.stream()` directly on a v2-flagged run).
"""
from uuid import uuid4
from langchain_core.outputs import ChatGenerationChunk
from langgraph.pregel._messages import StreamMessagesHandlerV2
emitted: list[Any] = []
handler = StreamMessagesHandlerV2(emitted.append, subgraphs=False)
run_id = uuid4()
# Register a fake run so `self.metadata.get(run_id)` would succeed for
# other callbacks — this makes sure the no-op is unconditional, not a
# side effect of missing metadata.
handler.metadata[run_id] = ((), {"langgraph_node": "x"})
handler.on_llm_new_token(
"hello",
chunk=ChatGenerationChunk(message=AIMessageChunk(content="hello")),
run_id=run_id,
)
assert emitted == [], (
"StreamMessagesHandlerV2.on_llm_new_token must not push to the "
"messages stream — it's the v2 marker's guarantee."
)
@@ -1,411 +0,0 @@
"""Tests for SubgraphTransformer namespace-based discovery."""
from __future__ import annotations
import operator
import time
from typing import Annotated, Any
import pytest
from langgraph.checkpoint.memory import InMemorySaver
from typing_extensions import TypedDict
from langgraph.constants import END, START
from langgraph.errors import GraphInterrupt
from langgraph.graph import StateGraph
from langgraph.stream._event_log import EventLog
from langgraph.stream._mux import StreamMux
from langgraph.stream._types import ProtocolEvent
from langgraph.stream.transformers import (
MessagesTransformer,
SubgraphRunStream,
SubgraphTransformer,
ValuesTransformer,
)
from langgraph.types import interrupt
TS = int(time.time() * 1000)
def _values(payload: dict[str, Any], *, namespace: list[str]) -> ProtocolEvent:
return {
"type": "event",
"method": "values",
"params": {
"namespace": namespace,
"timestamp": TS,
"data": payload,
},
}
def _subscribe(log: EventLog) -> None:
"""Flip `_subscribed = True` so pushes retain items for test inspection."""
log._subscribed = True
# ---------------------------------------------------------------------------
# Unit tests: feed events directly into the transformer
# ---------------------------------------------------------------------------
_FACTORIES = [ValuesTransformer, MessagesTransformer, SubgraphTransformer]
def _handle_values_items(handle: SubgraphRunStream) -> list:
return list(handle._mux.extensions["values"]._items) # type: ignore[attr-defined]
def _handle_subgraphs_items(handle: SubgraphRunStream) -> list:
return list(handle._mux.extensions["subgraphs"]._items) # type: ignore[attr-defined]
def _pre_subscribe_handle(handle: SubgraphRunStream) -> None:
"""Flip `_subscribed` on every EventLog inside the handle's mini-mux.
The mini-mux is built via `make_child` with the full factory list,
so values / messages / subgraphs logs all exist as projections.
Tests that feed events directly need them subscribed so pushes
retain items in the deque for `_items` inspection.
"""
for value in handle._mux.extensions.values():
if isinstance(value, EventLog):
_subscribe(value)
class TestSubgraphTransformerUnit:
def _mux(self) -> tuple[StreamMux, SubgraphTransformer]:
mux = StreamMux(factories=_FACTORIES, is_async=False)
transformer = mux.transformer_by_key("subgraphs")
assert isinstance(transformer, SubgraphTransformer)
_subscribe(transformer._root_log)
return mux, transformer
def _handle(self, transformer: SubgraphTransformer) -> SubgraphRunStream:
(handle,) = list(transformer._root_log._items)
return handle
def test_root_event_does_not_create_handle(self) -> None:
mux, transformer = self._mux()
mux.push(_values({"value": 1}, namespace=[]))
assert list(transformer._root_log._items) == []
assert transformer._by_ns == {}
def test_first_event_at_child_depth_yields_handle(self) -> None:
mux, transformer = self._mux()
mux.push(_values({"value": 1}, namespace=["child:task_a"]))
handle = self._handle(transformer)
assert handle.path == ("child:task_a",)
assert handle.graph_name == "child"
assert handle.trigger_call_id == "task_a"
assert handle.status == "started"
def test_handle_without_task_id_suffix(self) -> None:
mux, transformer = self._mux()
mux.push(_values({"value": 1}, namespace=["child"]))
handle = self._handle(transformer)
assert handle.path == ("child",)
assert handle.graph_name == "child"
assert handle.trigger_call_id is None
def test_discovery_is_method_agnostic(self) -> None:
mux, transformer = self._mux()
# Method-agnostic means "any event method triggers discovery".
# Use `updates` — neither ValuesTransformer nor
# MessagesTransformer care about it, so the test only exercises
# SubgraphTransformer's discovery path.
mux.push(
{
"type": "event",
"method": "updates",
"params": {"namespace": ["c:t"], "timestamp": TS, "data": "x"},
}
)
handle = self._handle(transformer)
assert handle.path == ("c:t",)
def test_grandchild_surfaces_under_child(self) -> None:
mux, transformer = self._mux()
mux.push(_values({"value": 1}, namespace=["child:t"]))
child = self._handle(transformer)
_pre_subscribe_handle(child)
mux.push(_values({"value": 2}, namespace=["child:t", "grand:u"]))
(grand,) = _handle_subgraphs_items(child)
assert grand.path == ("child:t", "grand:u")
assert grand.graph_name == "grand"
assert grand.trigger_call_id == "u"
def test_values_routed_into_handle(self) -> None:
mux, transformer = self._mux()
mux.push(_values({"value": 1}, namespace=["c:t"]))
handle = self._handle(transformer)
_pre_subscribe_handle(handle)
mux.push(_values({"value": 2}, namespace=["c:t"]))
# Both the discovery event and subsequent values land in the child.
assert _handle_values_items(handle) == [{"value": 1}, {"value": 2}]
assert handle.output == {"value": 2}
def test_root_values_not_routed(self) -> None:
mux, transformer = self._mux()
mux.push(_values({"value": 1}, namespace=["c:t"]))
handle = self._handle(transformer)
_pre_subscribe_handle(handle)
# Values event at root namespace — must not leak into child handle.
mux.push(_values({"value": "root"}, namespace=[]))
assert _handle_values_items(handle) == [{"value": 1}]
def test_finalize_closes_dangling(self) -> None:
mux, transformer = self._mux()
mux.push(_values({"value": 1}, namespace=["c:t"]))
handle = self._handle(transformer)
mux.close()
assert handle.status == "completed"
assert handle._mux.extensions["values"]._closed
assert handle._mux.extensions["subgraphs"]._closed
def test_fail_with_graph_interrupt_marks_interrupted(self) -> None:
mux, transformer = self._mux()
mux.push(_values({"value": 1}, namespace=["c:t"]))
handle = self._handle(transformer)
mux.fail(GraphInterrupt())
assert handle.status == "interrupted"
def test_fail_with_generic_error_marks_failed(self) -> None:
mux, transformer = self._mux()
mux.push(_values({"value": 1}, namespace=["c:t"]))
handle = self._handle(transformer)
mux.fail(RuntimeError("explode"))
assert handle.status == "failed"
assert handle.error == "explode"
def test_repeated_events_same_ns_single_handle(self) -> None:
mux, transformer = self._mux()
mux.push(_values({"value": 1}, namespace=["c:t"]))
mux.push(_values({"value": 2}, namespace=["c:t"]))
mux.push(_values({"value": 3}, namespace=["c:t"]))
handles = list(transformer._root_log._items)
assert len(handles) == 1
assert handles[0].path == ("c:t",)
# ---------------------------------------------------------------------------
# End-to-end tests via stream_v2 on real graphs
# ---------------------------------------------------------------------------
class SimpleState(TypedDict):
value: str
items: Annotated[list[str], operator.add]
def _build_nested_graph():
"""Parent graph with a compiled subgraph node."""
def inner_node(state: SimpleState) -> dict:
return {"value": state["value"] + "X", "items": ["x"]}
inner_builder = StateGraph(SimpleState)
inner_builder.add_node("inner_node", inner_node)
inner_builder.add_edge(START, "inner_node")
inner_builder.add_edge("inner_node", END)
inner = inner_builder.compile()
def outer_node(state: SimpleState) -> dict:
return {"value": state["value"] + "Y", "items": ["y"]}
outer_builder = StateGraph(SimpleState)
outer_builder.add_node("outer_node", outer_node)
outer_builder.add_node("sub", inner)
outer_builder.add_edge(START, "outer_node")
outer_builder.add_edge("outer_node", "sub")
outer_builder.add_edge("sub", END)
return outer_builder.compile()
class TestSubgraphTransformerEndToEnd:
def test_flat_graph_yields_no_subgraphs(self) -> None:
builder = StateGraph(SimpleState)
builder.add_node("n", lambda s: {"value": s["value"] + "!", "items": ["!"]})
builder.add_edge(START, "n")
builder.add_edge("n", END)
graph = builder.compile()
run = graph.stream_v2({"value": "", "items": []})
collected: list[SubgraphRunStream] = []
for sub in run.subgraphs:
collected.append(sub)
assert collected == []
# Output still resolves.
assert run.output is not None
def test_nested_graph_yields_one_child(self) -> None:
graph = _build_nested_graph()
run = graph.stream_v2({"value": "", "items": []})
collected: list[SubgraphRunStream] = []
for sub in run.subgraphs:
collected.append(sub)
assert len(collected) == 1
child = collected[0]
assert len(child.path) == 1
assert child.path[0].startswith("sub:")
assert child.status == "completed"
def test_error_in_subgraph_fails_child(self) -> None:
def boom(state: SimpleState) -> dict:
raise RuntimeError("subgraph_failed")
inner_builder = StateGraph(SimpleState)
inner_builder.add_node("inner", boom)
inner_builder.add_edge(START, "inner")
inner_builder.add_edge("inner", END)
inner = inner_builder.compile()
outer_builder = StateGraph(SimpleState)
outer_builder.add_node("sub", inner)
outer_builder.add_edge(START, "sub")
outer_builder.add_edge("sub", END)
graph = outer_builder.compile()
run = graph.stream_v2({"value": "", "items": []})
collected: list[SubgraphRunStream] = []
with pytest.raises(RuntimeError):
for sub in run.subgraphs:
collected.append(sub)
assert len(collected) == 1
assert collected[0].status == "failed"
class TestSubgraphTransformerAsyncEndToEnd:
@pytest.mark.anyio
async def test_nested_graph_yields_one_child(self) -> None:
async def inner(state: SimpleState) -> dict:
return {"value": state["value"] + "X", "items": ["x"]}
inner_builder = StateGraph(SimpleState)
inner_builder.add_node("inner", inner)
inner_builder.add_edge(START, "inner")
inner_builder.add_edge("inner", END)
inner_graph = inner_builder.compile()
outer_builder = StateGraph(SimpleState)
outer_builder.add_node("sub", inner_graph)
outer_builder.add_edge(START, "sub")
outer_builder.add_edge("sub", END)
graph = outer_builder.compile()
run = await graph.astream_v2({"value": "", "items": []})
collected: list[SubgraphRunStream] = []
async for sub in run.subgraphs:
collected.append(sub)
assert len(collected) == 1
child = collected[0]
assert child.status == "completed"
class TestSubgraphTriggerCallId:
"""Confirm `trigger_call_id` flows from real pregel metadata."""
def test_trigger_call_id_populated_end_to_end(self) -> None:
graph = _build_nested_graph()
run = graph.stream_v2({"value": "", "items": []})
collected: list[SubgraphRunStream] = list(run.subgraphs)
assert len(collected) == 1
child = collected[0]
# The child's single-segment path encodes `node_name:task_id`.
# Both the parsed task_id (`trigger_call_id`) and the segment
# should match the same task_id suffix.
assert ":" in child.path[0]
node_name, _, task_id = child.path[0].partition(":")
assert node_name == "sub"
assert task_id # non-empty
assert child.trigger_call_id == task_id
class TestSubgraphInterrupt:
"""Interrupts raised inside a subgraph surface as status=interrupted."""
def _build_interrupt_subgraph(self):
def inner_node(state: SimpleState) -> dict:
interrupt("need approval")
return {"value": state["value"] + "X", "items": ["x"]}
inner_builder = StateGraph(SimpleState)
inner_builder.add_node("inner_node", inner_node)
inner_builder.add_edge(START, "inner_node")
inner_builder.add_edge("inner_node", END)
inner = inner_builder.compile()
outer_builder = StateGraph(SimpleState)
outer_builder.add_node("sub", inner)
outer_builder.add_edge(START, "sub")
outer_builder.add_edge("sub", END)
return outer_builder.compile(checkpointer=InMemorySaver())
def test_interrupt_in_subgraph_marks_handle_interrupted(self) -> None:
graph = self._build_interrupt_subgraph()
run = graph.stream_v2(
{"value": "", "items": []},
config={"configurable": {"thread_id": "t1"}},
)
collected: list[SubgraphRunStream] = list(run.subgraphs)
assert run.interrupted is True
assert len(collected) == 1
assert collected[0].status == "interrupted"
class TestSubgraphNameCollision:
"""The subgraph's compiled `name` equaling its node name is detected.
Primary detector `name != langgraph_node` fails here; the
parent_run_id fallback in `_is_nested_pregel_start` is what keeps
the subgraph visible.
"""
def test_name_equals_node_name_still_detected(self) -> None:
def inner_node(state: SimpleState) -> dict:
return {"value": state["value"] + "X", "items": ["x"]}
inner_builder = StateGraph(SimpleState)
inner_builder.add_node("inner_node", inner_node)
inner_builder.add_edge(START, "inner_node")
inner_builder.add_edge("inner_node", END)
# Compile with the same name as the node it will be registered as.
inner = inner_builder.compile(name="sub")
outer_builder = StateGraph(SimpleState)
outer_builder.add_node("sub", inner)
outer_builder.add_edge(START, "sub")
outer_builder.add_edge("sub", END)
graph = outer_builder.compile()
run = graph.stream_v2({"value": "", "items": []})
collected: list[SubgraphRunStream] = list(run.subgraphs)
assert len(collected) == 1
child = collected[0]
assert child.graph_name == "sub"
assert child.status == "completed"
+791 -7
View File
@@ -37,6 +37,7 @@ def _checkpoint_summary(history: list) -> list[dict]:
Returns a list of dicts (newest-first, matching get_state_history order) with:
- id: short checkpoint id suffix (last 6 chars)
- parent_id: short parent checkpoint id suffix or None
- source: checkpoint metadata source (input, loop, fork, update)
- next: tuple of next node names
- values: channel values snapshot
"""
@@ -52,6 +53,7 @@ def _checkpoint_summary(history: list) -> list[dict]:
{
"id": cid[-6:],
"parent_id": pid[-6:] if pid else None,
"source": s.metadata.get("source"),
"next": s.next,
"values": s.values,
}
@@ -280,6 +282,116 @@ def test_replay_from_before_interrupt_refires(
assert call_count["node_b"] == 1 # NOT re-executed (after interrupt)
def test_replay_from_before_interrupt_then_resume(
sync_checkpointer: BaseCheckpointSaver,
) -> None:
"""Replay from checkpoint before interrupt node, then resume with a new
answer and verify the graph completes with the new value.
Graph: START --> node_a --> ask_human (interrupt) --> node_b --> END
Original run:
source=input next=(__start__,) values=[]
source=loop next=(node_a,) values=[]
source=loop next=(ask_human,) values=[a] <-- replay from here
source=loop next=(node_b,) values=[a, human:old_answer]
source=loop next=() values=[a, human:old_answer, b]
After replay (fork created) + resume with "new_answer":
source=input next=(__start__,) values=[]
source=loop next=(node_a,) values=[]
source=loop next=(ask_human,) values=[a] <-- branch point
source=loop next=(node_b,) values=[a, human:old_answer]
source=loop next=() values=[a, human:old_answer, b] (old branch)
source=fork next=(ask_human,) values=[a] <-- fork from branch point
source=loop next=(node_b,) values=[a, human:new_answer]
source=loop next=() values=[a, human:new_answer, b] (new branch)
"""
called: list[str] = []
def node_a(state: State) -> State:
called.append("node_a")
return {"value": ["a"]}
def ask_human(state: State) -> State:
called.append("ask_human")
answer = interrupt("What is your input?")
return {"value": [f"human:{answer}"]}
def node_b(state: State) -> State:
called.append("node_b")
return {"value": ["b"]}
graph = (
StateGraph(State)
.add_node("node_a", node_a)
.add_node("ask_human", ask_human)
.add_node("node_b", node_b)
.add_edge(START, "node_a")
.add_edge("node_a", "ask_human")
.add_edge("ask_human", "node_b")
.compile(checkpointer=sync_checkpointer)
)
config = {"configurable": {"thread_id": "1"}}
# --- Original run: invoke until interrupt, then resume to complete ---
graph.invoke({"value": []}, config)
graph.invoke(Command(resume="old_answer"), config)
original_history = list(graph.get_state_history(config))
original = _checkpoint_summary(original_history)
assert [(s["source"], s["next"], s["values"]) for s in original] == [
("loop", (), {"value": ["a", "human:old_answer", "b"]}),
("loop", ("node_b",), {"value": ["a", "human:old_answer"]}),
("loop", ("ask_human",), {"value": ["a"]}),
("loop", ("node_a",), {"value": []}),
("input", ("__start__",), {"value": []}),
]
# --- Replay from checkpoint before ask_human ---
before_ask = next(s for s in original_history if s.next == ("ask_human",))
called.clear()
replay_result = graph.invoke(None, before_ask.config)
assert replay_result["__interrupt__"][0].value == "What is your input?"
assert "ask_human" in called
assert "node_a" not in called # before the replay point, not re-executed
# A fork checkpoint is now the latest — it branches from the replay point
post_replay = _checkpoint_summary(list(graph.get_state_history(config)))
assert [(s["source"], s["next"]) for s in post_replay] == [
("fork", ("ask_human",)), # <-- new fork (latest)
("loop", ()), # original done
("loop", ("node_b",)),
("loop", ("ask_human",)), # branch point
("loop", ("node_a",)),
("input", ("__start__",)),
]
# --- Resume with a new answer ---
called.clear()
final_result = graph.invoke(Command(resume="new_answer"), config)
assert final_result["value"] == ["a", "human:new_answer", "b"]
assert "ask_human" in called
assert "node_b" in called
final = _checkpoint_summary(list(graph.get_state_history(config)))
assert [(s["source"], s["next"], s["values"]) for s in final] == [
# New branch (from fork)
("loop", (), {"value": ["a", "human:new_answer", "b"]}),
("loop", ("node_b",), {"value": ["a", "human:new_answer"]}),
("fork", ("ask_human",), {"value": ["a"]}),
# Original branch (preserved)
("loop", (), {"value": ["a", "human:old_answer", "b"]}),
("loop", ("node_b",), {"value": ["a", "human:old_answer"]}),
("loop", ("ask_human",), {"value": ["a"]}),
("loop", ("node_a",), {"value": []}),
("input", ("__start__",), {"value": []}),
]
def test_replay_interrupt_stable_across_replays(
sync_checkpointer: BaseCheckpointSaver,
) -> None:
@@ -320,8 +432,14 @@ def test_replay_interrupt_stable_across_replays(
r = graph.invoke(None, before_ask.config)
results.append(r)
assert all(r == results[0] for r in results)
assert "__interrupt__" in results[0]
# Each replay creates a fork with a unique interrupt ID, so we compare
# interrupt values and state values rather than full equality.
assert all("__interrupt__" in r for r in results)
assert all(
r["__interrupt__"][0].value == results[0]["__interrupt__"][0].value
for r in results
)
assert all(r["value"] == results[0]["value"] for r in results)
def test_fork_from_before_interrupt_refires(
@@ -854,6 +972,354 @@ def test_subgraph_interrupt_replay_from_interrupt_checkpoint(
assert "step_b" not in called
def test_subgraph_interrupt_replay_from_parent_then_resume(
sync_checkpointer: BaseCheckpointSaver,
) -> None:
"""Replay from the parent checkpoint where a subgraph interrupt fired,
then resume with a new answer. Verifies that a fork is created and the
full graph completes. Checks full checkpoint history at each stage."""
called: list[str] = []
def router(state: State) -> State:
called.append("router")
return {"value": ["routed"]}
def step_a(state: State) -> State:
called.append("step_a")
return {"value": ["sub_a"]}
def ask_human(state: State) -> State:
called.append("ask_human")
answer = interrupt("Provide input:")
return {"value": [f"human:{answer}"]}
def step_b(state: State) -> State:
called.append("step_b")
return {"value": ["sub_b"]}
subgraph = (
StateGraph(State)
.add_node("step_a", step_a)
.add_node("ask_human", ask_human)
.add_node("step_b", step_b)
.add_edge(START, "step_a")
.add_edge("step_a", "ask_human")
.add_edge("ask_human", "step_b")
.compile(checkpointer=True)
)
def post_process(state: State) -> State:
called.append("post_process")
return {"value": ["post"]}
graph = (
StateGraph(State)
.add_node("router", router)
.add_node("subgraph_node", subgraph)
.add_node("post_process", post_process)
.add_edge(START, "router")
.add_edge("router", "subgraph_node")
.add_edge("subgraph_node", "post_process")
.compile(checkpointer=sync_checkpointer)
)
config = {"configurable": {"thread_id": "1"}}
# Run until interrupt, then resume to complete
graph.invoke({"value": []}, config)
graph.invoke(Command(resume="old_answer"), config)
# Original parent history (newest first)
original_history = list(graph.get_state_history(config))
assert [s.next for s in original_history] == [
(), # done
("post_process",),
("subgraph_node",), # subgraph ran, interrupt fired here
("router",),
("__start__",),
]
# Find the parent checkpoint where the interrupt fired
interrupt_checkpoint = next(
s for s in original_history if s.next == ("subgraph_node",)
)
# Replay from parent checkpoint — subgraph re-executes, interrupt re-fires
called.clear()
replay_result = graph.invoke(None, interrupt_checkpoint.config)
assert "__interrupt__" in replay_result
assert replay_result["__interrupt__"][0].value == "Provide input:"
assert "step_a" in called
assert "ask_human" in called
assert "step_b" not in called
# Verify fork checkpoint was created
post_replay_history = list(graph.get_state_history(config))
assert [s.next for s in post_replay_history] == [
("subgraph_node",), # fork (interrupt pending)
(), # original done
("post_process",),
("subgraph_node",),
("router",),
("__start__",),
]
assert [s.metadata["source"] for s in post_replay_history] == [
"fork",
"loop",
"loop",
"loop",
"loop",
"input",
]
fork = post_replay_history[0]
assert (
fork.parent_config["configurable"]["checkpoint_id"]
== interrupt_checkpoint.config["configurable"]["checkpoint_id"]
)
# Resume with a new answer — full graph should complete
called.clear()
final_result = graph.invoke(Command(resume="new_answer"), config)
assert "__interrupt__" not in final_result
assert "human:new_answer" in final_result["value"]
assert "sub_b" in final_result["value"]
assert "post" in final_result["value"]
assert "ask_human" in called
assert "step_b" in called
assert "post_process" in called
# Final checkpoint history
final_history = list(graph.get_state_history(config))
assert [s.next for s in final_history] == [
(), # new branch done
("post_process",), # new branch post_process
("subgraph_node",), # fork
(), # original done
("post_process",),
("subgraph_node",),
("router",),
("__start__",),
]
assert [s.metadata["source"] for s in final_history] == [
"loop",
"loop",
"fork",
"loop",
"loop",
"loop",
"loop",
"input",
]
def test_subgraph_interrupt_resume_with_explicit_head_checkpoint_id(
sync_checkpointer: BaseCheckpointSaver,
) -> None:
"""Resume with Command(resume=...) plus the current head checkpoint_id
in config. The subgraph must continue from the interrupted node, not
restart from scratch. Explicit checkpoint_id triggers is_replaying but
this is a resume, not a time-travel, so ReplayState should not apply."""
called: list[str] = []
def step_a(state: State) -> State:
called.append("step_a")
return {"value": ["sub_a"]}
def ask_human(state: State) -> State:
called.append("ask_human")
answer = interrupt("Provide input:")
return {"value": [f"human:{answer}"]}
def step_b(state: State) -> State:
called.append("step_b")
return {"value": ["sub_b"]}
subgraph = (
StateGraph(State)
.add_node("step_a", step_a)
.add_node("ask_human", ask_human)
.add_node("step_b", step_b)
.add_edge(START, "step_a")
.add_edge("step_a", "ask_human")
.add_edge("ask_human", "step_b")
.compile(checkpointer=True)
)
graph = (
StateGraph(State)
.add_node("subgraph_node", subgraph)
.add_edge(START, "subgraph_node")
.compile(checkpointer=sync_checkpointer)
)
config = {"configurable": {"thread_id": "1"}}
# Run until interrupt fires in subgraph
graph.invoke({"value": []}, config)
assert called == ["step_a", "ask_human"]
# Resume with explicit head checkpoint_id in config
head_checkpoint_id = graph.get_state(config).config["configurable"]["checkpoint_id"]
called.clear()
resume_config = {
"configurable": {
"thread_id": "1",
"checkpoint_id": head_checkpoint_id,
"checkpoint_ns": "",
}
}
result = graph.invoke(Command(resume="answer"), resume_config)
assert called == ["ask_human", "step_b"]
assert "__interrupt__" not in result
assert result["value"] == ["sub_a", "human:answer", "sub_b"]
def test_subgraph_replay_loads_accumulated_state_then_resume(
sync_checkpointer: BaseCheckpointSaver,
) -> None:
"""Two parent invocations, then replay from before the subgraph in the
2nd invocation. The subgraph (checkpointer=True) should load its
accumulated state from the 1st invocation via ReplayState, re-fire
the interrupt, and then resume + complete.
This tests the ReplayState path: the parent is replaying and the
subgraph uses list(before=parent_checkpoint_id) to find its
corresponding checkpoint from the original execution.
"""
class SubState(TypedDict):
value: Annotated[list[str], operator.add]
class ParentState(TypedDict):
results: Annotated[list[str], operator.add]
started_state: list[dict] = []
def step_a(state: SubState) -> SubState:
started_state.append(dict(state))
answer = interrupt("question_a")
return {"value": [f"a:{answer}"]}
subgraph = (
StateGraph(SubState)
.add_node("step_a", step_a)
.add_edge(START, "step_a")
.compile(checkpointer=True)
)
def parent_node(state: ParentState) -> ParentState:
return {"results": ["p"]}
graph = (
StateGraph(ParentState)
.add_node("parent_node", parent_node)
.add_node("sub_node", subgraph)
.add_edge(START, "parent_node")
.add_edge("parent_node", "sub_node")
.compile(checkpointer=sync_checkpointer)
)
config = {"configurable": {"thread_id": "1"}}
# === 1st invocation: complete with answer "a1" ===
graph.invoke({"results": []}, config)
graph.invoke(Command(resume="a1"), config)
# step_a saw empty state (fresh subgraph)
assert started_state[0] == {"value": []}
# === 2nd invocation: complete with answer "a2" ===
started_state.clear()
graph.invoke({"results": []}, config)
graph.invoke(Command(resume="a2"), config)
# Stateful subgraph retained state from 1st invocation
assert started_state[0] == {"value": ["a:a1"]}
# Original history (newest first)
original_history = list(graph.get_state_history(config))
assert [s.next for s in original_history] == [
(), # 2nd done
("sub_node",), # 2nd sub_node
("parent_node",), # 2nd parent_node
("__start__",), # 2nd input
(), # 1st done
("sub_node",), # 1st sub_node
("parent_node",), # 1st parent_node
("__start__",), # 1st input
]
# Replay from before sub_node in 2nd invocation (newest match)
before_sub_2nd = [s for s in original_history if s.next == ("sub_node",)][0]
started_state.clear()
replay = graph.invoke(None, before_sub_2nd.config)
assert "__interrupt__" in replay
# Subgraph should see accumulated state from END of 1st invocation
assert started_state[0] == {"value": ["a:a1"]}
# Verify fork was created
post_replay_history = list(graph.get_state_history(config))
assert [s.next for s in post_replay_history] == [
("sub_node",), # fork (interrupt pending)
(), # 2nd done
("sub_node",), # 2nd sub_node
("parent_node",), # 2nd parent_node
("__start__",), # 2nd input
(), # 1st done
("sub_node",), # 1st sub_node
("parent_node",), # 1st parent_node
("__start__",), # 1st input
]
assert [s.metadata["source"] for s in post_replay_history] == [
"fork",
"loop",
"loop",
"loop",
"input",
"loop",
"loop",
"loop",
"input",
]
# Resume with a new answer
started_state.clear()
final = graph.invoke(Command(resume="a3"), config)
assert "__interrupt__" not in final
assert final["results"] == ["p", "p"]
# Final history
final_history = list(graph.get_state_history(config))
assert [s.next for s in final_history] == [
(), # new branch done
("sub_node",), # fork
(), # 2nd done
("sub_node",), # 2nd sub_node
("parent_node",), # 2nd parent_node
("__start__",), # 2nd input
(), # 1st done
("sub_node",), # 1st sub_node
("parent_node",), # 1st parent_node
("__start__",), # 1st input
]
assert [s.metadata["source"] for s in final_history] == [
"loop",
"fork",
"loop",
"loop",
"loop",
"input",
"loop",
"loop",
"loop",
"input",
]
def test_subgraph_interrupt_full_flow(
sync_checkpointer: BaseCheckpointSaver,
) -> None:
@@ -1290,6 +1756,321 @@ def test_subgraph_time_travel_to_second_interrupt(
assert "ask_1" not in called
def test_subgraph_time_travel_resume_from_first_interrupt(
sync_checkpointer: BaseCheckpointSaver,
) -> None:
"""Time travel to a subgraph checkpoint at the first interrupt, then
resume through both interrupts with new answers.
This verifies the key bug fix: after time-traveling to a subgraph
checkpoint with an interrupt, a fork checkpoint is created so that
subsequent resumes find the correct state (not the old branch tip).
Parent: START --> executor (subgraph, checkpointer=True) --> END
Executor: START --> step_a --> ask_1 (interrupt) --> ask_2 (interrupt) --> END
Parent history after original run completes:
source=input next=(__start__,) values=[]
source=loop next=(executor,) values=[]
source=loop next=() values=[step_a_done, ask_1:answer_1, ask_2:answer_2]
After time-traveling to 1st interrupt + resuming with new answers:
source=input next=(__start__,) values=[]
source=loop next=(executor,) values=[] <-- branch point
source=loop next=() values=[..., ask_2:answer_2] (old branch)
source=fork next=(executor,) values=[] <-- fork from time travel
source=loop next=() values=[..., ask_2:new_answer_2] (new branch)
"""
called: list[str] = []
def step_a(state: State) -> State:
called.append("step_a")
return {"value": ["step_a_done"]}
def ask_1(state: State) -> State:
called.append("ask_1")
answer = interrupt("Question 1?")
return {"value": [f"ask_1:{answer}"]}
def ask_2(state: State) -> State:
called.append("ask_2")
answer = interrupt("Question 2?")
return {"value": [f"ask_2:{answer}"]}
executor = (
StateGraph(State)
.add_node("step_a", step_a)
.add_node("ask_1", ask_1)
.add_node("ask_2", ask_2)
.add_edge(START, "step_a")
.add_edge("step_a", "ask_1")
.add_edge("ask_1", "ask_2")
.add_edge("ask_2", "__end__")
.compile(checkpointer=True)
)
graph = (
StateGraph(State)
.add_node("executor", executor)
.add_edge(START, "executor")
.compile(checkpointer=sync_checkpointer)
)
config = {"configurable": {"thread_id": "1"}}
# --- Original run: hit both interrupts and resume ---
graph.invoke({"value": []}, config)
sub_config_at_first = graph.get_state(config, subgraphs=True).tasks[0].state.config
graph.invoke(Command(resume="answer_1"), config)
graph.invoke(Command(resume="answer_2"), config)
original = _checkpoint_summary(list(graph.get_state_history(config)))
assert [(s["source"], s["next"], s["values"]) for s in original] == [
("loop", (), {"value": ["step_a_done", "ask_1:answer_1", "ask_2:answer_2"]}),
("loop", ("executor",), {"value": []}),
("input", ("__start__",), {"value": []}),
]
# --- Time travel to first interrupt's subgraph checkpoint ---
called.clear()
replay_result = graph.invoke(None, sub_config_at_first)
assert replay_result["__interrupt__"][0].value == "Question 1?"
assert "step_a" not in called # before interrupt, not re-executed
# Fork is now the latest parent checkpoint
post_tt = _checkpoint_summary(list(graph.get_state_history(config)))
assert [(s["source"], s["next"]) for s in post_tt] == [
("fork", ("executor",)), # <-- new fork (latest)
("loop", ()), # original done
("loop", ("executor",)),
("input", ("__start__",)),
]
# --- Resume both interrupts with new answers ---
called.clear()
resume_1 = graph.invoke(Command(resume="new_answer_1"), config)
assert resume_1["__interrupt__"][0].value == "Question 2?"
assert "ask_1" in called
called.clear()
resume_2 = graph.invoke(Command(resume="new_answer_2"), config)
assert resume_2["value"] == [
"step_a_done",
"ask_1:new_answer_1",
"ask_2:new_answer_2",
]
# Verify final history: original branch preserved, new branch appended
final = _checkpoint_summary(list(graph.get_state_history(config)))
assert [(s["source"], s["next"], s["values"]) for s in final] == [
# New branch (from time travel fork)
(
"loop",
(),
{"value": ["step_a_done", "ask_1:new_answer_1", "ask_2:new_answer_2"]},
),
("fork", ("executor",), {"value": []}),
# Original branch (preserved)
("loop", (), {"value": ["step_a_done", "ask_1:answer_1", "ask_2:answer_2"]}),
("loop", ("executor",), {"value": []}),
("input", ("__start__",), {"value": []}),
]
def test_subgraph_time_travel_resume_from_second_interrupt(
sync_checkpointer: BaseCheckpointSaver,
) -> None:
"""Time travel to a subgraph checkpoint at the second interrupt, then
resume with a new answer. The first interrupt's answer should be preserved.
Parent: START --> executor (subgraph, checkpointer=True) --> END
Executor: START --> step_a --> ask_1 (interrupt) --> ask_2 (interrupt) --> END
Key assertion: after resuming from a time-travel to the 2nd interrupt,
the final state keeps ask_1's original answer but uses the new ask_2 answer.
"""
called: list[str] = []
def step_a(state: State) -> State:
called.append("step_a")
return {"value": ["step_a_done"]}
def ask_1(state: State) -> State:
called.append("ask_1")
answer = interrupt("Question 1?")
return {"value": [f"ask_1:{answer}"]}
def ask_2(state: State) -> State:
called.append("ask_2")
answer = interrupt("Question 2?")
return {"value": [f"ask_2:{answer}"]}
executor = (
StateGraph(State)
.add_node("step_a", step_a)
.add_node("ask_1", ask_1)
.add_node("ask_2", ask_2)
.add_edge(START, "step_a")
.add_edge("step_a", "ask_1")
.add_edge("ask_1", "ask_2")
.add_edge("ask_2", "__end__")
.compile(checkpointer=True)
)
graph = (
StateGraph(State)
.add_node("executor", executor)
.add_edge(START, "executor")
.compile(checkpointer=sync_checkpointer)
)
config = {"configurable": {"thread_id": "1"}}
# --- Original run: hit both interrupts and resume ---
graph.invoke({"value": []}, config)
graph.invoke(Command(resume="answer_1"), config)
sub_config_at_second = graph.get_state(config, subgraphs=True).tasks[0].state.config
graph.invoke(Command(resume="answer_2"), config)
original = _checkpoint_summary(list(graph.get_state_history(config)))
assert [(s["source"], s["next"], s["values"]) for s in original] == [
("loop", (), {"value": ["step_a_done", "ask_1:answer_1", "ask_2:answer_2"]}),
("loop", ("executor",), {"value": []}),
("input", ("__start__",), {"value": []}),
]
# --- Time travel to second interrupt ---
called.clear()
replay_result = graph.invoke(None, sub_config_at_second)
assert replay_result["__interrupt__"][0].value == "Question 2?"
assert "step_a" not in called
assert "ask_1" not in called # already resolved, not re-executed
# Fork is now the latest parent checkpoint
post_tt = _checkpoint_summary(list(graph.get_state_history(config)))
assert [(s["source"], s["next"]) for s in post_tt] == [
("fork", ("executor",)), # <-- new fork (latest)
("loop", ()), # original done
("loop", ("executor",)),
("input", ("__start__",)),
]
# --- Resume with a new answer for ask_2 only ---
called.clear()
resume_result = graph.invoke(Command(resume="new_answer_2"), config)
# ask_1's original answer preserved, ask_2 uses the new answer
assert resume_result["value"] == [
"step_a_done",
"ask_1:answer_1",
"ask_2:new_answer_2",
]
# Verify final history: original branch preserved, new branch appended
final = _checkpoint_summary(list(graph.get_state_history(config)))
assert [(s["source"], s["next"], s["values"]) for s in final] == [
# New branch (from time travel fork)
(
"loop",
(),
{"value": ["step_a_done", "ask_1:answer_1", "ask_2:new_answer_2"]},
),
("fork", ("executor",), {"value": []}),
# Original branch (preserved)
("loop", (), {"value": ["step_a_done", "ask_1:answer_1", "ask_2:answer_2"]}),
("loop", ("executor",), {"value": []}),
("input", ("__start__",), {"value": []}),
]
def test_subgraph_time_travel_checkpoint_pattern(
sync_checkpointer: BaseCheckpointSaver,
) -> None:
"""Verify the checkpoint pattern created by time travel to a subgraph
interrupt. A fork checkpoint should branch from the replay point and
become the latest parent checkpoint.
Parent: START --> executor (subgraph, checkpointer=True) --> END
Executor: START --> ask (interrupt) --> END
Original run (after completing):
source=input next=(__start__,) values=[]
source=loop next=(executor,) values=[] <-- replay point
source=loop next=() values=[a:first]
After time travel to interrupt + resume with "second":
source=input next=(__start__,) values=[]
source=loop next=(executor,) values=[] <-- branch point
source=loop next=() values=[a:first] (old branch)
source=fork next=(executor,) values=[] <-- fork
source=loop next=() values=[a:second] (new branch)
"""
def ask(state: State) -> State:
answer = interrupt("Q?")
return {"value": [f"a:{answer}"]}
executor = (
StateGraph(State)
.add_node("ask", ask)
.add_edge(START, "ask")
.compile(checkpointer=True)
)
graph = (
StateGraph(State)
.add_node("executor", executor)
.add_edge(START, "executor")
.compile(checkpointer=sync_checkpointer)
)
config = {"configurable": {"thread_id": "1"}}
# Run until interrupt, then complete
graph.invoke({"value": []}, config)
sub_config = graph.get_state(config, subgraphs=True).tasks[0].state.config
graph.invoke(Command(resume="first"), config)
original = _checkpoint_summary(list(graph.get_state_history(config)))
assert [(s["source"], s["next"], s["values"]) for s in original] == [
("loop", (), {"value": ["a:first"]}),
("loop", ("executor",), {"value": []}),
("input", ("__start__",), {"value": []}),
]
# Time travel to the interrupt
graph.invoke(None, sub_config)
# Fork is now the latest, branching from the original replay point
post_tt = list(graph.get_state_history(config))
post_tt_summary = _checkpoint_summary(post_tt)
assert [(s["source"], s["next"]) for s in post_tt_summary] == [
("fork", ("executor",)), # <-- new fork (latest)
("loop", ()),
("loop", ("executor",)), # <-- replay point / fork parent
("input", ("__start__",)),
]
# Verify the fork's parent is the original replay point
replay_point_id = sub_config["configurable"]["checkpoint_map"][""]
assert post_tt[0].parent_config["configurable"]["checkpoint_id"] == replay_point_id
# Resume from the fork — graph completes with new answer
result = graph.invoke(Command(resume="second"), config)
assert result["value"] == ["a:second"]
final = _checkpoint_summary(list(graph.get_state_history(config)))
assert [(s["source"], s["next"], s["values"]) for s in final] == [
# New branch
("loop", (), {"value": ["a:second"]}),
("fork", ("executor",), {"value": []}),
# Original branch
("loop", (), {"value": ["a:first"]}),
("loop", ("executor",), {"value": []}),
("input", ("__start__",), {"value": []}),
]
def test_subgraph_time_travel_after_completion(
sync_checkpointer: BaseCheckpointSaver,
) -> None:
@@ -2283,14 +3064,16 @@ def test_replay_creates_branch_preserving_old_checkpoints(
# -- Post-replay checkpoint history (newest first) --
post_replay_history = list(graph.get_state_history(config))
post_summary = _checkpoint_summary(post_replay_history)
assert len(post_summary) == 7 # 5 original + 2 new branch checkpoints
# 5 original + 1 fork + 2 new branch checkpoints = 8
assert len(post_summary) == 8
# Verify the full shape after replay
assert [s["next"] for s in post_summary] == [
(), # new branch tip (C6)
("node_c",), # new branch (C5)
(), # old branch tip (C4)
("node_c",), # old (C3)
(), # new branch tip
("node_c",), # new branch
("node_b",), # fork from replay point
(), # old branch tip
("node_c",), # old
("node_b",), # branch point (C2)
("node_a",), # old (C1)
("__start__",), # old (C0)
@@ -2298,6 +3081,7 @@ def test_replay_creates_branch_preserving_old_checkpoints(
assert [s["values"] for s in post_summary] == [
{"value": ["a", "b2", "c"]}, # new branch tip
{"value": ["a", "b2"]}, # new: node_b re-ran with call_count=2
{"value": ["a"]}, # fork from replay point
{"value": ["a", "b1", "c"]}, # old branch tip preserved
{"value": ["a", "b1"]}, # old
{"value": ["a"]}, # branch point
+403 -7
View File
@@ -46,6 +46,7 @@ def _checkpoint_summary(history: list) -> list[dict]:
Returns a list of dicts (newest-first, matching get_state_history order) with:
- id: short checkpoint id suffix (last 6 chars)
- parent_id: short parent checkpoint id suffix or None
- source: checkpoint metadata source (input, loop, fork, update)
- next: tuple of next node names
- values: channel values snapshot
"""
@@ -61,6 +62,7 @@ def _checkpoint_summary(history: list) -> list[dict]:
{
"id": cid[-6:],
"parent_id": pid[-6:] if pid else None,
"source": s.metadata.get("source"),
"next": s.next,
"values": s.values,
}
@@ -335,8 +337,14 @@ async def test_replay_interrupt_stable_across_replays(
r = await graph.ainvoke(None, before_ask.config)
results.append(r)
assert all(r == results[0] for r in results)
assert "__interrupt__" in results[0]
# Each replay creates a fork with a unique interrupt ID, so we compare
# interrupt values and state values rather than full equality.
assert all("__interrupt__" in r for r in results)
assert all(
r["__interrupt__"][0].value == results[0]["__interrupt__"][0].value
for r in results
)
assert all(r["value"] == results[0]["value"] for r in results)
@NEEDS_CONTEXTVARS
@@ -1261,6 +1269,391 @@ async def test_subgraph_time_travel_after_completion_async(
assert "ask_2:answer_2" in replay_result["value"]
@NEEDS_CONTEXTVARS
async def test_replay_from_before_interrupt_then_resume_async(
async_checkpointer: BaseCheckpointSaver,
) -> None:
"""Replay from checkpoint before interrupt node, then resume with a new
answer and verify the graph completes with the new value.
Graph: START --> node_a --> ask_human (interrupt) --> node_b --> END
"""
called: list[str] = []
async def node_a(state: State) -> State:
called.append("node_a")
return {"value": ["a"]}
async def ask_human(state: State) -> State:
called.append("ask_human")
answer = interrupt("What is your input?")
return {"value": [f"human:{answer}"]}
async def node_b(state: State) -> State:
called.append("node_b")
return {"value": ["b"]}
graph = (
StateGraph(State)
.add_node("node_a", node_a)
.add_node("ask_human", ask_human)
.add_node("node_b", node_b)
.add_edge(START, "node_a")
.add_edge("node_a", "ask_human")
.add_edge("ask_human", "node_b")
.compile(checkpointer=async_checkpointer)
)
config = {"configurable": {"thread_id": "1"}}
# --- Original run: invoke until interrupt, then resume to complete ---
await graph.ainvoke({"value": []}, config)
await graph.ainvoke(Command(resume="old_answer"), config)
original_history = [s async for s in graph.aget_state_history(config)]
original = _checkpoint_summary(original_history)
assert [(s["source"], s["next"], s["values"]) for s in original] == [
("loop", (), {"value": ["a", "human:old_answer", "b"]}),
("loop", ("node_b",), {"value": ["a", "human:old_answer"]}),
("loop", ("ask_human",), {"value": ["a"]}),
("loop", ("node_a",), {"value": []}),
("input", ("__start__",), {"value": []}),
]
# --- Replay from checkpoint before ask_human ---
before_ask = next(s for s in original_history if s.next == ("ask_human",))
called.clear()
replay_result = await graph.ainvoke(None, before_ask.config)
assert replay_result["__interrupt__"][0].value == "What is your input?"
assert "ask_human" in called
assert "node_a" not in called
# A fork checkpoint is now the latest
post_replay = _checkpoint_summary(
[s async for s in graph.aget_state_history(config)]
)
assert [(s["source"], s["next"]) for s in post_replay] == [
("fork", ("ask_human",)),
("loop", ()),
("loop", ("node_b",)),
("loop", ("ask_human",)),
("loop", ("node_a",)),
("input", ("__start__",)),
]
# --- Resume with a new answer ---
called.clear()
final_result = await graph.ainvoke(Command(resume="new_answer"), config)
assert final_result["value"] == ["a", "human:new_answer", "b"]
assert "ask_human" in called
assert "node_b" in called
final = _checkpoint_summary([s async for s in graph.aget_state_history(config)])
assert [(s["source"], s["next"], s["values"]) for s in final] == [
# New branch (from fork)
("loop", (), {"value": ["a", "human:new_answer", "b"]}),
("loop", ("node_b",), {"value": ["a", "human:new_answer"]}),
("fork", ("ask_human",), {"value": ["a"]}),
# Original branch (preserved)
("loop", (), {"value": ["a", "human:old_answer", "b"]}),
("loop", ("node_b",), {"value": ["a", "human:old_answer"]}),
("loop", ("ask_human",), {"value": ["a"]}),
("loop", ("node_a",), {"value": []}),
("input", ("__start__",), {"value": []}),
]
@NEEDS_CONTEXTVARS
async def test_subgraph_time_travel_resume_from_first_interrupt_async(
async_checkpointer: BaseCheckpointSaver,
) -> None:
"""Time travel to a subgraph checkpoint at the first interrupt, then
resume through both interrupts with new answers.
Parent: START --> executor (subgraph, checkpointer=True) --> END
Executor: START --> step_a --> ask_1 (interrupt) --> ask_2 (interrupt) --> END
"""
called: list[str] = []
async def step_a(state: State) -> State:
called.append("step_a")
return {"value": ["step_a_done"]}
async def ask_1(state: State) -> State:
called.append("ask_1")
answer = interrupt("Question 1?")
return {"value": [f"ask_1:{answer}"]}
async def ask_2(state: State) -> State:
called.append("ask_2")
answer = interrupt("Question 2?")
return {"value": [f"ask_2:{answer}"]}
executor = (
StateGraph(State)
.add_node("step_a", step_a)
.add_node("ask_1", ask_1)
.add_node("ask_2", ask_2)
.add_edge(START, "step_a")
.add_edge("step_a", "ask_1")
.add_edge("ask_1", "ask_2")
.add_edge("ask_2", "__end__")
.compile(checkpointer=True)
)
graph = (
StateGraph(State)
.add_node("executor", executor)
.add_edge(START, "executor")
.compile(checkpointer=async_checkpointer)
)
config = {"configurable": {"thread_id": "1"}}
# --- Original run: hit both interrupts and resume ---
await graph.ainvoke({"value": []}, config)
sub_config_at_first = (
(await graph.aget_state(config, subgraphs=True)).tasks[0].state.config
)
await graph.ainvoke(Command(resume="answer_1"), config)
await graph.ainvoke(Command(resume="answer_2"), config)
original = _checkpoint_summary([s async for s in graph.aget_state_history(config)])
assert [(s["source"], s["next"], s["values"]) for s in original] == [
("loop", (), {"value": ["step_a_done", "ask_1:answer_1", "ask_2:answer_2"]}),
("loop", ("executor",), {"value": []}),
("input", ("__start__",), {"value": []}),
]
# --- Time travel to first interrupt's subgraph checkpoint ---
called.clear()
replay_result = await graph.ainvoke(None, sub_config_at_first)
assert replay_result["__interrupt__"][0].value == "Question 1?"
assert "step_a" not in called
# Fork is now the latest parent checkpoint
post_tt = _checkpoint_summary([s async for s in graph.aget_state_history(config)])
assert [(s["source"], s["next"]) for s in post_tt] == [
("fork", ("executor",)), # <-- new fork (latest)
("loop", ()), # original done
("loop", ("executor",)),
("input", ("__start__",)),
]
# --- Resume both interrupts with new answers ---
called.clear()
resume_1 = await graph.ainvoke(Command(resume="new_answer_1"), config)
assert resume_1["__interrupt__"][0].value == "Question 2?"
assert "ask_1" in called
called.clear()
resume_2 = await graph.ainvoke(Command(resume="new_answer_2"), config)
assert resume_2["value"] == [
"step_a_done",
"ask_1:new_answer_1",
"ask_2:new_answer_2",
]
# Verify final history: original branch preserved, new branch appended
final = _checkpoint_summary([s async for s in graph.aget_state_history(config)])
assert [(s["source"], s["next"], s["values"]) for s in final] == [
# New branch (from time travel fork)
(
"loop",
(),
{"value": ["step_a_done", "ask_1:new_answer_1", "ask_2:new_answer_2"]},
),
("fork", ("executor",), {"value": []}),
# Original branch (preserved)
("loop", (), {"value": ["step_a_done", "ask_1:answer_1", "ask_2:answer_2"]}),
("loop", ("executor",), {"value": []}),
("input", ("__start__",), {"value": []}),
]
@NEEDS_CONTEXTVARS
async def test_subgraph_time_travel_resume_from_second_interrupt_async(
async_checkpointer: BaseCheckpointSaver,
) -> None:
"""Time travel to a subgraph checkpoint at the second interrupt, then
resume with a new answer. The first interrupt's answer should be preserved.
Parent: START --> executor (subgraph, checkpointer=True) --> END
Executor: START --> step_a --> ask_1 (interrupt) --> ask_2 (interrupt) --> END
"""
called: list[str] = []
async def step_a(state: State) -> State:
called.append("step_a")
return {"value": ["step_a_done"]}
async def ask_1(state: State) -> State:
called.append("ask_1")
answer = interrupt("Question 1?")
return {"value": [f"ask_1:{answer}"]}
async def ask_2(state: State) -> State:
called.append("ask_2")
answer = interrupt("Question 2?")
return {"value": [f"ask_2:{answer}"]}
executor = (
StateGraph(State)
.add_node("step_a", step_a)
.add_node("ask_1", ask_1)
.add_node("ask_2", ask_2)
.add_edge(START, "step_a")
.add_edge("step_a", "ask_1")
.add_edge("ask_1", "ask_2")
.add_edge("ask_2", "__end__")
.compile(checkpointer=True)
)
graph = (
StateGraph(State)
.add_node("executor", executor)
.add_edge(START, "executor")
.compile(checkpointer=async_checkpointer)
)
config = {"configurable": {"thread_id": "1"}}
# --- Original run: hit both interrupts and resume ---
await graph.ainvoke({"value": []}, config)
await graph.ainvoke(Command(resume="answer_1"), config)
sub_config_at_second = (
(await graph.aget_state(config, subgraphs=True)).tasks[0].state.config
)
await graph.ainvoke(Command(resume="answer_2"), config)
original = _checkpoint_summary([s async for s in graph.aget_state_history(config)])
assert [(s["source"], s["next"], s["values"]) for s in original] == [
("loop", (), {"value": ["step_a_done", "ask_1:answer_1", "ask_2:answer_2"]}),
("loop", ("executor",), {"value": []}),
("input", ("__start__",), {"value": []}),
]
# --- Time travel to second interrupt ---
called.clear()
replay_result = await graph.ainvoke(None, sub_config_at_second)
assert replay_result["__interrupt__"][0].value == "Question 2?"
assert "step_a" not in called
assert "ask_1" not in called
# Fork is now the latest parent checkpoint
post_tt = _checkpoint_summary([s async for s in graph.aget_state_history(config)])
assert [(s["source"], s["next"]) for s in post_tt] == [
("fork", ("executor",)), # <-- new fork (latest)
("loop", ()), # original done
("loop", ("executor",)),
("input", ("__start__",)),
]
# --- Resume with a new answer for ask_2 only ---
called.clear()
resume_result = await graph.ainvoke(Command(resume="new_answer_2"), config)
assert resume_result["value"] == [
"step_a_done",
"ask_1:answer_1",
"ask_2:new_answer_2",
]
# Verify final history
final = _checkpoint_summary([s async for s in graph.aget_state_history(config)])
assert [(s["source"], s["next"], s["values"]) for s in final] == [
# New branch (from time travel fork)
(
"loop",
(),
{"value": ["step_a_done", "ask_1:answer_1", "ask_2:new_answer_2"]},
),
("fork", ("executor",), {"value": []}),
# Original branch (preserved)
("loop", (), {"value": ["step_a_done", "ask_1:answer_1", "ask_2:answer_2"]}),
("loop", ("executor",), {"value": []}),
("input", ("__start__",), {"value": []}),
]
@NEEDS_CONTEXTVARS
async def test_subgraph_time_travel_checkpoint_pattern_async(
async_checkpointer: BaseCheckpointSaver,
) -> None:
"""Verify the checkpoint pattern created by time travel to a subgraph
interrupt. A fork checkpoint should branch from the replay point.
Parent: START --> executor (subgraph, checkpointer=True) --> END
Executor: START --> ask (interrupt) --> END
"""
async def ask(state: State) -> State:
answer = interrupt("Q?")
return {"value": [f"a:{answer}"]}
executor = (
StateGraph(State)
.add_node("ask", ask)
.add_edge(START, "ask")
.compile(checkpointer=True)
)
graph = (
StateGraph(State)
.add_node("executor", executor)
.add_edge(START, "executor")
.compile(checkpointer=async_checkpointer)
)
config = {"configurable": {"thread_id": "1"}}
# Run until interrupt, then complete
await graph.ainvoke({"value": []}, config)
sub_config = (await graph.aget_state(config, subgraphs=True)).tasks[0].state.config
await graph.ainvoke(Command(resume="first"), config)
original = _checkpoint_summary([s async for s in graph.aget_state_history(config)])
assert [(s["source"], s["next"], s["values"]) for s in original] == [
("loop", (), {"value": ["a:first"]}),
("loop", ("executor",), {"value": []}),
("input", ("__start__",), {"value": []}),
]
# Time travel to the interrupt
await graph.ainvoke(None, sub_config)
# Fork is now the latest, branching from the original replay point
post_tt = [s async for s in graph.aget_state_history(config)]
post_tt_summary = _checkpoint_summary(post_tt)
assert [(s["source"], s["next"]) for s in post_tt_summary] == [
("fork", ("executor",)), # <-- new fork (latest)
("loop", ()),
("loop", ("executor",)), # <-- replay point / fork parent
("input", ("__start__",)),
]
# Verify the fork's parent is the original replay point
replay_point_id = sub_config["configurable"]["checkpoint_map"][""]
assert post_tt[0].parent_config["configurable"]["checkpoint_id"] == replay_point_id
# Resume from the fork
result = await graph.ainvoke(Command(resume="second"), config)
assert result["value"] == ["a:second"]
final = _checkpoint_summary([s async for s in graph.aget_state_history(config)])
assert [(s["source"], s["next"], s["values"]) for s in final] == [
# New branch
("loop", (), {"value": ["a:second"]}),
("fork", ("executor",), {"value": []}),
# Original branch
("loop", (), {"value": ["a:first"]}),
("loop", ("executor",), {"value": []}),
("input", ("__start__",), {"value": []}),
]
@NEEDS_CONTEXTVARS
async def test_3_levels_deep_time_travel_to_first_interrupt_async(
async_checkpointer: BaseCheckpointSaver,
@@ -2088,13 +2481,15 @@ async def test_replay_creates_branch_preserving_old_checkpoints(
# -- Post-replay checkpoint history (newest first) --
post_replay_history = [s async for s in graph.aget_state_history(config)]
post_summary = _checkpoint_summary(post_replay_history)
assert len(post_summary) == 7 # 5 original + 2 new branch checkpoints
# 5 original + 1 fork + 2 new branch checkpoints = 8
assert len(post_summary) == 8
assert [s["next"] for s in post_summary] == [
(), # new branch tip (C6)
("node_c",), # new branch (C5)
(), # old branch tip (C4)
("node_c",), # old (C3)
(), # new branch tip
("node_c",), # new branch
("node_b",), # fork from replay point
(), # old branch tip
("node_c",), # old
("node_b",), # branch point (C2)
("node_a",), # old (C1)
("__start__",), # old (C0)
@@ -2102,6 +2497,7 @@ async def test_replay_creates_branch_preserving_old_checkpoints(
assert [s["values"] for s in post_summary] == [
{"value": ["a", "b2", "c"]}, # new branch tip
{"value": ["a", "b2"]}, # new: node_b re-ran with call_count=2
{"value": ["a"]}, # fork from replay point
{"value": ["a", "b1", "c"]}, # old branch tip preserved
{"value": ["a", "b1"]}, # old
{"value": ["a"]}, # branch point
+12 -24
View File
@@ -1348,11 +1348,10 @@ wheels = [
[[package]]
name = "langchain-core"
version = "1.3.2"
version = "1.3.0"
source = { registry = "https://pypi.org/simple" }
dependencies = [
{ name = "jsonpatch" },
{ name = "langchain-protocol" },
{ name = "langsmith" },
{ name = "packaging" },
{ name = "pydantic" },
@@ -1361,26 +1360,14 @@ dependencies = [
{ name = "typing-extensions" },
{ name = "uuid-utils" },
]
sdist = { url = "https://files.pythonhosted.org/packages/a8/03/7219502e8ca728d65eb44d7a3eb60239230742a70dbfc9241b9bfd61c4ab/langchain_core-1.3.2.tar.gz", hash = "sha256:fd7a50b2f28ba561fd9d7f5d2760bc9e06cf00cdf820a3ccafe88a94ffa8d5b7", size = 911813, upload-time = "2026-04-24T15:49:23.699Z" }
sdist = { url = "https://files.pythonhosted.org/packages/92/fe/20190232d9b513242899dbb0c2bb77e31b4d61e343743adbe90ebc2603d2/langchain_core-1.3.0.tar.gz", hash = "sha256:14a39f528bf459aa3aa40d0a7f7f1bae7520d435ef991ae14a4ceb74d8c49046", size = 860755, upload-time = "2026-04-17T14:51:38.298Z" }
wheels = [
{ url = "https://files.pythonhosted.org/packages/7d/d5/8fa4431007cbb7cfed7590f4d6a5dea3ad724f4174d248f6642ef5ce7d05/langchain_core-1.3.2-py3-none-any.whl", hash = "sha256:d44a66127f9f8db735bdfd0ab9661bccb47a97113cfd3f2d89c74864422b7274", size = 542390, upload-time = "2026-04-24T15:49:21.991Z" },
]
[[package]]
name = "langchain-protocol"
version = "0.0.11"
source = { registry = "https://pypi.org/simple" }
dependencies = [
{ name = "typing-extensions" },
]
sdist = { url = "https://files.pythonhosted.org/packages/d0/bb/38b5eaefa41c67735eedd9f9a2568b11c9eb376fa129a5edd7cc3dcde071/langchain_protocol-0.0.11.tar.gz", hash = "sha256:c276e2373b5ac691fc7ac9a72019d55182444ce8e89385c3f7e9f0185d0aace7", size = 6622, upload-time = "2026-04-23T22:13:16.771Z" }
wheels = [
{ url = "https://files.pythonhosted.org/packages/0b/fa/6a8ecad8472b182f2caf9d83fd89f40fc1590cb96546d90089b7869b7f5e/langchain_protocol-0.0.11-py3-none-any.whl", hash = "sha256:364da1faf6f5d3001413bede792c1a822c0f23ae55d1ce1266ca7d8e80e79011", size = 6778, upload-time = "2026-04-23T22:13:15.712Z" },
{ url = "https://files.pythonhosted.org/packages/f8/e2/dbfa347aa072a6dc4cd38d6f9ebfc730b4c14c258c47f480f4c5c546f177/langchain_core-1.3.0-py3-none-any.whl", hash = "sha256:baf16ee028475df177b9ab8869a751c79406d64a6f12125b93802991b566cced", size = 515140, upload-time = "2026-04-17T14:51:36.274Z" },
]
[[package]]
name = "langgraph"
version = "1.1.7a2"
version = "1.1.9"
source = { editable = "." }
dependencies = [
{ name = "langchain-core" },
@@ -1452,7 +1439,7 @@ test = [
[package.metadata]
requires-dist = [
{ name = "langchain-core", specifier = "==1.3.2" },
{ name = "langchain-core", specifier = ">=1.3.0,<2" },
{ name = "langgraph-checkpoint", editable = "../checkpoint" },
{ name = "langgraph-prebuilt", editable = "../prebuilt" },
{ name = "langgraph-sdk", editable = "../sdk-py" },
@@ -1561,7 +1548,7 @@ wheels = [
[[package]]
name = "langgraph-checkpoint"
version = "4.0.1"
version = "4.0.2"
source = { editable = "../checkpoint" }
dependencies = [
{ name = "langchain-core" },
@@ -1719,7 +1706,7 @@ inmem = [
requires-dist = [
{ name = "click", specifier = ">=8.1.7" },
{ name = "httpx", specifier = ">=0.24.0" },
{ name = "langgraph-api", marker = "python_full_version >= '3.11' and extra == 'inmem'", specifier = ">=0.5.35,<0.8.0" },
{ name = "langgraph-api", marker = "python_full_version >= '3.11' and extra == 'inmem'", specifier = ">=0.5.35,<0.9.0" },
{ name = "langgraph-runtime-inmem", marker = "python_full_version >= '3.11' and extra == 'inmem'", specifier = ">=0.7" },
{ name = "langgraph-sdk", marker = "python_full_version >= '3.11'", specifier = ">=0.1.0" },
{ name = "pathspec", specifier = ">=0.11.0" },
@@ -1755,7 +1742,7 @@ test = [
[[package]]
name = "langgraph-prebuilt"
version = "1.0.9"
version = "1.0.10"
source = { editable = "../prebuilt" }
dependencies = [
{ name = "langchain-core" },
@@ -1865,7 +1852,7 @@ test = [
[[package]]
name = "langsmith"
version = "0.6.4"
version = "0.7.31"
source = { registry = "https://pypi.org/simple" }
dependencies = [
{ name = "httpx" },
@@ -1875,11 +1862,12 @@ dependencies = [
{ name = "requests" },
{ name = "requests-toolbelt" },
{ name = "uuid-utils" },
{ name = "xxhash" },
{ name = "zstandard" },
]
sdist = { url = "https://files.pythonhosted.org/packages/e7/85/9c7933052a997da1b85bc5c774f3865e9b1da1c8d71541ea133178b13229/langsmith-0.6.4.tar.gz", hash = "sha256:36f7223a01c218079fbb17da5e536ebbaf5c1468c028abe070aa3ae59bc99ec8", size = 919964, upload-time = "2026-01-15T20:02:28.873Z" }
sdist = { url = "https://files.pythonhosted.org/packages/e6/11/696019490992db5c87774dc20515529ef42a01e1d770fb754ed6d9b12fb0/langsmith-0.7.31.tar.gz", hash = "sha256:331ee4f7c26bb5be4022b9859b7d7b122cbf8c9d01d9f530114c1914b0349ffb", size = 1178480, upload-time = "2026-04-14T17:55:41.242Z" }
wheels = [
{ url = "https://files.pythonhosted.org/packages/66/0f/09a6637a7ba777eb307b7c80852d9ee26438e2bdafbad6fcc849ff9d9192/langsmith-0.6.4-py3-none-any.whl", hash = "sha256:ac4835860160be371042c7adbba3cb267bcf8d96a5ea976c33a8a4acad6c5486", size = 283503, upload-time = "2026-01-15T20:02:26.662Z" },
{ url = "https://files.pythonhosted.org/packages/1d/a1/a013cf458c301cda86a213dd153ce0a01c93f1ab5833f951e6a44c9763ce/langsmith-0.7.31-py3-none-any.whl", hash = "sha256:0291d49203f6e80dda011af1afda61eb0595a4d697adb684590a8805e1d61fb6", size = 373276, upload-time = "2026-04-14T17:55:39.677Z" },
]
[package.optional-dependencies]
+23 -7
View File
@@ -614,6 +614,7 @@ class _InjectedArgs:
store: str | None
runtime: str | None
all_injected_keys: set[str]
_optional_state_args: set[str]
class ToolNode(RunnableCallable):
@@ -807,6 +808,7 @@ class ToolNode(RunnableCallable):
context=runtime.context,
store=runtime.store,
stream_writer=runtime.stream_writer,
tools=list(self.tools_by_name.values()),
execution_info=runtime.execution_info,
server_info=runtime.server_info,
)
@@ -841,6 +843,7 @@ class ToolNode(RunnableCallable):
context=runtime.context,
store=runtime.store,
stream_writer=runtime.stream_writer,
tools=list(self.tools_by_name.values()),
execution_info=runtime.execution_info,
server_info=runtime.server_info,
)
@@ -1333,7 +1336,7 @@ class ToolNode(RunnableCallable):
return tool_call
tool_call_copy: ToolCall = copy(tool_call)
injected_args = {}
injected_args: dict[str, Any] = {}
# Inject state
if injected.state:
@@ -1361,14 +1364,20 @@ class ToolNode(RunnableCallable):
# Extract state values
if isinstance(state, dict):
for tool_arg, state_field in injected.state.items():
injected_args[tool_arg] = (
state[state_field] if state_field else state
)
if not state_field:
injected_args[tool_arg] = state
elif state_field in state:
injected_args[tool_arg] = state[state_field]
elif tool_arg not in injected._optional_state_args:
raise KeyError(state_field)
else:
for tool_arg, state_field in injected.state.items():
injected_args[tool_arg] = (
getattr(state, state_field) if state_field else state
)
if not state_field:
injected_args[tool_arg] = state
elif hasattr(state, state_field):
injected_args[tool_arg] = getattr(state, state_field)
elif tool_arg not in injected._optional_state_args:
raise AttributeError(state_field)
# Inject store
if injected.store:
@@ -1569,6 +1578,7 @@ class ToolRuntime(_DirectlyInjectedToolArg, Generic[ContextT, StateT]):
- `context`: Runtime context (shared with `Runtime`)
- `store`: `BaseStore` instance for persistent storage (shared with `Runtime`)
- `stream_writer`: `StreamWriter` for streaming output (shared with `Runtime`)
- `tools`: List of all available `BaseTool` instances
No `Annotated` wrapper is needed - just use `runtime: ToolRuntime`
as a parameter.
@@ -1611,6 +1621,7 @@ class ToolRuntime(_DirectlyInjectedToolArg, Generic[ContextT, StateT]):
context: ContextT
config: RunnableConfig
stream_writer: StreamWriter
tools: list[BaseTool]
tool_call_id: str | None
store: BaseStore | None
execution_info: ExecutionInfo | None = None
@@ -1859,6 +1870,7 @@ def _get_all_injected_args(tool: BaseTool) -> _InjectedArgs:
store_arg: str | None = None
runtime_arg: str | None = None
all_injected_keys: set[str] = set()
_optional_state_args: set[str] = set()
for name, type_ in all_annotations.items():
# Track all InjectedToolArg-annotated params (including custom subclasses)
@@ -1873,6 +1885,9 @@ def _get_all_injected_args(tool: BaseTool) -> _InjectedArgs:
if state_inj := _get_injection_from_type(type_, InjectedState):
if isinstance(state_inj, InjectedState) and state_inj.field:
state_args[name] = state_inj.field
field_info = full_schema.model_fields.get(name)
if field_info and not field_info.is_required():
_optional_state_args.add(name)
else:
state_args[name] = None
@@ -1889,4 +1904,5 @@ def _get_all_injected_args(tool: BaseTool) -> _InjectedArgs:
store=store_arg,
runtime=runtime_arg,
all_injected_keys=all_injected_keys,
_optional_state_args=_optional_state_args,
)
+1 -1
View File
@@ -4,7 +4,7 @@ build-backend = "hatchling.build"
[project]
name = "langgraph-prebuilt"
version = "1.0.9"
version = "1.0.10"
description = "Library with high-level APIs for creating and executing LangGraph agents and tools."
authors = []
requires-python = ">=3.10"
@@ -0,0 +1,285 @@
"""Test InjectedState with NotRequired state fields.
This tests the fix for https://github.com/langchain-ai/langchain/issues/35585
When using InjectedState(<field>) on a tool parameter, and the referenced field is
declared as NotRequired in the custom state schema, the ToolNode should gracefully
handle missing fields by injecting None instead of raising KeyError.
"""
import sys
from typing import Annotated
import pytest
from langchain_core.messages import AIMessage, AnyMessage, HumanMessage, ToolMessage
from langchain_core.tools import tool
from langgraph.graph.message import add_messages
from pydantic import BaseModel, Field
from typing_extensions import NotRequired
from langgraph.prebuilt import InjectedState, ToolNode, create_react_agent
from langgraph.prebuilt.chat_agent_executor import AgentState
from .model import FakeToolCallingModel
class CustomAgentStateWithNotRequired(AgentState):
"""Custom state with a NotRequired field (TypedDict style)."""
city: NotRequired[str]
class CustomAgentStatePydanticWithDefault(BaseModel):
"""Custom state with Optional field and default (Pydantic style)."""
messages: Annotated[list[AnyMessage], add_messages]
remaining_steps: int = Field(default=10)
city: str | None = Field(default=None)
@tool
def get_weather(city: Annotated[str | None, InjectedState("city")] = None) -> str:
"""Get weather for a given city."""
if city is None:
return "No city provided"
return f"It's always sunny in {city}!"
def _create_mock_runtime(
state: dict | None = None,
store=None,
):
"""Create a mock Runtime for testing ToolNode directly."""
from unittest.mock import Mock
from langgraph.runtime import Runtime
mock_runtime = Mock(spec=Runtime)
mock_runtime.context = {}
return mock_runtime
def _create_config_with_runtime(store=None, state=None):
"""Create a RunnableConfig with mocked runtime for direct ToolNode testing."""
from langgraph.prebuilt.tool_node import ToolRuntime
tool_runtime = ToolRuntime(
state=state or {},
config={},
context={},
store=store,
stream_writer=None,
tools=[],
tool_call_id="test_id",
)
return {
"configurable": {
"__pregel_runtime": _create_mock_runtime(),
"__tool_runtime__": tool_runtime,
}
}
@pytest.mark.skipif(
sys.version_info < (3, 11),
reason="InjectedState field extraction from Optional[Annotated[...]] not supported on Python <3.11",
)
def test_injected_state_not_required_field_missing_injects_none():
"""Test that InjectedState with NotRequired field injects None when field is missing.
This verifies the fix for https://github.com/langchain-ai/langchain/issues/35585
"""
tool_node = ToolNode([get_weather])
tool_call = {
"name": "get_weather",
"args": {},
"id": "call_1",
"type": "tool_call",
}
ai_msg = AIMessage("Let me check the weather", tool_calls=[tool_call])
# State WITHOUT the "city" field - should inject None instead of raising KeyError
state_without_city: CustomAgentStateWithNotRequired = {
"messages": [HumanMessage("What's the weather?"), ai_msg],
}
result = tool_node.invoke(
state_without_city,
config=_create_config_with_runtime(state=state_without_city),
)
assert len(result["messages"]) == 1
tool_msg = result["messages"][0]
assert isinstance(tool_msg, ToolMessage)
assert "No city provided" in tool_msg.content
@pytest.mark.skipif(
sys.version_info < (3, 11),
reason="InjectedState field extraction from Optional[Annotated[...]] not supported on Python <3.11",
)
def test_injected_state_not_required_field_present_works():
"""Test that InjectedState with NotRequired field works when field IS present."""
tool_node = ToolNode([get_weather])
tool_call = {
"name": "get_weather",
"args": {},
"id": "call_1",
"type": "tool_call",
}
ai_msg = AIMessage("Let me check the weather", tool_calls=[tool_call])
# State WITH the "city" field - this should work
state_with_city: CustomAgentStateWithNotRequired = {
"messages": [HumanMessage("What's the weather?"), ai_msg],
"city": "San Francisco",
}
result = tool_node.invoke(
state_with_city,
config=_create_config_with_runtime(state=state_with_city),
)
assert len(result["messages"]) == 1
tool_msg = result["messages"][0]
assert isinstance(tool_msg, ToolMessage)
assert "San Francisco" in tool_msg.content
@pytest.mark.skipif(
sys.version_info < (3, 11),
reason="InjectedState field extraction from Optional[Annotated[...]] not supported on Python <3.11",
)
def test_create_react_agent_injected_state_not_required_field_missing():
"""Test create_react_agent with InjectedState using NotRequired field that is missing.
This verifies the fix for https://github.com/langchain-ai/langchain/issues/35585
"""
model = FakeToolCallingModel(
tool_calls=[
[{"name": "get_weather", "args": {}, "id": "call_1"}],
[], # No more tool calls, agent should stop
]
)
agent = create_react_agent(
model,
tools=[get_weather],
state_schema=CustomAgentStateWithNotRequired,
)
# Invoke WITHOUT the city field - should work, injecting None
result = agent.invoke(
{"messages": [HumanMessage("What's the weather?")]},
)
# Check that the tool was called successfully with None injected
messages = result["messages"]
tool_messages = [m for m in messages if isinstance(m, ToolMessage)]
assert len(tool_messages) == 1
assert "No city provided" in tool_messages[0].content
@pytest.mark.skipif(
sys.version_info < (3, 11),
reason="InjectedState field extraction from Optional[Annotated[...]] not supported on Python <3.11",
)
def test_create_react_agent_injected_state_not_required_field_present():
"""Test create_react_agent with InjectedState using NotRequired field that IS present."""
model = FakeToolCallingModel(
tool_calls=[
[{"name": "get_weather", "args": {}, "id": "call_1"}],
[], # No more tool calls, agent should stop
]
)
agent = create_react_agent(
model,
tools=[get_weather],
state_schema=CustomAgentStateWithNotRequired,
)
# Invoke WITH the city field
result = agent.invoke(
{
"messages": [HumanMessage("What's the weather?")],
"city": "San Francisco",
},
)
# Check that the tool was called successfully
messages = result["messages"]
tool_messages = [m for m in messages if isinstance(m, ToolMessage)]
assert len(tool_messages) == 1
assert "San Francisco" in tool_messages[0].content
@tool
def get_weather_optional(city: Annotated[str | None, InjectedState("city")]) -> str:
"""Get weather for a given city (accepts None)."""
if city is None:
return "Please provide a city!"
return f"It's always sunny in {city}!"
def test_pydantic_state_with_default_field_missing_works():
"""Test that Pydantic state with Optional field and default=None works when field is missing.
This is the workaround suggested in the issue comments - using Pydantic BaseModel
with `city: Optional[str] = Field(default=None)` instead of TypedDict with NotRequired.
"""
model = FakeToolCallingModel(
tool_calls=[
[{"name": "get_weather_optional", "args": {}, "id": "call_1"}],
[], # No more tool calls, agent should stop
]
)
agent = create_react_agent(
model,
tools=[get_weather_optional],
state_schema=CustomAgentStatePydanticWithDefault,
)
# Invoke WITHOUT the city field - should work because Pydantic provides default
result = agent.invoke(
{"messages": [HumanMessage("What's the weather?")]},
)
# Check that the tool was called successfully with None
messages = result["messages"]
tool_messages = [m for m in messages if isinstance(m, ToolMessage)]
assert len(tool_messages) == 1
assert "Please provide a city!" in tool_messages[0].content
def test_pydantic_state_with_default_field_present_works():
"""Test that Pydantic state with Optional field works when field IS present."""
model = FakeToolCallingModel(
tool_calls=[
[{"name": "get_weather_optional", "args": {}, "id": "call_1"}],
[], # No more tool calls, agent should stop
]
)
agent = create_react_agent(
model,
tools=[get_weather_optional],
state_schema=CustomAgentStatePydanticWithDefault,
)
# Invoke WITH the city field
result = agent.invoke(
{
"messages": [HumanMessage("What's the weather?")],
"city": "San Francisco",
},
)
# Check that the tool was called successfully
messages = result["messages"]
tool_messages = [m for m in messages if isinstance(m, ToolMessage)]
assert len(tool_messages) == 1
assert "San Francisco" in tool_messages[0].content
+29 -8
View File
@@ -2016,8 +2016,8 @@ async def test_tool_node_inject_runtime_dynamic_tool_via_wrap_tool_call_async()
assert tool_message.tool_call_id == "call_dynamic_2"
def test_tool_runtime_forwards_execution_info_and_server_info() -> None:
"""Test that execution_info and server_info are forwarded from Runtime to ToolRuntime."""
def test_tool_runtime_forwards_execution_info_server_info_and_tools() -> None:
"""Test that execution_info, server_info, and tools are forwarded from Runtime to ToolRuntime."""
from langgraph.runtime import ExecutionInfo, ServerInfo
exec_info = ExecutionInfo(
@@ -2043,9 +2043,15 @@ def test_tool_runtime_forwards_execution_info_and_server_info() -> None:
"""Tool that captures runtime info."""
captured["execution_info"] = runtime.execution_info
captured["server_info"] = runtime.server_info
captured["tools"] = runtime.tools
return "ok"
node = ToolNode([info_tool])
@dec_tool
def other_tool(y: int) -> str:
"""Another tool available to the runtime."""
return str(y)
node = ToolNode([info_tool, other_tool])
tool_call = {
"name": "info_tool",
"args": {"x": 1},
@@ -2054,17 +2060,21 @@ def test_tool_runtime_forwards_execution_info_and_server_info() -> None:
}
msg = AIMessage("", tool_calls=[tool_call])
config: RunnableConfig = {"configurable": {"__pregel_runtime": mock_runtime}}
node.invoke({"messages": [msg]}, config=config)
result = node.invoke({"messages": [msg]}, config=config)
assert result["messages"][-1].content == "ok"
assert captured["execution_info"] is exec_info
assert captured["execution_info"].thread_id == "t-1"
assert captured["execution_info"].task_id == "tk-1"
assert captured["server_info"] is server_info
assert captured["server_info"].assistant_id == "asst-1"
assert [tool.name for tool in captured["tools"]] == ["info_tool", "other_tool"]
async def test_tool_runtime_forwards_execution_info_and_server_info_async() -> None:
"""Test that execution_info and server_info are forwarded in async path."""
async def test_tool_runtime_forwards_execution_info_server_info_and_tools_async() -> (
None
):
"""Test that execution_info, server_info, and tools are forwarded in async path."""
from langgraph.runtime import ExecutionInfo, ServerInfo
exec_info = ExecutionInfo(
@@ -2090,9 +2100,15 @@ async def test_tool_runtime_forwards_execution_info_and_server_info_async() -> N
"""Async tool that captures runtime info."""
captured["execution_info"] = runtime.execution_info
captured["server_info"] = runtime.server_info
captured["tools"] = runtime.tools
return "ok"
node = ToolNode([info_tool_async])
@dec_tool
async def other_tool_async(y: int) -> str:
"""Another async tool available to the runtime."""
return str(y)
node = ToolNode([info_tool_async, other_tool_async])
tool_call = {
"name": "info_tool_async",
"args": {"x": 1},
@@ -2101,12 +2117,17 @@ async def test_tool_runtime_forwards_execution_info_and_server_info_async() -> N
}
msg = AIMessage("", tool_calls=[tool_call])
config: RunnableConfig = {"configurable": {"__pregel_runtime": mock_runtime}}
await node.ainvoke({"messages": [msg]}, config=config)
result = await node.ainvoke({"messages": [msg]}, config=config)
assert result["messages"][-1].content == "ok"
assert captured["execution_info"] is exec_info
assert captured["execution_info"].thread_id == "t-2"
assert captured["server_info"] is server_info
assert captured["server_info"].graph_id == "graph-2"
assert [tool.name for tool in captured["tools"]] == [
"info_tool_async",
"other_tool_async",
]
# --- InjectedToolArg security tests ---
+11 -23
View File
@@ -249,11 +249,10 @@ wheels = [
[[package]]
name = "langchain-core"
version = "1.3.2"
version = "1.3.0"
source = { registry = "https://pypi.org/simple" }
dependencies = [
{ name = "jsonpatch" },
{ name = "langchain-protocol" },
{ name = "langsmith" },
{ name = "packaging" },
{ name = "pydantic" },
@@ -262,26 +261,14 @@ dependencies = [
{ name = "typing-extensions" },
{ name = "uuid-utils" },
]
sdist = { url = "https://files.pythonhosted.org/packages/a8/03/7219502e8ca728d65eb44d7a3eb60239230742a70dbfc9241b9bfd61c4ab/langchain_core-1.3.2.tar.gz", hash = "sha256:fd7a50b2f28ba561fd9d7f5d2760bc9e06cf00cdf820a3ccafe88a94ffa8d5b7", size = 911813, upload-time = "2026-04-24T15:49:23.699Z" }
sdist = { url = "https://files.pythonhosted.org/packages/92/fe/20190232d9b513242899dbb0c2bb77e31b4d61e343743adbe90ebc2603d2/langchain_core-1.3.0.tar.gz", hash = "sha256:14a39f528bf459aa3aa40d0a7f7f1bae7520d435ef991ae14a4ceb74d8c49046", size = 860755, upload-time = "2026-04-17T14:51:38.298Z" }
wheels = [
{ url = "https://files.pythonhosted.org/packages/7d/d5/8fa4431007cbb7cfed7590f4d6a5dea3ad724f4174d248f6642ef5ce7d05/langchain_core-1.3.2-py3-none-any.whl", hash = "sha256:d44a66127f9f8db735bdfd0ab9661bccb47a97113cfd3f2d89c74864422b7274", size = 542390, upload-time = "2026-04-24T15:49:21.991Z" },
]
[[package]]
name = "langchain-protocol"
version = "0.0.11"
source = { registry = "https://pypi.org/simple" }
dependencies = [
{ name = "typing-extensions" },
]
sdist = { url = "https://files.pythonhosted.org/packages/d0/bb/38b5eaefa41c67735eedd9f9a2568b11c9eb376fa129a5edd7cc3dcde071/langchain_protocol-0.0.11.tar.gz", hash = "sha256:c276e2373b5ac691fc7ac9a72019d55182444ce8e89385c3f7e9f0185d0aace7", size = 6622, upload-time = "2026-04-23T22:13:16.771Z" }
wheels = [
{ url = "https://files.pythonhosted.org/packages/0b/fa/6a8ecad8472b182f2caf9d83fd89f40fc1590cb96546d90089b7869b7f5e/langchain_protocol-0.0.11-py3-none-any.whl", hash = "sha256:364da1faf6f5d3001413bede792c1a822c0f23ae55d1ce1266ca7d8e80e79011", size = 6778, upload-time = "2026-04-23T22:13:15.712Z" },
{ url = "https://files.pythonhosted.org/packages/f8/e2/dbfa347aa072a6dc4cd38d6f9ebfc730b4c14c258c47f480f4c5c546f177/langchain_core-1.3.0-py3-none-any.whl", hash = "sha256:baf16ee028475df177b9ab8869a751c79406d64a6f12125b93802991b566cced", size = 515140, upload-time = "2026-04-17T14:51:36.274Z" },
]
[[package]]
name = "langgraph"
version = "1.1.7a2"
version = "1.1.9"
source = { editable = "../langgraph" }
dependencies = [
{ name = "langchain-core" },
@@ -294,7 +281,7 @@ dependencies = [
[package.metadata]
requires-dist = [
{ name = "langchain-core", specifier = "==1.3.2" },
{ name = "langchain-core", specifier = ">=1.3.0,<2" },
{ name = "langgraph-checkpoint", editable = "../checkpoint" },
{ name = "langgraph-prebuilt", editable = "." },
{ name = "langgraph-sdk", editable = "../sdk-py" },
@@ -365,7 +352,7 @@ test = [
[[package]]
name = "langgraph-checkpoint"
version = "4.0.1"
version = "4.0.2"
source = { editable = "../checkpoint" }
dependencies = [
{ name = "langchain-core" },
@@ -503,7 +490,7 @@ test = [
[[package]]
name = "langgraph-prebuilt"
version = "1.0.9"
version = "1.0.10"
source = { editable = "." }
dependencies = [
{ name = "langchain-core" },
@@ -632,7 +619,7 @@ test = [
[[package]]
name = "langsmith"
version = "0.6.4"
version = "0.7.31"
source = { registry = "https://pypi.org/simple" }
dependencies = [
{ name = "httpx" },
@@ -642,11 +629,12 @@ dependencies = [
{ name = "requests" },
{ name = "requests-toolbelt" },
{ name = "uuid-utils" },
{ name = "xxhash" },
{ name = "zstandard" },
]
sdist = { url = "https://files.pythonhosted.org/packages/e7/85/9c7933052a997da1b85bc5c774f3865e9b1da1c8d71541ea133178b13229/langsmith-0.6.4.tar.gz", hash = "sha256:36f7223a01c218079fbb17da5e536ebbaf5c1468c028abe070aa3ae59bc99ec8", size = 919964, upload-time = "2026-01-15T20:02:28.873Z" }
sdist = { url = "https://files.pythonhosted.org/packages/e6/11/696019490992db5c87774dc20515529ef42a01e1d770fb754ed6d9b12fb0/langsmith-0.7.31.tar.gz", hash = "sha256:331ee4f7c26bb5be4022b9859b7d7b122cbf8c9d01d9f530114c1914b0349ffb", size = 1178480, upload-time = "2026-04-14T17:55:41.242Z" }
wheels = [
{ url = "https://files.pythonhosted.org/packages/66/0f/09a6637a7ba777eb307b7c80852d9ee26438e2bdafbad6fcc849ff9d9192/langsmith-0.6.4-py3-none-any.whl", hash = "sha256:ac4835860160be371042c7adbba3cb267bcf8d96a5ea976c33a8a4acad6c5486", size = 283503, upload-time = "2026-01-15T20:02:26.662Z" },
{ url = "https://files.pythonhosted.org/packages/1d/a1/a013cf458c301cda86a213dd153ce0a01c93f1ab5833f951e6a44c9763ce/langsmith-0.7.31-py3-none-any.whl", hash = "sha256:0291d49203f6e80dda011af1afda61eb0595a4d697adb684590a8805e1d61fb6", size = 373276, upload-time = "2026-04-14T17:55:39.677Z" },
]
[[package]]
+10 -23
View File
@@ -262,11 +262,10 @@ wheels = [
[[package]]
name = "langchain-core"
version = "1.3.2"
version = "1.3.0"
source = { registry = "https://pypi.org/simple" }
dependencies = [
{ name = "jsonpatch" },
{ name = "langchain-protocol" },
{ name = "langsmith" },
{ name = "packaging" },
{ name = "pydantic" },
@@ -275,26 +274,14 @@ dependencies = [
{ name = "typing-extensions" },
{ name = "uuid-utils" },
]
sdist = { url = "https://files.pythonhosted.org/packages/a8/03/7219502e8ca728d65eb44d7a3eb60239230742a70dbfc9241b9bfd61c4ab/langchain_core-1.3.2.tar.gz", hash = "sha256:fd7a50b2f28ba561fd9d7f5d2760bc9e06cf00cdf820a3ccafe88a94ffa8d5b7", size = 911813, upload-time = "2026-04-24T15:49:23.699Z" }
sdist = { url = "https://files.pythonhosted.org/packages/92/fe/20190232d9b513242899dbb0c2bb77e31b4d61e343743adbe90ebc2603d2/langchain_core-1.3.0.tar.gz", hash = "sha256:14a39f528bf459aa3aa40d0a7f7f1bae7520d435ef991ae14a4ceb74d8c49046", size = 860755, upload-time = "2026-04-17T14:51:38.298Z" }
wheels = [
{ url = "https://files.pythonhosted.org/packages/7d/d5/8fa4431007cbb7cfed7590f4d6a5dea3ad724f4174d248f6642ef5ce7d05/langchain_core-1.3.2-py3-none-any.whl", hash = "sha256:d44a66127f9f8db735bdfd0ab9661bccb47a97113cfd3f2d89c74864422b7274", size = 542390, upload-time = "2026-04-24T15:49:21.991Z" },
]
[[package]]
name = "langchain-protocol"
version = "0.0.11"
source = { registry = "https://pypi.org/simple" }
dependencies = [
{ name = "typing-extensions" },
]
sdist = { url = "https://files.pythonhosted.org/packages/d0/bb/38b5eaefa41c67735eedd9f9a2568b11c9eb376fa129a5edd7cc3dcde071/langchain_protocol-0.0.11.tar.gz", hash = "sha256:c276e2373b5ac691fc7ac9a72019d55182444ce8e89385c3f7e9f0185d0aace7", size = 6622, upload-time = "2026-04-23T22:13:16.771Z" }
wheels = [
{ url = "https://files.pythonhosted.org/packages/0b/fa/6a8ecad8472b182f2caf9d83fd89f40fc1590cb96546d90089b7869b7f5e/langchain_protocol-0.0.11-py3-none-any.whl", hash = "sha256:364da1faf6f5d3001413bede792c1a822c0f23ae55d1ce1266ca7d8e80e79011", size = 6778, upload-time = "2026-04-23T22:13:15.712Z" },
{ url = "https://files.pythonhosted.org/packages/f8/e2/dbfa347aa072a6dc4cd38d6f9ebfc730b4c14c258c47f480f4c5c546f177/langchain_core-1.3.0-py3-none-any.whl", hash = "sha256:baf16ee028475df177b9ab8869a751c79406d64a6f12125b93802991b566cced", size = 515140, upload-time = "2026-04-17T14:51:36.274Z" },
]
[[package]]
name = "langgraph"
version = "1.1.7a2"
version = "1.1.9"
source = { editable = "../langgraph" }
dependencies = [
{ name = "langchain-core" },
@@ -307,7 +294,7 @@ dependencies = [
[package.metadata]
requires-dist = [
{ name = "langchain-core", specifier = "==1.3.2" },
{ name = "langchain-core", specifier = ">=1.3.0,<2" },
{ name = "langgraph-checkpoint", editable = "../checkpoint" },
{ name = "langgraph-prebuilt", editable = "../prebuilt" },
{ name = "langgraph-sdk", editable = "." },
@@ -378,7 +365,7 @@ test = [
[[package]]
name = "langgraph-checkpoint"
version = "4.0.1"
version = "4.0.2"
source = { editable = "../checkpoint" }
dependencies = [
{ name = "langchain-core" },
@@ -426,7 +413,7 @@ test = [
[[package]]
name = "langgraph-prebuilt"
version = "1.0.9"
version = "1.0.10"
source = { editable = "../prebuilt" }
dependencies = [
{ name = "langchain-core" },
@@ -547,7 +534,7 @@ test = [
[[package]]
name = "langsmith"
version = "0.7.20"
version = "0.7.31"
source = { registry = "https://pypi.org/simple" }
dependencies = [
{ name = "httpx" },
@@ -560,9 +547,9 @@ dependencies = [
{ name = "xxhash" },
{ name = "zstandard" },
]
sdist = { url = "https://files.pythonhosted.org/packages/80/c6/cbdc6638207f68a3c61ec0b64fa593f6b11de3170d03c852238c31b54960/langsmith-0.7.20.tar.gz", hash = "sha256:fa983a74f75648ee0e80d3f9751162b6f9a438896d5f9bdb6cba9abda451e234", size = 1134732, upload-time = "2026-03-18T00:03:39.129Z" }
sdist = { url = "https://files.pythonhosted.org/packages/e6/11/696019490992db5c87774dc20515529ef42a01e1d770fb754ed6d9b12fb0/langsmith-0.7.31.tar.gz", hash = "sha256:331ee4f7c26bb5be4022b9859b7d7b122cbf8c9d01d9f530114c1914b0349ffb", size = 1178480, upload-time = "2026-04-14T17:55:41.242Z" }
wheels = [
{ url = "https://files.pythonhosted.org/packages/e8/46/9294d4f49de6a8f08e8b83907713ca545459d87d474c6add15d31a36f5dc/langsmith-0.7.20-py3-none-any.whl", hash = "sha256:0162faf791ea48d69009a12a3da917468556b99cf5d5fcacbb8cda064262e118", size = 359314, upload-time = "2026-03-18T00:03:37.59Z" },
{ url = "https://files.pythonhosted.org/packages/1d/a1/a013cf458c301cda86a213dd153ce0a01c93f1ab5833f951e6a44c9763ce/langsmith-0.7.31-py3-none-any.whl", hash = "sha256:0291d49203f6e80dda011af1afda61eb0595a4d697adb684590a8805e1d61fb6", size = 373276, upload-time = "2026-04-14T17:55:39.677Z" },
]
[[package]]
+137
View File
@@ -0,0 +1,137 @@
# Delta-channel reconstruction: query strategy benchmark
**Branch:** `delta-channel-writes-based`
**Question (Nuno):** Is the recursive CTE the right query shape for reconstructing a delta channel inside `get_tuple`, or would a plain `SELECT WHERE` be cheaper even though it returns more rows?
**Answer:** Plain `SELECT WHERE` wins at every realistic depth. The recursion isn't the problem — the JSON-expression join inside the CTE is.
## Setup
- Postgres 16 on `localhost:5441` (the `compose-postgres.yml` instance, run directly without docker for this round)
- Single delta channel `messages`, one write per checkpoint, `DELTA_SENTINEL` blob per checkpoint
- Linear chain (`branch=1`) and 5-way branching at every step (`branch=5`) — branching is the case where plain over-fetches sibling rows
- Median of 20 timed runs after 3 warmups, fresh psycopg cursor per strategy
- Bench script: `bench_get_tuple_strategies.py` at repo root
Three strategies compared:
| name | roundtrips | shape |
|------|-----------|-------|
| `cte` | 1 | Current prod: recursive CTE walks ancestors, LEFT JOINs writes + blobs |
| `plain` | 3 | Nuno's suggestion: thread-wide `SELECT WHERE` per table, Python walks parent chain and filters |
| `cte+narrow` | 2 | CTE returns ancestor IDs only, then one `UNION ALL` of writes + blobs filtered by `ANY(ids)` |
## Results (ms per get_tuple, median of 20)
```
depth branch cte plain cte+narrow rows_cte rows_plain plain/cte
10 1 0.14ms 0.26ms 0.24ms 9 30 1.89x
10 5 0.21ms 0.23ms 0.17ms 9 110 1.11x
50 1 0.89ms 0.27ms 0.33ms 49 150 0.31x
50 5 2.35ms 0.66ms 0.51ms 49 550 0.28x
200 1 11.61ms 0.78ms 1.30ms 199 600 0.07x
200 5 34.79ms 2.33ms 3.07ms 199 2200 0.07x
1000 1 274.60ms 2.59ms 13.29ms 999 3000 0.01x
1000 5 856.01ms 10.14ms 15.31ms 999 11000 0.01x
```
Lower is better. `plain/cte < 1` means plain is faster.
### Headline numbers
- depth 50: plain is **3x** faster
- depth 200: plain is **15x** faster
- depth 1000: plain is **~100x** faster
- Branching makes plain over-fetch (3000 rows → 11000 rows at d=1000), but it remains ~85x faster than the CTE
## Why the CTE collapses
`EXPLAIN (ANALYZE, BUFFERS)` of the CTE at depth 1000 (linear). Excerpt with the load-bearing nodes:
```
Sort ... actual time=137.798..137.827 rows=999
CTE ancestors
-> Recursive Union ... actual time=0.005..2.443 rows=999
^^^^^^
recursion is 2.4 ms — fine
-> Nested Loop Left Join ... actual time=2.676..137.529 rows=999
Join Filter: (cw.checkpoint_id = a.cid)
Rows Removed by Join Filter: 998001
^^^^^^^
999 ancestors x ~1000 writes
-> Nested Loop Left Join ... actual time=2.669..85.061 rows=999
Join Filter: (bl.version = ((c.checkpoint -> 'channel_versions'::text) ->> bl.channel))
Rows Removed by Join Filter: 998001
^^^^^^^
same quadratic blow-up on the blob join
```
Two pathological things are happening:
1. **The blob join filter is on a JSON expression**: `bl.version = (c.checkpoint -> 'channel_versions' ->> bl.channel)`. The planner cannot push this into an index lookup, so it materializes `checkpoint_blobs` for the thread and does a nested-loop comparison against every ancestor — a Cartesian product that grows as `O(ancestors × blobs_in_thread)`.
2. **The writes join is similar**: writes for the thread are materialized once, then nested-loop joined against ancestors with a `Join Filter` rather than a hash/merge join over the indexed `checkpoint_id`.
At depth 1000 that's **~2 million rows evaluated, 99.9% of them discarded**. The recursion itself is a rounding error.
For comparison, the plain Q1 (`SELECT … FROM checkpoints WHERE thread_id=? AND checkpoint_ns=?`) at depth 1000:
```
Seq Scan on checkpoints ... actual time=0.012..0.121 rows=1000
Execution Time: 0.140 ms
```
A simple seq scan over 57 buffers. Q2 and Q3 follow the same shape and complete in well under 1 ms each.
## Crossover and remote-DB reasoning
- Pure local Postgres: plain wins from depth ~30 onward; CTE wins by fractions of a ms below that
- Remote Postgres at ~5 ms RTT adds ~10 ms to plain (3 roundtrips vs 1). Crossover shifts to ~depth 30. Above that, the CTE's quadratic SQL cost still dominates the RTT savings.
There is no realistic conversation depth where the CTE wins on a remote DB. At depth 200+ (anything resembling a real multi-turn agent run) plain is faster regardless of network.
## Recommendation
**Switch to plain SELECT WHERE, one delta channel at a time.**
Three indexed queries per delta channel:
```sql
-- Q1: parent chain + per-checkpoint version of this channel
SELECT checkpoint_id,
parent_checkpoint_id,
checkpoint -> 'channel_versions' ->> 'channel_name' AS ver
FROM checkpoints
WHERE thread_id = ? AND checkpoint_ns = ?;
-- Q2: writes for this channel, anywhere in the thread
SELECT checkpoint_id, type, blob, task_id, idx
FROM checkpoint_writes
WHERE thread_id = ? AND checkpoint_ns = ? AND channel = ?;
-- Q3: blobs for this channel, anywhere in the thread
SELECT version, type, blob
FROM checkpoint_blobs
WHERE thread_id = ? AND checkpoint_ns = ? AND channel = ?;
```
Python then:
- Builds `parent_of: dict[cid, parent_cid]` from Q1
- Walks from target's parent newest → oldest
- Filters Q2 rows by `ancestor_set`, processes oldest → newest, applies overwrite-terminator
- Picks seed blob via the per-ancestor `ver` map, terminates at first non-sentinel blob
All O(n) on n = thread checkpoints, with tight constants (dict lookups). No recursion, no JSON-expression joins, no quadratic plans.
If the 3-roundtrip cost ever shows up on remote-DB benchmarks, fold Q2 + Q3 into one `UNION ALL` to get back to 2 roundtrips. Bench says it isn't worth the SQL complexity right now.
## Bonus: code simplification from single-channel scope
Multi-channel reconstruction in the current `_reconstruct_delta_channels_cur` carries:
- `rows_by_cid` nested dicts, keyed by cid then channel
- `seen_blob: set[(cid, channel)]` and `seen_write: set[(cid, channel, task_id, idx)]` dedup
- `collected: dict[channel, list]`, `done: set[channel]`, `seeds: dict[channel, value]`
- Inner `for ch in channels_list` loops and an early-exit `if len(done) == len(channels_list)`
Single-channel collapses these to a single list, a single bool, and one `Optional[Any]`. Roughly half the Python in that function, plus an obvious shape for splitting pure post-processing into `base.py` so sync and async stop duplicating it.
If multi-channel coalescing turns out to matter later, it can come back as a SQL-level optimization without re-introducing the bookkeeping in Python.