External services consuming the resource server API may need items they
create to carry specific attributes, typically quota_excluded so their
uploads do not consume the user quota.
The new EXTERNAL_API_AUD_ITEM_ATTRIBUTES setting maps a token audience
to the attributes applied at creation. The lookup is a hook on the item
viewset overridden only in the resource server viewset, so both the
root create and children creation paths are covered while the regular
API remains unaffected.
Sharing a folder with many contacts one call at a time is slow and can
leave a partial state when one of them fails. This endpoint accepts a
list of email/role rows, validates them all before any write, then
creates accesses for known users and invitations for unknown emails.
Rows targeting users already covered by an equal or higher role are
skipped and reported so the client can inform the user.
The feature is disabled by default and gated by the new
ALLOW_SHARE_IMPORT_FILE setting exposed in the config endpoint.
Deployments without a DeployCenter service had no way to enforce
storage quotas. This backend computes each user's usage locally and
applies a configurable default limit, overridable per user in the
admin (0 meaning unlimited) and skippable for users created before a
cutoff so quotas can be rolled out to new users only. The quota is
soft: can_upload runs before the file size is known, so one upload
can overshoot before the next one is blocked.
An empty AWS_S3_UPLOAD_ACL in an env file is easy to mistake for
an unset variable, and unset must keep meaning "private". The
explicit "default" value states the intent: let the bucket's
default object ACL apply. The frontend translates it where the
config is read, so the drivers keep sending exactly the ACL they
are given.
GCS based storages like S3NS reject presigned PUT URLs signing the
x-amz-acl header: their S3 compatibility layer consumes x-amz-*
headers having an x-goog-* equivalent, so the signature check no
longer finds the header in the request. Objects are private by
default on S3 and GCS, so the ACL can simply be dropped by setting
AWS_S3_UPLOAD_ACL to an empty string. The setting is exposed in the
config endpoint so the frontend can send a header matching the
signature.
See https://documentation.s3ns.fr/storage/docs/aws-simple-migration
Add a dict-shaped config exposed to the frontend so operators can
enable or disable entitlement disclaimers per reason without code
changes. Also pin the Test settings to the local entitlements backend
so test runs do not depend on external state.
Add account_type=organization support to the usage metrics
endpoint. When requested, the API filters users by a shared
OIDC claim (account_id_key/account_id_value) and returns
aggregated storage across all matching users.
Add handling of account_id_key/account_id_value pair, allowing
filtering on allowed parameters in a more generic way. At the
moment `account_email` filter is still needed.
Also include the context of the entitlement in the /entitlements
response.
Restructure the entitlements module from flat files into a
backends/ package with a factory pattern. This enables
constructor kwargs injection for backend configuration.
Rename ANCT backend to DeployCenter to match the actual
service name, and rename Dummy backend to Static for clarity.
The factory now passes ENTITLEMENTS_BACKEND_PARAMETERS to
the backend constructor, making configuration more explicit
and testable.
Some request parameters get updated because of changes on
deploy center side that need to be reflected in services.
Share invitation emails embedded `Site.domain` directly, producing
schemeless links that mail clients failed to resolve. Introduce a
`DJANGO_EMAIL_URL_APP` environment variable holding the app's
absolute URL used when building invitation links, with a fallback
on the current Site domain when unset. Wire it in dev env, helm
dev values, the helm example and the env reference.
The invitation validity duration was hardcoded to 7 days. Expose it as
an INVITATION_VALIDITY_DURATION env var so operators can tune it per
deployment without patching settings. Default remains 7 days.
Add an architecture doc covering the rationale behind the selection
store refactor: why React Context did not scale to 200+ rows during a
marquee drag, why useSyncExternalStore with per-id listeners enables
automatic bailout, and how global vs embedded scopes are wired.
Lives under docs/architecture/ for long-term reference.
An item is considered hard deleted in two cases:
- hard deletion was forced by the user while sitting in the trashbin
- the item was only soft deleted but the trashbin retention period
has expired
Thus, a hard deleted item still exists in database with an associated
file in S3. What the asynchronous task is doing is another action:
permanently removing the item from database and its associated file
from S3. I propose to rename this action "purge".
We refactor the task to improve its resilience:
- don't recursively generate asynchronous tasks. We are already
outside of a request cycle so we don't need to spawn new
asynchronous tasks.
- purge the subtree from the bottom to the top: leaves first,
root last.
- fail fast if something goes wrong and make sure the task will
continue when we retry it.
The new CSS and JS injection features needed documentation.
We merged the old theming doc into a broader customization
guide covering runtime CSS, JS injection, and footer config.
The mirror_file task is reponsible to copy an object from a bucket to an
other one. The buckets can be in different location, in different cloud
providers so we have to download it to reupload it. The `copy` method on
the s3 client only work when the bucket are in the cloud provider infra,
like for AWS.
Add a new configuration option to define an external home URL. When set,
the frontend will redirect users to this URL instead of displaying the
default home page. This allows deployments to use a custom landing page.
The celery beat configuration was not working correctly, we installed
the database scheduler but we didn't configure it correctly. We prefer
removing this scheduler to use the default one. Instead, the crontab
part can be configured using environment variable. By default it will be
run every night at 3AM.
We want to add documentation showing how to use DS_Proxy with Drive.
With proxy is fully optionnal and is here if you want to an encryption
layer between Drive and the object storage.
Allong the allowed extension list we added an allowed mimetype list. We
combine both to ensure that the user is allowed to make the upload. If
the mimetype is not allowed, then the file and the item are deleted.
We want to limit what type of file can be uploaded. For this we use an
allowed list of file extension. This allowed list can be disabled using
a setting and this list is configurable using a setting. When enabled,
file with no extension and hidden file are also rejected.
Add setting FEATURES_INDEXED_SEARCH that allows to disable the
search of indexed files while using the indexation tools.
Signed-off-by: Fabre Florian <ffabre@hybird.org>
Use SEARCH_INDEXER_CONTENT_MAX_SIZE as limit (in bytes) for the file content.
Fix default configuration of OIDC_STORE_ACCESS_TOKEN
Signed-off-by: Fabre Florian <ffabre@hybird.org>
Use nb_results instead of page/page_size argument for /search API.
Add --batch-size argument to the index command.
Fix an issue in SearchIndexer.has_text when item.mimetype is empty.
Signed-off-by: Fabre Florian <ffabre@hybird.org>
Set SEARCH_INDEXER_CLASS=None as default configuration for dev.
Add documentation for Find service setup.
Signed-off-by: Fabre Florian <ffabre@hybird.org>
We have some instances that requires more open resource server features.
This change adds a setting that allows to cherry-pick
which api routes needs to be available or not.
The pagination class Viewsets.Pagination has the max_page_size set to
200. We want to be able to configure this value using a setting. A new
one MAX_PAGE_SIZE is created for this. By default its value is 200 to
not change the current behavior.
We support installation on kubernets. This dopcumentation is here to
show how to deploy on a developement cluster in order to understand what
is needed to do it later in production.
We need to adapt the frontend code to use the PUT method to upload a
file instead of the POST method.
The headers used to sign the url must be added to the request sent. For
now the X-ams-acl header is used.
Nathan Vasse