Merge pull request #1537 from volatilityfoundation/issues/improve-jsonschema-speed

Issues/improve jsonschema speed

pyproject.toml

@@ -1,7 +1,15 @@
 [project]
 name = "volatility3"
 description = "Memory forensics framework"
-keywords = ["volatility", "memory", "forensics", "framework", "windows", "linux", "volshell"]
+keywords = [
+    "volatility",
+    "memory",
+    "forensics",
+    "framework",
+    "windows",
+    "linux",
+    "volshell",
+]
 readme = "README.md"
 authors = [
     { name = "Volatility Foundation", email = "volatility@volatilityfoundation.org" },
@@ -10,9 +18,7 @@ requires-python = ">=3.8.0"
 license = { text = "VSL" }
 dynamic = ["version"]
 
-dependencies = [
-    "pefile>=2024.8.26",
-]
+dependencies = ["pefile>=2024.8.26"]
 
 [project.optional-dependencies]
 full = [
@@ -26,10 +32,7 @@ full = [
     "pillow>=10.0.0,<11.0.0",
 ]
 
-cloud = [
-    "gcsfs>=2024.10.0",
-    "s3fs>=2024.10.0",
-]
+cloud = ["gcsfs>=2024.10.0", "s3fs>=2024.10.0"]
 
 dev = [
     "volatility3[full,cloud]",
@@ -79,16 +82,16 @@ target-version = "py38"
 
 [tool.ruff.lint]
 select = [
-	"F",       # pyflakes
-	"E",       # pycodestyle errors
-	"W",       # pycodestyle warnings
-	"G",       # flake8-logging-format
-	"PIE",     # flake8-pie
-	"UP",      # pyupgrade
+    "F",   # pyflakes
+    "E",   # pycodestyle errors
+    "W",   # pycodestyle warnings
+    "G",   # flake8-logging-format
+    "PIE", # flake8-pie
+    "UP",  # pyupgrade
 ]
 
 ignore = [
-    "E501",    # ignore due to conflict with formatter
+    "E501", # ignore due to conflict with formatter
 ]
 
 [build-system]

volatility3/schemas/__init__.py

@@ -14,6 +14,8 @@ vollog = logging.getLogger(__name__)
 
 cached_validation_filepath = os.path.join(constants.CACHE_PATH, "valid_isf.hashcache")
 
+validators = {}
+
 
 def load_cached_validations() -> Set[str]:
     """Loads up the list of successfully cached json objects, so we don't need
@@ -93,6 +95,13 @@ def valid(
         return True
     try:
         import jsonschema
+
+        schema_key = json.dumps(schema, sort_keys=True)
+        if schema_key not in validators:
+            validator_class = jsonschema.validators.validator_for(schema)
+            validator_class.check_schema(schema)
+            validator = validator_class(schema)
+            validators[schema_key] = validator
     except ImportError:
         vollog.info("Dependency for validation unavailable: jsonschema")
         vollog.debug("All validations will report success, even with malformed input")
@@ -100,7 +109,7 @@ def valid(
 
     try:
         vollog.debug("Validating JSON against schema...")
-        jsonschema.validate(input, schema)
+        validators[schema_key].validate(input)
         cached_validations.add(input_hash)
         vollog.debug("JSON validated against schema (result cached)")
     except jsonschema.exceptions.SchemaError: