Updates

ETB-API/venv/lib/python3.12/site-packages/jsonschema/benchmarks/__init__.py

@@ -0,0 +1,5 @@
+"""
+Benchmarks for validation.
+
+This package is *not* public API.
+"""

ETB-API/venv/lib/python3.12/site-packages/jsonschema/benchmarks/const_vs_enum.py

@@ -0,0 +1,30 @@
+"""
+A benchmark for comparing equivalent validation of `const` and `enum`.
+"""
+
+from pyperf import Runner
+
+from jsonschema import Draft202012Validator
+
+value = [37] * 100
+const_schema = {"const": list(value)}
+enum_schema = {"enum": [list(value)]}
+
+valid = list(value)
+invalid = [*valid, 73]
+
+const = Draft202012Validator(const_schema)
+enum = Draft202012Validator(enum_schema)
+
+assert const.is_valid(valid)
+assert enum.is_valid(valid)
+assert not const.is_valid(invalid)
+assert not enum.is_valid(invalid)
+
+
+if __name__ == "__main__":
+    runner = Runner()
+    runner.bench_func("const valid", lambda: const.is_valid(valid))
+    runner.bench_func("const invalid", lambda: const.is_valid(invalid))
+    runner.bench_func("enum valid", lambda: enum.is_valid(valid))
+    runner.bench_func("enum invalid", lambda: enum.is_valid(invalid))

ETB-API/venv/lib/python3.12/site-packages/jsonschema/benchmarks/contains.py

@@ -0,0 +1,28 @@
+"""
+A benchmark for validation of the `contains` keyword.
+"""
+
+from pyperf import Runner
+
+from jsonschema import Draft202012Validator
+
+schema = {
+    "type": "array",
+    "contains": {"const": 37},
+}
+validator = Draft202012Validator(schema)
+
+size = 1000
+beginning = [37] + [0] * (size - 1)
+middle = [0] * (size // 2) + [37] + [0] * (size // 2)
+end = [0] * (size - 1) + [37]
+invalid = [0] * size
+
+
+if __name__ == "__main__":
+    runner = Runner()
+    runner.bench_func("baseline", lambda: validator.is_valid([]))
+    runner.bench_func("beginning", lambda: validator.is_valid(beginning))
+    runner.bench_func("middle", lambda: validator.is_valid(middle))
+    runner.bench_func("end", lambda: validator.is_valid(end))
+    runner.bench_func("invalid", lambda: validator.is_valid(invalid))

ETB-API/venv/lib/python3.12/site-packages/jsonschema/benchmarks/issue232.py

@@ -0,0 +1,25 @@
+"""
+A performance benchmark using the example from issue #232.
+
+See https://github.com/python-jsonschema/jsonschema/pull/232.
+"""
+from pathlib import Path
+
+from pyperf import Runner
+from referencing import Registry
+
+from jsonschema.tests._suite import Version
+import jsonschema
+
+issue232 = Version(
+    path=Path(__file__).parent / "issue232",
+    remotes=Registry(),
+    name="issue232",
+)
+
+
+if __name__ == "__main__":
+    issue232.benchmark(
+        runner=Runner(),
+        Validator=jsonschema.Draft4Validator,
+    )

ETB-API/venv/lib/python3.12/site-packages/jsonschema/benchmarks/json_schema_test_suite.py

@@ -0,0 +1,12 @@
+"""
+A performance benchmark using the official test suite.
+
+This benchmarks jsonschema using every valid example in the
+JSON-Schema-Test-Suite. It will take some time to complete.
+"""
+from pyperf import Runner
+
+from jsonschema.tests._suite import Suite
+
+if __name__ == "__main__":
+    Suite().benchmark(runner=Runner())

ETB-API/venv/lib/python3.12/site-packages/jsonschema/benchmarks/nested_schemas.py

@@ -0,0 +1,56 @@
+"""
+Validating highly nested schemas shouldn't cause exponential time blowups.
+
+See https://github.com/python-jsonschema/jsonschema/issues/1097.
+"""
+from itertools import cycle
+
+from jsonschema.validators import validator_for
+
+metaschemaish = {
+    "$id": "https://example.com/draft/2020-12/schema/strict",
+    "$schema": "https://json-schema.org/draft/2020-12/schema",
+
+    "$vocabulary": {
+        "https://json-schema.org/draft/2020-12/vocab/core": True,
+        "https://json-schema.org/draft/2020-12/vocab/applicator": True,
+        "https://json-schema.org/draft/2020-12/vocab/unevaluated": True,
+        "https://json-schema.org/draft/2020-12/vocab/validation": True,
+        "https://json-schema.org/draft/2020-12/vocab/meta-data": True,
+        "https://json-schema.org/draft/2020-12/vocab/format-annotation": True,
+        "https://json-schema.org/draft/2020-12/vocab/content": True,
+    },
+    "$dynamicAnchor": "meta",
+
+    "$ref": "https://json-schema.org/draft/2020-12/schema",
+    "unevaluatedProperties": False,
+}
+
+
+def nested_schema(levels):
+    """
+    Produce a schema which validates deeply nested objects and arrays.
+    """
+
+    names = cycle(["foo", "bar", "baz", "quux", "spam", "eggs"])
+    schema = {"type": "object", "properties": {"ham": {"type": "string"}}}
+    for _, name in zip(range(levels - 1), names):
+        schema = {"type": "object", "properties": {name: schema}}
+    return schema
+
+
+validator = validator_for(metaschemaish)(metaschemaish)
+
+if __name__ == "__main__":
+    from pyperf import Runner
+    runner = Runner()
+
+    not_nested = nested_schema(levels=1)
+    runner.bench_func("not nested", lambda: validator.is_valid(not_nested))
+
+    for levels in range(1, 11, 3):
+        schema = nested_schema(levels=levels)
+        runner.bench_func(
+            f"nested * {levels}",
+            lambda schema=schema: validator.is_valid(schema),
+        )

ETB-API/venv/lib/python3.12/site-packages/jsonschema/benchmarks/subcomponents.py

@@ -0,0 +1,42 @@
+"""
+A benchmark which tries to compare the possible slow subparts of validation.
+"""
+from referencing import Registry
+from referencing.jsonschema import DRAFT202012
+from rpds import HashTrieMap, HashTrieSet
+
+from jsonschema import Draft202012Validator
+
+schema = {
+    "type": "array",
+    "minLength": 1,
+    "maxLength": 1,
+    "items": {"type": "integer"},
+}
+
+hmap = HashTrieMap()
+hset = HashTrieSet()
+
+registry = Registry()
+
+v = Draft202012Validator(schema)
+
+
+def registry_data_structures():
+    return hmap.insert("foo", "bar"), hset.insert("foo")
+
+
+def registry_add():
+    resource = DRAFT202012.create_resource(schema)
+    return registry.with_resource(uri="urn:example", resource=resource)
+
+
+if __name__ == "__main__":
+    from pyperf import Runner
+    runner = Runner()
+
+    runner.bench_func("HashMap/HashSet insertion", registry_data_structures)
+    runner.bench_func("Registry insertion", registry_add)
+    runner.bench_func("Success", lambda: v.is_valid([1]))
+    runner.bench_func("Failure", lambda: v.is_valid(["foo"]))
+    runner.bench_func("Metaschema validation", lambda: v.check_schema(schema))

ETB-API/venv/lib/python3.12/site-packages/jsonschema/benchmarks/unused_registry.py

@@ -0,0 +1,35 @@
+"""
+An unused schema registry should not cause slower validation.
+
+"Unused" here means one where no reference resolution is occurring anyhow.
+
+See https://github.com/python-jsonschema/jsonschema/issues/1088.
+"""
+from pyperf import Runner
+from referencing import Registry
+from referencing.jsonschema import DRAFT201909
+
+from jsonschema import Draft201909Validator
+
+registry = Registry().with_resource(
+    "urn:example:foo",
+    DRAFT201909.create_resource({}),
+)
+
+schema = {"$ref": "https://json-schema.org/draft/2019-09/schema"}
+instance = {"maxLength": 4}
+
+no_registry = Draft201909Validator(schema)
+with_useless_registry = Draft201909Validator(schema, registry=registry)
+
+if __name__ == "__main__":
+    runner = Runner()
+
+    runner.bench_func(
+        "no registry",
+        lambda: no_registry.is_valid(instance),
+    )
+    runner.bench_func(
+        "useless registry",
+        lambda: with_useless_registry.is_valid(instance),
+    )

ETB-API/venv/lib/python3.12/site-packages/jsonschema/benchmarks/useless_applicator_schemas.py

@@ -0,0 +1,106 @@
+
+"""
+A benchmark for validation of applicators containing lots of useless schemas.
+
+Signals a small possible optimization to remove all such schemas ahead of time.
+"""
+
+from pyperf import Runner
+
+from jsonschema import Draft202012Validator as Validator
+
+NUM_USELESS = 100000
+
+subschema = {"const": 37}
+
+valid = 37
+invalid = 12
+
+baseline = Validator(subschema)
+
+
+# These should be indistinguishable from just `subschema`
+by_name = {
+    "single subschema": {
+        "anyOf": Validator({"anyOf": [subschema]}),
+        "allOf": Validator({"allOf": [subschema]}),
+        "oneOf": Validator({"oneOf": [subschema]}),
+    },
+    "redundant subschemas": {
+        "anyOf": Validator({"anyOf": [subschema] * NUM_USELESS}),
+        "allOf": Validator({"allOf": [subschema] * NUM_USELESS}),
+    },
+    "useless successful subschemas (beginning)": {
+        "anyOf": Validator({"anyOf": [subschema, *[True] * NUM_USELESS]}),
+        "allOf": Validator({"allOf": [subschema, *[True] * NUM_USELESS]}),
+    },
+    "useless successful subschemas (middle)": {
+        "anyOf": Validator(
+            {
+                "anyOf": [
+                    *[True] * (NUM_USELESS // 2),
+                    subschema,
+                    *[True] * (NUM_USELESS // 2),
+                ],
+            },
+        ),
+        "allOf": Validator(
+            {
+                "allOf": [
+                    *[True] * (NUM_USELESS // 2),
+                    subschema,
+                    *[True] * (NUM_USELESS // 2),
+                ],
+            },
+        ),
+    },
+    "useless successful subschemas (end)": {
+        "anyOf": Validator({"anyOf": [*[True] * NUM_USELESS, subschema]}),
+        "allOf": Validator({"allOf": [*[True] * NUM_USELESS, subschema]}),
+    },
+    "useless failing subschemas (beginning)": {
+        "anyOf": Validator({"anyOf": [subschema, *[False] * NUM_USELESS]}),
+        "oneOf": Validator({"oneOf": [subschema, *[False] * NUM_USELESS]}),
+    },
+    "useless failing subschemas (middle)": {
+        "anyOf": Validator(
+            {
+                "anyOf": [
+                    *[False] * (NUM_USELESS // 2),
+                    subschema,
+                    *[False] * (NUM_USELESS // 2),
+                ],
+            },
+        ),
+        "oneOf": Validator(
+            {
+                "oneOf": [
+                    *[False] * (NUM_USELESS // 2),
+                    subschema,
+                    *[False] * (NUM_USELESS // 2),
+                ],
+            },
+        ),
+    },
+    "useless failing subschemas (end)": {
+        "anyOf": Validator({"anyOf": [*[False] * NUM_USELESS, subschema]}),
+        "oneOf": Validator({"oneOf": [*[False] * NUM_USELESS, subschema]}),
+    },
+}
+
+if __name__ == "__main__":
+    runner = Runner()
+
+    runner.bench_func("baseline valid", lambda: baseline.is_valid(valid))
+    runner.bench_func("baseline invalid", lambda: baseline.is_valid(invalid))
+
+    for group, applicators in by_name.items():
+        for applicator, validator in applicators.items():
+            runner.bench_func(
+                f"{group}: {applicator} valid",
+                lambda validator=validator: validator.is_valid(valid),
+            )
+            runner.bench_func(
+                f"{group}: {applicator} invalid",
+                lambda validator=validator: validator.is_valid(invalid),
+            )

ETB-API/venv/lib/python3.12/site-packages/jsonschema/benchmarks/useless_keywords.py

@@ -0,0 +1,32 @@
+"""
+A benchmark for validation of schemas containing lots of useless keywords.
+
+Checks we filter them out once, ahead of time.
+"""
+
+from pyperf import Runner
+
+from jsonschema import Draft202012Validator
+
+NUM_USELESS = 100000
+schema = dict(
+    [
+        ("not", {"const": 42}),
+        *((str(i), i) for i in range(NUM_USELESS)),
+        ("type", "integer"),
+        *((str(i), i) for i in range(NUM_USELESS, NUM_USELESS)),
+        ("minimum", 37),
+    ],
+)
+validator = Draft202012Validator(schema)
+
+valid = 3737
+invalid = 12
+
+
+if __name__ == "__main__":
+    runner = Runner()
+    runner.bench_func("beginning of schema", lambda: validator.is_valid(42))
+    runner.bench_func("middle of schema", lambda: validator.is_valid("foo"))
+    runner.bench_func("end of schema", lambda: validator.is_valid(12))
+    runner.bench_func("valid", lambda: validator.is_valid(3737))

ETB-API/venv/lib/python3.12/site-packages/jsonschema/benchmarks/validator_creation.py

@@ -0,0 +1,14 @@
+from pyperf import Runner
+
+from jsonschema import Draft202012Validator
+
+schema = {
+    "type": "array",
+    "minLength": 1,
+    "maxLength": 1,
+    "items": {"type": "integer"},
+}
+
+
+if __name__ == "__main__":
+    Runner().bench_func("validator creation", Draft202012Validator, schema)