WIP squash me or drop, get pydantic check working

Author: aabmass

util/opentelemetry-util-genai/pyproject.toml

@@ -34,7 +34,7 @@ dependencies = [
 upload = "opentelemetry.util.genai._upload:upload_completion_hook"
 
 [project.optional-dependencies]
-test = ["pytest>=7.0.0"]
+test = ["pytest>=7.0.0", "pydantic>=2.0"]
 upload = ["fsspec>=2025.9.0"]
 
 [project.urls]

util/opentelemetry-util-genai/src/opentelemetry/util/genai/utils.py

@@ -17,8 +17,9 @@
 import logging
 import os
 from base64 import b64encode
+from collections.abc import Iterator
 from functools import partial
-from typing import Any
+from typing import Any, Protocol, runtime_checkable
 
 from opentelemetry.instrumentation._semconv import (
     _OpenTelemetrySemanticConventionStability,
@@ -65,11 +66,24 @@ def get_content_capturing_mode() -> ContentCapturingMode:
         return ContentCapturingMode.NO_CONTENT
 
 
+@runtime_checkable
+class _PydanticModelDumpable(Protocol):
+    """Checkable protocol for pydantic model dump-able object.
+
+    See https://docs.pydantic.dev/latest/api/base_model/#pydantic.BaseModel.model_dump
+    """
+
+    def model_dump_json(self, **kwargs: Any) -> str: ...
+
+
 class _GenAiJsonEncoder(json.JSONEncoder):
+    def encode(self, o: Any) -> str:
+        return super().encode(o)
+
     def default(self, o: Any) -> Any:
         if isinstance(o, bytes):
             return b64encode(o).decode()
-        elif isinstance(o, (datetime.datetime, datetime.date)):
+        if isinstance(o, (datetime.datetime, datetime.date)):
             return o.isoformat()
 
         try:

util/opentelemetry-util-genai/test-requirements.txt

@@ -1,3 +1,4 @@
 pytest==7.4.4
 fsspec==2025.9.0
+pydantic==2.11.4
 -e opentelemetry-instrumentation

util/opentelemetry-util-genai/tests/test_utils.py

@@ -12,10 +12,14 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
+import datetime
+import io
 import json
 import os
 import unittest
-from unittest.mock import patch
+from unittest.mock import Mock, patch
+
+from pydantic import BaseModel
 
 from opentelemetry import trace
 from opentelemetry.instrumentation._semconv import (
@@ -42,7 +46,11 @@
     OutputMessage,
     Text,
 )
-from opentelemetry.util.genai.utils import get_content_capturing_mode
+from opentelemetry.util.genai.utils import (
+    gen_ai_json_dump,
+    gen_ai_json_dumps,
+    get_content_capturing_mode,
+)
 
 
 def patch_env_vars(stability_mode, content_capturing):
@@ -70,7 +78,9 @@ class TestVersion(unittest.TestCase):
         stability_mode="gen_ai_latest_experimental",
         content_capturing="SPAN_ONLY",
     )
-    def test_get_content_capturing_mode_parses_valid_envvar(self):  # pylint: disable=no-self-use
+    def test_get_content_capturing_mode_parses_valid_envvar(
+        self,
+    ):  # pylint: disable=no-self-use
         assert get_content_capturing_mode() == ContentCapturingMode.SPAN_ONLY
 
     @patch_env_vars(
@@ -287,3 +297,68 @@ class BoomError(RuntimeError):
         assert span.start_time is not None
         assert span.end_time is not None
         assert span.end_time >= span.start_time
+
+
+class TestGenAiJson(unittest.TestCase):
+    def test_gen_ai_json_dumps(self):
+        class Unserializable:
+            # pylint: disable=no-self-use
+            def __str__(self):
+                return "unserializable"
+
+        obj = {
+            "bytes": b"test",
+            "datetime": datetime.datetime(2023, 1, 1, 12, 0, 0),
+            "date": datetime.date(2023, 1, 1),
+            "unserializable": Unserializable(),
+        }
+        result = gen_ai_json_dumps(obj)
+        expected = '{"bytes":"dGVzdA==","datetime":"2023-01-01T12:00:00","date":"2023-01-01","unserializable":"unserializable"}'
+        self.assertEqual(result, expected)
+
+    def test_gen_ai_json_dumps_pydantic(self):
+        class ExamplePydanticModel(BaseModel):
+            datetime_field: datetime.datetime
+            string_field: str
+
+        pydantic_model = ExamplePydanticModel(
+            datetime_field=datetime.datetime(2023, 1, 1, 12, 0, 0),
+            string_field="test",
+        )
+        # spy the model
+        pydantic_model = Mock(wraps=pydantic_model)
+
+        result = gen_ai_json_dumps({"some": {"pydantic": [pydantic_model]}})
+        self.assertEqual(result, '{"key":"pydantic"}')
+
+        pydantic_model.model_dump_json.assert_called_once()
+
+    def test_gen_ai_json_dump(self):
+        class Unserializable:
+            # pylint: disable=no-self-use
+            def __str__(self):
+                return "unserializable"
+
+        obj = {
+            "bytes": b"test",
+            "datetime": datetime.datetime(2023, 1, 1, 12, 0, 0),
+            "date": datetime.date(2023, 1, 1),
+            "unserializable": Unserializable(),
+        }
+        string_io = io.StringIO()
+        gen_ai_json_dump(obj, string_io)
+        result = string_io.getvalue()
+        expected = '{"bytes":"dGVzdA==","datetime":"2023-01-01T12:00:00","date":"2023-01-01","unserializable":"unserializable"}'
+        self.assertEqual(result, expected)
+
+    def test_gen_ai_json_dump_pydantic(self):
+        class PydanticModel:
+            # pylint: disable=no-self-use
+            def model_dump_json(self, **kwargs):
+                return '{"key":"pydantic"}'
+
+        obj = PydanticModel()
+        string_io = io.StringIO()
+        gen_ai_json_dump(obj, string_io)
+        result = string_io.getvalue()
+        self.assertEqual(result, '{"key":"pydantic"}')