fix(telemetry): serialize Pydantic BaseModel in _safe_json_serialize

[giulio-leone]

Problem

_safe_json_serialize silently replaces Pydantic BaseModel instances with the string '<not serializable>' when serializing tool responses for tracing spans.

Any tool that returns a Pydantic model rather than a plain dict executes correctly, but the traced output is lost:

tool_response = {'result': SearchResults(query='test', total=2, items=['doc1', 'doc2'])}
_safe_json_serialize(tool_response)
# Output: {"result": "<not serializable>"}
# Expected: {"result": {"query": "test", "total": 2, "items": ["doc1", "doc2"]}}

Root Cause

The default lambda in json.dumps unconditionally returns '<not serializable>' for any object that isn't natively JSON-serializable, including Pydantic models.

Fix

Replace the generic lambda with a _default function that calls model_dump(mode='json') for BaseModel instances before falling back to '<not serializable>'. BaseModel is already imported in tracing.py (line 63).

Test

56 telemetry tests pass, 2 consecutive clean runs.

Fixes #4629

[gemini-code-assist]

Summary of Changes

Hello, I'm Gemini Code Assist¹! I'm currently reviewing this pull request and will post my feedback shortly. In the meantime, here's a summary to help you and other reviewers quickly get up to speed!

This pull request resolves a critical data serialization problem within the telemetry system. Previously, Pydantic BaseModel objects were not properly converted to JSON, resulting in their replacement with a generic non-serializable string. The change introduces a custom default serialization handler that intelligently processes Pydantic models, thereby preserving valuable tracing information and enhancing the reliability of telemetry data.

Highlights

• Pydantic BaseModel Serialization Fix: Addressed an issue where Pydantic BaseModel instances were not correctly serialized by _safe_json_serialize, leading to data loss in tracing spans.
• Improved Telemetry Data: Ensured that tool responses containing Pydantic models are now fully serialized, providing complete and accurate data for telemetry and tracing.

🧠 New Feature in Public Preview: You can now enable Memory to help Gemini Code Assist learn from your team's feedback. This makes future code reviews more consistent and personalized to your project's style. Click here to enable Memory in your admin console.

Changelog

• src/google/adk/telemetry/tracing.py
  • Modified the _safe_json_serialize function to correctly handle Pydantic BaseModel instances during JSON serialization.

Activity

• No specific activity (comments, reviews, etc.) was provided in the context for this pull request.

Using Gemini Code Assist

The full guide for Gemini Code Assist can be found on our documentation page, here are some quick tips.

Invoking Gemini

You can request assistance from Gemini at any point by creating a comment using either /gemini <command> or @gemini-code-assist <command>. Below is a summary of the supported commands on the current page.

Feature	Command	Description
Code Review	/gemini review	Performs a code review for the current pull request in its current state.
Pull Request Summary	/gemini summary	Provides a summary of the current pull request in its current state.
Comment	@gemini-code-assist	Responds in comments when explicitly tagged, both in pull request comments and review comments.
Help	/gemini help	Displays a list of available commands.

Customization

To customize Gemini Code Assist for GitHub experience, repository maintainers can create a configuration file and/or provide a custom code review style guide (such as PEP-8 for Python) by creating and adding files to a .gemini/ folder in the base of the repository. Detailed instructions can be found here.

Limitations & Feedback

Gemini Code Assist may make mistakes. Please leave feedback on any instances where its feedback is incorrect or counter productive. You can react with 👍 and 👎 on @gemini-code-assist comments. If you're interested in giving your feedback about your experience with Gemini Code Assist for Github and other Google products, sign up here.

You can also get AI-powered code generation, chat, as well as code reviews directly in the IDE at no cost with the Gemini Code Assist IDE Extension.

Footnotes

1. Review the Privacy Notices, Generative AI Prohibited Use Policy, Terms of Service, and learn how to configure Gemini Code Assist in GitHub here. Gemini can make mistakes, so double check it and use code with caution. ↩

[google-cla]

Thanks for your pull request! It looks like this may be your first contribution to a Google open source project. Before we can look at your pull request, you'll need to sign a Contributor License Agreement (CLA).

View this failed invocation of the CLA check for more information.

For the most up to date status, view the checks section at the bottom of the pull request.

[adk-bot]

Response from ADK Triaging Agent

Hello @giulio-leone, thank you for creating this PR!

Before we can merge this, you'll need to sign the Contributor License Agreement (CLA). You can find more information in the "Details" link of the cla/google check at the bottom of the PR.

Also, could you please add a testing plan section to your PR description to describe how you tested your changes? This will help reviewers to review your PR more efficiently.

Thanks!

[gemini-code-assist]

Code Review

This pull request correctly addresses the issue of serializing Pydantic BaseModel instances in _safe_json_serialize by introducing a custom default handler for json.dumps. The fix is well-targeted and effective.

My main feedback is that a very similar function, _safe_json_serialize_no_whitespaces in src/google/adk/telemetry/_experimental_semconv.py, has the same serialization problem but was not updated. To ensure consistency and prevent future bugs, I recommend applying the same fix there. To avoid code duplication, the new default handler logic could be extracted into a reusable, module-level function. I've left a specific comment with a suggestion on how to refactor this.

[gemini-code-assist]

This correctly handles Pydantic model serialization. To improve reusability and fix a similar issue in another file, I suggest extracting the _default logic into a module-level function, for example _pydantic_json_default.

# At module level in tracing.py
def _pydantic_json_default(o: Any) -> Any:
  if isinstance(o, BaseModel):
    return o.model_dump(mode='json')
  return '<not serializable>'

A nearly identical function, _safe_json_serialize_no_whitespaces in src/google/adk/telemetry/_experimental_semconv.py, still uses the old lambda and will also need this fix. Making the helper function module-level will allow you to import and reuse it there, avoiding code duplication.

Suggested change

	def _default(o: Any) -> Any:
	if isinstance(o, BaseModel):
	return o.model_dump(mode='json')
	return '<not serializable>'
	try:
	# Try direct JSON serialization first
	return json.dumps(
	obj, ensure_ascii=False, default=lambda o: '<not serializable>'
	)
	return json.dumps(obj, ensure_ascii=False, default=_default)
	try:
	# Try direct JSON serialization first
	return json.dumps(obj, ensure_ascii=False, default=_pydantic_json_default)

[giulio-leone]

Closing — CLA not yet signed. Will resubmit when ready.