fix(google-generativeai): emit choice events from the last chunk

[chiruu12]

Both streaming builders passed the generator to emit_choice_events after the for item in response loop had already drained it. emit_choice_events reads response.candidates, which a generator does not have, so with events enabled every streamed Gemini call raised AttributeError out of the generator into the caller, and the span was never ended.

The non-events branch three lines below already used last_chunk. This makes the events branch do the same, and skips emitting when the stream produced no chunks at all.

Neither builder is decorated @dont_throw and neither ends the span in a finally, which is why this surfaced in user code rather than a log line. Worth fixing, but out of scope here.

Tests in tests/test_streaming_events.py cover the sync builder, the async builder and the empty stream. All three fail on main.

Fixes #4417

Summary by CodeRabbit

• Bug Fixes

  • Improved synchronous and asynchronous streaming event handling by using the final response chunk.
  • Prevented choice events from being emitted when streaming responses contain no chunks.

• Tests

  • Added coverage for final-chunk event handling in synchronous and asynchronous streams.
  • Added validation for empty streaming responses.

[Copilot]

Copilot was unable to review this pull request because the user who requested the review has reached their quota limit.

[coderabbitai]

📝 Walkthrough

Walkthrough

Synchronous and asynchronous Google Generative AI streaming builders now emit choice events from the final response chunk and skip emission when a stream is empty.

Changes

Google Generative AI streaming events

Layer / File(s)	Summary
Use the final streamed chunk packages/opentelemetry-instrumentation-google-generativeai/opentelemetry/instrumentation/google_generativeai/__init__.py	Synchronous and asynchronous streaming paths now use last_chunk for choice events and avoid emission when no chunk exists.
Validate streaming event behavior packages/opentelemetry-instrumentation-google-generativeai/tests/test_streaming_events.py	Tests cover final-chunk event emission for synchronous and asynchronous streams and verify that empty streams emit no events.

Estimated code review effort: 2 (Simple) | ~10 minutes

Mergeability Score: ⚪ Minimal · up to fb0f8

The change is localized to streamed event emission and avoids the reported failure on empty streams; no actionable merge-blocking risk remains beyond normal checks and review.

Possibly related PRs

• traceloop/openllmetry#4372: Both changes derive streaming events or attributes from final streamed data, but they affect different instrumentation modules.

🚥 Pre-merge checks | ✅ 5

✅ Passed checks (5 passed)

Check name	Status	Explanation
Description Check	✅ Passed	Check skipped - CodeRabbit’s high-level summary is enabled.
Title check	✅ Passed	The title clearly describes the primary fix: emitting Google Generative AI choice events from the final stream chunk.
Linked Issues check	✅ Passed	The changes fix the drained-generator error, use the final chunk for sync and async events, guard empty streams, and add matching tests [#4417].
Out of Scope Changes check	✅ Passed	All changes are limited to Google Generative AI streaming event handling and related tests required by the linked issue.
Docstring Coverage	✅ Passed	No functions found in the changed files to evaluate docstring coverage. Skipping docstring coverage check.

✨ Finishing Touches

🧪 Generate unit tests (beta)

• Create PR with unit tests

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands.}

[coderabbitai]

🧹 Nitpick comments (1)

packages/opentelemetry-instrumentation-google-generativeai/tests/test_streaming_events.py (1)

93-101: 🎯 Functional Correctness | 🔵 Trivial | ⚡ Quick win

Add an empty asynchronous stream test.

The current test covers only the synchronous empty-stream branch. Add the equivalent asynchronous case to protect _abuild_from_streaming_response.

Proposed async empty-stream regression test

+@pytest.mark.asyncio
+async def test_async_streaming_with_no_chunks_does_not_emit(monkeypatch):
+    seen = _streaming_setup(monkeypatch)
+
+    async def agen():
+        for chunk in []:
+            yield chunk
+
+    generator = _abuild_from_streaming_response(
+        MagicMock(), agen(), "gemini-2.0-flash", MagicMock(), None
+    )
+    async for _ in generator:
+        pass
+
+    assert "response" not in seen

🤖 Prompt for AI Agents

Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

In
`@packages/opentelemetry-instrumentation-google-generativeai/tests/test_streaming_events.py`
around lines 93 - 101, Add an asynchronous counterpart to
test_sync_streaming_with_no_chunks_does_not_emit that exercises
_abuild_from_streaming_response with an empty async iterator, consumes the
result, and asserts "response" is absent from seen while reusing the existing
_streaming_setup setup.

🤖 Prompt for all review comments with AI agents

Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

Nitpick comments:
In
`@packages/opentelemetry-instrumentation-google-generativeai/tests/test_streaming_events.py`:
- Around line 93-101: Add an asynchronous counterpart to
test_sync_streaming_with_no_chunks_does_not_emit that exercises
_abuild_from_streaming_response with an empty async iterator, consumes the
result, and asserts "response" is absent from seen while reusing the existing
_streaming_setup setup.

---

ℹ️ Review info

⚙️ Run configuration

Configuration used: defaults

Review profile: CHILL

Plan: Pro Plus

Run ID: 3776eafc-1614-4d1c-a2bb-390b95142320

📥 Commits

Reviewing files that changed from the base of the PR and between 62e24c2 and fb0f81b.

📒 Files selected for processing (2)

• packages/opentelemetry-instrumentation-google-generativeai/opentelemetry/instrumentation/google_generativeai/__init__.py
• packages/opentelemetry-instrumentation-google-generativeai/tests/test_streaming_events.py