fix(openai): report request attributes in responses API instrumentation #3471
Conversation
WalkthroughIntroduced a new helper function Changes
Sequence Diagram(s)sequenceDiagram
participant Caller
participant Wrapper as Wrapper<br/>(sync/async)
participant Helper as prepare_kwargs_for_shared_attributes
participant SetAttrs as _set_request_attributes
participant Span
participant SharedModule as Shared Module
Caller->>Wrapper: Invoke wrapper
Wrapper->>Helper: prepare_kwargs_for_shared_attributes(kwargs)
Helper->>Helper: Clone kwargs,<br/>remap max_output_tokens→max_tokens
Helper-->>Wrapper: Return prepared kwargs
Wrapper->>SetAttrs: _set_request_attributes(span,<br/>prepared_kwargs, instance)
SetAttrs->>SharedModule: Extract model name via<br/>_extract_model_name_from_provider_format
SetAttrs->>Span: Set GEN_AI_RESPONSE_MODEL
SetAttrs->>Span: Set GEN_AI_REQUEST_* attributes<br/>(temperature, max_tokens, top_p)
Wrapper->>Wrapper: Execute OpenAI request
Wrapper->>Span: Set usage metrics<br/>(tokens, reasoning)
Wrapper-->>Caller: Return response
Estimated code review effort🎯 3 (Moderate) | ⏱️ ~20 minutes
Possibly related PRs
Suggested reviewers
Poem
Pre-merge checks and finishing touches❌ Failed checks (1 warning)
✅ Passed checks (2 passed)
✨ Finishing touches
🧪 Generate unit tests (beta)
📜 Recent review detailsConfiguration used: CodeRabbit UI Review profile: CHILL Plan: Pro Disabled knowledge base sources:
📒 Files selected for processing (3)
🧰 Additional context used📓 Path-based instructions (2)**/cassettes/**/*.{yaml,yml,json}📄 CodeRabbit inference engine (CLAUDE.md)
Files:
**/*.py📄 CodeRabbit inference engine (CLAUDE.md)
Files:
🧬 Code graph analysis (1)packages/opentelemetry-instrumentation-openai/tests/traces/test_responses.py (2)
🪛 Ruff (0.14.5)packages/opentelemetry-instrumentation-openai/tests/traces/test_responses.py34-34: Unused function argument: (ARG001) ⏰ Context from checks skipped due to timeout of 90000ms. You can increase the timeout in your CodeRabbit configuration to a maximum of 15 minutes (900000ms). (6)
🔇 Additional comments (6)
Comment |
There was a problem hiding this comment.
Important
Looks good to me! 👍
Reviewed everything up to c739dad in 2 minutes and 3 seconds. Click for details.
- Reviewed
261lines of code in3files - Skipped
0files when reviewing. - Skipped posting
6draft comments. View those below. - Modify your settings and rules to customize what types of comments Ellipsis leaves. And don't forget to react with 👍 or 👎 to teach Ellipsis.
1. packages/opentelemetry-instrumentation-openai/opentelemetry/instrumentation/openai/v1/responses_wrappers.py:189
- Draft comment:
The new function prepare_kwargs_for_shared_attributes maps only 'max_output_tokens' to 'max_tokens'. Consider documenting or expanding this mapping if additional parameters need conversion. - Reason this comment was not posted:
Confidence changes required:50%<= threshold50%None
2. packages/opentelemetry-instrumentation-openai/opentelemetry/instrumentation/openai/v1/responses_wrappers.py:208
- Draft comment:
The explicit setting of the 'gen_ai.system' (and previously 'gen_ai.request.model') attribute in set_data_attributes has been removed. Ensure that these required attributes are now set via _set_request_attributes in the wrappers to avoid regression. - Reason this comment was not posted:
Comment looked like it was already resolved.
3. packages/opentelemetry-instrumentation-openai/opentelemetry/instrumentation/openai/v1/responses_wrappers.py:445
- Draft comment:
Multiple wrappers (responses_get_or_create, async_responses_get_or_create, cancel wrappers) call _set_request_attributes with prepare_kwargs_for_shared_attributes. This duplication is acceptable, but consider consolidating the call into a helper if additional shared logic emerges. - Reason this comment was not posted:
Comment was not on a location in the diff, so it can't be submitted as a review comment.
4. packages/opentelemetry-instrumentation-openai/tests/traces/cassettes/test_responses/test_responses_with_request_params.yaml:1
- Draft comment:
New cassette file correctly includes a test case for capturing request parameters. Ensure that any changes in response serialization are kept in sync with expected binary responses. - Reason this comment was not posted:
Confidence changes required:50%<= threshold50%None
5. packages/opentelemetry-instrumentation-openai/tests/traces/test_responses.py:32
- Draft comment:
The test 'test_responses_with_request_params' properly verifies that parameters (temperature, max_tokens, top_p) are reported. Confirm that _set_request_attributes in the shared logic processes these keys as expected. - Reason this comment was not posted:
Confidence changes required:30%<= threshold50%None
6. packages/opentelemetry-instrumentation-openai/tests/traces/test_responses.py:465
- Draft comment:
In test_response_stream_init_with_none_tools, the final assertion allows tools to be an empty list or None. Since the comment suggests tools should default to an empty list, consider enforcing a consistent default instead of allowing None. - Reason this comment was not posted:
Confidence changes required:50%<= threshold50%None
Workflow ID: wflow_ZqsVlCWBWqPPxnkv
You can customize by changing your verbosity settings, reacting with 👍 or 👎, replying to comments, or adding code review rules.
Important
Enhance OpenAI API instrumentation by adding request attribute reporting and tests for capturing request parameters.
_set_request_attributestoresponses_get_or_create_wrapper,async_responses_get_or_create_wrapper, andresponses_cancel_wrapperto report request attributes.prepare_kwargs_for_shared_attributes()to mapmax_output_tokenstomax_tokens._extract_model_name_from_provider_formatand_set_request_attributestoresponses_wrappers.py.test_responses_with_request_paramsintest_responses.pyto verify capturing of request parameters liketemperature,max_tokens, andtop_p.test_responses_with_request_params.yamlcassette for VCR testing.This description was created by
for c739dad. You can customize this summary. It will automatically update as commits are pushed.
Summary by CodeRabbit
✏️ Tip: You can customize this high-level summary in your review settings.