atiaomar1978-hub commented on PR #26148: URL: https://github.com/apache/camel/pull/26148#issuecomment-5562648345
## Review follow-up (Bugbot + Grok) _AI-generated on behalf of atiaomar1978-hub_ We ran automated review on the CAMEL-23395 branch before opening this upstream PR. Summary of findings and what was addressed: ### Bugbot | Finding | Resolution | |---|---| | Trace missing on failure paths — token budget / max iterations left `CamelOpenAIAgenticTrace` unset and no `LoopCompleted` event | Fixed: `try/finally` in `OpenAIProducer.processNonStreamingAgentic()` calls `finalizeObservability()` so trace + terminal event are always published | **Bugbot re-run after fixes:** no bugs found. ### Grok (senior committer-style review) | Finding | Resolution | |---|---| | Always publish trace + terminal event on failure | ✅ `finalizeObservability()` in `finally` block | | Record LLM round before token-budget abort | ✅ `recordFinalIteration()` before throw | | Duplicate / ambiguous iteration ids in trace | ✅ unique 1-based `modelCall` sequence per LLM call (`d926c88eb96`) | | Do not implement `CamelEvent.ExchangeEvent` | ✅ `AbstractOpenAIExchangeEvent` implements `CamelEvent` only (Custom type) | | Move `AGENTIC_TRACE` metadata to properties section | ✅ relocated in `OpenAIConstants` | | Add `toString()` on events | ✅ added to all three event types | | Skip `notify()` when no event notifiers; log failures | ✅ early return + debug log on notify failure | | Parallel timeout `durationMs = 0` | ✅ `timeoutResult()` reports configured timeout | | `stopReason` inferred from exception message text | ✅ `token_budget_exceeded` set at throw site | | Failure-path test coverage | ✅ token budget + max iterations in `OpenAIAgenticTraceTest` / `OpenAIAgenticEventNotifierTest` | ### Deferred (suggested follow-up, not blocking this PR) - `asJSon()` on events for Event console structured details (CAMEL-24377 alignment) - Trace rows when `FAIL_EXCHANGE` throws mid tool batch - Opt-out producer flag for trace collection - Dedicated tests for `returnDirect` trace shape and `success=false` tool paths ### Verification ```bash ./mvnw -pl components/camel-ai/camel-openai test -Dtest=OpenAIAgentic* ``` All `OpenAIAgentic*` tests pass locally. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
