The GitHub Actions job "Tests (AMD)" on airflow.git/llm-file-analysis-serialize-output-bug has succeeded. Run started by GitHub user kaxil (triggered by kaxil).
Head commit for run: 0ced070edd1ae84367ec1299cbf5fa327a32138d / Kaxil Naik <[email protected]> common-ai: Honour serialize_output=True on LLMFileAnalysisOperator LLMFileAnalysisOperator overrides execute() rather than calling super().execute(), so it never reached the model_dump() branch that LLMOperator added for the serialize_output=True / older-Airflow fallback. serialize_output=True on this operator silently returned a Pydantic instance instead of the dict the user asked for. Mirror the same isinstance(output, BaseModel) -> model_dump() block in LLMFileAnalysisOperator.execute(). Add a regression test that constructs the operator with serialize_output=True and asserts the return value is a dict, not a BaseModel. Report URL: https://github.com/apache/airflow/actions/runs/26772897606 With regards, GitHub Actions via GitBox --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
