guan404ming opened a new pull request, #71051:
URL: https://github.com/apache/airflow/pull/71051

   ## Why
   
   - `LLMSchemaCompareOperator` was the last LLM operator still rejecting 
`require_approval=True` (left over from #70069), so its comparison results 
could not be human-gated.
   
   ## How
   
   - Drop the rejection; pause via `LLMApprovalMixin` after the comparison, 
with a review body showing the compatibility verdict, mismatch severity counts, 
and the full result JSON.
   - Force `output_type=SchemaCompareResult` + `serialize_output=True` so the 
approved resume path returns the same dict shape as `execute()`.
   
   ## E2E
   
   Verified against a live Airflow (breeze, pydantic-ai TestModel, two sqlite 
sources with drifted schemas): trigger, pause in `awaiting_input`, review, 
approve, task success with dict XCom.
   
   
![e2e](https://raw.githubusercontent.com/guan404ming/airflow/e2e-assets-llm-schema-compare/e2e_llm_schema_compare_approval.gif)
   
   related: #70069
   
   ---
   
   ##### Was generative AI tooling used to co-author this PR?
   
   - [X] Yes — Claude Code (Fable 5)
   
   Generated-by: Claude Code (Fable 5) following [the 
guidelines](https://github.com/apache/airflow/blob/main/contributing-docs/05_pull_requests.rst#gen-ai-assisted-contributions)


-- 
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]

Reply via email to