hutiefang76 opened a new pull request, #12632:
URL: https://github.com/apache/seatunnel/pull/12632

   ### Purpose of this pull request
   
   Resuming without an ID could select the newest-created session even after an 
older session was resumed and saved more recently. Sort valid session metadata 
by last_active before applying the limit, and use that same order for the 
default resume target. Skip corrupt files and malformed metadata.
   
   ### Does this PR introduce _any_ user-facing change?
   
   Yes, in dev. The CLI resumes the most recently active valid session; legacy 
files fall back to created_at. Session IDs and creation times are preserved.
   
   ### How was this patch tested?
   
   - `python3 -m pytest -q`: 214 passed, 3 subtests passed. Regression coverage 
includes save/resume/save ordering, limit handling, corrupt files, legacy 
metadata and invalid metadata types.
   - Root `-N spotless:apply` and `-N -DskipTests verify` passed. Modified 
Python files passed the fatal-error Ruff checks.
   - Full-repository `-DskipTests verify` was attempted but stopped at 
`seatunnel-engine-ui`: the npm mirror returned HTTP 403 for `tslib-2.3.0.tgz`. 
This is not a successful full-repository build.
   
   ### Check list
   
   * [x] No new JAR binary or dependency is added.
   * [x] EN/ZH documentation updated.
   * [ ] Incompatibility guide updated. Not applicable: no schema or 
configuration incompatibility.
   * [ ] Connector registration updates: not applicable.
   
   Implemented with Codex assistance; the reproduction and checks above were 
run against this branch.
   
   ---
   [![Compound 
Engineering](https://img.shields.io/badge/Built_with-Compound_Engineering-6366f1)](https://github.com/EveryInc/compound-engineering-plugin)
   


-- 
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