aminghadersohi commented on PR #44285:
URL: https://github.com/apache/superset/pull/44285#issuecomment-5687521749

   ## Rebased onto current master — CI re-run fully attributed
   
   Head is now `520c54969f65a38a69996c429b0eb252d4645278`, rebased cleanly (no 
conflicts) onto master `a184395eca7c291d0b8b438f00690c633112f4d7`. The 
five-file diff is unchanged; the branch still touches **zero** migrations (`git 
diff origin/master -- superset/migrations/` is empty).
   
   ### CI is not green, and every failure is attributed
   
   Terminal rollup: **66 pass, 15 fail, 12 skipping, 0 pending.** I inspected 
the log of every failing job rather than assuming a shared cause:
   
   - **12 jobs** fail on the upstream Alembic two-head condition, with the 
literal error in-log: `Multiple head revisions are present for given argument 
'head'` / `The script directory has multiple heads (due to branching)` — 
`docker-build (dev)`, `test-sqlite`, `test-mysql`, `test-postgres (current)`, 
`test-postgres-hive`, `test-postgres-presto`, `test-load-examples`, 
`cypress-matrix (chrome)`, and the four `playwright-tests*` matrix jobs.
   - **3 jobs** are aggregation gates that only re-report those children: 
`test-postgres-required`, `cypress-matrix-required`, 
`playwright-tests-required`.
   
   This is the identical 15-job failure set as the pre-rebase head 
`0432db3c00`, so the rebase introduced nothing new. The 12 skipping checks are 
frontend/bot/master-push-only contexts excluded by the Python-and-docs path 
filter; none is a test gate for this change.
   
   Master still carries two heads — `c7f53d184ea2` 
(`coordinate_purge_audit_pruning`) and `88a01c781622` 
(`index_ab_user_lower_username`) — verified by parsing the full 380-revision 
graph at `a184395eca`.
   
   ### Causality is confirmed by a control PR, not by inference
   
   #44288 changes **only** the head merge, and its CI is fully green — 
including the exact jobs failing here: `test-sqlite`, `test-load-examples`, 
`test-mysql`, `test-postgres (current)`, `test-postgres-hive`, 
**`test-postgres-presto`**, `cypress-matrix`, and all `docker-build` variants 
(65 pass / 0 fail). That isolates the blocker to the head condition and 
excludes this PR's dependency change as a contributor.
   
   ### Exact human action required
   
   **A maintainer with write access must review and merge 
[#44288](https://github.com/apache/superset/pull/44288)** 
(`d6f00dcffec9beb74bc9480b40f38515c5527a5a`). It is OPEN, non-draft, CI-green, 
and `REVIEW_REQUIRED` with @eschutho requested; it is the only open head-merge 
PR (the duplicate #44317 is closed). #44283 adds a ClickHouse migration and 
will need to reparent onto whichever head lands. Once master reports a single 
head, I will rebase this PR again and re-run CI.
   
   No migration workaround is being added to this Presto dependency PR — that 
would duplicate #44288 and conflict when it lands. Nothing has been drafted or 
merged.
   
   ### Change verification on the rebased tree
   
   - `tests/unit_tests/db_engine_specs/test_presto.py`: **135 passed**.
   - Scoped `pre-commit` over all five changed files: all applicable hooks 
passed, including mypy, Ruff, pylint, and engine-spec metadata validation.
   - The `version_requirements` string in `docs/src/data/databases.json` 
matches `PrestoEngineSpec` byte-for-byte, and its field ordering matches the 
generator's convention (as in the existing Firebird entry).
   - Floor causality re-verified in a clean Python 3.11 venv: SQLAlchemy 2.0.52 
+ `pyhive[presto]==0.6.5` → `ModuleNotFoundError: No module named 
'sqlalchemy.databases'`; `==0.7.0` → dialect loads as `PrestoDialect` and 
compiles `SELECT :value` to `SELECT %(value)s` with no socket use.
   
   Review status: zero submitted reviews, zero inline threads, zero human 
comments. Human approval is still required.
   


-- 
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]


---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to