richardfogaca commented on PR #40128:
URL: https://github.com/apache/superset/pull/40128#issuecomment-4635411416
One follow-up not covered in my earlier review, verified against current
HEAD `9794041`:
-
**`superset/migrations/versions/2026-05-08_12-05_9e1f3b8c4d2a_add_deleted_at_to_dashboards.py:143`**
On MySQL, the migration drops `idx_unique_slug` before creating the
functional replacement. Since MySQL DDL autocommits, a failure in the following
`CREATE UNIQUE INDEX` would leave the table without slug uniqueness; the
downgrade has the same shape at line 172.
WDYT - could we create the replacement unique index first, then drop the
previous one only after it succeeds? That keeps the stricter existing
uniqueness in place if the functional-index DDL fails on an unexpected
MySQL/MariaDB variant.
- **`UPDATING.md:74`**
The docs still say MySQL `8.0+` gets the functional index and line 76
repeats the MySQL `8.0+` slug-reuse behavior, but the migration gates
functional indexes at MySQL `8.0.13+` and excludes MariaDB.
Small suggestion: could we update this to say MySQL 8.0.13+ excluding
MariaDB, with MySQL 8.0.0-8.0.12 and MariaDB in the full-constraint bucket?
--
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]
