rusackas opened a new pull request, #44111:
URL: https://github.com/apache/superset/pull/44111

   ### SUMMARY
   **Draft, stacked on #44103** — this serves double duty: it's a real fix for 
a visual regression Evan spotted (control-panel `Select` controls sitting 
taller than sibling `Input` controls), and it doubles as the first live test of 
#44103's new Chromatic workflow, since GitHub evaluates 
`pull_request`-triggered workflow files from the PR's own head ref — so 
Chromatic will actually run here even before #44103 merges to master. Rebase 
onto master once #44103 lands.
   
   The antd v5→v6 upgrade (#41636) renamed the CSS selectors targeting 
rc-select/rc-overflow's multi-value tag container to class names that don't 
exist in the antd 6 / rc-select build actually bundled here:
   
   | Used in code | Actually rendered |
   |---|---|
   | `.ant-select-content` | `.ant-select-selection-overflow` |
   | `.ant-select-content-item` | `.ant-select-selection-overflow-item` |
   | `.ant-select-content-item-suffix` | 
`.ant-select-selection-overflow-item-suffix` |
   | `.ant-select-input` | `.ant-select-selection-search` |
   | `.ant-select-placeholder` | `.ant-select-selection-placeholder` |
   
   Verified against the installed `[email protected]` / `rc-overflow` source 
directly (`prefixCls` string concatenation in `Selector/MultipleSelector.js`, 
`Selector/SingleSelector.js`, and rc-overflow's `Overflow.js`) — none of the 
old selectors exist anywhere in the bundled JS.
   
   Because the `oneLine` rule's `flex-wrap: nowrap` never actually applied (it 
targeted a class that's never rendered), multi-value `Select` controls using 
`oneLine` silently stopped being capped to a single row and grew taller than 
sibling controls whenever their tags wrapped — visible in the Explore control 
panel as Select-based controls sitting noticeably taller than adjacent 
Input-based controls.
   
   A follow-up fix (#42600) touched the same rule but only removed an 
already-dead `.ant-select-content` reference from the `max-height` selector 
list, without addressing the real class-name drift.
   
   Adds a `OneLineHeightParity` story pairing an `oneLine` multi-Select with a 
plain Input at matching width, so this regression is visually covered going 
forward rather than only catchable by eye.
   
   ### BEFORE/AFTER SCREENSHOTS OR ANIMATED GIF
   The new `OneLineHeightParity` Storybook story (`Components/Select`) shows 
this directly; Chromatic should also pick up the height change on this PR.
   
   ### TESTING INSTRUCTIONS
   - `tsc --noEmit` on `superset-ui-core` is clean (0 errors).
   - `jest packages/superset-ui-core/src/components/Select` shows the identical 
64-failed/148-passed baseline as unmodified master — those are pre-existing, 
unrelated stale-class failures in `AsyncSelect.test.tsx`'s own test helpers 
(querying `.ant-select-dropdown-list`, another leftover from the same antd v6 
upgrade), a separate follow-up worth filing, not touched here.
   - `oxlint` clean on both changed files.
   - Storybook: `Components/Select` → `OneLineHeightParity` — the Select and 
Input should render at matching heights.
   
   ### ADDITIONAL INFORMATION
   - [ ] Has associated issue:
   - [ ] Required feature flags:
   - [x] Changes UI
   - [ ] Includes DB Migration (follow approval process in 
[SIP-59](https://github.com/apache/superset/issues/13351))
     - [ ] Migration is atomic, supports rollback & is backwards-compatible
     - [ ] Confirm DB migration upgrade and downgrade tested
     - [ ] Runtime estimates and downtime expectations provided
   - [ ] Introduces new feature or API
   - [ ] Removes existing feature or API
   
   🤖 Generated with [Claude Code](https://claude.com/claude-code)


-- 
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