Wait for collapsed reasoning to let go of its text before counting the column - #86
Merged
splitbrain merged 1 commit intoOct 3, 2026
Conversation
…e column The row-run layout test counted the turn's rows and paragraphs once every reasoning panel had height 0. A closing panel reaches height 0 one frame before Radix removes its children, and the reasoning text inside is markdown the count took for prose. The test then found 7 or 8 elements instead of 6, about one run in twenty. The wait now ends when no panel holds markdown any more.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Fixes the flaky
e2e/ux.test.ts › a row run stays a list across a message that speaks, which failed onmainin 37118036046 (expected 7 to be 6) and 37150134723 (expected 8 to be 6), on commits that did not touch the thread view.Cause
The test measures the turn's layout once the reasoning panels have collapsed, and waited for every panel to have height 0. Radix closes a panel in two steps: the animation brings the height to 0, and only when it has ended are the panel's children removed. In the frame between, a panel is at height 0 and still holds its text. That text is markdown with the
aui-mdclass, one of the selectors the test counts as prose, so a measurement in that window counts one or two reasoning texts on top of the six seams.Confirmed with an instrumented copy of the test: in a failing run both panels were at height 0 and still held their markdown, and the extra entries were the two reasoning texts, inside
[data-slot="reasoning-content"].Fix
The wait now ends when no reasoning panel contains
.aui-mdany more, which is when the collapse is finished. The other test with a height-based wait counts only triggers, which a zero-height panel does not move, so it stays.Locally the flake hit about one run in twenty before, and six runs passed after.