Fix numeric color sorting in parcats bundles - #7959
Open
CAOShurong wants to merge 3 commits into
Open
Conversation
camdecoster
reviewed
Aug 18, 2026
camdecoster
left a comment
Contributor
There was a problem hiding this comment.
This looks like a good fix to me! I left one comment for you to consider. After that's resolved, I'll approve.
camdecoster
reviewed
Aug 18, 2026
| } | ||
|
|
||
| return a.length - b.length; | ||
| } |
Contributor
There was a problem hiding this comment.
What do you think of adding a JSDoc comment and a block to handle NaN values? This likely wouldn't happen in practice, but it could.
Suggested change
| } | |
| /** | |
| * Compare two sort arrays element by element in ascending order. | |
| * Values that do not order against each other, for example NaN, sort last. | |
| * The shorter array sorts first when one array is a prefix of the other. | |
| * | |
| * @param {Array} a | |
| * @param {Array} b | |
| */ | |
| function compareArrays(a, b) { | |
| for (let i = 0; i < Math.min(a.length, b.length); i++) { | |
| const valA = a[i]; | |
| const valB = b[i]; | |
| if (valA < valB) return -1; | |
| if (valA > valB) return 1; | |
| // Handle values that do not order against each other (NaN, undefined, etc.) | |
| if (valA !== valB) { | |
| // Sort these after every orderable value. | |
| const badA = isNaN(valA); | |
| const badB = isNaN(valB); | |
| if (badA !== badB) return badA ? 1 : -1; | |
| } | |
| } | |
| return a.length - b.length; | |
| } |
Per review suggestion from camdecoster: - Add JSDoc comment documenting ascending-order comparison, NaN-sorting behavior, and prefix handling. - Handle values that do not order against each other (NaN, undefined) by sorting them after every orderable value. - Add test verifying NaN color values sort after orderable values in bundled parallel-categories paths. The 4 pre-existing drag/reorder test failures are unchanged by this commit (27->28 passing, same 4 baseline failures).
Author
|
Thanks @camdecoster for the suggestion! I've implemented it in commit \�ec4e475:
The new test and the existing numeric-color test both pass. The 4 pre-existing drag/reorder test failures are unchanged by this commit (they fail identically on the prior head). |
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
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.
Closes #7952.
When
bundlecolorsis enabled, the parcats path sorter prepends each path's raw color to an array of category indices, then compares those arrays with JavaScript's<and>operators. That coerces the arrays to comma-separated strings, so numeric colors such as1, 10, 2, 20are ordered lexicographically.This change compares the sort arrays element by element. Numeric colors now retain numeric order, while display-category indices and the existing value-index tie-breaker keep their current roles. A browser regression test renders a bundled parcats trace and checks the resulting path view-model order.
Verification:
npm run lint— passed (2,479 files)npm run cibuild— passednpm run test-jasmine -- parcats --nowatch --report-spec— the new regression passed; the run had 27 successes and the same four pre-existing Chrome 151/Windows drag-callback failures reproduced before the implementationnpm run test-bundle— 10 bundle suites passed; the unrelated MathJax v2 config suite timed out on Chrome 151/WindowsThe repository does not define a type-check script. The Windows
test-syntaxand fullbuildcommands also hit an existing glob portability problem (0source files found), so I am relying on the successful CI build above and the repository's hosted CI for those Linux gates.AI assistance disclosure: This contribution was developed and tested with OpenAI Codex. The implementation, regression coverage, and observed command results are disclosed above.