chore: release 0.6.2#13
Merged
Merged
Conversation
0.6.1 threaded a label-only group header only when the value rows could reach its column -- the maze scans the value cell's own column and the row-label columns. A label-only band groups a ROW RANGE, but when the header sits in a leading stub / line-number column while the sub-rows leave that column empty and carry their identity in a different column (a numbered schedule with plan x cover value columns), no scanned column touched the header's column, so the band was unreachable and the header was dropped entirely instead of threaded. The maze now scans ALL columns for label-only bands (those carrying an explicit rowgroup_extent_end). Full-width and source scope="rowgroup" bands keep the column-restricted scan (own column + row-label columns) so unrelated stub dividers don't cross-attach. Extent still closes at the next group, so one line-item's header does not leak onto the next. 10. Travel Delay > Adult under 70 | PLANS > BASIC: 100 Also recovers data in three real-world corpus tables (pubtabnet-180372, -357665, -374857) whose year / cohort group headers -- in a stub column the data rows leave empty -- were silently dropped in 0.6.1; the 0.6.1 value-multiset check missed it because the lost text was a band label, not a cell value. New fixture matrix/label-only-rowgroup-stub-column. 3 golds recovered. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Release 0.6.2.
What's in this release
fix(core): thread label-only row-group headers across columns
Follow-up to 0.6.1's label-only-row threading. A label-only band groups a row
range, but 0.6.1 only reached it from the value cell's own column and the
row-label columns. When a group header sits in a leading stub / line-number
column while the sub-rows leave that column empty and carry their identity in a
different column (a numbered schedule with
plan × covervalue columns), theband was unreachable — so the header was dropped entirely instead of
threaded.
The maze now scans all columns for label-only bands; full-width and source
scope="rowgroup"bands keep the column-restricted scan so unrelated stubdividers don't cross-attach. Extent still closes at the next group.
This also recovers data in three real-world corpus tables (
pubtabnet-180372,-357665,-374857) whose year / cohort group headers — in a stub column thedata rows leave empty — were silently dropped in 0.6.1.
Verification
matrix/label-only-rowgroup-stub-column.🤖 Generated with Claude Code