Skip to content

fix(cpp): aligned VECTOR row-offset skip only when time/value counts …#778

Merged
jt2594838 merged 1 commit intoapache:developfrom
hongzhi-gao:fix/py-dataframe-query-row
Apr 12, 2026
Merged

fix(cpp): aligned VECTOR row-offset skip only when time/value counts …#778
jt2594838 merged 1 commit intoapache:developfrom
hongzhi-gao:fix/py-dataframe-query-row

Conversation

@hongzhi-gao
Copy link
Copy Markdown
Contributor

Whole-chunk and whole-page skips by statistic count previously used only the value side for aligned series, which could desynchronize row_offset from decoded rows when ChunkMeta or page header counts differed.

Require both time and value statistics to be present, positive, and equal before applying count-based skip; otherwise decode and rely on page/row handling.

…match

Whole-chunk and whole-page skips by statistic count previously used only the
value side for aligned series, which could desynchronize row_offset from
decoded rows when ChunkMeta or page header counts differed.

Require both time and value statistics to be present, positive, and equal
before applying count-based skip; otherwise decode and rely on page/row
handling.

Made-with: Cursor
@jt2594838 jt2594838 merged commit fa3e01f into apache:develop Apr 12, 2026
16 checks passed
@jt2594838 jt2594838 deleted the fix/py-dataframe-query-row branch April 12, 2026 07:57
ycycse pushed a commit to ycycse/tsfile that referenced this pull request Apr 12, 2026
…match (apache#778)

Whole-chunk and whole-page skips by statistic count previously used only the
value side for aligned series, which could desynchronize row_offset from
decoded rows when ChunkMeta or page header counts differed.

Require both time and value statistics to be present, positive, and equal
before applying count-based skip; otherwise decode and rely on page/row
handling.

Made-with: Cursor
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants