Repository navigation
Conversation
Allow row-tracked Parquet append tables to apply scalar conditions and snapshot visibility before full-text Top-K in FAST mode. Preserve existing scalar search modes and Rust defaults. Cover non-DE filtering, index coverage, deletion vectors, and pinned snapshots with regression tests. Signed-off-by: QuakeWang <wangfuzheng0814@foxmail.com>
Signed-off-by: QuakeWang <wangfuzheng0814@foxmail.com>
JeonDaehong
approved these changes
Oct 21, 2026
JeonDaehong
left a comment
JeonDaehong
left a comment
Contributor
There was a problem hiding this comment.
Ran a parity check on 39337d4. Same ten rows (labels with NULLs, keep / keeper / drop / other), full-text index on text, and a btree or bitmap index on label. I compared a data-evolution table against the new row-tracked Parquet append table in FAST mode, with refine-from-data true and false. Filters: =, NOT(=), !=, IS [NOT] NULL, IN, NOT IN, >, starts_with, and AND / OR with an unindexed id. All 44 combinations return the same row ids on both tables.
The only differences between filters are FAST-mode ones, and they match the DE side: NOT(...) and OR with the unindexed id return nothing, and AND id >= 5 returns nothing once refinement is off.
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Purpose
Non-DE full-text scans currently reject data-column predicates. Enable scalar and live-row filtering before full-text Top-K.
Brief change log
Support FAST-mode prefilters on row-tracked Parquet append tables by reusing scalar coverage and exact refinement. Preserve scalar search modes and Rust defaults.
Tests
Seven non-DE regressions; full core unit suite, workspace Clippy, and formatting checks.
API and Format
Existing builder API; no storage format changes.
Documentation
Document supported tables, search modes, and SQL pushdown limitations.