Add list-aware physical scan split boundaries - #9578
Draft
mhk197 wants to merge 1 commit into
Draft
Conversation
Signed-off-by: Matt Katz <mhkatz97@gmail.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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Summary
List layouts now persist exact scan split boundaries when an elements physical chunk end coincides with an enclosing list offset.
ChunkedLayoutseam before nested structural writers run.(outer_row_end, element_row_end)pairs in list metadata and use them directly during scan splitting; legacy layouts retain the existing heuristic fallback.Why
Previously, an elements layout could independently repartition in element-row space. Its physical endpoints almost never coincided with variable-length list input-chunk ends, leaving the parent list writer with few or no exact split points.
The parent list writer now owns a sparse, offset-aligned fence schedule. It does not retain a whole column of offsets: it translates only producer-selected candidates for each input chunk, and persists a candidate only after the completed child writer confirms that the fence stayed physical.
For nested
list<list<...>>andlist<struct<list<...>>>, each enclosing list materializes its own fences before invoking its structural elements writer. Consequently, inner writers cannot merge through an outer fence, and every metadata boundary remains in the appropriate enclosing list row space.Test plan
cargo +nightly fmt --allcargo test -p vortex-layout --no-fail-fastcargo test -p vortex-file --no-fail-fastcargo clippy -p vortex-layout -p vortex-file --all-targets --all-features -- -D warningsIncludes unit coverage for offset-aligned repartitioning, writer-report propagation, persisted-boundary reading, nested lists, and list-of-struct-of-list file splits.