Repository navigation
Return UUID column statistics to Java as bytes - #11299
Merged
Merged
Conversation
A writer that wrote a UUID column failed in finish(), because the min and max of a fixed-size list of u8 had no Java conversion. These bounds now return to Java as byte arrays. Signed-off-by: Robert Kruszewski <[email protected]>
Merging this PR will degrade performance by 10.08%1 warning
Warning Please fix the performance issues or acknowledge them on CodSpeed. Performance Changes
Tip Investigate this regression by commenting Comparing 
Footnotes
|
myrrc
approved these changes
Nov 6, 2026
dfa1
added a commit
to dfa1/vortex-java
that referenced
this pull request
Nov 12, 2026
From 0.92 Rust imports arrow.uuid as vortex.uuid and records its min and max in the file statistics: a FixedSizeList(U8, 16) scalar, exact, in unsigned byte order (vortex-data/vortex#11299 let vortex-jni hand them back to Java). FileStatsAccumulator now does the same, so a uuid column's footer matches vortex-jni's byte for byte, and WriteSummary reports the bounds as UUIDs. UUID.compareTo is not that order: it compares signed longs. 0.93 also unblocks two uuid cross-checks that were skipped: - DictAllTypesInteropIntegrationTest's uuid case: vortex-jni 0.90 could not close a writer with a uuid column; - JavaWritesRustReadsIntegrationTest#javaWriter_rustReader_nullable_uuid, disabled since vortex-jni 0.74, which could not export vortex.uuid to Arrow. Rust now exports it as Arrow's arrow.uuid (vortex-data/vortex#11224), so the test reads UUIDs instead of 16-byte binaries. Rust 0.92's canonical-decimal change (vortex-data/vortex#10995) needs nothing here: vortex-java writes vortex.decimal at its precision's canonical width and narrows only through vortex.decimal_byte_parts, as Rust does. Co-Authored-By: Claude Opus 5.5 <[email protected]> Claude-Session: https://claude.ai/code/session_011fRbYvRdQi45rxQgF3QMMP
dfa1
added a commit
to dfa1/vortex-java
that referenced
this pull request
Nov 12, 2026
From 0.92 Rust imports arrow.uuid as vortex.uuid and records its min and max in the file statistics: a FixedSizeList(U8, 16) scalar, exact, in unsigned byte order (vortex-data/vortex#11299 let vortex-jni hand them back to Java). FileStatsAccumulator now does the same, so a uuid column's footer matches vortex-jni's byte for byte, and WriteSummary reports the bounds as UUIDs. UUID.compareTo is not that order: it compares signed longs. 0.93 also unblocks two uuid cross-checks that were skipped: - DictAllTypesInteropIntegrationTest's uuid case: vortex-jni 0.90 could not close a writer with a uuid column; - JavaWritesRustReadsIntegrationTest#javaWriter_rustReader_nullable_uuid, disabled since vortex-jni 0.74, which could not export vortex.uuid to Arrow. Rust now exports it as Arrow's arrow.uuid (vortex-data/vortex#11224), so the test reads UUIDs instead of 16-byte binaries. Rust 0.92's canonical-decimal change (vortex-data/vortex#10995) needs nothing here: vortex-java writes vortex.decimal at its precision's canonical width and narrows only through vortex.decimal_byte_parts, as Rust does. Co-Authored-By: Claude Opus 5.5 <[email protected]> Claude-Session: https://claude.ai/code/session_011fRbYvRdQi45rxQgF3QMMP
dfa1
added a commit
to dfa1/vortex-java
that referenced
this pull request
Nov 12, 2026
From 0.92 Rust imports arrow.uuid as vortex.uuid and records its min and max in the file statistics: a FixedSizeList(U8, 16) scalar, exact, in unsigned byte order (vortex-data/vortex#11299 let vortex-jni hand them back to Java). FileStatsAccumulator now does the same, so a uuid column's footer matches vortex-jni's byte for byte, and WriteSummary reports the bounds as UUIDs. UUID.compareTo is not that order: it compares signed longs. 0.93 also unblocks two uuid cross-checks that were skipped: - DictAllTypesInteropIntegrationTest's uuid case: vortex-jni 0.90 could not close a writer with a uuid column; - JavaWritesRustReadsIntegrationTest#javaWriter_rustReader_nullable_uuid, disabled since vortex-jni 0.74, which could not export vortex.uuid to Arrow. Rust now exports it as Arrow's arrow.uuid (vortex-data/vortex#11224), so the test reads UUIDs instead of 16-byte binaries. Rust 0.92's canonical-decimal change (vortex-data/vortex#10995) needs nothing here: vortex-java writes vortex.decimal at its precision's canonical width and narrows only through vortex.decimal_byte_parts, as Rust does. Co-Authored-By: Claude Opus 5.5 <[email protected]> Claude-Session: https://claude.ai/code/session_011fRbYvRdQi45rxQgF3QMMP
dfa1
added a commit
to dfa1/vortex-java
that referenced
this pull request
Nov 12, 2026
From 0.92 Rust imports arrow.uuid as vortex.uuid and records its min and max in the file statistics: a FixedSizeList(U8, 16) scalar, exact, in unsigned byte order (vortex-data/vortex#11299 let vortex-jni hand them back to Java). FileStatsAccumulator now does the same, so a uuid column's footer matches vortex-jni's byte for byte, and WriteSummary reports the bounds as UUIDs. UUID.compareTo is not that order: it compares signed longs. 0.93 also unblocks two uuid cross-checks that were skipped: - DictAllTypesInteropIntegrationTest's uuid case: vortex-jni 0.90 could not close a writer with a uuid column; - JavaWritesRustReadsIntegrationTest#javaWriter_rustReader_nullable_uuid, disabled since vortex-jni 0.74, which could not export vortex.uuid to Arrow. Rust now exports it as Arrow's arrow.uuid (vortex-data/vortex#11224), so the test reads UUIDs instead of 16-byte binaries. Rust 0.92's canonical-decimal change (vortex-data/vortex#10995) needs nothing here: vortex-java writes vortex.decimal at its precision's canonical width and narrows only through vortex.decimal_byte_parts, as Rust does. Co-Authored-By: Claude Opus 5.5 <[email protected]> Claude-Session: https://claude.ai/code/session_011fRbYvRdQi45rxQgF3QMMP
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.
In 0.92.0,
VortexWriter.finish()fails for any file that has a UUID column:The writer now records min and max for the
arrow.uuidextension type, butscalar_to_javacan return only flat scalars to Java. The whole write fails because one bound cannot be represented. Iceberg found this while upgrading to 0.92.0.