Glue: honor Segment in get_partitions - #10123
Open
parisni wants to merge 2 commits into
Open
Conversation
A segmented GetPartitions (used by the SDK-v1 Hive metastore Glue client for parallel partition scans) must return a disjoint slice per segment whose union is the full set with no duplicates. moto ignored the Segment parameter and returned every partition for each segment, so a client issuing TotalSegments=N saw each partition N times. This made Spark read every partitioned table row N times when reading a Delta/Hudi table synced into the moto Glue catalog. Assign each partition to exactly one segment via a stable md5 hash of its values, mirroring real AWS Glue segmentation semantics.
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
A segmented
GetPartitions(theSegmentparameter,{SegmentNumber, TotalSegments}) is used by parallel scanners such as the Hive metastore Glueclient. Per the API contract each segment must return a disjoint slice of the
partitions, and the union across all segments is the full set with no
duplicates.
moto currently ignores
Segmentand returns every partition for every segment,so a client issuing
TotalSegments=Nsees each partition N times. When Sparkreads a partitioned table backed by the moto Glue catalog this multiplies every
row by N.
This assigns each partition to exactly one segment via a stable md5 hash of its
values, mirroring real AWS Glue segmentation.
Test added in
tests/test_glue/test_datacatalog.py.