parquet: Allow setting parquet rowgroup row count limit#16317
Open
sqd wants to merge 1 commit into
Open
Conversation
parquet-java/parquet-mr supports this limit (config key parquet.block.row.count.limit) but Iceberg doesn't support setting it yet. This commit simply wires it up. Parquet.WriteBuilder has two terminal paths: when callers supply a createWriterFunc (Iceberg's ParquetValueWriter — the path every production engine integration uses, such as rewrite_data_files), the build returns Iceberg's own ParquetWriter, which manages the row-group lifecycle itself and ignores parquet-mr's auto-roll. When callers supply a WriteSupport instead, we delegate to parquet-mr's ParquetWriter, which enforces row-group limits internally. The new property has to be wired into both: on the Iceberg path it is carried via ParquetProperties and consumed by an explicit recordCount check in ParquetWriter.checkSize(); on the parquet-mr path it is passed through ParquetWriteBuilder.withRowGroupRowCountLimit() and enforced by parquet-mr.
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
parquet-java/parquet-mr supports this limit (config key parquet.block.row.count.limit) but Iceberg doesn't support setting it yet. This commit simply wires it up.
Parquet.WriteBuilder has two terminal paths: when callers supply a createWriterFunc (Iceberg's ParquetValueWriter — the path every production engine integration uses, such as Spark rewrite_data_files), the build returns Iceberg's own ParquetWriter, which manages the row-group lifecycle itself and ignores parquet-mr's auto-roll. When callers supply a WriteSupport instead, we delegate to parquet-mr's ParquetWriter, which enforces row-group limits internally.
The new property has to be wired into both: on the Iceberg path it is carried via ParquetProperties and consumed by an explicit recordCount check in ParquetWriter.checkSize(); on the parquet-mr path it is passed through ParquetWriteBuilder.withRowGroupRowCountLimit() and enforced by parquet-mr.