[Kernel][Defaults] Handle legacy map types in Parquet files #3097
Conversation
This file contains 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
| int initialBatchSize, | ||
| MapType typeFromClient, | ||
| GroupType typeFromFile) { | ||
| // Repeated element can be any name. Latest Parquet versions use "key_value" as the name, |
There was a problem hiding this comment.
Can we check which version it was written as?
i.e. latest -> use key_value. older? only then allow arbitrary name
There was a problem hiding this comment.
As far as I know, there is no clear or confirmed way to identify which version (i.e., Parquet format version) the file was written with. Each writer followed its own way to add the metadata.
scottsand-db
approved these changes
May 14, 2024
vkorukanti
added a commit
that referenced
this pull request
May 14, 2024
Currently, Kernel's Parquet reader explicitly looks for the `key_value` repeated group under the Parquet map type, but the older versions of Parquet writers wrote any name for the repeated group. Instead of looking for the explicit `key_value` element, fetch the first element in the list. See [here](https://github.com/apache/parquet-format/blob/master/LogicalTypes.md#maps) for more details. The [test](https://github.com/apache/spark/blob/master/sql/core/src/test/scala/org/apache/spark/sql/execution/datasources/parquet/ParquetThriftCompatibilitySuite.scala#L29) and sample file written by legacy writers are taken from Apache Spark™. Some columns (arrays with 2-level encoding, another legacy format) from the test file are currently not supported. I will follow up with a separate PR. It involves bit refactoring on the ArrayColumnReader.
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.
Description
Currently, Kernel's Parquet reader explicitly looks for the
key_valuerepeated group under the Parquet map type, but the older versions of Parquet writers wrote any name for the repeated group. Instead of looking for the explicitkey_valueelement, fetch the first element in the list. See here for more details.How was this patch tested?
The test and sample file written by legacy writers are taken from Apache Spark™.
Some columns (arrays with 2-level encoding, another legacy format) from the test file are currently not supported. I will follow up with a separate PR. It involves bit refactoring on the ArrayColumnReader.