[BUG] Parquet with RLE encoded booleans loads corrupted data #8630
Labels
bug
Something isn't working
cudf_dependency
An issue or PR with this label depends on a new feature in cudf
Describe the bug
The RAPIDS Accelerator fails to properly load a Parquet file with RLE encoded booleans. Some of the boolean values should be null, but none of the loaded values are null.
Steps/Code to reproduce bug
Load the rle_boolean_encoding.parquet file from parquet-testing
Expected behavior
The data should match what is loaded from the same file via Spark CPU.
The text was updated successfully, but these errors were encountered: