Rich-T-kid opened a new issue, #12160: URL: https://github.com/apache/arrow-rs/issues/12160
When a Parquet BYTE_ARRAY column has no UTF8 annotation, supplying Dictionary(Int32, Utf8) currently does not preserve the requested value type through dictionary decoding. The reader decodes dictionary values as Binary, and callers must cast resulting batches after decoding. For pushed predicates this can rebuild the dictionary for every batch before evaluating the predicate. The Parquet dictionary reader should convert each dictionary page values to the requested compatible string or binary type while preserving its keys. Invalid UTF-8 must still return an error, and plain-encoded fallback pages must produce the same requested dictionary type. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
