Details
-
Bug
-
Status: Resolved
-
Minor
-
Resolution: Fixed
-
None
Description
Getting the range of bytes of a column chunk inside a parquet file can be useful for external crates (for instance if they want to pre-fetch the columns), and is not completely obvious (it is enough to take a look at [1] and [2] to see that things can quickly get messy).
I think it would be nice to move this logic in the metadata definition rather than have lost it in the middle of the reader implem.
[1] https://stackoverflow.com/questions/55225108/why-is-dictionary-page-offset-0-for-plain-dictionary-encoding/
[2] https://issues.apache.org/jira/browse/PARQUET-816
Attachments
Issue Links
- links to