QuakeWang opened a new pull request, #5:
URL: https://github.com/apache/paimon-mosaic/pull/5
Projected reads already avoided unnecessary IO for sparse paged buckets, but
the reader still copied buffers at several internal boundaries.
Before this PR:
- `read_ranges` coalesced IO ranges, then copied each logical range into
its own `Vec<u8>`.
- Paged projection copied selected slot bytes into an intermediate map
before parsing.
- `parse_column_slot` copied the decompressed page payload after parsing
its header.
This PR keeps the existing format and `read_ranges` API, but adds a shared
range buffer path for projected reads. The reader now borrows slices from
coalesced buffers during parsing, stores paged slot locations instead of slot
bytes, and lets `ColumnPageReader` parse from an offset inside the owned
decompressed page buffer.
--
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]
