dgvj-work opened a new issue, #2370: URL: https://github.com/apache/iceberg-go/issues/2370
Incremental changelog planning reads every live data manifest that the partition filter keeps for each snapshot that adds delete files. Equality deletes, and position deletes that do not name a data file, cannot be limited to one path, so that scan is what the planner does today. The remaining cost is narrowing those reads to manifests that can contain the referenced data files, instead of opening every manifest the partition filter keeps. Decoded manifest entries are already cached by path. This came out of the review on #2169. It is separate from the correctness of the planned tasks. -- 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] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
