rangareddy commented on issue #19862: URL: https://github.com/apache/hudi/issues/19862#issuecomment-5695665951
Opened #19974 (draft) for this. Confirmed all ten sites you listed, and the six other `applyFilter` procedures are genuinely unaffected (they take no `limit`), so the list is complete. One thing worth flagging from implementing it: the ten are not one shape. Four only truncate an already-materialised list, three bound at the source (`df.orderBy(...).limit(n)`, `rdd.take(n)`), but in `show_cleans`, `show_timeline` and `show_logfile_records` the `limit` is a real work bound, capping how many clean instants get their metadata read and how many records are pulled out of log files. Filtering first without care would make those unbounded, so the early bound is lifted only when a filter is actually present; an unfiltered call does exactly the work it did before. `show_timeline` also needed an extra condition: `getTimelineEntries` deliberately ignores `limit` when both `startTime` and `endTime` are given, so reapplying the bound unconditionally truncated a fully specified range to the default of 20. That is pinned by a new case across all four V1/V2 x COW/MOR variants. -- 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]
