zenoyang commented on pull request #8737: URL: https://github.com/apache/incubator-doris/pull/8737#issuecomment-1082698060
> 1 Not only `all pages are dict encoding`, but also `some pages are dict encoding` can also use ColumnDict to read and evaluate predicate. 2 Even considering the conversion overhead, this conversion only need to convert once in `SegmentIterator.next`, so it may not have a particularly noticeable cost.If current ColumnDict design can not do this, I think this is a point worth optimizing . 3 ` If the last data_page is DICT_ENCODING, it means that all data_pages are DICT_ENCODING`, this is a too trick design, computational logic and data storage logic should not be coupled. Otherwise we need to think too much when we redesign data structures on disk. 1. Yes. 2. The current ColumnDict can do 3. It is really tricky. It is better to write encoding info into the segment footer, but it involves changes to the segment file structure, and I don't think it is necessary for the time being. In summary, this PR will be closed first. -- 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]
