akshat-lakhera commented on issue #20785: URL: https://github.com/apache/druid/issues/20785#issuecomment-6497900936
I'd like to work on this. I read through #17192 and your list of gaps, and my plan is to start from that PR's stringMin and stringMax aggregators on current master and fix them in the order you laid out. That means a dedicated string base class for the null handling, comparing UTF-8 bytes so the order matches the dictionaries, truncating on character boundaries, tracking the min and max dictionary id per group in the vector path for dictionary encoded columns, and turning off vectorization for expression inputs until there's a proper selector for them. I'd wire MIN and MAX on VARCHAR to them in SQL and add vector aggregator tests plus CalciteQueryTest cases for nulls, multi value columns and expressions, and leave ingestion support for a separate PR as you suggested. For multi value strings I'd take the min and max across the individual values. Does that sound right to you @FrankChen021? -- 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]
