paleolimbot commented on PR #494: URL: https://github.com/apache/parquet-format/pull/494#issuecomment-2814427804
Thank you for the summary! I like the place we're at right now: readers can confidently skip files or row groups for all cases I envision being important. The perfectionist in me would love to handle the all-empty-but-not-null case (3) but I would be surprised if this was important (if it is, one can demonstrate a realistic scenario and propose to add something to handle that case). My proposal was hacky and I'd love to avoid it 🙂 ...it's a great point that using a "missing" thift element, an NaN, and/or Inf values for this concept are easily misinterpreted because they carry other meaning in a Parquet context. I've updated C++'s GeoStatistics to better communicate the meaning of various combinations of empty, uncalculated, or invalid. -- 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]
