rdblue commented on PR #13227: URL: https://github.com/apache/iceberg/pull/13227#issuecomment-3046661554
> Just a note that I don't think this clarification is needed...a Cartesian bounding box like the one described here is perfectly legal in for the geometry type in GeoArrow's box type, GeoParquet's file metadata, GeoJSON, Parquet GeoStatistics, and the current Iceberg spec (even though all of these allow for the wraparound case as well). How does a writer know what bounding box to produce? In all of those specs, is it the writer's decision whether to use a purely cartesian bounding box or one that wraps around? What I want to avoid is needing to interpret the CRS to know how to compute the bounding box. If writers are allowed to produce a box that is cartesian and ignore whether longitude is periodic, then maybe that is fine because it is the norm. But my expectation is that this would be determined by the CRS and if we wanted to produce a simpler bounding box it would be incorrect without this change. -- 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]
