C-Loftus commented on PR #1504: URL: https://github.com/apache/arrow-go/pull/1504#issuecomment-6777071019
Thanks for your feedback! Here are the changes I made: - PopulateCryptoData now clears GeospatialStatistics from the cut-down plaintext ColumnMetaData, like it already does for Statistics and EncodingStats. `TestGeoStatisticsRedactedFromPlaintextFooter` checks that the bbox and geometry types are no longer in the footer when a column has its own key and the footer is plaintext. - I removed GeoStatistics.Merge, Reset and UpdateWKB. The exported API is now just what the writer and readers use. Merge can come back when file-level aggregation needs it. - NaN values are now skipped per dimension instead of invalidating the chunk, so they never end up in the stats and the other geometries can still be used for pruning. A new test recomputes stats from geospatial-with-nan.parquet and gets the same result as the reference writer `(x=[10,130] y=[20,140] types=[3001 3002])`. -- 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]
