You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
The upcoming GeoParquet 1.1 spec has support for bounding box columns (opengeospatial/geoparquet#191). This enables readers to take advantage of any spatial partitioning present in the dataset (by looking at the Parquet statistics for each bounding box column) while maintaining interoperability because the geometries are still stored as WKB, which many tools are able to read and write.
GeoPandas hasn't yet added support for this, and we may need a special case here because we already have the bbox columns. But we should include the GeoParquet metadata on this to ensure that those bbox columns are found.
Does this mean that we should implement our own wrapper for writing GeoParquet here, so that we can manually specify metadata?
The text was updated successfully, but these errors were encountered:
The write side is closed by #40. At the moment, we don't require our own read-side spatial filtering inside stac-geoparquet, so I don't think we need a special reader.
The upcoming GeoParquet 1.1 spec has support for bounding box columns (opengeospatial/geoparquet#191). This enables readers to take advantage of any spatial partitioning present in the dataset (by looking at the Parquet statistics for each bounding box column) while maintaining interoperability because the geometries are still stored as WKB, which many tools are able to read and write.
GeoPandas hasn't yet added support for this, and we may need a special case here because we already have the bbox columns. But we should include the GeoParquet metadata on this to ensure that those bbox columns are found.
Does this mean that we should implement our own wrapper for writing GeoParquet here, so that we can manually specify metadata?
The text was updated successfully, but these errors were encountered: