Hi, I've been playing around a little with the MVT driver and noticed in the test suite, features are written first via the Memory driver and then gdal.VectorTranslate is used to create the MVTs.
Is there any advantage to doing this vs writing features directly to a MVT datasource? From what I could tell in experimenting, when writing directly to a MVT dataset, the driver writes all the features to a sqlite db per tile, and then when the dataset is closed, the actual MVTs are created; wasn't sure if there's any optimizations taken if you go through a Memory dataset first. Also, is there a way to create just a single MVT with the driver? I think the answer is no (always creates all the tiles that cover the geometries) but I might have missed an option. Thanks! Patrick
_______________________________________________ gdal-dev mailing list gdal-dev@lists.osgeo.org https://lists.osgeo.org/mailman/listinfo/gdal-dev
