james-willis opened a new pull request, #1211: URL: https://github.com/apache/sedona-db/pull/1211
Follow-up to #1203, addressing @paleolimbot's review comment on `DBEngine.create_random_raster_view`: the grid was placed only by a raw GDAL 6-tuple (`gdal_transform=(100.0, 2.0, 0.0, 500.0, 0.0, -3.0)`), which is hard to read in tests. - `create_random_raster_view` now takes `bbox=(minx, miny, maxx, maxy)` (north-up, no skew — the transform is derived from the bbox and the grid shape, as `rasterio.transform.from_bounds` does). `gdal_transform` remains for grids a bbox cannot express (skew, south-up); passing both raises. - The default is now spelled as `bbox=(100.0, 482.0, 114.0, 500.0)`, which with the default 7x6 grid derives the exact historical transform, so existing fixtures are byte-identical. - The fixture writers `write_geotiff` / `write_random_geotiff` / `write_grid_geotiff` in `sedonadb.raster_testing` accept the same `bbox`/`gdal_transform` pair (exactly one), so the rasterio-oracle tests can use the readable spelling too. - New `python/sedonadb/tests/test_raster_testing.py` pins the bbox math to `rasterio.transform.from_bounds` and the historical default transform — the parity suite compares engines on the same file, so it could not catch a wrong bbox-derived transform on its own. No spark-parity callers passed `gdal_transform`, so none needed updating; they all go through the bbox default now. Verified: `integration/spark-parity` suite green (21 passed, 13 xfailed — the documented divergences) with pyspark 4.1.3 + JDK 17; `python/sedonadb/tests/functions` raster tests green (312 passed, 28 skipped); ruff 0.14.6 check + format clean. -- 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]
