jiayuasu commented on issue #3161:
URL: https://github.com/apache/sedona/issues/3161#issuecomment-5100660917

   Hey @umartin ,
   
   Thanks for reporting this. We traced the regression to the incomplete RT90 
datum registry in the proj4sedona version bundled with Sedona 1.9.0. EPSG:3021 
was being transformed without its required Helmert parameters.
   
   The fix is merged in [#3191](https://github.com/apache/sedona/pull/3191) and 
will be included in the next Sedona release next week.
   
   Until then, the safest workaround is to provide the complete EPSG:3021 PROJ 
definition:
   
   ```sql
   SELECT ST_AsText(ST_Transform(
     ST_Point(1272691.622, 6404578.16),
     '+proj=tmerc +lat_0=0 +lon_0=15.8082777777778 +k=1 +x_0=1500000 +y_0=0 
+ellps=bessel +towgs84=414.1,41.3,603.1,-0.855,2.141,-7.023,0 +units=m +no_defs 
+type=crs',
     'EPSG:3006'
   ));
   ```
   
   Verified with the released Sedona 1.9.0 shaded artifact, this returns:
   
   ```text
   POINT (320728.44351324247 6400228.104394391)
   ```
   
   This matches the reported PostGIS result. For real data, replace 
`ST_Point(...)` with the geometry column.
   
   Alternatively, if the GeoTools wrapper is installed, temporarily use the 
legacy backend:
   
   ```sql
   spark.sedona.crs.geotools=all;
   ```
   
   Note that this routes every vector `ST_Transform` call through GeoTools.


-- 
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]

Reply via email to