jiayuasu opened a new issue, #3061: URL: https://github.com/apache/sedona/issues/3061
Part of #3054. ## Scope Expose the geography constructor/conversion functions in SedonaFlink, registering them so they are callable from the Flink Table API / SQL — reaching parity with the 9 functions Spark exposes. Depends on #3058 (GeographyTypeSerializer), now merged. ## Functions - [ ] `ST_GeogFromWKT` (wkt, [srid]) - [ ] `ST_GeogFromText` (wkt, [srid]) - [ ] `ST_GeogFromEWKT` (ewkt) - [ ] `ST_GeogCollFromText` (wkt, [srid]) - [ ] `ST_GeogFromWKB` (wkb, [srid]) - [ ] `ST_GeogFromEWKB` (wkb) - [ ] `ST_GeogFromGeoHash` (geohash, [precision]) - [ ] `ST_GeogToGeometry` (geography) → geometry - [ ] `ST_GeomToGeography` (geometry) → geography ## Tasks - [ ] Add `ScalarFunction` wrappers in `flink/src/main/java/org/apache/sedona/flink/expressions/GeographyConstructors.java`, each delegating to `org.apache.sedona.common.geography.Constructors` with `@DataTypeHint(value = "RAW", rawSerializer = GeographyTypeSerializer.class, bridgedTo = Geography.class)`, matching Spark's optional-argument defaults (srid `0`, geohash precision `null`). - [ ] Register all 9 in `flink/src/main/java/org/apache/sedona/flink/Catalog.java`. - [ ] Add tests round-tripping each function through the Flink type system (`GeographyConstructorTest`). Docs are tracked separately as the next sub-task. -- 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]
