This is an automated email from the ASF dual-hosted git repository.
jiayu pushed a commit to branch master
in repository https://gitbox.apache.org/repos/asf/sedona.git
The following commit(s) were added to refs/heads/master by this push:
new 179c87cadf [GH-2062] Fix failing Geopandas tests after recent PR
merges (#2063)
179c87cadf is described below
commit 179c87cadf2e57858841b095495abe0b0930ef0c
Author: Peter Nguyen <[email protected]>
AuthorDate: Sat Jul 5 21:10:24 2025 -0700
[GH-2062] Fix failing Geopandas tests after recent PR merges (#2063)
* Convert to wkb for ps.Series case and don't pass self.crs in to_geopandas
* Catch np.nan for .crs
---
python/sedona/geopandas/geoseries.py | 19 +++++++++++++++++--
1 file changed, 17 insertions(+), 2 deletions(-)
diff --git a/python/sedona/geopandas/geoseries.py
b/python/sedona/geopandas/geoseries.py
index 805b827fb8..f7ad400806 100644
--- a/python/sedona/geopandas/geoseries.py
+++ b/python/sedona/geopandas/geoseries.py
@@ -140,8 +140,23 @@ class GeoSeries(GeoFrame, pspd.Series):
# This is NOT a scalable solution since we call to_pandas() on the
data and is a hacky solution
# but this should be resolved if/once
https://github.com/apache/spark/pull/51300 is merged in.
# For now, we reset self._anchor = data to have keep the geometry
information (e.g crs) that's lost in to_pandas()
+
+ pd_data = data.to_pandas()
+
+ # If has shapely geometries, convert to wkb since
pandas-on-pyspark can't understand shapely geometries
+ if (
+ isinstance(pd_data, pd.Series)
+ and any(isinstance(x, BaseGeometry) for x in pd_data)
+ ) or (
+ isinstance(pd_data, pd.DataFrame)
+ and any(isinstance(x, BaseGeometry) for x in
pd_data.values.ravel())
+ ):
+ pd_data = pd_data.apply(
+ lambda geom: geom.wkb if geom is not None else None
+ )
+
super().__init__(
- data=data.to_pandas(),
+ data=pd_data,
index=index,
dtype=dtype,
name=name,
@@ -214,7 +229,7 @@ class GeoSeries(GeoFrame, pspd.Series):
tmp_df = self._process_geometry_column("ST_SRID", rename="crs")
srid = tmp_df.take([0])[0]
# Sedona returns 0 if doesn't exist
- return CRS.from_user_input(srid) if srid else None
+ return CRS.from_user_input(srid) if srid != 0 and not pd.isna(srid)
else None
@crs.setter
def crs(self, value: Union["CRS", None]):
