Frequently I have a large list of geometry objects I need to get reprojected for analysis. A while back I discovered the `reprojectAs`

method of the `arcpy.Geometry`

object. While useful, I still needed a way to handle a few edge cases to make a more universal way to reproject. These edge cases include varying spatial references across the list of geometry objects, and applying a transformation when needed - when the geographic coordinate systems for the input and output are different. The Gist below is what I came up with to accomplish this task.

Most commonly I use this when trying to quickly work with a large volume of points collected from a variety of sources online, and get them in the same spatial reference for loading into a feature class, or more commonly perform a few analysis steps to validate, and clean up the data. Simply to play with this and test it is relatively easy to just load geometries from a feature class using a data analysis search cursor, and applying this function such as the example below, going from Web Mercator to just WGS84.

```
fc_path = r'.\test01.gdb\test_points'
sr_out = arcpy.SpatialReference(3857)
geom_list = [_[0] for _ in arcpy.da.SearchCursor(fc_path, '[email protected]')]
print(set(geom.spatialReference.name for geom in geom_list))
geom_list = [reproject_geometry_object(geom, sr_out) for geom in geom_list]
print(set(geom.spatialReference.name for geom in geom_list))
```

The output simply prints out the starting spatial references, and the ending spatial references. In my test case, there is only one for both. Using this function, the output will *always* be a single spatial reference anyway. Here is the example output for my test.

```
{'GCS_WGS_1984'}
{'WGS_1984_Web_Mercator_Auxiliary_Sphere'}
```

Incidentally, if you are wondering why I am using these lists of geometry objects, it is a lot faster than performing analysis with entire feature class datasets in geoprocessing tools. A good place to start understanding how to do this is available in the ArcGIS Documentation, *Using Geometry Objects with Geoprocessing Tools*.

While a bit of an esoteric topic, if this is something you find yourself in need of, typically you *really* need it. Hence, hopefully my struggles make your life easier!