Hi folks,
We observed a segfault in STRtree within the geopandas CI. See run here. Setting empty geometries to None before building the tree resolved the issue.
Strangely, the issue seems to only happen with Pandas == 0.24, so it might not be pygeos' fault, but I'm going to report it here because the segfault happens within pygeos.
I was not able to reproduce this locally using the latest Python 3.6 and Pandas == 0.24, so it might also be related to the specific Python binary or version Travis uses (I think it's not the latest).
If someone does decide to look into this, let me know if there is anything you need from me.
Hi folks,
We observed a segfault in STRtree within the geopandas CI. See run here. Setting empty geometries to
Nonebefore building the tree resolved the issue.Strangely, the issue seems to only happen with
Pandas == 0.24, so it might not be pygeos' fault, but I'm going to report it here because the segfault happens within pygeos.I was not able to reproduce this locally using the latest Python 3.6 and Pandas == 0.24, so it might also be related to the specific Python binary or version Travis uses (I think it's not the latest).
If someone does decide to look into this, let me know if there is anything you need from me.