Closed vlandau closed 1 year ago
I'm now noticing that passing a resolution results in a DataArray that has two additional empty indices, each with length 0 (band and time). It would appear there is something wrong with the STAC items.
Thanks @vlandau, I can reproduce it. Because of the incorrect metadata, it looks to stackstac like the item doesn't overlap with the bounds you've specified. However, there's a bug in stackstac when 0 items overlap. I'll look into how to fix that.
Thanks for looking into this @gjoseph92! Let me know if there's anything I can do to help out.
Just merged a fix. I confirmed it works for your example, so no need to test it, but if you need something that works, go ahead and install from main for now.
Awesome, thanks for such a speedy resolution to this @gjoseph92, and that's for all your work on this software package!
I'm currently working with some elevation data from Microsoft's Planetary Computer STAC, and I'm getting a stackstac error only for a specific area/tile (other AOIs seem to work).
To reproduce:
Output:
As you might expect based on the stacktrace, manually setting the resolution in the
stack
call avoids the issue. Notably, it is successful when the bbox is set to(-97.73474, 33.785180000000004, -97.71473999999999, 33.80518)
Could something be wrong with this particular item in the STAC?
Version/environment info:
Python: 3.10.6
Package versions: