Bug #109 forcing with DatetimeNoLeap #110
Conversation
|
@DirkEilander : could you review this small bugfix? You reviewed this new change from #97 so you should already know more or less what it's about :) Many thanks! |
|
Tests should be fixed when merging Deltares/hydromt#211 in hydromt |
The test fails on another issue it seems, see #113 |
There was a problem hiding this comment.
Thanks for these fixes @hboisgon.
I think it's basically good to go. But strongly suggest to remove the lines that assume self.crs still has the rasterio CRS dtype as these will never be hit (see comment)
hydromt_wflow/wflow.py
Outdated
| elif hasattr(self.crs, "is_epsg_code"): | ||
| if self.crs.is_epsg_code: | ||
| code = int(self.crs["init"].lstrip("epsg:")) |
There was a problem hiding this comment.
I think these lines can be removed because self.crs is always of pyproj.CRS type.
For the RasterModel see https://github.com/Deltares/hydromt/blob/075ac08d91f7e06bdc0504c0163d450e7b23a955/hydromt/models/model_grid.py#L181
For the Model see
https://github.com/Deltares/hydromt/blob/075ac08d91f7e06bdc0504c0163d450e7b23a955/hydromt/models/model_api.py#L1278
There was a problem hiding this comment.
This is true for hydromt >= 0.5.0 but not for older version I think. I don't mind removing I think we put pins everywhere already in the different env files.
See #109 : solution is to use xr.to_datetimeindex instead of pd.to_datetime to convert times in forcing DataArrays