Huite

Fixed by d0b9459ed05b51e94a847fb44fe5efcaabb4f41d numba_celltree needs to be updated in terms of its supported Python versions.

Good point, re-opening...

1. `xr.DataArray.notnull(uda)` seems quite obscure, generally only `uda.notnull()` will be called. 2. The xarray top level functions should be explicitly wrapped, I've made a start with ones_like, zeros_like, etc. 3....

I must admit I'm a little confused by the introduction of these attributes: they seem fully redundant, because the dimension already associates data with a specific aspect (node, edge, face)...

Regarding this point: > D-EcoImpact also uses automatic topology dimension matching like xugrid. In that case please point us to the code that does this. Xugrid infers the dimension based...

Ah okay, with your example I get what you're after. I considered something like that in the past, I think you can get all of this in a much nicer...

(Mostly a drive-by comment, saw this pop up in my notifications...) This makes some sense for CSV: every digit is a byte, but the gains for netCDF seem pretty marginal....

The easiest way to identify this is via `grid_mapping_name` as far as I can tell. Best to check cf-xarray and rioxarray.

The bounds attr should be present in the face_x and face_y coordinate rather than the node coordinates. The following works for visualizing in FEWS: ```python grid = da_ptid.ugrid.grid bounds =...

From the vector conversion examples: ```python lines = gpd.GeoDataFrame(geometry=provinces.exterior) length = xu.length_of_intersection(lines, uda) fig, ax = plt.subplots() length.ugrid.plot(ax=ax) length.ugrid.plot.line(ax=ax, edgecolor="black", linewidth=0.5) provinces.plot(ax=ax, edgecolor="red", facecolor="none", linewidth=1.5) ax.set_xlim(xmin, xmax) ax.set_ylim(ymin, ymax) ```...