Issue #1437 fix recharge from imod5 cap data 2d #1448
Conversation
imod/tests/test_mf6/test_mf6_rch.py
Outdated
| assert np.isnan(rate[:, -1]).all() | ||
| assert np.isnan(rate[0, :]).all() | ||
| assert np.isnan(rate[-1, :]).all() | ||
| assert np.isnan(rate[100, 100]).all() |
There was a problem hiding this comment.
Isn't this one already tested by rate[-1, :] and rate[:, -1]?
There was a problem hiding this comment.
No, the grid is size 184, 184. So this is a central cell where areas have been set in such a way the cell needs to be deactivated. I set this more explicit now in the test by computing a midpoint cell and assigning values to that.
| assert np.isnan(rate[0, 0, :]).all() | ||
| assert np.isnan(rate[0, -1, :]).all() | ||
| assert np.isnan(rate[:, 100, 100]).all() | ||
| assert np.isnan(rate[:, 0]).all() |
There was a problem hiding this comment.
What happend to the extra dimension? Are we using a disv grid now? Or did you drop a coordinate?
There was a problem hiding this comment.
We did drop the dimension, only the coordinate label is preserved. That reminds me: probably good to test for that as well!
| assert np.isnan(rate[0, 0, :]).all() | ||
| assert np.isnan(rate[0, -1, :]).all() | ||
| assert np.isnan(rate[:, 100, 100]).all() | ||
| assert np.isnan(rate[:, 0]).all() |
There was a problem hiding this comment.
We did drop the dimension, only the coordinate label is preserved. That reminds me: probably good to test for that as well!
imod/tests/test_mf6/test_mf6_rch.py
Outdated
| assert np.isnan(rate[:, -1]).all() | ||
| assert np.isnan(rate[0, :]).all() | ||
| assert np.isnan(rate[-1, :]).all() | ||
| assert np.isnan(rate[100, 100]).all() |
There was a problem hiding this comment.
No, the grid is size 184, 184. So this is a central cell where areas have been set in such a way the cell needs to be deactivated. I set this more explicit now in the test by computing a midpoint cell and assigning values to that.
|
Fixes #1437
Description
primodexpects a 2D rate array or single layer array for the dummy MODFLOW6 package, as otherwise thesqueeze("layer")it does returns an error. I think the safest for iMOD Python is to return a 2D array with extra layer coordinate. I added some extra tests to test if this RCH package could be clipped and regridded, as in most/all fixtures for clipping/regridding I added packages with layer dimension. Luckily these two extra tests turned out to be mostly for my own peace of mind, as they ran without problems.Checklist
Issue #nr, e.g.Issue #737