Pytest style analysis/test_distances.py #1541
Conversation
| assert_equal(contacts.shape, self.shape, | ||
| "wrong shape (should be {0})".format(self.shape)) | ||
| contacts = MDAnalysis.analysis.distances.contact_matrix(self.coord, cutoff=1, returntype="numpy") | ||
| assert contacts.shape == self.shape, "wrong shape (should be {0})".format(self.shape) |
There was a problem hiding this comment.
This looks like you break the 80 char line limit. Also these tests look like they could be parametrized.
There was a problem hiding this comment.
@kain88-de All the 4 methods have different check in the assert_equal statement. If you want to parametrize, I'll have to send both the values that will be compared. I don't think that's a good idea.
There was a problem hiding this comment.
called np.asarray(...) on the contacts should also work for the sparse case
| [1, 0, 1, 1, 1]], dtype=np.bool) | ||
|
|
||
| class TestContactMatrix(object): | ||
| coord = np.array([[1, 1, 1], |
There was a problem hiding this comment.
The problem with declaring arrays in the class namespace like this is they can be accidentally modified by a test, and they will remain changed for all other tests (I think?)
| del self.expected | ||
|
|
||
| class TestDist(object): | ||
| u = MDAnalysis.Universe(GRO) |
There was a problem hiding this comment.
Same here, we shouldn't just be loading everything into the class namespace
|
@richardjgowers Better now? |
utkbansal
changed the title
|
@utkbansal the coverage went down with the latest changes |
|
@kain88-de I don't see how that happened, have restarted the build. |
Fixes #
Changes made in this Pull Request:
PR Checklist