Skip to content

Fix divide by 0 runtime warning #23430

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 1 commit into from
Jul 18, 2022
Merged

Fix divide by 0 runtime warning #23430

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 1 addition & 1 deletion lib/mpl_toolkits/mplot3d/proj3d.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -21,7 +21,7 @@ def _line2d_seg_dist(p1, p2, p0):

x01 = np.asarray(p0[0]) - p1[0]
y01 = np.asarray(p0[1]) - p1[1]
if np.all(p1 == p2):
if np.all(p1[0:2] == p2[0:2]):
return np.hypot(x01, y01)

x21 = p2[0] - p1[0]
Expand Down
10 changes: 6 additions & 4 deletions lib/mpl_toolkits/tests/test_mplot3d.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1011,11 +1011,13 @@ def test_lines_dists():
def test_lines_dists_nowarning():
# Smoke test to see that no RuntimeWarning is emitted when two first
# arguments are the same, see GH#22624
p0 = (10, 30)
p1 = (20, 150)
proj3d._line2d_seg_dist(p0, p0, p1)
p0 = (10, 30, 50)
p1 = (10, 30, 20)
p2 = (20, 150)
proj3d._line2d_seg_dist(p0, p0, p2)
proj3d._line2d_seg_dist(p0, p1, p2)
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Um, this now accepts a 3d line and a 2d point? I'm not clear what the semantics here is or whether this is desired. At the very least we should document the expected input dimensions for the function.

Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

This is an internal helper that I think we pulled out.

Given that @scottshambaugh was seeing this warning we are apparently passing in 3D points in practice.

While this is an odd internal wart, I do not think it is worth the effort we have collectively thrown at this function recently!

Copy link
Member

@timhoffm timhoffm Jul 18, 2022 •
edited
Loading

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

My point is that we are doing some ill-defined things in the function. Note that this function is only used in format_coord and I wouldn't be surprised if it's the cause of #22775.

The earlier work #22628 was defensive in only bailing out when both line-defining points were equal in 3d. That does not alter the behavior - it's not making anything worse at least.

I don't know what the correct behavior here should be, but I believe that neglecting the z-direction will add another quirk to the coord determination.
Edit: Rethinking this, since we are anyway only measuring the distance in x, y. This check is not making it worse either. - As little as that is worth.

More generally, I support @anntzer's view that we should stop displaying 3D coords, because we can't get them from a 2d pointer consistently.

p0 = np.array(p0)
proj3d._line2d_seg_dist(p0, p0, p1)
proj3d._line2d_seg_dist(p0, p0, p2)


def test_autoscale():
Expand Down