Fix bar() named shape-mismatch error under NumPy nightlies - #32431
Open
mitre88 wants to merge 1 commit into
Open
mitre88 wants to merge 1 commit into
mitre88 wants to merge 1 commit into
Conversation
Copy link
Copy Markdown
Sorry, something went wrong.
NumPy nightlies changed broadcast ValueError text so it no longer includes "arg N", which broke the string-replace remapping in Axes.bar and failed test_bar_shape_mismatch (matplotlib#32430). Validate broadcastability with np.broadcast_shapes and raise a named ValueError from the first non-broadcastable argument pair, instead of parsing NumPy's wording. Successful broadcasts are unchanged. Fixes matplotlib#32430
timhoffm
reviewed
Oct 4, 2026
timhoffm
left a comment
Copy link
Copy Markdown
Member
There was a problem hiding this comment.
Let's keep this simpler. Note: We do not guarantee stability of exception messages.
Sorry, something went wrong.
Comment thread
lib/matplotlib/axes/_axes.py
Comment on lines
+2531
to
+2546
| np.broadcast_shapes(*shapes) | ||
| except ValueError: | ||
| # Report the first pair (in argument order) that cannot broadcast, | ||
| # matching the historical named message users and tests expect. | ||
| for i in range(len(shapes)): | ||
| for j in range(i + 1, len(shapes)): | ||
| try: | ||
| np.broadcast_shapes(shapes[i], shapes[j]) | ||
| except ValueError: | ||
| raise ValueError( | ||
| "shape mismatch: objects cannot be broadcast to a " | ||
| "single shape. Mismatch is between " | ||
| f"'{arg_names[i]}' with shape {shapes[i]} and " | ||
| f"'{arg_names[j]}' with shape {shapes[j]}." | ||
| ) from None | ||
| raise |
Copy link
Copy Markdown
Member
There was a problem hiding this comment.
I think this is too much effort. Instead we should:
- still directly to np.broadcast_arrays - it's better to ask for foregiveness than for permission
- do not try to interpret the original error string
- raise a ValueError from the original error only providing the context
i.e.
except ValueError as e:
non_trivial_inputs = ", ".join(
f"{name} {np.shape(arg)}" for arg, name in zip(args, arg_names)
# filter out the args that could contribute to a broadcast error (i.e. non-trival shapes)
if any(dim != 1 for dim in np.shape(arg))
)
raise ValueError(f"bar() failed on broadcasting inputs: {non_trivial_inputs}") from e
Sorry, something went wrong.
This branch has not been deployed
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Reviewers
At least 1 approving review is required to merge this pull request.
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
PR summary
Fixes #32430.
Axes.bar used to catch NumPy's broadcast ValueError and rewrite "arg 0" → "'x'", "arg 1" → "'height'", etc. NumPy nightlies changed that message to something like operands could not be broadcast together with shapes (2,) (3,) () () () () (no arg N), so the remapping no longer ran and test_bar_shape_mismatch failed with the cryptic raw text.
This PR stops parsing NumPy's wording. Before np.broadcast_arrays, it checks shapes with np.broadcast_shapes and, on failure, finds the first non-broadcastable argument pair in order and raises:
shape mismatch: objects cannot be broadcast to a single shape. Mismatch is between 'x' with shape (2,) and 'height' with shape (3,).
Successful broadcasts are unchanged. test_bar_shape_mismatch keeps the existing x/height case and adds a width mismatch case.
AI Disclosure
AI-assisted (Cursor agent under my direction). I specified the root cause, required fix (shape-based named error, not a test-only regex change), and reviewed the resulting patch and tests myself.
PR quality check