Thanks to visit codestin.com
Credit goes to github.com

Skip to content

Fix #21409: Make twin axes inherit parent position #31353

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
QuLogic merged 7 commits into from
Apr 10, 2026
Merged

Fix #21409: Make twin axes inherit parent position #31353

Show file tree
Hide file tree
Changes from 2 commits
Commits
Show all changes
7 commits
Select commit Hold shift + click to select a range
d9ef7af
Fix #21409: Make twin axes inherit parent position
maf2310 Mar 23, 2026
88d4ec5
Refine twin position fix to preserve layout participation
maf2310 Mar 24, 2026
a5435b5
Address review: preserve original/active positions and fix test + lint
maf2310 Mar 25, 2026
ce749b4
Clarify manual-position twin sync in code comment
maf2310 Mar 25, 2026
5876d06
Add regression test for twinx top spine position
maf2310 Mar 27, 2026
e8df9e5
Add test to ensure twin axes remain aligned after tight_layout
maf2310 Mar 29, 2026
c2a343c
Add tests for twin axes alignment with manual positioning_layout
maf2310 Mar 30, 2026
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
3 changes: 3 additions & 0 deletions lib/matplotlib/axes/_base.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -4657,6 +4657,9 @@
twin.set_zorder(self.zorder)

self._twinned_axes.join(self, twin)
if not self.get_in_layout():
Comment thread
maf2310 marked this conversation as resolved.
Copy link
Copy Markdown
Member

@jklymak jklymak Mar 27, 2026

Choose a reason for hiding this comment

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

I think this is all fine, but needs to be tested.

Copy link
Copy Markdown
Contributor Author

@maf2310 maf2310 Mar 27, 2026

Choose a reason for hiding this comment

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

Thanks! I’ll add additional tests to cover this case.

Copy link
Copy Markdown
Contributor Author

@maf2310 maf2310 Mar 27, 2026

Choose a reason for hiding this comment

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

I added tests to cover both manual positioning and layout-managed cases. Let me know if anything else should be tested.

Copy link
Copy Markdown
Member

@jklymak jklymak Mar 29, 2026

Choose a reason for hiding this comment

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

I'd prefer an actual test that the positions match after layout is applied.

Copy link
Copy Markdown
Contributor Author

@maf2310 maf2310 Mar 29, 2026

Choose a reason for hiding this comment

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

ok, I'll add it

Copy link
Copy Markdown
Contributor Author

@maf2310 maf2310 Mar 29, 2026

Choose a reason for hiding this comment

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

I added a test that explicitly checks that the twin and parent positions
still match after tight_layout() is applied.
Let me know if I should add anything else.

twin.set_position(self.get_position())
Copy link
Copy Markdown
Contributor

@sanrishi sanrishi Mar 24, 2026

Choose a reason for hiding this comment

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

Thanks for the update the not self.get_in_layout() guard looks right and keeps the fix scoped to manual positioning. One small concern: twin.set_position(self.get_position()) sets both original and active positions. If the parent only changed the active position (set_position(..., which="active")), this would also overwrite the twin’s original.

Should we consider _set_position(..., which="active") or explicitly using get_position(original=True) depending on intended semantics?


return twin

def twinx(self, axes_class=None, **kwargs):
Expand Down Expand Up @@ -4723,7 +4726,7 @@
axes_class : subclass type of `~.axes.Axes`, optional
The `.axes.Axes` subclass that is instantiated. This parameter
is incompatible with *projection* and *polar*. See
:ref:`axisartist_users-guide-index` for examples.

Check warning on line 4729 in lib/matplotlib/axes/_base.py

View workflow job for this annotation

GitHub Actions / ruff 

[rdjson] reported by reviewdog 🐶
Trailing whitespace

Raw Output:
message:"Trailing whitespace"  location:{path:"/home/runner/work/matplotlib/matplotlib/lib/matplotlib/axes/_base.py"  range:{start:{line:4729  column:51}  end:{line:4729  column:52}}}  severity:WARNING  source:{name:"ruff"  url:"https://docs.astral.sh/ruff"}  code:{value:"W291"  url:"https://docs.astral.sh/ruff/rules/trailing-whitespace"}  suggestions:{range:{start:{line:4729  column:51}  end:{line:4729  column:52}}}

By default, `~.axes.Axes` is used.

Expand Down
10 changes: 10 additions & 0 deletions lib/matplotlib/tests/test_axes.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -477,6 +477,16 @@ def test_twin_inherit_autoscale_setting():
assert not ax_y_off.get_autoscaley_on()


@pytest.mark.parametrize("twin", ("x", "y"))
def test_twin_respects_position_after_set_position(twin):
fig, ax = plt.subplots()

ax.set_position([0.2, 0.2, 0.5, 0.5])
ax2 = getattr(ax, f"twin{twin}")()

assert_allclose(ax.bbox.bounds, ax2.bbox.bounds)
Copy link
Copy Markdown
Contributor

@sanrishi sanrishi Mar 24, 2026

Choose a reason for hiding this comment

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

Also, the new test compares ax.bbox.bounds without forcing a draw; that can be backend‑sensitive. Would you consider comparing ax.get_position().bounds instead, or calling fig.canvas.draw() before comparing bboxes?



def test_inverted_cla():
# GitHub PR #5450. Setting autoscale should reset
# axes to be non-inverted.
Expand Down
Loading