Hatch linewidths in an rcParam context #21325
Conversation
in accordance what we've already done in the top-level menu.
- use a marker and no line for the data points - hard-code yerr instead of using random with a fixed seed - Slightly move y so that the data points don't always fall on the grid - Remove "specialized" from section title. There are no more special than any other plots.
The original thumbnail margin in gallery.css is 5px. We reduce the horizontal margin to 2px so that four columns fit within the available width in the pydata-sphinx-theme.
…ypes Improve errorbar plot types example
…arg-support-mplot3d [ENH]: data kwarg support for mplot3d matplotlib#20912
Fix make_norm_from_scale `__name__` when used inline.
Rename section title Gallery -> Examples
Co-authored-by: Thomas A Caswell <tcaswell@gmail.com>
This has been the case since at least mpl2.0, so let's just fix the doc for now. Likely there was some confusion with zoom/pan mode, where CTRL does fix the aspect.
- Most 2D data examples don't need Z.min() subtraction (the normalization adjusts itself as needed). - Tweak spacing in the expression for Z to be the same everywhere. - Use shorter .min(), .max() methods instead of free functions.
Capitalization fixes in example section titles.
Accepting to call _process_plot_format() (which is unlikely to be the slowest step here) even in the default case makes the code logic much simpler.
Simplify/uniformize sample data setup in plot_types examples.
When using the default padding and the pydata-sphinx-theme, the image is shifted away from the center. While that should be fixed in sphinx-gallery (and I will report it there), this serves as a workaround on the matplotlib side, because I don't think they will have it fixed ready for the 3.5 release.
…ding DOC: Fix decenter of image in gallery thumbnails
…riented DOC: clarify what we mean by object oriented in pyplot api
- remove contour plot for simplicity. - don't set zorder. - Z doesn't need to be made positive.
Add contour and tricontour plots to plot types
Tweak streamplot plot_types example.
- `__scale = "manual"` doesn't exist anymore since d1c5a6a. - Shorten Boundary/NoNorm inversion and norm._scale check. - No need to define `x` well before its use point, and no need to make it an array explicitly.
Fix polar() regression on second call failure
The zoom and pan funcitons change the vmin/vmax of the norm attached to the colorbar. The colorbar is rendered as an inset axis, but the event handler is implemented on the parent axis.
This helps for subclasses in finding the zoom/pan locations by not having to duplicate the code used to determine the x/y locations of the zoom or pan.
Setting the zoom selector rectangle to the vertical/horizontal limits of the colorbar, depending on the orientation. Remove some mappable types from colorbar navigation. Certain mappables, like categoricals and contours shouldn't be mapped by default due to the limits of an axis carrying certain meaning. So, turn that off for now and potentially revisit in the future.
Adding tests for vertical and horizontal placements, zoom in, zoom out, and pan. Also verifying that a colorbar on a Contourset is not able to be interacted with.
Pin macOS to 10.15 for wheels
Make `Path.__deepcopy__` interact better with subclasses, e.g. TextPath.
Doing out-of-tree builds makes a copy that breaks symlinks in the git tree, causing PyPy wheels to get a version that signifies a dirty tree when they shouldn't. This will change in pip soon, but changing this will fix the version for our release now.
Use in-tree builds for PyPy wheels
Update link to Agg website
Simplify `Colormap.__call__` a bit.
![github-actions[bot]](https://avatars.githubusercontent.com/in/15368?s=60&v=4){kind=small}
There was a problem hiding this comment.
Thank you for opening your first PR into Matplotlib!
If you have not heard from us in a while, please feel free to ping @matplotlib/developers or anyone who has commented on the PR. Most of our reviewers are volunteers and sometimes things fall through the cracks.
You can also join us on gitter for real-time discussion.
For details on testing, writing docs, and our review process, please see the developer guide
We strive to be a welcoming and open project. Please follow our Code of Conduct.
|
Your test failure seems to mean that your hatch output is not the same each time? Note you will also need to add tests to this to test the behaviour change, and at some point it will need a whats-new. Is it also backwards incompatible? If so, you need to work out how to make the transition. Just applying @anntzer's patch is not enough here. |
|
This PR is affected by a re-writing of our history to remove a large number of accidentally committed files see discourse for details. To recover this PR it will need be rebased onto the new default branch (main). There are several ways to accomplish this, but we recommend (assuming that you call the matplotlib/matplotlib remote git remote update
git checkout main
git merge --ff-only upstream/main
git checkout YOUR_BRANCH
git rebase --onto=main upstream/old_master
# git rebase -i main # if you prefer
git push --force-with-lease # assuming you are tracking your branchIf you do not feel comfortable doing this or need any help please reach out to any of the Matplotlib developers. We can either help you with the process or do it for you. Thank you for your contributions to Matplotlib and sorry for the inconvenience. |
|
Closing in favour of #25367. |
PR Summary
Have added the suggested additions in order to solve Issue #21108
PR Checklist
pytestpasses).flake8on changed files to check).flake8-docstringsand runflake8 --docstring-convention=all).doc/users/next_whats_new/(follow instructions in README.rst there).doc/api/next_api_changes/(follow instructions in README.rst there).