Hi all,
We are pleased to announce the release of 3.10.0.
This is the first release of the 3.10.x series. Pre-built wheels are available for most major platforms, and can be installed using `pip install matplotlib==3.10.0`. Other packages may also be available already; please check with your preferred source.
We thank the 128 authors for the 337 pull requests that comprise the 3.10.0 release.
Highlights of this release include:
- Preliminary support for free-threaded CPython 3.13
- New more-accessible color cycle
- Dark-mode diverging colormaps
- Exception handling control
- InsetIndicator artist
- FillBetweenPolyCollection
- Fill between 3D lines
- Data in 3D plots can now be dynamically clipped to the axes view limits
- Rotating 3d plots with the mouse
- Increased Figure limits with Agg renderer
- Subfigures no longer provisional
- Subfigures are now added in row-major order
Please see the milestone on GitHub for more details:
https://github.com/matplotlib/matplotlib/milestone/84?closed=1
This release is the third to have published attestations on both GitHub and PyPI. At this time, the infrastructure issue that affect 3.9.3 appears to be fixed, so attestations should be available for all wheels and the sdist.
--
Kyle Sunden
Hi all,
We are pleased to announce the release of 3.9.4.
This is the fourth (and final planned) bugfix release of the 3.9.x
series. Pre-built wheels are available for most major platforms, and can
be installed using `pip install matplotlib==3.9.4`. Other packages may
also be available already; please check with your preferred source.
We thank the 3 authors for the 3 pull requests that comprise the 3.9.4
release.
This release contains two bug-fixes and adjustments:
* Fix toolbar icons in GTK backend
* Fix Poly3DCollection initialization with list of lists
As this release is mostly a packaging fix for the toolbar icons, and the
3.10 release is imminent, I did not build new docs for 3.9.4. Please see
the milestone on GitHub for more details:
https://github.com/matplotlib/matplotlib/milestone/95?closed=1
This release is the second to have published attestations on both GitHub
and PyPI. At this time, the infrastructure issue that affect 3.9.3
appears to be fixed, so attestations should be available for all wheels
and the sdist.
--
Elliott
Hi all,
Apologies for the delay in these release announcements.
We are pleased to announce the release of 3.9.3 (and 3.9.2, below).
This is the third bugfix release of the 3.9.x series. Pre-built wheels
are available for most major platforms, and can be installed using `pip
install matplotlib==3.9.3`. Other packages may also be available
already; please check with your preferred source.
We thank the 18 authors for the 62 pull requests that comprise the 3.9.3
release.
This release contains several bug-fixes and adjustments:
* Fix |axline| with extremely small slopes
* Fix |axline| with non-linear axis scales
* Fix |minimumSizeHint| with Qt backend
* Fix config directory usage when it's behind a symlink
* Fix draggable legend when blitting is enabled
* Fix high CPU utilization in the |macosx| backend
* Fix multiple hatch |edgecolors| passed to |contourf|
* Improve compatibility with |pytest| 8.2.0
For further details, please see the Release notes for Matplotlib 3.9.3:
https://matplotlib.org/3.9.3/users/release_notes.html#version-3-9
and the milestone on GitHub:
https://github.com/matplotlib/matplotlib/milestone/95?closed=1
This release is the first to have published attestations on both GitHub
and PyPI. Unfortunately, due to an infrastructure issue, attestations
for PyPy wheels and the sdist are not available on PyPI, though those
for all CPython wheels are available. For those that need them,
alternative attestations are available from GitHub:
https://github.com/matplotlib/matplotlib/attestations We hope that said
infrastructure issues will be fixed for the 3.10.0 release.
The following are the release notes for 3.9.2:
* Be more resilient to I/O failures when writing font cache
* Fix nondeterministic behavior with subplot spacing and constrained
layout
* Fix sticky edge tolerance relative to data range
* Improve formatting of image values in cases of singular norms
Windows wheels now bundle the MSVC runtime DLL statically to avoid
inconsistencies with other wheels and random crashes depending on import
order.
For further details, please see the Release notes for Matplotlib 3.9.2:
https://matplotlib.org/3.9.2/users/release_notes.html#version-3-9
and the milestone on GitHub:
https://github.com/matplotlib/matplotlib/milestone/93?closed=1
--
Elliott
