@@ -79,7 +79,6 @@ As usual, install using pip:
79
79
80
80
Note that wheels are not available for macOS<10.13, because the libc++ included
81
81
with these versions is too old and vendoring of libc++ appears to be fragile.
82
- Help for packaging would be welcome.
83
82
84
83
mplcairo can use Raqm _ (≥0.7.0; ≥0.7.2 is recommended as it provides better
85
84
emoji support, especially in the presence of ligatures) for complex text layout
@@ -198,9 +197,6 @@ delocate-wheel_ (to vendor a recent version of libc++). Currently, it can only
198
197
be built from a Homebrew-clang wheel, not a conda-clang wheel (due to some path
199
198
intricacies...).
200
199
201
- As I can personally only test the macOS build on CI, any help with the build
202
- and the packaging on that platform would be welcome.
203
-
204
200
.. _delocate-wheel : https://github.com/matthew-brett/delocate
205
201
206
202
Windows
@@ -286,10 +282,10 @@ specifically, the following backends are provided:
286
282
- ``module://mplcairo.macosx `` (macOS widget, copying data from a cairo image
287
283
surface).
288
284
289
- On macOS, **it is necessary to explicitly import mplcairo before importing
290
- Matplotlib ** due to incompatibilities associated with the use of a recent
291
- libc++. As such, the most practical option is to import mplcairo, then call
292
- e.g. ``matplotlib.use("module://mplcairo.macosx") ``.
285
+ On macOS, prior to Matplotlib 3.8, **it was necessary to explicitly import
286
+ mplcairo before importing Matplotlib ** (unless your Matplotlib is built with
287
+ `` system_freetype = True ``). A practical option was to import mplcairo, then
288
+ call e.g. ``matplotlib.use("module://mplcairo.macosx") ``.
293
289
294
290
Jupyter is entirely unsupported (patches would be appreciated). One
295
291
possibility is to set the ``MPLCAIRO_PATCH_AGG `` environment variable to a
0 commit comments