Skip to content

ENH: switch mpl_toolkits to implicit namespace package (PEP 420) #25381

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.

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
merged 1 commit into from
Jun 8, 2023
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
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
6 changes: 6 additions & 0 deletions doc/users/next_whats_new/mpl_toolkit_pep420.rst
Original file line number Diff line number Diff line change
@@ -0,0 +1,6 @@
``matplotlib.mpl_toolkits`` is now an implicit namespace package
----------------------------------------------------------------

Following the deprecation of ``pkg_resources.declare_namespace`` in ``setuptools`` 67.3.0,
``matplotlib.mpl_toolkits`` is now implemented as an implicit namespace, following
`PEP 420 <https://peps.python.org/pep-0420/>`_.
4 changes: 0 additions & 4 deletions lib/mpl_toolkits/__init__.py

This file was deleted.

8 changes: 5 additions & 3 deletions setup.py
Original file line number Diff line number Diff line change
Expand Up @@ -29,7 +29,7 @@
import shutil
import subprocess

from setuptools import setup, find_packages, Distribution, Extension
from setuptools import setup, find_namespace_packages, Distribution, Extension
import setuptools.command.build_ext
import setuptools.command.build_py
import setuptools.command.sdist
Expand Down Expand Up @@ -301,8 +301,10 @@ def make_release_tree(self, base_dir, files):
],

package_dir={"": "lib"},
packages=find_packages("lib"),
namespace_packages=["mpl_toolkits"],
packages=find_namespace_packages(
Copy link
Member

Choose a reason for hiding this comment

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

Not sure I read https://setuptools.pypa.io/en/latest/userguide/package_discovery.html#finding-simple-packages correctly, but could we use find_packages(…, include=..)? The exclude has the danger of unintentionally picking up other files we add in the future.

Copy link
Contributor Author

Choose a reason for hiding this comment

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

I experimented a bit with it and so far I haven't been able to build matplotlib by supplying include instead of exclude. Even building an exhaustive list of all modules that should be included doesn't appear to work. I don't know if that's a bug in setuptools or if there's something I don't understand about this function.

Copy link
Member

Choose a reason for hiding this comment

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

I think the key difference between include and exclude are a) liklihood of adding things that we want to go in each list. and b) ease of drawing the lines around the things we want included/excluded.

for a), I think it is more likely that we add python modules than folders we wish to exclude from being importable python modules (both could happen, but I'd expect python modules to be more common), so excludes are less likely to be updated

for b) I think that the things excluded appear directly next to things that want to be included, so I think it is much easier to specify exclusions.

I mean, it is just a list of strings at the end of the day, so if there is an easier way to produce the list we want, I'm not opposed, but lean towards using the tools provided to produce the lists.

Copy link
Member

Choose a reason for hiding this comment

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

This will change under meson so lets not over think this.

where="lib",
exclude=["*baseline_images*", "*tinypages*", "*mpl-data*", "*web_backend*"],
),
py_modules=["pylab"],
# Dummy extension to trigger build_ext, which will swap it out with
# real extensions that can depend on numpy for the build.
Expand Down