Skip to content

Conversation

yashuatla
Copy link
Owner

This PR contains changes from a range of commits from the original repository.

Commit Range: b18407b..7fa82d7
Files Changed: 41 (29 programming files)
Programming Ratio: 70.7%

Commits included:

oscargus and others added 18 commits May 18, 2025 15:53
... instead of having to go through ft2font in createType1Descriptor
just to extract the font bbox, ascender and descender.

FontBBox is gauranteed to exist in the type1 font definition by the
standard; its parsing as a size-4 array matches freetype's behavior (see
ps_parser_load_field); and using bbox entries as ascender and descender
also matches freetype's behavior (T1_Face_Init directly assigns
`root->ascender = (FT_Short)(root->bbox.yMax)` and likewise for the
descender; see also the docs for ascender and descender in FT_FaceRec).
Provide a simple macro to call a FreeType function and throw
an exception if an error is returned, while also including the
source file and line location for the error.  For example, trying
`FT2Font(open("pyproject.toml", "rb"))` now raises "FT_Open_Face
(ft2font.cpp line 220) failed with error 0x2: unknown file format"
instead of "Can not load face (unknown file format; error code 0x2)"
The previous implementation was clearly wrong (the isinstance check
would raise TypeError as the second argument would be a bool), but the
tests didn't catch that because the bug led to _is_tensorflow_array
returning False, then _unpack_to_numpy returning the original input,
and then assert_array_equal implicitly converting `result` by calling
`__array__` on it.  Fix the test by explicitly checking that `result`
is indeed a numpy array, and also fix _is_tensorflow_array with more
restrictive exception catching (also applied to _is_torch_array,
_is_jax_array, and _is_pandas_dataframe, while we're at it).
Simplify/improve error reporting from ft2font.
Fix tight-bbox computation of HostAxes.
Mypy 1.16.0 flags errors here:

lib/matplotlib/_mathtext.py:2531: error: "Node" has no attribute "width"  [attr-defined]
lib/matplotlib/_mathtext.py:2608: error: List item 0 has incompatible type "Kern"; expected "Hlist | Vlist"  [list-item]

The check for the attribute _metrics is equivalent to checking for an
instance of Char, since only Char and its subclasses set self._metrics.

Mypy infers an unnecessarily tight type list[Hlist | Vlist] for
spaced_nucleus so we give it a more general one.
Since AFM is now private, we can delete unused methods without
deprecation.

Additionally, add `AFM.get_glyph_name` so that the PostScript mathtext
code doesn't need to special case AFM files.
Deprecate point_at_t and document that a BezierSegment can be called
* Remove apply_theta_transforms argument

* Improve formatting

* Rename xxxxxx-DS.rst to 30004-DS.rst

* Delete extra line
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

7 participants