Fix issue with tex-encoding on non-Unicode platforms (matplotlib#23033)

Author: oscargus

lib/matplotlib/tests/test_texmanager.py

@@ -3,6 +3,7 @@
 
 import matplotlib.pyplot as plt
 from matplotlib.texmanager import TexManager
+from matplotlib.testing._markers import needs_usetex
 import pytest
 
 
@@ -39,3 +40,20 @@ def test_font_selection(rc, preamble, family):
     src = Path(tm.make_tex("hello, world", fontsize=12)).read_text()
     assert preamble in src
     assert [*re.findall(r"\\\w+family", src)] == [family]
+
+
+@needs_usetex
+def test_unicode_characters():
+    # Smoke test to see that Unicode characters does not cause issues
+    # See #23019
+    plt.rcParams['text.usetex'] = True
+    fig, ax = plt.subplots()
+    ax.set_ylabel('\\textit{Velocity (\N{DEGREE SIGN}/sec)}')
+    ax.set_xlabel('\N{VULGAR FRACTION ONE QUARTER}Öøæ')
+    fig.canvas.draw()
+
+    # But not all characters.
+    # Should raise RuntimeError, not UnicodeDecodeError
+    with pytest.raises(RuntimeError):
+        ax.set_title('\N{SNOWMAN}')
+        fig.canvas.draw()

lib/matplotlib/texmanager.py

@@ -243,7 +243,8 @@ def make_tex(cls, tex, fontsize):
         Return the file name.
         """
         texfile = cls.get_basefile(tex, fontsize) + ".tex"
-        Path(texfile).write_text(cls._get_tex_source(tex, fontsize))
+        Path(texfile).write_text(cls._get_tex_source(tex, fontsize),
+                                 encoding='utf-8')
         return texfile
 
     @classmethod
@@ -267,7 +268,8 @@ def _run_checked_subprocess(cls, command, tex, *, cwd=None):
                     prog=command[0],
                     format_command=cbook._pformat_subprocess(command),
                     tex=tex.encode('unicode_escape'),
-                    exc=exc.output.decode('utf-8'))) from None
+                    exc=exc.output.decode('utf-8', 'backslashreplace'))
+                ) from None
         _log.debug(report)
         return report