Merge pull request #1713 from pre-commit/pyvenv_utf8_windows

fix for base executable with non-ascii characters on windows

Author: asottile

pre_commit/languages/python.py

@@ -36,7 +36,7 @@ def _version_info(exe: str) -> str:
 
 def _read_pyvenv_cfg(filename: str) -> Dict[str, str]:
     ret = {}
-    with open(filename) as f:
+    with open(filename, encoding='UTF-8') as f:
         for line in f:
             try:
                 k, v = line.split('=')

tests/languages/python_test.py

@@ -23,6 +23,13 @@ def test_read_pyvenv_cfg(tmpdir):
     assert python._read_pyvenv_cfg(pyvenv_cfg) == expected
 
 
+def test_read_pyvenv_cfg_non_utf8(tmpdir):
+    pyvenv_cfg = tmpdir.join('pyvenv_cfg')
+    pyvenv_cfg.write_binary('hello = hello john.š\n'.encode())
+    expected = {'hello': 'hello john.š'}
+    assert python._read_pyvenv_cfg(pyvenv_cfg) == expected
+
+
 def test_norm_version_expanduser():
     home = os.path.expanduser('~')
     if os.name == 'nt':  # pragma: nt cover