1 files changed, 63 insertions, 0 deletions
diff --git a/Lib/test/support.py b/Lib/test/support.py
index c5640e0a08..ec4b47d9fc 100644
--- a/Lib/test/support.py
+++ b/Lib/test/support.py
@@ -603,6 +603,49 @@ else:
 # module name.
 TESTFN = "{}_{}_tmp".format(TESTFN, os.getpid())
 
+# FS_NONASCII: non-ASCII character encodable by os.fsencode(),
+# or None if there is no such character.
+FS_NONASCII = None
+for character in (
+    # First try printable and common characters to have a readable filename.
+    # For each character, the encoding list are just example of encodings able
+    # to encode the character (the list is not exhaustive).
+
+    # U+00E6 (Latin Small Letter Ae): cp1252, iso-8859-1
+    '\u00E6',
+    # U+0130 (Latin Capital Letter I With Dot Above): cp1254, iso8859_3
+    '\u0130',
+    # U+0141 (Latin Capital Letter L With Stroke): cp1250, cp1257
+    '\u0141',
+    # U+03C6 (Greek Small Letter Phi): cp1253
+    '\u03C6',
+    # U+041A (Cyrillic Capital Letter Ka): cp1251
+    '\u041A',
+    # U+05D0 (Hebrew Letter Alef): Encodable to cp424
+    '\u05D0',
+    # U+060C (Arabic Comma): cp864, cp1006, iso8859_6, mac_arabic
+    '\u060C',
+    # U+062A (Arabic Letter Teh): cp720
+    '\u062A',
+    # U+0E01 (Thai Character Ko Kai): cp874
+    '\u0E01',
+
+    # Then try more "special" characters. "special" because they may be
+    # interpreted or displayed differently depending on the exact locale
+    # encoding and the font.
+
+    # U+00A0 (No-Break Space)
+    '\u00A0',
+    # U+20AC (Euro Sign)
+    '\u20AC',
+):
+    try:
+        os.fsdecode(os.fsencode(character))
+    except UnicodeError:
+        pass
+    else:
+        FS_NONASCII = character
+        break
 
 # TESTFN_UNICODE is a non-ascii filename
 TESTFN_UNICODE = TESTFN + "-\xe0\xf2\u0258\u0141\u011f"
@@ -647,6 +690,26 @@ elif sys.platform != 'darwin':
         # the byte 0xff. Skip some unicode filename tests.
         pass
 
+# TESTFN_UNDECODABLE is a filename (bytes type) that should *not* be able to be
+# decoded from the filesystem encoding (in strict mode). It can be None if we
+# cannot generate such filename.
+TESTFN_UNDECODABLE = None
+# b'\xff' is not decodable by os.fsdecode() with code page 932. Windows
+# accepts it to create a file or a directory, or don't accept to enter to
+# such directory (when the bytes name is used). So test b'\xe7' first: it is
+# not decodable from cp932.
+for name in (b'\xe7w\xf0', b'abc\xff'):
+    try:
+        os.fsdecode(name)
+    except UnicodeDecodeError:
+        TESTFN_UNDECODABLE = name
+        break
+
+if FS_NONASCII:
+    TESTFN_NONASCII = TESTFN + '-' + FS_NONASCII
+else:
+    TESTFN_NONASCII = None
+
 # Save the initial cwd
 SAVEDCWD = os.getcwd()