diff options
author | Victor Stinner <victor.stinner@gmail.com> | 2012-11-10 12:07:39 +0100 |
---|---|---|
committer | Victor Stinner <victor.stinner@gmail.com> | 2012-11-10 12:07:39 +0100 |
commit | ff3d515952aa6eabdb82a90b9fe6f9dc33a406e5 (patch) | |
tree | 86bb37049fad83ab74cf568623274bdd97c115f0 /Lib/test/support.py | |
parent | ef3971d3e57ea8364a90581abb36c746a91f8001 (diff) | |
download | cpython-git-ff3d515952aa6eabdb82a90b9fe6f9dc33a406e5.tar.gz |
Issue #16444, #16218: Use TESTFN_UNDECODABLE on UNIX
Check if data is decoded by os.fsdecode() (filesystem encoding with
surrogateescape error handler, PEP 383), not by UTF-8 or the filesystem
encoding in strict mode.
Use TESTFN_UNDECODABLE in test_cmd_line_script.test_non_ascii() on UNIX.
Diffstat (limited to 'Lib/test/support.py')
-rw-r--r-- | Lib/test/support.py | 28 |
1 files changed, 20 insertions, 8 deletions
diff --git a/Lib/test/support.py b/Lib/test/support.py index ec4b47d9fc..d0a37ea926 100644 --- a/Lib/test/support.py +++ b/Lib/test/support.py @@ -692,17 +692,29 @@ elif sys.platform != 'darwin': # 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. +# cannot generate such filename (ex: the latin1 encoding can decode any byte +# sequence). On UNIX, TESTFN_UNDECODABLE can be decoded by os.fsdecode() thanks +# to the surrogateescape error handler (PEP 383), but not from the filesystem +# encoding in strict mode. 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'): +for name in ( + # 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. + b'\xe7w\xf0', + # undecodable from ASCII, UTF-8 + b'\xff', + # undecodable from iso8859-3, iso8859-6, iso8859-7, cp424, iso8859-8, cp856 + # and cp857 + b'\xae\xd5' + # undecodable from UTF-8 (UNIX and Mac OS X) + b'\xed\xb2\x80', b'\xed\xb4\x80', +): try: - os.fsdecode(name) + name.decode(TESTFN_ENCODING) except UnicodeDecodeError: - TESTFN_UNDECODABLE = name + TESTFN_UNDECODABLE = os.fsencode(TESTFN) + name break if FS_NONASCII: |