[Python-checkins] cpython: Issue #16414: Add support.FS_NONASCII and support.TESTFN_NONASCII

Tue Nov 6 23:23:28 CET 2012

http://hg.python.org/cpython/rev/de8cf1ece068
changeset: 80278:de8cf1ece068
user: Victor Stinner <victor.stinner at gmail.com>
date: Tue Nov 06 23:23:43 2012 +0100
summary:
 Issue #16414: Add support.FS_NONASCII and support.TESTFN_NONASCII
These constants are used to test functions with non-ASCII data, especially
filenames.
files:
 Lib/test/support.py | 31 ++++++++++++++++++++
 Lib/test/test_cmd_line.py | 6 +-
 Lib/test/test_cmd_line_script.py | 11 +-----
 Lib/test/test_genericpath.py | 2 +
 Lib/test/test_os.py | 4 ++
 5 files changed, 42 insertions(+), 12 deletions(-)

diff --git a/Lib/test/support.py b/Lib/test/support.py
--- a/Lib/test/support.py
+++ b/Lib/test/support.py
@@ -603,6 +603,32 @@
 # 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 (
+ # U+00E6 (Latin small letter AE): Encodable to cp1252, cp1254, cp1257, iso-8859-1
+ '\u00E6',
+ # U+0141 (Latin capital letter L with stroke): Encodable to cp1250, cp1257
+ '\u0141',
+ # U+041A (Cyrillic capital letter KA): Encodable to cp932, cp950, cp1251
+ '\u041A',
+ # U+05D0 (Hebrew Letter Alef): Encodable to cp424, cp1255
+ '\u05D0',
+ # U+06A9 (Arabic letter KEHEH): Encodable to cp1256
+ '\u06A9',
+ # U+03A9 (Greek capital letter OMEGA): Encodable to cp932, cp950, cp1253
+ '\u03A9',
+ # U+0E01 (Thai character KO KAI): Encodable to cp874
+ '\u0E01',
+):
+ 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"
@@ -658,6 +684,11 @@
 TESTFN_UNDECODABLE = name
 break
 
+if FS_NONASCII:
+ TESTFN_NONASCII = TESTFN + '- ' + FS_NONASCII
+else:
+ TESTFN_NONASCII = None
+
 # Save the initial cwd
 SAVEDCWD = os.getcwd()
 
diff --git a/Lib/test/test_cmd_line.py b/Lib/test/test_cmd_line.py
--- a/Lib/test/test_cmd_line.py
+++ b/Lib/test/test_cmd_line.py
@@ -93,15 +93,15 @@
 # All good if execution is successful
 assert_python_ok('-c', 'pass')
 
- @unittest.skipIf(sys.getfilesystemencoding() == 'ascii',
- 'need a filesystem encoding different than ASCII')
+ @unittest.skipUnless(test.support.FS_NONASCII, 'need support.FS_NONASCII')
 def test_non_ascii(self):
 # Test handling of non-ascii data
 if test.support.verbose:
 import locale
 print('locale encoding = %s, filesystem encoding = %s'
 % (locale.getpreferredencoding(), sys.getfilesystemencoding()))
- command = "assert(ord('\xe9') == 0xe9)"
+ command = ("assert(ord(%r) == %s)"
+ % (test.support.FS_NONASCII, ord(test.support.FS_NONASCII)))
 assert_python_ok('-c', command)
 
 # On Windows, pass bytes to subprocess doesn't test how Python decodes the
diff --git a/Lib/test/test_cmd_line_script.py b/Lib/test/test_cmd_line_script.py
--- a/Lib/test/test_cmd_line_script.py
+++ b/Lib/test/test_cmd_line_script.py
@@ -363,19 +363,12 @@
 self.assertTrue(text[1].startswith(' File '))
 self.assertTrue(text[3].startswith('NameError'))
 
+ @unittest.skipUnless(support.TESTFN_NONASCII, 'need support.TESTFN_NONASCII')
 def test_non_ascii(self):
 # Issue #16218
 # non-ascii filename encodable to cp1252, cp932, latin1 and utf8
- filename = support.TESTFN + '\xa3'
- try:
- os.fsencode(filename)
- except UnicodeEncodeError:
- self.skipTest(
- "Filesystem encoding %r cannot encode "
- "the filename: %a"
- % (sys.getfilesystemencoding(), filename))
 source = 'print(ascii(__file__))\n'
- script_name = _make_test_script(os.curdir, filename, source)
+ script_name = _make_test_script(os.curdir, support.TESTFN_NONASCII, source)
 self.addCleanup(support.unlink, script_name)
 rc, stdout, stderr = assert_python_ok(script_name)
 self.assertEqual(
diff --git a/Lib/test/test_genericpath.py b/Lib/test/test_genericpath.py
--- a/Lib/test/test_genericpath.py
+++ b/Lib/test/test_genericpath.py
@@ -313,6 +313,8 @@
 def test_nonascii_abspath(self):
 if support.TESTFN_UNDECODABLE:
 name = support.TESTFN_UNDECODABLE
+ elif support.TESTFN_NONASCII:
+ name = support.TESTFN_NONASCII
 else:
 name = b'a\xffb\xe7w\xf0'
 
diff --git a/Lib/test/test_os.py b/Lib/test/test_os.py
--- a/Lib/test/test_os.py
+++ b/Lib/test/test_os.py
@@ -1243,6 +1243,8 @@
 def setUp(self):
 if support.TESTFN_UNENCODABLE:
 self.dir = support.TESTFN_UNENCODABLE
+ elif support.TESTFN_NONASCII:
+ self.dir = support.TESTFN_NONASCII
 else:
 self.dir = support.TESTFN
 self.bdir = os.fsencode(self.dir)
@@ -1257,6 +1259,8 @@
 add_filename(support.TESTFN_UNICODE)
 if support.TESTFN_UNENCODABLE:
 add_filename(support.TESTFN_UNENCODABLE)
+ if support.TESTFN_NONASCII:
+ add_filename(support.TESTFN_NONASCII)
 if not bytesfn:
 self.skipTest("couldn't create any non-ascii filename")
 
-- 
Repository URL: http://hg.python.org/cpython
