From: Éric Araujo <merwok@netwok.org>
Date: Tue, 9 Aug 2011 21:18:06 +0000 (+0200)
Subject: Avoid unwanted behavior change in shlex.quote (see #9723).
X-Git-Tag: v3.3.0a1~1712^2~5
X-Git-Url: http://git.ipfire.org/gitweb.cgi?a=commitdiff_plain;h=7fc0394a12d4bb5632424bfdfd4bb35590dec2e5;p=thirdparty%2FPython%2Fcpython.git

Avoid unwanted behavior change in shlex.quote (see #9723).

I simplified the quote code to use a regex instead of a loop+test when I
moved pipes.quote to shlex in 5966eeb0457d; Ezio Melotti pointed out
that my regex contained redundant parts (now removed) and allowed
non-ASCII characters (now disallowed).

I think common UNIX shells don’t quote non-ASCII characters, but there’s
no harm in doing so.  We’ll see if users request a change.
---

7fc0394a12d4bb5632424bfdfd4bb35590dec2e5
diff --cc Lib/shlex.py
index 279ab4840595,3edd3db1ed9e..92c49c360842
--- a/Lib/shlex.py
+++ b/Lib/shlex.py
@@@ -275,21 -274,6 +275,21 @@@ def split(s, comments=False, posix=True
          lex.commenters = ''
      return list(lex)
  
 +
- _find_unsafe = re.compile(r'[^\w\d@%_\-\+=:,\./]').search
++_find_unsafe = re.compile(r'[^\w@%\-\+=:,\./]', re.ASCII).search
 +
 +def quote(s):
 +    """Return a shell-escaped version of the string *s*."""
 +    if not s:
 +        return "''"
 +    if _find_unsafe(s) is None:
 +        return s
 +
 +    # use single quotes, and put single quotes into double quotes
 +    # the string $'b is then quoted as '$'"'"'b'
 +    return "'" + s.replace("'", "'\"'\"'") + "'"
 +
 +
  if __name__ == '__main__':
      if len(sys.argv) == 1:
          lexer = shlex()
diff --cc Lib/test/test_shlex.py
index ea3d777ceba3,25e4b6df6c5d..d4463f300481
--- a/Lib/test/test_shlex.py
+++ b/Lib/test/test_shlex.py
@@@ -174,21 -173,6 +174,22 @@@ class ShlexTest(unittest.TestCase)
                               "%s: %s != %s" %
                               (self.data[i][0], l, self.data[i][1:]))
  
 +    def testQuote(self):
 +        safeunquoted = string.ascii_letters + string.digits + '@%_-+=:,./'
-         unsafe = '"`$\\!'
++        unicode_sample = '\xe9\xe0\xdf'  # e + acute accent, a + grave, sharp s
++        unsafe = '"`$\\!' + unicode_sample
 +
 +        self.assertEqual(shlex.quote(''), "''")
 +        self.assertEqual(shlex.quote(safeunquoted), safeunquoted)
 +        self.assertEqual(shlex.quote('test file name'), "'test file name'")
 +        for u in unsafe:
 +            self.assertEqual(shlex.quote('test%sname' % u),
 +                             "'test%sname'" % u)
 +        for u in unsafe:
 +            self.assertEqual(shlex.quote("test%s'name'" % u),
 +                             "'test%s'\"'\"'name'\"'\"''" % u)
 +
 +
  # Allow this test to be used with old shlex.py
  if not getattr(shlex, "split", None):
      for methname in dir(ShlexTest):