Python 3: shlex.split() wants unicode

On Python 2, shlex.split() raises if you pass it a unicode object with
non-ASCII characters in it.  The Ansible codebase copes by explicitly
converting the string using to_bytes() before passing it to
shlex.split().

On Python 3, shlex.split() raises ('bytes' object has no attribute 'read')
if you pass a bytes object.  Oops.

This commit introduces a new wrapper function, shlex_split, that
transparently performs the to_bytes/to_unicode conversions only on
Python 2.

Currently I've only converted one call site (the one that was causing a
unit test to fail on Python 3).  If this approach is deemed suitable,
I'll convert them all.

lib/ansible/inventory/ini.py

@@ -20,7 +20,6 @@
 __metaclass__ = type
 
 import ast
-import shlex
 import re
 
 from ansible import constants as C
@@ -30,7 +29,8 @@
 from ansible.inventory.expand_hosts import detect_range
 from ansible.inventory.expand_hosts import expand_hostname_range
 from ansible.parsing.utils.addresses import parse_address
-from ansible.utils.unicode import to_unicode, to_bytes
+from ansible.utils.shlex import shlex_split
+from ansible.utils.unicode import to_unicode
 
 class InventoryParser(object):
     """
@@ -231,13 +231,11 @@ def _parse_host_definition(self, line):
         # beta:2345 user=admin      # we'll tell shlex
         # gamma sudo=True user=root # to ignore comments
 
-        line = to_bytes(line)
         try:
-            tokens = shlex.split(line, comments=True)
+            tokens = shlex_split(line, comments=True)
         except ValueError as e:
             self._raise_error("Error parsing host definition '%s': %s" % (varstring, e))
 
-        tokens = [ to_unicode(t) for t in tokens]
         (hostnames, port) = self._expand_hostpattern(tokens[0])
         hosts = self._Hosts(hostnames, port)
 

lib/ansible/utils/shlex.py

@@ -0,0 +1,33 @@
+# (c) 2015, Marius Gedminas <[email protected]>
+#
+# This file is part of Ansible
+#
+# Ansible is free software: you can redistribute it and/or modify
+# it under the terms of the GNU General Public License as published by
+# the Free Software Foundation, either version 3 of the License, or
+# (at your option) any later version.
+#
+# Ansible is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+# GNU General Public License for more details.
+#
+# You should have received a copy of the GNU General Public License
+# alongwith Ansible.  If not, see <http://www.gnu.org/licenses/>.
+
+from __future__ import absolute_import
+
+import shlex
+from six import PY3
+
+from ansible.utils.unicode import to_bytes, to_unicode
+
+
+if PY3:
+    # shlex.split() wants Unicode (i.e. ``str``) input on Python 3
+    shlex_split = shlex.split
+else:
+    # shlex.split() wants bytes (i.e. ``str``) input on Python 2
+    def shlex_split(s, comments=False, posix=True):
+        return map(to_unicode, shlex.split(to_bytes(s), comments, posix))
+    shlex_split.__doc__ = shlex.split.__doc__

test/units/utils/test_shlex.py

@@ -0,0 +1,39 @@
+# (c) 2015, Marius Gedminas <[email protected]>
+#
+# This file is part of Ansible
+#
+# Ansible is free software: you can redistribute it and/or modify
+# it under the terms of the GNU General Public License as published by
+# the Free Software Foundation, either version 3 of the License, or
+# (at your option) any later version.
+#
+# Ansible is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+# GNU General Public License for more details.
+#
+# You should have received a copy of the GNU General Public License
+# along with Ansible.  If not, see <http://www.gnu.org/licenses/>.
+
+import unittest
+
+from ansible.utils.shlex import shlex_split
+
+
+class TestSplit(unittest.TestCase):
+
+    def test_trivial(self):
+        self.assertEqual(shlex_split("a b c"), ["a", "b", "c"])
+
+    def test_unicode(self):
+        self.assertEqual(shlex_split(u"a b \u010D"), [u"a", u"b", u"\u010D"])
+
+    def test_quoted(self):
+        self.assertEqual(shlex_split('"a b" c'), ["a b", "c"])
+
+    def test_comments(self):
+        self.assertEqual(shlex_split('"a b" c # d', comments=True), ["a b", "c"])
+
+    def test_error(self):
+        self.assertRaises(ValueError, shlex_split, 'a "b')
+