BUG: when given unicode inputs, arg_split should return unicode outputs. Always use utf-8 to encode the string instead of relying on sys.stdin.encoding, which may not be able to accept the full range of Unicode characters. When given unicode strings, arg_split is probably not receiving input from a terminal.

IPython/utils/process.py

@@ -1,140 +1,146
 # encoding: utf-8
 """
 Utilities for working with external processes.
 """
 
 #-----------------------------------------------------------------------------
 #  Copyright (C) 2008-2009  The IPython Development Team
 #
 #  Distributed under the terms of the BSD License.  The full license is in
 #  the file COPYING, distributed as part of this software.
 #-----------------------------------------------------------------------------
 
 #-----------------------------------------------------------------------------
 # Imports
 #-----------------------------------------------------------------------------
 from __future__ import print_function
 
 # Stdlib
 import os
 import sys
 import shlex
 
 # Our own
 if sys.platform == 'win32':
     from ._process_win32 import _find_cmd, system, getoutput, AvoidUNCPath
 else:
     from ._process_posix import _find_cmd, system, getoutput
 
 from ._process_common import getoutputerror
 
 #-----------------------------------------------------------------------------
 # Code
 #-----------------------------------------------------------------------------
 
 
 class FindCmdError(Exception):
     pass
 
 
 def find_cmd(cmd):
     """Find absolute path to executable cmd in a cross platform manner.
     
     This function tries to determine the full path to a command line program
     using `which` on Unix/Linux/OS X and `win32api` on Windows.  Most of the
     time it will use the version that is first on the users `PATH`.  If
     cmd is `python` return `sys.executable`.
 
     Warning, don't use this to find IPython command line programs as there
     is a risk you will find the wrong one.  Instead find those using the
     following code and looking for the application itself::
     
         from IPython.utils.path import get_ipython_module_path
         from IPython.utils.process import pycmd2argv
         argv = pycmd2argv(get_ipython_module_path('IPython.frontend.terminal.ipapp'))
 
     Parameters
     ----------
     cmd : str
         The command line program to look for.
     """
     if cmd == 'python':
         return os.path.abspath(sys.executable)
     try:
         path = _find_cmd(cmd).rstrip()
     except OSError:
         raise FindCmdError('command could not be found: %s' % cmd)
     # which returns empty if not found
     if path == '':
         raise FindCmdError('command could not be found: %s' % cmd)
     return os.path.abspath(path)
 
 
 def pycmd2argv(cmd):
     r"""Take the path of a python command and return a list (argv-style).
 
     This only works on Python based command line programs and will find the
     location of the ``python`` executable using ``sys.executable`` to make
     sure the right version is used.
 
     For a given path ``cmd``, this returns [cmd] if cmd's extension is .exe,
     .com or .bat, and [, cmd] otherwise.  
 
     Parameters
     ----------
     cmd : string
       The path of the command.
 
     Returns
     -------
     argv-style list.
     """
     ext = os.path.splitext(cmd)[1]
     if ext in ['.exe', '.com', '.bat']:
         return [cmd]
     else:
         if sys.platform == 'win32':
             # The -u option here turns on unbuffered output, which is required
             # on Win32 to prevent wierd conflict and problems with Twisted.
             # Also, use sys.executable to make sure we are picking up the 
             # right python exe.
             return [sys.executable, '-u', cmd]
         else:
             return [sys.executable, cmd]
 
 
 def arg_split(s, posix=False):
     """Split a command line's arguments in a shell-like manner.
 
     This is a modified version of the standard library's shlex.split()
     function, but with a default of posix=False for splitting, so that quotes
     in inputs are respected."""
 
     # Unfortunately, python's shlex module is buggy with unicode input:
     # http://bugs.python.org/issue1170
     # At least encoding the input when it's unicode seems to help, but there
     # may be more problems lurking.  Apparently this is fixed in python3.
+    is_unicode = False
     if isinstance(s, unicode):
-        s = s.encode(sys.stdin.encoding)
+        is_unicode = True
+        s = s.encode('utf-8')
     lex = shlex.shlex(s, posix=posix)
     lex.whitespace_split = True
-    return list(lex)
+    tokens = list(lex)
+    if is_unicode:
+        # Convert the tokens back to unicode.
+        tokens = [x.decode('utf-8') for x in tokens]
+    return tokens
 
 
 def abbrev_cwd():
     """ Return abbreviated version of cwd, e.g. d:mydir """
     cwd = os.getcwd().replace('\\','/')
     drivepart = ''
     tail = cwd
     if sys.platform == 'win32':
         if len(cwd) < 4:
             return cwd
         drivepart,tail = os.path.splitdrive(cwd)
 
 
     parts = tail.split('/')
     if len(parts) > 2:
         tail = '/'.join(parts[-2:])
 
     return (drivepart + (
         cwd == '/' and '/' or tail))

IPython/utils/tests/test_process.py

@@ -1,95 +1,98
 # encoding: utf-8
 """
 Tests for platutils.py
 """
 
 #-----------------------------------------------------------------------------
 #  Copyright (C) 2008-2009  The IPython Development Team
 #
 #  Distributed under the terms of the BSD License.  The full license is in
 #  the file COPYING, distributed as part of this software.
 #-----------------------------------------------------------------------------
 
 #-----------------------------------------------------------------------------
 # Imports
 #-----------------------------------------------------------------------------
 
 import sys
 from unittest import TestCase
 
 import nose.tools as nt
 
 from IPython.utils.process import (find_cmd, FindCmdError, arg_split,
                                    system, getoutput, getoutputerror)
 from IPython.testing import decorators as dec
 from IPython.testing import tools as tt
 
 #-----------------------------------------------------------------------------
 # Tests
 #-----------------------------------------------------------------------------
 
 def test_find_cmd_python():
     """Make sure we find sys.exectable for python."""
     nt.assert_equals(find_cmd('python'), sys.executable)
 
     
 @dec.skip_win32
 def test_find_cmd_ls():
     """Make sure we can find the full path to ls."""
     path = find_cmd('ls')
     nt.assert_true(path.endswith('ls'))
 
     
 def has_pywin32():
     try:
         import win32api
     except ImportError:
         return False
     return True
 
 
 @dec.onlyif(has_pywin32, "This test requires win32api to run")
 def test_find_cmd_pythonw():
     """Try to find pythonw on Windows."""
     path = find_cmd('pythonw')
     nt.assert_true(path.endswith('pythonw.exe'))
 
 
 @dec.onlyif(lambda : sys.platform != 'win32' or has_pywin32(),
             "This test runs on posix or in win32 with win32api installed")
 def test_find_cmd_fail():
     """Make sure that FindCmdError is raised if we can't find the cmd."""
     nt.assert_raises(FindCmdError,find_cmd,'asdfasdf')
 
     
 def test_arg_split():
     """Ensure that argument lines are correctly split like in a shell."""
     tests = [['hi', ['hi']],
              [u'hi', [u'hi']],
+             ['hello there', ['hello', 'there']],
+             [u'h\N{LATIN SMALL LETTER A WITH CARON}llo', [u'h\N{LATIN SMALL LETTER A WITH CARON}llo']],
+             ['something "with quotes"', ['something', '"with quotes"']],
              ]
     for argstr, argv in tests:
         nt.assert_equal(arg_split(argstr), argv)
 
 
 class SubProcessTestCase(TestCase, tt.TempFileMixin):
     def setUp(self):
         """Make a valid python temp file."""
         lines = ["from __future__ import print_function",
                  "import sys",
                  "print('on stdout', end='', file=sys.stdout)",
                  "print('on stderr', end='', file=sys.stderr)",
                  "sys.stdout.flush()",
                  "sys.stderr.flush()"]
         self.mktmp('\n'.join(lines))
 
     def test_system(self):
         system('python "%s"' % self.fname)
 
     def test_getoutput(self):
         out = getoutput('python "%s"' % self.fname)
         self.assertEquals(out, 'on stdout')
 
     def test_getoutput(self):
         out, err = getoutputerror('python "%s"' % self.fname)
         self.assertEquals(out, 'on stdout')
         self.assertEquals(err, 'on stderr')