[PATCH] py3: have a bytes version of shlex.split()

[Pulkit Goyal]

I don't know who did it, but this one is already
pushed(https://www.mercurial-scm.org/repo/hg-committed/rev/caf7e1c5efe4).
I think this will prevent any further confusion.

On Sun, Dec 25, 2016 at 5:46 AM, Pulkit Goyal <7895pulkit at gmail.com> wrote:
> # HG changeset patch
> # User Pulkit Goyal <7895pulkit at gmail.com>
> # Date 1482615415 -19800
> #      Sun Dec 25 03:06:55 2016 +0530
> # Node ID 4b6b316bc3d87df1ce7deffbda1cb813b955d658
> # Parent  81bf1a686b999c3c2d743135f8ee316546ac77c9
> py3: have a bytes version of shlex.split()
>
> shlex.split() only accepts unicodes on Python 3. After this patch we will be
> using pycompat.shlexsplit(). This patch also replaces existing occurences of
> shlex.split with pycompat.shlexsplit.
>
> diff -r 81bf1a686b99 -r 4b6b316bc3d8 hgext/extdiff.py
> --- a/hgext/extdiff.py  Sun Dec 25 02:42:46 2016 +0530
> +++ b/hgext/extdiff.py  Sun Dec 25 03:06:55 2016 +0530
> @@ -64,7 +64,6 @@
>
>  import os
>  import re
> -import shlex
>  import shutil
>  import tempfile
>  from mercurial.i18n import _
> @@ -78,6 +77,7 @@
>      commands,
>      error,
>      filemerge,
> +    pycompat,
>      scmutil,
>      util,
>  )
> @@ -371,7 +371,7 @@
>              if path:
>                  # case "cmd = path opts"
>                  cmdline = path
> -                diffopts = len(shlex.split(cmdline)) > 1
> +                diffopts = len(pycompat.shlexsplit(cmdline)) > 1
>              else:
>                  # case "cmd ="
>                  path = util.findexe(cmd)
> diff -r 81bf1a686b99 -r 4b6b316bc3d8 mercurial/commands.py
> --- a/mercurial/commands.py     Sun Dec 25 02:42:46 2016 +0530
> +++ b/mercurial/commands.py     Sun Dec 25 03:06:55 2016 +0530
> @@ -11,7 +11,6 @@
>  import errno
>  import os
>  import re
> -import shlex
>  import socket
>  import string
>  import sys
> @@ -1981,7 +1980,7 @@
>      editor = ui.geteditor()
>      editor = util.expandpath(editor)
>      fm.write('editor', _("checking commit editor... (%s)\n"), editor)
> -    cmdpath = util.findexe(shlex.split(editor)[0])
> +    cmdpath = util.findexe(pycompat.shlexsplit(editor)[0])
>      fm.condwrite(not cmdpath and editor == 'vi', 'vinotfound',
>                   _(" No commit editor set and can't find %s in PATH\n"
>                     " (specify a commit editor in your configuration"
> diff -r 81bf1a686b99 -r 4b6b316bc3d8 mercurial/dispatch.py
> --- a/mercurial/dispatch.py     Sun Dec 25 02:42:46 2016 +0530
> +++ b/mercurial/dispatch.py     Sun Dec 25 03:06:55 2016 +0530
> @@ -14,7 +14,6 @@
>  import os
>  import pdb
>  import re
> -import shlex
>  import signal
>  import sys
>  import time
> @@ -279,7 +278,7 @@
>          cmd = re.sub(r'\$(\d+|\$)', replacer, cmd)
>          givenargs = [x for i, x in enumerate(givenargs)
>                       if i not in nums]
> -        args = shlex.split(cmd)
> +        args = pycompat.shlexsplit(cmd)
>      return args + givenargs
>
>  def aliasinterpolate(name, args, cmd):
> @@ -351,7 +350,7 @@
>              return
>
>          try:
> -            args = shlex.split(self.definition)
> +            args = pycompat.shlexsplit(self.definition)
>          except ValueError as inst:
>              self.badalias = (_("error in definition for alias '%s': %s")
>                               % (self.name, inst))
> @@ -461,7 +460,7 @@
>          args = aliasargs(entry[0], args)
>          defaults = ui.config("defaults", cmd)
>          if defaults:
> -            args = map(util.expandpath, shlex.split(defaults)) + args
> +            args = map(util.expandpath, pycompat.shlexsplit(defaults)) + args
>          c = list(entry[1])
>      else:
>          cmd = None
> diff -r 81bf1a686b99 -r 4b6b316bc3d8 mercurial/pycompat.py
> --- a/mercurial/pycompat.py     Sun Dec 25 02:42:46 2016 +0530
> +++ b/mercurial/pycompat.py     Sun Dec 25 03:06:55 2016 +0530
> @@ -12,6 +12,7 @@
>
>  import getopt
>  import os
> +import shlex
>  import sys
>
>  ispy3 = (sys.version_info[0] >= 3)
> @@ -122,6 +123,14 @@
>          dic = dict((k.encode('latin-1'), v) for k, v in dic.iteritems())
>          return dic
>
> +    # shlex.split() accepts unicodes on Python 3. This function takes bytes
> +    # argument, convert it into unicodes, pass into shlex.split(), convert the
> +    # returned value to bytes and return that.
> +    # TODO: handle shlex.shlex().
> +    def shlexsplit(s):
> +        ret = shlex.split(s.decode('latin-1'))
> +        return [a.encode('latin-1') for a in ret]
> +
>  else:
>      def sysstr(s):
>          return s
> @@ -162,6 +171,7 @@
>      getcwd = os.getcwd
>      osgetenv = os.getenv
>      sysexecutable = sys.executable
> +    shlexsplit = shlex.split
>
>  stringio = io.StringIO
>  empty = _queue.Empty