[MERGE REVIEW] All ini files as utf-8

[Martin Pool]

On 2006-03-06 Mon, at 1:42 PM, Robey Pointer wrote:

>
> On 3 Mar 2006, at 17:31, Aaron Bentley wrote:
>
>> |>This patch updates the config parsing to treat the config data  
>> as utf-8.
>> |> This should fix the problems reported by Erik Bågfors and Carlos
>> |>Perelló Marín.

+1
>
>> === modified file 'a/bzrlib/tests/test_config.py'
>> --- a/bzrlib/tests/test_config.py	
>> +++ b/bzrlib/tests/test_config.py	
>> @@ -1,3 +1,4 @@
>> +# -*- coding: utf-8 -*-
>>  # Copyright (C) 2005 by Canonical Ltd
>>  #   Authors: Robert Collins <robert.collins at canonical.com>
>>  #
>> @@ -30,7 +31,7 @@
>>
>>  sample_long_alias="log -r-15..-1 --line"
>>  sample_config_text = ("[DEFAULT]\n"
>> -                      "email=Robert Collins <robertc at example.com>\n"
>> +                      u"email=Erik Bågfors <erik at bagfors.nu>\n"
>>                        "editor=vim\n"
>>                        "gpg_signing_command=gnome-gpg\n"
>>                        "log_format=short\n"
>
> I don't know how good everyone's editor is, so maybe it won't be a  
> problem converting the file to utf8.  However, you can get the same  
> effect by using a unicode \uxxxx escape and leaving the file in ascii.

Yes, I think that'd be better.  You can also say

   u"email=Erik B\N{LATIN SMALL LETTER A WITH RING ABOVE}gfors  
<erik at bagfors.nu>\n"

for the sake of legibility.  (Note that this only works in u-prefixed  
unicode literals.)

-- 
Martin Pool