On Mon, Dec 23, 2013 at 5:40 AM, Ron Savage <[email protected]> wrote:

> On Monday, 23 December 2013 14:26:28 UTC+11, rns wrote:
>
>> On Mon, Dec 23, 2013 at 4:56 AM, Jeffrey Kegler <jeffre...@jeffreykegler.
>> com> wrote:
>>
>>>  Any idea why this is the fix?
>>>
>> None, unfortunately.
>>
>>
>>> Just a peculiarity of the Windows port?
>>>
>> Looks like it.
>>
>>
>>> Is there a reason the order should matter? -- jeffrey
>>>
>> It's not mentioned explicitly in the doc, but seems to be tacitly assumed
>> that 'use utf8' must be at the top of the code. Looks like "at the top"
>> also means "before use open..." that is frequently the case
>> http://goo.gl/3bABVr
>>
>
> I'd take that as a bug in Perl. Any like to submit a bug report? A tiny
> demo program would be nice in that case.
>
The attached demo fails under 5.14.2 (compiled from source on winxp sp3
with cl 13.10.3077 (VS .NET 2003)), but passes (albeit with 'Wide character
in print at c:/Perl/lib/Test/Builder.pm line 1172.') under 5.8.7 (from a
binary dist with Apache and mod_perl).

-- 
>  You received this message because you are subscribed to the Google Groups
> "marpa parser" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to [email protected].
> For more options, visit https://groups.google.com/groups/opt_out.
>

-- 
You received this message because you are subscribed to the Google Groups 
"marpa parser" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
For more options, visit https://groups.google.com/groups/opt_out.

Attachment: utf8_after_use_open_encoding.pl
Description: Binary data

Reply via email to