Front page | perl.perl5.porters |
Postings from June 2004
Stas Bekman <stas@stason.org> writes:
>SADAHIRO Tomoyuki wrote:
>> On Sun, 30 May 2004 21:43:38 -0700
>> Stas Bekman <stas@stason.org> wrote:
>>
>>
>>>So it's unclear whose fault it is, the code that populates sv with pv under -T
>>>or Perl_sv_utf8_decode(), which may need to check pPOK instead of POK?
>>
>>
>> You can use <$_ = Encode::decode('utf8', $_)>
>> instead of <utf8::decode $_>.
>> Encode::decode('utf8') is based on Encode::utf8::decode_xs() and
>> it does not use "broken" sv_utf8_decode().
>> I think it should work well even if taint mode.
>
>Thank you, Sadahiro. It indeed fixes my simple test case (I haven't tested
>that solution in the application yet).
>
>perl-5.8.0 -T -MEncode -MDevel::Peek -le '$_ = shift; Dump $_; $_ =
>Encode::decode('utf8', $_); Dump $_' `perl -le 'binmode STDOUT, ":utf8"; print
>"\x{0432}\x{0430}"'`
>
>Shouldn't sv_utf8_decode() (and potentially other utf8 functions) then be
>replaced with Encode::utf8::decode_xs() (and equivalents from Encode?).
sv_utf8_decode is an internals thing that is exported to help out
XS code at perl5.6 or so. We have discussed the pPOK vs POK thing before
and I forget the final outcome.
Normally you have to do SvGETMAGIC() to get POK flag on, and I think
internals routine assumes caller does that.
We could change that, but that won't help older perls.
Thread Previous
|
Thread Next