SOAP::Lite UTF-8

[Tatsuhiko Miyagawa]

There is a patch to handle utf-8 strings in SOAP/XMLRPC::Lite
correctly, without some black magic.
http://blog.livedoor.jp/dankogai/archives/50375819.html

By black magic I mean SOAP::Data->type(string => $string) to suppress
the b64'ing the content.

HTH.

On 10/24/06, Nigel Rantor <wiggly at wiggly.org> wrote:
> Stephen Collyer wrote:
> > Nigel Rantor wrote:
> >
> >>> Is that what you're referring to ? In this case, stripping the
> >>> UTF-8 flag seems to make no difference.
> >>
> >>
> >> Nope, I believe that is the PerlIO layer complaining because it didn't
> >> think it was going to be outputting wide characters.
> >
> > Right. But it's not clear to me why it's not handled auto-magically.
> >
> >> I thought that saying;
> >>
> >> use open ':utf8';
> >  >
> >> should have solved that problem, but it didn't.
> >
> > As far as I understand it (which may not be far enough)
> > that will only affect fds returned by "open" and similar
> > calls, rather than "socket".
>
> Yes, I had thought that might be it but I held out hope that it might be
> automagically and if not that the use open pragma would affect all IO
> operations.
>
> >> However saying;
> >>
> >> use encoding ':utf8';
> >>
> >> does.
> >
> > If that fixes it for you, then I'm pretty surprised as AFAIK
> > that merely relates to the script encoding. Maybe I'm behind
> > the times. (It's less than useful for me - seems to cause another
> > problem).
>
> Yes, as was I, my understanding was the same as yours, that it allows
> you to have scripts containing characters in that encoding, but it gets
> rid of the wide character in output warning so. What problem does it
> cause you?
>
> If I get around to digging into the SOAP::Lite classes to fix the
> problem there I'll let you know how I get on.
>
> I suppose I should have mentioned this earlier, I'm using Perl 5.8.7, so
> fairly recent.
>
>   n
>
>


-- 
Tatsuhiko Miyagawa