I worked around this problem, but I well think that it should considered a serious bug in XPathScript; please, let me know if I should file a (another) bug report on rt.cpan.org and if I should put it in AxKit or XML::XPathScript (or both, or... or?).

By the way, I worked around the bug this way:

* first of all, note that all my XML docs are ISO-8859-1 encoded;
* I import the Encode module on top of all XPS stylesheets
* when I extract a text node in a variable and it gets converted to
  UTF-8, I convert it back to ISO this way:

  my $pagetitle = findvalue('/bs/meta/title/text()') ;
  Encode::from_to($pagetitle,"utf8","iso-8859-1") ;

* on top of the stylesheet, I modify the content-type header this way:
  $r->content_type("text/html; charset=ISO-8859-1") ;

This fixes the problem but it may not be a general and/or optimal solution.

Ciao
--Marco

--
Marco Marongiu                            Email: [EMAIL PROTECTED]
System Administrator                      Phone: +39 070 460 1684
Tiscali S.p.A.                            Fax:   +39 070 460 9684
International IT Services


--------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]



Reply via email to