HTML to XML in Perl?

[Ilya Zakharevich]

Suppose I want to translate an HTML to a XML-well-formed HTML (so that
I can, e.g., apply xsltproc to the result). E.g., HTML::TreeBuilder
can apply "usual heuristics" to parse HTML; how to get XML out of it?

Thanks,
Ilya

[John Bokma]

Ilya Zakharevich <nospam-abuse@ilyaz.org> wrote:

> Suppose I want to translate an HTML to a XML-well-formed HTML (so that
> I can, e.g., apply xsltproc to the result). E.g., HTML::TreeBuilder
> can apply "usual heuristics" to parse HTML; how to get XML out of it?

Question: I use XML, not XHTML, at home, and use XML::Twig to convert it
to HTML. I can use xsltproc if I want to on the XML file.

You might want to traverse the parse tree HTML::TreeBuilder generates.
Also, not 100% sure, but it might me that HTML tidy can do the XHTML
conversion for you:

Google...

"Validator fixes errors in HTML and XHTML. Converts HTML to XHTML. Free
Software."

[url]http://www.google.com/search?q=html%20tidy%20xhtml[/url]

Sounds like it does :-D.

--
John Bokma Freelance software developer
&
Experienced Perl programmer: [url]http://castleamber.com/[/url]

[mirod]

Suppose I want to translate an HTML to a XML-well-formed HTML (so that
I can, e.g., apply xsltproc to the result). E.g., HTML::TreeBuilder
can apply "usual heuristics" to parse HTML; how to get XML out of it?

HTML::TreeBuilder has an as_XML method that does a pretty good job at getting proper XML from any HTML. It's not 100% reliable, but it mostly works.

I found HTML::TreeBuilder to be more useful for converting HTML to XML than XML::LibXML, which does not parse a lot of crappy HTML, and than HTML::Tidy, which is (was?) a pain to use.