Description: Turn messy HTML into standards-compliant HTML with all close tags properly noted. Takes STDIN, or a list of files, and result goes on STDOUT.
#!/usr/bin/perl
use XML::LibXML;
print XML::LibXML->new->parse_html_string(join "", <>)->toStringHTML;
Replies are listed 'Best First'.
Re: HTML tidy, using XML::LibXML
by Matts (Deacon) on Jan 04, 2003 at 17:10 UTC
    If you've got sloppy HTML that needs tidying up and XML::LibXML barfs on it, you can change that one-liner to the following three-liner:
    my $parser = XML::LibXML->new(); $parser->recover(1); # Set recovery on error flag $parser->parse_html_string(join "", <>)->toStringHTML;
    XML::LibXML can also parse SGML. Most people don't know that (probably sucky docs ;-)

      Don't know if it's accurate but the docs say that won't work-

      This switch (recover) will only work with XML data rather than HTML data.

      (Update, July 2010: this does work and has for at least a few years -- don't know if it was always the case; I should have tested back when I initially responded.)