The Perl Toolchain Summit needs more sponsors. If your company depends on Perl, please support this very important event.

NAME

HTML::T5 - HTML validation in a Perl object

VERSION

Version 0.013

SYNOPSIS

    use HTML::T5;

    my $tidy = HTML::T5->new( {config_file => 'path/to/config'} );
    $tidy->ignore( type => TIDY_WARNING, type => TIDY_INFO );
    $tidy->parse( "foo.html", $contents_of_foo );

    for my $message ( $tidy->messages ) {
        print $message->as_string;
    }

DESCRIPTION

HTML::T5 is an HTML checker in a handy dandy object. It's meant as a replacement for HTML::Lint. If you're currently an HTML::Lint user looking to migrate, see the section "Converting from HTML::Lint".

HTML::T5 was forked from HTML::Tidy5 by Andy Lester (PETDANCE), thanks.

EXPORTS

Message types TIDY_ERROR, TIDY_WARNING and TIDY_INFO.

Everything else is an object method.

METHODS

new()

Create an HTML::T5 object.

    my $tidy = HTML::T5->new();

Optionally you can give a hashref of configuration parms.

    my $tidy = HTML::T5->new( {config_file => 'path/to/tidy.cfg'} );

This configuration file will be read and used when you clean or parse an HTML file.

You can also pass options directly to tidy.

    my $tidy = HTML::T5->new( {
                                    output_xhtml => 1,
                                    tidy_mark => 0,
                                } );

See tidy -help-config for the list of options supported by tidy.

The following options are not supported by HTML::T5:

  • quiet

messages()

Returns the messages accumulated.

clear_messages()

Clears the list of messages, in case you want to print and clear, print and clear. If you don't clear the messages, then each time you call parse() you'll be accumulating more in the list.

ignore( parm => value [, parm => value ] )

Specify types of messages to ignore. Note that the ignore flags must be set before calling parse(). You can call ignore() as many times as necessary to set up all your restrictions; the options will stack up.

  • type => TIDY_INFO|TIDY_WARNING|TIDY_ERROR

    Specifies the type of messages you want to ignore, either info or warnings or errors. If you wanted, you could call ignore on all three and get no messages at all.

        $tidy->ignore( type => TIDY_WARNING );
  • text => qr/regex/

  • text => [ qr/regex1/, qr/regex2/, ... ]

    Checks the text of the message against the specified regex or regexes, and ignores the message if there's a match. The value for the text parm may be either a regex, or a reference to a list of regexes.

        $tidy->ignore( text => qr/DOCTYPE/ );
        $tidy->ignore( text => [ qr/unsupported/, qr/proprietary/i ] );

parse( $filename, $str [, $str...] )

Parses a string, or list of strings, that make up a single HTML file.

The $filename parm is only used as an identifier for your use. The file is not actually read and opened.

Returns true if all went OK, or false if there was some problem calling tidy, or parsing tidy's output.

clean( $str [, $str...] )

Cleans a string, or list of strings, that make up a single HTML file.

Returns the cleaned string as a single string.

tidy_library_version()

Returns the version of the underling tidy library.

INSTALLING TIDY

HTML::T5 requires that Alien::TidyHTML5, which provides the C html-tidy library or ascertains its presence, be installed on your system. Normally this happens by installing HTML::T5 from CPAN, but you may need to have a C compiler installed.

You can probably obtain html-tidy through your distribution's package manager (make sure you install the development package with headers). You can also check the html-tidy home page is http://www.html-tidy.org/.

CONVERTING FROM HTML::Lint

HTML::T5 is different from HTML::Lint in a number of crucial ways.

  • It's not pure Perl

    HTML::T5 is mostly a happy wrapper around the html-tidy library.

  • The real work is done by someone else

    Changes to tidy may come down the pipe that I don't have control over. That's the price we pay for having it do a darn good job.

  • It's no longer bundled with its Test:: counterpart

    HTML::Lint came bundled with Test::HTML::Lint, but Test::HTML::T5 is a separate distribution. This saves the people who don't want the Test:: framework from pulling it in, and all its prerequisite modules.

BUGS & FEEDBACK

Please report any bugs or feature requests at the issue tracker on github http://github.com/shlomif/html-tidy5/issues. I will be notified, and then you'll automatically be notified of progress on your bug as I make changes.

Please do NOT use http://rt.cpan.org.

SUPPORT

You can find documentation for this module with the perldoc command.

    perldoc HTML::T5

You can also look for information at:

MOTIVATION

HTML::Tidy5 is painful to install on https://en.wikipedia.org/wiki/Continuous_integration services, and even my https://github.com/shlomif/ci-gen-framework cannot help enough. We hope that this fork based on Alien::TidyHTML5 will ease the installation pains and allow for less CPAN Testers failure.

MEDIA LINKS

Getting the green test bar (= for success) reminds me of this performance of "Green Light" by Lorde, live at Glastonbury which used to be on YouTube but the account was terminated.

ACKNOWLEDGEMENTS

Thanks to Rufus Cable, Jonathan Rockway, and Robert Bachmann for contributions.

Thanks to https://github.com/robrwo for the code to use Alien::TidyHTML5 .

Thanks to Andy Lester for maintaining HTML::Tidy5 and HTML::Tidy.

AUTHOR

Shlomi Fish, https://www.shlomifish.org/, https://metacpan.org/author/SHLOMIF

COPYRIGHT & LICENSE

Copyright (C) 2005-2018 by Andy Lester

This library is free software. You may modify or distribute it under the Artistic License v2.0.