package Daizu::Plugin::XHTMLArticle;
use warnings;
use strict;
use XML::LibXML;
use Carp qw( croak );
use Encode qw( decode );
use Daizu::Util qw(
url_encode daizu_data_dir
);
=head1 NAME
Daizu::Plugin::XHTMLArticle - plugin for loading articles written in XHTML
=head1 DESCRIPTION
This article loader plugin allows you to write Daizu articles in XHTML
format. The content of the files it is used for must be well-formed XML,
except that there is no need to have a single root element. The content
is wrapped in a C element before being parsed. This plugin will
fail if there are any errors during parsing.
The default namespace declared on the root C element is the XHTML
namespace. The namespace prefix C is also declared and mapped to
the Daizu HTML extension namespace.
A DTD is also included automatically. It doesn't validate the input,
but it does provide all the standard HTML entity references, so for
example you can use C< > to get a non-breaking space.
TODO - link to a page describing the HTML extensions.
TODO - describe the XInclude support and daizu: URI scheme.
=head1 CONFIGURATION
To turn on this plugin, include the following in your Daizu CMS configuration
file:
=for syntax-highlight xml
=head1 METHODS
=over
=item Daizu::Plugin::XHTMLArticle-Eregister($cms, $whole_config, $plugin_config, $path)
Called by Daizu CMS when the plugin is registered. It registers the
Lload_article($cms, $file)> method as
an article loader for the MIME types 'text/html' and 'application/xhtml+xml'.
The configuration is currently ignored.
=cut
sub register
{
my ($class, $cms, $whole_config, $plugin_config, $path) = @_;
my $self = bless {}, $class;
$cms->add_article_loader($_, '', $self => 'load_article')
for qw( text/html application/xhtml+xml );
}
=item $self-Eload_article($cms, $file)
Does the actual parsing of the XHTML content of C<$file> (which should
be a L object), and returns the appropriate content as an XHTML
DOM of the file.
Never rejects a file, and therefore always returns true.
=cut
sub load_article
{
my ($self, $cms, $file) = @_;
my $parser = XML::LibXML->new;
$parser->pedantic_parser(1);
$parser->validation(0);
$parser->line_numbers(1);
my $xml_dir = url_encode(daizu_data_dir('xml')->stringify);
my $doc = $parser->parse_string(
'' .
qq{} .
qq{} .
decode('UTF-8', ${$file->data}, Encode::FB_CROAK) .
''
);
return { content => $doc };
}
=back
=head1 COPYRIGHT
This software is copyright 2006 Geoff Richards Egeoff@laxan.comE.
For licensing information see this page:
L
=cut
1;
# vi:ts=4 sw=4 expandtab