Filter::FormatText#2

HTMLの本文を整形したPlainテキストに変換します。

サンプル

- module: Filter::FormatText

  config:

    bytes_per_line: 1行の長さ(optional def.72)

    always: HTMLテキスト以外でも実行するか？(optional def.0)

使い方や目的はこちら

ダウンロード: FormatText.pm

 package Plagger::Plugin::Filter::FormatText;
 use HTML::TreeBuilder;

 use HTML::FormatText;

 use Jcode;
 use strict;

 use warnings;

 use base qw( Plagger::Plugin );
 our $VERSION = 0.03;
 sub register {

     my($self, $context) = @_;

     $context->register_hook(

         $self,

         'update.entry.fixup' => \&filter,

     );

 }
 sub filter {

     my($self, $context, $args) = @_;

     my $entry = $args->{entry};

     my $body      = $entry->body;

     my $body_text = $entry->body;

     $body_text =~ s/<[^>]*>//g;

     unless ($self->conf->{always} || $body eq $body_text){

       my $tree      = HTML::TreeBuilder->new()->parse($entry->body);

       my $formatter = HTML::FormatText->new( lm => 0, rm => 998 );

       my $line = $self->conf->{bytes_per_line} || 72;

       # http://72.14.209.104/search?q=cache:ENkjSG7daIIJ:rcl.hp.infoseek.co.jp/cgi-bin/wiki.cgi%3Fpage%3DPerl%252FTIPS+jfold+%E6%94%B9%E8%A1%8C

       my $body = join "\n", map {

           jcode($_)->jfold( $line, "\n")

       } split /\n/mx, $formatter->format($tree);

       utf8::encode($body);

       $entry->body($body);
       $context->log(info => "format $entry->{link}") if $entry->{link};

     }

 }
 1;
 __END__
 =head1 NAME
 Plagger::Plugin::Filter::FormatText - Format HTML as plaintext
 =head1 SYNOPSIS
   - module: Filter::FormatText
 =head1 DESCRIPTION
 This plugin is formatted into plaintext from HTML using HTML::Format.
 =head1 CONFiG
 =over 4
 =item bytes_per_line (Optional)
 Default 72.
 =item always (Optional)
 When set to 1, always format. Default to 0.
 =back
 =head1 AUTHOR
 Nobuhito Sato
 =head1 SEE ALSO
 L<Plagger>, L<HTML::FormatText>
 =cut