MediaWiki extension: HtmlToWiki

InfoDabble > Tech Notes > MediaWiki > MediaWiki extension: HtmlToWiki

This project is still under development. See Writing a new special page

**This is a MediaWiki Extension**
Release status: beta
Description:	HTML to wikitext converter
Author:	Eric Hartwell
Version:	0.5 (2007.8.10)
MediaWiki:	1.10
Download:	Code

1 What can this extension do?
2 Usage
3 Installation
4 Implementation
- 4.1 Tips, tricks, and hoops
- 4.2 Version history
5 See also

[edit] What can this extension do?

This extension is yet another HTML to wikitext converter (Try it!). What's different about this one?

Completely written in PHP
Installs as a special page in your wiki
Uses HTML Purifier, a standards-compliant HTML filter library written in PHP, to parse almost any HTML input source
Converts a basic subset of HTML tags to MediaWiki wikitext format
- Easy to modify handling of particular tags and attributes

For a more powerful online converter, use Diberri's Perl-based online HTML::WikiConverter page.

[edit] Usage

[edit] Installation

[edit] Parameters

[edit] Changes to LocalSettings.php

require_once("$IP/extensions/HtmlToWiki/HtmlToWiki.php");

[edit] Code

[edit] HtmlToWiki.php

<?php
# Not a valid entry point, skip unless MEDIAWIKI is defined
if (!defined('MEDIAWIKI')) {
        echo <<<EOT
To install my extension, put the following line in LocalSettings.php:
require_once( "$IP/extensions/HtmlToWiki/HtmlToWiki.php" );
EOT;
        exit( 1 );
}
 
$wgAutoloadClasses['HtmlToWiki'] = dirname(__FILE__) . '/HtmlToWiki_body.php';
$wgSpecialPages['HtmlToWiki'] = 'HtmlToWiki';
$wgHooks['LoadAllMessages'][] = 'HtmlToWiki::loadMessages';
$wgHooks['LangugeGetSpecialPageAliases'][] = 'HtmlToWikiLocalizedPageName';
 
   $wgExtensionCredits['specialpage'][] = array(
       'name' => 'HtmlToWiki',
       'author' =>'Eric Hartwell', 
       'url' => 'http://www.ehartwell.com',
       'description' => 'HTML to wikitext converter'
       );
 
function HtmlToWikiLocalizedPageName(&$specialPageArray, $code) {
  # The localized title of the special page is among the messages of the
  # extension:
  HtmlToWiki::loadMessages();
  $text = wfMsg('htmltowiki');
 
  # Convert from title in text form to DBKey and put it into the alias array:
  $title = Title::newFromText($text);
  $specialPageArray[wfMsg('htmltowiki')][] = $title->getDBKey();
 
  return true;
}
 
?>

[edit] HtmlToWiki_body.php

<?php
class HtmlToWiki extends SpecialPage
{
	function HtmlToWiki() {
		self::loadMessages();
		SpecialPage::SpecialPage(wfMsg('htmltowiki'));
	}
 
    function execute( $par ) {
		global $wgRequest, $wgOut;
 
		$this->setHeaders();
		$self = Title::makeTitle( NS_SPECIAL, wfMsg('htmltowiki') );
 
		# Get request data from, e.g.
		$param = $wgRequest->getText('param');
		$html = $wgRequest->getText('convert');
 
		# Only load the HTML Purifier code if we're going to use it
		require( dirname( __FILE__ ) . '/HtmlToWikiPurifier.php' );
 
		# Build the input form using Xml.php (see: Special:LinkSearch)
		$wgOut->addWikiText( wfMsg('htmltowiki-text') );
		$wgOut->addHtml( 
						Xml::openElement( 'form', array( 'id' => 'mw-htmltowiki-form', 
										  'method' => 'post', 'action' => $self->escapeLocalURL() ) ) .
						xml_startfieldset( 'input', wfMsg( 'htmltowiki-input' ) ) .
						xml_textarea( 'convert', ($html != '') ? 6 : 20, 100, $html ) . 
						Xml::submitButton( wfMsg( 'htmltowiki-convert' ) ) . 
						xml_endfieldset() 
					);
 
		// Show the converted wikitext if there's something to convert
		if ( $html != '' ) {
			$parser = new HTMLPurifierWikitext();
			$out = $parser->purifyWikitext( $html );
			$wgOut->addHtml( 
					xml_startfieldset( 'output', wfMsg( 'htmltowiki-output' ) ) .
					xml_textarea( 'wikitext', 15, 100, $out ) . 
					xml_endfieldset()  
				);
		}
 
		// Finish off the input/output form
		$wgOut->addHtml( Xml::closeElement( 'form' ) );
		$wgOut->addWikiText( wfMsg('htmltowiki-credits') );
    }
 
    function loadMessages() {
		static $messagesLoaded = false;
		global $wgMessageCache;
		if ( $messagesLoaded ) return;
		$messagesLoaded = true;
 
		require( dirname( __FILE__ ) . '/HtmlToWiki.i18n.php' );
		foreach ( $allMessages as $lang => $langMessages ) {
				$wgMessageCache->addMessages( $langMessages, $lang );
		}
    }
}
 
 
/**
 * Convenience functions to build an HTML text input field
 * @return string HTML
 */
 ## Note: <textarea value=xxx /> doesn't work with IE
function xml_textarea( $name, $rows=false, $cols=false, $value=false, $attribs=array() ) {
	return( 
			Xml::openElement( 'textarea', 
							  array( 'name' => $name, 'cols' => $cols, 'rows' => $rows ) + $attribs,
							  $value )
			. $value
			. Xml::closeElement('textarea') );
}
 
function xml_startfieldset( $id=false, $label=false ) {
 	$out = Xml::openElement( 'fieldset', array( 'id' => $id ));
 	if ( $label )	$out .= Xml::element( 'legend', array(), $label );
	return ( $out );
}
 
function xml_endfieldset() {
	return(	Xml::closeElement( 'fieldset' ) );
}
 
?>

[edit] HtmlToWiki.i18n.php

<?php
$allMessages = array(
        'en' => array( 
                'htmltowiki' 			=> 'HtmlToWiki',
				'htmltowiki-text'		=> 'Convert HTML to wikitext:',
				'htmltowiki-convert'	=> 'Convert to wikitext',
				'htmltowiki-input'		=> 'HTML source',
				'htmltowiki-output'		=> 'Converted wikitext',
				'htmltowiki-credits'	=> 'For more information about this utility, see [http://www.ehartwell.com/InfoDabble/MediaWiki_extension:_HtmlToWiki HtmlToWiki]. This utility uses the [http://htmlpurifier.org/ HTML Purifier] library, which is licensed under the LGPL v2.1+.',
        ),
);
?>

[edit] HtmlToWikiPurifier.php

<?php
class HtmlToWiki extends SpecialPage
{
	function HtmlToWiki() {
		self::loadMessages();
		SpecialPage::SpecialPage(wfMsg('htmltowiki'));
	}
 
    function execute( $par ) {
		global $wgRequest, $wgOut;
 
		$this->setHeaders();
		$self = Title::makeTitle( NS_SPECIAL, wfMsg('htmltowiki') );
 
		# Get request data from, e.g.
		$param = $wgRequest->getText('param');
		$html = $wgRequest->getText('convert');
 
		# Only load the HTML Purifier code if we're going to use it
		require( dirname( __FILE__ ) . '/HtmlToWikiPurifier.php' );
 
		# Build the input form using Xml.php (see: Special:LinkSearch)
		$wgOut->addWikiText( wfMsg('htmltowiki-text') );
		$wgOut->addHtml( 
						Xml::openElement( 'form', array( 'id' => 'mw-htmltowiki-form', 
										  'method' => 'post', 'action' => $self->escapeLocalURL() ) ) .
						xml_startfieldset( 'input', wfMsg( 'htmltowiki-input' ) ) .
						xml_textarea( 'convert', ($html != '') ? 6 : 20, 100, $html ) . 
						Xml::submitButton( wfMsg( 'htmltowiki-convert' ) ) . 
						xml_endfieldset() 
					);
 
		// Show the converted wikitext if there's something to convert
		if ( $html != '' ) {
			$parser = new HTMLPurifierWikitext();
			$out = $parser->purifyWikitext( $html );
			$wgOut->addHtml( 
					xml_startfieldset( 'output', wfMsg( 'htmltowiki-output' ) ) .
					xml_textarea( 'wikitext', 15, 100, $out ) . 
					xml_endfieldset()  
				);
		}
 
		// Finish off the input/output form
		$wgOut->addHtml( Xml::closeElement( 'form' ) );
		$wgOut->addWikiText( wfMsg('htmltowiki-credits') );
    }
 
    function loadMessages() {
		static $messagesLoaded = false;
		global $wgMessageCache;
		if ( $messagesLoaded ) return;
		$messagesLoaded = true;
 
		require( dirname( __FILE__ ) . '/HtmlToWiki.i18n.php' );
		foreach ( $allMessages as $lang => $langMessages ) {
				$wgMessageCache->addMessages( $langMessages, $lang );
		}
    }
}
 
 
/**
 * Convenience functions to build an HTML text input field
 * @return string HTML
 */
 ## Note: <textarea value=xxx /> doesn't work with IE
function xml_textarea( $name, $rows=false, $cols=false, $value=false, $attribs=array() ) {
	return( 
			Xml::openElement( 'textarea', 
							  array( 'name' => $name, 'cols' => $cols, 'rows' => $rows ) + $attribs,
							  $value )
			. $value
			. Xml::closeElement('textarea') );
}
 
function xml_startfieldset( $id=false, $label=false ) {
 	$out = Xml::openElement( 'fieldset', array( 'id' => $id ));
 	if ( $label )	$out .= Xml::element( 'legend', array(), $label );
	return ( $out );
}
 
function xml_endfieldset() {
	return(	Xml::closeElement( 'fieldset' ) );
}
 
?>

[edit] Implementation

Suprisingly enough, it's hard to find a current HTML parser written in PHP.

Finally I found HTML Purifier, which bills itself as a standards-compliant HTML filter library.

[edit] Tips, tricks, and hoops

Initially I hoped to use PHP's DOM functions, but they are only suitable for use with clean XML. For example, it doesn't tokenize common nested tags such as:
<p>This is <b>typical</b> HTML markup.</p>
"[Y]ou save my day by allowing me not to write another damned HTML parser."

[edit] Version history

August 10, 2007 - initial version

[edit] See also

For a more powerful online converter, use Diberri's Perl-based online HTML::WikiConverter page
Send2Wiki

Other HTML parsers and converters:

Magic HTML Parser (PHP Classes) Parse HTML documents and extract keywords
HTML Parser Class (Totally PHP) HTML parser class that it supports the innerHTML property
DOM Functions (PHP.net) operate on XML documents through the DOM API with PHP 5.
HTML table to wiki converter (DZone) to html python wiki conversion
HTML Parser for PHP 4 (SourceForge) Object oriented PHP based HTML parser

Retrieved from "http://www.ehartwell.com/InfoDabble/MediaWiki_extension:_HtmlToWiki"

Navigation

Search

MediaWiki extension: HtmlToWiki

Contents

[edit] What can this extension do?

[edit] Usage

[edit] Installation

[edit] Parameters

[edit] Changes to LocalSettings.php

[edit] Code

[edit] HtmlToWiki.php

[edit] HtmlToWiki_body.php

[edit] HtmlToWiki.i18n.php

[edit] HtmlToWikiPurifier.php

[edit] Implementation

[edit] Tips, tricks, and hoops

[edit] Version history

[edit] See also

<Footer>

Navigation

Search

MediaWiki extension: HtmlToWiki

Contents

[edit] What can this extension do?

[edit] Usage

[edit] Installation

[edit] Parameters

[edit] Changes to LocalSettings.php

[edit] Code

[edit] HtmlToWiki.php

[edit] HtmlToWiki_body.php

[edit] HtmlToWiki.i18n.php

[edit] HtmlToWikiPurifier.php

[edit] Implementation

[edit] Tips, tricks, and hoops

[edit] Version history

[edit] See also

&lt;Footer&gt;

<Footer>