githubFork/roundcubemail.git

alecpl

2008-10-03 f613a1e4e004dd03d7de80ea5a45bee71d3a3454

commit \| author \| age
4e17e6	1	<?php
T	2
	3	/*************************************************************************
8ac6fd	4	* *
A	5	* class.html2text.inc *
	6	* *
	7	*************************************************************************
	8	* *
	9	* Converts HTML to formatted plain text *
	10	* *
	11	* Copyright (c) 2005-2007 Jon Abernathy <jon@chuggnutt.com> *
	12	* All rights reserved. *
	13	* *
	14	* This script is free software; you can redistribute it and/or modify *
	15	* it under the terms of the GNU General Public License as published by *
	16	* the Free Software Foundation; either version 2 of the License, or *
	17	* (at your option) any later version. *
	18	* *
	19	* The GNU General Public License can be found at *
	20	* http://www.gnu.org/copyleft/gpl.html. *
	21	* *
	22	* This script is distributed in the hope that it will be useful, *
	23	* but WITHOUT ANY WARRANTY; without even the implied warranty of *
	24	* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the *
	25	* GNU General Public License for more details. *
	26	* *
	27	* Author(s): Jon Abernathy <jon@chuggnutt.com> *
	28	* *
	29	* Last modified: 08/08/07 *
	30	* *
	31	*************************************************************************/
4e17e6	32
T	33
	34	/**
8ac6fd	35	* Takes HTML and converts it to formatted, plain text.
A	36	*
	37	* Thanks to Alexander Krug (http://www.krugar.de/) to pointing out and
	38	* correcting an error in the regexp search array. Fixed 7/30/03.
	39	*
	40	* Updated set_html() function's file reading mechanism, 9/25/03.
	41	*
	42	* Thanks to Joss Sanglier (http://www.dancingbear.co.uk/) for adding
	43	* several more HTML entity codes to the $search and $replace arrays.
	44	* Updated 11/7/03.
	45	*
	46	* Thanks to Darius Kasperavicius (http://www.dar.dar.lt/) for
	47	* suggesting the addition of $allowed_tags and its supporting function
	48	* (which I slightly modified). Updated 3/12/04.
	49	*
	50	* Thanks to Justin Dearing for pointing out that a replacement for the
	51	* <TH> tag was missing, and suggesting an appropriate fix.
	52	* Updated 8/25/04.
	53	*
	54	* Thanks to Mathieu Collas (http://www.myefarm.com/) for finding a
	55	* display/formatting bug in the _build_link_list() function: email
	56	* readers would show the left bracket and number ("[1") as part of the
	57	* rendered email address.
	58	* Updated 12/16/04.
	59	*
	60	* Thanks to Wojciech Bajon (http://histeria.pl/) for submitting code
	61	* to handle relative links, which I hadn't considered. I modified his
	62	* code a bit to handle normal HTTP links and MAILTO links. Also for
	63	* suggesting three additional HTML entity codes to search for.
	64	* Updated 03/02/05.
	65	*
	66	* Thanks to Jacob Chandler for pointing out another link condition
	67	* for the _build_link_list() function: "https".
	68	* Updated 04/06/05.
	69	*
	70	* Thanks to Marc Bertrand (http://www.dresdensky.com/) for
	71	* suggesting a revision to the word wrapping functionality; if you
	72	* specify a $width of 0 or less, word wrapping will be ignored.
	73	* Updated 11/02/06.
	74	*
	75	* *** Big housecleaning updates below:
	76	*
	77	* Thanks to Colin Brown (http://www.sparkdriver.co.uk/) for
	78	* suggesting the fix to handle </li> and blank lines (whitespace).
	79	* Christian Basedau (http://www.movetheweb.de/) also suggested the
	80	* blank lines fix.
	81	*
	82	* Special thanks to Marcus Bointon (http://www.synchromedia.co.uk/),
	83	* Christian Basedau, Norbert Laposa (http://ln5.co.uk/),
	84	* Bas van de Weijer, and Marijn van Butselaar
	85	* for pointing out my glaring error in the <th> handling. Marcus also
	86	* supplied a host of fixes.
	87	*
	88	* Thanks to Jeffrey Silverman (http://www.newtnotes.com/) for pointing
	89	* out that extra spaces should be compressed--a problem addressed with
	90	* Marcus Bointon's fixes but that I had not yet incorporated.
	91	*
	92	* Thanks to Daniel Schledermann (http://www.typoconsult.dk/) for
	93	* suggesting a valuable fix with <a> tag handling.
	94	*
	95	* Thanks to Wojciech Bajon (again!) for suggesting fixes and additions,
	96	* including the <a> tag handling that Daniel Schledermann pointed
	97	* out but that I had not yet incorporated. I haven't (yet)
	98	* incorporated all of Wojciech's changes, though I may at some
	99	* future time.
	100	*
	101	* *** End of the housecleaning updates. Updated 08/08/07.
	102	*
	103	* @author Jon Abernathy <jon@chuggnutt.com>
	104	* @version 1.0.0
	105	* @since PHP 4.0.2
	106	*/
4e17e6	107	class html2text
T	108	{
	109
	110	/**
	111	* Contains the HTML content to convert.
	112	*
	113	* @var string $html
	114	* @access public
	115	*/
	116	var $html;
	117
	118	/**
	119	* Contains the converted, formatted text.
	120	*
	121	* @var string $text
	122	* @access public
	123	*/
	124	var $text;
	125
	126	/**
	127	* Maximum width of the formatted text, in columns.
	128	*
8ac6fd	129	* Set this value to 0 (or less) to ignore word wrapping
A	130	* and not constrain text to a fixed-width column.
	131	*
4e17e6	132	* @var integer $width
T	133	* @access public
	134	*/
	135	var $width = 70;
	136
	137	/**
	138	* List of preg* regular expression patterns to search for,
	139	* used in conjunction with $replace.
	140	*
	141	* @var array $search
	142	* @access public
	143	* @see $replace
	144	*/
	145	var $search = array(
	146	"/\r/", // Non-legal carriage return
	147	"/[\n\t]+/", // Newlines and tabs
8ac6fd	148	'/[ ]{2,}/', // Runs of spaces, pre-handling
4e17e6	149	'/<script[^>]>.?<\/script>/i', // <script>s -- which strip_tags supposedly has problems with
8ac6fd	150	'/<style[^>]>.?<\/style>/i', // <style>s -- which strip_tags supposedly has problems with
4e17e6	151	//'/<!-- .* -->/', // Comments -- which strip_tags might have problem a with
8ac6fd	152	'/<h[123][^>]>(.?)<\/h[123]>/ie', // H1 - H3
A	153	'/<h[456][^>]>(.?)<\/h[456]>/ie', // H4 - H6
4e17e6	154	'/<p[^>]*>/i', // <P>
T	155	'/<br[^>]*>/i', // <br>
8ac6fd	156	'/<b[^>]>(.?)<\/b>/ie', // <b>
A	157	'/<strong[^>]>(.?)<\/strong>/ie', // <strong>
	158	'/<i[^>]>(.?)<\/i>/i', // <i>
	159	'/<em[^>]>(.?)<\/em>/i', // <em>
4e17e6	160	'/(<ul[^>]*>\|<\/ul>)/i', // <ul> and </ul>
T	161	'/(<ol[^>]*>\|<\/ol>)/i', // <ol> and </ol>
8ac6fd	162	'/<li[^>]>(.?)<\/li>/i', // <li> and </li>
4e17e6	163	'/<li[^>]*>/i', // <li>
8ac6fd	164	'/<a [^>]href=("\|\')([^"\']+)\1[^>]>(.*?)<\/a>/ie',
A	165	// <a href="">
4e17e6	166	'/<hr[^>]*>/i', // <hr>
T	167	'/(<table[^>]*>\|<\/table>)/i', // <table> and </table>
	168	'/(<tr[^>]*>\|<\/tr>)/i', // <tr> and </tr>
8ac6fd	169	'/<td[^>]>(.?)<\/td>/i', // <td> and </td>
A	170	'/<th[^>]>(.?)<\/th>/ie', // <th> and </th>
	171	'/&(nbsp\|#160);/i', // Non-breaking space
	172	'/&(quot\|rdquo\|ldquo\|#8220\|#8221\|#147\|#148);/i',
	173	// Double quotes
	174	'/&(apos\|rsquo\|lsquo\|#8216\|#8217);/i', // Single quotes
	175	'/>/i', // Greater-than
	176	'/</i', // Less-than
	177	'/&(amp\|#38);/i', // Ampersand
	178	'/&(copy\|#169);/i', // Copyright
	179	'/&(trade\|#8482\|#153);/i', // Trademark
	180	'/&(reg\|#174);/i', // Registered
	181	'/&(mdash\|#151\|#8212);/i', // mdash
	182	'/&(ndash\|minus\|#8211\|#8722);/i', // ndash
	183	'/&(bull\|#149\|#8226);/i', // Bullet
	184	'/&(pound\|#163);/i', // Pound sign
	185	'/&(euro\|#8364);/i', // Euro sign
	186	'/&[^&;]+;/i', // Unknown/unhandled entities
	187	'/[ ]{2,}/' // Runs of spaces, post-handling
4e17e6	188	);
T	189
	190	/**
	191	* List of pattern replacements corresponding to patterns searched.
	192	*
	193	* @var array $replace
	194	* @access public
	195	* @see $search
	196	*/
	197	var $replace = array(
	198	'', // Non-legal carriage return
	199	' ', // Newlines and tabs
8ac6fd	200	' ', // Runs of spaces, pre-handling
4e17e6	201	'', // <script>s -- which strip_tags supposedly has problems with
8ac6fd	202	'', // <style>s -- which strip_tags supposedly has problems with
A	203	//'', // Comments -- which strip_tags might have problem a with
4e17e6	204	"strtoupper(\"\n\n\\1\n\n\")", // H1 - H3
8ac6fd	205	"ucwords(\"\n\n\\1\n\")", // H4 - H6
A	206	"\n\n", // <P>
4e17e6	207	"\n", // <br>
T	208	'strtoupper("\\1")', // <b>
8ac6fd	209	'strtoupper("\\1")', // <strong>
4e17e6	210	'_\\1_', // <i>
8ac6fd	211	'_\\1_', // <em>
4e17e6	212	"\n\n", // <ul> and </ul>
T	213	"\n\n", // <ol> and </ol>
8ac6fd	214	"\t* \\1\n", // <li> and </li>
A	215	"\n\t* ", // <li>
	216	'$this->_build_link_list("\\2", "\\3")',
	217	// <a href="">
	218	"\n-------------------------\n", // <hr>
	219	"\n\n", // <table> and </table>
4e17e6	220	"\n", // <tr> and </tr>
T	221	"\t\t\\1\n", // <td> and </td>
	222	"strtoupper(\"\t\t\\1\n\")", // <th> and </th>
8ac6fd	223	' ', // Non-breaking space
A	224	'"', // Double quotes
	225	"'", // Single quotes
4e17e6	226	'>',
T	227	'<',
	228	'&',
	229	'(c)',
	230	'(tm)',
	231	'(R)',
8ac6fd	232	'--',
A	233	'-',
4e17e6	234	'*',
8ac6fd	235	'£',
A	236	'EUR', // Euro sign. ?
	237	'', // Unknown/unhandled entities
	238	' ' // Runs of spaces, post-handling
7353fa	239	);
A	240
8ac6fd	241	/**
A	242	* List of preg* regular expression patterns to search for in PRE body,
	243	* used in conjunction with $pre_replace.
	244	*
	245	* @var array $pre_search
	246	* @access public
	247	* @see $pre_replace
	248	*/
7789fd	249	var $pre_search = array(
7353fa	250	"/\n/",
A	251	"/\t/",
	252	'/ /',
	253	'/<pre[^>]*>/',
	254	'/<\/pre>/'
	255	);
	256
	257	/**
	258	* List of pattern replacements corresponding to patterns searched for PRE body.
	259	*
	260	* @var array $pre_replace
	261	* @access public
	262	* @see $pre_search
	263	*/
7789fd	264	var $pre_replace = array(
7353fa	265	'<br>',
A	266	'    ',
	267	' ',
	268	'',
	269	''
4e17e6	270	);
T	271
	272	/**
	273	* Contains a list of HTML tags to allow in the resulting text.
	274	*
	275	* @var string $allowed_tags
	276	* @access public
	277	* @see set_allowed_tags()
	278	*/
	279	var $allowed_tags = '';
	280
	281	/**
	282	* Contains the base URL that relative links should resolve to.
	283	*
	284	* @var string $url
	285	* @access public
	286	*/
	287	var $url;
	288
	289	/**
	290	* Indicates whether content in the $html variable has been converted yet.
	291	*
8ac6fd	292	* @var boolean $_converted
4e17e6	293	* @access private
T	294	* @see $html, $text
	295	*/
	296	var $_converted = false;
	297
	298	/**
	299	* Contains URL addresses from links to be rendered in plain text.
	300	*
8ac6fd	301	* @var string $_link_list
4e17e6	302	* @access private
T	303	* @see _build_link_list()
	304	*/
8ac6fd	305	var $_link_list = '';
dd792e	306
S	307	/**
8ac6fd	308	* Number of valid links detected in the text, used for plain text
A	309	* display (rendered similar to footnotes).
	310	*
	311	* @var integer $_link_count
	312	* @access private
	313	* @see _build_link_list()
dd792e	314	*/
8ac6fd	315	var $_link_count = 0;
4e17e6	316
e7f85b	317	/**
A	318	* Boolean flag, true if a table of link URLs should be listed after the text.
	319	*
	320	* @var boolean $_do_links
	321	* @access private
	322	* @see html2text()
	323	*/
	324	var $_do_links = true;
	325
4e17e6	326	/**
T	327	* Constructor.
	328	*
	329	* If the HTML source string (or file) is supplied, the class
	330	* will instantiate with that source propagated, all that has
	331	* to be done it to call get_text().
	332	*
	333	* @param string $source HTML content
	334	* @param boolean $from_file Indicates $source is a file to pull content from
e7f85b	335	* @param boolean $do_links Indicate whether a table of link URLs is desired
A	336	* @param integer $width Maximum width of the formatted text, 0 for no limit
4e17e6	337	* @access public
T	338	* @return void
	339	*/
e7f85b	340	function html2text( $source = '', $from_file = false, $do_links = true, $width = 75 )
4e17e6	341	{
T	342	if ( !empty($source) ) {
	343	$this->set_html($source, $from_file);
	344	}
e7f85b	345
4e17e6	346	$this->set_base_url();
e7f85b	347	$this->_do_links = $do_links;
A	348	$this->width = $width;
4e17e6	349	}
T	350
	351	/**
	352	* Loads source HTML into memory, either from $source string or a file.
	353	*
	354	* @param string $source HTML content
	355	* @param boolean $from_file Indicates $source is a file to pull content from
	356	* @access public
	357	* @return void
	358	*/
	359	function set_html( $source, $from_file = false )
	360	{
	361	if ( $from_file && file_exists($source) ) {
8ac6fd	362	$this->html = file_get_contents($source);
4e17e6	363	}
8ac6fd	364	else
A	365	$this->html = $source;
4e17e6	366
T	367	$this->_converted = false;
	368	}
	369
	370	/**
	371	* Returns the text, converted from HTML.
	372	*
	373	* @access public
	374	* @return string
	375	*/
	376	function get_text()
	377	{
	378	if ( !$this->_converted ) {
	379	$this->_convert();
	380	}
	381
	382	return $this->text;
	383	}
	384
	385	/**
	386	* Prints the text, converted from HTML.
	387	*
	388	* @access public
	389	* @return void
	390	*/
	391	function print_text()
	392	{
	393	print $this->get_text();
	394	}
	395
	396	/**
	397	* Alias to print_text(), operates identically.
	398	*
	399	* @access public
	400	* @return void
	401	* @see print_text()
	402	*/
	403	function p()
	404	{
	405	print $this->get_text();
	406	}
	407
	408	/**
	409	* Sets the allowed HTML tags to pass through to the resulting text.
	410	*
	411	* Tags should be in the form "<p>", with no corresponding closing tag.
	412	*
	413	* @access public
	414	* @return void
	415	*/
	416	function set_allowed_tags( $allowed_tags = '' )
	417	{
	418	if ( !empty($allowed_tags) ) {
	419	$this->allowed_tags = $allowed_tags;
	420	}
	421	}
	422
	423	/**
	424	* Sets a base URL to handle relative links.
	425	*
	426	* @access public
	427	* @return void
	428	*/
	429	function set_base_url( $url = '' )
	430	{
	431	if ( empty($url) ) {
8ac6fd	432	if ( !empty($_SERVER['HTTP_HOST']) ) {
A	433	$this->url = 'http://' . $_SERVER['HTTP_HOST'];
	434	} else {
	435	$this->url = '';
	436	}
4e17e6	437	} else {
T	438	// Strip any trailing slashes for consistency (relative
	439	// URLs may already start with a slash like "/file.html")
	440	if ( substr($url, -1) == '/' ) {
	441	$url = substr($url, 0, -1);
	442	}
	443	$this->url = $url;
	444	}
	445	}
	446
	447	/**
	448	* Workhorse function that does actual conversion.
	449	*
	450	* First performs custom tag replacement specified by $search and
	451	* $replace arrays. Then strips any remaining HTML tags, reduces whitespace
	452	* and newlines to a readable format, and word wraps the text to
	453	* $width characters.
	454	*
	455	* @access private
	456	* @return void
	457	*/
	458	function _convert()
	459	{
	460	// Variables used for building the link list
8ac6fd	461	$this->_link_count = 0;
A	462	$this->_link_list = '';
4e17e6	463
T	464	$text = trim(stripslashes($this->html));
	465
7353fa	466	// Convert <PRE>
8ac6fd	467	$this->_convert_pre($text);
A	468
4e17e6	469	// Run our defined search-and-replace
T	470	$text = preg_replace($this->search, $this->replace, $text);
	471
	472	// Strip any other HTML tags
	473	$text = strip_tags($text, $this->allowed_tags);
	474
	475	// Bring down number of empty lines to 2 max
8ac6fd	476	$text = preg_replace("/\n\s+\n/", "\n\n", $text);
4e17e6	477	$text = preg_replace("/[\n]{3,}/", "\n\n", $text);
T	478
	479	// Add link list
8ac6fd	480	if ( !empty($this->_link_list) ) {
A	481	$text .= "\n\nLinks:\n------\n" . $this->_link_list;
4e17e6	482	}
T	483
	484	// Wrap the text to a readable format
	485	// for PHP versions >= 4.0.2. Default width is 75
8ac6fd	486	// If width is 0 or less, don't wrap the text.
A	487	if ( $this->width > 0 ) {
	488	$text = wordwrap($text, $this->width);
	489	}
4e17e6	490
T	491	$this->text = $text;
	492
	493	$this->_converted = true;
	494	}
	495
	496	/**
	497	* Helper function called by preg_replace() on link replacement.
	498	*
	499	* Maintains an internal list of links to be displayed at the end of the
	500	* text, with numeric indices to the original point in the text they
	501	* appeared. Also makes an effort at identifying and handling absolute
	502	* and relative links.
	503	*
	504	* @param string $link URL of the link
	505	* @param string $display Part of the text to associate number with
	506	* @access private
	507	* @return string
8ac6fd	508	*/
A	509	function _build_link_list( $link, $display )
	510	{
e7f85b	511	if ( !$this->_do_links ) return $display;
A	512
	513	if ( substr($link, 0, 7) == 'http://' \|\| substr($link, 0, 8) == 'https://' \|\|
8ac6fd	514	substr($link, 0, 7) == 'mailto:' ) {
A	515	$this->_link_count++;
	516	$this->_link_list .= "[" . $this->_link_count . "] $link\n";
	517	$additional = ' [' . $this->_link_count . ']';
	518	} elseif ( substr($link, 0, 11) == 'javascript:' ) {
	519	// Don't count the link; ignore it
	520	$additional = '';
	521	// what about href="#anchor" ?
	522	} else {
	523	$this->_link_count++;
	524	$this->_link_list .= "[" . $this->_link_count . "] " . $this->url;
	525	if ( substr($link, 0, 1) != '/' ) {
	526	$this->_link_list .= '/';
4e17e6	527	}
8ac6fd	528	$this->_link_list .= "$link\n";
A	529	$additional = ' [' . $this->_link_count . ']';
4e17e6	530	}
T	531
8ac6fd	532	return $display . $additional;
A	533	}
	534
7353fa	535	/**
A	536	* Helper function for PRE body conversion.
	537	*
	538	* @param string HTML content
	539	* @access private
8ac6fd	540	*/
7353fa	541	function _convert_pre(&$text)
8ac6fd	542	{
A	543	while(preg_match('/<pre[^>]>(.)<\/pre>/ismU', $text, $matches))
	544	{
	545	$result = preg_replace($this->pre_search, $this->pre_replace, $matches[1]);
	546	$text = preg_replace('/<pre[^>]>.<\/pre>/ismU', '<div><br>' . $result . '<br></div>', $text);
	547	}
	548	}
4e17e6	549	}
T	550
7353fa	551	?>