library/classes/html2text.class.php

   1 <?php
   2
   3 /*************************************************************************
   4  *                                                                       *
   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  *************************************************************************/
  32
  33
  34 /**
  35  *  Takes HTML and converts it to formatted, plain text.
  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  *  Modified 2015-10-04 by Rod Roark <rod@sunsetsystems.com>
 104  *  Added an argument to the _convert function to allow omission of the
 105  *  links at the end.
 106  *
 107  *  @author Jon Abernathy <jon@chuggnutt.com>
 108  *  @version 1.0.0
 109  *  @since PHP 4.0.2
 110  */
 111 class html2text
 112 {
 113
 114     /**
 115      *  Contains the HTML content to convert.
 116      *
 117      *  @var string $html
 118      *  @access public
 119      */
 120     var $html;
 121
 122     /**
 123      *  Contains the converted, formatted text.
 124      *
 125      *  @var string $text
 126      *  @access public
 127      */
 128     var $text;
 129
 130     /**
 131      *  Maximum width of the formatted text, in columns.
 132      *
 133      *  Set this value to 0 (or less) to ignore word wrapping
 134      *  and not constrain text to a fixed-width column.
 135      *
 136      *  @var integer $width
 137      *  @access public
 138      */
 139     var $width = 70;
 140
 141     /**
 142      *  List of preg* regular expression patterns to search for,
 143      *  used in conjunction with $replace.
 144      *
 145      *  @var array $search
 146      *  @access public
 147      *  @see $replace
 148      */
 149     var $search = array(
 150         "/\r/",                                  // Non-legal carriage return
 151         "/[\n\t]+/",                             // Newlines and tabs
 152         '/[ ]{2,}/',                             // Runs of spaces, pre-handling
 153         '/<script[^>]*>.*?<\/script>/i',         // <script>s -- which strip_tags supposedly has problems with
 154         '/<style[^>]*>.*?<\/style>/i',           // <style>s -- which strip_tags supposedly has problems with
 155         //'/<!-- .* -->/',                         // Comments -- which strip_tags might have problem a with
 156         '/<h[123][^>]*>(.*?)<\/h[123]>/ie',      // H1 - H3
 157         '/<h[456][^>]*>(.*?)<\/h[456]>/ie',      // H4 - H6
 158         '/<p[^>]*>/i',                           // <P>
 159         '/<br[^>]*>/i',                          // <br>
 160         '/<b[^>]*>(.*?)<\/b>/ie',                // <b>
 161         '/<strong[^>]*>(.*?)<\/strong>/ie',      // <strong>
 162         '/<i[^>]*>(.*?)<\/i>/i',                 // <i>
 163         '/<em[^>]*>(.*?)<\/em>/i',               // <em>
 164         '/(<ul[^>]*>|<\/ul>)/i',                 // <ul> and </ul>
 165         '/(<ol[^>]*>|<\/ol>)/i',                 // <ol> and </ol>
 166         '/<li[^>]*>(.*?)<\/li>/i',               // <li> and </li>
 167         '/<li[^>]*>/i',                          // <li>
 168         '/<a [^>]*href="([^"]+)"[^>]*>(.*?)<\/a>/ie',
 169                                                  // <a href="">
 170         '/<hr[^>]*>/i',                          // <hr>
 171         '/(<table[^>]*>|<\/table>)/i',           // <table> and </table>
 172         '/(<tr[^>]*>|<\/tr>)/i',                 // <tr> and </tr>
 173         '/<td[^>]*>(.*?)<\/td>/i',               // <td> and </td>
 174         '/<th[^>]*>(.*?)<\/th>/ie',              // <th> and </th>
 175         '/&(nbsp|#160);/i',                      // Non-breaking space
 176         '/&(quot|rdquo|ldquo|#8220|#8221|#147|#148);/i',
 177                                                          // Double quotes
 178         '/&(apos|rsquo|lsquo|#8216|#8217);/i',   // Single quotes
 179         '/&gt;/i',                               // Greater-than
 180         '/&lt;/i',                               // Less-than
 181         '/&(amp|#38);/i',                        // Ampersand
 182         '/&(copy|#169);/i',                      // Copyright
 183         '/&(trade|#8482|#153);/i',               // Trademark
 184         '/&(reg|#174);/i',                       // Registered
 185         '/&(mdash|#151|#8212);/i',               // mdash
 186         '/&(ndash|minus|#8211|#8722);/i',        // ndash
 187         '/&(bull|#149|#8226);/i',                // Bullet
 188         '/&(pound|#163);/i',                     // Pound sign
 189         '/&(euro|#8364);/i',                     // Euro sign
 190         '/&[^&;]+;/i',                           // Unknown/unhandled entities
 191         '/[ ]{2,}/'                              // Runs of spaces, post-handling
 192     );
 193
 194     /**
 195      *  List of pattern replacements corresponding to patterns searched.
 196      *
 197      *  @var array $replace
 198      *  @access public
 199      *  @see $search
 200      */
 201     var $replace = array(
 202         '',                                     // Non-legal carriage return
 203         ' ',                                    // Newlines and tabs
 204         ' ',                                    // Runs of spaces, pre-handling
 205         '',                                     // <script>s -- which strip_tags supposedly has problems with
 206         '',                                     // <style>s -- which strip_tags supposedly has problems with
 207         //'',                                     // Comments -- which strip_tags might have problem a with
 208         "strtoupper(\"\n\n\\1\n\n\")",          // H1 - H3
 209         "ucwords(\"\n\n\\1\n\n\")",             // H4 - H6
 210         "\n\n\t",                               // <P>
 211         "\n",                                   // <br>
 212         'strtoupper("\\1")',                    // <b>
 213         'strtoupper("\\1")',                    // <strong>
 214         '_\\1_',                                // <i>
 215         '_\\1_',                                // <em>
 216         "\n\n",                                 // <ul> and </ul>
 217         "\n\n",                                 // <ol> and </ol>
 218         "\t* \\1\n",                            // <li> and </li>
 219         "\n\t* ",                               // <li>
 220         '$this->_build_link_list("\\1", "\\2")',
 221                                                 // <a href="">
 222         "\n-------------------------\n",        // <hr>
 223         "\n\n",                                 // <table> and </table>
 224         "\n",                                   // <tr> and </tr>
 225         "\t\t\\1\n",                            // <td> and </td>
 226         "strtoupper(\"\t\t\\1\n\")",            // <th> and </th>
 227         ' ',                                    // Non-breaking space
 228         '"',                                    // Double quotes
 229         "'",                                    // Single quotes
 230         '>',
 231         '<',
 232         '&',
 233         '(c)',
 234         '(tm)',
 235         '(R)',
 236         '--',
 237         '-',
 238         '*',
 239         '£',
 240         'EUR',                                  // Euro sign. € ?
 241         '',                                     // Unknown/unhandled entities
 242         ' '                                     // Runs of spaces, post-handling
 243     );
 244
 245     /**
 246      *  Contains a list of HTML tags to allow in the resulting text.
 247      *
 248      *  @var string $allowed_tags
 249      *  @access public
 250      *  @see set_allowed_tags()
 251      */
 252     var $allowed_tags = '';
 253
 254     /**
 255      *  Contains the base URL that relative links should resolve to.
 256      *
 257      *  @var string $url
 258      *  @access public
 259      */
 260     var $url;
 261
 262     /**
 263      *  Indicates whether content in the $html variable has been converted yet.
 264      *
 265      *  @var boolean $_converted
 266      *  @access private
 267      *  @see $html, $text
 268      */
 269     var $_converted = false;
 270
 271     /**
 272      *  Contains URL addresses from links to be rendered in plain text.
 273      *
 274      *  @var string $_link_list
 275      *  @access private
 276      *  @see _build_link_list()
 277      */
 278     var $_link_list = '';
 279
 280     /**
 281      *  Number of valid links detected in the text, used for plain text
 282      *  display (rendered similar to footnotes).
 283      *
 284      *  @var integer $_link_count
 285      *  @access private
 286      *  @see _build_link_list()
 287      */
 288     var $_link_count = 0;
 289
 290     /**
 291      *  Constructor.
 292      *
 293      *  If the HTML source string (or file) is supplied, the class
 294      *  will instantiate with that source propagated, all that has
 295      *  to be done it to call get_text().
 296      *
 297      *  @param string $source HTML content
 298      *  @param boolean $from_file Indicates $source is a file to pull content from
 299      *  @access public
 300      *  @return void
 301      */
 302     function __construct( $source = '', $from_file = false )
 303     {
 304         if ( !empty($source) ) {
 305             $this->set_html($source, $from_file);
 306         }
 307         $this->set_base_url();
 308     }
 309
 310     /**
 311      *  Loads source HTML into memory, either from $source string or a file.
 312      *
 313      *  @param string $source HTML content
 314      *  @param boolean $from_file Indicates $source is a file to pull content from
 315      *  @access public
 316      *  @return void
 317      */
 318     function set_html( $source, $from_file = false )
 319     {
 320         $this->html = $source;
 321
 322         if ( $from_file && file_exists($source) ) {
 323             $fp = fopen($source, 'r');
 324             $this->html = fread($fp, filesize($source));
 325             fclose($fp);
 326         }
 327
 328         $this->_converted = false;
 329     }
 330
 331     /**
 332      *  Returns the text, converted from HTML.
 333      *
 334      *  @access public
 335      *  @return string
 336      */
 337     function get_text()
 338     {
 339         if ( !$this->_converted ) {
 340             $this->_convert();
 341         }
 342
 343         return $this->text;
 344     }
 345
 346     /**
 347      *  Prints the text, converted from HTML.
 348      *
 349      *  @access public
 350      *  @return void
 351      */
 352     function print_text()
 353     {
 354         print $this->get_text();
 355     }
 356
 357     /**
 358      *  Alias to print_text(), operates identically.
 359      *
 360      *  @access public
 361      *  @return void
 362      *  @see print_text()
 363      */
 364     function p()
 365     {
 366         print $this->get_text();
 367     }
 368
 369     /**
 370      *  Sets the allowed HTML tags to pass through to the resulting text.
 371      *
 372      *  Tags should be in the form "<p>", with no corresponding closing tag.
 373      *
 374      *  @access public
 375      *  @return void
 376      */
 377     function set_allowed_tags( $allowed_tags = '' )
 378     {
 379         if ( !empty($allowed_tags) ) {
 380             $this->allowed_tags = $allowed_tags;
 381         }
 382     }
 383
 384     /**
 385      *  Sets a base URL to handle relative links.
 386      *
 387      *  @access public
 388      *  @return void
 389      */
 390     function set_base_url( $url = '' )
 391     {
 392         if ( empty($url) ) {
 393                 if ( !empty($_SERVER['HTTP_HOST']) ) {
 394                     $this->url = 'http://' . $_SERVER['HTTP_HOST'];
 395                 } else {
 396                     $this->url = '';
 397                 }
 398         } else {
 399             // Strip any trailing slashes for consistency (relative
 400             // URLs may already start with a slash like "/file.html")
 401             if ( substr($url, -1) == '/' ) {
 402                 $url = substr($url, 0, -1);
 403             }
 404             $this->url = $url;
 405         }
 406     }
 407
 408     /**
 409      *  Workhorse function that does actual conversion.
 410      *
 411      *  First performs custom tag replacement specified by $search and
 412      *  $replace arrays. Then strips any remaining HTML tags, reduces whitespace
 413      *  and newlines to a readable format, and word wraps the text to
 414      *  $width characters.
 415      *
 416      *  @access private
 417      *  @return void
 418      */
 419     function _convert($ll=true)
 420     {
 421         // Variables used for building the link list
 422         $this->_link_count = 0;
 423         $this->_link_list = '';
 424
 425         $text = trim(stripslashes($this->html));
 426
 427         // Run our defined search-and-replace
 428         $text = preg_replace($this->search, $this->replace, $text);
 429
 430         // Strip any other HTML tags
 431         $text = strip_tags($text, $this->allowed_tags);
 432
 433         // Bring down number of empty lines to 2 max
 434         $text = preg_replace("/\n\s+\n/", "\n\n", $text);
 435         $text = preg_replace("/[\n]{3,}/", "\n\n", $text);
 436
 437         // Add link list
 438         if ( $ll && !empty($this->_link_list) ) {
 439             $text .= "\n\nLinks:\n------\n" . $this->_link_list;
 440         }
 441
 442         // Wrap the text to a readable format
 443         // for PHP versions >= 4.0.2. Default width is 75
 444         // If width is 0 or less, don't wrap the text.
 445         if ( $this->width > 0 ) {
 446                 $text = wordwrap($text, $this->width);
 447         }
 448
 449         $this->text = $text;
 450
 451         $this->_converted = true;
 452     }
 453
 454     /**
 455      *  Helper function called by preg_replace() on link replacement.
 456      *
 457      *  Maintains an internal list of links to be displayed at the end of the
 458      *  text, with numeric indices to the original point in the text they
 459      *  appeared. Also makes an effort at identifying and handling absolute
 460      *  and relative links.
 461      *
 462      *  @param string $link URL of the link
 463      *  @param string $display Part of the text to associate number with
 464      *  @access private
 465      *  @return string
 466      */
 467     function _build_link_list( $link, $display )
 468     {
 469                 if ( substr($link, 0, 7) == 'http://' || substr($link, 0, 8) == 'https://' ||
 470              substr($link, 0, 7) == 'mailto:' ) {
 471             $this->_link_count++;
 472             $this->_link_list .= "[" . $this->_link_count . "] $link\n";
 473             $additional = ' [' . $this->_link_count . ']';
 474                 } elseif ( substr($link, 0, 11) == 'javascript:' ) {
 475                         // Don't count the link; ignore it
 476                         $additional = '';
 477                 // what about href="#anchor" ?
 478         } else {
 479             $this->_link_count++;
 480             $this->_link_list .= "[" . $this->_link_count . "] " . $this->url;
 481             if ( substr($link, 0, 1) != '/' ) {
 482                 $this->_link_list .= '/';
 483             }
 484             $this->_link_list .= "$link\n";
 485             $additional = ' [' . $this->_link_count . ']';
 486         }
 487
 488         return $display . $additional;
 489     }
 490
 491 }
 492
 493 ?>