| 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 | |
| 148 | '/<script[^>]*>.*?<\/script>/i', // <script>s -- which strip_tags supposedly has problems with | |
| 8ac6fd | 149 | '/<style[^>]*>.*?<\/style>/i', // <style>s -- which strip_tags supposedly has problems with |
| 4e17e6 | 150 | '/<p[^>]*>/i', // <P> |
| T | 151 | '/<br[^>]*>/i', // <br> |
| 8ac6fd | 152 | '/<i[^>]*>(.*?)<\/i>/i', // <i> |
| A | 153 | '/<em[^>]*>(.*?)<\/em>/i', // <em> |
| 4e17e6 | 154 | '/(<ul[^>]*>|<\/ul>)/i', // <ul> and </ul> |
| T | 155 | '/(<ol[^>]*>|<\/ol>)/i', // <ol> and </ol> |
| 8ac6fd | 156 | '/<li[^>]*>(.*?)<\/li>/i', // <li> and </li> |
| 4e17e6 | 157 | '/<li[^>]*>/i', // <li> |
| T | 158 | '/<hr[^>]*>/i', // <hr> |
| f6b282 | 159 | '/<div[^>]*>/i', // <div> |
| 4e17e6 | 160 | '/(<table[^>]*>|<\/table>)/i', // <table> and </table> |
| T | 161 | '/(<tr[^>]*>|<\/tr>)/i', // <tr> and </tr> |
| 8ac6fd | 162 | '/<td[^>]*>(.*?)<\/td>/i', // <td> and </td> |
| 4e17e6 | 163 | ); |
| T | 164 | |
| 165 | /** | |
| 166 | * List of pattern replacements corresponding to patterns searched. | |
| 167 | * | |
| 168 | * @var array $replace | |
| 169 | * @access public | |
| 170 | * @see $search | |
| 171 | */ | |
| 172 | var $replace = array( | |
| 173 | '', // Non-legal carriage return | |
| 174 | ' ', // Newlines and tabs | |
| 175 | '', // <script>s -- which strip_tags supposedly has problems with | |
| 8ac6fd | 176 | '', // <style>s -- which strip_tags supposedly has problems with |
| f6b282 | 177 | "\n\n", // <P> |
| 4e17e6 | 178 | "\n", // <br> |
| T | 179 | '_\\1_', // <i> |
| 8ac6fd | 180 | '_\\1_', // <em> |
| 4e17e6 | 181 | "\n\n", // <ul> and </ul> |
| T | 182 | "\n\n", // <ol> and </ol> |
| 8ac6fd | 183 | "\t* \\1\n", // <li> and </li> |
| A | 184 | "\n\t* ", // <li> |
| 6972cc | 185 | "\n-------------------------\n", // <hr> |
| 11bcac | 186 | "<div>\n", // <div> |
| 6972cc | 187 | "\n\n", // <table> and </table> |
| 4e17e6 | 188 | "\n", // <tr> and </tr> |
| T | 189 | "\t\t\\1\n", // <td> and </td> |
| ca0cd0 | 190 | ); |
| A | 191 | |
| 192 | /** | |
| 193 | * List of preg* regular expression patterns to search for, | |
| 194 | * used in conjunction with $ent_replace. | |
| 195 | * | |
| 196 | * @var array $ent_search | |
| 197 | * @access public | |
| 198 | * @see $ent_replace | |
| 199 | */ | |
| 200 | var $ent_search = array( | |
| 201 | '/&(nbsp|#160);/i', // Non-breaking space | |
| 202 | '/&(quot|rdquo|ldquo|#8220|#8221|#147|#148);/i', | |
| 203 | // Double quotes | |
| 204 | '/&(apos|rsquo|lsquo|#8216|#8217);/i', // Single quotes | |
| 205 | '/>/i', // Greater-than | |
| 206 | '/</i', // Less-than | |
| 207 | '/&(copy|#169);/i', // Copyright | |
| 208 | '/&(trade|#8482|#153);/i', // Trademark | |
| 209 | '/&(reg|#174);/i', // Registered | |
| 210 | '/&(mdash|#151|#8212);/i', // mdash | |
| 211 | '/&(ndash|minus|#8211|#8722);/i', // ndash | |
| 212 | '/&(bull|#149|#8226);/i', // Bullet | |
| 213 | '/&(pound|#163);/i', // Pound sign | |
| 214 | '/&(euro|#8364);/i', // Euro sign | |
| 215 | '/&(amp|#38);/i', // Ampersand: see _converter() | |
| 216 | '/[ ]{2,}/', // Runs of spaces, post-handling | |
| 217 | ); | |
| 218 | ||
| 219 | /** | |
| 220 | * List of pattern replacements corresponding to patterns searched. | |
| 221 | * | |
| 222 | * @var array $ent_replace | |
| 223 | * @access public | |
| 224 | * @see $ent_search | |
| 225 | */ | |
| 226 | var $ent_replace = array( | |
| 8ac6fd | 227 | ' ', // Non-breaking space |
| A | 228 | '"', // Double quotes |
| 229 | "'", // Single quotes | |
| 4e17e6 | 230 | '>', |
| T | 231 | '<', |
| 232 | '(c)', | |
| 233 | '(tm)', | |
| 234 | '(R)', | |
| 8ac6fd | 235 | '--', |
| A | 236 | '-', |
| 4e17e6 | 237 | '*', |
| 300fc6 | 238 | '£', |
| 8ac6fd | 239 | 'EUR', // Euro sign. € ? |
| 6084d7 | 240 | '|+|amp|+|', // Ampersand: see _converter() |
| ca0cd0 | 241 | ' ', // Runs of spaces, post-handling |
| f50cc7 | 242 | ); |
| A | 243 | |
| 244 | /** | |
| 245 | * List of preg* regular expression patterns to search for | |
| 246 | * and replace using callback function. | |
| 247 | * | |
| 248 | * @var array $callback_search | |
| 249 | * @access public | |
| 250 | */ | |
| 251 | var $callback_search = array( | |
| f35995 | 252 | '/<(a) [^>]*href=("|\')([^"\']+)\2[^>]*>(.*?)<\/a>/i', |
| A | 253 | // <a href=""> |
| f50cc7 | 254 | '/<(h)[123456][^>]*>(.*?)<\/h[123456]>/i', // H1 - H3 |
| A | 255 | '/<(b)[^>]*>(.*?)<\/b>/i', // <b> |
| 256 | '/<(strong)[^>]*>(.*?)<\/strong>/i', // <strong> | |
| 257 | '/<(th)[^>]*>(.*?)<\/th>/i', // <th> and </th> | |
| 7353fa | 258 | ); |
| A | 259 | |
| 8ac6fd | 260 | /** |
| A | 261 | * List of preg* regular expression patterns to search for in PRE body, |
| 262 | * used in conjunction with $pre_replace. | |
| 263 | * | |
| 264 | * @var array $pre_search | |
| 265 | * @access public | |
| 266 | * @see $pre_replace | |
| 267 | */ | |
| 7789fd | 268 | var $pre_search = array( |
| 6972cc | 269 | "/\n/", |
| T | 270 | "/\t/", |
| 271 | '/ /', | |
| 272 | '/<pre[^>]*>/', | |
| 273 | '/<\/pre>/' | |
| 7353fa | 274 | ); |
| A | 275 | |
| 276 | /** | |
| 277 | * List of pattern replacements corresponding to patterns searched for PRE body. | |
| 278 | * | |
| 279 | * @var array $pre_replace | |
| 280 | * @access public | |
| 281 | * @see $pre_search | |
| 282 | */ | |
| 7789fd | 283 | var $pre_replace = array( |
| 6972cc | 284 | '<br>', |
| T | 285 | ' ', |
| 286 | ' ', | |
| 287 | '', | |
| 288 | '' | |
| 4e17e6 | 289 | ); |
| T | 290 | |
| 291 | /** | |
| 292 | * Contains a list of HTML tags to allow in the resulting text. | |
| 293 | * | |
| 294 | * @var string $allowed_tags | |
| 295 | * @access public | |
| 296 | * @see set_allowed_tags() | |
| 297 | */ | |
| 298 | var $allowed_tags = ''; | |
| 299 | ||
| 300 | /** | |
| 301 | * Contains the base URL that relative links should resolve to. | |
| 302 | * | |
| 303 | * @var string $url | |
| 304 | * @access public | |
| 305 | */ | |
| 306 | var $url; | |
| 307 | ||
| 308 | /** | |
| 309 | * Indicates whether content in the $html variable has been converted yet. | |
| 310 | * | |
| 8ac6fd | 311 | * @var boolean $_converted |
| 4e17e6 | 312 | * @access private |
| T | 313 | * @see $html, $text |
| 314 | */ | |
| 315 | var $_converted = false; | |
| 316 | ||
| 317 | /** | |
| 318 | * Contains URL addresses from links to be rendered in plain text. | |
| 319 | * | |
| 43c40f | 320 | * @var array $_link_list |
| 4e17e6 | 321 | * @access private |
| T | 322 | * @see _build_link_list() |
| 323 | */ | |
| 43c40f | 324 | var $_link_list = array(); |
| 4e17e6 | 325 | |
| ca0cd0 | 326 | /** |
| A | 327 | * Boolean flag, true if a table of link URLs should be listed after the text. |
| 328 | * | |
| 329 | * @var boolean $_do_links | |
| 330 | * @access private | |
| 331 | * @see html2text() | |
| e7f85b | 332 | */ |
| A | 333 | var $_do_links = true; |
| ca0cd0 | 334 | |
| 4e17e6 | 335 | /** |
| T | 336 | * Constructor. |
| 337 | * | |
| 338 | * If the HTML source string (or file) is supplied, the class | |
| 339 | * will instantiate with that source propagated, all that has | |
| 340 | * to be done it to call get_text(). | |
| 341 | * | |
| 342 | * @param string $source HTML content | |
| 343 | * @param boolean $from_file Indicates $source is a file to pull content from | |
| e7f85b | 344 | * @param boolean $do_links Indicate whether a table of link URLs is desired |
| A | 345 | * @param integer $width Maximum width of the formatted text, 0 for no limit |
| 4e17e6 | 346 | * @access public |
| T | 347 | * @return void |
| 348 | */ | |
| e7f85b | 349 | function html2text( $source = '', $from_file = false, $do_links = true, $width = 75 ) |
| 4e17e6 | 350 | { |
| T | 351 | if ( !empty($source) ) { |
| 352 | $this->set_html($source, $from_file); | |
| 353 | } | |
| 6972cc | 354 | |
| 4e17e6 | 355 | $this->set_base_url(); |
| 6972cc | 356 | $this->_do_links = $do_links; |
| T | 357 | $this->width = $width; |
| 4e17e6 | 358 | } |
| T | 359 | |
| 360 | /** | |
| 361 | * Loads source HTML into memory, either from $source string or a file. | |
| 362 | * | |
| 363 | * @param string $source HTML content | |
| 364 | * @param boolean $from_file Indicates $source is a file to pull content from | |
| 365 | * @access public | |
| 366 | * @return void | |
| 367 | */ | |
| 368 | function set_html( $source, $from_file = false ) | |
| 369 | { | |
| 370 | if ( $from_file && file_exists($source) ) { | |
| 6972cc | 371 | $this->html = file_get_contents($source); |
| 4e17e6 | 372 | } |
| 8ac6fd | 373 | else |
| 6972cc | 374 | $this->html = $source; |
| 4e17e6 | 375 | |
| T | 376 | $this->_converted = false; |
| 377 | } | |
| 378 | ||
| 379 | /** | |
| 380 | * Returns the text, converted from HTML. | |
| 381 | * | |
| 382 | * @access public | |
| 383 | * @return string | |
| 384 | */ | |
| 385 | function get_text() | |
| 386 | { | |
| 387 | if ( !$this->_converted ) { | |
| 388 | $this->_convert(); | |
| 389 | } | |
| 390 | ||
| 391 | return $this->text; | |
| 392 | } | |
| 393 | ||
| 394 | /** | |
| 395 | * Prints the text, converted from HTML. | |
| 396 | * | |
| 397 | * @access public | |
| 398 | * @return void | |
| 399 | */ | |
| 400 | function print_text() | |
| 401 | { | |
| 402 | print $this->get_text(); | |
| 403 | } | |
| 404 | ||
| 405 | /** | |
| 406 | * Alias to print_text(), operates identically. | |
| 407 | * | |
| 408 | * @access public | |
| 409 | * @return void | |
| 410 | * @see print_text() | |
| 411 | */ | |
| 412 | function p() | |
| 413 | { | |
| 414 | print $this->get_text(); | |
| 415 | } | |
| 416 | ||
| 417 | /** | |
| 418 | * Sets the allowed HTML tags to pass through to the resulting text. | |
| 419 | * | |
| 420 | * Tags should be in the form "<p>", with no corresponding closing tag. | |
| 421 | * | |
| 422 | * @access public | |
| 423 | * @return void | |
| 424 | */ | |
| 425 | function set_allowed_tags( $allowed_tags = '' ) | |
| 426 | { | |
| 427 | if ( !empty($allowed_tags) ) { | |
| 428 | $this->allowed_tags = $allowed_tags; | |
| 429 | } | |
| 430 | } | |
| 431 | ||
| 432 | /** | |
| 433 | * Sets a base URL to handle relative links. | |
| 434 | * | |
| 435 | * @access public | |
| 436 | * @return void | |
| 437 | */ | |
| 438 | function set_base_url( $url = '' ) | |
| 439 | { | |
| 440 | if ( empty($url) ) { | |
| 8ac6fd | 441 | if ( !empty($_SERVER['HTTP_HOST']) ) { |
| A | 442 | $this->url = 'http://' . $_SERVER['HTTP_HOST']; |
| 443 | } else { | |
| 444 | $this->url = ''; | |
| 445 | } | |
| 4e17e6 | 446 | } else { |
| T | 447 | // Strip any trailing slashes for consistency (relative |
| 448 | // URLs may already start with a slash like "/file.html") | |
| 449 | if ( substr($url, -1) == '/' ) { | |
| 450 | $url = substr($url, 0, -1); | |
| 451 | } | |
| 452 | $this->url = $url; | |
| 453 | } | |
| 454 | } | |
| 455 | ||
| 456 | /** | |
| 11bcac | 457 | * Workhorse function that does actual conversion (calls _converter() method). |
| 4e17e6 | 458 | * |
| T | 459 | * @access private |
| 460 | * @return void | |
| 461 | */ | |
| 462 | function _convert() | |
| 463 | { | |
| 464 | // Variables used for building the link list | |
| 43c40f | 465 | $this->_link_list = array(); |
| 4e17e6 | 466 | |
| T | 467 | $text = trim(stripslashes($this->html)); |
| 11bcac | 468 | |
| A | 469 | // Convert HTML to TXT |
| 470 | $this->_converter($text); | |
| 471 | ||
| 472 | // Add link list | |
| 43c40f | 473 | if (!empty($this->_link_list)) { |
| A | 474 | $text .= "\n\nLinks:\n------\n"; |
| 475 | foreach ($this->_link_list as $idx => $url) { | |
| 476 | $text .= '[' . ($idx+1) . '] ' . $url . "\n"; | |
| 477 | } | |
| 11bcac | 478 | } |
| A | 479 | |
| 480 | $this->text = $text; | |
| 481 | ||
| 482 | $this->_converted = true; | |
| 483 | } | |
| 484 | ||
| 485 | /** | |
| 486 | * Workhorse function that does actual conversion. | |
| 487 | * | |
| 488 | * First performs custom tag replacement specified by $search and | |
| 489 | * $replace arrays. Then strips any remaining HTML tags, reduces whitespace | |
| 490 | * and newlines to a readable format, and word wraps the text to | |
| 491 | * $width characters. | |
| 492 | * | |
| 493 | * @param string Reference to HTML content string | |
| 494 | * | |
| 495 | * @access private | |
| 496 | * @return void | |
| 497 | */ | |
| 498 | function _converter(&$text) | |
| 499 | { | |
| 500 | // Convert <BLOCKQUOTE> (before PRE!) | |
| 501 | $this->_convert_blockquotes($text); | |
| 4e17e6 | 502 | |
| 6972cc | 503 | // Convert <PRE> |
| 8ac6fd | 504 | $this->_convert_pre($text); |
| 300fc6 | 505 | |
| ca0cd0 | 506 | // Run our defined tags search-and-replace |
| 4e17e6 | 507 | $text = preg_replace($this->search, $this->replace, $text); |
| ca0cd0 | 508 | |
| A | 509 | // Run our defined tags search-and-replace with callback |
| 510 | $text = preg_replace_callback($this->callback_search, array('html2text', '_preg_callback'), $text); | |
| 511 | ||
| 512 | // Strip any other HTML tags | |
| 513 | $text = strip_tags($text, $this->allowed_tags); | |
| 514 | ||
| 515 | // Run our defined entities/characters search-and-replace | |
| 516 | $text = preg_replace($this->ent_search, $this->ent_replace, $text); | |
| 4e17e6 | 517 | |
| 6972cc | 518 | // Replace known html entities |
| T | 519 | $text = html_entity_decode($text, ENT_COMPAT, 'UTF-8'); |
| 4e0419 | 520 | |
| 755900 | 521 | // Remove unknown/unhandled entities (this cannot be done in search-and-replace block) |
| 6084d7 | 522 | $text = preg_replace('/&([a-zA-Z0-9]{2,6}|#[0-9]{2,4});/', '', $text); |
| A | 523 | |
| 524 | // Convert "|+|amp|+|" into "&", need to be done after handling of unknown entities | |
| 525 | // This properly handles situation of "&quot;" in input string | |
| 526 | $text = str_replace('|+|amp|+|', '&', $text); | |
| 755900 | 527 | |
| 4e17e6 | 528 | // Bring down number of empty lines to 2 max |
| 8ac6fd | 529 | $text = preg_replace("/\n\s+\n/", "\n\n", $text); |
| 4e17e6 | 530 | $text = preg_replace("/[\n]{3,}/", "\n\n", $text); |
| T | 531 | |
| 4d7fbd | 532 | // remove leading empty lines (can be produced by eg. P tag on the beginning) |
| ca0cd0 | 533 | $text = ltrim($text, "\n"); |
| 4d7fbd | 534 | |
| 4e17e6 | 535 | // Wrap the text to a readable format |
| T | 536 | // for PHP versions >= 4.0.2. Default width is 75 |
| 8ac6fd | 537 | // If width is 0 or less, don't wrap the text. |
| A | 538 | if ( $this->width > 0 ) { |
| 539 | $text = wordwrap($text, $this->width); | |
| 540 | } | |
| 4e17e6 | 541 | } |
| T | 542 | |
| 543 | /** | |
| 544 | * Helper function called by preg_replace() on link replacement. | |
| 545 | * | |
| 546 | * Maintains an internal list of links to be displayed at the end of the | |
| 547 | * text, with numeric indices to the original point in the text they | |
| 548 | * appeared. Also makes an effort at identifying and handling absolute | |
| 549 | * and relative links. | |
| 550 | * | |
| 551 | * @param string $link URL of the link | |
| 552 | * @param string $display Part of the text to associate number with | |
| 553 | * @access private | |
| 554 | * @return string | |
| 8ac6fd | 555 | */ |
| A | 556 | function _build_link_list( $link, $display ) |
| 557 | { | |
| 43c40f | 558 | if (!$this->_do_links || empty($link)) { |
| 11bcac | 559 | return $display; |
| 43c40f | 560 | } |
| 11bcac | 561 | |
| 43c40f | 562 | // Ignored link types |
| A | 563 | if (preg_match('!^(javascript|mailto|#):!i', $link)) { |
| 564 | return $display; | |
| 4e17e6 | 565 | } |
| T | 566 | |
| 43c40f | 567 | if (preg_match('!^(https?://)!i', $link)) { |
| A | 568 | $url = $link; |
| 569 | } | |
| 570 | else { | |
| 571 | $url = $this->url; | |
| 572 | if (substr($link, 0, 1) != '/') { | |
| 573 | $url .= '/'; | |
| 574 | } | |
| 575 | $url .= "$link"; | |
| 576 | } | |
| 577 | ||
| 578 | if (($index = array_search($url, $this->_link_list)) === false) { | |
| 579 | $this->_link_list[] = $url; | |
| 580 | $index = count($this->_link_list); | |
| 581 | } | |
| 582 | ||
| 583 | return $display . ' [' . ($index+1) . ']'; | |
| 8ac6fd | 584 | } |
| 11bcac | 585 | |
| 7353fa | 586 | /** |
| A | 587 | * Helper function for PRE body conversion. |
| 588 | * | |
| 589 | * @param string HTML content | |
| 590 | * @access private | |
| 8ac6fd | 591 | */ |
| 7353fa | 592 | function _convert_pre(&$text) |
| 8ac6fd | 593 | { |
| d483cd | 594 | // get the content of PRE element |
| 11bcac | 595 | while (preg_match('/<pre[^>]*>(.*)<\/pre>/ismU', $text, $matches)) { |
| d483cd | 596 | // convert the content |
| A | 597 | $this->pre_content = sprintf('<div><br>%s<br></div>', |
| 598 | preg_replace($this->pre_search, $this->pre_replace, $matches[1])); | |
| 599 | // replace the content (use callback because content can contain $0 variable) | |
| 600 | $text = preg_replace_callback('/<pre[^>]*>.*<\/pre>/ismU', | |
| 601 | array('html2text', '_preg_pre_callback'), $text, 1); | |
| 602 | // free memory | |
| 603 | $this->pre_content = ''; | |
| 11bcac | 604 | } |
| A | 605 | } |
| 606 | ||
| 607 | /** | |
| 608 | * Helper function for BLOCKQUOTE body conversion. | |
| 609 | * | |
| 610 | * @param string HTML content | |
| 611 | * @access private | |
| 612 | */ | |
| 613 | function _convert_blockquotes(&$text) | |
| 614 | { | |
| 615 | if (preg_match_all('/<\/*blockquote[^>]*>/i', $text, $matches, PREG_OFFSET_CAPTURE)) { | |
| 616 | $level = 0; | |
| 617 | $diff = 0; | |
| 618 | foreach ($matches[0] as $m) { | |
| 619 | if ($m[0][0] == '<' && $m[0][1] == '/') { | |
| 620 | $level--; | |
| 621 | if ($level < 0) { | |
| 622 | $level = 0; // malformed HTML: go to next blockquote | |
| 623 | } | |
| 624 | else if ($level > 0) { | |
| 625 | // skip inner blockquote | |
| 626 | } | |
| 627 | else { | |
| 628 | $end = $m[1]; | |
| 629 | $len = $end - $taglen - $start; | |
| 630 | // Get blockquote content | |
| 631 | $body = substr($text, $start + $taglen - $diff, $len); | |
| 632 | ||
| 633 | // Set text width | |
| 634 | $p_width = $this->width; | |
| 635 | if ($this->width > 0) $this->width -= 2; | |
| 636 | // Convert blockquote content | |
| 637 | $body = trim($body); | |
| 638 | $this->_converter($body); | |
| 639 | // Add citation markers and create PRE block | |
| 640 | $body = preg_replace('/((^|\n)>*)/', '\\1> ', trim($body)); | |
| 641 | $body = '<pre>' . htmlspecialchars($body) . '</pre>'; | |
| 642 | // Re-set text width | |
| 643 | $this->width = $p_width; | |
| 644 | // Replace content | |
| 645 | $text = substr($text, 0, $start - $diff) | |
| 646 | . $body . substr($text, $end + strlen($m[0]) - $diff); | |
| 647 | ||
| 648 | $diff = $len + $taglen + strlen($m[0]) - strlen($body); | |
| 649 | unset($body); | |
| 650 | } | |
| 651 | } | |
| 652 | else { | |
| 653 | if ($level == 0) { | |
| 654 | $start = $m[1]; | |
| 655 | $taglen = strlen($m[0]); | |
| 656 | } | |
| 657 | $level ++; | |
| 658 | } | |
| 659 | } | |
| 6972cc | 660 | } |
| 8ac6fd | 661 | } |
| f50cc7 | 662 | |
| A | 663 | /** |
| 664 | * Callback function for preg_replace_callback use. | |
| 665 | * | |
| 666 | * @param array PREG matches | |
| 667 | * @return string | |
| 668 | */ | |
| d483cd | 669 | private function _preg_callback($matches) |
| f50cc7 | 670 | { |
| 6972cc | 671 | switch($matches[1]) { |
| T | 672 | case 'b': |
| 673 | case 'strong': | |
| f35995 | 674 | return $this->_toupper($matches[2]); |
| da1722 | 675 | case 'th': |
| f35995 | 676 | return $this->_toupper("\t\t". $matches[2] ."\n"); |
| 6972cc | 677 | case 'h': |
| f35995 | 678 | return $this->_toupper("\n\n". $matches[2] ."\n\n"); |
| 6972cc | 679 | case 'a': |
| 29c542 | 680 | // Remove spaces in URL (#1487805) |
| A | 681 | $url = str_replace(' ', '', $matches[3]); |
| 682 | return $this->_build_link_list($url, $matches[4]); | |
| 6972cc | 683 | } |
| f50cc7 | 684 | } |
| 29c542 | 685 | |
| f50cc7 | 686 | /** |
| d483cd | 687 | * Callback function for preg_replace_callback use in PRE content handler. |
| A | 688 | * |
| 689 | * @param array PREG matches | |
| 690 | * @return string | |
| 691 | */ | |
| 692 | private function _preg_pre_callback($matches) | |
| 693 | { | |
| 694 | return $this->pre_content; | |
| 695 | } | |
| 696 | ||
| 697 | /** | |
| f35995 | 698 | * Strtoupper function with HTML tags and entities handling. |
| f50cc7 | 699 | * |
| f35995 | 700 | * @param string $str Text to convert |
| A | 701 | * @return string Converted text |
| 702 | */ | |
| 703 | private function _toupper($str) | |
| 704 | { | |
| 705 | // string can containg HTML tags | |
| 706 | $chunks = preg_split('/(<[^>]*>)/', $str, null, PREG_SPLIT_NO_EMPTY | PREG_SPLIT_DELIM_CAPTURE); | |
| 707 | ||
| 708 | // convert toupper only the text between HTML tags | |
| 709 | foreach ($chunks as $idx => $chunk) { | |
| 710 | if ($chunk[0] != '<') { | |
| 711 | $chunks[$idx] = $this->_strtoupper($chunk); | |
| 712 | } | |
| 713 | } | |
| 714 | ||
| 715 | return implode($chunks); | |
| 716 | } | |
| 717 | ||
| 718 | /** | |
| 719 | * Strtoupper multibyte wrapper function with HTML entities handling. | |
| 720 | * | |
| 721 | * @param string $str Text to convert | |
| 722 | * @return string Converted text | |
| f50cc7 | 723 | */ |
| d483cd | 724 | private function _strtoupper($str) |
| f50cc7 | 725 | { |
| 67e592 | 726 | $str = html_entity_decode($str, ENT_COMPAT, RCMAIL_CHARSET); |
| A | 727 | |
| 6972cc | 728 | if (function_exists('mb_strtoupper')) |
| 67e592 | 729 | $str = mb_strtoupper($str); |
| 6972cc | 730 | else |
| 67e592 | 731 | $str = strtoupper($str); |
| A | 732 | |
| 733 | $str = htmlspecialchars($str, ENT_COMPAT, RCMAIL_CHARSET); | |
| 734 | ||
| 735 | return $str; | |
| f50cc7 | 736 | } |
| 4e17e6 | 737 | } |
