403 Forbidden

Previously run action hooks

Hooks that have been called previously.

"; 580 | print_r(self::$run_actions); 581 | echo "

12 | # 13 | 14 | 15 | define( 'MARKDOWN_VERSION', "1.0.1n" ); # Sat 10 Oct 2009 16 | define( 'MARKDOWNEXTRA_VERSION', "1.2.4" ); # Sat 10 Oct 2009 17 | 18 | 19 | # 20 | # Global default settings: 21 | # 22 | 23 | # Change to ">" for HTML output 24 | @define( 'MARKDOWN_EMPTY_ELEMENT_SUFFIX', " />"); 25 | 26 | # Define the width of a tab for code blocks. 27 | @define( 'MARKDOWN_TAB_WIDTH', 4 ); 28 | 29 | # Optional title attribute for footnote links and backlinks. 30 | @define( 'MARKDOWN_FN_LINK_TITLE', "" ); 31 | @define( 'MARKDOWN_FN_BACKLINK_TITLE', "" ); 32 | 33 | # Optional class attribute for footnote links and backlinks. 34 | @define( 'MARKDOWN_FN_LINK_CLASS', "" ); 35 | @define( 'MARKDOWN_FN_BACKLINK_CLASS', "" ); 36 | 37 | 38 | # 39 | # WordPress settings: 40 | # 41 | 42 | # Change to false to remove Markdown from posts and/or comments. 43 | @define( 'MARKDOWN_WP_POSTS', true ); 44 | @define( 'MARKDOWN_WP_COMMENTS', true ); 45 | 46 | 47 | 48 | ### Standard Function Interface ### 49 | 50 | @define( 'MARKDOWN_PARSER_CLASS', 'MarkdownExtra_Parser' ); 51 | 52 | function Markdown($text) { 53 | # 54 | # Initialize the parser and return the result of its transform method. 55 | # 56 | # Setup static parser variable. 57 | static $parser; 58 | if (!isset($parser)) { 59 | $parser_class = MARKDOWN_PARSER_CLASS; 60 | $parser = new $parser_class; 61 | } 62 | 63 | # Transform text using parser. 64 | return $parser->transform($text); 65 | } 66 | 67 | ### Smarty Modifier Interface ### 68 | 69 | function smarty_modifier_markdown($text) { 70 | return Markdown($text); 71 | } 72 | 73 | # 74 | # Markdown Parser Class 75 | # 76 | 77 | class Markdown_Parser { 78 | 79 | # Regex to match balanced [brackets]. 80 | # Needed to insert a maximum bracked depth while converting to PHP. 81 | var $nested_brackets_depth = 6; 82 | var $nested_brackets_re; 83 | 84 | var $nested_url_parenthesis_depth = 4; 85 | var $nested_url_parenthesis_re; 86 | 87 | # Table of hash values for escaped characters: 88 | var $escape_chars = '\`*_{}[]()>#+-.!'; 89 | var $escape_chars_re; 90 | 91 | # Change to ">" for HTML output. 92 | var $empty_element_suffix = MARKDOWN_EMPTY_ELEMENT_SUFFIX; 93 | var $tab_width = MARKDOWN_TAB_WIDTH; 94 | 95 | # Change to `true` to disallow markup or entities. 96 | var $no_markup = false; 97 | var $no_entities = false; 98 | 99 | # Predefined urls and titles for reference links and images. 100 | var $predef_urls = array(); 101 | var $predef_titles = array(); 102 | 103 | 104 | function Markdown_Parser() { 105 | # 106 | # Constructor function. Initialize appropriate member variables. 107 | # 108 | $this->_initDetab(); 109 | $this->prepareItalicsAndBold(); 110 | 111 | $this->nested_brackets_re = 112 | str_repeat('(?>[^\[\]]+|\[', $this->nested_brackets_depth). 113 | str_repeat('\])*', $this->nested_brackets_depth); 114 | 115 | $this->nested_url_parenthesis_re = 116 | str_repeat('(?>[^()\s]+|$', $this->nested_url_parenthesis_depth). 117 | str_repeat('(?>$))*', $this->nested_url_parenthesis_depth); 118 | 119 | $this->escape_chars_re = '['.preg_quote($this->escape_chars).']'; 120 | 121 | # Sort document, block, and span gamut in ascendent priority order. 122 | asort($this->document_gamut); 123 | asort($this->block_gamut); 124 | asort($this->span_gamut); 125 | } 126 | 127 | 128 | # Internal hashes used during transformation. 129 | var $urls = array(); 130 | var $titles = array(); 131 | var $html_hashes = array(); 132 | 133 | # Status flag to avoid invalid nesting. 134 | var $in_anchor = false; 135 | 136 | 137 | function setup() { 138 | # 139 | # Called before the transformation process starts to setup parser 140 | # states. 141 | # 142 | # Clear global hashes. 143 | $this->urls = $this->predef_urls; 144 | $this->titles = $this->predef_titles; 145 | $this->html_hashes = array(); 146 | 147 | $in_anchor = false; 148 | } 149 | 150 | function teardown() { 151 | # 152 | # Called after the transformation process to clear any variable 153 | # which may be taking up memory unnecessarly. 154 | # 155 | $this->urls = array(); 156 | $this->titles = array(); 157 | $this->html_hashes = array(); 158 | } 159 | 160 | 161 | function transform($text) { 162 | # 163 | # Main function. Performs some preprocessing on the input text 164 | # and pass it through the document gamut. 165 | # 166 | $this->setup(); 167 | 168 | # Remove UTF-8 BOM and marker character in input, if present. 169 | $text = preg_replace('{^\xEF\xBB\xBF|\x1A}', '', $text); 170 | 171 | # Standardize line endings: 172 | # DOS to Unix and Mac to Unix 173 | $text = preg_replace('{\r\n?}', "\n", $text); 174 | 175 | # Make sure $text ends with a couple of newlines: 176 | $text .= "\n\n"; 177 | 178 | # Convert all tabs to spaces. 179 | $text = $this->detab($text); 180 | 181 | # Turn block-level HTML blocks into hash entries 182 | $text = $this->hashHTMLBlocks($text); 183 | 184 | # Strip any lines consisting only of spaces and tabs. 185 | # This makes subsequent regexen easier to write, because we can 186 | # match consecutive blank lines with /\n+/ instead of something 187 | # contorted like /[ ]*\n+/ . 188 | $text = preg_replace('/^[ ]+$/m', '', $text); 189 | 190 | # Run document gamut methods. 191 | foreach ($this->document_gamut as $method => $priority) { 192 | $text = $this->$method($text); 193 | } 194 | 195 | $this->teardown(); 196 | 197 | return $text . "\n"; 198 | } 199 | 200 | var $document_gamut = array( 201 | # Strip link definitions, store in hashes. 202 | "stripLinkDefinitions" => 20, 203 | 204 | "runBasicBlockGamut" => 30, 205 | ); 206 | 207 | 208 | function stripLinkDefinitions($text) { 209 | # 210 | # Strips link definitions from text, stores the URLs and titles in 211 | # hash references. 212 | # 213 | $less_than_tab = $this->tab_width - 1; 214 | 215 | # Link defs are in the form: ^[id]: url "optional title" 216 | $text = preg_replace_callback('{ 217 | ^[ ]{0,'.$less_than_tab.'}\[(.+)\][ ]?: # id = $1 218 | [ ]* 219 | \n? # maybe *one* newline 220 | [ ]* 221 | (?: 222 | <(.+?)> # url = $2 223 | | 224 | (\S+?) # url = $3 225 | ) 226 | [ ]* 227 | \n? # maybe one newline 228 | [ ]* 229 | (?: 230 | (?<=\s) # lookbehind for whitespace 231 | ["(] 232 | (.*?) # title = $4 233 | [")] 234 | [ ]* 235 | )? # title is optional 236 | (?:\n+|\Z) 237 | }xm', 238 | array(&$this, '_stripLinkDefinitions_callback'), 239 | $text); 240 | return $text; 241 | } 242 | function _stripLinkDefinitions_callback($matches) { 243 | $link_id = strtolower($matches[1]); 244 | $url = $matches[2] == '' ? $matches[3] : $matches[2]; 245 | $this->urls[$link_id] = $url; 246 | $this->titles[$link_id] =& $matches[4]; 247 | return ''; # String that will replace the block 248 | } 249 | 250 | 251 | function hashHTMLBlocks($text) { 252 | if ($this->no_markup) return $text; 253 | 254 | $less_than_tab = $this->tab_width - 1; 255 | 256 | # Hashify HTML blocks: 257 | # We only want to do this for block-level HTML tags, such as headers, 258 | # lists, and tables. That's because we still want to wrap

s around 259 | # "paragraphs" that are wrapped in non-block-level tags, such as anchors, 260 | # phrase emphasis, and spans. The list of tags we're looking for is 261 | # hard-coded: 262 | # 263 | # * List "a" is made of tags which can be both inline or block-level. 264 | # These will be treated block-level when the start tag is alone on 265 | # its line, otherwise they're not matched here and will be taken as 266 | # inline later. 267 | # * List "b" is made of tags which are always block-level; 268 | # 269 | $block_tags_a_re = 'ins|del'; 270 | $block_tags_b_re = 'p|div|h[1-6]|blockquote|pre|table|dl|ol|ul|address|'. 271 | 'script|noscript|form|fieldset|iframe|math'; 272 | 273 | # Regular expression for the content of a block tag. 274 | $nested_tags_level = 4; 275 | $attr = ' 276 | (?> # optional tag attributes 277 | \s # starts with whitespace 278 | (?> 279 | [^>"/]+ # text outside quotes 280 | | 281 | /+(?!>) # slash not followed by ">" 282 | | 283 | "[^"]*" # text inside double quotes (tolerate ">") 284 | | 285 | \'[^\']*\' # text inside single quotes (tolerate ">") 286 | )* 287 | )? 288 | '; 289 | $content = 290 | str_repeat(' 291 | (?> 292 | [^<]+ # content without tag 293 | | 294 | <\2 # nested opening tag 295 | '.$attr.' # attributes 296 | (?> 297 | /> 298 | | 299 | >', $nested_tags_level). # end of opening tag 300 | '.*?'. # last level nested tag content 301 | str_repeat(' 302 | # closing nested tag 303 | ) 304 | | 305 | <(?!/\2\s*> # other tags with a different name 306 | ) 307 | )*', 308 | $nested_tags_level); 309 | $content2 = str_replace('\2', '\3', $content); 310 | 311 | # First, look for nested blocks, e.g.: 312 | #

313 | #

314 | # tags for inner block must be indented. 315 | #

316 | #

` and stop at the first `

`. 322 | $text = preg_replace_callback('{(?> 323 | (?> 324 | (?<=\n\n) # Starting after a blank line 325 | | # or 326 | \A\n? # the beginning of the doc 327 | ) 328 | ( # save in $1 329 | 330 | # Match from `\n` to `\n`, handling nested tags 331 | # in between. 332 | 333 | [ ]{0,'.$less_than_tab.'} 334 | <('.$block_tags_b_re.')# start tag = $2 335 | '.$attr.'> # attributes followed by > and \n 336 | '.$content.' # content, support nesting 337 | # the matching end tag 338 | [ ]* # trailing spaces/tabs 339 | (?=\n+|\Z) # followed by a newline or end of document 340 | 341 | | # Special version for tags of group a. 342 | 343 | [ ]{0,'.$less_than_tab.'} 344 | <('.$block_tags_a_re.')# start tag = $3 345 | '.$attr.'>[ ]*\n # attributes followed by > 346 | '.$content2.' # content, support nesting 347 | # the matching end tag 348 | [ ]* # trailing spaces/tabs 349 | (?=\n+|\Z) # followed by a newline or end of document 350 | 351 | | # Special case just for

. It was easier to make a special 352 | # case than to make the other regex more complicated. 353 | 354 | [ ]{0,'.$less_than_tab.'} 355 | <(hr) # start tag = $2 356 | '.$attr.' # attributes 357 | /?> # the matching end tag 358 | [ ]* 359 | (?=\n{2,}|\Z) # followed by a blank line or end of document 360 | 361 | | # Special case for standalone HTML comments: 362 | 363 | [ ]{0,'.$less_than_tab.'} 364 | (?s: 365 | 366 | ) 367 | [ ]* 368 | (?=\n{2,}|\Z) # followed by a blank line or end of document 369 | 370 | | # PHP and ASP-style processor instructions ( 377 | ) 378 | [ ]* 379 | (?=\n{2,}|\Z) # followed by a blank line or end of document 380 | 381 | ) 382 | )}Sxmi', 383 | array(&$this, '_hashHTMLBlocks_callback'), 384 | $text); 385 | 386 | return $text; 387 | } 388 | function _hashHTMLBlocks_callback($matches) { 389 | $text = $matches[1]; 390 | $key = $this->hashBlock($text); 391 | return "\n\n$key\n\n"; 392 | } 393 | 394 | 395 | function hashPart($text, $boundary = 'X') { 396 | # 397 | # Called whenever a tag must be hashed when a function insert an atomic 398 | # element in the text stream. Passing $text to through this function gives 399 | # a unique text-token which will be reverted back when calling unhash. 400 | # 401 | # The $boundary argument specify what character should be used to surround 402 | # the token. By convension, "B" is used for block elements that needs not 403 | # to be wrapped into paragraph tags at the end, ":" is used for elements 404 | # that are word separators and "X" is used in the general case. 405 | # 406 | # Swap back any tag hash found in $text so we do not have to `unhash` 407 | # multiple times at the end. 408 | $text = $this->unhash($text); 409 | 410 | # Then hash the block. 411 | static $i = 0; 412 | $key = "$boundary\x1A" . ++$i . $boundary; 413 | $this->html_hashes[$key] = $text; 414 | return $key; # String that will replace the tag. 415 | } 416 | 417 | 418 | function hashBlock($text) { 419 | # 420 | # Shortcut function for hashPart with block-level boundaries. 421 | # 422 | return $this->hashPart($text, 'B'); 423 | } 424 | 425 | 426 | var $block_gamut = array( 427 | # 428 | # These are all the transformations that form block-level 429 | # tags like paragraphs, headers, and list items. 430 | # 431 | "doHeaders" => 10, 432 | "doHorizontalRules" => 20, 433 | 434 | "doLists" => 40, 435 | "doCodeBlocks" => 50, 436 | "doBlockQuotes" => 60, 437 | ); 438 | 439 | function runBlockGamut($text) { 440 | # 441 | # Run block gamut tranformations. 442 | # 443 | # We need to escape raw HTML in Markdown source before doing anything 444 | # else. This need to be done for each block, and not only at the 445 | # begining in the Markdown function since hashed blocks can be part of 446 | # list items and could have been indented. Indented blocks would have 447 | # been seen as a code block in a previous pass of hashHTMLBlocks. 448 | $text = $this->hashHTMLBlocks($text); 449 | 450 | return $this->runBasicBlockGamut($text); 451 | } 452 | 453 | function runBasicBlockGamut($text) { 454 | # 455 | # Run block gamut tranformations, without hashing HTML blocks. This is 456 | # useful when HTML blocks are known to be already hashed, like in the first 457 | # whole-document pass. 458 | # 459 | foreach ($this->block_gamut as $method => $priority) { 460 | $text = $this->$method($text); 461 | } 462 | 463 | # Finally form paragraph and restore hashed blocks. 464 | $text = $this->formParagraphs($text); 465 | 466 | return $text; 467 | } 468 | 469 | 470 | function doHorizontalRules($text) { 471 | # Do Horizontal Rules: 472 | return preg_replace( 473 | '{ 474 | ^[ ]{0,3} # Leading space 475 | ([-*_]) # $1: First marker 476 | (?> # Repeated marker group 477 | [ ]{0,2} # Zero, one, or two spaces. 478 | \1 # Marker character 479 | ){2,} # Group repeated at least twice 480 | [ ]* # Tailing spaces 481 | $ # End of line. 482 | }mx', 483 | "\n".$this->hashBlock("empty_element_suffix")."\n", 484 | $text); 485 | } 486 | 487 | 488 | var $span_gamut = array( 489 | # 490 | # These are all the transformations that occur *within* block-level 491 | # tags like paragraphs, headers, and list items. 492 | # 493 | # Process character escapes, code spans, and inline HTML 494 | # in one shot. 495 | "parseSpan" => -30, 496 | 497 | # Process anchor and image tags. Images must come first, 498 | # because ![foo][f] looks like an anchor. 499 | "doImages" => 10, 500 | "doAnchors" => 20, 501 | 502 | # Make links out of things like `` 503 | # Must come after doAnchors, because you can use < and > 504 | # delimiters in inline links like [this](). 505 | "doAutoLinks" => 30, 506 | "encodeAmpsAndAngles" => 40, 507 | 508 | "doItalicsAndBold" => 50, 509 | "doHardBreaks" => 60, 510 | ); 511 | 512 | function runSpanGamut($text) { 513 | # 514 | # Run span gamut tranformations. 515 | # 516 | foreach ($this->span_gamut as $method => $priority) { 517 | $text = $this->$method($text); 518 | } 519 | 520 | return $text; 521 | } 522 | 523 | 524 | function doHardBreaks($text) { 525 | # Do hard breaks: 526 | return preg_replace_callback('/ {2,}\n/', 527 | array(&$this, '_doHardBreaks_callback'), $text); 528 | } 529 | function _doHardBreaks_callback($matches) { 530 | return $this->hashPart("empty_element_suffix\n"); 531 | } 532 | 533 | 534 | function doAnchors($text) { 535 | # 536 | # Turn Markdown link shortcuts into XHTML tags. 537 | # 538 | if ($this->in_anchor) return $text; 539 | $this->in_anchor = true; 540 | 541 | # 542 | # First, handle reference-style links: [link text] [id] 543 | # 544 | $text = preg_replace_callback('{ 545 | ( # wrap whole match in $1 546 | \[ 547 | ('.$this->nested_brackets_re.') # link text = $2 548 | \] 549 | 550 | [ ]? # one optional space 551 | (?:\n[ ]*)? # one optional newline followed by spaces 552 | 553 | \[ 554 | (.*?) # id = $3 555 | \] 556 | ) 557 | }xs', 558 | array(&$this, '_doAnchors_reference_callback'), $text); 559 | 560 | # 561 | # Next, inline-style links: [link text](url "optional title") 562 | # 563 | $text = preg_replace_callback('{ 564 | ( # wrap whole match in $1 565 | \[ 566 | ('.$this->nested_brackets_re.') # link text = $2 567 | \] 568 | $ # literal paren 569 | [ \n]* 570 | (?: 571 | <(.+?)> # href = $3 572 | | 573 | ('.$this->nested_url_parenthesis_re.') # href = $4 574 | ) 575 | [ \n]* 576 | ( # $5 577 | ([\'"]) # quote char = $6 578 | (.*?) # Title = $7 579 | \6 # matching quote 580 | [ \n]* # ignore any spaces/tabs between closing quote and ) 581 | )? # title is optional 582 | $ 583 | ) 584 | }xs', 585 | array(&$this, '_doAnchors_inline_callback'), $text); 586 | 587 | # 588 | # Last, handle reference-style shortcuts: [link text] 589 | # These must come last in case you've also got [link text][1] 590 | # or [link text](/foo) 591 | # 592 | $text = preg_replace_callback('{ 593 | ( # wrap whole match in $1 594 | \[ 595 | ([^\[\]]+) # link text = $2; can\'t contain [ or ] 596 | \] 597 | ) 598 | }xs', 599 | array(&$this, '_doAnchors_reference_callback'), $text); 600 | 601 | $this->in_anchor = false; 602 | return $text; 603 | } 604 | function _doAnchors_reference_callback($matches) { 605 | $whole_match = $matches[1]; 606 | $link_text = $matches[2]; 607 | $link_id =& $matches[3]; 608 | 609 | if ($link_id == "") { 610 | # for shortcut links like [this][] or [this]. 611 | $link_id = $link_text; 612 | } 613 | 614 | # lower-case and turn embedded newlines into spaces 615 | $link_id = strtolower($link_id); 616 | $link_id = preg_replace('{[ ]?\n}', ' ', $link_id); 617 | 618 | if (isset($this->urls[$link_id])) { 619 | $url = $this->urls[$link_id]; 620 | $url = $this->encodeAttribute($url); 621 | 622 | $result = "titles[$link_id] ) ) { 624 | $title = $this->titles[$link_id]; 625 | $title = $this->encodeAttribute($title); 626 | $result .= " title=\"$title\""; 627 | } 628 | 629 | $link_text = $this->runSpanGamut($link_text); 630 | $result .= ">$link_text"; 631 | $result = $this->hashPart($result); 632 | } 633 | else { 634 | $result = $whole_match; 635 | } 636 | return $result; 637 | } 638 | function _doAnchors_inline_callback($matches) { 639 | $whole_match = $matches[1]; 640 | $link_text = $this->runSpanGamut($matches[2]); 641 | $url = $matches[3] == '' ? $matches[4] : $matches[3]; 642 | $title =& $matches[7]; 643 | 644 | $url = $this->encodeAttribute($url); 645 | 646 | $result = "encodeAttribute($title); 649 | $result .= " title=\"$title\""; 650 | } 651 | 652 | $link_text = $this->runSpanGamut($link_text); 653 | $result .= ">$link_text"; 654 | 655 | return $this->hashPart($result); 656 | } 657 | 658 | 659 | function doImages($text) { 660 | # 661 | # Turn Markdown image shortcuts into tags. 662 | # 663 | # 664 | # First, handle reference-style labeled images: ![alt text][id] 665 | # 666 | $text = preg_replace_callback('{ 667 | ( # wrap whole match in $1 668 | !\[ 669 | ('.$this->nested_brackets_re.') # alt text = $2 670 | \] 671 | 672 | [ ]? # one optional space 673 | (?:\n[ ]*)? # one optional newline followed by spaces 674 | 675 | \[ 676 | (.*?) # id = $3 677 | \] 678 | 679 | ) 680 | }xs', 681 | array(&$this, '_doImages_reference_callback'), $text); 682 | 683 | # 684 | # Next, handle inline images: ![alt text](url "optional title") 685 | # Don't forget: encode * and _ 686 | # 687 | $text = preg_replace_callback('{ 688 | ( # wrap whole match in $1 689 | !\[ 690 | ('.$this->nested_brackets_re.') # alt text = $2 691 | \] 692 | \s? # One optional whitespace character 693 | $ # literal paren 694 | [ \n]* 695 | (?: 696 | <(\S*)> # src url = $3 697 | | 698 | ('.$this->nested_url_parenthesis_re.') # src url = $4 699 | ) 700 | [ \n]* 701 | ( # $5 702 | ([\'"]) # quote char = $6 703 | (.*?) # title = $7 704 | \6 # matching quote 705 | [ \n]* 706 | )? # title is optional 707 | $ 708 | ) 709 | }xs', 710 | array(&$this, '_doImages_inline_callback'), $text); 711 | 712 | return $text; 713 | } 714 | function _doImages_reference_callback($matches) { 715 | $whole_match = $matches[1]; 716 | $alt_text = $matches[2]; 717 | $link_id = strtolower($matches[3]); 718 | 719 | if ($link_id == "") { 720 | $link_id = strtolower($alt_text); # for shortcut links like ![this][]. 721 | } 722 | 723 | $alt_text = $this->encodeAttribute($alt_text); 724 | if (isset($this->urls[$link_id])) { 725 | $url = $this->encodeAttribute($this->urls[$link_id]); 726 | $result = " $\"$alt_text\"";$ titles[$link_id])) { 728 | $title = $this->titles[$link_id]; 729 | $title = $this->encodeAttribute($title); 730 | $result .= " title=\"$title\""; 731 | } 732 | $result .= $this->empty_element_suffix; 733 | $result = $this->hashPart($result); 734 | } 735 | else { 736 | # If there's no such link ID, leave intact: 737 | $result = $whole_match; 738 | } 739 | 740 | return $result; 741 | } 742 | function _doImages_inline_callback($matches) { 743 | $whole_match = $matches[1]; 744 | $alt_text = $matches[2]; 745 | $url = $matches[3] == '' ? $matches[4] : $matches[3]; 746 | $title =& $matches[7]; 747 | 748 | $alt_text = $this->encodeAttribute($alt_text); 749 | $url = $this->encodeAttribute($url); 750 | $result = " $\"$alt_text\"";$ encodeAttribute($title); 753 | $result .= " title=\"$title\""; # $title already quoted 754 | } 755 | $result .= $this->empty_element_suffix; 756 | 757 | return $this->hashPart($result); 758 | } 759 | 760 | 761 | function doHeaders($text) { 762 | # Setext-style headers: 763 | # Header 1 764 | # ======== 765 | # 766 | # Header 2 767 | # -------- 768 | # 769 | $text = preg_replace_callback('{ ^(.+?)[ ]*\n(=+|-+)[ ]*\n+ }mx', 770 | array(&$this, '_doHeaders_callback_setext'), $text); 771 | 772 | # atx-style headers: 773 | # # Header 1 774 | # ## Header 2 775 | # ## Header 2 with closing hashes ## 776 | # ... 777 | # ###### Header 6 778 | # 779 | $text = preg_replace_callback('{ 780 | ^(\#{1,6}) # $1 = string of #\'s 781 | [ ]* 782 | (.+?) # $2 = Header text 783 | [ ]* 784 | \#* # optional closing #\'s (not counted) 785 | \n+ 786 | }xm', 787 | array(&$this, '_doHeaders_callback_atx'), $text); 788 | 789 | return $text; 790 | } 791 | function _doHeaders_callback_setext($matches) { 792 | # Terrible hack to check we haven't found an empty list item. 793 | if ($matches[2] == '-' && preg_match('{^-(?: |$)}', $matches[1])) 794 | return $matches[0]; 795 | 796 | $level = $matches[2]{0} == '=' ? 1 : 2; 797 | $block = "".$this->runSpanGamut($matches[1]).""; 798 | return "\n" . $this->hashBlock($block) . "\n\n"; 799 | } 800 | function _doHeaders_callback_atx($matches) { 801 | $level = strlen($matches[1]); 802 | $block = "".$this->runSpanGamut($matches[2]).""; 803 | return "\n" . $this->hashBlock($block) . "\n\n"; 804 | } 805 | 806 | 807 | function doLists($text) { 808 | # 809 | # Form HTML ordered (numbered) and unordered (bulleted) lists. 810 | # 811 | $less_than_tab = $this->tab_width - 1; 812 | 813 | # Re-usable patterns to match list item bullets and number markers: 814 | $marker_ul_re = '[*+-]'; 815 | $marker_ol_re = '\d+[.]'; 816 | $marker_any_re = "(?:$marker_ul_re|$marker_ol_re)"; 817 | 818 | $markers_relist = array( 819 | $marker_ul_re => $marker_ol_re, 820 | $marker_ol_re => $marker_ul_re, 821 | ); 822 | 823 | foreach ($markers_relist as $marker_re => $other_marker_re) { 824 | # Re-usable pattern to match any entirel ul or ol list: 825 | $whole_list_re = ' 826 | ( # $1 = whole list 827 | ( # $2 828 | ([ ]{0,'.$less_than_tab.'}) # $3 = number of spaces 829 | ('.$marker_re.') # $4 = first list item marker 830 | [ ]+ 831 | ) 832 | (?s:.+?) 833 | ( # $5 834 | \z 835 | | 836 | \n{2,} 837 | (?=\S) 838 | (?! # Negative lookahead for another list item marker 839 | [ ]* 840 | '.$marker_re.'[ ]+ 841 | ) 842 | | 843 | (?= # Lookahead for another kind of list 844 | \n 845 | \3 # Must have the same indentation 846 | '.$other_marker_re.'[ ]+ 847 | ) 848 | ) 849 | ) 850 | '; // mx 851 | 852 | # We use a different prefix before nested lists than top-level lists. 853 | # See extended comment in _ProcessListItems(). 854 | 855 | if ($this->list_level) { 856 | $text = preg_replace_callback('{ 857 | ^ 858 | '.$whole_list_re.' 859 | }mx', 860 | array(&$this, '_doLists_callback'), $text); 861 | } 862 | else { 863 | $text = preg_replace_callback('{ 864 | (?:(?<=\n)\n|\A\n?) # Must eat the newline 865 | '.$whole_list_re.' 866 | }mx', 867 | array(&$this, '_doLists_callback'), $text); 868 | } 869 | } 870 | 871 | return $text; 872 | } 873 | function _doLists_callback($matches) { 874 | # Re-usable patterns to match list item bullets and number markers: 875 | $marker_ul_re = '[*+-]'; 876 | $marker_ol_re = '\d+[.]'; 877 | $marker_any_re = "(?:$marker_ul_re|$marker_ol_re)"; 878 | 879 | $list = $matches[1]; 880 | $list_type = preg_match("/$marker_ul_re/", $matches[4]) ? "ul" : "ol"; 881 | 882 | $marker_any_re = ( $list_type == "ul" ? $marker_ul_re : $marker_ol_re ); 883 | 884 | $list .= "\n"; 885 | $result = $this->processListItems($list, $marker_any_re); 886 | 887 | $result = $this->hashBlock("<$list_type>\n" . $result . ""); 888 | return "\n". $result ."\n\n"; 889 | } 890 | 891 | var $list_level = 0; 892 | 893 | function processListItems($list_str, $marker_any_re) { 894 | # 895 | # Process the contents of a single ordered or unordered list, splitting it 896 | # into individual list items. 897 | # 898 | # The $this->list_level global keeps track of when we're inside a list. 899 | # Each time we enter a list, we increment it; when we leave a list, 900 | # we decrement. If it's zero, we're not in a list anymore. 901 | # 902 | # We do this because when we're not inside a list, we want to treat 903 | # something like this: 904 | # 905 | # I recommend upgrading to version 906 | # 8. Oops, now this line is treated 907 | # as a sub-list. 908 | # 909 | # As a single paragraph, despite the fact that the second line starts 910 | # with a digit-period-space sequence. 911 | # 912 | # Whereas when we're inside a list (or sub-list), that line will be 913 | # treated as the start of a sub-list. What a kludge, huh? This is 914 | # an aspect of Markdown's syntax that's hard to parse perfectly 915 | # without resorting to mind-reading. Perhaps the solution is to 916 | # change the syntax rules such that sub-lists must start with a 917 | # starting cardinal number; e.g. "1." or "a.". 918 | 919 | $this->list_level++; 920 | 921 | # trim trailing blank lines: 922 | $list_str = preg_replace("/\n{2,}\\z/", "\n", $list_str); 923 | 924 | $list_str = preg_replace_callback('{ 925 | (\n)? # leading line = $1 926 | (^[ ]*) # leading whitespace = $2 927 | ('.$marker_any_re.' # list marker and space = $3 928 | (?:[ ]+|(?=\n)) # space only required if item is not empty 929 | ) 930 | ((?s:.*?)) # list item text = $4 931 | (?:(\n+(?=\n))|\n) # tailing blank line = $5 932 | (?= \n* (\z | \2 ('.$marker_any_re.') (?:[ ]+|(?=\n)))) 933 | }xm', 934 | array(&$this, '_processListItems_callback'), $list_str); 935 | 936 | $this->list_level--; 937 | return $list_str; 938 | } 939 | function _processListItems_callback($matches) { 940 | $item = $matches[4]; 941 | $leading_line =& $matches[1]; 942 | $leading_space =& $matches[2]; 943 | $marker_space = $matches[3]; 944 | $tailing_blank_line =& $matches[5]; 945 | 946 | if ($leading_line || $tailing_blank_line || 947 | preg_match('/\n{2,}/', $item)) 948 | { 949 | # Replace marker with the appropriate whitespace indentation 950 | $item = $leading_space . str_repeat(' ', strlen($marker_space)) . $item; 951 | $item = $this->runBlockGamut($this->outdent($item)."\n"); 952 | } 953 | else { 954 | # Recursion for sub-lists: 955 | $item = $this->doLists($this->outdent($item)); 956 | $item = preg_replace('/\n+$/', '', $item); 957 | $item = $this->runSpanGamut($item); 958 | } 959 | 960 | return "

" . $item . "

\n"; 961 | } 962 | 963 | 964 | function doCodeBlocks($text) { 965 | # 966 | # Process Markdown `

` blocks.
 967 | 	#
 968 | 		$text = preg_replace_callback('{
 969 | 				(?:\n\n|\A\n?)
 970 | 				(	            # $1 = the code block -- one or more lines, starting with a space/tab
 971 | 				  (?>
 972 | 					[ ]{'.$this->tab_width.'}  # Lines must start with a tab or a tab-width of spaces
 973 | 					.*\n+
 974 | 				  )+
 975 | 				)
 976 | 				((?=^[ ]{0,'.$this->tab_width.'}\S)|\Z)	# Lookahead for non-space at line-start, or end of doc
 977 | 			}xm',
 978 | 			array(&$this, '_doCodeBlocks_callback'), $text);
 979 | 
 980 | 		return $text;
 981 | 	}
 982 | 	function _doCodeBlocks_callback($matches) {
 983 | 		$codeblock = $matches[1];
 984 | 
 985 | 		$codeblock = $this->outdent($codeblock);
 986 | 		$codeblock = htmlspecialchars($codeblock, ENT_NOQUOTES);
 987 | 
 988 | 		# trim leading newlines and trailing newlines
 989 | 		$codeblock = preg_replace('/\A\n+|\n+\z/', '', $codeblock);
 990 | 
 991 | 		$codeblock = "$codeblock\n";
 992 | 		return "\n\n".$this->hashBlock($codeblock)."\n\n";
 993 | 	}
 994 | 
 995 | 
 996 | 	function makeCodeSpan($code) {
 997 | 	#
 998 | 	# Create a code span markup for $code. Called from handleSpanToken.
 999 | 	#
1000 | 		$code = htmlspecialchars(trim($code), ENT_NOQUOTES);
1001 | 		return $this->hashPart("$code");
1002 | 	}
1003 | 
1004 | 
1005 | 	var $em_relist = array(
1006 | 		''  => '(?:(? '(?<=\S|^)(? '(?<=\S|^)(? '(?:(? '(?<=\S|^)(? '(?<=\S|^)(? '(?:(? '(?<=\S|^)(? '(?<=\S|^)(?em_relist as $em => $em_re) {
1028 | 			foreach ($this->strong_relist as $strong => $strong_re) {
1029 | 				# Construct list of allowed token expressions.
1030 | 				$token_relist = array();
1031 | 				if (isset($this->em_strong_relist["$em$strong"])) {
1032 | 					$token_relist[] = $this->em_strong_relist["$em$strong"];
1033 | 				}
1034 | 				$token_relist[] = $em_re;
1035 | 				$token_relist[] = $strong_re;
1036 | 				
1037 | 				# Construct master expression from list.
1038 | 				$token_re = '{('. implode('|', $token_relist) .')}';
1039 | 				$this->em_strong_prepared_relist["$em$strong"] = $token_re;
1040 | 			}
1041 | 		}
1042 | 	}
1043 | 	
1044 | 	function doItalicsAndBold($text) {
1045 | 		$token_stack = array('');
1046 | 		$text_stack = array('');
1047 | 		$em = '';
1048 | 		$strong = '';
1049 | 		$tree_char_em = false;
1050 | 		
1051 | 		while (1) {
1052 | 			#
1053 | 			# Get prepared regular expression for seraching emphasis tokens
1054 | 			# in current context.
1055 | 			#
1056 | 			$token_re = $this->em_strong_prepared_relist["$em$strong"];
1057 | 			
1058 | 			#
1059 | 			# Each loop iteration search for the next emphasis token. 
1060 | 			# Each token is then passed to handleSpanToken.
1061 | 			#
1062 | 			$parts = preg_split($token_re, $text, 2, PREG_SPLIT_DELIM_CAPTURE);
1063 | 			$text_stack[0] .= $parts[0];
1064 | 			$token =& $parts[1];
1065 | 			$text =& $parts[2];
1066 | 			
1067 | 			if (empty($token)) {
1068 | 				# Reached end of text span: empty stack without emitting.
1069 | 				# any more emphasis.
1070 | 				while ($token_stack[0]) {
1071 | 					$text_stack[1] .= array_shift($token_stack);
1072 | 					$text_stack[0] .= array_shift($text_stack);
1073 | 				}
1074 | 				break;
1075 | 			}
1076 | 			
1077 | 			$token_len = strlen($token);
1078 | 			if ($tree_char_em) {
1079 | 				# Reached closing marker while inside a three-char emphasis.
1080 | 				if ($token_len == 3) {
1081 | 					# Three-char closing marker, close em and strong.
1082 | 					array_shift($token_stack);
1083 | 					$span = array_shift($text_stack);
1084 | 					$span = $this->runSpanGamut($span);
1085 | 					$span = "$span";
1086 | 					$text_stack[0] .= $this->hashPart($span);
1087 | 					$em = '';
1088 | 					$strong = '';
1089 | 				} else {
1090 | 					# Other closing marker: close one em or strong and
1091 | 					# change current token state to match the other
1092 | 					$token_stack[0] = str_repeat($token{0}, 3-$token_len);
1093 | 					$tag = $token_len == 2 ? "strong" : "em";
1094 | 					$span = $text_stack[0];
1095 | 					$span = $this->runSpanGamut($span);
1096 | 					$span = "<$tag>$span";
1097 | 					$text_stack[0] = $this->hashPart($span);
1098 | 					$$tag = ''; # $$tag stands for $em or $strong
1099 | 				}
1100 | 				$tree_char_em = false;
1101 | 			} else if ($token_len == 3) {
1102 | 				if ($em) {
1103 | 					# Reached closing marker for both em and strong.
1104 | 					# Closing strong marker:
1105 | 					for ($i = 0; $i < 2; ++$i) {
1106 | 						$shifted_token = array_shift($token_stack);
1107 | 						$tag = strlen($shifted_token) == 2 ? "strong" : "em";
1108 | 						$span = array_shift($text_stack);
1109 | 						$span = $this->runSpanGamut($span);
1110 | 						$span = "<$tag>$span";
1111 | 						$text_stack[0] .= $this->hashPart($span);
1112 | 						$$tag = ''; # $$tag stands for $em or $strong
1113 | 					}
1114 | 				} else {
1115 | 					# Reached opening three-char emphasis marker. Push on token 
1116 | 					# stack; will be handled by the special condition above.
1117 | 					$em = $token{0};
1118 | 					$strong = "$em$em";
1119 | 					array_unshift($token_stack, $token);
1120 | 					array_unshift($text_stack, '');
1121 | 					$tree_char_em = true;
1122 | 				}
1123 | 			} else if ($token_len == 2) {
1124 | 				if ($strong) {
1125 | 					# Unwind any dangling emphasis marker:
1126 | 					if (strlen($token_stack[0]) == 1) {
1127 | 						$text_stack[1] .= array_shift($token_stack);
1128 | 						$text_stack[0] .= array_shift($text_stack);
1129 | 					}
1130 | 					# Closing strong marker:
1131 | 					array_shift($token_stack);
1132 | 					$span = array_shift($text_stack);
1133 | 					$span = $this->runSpanGamut($span);
1134 | 					$span = "$span";
1135 | 					$text_stack[0] .= $this->hashPart($span);
1136 | 					$strong = '';
1137 | 				} else {
1138 | 					array_unshift($token_stack, $token);
1139 | 					array_unshift($text_stack, '');
1140 | 					$strong = $token;
1141 | 				}
1142 | 			} else {
1143 | 				# Here $token_len == 1
1144 | 				if ($em) {
1145 | 					if (strlen($token_stack[0]) == 1) {
1146 | 						# Closing emphasis marker:
1147 | 						array_shift($token_stack);
1148 | 						$span = array_shift($text_stack);
1149 | 						$span = $this->runSpanGamut($span);
1150 | 						$span = "$span";
1151 | 						$text_stack[0] .= $this->hashPart($span);
1152 | 						$em = '';
1153 | 					} else {
1154 | 						$text_stack[0] .= $token;
1155 | 					}
1156 | 				} else {
1157 | 					array_unshift($token_stack, $token);
1158 | 					array_unshift($text_stack, '');
1159 | 					$em = $token;
1160 | 				}
1161 | 			}
1162 | 		}
1163 | 		return $text_stack[0];
1164 | 	}
1165 | 
1166 | 
1167 | 	function doBlockQuotes($text) {
1168 | 		$text = preg_replace_callback('/
1169 | 			  (								# Wrap whole match in $1
1170 | 				(?>
1171 | 				  ^[ ]*>[ ]?			# ">" at the start of a line
1172 | 					.+\n					# rest of the first line
1173 | 				  (.+\n)*					# subsequent consecutive lines
1174 | 				  \n*						# blanks
1175 | 				)+
1176 | 			  )
1177 | 			/xm',
1178 | 			array(&$this, '_doBlockQuotes_callback'), $text);
1179 | 
1180 | 		return $text;
1181 | 	}
1182 | 	function _doBlockQuotes_callback($matches) {
1183 | 		$bq = $matches[1];
1184 | 		# trim one level of quoting - trim whitespace-only lines
1185 | 		$bq = preg_replace('/^[ ]*>[ ]?|^[ ]+$/m', '', $bq);
1186 | 		$bq = $this->runBlockGamut($bq);		# recurse
1187 | 
1188 | 		$bq = preg_replace('/^/m', "  ", $bq);
1189 | 		# These leading spaces cause problem with  content, 
1190 | 		# so we need to fix that:
1191 | 		$bq = preg_replace_callback('{(\s*.+?)}sx', 
1192 | 			array(&$this, '_doBlockQuotes_callback2'), $bq);
1193 | 
1194 | 		return "\n". $this->hashBlock("\n$bq\n")."\n\n";
1195 | 	}
1196 | 	function _doBlockQuotes_callback2($matches) {
1197 | 		$pre = $matches[1];
1198 | 		$pre = preg_replace('/^  /m', '', $pre);
1199 | 		return $pre;
1200 | 	}
1201 | 
1202 | 
1203 | 	function formParagraphs($text) {
1204 | 	#
1205 | 	#	Params:
1206 | 	#		$text - string to process with html  tags
1207 | 	#
1208 | 		# Strip leading and trailing lines:
1209 | 		$text = preg_replace('/\A\n+|\n+\z/', '', $text);
1210 | 
1211 | 		$grafs = preg_split('/\n{2,}/', $text, -1, PREG_SPLIT_NO_EMPTY);
1212 | 
1213 | 		#
1214 | 		# Wrap 
 tags and unhashify HTML blocks
1215 | 		#
1216 | 		foreach ($grafs as $key => $value) {
1217 | 			if (!preg_match('/^B\x1A[0-9]+B$/', $value)) {
1218 | 				# Is a paragraph.
1219 | 				$value = $this->runSpanGamut($value);
1220 | 				$value = preg_replace('/^([ ]*)/', "
", $value);
1221 | 				$value .= "";
1222 | 				$grafs[$key] = $this->unhash($value);
1223 | 			}
1224 | 			else {
1225 | 				# Is a block.
1226 | 				# Modify elements of @grafs in-place...
1227 | 				$graf = $value;
1228 | 				$block = $this->html_hashes[$graf];
1229 | 				$graf = $block;
1230 | //				if (preg_match('{
1231 | //					\A
1232 | //					(							# $1 =  tag
1233 | //					  ]*
1235 | //					  \b
1236 | //					  markdown\s*=\s*  ([\'"])	#	$2 = attr quote char
1237 | //					  1
1238 | //					  \2
1239 | //					  [^>]*
1240 | //					  >
1241 | //					)
1242 | //					(							# $3 = contents
1243 | //					.*
1244 | //					)
1245 | //					()					# $4 = closing tag
1246 | //					\z
1247 | //					}xs', $block, $matches))
1248 | //				{
1249 | //					list(, $div_open, , $div_content, $div_close) = $matches;
1250 | //
1251 | //					# We can't call Markdown(), because that resets the hash;
1252 | //					# that initialization code should be pulled into its own sub, though.
1253 | //					$div_content = $this->hashHTMLBlocks($div_content);
1254 | //					
1255 | //					# Run document gamut methods on the content.
1256 | //					foreach ($this->document_gamut as $method => $priority) {
1257 | //						$div_content = $this->$method($div_content);
1258 | //					}
1259 | //
1260 | //					$div_open = preg_replace(
1261 | //						'{\smarkdown\s*=\s*([\'"]).+?\1}', '', $div_open);
1262 | //
1263 | //					$graf = $div_open . "\n" . $div_content . "\n" . $div_close;
1264 | //				}
1265 | 				$grafs[$key] = $graf;
1266 | 			}
1267 | 		}
1268 | 
1269 | 		return implode("\n\n", $grafs);
1270 | 	}
1271 | 
1272 | 
1273 | 	function encodeAttribute($text) {
1274 | 	#
1275 | 	# Encode text for a double-quoted HTML attribute. This function
1276 | 	# is *not* suitable for attributes enclosed in single quotes.
1277 | 	#
1278 | 		$text = $this->encodeAmpsAndAngles($text);
1279 | 		$text = str_replace('"', '"', $text);
1280 | 		return $text;
1281 | 	}
1282 | 	
1283 | 	
1284 | 	function encodeAmpsAndAngles($text) {
1285 | 	#
1286 | 	# Smart processing for ampersands and angle brackets that need to 
1287 | 	# be encoded. Valid character entities are left alone unless the
1288 | 	# no-entities mode is set.
1289 | 	#
1290 | 		if ($this->no_entities) {
1291 | 			$text = str_replace('&', '&', $text);
1292 | 		} else {
1293 | 			# Ampersand-encoding based entirely on Nat Irons's Amputator
1294 | 			# MT plugin: 
1295 | 			$text = preg_replace('/&(?!#?[xX]?(?:[0-9a-fA-F]+|\w+);)/', 
1296 | 								'&', $text);;
1297 | 		}
1298 | 		# Encode remaining <'s
1299 | 		$text = str_replace('<', '<', $text);
1300 | 
1301 | 		return $text;
1302 | 	}
1303 | 
1304 | 
1305 | 	function doAutoLinks($text) {
1306 | 		$text = preg_replace_callback('{<((https?|ftp|dict):[^\'">\s]+)>}i', 
1307 | 			array(&$this, '_doAutoLinks_url_callback'), $text);
1308 | 
1309 | 		# Email addresses: 
1310 | 		$text = preg_replace_callback('{
1311 | 			<
1312 | 			(?:mailto:)?
1313 | 			(
1314 | 				(?:
1315 | 					[-!#$%&\'*+/=?^_`.{|}~\w\x80-\xFF]+
1316 | 				|
1317 | 					".*?"
1318 | 				)
1319 | 				\@
1320 | 				(?:
1321 | 					[-a-z0-9\x80-\xFF]+(\.[-a-z0-9\x80-\xFF]+)*\.[a-z]+
1322 | 				|
1323 | 					\[[\d.a-fA-F:]+\]	# IPv4 & IPv6
1324 | 				)
1325 | 			)
1326 | 			>
1327 | 			}xi',
1328 | 			array(&$this, '_doAutoLinks_email_callback'), $text);
1329 | 
1330 | 		return $text;
1331 | 	}
1332 | 	function _doAutoLinks_url_callback($matches) {
1333 | 		$url = $this->encodeAttribute($matches[1]);
1334 | 		$link = "$url";
1335 | 		return $this->hashPart($link);
1336 | 	}
1337 | 	function _doAutoLinks_email_callback($matches) {
1338 | 		$address = $matches[1];
1339 | 		$link = $this->encodeEmailAddress($address);
1340 | 		return $this->hashPart($link);
1341 | 	}
1342 | 
1343 | 
1344 | 	function encodeEmailAddress($addr) {
1345 | 	#
1346 | 	#	Input: an email address, e.g. "foo@example.com"
1347 | 	#
1348 | 	#	Output: the email address as a mailto link, with each character
1349 | 	#		of the address encoded as either a decimal or hex entity, in
1350 | 	#		the hopes of foiling most address harvesting spam bots. E.g.:
1351 | 	#
1352 | 	#	  foo@exampl
1355 | 	#        e.com
1356 | 	#
1357 | 	#	Based by a filter by Matthew Wickline, posted to BBEdit-Talk.
1358 | 	#   With some optimizations by Milian Wolff.
1359 | 	#
1360 | 		$addr = "mailto:" . $addr;
1361 | 		$chars = preg_split('/(? $char) {
1365 | 			$ord = ord($char);
1366 | 			# Ignore non-ascii chars.
1367 | 			if ($ord < 128) {
1368 | 				$r = ($seed * (1 + $key)) % 100; # Pseudo-random function.
1369 | 				# roughly 10% raw, 45% hex, 45% dec
1370 | 				# '@' *must* be encoded. I insist.
1371 | 				if ($r > 90 && $char != '@') /* do nothing */;
1372 | 				else if ($r < 45) $chars[$key] = '&#x'.dechex($ord).';';
1373 | 				else              $chars[$key] = '&#'.$ord.';';
1374 | 			}
1375 | 		}
1376 | 		
1377 | 		$addr = implode('', $chars);
1378 | 		$text = implode('', array_slice($chars, 7)); # text without `mailto:`
1379 | 		$addr = "$text";
1380 | 
1381 | 		return $addr;
1382 | 	}
1383 | 
1384 | 
1385 | 	function parseSpan($str) {
1386 | 	#
1387 | 	# Take the string $str and parse it into tokens, hashing embeded HTML,
1388 | 	# escaped characters and handling code spans.
1389 | 	#
1390 | 		$output = '';
1391 | 		
1392 | 		$span_re = '{
1393 | 				(
1394 | 					\\\\'.$this->escape_chars_re.'
1395 | 				|
1396 | 					(?no_markup ? '' : '
1399 | 				|
1400 | 							# comment
1401 | 				|
1402 | 					<\?.*?\?> | <%.*?%>		# processing instruction
1403 | 				|
1404 | 					<[/!$]?[-a-zA-Z0-9:_]+	# regular tags
1405 | 					(?>
1406 | 						\s
1407 | 						(?>[^"\'>]+|"[^"]*"|\'[^\']*\')*
1408 | 					)?
1409 | 					>
1410 | 			').'
1411 | 				)
1412 | 				}xs';
1413 | 
1414 | 		while (1) {
1415 | 			#
1416 | 			# Each loop iteration seach for either the next tag, the next 
1417 | 			# openning code span marker, or the next escaped character. 
1418 | 			# Each token is then passed to handleSpanToken.
1419 | 			#
1420 | 			$parts = preg_split($span_re, $str, 2, PREG_SPLIT_DELIM_CAPTURE);
1421 | 			
1422 | 			# Create token from text preceding tag.
1423 | 			if ($parts[0] != "") {
1424 | 				$output .= $parts[0];
1425 | 			}
1426 | 			
1427 | 			# Check if we reach the end.
1428 | 			if (isset($parts[1])) {
1429 | 				$output .= $this->handleSpanToken($parts[1], $parts[2]);
1430 | 				$str = $parts[2];
1431 | 			}
1432 | 			else {
1433 | 				break;
1434 | 			}
1435 | 		}
1436 | 		
1437 | 		return $output;
1438 | 	}
1439 | 	
1440 | 	
1441 | 	function handleSpanToken($token, &$str) {
1442 | 	#
1443 | 	# Handle $token provided by parseSpan by determining its nature and 
1444 | 	# returning the corresponding value that should replace it.
1445 | 	#
1446 | 		switch ($token{0}) {
1447 | 			case "\\":
1448 | 				return $this->hashPart("&#". ord($token{1}). ";");
1449 | 			case "`":
1450 | 				# Search for end marker in remaining text.
1451 | 				if (preg_match('/^(.*?[^`])'.preg_quote($token).'(?!`)(.*)$/sm', 
1452 | 					$str, $matches))
1453 | 				{
1454 | 					$str = $matches[2];
1455 | 					$codespan = $this->makeCodeSpan($matches[1]);
1456 | 					return $this->hashPart($codespan);
1457 | 				}
1458 | 				return $token; // return as text since no ending marker found.
1459 | 			default:
1460 | 				return $this->hashPart($token);
1461 | 		}
1462 | 	}
1463 | 
1464 | 
1465 | 	function outdent($text) {
1466 | 	#
1467 | 	# Remove one level of line-leading tabs or spaces
1468 | 	#
1469 | 		return preg_replace('/^(\t|[ ]{1,'.$this->tab_width.'})/m', '', $text);
1470 | 	}
1471 | 
1472 | 
1473 | 	# String length function for detab. `_initDetab` will create a function to 
1474 | 	# hanlde UTF-8 if the default function does not exist.
1475 | 	var $utf8_strlen = 'mb_strlen';
1476 | 	
1477 | 	function detab($text) {
1478 | 	#
1479 | 	# Replace tabs with the appropriate amount of space.
1480 | 	#
1481 | 		# For each line we separate the line in blocks delemited by
1482 | 		# tab characters. Then we reconstruct every line by adding the 
1483 | 		# appropriate number of space between each blocks.
1484 | 		
1485 | 		$text = preg_replace_callback('/^.*\t.*$/m',
1486 | 			array(&$this, '_detab_callback'), $text);
1487 | 
1488 | 		return $text;
1489 | 	}
1490 | 	function _detab_callback($matches) {
1491 | 		$line = $matches[0];
1492 | 		$strlen = $this->utf8_strlen; # strlen function for UTF-8.
1493 | 		
1494 | 		# Split in blocks.
1495 | 		$blocks = explode("\t", $line);
1496 | 		# Add each blocks to the line.
1497 | 		$line = $blocks[0];
1498 | 		unset($blocks[0]); # Do not add first block twice.
1499 | 		foreach ($blocks as $block) {
1500 | 			# Calculate amount of space, insert spaces, insert block.
1501 | 			$amount = $this->tab_width - 
1502 | 				$strlen($line, 'UTF-8') % $this->tab_width;
1503 | 			$line .= str_repeat(" ", $amount) . $block;
1504 | 		}
1505 | 		return $line;
1506 | 	}
1507 | 	function _initDetab() {
1508 | 	#
1509 | 	# Check for the availability of the function in the `utf8_strlen` property
1510 | 	# (initially `mb_strlen`). If the function is not available, create a 
1511 | 	# function that will loosely count the number of UTF-8 characters with a
1512 | 	# regular expression.
1513 | 	#
1514 | 		if (function_exists($this->utf8_strlen)) return;
1515 | 		$this->utf8_strlen = create_function('$text', 'return preg_match_all(
1516 | 			"/[\\\\x00-\\\\xBF]|[\\\\xC0-\\\\xFF][\\\\x80-\\\\xBF]*/", 
1517 | 			$text, $m);');
1518 | 	}
1519 | 
1520 | 
1521 | 	function unhash($text) {
1522 | 	#
1523 | 	# Swap back in all the tags hashed by _HashHTMLBlocks.
1524 | 	#
1525 | 		return preg_replace_callback('/(.)\x1A[0-9]+\1/', 
1526 | 			array(&$this, '_unhash_callback'), $text);
1527 | 	}
1528 | 	function _unhash_callback($matches) {
1529 | 		return $this->html_hashes[$matches[0]];
1530 | 	}
1531 | 
1532 | }
1533 | 
1534 | 
1535 | #
1536 | # Markdown Extra Parser Class
1537 | #
1538 | 
1539 | class MarkdownExtra_Parser extends Markdown_Parser {
1540 | 
1541 | 	# Prefix for footnote ids.
1542 | 	var $fn_id_prefix = "";
1543 | 	
1544 | 	# Optional title attribute for footnote links and backlinks.
1545 | 	var $fn_link_title = MARKDOWN_FN_LINK_TITLE;
1546 | 	var $fn_backlink_title = MARKDOWN_FN_BACKLINK_TITLE;
1547 | 	
1548 | 	# Optional class attribute for footnote links and backlinks.
1549 | 	var $fn_link_class = MARKDOWN_FN_LINK_CLASS;
1550 | 	var $fn_backlink_class = MARKDOWN_FN_BACKLINK_CLASS;
1551 | 	
1552 | 	# Predefined abbreviations.
1553 | 	var $predef_abbr = array();
1554 | 
1555 | 
1556 | 	function MarkdownExtra_Parser() {
1557 | 	#
1558 | 	# Constructor function. Initialize the parser object.
1559 | 	#
1560 | 		# Add extra escapable characters before parent constructor 
1561 | 		# initialize the table.
1562 | 		$this->escape_chars .= ':|';
1563 | 		
1564 | 		# Insert extra document, block, and span transformations. 
1565 | 		# Parent constructor will do the sorting.
1566 | 		$this->document_gamut += array(
1567 | 			"doFencedCodeBlocks" => 5,
1568 | 			"stripFootnotes"     => 15,
1569 | 			"stripAbbreviations" => 25,
1570 | 			"appendFootnotes"    => 50,
1571 | 			);
1572 | 		$this->block_gamut += array(
1573 | 			"doFencedCodeBlocks" => 5,
1574 | 			"doTables"           => 15,
1575 | 			"doDefLists"         => 45,
1576 | 			);
1577 | 		$this->span_gamut += array(
1578 | 			"doFootnotes"        => 5,
1579 | 			"doAbbreviations"    => 70,
1580 | 			);
1581 | 		
1582 | 		parent::Markdown_Parser();
1583 | 	}
1584 | 	
1585 | 	
1586 | 	# Extra variables used during extra transformations.
1587 | 	var $footnotes = array();
1588 | 	var $footnotes_ordered = array();
1589 | 	var $abbr_desciptions = array();
1590 | 	var $abbr_word_re = '';
1591 | 	
1592 | 	# Give the current footnote number.
1593 | 	var $footnote_counter = 1;
1594 | 	
1595 | 	
1596 | 	function setup() {
1597 | 	#
1598 | 	# Setting up Extra-specific variables.
1599 | 	#
1600 | 		parent::setup();
1601 | 		
1602 | 		$this->footnotes = array();
1603 | 		$this->footnotes_ordered = array();
1604 | 		$this->abbr_desciptions = array();
1605 | 		$this->abbr_word_re = '';
1606 | 		$this->footnote_counter = 1;
1607 | 		
1608 | 		foreach ($this->predef_abbr as $abbr_word => $abbr_desc) {
1609 | 			if ($this->abbr_word_re)
1610 | 				$this->abbr_word_re .= '|';
1611 | 			$this->abbr_word_re .= preg_quote($abbr_word);
1612 | 			$this->abbr_desciptions[$abbr_word] = trim($abbr_desc);
1613 | 		}
1614 | 	}
1615 | 	
1616 | 	function teardown() {
1617 | 	#
1618 | 	# Clearing Extra-specific variables.
1619 | 	#
1620 | 		$this->footnotes = array();
1621 | 		$this->footnotes_ordered = array();
1622 | 		$this->abbr_desciptions = array();
1623 | 		$this->abbr_word_re = '';
1624 | 		
1625 | 		parent::teardown();
1626 | 	}
1627 | 	
1628 | 	
1629 | 	### HTML Block Parser ###
1630 | 	
1631 | 	# Tags that are always treated as block tags:
1632 | 	var $block_tags_re = 'p|div|h[1-6]|blockquote|pre|table|dl|ol|ul|address|form|fieldset|iframe|hr|legend';
1633 | 	
1634 | 	# Tags treated as block tags only if the opening tag is alone on it's line:
1635 | 	var $context_block_tags_re = 'script|noscript|math|ins|del';
1636 | 	
1637 | 	# Tags where markdown="1" default to span mode:
1638 | 	var $contain_span_tags_re = 'p|h[1-6]|li|dd|dt|td|th|legend|address';
1639 | 	
1640 | 	# Tags which must not have their contents modified, no matter where 
1641 | 	# they appear:
1642 | 	var $clean_tags_re = 'script|math';
1643 | 	
1644 | 	# Tags that do not need to be closed.
1645 | 	var $auto_close_tags_re = 'hr|img';
1646 | 	
1647 | 
1648 | 	function hashHTMLBlocks($text) {
1649 | 	#
1650 | 	# Hashify HTML Blocks and "clean tags".
1651 | 	#
1652 | 	# We only want to do this for block-level HTML tags, such as headers,
1653 | 	# lists, and tables. That's because we still want to wrap s around
1654 | 	# "paragraphs" that are wrapped in non-block-level tags, such as anchors,
1655 | 	# phrase emphasis, and spans. The list of tags we're looking for is
1656 | 	# hard-coded.
1657 | 	#
1658 | 	# This works by calling _HashHTMLBlocks_InMarkdown, which then calls
1659 | 	# _HashHTMLBlocks_InHTML when it encounter block tags. When the markdown="1" 
1660 | 	# attribute is found whitin a tag, _HashHTMLBlocks_InHTML calls back
1661 | 	#  _HashHTMLBlocks_InMarkdown to handle the Markdown syntax within the tag.
1662 | 	# These two functions are calling each other. It's recursive!
1663 | 	#
1664 | 		#
1665 | 		# Call the HTML-in-Markdown hasher.
1666 | 		#
1667 | 		list($text, ) = $this->_hashHTMLBlocks_inMarkdown($text);
1668 | 		
1669 | 		return $text;
1670 | 	}
1671 | 	function _hashHTMLBlocks_inMarkdown($text, $indent = 0, 
1672 | 										$enclosing_tag_re = '', $span = false)
1673 | 	{
1674 | 	#
1675 | 	# Parse markdown text, calling _HashHTMLBlocks_InHTML for block tags.
1676 | 	#
1677 | 	# *   $indent is the number of space to be ignored when checking for code 
1678 | 	#     blocks. This is important because if we don't take the indent into 
1679 | 	#     account, something like this (which looks right) won't work as expected:
1680 | 	#
1681 | 	#     

1682 | 	#         
1683 | 	#         Hello World.  <-- Is this a Markdown code block or text?
1684 | 	#           <-- Is this a Markdown code block or a real tag?
1685 | 	#     
1686 | 	#
1687 | 	#     If you don't like this, just don't indent the tag on which
1688 | 	#     you apply the markdown="1" attribute.
1689 | 	#
1690 | 	# *   If $enclosing_tag_re is not empty, stops at the first unmatched closing 
1691 | 	#     tag with that name. Nested tags supported.
1692 | 	#
1693 | 	# *   If $span is true, text inside must treated as span. So any double 
1694 | 	#     newline will be replaced by a single newline so that it does not create 
1695 | 	#     paragraphs.
1696 | 	#
1697 | 	# Returns an array of that form: ( processed text , remaining text )
1698 | 	#
1699 | 		if ($text === '') return array('', '');
1700 | 
1701 | 		# Regex to check for the presense of newlines around a block tag.
1702 | 		$newline_before_re = '/(?:^\n?|\n\n)*$/';
1703 | 		$newline_after_re = 
1704 | 			'{
1705 | 				^						# Start of text following the tag.
1706 | 				(?>[ ]*)?		# Optional comment.
1707 | 				[ ]*\n					# Must be followed by newline.
1708 | 			}xs';
1709 | 		
1710 | 		# Regex to match any tag.
1711 | 		$block_tag_re =
1712 | 			'{
1713 | 				(					# $2: Capture hole tag.
1714 | 									# Tag name.
1716 | 							'.$this->block_tags_re.'			|
1717 | 							'.$this->context_block_tags_re.'	|
1718 | 							'.$this->clean_tags_re.'        	|
1719 | 							(?!\s)'.$enclosing_tag_re.'
1720 | 						)
1721 | 						(?:
1722 | 							(?=[\s"\'/a-zA-Z0-9])	# Allowed characters after tag name.
1723 | 							(?>
1724 | 								".*?"		|	# Double quotes (can contain `>`)
1725 | 								\'.*?\'   	|	# Single quotes (can contain `>`)
1726 | 								.+?				# Anything but quotes and `>`.
1727 | 							)*?
1728 | 						)?
1729 | 					>					# End of tag.
1730 | 				|
1731 | 						# HTML Comment
1732 | 				|
1733 | 					<\?.*?\?> | <%.*?%>	# Processing instruction
1734 | 				|
1735 | 						# CData Block
1736 | 				|
1737 | 					# Code span marker
1738 | 					`+
1739 | 				'. ( !$span ? ' # If not in span.
1740 | 				|
1741 | 					# Indented code block
1742 | 					(?: ^[ ]*\n | ^ | \n[ ]*\n )
1743 | 					[ ]{'.($indent+4).'}[^\n]* \n
1744 | 					(?>
1745 | 						(?: [ ]{'.($indent+4).'}[^\n]* | [ ]* ) \n
1746 | 					)*
1747 | 				|
1748 | 					# Fenced code block marker
1749 | 					(?> ^ | \n )
1750 | 					[ ]{'.($indent).'}~~~+[ ]*\n
1751 | 				' : '' ). ' # End (if not is span).
1752 | 				)
1753 | 			}xs';
1754 | 
1755 | 		
1756 | 		$depth = 0;		# Current depth inside the tag tree.
1757 | 		$parsed = "";	# Parsed text that will be returned.
1758 | 
1759 | 		#
1760 | 		# Loop through every tag until we find the closing tag of the parent
1761 | 		# or loop until reaching the end of text if no parent tag specified.
1762 | 		#
1763 | 		do {
1764 | 			#
1765 | 			# Split the text using the first $tag_match pattern found.
1766 | 			# Text before  pattern will be first in the array, text after
1767 | 			# pattern will be at the end, and between will be any catches made 
1768 | 			# by the pattern.
1769 | 			#
1770 | 			$parts = preg_split($block_tag_re, $text, 2, 
1771 | 								PREG_SPLIT_DELIM_CAPTURE);
1772 | 			
1773 | 			# If in Markdown span mode, add a empty-string span-level hash 
1774 | 			# after each newline to prevent triggering any block element.
1775 | 			if ($span) {
1776 | 				$void = $this->hashPart("", ':');
1777 | 				$newline = "$void\n";
1778 | 				$parts[0] = $void . str_replace("\n", $newline, $parts[0]) . $void;
1779 | 			}
1780 | 			
1781 | 			$parsed .= $parts[0]; # Text before current tag.
1782 | 			
1783 | 			# If end of $text has been reached. Stop loop.
1784 | 			if (count($parts) < 3) {
1785 | 				$text = "";
1786 | 				break;
1787 | 			}
1788 | 			
1789 | 			$tag  = $parts[1]; # Tag to handle.
1790 | 			$text = $parts[2]; # Remaining text after current tag.
1791 | 			$tag_re = preg_quote($tag); # For use in a regular expression.
1792 | 			
1793 | 			#
1794 | 			# Check for: Code span marker
1795 | 			#
1796 | 			if ($tag{0} == "`") {
1797 | 				# Find corresponding end marker.
1798 | 				$tag_re = preg_quote($tag);
1799 | 				if (preg_match('{^(?>.+?|\n(?!\n))*?(?.*\n)+?'.$tag_re.' *\n}', $text, 
1826 | 					$matches)) 
1827 | 				{
1828 | 					# End marker found: pass text unchanged until marker.
1829 | 					$parsed .= $tag . $matches[0];
1830 | 					$text = substr($text, strlen($matches[0]));
1831 | 				}
1832 | 				else {
1833 | 					# No end marker: just skip it.
1834 | 					$parsed .= $tag;
1835 | 				}
1836 | 			}
1837 | 			#
1838 | 			# Check for: Opening Block level tag or
1839 | 			#            Opening Context Block tag (like ins and del) 
1840 | 			#               used as a block tag (tag is alone on it's line).
1841 | 			#
1842 | 			else if (preg_match('{^<(?:'.$this->block_tags_re.')\b}', $tag) ||
1843 | 				(	preg_match('{^<(?:'.$this->context_block_tags_re.')\b}', $tag) &&
1844 | 					preg_match($newline_before_re, $parsed) &&
1845 | 					preg_match($newline_after_re, $text)	)
1846 | 				)
1847 | 			{
1848 | 				# Need to parse tag and following text using the HTML parser.
1849 | 				list($block_text, $text) = 
1850 | 					$this->_hashHTMLBlocks_inHTML($tag . $text, "hashBlock", true);
1851 | 				
1852 | 				# Make sure it stays outside of any paragraph by adding newlines.
1853 | 				$parsed .= "\n\n$block_text\n\n";
1854 | 			}
1855 | 			#
1856 | 			# Check for: Clean tag (like script, math)
1857 | 			#            HTML Comments, processing instructions.
1858 | 			#
1859 | 			else if (preg_match('{^<(?:'.$this->clean_tags_re.')\b}', $tag) ||
1860 | 				$tag{1} == '!' || $tag{1} == '?')
1861 | 			{
1862 | 				# Need to parse tag and following text using the HTML parser.
1863 | 				# (don't check for markdown attribute)
1864 | 				list($block_text, $text) = 
1865 | 					$this->_hashHTMLBlocks_inHTML($tag . $text, "hashClean", false);
1866 | 				
1867 | 				$parsed .= $block_text;
1868 | 			}
1869 | 			#
1870 | 			# Check for: Tag with same name as enclosing tag.
1871 | 			#
1872 | 			else if ($enclosing_tag_re !== '' &&
1873 | 				# Same name as enclosing tag.
1874 | 				preg_match('{^= 0);
1897 | 		
1898 | 		return array($parsed, $text);
1899 | 	}
1900 | 	function _hashHTMLBlocks_inHTML($text, $hash_method, $md_attr) {
1901 | 	#
1902 | 	# Parse HTML, calling _HashHTMLBlocks_InMarkdown for block tags.
1903 | 	#
1904 | 	# *   Calls $hash_method to convert any blocks.
1905 | 	# *   Stops when the first opening tag closes.
1906 | 	# *   $md_attr indicate if the use of the `markdown="1"` attribute is allowed.
1907 | 	#     (it is not inside clean tags)
1908 | 	#
1909 | 	# Returns an array of that form: ( processed text , remaining text )
1910 | 	#
1911 | 		if ($text === '') return array('', '');
1912 | 		
1913 | 		# Regex to match `markdown` attribute inside of a tag.
1914 | 		$markdown_attr_re = '
1915 | 			{
1916 | 				\s*			# Eat whitespace before the `markdown` attribute
1917 | 				markdown
1918 | 				\s*=\s*
1919 | 				(?>
1920 | 					(["\'])		# $1: quote delimiter		
1921 | 					(.*?)		# $2: attribute value
1922 | 					\1			# matching delimiter	
1923 | 				|
1924 | 					([^\s>]*)	# $3: unquoted attribute value
1925 | 				)
1926 | 				()				# $4: make $3 always defined (avoid warnings)
1927 | 			}xs';
1928 | 		
1929 | 		# Regex to match any tag.
1930 | 		$tag_re = '{
1931 | 				(					# $2: Capture hole tag.
1932 | 					
1937 | 								".*?"		|	# Double quotes (can contain `>`)
1938 | 								\'.*?\'   	|	# Single quotes (can contain `>`)
1939 | 								.+?				# Anything but quotes and `>`.
1940 | 							)*?
1941 | 						)?
1942 | 					>					# End of tag.
1943 | 				|
1944 | 						# HTML Comment
1945 | 				|
1946 | 					<\?.*?\?> | <%.*?%>	# Processing instruction
1947 | 				|
1948 | 						# CData Block
1949 | 				)
1950 | 			}xs';
1951 | 		
1952 | 		$original_text = $text;		# Save original text in case of faliure.
1953 | 		
1954 | 		$depth		= 0;	# Current depth inside the tag tree.
1955 | 		$block_text	= "";	# Temporary text holder for current text.
1956 | 		$parsed		= "";	# Parsed text that will be returned.
1957 | 
1958 | 		#
1959 | 		# Get the name of the starting tag.
1960 | 		# (This pattern makes $base_tag_name_re safe without quoting.)
1961 | 		#
1962 | 		if (preg_match('/^<([\w:$]*)\b/', $text, $matches))
1963 | 			$base_tag_name_re = $matches[1];
1964 | 
1965 | 		#
1966 | 		# Loop through every tag until we find the corresponding closing tag.
1967 | 		#
1968 | 		do {
1969 | 			#
1970 | 			# Split the text using the first $tag_match pattern found.
1971 | 			# Text before  pattern will be first in the array, text after
1972 | 			# pattern will be at the end, and between will be any catches made 
1973 | 			# by the pattern.
1974 | 			#
1975 | 			$parts = preg_split($tag_re, $text, 2, PREG_SPLIT_DELIM_CAPTURE);
1976 | 			
1977 | 			if (count($parts) < 3) {
1978 | 				#
1979 | 				# End of $text reached with unbalenced tag(s).
1980 | 				# In that case, we return original text unchanged and pass the
1981 | 				# first character as filtered to prevent an infinite loop in the 
1982 | 				# parent function.
1983 | 				#
1984 | 				return array($original_text{0}, substr($original_text, 1));
1985 | 			}
1986 | 			
1987 | 			$block_text .= $parts[0]; # Text before current tag.
1988 | 			$tag         = $parts[1]; # Tag to handle.
1989 | 			$text        = $parts[2]; # Remaining text after current tag.
1990 | 			
1991 | 			#
1992 | 			# Check for: Auto-close tag (like )
1993 | 			#			 Comments and Processing Instructions.
1994 | 			#
1995 | 			if (preg_match('{^auto_close_tags_re.')\b}', $tag) ||
1996 | 				$tag{1} == '!' || $tag{1} == '?')
1997 | 			{
1998 | 				# Just add the tag to the block as if it was text.
1999 | 				$block_text .= $tag;
2000 | 			}
2001 | 			else {
2002 | 				#
2003 | 				# Increase/decrease nested tag count. Only do so if
2004 | 				# the tag's name match base tag's.
2005 | 				#
2006 | 				if (preg_match('{^mode = $attr_m[2] . $attr_m[3];
2023 | 					$span_mode = $this->mode == 'span' || $this->mode != 'block' &&
2024 | 						preg_match('{^<(?:'.$this->contain_span_tags_re.')\b}', $tag);
2025 | 					
2026 | 					# Calculate indent before tag.
2027 | 					if (preg_match('/(?:^|\n)( *?)(?! ).*?$/', $block_text, $matches)) {
2028 | 						$strlen = $this->utf8_strlen;
2029 | 						$indent = $strlen($matches[1], 'UTF-8');
2030 | 					} else {
2031 | 						$indent = 0;
2032 | 					}
2033 | 					
2034 | 					# End preceding block with this tag.
2035 | 					$block_text .= $tag;
2036 | 					$parsed .= $this->$hash_method($block_text);
2037 | 					
2038 | 					# Get enclosing tag name for the ParseMarkdown function.
2039 | 					# (This pattern makes $tag_name_re safe without quoting.)
2040 | 					preg_match('/^<([\w:$]*)\b/', $tag, $matches);
2041 | 					$tag_name_re = $matches[1];
2042 | 					
2043 | 					# Parse the content using the HTML-in-Markdown parser.
2044 | 					list ($block_text, $text)
2045 | 						= $this->_hashHTMLBlocks_inMarkdown($text, $indent, 
2046 | 							$tag_name_re, $span_mode);
2047 | 					
2048 | 					# Outdent markdown text.
2049 | 					if ($indent > 0) {
2050 | 						$block_text = preg_replace("/^[ ]{1,$indent}/m", "", 
2051 | 													$block_text);
2052 | 					}
2053 | 					
2054 | 					# Append tag content to parsed text.
2055 | 					if (!$span_mode)	$parsed .= "\n\n$block_text\n\n";
2056 | 					else				$parsed .= "$block_text";
2057 | 					
2058 | 					# Start over a new block.
2059 | 					$block_text = "";
2060 | 				}
2061 | 				else $block_text .= $tag;
2062 | 			}
2063 | 			
2064 | 		} while ($depth > 0);
2065 | 		
2066 | 		#
2067 | 		# Hash last block text that wasn't processed inside the loop.
2068 | 		#
2069 | 		$parsed .= $this->$hash_method($block_text);
2070 | 		
2071 | 		return array($parsed, $text);
2072 | 	}
2073 | 
2074 | 
2075 | 	function hashClean($text) {
2076 | 	#
2077 | 	# Called whenever a tag must be hashed when a function insert a "clean" tag
2078 | 	# in $text, it pass through this function and is automaticaly escaped, 
2079 | 	# blocking invalid nested overlap.
2080 | 	#
2081 | 		return $this->hashPart($text, 'C');
2082 | 	}
2083 | 
2084 | 
2085 | 	function doHeaders($text) {
2086 | 	#
2087 | 	# Redefined to add id attribute support.
2088 | 	#
2089 | 		# Setext-style headers:
2090 | 		#	  Header 1  {#header1}
2091 | 		#	  ========
2092 | 		#  
2093 | 		#	  Header 2  {#header2}
2094 | 		#	  --------
2095 | 		#
2096 | 		$text = preg_replace_callback(
2097 | 			'{
2098 | 				(^.+?)								# $1: Header text
2099 | 				(?:[ ]+\{\#([-_:a-zA-Z0-9]+)\})?	# $2: Id attribute
2100 | 				[ ]*\n(=+|-+)[ ]*\n+				# $3: Header footer
2101 | 			}mx',
2102 | 			array(&$this, '_doHeaders_callback_setext'), $text);
2103 | 
2104 | 		# atx-style headers:
2105 | 		#	# Header 1        {#header1}
2106 | 		#	## Header 2       {#header2}
2107 | 		#	## Header 2 with closing hashes ##  {#header3}
2108 | 		#	...
2109 | 		#	###### Header 6   {#header2}
2110 | 		#
2111 | 		$text = preg_replace_callback('{
2112 | 				^(\#{1,6})	# $1 = string of #\'s
2113 | 				[ ]*
2114 | 				(.+?)		# $2 = Header text
2115 | 				[ ]*
2116 | 				\#*			# optional closing #\'s (not counted)
2117 | 				(?:[ ]+\{\#([-_:a-zA-Z0-9]+)\})? # id attribute
2118 | 				[ ]*
2119 | 				\n+
2120 | 			}xm',
2121 | 			array(&$this, '_doHeaders_callback_atx'), $text);
2122 | 
2123 | 		return $text;
2124 | 	}
2125 | 	function _doHeaders_attr($attr) {
2126 | 		if (empty($attr))  return "";
2127 | 		return " id=\"$attr\"";
2128 | 	}
2129 | 	function _doHeaders_callback_setext($matches) {
2130 | 		if ($matches[3] == '-' && preg_match('{^- }', $matches[1]))
2131 | 			return $matches[0];
2132 | 		$level = $matches[3]{0} == '=' ? 1 : 2;
2133 | 		$attr  = $this->_doHeaders_attr($id =& $matches[2]);
2134 | 		$block = "".$this->runSpanGamut($matches[1])."";
2135 | 		return "\n" . $this->hashBlock($block) . "\n\n";
2136 | 	}
2137 | 	function _doHeaders_callback_atx($matches) {
2138 | 		$level = strlen($matches[1]);
2139 | 		$attr  = $this->_doHeaders_attr($id =& $matches[3]);
2140 | 		$block = "".$this->runSpanGamut($matches[2])."";
2141 | 		return "\n" . $this->hashBlock($block) . "\n\n";
2142 | 	}
2143 | 
2144 | 
2145 | 	function doTables($text) {
2146 | 	#
2147 | 	# Form HTML tables.
2148 | 	#
2149 | 		$less_than_tab = $this->tab_width - 1;
2150 | 		#
2151 | 		# Find tables with leading pipe.
2152 | 		#
2153 | 		#	| Header 1 | Header 2
2154 | 		#	| -------- | --------
2155 | 		#	| Cell 1   | Cell 2
2156 | 		#	| Cell 3   | Cell 4
2157 | 		#
2158 | 		$text = preg_replace_callback('
2159 | 			{
2160 | 				^							# Start of a line
2161 | 				[ ]{0,'.$less_than_tab.'}	# Allowed whitespace.
2162 | 				[|]							# Optional leading pipe (present)
2163 | 				(.+) \n						# $1: Header row (at least one pipe)
2164 | 				
2165 | 				[ ]{0,'.$less_than_tab.'}	# Allowed whitespace.
2166 | 				[|] ([ ]*[-:]+[-| :]*) \n	# $2: Header underline
2167 | 				
2168 | 				(							# $3: Cells
2169 | 					(?>
2170 | 						[ ]*				# Allowed whitespace.
2171 | 						[|] .* \n			# Row content.
2172 | 					)*
2173 | 				)
2174 | 				(?=\n|\Z)					# Stop at final double newline.
2175 | 			}xm',
2176 | 			array(&$this, '_doTable_leadingPipe_callback'), $text);
2177 | 		
2178 | 		#
2179 | 		# Find tables without leading pipe.
2180 | 		#
2181 | 		#	Header 1 | Header 2
2182 | 		#	-------- | --------
2183 | 		#	Cell 1   | Cell 2
2184 | 		#	Cell 3   | Cell 4
2185 | 		#
2186 | 		$text = preg_replace_callback('
2187 | 			{
2188 | 				^							# Start of a line
2189 | 				[ ]{0,'.$less_than_tab.'}	# Allowed whitespace.
2190 | 				(\S.*[|].*) \n				# $1: Header row (at least one pipe)
2191 | 				
2192 | 				[ ]{0,'.$less_than_tab.'}	# Allowed whitespace.
2193 | 				([-:]+[ ]*[|][-| :]*) \n	# $2: Header underline
2194 | 				
2195 | 				(							# $3: Cells
2196 | 					(?>
2197 | 						.* [|] .* \n		# Row content
2198 | 					)*
2199 | 				)
2200 | 				(?=\n|\Z)					# Stop at final double newline.
2201 | 			}xm',
2202 | 			array(&$this, '_DoTable_callback'), $text);
2203 | 
2204 | 		return $text;
2205 | 	}
2206 | 	function _doTable_leadingPipe_callback($matches) {
2207 | 		$head		= $matches[1];
2208 | 		$underline	= $matches[2];
2209 | 		$content	= $matches[3];
2210 | 		
2211 | 		# Remove leading pipe for each row.
2212 | 		$content	= preg_replace('/^ *[|]/m', '', $content);
2213 | 		
2214 | 		return $this->_doTable_callback(array($matches[0], $head, $underline, $content));
2215 | 	}
2216 | 	function _doTable_callback($matches) {
2217 | 		$head		= $matches[1];
2218 | 		$underline	= $matches[2];
2219 | 		$content	= $matches[3];
2220 | 
2221 | 		# Remove any tailing pipes for each line.
2222 | 		$head		= preg_replace('/[|] *$/m', '', $head);
2223 | 		$underline	= preg_replace('/[|] *$/m', '', $underline);
2224 | 		$content	= preg_replace('/[|] *$/m', '', $content);
2225 | 		
2226 | 		# Reading alignement from header underline.
2227 | 		$separators	= preg_split('/ *[|] */', $underline);
2228 | 		foreach ($separators as $n => $s) {
2229 | 			if (preg_match('/^ *-+: *$/', $s))		$attr[$n] = ' align="right"';
2230 | 			else if (preg_match('/^ *:-+: *$/', $s))$attr[$n] = ' align="center"';
2231 | 			else if (preg_match('/^ *:-+ *$/', $s))	$attr[$n] = ' align="left"';
2232 | 			else									$attr[$n] = '';
2233 | 		}
2234 | 		
2235 | 		# Parsing span elements, including code spans, character escapes, 
2236 | 		# and inline HTML tags, so that pipes inside those gets ignored.
2237 | 		$head		= $this->parseSpan($head);
2238 | 		$headers	= preg_split('/ *[|] */', $head);
2239 | 		$col_count	= count($headers);
2240 | 		
2241 | 		# Write column headers.
2242 | 		$text = "\n";
2243 | 		$text .= "\n";
2244 | 		$text .= "\n";
2245 | 		foreach ($headers as $n => $header)
2246 | 			$text .= "  ".$this->runSpanGamut(trim($header))."\n";
2247 | 		$text .= "\n";
2248 | 		$text .= "\n";
2249 | 		
2250 | 		# Split content by row.
2251 | 		$rows = explode("\n", trim($content, "\n"));
2252 | 		
2253 | 		$text .= "\n";
2254 | 		foreach ($rows as $row) {
2255 | 			# Parsing span elements, including code spans, character escapes, 
2256 | 			# and inline HTML tags, so that pipes inside those gets ignored.
2257 | 			$row = $this->parseSpan($row);
2258 | 			
2259 | 			# Split row by cell.
2260 | 			$row_cells = preg_split('/ *[|] */', $row, $col_count);
2261 | 			$row_cells = array_pad($row_cells, $col_count, '');
2262 | 			
2263 | 			$text .= "\n";
2264 | 			foreach ($row_cells as $n => $cell)
2265 | 				$text .= "  ".$this->runSpanGamut(trim($cell))."\n";
2266 | 			$text .= "\n";
2267 | 		}
2268 | 		$text .= "\n";
2269 | 		$text .= "
";
2270 | 		
2271 | 		return $this->hashBlock($text) . "\n";
2272 | 	}
2273 | 
2274 | 	
2275 | 	function doDefLists($text) {
2276 | 	#
2277 | 	# Form HTML definition lists.
2278 | 	#
2279 | 		$less_than_tab = $this->tab_width - 1;
2280 | 
2281 | 		# Re-usable pattern to match any entire dl list:
2282 | 		$whole_list_re = '(?>
2283 | 			(								# $1 = whole list
2284 | 			  (								# $2
2285 | 				[ ]{0,'.$less_than_tab.'}
2286 | 				((?>.*\S.*\n)+)				# $3 = defined term
2287 | 				\n?
2288 | 				[ ]{0,'.$less_than_tab.'}:[ ]+ # colon starting definition
2289 | 			  )
2290 | 			  (?s:.+?)
2291 | 			  (								# $4
2292 | 				  \z
2293 | 				|
2294 | 				  \n{2,}
2295 | 				  (?=\S)
2296 | 				  (?!						# Negative lookahead for another term
2297 | 					[ ]{0,'.$less_than_tab.'}
2298 | 					(?: \S.*\n )+?			# defined term
2299 | 					\n?
2300 | 					[ ]{0,'.$less_than_tab.'}:[ ]+ # colon starting definition
2301 | 				  )
2302 | 				  (?!						# Negative lookahead for another definition
2303 | 					[ ]{0,'.$less_than_tab.'}:[ ]+ # colon starting definition
2304 | 				  )
2305 | 			  )
2306 | 			)
2307 | 		)'; // mx
2308 | 
2309 | 		$text = preg_replace_callback('{
2310 | 				(?>\A\n?|(?<=\n\n))
2311 | 				'.$whole_list_re.'
2312 | 			}mx',
2313 | 			array(&$this, '_doDefLists_callback'), $text);
2314 | 
2315 | 		return $text;
2316 | 	}
2317 | 	function _doDefLists_callback($matches) {
2318 | 		# Re-usable patterns to match list item bullets and number markers:
2319 | 		$list = $matches[1];
2320 | 		
2321 | 		# Turn double returns into triple returns, so that we can make a
2322 | 		# paragraph for the last item in a list, if necessary:
2323 | 		$result = trim($this->processDefListItems($list));
2324 | 		$result = "\n" . $result . "\n";
2325 | 		return $this->hashBlock($result) . "\n\n";
2326 | 	}
2327 | 
2328 | 
2329 | 	function processDefListItems($list_str) {
2330 | 	#
2331 | 	#	Process the contents of a single definition list, splitting it
2332 | 	#	into individual term and definition list items.
2333 | 	#
2334 | 		$less_than_tab = $this->tab_width - 1;
2335 | 		
2336 | 		# trim trailing blank lines:
2337 | 		$list_str = preg_replace("/\n{2,}\\z/", "\n", $list_str);
2338 | 
2339 | 		# Process definition terms.
2340 | 		$list_str = preg_replace_callback('{
2341 | 			(?>\A\n?|\n\n+)					# leading line
2342 | 			(								# definition terms = $1
2343 | 				[ ]{0,'.$less_than_tab.'}	# leading whitespace
2344 | 				(?![:][ ]|[ ])				# negative lookahead for a definition 
2345 | 											#   mark (colon) or more whitespace.
2346 | 				(?> \S.* \n)+?				# actual term (not whitespace).	
2347 | 			)			
2348 | 			(?=\n?[ ]{0,3}:[ ])				# lookahead for following line feed 
2349 | 											#   with a definition mark.
2350 | 			}xm',
2351 | 			array(&$this, '_processDefListItems_callback_dt'), $list_str);
2352 | 
2353 | 		# Process actual definitions.
2354 | 		$list_str = preg_replace_callback('{
2355 | 			\n(\n+)?						# leading line = $1
2356 | 			(								# marker space = $2
2357 | 				[ ]{0,'.$less_than_tab.'}	# whitespace before colon
2358 | 				[:][ ]+						# definition mark (colon)
2359 | 			)
2360 | 			((?s:.+?))						# definition text = $3
2361 | 			(?= \n+ 						# stop at next definition mark,
2362 | 				(?:							# next term or end of text
2363 | 					[ ]{0,'.$less_than_tab.'} [:][ ]	|
2364 | 					 | \z
2365 | 				)						
2366 | 			)					
2367 | 			}xm',
2368 | 			array(&$this, '_processDefListItems_callback_dd'), $list_str);
2369 | 
2370 | 		return $list_str;
2371 | 	}
2372 | 	function _processDefListItems_callback_dt($matches) {
2373 | 		$terms = explode("\n", trim($matches[1]));
2374 | 		$text = '';
2375 | 		foreach ($terms as $term) {
2376 | 			$term = $this->runSpanGamut(trim($term));
2377 | 			$text .= "\n
" . $term . "";
2378 | 		}
2379 | 		return $text . "\n";
2380 | 	}
2381 | 	function _processDefListItems_callback_dd($matches) {
2382 | 		$leading_line	= $matches[1];
2383 | 		$marker_space	= $matches[2];
2384 | 		$def			= $matches[3];
2385 | 
2386 | 		if ($leading_line || preg_match('/\n{2,}/', $def)) {
2387 | 			# Replace marker with the appropriate whitespace indentation
2388 | 			$def = str_repeat(' ', strlen($marker_space)) . $def;
2389 | 			$def = $this->runBlockGamut($this->outdent($def . "\n\n"));
2390 | 			$def = "\n". $def ."\n";
2391 | 		}
2392 | 		else {
2393 | 			$def = rtrim($def);
2394 | 			$def = $this->runSpanGamut($this->outdent($def));
2395 | 		}
2396 | 
2397 | 		return "\n" . $def . "\n";
2398 | 	}
2399 | 
2400 | 
2401 | 	function doFencedCodeBlocks($text) {
2402 | 	#
2403 | 	# Adding the fenced code block syntax to regular Markdown:
2404 | 	#
2405 | 	# ~~~
2406 | 	# Code block
2407 | 	# ~~~
2408 | 	#
2409 | 		$less_than_tab = $this->tab_width;
2410 | 		
2411 | 		$text = preg_replace_callback('{
2412 | 				(?:\n|\A)
2413 | 				# 1: Opening marker
2414 | 				(
2415 | 					~{3,} # Marker: three tilde or more.
2416 | 				)
2417 | 				[ ]* \n # Whitespace and newline following marker.
2418 | 				
2419 | 				# 2: Content
2420 | 				(
2421 | 					(?>
2422 | 						(?!\1 [ ]* \n)	# Not a closing marker.
2423 | 						.*\n+
2424 | 					)+
2425 | 				)
2426 | 				
2427 | 				# Closing marker.
2428 | 				\1 [ ]* \n
2429 | 			}xm',
2430 | 			array(&$this, '_doFencedCodeBlocks_callback'), $text);
2431 | 
2432 | 		return $text;
2433 | 	}
2434 | 	function _doFencedCodeBlocks_callback($matches) {
2435 | 		$codeblock = $matches[2];
2436 | 		$codeblock = htmlspecialchars($codeblock, ENT_NOQUOTES);
2437 | 		$codeblock = preg_replace_callback('/^\n+/',
2438 | 			array(&$this, '_doFencedCodeBlocks_newlines'), $codeblock);
2439 | 		$codeblock = "$codeblock";
2440 | 		return "\n\n".$this->hashBlock($codeblock)."\n\n";
2441 | 	}
2442 | 	function _doFencedCodeBlocks_newlines($matches) {
2443 | 		return str_repeat("empty_element_suffix", 
2444 | 			strlen($matches[0]));
2445 | 	}
2446 | 
2447 | 
2448 | 	#
2449 | 	# Redefining emphasis markers so that emphasis by underscore does not
2450 | 	# work in the middle of a word.
2451 | 	#
2452 | 	var $em_relist = array(
2453 | 		''  => '(?:(? '(?<=\S|^)(? '(?<=\S|^)(? '(?:(? '(?<=\S|^)(? '(?<=\S|^)(? '(?:(? '(?<=\S|^)(? '(?<=\S|^)(? tags
2473 | 	#
2474 | 		# Strip leading and trailing lines:
2475 | 		$text = preg_replace('/\A\n+|\n+\z/', '', $text);
2476 | 		
2477 | 		$grafs = preg_split('/\n{2,}/', $text, -1, PREG_SPLIT_NO_EMPTY);
2478 | 
2479 | 		#
2480 | 		# Wrap  tags and unhashify HTML blocks
2481 | 		#
2482 | 		foreach ($grafs as $key => $value) {
2483 | 			$value = trim($this->runSpanGamut($value));
2484 | 			
2485 | 			# Check if this should be enclosed in a paragraph.
2486 | 			# Clean tag hashes & block tag hashes are left alone.
2487 | 			$is_p = !preg_match('/^B\x1A[0-9]+B|^C\x1A[0-9]+C$/', $value);
2488 | 			
2489 | 			if ($is_p) {
2490 | 				$value = "
$value";
2491 | 			}
2492 | 			$grafs[$key] = $value;
2493 | 		}
2494 | 		
2495 | 		# Join grafs in one text, then unhash HTML tags. 
2496 | 		$text = implode("\n\n", $grafs);
2497 | 		
2498 | 		# Finish by removing any tag hashes still present in $text.
2499 | 		$text = $this->unhash($text);
2500 | 		
2501 | 		return $text;
2502 | 	}
2503 | 	
2504 | 	
2505 | 	### Footnotes
2506 | 	
2507 | 	function stripFootnotes($text) {
2508 | 	#
2509 | 	# Strips link definitions from text, stores the URLs and titles in
2510 | 	# hash references.
2511 | 	#
2512 | 		$less_than_tab = $this->tab_width - 1;
2513 | 
2514 | 		# Link defs are in the form: [^id]: url "optional title"
2515 | 		$text = preg_replace_callback('{
2516 | 			^[ ]{0,'.$less_than_tab.'}\[\^(.+?)\][ ]?:	# note_id = $1
2517 | 			  [ ]*
2518 | 			  \n?					# maybe *one* newline
2519 | 			(						# text = $2 (no blank lines allowed)
2520 | 				(?:					
2521 | 					.+				# actual text
2522 | 				|
2523 | 					\n				# newlines but 
2524 | 					(?!\[\^.+?\]:\s)# negative lookahead for footnote marker.
2525 | 					(?!\n+[ ]{0,3}\S)# ensure line is not blank and followed 
2526 | 									# by non-indented content
2527 | 				)*
2528 | 			)		
2529 | 			}xm',
2530 | 			array(&$this, '_stripFootnotes_callback'),
2531 | 			$text);
2532 | 		return $text;
2533 | 	}
2534 | 	function _stripFootnotes_callback($matches) {
2535 | 		$note_id = $this->fn_id_prefix . $matches[1];
2536 | 		$this->footnotes[$note_id] = $this->outdent($matches[2]);
2537 | 		return ''; # String that will replace the block
2538 | 	}
2539 | 
2540 | 
2541 | 	function doFootnotes($text) {
2542 | 	#
2543 | 	# Replace footnote references in $text [^id] with a special text-token 
2544 | 	# which will be replaced by the actual footnote marker in appendFootnotes.
2545 | 	#
2546 | 		if (!$this->in_anchor) {
2547 | 			$text = preg_replace('{\[\^(.+?)\]}', "F\x1Afn:\\1\x1A:", $text);
2548 | 		}
2549 | 		return $text;
2550 | 	}
2551 | 
2552 | 	
2553 | 	function appendFootnotes($text) {
2554 | 	#
2555 | 	# Append footnote list to text.
2556 | 	#
2557 | 		$text = preg_replace_callback('{F\x1Afn:(.*?)\x1A:}', 
2558 | 			array(&$this, '_appendFootnotes_callback'), $text);
2559 | 	
2560 | 		if (!empty($this->footnotes_ordered)) {
2561 | 			$text .= "\n\n";
2562 | 			$text .= "\n";
2563 | 			$text .= "empty_element_suffix ."\n";
2564 | 			$text .= "\n\n";
2565 | 			
2566 | 			$attr = " rev=\"footnote\"";
2567 | 			if ($this->fn_backlink_class != "") {
2568 | 				$class = $this->fn_backlink_class;
2569 | 				$class = $this->encodeAttribute($class);
2570 | 				$attr .= " class=\"$class\"";
2571 | 			}
2572 | 			if ($this->fn_backlink_title != "") {
2573 | 				$title = $this->fn_backlink_title;
2574 | 				$title = $this->encodeAttribute($title);
2575 | 				$attr .= " title=\"$title\"";
2576 | 			}
2577 | 			$num = 0;
2578 | 			
2579 | 			while (!empty($this->footnotes_ordered)) {
2580 | 				$footnote = reset($this->footnotes_ordered);
2581 | 				$note_id = key($this->footnotes_ordered);
2582 | 				unset($this->footnotes_ordered[$note_id]);
2583 | 				
2584 | 				$footnote .= "\n"; # Need to append newline before parsing.
2585 | 				$footnote = $this->runBlockGamut("$footnote\n");				
2586 | 				$footnote = preg_replace_callback('{F\x1Afn:(.*?)\x1A:}', 
2587 | 					array(&$this, '_appendFootnotes_callback'), $footnote);
2588 | 				
2589 | 				$attr = str_replace("%%", ++$num, $attr);
2590 | 				$note_id = $this->encodeAttribute($note_id);
2591 | 				
2592 | 				# Add backlink to last paragraph; create new paragraph if needed.
2593 | 				$backlink = "↩";
2594 | 				if (preg_match('{$}', $footnote)) {
2595 | 					$footnote = substr($footnote, 0, -4) . " $backlink";
2596 | 				} else {
2597 | 					$footnote .= "\n\n$backlink";
2598 | 				}
2599 | 				
2600 | 				$text .= "\n";
2601 | 				$text .= $footnote . "\n";
2602 | 				$text .= "\n\n";
2603 | 			}
2604 | 			
2605 | 			$text .= "\n";
2606 | 			$text .= "";
2607 | 		}
2608 | 		return $text;
2609 | 	}
2610 | 	function _appendFootnotes_callback($matches) {
2611 | 		$node_id = $this->fn_id_prefix . $matches[1];
2612 | 		
2613 | 		# Create footnote marker only if it has a corresponding footnote *and*
2614 | 		# the footnote hasn't been used by another marker.
2615 | 		if (isset($this->footnotes[$node_id])) {
2616 | 			# Transfert footnote content to the ordered list.
2617 | 			$this->footnotes_ordered[$node_id] = $this->footnotes[$node_id];
2618 | 			unset($this->footnotes[$node_id]);
2619 | 			
2620 | 			$num = $this->footnote_counter++;
2621 | 			$attr = " rel=\"footnote\"";
2622 | 			if ($this->fn_link_class != "") {
2623 | 				$class = $this->fn_link_class;
2624 | 				$class = $this->encodeAttribute($class);
2625 | 				$attr .= " class=\"$class\"";
2626 | 			}
2627 | 			if ($this->fn_link_title != "") {
2628 | 				$title = $this->fn_link_title;
2629 | 				$title = $this->encodeAttribute($title);
2630 | 				$attr .= " title=\"$title\"";
2631 | 			}
2632 | 			
2633 | 			$attr = str_replace("%%", $num, $attr);
2634 | 			$node_id = $this->encodeAttribute($node_id);
2635 | 			
2636 | 			return
2637 | 				"^{".
2638 | 				"$num".
2639 | 				"}";
2640 | 		}
2641 | 		
2642 | 		return "[^".$matches[1]."]";
2643 | 	}
2644 | 		
2645 | 	
2646 | 	### Abbreviations ###
2647 | 	
2648 | 	function stripAbbreviations($text) {
2649 | 	#
2650 | 	# Strips abbreviations from text, stores titles in hash references.
2651 | 	#
2652 | 		$less_than_tab = $this->tab_width - 1;
2653 | 
2654 | 		# Link defs are in the form: [id]*: url "optional title"
2655 | 		$text = preg_replace_callback('{
2656 | 			^[ ]{0,'.$less_than_tab.'}\*\[(.+?)\][ ]?:	# abbr_id = $1
2657 | 			(.*)					# text = $2 (no blank lines allowed)	
2658 | 			}xm',
2659 | 			array(&$this, '_stripAbbreviations_callback'),
2660 | 			$text);
2661 | 		return $text;
2662 | 	}
2663 | 	function _stripAbbreviations_callback($matches) {
2664 | 		$abbr_word = $matches[1];
2665 | 		$abbr_desc = $matches[2];
2666 | 		if ($this->abbr_word_re)
2667 | 			$this->abbr_word_re .= '|';
2668 | 		$this->abbr_word_re .= preg_quote($abbr_word);
2669 | 		$this->abbr_desciptions[$abbr_word] = trim($abbr_desc);
2670 | 		return ''; # String that will replace the block
2671 | 	}
2672 | 	
2673 | 	
2674 | 	function doAbbreviations($text) {
2675 | 	#
2676 | 	# Find defined abbreviations in text and wrap them in  elements.
2677 | 	#
2678 | 		if ($this->abbr_word_re) {
2679 | 			// cannot use the /x modifier because abbr_word_re may 
2680 | 			// contain significant spaces:
2681 | 			$text = preg_replace_callback('{'.
2682 | 				'(?abbr_word_re.')'.
2684 | 				'(?![\w\x1A])'.
2685 | 				'}', 
2686 | 				array(&$this, '_doAbbreviations_callback'), $text);
2687 | 		}
2688 | 		return $text;
2689 | 	}
2690 | 	function _doAbbreviations_callback($matches) {
2691 | 		$abbr = $matches[0];
2692 | 		if (isset($this->abbr_desciptions[$abbr])) {
2693 | 			$desc = $this->abbr_desciptions[$abbr];
2694 | 			if (empty($desc)) {
2695 | 				return $this->hashPart("$abbr");
2696 | 			} else {
2697 | 				$desc = $this->encodeAttribute($desc);
2698 | 				return $this->hashPart("$abbr");
2699 | 			}
2700 | 		} else {
2701 | 			return $matches[0];
2702 | 		}
2703 | 	}
2704 | 
2705 | }
2706 | 
2707 | 
2708 | /*
2709 | 
2710 | PHP Markdown Extra
2711 | ==================
2712 | 
2713 | Description
2714 | -----------
2715 | 
2716 | This is a PHP port of the original Markdown formatter written in Perl 
2717 | by John Gruber. This special "Extra" version of PHP Markdown features 
2718 | further enhancements to the syntax for making additional constructs 
2719 | such as tables and definition list.
2720 | 
2721 | Markdown is a text-to-HTML filter; it translates an easy-to-read /
2722 | easy-to-write structured text format into HTML. Markdown's text format
2723 | is most similar to that of plain text email, and supports features such
2724 | as headers, *emphasis*, code blocks, blockquotes, and links.
2725 | 
2726 | Markdown's syntax is designed not as a generic markup language, but
2727 | specifically to serve as a front-end to (X)HTML. You can use span-level
2728 | HTML tags anywhere in a Markdown document, and you can use block level
2729 | HTML tags (like  and  as well).
2730 | 
2731 | For more information about Markdown's syntax, see:
2732 | 
2733 | 
2734 | 
2735 | 
2736 | Bugs
2737 | ----
2738 | 
2739 | To file bug reports please send email to:
2740 | 
2741 | 
2742 | 
2743 | Please include with your report: (1) the example input; (2) the output you
2744 | expected; (3) the output Markdown actually produced.
2745 | 
2746 | 
2747 | Version History
2748 | --------------- 
2749 | 
2750 | See the readme file for detailed release notes for this version.
2751 | 
2752 | 
2753 | Copyright and License
2754 | ---------------------
2755 | 
2756 | PHP Markdown & Extra  
2757 | Copyright (c) 2004-2009 Michel Fortin  
2758 |   
2759 | All rights reserved.
2760 | 
2761 | Based on Markdown  
2762 | Copyright (c) 2003-2006 John Gruber   
2763 |    
2764 | All rights reserved.
2765 | 
2766 | Redistribution and use in source and binary forms, with or without
2767 | modification, are permitted provided that the following conditions are
2768 | met:
2769 | 
2770 | *	Redistributions of source code must retain the above copyright notice,
2771 | 	this list of conditions and the following disclaimer.
2772 | 
2773 | *	Redistributions in binary form must reproduce the above copyright
2774 | 	notice, this list of conditions and the following disclaimer in the
2775 | 	documentation and/or other materials provided with the distribution.
2776 | 
2777 | *	Neither the name "Markdown" nor the names of its contributors may
2778 | 	be used to endorse or promote products derived from this software
2779 | 	without specific prior written permission.
2780 | 
2781 | This software is provided by the copyright holders and contributors "as
2782 | is" and any express or implied warranties, including, but not limited
2783 | to, the implied warranties of merchantability and fitness for a
2784 | particular purpose are disclaimed. In no event shall the copyright owner
2785 | or contributors be liable for any direct, indirect, incidental, special,
2786 | exemplary, or consequential damages (including, but not limited to,
2787 | procurement of substitute goods or services; loss of use, data, or
2788 | profits; or business interruption) however caused and on any theory of
2789 | liability, whether in contract, strict liability, or tort (including
2790 | negligence or otherwise) arising in any way out of the use of this
2791 | software, even if advised of the possibility of such damage.
2792 | 
2793 | */
2794 | ?>


--------------------------------------------------------------------------------

Found plugins

Activated plugins

Register action hooks

Previously run action hooks