| 1 |
<?php |
| 2 |
# |
| 3 |
# Markdown Extra - A text-to-HTML conversion tool for web writers |
| 4 |
# |
| 5 |
# PHP Markdown & Extra |
| 6 |
# Copyright (c) 2004-2013 Michel Fortin |
| 7 |
# <http://michelf.ca/projects/php-markdown/> |
| 8 |
# |
| 9 |
# Original Markdown |
| 10 |
# Copyright (c) 2004-2006 John Gruber |
| 11 |
# <http://daringfireball.net/projects/markdown/> |
| 12 |
# |
| 13 |
# Tweaked to remove WordPress interface |
| 14 |
|
| 15 |
|
| 16 |
define( 'MARKDOWN_VERSION', "1.0.2" ); # 29 Nov 2013 |
| 17 |
define( 'MARKDOWNEXTRA_VERSION', "1.2.8" ); # 29 Nov 2013 |
| 18 |
|
| 19 |
|
| 20 |
# |
| 21 |
# Global default settings: |
| 22 |
# |
| 23 |
|
| 24 |
# Change to ">" for HTML output |
| 25 |
@define( 'MARKDOWN_EMPTY_ELEMENT_SUFFIX', " />"); |
| 26 |
|
| 27 |
# Define the width of a tab for code blocks. |
| 28 |
@define( 'MARKDOWN_TAB_WIDTH', 4 ); |
| 29 |
|
| 30 |
# Optional title attribute for footnote links and backlinks. |
| 31 |
@define( 'MARKDOWN_FN_LINK_TITLE', "" ); |
| 32 |
@define( 'MARKDOWN_FN_BACKLINK_TITLE', "" ); |
| 33 |
|
| 34 |
# Optional class attribute for footnote links and backlinks. |
| 35 |
@define( 'MARKDOWN_FN_LINK_CLASS', "jetpack-footnote" ); |
| 36 |
@define( 'MARKDOWN_FN_BACKLINK_CLASS', "" ); |
| 37 |
|
| 38 |
# Optional class prefix for fenced code block. |
| 39 |
@define( 'MARKDOWN_CODE_CLASS_PREFIX', "" ); |
| 40 |
|
| 41 |
# Class attribute for code blocks goes on the `code` tag; |
| 42 |
# setting this to true will put attributes on the `pre` tag instead. |
| 43 |
@define( 'MARKDOWN_CODE_ATTR_ON_PRE', false ); |
| 44 |
|
| 45 |
|
| 46 |
|
| 47 |
### Standard Function Interface ### |
| 48 |
|
| 49 |
@define( 'MARKDOWN_PARSER_CLASS', 'MarkdownExtra_Parser' ); |
| 50 |
|
| 51 |
function Markdown($text) { |
| 52 |
# |
| 53 |
# Initialize the parser and return the result of its transform method. |
| 54 |
# |
| 55 |
# Setup static parser variable. |
| 56 |
static $parser; |
| 57 |
if (!isset($parser)) { |
| 58 |
$parser_class = MARKDOWN_PARSER_CLASS; |
| 59 |
$parser = new $parser_class; |
| 60 |
} |
| 61 |
|
| 62 |
# Transform text using parser. |
| 63 |
return $parser->transform($text); |
| 64 |
} |
| 65 |
|
| 66 |
/** |
| 67 |
* Returns the length of $text loosely counting the number of UTF-8 characters with regular expression. |
| 68 |
* Used by the Markdown_Parser class when mb_strlen is not available. |
| 69 |
* |
| 70 |
* @since 5.9 |
| 71 |
* |
| 72 |
* @return string Length of the multibyte string |
| 73 |
* |
| 74 |
*/ |
| 75 |
function jetpack_utf8_strlen( $text ) { |
| 76 |
return preg_match_all( "/[\\x00-\\xBF]|[\\xC0-\\xFF][\\x80-\\xBF]*/", $text, $m ); |
| 77 |
} |
| 78 |
|
| 79 |
# |
| 80 |
# Markdown Parser Class |
| 81 |
# |
| 82 |
|
| 83 |
class Markdown_Parser { |
| 84 |
|
| 85 |
### Configuration Variables ### |
| 86 |
|
| 87 |
# Change to ">" for HTML output. |
| 88 |
public $empty_element_suffix = MARKDOWN_EMPTY_ELEMENT_SUFFIX; |
| 89 |
public $tab_width = MARKDOWN_TAB_WIDTH; |
| 90 |
|
| 91 |
# Change to `true` to disallow markup or entities. |
| 92 |
public $no_markup = false; |
| 93 |
public $no_entities = false; |
| 94 |
|
| 95 |
# Predefined urls and titles for reference links and images. |
| 96 |
public $predef_urls = array(); |
| 97 |
public $predef_titles = array(); |
| 98 |
|
| 99 |
|
| 100 |
### Parser Implementation ### |
| 101 |
|
| 102 |
# Regex to match balanced [brackets]. |
| 103 |
# Needed to insert a maximum bracked depth while converting to PHP. |
| 104 |
public $nested_brackets_depth = 6; |
| 105 |
public $nested_brackets_re; |
| 106 |
|
| 107 |
public $nested_url_parenthesis_depth = 4; |
| 108 |
public $nested_url_parenthesis_re; |
| 109 |
|
| 110 |
# Table of hash values for escaped characters: |
| 111 |
public $escape_chars = '\`*_{}[]()>#+-.!'; |
| 112 |
public $escape_chars_re; |
| 113 |
|
| 114 |
|
| 115 |
function __construct() { |
| 116 |
# |
| 117 |
# Constructor function. Initialize appropriate member variables. |
| 118 |
# |
| 119 |
$this->_initDetab(); |
| 120 |
$this->prepareItalicsAndBold(); |
| 121 |
|
| 122 |
$this->nested_brackets_re = |
| 123 |
str_repeat('(?>[^\[\]]+|\[', $this->nested_brackets_depth). |
| 124 |
str_repeat('\])*', $this->nested_brackets_depth); |
| 125 |
|
| 126 |
$this->nested_url_parenthesis_re = |
| 127 |
str_repeat('(?>[^()\s]+|\(', $this->nested_url_parenthesis_depth). |
| 128 |
str_repeat('(?>\)))*', $this->nested_url_parenthesis_depth); |
| 129 |
|
| 130 |
$this->escape_chars_re = '['.preg_quote($this->escape_chars).']'; |
| 131 |
|
| 132 |
# Sort document, block, and span gamut in ascendent priority order. |
| 133 |
asort($this->document_gamut); |
| 134 |
asort($this->block_gamut); |
| 135 |
asort($this->span_gamut); |
| 136 |
} |
| 137 |
|
| 138 |
|
| 139 |
# Internal hashes used during transformation. |
| 140 |
public $urls = array(); |
| 141 |
public $titles = array(); |
| 142 |
public $html_hashes = array(); |
| 143 |
|
| 144 |
# Status flag to avoid invalid nesting. |
| 145 |
public $in_anchor = false; |
| 146 |
|
| 147 |
|
| 148 |
function setup() { |
| 149 |
# |
| 150 |
# Called before the transformation process starts to setup parser |
| 151 |
# states. |
| 152 |
# |
| 153 |
# Clear global hashes. |
| 154 |
$this->urls = $this->predef_urls; |
| 155 |
$this->titles = $this->predef_titles; |
| 156 |
$this->html_hashes = array(); |
| 157 |
|
| 158 |
$this->in_anchor = false; |
| 159 |
} |
| 160 |
|
| 161 |
function teardown() { |
| 162 |
# |
| 163 |
# Called after the transformation process to clear any variable |
| 164 |
# which may be taking up memory unnecessarly. |
| 165 |
# |
| 166 |
$this->urls = array(); |
| 167 |
$this->titles = array(); |
| 168 |
$this->html_hashes = array(); |
| 169 |
} |
| 170 |
|
| 171 |
|
| 172 |
function transform($text) { |
| 173 |
# |
| 174 |
# Main function. Performs some preprocessing on the input text |
| 175 |
# and pass it through the document gamut. |
| 176 |
# |
| 177 |
$this->setup(); |
| 178 |
|
| 179 |
# Remove UTF-8 BOM and marker character in input, if present. |
| 180 |
$text = preg_replace('{^\xEF\xBB\xBF|\x1A}', '', $text); |
| 181 |
|
| 182 |
# Standardize line endings: |
| 183 |
# DOS to Unix and Mac to Unix |
| 184 |
$text = preg_replace('{\r\n?}', "\n", $text); |
| 185 |
|
| 186 |
# Make sure $text ends with a couple of newlines: |
| 187 |
$text .= "\n\n"; |
| 188 |
|
| 189 |
# Convert all tabs to spaces. |
| 190 |
$text = $this->detab($text); |
| 191 |
|
| 192 |
# Turn block-level HTML blocks into hash entries |
| 193 |
$text = $this->hashHTMLBlocks($text); |
| 194 |
|
| 195 |
# Strip any lines consisting only of spaces and tabs. |
| 196 |
# This makes subsequent regexen easier to write, because we can |
| 197 |
# match consecutive blank lines with /\n+/ instead of something |
| 198 |
# contorted like /[ ]*\n+/ . |
| 199 |
$text = preg_replace('/^[ ]+$/m', '', $text); |
| 200 |
|
| 201 |
# Run document gamut methods. |
| 202 |
foreach ($this->document_gamut as $method => $priority) { |
| 203 |
$text = $this->$method($text); |
| 204 |
} |
| 205 |
|
| 206 |
$this->teardown(); |
| 207 |
|
| 208 |
return $text . "\n"; |
| 209 |
} |
| 210 |
|
| 211 |
public $document_gamut = array( |
| 212 |
# Strip link definitions, store in hashes. |
| 213 |
"stripLinkDefinitions" => 20, |
| 214 |
|
| 215 |
"runBasicBlockGamut" => 30, |
| 216 |
); |
| 217 |
|
| 218 |
|
| 219 |
function stripLinkDefinitions($text) { |
| 220 |
# |
| 221 |
# Strips link definitions from text, stores the URLs and titles in |
| 222 |
# hash references. |
| 223 |
# |
| 224 |
$less_than_tab = $this->tab_width - 1; |
| 225 |
|
| 226 |
# Link defs are in the form: ^[id]: url "optional title" |
| 227 |
$text = preg_replace_callback('{ |
| 228 |
^[ ]{0,'.$less_than_tab.'}\[(.+)\][ ]?: # id = $1 |
| 229 |
[ ]* |
| 230 |
\n? # maybe *one* newline |
| 231 |
[ ]* |
| 232 |
(?: |
| 233 |
<(.+?)> # url = $2 |
| 234 |
| |
| 235 |
(\S+?) # url = $3 |
| 236 |
) |
| 237 |
[ ]* |
| 238 |
\n? # maybe one newline |
| 239 |
[ ]* |
| 240 |
(?: |
| 241 |
(?<=\s) # lookbehind for whitespace |
| 242 |
["(] |
| 243 |
(.*?) # title = $4 |
| 244 |
[")] |
| 245 |
[ ]* |
| 246 |
)? # title is optional |
| 247 |
(?:\n+|\Z) |
| 248 |
}xm', |
| 249 |
array(&$this, '_stripLinkDefinitions_callback'), |
| 250 |
$text); |
| 251 |
return $text; |
| 252 |
} |
| 253 |
function _stripLinkDefinitions_callback($matches) { |
| 254 |
$link_id = strtolower($matches[1]); |
| 255 |
$url = $matches[2] == '' ? $matches[3] : $matches[2]; |
| 256 |
$this->urls[$link_id] = $url; |
| 257 |
$this->titles[$link_id] =& $matches[4]; |
| 258 |
return ''; # String that will replace the block |
| 259 |
} |
| 260 |
|
| 261 |
|
| 262 |
function hashHTMLBlocks($text) { |
| 263 |
if ($this->no_markup) return $text; |
| 264 |
|
| 265 |
$less_than_tab = $this->tab_width - 1; |
| 266 |
|
| 267 |
# Hashify HTML blocks: |
| 268 |
# We only want to do this for block-level HTML tags, such as headers, |
| 269 |
# lists, and tables. That's because we still want to wrap <p>s around |
| 270 |
# "paragraphs" that are wrapped in non-block-level tags, such as anchors, |
| 271 |
# phrase emphasis, and spans. The list of tags we're looking for is |
| 272 |
# hard-coded: |
| 273 |
# |
| 274 |
# * List "a" is made of tags which can be both inline or block-level. |
| 275 |
# These will be treated block-level when the start tag is alone on |
| 276 |
# its line, otherwise they're not matched here and will be taken as |
| 277 |
# inline later. |
| 278 |
# * List "b" is made of tags which are always block-level; |
| 279 |
# |
| 280 |
$block_tags_a_re = 'ins|del'; |
| 281 |
$block_tags_b_re = 'p|div|h[1-6]|blockquote|pre|table|dl|ol|ul|address|'. |
| 282 |
'script|noscript|form|fieldset|iframe|math|svg|'. |
| 283 |
'article|section|nav|aside|hgroup|header|footer|'. |
| 284 |
'figure'; |
| 285 |
|
| 286 |
# Regular expression for the content of a block tag. |
| 287 |
$nested_tags_level = 4; |
| 288 |
$attr = ' |
| 289 |
(?> # optional tag attributes |
| 290 |
\s # starts with whitespace |
| 291 |
(?> |
| 292 |
[^>"/]+ # text outside quotes |
| 293 |
| |
| 294 |
/+(?!>) # slash not followed by ">" |
| 295 |
| |
| 296 |
"[^"]*" # text inside double quotes (tolerate ">") |
| 297 |
| |
| 298 |
\'[^\']*\' # text inside single quotes (tolerate ">") |
| 299 |
)* |
| 300 |
)? |
| 301 |
'; |
| 302 |
$content = |
| 303 |
str_repeat(' |
| 304 |
(?> |
| 305 |
[^<]+ # content without tag |
| 306 |
| |
| 307 |
<\2 # nested opening tag |
| 308 |
'.$attr.' # attributes |
| 309 |
(?> |
| 310 |
/> |
| 311 |
| |
| 312 |
>', $nested_tags_level). # end of opening tag |
| 313 |
'.*?'. # last level nested tag content |
| 314 |
str_repeat(' |
| 315 |
</\2\s*> # closing nested tag |
| 316 |
) |
| 317 |
| |
| 318 |
<(?!/\2\s*> # other tags with a different name |
| 319 |
) |
| 320 |
)*', |
| 321 |
$nested_tags_level); |
| 322 |
$content2 = str_replace('\2', '\3', $content); |
| 323 |
|
| 324 |
# First, look for nested blocks, e.g.: |
| 325 |
# <div> |
| 326 |
# <div> |
| 327 |
# tags for inner block must be indented. |
| 328 |
# </div> |
| 329 |
# </div> |
| 330 |
# |
| 331 |
# The outermost tags must start at the left margin for this to match, and |
| 332 |
# the inner nested divs must be indented. |
| 333 |
# We need to do this before the next, more liberal match, because the next |
| 334 |
# match will start at the first `<div>` and stop at the first `</div>`. |
| 335 |
$text = preg_replace_callback('{(?> |
| 336 |
(?> |
| 337 |
(?<=\n\n) # Starting after a blank line |
| 338 |
| # or |
| 339 |
\A\n? # the beginning of the doc |
| 340 |
) |
| 341 |
( # save in $1 |
| 342 |
|
| 343 |
# Match from `\n<tag>` to `</tag>\n`, handling nested tags |
| 344 |
# in between. |
| 345 |
|
| 346 |
[ ]{0,'.$less_than_tab.'} |
| 347 |
<('.$block_tags_b_re.')# start tag = $2 |
| 348 |
'.$attr.'> # attributes followed by > and \n |
| 349 |
'.$content.' # content, support nesting |
| 350 |
</\2> # the matching end tag |
| 351 |
[ ]* # trailing spaces/tabs |
| 352 |
(?=\n+|\Z) # followed by a newline or end of document |
| 353 |
|
| 354 |
| # Special version for tags of group a. |
| 355 |
|
| 356 |
[ ]{0,'.$less_than_tab.'} |
| 357 |
<('.$block_tags_a_re.')# start tag = $3 |
| 358 |
'.$attr.'>[ ]*\n # attributes followed by > |
| 359 |
'.$content2.' # content, support nesting |
| 360 |
</\3> # the matching end tag |
| 361 |
[ ]* # trailing spaces/tabs |
| 362 |
(?=\n+|\Z) # followed by a newline or end of document |
| 363 |
|
| 364 |
| # Special case just for <hr />. It was easier to make a special |
| 365 |
# case than to make the other regex more complicated. |
| 366 |
|
| 367 |
[ ]{0,'.$less_than_tab.'} |
| 368 |
<(hr) # start tag = $2 |
| 369 |
'.$attr.' # attributes |
| 370 |
/?> # the matching end tag |
| 371 |
[ ]* |
| 372 |
(?=\n{2,}|\Z) # followed by a blank line or end of document |
| 373 |
|
| 374 |
| # Special case for standalone HTML comments: |
| 375 |
|
| 376 |
[ ]{0,'.$less_than_tab.'} |
| 377 |
(?s: |
| 378 |
<!-- .*? --> |
| 379 |
) |
| 380 |
[ ]* |
| 381 |
(?=\n{2,}|\Z) # followed by a blank line or end of document |
| 382 |
|
| 383 |
| # PHP and ASP-style processor instructions (<? and <%) |
| 384 |
|
| 385 |
[ ]{0,'.$less_than_tab.'} |
| 386 |
(?s: |
| 387 |
<([?%]) # $2 |
| 388 |
.*? |
| 389 |
\2> |
| 390 |
) |
| 391 |
[ ]* |
| 392 |
(?=\n{2,}|\Z) # followed by a blank line or end of document |
| 393 |
|
| 394 |
) |
| 395 |
)}Sxmi', |
| 396 |
array(&$this, '_hashHTMLBlocks_callback'), |
| 397 |
$text); |
| 398 |
|
| 399 |
return $text; |
| 400 |
} |
| 401 |
function _hashHTMLBlocks_callback($matches) { |
| 402 |
$text = $matches[1]; |
| 403 |
$key = $this->hashBlock($text); |
| 404 |
return "\n\n$key\n\n"; |
| 405 |
} |
| 406 |
|
| 407 |
|
| 408 |
function hashPart($text, $boundary = 'X') { |
| 409 |
# |
| 410 |
# Called whenever a tag must be hashed when a function insert an atomic |
| 411 |
# element in the text stream. Passing $text to through this function gives |
| 412 |
# a unique text-token which will be reverted back when calling unhash. |
| 413 |
# |
| 414 |
# The $boundary argument specify what character should be used to surround |
| 415 |
# the token. By convension, "B" is used for block elements that needs not |
| 416 |
# to be wrapped into paragraph tags at the end, ":" is used for elements |
| 417 |
# that are word separators and "X" is used in the general case. |
| 418 |
# |
| 419 |
# Swap back any tag hash found in $text so we do not have to `unhash` |
| 420 |
# multiple times at the end. |
| 421 |
$text = $this->unhash($text); |
| 422 |
|
| 423 |
# Then hash the block. |
| 424 |
static $i = 0; |
| 425 |
$key = "$boundary\x1A" . ++$i . $boundary; |
| 426 |
$this->html_hashes[$key] = $text; |
| 427 |
return $key; # String that will replace the tag. |
| 428 |
} |
| 429 |
|
| 430 |
|
| 431 |
function hashBlock($text) { |
| 432 |
# |
| 433 |
# Shortcut function for hashPart with block-level boundaries. |
| 434 |
# |
| 435 |
return $this->hashPart($text, 'B'); |
| 436 |
} |
| 437 |
|
| 438 |
|
| 439 |
public $block_gamut = array( |
| 440 |
# |
| 441 |
# These are all the transformations that form block-level |
| 442 |
# tags like paragraphs, headers, and list items. |
| 443 |
# |
| 444 |
"doHeaders" => 10, |
| 445 |
"doHorizontalRules" => 20, |
| 446 |
|
| 447 |
"doLists" => 40, |
| 448 |
"doCodeBlocks" => 50, |
| 449 |
"doBlockQuotes" => 60, |
| 450 |
); |
| 451 |
|
| 452 |
function runBlockGamut($text) { |
| 453 |
# |
| 454 |
# Run block gamut tranformations. |
| 455 |
# |
| 456 |
# We need to escape raw HTML in Markdown source before doing anything |
| 457 |
# else. This need to be done for each block, and not only at the |
| 458 |
# beginning in the Markdown function since hashed blocks can be part of |
| 459 |
# list items and could have been indented. Indented blocks would have |
| 460 |
# been seen as a code block in a previous pass of hashHTMLBlocks. |
| 461 |
$text = $this->hashHTMLBlocks($text); |
| 462 |
|
| 463 |
return $this->runBasicBlockGamut($text); |
| 464 |
} |
| 465 |
|
| 466 |
function runBasicBlockGamut($text) { |
| 467 |
# |
| 468 |
# Run block gamut tranformations, without hashing HTML blocks. This is |
| 469 |
# useful when HTML blocks are known to be already hashed, like in the first |
| 470 |
# whole-document pass. |
| 471 |
# |
| 472 |
foreach ($this->block_gamut as $method => $priority) { |
| 473 |
$text = $this->$method($text); |
| 474 |
} |
| 475 |
|
| 476 |
# Finally form paragraph and restore hashed blocks. |
| 477 |
$text = $this->formParagraphs($text); |
| 478 |
|
| 479 |
return $text; |
| 480 |
} |
| 481 |
|
| 482 |
|
| 483 |
function doHorizontalRules($text) { |
| 484 |
# Do Horizontal Rules: |
| 485 |
return preg_replace( |
| 486 |
'{ |
| 487 |
^[ ]{0,3} # Leading space |
| 488 |
([-*_]) # $1: First marker |
| 489 |
(?> # Repeated marker group |
| 490 |
[ ]{0,2} # Zero, one, or two spaces. |
| 491 |
\1 # Marker character |
| 492 |
){2,} # Group repeated at least twice |
| 493 |
[ ]* # Tailing spaces |
| 494 |
$ # End of line. |
| 495 |
}mx', |
| 496 |
"\n".$this->hashBlock("<hr$this->empty_element_suffix")."\n", |
| 497 |
$text); |
| 498 |
} |
| 499 |
|
| 500 |
|
| 501 |
public $span_gamut = array( |
| 502 |
# |
| 503 |
# These are all the transformations that occur *within* block-level |
| 504 |
# tags like paragraphs, headers, and list items. |
| 505 |
# |
| 506 |
# Process character escapes, code spans, and inline HTML |
| 507 |
# in one shot. |
| 508 |
"parseSpan" => -30, |
| 509 |
|
| 510 |
# Process anchor and image tags. Images must come first, |
| 511 |
# because ![foo][f] looks like an anchor. |
| 512 |
"doImages" => 10, |
| 513 |
"doAnchors" => 20, |
| 514 |
|
| 515 |
# Make links out of things like `<http://example.com/>` |
| 516 |
# Must come after doAnchors, because you can use < and > |
| 517 |
# delimiters in inline links like [this](<url>). |
| 518 |
"doAutoLinks" => 30, |
| 519 |
"encodeAmpsAndAngles" => 40, |
| 520 |
|
| 521 |
"doItalicsAndBold" => 50, |
| 522 |
"doHardBreaks" => 60, |
| 523 |
); |
| 524 |
|
| 525 |
function runSpanGamut($text) { |
| 526 |
# |
| 527 |
# Run span gamut tranformations. |
| 528 |
# |
| 529 |
foreach ($this->span_gamut as $method => $priority) { |
| 530 |
$text = $this->$method($text); |
| 531 |
} |
| 532 |
|
| 533 |
return $text; |
| 534 |
} |
| 535 |
|
| 536 |
|
| 537 |
function doHardBreaks($text) { |
| 538 |
# Do hard breaks: |
| 539 |
return preg_replace_callback('/ {2,}\n/', |
| 540 |
array(&$this, '_doHardBreaks_callback'), $text); |
| 541 |
} |
| 542 |
function _doHardBreaks_callback($matches) { |
| 543 |
return $this->hashPart("<br$this->empty_element_suffix\n"); |
| 544 |
} |
| 545 |
|
| 546 |
|
| 547 |
function doAnchors($text) { |
| 548 |
# |
| 549 |
# Turn Markdown link shortcuts into XHTML <a> tags. |
| 550 |
# |
| 551 |
if ($this->in_anchor) return $text; |
| 552 |
$this->in_anchor = true; |
| 553 |
|
| 554 |
# |
| 555 |
# First, handle reference-style links: [link text] [id] |
| 556 |
# |
| 557 |
$text = preg_replace_callback('{ |
| 558 |
( # wrap whole match in $1 |
| 559 |
\[ |
| 560 |
('.$this->nested_brackets_re.') # link text = $2 |
| 561 |
\] |
| 562 |
|
| 563 |
[ ]? # one optional space |
| 564 |
(?:\n[ ]*)? # one optional newline followed by spaces |
| 565 |
|
| 566 |
\[ |
| 567 |
(.*?) # id = $3 |
| 568 |
\] |
| 569 |
) |
| 570 |
}xs', |
| 571 |
array(&$this, '_doAnchors_reference_callback'), $text); |
| 572 |
|
| 573 |
# |
| 574 |
# Next, inline-style links: [link text](url "optional title") |
| 575 |
# |
| 576 |
$text = preg_replace_callback('{ |
| 577 |
( # wrap whole match in $1 |
| 578 |
\[ |
| 579 |
('.$this->nested_brackets_re.') # link text = $2 |
| 580 |
\] |
| 581 |
\( # literal paren |
| 582 |
[ \n]* |
| 583 |
(?: |
| 584 |
<(.+?)> # href = $3 |
| 585 |
| |
| 586 |
('.$this->nested_url_parenthesis_re.') # href = $4 |
| 587 |
) |
| 588 |
[ \n]* |
| 589 |
( # $5 |
| 590 |
([\'"]) # quote char = $6 |
| 591 |
(.*?) # Title = $7 |
| 592 |
\6 # matching quote |
| 593 |
[ \n]* # ignore any spaces/tabs between closing quote and ) |
| 594 |
)? # title is optional |
| 595 |
\) |
| 596 |
) |
| 597 |
}xs', |
| 598 |
array(&$this, '_doAnchors_inline_callback'), $text); |
| 599 |
|
| 600 |
# |
| 601 |
# Last, handle reference-style shortcuts: [link text] |
| 602 |
# These must come last in case you've also got [link text][1] |
| 603 |
# or [link text](/foo) |
| 604 |
# |
| 605 |
$text = preg_replace_callback('{ |
| 606 |
( # wrap whole match in $1 |
| 607 |
\[ |
| 608 |
([^\[\]]+) # link text = $2; can\'t contain [ or ] |
| 609 |
\] |
| 610 |
) |
| 611 |
}xs', |
| 612 |
array(&$this, '_doAnchors_reference_callback'), $text); |
| 613 |
|
| 614 |
$this->in_anchor = false; |
| 615 |
return $text; |
| 616 |
} |
| 617 |
function _doAnchors_reference_callback($matches) { |
| 618 |
$whole_match = $matches[1]; |
| 619 |
$link_text = $matches[2]; |
| 620 |
$link_id =& $matches[3]; |
| 621 |
|
| 622 |
if ($link_id == "") { |
| 623 |
# for shortcut links like [this][] or [this]. |
| 624 |
$link_id = $link_text; |
| 625 |
} |
| 626 |
|
| 627 |
# lower-case and turn embedded newlines into spaces |
| 628 |
$link_id = strtolower($link_id); |
| 629 |
$link_id = preg_replace('{[ ]?\n}', ' ', $link_id); |
| 630 |
|
| 631 |
if (isset($this->urls[$link_id])) { |
| 632 |
$url = $this->urls[$link_id]; |
| 633 |
$url = $this->encodeAttribute($url); |
| 634 |
|
| 635 |
$result = "<a href=\"$url\""; |
| 636 |
if ( isset( $this->titles[$link_id] ) ) { |
| 637 |
$title = $this->titles[$link_id]; |
| 638 |
$title = $this->encodeAttribute($title); |
| 639 |
$result .= " title=\"$title\""; |
| 640 |
} |
| 641 |
|
| 642 |
$link_text = $this->runSpanGamut($link_text); |
| 643 |
$result .= ">$link_text</a>"; |
| 644 |
$result = $this->hashPart($result); |
| 645 |
} |
| 646 |
else { |
| 647 |
$result = $whole_match; |
| 648 |
} |
| 649 |
return $result; |
| 650 |
} |
| 651 |
function _doAnchors_inline_callback($matches) { |
| 652 |
$whole_match = $matches[1]; |
| 653 |
$link_text = $this->runSpanGamut($matches[2]); |
| 654 |
$url = $matches[3] == '' ? $matches[4] : $matches[3]; |
| 655 |
$title =& $matches[7]; |
| 656 |
|
| 657 |
$url = $this->encodeAttribute($url); |
| 658 |
|
| 659 |
$result = "<a href=\"$url\""; |
| 660 |
if (isset($title)) { |
| 661 |
$title = $this->encodeAttribute($title); |
| 662 |
$result .= " title=\"$title\""; |
| 663 |
} |
| 664 |
|
| 665 |
$link_text = $this->runSpanGamut($link_text); |
| 666 |
$result .= ">$link_text</a>"; |
| 667 |
|
| 668 |
return $this->hashPart($result); |
| 669 |
} |
| 670 |
|
| 671 |
|
| 672 |
function doImages($text) { |
| 673 |
# |
| 674 |
# Turn Markdown image shortcuts into <img> tags. |
| 675 |
# |
| 676 |
# |
| 677 |
# First, handle reference-style labeled images: ![alt text][id] |
| 678 |
# |
| 679 |
$text = preg_replace_callback('{ |
| 680 |
( # wrap whole match in $1 |
| 681 |
!\[ |
| 682 |
('.$this->nested_brackets_re.') # alt text = $2 |
| 683 |
\] |
| 684 |
|
| 685 |
[ ]? # one optional space |
| 686 |
(?:\n[ ]*)? # one optional newline followed by spaces |
| 687 |
|
| 688 |
\[ |
| 689 |
(.*?) # id = $3 |
| 690 |
\] |
| 691 |
|
| 692 |
) |
| 693 |
}xs', |
| 694 |
array(&$this, '_doImages_reference_callback'), $text); |
| 695 |
|
| 696 |
# |
| 697 |
# Next, handle inline images:  |
| 698 |
# Don't forget: encode * and _ |
| 699 |
# |
| 700 |
$text = preg_replace_callback('{ |
| 701 |
( # wrap whole match in $1 |
| 702 |
!\[ |
| 703 |
('.$this->nested_brackets_re.') # alt text = $2 |
| 704 |
\] |
| 705 |
\s? # One optional whitespace character |
| 706 |
\( # literal paren |
| 707 |
[ \n]* |
| 708 |
(?: |
| 709 |
<(\S*)> # src url = $3 |
| 710 |
| |
| 711 |
('.$this->nested_url_parenthesis_re.') # src url = $4 |
| 712 |
) |
| 713 |
[ \n]* |
| 714 |
( # $5 |
| 715 |
([\'"]) # quote char = $6 |
| 716 |
(.*?) # title = $7 |
| 717 |
\6 # matching quote |
| 718 |
[ \n]* |
| 719 |
)? # title is optional |
| 720 |
\) |
| 721 |
) |
| 722 |
}xs', |
| 723 |
array(&$this, '_doImages_inline_callback'), $text); |
| 724 |
|
| 725 |
return $text; |
| 726 |
} |
| 727 |
function _doImages_reference_callback($matches) { |
| 728 |
$whole_match = $matches[1]; |
| 729 |
$alt_text = $matches[2]; |
| 730 |
$link_id = strtolower($matches[3]); |
| 731 |
|
| 732 |
if ($link_id == "") { |
| 733 |
$link_id = strtolower($alt_text); # for shortcut links like ![this][]. |
| 734 |
} |
| 735 |
|
| 736 |
$alt_text = $this->encodeAttribute($alt_text); |
| 737 |
if (isset($this->urls[$link_id])) { |
| 738 |
$url = $this->encodeAttribute($this->urls[$link_id]); |
| 739 |
$result = "<img src=\"$url\" alt=\"$alt_text\""; |
| 740 |
if (isset($this->titles[$link_id])) { |
| 741 |
$title = $this->titles[$link_id]; |
| 742 |
$title = $this->encodeAttribute($title); |
| 743 |
$result .= " title=\"$title\""; |
| 744 |
} |
| 745 |
$result .= $this->empty_element_suffix; |
| 746 |
$result = $this->hashPart($result); |
| 747 |
} |
| 748 |
else { |
| 749 |
# If there's no such link ID, leave intact: |
| 750 |
$result = $whole_match; |
| 751 |
} |
| 752 |
|
| 753 |
return $result; |
| 754 |
} |
| 755 |
function _doImages_inline_callback($matches) { |
| 756 |
$whole_match = $matches[1]; |
| 757 |
$alt_text = $matches[2]; |
| 758 |
$url = $matches[3] == '' ? $matches[4] : $matches[3]; |
| 759 |
$title =& $matches[7]; |
| 760 |
|
| 761 |
$alt_text = $this->encodeAttribute($alt_text); |
| 762 |
$url = $this->encodeAttribute($url); |
| 763 |
$result = "<img src=\"$url\" alt=\"$alt_text\""; |
| 764 |
if (isset($title)) { |
| 765 |
$title = $this->encodeAttribute($title); |
| 766 |
$result .= " title=\"$title\""; # $title already quoted |
| 767 |
} |
| 768 |
$result .= $this->empty_element_suffix; |
| 769 |
|
| 770 |
return $this->hashPart($result); |
| 771 |
} |
| 772 |
|
| 773 |
|
| 774 |
function doHeaders($text) { |
| 775 |
# Setext-style headers: |
| 776 |
# Header 1 |
| 777 |
# ======== |
| 778 |
# |
| 779 |
# Header 2 |
| 780 |
# -------- |
| 781 |
# |
| 782 |
$text = preg_replace_callback('{ ^(.+?)[ ]*\n(=+|-+)[ ]*\n+ }mx', |
| 783 |
array(&$this, '_doHeaders_callback_setext'), $text); |
| 784 |
|
| 785 |
# atx-style headers: |
| 786 |
# # Header 1 |
| 787 |
# ## Header 2 |
| 788 |
# ## Header 2 with closing hashes ## |
| 789 |
# ... |
| 790 |
# ###### Header 6 |
| 791 |
# |
| 792 |
$text = preg_replace_callback('{ |
| 793 |
^(\#{1,6}) # $1 = string of #\'s |
| 794 |
[ ]* |
| 795 |
(.+?) # $2 = Header text |
| 796 |
[ ]* |
| 797 |
\#* # optional closing #\'s (not counted) |
| 798 |
\n+ |
| 799 |
}xm', |
| 800 |
array(&$this, '_doHeaders_callback_atx'), $text); |
| 801 |
|
| 802 |
return $text; |
| 803 |
} |
| 804 |
function _doHeaders_callback_setext($matches) { |
| 805 |
# Terrible hack to check we haven't found an empty list item. |
| 806 |
if ($matches[2] == '-' && preg_match('{^-(?: |$)}', $matches[1])) |
| 807 |
return $matches[0]; |
| 808 |
|
| 809 |
$level = $matches[2]{0} == '=' ? 1 : 2; |
| 810 |
$block = "<h$level>".$this->runSpanGamut($matches[1])."</h$level>"; |
| 811 |
return "\n" . $this->hashBlock($block) . "\n\n"; |
| 812 |
} |
| 813 |
function _doHeaders_callback_atx($matches) { |
| 814 |
$level = strlen($matches[1]); |
| 815 |
$block = "<h$level>".$this->runSpanGamut($matches[2])."</h$level>"; |
| 816 |
return "\n" . $this->hashBlock($block) . "\n\n"; |
| 817 |
} |
| 818 |
|
| 819 |
|
| 820 |
function doLists($text) { |
| 821 |
# |
| 822 |
# Form HTML ordered (numbered) and unordered (bulleted) lists. |
| 823 |
# |
| 824 |
$less_than_tab = $this->tab_width - 1; |
| 825 |
|
| 826 |
# Re-usable patterns to match list item bullets and number markers: |
| 827 |
$marker_ul_re = '[*+-]'; |
| 828 |
$marker_ol_re = '\d+[\.]'; |
| 829 |
$marker_any_re = "(?:$marker_ul_re|$marker_ol_re)"; |
| 830 |
|
| 831 |
$markers_relist = array( |
| 832 |
$marker_ul_re => $marker_ol_re, |
| 833 |
$marker_ol_re => $marker_ul_re, |
| 834 |
); |
| 835 |
|
| 836 |
foreach ($markers_relist as $marker_re => $other_marker_re) { |
| 837 |
# Re-usable pattern to match any entirel ul or ol list: |
| 838 |
$whole_list_re = ' |
| 839 |
( # $1 = whole list |
| 840 |
( # $2 |
| 841 |
([ ]{0,'.$less_than_tab.'}) # $3 = number of spaces |
| 842 |
('.$marker_re.') # $4 = first list item marker |
| 843 |
[ ]+ |
| 844 |
) |
| 845 |
(?s:.+?) |
| 846 |
( # $5 |
| 847 |
\z |
| 848 |
| |
| 849 |
\n{2,} |
| 850 |
(?=\S) |
| 851 |
(?! # Negative lookahead for another list item marker |
| 852 |
[ ]* |
| 853 |
'.$marker_re.'[ ]+ |
| 854 |
) |
| 855 |
| |
| 856 |
(?= # Lookahead for another kind of list |
| 857 |
\n |
| 858 |
\3 # Must have the same indentation |
| 859 |
'.$other_marker_re.'[ ]+ |
| 860 |
) |
| 861 |
) |
| 862 |
) |
| 863 |
'; // mx |
| 864 |
|
| 865 |
# We use a different prefix before nested lists than top-level lists. |
| 866 |
# See extended comment in _ProcessListItems(). |
| 867 |
|
| 868 |
if ($this->list_level) { |
| 869 |
$text = preg_replace_callback('{ |
| 870 |
^ |
| 871 |
'.$whole_list_re.' |
| 872 |
}mx', |
| 873 |
array(&$this, '_doLists_callback'), $text); |
| 874 |
} |
| 875 |
else { |
| 876 |
$text = preg_replace_callback('{ |
| 877 |
(?:(?<=\n)\n|\A\n?) # Must eat the newline |
| 878 |
'.$whole_list_re.' |
| 879 |
}mx', |
| 880 |
array(&$this, '_doLists_callback'), $text); |
| 881 |
} |
| 882 |
} |
| 883 |
|
| 884 |
return $text; |
| 885 |
} |
| 886 |
function _doLists_callback($matches) { |
| 887 |
# Re-usable patterns to match list item bullets and number markers: |
| 888 |
$marker_ul_re = '[*+-]'; |
| 889 |
$marker_ol_re = '\d+[\.]'; |
| 890 |
$marker_any_re = "(?:$marker_ul_re|$marker_ol_re)"; |
| 891 |
|
| 892 |
$list = $matches[1]; |
| 893 |
$list_type = preg_match("/$marker_ul_re/", $matches[4]) ? "ul" : "ol"; |
| 894 |
|
| 895 |
$marker_any_re = ( $list_type == "ul" ? $marker_ul_re : $marker_ol_re ); |
| 896 |
|
| 897 |
$list .= "\n"; |
| 898 |
$result = $this->processListItems($list, $marker_any_re); |
| 899 |
|
| 900 |
$result = $this->hashBlock("<$list_type>\n" . $result . "</$list_type>"); |
| 901 |
return "\n". $result ."\n\n"; |
| 902 |
} |
| 903 |
|
| 904 |
public $list_level = 0; |
| 905 |
|
| 906 |
function processListItems($list_str, $marker_any_re) { |
| 907 |
# |
| 908 |
# Process the contents of a single ordered or unordered list, splitting it |
| 909 |
# into individual list items. |
| 910 |
# |
| 911 |
# The $this->list_level global keeps track of when we're inside a list. |
| 912 |
# Each time we enter a list, we increment it; when we leave a list, |
| 913 |
# we decrement. If it's zero, we're not in a list anymore. |
| 914 |
# |
| 915 |
# We do this because when we're not inside a list, we want to treat |
| 916 |
# something like this: |
| 917 |
# |
| 918 |
# I recommend upgrading to version |
| 919 |
# 8. Oops, now this line is treated |
| 920 |
# as a sub-list. |
| 921 |
# |
| 922 |
# As a single paragraph, despite the fact that the second line starts |
| 923 |
# with a digit-period-space sequence. |
| 924 |
# |
| 925 |
# Whereas when we're inside a list (or sub-list), that line will be |
| 926 |
# treated as the start of a sub-list. What a kludge, huh? This is |
| 927 |
# an aspect of Markdown's syntax that's hard to parse perfectly |
| 928 |
# without resorting to mind-reading. Perhaps the solution is to |
| 929 |
# change the syntax rules such that sub-lists must start with a |
| 930 |
# starting cardinal number; e.g. "1." or "a.". |
| 931 |
|
| 932 |
$this->list_level++; |
| 933 |
|
| 934 |
# trim trailing blank lines: |
| 935 |
$list_str = preg_replace("/\n{2,}\\z/", "\n", $list_str); |
| 936 |
|
| 937 |
$list_str = preg_replace_callback('{ |
| 938 |
(\n)? # leading line = $1 |
| 939 |
(^[ ]*) # leading whitespace = $2 |
| 940 |
('.$marker_any_re.' # list marker and space = $3 |
| 941 |
(?:[ ]+|(?=\n)) # space only required if item is not empty |
| 942 |
) |
| 943 |
((?s:.*?)) # list item text = $4 |
| 944 |
(?:(\n+(?=\n))|\n) # tailing blank line = $5 |
| 945 |
(?= \n* (\z | \2 ('.$marker_any_re.') (?:[ ]+|(?=\n)))) |
| 946 |
}xm', |
| 947 |
array(&$this, '_processListItems_callback'), $list_str); |
| 948 |
|
| 949 |
$this->list_level--; |
| 950 |
return $list_str; |
| 951 |
} |
| 952 |
function _processListItems_callback($matches) { |
| 953 |
$item = $matches[4]; |
| 954 |
$leading_line =& $matches[1]; |
| 955 |
$leading_space =& $matches[2]; |
| 956 |
$marker_space = $matches[3]; |
| 957 |
$tailing_blank_line =& $matches[5]; |
| 958 |
|
| 959 |
if ($leading_line || $tailing_blank_line || |
| 960 |
preg_match('/\n{2,}/', $item)) |
| 961 |
{ |
| 962 |
# Replace marker with the appropriate whitespace indentation |
| 963 |
$item = $leading_space . str_repeat(' ', strlen($marker_space)) . $item; |
| 964 |
$item = $this->runBlockGamut($this->outdent($item)."\n"); |
| 965 |
} |
| 966 |
else { |
| 967 |
# Recursion for sub-lists: |
| 968 |
$item = $this->doLists($this->outdent($item)); |
| 969 |
$item = preg_replace('/\n+$/', '', $item); |
| 970 |
$item = $this->runSpanGamut($item); |
| 971 |
} |
| 972 |
|
| 973 |
return "<li>" . $item . "</li>\n"; |
| 974 |
} |
| 975 |
|
| 976 |
|
| 977 |
function doCodeBlocks($text) { |
| 978 |
# |
| 979 |
# Process Markdown `<pre><code>` blocks. |
| 980 |
# |
| 981 |
$text = preg_replace_callback('{ |
| 982 |
(?:\n\n|\A\n?) |
| 983 |
( # $1 = the code block -- one or more lines, starting with a space/tab |
| 984 |
(?> |
| 985 |
[ ]{'.$this->tab_width.'} # Lines must start with a tab or a tab-width of spaces |
| 986 |
.*\n+ |
| 987 |
)+ |
| 988 |
) |
| 989 |
((?=^[ ]{0,'.$this->tab_width.'}\S)|\Z) # Lookahead for non-space at line-start, or end of doc |
| 990 |
}xm', |
| 991 |
array(&$this, '_doCodeBlocks_callback'), $text); |
| 992 |
|
| 993 |
return $text; |
| 994 |
} |
| 995 |
function _doCodeBlocks_callback($matches) { |
| 996 |
$codeblock = $matches[1]; |
| 997 |
|
| 998 |
$codeblock = $this->outdent($codeblock); |
| 999 |
$codeblock = htmlspecialchars($codeblock, ENT_NOQUOTES); |
| 1000 |
|
| 1001 |
# trim leading newlines and trailing newlines |
| 1002 |
$codeblock = preg_replace('/\A\n+|\n+\z/', '', $codeblock); |
| 1003 |
|
| 1004 |
$codeblock = "<pre><code>$codeblock\n</code></pre>"; |
| 1005 |
return "\n\n".$this->hashBlock($codeblock)."\n\n"; |
| 1006 |
} |
| 1007 |
|
| 1008 |
|
| 1009 |
function makeCodeSpan($code) { |
| 1010 |
# |
| 1011 |
# Create a code span markup for $code. Called from handleSpanToken. |
| 1012 |
# |
| 1013 |
$code = htmlspecialchars(trim($code), ENT_NOQUOTES); |
| 1014 |
return $this->hashPart("<code>$code</code>"); |
| 1015 |
} |
| 1016 |
|
| 1017 |
|
| 1018 |
public $em_relist = array( |
| 1019 |
'' => '(?:(?<!\*)\*(?!\*)|(?<!_)_(?!_))(?=\S|$)(?![\.,:;]\s)', |
| 1020 |
'*' => '(?<=\S|^)(?<!\*)\*(?!\*)', |
| 1021 |
'_' => '(?<=\S|^)(?<!_)_(?!_)', |
| 1022 |
); |
| 1023 |
public $strong_relist = array( |
| 1024 |
'' => '(?:(?<!\*)\*\*(?!\*)|(?<!_)__(?!_))(?=\S|$)(?![\.,:;]\s)', |
| 1025 |
'**' => '(?<=\S|^)(?<!\*)\*\*(?!\*)', |
| 1026 |
'__' => '(?<=\S|^)(?<!_)__(?!_)', |
| 1027 |
); |
| 1028 |
public $em_strong_relist = array( |
| 1029 |
'' => '(?:(?<!\*)\*\*\*(?!\*)|(?<!_)___(?!_))(?=\S|$)(?![\.,:;]\s)', |
| 1030 |
'***' => '(?<=\S|^)(?<!\*)\*\*\*(?!\*)', |
| 1031 |
'___' => '(?<=\S|^)(?<!_)___(?!_)', |
| 1032 |
); |
| 1033 |
public $em_strong_prepared_relist; |
| 1034 |
|
| 1035 |
function prepareItalicsAndBold() { |
| 1036 |
# |
| 1037 |
# Prepare regular expressions for searching emphasis tokens in any |
| 1038 |
# context. |
| 1039 |
# |
| 1040 |
foreach ($this->em_relist as $em => $em_re) { |
| 1041 |
foreach ($this->strong_relist as $strong => $strong_re) { |
| 1042 |
# Construct list of allowed token expressions. |
| 1043 |
$token_relist = array(); |
| 1044 |
if (isset($this->em_strong_relist["$em$strong"])) { |
| 1045 |
$token_relist[] = $this->em_strong_relist["$em$strong"]; |
| 1046 |
} |
| 1047 |
$token_relist[] = $em_re; |
| 1048 |
$token_relist[] = $strong_re; |
| 1049 |
|
| 1050 |
# Construct master expression from list. |
| 1051 |
$token_re = '{('. implode('|', $token_relist) .')}'; |
| 1052 |
$this->em_strong_prepared_relist["$em$strong"] = $token_re; |
| 1053 |
} |
| 1054 |
} |
| 1055 |
} |
| 1056 |
|
| 1057 |
function doItalicsAndBold($text) { |
| 1058 |
$token_stack = array(''); |
| 1059 |
$text_stack = array(''); |
| 1060 |
$em = ''; |
| 1061 |
$strong = ''; |
| 1062 |
$tree_char_em = false; |
| 1063 |
|
| 1064 |
while (1) { |
| 1065 |
# |
| 1066 |
# Get prepared regular expression for seraching emphasis tokens |
| 1067 |
# in current context. |
| 1068 |
# |
| 1069 |
$token_re = $this->em_strong_prepared_relist["$em$strong"]; |
| 1070 |
|
| 1071 |
# |
| 1072 |
# Each loop iteration search for the next emphasis token. |
| 1073 |
# Each token is then passed to handleSpanToken. |
| 1074 |
# |
| 1075 |
$parts = preg_split($token_re, $text, 2, PREG_SPLIT_DELIM_CAPTURE); |
| 1076 |
$text_stack[0] .= $parts[0]; |
| 1077 |
$token =& $parts[1]; |
| 1078 |
$text =& $parts[2]; |
| 1079 |
|
| 1080 |
if (empty($token)) { |
| 1081 |
# Reached end of text span: empty stack without emitting. |
| 1082 |
# any more emphasis. |
| 1083 |
while ($token_stack[0]) { |
| 1084 |
$text_stack[1] .= array_shift($token_stack); |
| 1085 |
$text_stack[0] .= array_shift($text_stack); |
| 1086 |
} |
| 1087 |
break; |
| 1088 |
} |
| 1089 |
|
| 1090 |
$token_len = strlen($token); |
| 1091 |
if ($tree_char_em) { |
| 1092 |
# Reached closing marker while inside a three-char emphasis. |
| 1093 |
if ($token_len == 3) { |
| 1094 |
# Three-char closing marker, close em and strong. |
| 1095 |
array_shift($token_stack); |
| 1096 |
$span = array_shift($text_stack); |
| 1097 |
$span = $this->runSpanGamut($span); |
| 1098 |
$span = "<strong><em>$span</em></strong>"; |
| 1099 |
$text_stack[0] .= $this->hashPart($span); |
| 1100 |
$em = ''; |
| 1101 |
$strong = ''; |
| 1102 |
} else { |
| 1103 |
# Other closing marker: close one em or strong and |
| 1104 |
# change current token state to match the other |
| 1105 |
$token_stack[0] = str_repeat($token{0}, 3-$token_len); |
| 1106 |
$tag = $token_len == 2 ? "strong" : "em"; |
| 1107 |
$span = $text_stack[0]; |
| 1108 |
$span = $this->runSpanGamut($span); |
| 1109 |
$span = "<$tag>$span</$tag>"; |
| 1110 |
$text_stack[0] = $this->hashPart($span); |
| 1111 |
$$tag = ''; # $$tag stands for $em or $strong |
| 1112 |
} |
| 1113 |
$tree_char_em = false; |
| 1114 |
} else if ($token_len == 3) { |
| 1115 |
if ($em) { |
| 1116 |
# Reached closing marker for both em and strong. |
| 1117 |
# Closing strong marker: |
| 1118 |
for ($i = 0; $i < 2; ++$i) { |
| 1119 |
$shifted_token = array_shift($token_stack); |
| 1120 |
$tag = strlen($shifted_token) == 2 ? "strong" : "em"; |
| 1121 |
$span = array_shift($text_stack); |
| 1122 |
$span = $this->runSpanGamut($span); |
| 1123 |
$span = "<$tag>$span</$tag>"; |
| 1124 |
$text_stack[0] .= $this->hashPart($span); |
| 1125 |
$$tag = ''; # $$tag stands for $em or $strong |
| 1126 |
} |
| 1127 |
} else { |
| 1128 |
# Reached opening three-char emphasis marker. Push on token |
| 1129 |
# stack; will be handled by the special condition above. |
| 1130 |
$em = $token{0}; |
| 1131 |
$strong = "$em$em"; |
| 1132 |
array_unshift($token_stack, $token); |
| 1133 |
array_unshift($text_stack, ''); |
| 1134 |
$tree_char_em = true; |
| 1135 |
} |
| 1136 |
} else if ($token_len == 2) { |
| 1137 |
if ($strong) { |
| 1138 |
# Unwind any dangling emphasis marker: |
| 1139 |
if (strlen($token_stack[0]) == 1) { |
| 1140 |
$text_stack[1] .= array_shift($token_stack); |
| 1141 |
$text_stack[0] .= array_shift($text_stack); |
| 1142 |
} |
| 1143 |
# Closing strong marker: |
| 1144 |
array_shift($token_stack); |
| 1145 |
$span = array_shift($text_stack); |
| 1146 |
$span = $this->runSpanGamut($span); |
| 1147 |
$span = "<strong>$span</strong>"; |
| 1148 |
$text_stack[0] .= $this->hashPart($span); |
| 1149 |
$strong = ''; |
| 1150 |
} else { |
| 1151 |
array_unshift($token_stack, $token); |
| 1152 |
array_unshift($text_stack, ''); |
| 1153 |
$strong = $token; |
| 1154 |
} |
| 1155 |
} else { |
| 1156 |
# Here $token_len == 1 |
| 1157 |
if ($em) { |
| 1158 |
if (strlen($token_stack[0]) == 1) { |
| 1159 |
# Closing emphasis marker: |
| 1160 |
array_shift($token_stack); |
| 1161 |
$span = array_shift($text_stack); |
| 1162 |
$span = $this->runSpanGamut($span); |
| 1163 |
$span = "<em>$span</em>"; |
| 1164 |
$text_stack[0] .= $this->hashPart($span); |
| 1165 |
$em = ''; |
| 1166 |
} else { |
| 1167 |
$text_stack[0] .= $token; |
| 1168 |
} |
| 1169 |
} else { |
| 1170 |
array_unshift($token_stack, $token); |
| 1171 |
array_unshift($text_stack, ''); |
| 1172 |
$em = $token; |
| 1173 |
} |
| 1174 |
} |
| 1175 |
} |
| 1176 |
return $text_stack[0]; |
| 1177 |
} |
| 1178 |
|
| 1179 |
|
| 1180 |
function doBlockQuotes($text) { |
| 1181 |
$text = preg_replace_callback('/ |
| 1182 |
( # Wrap whole match in $1 |
| 1183 |
(?> |
| 1184 |
^[ ]*>[ ]? # ">" at the start of a line |
| 1185 |
.+\n # rest of the first line |
| 1186 |
(.+\n)* # subsequent consecutive lines |
| 1187 |
\n* # blanks |
| 1188 |
)+ |
| 1189 |
) |
| 1190 |
/xm', |
| 1191 |
array(&$this, '_doBlockQuotes_callback'), $text); |
| 1192 |
|
| 1193 |
return $text; |
| 1194 |
} |
| 1195 |
function _doBlockQuotes_callback($matches) { |
| 1196 |
$bq = $matches[1]; |
| 1197 |
# trim one level of quoting - trim whitespace-only lines |
| 1198 |
$bq = preg_replace('/^[ ]*>[ ]?|^[ ]+$/m', '', $bq); |
| 1199 |
$bq = $this->runBlockGamut($bq); # recurse |
| 1200 |
|
| 1201 |
$bq = preg_replace('/^/m', " ", $bq); |
| 1202 |
# These leading spaces cause problem with <pre> content, |
| 1203 |
# so we need to fix that: |
| 1204 |
$bq = preg_replace_callback('{(\s*<pre>.+?</pre>)}sx', |
| 1205 |
array(&$this, '_doBlockQuotes_callback2'), $bq); |
| 1206 |
|
| 1207 |
return "\n". $this->hashBlock("<blockquote>\n$bq\n</blockquote>")."\n\n"; |
| 1208 |
} |
| 1209 |
function _doBlockQuotes_callback2($matches) { |
| 1210 |
$pre = $matches[1]; |
| 1211 |
$pre = preg_replace('/^ /m', '', $pre); |
| 1212 |
return $pre; |
| 1213 |
} |
| 1214 |
|
| 1215 |
|
| 1216 |
function formParagraphs($text) { |
| 1217 |
# |
| 1218 |
# Params: |
| 1219 |
# $text - string to process with html <p> tags |
| 1220 |
# |
| 1221 |
# Strip leading and trailing lines: |
| 1222 |
$text = preg_replace('/\A\n+|\n+\z/', '', $text); |
| 1223 |
|
| 1224 |
$grafs = preg_split('/\n{2,}/', $text, -1, PREG_SPLIT_NO_EMPTY); |
| 1225 |
|
| 1226 |
# |
| 1227 |
# Wrap <p> tags and unhashify HTML blocks |
| 1228 |
# |
| 1229 |
foreach ($grafs as $key => $value) { |
| 1230 |
if (!preg_match('/^B\x1A[0-9]+B$/', $value)) { |
| 1231 |
# Is a paragraph. |
| 1232 |
$value = $this->runSpanGamut($value); |
| 1233 |
$value = preg_replace('/^([ ]*)/', "<p>", $value); |
| 1234 |
$value .= "</p>"; |
| 1235 |
$grafs[$key] = $this->unhash($value); |
| 1236 |
} |
| 1237 |
else { |
| 1238 |
# Is a block. |
| 1239 |
# Modify elements of @grafs in-place... |
| 1240 |
$graf = $value; |
| 1241 |
$block = $this->html_hashes[$graf]; |
| 1242 |
$graf = $block; |
| 1243 |
// if (preg_match('{ |
| 1244 |
// \A |
| 1245 |
// ( # $1 = <div> tag |
| 1246 |
// <div \s+ |
| 1247 |
// [^>]* |
| 1248 |
// \b |
| 1249 |
// markdown\s*=\s* ([\'"]) # $2 = attr quote char |
| 1250 |
// 1 |
| 1251 |
// \2 |
| 1252 |
// [^>]* |
| 1253 |
// > |
| 1254 |
// ) |
| 1255 |
// ( # $3 = contents |
| 1256 |
// .* |
| 1257 |
// ) |
| 1258 |
// (</div>) # $4 = closing tag |
| 1259 |
// \z |
| 1260 |
// }xs', $block, $matches)) |
| 1261 |
// { |
| 1262 |
// list(, $div_open, , $div_content, $div_close) = $matches; |
| 1263 |
// |
| 1264 |
// # We can't call Markdown(), because that resets the hash; |
| 1265 |
// # that initialization code should be pulled into its own sub, though. |
| 1266 |
// $div_content = $this->hashHTMLBlocks($div_content); |
| 1267 |
// |
| 1268 |
// # Run document gamut methods on the content. |
| 1269 |
// foreach ($this->document_gamut as $method => $priority) { |
| 1270 |
// $div_content = $this->$method($div_content); |
| 1271 |
// } |
| 1272 |
// |
| 1273 |
// $div_open = preg_replace( |
| 1274 |
// '{\smarkdown\s*=\s*([\'"]).+?\1}', '', $div_open); |
| 1275 |
// |
| 1276 |
// $graf = $div_open . "\n" . $div_content . "\n" . $div_close; |
| 1277 |
// } |
| 1278 |
$grafs[$key] = $graf; |
| 1279 |
} |
| 1280 |
} |
| 1281 |
|
| 1282 |
return implode("\n\n", $grafs); |
| 1283 |
} |
| 1284 |
|
| 1285 |
|
| 1286 |
function encodeAttribute($text) { |
| 1287 |
# |
| 1288 |
# Encode text for a double-quoted HTML attribute. This function |
| 1289 |
# is *not* suitable for attributes enclosed in single quotes. |
| 1290 |
# |
| 1291 |
$text = $this->encodeAmpsAndAngles($text); |
| 1292 |
$text = str_replace('"', '"', $text); |
| 1293 |
return $text; |
| 1294 |
} |
| 1295 |
|
| 1296 |
|
| 1297 |
function encodeAmpsAndAngles($text) { |
| 1298 |
# |
| 1299 |
# Smart processing for ampersands and angle brackets that need to |
| 1300 |
# be encoded. Valid character entities are left alone unless the |
| 1301 |
# no-entities mode is set. |
| 1302 |
# |
| 1303 |
if ($this->no_entities) { |
| 1304 |
$text = str_replace('&', '&', $text); |
| 1305 |
} else { |
| 1306 |
# Ampersand-encoding based entirely on Nat Irons's Amputator |
| 1307 |
# MT plugin: <http://bumppo.net/projects/amputator/> |
| 1308 |
$text = preg_replace('/&(?!#?[xX]?(?:[0-9a-fA-F]+|\w+);)/', |
| 1309 |
'&', $text);; |
| 1310 |
} |
| 1311 |
# Encode remaining <'s |
| 1312 |
$text = str_replace('<', '<', $text); |
| 1313 |
|
| 1314 |
return $text; |
| 1315 |
} |
| 1316 |
|
| 1317 |
|
| 1318 |
function doAutoLinks($text) { |
| 1319 |
$text = preg_replace_callback('{<((https?|ftp|dict):[^\'">\s]+)>}i', |
| 1320 |
array(&$this, '_doAutoLinks_url_callback'), $text); |
| 1321 |
|
| 1322 |
# Email addresses: <address@domain.foo> |
| 1323 |
$text = preg_replace_callback('{ |
| 1324 |
< |
| 1325 |
(?:mailto:)? |
| 1326 |
( |
| 1327 |
(?: |
| 1328 |
[-!#$%&\'*+/=?^_`.{|}~\w\x80-\xFF]+ |
| 1329 |
| |
| 1330 |
".*?" |
| 1331 |
) |
| 1332 |
\@ |
| 1333 |
(?: |
| 1334 |
[-a-z0-9\x80-\xFF]+(\.[-a-z0-9\x80-\xFF]+)*\.[a-z]+ |
| 1335 |
| |
| 1336 |
\[[\d.a-fA-F:]+\] # IPv4 & IPv6 |
| 1337 |
) |
| 1338 |
) |
| 1339 |
> |
| 1340 |
}xi', |
| 1341 |
array(&$this, '_doAutoLinks_email_callback'), $text); |
| 1342 |
$text = preg_replace_callback('{<(tel:([^\'">\s]+))>}i',array(&$this, '_doAutoLinks_tel_callback'), $text); |
| 1343 |
|
| 1344 |
return $text; |
| 1345 |
} |
| 1346 |
function _doAutoLinks_tel_callback($matches) { |
| 1347 |
$url = $this->encodeAttribute($matches[1]); |
| 1348 |
$tel = $this->encodeAttribute($matches[2]); |
| 1349 |
$link = "<a href=\"$url\">$tel</a>"; |
| 1350 |
return $this->hashPart($link); |
| 1351 |
} |
| 1352 |
function _doAutoLinks_url_callback($matches) { |
| 1353 |
$url = $this->encodeAttribute($matches[1]); |
| 1354 |
$link = "<a href=\"$url\">$url</a>"; |
| 1355 |
return $this->hashPart($link); |
| 1356 |
} |
| 1357 |
function _doAutoLinks_email_callback($matches) { |
| 1358 |
$address = $matches[1]; |
| 1359 |
$link = $this->encodeEmailAddress($address); |
| 1360 |
return $this->hashPart($link); |
| 1361 |
} |
| 1362 |
|
| 1363 |
|
| 1364 |
function encodeEmailAddress($addr) { |
| 1365 |
# |
| 1366 |
# Input: an email address, e.g. "foo@example.com" |
| 1367 |
# |
| 1368 |
# Output: the email address as a mailto link, with each character |
| 1369 |
# of the address encoded as either a decimal or hex entity, in |
| 1370 |
# the hopes of foiling most address harvesting spam bots. E.g.: |
| 1371 |
# |
| 1372 |
# <p><a href="mailto:foo |
| 1373 |
# @example.co |
| 1374 |
# m">foo@exampl |
| 1375 |
# e.com</a></p> |
| 1376 |
# |
| 1377 |
# Based by a filter by Matthew Wickline, posted to BBEdit-Talk. |
| 1378 |
# With some optimizations by Milian Wolff. |
| 1379 |
# |
| 1380 |
$addr = "mailto:" . $addr; |
| 1381 |
$chars = preg_split('/(?<!^)(?!$)/', $addr); |
| 1382 |
$seed = (int)abs(crc32($addr) / strlen($addr)); # Deterministic seed. |
| 1383 |
|
| 1384 |
foreach ($chars as $key => $char) { |
| 1385 |
$ord = ord($char); |
| 1386 |
# Ignore non-ascii chars. |
| 1387 |
if ($ord < 128) { |
| 1388 |
$r = ($seed * (1 + $key)) % 100; # Pseudo-random function. |
| 1389 |
# roughly 10% raw, 45% hex, 45% dec |
| 1390 |
# '@' *must* be encoded. I insist. |
| 1391 |
if ($r > 90 && $char != '@') /* do nothing */; |
| 1392 |
else if ($r < 45) $chars[$key] = '&#x'.dechex($ord).';'; |
| 1393 |
else $chars[$key] = '&#'.$ord.';'; |
| 1394 |
} |
| 1395 |
} |
| 1396 |
|
| 1397 |
$addr = implode('', $chars); |
| 1398 |
$text = implode('', array_slice($chars, 7)); # text without `mailto:` |
| 1399 |
$addr = "<a href=\"$addr\">$text</a>"; |
| 1400 |
|
| 1401 |
return $addr; |
| 1402 |
} |
| 1403 |
|
| 1404 |
|
| 1405 |
function parseSpan($str) { |
| 1406 |
# |
| 1407 |
# Take the string $str and parse it into tokens, hashing embedded HTML, |
| 1408 |
# escaped characters and handling code spans. |
| 1409 |
# |
| 1410 |
$output = ''; |
| 1411 |
|
| 1412 |
$span_re = '{ |
| 1413 |
( |
| 1414 |
\\\\'.$this->escape_chars_re.' |
| 1415 |
| |
| 1416 |
(?<![`\\\\]) |
| 1417 |
`+ # code span marker |
| 1418 |
'.( $this->no_markup ? '' : ' |
| 1419 |
| |
| 1420 |
<!-- .*? --> # comment |
| 1421 |
| |
| 1422 |
<\?.*?\?> | <%.*?%> # processing instruction |
| 1423 |
| |
| 1424 |
<[!$]?[-a-zA-Z0-9:_]+ # regular tags |
| 1425 |
(?> |
| 1426 |
\s |
| 1427 |
(?>[^"\'>]+|"[^"]*"|\'[^\']*\')* |
| 1428 |
)? |
| 1429 |
> |
| 1430 |
| |
| 1431 |
<[-a-zA-Z0-9:_]+\s*/> # xml-style empty tag |
| 1432 |
| |
| 1433 |
</[-a-zA-Z0-9:_]+\s*> # closing tag |
| 1434 |
').' |
| 1435 |
) |
| 1436 |
}xs'; |
| 1437 |
|
| 1438 |
while (1) { |
| 1439 |
# |
| 1440 |
# Each loop iteration search for either the next tag, the next |
| 1441 |
# openning code span marker, or the next escaped character. |
| 1442 |
# Each token is then passed to handleSpanToken. |
| 1443 |
# |
| 1444 |
$parts = preg_split($span_re, $str, 2, PREG_SPLIT_DELIM_CAPTURE); |
| 1445 |
|
| 1446 |
# Create token from text preceding tag. |
| 1447 |
if ($parts[0] != "") { |
| 1448 |
$output .= $parts[0]; |
| 1449 |
} |
| 1450 |
|
| 1451 |
# Check if we reach the end. |
| 1452 |
if (isset($parts[1])) { |
| 1453 |
$output .= $this->handleSpanToken($parts[1], $parts[2]); |
| 1454 |
$str = $parts[2]; |
| 1455 |
} |
| 1456 |
else { |
| 1457 |
break; |
| 1458 |
} |
| 1459 |
} |
| 1460 |
|
| 1461 |
return $output; |
| 1462 |
} |
| 1463 |
|
| 1464 |
|
| 1465 |
function handleSpanToken($token, &$str) { |
| 1466 |
# |
| 1467 |
# Handle $token provided by parseSpan by determining its nature and |
| 1468 |
# returning the corresponding value that should replace it. |
| 1469 |
# |
| 1470 |
switch ($token{0}) { |
| 1471 |
case "\\": |
| 1472 |
return $this->hashPart("&#". ord($token{1}). ";"); |
| 1473 |
case "`": |
| 1474 |
# Search for end marker in remaining text. |
| 1475 |
if (preg_match('/^(.*?[^`])'.preg_quote($token).'(?!`)(.*)$/sm', |
| 1476 |
$str, $matches)) |
| 1477 |
{ |
| 1478 |
$str = $matches[2]; |
| 1479 |
$codespan = $this->makeCodeSpan($matches[1]); |
| 1480 |
return $this->hashPart($codespan); |
| 1481 |
} |
| 1482 |
return $token; // return as text since no ending marker found. |
| 1483 |
default: |
| 1484 |
return $this->hashPart($token); |
| 1485 |
} |
| 1486 |
} |
| 1487 |
|
| 1488 |
|
| 1489 |
function outdent($text) { |
| 1490 |
# |
| 1491 |
# Remove one level of line-leading tabs or spaces |
| 1492 |
# |
| 1493 |
return preg_replace('/^(\t|[ ]{1,'.$this->tab_width.'})/m', '', $text); |
| 1494 |
} |
| 1495 |
|
| 1496 |
|
| 1497 |
# String length function for detab. `_initDetab` will create a function to |
| 1498 |
# hanlde UTF-8 if the default function does not exist. |
| 1499 |
public $utf8_strlen = 'mb_strlen'; |
| 1500 |
|
| 1501 |
function detab($text) { |
| 1502 |
# |
| 1503 |
# Replace tabs with the appropriate amount of space. |
| 1504 |
# |
| 1505 |
# For each line we separate the line in blocks delemited by |
| 1506 |
# tab characters. Then we reconstruct every line by adding the |
| 1507 |
# appropriate number of space between each blocks. |
| 1508 |
|
| 1509 |
$text = preg_replace_callback('/^.*\t.*$/m', |
| 1510 |
array(&$this, '_detab_callback'), $text); |
| 1511 |
|
| 1512 |
return $text; |
| 1513 |
} |
| 1514 |
function _detab_callback($matches) { |
| 1515 |
$line = $matches[0]; |
| 1516 |
$strlen = $this->utf8_strlen; # strlen function for UTF-8. |
| 1517 |
|
| 1518 |
# Split in blocks. |
| 1519 |
$blocks = explode("\t", $line); |
| 1520 |
# Add each blocks to the line. |
| 1521 |
$line = $blocks[0]; |
| 1522 |
unset($blocks[0]); # Do not add first block twice. |
| 1523 |
foreach ($blocks as $block) { |
| 1524 |
# Calculate amount of space, insert spaces, insert block. |
| 1525 |
$amount = $this->tab_width - |
| 1526 |
$strlen($line, 'UTF-8') % $this->tab_width; |
| 1527 |
$line .= str_repeat(" ", $amount) . $block; |
| 1528 |
} |
| 1529 |
return $line; |
| 1530 |
} |
| 1531 |
function _initDetab() { |
| 1532 |
# |
| 1533 |
# Check for the availability of the function in the `utf8_strlen` property |
| 1534 |
# (initially `mb_strlen`). If the function is not available, use jetpack_utf8_strlen |
| 1535 |
# that will loosely count the number of UTF-8 characters with a |
| 1536 |
# regular expression. |
| 1537 |
# |
| 1538 |
if ( function_exists( $this->utf8_strlen ) ) { |
| 1539 |
return; |
| 1540 |
} |
| 1541 |
$this->utf8_strlen = 'jetpack_utf8_strlen'; |
| 1542 |
} |
| 1543 |
|
| 1544 |
|
| 1545 |
function unhash($text) { |
| 1546 |
# |
| 1547 |
# Swap back in all the tags hashed by _HashHTMLBlocks. |
| 1548 |
# |
| 1549 |
return preg_replace_callback('/(.)\x1A[0-9]+\1/', |
| 1550 |
array(&$this, '_unhash_callback'), $text); |
| 1551 |
} |
| 1552 |
function _unhash_callback($matches) { |
| 1553 |
return $this->html_hashes[$matches[0]]; |
| 1554 |
} |
| 1555 |
|
| 1556 |
} |
| 1557 |
|
| 1558 |
|
| 1559 |
# |
| 1560 |
# Markdown Extra Parser Class |
| 1561 |
# |
| 1562 |
|
| 1563 |
class MarkdownExtra_Parser extends Markdown_Parser { |
| 1564 |
|
| 1565 |
### Configuration Variables ### |
| 1566 |
|
| 1567 |
# Prefix for footnote ids. |
| 1568 |
public $fn_id_prefix = ""; |
| 1569 |
|
| 1570 |
# Optional title attribute for footnote links and backlinks. |
| 1571 |
public $fn_link_title = MARKDOWN_FN_LINK_TITLE; |
| 1572 |
public $fn_backlink_title = MARKDOWN_FN_BACKLINK_TITLE; |
| 1573 |
|
| 1574 |
# Optional class attribute for footnote links and backlinks. |
| 1575 |
public $fn_link_class = MARKDOWN_FN_LINK_CLASS; |
| 1576 |
public $fn_backlink_class = MARKDOWN_FN_BACKLINK_CLASS; |
| 1577 |
|
| 1578 |
# Optional class prefix for fenced code block. |
| 1579 |
public $code_class_prefix = MARKDOWN_CODE_CLASS_PREFIX; |
| 1580 |
# Class attribute for code blocks goes on the `code` tag; |
| 1581 |
# setting this to true will put attributes on the `pre` tag instead. |
| 1582 |
public $code_attr_on_pre = MARKDOWN_CODE_ATTR_ON_PRE; |
| 1583 |
|
| 1584 |
# Predefined abbreviations. |
| 1585 |
public $predef_abbr = array(); |
| 1586 |
|
| 1587 |
|
| 1588 |
### Parser Implementation ### |
| 1589 |
|
| 1590 |
function __construct() { |
| 1591 |
# |
| 1592 |
# Constructor function. Initialize the parser object. |
| 1593 |
# |
| 1594 |
# Add extra escapable characters before parent constructor |
| 1595 |
# initialize the table. |
| 1596 |
$this->escape_chars .= ':|'; |
| 1597 |
|
| 1598 |
# Insert extra document, block, and span transformations. |
| 1599 |
# Parent constructor will do the sorting. |
| 1600 |
$this->document_gamut += array( |
| 1601 |
"doFencedCodeBlocks" => 5, |
| 1602 |
"stripFootnotes" => 15, |
| 1603 |
"stripAbbreviations" => 25, |
| 1604 |
"appendFootnotes" => 50, |
| 1605 |
); |
| 1606 |
$this->block_gamut += array( |
| 1607 |
"doFencedCodeBlocks" => 5, |
| 1608 |
"doTables" => 15, |
| 1609 |
"doDefLists" => 45, |
| 1610 |
); |
| 1611 |
$this->span_gamut += array( |
| 1612 |
"doFootnotes" => 5, |
| 1613 |
"doAbbreviations" => 70, |
| 1614 |
); |
| 1615 |
|
| 1616 |
parent::__construct(); |
| 1617 |
} |
| 1618 |
|
| 1619 |
|
| 1620 |
# Extra variables used during extra transformations. |
| 1621 |
public $footnotes = array(); |
| 1622 |
public $footnotes_ordered = array(); |
| 1623 |
public $footnotes_ref_count = array(); |
| 1624 |
public $footnotes_numbers = array(); |
| 1625 |
public $abbr_desciptions = array(); |
| 1626 |
public $abbr_word_re = ''; |
| 1627 |
|
| 1628 |
# Give the current footnote number. |
| 1629 |
public $footnote_counter = 1; |
| 1630 |
|
| 1631 |
|
| 1632 |
function setup() { |
| 1633 |
# |
| 1634 |
# Setting up Extra-specific variables. |
| 1635 |
# |
| 1636 |
parent::setup(); |
| 1637 |
|
| 1638 |
$this->footnotes = array(); |
| 1639 |
$this->footnotes_ordered = array(); |
| 1640 |
$this->footnotes_ref_count = array(); |
| 1641 |
$this->footnotes_numbers = array(); |
| 1642 |
$this->abbr_desciptions = array(); |
| 1643 |
$this->abbr_word_re = ''; |
| 1644 |
$this->footnote_counter = 1; |
| 1645 |
|
| 1646 |
foreach ($this->predef_abbr as $abbr_word => $abbr_desc) { |
| 1647 |
if ($this->abbr_word_re) |
| 1648 |
$this->abbr_word_re .= '|'; |
| 1649 |
$this->abbr_word_re .= preg_quote($abbr_word); |
| 1650 |
$this->abbr_desciptions[$abbr_word] = trim($abbr_desc); |
| 1651 |
} |
| 1652 |
} |
| 1653 |
|
| 1654 |
function teardown() { |
| 1655 |
# |
| 1656 |
# Clearing Extra-specific variables. |
| 1657 |
# |
| 1658 |
$this->footnotes = array(); |
| 1659 |
$this->footnotes_ordered = array(); |
| 1660 |
$this->footnotes_ref_count = array(); |
| 1661 |
$this->footnotes_numbers = array(); |
| 1662 |
$this->abbr_desciptions = array(); |
| 1663 |
$this->abbr_word_re = ''; |
| 1664 |
|
| 1665 |
parent::teardown(); |
| 1666 |
} |
| 1667 |
|
| 1668 |
|
| 1669 |
### Extra Attribute Parser ### |
| 1670 |
|
| 1671 |
# Expression to use to catch attributes (includes the braces) |
| 1672 |
public $id_class_attr_catch_re = '\{((?:[ ]*[#.][-_:a-zA-Z0-9]+){1,})[ ]*\}'; |
| 1673 |
# Expression to use when parsing in a context when no capture is desired |
| 1674 |
public $id_class_attr_nocatch_re = '\{(?:[ ]*[#.][-_:a-zA-Z0-9]+){1,}[ ]*\}'; |
| 1675 |
|
| 1676 |
function doExtraAttributes($tag_name, $attr) { |
| 1677 |
# |
| 1678 |
# Parse attributes caught by the $this->id_class_attr_catch_re expression |
| 1679 |
# and return the HTML-formatted list of attributes. |
| 1680 |
# |
| 1681 |
# Currently supported attributes are .class and #id. |
| 1682 |
# |
| 1683 |
if (empty($attr)) return ""; |
| 1684 |
|
| 1685 |
# Split on components |
| 1686 |
preg_match_all('/[#.][-_:a-zA-Z0-9]+/', $attr, $matches); |
| 1687 |
$elements = $matches[0]; |
| 1688 |
|
| 1689 |
# handle classes and ids (only first id taken into account) |
| 1690 |
$classes = array(); |
| 1691 |
$id = false; |
| 1692 |
foreach ($elements as $element) { |
| 1693 |
if ($element{0} == '.') { |
| 1694 |
$classes[] = substr($element, 1); |
| 1695 |
} else if ($element{0} == '#') { |
| 1696 |
if ($id === false) $id = substr($element, 1); |
| 1697 |
} |
| 1698 |
} |
| 1699 |
|
| 1700 |
# compose attributes as string |
| 1701 |
$attr_str = ""; |
| 1702 |
if (!empty($id)) { |
| 1703 |
$attr_str .= ' id="'.$id.'"'; |
| 1704 |
} |
| 1705 |
if (!empty($classes)) { |
| 1706 |
$attr_str .= ' class="'.implode(" ", $classes).'"'; |
| 1707 |
} |
| 1708 |
return $attr_str; |
| 1709 |
} |
| 1710 |
|
| 1711 |
|
| 1712 |
function stripLinkDefinitions($text) { |
| 1713 |
# |
| 1714 |
# Strips link definitions from text, stores the URLs and titles in |
| 1715 |
# hash references. |
| 1716 |
# |
| 1717 |
$less_than_tab = $this->tab_width - 1; |
| 1718 |
|
| 1719 |
# Link defs are in the form: ^[id]: url "optional title" |
| 1720 |
$text = preg_replace_callback('{ |
| 1721 |
^[ ]{0,'.$less_than_tab.'}\[(.+)\][ ]?: # id = $1 |
| 1722 |
[ ]* |
| 1723 |
\n? # maybe *one* newline |
| 1724 |
[ ]* |
| 1725 |
(?: |
| 1726 |
<(.+?)> # url = $2 |
| 1727 |
| |
| 1728 |
(\S+?) # url = $3 |
| 1729 |
) |
| 1730 |
[ ]* |
| 1731 |
\n? # maybe one newline |
| 1732 |
[ ]* |
| 1733 |
(?: |
| 1734 |
(?<=\s) # lookbehind for whitespace |
| 1735 |
["(] |
| 1736 |
(.*?) # title = $4 |
| 1737 |
[")] |
| 1738 |
[ ]* |
| 1739 |
)? # title is optional |
| 1740 |
(?:[ ]* '.$this->id_class_attr_catch_re.' )? # $5 = extra id & class attr |
| 1741 |
(?:\n+|\Z) |
| 1742 |
}xm', |
| 1743 |
array(&$this, '_stripLinkDefinitions_callback'), |
| 1744 |
$text); |
| 1745 |
return $text; |
| 1746 |
} |
| 1747 |
function _stripLinkDefinitions_callback($matches) { |
| 1748 |
$link_id = strtolower($matches[1]); |
| 1749 |
$url = $matches[2] == '' ? $matches[3] : $matches[2]; |
| 1750 |
$this->urls[$link_id] = $url; |
| 1751 |
$this->titles[$link_id] =& $matches[4]; |
| 1752 |
$this->ref_attr[$link_id] = $this->doExtraAttributes("", $dummy =& $matches[5]); |
| 1753 |
return ''; # String that will replace the block |
| 1754 |
} |
| 1755 |
|
| 1756 |
|
| 1757 |
### HTML Block Parser ### |
| 1758 |
|
| 1759 |
# Tags that are always treated as block tags: |
| 1760 |
public $block_tags_re = 'p|div|h[1-6]|blockquote|pre|table|dl|ol|ul|address|form|fieldset|iframe|hr|legend|article|section|nav|aside|hgroup|header|footer|figcaption'; |
| 1761 |
|
| 1762 |
# Tags treated as block tags only if the opening tag is alone on its line: |
| 1763 |
public $context_block_tags_re = 'script|noscript|ins|del|iframe|object|source|track|param|math|svg|canvas|audio|video'; |
| 1764 |
|
| 1765 |
# Tags where markdown="1" default to span mode: |
| 1766 |
public $contain_span_tags_re = 'p|h[1-6]|li|dd|dt|td|th|legend|address'; |
| 1767 |
|
| 1768 |
# Tags which must not have their contents modified, no matter where |
| 1769 |
# they appear: |
| 1770 |
public $clean_tags_re = 'script|math|svg'; |
| 1771 |
|
| 1772 |
# Tags that do not need to be closed. |
| 1773 |
public $auto_close_tags_re = 'hr|img|param|source|track'; |
| 1774 |
|
| 1775 |
|
| 1776 |
function hashHTMLBlocks($text) { |
| 1777 |
# |
| 1778 |
# Hashify HTML Blocks and "clean tags". |
| 1779 |
# |
| 1780 |
# We only want to do this for block-level HTML tags, such as headers, |
| 1781 |
# lists, and tables. That's because we still want to wrap <p>s around |
| 1782 |
# "paragraphs" that are wrapped in non-block-level tags, such as anchors, |
| 1783 |
# phrase emphasis, and spans. The list of tags we're looking for is |
| 1784 |
# hard-coded. |
| 1785 |
# |
| 1786 |
# This works by calling _HashHTMLBlocks_InMarkdown, which then calls |
| 1787 |
# _HashHTMLBlocks_InHTML when it encounter block tags. When the markdown="1" |
| 1788 |
# attribute is found within a tag, _HashHTMLBlocks_InHTML calls back |
| 1789 |
# _HashHTMLBlocks_InMarkdown to handle the Markdown syntax within the tag. |
| 1790 |
# These two functions are calling each other. It's recursive! |
| 1791 |
# |
| 1792 |
if ($this->no_markup) return $text; |
| 1793 |
|
| 1794 |
# |
| 1795 |
# Call the HTML-in-Markdown hasher. |
| 1796 |
# |
| 1797 |
list($text, ) = $this->_hashHTMLBlocks_inMarkdown($text); |
| 1798 |
|
| 1799 |
return $text; |
| 1800 |
} |
| 1801 |
function _hashHTMLBlocks_inMarkdown($text, $indent = 0, |
| 1802 |
$enclosing_tag_re = '', $span = false) |
| 1803 |
{ |
| 1804 |
# |
| 1805 |
# Parse markdown text, calling _HashHTMLBlocks_InHTML for block tags. |
| 1806 |
# |
| 1807 |
# * $indent is the number of space to be ignored when checking for code |
| 1808 |
# blocks. This is important because if we don't take the indent into |
| 1809 |
# account, something like this (which looks right) won't work as expected: |
| 1810 |
# |
| 1811 |
# <div> |
| 1812 |
# <div markdown="1"> |
| 1813 |
# Hello World. <-- Is this a Markdown code block or text? |
| 1814 |
# </div> <-- Is this a Markdown code block or a real tag? |
| 1815 |
# <div> |
| 1816 |
# |
| 1817 |
# If you don't like this, just don't indent the tag on which |
| 1818 |
# you apply the markdown="1" attribute. |
| 1819 |
# |
| 1820 |
# * If $enclosing_tag_re is not empty, stops at the first unmatched closing |
| 1821 |
# tag with that name. Nested tags supported. |
| 1822 |
# |
| 1823 |
# * If $span is true, text inside must treated as span. So any double |
| 1824 |
# newline will be replaced by a single newline so that it does not create |
| 1825 |
# paragraphs. |
| 1826 |
# |
| 1827 |
# Returns an array of that form: ( processed text , remaining text ) |
| 1828 |
# |
| 1829 |
if ($text === '') return array('', ''); |
| 1830 |
|
| 1831 |
# Regex to check for the presence of newlines around a block tag. |
| 1832 |
$newline_before_re = '/(?:^\n?|\n\n)*$/'; |
| 1833 |
$newline_after_re = |
| 1834 |
'{ |
| 1835 |
^ # Start of text following the tag. |
| 1836 |
(?>[ ]*<!--.*?-->)? # Optional comment. |
| 1837 |
[ ]*\n # Must be followed by newline. |
| 1838 |
}xs'; |
| 1839 |
|
| 1840 |
# Regex to match any tag. |
| 1841 |
$block_tag_re = |
| 1842 |
'{ |
| 1843 |
( # $2: Capture whole tag. |
| 1844 |
</? # Any opening or closing tag. |
| 1845 |
(?> # Tag name. |
| 1846 |
'.$this->block_tags_re.' | |
| 1847 |
'.$this->context_block_tags_re.' | |
| 1848 |
'.$this->clean_tags_re.' | |
| 1849 |
(?!\s)'.$enclosing_tag_re.' |
| 1850 |
) |
| 1851 |
(?: |
| 1852 |
(?=[\s"\'/a-zA-Z0-9]) # Allowed characters after tag name. |
| 1853 |
(?> |
| 1854 |
".*?" | # Double quotes (can contain `>`) |
| 1855 |
\'.*?\' | # Single quotes (can contain `>`) |
| 1856 |
.+? # Anything but quotes and `>`. |
| 1857 |
)*? |
| 1858 |
)? |
| 1859 |
> # End of tag. |
| 1860 |
| |
| 1861 |
<!-- .*? --> # HTML Comment |
| 1862 |
| |
| 1863 |
<\?.*?\?> | <%.*?%> # Processing instruction |
| 1864 |
| |
| 1865 |
<!\[CDATA\[.*?\]\]> # CData Block |
| 1866 |
'. ( !$span ? ' # If not in span. |
| 1867 |
| |
| 1868 |
# Indented code block |
| 1869 |
(?: ^[ ]*\n | ^ | \n[ ]*\n ) |
| 1870 |
[ ]{'.($indent+4).'}[^\n]* \n |
| 1871 |
(?> |
| 1872 |
(?: [ ]{'.($indent+4).'}[^\n]* | [ ]* ) \n |
| 1873 |
)* |
| 1874 |
| |
| 1875 |
# Fenced code block marker |
| 1876 |
(?<= ^ | \n ) |
| 1877 |
[ ]{0,'.($indent+3).'}(?:~{3,}|`{3,}) |
| 1878 |
[ ]* |
| 1879 |
(?: |
| 1880 |
\.?[-_:a-zA-Z0-9]+ # standalone class name |
| 1881 |
| |
| 1882 |
'.$this->id_class_attr_nocatch_re.' # extra attributes |
| 1883 |
)? |
| 1884 |
[ ]* |
| 1885 |
(?= \n ) |
| 1886 |
' : '' ). ' # End (if not is span). |
| 1887 |
| |
| 1888 |
# Code span marker |
| 1889 |
# Note, this regex needs to go after backtick fenced |
| 1890 |
# code blocks but it should also be kept outside of the |
| 1891 |
# "if not in span" condition adding backticks to the parser |
| 1892 |
`+ |
| 1893 |
) |
| 1894 |
}xs'; |
| 1895 |
|
| 1896 |
|
| 1897 |
$depth = 0; # Current depth inside the tag tree. |
| 1898 |
$parsed = ""; # Parsed text that will be returned. |
| 1899 |
|
| 1900 |
# |
| 1901 |
# Loop through every tag until we find the closing tag of the parent |
| 1902 |
# or loop until reaching the end of text if no parent tag specified. |
| 1903 |
# |
| 1904 |
do { |
| 1905 |
# |
| 1906 |
# Split the text using the first $tag_match pattern found. |
| 1907 |
# Text before pattern will be first in the array, text after |
| 1908 |
# pattern will be at the end, and between will be any catches made |
| 1909 |
# by the pattern. |
| 1910 |
# |
| 1911 |
$parts = preg_split($block_tag_re, $text, 2, |
| 1912 |
PREG_SPLIT_DELIM_CAPTURE); |
| 1913 |
|
| 1914 |
# If in Markdown span mode, add a empty-string span-level hash |
| 1915 |
# after each newline to prevent triggering any block element. |
| 1916 |
if ($span) { |
| 1917 |
$void = $this->hashPart("", ':'); |
| 1918 |
$newline = "$void\n"; |
| 1919 |
$parts[0] = $void . str_replace("\n", $newline, $parts[0]) . $void; |
| 1920 |
} |
| 1921 |
|
| 1922 |
$parsed .= $parts[0]; # Text before current tag. |
| 1923 |
|
| 1924 |
# If end of $text has been reached. Stop loop. |
| 1925 |
if (count($parts) < 3) { |
| 1926 |
$text = ""; |
| 1927 |
break; |
| 1928 |
} |
| 1929 |
|
| 1930 |
$tag = $parts[1]; # Tag to handle. |
| 1931 |
$text = $parts[2]; # Remaining text after current tag. |
| 1932 |
$tag_re = preg_quote($tag); # For use in a regular expression. |
| 1933 |
|
| 1934 |
# |
| 1935 |
# Check for: Fenced code block marker. |
| 1936 |
# Note: need to recheck the whole tag to disambiguate backtick |
| 1937 |
# fences from code spans |
| 1938 |
# |
| 1939 |
if (preg_match('{^\n?([ ]{0,'.($indent+3).'})(~{3,}|`{3,})[ ]*(?:\.?[-_:a-zA-Z0-9]+|'.$this->id_class_attr_nocatch_re.')?[ ]*\n?$}', $tag, $capture)) { |
| 1940 |
# Fenced code block marker: find matching end marker. |
| 1941 |
$fence_indent = strlen($capture[1]); # use captured indent in re |
| 1942 |
$fence_re = $capture[2]; # use captured fence in re |
| 1943 |
if (preg_match('{^(?>.*\n)*?[ ]{'.($fence_indent).'}'.$fence_re.'[ ]*(?:\n|$)}', $text, |
| 1944 |
$matches)) |
| 1945 |
{ |
| 1946 |
# End marker found: pass text unchanged until marker. |
| 1947 |
$parsed .= $tag . $matches[0]; |
| 1948 |
$text = substr($text, strlen($matches[0])); |
| 1949 |
} |
| 1950 |
else { |
| 1951 |
# No end marker: just skip it. |
| 1952 |
$parsed .= $tag; |
| 1953 |
} |
| 1954 |
} |
| 1955 |
# |
| 1956 |
# Check for: Indented code block. |
| 1957 |
# |
| 1958 |
else if ($tag{0} == "\n" || $tag{0} == " ") { |
| 1959 |
# Indented code block: pass it unchanged, will be handled |
| 1960 |
# later. |
| 1961 |
$parsed .= $tag; |
| 1962 |
} |
| 1963 |
# |
| 1964 |
# Check for: Code span marker |
| 1965 |
# Note: need to check this after backtick fenced code blocks |
| 1966 |
# |
| 1967 |
else if ($tag{0} == "`") { |
| 1968 |
# Find corresponding end marker. |
| 1969 |
$tag_re = preg_quote($tag); |
| 1970 |
if (preg_match('{^(?>.+?|\n(?!\n))*?(?<!`)'.$tag_re.'(?!`)}', |
| 1971 |
$text, $matches)) |
| 1972 |
{ |
| 1973 |
# End marker found: pass text unchanged until marker. |
| 1974 |
$parsed .= $tag . $matches[0]; |
| 1975 |
$text = substr($text, strlen($matches[0])); |
| 1976 |
} |
| 1977 |
else { |
| 1978 |
# Unmatched marker: just skip it. |
| 1979 |
$parsed .= $tag; |
| 1980 |
} |
| 1981 |
} |
| 1982 |
# |
| 1983 |
# Check for: Opening Block level tag or |
| 1984 |
# Opening Context Block tag (like ins and del) |
| 1985 |
# used as a block tag (tag is alone on it's line). |
| 1986 |
# |
| 1987 |
else if (preg_match('{^<(?:'.$this->block_tags_re.')\b}', $tag) || |
| 1988 |
( preg_match('{^<(?:'.$this->context_block_tags_re.')\b}', $tag) && |
| 1989 |
preg_match($newline_before_re, $parsed) && |
| 1990 |
preg_match($newline_after_re, $text) ) |
| 1991 |
) |
| 1992 |
{ |
| 1993 |
# Need to parse tag and following text using the HTML parser. |
| 1994 |
list($block_text, $text) = |
| 1995 |
$this->_hashHTMLBlocks_inHTML($tag . $text, "hashBlock", true); |
| 1996 |
|
| 1997 |
# Make sure it stays outside of any paragraph by adding newlines. |
| 1998 |
$parsed .= "\n\n$block_text\n\n"; |
| 1999 |
} |
| 2000 |
# |
| 2001 |
# Check for: Clean tag (like script, math) |
| 2002 |
# HTML Comments, processing instructions. |
| 2003 |
# |
| 2004 |
else if (preg_match('{^<(?:'.$this->clean_tags_re.')\b}', $tag) || |
| 2005 |
$tag{1} == '!' || $tag{1} == '?') |
| 2006 |
{ |
| 2007 |
# Need to parse tag and following text using the HTML parser. |
| 2008 |
# (don't check for markdown attribute) |
| 2009 |
list($block_text, $text) = |
| 2010 |
$this->_hashHTMLBlocks_inHTML($tag . $text, "hashClean", false); |
| 2011 |
|
| 2012 |
$parsed .= $block_text; |
| 2013 |
} |
| 2014 |
# |
| 2015 |
# Check for: Tag with same name as enclosing tag. |
| 2016 |
# |
| 2017 |
else if ($enclosing_tag_re !== '' && |
| 2018 |
# Same name as enclosing tag. |
| 2019 |
preg_match('{^</?(?:'.$enclosing_tag_re.')\b}', $tag)) |
| 2020 |
{ |
| 2021 |
# |
| 2022 |
# Increase/decrease nested tag count. |
| 2023 |
# |
| 2024 |
if ($tag{1} == '/') $depth--; |
| 2025 |
else if ($tag{strlen($tag)-2} != '/') $depth++; |
| 2026 |
|
| 2027 |
if ($depth < 0) { |
| 2028 |
# |
| 2029 |
# Going out of parent element. Clean up and break so we |
| 2030 |
# return to the calling function. |
| 2031 |
# |
| 2032 |
$text = $tag . $text; |
| 2033 |
break; |
| 2034 |
} |
| 2035 |
|
| 2036 |
$parsed .= $tag; |
| 2037 |
} |
| 2038 |
else { |
| 2039 |
$parsed .= $tag; |
| 2040 |
} |
| 2041 |
} while ($depth >= 0); |
| 2042 |
|
| 2043 |
return array($parsed, $text); |
| 2044 |
} |
| 2045 |
function _hashHTMLBlocks_inHTML($text, $hash_method, $md_attr) { |
| 2046 |
# |
| 2047 |
# Parse HTML, calling _HashHTMLBlocks_InMarkdown for block tags. |
| 2048 |
# |
| 2049 |
# * Calls $hash_method to convert any blocks. |
| 2050 |
# * Stops when the first opening tag closes. |
| 2051 |
# * $md_attr indicate if the use of the `markdown="1"` attribute is allowed. |
| 2052 |
# (it is not inside clean tags) |
| 2053 |
# |
| 2054 |
# Returns an array of that form: ( processed text , remaining text ) |
| 2055 |
# |
| 2056 |
if ($text === '') return array('', ''); |
| 2057 |
|
| 2058 |
# Regex to match `markdown` attribute inside of a tag. |
| 2059 |
$markdown_attr_re = ' |
| 2060 |
{ |
| 2061 |
\s* # Eat whitespace before the `markdown` attribute |
| 2062 |
markdown |
| 2063 |
\s*=\s* |
| 2064 |
(?> |
| 2065 |
(["\']) # $1: quote delimiter |
| 2066 |
(.*?) # $2: attribute value |
| 2067 |
\1 # matching delimiter |
| 2068 |
| |
| 2069 |
([^\s>]*) # $3: unquoted attribute value |
| 2070 |
) |
| 2071 |
() # $4: make $3 always defined (avoid warnings) |
| 2072 |
}xs'; |
| 2073 |
|
| 2074 |
# Regex to match any tag. |
| 2075 |
$tag_re = '{ |
| 2076 |
( # $2: Capture whole tag. |
| 2077 |
</? # Any opening or closing tag. |
| 2078 |
[\w:$]+ # Tag name. |
| 2079 |
(?: |
| 2080 |
(?=[\s"\'/a-zA-Z0-9]) # Allowed characters after tag name. |
| 2081 |
(?> |
| 2082 |
".*?" | # Double quotes (can contain `>`) |
| 2083 |
\'.*?\' | # Single quotes (can contain `>`) |
| 2084 |
.+? # Anything but quotes and `>`. |
| 2085 |
)*? |
| 2086 |
)? |
| 2087 |
> # End of tag. |
| 2088 |
| |
| 2089 |
<!-- .*? --> # HTML Comment |
| 2090 |
| |
| 2091 |
<\?.*?\?> | <%.*?%> # Processing instruction |
| 2092 |
| |
| 2093 |
<!\[CDATA\[.*?\]\]> # CData Block |
| 2094 |
) |
| 2095 |
}xs'; |
| 2096 |
|
| 2097 |
$original_text = $text; # Save original text in case of faliure. |
| 2098 |
|
| 2099 |
$depth = 0; # Current depth inside the tag tree. |
| 2100 |
$block_text = ""; # Temporary text holder for current text. |
| 2101 |
$parsed = ""; # Parsed text that will be returned. |
| 2102 |
|
| 2103 |
# |
| 2104 |
# Get the name of the starting tag. |
| 2105 |
# (This pattern makes $base_tag_name_re safe without quoting.) |
| 2106 |
# |
| 2107 |
if (preg_match('/^<([\w:$]*)\b/', $text, $matches)) |
| 2108 |
$base_tag_name_re = $matches[1]; |
| 2109 |
|
| 2110 |
# |
| 2111 |
# Loop through every tag until we find the corresponding closing tag. |
| 2112 |
# |
| 2113 |
do { |
| 2114 |
# |
| 2115 |
# Split the text using the first $tag_match pattern found. |
| 2116 |
# Text before pattern will be first in the array, text after |
| 2117 |
# pattern will be at the end, and between will be any catches made |
| 2118 |
# by the pattern. |
| 2119 |
# |
| 2120 |
$parts = preg_split($tag_re, $text, 2, PREG_SPLIT_DELIM_CAPTURE); |
| 2121 |
|
| 2122 |
if (count($parts) < 3) { |
| 2123 |
# |
| 2124 |
# End of $text reached with unbalenced tag(s). |
| 2125 |
# In that case, we return original text unchanged and pass the |
| 2126 |
# first character as filtered to prevent an infinite loop in the |
| 2127 |
# parent function. |
| 2128 |
# |
| 2129 |
return array($original_text{0}, substr($original_text, 1)); |
| 2130 |
} |
| 2131 |
|
| 2132 |
$block_text .= $parts[0]; # Text before current tag. |
| 2133 |
$tag = $parts[1]; # Tag to handle. |
| 2134 |
$text = $parts[2]; # Remaining text after current tag. |
| 2135 |
|
| 2136 |
# |
| 2137 |
# Check for: Auto-close tag (like <hr/>) |
| 2138 |
# Comments and Processing Instructions. |
| 2139 |
# |
| 2140 |
if (preg_match('{^</?(?:'.$this->auto_close_tags_re.')\b}', $tag) || |
| 2141 |
$tag{1} == '!' || $tag{1} == '?') |
| 2142 |
{ |
| 2143 |
# Just add the tag to the block as if it was text. |
| 2144 |
$block_text .= $tag; |
| 2145 |
} |
| 2146 |
else { |
| 2147 |
# |
| 2148 |
# Increase/decrease nested tag count. Only do so if |
| 2149 |
# the tag's name match base tag's. |
| 2150 |
# |
| 2151 |
if (preg_match('{^</?'.$base_tag_name_re.'\b}', $tag)) { |
| 2152 |
if ($tag{1} == '/') $depth--; |
| 2153 |
else if ($tag{strlen($tag)-2} != '/') $depth++; |
| 2154 |
} |
| 2155 |
|
| 2156 |
# |
| 2157 |
# Check for `markdown="1"` attribute and handle it. |
| 2158 |
# |
| 2159 |
if ($md_attr && |
| 2160 |
preg_match($markdown_attr_re, $tag, $attr_m) && |
| 2161 |
preg_match('/^1|block|span$/', $attr_m[2] . $attr_m[3])) |
| 2162 |
{ |
| 2163 |
# Remove `markdown` attribute from opening tag. |
| 2164 |
$tag = preg_replace($markdown_attr_re, '', $tag); |
| 2165 |
|
| 2166 |
# Check if text inside this tag must be parsed in span mode. |
| 2167 |
$this->mode = $attr_m[2] . $attr_m[3]; |
| 2168 |
$span_mode = $this->mode == 'span' || $this->mode != 'block' && |
| 2169 |
preg_match('{^<(?:'.$this->contain_span_tags_re.')\b}', $tag); |
| 2170 |
|
| 2171 |
# Calculate indent before tag. |
| 2172 |
if (preg_match('/(?:^|\n)( *?)(?! ).*?$/', $block_text, $matches)) { |
| 2173 |
$strlen = $this->utf8_strlen; |
| 2174 |
$indent = $strlen($matches[1], 'UTF-8'); |
| 2175 |
} else { |
| 2176 |
$indent = 0; |
| 2177 |
} |
| 2178 |
|
| 2179 |
# End preceding block with this tag. |
| 2180 |
$block_text .= $tag; |
| 2181 |
$parsed .= $this->$hash_method($block_text); |
| 2182 |
|
| 2183 |
# Get enclosing tag name for the ParseMarkdown function. |
| 2184 |
# (This pattern makes $tag_name_re safe without quoting.) |
| 2185 |
preg_match('/^<([\w:$]*)\b/', $tag, $matches); |
| 2186 |
$tag_name_re = $matches[1]; |
| 2187 |
|
| 2188 |
# Parse the content using the HTML-in-Markdown parser. |
| 2189 |
list ($block_text, $text) |
| 2190 |
= $this->_hashHTMLBlocks_inMarkdown($text, $indent, |
| 2191 |
$tag_name_re, $span_mode); |
| 2192 |
|
| 2193 |
# Outdent markdown text. |
| 2194 |
if ($indent > 0) { |
| 2195 |
$block_text = preg_replace("/^[ ]{1,$indent}/m", "", |
| 2196 |
$block_text); |
| 2197 |
} |
| 2198 |
|
| 2199 |
# Append tag content to parsed text. |
| 2200 |
if (!$span_mode) $parsed .= "\n\n$block_text\n\n"; |
| 2201 |
else $parsed .= "$block_text"; |
| 2202 |
|
| 2203 |
# Start over with a new block. |
| 2204 |
$block_text = ""; |
| 2205 |
} |
| 2206 |
else $block_text .= $tag; |
| 2207 |
} |
| 2208 |
|
| 2209 |
} while ($depth > 0); |
| 2210 |
|
| 2211 |
# |
| 2212 |
# Hash last block text that wasn't processed inside the loop. |
| 2213 |
# |
| 2214 |
$parsed .= $this->$hash_method($block_text); |
| 2215 |
|
| 2216 |
return array($parsed, $text); |
| 2217 |
} |
| 2218 |
|
| 2219 |
|
| 2220 |
function hashClean($text) { |
| 2221 |
# |
| 2222 |
# Called whenever a tag must be hashed when a function inserts a "clean" tag |
| 2223 |
# in $text, it passes through this function and is automaticaly escaped, |
| 2224 |
# blocking invalid nested overlap. |
| 2225 |
# |
| 2226 |
return $this->hashPart($text, 'C'); |
| 2227 |
} |
| 2228 |
|
| 2229 |
|
| 2230 |
function doAnchors($text) { |
| 2231 |
# |
| 2232 |
# Turn Markdown link shortcuts into XHTML <a> tags. |
| 2233 |
# |
| 2234 |
if ($this->in_anchor) return $text; |
| 2235 |
$this->in_anchor = true; |
| 2236 |
|
| 2237 |
# |
| 2238 |
# First, handle reference-style links: [link text] [id] |
| 2239 |
# |
| 2240 |
$text = preg_replace_callback('{ |
| 2241 |
( # wrap whole match in $1 |
| 2242 |
\[ |
| 2243 |
('.$this->nested_brackets_re.') # link text = $2 |
| 2244 |
\] |
| 2245 |
|
| 2246 |
[ ]? # one optional space |
| 2247 |
(?:\n[ ]*)? # one optional newline followed by spaces |
| 2248 |
|
| 2249 |
\[ |
| 2250 |
(.*?) # id = $3 |
| 2251 |
\] |
| 2252 |
) |
| 2253 |
}xs', |
| 2254 |
array(&$this, '_doAnchors_reference_callback'), $text); |
| 2255 |
|
| 2256 |
# |
| 2257 |
# Next, inline-style links: [link text](url "optional title") |
| 2258 |
# |
| 2259 |
$text = preg_replace_callback('{ |
| 2260 |
( # wrap whole match in $1 |
| 2261 |
\[ |
| 2262 |
('.$this->nested_brackets_re.') # link text = $2 |
| 2263 |
\] |
| 2264 |
\( # literal paren |
| 2265 |
[ \n]* |
| 2266 |
(?: |
| 2267 |
<(.+?)> # href = $3 |
| 2268 |
| |
| 2269 |
('.$this->nested_url_parenthesis_re.') # href = $4 |
| 2270 |
) |
| 2271 |
[ \n]* |
| 2272 |
( # $5 |
| 2273 |
([\'"]) # quote char = $6 |
| 2274 |
(.*?) # Title = $7 |
| 2275 |
\6 # matching quote |
| 2276 |
[ \n]* # ignore any spaces/tabs between closing quote and ) |
| 2277 |
)? # title is optional |
| 2278 |
\) |
| 2279 |
(?:[ ]? '.$this->id_class_attr_catch_re.' )? # $8 = id/class attributes |
| 2280 |
) |
| 2281 |
}xs', |
| 2282 |
array(&$this, '_doAnchors_inline_callback'), $text); |
| 2283 |
|
| 2284 |
# |
| 2285 |
# Last, handle reference-style shortcuts: [link text] |
| 2286 |
# These must come last in case you've also got [link text][1] |
| 2287 |
# or [link text](/foo) |
| 2288 |
# |
| 2289 |
$text = preg_replace_callback('{ |
| 2290 |
( # wrap whole match in $1 |
| 2291 |
\[ |
| 2292 |
([^\[\]]+) # link text = $2; can\'t contain [ or ] |
| 2293 |
\] |
| 2294 |
) |
| 2295 |
}xs', |
| 2296 |
array(&$this, '_doAnchors_reference_callback'), $text); |
| 2297 |
|
| 2298 |
$this->in_anchor = false; |
| 2299 |
return $text; |
| 2300 |
} |
| 2301 |
function _doAnchors_reference_callback($matches) { |
| 2302 |
$whole_match = $matches[1]; |
| 2303 |
$link_text = $matches[2]; |
| 2304 |
$link_id =& $matches[3]; |
| 2305 |
|
| 2306 |
if ($link_id == "") { |
| 2307 |
# for shortcut links like [this][] or [this]. |
| 2308 |
$link_id = $link_text; |
| 2309 |
} |
| 2310 |
|
| 2311 |
# lower-case and turn embedded newlines into spaces |
| 2312 |
$link_id = strtolower($link_id); |
| 2313 |
$link_id = preg_replace('{[ ]?\n}', ' ', $link_id); |
| 2314 |
|
| 2315 |
if (isset($this->urls[$link_id])) { |
| 2316 |
$url = $this->urls[$link_id]; |
| 2317 |
$url = $this->encodeAttribute($url); |
| 2318 |
|
| 2319 |
$result = "<a href=\"$url\""; |
| 2320 |
if ( isset( $this->titles[$link_id] ) ) { |
| 2321 |
$title = $this->titles[$link_id]; |
| 2322 |
$title = $this->encodeAttribute($title); |
| 2323 |
$result .= " title=\"$title\""; |
| 2324 |
} |
| 2325 |
if (isset($this->ref_attr[$link_id])) |
| 2326 |
$result .= $this->ref_attr[$link_id]; |
| 2327 |
|
| 2328 |
$link_text = $this->runSpanGamut($link_text); |
| 2329 |
$result .= ">$link_text</a>"; |
| 2330 |
$result = $this->hashPart($result); |
| 2331 |
} |
| 2332 |
else { |
| 2333 |
$result = $whole_match; |
| 2334 |
} |
| 2335 |
return $result; |
| 2336 |
} |
| 2337 |
function _doAnchors_inline_callback($matches) { |
| 2338 |
$whole_match = $matches[1]; |
| 2339 |
$link_text = $this->runSpanGamut($matches[2]); |
| 2340 |
$url = $matches[3] == '' ? $matches[4] : $matches[3]; |
| 2341 |
$title =& $matches[7]; |
| 2342 |
$attr = $this->doExtraAttributes("a", $dummy =& $matches[8]); |
| 2343 |
|
| 2344 |
|
| 2345 |
$url = $this->encodeAttribute($url); |
| 2346 |
|
| 2347 |
$result = "<a href=\"$url\""; |
| 2348 |
if (isset($title)) { |
| 2349 |
$title = $this->encodeAttribute($title); |
| 2350 |
$result .= " title=\"$title\""; |
| 2351 |
} |
| 2352 |
$result .= $attr; |
| 2353 |
|
| 2354 |
$link_text = $this->runSpanGamut($link_text); |
| 2355 |
$result .= ">$link_text</a>"; |
| 2356 |
|
| 2357 |
return $this->hashPart($result); |
| 2358 |
} |
| 2359 |
|
| 2360 |
|
| 2361 |
function doImages($text) { |
| 2362 |
# |
| 2363 |
# Turn Markdown image shortcuts into <img> tags. |
| 2364 |
# |
| 2365 |
# |
| 2366 |
# First, handle reference-style labeled images: ![alt text][id] |
| 2367 |
# |
| 2368 |
$text = preg_replace_callback('{ |
| 2369 |
( # wrap whole match in $1 |
| 2370 |
!\[ |
| 2371 |
('.$this->nested_brackets_re.') # alt text = $2 |
| 2372 |
\] |
| 2373 |
|
| 2374 |
[ ]? # one optional space |
| 2375 |
(?:\n[ ]*)? # one optional newline followed by spaces |
| 2376 |
|
| 2377 |
\[ |
| 2378 |
(.*?) # id = $3 |
| 2379 |
\] |
| 2380 |
|
| 2381 |
) |
| 2382 |
}xs', |
| 2383 |
array(&$this, '_doImages_reference_callback'), $text); |
| 2384 |
|
| 2385 |
# |
| 2386 |
# Next, handle inline images:  |
| 2387 |
# Don't forget: encode * and _ |
| 2388 |
# |
| 2389 |
$text = preg_replace_callback('{ |
| 2390 |
( # wrap whole match in $1 |
| 2391 |
!\[ |
| 2392 |
('.$this->nested_brackets_re.') # alt text = $2 |
| 2393 |
\] |
| 2394 |
\s? # One optional whitespace character |
| 2395 |
\( # literal paren |
| 2396 |
[ \n]* |
| 2397 |
(?: |
| 2398 |
<(\S*)> # src url = $3 |
| 2399 |
| |
| 2400 |
('.$this->nested_url_parenthesis_re.') # src url = $4 |
| 2401 |
) |
| 2402 |
[ \n]* |
| 2403 |
( # $5 |
| 2404 |
([\'"]) # quote char = $6 |
| 2405 |
(.*?) # title = $7 |
| 2406 |
\6 # matching quote |
| 2407 |
[ \n]* |
| 2408 |
)? # title is optional |
| 2409 |
\) |
| 2410 |
(?:[ ]? '.$this->id_class_attr_catch_re.' )? # $8 = id/class attributes |
| 2411 |
) |
| 2412 |
}xs', |
| 2413 |
array(&$this, '_doImages_inline_callback'), $text); |
| 2414 |
|
| 2415 |
return $text; |
| 2416 |
} |
| 2417 |
function _doImages_reference_callback($matches) { |
| 2418 |
$whole_match = $matches[1]; |
| 2419 |
$alt_text = $matches[2]; |
| 2420 |
$link_id = strtolower($matches[3]); |
| 2421 |
|
| 2422 |
if ($link_id == "") { |
| 2423 |
$link_id = strtolower($alt_text); # for shortcut links like ![this][]. |
| 2424 |
} |
| 2425 |
|
| 2426 |
$alt_text = $this->encodeAttribute($alt_text); |
| 2427 |
if (isset($this->urls[$link_id])) { |
| 2428 |
$url = $this->encodeAttribute($this->urls[$link_id]); |
| 2429 |
$result = "<img src=\"$url\" alt=\"$alt_text\""; |
| 2430 |
if (isset($this->titles[$link_id])) { |
| 2431 |
$title = $this->titles[$link_id]; |
| 2432 |
$title = $this->encodeAttribute($title); |
| 2433 |
$result .= " title=\"$title\""; |
| 2434 |
} |
| 2435 |
if (isset($this->ref_attr[$link_id])) |
| 2436 |
$result .= $this->ref_attr[$link_id]; |
| 2437 |
$result .= $this->empty_element_suffix; |
| 2438 |
$result = $this->hashPart($result); |
| 2439 |
} |
| 2440 |
else { |
| 2441 |
# If there's no such link ID, leave intact: |
| 2442 |
$result = $whole_match; |
| 2443 |
} |
| 2444 |
|
| 2445 |
return $result; |
| 2446 |
} |
| 2447 |
function _doImages_inline_callback($matches) { |
| 2448 |
$whole_match = $matches[1]; |
| 2449 |
$alt_text = $matches[2]; |
| 2450 |
$url = $matches[3] == '' ? $matches[4] : $matches[3]; |
| 2451 |
$title =& $matches[7]; |
| 2452 |
$attr = $this->doExtraAttributes("img", $dummy =& $matches[8]); |
| 2453 |
|
| 2454 |
$alt_text = $this->encodeAttribute($alt_text); |
| 2455 |
$url = $this->encodeAttribute($url); |
| 2456 |
$result = "<img src=\"$url\" alt=\"$alt_text\""; |
| 2457 |
if (isset($title)) { |
| 2458 |
$title = $this->encodeAttribute($title); |
| 2459 |
$result .= " title=\"$title\""; # $title already quoted |
| 2460 |
} |
| 2461 |
$result .= $attr; |
| 2462 |
$result .= $this->empty_element_suffix; |
| 2463 |
|
| 2464 |
return $this->hashPart($result); |
| 2465 |
} |
| 2466 |
|
| 2467 |
|
| 2468 |
function doHeaders($text) { |
| 2469 |
# |
| 2470 |
# Redefined to add id and class attribute support. |
| 2471 |
# |
| 2472 |
# Setext-style headers: |
| 2473 |
# Header 1 {#header1} |
| 2474 |
# ======== |
| 2475 |
# |
| 2476 |
# Header 2 {#header2 .class1 .class2} |
| 2477 |
# -------- |
| 2478 |
# |
| 2479 |
$text = preg_replace_callback( |
| 2480 |
'{ |
| 2481 |
(^.+?) # $1: Header text |
| 2482 |
(?:[ ]+ '.$this->id_class_attr_catch_re.' )? # $3 = id/class attributes |
| 2483 |
[ ]*\n(=+|-+)[ ]*\n+ # $3: Header footer |
| 2484 |
}mx', |
| 2485 |
array(&$this, '_doHeaders_callback_setext'), $text); |
| 2486 |
|
| 2487 |
# atx-style headers: |
| 2488 |
# # Header 1 {#header1} |
| 2489 |
# ## Header 2 {#header2} |
| 2490 |
# ## Header 2 with closing hashes ## {#header3.class1.class2} |
| 2491 |
# ... |
| 2492 |
# ###### Header 6 {.class2} |
| 2493 |
# |
| 2494 |
$text = preg_replace_callback('{ |
| 2495 |
^(\#{1,6}) # $1 = string of #\'s |
| 2496 |
[ ]* |
| 2497 |
(.+?) # $2 = Header text |
| 2498 |
[ ]* |
| 2499 |
\#* # optional closing #\'s (not counted) |
| 2500 |
(?:[ ]+ '.$this->id_class_attr_catch_re.' )? # $3 = id/class attributes |
| 2501 |
[ ]* |
| 2502 |
\n+ |
| 2503 |
}xm', |
| 2504 |
array(&$this, '_doHeaders_callback_atx'), $text); |
| 2505 |
|
| 2506 |
return $text; |
| 2507 |
} |
| 2508 |
function _doHeaders_callback_setext($matches) { |
| 2509 |
if ($matches[3] == '-' && preg_match('{^- }', $matches[1])) |
| 2510 |
return $matches[0]; |
| 2511 |
$level = $matches[3]{0} == '=' ? 1 : 2; |
| 2512 |
$attr = $this->doExtraAttributes("h$level", $dummy =& $matches[2]); |
| 2513 |
$block = "<h$level$attr>".$this->runSpanGamut($matches[1])."</h$level>"; |
| 2514 |
return "\n" . $this->hashBlock($block) . "\n\n"; |
| 2515 |
} |
| 2516 |
function _doHeaders_callback_atx($matches) { |
| 2517 |
$level = strlen($matches[1]); |
| 2518 |
$attr = $this->doExtraAttributes("h$level", $dummy =& $matches[3]); |
| 2519 |
$block = "<h$level$attr>".$this->runSpanGamut($matches[2])."</h$level>"; |
| 2520 |
return "\n" . $this->hashBlock($block) . "\n\n"; |
| 2521 |
} |
| 2522 |
|
| 2523 |
|
| 2524 |
function doTables($text) { |
| 2525 |
# |
| 2526 |
# Form HTML tables. |
| 2527 |
# |
| 2528 |
$less_than_tab = $this->tab_width - 1; |
| 2529 |
# |
| 2530 |
# Find tables with leading pipe. |
| 2531 |
# |
| 2532 |
# | Header 1 | Header 2 |
| 2533 |
# | -------- | -------- |
| 2534 |
# | Cell 1 | Cell 2 |
| 2535 |
# | Cell 3 | Cell 4 |
| 2536 |
# |
| 2537 |
$text = preg_replace_callback(' |
| 2538 |
{ |
| 2539 |
^ # Start of a line |
| 2540 |
[ ]{0,'.$less_than_tab.'} # Allowed whitespace. |
| 2541 |
[|] # Optional leading pipe (present) |
| 2542 |
(.+) \n # $1: Header row (at least one pipe) |
| 2543 |
|
| 2544 |
[ ]{0,'.$less_than_tab.'} # Allowed whitespace. |
| 2545 |
[|] ([ ]*[-:]+[-| :]*) \n # $2: Header underline |
| 2546 |
|
| 2547 |
( # $3: Cells |
| 2548 |
(?> |
| 2549 |
[ ]* # Allowed whitespace. |
| 2550 |
[|] .* \n # Row content. |
| 2551 |
)* |
| 2552 |
) |
| 2553 |
(?=\n|\Z) # Stop at final double newline. |
| 2554 |
}xm', |
| 2555 |
array(&$this, '_doTable_leadingPipe_callback'), $text); |
| 2556 |
|
| 2557 |
# |
| 2558 |
# Find tables without leading pipe. |
| 2559 |
# |
| 2560 |
# Header 1 | Header 2 |
| 2561 |
# -------- | -------- |
| 2562 |
# Cell 1 | Cell 2 |
| 2563 |
# Cell 3 | Cell 4 |
| 2564 |
# |
| 2565 |
$text = preg_replace_callback(' |
| 2566 |
{ |
| 2567 |
^ # Start of a line |
| 2568 |
[ ]{0,'.$less_than_tab.'} # Allowed whitespace. |
| 2569 |
(\S.*[|].*) \n # $1: Header row (at least one pipe) |
| 2570 |
|
| 2571 |
[ ]{0,'.$less_than_tab.'} # Allowed whitespace. |
| 2572 |
([-:]+[ ]*[|][-| :]*) \n # $2: Header underline |
| 2573 |
|
| 2574 |
( # $3: Cells |
| 2575 |
(?> |
| 2576 |
.* [|] .* \n # Row content |
| 2577 |
)* |
| 2578 |
) |
| 2579 |
(?=\n|\Z) # Stop at final double newline. |
| 2580 |
}xm', |
| 2581 |
array(&$this, '_DoTable_callback'), $text); |
| 2582 |
|
| 2583 |
return $text; |
| 2584 |
} |
| 2585 |
function _doTable_leadingPipe_callback($matches) { |
| 2586 |
$head = $matches[1]; |
| 2587 |
$underline = $matches[2]; |
| 2588 |
$content = $matches[3]; |
| 2589 |
|
| 2590 |
# Remove leading pipe for each row. |
| 2591 |
$content = preg_replace('/^ *[|]/m', '', $content); |
| 2592 |
|
| 2593 |
return $this->_doTable_callback(array($matches[0], $head, $underline, $content)); |
| 2594 |
} |
| 2595 |
function _doTable_callback($matches) { |
| 2596 |
$head = $matches[1]; |
| 2597 |
$underline = $matches[2]; |
| 2598 |
$content = $matches[3]; |
| 2599 |
|
| 2600 |
# Remove any tailing pipes for each line. |
| 2601 |
$head = preg_replace('/[|] *$/m', '', $head); |
| 2602 |
$underline = preg_replace('/[|] *$/m', '', $underline); |
| 2603 |
$content = preg_replace('/[|] *$/m', '', $content); |
| 2604 |
|
| 2605 |
# Reading alignement from header underline. |
| 2606 |
$separators = preg_split('/ *[|] */', $underline); |
| 2607 |
foreach ($separators as $n => $s) { |
| 2608 |
if (preg_match('/^ *-+: *$/', $s)) $attr[$n] = ' align="right"'; |
| 2609 |
else if (preg_match('/^ *:-+: *$/', $s))$attr[$n] = ' align="center"'; |
| 2610 |
else if (preg_match('/^ *:-+ *$/', $s)) $attr[$n] = ' align="left"'; |
| 2611 |
else $attr[$n] = ''; |
| 2612 |
} |
| 2613 |
|
| 2614 |
# Parsing span elements, including code spans, character escapes, |
| 2615 |
# and inline HTML tags, so that pipes inside those gets ignored. |
| 2616 |
$head = $this->parseSpan($head); |
| 2617 |
$headers = preg_split('/ *[|] */', $head); |
| 2618 |
$col_count = count($headers); |
| 2619 |
$attr = array_pad($attr, $col_count, ''); |
| 2620 |
|
| 2621 |
# Write column headers. |
| 2622 |
$text = "<table>\n"; |
| 2623 |
$text .= "<thead>\n"; |
| 2624 |
$text .= "<tr>\n"; |
| 2625 |
foreach ($headers as $n => $header) |
| 2626 |
$text .= " <th$attr[$n]>".$this->runSpanGamut(trim($header))."</th>\n"; |
| 2627 |
$text .= "</tr>\n"; |
| 2628 |
$text .= "</thead>\n"; |
| 2629 |
|
| 2630 |
# Split content by row. |
| 2631 |
$rows = explode("\n", trim($content, "\n")); |
| 2632 |
|
| 2633 |
$text .= "<tbody>\n"; |
| 2634 |
foreach ($rows as $row) { |
| 2635 |
# Parsing span elements, including code spans, character escapes, |
| 2636 |
# and inline HTML tags, so that pipes inside those gets ignored. |
| 2637 |
$row = $this->parseSpan($row); |
| 2638 |
|
| 2639 |
# Split row by cell. |
| 2640 |
$row_cells = preg_split('/ *[|] */', $row, $col_count); |
| 2641 |
$row_cells = array_pad($row_cells, $col_count, ''); |
| 2642 |
|
| 2643 |
$text .= "<tr>\n"; |
| 2644 |
foreach ($row_cells as $n => $cell) |
| 2645 |
$text .= " <td$attr[$n]>".$this->runSpanGamut(trim($cell))."</td>\n"; |
| 2646 |
$text .= "</tr>\n"; |
| 2647 |
} |
| 2648 |
$text .= "</tbody>\n"; |
| 2649 |
$text .= "</table>"; |
| 2650 |
|
| 2651 |
return $this->hashBlock($text) . "\n"; |
| 2652 |
} |
| 2653 |
|
| 2654 |
|
| 2655 |
function doDefLists($text) { |
| 2656 |
# |
| 2657 |
# Form HTML definition lists. |
| 2658 |
# |
| 2659 |
$less_than_tab = $this->tab_width - 1; |
| 2660 |
|
| 2661 |
# Re-usable pattern to match any entire dl list: |
| 2662 |
$whole_list_re = '(?> |
| 2663 |
( # $1 = whole list |
| 2664 |
( # $2 |
| 2665 |
[ ]{0,'.$less_than_tab.'} |
| 2666 |
((?>.*\S.*\n)+) # $3 = defined term |
| 2667 |
\n? |
| 2668 |
[ ]{0,'.$less_than_tab.'}:[ ]+ # colon starting definition |
| 2669 |
) |
| 2670 |
(?s:.+?) |
| 2671 |
( # $4 |
| 2672 |
\z |
| 2673 |
| |
| 2674 |
\n{2,} |
| 2675 |
(?=\S) |
| 2676 |
(?! # Negative lookahead for another term |
| 2677 |
[ ]{0,'.$less_than_tab.'} |
| 2678 |
(?: \S.*\n )+? # defined term |
| 2679 |
\n? |
| 2680 |
[ ]{0,'.$less_than_tab.'}:[ ]+ # colon starting definition |
| 2681 |
) |
| 2682 |
(?! # Negative lookahead for another definition |
| 2683 |
[ ]{0,'.$less_than_tab.'}:[ ]+ # colon starting definition |
| 2684 |
) |
| 2685 |
) |
| 2686 |
) |
| 2687 |
)'; // mx |
| 2688 |
|
| 2689 |
$text = preg_replace_callback('{ |
| 2690 |
(?>\A\n?|(?<=\n\n)) |
| 2691 |
'.$whole_list_re.' |
| 2692 |
}mx', |
| 2693 |
array(&$this, '_doDefLists_callback'), $text); |
| 2694 |
|
| 2695 |
return $text; |
| 2696 |
} |
| 2697 |
function _doDefLists_callback($matches) { |
| 2698 |
# Re-usable patterns to match list item bullets and number markers: |
| 2699 |
$list = $matches[1]; |
| 2700 |
|
| 2701 |
# Turn double returns into triple returns, so that we can make a |
| 2702 |
# paragraph for the last item in a list, if necessary: |
| 2703 |
$result = trim($this->processDefListItems($list)); |
| 2704 |
$result = "<dl>\n" . $result . "\n</dl>"; |
| 2705 |
return $this->hashBlock($result) . "\n\n"; |
| 2706 |
} |
| 2707 |
|
| 2708 |
|
| 2709 |
function processDefListItems($list_str) { |
| 2710 |
# |
| 2711 |
# Process the contents of a single definition list, splitting it |
| 2712 |
# into individual term and definition list items. |
| 2713 |
# |
| 2714 |
$less_than_tab = $this->tab_width - 1; |
| 2715 |
|
| 2716 |
# trim trailing blank lines: |
| 2717 |
$list_str = preg_replace("/\n{2,}\\z/", "\n", $list_str); |
| 2718 |
|
| 2719 |
# Process definition terms. |
| 2720 |
$list_str = preg_replace_callback('{ |
| 2721 |
(?>\A\n?|\n\n+) # leading line |
| 2722 |
( # definition terms = $1 |
| 2723 |
[ ]{0,'.$less_than_tab.'} # leading whitespace |
| 2724 |
(?!\:[ ]|[ ]) # negative lookahead for a definition |
| 2725 |
# mark (colon) or more whitespace. |
| 2726 |
(?> \S.* \n)+? # actual term (not whitespace). |
| 2727 |
) |
| 2728 |
(?=\n?[ ]{0,3}:[ ]) # lookahead for following line feed |
| 2729 |
# with a definition mark. |
| 2730 |
}xm', |
| 2731 |
array(&$this, '_processDefListItems_callback_dt'), $list_str); |
| 2732 |
|
| 2733 |
# Process actual definitions. |
| 2734 |
$list_str = preg_replace_callback('{ |
| 2735 |
\n(\n+)? # leading line = $1 |
| 2736 |
( # marker space = $2 |
| 2737 |
[ ]{0,'.$less_than_tab.'} # whitespace before colon |
| 2738 |
\:[ ]+ # definition mark (colon) |
| 2739 |
) |
| 2740 |
((?s:.+?)) # definition text = $3 |
| 2741 |
(?= \n+ # stop at next definition mark, |
| 2742 |
(?: # next term or end of text |
| 2743 |
[ ]{0,'.$less_than_tab.'} \:[ ] | |
| 2744 |
<dt> | \z |
| 2745 |
) |
| 2746 |
) |
| 2747 |
}xm', |
| 2748 |
array(&$this, '_processDefListItems_callback_dd'), $list_str); |
| 2749 |
|
| 2750 |
return $list_str; |
| 2751 |
} |
| 2752 |
function _processDefListItems_callback_dt($matches) { |
| 2753 |
$terms = explode("\n", trim($matches[1])); |
| 2754 |
$text = ''; |
| 2755 |
foreach ($terms as $term) { |
| 2756 |
$term = $this->runSpanGamut(trim($term)); |
| 2757 |
$text .= "\n<dt>" . $term . "</dt>"; |
| 2758 |
} |
| 2759 |
return $text . "\n"; |
| 2760 |
} |
| 2761 |
function _processDefListItems_callback_dd($matches) { |
| 2762 |
$leading_line = $matches[1]; |
| 2763 |
$marker_space = $matches[2]; |
| 2764 |
$def = $matches[3]; |
| 2765 |
|
| 2766 |
if ($leading_line || preg_match('/\n{2,}/', $def)) { |
| 2767 |
# Replace marker with the appropriate whitespace indentation |
| 2768 |
$def = str_repeat(' ', strlen($marker_space)) . $def; |
| 2769 |
$def = $this->runBlockGamut($this->outdent($def . "\n\n")); |
| 2770 |
$def = "\n". $def ."\n"; |
| 2771 |
} |
| 2772 |
else { |
| 2773 |
$def = rtrim($def); |
| 2774 |
$def = $this->runSpanGamut($this->outdent($def)); |
| 2775 |
} |
| 2776 |
|
| 2777 |
return "\n<dd>" . $def . "</dd>\n"; |
| 2778 |
} |
| 2779 |
|
| 2780 |
|
| 2781 |
function doFencedCodeBlocks($text) { |
| 2782 |
# |
| 2783 |
# Adding the fenced code block syntax to regular Markdown: |
| 2784 |
# |
| 2785 |
# ~~~ |
| 2786 |
# Code block |
| 2787 |
# ~~~ |
| 2788 |
# |
| 2789 |
$less_than_tab = $this->tab_width; |
| 2790 |
|
| 2791 |
$text = preg_replace_callback('{ |
| 2792 |
(?:\n|\A) |
| 2793 |
# 1: Opening marker |
| 2794 |
( |
| 2795 |
(?:~{3,}|`{3,}) # 3 or more tildes/backticks. |
| 2796 |
) |
| 2797 |
[ ]* |
| 2798 |
(?: |
| 2799 |
\.?([-_:a-zA-Z0-9]+) # 2: standalone class name |
| 2800 |
| |
| 2801 |
'.$this->id_class_attr_catch_re.' # 3: Extra attributes |
| 2802 |
)? |
| 2803 |
[ ]* \n # Whitespace and newline following marker. |
| 2804 |
|
| 2805 |
# 4: Content |
| 2806 |
( |
| 2807 |
(?> |
| 2808 |
(?!\1 [ ]* \n) # Not a closing marker. |
| 2809 |
.*\n+ |
| 2810 |
)+ |
| 2811 |
) |
| 2812 |
|
| 2813 |
# Closing marker. |
| 2814 |
\1 [ ]* (?= \n ) |
| 2815 |
}xm', |
| 2816 |
array(&$this, '_doFencedCodeBlocks_callback'), $text); |
| 2817 |
|
| 2818 |
return $text; |
| 2819 |
} |
| 2820 |
function _doFencedCodeBlocks_callback($matches) { |
| 2821 |
$classname =& $matches[2]; |
| 2822 |
$attrs =& $matches[3]; |
| 2823 |
$codeblock = $matches[4]; |
| 2824 |
$codeblock = htmlspecialchars($codeblock, ENT_NOQUOTES); |
| 2825 |
$codeblock = preg_replace_callback('/^\n+/', |
| 2826 |
array(&$this, '_doFencedCodeBlocks_newlines'), $codeblock); |
| 2827 |
|
| 2828 |
if ($classname != "") { |
| 2829 |
if ($classname{0} == '.') |
| 2830 |
$classname = substr($classname, 1); |
| 2831 |
$attr_str = ' class="'.$this->code_class_prefix.$classname.'"'; |
| 2832 |
} else { |
| 2833 |
$attr_str = $this->doExtraAttributes($this->code_attr_on_pre ? "pre" : "code", $attrs); |
| 2834 |
} |
| 2835 |
$pre_attr_str = $this->code_attr_on_pre ? $attr_str : ''; |
| 2836 |
$code_attr_str = $this->code_attr_on_pre ? '' : $attr_str; |
| 2837 |
$codeblock = "<pre$pre_attr_str><code$code_attr_str>$codeblock</code></pre>"; |
| 2838 |
|
| 2839 |
return "\n\n".$this->hashBlock($codeblock)."\n\n"; |
| 2840 |
} |
| 2841 |
function _doFencedCodeBlocks_newlines($matches) { |
| 2842 |
return str_repeat("<br$this->empty_element_suffix", |
| 2843 |
strlen($matches[0])); |
| 2844 |
} |
| 2845 |
|
| 2846 |
|
| 2847 |
# |
| 2848 |
# Redefining emphasis markers so that emphasis by underscore does not |
| 2849 |
# work in the middle of a word. |
| 2850 |
# |
| 2851 |
public $em_relist = array( |
| 2852 |
'' => '(?:(?<!\*)\*(?!\*)|(?<![a-zA-Z0-9_])_(?!_))(?=\S|$)(?![\.,:;]\s)', |
| 2853 |
'*' => '(?<=\S|^)(?<!\*)\*(?!\*)', |
| 2854 |
'_' => '(?<=\S|^)(?<!_)_(?![a-zA-Z0-9_])', |
| 2855 |
); |
| 2856 |
public $strong_relist = array( |
| 2857 |
'' => '(?:(?<!\*)\*\*(?!\*)|(?<![a-zA-Z0-9_])__(?!_))(?=\S|$)(?![\.,:;]\s)', |
| 2858 |
'**' => '(?<=\S|^)(?<!\*)\*\*(?!\*)', |
| 2859 |
'__' => '(?<=\S|^)(?<!_)__(?![a-zA-Z0-9_])', |
| 2860 |
); |
| 2861 |
public $em_strong_relist = array( |
| 2862 |
'' => '(?:(?<!\*)\*\*\*(?!\*)|(?<![a-zA-Z0-9_])___(?!_))(?=\S|$)(?![\.,:;]\s)', |
| 2863 |
'***' => '(?<=\S|^)(?<!\*)\*\*\*(?!\*)', |
| 2864 |
'___' => '(?<=\S|^)(?<!_)___(?![a-zA-Z0-9_])', |
| 2865 |
); |
| 2866 |
|
| 2867 |
|
| 2868 |
function formParagraphs($text) { |
| 2869 |
# |
| 2870 |
# Params: |
| 2871 |
# $text - string to process with html <p> tags |
| 2872 |
# |
| 2873 |
# Strip leading and trailing lines: |
| 2874 |
$text = preg_replace('/\A\n+|\n+\z/', '', $text); |
| 2875 |
|
| 2876 |
$grafs = preg_split('/\n{2,}/', $text, -1, PREG_SPLIT_NO_EMPTY); |
| 2877 |
|
| 2878 |
# |
| 2879 |
# Wrap <p> tags and unhashify HTML blocks |
| 2880 |
# |
| 2881 |
foreach ($grafs as $key => $value) { |
| 2882 |
$value = trim($this->runSpanGamut($value)); |
| 2883 |
|
| 2884 |
# Check if this should be enclosed in a paragraph. |
| 2885 |
# Clean tag hashes & block tag hashes are left alone. |
| 2886 |
$is_p = !preg_match('/^B\x1A[0-9]+B|^C\x1A[0-9]+C$/', $value); |
| 2887 |
|
| 2888 |
if ($is_p) { |
| 2889 |
$value = "<p>$value</p>"; |
| 2890 |
} |
| 2891 |
$grafs[$key] = $value; |
| 2892 |
} |
| 2893 |
|
| 2894 |
# Join grafs in one text, then unhash HTML tags. |
| 2895 |
$text = implode("\n\n", $grafs); |
| 2896 |
|
| 2897 |
# Finish by removing any tag hashes still present in $text. |
| 2898 |
$text = $this->unhash($text); |
| 2899 |
|
| 2900 |
return $text; |
| 2901 |
} |
| 2902 |
|
| 2903 |
|
| 2904 |
### Footnotes |
| 2905 |
|
| 2906 |
function stripFootnotes($text) { |
| 2907 |
# |
| 2908 |
# Strips link definitions from text, stores the URLs and titles in |
| 2909 |
# hash references. |
| 2910 |
# |
| 2911 |
$less_than_tab = $this->tab_width - 1; |
| 2912 |
|
| 2913 |
# Link defs are in the form: [^id]: url "optional title" |
| 2914 |
$text = preg_replace_callback('{ |
| 2915 |
^[ ]{0,'.$less_than_tab.'}\[\^(.+?)\][ ]?: # note_id = $1 |
| 2916 |
[ ]* |
| 2917 |
\n? # maybe *one* newline |
| 2918 |
( # text = $2 (no blank lines allowed) |
| 2919 |
(?: |
| 2920 |
.+ # actual text |
| 2921 |
| |
| 2922 |
\n # newlines but |
| 2923 |
(?!\[\^.+?\]:\s)# negative lookahead for footnote marker. |
| 2924 |
(?!\n+[ ]{0,3}\S)# ensure line is not blank and followed |
| 2925 |
# by non-indented content |
| 2926 |
)* |
| 2927 |
) |
| 2928 |
}xm', |
| 2929 |
array(&$this, '_stripFootnotes_callback'), |
| 2930 |
$text); |
| 2931 |
return $text; |
| 2932 |
} |
| 2933 |
function _stripFootnotes_callback($matches) { |
| 2934 |
$note_id = $this->fn_id_prefix . $matches[1]; |
| 2935 |
$this->footnotes[$note_id] = $this->outdent($matches[2]); |
| 2936 |
return ''; # String that will replace the block |
| 2937 |
} |
| 2938 |
|
| 2939 |
|
| 2940 |
function doFootnotes($text) { |
| 2941 |
# |
| 2942 |
# Replace footnote references in $text [^id] with a special text-token |
| 2943 |
# which will be replaced by the actual footnote marker in appendFootnotes. |
| 2944 |
# |
| 2945 |
if (!$this->in_anchor) { |
| 2946 |
$text = preg_replace('{\[\^(.+?)\]}', "F\x1Afn:\\1\x1A:", $text); |
| 2947 |
} |
| 2948 |
return $text; |
| 2949 |
} |
| 2950 |
|
| 2951 |
|
| 2952 |
function appendFootnotes($text) { |
| 2953 |
# |
| 2954 |
# Append footnote list to text. |
| 2955 |
# |
| 2956 |
$text = preg_replace_callback('{F\x1Afn:(.*?)\x1A:}', |
| 2957 |
array(&$this, '_appendFootnotes_callback'), $text); |
| 2958 |
|
| 2959 |
if (!empty($this->footnotes_ordered)) { |
| 2960 |
$text .= "\n\n"; |
| 2961 |
$text .= "<div class=\"footnotes\">\n"; |
| 2962 |
$text .= "<hr". $this->empty_element_suffix ."\n"; |
| 2963 |
$text .= "<ol>\n\n"; |
| 2964 |
|
| 2965 |
$attr = ""; |
| 2966 |
if ($this->fn_backlink_class != "") { |
| 2967 |
$class = $this->fn_backlink_class; |
| 2968 |
$class = $this->encodeAttribute($class); |
| 2969 |
$attr .= " class=\"$class\""; |
| 2970 |
} |
| 2971 |
if ($this->fn_backlink_title != "") { |
| 2972 |
$title = $this->fn_backlink_title; |
| 2973 |
$title = $this->encodeAttribute($title); |
| 2974 |
$attr .= " title=\"$title\""; |
| 2975 |
} |
| 2976 |
$num = 0; |
| 2977 |
|
| 2978 |
while (!empty($this->footnotes_ordered)) { |
| 2979 |
$footnote = reset($this->footnotes_ordered); |
| 2980 |
$note_id = key($this->footnotes_ordered); |
| 2981 |
unset($this->footnotes_ordered[$note_id]); |
| 2982 |
$ref_count = $this->footnotes_ref_count[$note_id]; |
| 2983 |
unset($this->footnotes_ref_count[$note_id]); |
| 2984 |
unset($this->footnotes[$note_id]); |
| 2985 |
|
| 2986 |
$footnote .= "\n"; # Need to append newline before parsing. |
| 2987 |
$footnote = $this->runBlockGamut("$footnote\n"); |
| 2988 |
$footnote = preg_replace_callback('{F\x1Afn:(.*?)\x1A:}', |
| 2989 |
array(&$this, '_appendFootnotes_callback'), $footnote); |
| 2990 |
|
| 2991 |
$attr = str_replace("%%", ++$num, $attr); |
| 2992 |
$note_id = $this->encodeAttribute($note_id); |
| 2993 |
|
| 2994 |
# Prepare backlink, multiple backlinks if multiple references |
| 2995 |
$backlink = "<a href=\"#fnref:$note_id\"$attr>↩</a>"; |
| 2996 |
for ($ref_num = 2; $ref_num <= $ref_count; ++$ref_num) { |
| 2997 |
$backlink .= " <a href=\"#fnref$ref_num:$note_id\"$attr>↩</a>"; |
| 2998 |
} |
| 2999 |
# Add backlink to last paragraph; create new paragraph if needed. |
| 3000 |
if (preg_match('{</p>$}', $footnote)) { |
| 3001 |
$footnote = substr($footnote, 0, -4) . " $backlink</p>"; |
| 3002 |
} else { |
| 3003 |
$footnote .= "\n\n<p>$backlink</p>"; |
| 3004 |
} |
| 3005 |
|
| 3006 |
$text .= "<li id=\"fn:$note_id\">\n"; |
| 3007 |
$text .= $footnote . "\n"; |
| 3008 |
$text .= "</li>\n\n"; |
| 3009 |
} |
| 3010 |
|
| 3011 |
$text .= "</ol>\n"; |
| 3012 |
$text .= "</div>"; |
| 3013 |
} |
| 3014 |
return $text; |
| 3015 |
} |
| 3016 |
function _appendFootnotes_callback($matches) { |
| 3017 |
$node_id = $this->fn_id_prefix . $matches[1]; |
| 3018 |
|
| 3019 |
# Create footnote marker only if it has a corresponding footnote *and* |
| 3020 |
# the footnote hasn't been used by another marker. |
| 3021 |
if (isset($this->footnotes[$node_id])) { |
| 3022 |
$num =& $this->footnotes_numbers[$node_id]; |
| 3023 |
if (!isset($num)) { |
| 3024 |
# Transfer footnote content to the ordered list and give it its |
| 3025 |
# number |
| 3026 |
$this->footnotes_ordered[$node_id] = $this->footnotes[$node_id]; |
| 3027 |
$this->footnotes_ref_count[$node_id] = 1; |
| 3028 |
$num = $this->footnote_counter++; |
| 3029 |
$ref_count_mark = ''; |
| 3030 |
} else { |
| 3031 |
$ref_count_mark = $this->footnotes_ref_count[$node_id] += 1; |
| 3032 |
} |
| 3033 |
|
| 3034 |
$attr = ""; |
| 3035 |
if ($this->fn_link_class != "") { |
| 3036 |
$class = $this->fn_link_class; |
| 3037 |
$class = $this->encodeAttribute($class); |
| 3038 |
$attr .= " class=\"$class\""; |
| 3039 |
} |
| 3040 |
if ($this->fn_link_title != "") { |
| 3041 |
$title = $this->fn_link_title; |
| 3042 |
$title = $this->encodeAttribute($title); |
| 3043 |
$attr .= " title=\"$title\""; |
| 3044 |
} |
| 3045 |
|
| 3046 |
$attr = str_replace("%%", $num, $attr); |
| 3047 |
$node_id = $this->encodeAttribute($node_id); |
| 3048 |
|
| 3049 |
return |
| 3050 |
"<sup id=\"fnref$ref_count_mark:$node_id\">". |
| 3051 |
"<a href=\"#fn:$node_id\"$attr>$num</a>". |
| 3052 |
"</sup>"; |
| 3053 |
} |
| 3054 |
|
| 3055 |
return "[^".$matches[1]."]"; |
| 3056 |
} |
| 3057 |
|
| 3058 |
|
| 3059 |
### Abbreviations ### |
| 3060 |
|
| 3061 |
function stripAbbreviations($text) { |
| 3062 |
# |
| 3063 |
# Strips abbreviations from text, stores titles in hash references. |
| 3064 |
# |
| 3065 |
$less_than_tab = $this->tab_width - 1; |
| 3066 |
|
| 3067 |
# Link defs are in the form: [id]*: url "optional title" |
| 3068 |
$text = preg_replace_callback('{ |
| 3069 |
^[ ]{0,'.$less_than_tab.'}\*\[(.+?)\][ ]?: # abbr_id = $1 |
| 3070 |
(.*) # text = $2 (no blank lines allowed) |
| 3071 |
}xm', |
| 3072 |
array(&$this, '_stripAbbreviations_callback'), |
| 3073 |
$text); |
| 3074 |
return $text; |
| 3075 |
} |
| 3076 |
function _stripAbbreviations_callback($matches) { |
| 3077 |
$abbr_word = $matches[1]; |
| 3078 |
$abbr_desc = $matches[2]; |
| 3079 |
if ($this->abbr_word_re) |
| 3080 |
$this->abbr_word_re .= '|'; |
| 3081 |
$this->abbr_word_re .= preg_quote($abbr_word); |
| 3082 |
$this->abbr_desciptions[$abbr_word] = trim($abbr_desc); |
| 3083 |
return ''; # String that will replace the block |
| 3084 |
} |
| 3085 |
|
| 3086 |
|
| 3087 |
function doAbbreviations($text) { |
| 3088 |
# |
| 3089 |
# Find defined abbreviations in text and wrap them in <abbr> elements. |
| 3090 |
# |
| 3091 |
if ($this->abbr_word_re) { |
| 3092 |
// cannot use the /x modifier because abbr_word_re may |
| 3093 |
// contain significant spaces: |
| 3094 |
$text = preg_replace_callback('{'. |
| 3095 |
'(?<![\w\x1A])'. |
| 3096 |
'(?:'.$this->abbr_word_re.')'. |
| 3097 |
'(?![\w\x1A])'. |
| 3098 |
'}', |
| 3099 |
array(&$this, '_doAbbreviations_callback'), $text); |
| 3100 |
} |
| 3101 |
return $text; |
| 3102 |
} |
| 3103 |
function _doAbbreviations_callback($matches) { |
| 3104 |
$abbr = $matches[0]; |
| 3105 |
if (isset($this->abbr_desciptions[$abbr])) { |
| 3106 |
$desc = $this->abbr_desciptions[$abbr]; |
| 3107 |
if (empty($desc)) { |
| 3108 |
return $this->hashPart("<abbr>$abbr</abbr>"); |
| 3109 |
} else { |
| 3110 |
$desc = $this->encodeAttribute($desc); |
| 3111 |
return $this->hashPart("<abbr title=\"$desc\">$abbr</abbr>"); |
| 3112 |
} |
| 3113 |
} else { |
| 3114 |
return $matches[0]; |
| 3115 |
} |
| 3116 |
} |
| 3117 |
|
| 3118 |
} |
| 3119 |
|
| 3120 |
|
| 3121 |
/* |
| 3122 |
|
| 3123 |
PHP Markdown Extra |
| 3124 |
================== |
| 3125 |
|
| 3126 |
Description |
| 3127 |
----------- |
| 3128 |
|
| 3129 |
This is a PHP port of the original Markdown formatter written in Perl |
| 3130 |
by John Gruber. This special "Extra" version of PHP Markdown features |
| 3131 |
further enhancements to the syntax for making additional constructs |
| 3132 |
such as tables and definition list. |
| 3133 |
|
| 3134 |
Markdown is a text-to-HTML filter; it translates an easy-to-read / |
| 3135 |
easy-to-write structured text format into HTML. Markdown's text format |
| 3136 |
is mostly similar to that of plain text email, and supports features such |
| 3137 |
as headers, *emphasis*, code blocks, blockquotes, and links. |
| 3138 |
|
| 3139 |
Markdown's syntax is designed not as a generic markup language, but |
| 3140 |
specifically to serve as a front-end to (X)HTML. You can use span-level |
| 3141 |
HTML tags anywhere in a Markdown document, and you can use block level |
| 3142 |
HTML tags (like <div> and <table> as well). |
| 3143 |
|
| 3144 |
For more information about Markdown's syntax, see: |
| 3145 |
|
| 3146 |
<http://daringfireball.net/projects/markdown/> |
| 3147 |
|
| 3148 |
|
| 3149 |
Bugs |
| 3150 |
---- |
| 3151 |
|
| 3152 |
To file bug reports please send email to: |
| 3153 |
|
| 3154 |
<michel.fortin@michelf.ca> |
| 3155 |
|
| 3156 |
Please include with your report: (1) the example input; (2) the output you |
| 3157 |
expected; (3) the output Markdown actually produced. |
| 3158 |
|
| 3159 |
|
| 3160 |
Version History |
| 3161 |
--------------- |
| 3162 |
|
| 3163 |
See the readme file for detailed release notes for this version. |
| 3164 |
|
| 3165 |
|
| 3166 |
Copyright and License |
| 3167 |
--------------------- |
| 3168 |
|
| 3169 |
PHP Markdown & Extra |
| 3170 |
Copyright (c) 2004-2013 Michel Fortin |
| 3171 |
<http://michelf.ca/> |
| 3172 |
All rights reserved. |
| 3173 |
|
| 3174 |
Based on Markdown |
| 3175 |
Copyright (c) 2003-2006 John Gruber |
| 3176 |
<http://daringfireball.net/> |
| 3177 |
All rights reserved. |
| 3178 |
|
| 3179 |
Redistribution and use in source and binary forms, with or without |
| 3180 |
modification, are permitted provided that the following conditions are |
| 3181 |
met: |
| 3182 |
|
| 3183 |
* Redistributions of source code must retain the above copyright notice, |
| 3184 |
this list of conditions and the following disclaimer. |
| 3185 |
|
| 3186 |
* Redistributions in binary form must reproduce the above copyright |
| 3187 |
notice, this list of conditions and the following disclaimer in the |
| 3188 |
documentation and/or other materials provided with the distribution. |
| 3189 |
|
| 3190 |
* Neither the name "Markdown" nor the names of its contributors may |
| 3191 |
be used to endorse or promote products derived from this software |
| 3192 |
without specific prior written permission. |
| 3193 |
|
| 3194 |
This software is provided by the copyright holders and contributors "as |
| 3195 |
is" and any express or implied warranties, including, but not limited |
| 3196 |
to, the implied warranties of merchantability and fitness for a |
| 3197 |
particular purpose are disclaimed. In no event shall the copyright owner |
| 3198 |
or contributors be liable for any direct, indirect, incidental, special, |
| 3199 |
exemplary, or consequential damages (including, but not limited to, |
| 3200 |
procurement of substitute goods or services; loss of use, data, or |
| 3201 |
profits; or business interruption) however caused and on any theory of |
| 3202 |
liability, whether in contract, strict liability, or tort (including |
| 3203 |
negligence or otherwise) arising in any way out of the use of this |
| 3204 |
software, even if advised of the possibility of such damage. |
| 3205 |
|
| 3206 |
*/ |
| 3207 |
?> |
| 3208 |
|
