| 1 |
<?php |
| 2 |
/** |
| 3 |
* PHP Markdown class. |
| 4 |
* |
| 5 |
* Copyright {@link http://www.michelf.com/projects/php-markdown/ Michel Fortin}. |
| 6 |
* Original Markdown. Copyright {@link http://daringfireball.net/projects/markdown/ John Gruber}. |
| 7 |
* |
| 8 |
* Modified by {@link http://www.websharks-inc.com/ WebSharks, Inc.}. |
| 9 |
* Excludes WordPress® and all other interfaces. |
| 10 |
* Uses a custom class name and interface. |
| 11 |
* |
| 12 |
* This file is included with all WordPress® themes/plugins by WebSharks, Inc. |
| 13 |
* |
| 14 |
* @package WebSharks\Xtnls\Markdown |
| 15 |
* @since x.xx |
| 16 |
*/ |
| 17 |
/** |
| 18 |
* PHP Markdown interface. |
| 19 |
* |
| 20 |
* @package WebSharks\Xtnls\Markdown |
| 21 |
* @since x.xx |
| 22 |
* |
| 23 |
* @param str $text Text to be parsed by the Markdown class. |
| 24 |
* @return str HTML output; after having been parsed by the Markdown class. |
| 25 |
*/ |
| 26 |
function NC_Markdown($text) { |
| 27 |
|
| 28 |
static $parser; |
| 29 |
if (!isset($parser)) { |
| 30 |
$parser_class = NC_Markdown_Parser; |
| 31 |
$parser = new $parser_class; |
| 32 |
} |
| 33 |
|
| 34 |
return $parser->transform($text); |
| 35 |
} |
| 36 |
/** |
| 37 |
* PHP Markdown class. |
| 38 |
* @package Xtnls\Markdown |
| 39 |
* @since x.xx |
| 40 |
*/ |
| 41 |
class NC_Markdown_Parser { |
| 42 |
|
| 43 |
# Regex to match balanced [brackets]. |
| 44 |
# Needed to insert a maximum bracked depth while converting to PHP. |
| 45 |
var $nested_brackets_depth = 6; |
| 46 |
var $nested_brackets_re; |
| 47 |
|
| 48 |
var $nested_url_parenthesis_depth = 4; |
| 49 |
var $nested_url_parenthesis_re; |
| 50 |
|
| 51 |
# Table of hash values for escaped characters: |
| 52 |
var $escape_chars = '\`*_{}[]()>#+-.!'; |
| 53 |
var $escape_chars_re; |
| 54 |
|
| 55 |
# Change to ">" for HTML output. |
| 56 |
var $empty_element_suffix = " />"; |
| 57 |
var $tab_width = 4; |
| 58 |
|
| 59 |
# Change to `true` to disallow markup or entities. |
| 60 |
var $no_markup = false; |
| 61 |
var $no_entities = false; |
| 62 |
|
| 63 |
# Predefined urls and titles for reference links and images. |
| 64 |
var $predef_urls = array(); |
| 65 |
var $predef_titles = array(); |
| 66 |
|
| 67 |
|
| 68 |
function NC_Markdown_Parser() { |
| 69 |
# |
| 70 |
# Constructor function. Initialize appropriate member variables. |
| 71 |
# |
| 72 |
$this->_initDetab(); |
| 73 |
$this->prepareItalicsAndBold(); |
| 74 |
|
| 75 |
$this->nested_brackets_re = |
| 76 |
str_repeat('(?>[^\[\]]+|\[', $this->nested_brackets_depth). |
| 77 |
str_repeat('\])*', $this->nested_brackets_depth); |
| 78 |
|
| 79 |
$this->nested_url_parenthesis_re = |
| 80 |
str_repeat('(?>[^()\s]+|\(', $this->nested_url_parenthesis_depth). |
| 81 |
str_repeat('(?>\)))*', $this->nested_url_parenthesis_depth); |
| 82 |
|
| 83 |
$this->escape_chars_re = '['.preg_quote($this->escape_chars).']'; |
| 84 |
|
| 85 |
# Sort document, block, and span gamut in ascendent priority order. |
| 86 |
asort($this->document_gamut); |
| 87 |
asort($this->block_gamut); |
| 88 |
asort($this->span_gamut); |
| 89 |
} |
| 90 |
|
| 91 |
|
| 92 |
# Internal hashes used during transformation. |
| 93 |
var $urls = array(); |
| 94 |
var $titles = array(); |
| 95 |
var $html_hashes = array(); |
| 96 |
|
| 97 |
# Status flag to avoid invalid nesting. |
| 98 |
var $in_anchor = false; |
| 99 |
|
| 100 |
|
| 101 |
function setup() { |
| 102 |
# |
| 103 |
# Called before the transformation process starts to setup parser |
| 104 |
# states. |
| 105 |
# |
| 106 |
# Clear global hashes. |
| 107 |
$this->urls = $this->predef_urls; |
| 108 |
$this->titles = $this->predef_titles; |
| 109 |
$this->html_hashes = array(); |
| 110 |
|
| 111 |
$in_anchor = false; |
| 112 |
} |
| 113 |
|
| 114 |
function teardown() { |
| 115 |
# |
| 116 |
# Called after the transformation process to clear any variable |
| 117 |
# which may be taking up memory unnecessarly. |
| 118 |
# |
| 119 |
$this->urls = array(); |
| 120 |
$this->titles = array(); |
| 121 |
$this->html_hashes = array(); |
| 122 |
} |
| 123 |
|
| 124 |
|
| 125 |
function transform($text) { |
| 126 |
# |
| 127 |
# Main function. Performs some preprocessing on the input text |
| 128 |
# and pass it through the document gamut. |
| 129 |
# |
| 130 |
$this->setup(); |
| 131 |
|
| 132 |
# Remove UTF-8 BOM and marker character in input, if present. |
| 133 |
$text = preg_replace('{^\xEF\xBB\xBF|\x1A}', '', $text); |
| 134 |
|
| 135 |
# Standardize line endings: |
| 136 |
# DOS to Unix and Mac to Unix |
| 137 |
$text = preg_replace('{\r\n?}', "\n", $text); |
| 138 |
|
| 139 |
# Make sure $text ends with a couple of newlines: |
| 140 |
$text .= "\n\n"; |
| 141 |
|
| 142 |
# Convert all tabs to spaces. |
| 143 |
$text = $this->detab($text); |
| 144 |
|
| 145 |
# Turn block-level HTML blocks into hash entries |
| 146 |
$text = $this->hashHTMLBlocks($text); |
| 147 |
|
| 148 |
# Strip any lines consisting only of spaces and tabs. |
| 149 |
# This makes subsequent regexen easier to write, because we can |
| 150 |
# match consecutive blank lines with /\n+/ instead of something |
| 151 |
# contorted like /[ ]*\n+/ . |
| 152 |
$text = preg_replace('/^[ ]+$/m', '', $text); |
| 153 |
|
| 154 |
# Run document gamut methods. |
| 155 |
foreach ($this->document_gamut as $method => $priority) { |
| 156 |
$text = $this->$method($text); |
| 157 |
} |
| 158 |
|
| 159 |
$this->teardown(); |
| 160 |
|
| 161 |
return $text . "\n"; |
| 162 |
} |
| 163 |
|
| 164 |
var $document_gamut = array( |
| 165 |
# Strip link definitions, store in hashes. |
| 166 |
"stripLinkDefinitions" => 20, |
| 167 |
|
| 168 |
"runBasicBlockGamut" => 30, |
| 169 |
); |
| 170 |
|
| 171 |
|
| 172 |
function stripLinkDefinitions($text) { |
| 173 |
# |
| 174 |
# Strips link definitions from text, stores the URLs and titles in |
| 175 |
# hash references. |
| 176 |
# |
| 177 |
$less_than_tab = $this->tab_width - 1; |
| 178 |
|
| 179 |
# Link defs are in the form: ^[id]: url "optional title" |
| 180 |
$text = preg_replace_callback('{ |
| 181 |
^[ ]{0,'.$less_than_tab.'}\[(.+)\][ ]?: # id = $1 |
| 182 |
[ ]* |
| 183 |
\n? # maybe *one* newline |
| 184 |
[ ]* |
| 185 |
<?(\S+?)>? # url = $2 |
| 186 |
[ ]* |
| 187 |
\n? # maybe one newline |
| 188 |
[ ]* |
| 189 |
(?: |
| 190 |
(?<=\s) # lookbehind for whitespace |
| 191 |
["(] |
| 192 |
(.*?) # title = $3 |
| 193 |
[")] |
| 194 |
[ ]* |
| 195 |
)? # title is optional |
| 196 |
(?:\n+|\Z) |
| 197 |
}xm', |
| 198 |
array(&$this, '_stripLinkDefinitions_callback'), |
| 199 |
$text); |
| 200 |
return $text; |
| 201 |
} |
| 202 |
function _stripLinkDefinitions_callback($matches) { |
| 203 |
$link_id = strtolower($matches[1]); |
| 204 |
$this->urls[$link_id] = $matches[2]; |
| 205 |
$this->titles[$link_id] =& $matches[3]; |
| 206 |
return ''; # String that will replace the block |
| 207 |
} |
| 208 |
|
| 209 |
|
| 210 |
function hashHTMLBlocks($text) { |
| 211 |
if ($this->no_markup) return $text; |
| 212 |
|
| 213 |
$less_than_tab = $this->tab_width - 1; |
| 214 |
|
| 215 |
# Hashify HTML blocks: |
| 216 |
# We only want to do this for block-level HTML tags, such as headers, |
| 217 |
# lists, and tables. That's because we still want to wrap <p>s around |
| 218 |
# "paragraphs" that are wrapped in non-block-level tags, such as anchors, |
| 219 |
# phrase emphasis, and spans. The list of tags we're looking for is |
| 220 |
# hard-coded: |
| 221 |
# |
| 222 |
# * List "a" is made of tags which can be both inline or block-level. |
| 223 |
# These will be treated block-level when the start tag is alone on |
| 224 |
# its line, otherwise they're not matched here and will be taken as |
| 225 |
# inline later. |
| 226 |
# * List "b" is made of tags which are always block-level; |
| 227 |
# |
| 228 |
$block_tags_a_re = 'ins|del'; |
| 229 |
$block_tags_b_re = 'p|div|h[1-6]|blockquote|pre|table|dl|ol|ul|address|'. |
| 230 |
'script|noscript|form|fieldset|iframe|math'; |
| 231 |
|
| 232 |
# Regular expression for the content of a block tag. |
| 233 |
$nested_tags_level = 4; |
| 234 |
$attr = ' |
| 235 |
(?> # optional tag attributes |
| 236 |
\s # starts with whitespace |
| 237 |
(?> |
| 238 |
[^>"/]+ # text outside quotes |
| 239 |
| |
| 240 |
/+(?!>) # slash not followed by ">" |
| 241 |
| |
| 242 |
"[^"]*" # text inside double quotes (tolerate ">") |
| 243 |
| |
| 244 |
\'[^\']*\' # text inside single quotes (tolerate ">") |
| 245 |
)* |
| 246 |
)? |
| 247 |
'; |
| 248 |
$content = |
| 249 |
str_repeat(' |
| 250 |
(?> |
| 251 |
[^<]+ # content without tag |
| 252 |
| |
| 253 |
<\2 # nested opening tag |
| 254 |
'.$attr.' # attributes |
| 255 |
(?> |
| 256 |
/> |
| 257 |
| |
| 258 |
>', $nested_tags_level). # end of opening tag |
| 259 |
'.*?'. # last level nested tag content |
| 260 |
str_repeat(' |
| 261 |
</\2\s*> # closing nested tag |
| 262 |
) |
| 263 |
| |
| 264 |
<(?!/\2\s*> # other tags with a different name |
| 265 |
) |
| 266 |
)*', |
| 267 |
$nested_tags_level); |
| 268 |
$content2 = str_replace('\2', '\3', $content); |
| 269 |
|
| 270 |
# First, look for nested blocks, e.g.: |
| 271 |
# <div> |
| 272 |
# <div> |
| 273 |
# tags for inner block must be indented. |
| 274 |
# </div> |
| 275 |
# </div> |
| 276 |
# |
| 277 |
# The outermost tags must start at the left margin for this to match, and |
| 278 |
# the inner nested divs must be indented. |
| 279 |
# We need to do this before the next, more liberal match, because the next |
| 280 |
# match will start at the first `<div>` and stop at the first `</div>`. |
| 281 |
$text = preg_replace_callback('{(?> |
| 282 |
(?> |
| 283 |
(?<=\n\n) # Starting after a blank line |
| 284 |
| # or |
| 285 |
\A\n? # the beginning of the doc |
| 286 |
) |
| 287 |
( # save in $1 |
| 288 |
|
| 289 |
# Match from `\n<tag>` to `</tag>\n`, handling nested tags |
| 290 |
# in between. |
| 291 |
|
| 292 |
[ ]{0,'.$less_than_tab.'} |
| 293 |
<('.$block_tags_b_re.')# start tag = $2 |
| 294 |
'.$attr.'> # attributes followed by > and \n |
| 295 |
'.$content.' # content, support nesting |
| 296 |
</\2> # the matching end tag |
| 297 |
[ ]* # trailing spaces/tabs |
| 298 |
(?=\n+|\Z) # followed by a newline or end of document |
| 299 |
|
| 300 |
| # Special version for tags of group a. |
| 301 |
|
| 302 |
[ ]{0,'.$less_than_tab.'} |
| 303 |
<('.$block_tags_a_re.')# start tag = $3 |
| 304 |
'.$attr.'>[ ]*\n # attributes followed by > |
| 305 |
'.$content2.' # content, support nesting |
| 306 |
</\3> # the matching end tag |
| 307 |
[ ]* # trailing spaces/tabs |
| 308 |
(?=\n+|\Z) # followed by a newline or end of document |
| 309 |
|
| 310 |
| # Special case just for <hr />. It was easier to make a special |
| 311 |
# case than to make the other regex more complicated. |
| 312 |
|
| 313 |
[ ]{0,'.$less_than_tab.'} |
| 314 |
<(hr) # start tag = $2 |
| 315 |
'.$attr.' # attributes |
| 316 |
/?> # the matching end tag |
| 317 |
[ ]* |
| 318 |
(?=\n{2,}|\Z) # followed by a blank line or end of document |
| 319 |
|
| 320 |
| # Special case for standalone HTML comments: |
| 321 |
|
| 322 |
[ ]{0,'.$less_than_tab.'} |
| 323 |
(?s: |
| 324 |
<!-- .*? --> |
| 325 |
) |
| 326 |
[ ]* |
| 327 |
(?=\n{2,}|\Z) # followed by a blank line or end of document |
| 328 |
|
| 329 |
| # PHP and ASP-style processor instructions (<? and <%) |
| 330 |
|
| 331 |
[ ]{0,'.$less_than_tab.'} |
| 332 |
(?s: |
| 333 |
<([?%]) # $2 |
| 334 |
.*? |
| 335 |
\2> |
| 336 |
) |
| 337 |
[ ]* |
| 338 |
(?=\n{2,}|\Z) # followed by a blank line or end of document |
| 339 |
|
| 340 |
) |
| 341 |
)}Sxmi', |
| 342 |
array(&$this, '_hashHTMLBlocks_callback'), |
| 343 |
$text); |
| 344 |
|
| 345 |
return $text; |
| 346 |
} |
| 347 |
function _hashHTMLBlocks_callback($matches) { |
| 348 |
$text = $matches[1]; |
| 349 |
$key = $this->hashBlock($text); |
| 350 |
return "\n\n$key\n\n"; |
| 351 |
} |
| 352 |
|
| 353 |
|
| 354 |
function hashPart($text, $boundary = 'X') { |
| 355 |
# |
| 356 |
# Called whenever a tag must be hashed when a function insert an atomic |
| 357 |
# element in the text stream. Passing $text to through this function gives |
| 358 |
# a unique text-token which will be reverted back when calling unhash. |
| 359 |
# |
| 360 |
# The $boundary argument specify what character should be used to surround |
| 361 |
# the token. By convension, "B" is used for block elements that needs not |
| 362 |
# to be wrapped into paragraph tags at the end, ":" is used for elements |
| 363 |
# that are word separators and "X" is used in the general case. |
| 364 |
# |
| 365 |
# Swap back any tag hash found in $text so we do not have to `unhash` |
| 366 |
# multiple times at the end. |
| 367 |
$text = $this->unhash($text); |
| 368 |
|
| 369 |
# Then hash the block. |
| 370 |
static $i = 0; |
| 371 |
$key = "$boundary\x1A" . ++$i . $boundary; |
| 372 |
$this->html_hashes[$key] = $text; |
| 373 |
return $key; # String that will replace the tag. |
| 374 |
} |
| 375 |
|
| 376 |
|
| 377 |
function hashBlock($text) { |
| 378 |
# |
| 379 |
# Shortcut function for hashPart with block-level boundaries. |
| 380 |
# |
| 381 |
return $this->hashPart($text, 'B'); |
| 382 |
} |
| 383 |
|
| 384 |
|
| 385 |
var $block_gamut = array( |
| 386 |
# |
| 387 |
# These are all the transformations that form block-level |
| 388 |
# tags like paragraphs, headers, and list items. |
| 389 |
# |
| 390 |
"doHeaders" => 10, |
| 391 |
"doHorizontalRules" => 20, |
| 392 |
|
| 393 |
"doLists" => 40, |
| 394 |
"doCodeBlocks" => 50, |
| 395 |
"doBlockQuotes" => 60, |
| 396 |
); |
| 397 |
|
| 398 |
function runBlockGamut($text) { |
| 399 |
# |
| 400 |
# Run block gamut tranformations. |
| 401 |
# |
| 402 |
# We need to escape raw HTML in Markdown source before doing anything |
| 403 |
# else. This need to be done for each block, and not only at the |
| 404 |
# begining in the Markdown function since hashed blocks can be part of |
| 405 |
# list items and could have been indented. Indented blocks would have |
| 406 |
# been seen as a code block in a previous pass of hashHTMLBlocks. |
| 407 |
$text = $this->hashHTMLBlocks($text); |
| 408 |
|
| 409 |
return $this->runBasicBlockGamut($text); |
| 410 |
} |
| 411 |
|
| 412 |
function runBasicBlockGamut($text) { |
| 413 |
# |
| 414 |
# Run block gamut tranformations, without hashing HTML blocks. This is |
| 415 |
# useful when HTML blocks are known to be already hashed, like in the first |
| 416 |
# whole-document pass. |
| 417 |
# |
| 418 |
foreach ($this->block_gamut as $method => $priority) { |
| 419 |
$text = $this->$method($text); |
| 420 |
} |
| 421 |
|
| 422 |
# Finally form paragraph and restore hashed blocks. |
| 423 |
$text = $this->formParagraphs($text); |
| 424 |
|
| 425 |
return $text; |
| 426 |
} |
| 427 |
|
| 428 |
|
| 429 |
function doHorizontalRules($text) { |
| 430 |
# Do Horizontal Rules: |
| 431 |
return preg_replace( |
| 432 |
'{ |
| 433 |
^[ ]{0,3} # Leading space |
| 434 |
([-*_]) # $1: First marker |
| 435 |
(?> # Repeated marker group |
| 436 |
[ ]{0,2} # Zero, one, or two spaces. |
| 437 |
\1 # Marker character |
| 438 |
){2,} # Group repeated at least twice |
| 439 |
[ ]* # Tailing spaces |
| 440 |
$ # End of line. |
| 441 |
}mx', |
| 442 |
"\n".$this->hashBlock("<hr$this->empty_element_suffix")."\n", |
| 443 |
$text); |
| 444 |
} |
| 445 |
|
| 446 |
|
| 447 |
var $span_gamut = array( |
| 448 |
# |
| 449 |
# These are all the transformations that occur *within* block-level |
| 450 |
# tags like paragraphs, headers, and list items. |
| 451 |
# |
| 452 |
# Process character escapes, code spans, and inline HTML |
| 453 |
# in one shot. |
| 454 |
"parseSpan" => -30, |
| 455 |
|
| 456 |
# Process anchor and image tags. Images must come first, |
| 457 |
# because ![foo][f] looks like an anchor. |
| 458 |
"doImages" => 10, |
| 459 |
"doAnchors" => 20, |
| 460 |
|
| 461 |
# Make links out of things like `<http://example.com/>` |
| 462 |
# Must come after doAnchors, because you can use < and > |
| 463 |
# delimiters in inline links like [this](<url>). |
| 464 |
"doAutoLinks" => 30, |
| 465 |
"encodeAmpsAndAngles" => 40, |
| 466 |
|
| 467 |
"doItalicsAndBold" => 50, |
| 468 |
"doHardBreaks" => 60, |
| 469 |
); |
| 470 |
|
| 471 |
function runSpanGamut($text) { |
| 472 |
# |
| 473 |
# Run span gamut tranformations. |
| 474 |
# |
| 475 |
foreach ($this->span_gamut as $method => $priority) { |
| 476 |
$text = $this->$method($text); |
| 477 |
} |
| 478 |
|
| 479 |
return $text; |
| 480 |
} |
| 481 |
|
| 482 |
|
| 483 |
function doHardBreaks($text) { |
| 484 |
# Do hard breaks: |
| 485 |
return preg_replace_callback('/ {2,}\n/', |
| 486 |
array(&$this, '_doHardBreaks_callback'), $text); |
| 487 |
} |
| 488 |
function _doHardBreaks_callback($matches) { |
| 489 |
return $this->hashPart("<br$this->empty_element_suffix\n"); |
| 490 |
} |
| 491 |
|
| 492 |
|
| 493 |
function doAnchors($text) { |
| 494 |
# |
| 495 |
# Turn Markdown link shortcuts into XHTML <a> tags. |
| 496 |
# |
| 497 |
if ($this->in_anchor) return $text; |
| 498 |
$this->in_anchor = true; |
| 499 |
|
| 500 |
# |
| 501 |
# First, handle reference-style links: [link text] [id] |
| 502 |
# |
| 503 |
$text = preg_replace_callback('{ |
| 504 |
( # wrap whole match in $1 |
| 505 |
\[ |
| 506 |
('.$this->nested_brackets_re.') # link text = $2 |
| 507 |
\] |
| 508 |
|
| 509 |
[ ]? # one optional space |
| 510 |
(?:\n[ ]*)? # one optional newline followed by spaces |
| 511 |
|
| 512 |
\[ |
| 513 |
(.*?) # id = $3 |
| 514 |
\] |
| 515 |
) |
| 516 |
}xs', |
| 517 |
array(&$this, '_doAnchors_reference_callback'), $text); |
| 518 |
|
| 519 |
# |
| 520 |
# Next, inline-style links: [link text](url "optional title") |
| 521 |
# |
| 522 |
$text = preg_replace_callback('{ |
| 523 |
( # wrap whole match in $1 |
| 524 |
\[ |
| 525 |
('.$this->nested_brackets_re.') # link text = $2 |
| 526 |
\] |
| 527 |
\( # literal paren |
| 528 |
[ ]* |
| 529 |
(?: |
| 530 |
<(\S*)> # href = $3 |
| 531 |
| |
| 532 |
('.$this->nested_url_parenthesis_re.') # href = $4 |
| 533 |
) |
| 534 |
[ ]* |
| 535 |
( # $5 |
| 536 |
([\'"]) # quote char = $6 |
| 537 |
(.*?) # Title = $7 |
| 538 |
\6 # matching quote |
| 539 |
[ ]* # ignore any spaces/tabs between closing quote and ) |
| 540 |
)? # title is optional |
| 541 |
\) |
| 542 |
) |
| 543 |
}xs', |
| 544 |
array(&$this, '_DoAnchors_inline_callback'), $text); |
| 545 |
|
| 546 |
# |
| 547 |
# Last, handle reference-style shortcuts: [link text] |
| 548 |
# These must come last in case you've also got [link test][1] |
| 549 |
# or [link test](/foo) |
| 550 |
# |
| 551 |
// $text = preg_replace_callback('{ |
| 552 |
// ( # wrap whole match in $1 |
| 553 |
// \[ |
| 554 |
// ([^\[\]]+) # link text = $2; can\'t contain [ or ] |
| 555 |
// \] |
| 556 |
// ) |
| 557 |
// }xs', |
| 558 |
// array(&$this, '_doAnchors_reference_callback'), $text); |
| 559 |
|
| 560 |
$this->in_anchor = false; |
| 561 |
return $text; |
| 562 |
} |
| 563 |
function _doAnchors_reference_callback($matches) { |
| 564 |
$whole_match = $matches[1]; |
| 565 |
$link_text = $matches[2]; |
| 566 |
$link_id =& $matches[3]; |
| 567 |
|
| 568 |
if ($link_id == "") { |
| 569 |
# for shortcut links like [this][] or [this]. |
| 570 |
$link_id = $link_text; |
| 571 |
} |
| 572 |
|
| 573 |
# lower-case and turn embedded newlines into spaces |
| 574 |
$link_id = strtolower($link_id); |
| 575 |
$link_id = preg_replace('{[ ]?\n}', ' ', $link_id); |
| 576 |
|
| 577 |
if (isset($this->urls[$link_id])) { |
| 578 |
$url = $this->urls[$link_id]; |
| 579 |
$url = $this->encodeAttribute($url); |
| 580 |
|
| 581 |
$result = "<a href=\"$url\""; |
| 582 |
if ( isset( $this->titles[$link_id] ) ) { |
| 583 |
$title = $this->titles[$link_id]; |
| 584 |
$title = $this->encodeAttribute($title); |
| 585 |
$result .= " title=\"$title\""; |
| 586 |
} |
| 587 |
|
| 588 |
$link_text = $this->runSpanGamut($link_text); |
| 589 |
$result .= ">$link_text</a>"; |
| 590 |
$result = $this->hashPart($result); |
| 591 |
} |
| 592 |
else { |
| 593 |
$result = $whole_match; |
| 594 |
} |
| 595 |
return $result; |
| 596 |
} |
| 597 |
function _doAnchors_inline_callback($matches) { |
| 598 |
$whole_match = $matches[1]; |
| 599 |
$link_text = $this->runSpanGamut($matches[2]); |
| 600 |
$url = $matches[3] == '' ? $matches[4] : $matches[3]; |
| 601 |
$title =& $matches[7]; |
| 602 |
|
| 603 |
$url = $this->encodeAttribute($url); |
| 604 |
|
| 605 |
$result = "<a href=\"$url\""; |
| 606 |
if (isset($title)) { |
| 607 |
$title = $this->encodeAttribute($title); |
| 608 |
$result .= " title=\"$title\""; |
| 609 |
} |
| 610 |
|
| 611 |
$link_text = $this->runSpanGamut($link_text); |
| 612 |
$result .= ">$link_text</a>"; |
| 613 |
|
| 614 |
return $this->hashPart($result); |
| 615 |
} |
| 616 |
|
| 617 |
|
| 618 |
function doImages($text) { |
| 619 |
# |
| 620 |
# Turn Markdown image shortcuts into <img> tags. |
| 621 |
# |
| 622 |
# |
| 623 |
# First, handle reference-style labeled images: ![alt text][id] |
| 624 |
# |
| 625 |
$text = preg_replace_callback('{ |
| 626 |
( # wrap whole match in $1 |
| 627 |
!\[ |
| 628 |
('.$this->nested_brackets_re.') # alt text = $2 |
| 629 |
\] |
| 630 |
|
| 631 |
[ ]? # one optional space |
| 632 |
(?:\n[ ]*)? # one optional newline followed by spaces |
| 633 |
|
| 634 |
\[ |
| 635 |
(.*?) # id = $3 |
| 636 |
\] |
| 637 |
|
| 638 |
) |
| 639 |
}xs', |
| 640 |
array(&$this, '_doImages_reference_callback'), $text); |
| 641 |
|
| 642 |
# |
| 643 |
# Next, handle inline images:  |
| 644 |
# Don't forget: encode * and _ |
| 645 |
# |
| 646 |
$text = preg_replace_callback('{ |
| 647 |
( # wrap whole match in $1 |
| 648 |
!\[ |
| 649 |
('.$this->nested_brackets_re.') # alt text = $2 |
| 650 |
\] |
| 651 |
\s? # One optional whitespace character |
| 652 |
\( # literal paren |
| 653 |
[ ]* |
| 654 |
(?: |
| 655 |
<(\S*)> # src url = $3 |
| 656 |
| |
| 657 |
('.$this->nested_url_parenthesis_re.') # src url = $4 |
| 658 |
) |
| 659 |
[ ]* |
| 660 |
( # $5 |
| 661 |
([\'"]) # quote char = $6 |
| 662 |
(.*?) # title = $7 |
| 663 |
\6 # matching quote |
| 664 |
[ ]* |
| 665 |
)? # title is optional |
| 666 |
\) |
| 667 |
) |
| 668 |
}xs', |
| 669 |
array(&$this, '_doImages_inline_callback'), $text); |
| 670 |
|
| 671 |
return $text; |
| 672 |
} |
| 673 |
function _doImages_reference_callback($matches) { |
| 674 |
$whole_match = $matches[1]; |
| 675 |
$alt_text = $matches[2]; |
| 676 |
$link_id = strtolower($matches[3]); |
| 677 |
|
| 678 |
if ($link_id == "") { |
| 679 |
$link_id = strtolower($alt_text); # for shortcut links like ![this][]. |
| 680 |
} |
| 681 |
|
| 682 |
$alt_text = $this->encodeAttribute($alt_text); |
| 683 |
if (isset($this->urls[$link_id])) { |
| 684 |
$url = $this->encodeAttribute($this->urls[$link_id]); |
| 685 |
$result = "<img src=\"$url\" alt=\"$alt_text\""; |
| 686 |
if (isset($this->titles[$link_id])) { |
| 687 |
$title = $this->titles[$link_id]; |
| 688 |
$title = $this->encodeAttribute($title); |
| 689 |
$result .= " title=\"$title\""; |
| 690 |
} |
| 691 |
$result .= $this->empty_element_suffix; |
| 692 |
$result = $this->hashPart($result); |
| 693 |
} |
| 694 |
else { |
| 695 |
# If there's no such link ID, leave intact: |
| 696 |
$result = $whole_match; |
| 697 |
} |
| 698 |
|
| 699 |
return $result; |
| 700 |
} |
| 701 |
function _doImages_inline_callback($matches) { |
| 702 |
$whole_match = $matches[1]; |
| 703 |
$alt_text = $matches[2]; |
| 704 |
$url = $matches[3] == '' ? $matches[4] : $matches[3]; |
| 705 |
$title =& $matches[7]; |
| 706 |
|
| 707 |
$alt_text = $this->encodeAttribute($alt_text); |
| 708 |
$url = $this->encodeAttribute($url); |
| 709 |
$result = "<img src=\"$url\" alt=\"$alt_text\""; |
| 710 |
if (isset($title)) { |
| 711 |
$title = $this->encodeAttribute($title); |
| 712 |
$result .= " title=\"$title\""; # $title already quoted |
| 713 |
} |
| 714 |
$result .= $this->empty_element_suffix; |
| 715 |
|
| 716 |
return $this->hashPart($result); |
| 717 |
} |
| 718 |
|
| 719 |
|
| 720 |
function doHeaders($text) { |
| 721 |
# Setext-style headers: |
| 722 |
# Header 1 |
| 723 |
# ======== |
| 724 |
# |
| 725 |
# Header 2 |
| 726 |
# -------- |
| 727 |
# |
| 728 |
$text = preg_replace_callback('{ ^(.+?)[ ]*\n(=+|-+)[ ]*\n+ }mx', |
| 729 |
array(&$this, '_doHeaders_callback_setext'), $text); |
| 730 |
|
| 731 |
# atx-style headers: |
| 732 |
# # Header 1 |
| 733 |
# ## Header 2 |
| 734 |
# ## Header 2 with closing hashes ## |
| 735 |
# ... |
| 736 |
# ###### Header 6 |
| 737 |
# |
| 738 |
$text = preg_replace_callback('{ |
| 739 |
^(\#{1,6}) # $1 = string of #\'s |
| 740 |
[ ]* |
| 741 |
(.+?) # $2 = Header text |
| 742 |
[ ]* |
| 743 |
\#* # optional closing #\'s (not counted) |
| 744 |
\n+ |
| 745 |
}xm', |
| 746 |
array(&$this, '_doHeaders_callback_atx'), $text); |
| 747 |
|
| 748 |
return $text; |
| 749 |
} |
| 750 |
function _doHeaders_callback_setext($matches) { |
| 751 |
# Terrible hack to check we haven't found an empty list item. |
| 752 |
if ($matches[2] == '-' && preg_match('{^-(?: |$)}', $matches[1])) |
| 753 |
return $matches[0]; |
| 754 |
|
| 755 |
$level = $matches[2]{0} == '=' ? 1 : 2; |
| 756 |
$block = "<h$level>".$this->runSpanGamut($matches[1])."</h$level>"; |
| 757 |
return "\n" . $this->hashBlock($block) . "\n\n"; |
| 758 |
} |
| 759 |
function _doHeaders_callback_atx($matches) { |
| 760 |
$level = strlen($matches[1]); |
| 761 |
$block = "<h$level>".$this->runSpanGamut($matches[2])."</h$level>"; |
| 762 |
return "\n" . $this->hashBlock($block) . "\n\n"; |
| 763 |
} |
| 764 |
|
| 765 |
|
| 766 |
function doLists($text) { |
| 767 |
# |
| 768 |
# Form HTML ordered (numbered) and unordered (bulleted) lists. |
| 769 |
# |
| 770 |
$less_than_tab = $this->tab_width - 1; |
| 771 |
|
| 772 |
# Re-usable patterns to match list item bullets and number markers: |
| 773 |
$marker_ul_re = '[*+-]'; |
| 774 |
$marker_ol_re = '\d+[.]'; |
| 775 |
$marker_any_re = "(?:$marker_ul_re|$marker_ol_re)"; |
| 776 |
|
| 777 |
$markers_relist = array($marker_ul_re, $marker_ol_re); |
| 778 |
|
| 779 |
foreach ($markers_relist as $marker_re) { |
| 780 |
# Re-usable pattern to match any entirel ul or ol list: |
| 781 |
$whole_list_re = ' |
| 782 |
( # $1 = whole list |
| 783 |
( # $2 |
| 784 |
[ ]{0,'.$less_than_tab.'} |
| 785 |
('.$marker_re.') # $3 = first list item marker |
| 786 |
[ ]+ |
| 787 |
) |
| 788 |
(?s:.+?) |
| 789 |
( # $4 |
| 790 |
\z |
| 791 |
| |
| 792 |
\n{2,} |
| 793 |
(?=\S) |
| 794 |
(?! # Negative lookahead for another list item marker |
| 795 |
[ ]* |
| 796 |
'.$marker_re.'[ ]+ |
| 797 |
) |
| 798 |
) |
| 799 |
) |
| 800 |
'; // mx |
| 801 |
|
| 802 |
# We use a different prefix before nested lists than top-level lists. |
| 803 |
# See extended comment in _ProcessListItems(). |
| 804 |
|
| 805 |
if ($this->list_level) { |
| 806 |
$text = preg_replace_callback('{ |
| 807 |
^ |
| 808 |
'.$whole_list_re.' |
| 809 |
}mx', |
| 810 |
array(&$this, '_doLists_callback'), $text); |
| 811 |
} |
| 812 |
else { |
| 813 |
$text = preg_replace_callback('{ |
| 814 |
(?:(?<=\n)\n|\A\n?) # Must eat the newline |
| 815 |
'.$whole_list_re.' |
| 816 |
}mx', |
| 817 |
array(&$this, '_doLists_callback'), $text); |
| 818 |
} |
| 819 |
} |
| 820 |
|
| 821 |
return $text; |
| 822 |
} |
| 823 |
function _doLists_callback($matches) { |
| 824 |
# Re-usable patterns to match list item bullets and number markers: |
| 825 |
$marker_ul_re = '[*+-]'; |
| 826 |
$marker_ol_re = '\d+[.]'; |
| 827 |
$marker_any_re = "(?:$marker_ul_re|$marker_ol_re)"; |
| 828 |
|
| 829 |
$list = $matches[1]; |
| 830 |
$list_type = preg_match("/$marker_ul_re/", $matches[3]) ? "ul" : "ol"; |
| 831 |
|
| 832 |
$marker_any_re = ( $list_type == "ul" ? $marker_ul_re : $marker_ol_re ); |
| 833 |
|
| 834 |
$list .= "\n"; |
| 835 |
$result = $this->processListItems($list, $marker_any_re); |
| 836 |
|
| 837 |
$result = $this->hashBlock("<$list_type>\n" . $result . "</$list_type>"); |
| 838 |
return "\n". $result ."\n\n"; |
| 839 |
} |
| 840 |
|
| 841 |
var $list_level = 0; |
| 842 |
|
| 843 |
function processListItems($list_str, $marker_any_re) { |
| 844 |
# |
| 845 |
# Process the contents of a single ordered or unordered list, splitting it |
| 846 |
# into individual list items. |
| 847 |
# |
| 848 |
# The $this->list_level global keeps track of when we're inside a list. |
| 849 |
# Each time we enter a list, we increment it; when we leave a list, |
| 850 |
# we decrement. If it's zero, we're not in a list anymore. |
| 851 |
# |
| 852 |
# We do this because when we're not inside a list, we want to treat |
| 853 |
# something like this: |
| 854 |
# |
| 855 |
# I recommend upgrading to version |
| 856 |
# 8. Oops, now this line is treated |
| 857 |
# as a sub-list. |
| 858 |
# |
| 859 |
# As a single paragraph, despite the fact that the second line starts |
| 860 |
# with a digit-period-space sequence. |
| 861 |
# |
| 862 |
# Whereas when we're inside a list (or sub-list), that line will be |
| 863 |
# treated as the start of a sub-list. What a kludge, huh? This is |
| 864 |
# an aspect of Markdown's syntax that's hard to parse perfectly |
| 865 |
# without resorting to mind-reading. Perhaps the solution is to |
| 866 |
# change the syntax rules such that sub-lists must start with a |
| 867 |
# starting cardinal number; e.g. "1." or "a.". |
| 868 |
|
| 869 |
$this->list_level++; |
| 870 |
|
| 871 |
# trim trailing blank lines: |
| 872 |
$list_str = preg_replace("/\n{2,}\\z/", "\n", $list_str); |
| 873 |
|
| 874 |
$list_str = preg_replace_callback('{ |
| 875 |
(\n)? # leading line = $1 |
| 876 |
(^[ ]*) # leading whitespace = $2 |
| 877 |
('.$marker_any_re.' # list marker and space = $3 |
| 878 |
(?:[ ]+|(?=\n)) # space only required if item is not empty |
| 879 |
) |
| 880 |
((?s:.*?)) # list item text = $4 |
| 881 |
(?:(\n+(?=\n))|\n) # tailing blank line = $5 |
| 882 |
(?= \n* (\z | \2 ('.$marker_any_re.') (?:[ ]+|(?=\n)))) |
| 883 |
}xm', |
| 884 |
array(&$this, '_processListItems_callback'), $list_str); |
| 885 |
|
| 886 |
$this->list_level--; |
| 887 |
return $list_str; |
| 888 |
} |
| 889 |
function _processListItems_callback($matches) { |
| 890 |
$item = $matches[4]; |
| 891 |
$leading_line =& $matches[1]; |
| 892 |
$leading_space =& $matches[2]; |
| 893 |
$marker_space = $matches[3]; |
| 894 |
$tailing_blank_line =& $matches[5]; |
| 895 |
|
| 896 |
if ($leading_line || $tailing_blank_line || |
| 897 |
preg_match('/\n{2,}/', $item)) |
| 898 |
{ |
| 899 |
# Replace marker with the appropriate whitespace indentation |
| 900 |
$item = $leading_space . str_repeat(' ', strlen($marker_space)) . $item; |
| 901 |
$item = $this->runBlockGamut($this->outdent($item)."\n"); |
| 902 |
} |
| 903 |
else { |
| 904 |
# Recursion for sub-lists: |
| 905 |
$item = $this->doLists($this->outdent($item)); |
| 906 |
$item = preg_replace('/\n+$/', '', $item); |
| 907 |
$item = $this->runSpanGamut($item); |
| 908 |
} |
| 909 |
|
| 910 |
return "<li>" . $item . "</li>\n"; |
| 911 |
} |
| 912 |
|
| 913 |
|
| 914 |
function doCodeBlocks($text) { |
| 915 |
# |
| 916 |
# Process Markdown `<pre><code>` blocks. |
| 917 |
# |
| 918 |
$text = preg_replace_callback('{ |
| 919 |
(?:\n\n|\A\n?) |
| 920 |
( # $1 = the code block -- one or more lines, starting with a space/tab |
| 921 |
(?> |
| 922 |
[ ]{'.$this->tab_width.'} # Lines must start with a tab or a tab-width of spaces |
| 923 |
.*\n+ |
| 924 |
)+ |
| 925 |
) |
| 926 |
((?=^[ ]{0,'.$this->tab_width.'}\S)|\Z) # Lookahead for non-space at line-start, or end of doc |
| 927 |
}xm', |
| 928 |
array(&$this, '_doCodeBlocks_callback'), $text); |
| 929 |
|
| 930 |
return $text; |
| 931 |
} |
| 932 |
function _doCodeBlocks_callback($matches) { |
| 933 |
$codeblock = $matches[1]; |
| 934 |
|
| 935 |
$codeblock = $this->outdent($codeblock); |
| 936 |
$codeblock = htmlspecialchars($codeblock, ENT_NOQUOTES); |
| 937 |
|
| 938 |
# trim leading newlines and trailing newlines |
| 939 |
$codeblock = preg_replace('/\A\n+|\n+\z/', '', $codeblock); |
| 940 |
|
| 941 |
$codeblock = "<pre><code>$codeblock\n</code></pre>"; |
| 942 |
return "\n\n".$this->hashBlock($codeblock)."\n\n"; |
| 943 |
} |
| 944 |
|
| 945 |
|
| 946 |
function makeCodeSpan($code) { |
| 947 |
# |
| 948 |
# Create a code span markup for $code. Called from handleSpanToken. |
| 949 |
# |
| 950 |
$code = htmlspecialchars(trim($code), ENT_NOQUOTES); |
| 951 |
return $this->hashPart("<code>$code</code>"); |
| 952 |
} |
| 953 |
|
| 954 |
|
| 955 |
var $em_relist = array( |
| 956 |
'' => '(?:(?<!\*)\*(?!\*)|(?<!_)_(?!_))(?=\S)(?![.,:;]\s)', |
| 957 |
'*' => '(?<=\S)(?<!\*)\*(?!\*)', |
| 958 |
'_' => '(?<=\S)(?<!_)_(?!_)', |
| 959 |
); |
| 960 |
var $strong_relist = array( |
| 961 |
'' => '(?:(?<!\*)\*\*(?!\*)|(?<!_)__(?!_))(?=\S)(?![.,:;]\s)', |
| 962 |
'**' => '(?<=\S)(?<!\*)\*\*(?!\*)', |
| 963 |
'__' => '(?<=\S)(?<!_)__(?!_)', |
| 964 |
); |
| 965 |
var $em_strong_relist = array( |
| 966 |
'' => '(?:(?<!\*)\*\*\*(?!\*)|(?<!_)___(?!_))(?=\S)(?![.,:;]\s)', |
| 967 |
'***' => '(?<=\S)(?<!\*)\*\*\*(?!\*)', |
| 968 |
'___' => '(?<=\S)(?<!_)___(?!_)', |
| 969 |
); |
| 970 |
var $em_strong_prepared_relist; |
| 971 |
|
| 972 |
function prepareItalicsAndBold() { |
| 973 |
# |
| 974 |
# Prepare regular expressions for seraching emphasis tokens in any |
| 975 |
# context. |
| 976 |
# |
| 977 |
foreach ($this->em_relist as $em => $em_re) { |
| 978 |
foreach ($this->strong_relist as $strong => $strong_re) { |
| 979 |
# Construct list of allowed token expressions. |
| 980 |
$token_relist = array(); |
| 981 |
if (isset($this->em_strong_relist["$em$strong"])) { |
| 982 |
$token_relist[] = $this->em_strong_relist["$em$strong"]; |
| 983 |
} |
| 984 |
$token_relist[] = $em_re; |
| 985 |
$token_relist[] = $strong_re; |
| 986 |
|
| 987 |
# Construct master expression from list. |
| 988 |
$token_re = '{('. implode('|', $token_relist) .')}'; |
| 989 |
$this->em_strong_prepared_relist["$em$strong"] = $token_re; |
| 990 |
} |
| 991 |
} |
| 992 |
} |
| 993 |
|
| 994 |
function doItalicsAndBold($text) { |
| 995 |
$token_stack = array(''); |
| 996 |
$text_stack = array(''); |
| 997 |
$em = ''; |
| 998 |
$strong = ''; |
| 999 |
$tree_char_em = false; |
| 1000 |
|
| 1001 |
while (1) { |
| 1002 |
# |
| 1003 |
# Get prepared regular expression for seraching emphasis tokens |
| 1004 |
# in current context. |
| 1005 |
# |
| 1006 |
$token_re = $this->em_strong_prepared_relist["$em$strong"]; |
| 1007 |
|
| 1008 |
# |
| 1009 |
# Each loop iteration seach for the next emphasis token. |
| 1010 |
# Each token is then passed to handleSpanToken. |
| 1011 |
# |
| 1012 |
$parts = preg_split($token_re, $text, 2, PREG_SPLIT_DELIM_CAPTURE); |
| 1013 |
$text_stack[0] .= $parts[0]; |
| 1014 |
$token =& $parts[1]; |
| 1015 |
$text =& $parts[2]; |
| 1016 |
|
| 1017 |
if (empty($token)) { |
| 1018 |
# Reached end of text span: empty stack without emitting. |
| 1019 |
# any more emphasis. |
| 1020 |
while ($token_stack[0]) { |
| 1021 |
$text_stack[1] .= array_shift($token_stack); |
| 1022 |
$text_stack[0] .= array_shift($text_stack); |
| 1023 |
} |
| 1024 |
break; |
| 1025 |
} |
| 1026 |
|
| 1027 |
$token_len = strlen($token); |
| 1028 |
if ($tree_char_em) { |
| 1029 |
# Reached closing marker while inside a three-char emphasis. |
| 1030 |
if ($token_len == 3) { |
| 1031 |
# Three-char closing marker, close em and strong. |
| 1032 |
array_shift($token_stack); |
| 1033 |
$span = array_shift($text_stack); |
| 1034 |
$span = $this->runSpanGamut($span); |
| 1035 |
$span = "<strong><em>$span</em></strong>"; |
| 1036 |
$text_stack[0] .= $this->hashPart($span); |
| 1037 |
$em = ''; |
| 1038 |
$strong = ''; |
| 1039 |
} else { |
| 1040 |
# Other closing marker: close one em or strong and |
| 1041 |
# change current token state to match the other |
| 1042 |
$token_stack[0] = str_repeat($token{0}, 3-$token_len); |
| 1043 |
$tag = $token_len == 2 ? "strong" : "em"; |
| 1044 |
$span = $text_stack[0]; |
| 1045 |
$span = $this->runSpanGamut($span); |
| 1046 |
$span = "<$tag>$span</$tag>"; |
| 1047 |
$text_stack[0] = $this->hashPart($span); |
| 1048 |
$$tag = ''; # $$tag stands for $em or $strong |
| 1049 |
} |
| 1050 |
$tree_char_em = false; |
| 1051 |
} else if ($token_len == 3) { |
| 1052 |
if ($em) { |
| 1053 |
# Reached closing marker for both em and strong. |
| 1054 |
# Closing strong marker: |
| 1055 |
for ($i = 0; $i < 2; ++$i) { |
| 1056 |
$shifted_token = array_shift($token_stack); |
| 1057 |
$tag = strlen($shifted_token) == 2 ? "strong" : "em"; |
| 1058 |
$span = array_shift($text_stack); |
| 1059 |
$span = $this->runSpanGamut($span); |
| 1060 |
$span = "<$tag>$span</$tag>"; |
| 1061 |
$text_stack[0] .= $this->hashPart($span); |
| 1062 |
$$tag = ''; # $$tag stands for $em or $strong |
| 1063 |
} |
| 1064 |
} else { |
| 1065 |
# Reached opening three-char emphasis marker. Push on token |
| 1066 |
# stack; will be handled by the special condition above. |
| 1067 |
$em = $token{0}; |
| 1068 |
$strong = "$em$em"; |
| 1069 |
array_unshift($token_stack, $token); |
| 1070 |
array_unshift($text_stack, ''); |
| 1071 |
$tree_char_em = true; |
| 1072 |
} |
| 1073 |
} else if ($token_len == 2) { |
| 1074 |
if ($strong) { |
| 1075 |
# Unwind any dangling emphasis marker: |
| 1076 |
if (strlen($token_stack[0]) == 1) { |
| 1077 |
$text_stack[1] .= array_shift($token_stack); |
| 1078 |
$text_stack[0] .= array_shift($text_stack); |
| 1079 |
} |
| 1080 |
# Closing strong marker: |
| 1081 |
array_shift($token_stack); |
| 1082 |
$span = array_shift($text_stack); |
| 1083 |
$span = $this->runSpanGamut($span); |
| 1084 |
$span = "<strong>$span</strong>"; |
| 1085 |
$text_stack[0] .= $this->hashPart($span); |
| 1086 |
$strong = ''; |
| 1087 |
} else { |
| 1088 |
array_unshift($token_stack, $token); |
| 1089 |
array_unshift($text_stack, ''); |
| 1090 |
$strong = $token; |
| 1091 |
} |
| 1092 |
} else { |
| 1093 |
# Here $token_len == 1 |
| 1094 |
if ($em) { |
| 1095 |
if (strlen($token_stack[0]) == 1) { |
| 1096 |
# Closing emphasis marker: |
| 1097 |
array_shift($token_stack); |
| 1098 |
$span = array_shift($text_stack); |
| 1099 |
$span = $this->runSpanGamut($span); |
| 1100 |
$span = "<em>$span</em>"; |
| 1101 |
$text_stack[0] .= $this->hashPart($span); |
| 1102 |
$em = ''; |
| 1103 |
} else { |
| 1104 |
$text_stack[0] .= $token; |
| 1105 |
} |
| 1106 |
} else { |
| 1107 |
array_unshift($token_stack, $token); |
| 1108 |
array_unshift($text_stack, ''); |
| 1109 |
$em = $token; |
| 1110 |
} |
| 1111 |
} |
| 1112 |
} |
| 1113 |
return $text_stack[0]; |
| 1114 |
} |
| 1115 |
|
| 1116 |
|
| 1117 |
function doBlockQuotes($text) { |
| 1118 |
$text = preg_replace_callback('/ |
| 1119 |
( # Wrap whole match in $1 |
| 1120 |
(?> |
| 1121 |
^[ ]*>[ ]? # ">" at the start of a line |
| 1122 |
.+\n # rest of the first line |
| 1123 |
(.+\n)* # subsequent consecutive lines |
| 1124 |
\n* # blanks |
| 1125 |
)+ |
| 1126 |
) |
| 1127 |
/xm', |
| 1128 |
array(&$this, '_doBlockQuotes_callback'), $text); |
| 1129 |
|
| 1130 |
return $text; |
| 1131 |
} |
| 1132 |
function _doBlockQuotes_callback($matches) { |
| 1133 |
$bq = $matches[1]; |
| 1134 |
# trim one level of quoting - trim whitespace-only lines |
| 1135 |
$bq = preg_replace('/^[ ]*>[ ]?|^[ ]+$/m', '', $bq); |
| 1136 |
$bq = $this->runBlockGamut($bq); # recurse |
| 1137 |
|
| 1138 |
$bq = preg_replace('/^/m', " ", $bq); |
| 1139 |
# These leading spaces cause problem with <pre> content, |
| 1140 |
# so we need to fix that: |
| 1141 |
$bq = preg_replace_callback('{(\s*<pre>.+?</pre>)}sx', |
| 1142 |
array(&$this, '_DoBlockQuotes_callback2'), $bq); |
| 1143 |
|
| 1144 |
return "\n". $this->hashBlock("<blockquote>\n$bq\n</blockquote>")."\n\n"; |
| 1145 |
} |
| 1146 |
function _doBlockQuotes_callback2($matches) { |
| 1147 |
$pre = $matches[1]; |
| 1148 |
$pre = preg_replace('/^ /m', '', $pre); |
| 1149 |
return $pre; |
| 1150 |
} |
| 1151 |
|
| 1152 |
|
| 1153 |
function formParagraphs($text) { |
| 1154 |
# |
| 1155 |
# Params: |
| 1156 |
# $text - string to process with html <p> tags |
| 1157 |
# |
| 1158 |
# Strip leading and trailing lines: |
| 1159 |
$text = preg_replace('/\A\n+|\n+\z/', '', $text); |
| 1160 |
|
| 1161 |
$grafs = preg_split('/\n{2,}/', $text, -1, PREG_SPLIT_NO_EMPTY); |
| 1162 |
|
| 1163 |
# |
| 1164 |
# Wrap <p> tags and unhashify HTML blocks |
| 1165 |
# |
| 1166 |
foreach ($grafs as $key => $value) { |
| 1167 |
if (!preg_match('/^B\x1A[0-9]+B$/', $value)) { |
| 1168 |
# Is a paragraph. |
| 1169 |
$value = $this->runSpanGamut($value); |
| 1170 |
$value = preg_replace('/^([ ]*)/', "<p>", $value); |
| 1171 |
$value .= "</p>"; |
| 1172 |
$grafs[$key] = $this->unhash($value); |
| 1173 |
} |
| 1174 |
else { |
| 1175 |
# Is a block. |
| 1176 |
# Modify elements of @grafs in-place... |
| 1177 |
$graf = $value; |
| 1178 |
$block = $this->html_hashes[$graf]; |
| 1179 |
$graf = $block; |
| 1180 |
// if (preg_match('{ |
| 1181 |
// \A |
| 1182 |
// ( # $1 = <div> tag |
| 1183 |
// <div \s+ |
| 1184 |
// [^>]* |
| 1185 |
// \b |
| 1186 |
// markdown\s*=\s* ([\'"]) # $2 = attr quote char |
| 1187 |
// 1 |
| 1188 |
// \2 |
| 1189 |
// [^>]* |
| 1190 |
// > |
| 1191 |
// ) |
| 1192 |
// ( # $3 = contents |
| 1193 |
// .* |
| 1194 |
// ) |
| 1195 |
// (</div>) # $4 = closing tag |
| 1196 |
// \z |
| 1197 |
// }xs', $block, $matches)) |
| 1198 |
// { |
| 1199 |
// list(, $div_open, , $div_content, $div_close) = $matches; |
| 1200 |
// |
| 1201 |
// # We can't call Markdown(), because that resets the hash; |
| 1202 |
// # that initialization code should be pulled into its own sub, though. |
| 1203 |
// $div_content = $this->hashHTMLBlocks($div_content); |
| 1204 |
// |
| 1205 |
// # Run document gamut methods on the content. |
| 1206 |
// foreach ($this->document_gamut as $method => $priority) { |
| 1207 |
// $div_content = $this->$method($div_content); |
| 1208 |
// } |
| 1209 |
// |
| 1210 |
// $div_open = preg_replace( |
| 1211 |
// '{\smarkdown\s*=\s*([\'"]).+?\1}', '', $div_open); |
| 1212 |
// |
| 1213 |
// $graf = $div_open . "\n" . $div_content . "\n" . $div_close; |
| 1214 |
// } |
| 1215 |
$grafs[$key] = $graf; |
| 1216 |
} |
| 1217 |
} |
| 1218 |
|
| 1219 |
return implode("\n\n", $grafs); |
| 1220 |
} |
| 1221 |
|
| 1222 |
|
| 1223 |
function encodeAttribute($text) { |
| 1224 |
# |
| 1225 |
# Encode text for a double-quoted HTML attribute. This function |
| 1226 |
# is *not* suitable for attributes enclosed in single quotes. |
| 1227 |
# |
| 1228 |
$text = $this->encodeAmpsAndAngles($text); |
| 1229 |
$text = str_replace('"', '"', $text); |
| 1230 |
return $text; |
| 1231 |
} |
| 1232 |
|
| 1233 |
|
| 1234 |
function encodeAmpsAndAngles($text) { |
| 1235 |
# |
| 1236 |
# Smart processing for ampersands and angle brackets that need to |
| 1237 |
# be encoded. Valid character entities are left alone unless the |
| 1238 |
# no-entities mode is set. |
| 1239 |
# |
| 1240 |
if ($this->no_entities) { |
| 1241 |
$text = str_replace('&', '&', $text); |
| 1242 |
} else { |
| 1243 |
# Ampersand-encoding based entirely on Nat Irons's Amputator |
| 1244 |
# MT plugin: <http://bumppo.net/projects/amputator/> |
| 1245 |
$text = preg_replace('/&(?!#?[xX]?(?:[0-9a-fA-F]+|\w+);)/', |
| 1246 |
'&', $text);; |
| 1247 |
} |
| 1248 |
# Encode remaining <'s |
| 1249 |
$text = str_replace('<', '<', $text); |
| 1250 |
|
| 1251 |
return $text; |
| 1252 |
} |
| 1253 |
|
| 1254 |
|
| 1255 |
function doAutoLinks($text) { |
| 1256 |
$text = preg_replace_callback('{<((https?|ftp|dict):[^\'">\s]+)>}i', |
| 1257 |
array(&$this, '_doAutoLinks_url_callback'), $text); |
| 1258 |
|
| 1259 |
# Email addresses: <address@domain.foo> |
| 1260 |
$text = preg_replace_callback('{ |
| 1261 |
< |
| 1262 |
(?:mailto:)? |
| 1263 |
( |
| 1264 |
[-.\w\x80-\xFF]+ |
| 1265 |
\@ |
| 1266 |
[-a-z0-9\x80-\xFF]+(\.[-a-z0-9\x80-\xFF]+)*\.[a-z]+ |
| 1267 |
) |
| 1268 |
> |
| 1269 |
}xi', |
| 1270 |
array(&$this, '_doAutoLinks_email_callback'), $text); |
| 1271 |
|
| 1272 |
return $text; |
| 1273 |
} |
| 1274 |
function _doAutoLinks_url_callback($matches) { |
| 1275 |
$url = $this->encodeAttribute($matches[1]); |
| 1276 |
$link = "<a href=\"$url\">$url</a>"; |
| 1277 |
return $this->hashPart($link); |
| 1278 |
} |
| 1279 |
function _doAutoLinks_email_callback($matches) { |
| 1280 |
$address = $matches[1]; |
| 1281 |
$link = $this->encodeEmailAddress($address); |
| 1282 |
return $this->hashPart($link); |
| 1283 |
} |
| 1284 |
|
| 1285 |
|
| 1286 |
function encodeEmailAddress($addr) { |
| 1287 |
# |
| 1288 |
# Input: an email address, e.g. "foo@example.com" |
| 1289 |
# |
| 1290 |
# Output: the email address as a mailto link, with each character |
| 1291 |
# of the address encoded as either a decimal or hex entity, in |
| 1292 |
# the hopes of foiling most address harvesting spam bots. E.g.: |
| 1293 |
# |
| 1294 |
# <p><a href="mailto:foo |
| 1295 |
# @example.co |
| 1296 |
# m">foo@exampl |
| 1297 |
# e.com</a></p> |
| 1298 |
# |
| 1299 |
# Based by a filter by Matthew Wickline, posted to BBEdit-Talk. |
| 1300 |
# With some optimizations by Milian Wolff. |
| 1301 |
# |
| 1302 |
$addr = "mailto:" . $addr; |
| 1303 |
$chars = preg_split('/(?<!^)(?!$)/', $addr); |
| 1304 |
$seed = (int)abs(crc32($addr) / strlen($addr)); # Deterministic seed. |
| 1305 |
|
| 1306 |
foreach ($chars as $key => $char) { |
| 1307 |
$ord = ord($char); |
| 1308 |
# Ignore non-ascii chars. |
| 1309 |
if ($ord < 128) { |
| 1310 |
$r = ($seed * (1 + $key)) % 100; # Pseudo-random function. |
| 1311 |
# roughly 10% raw, 45% hex, 45% dec |
| 1312 |
# '@' *must* be encoded. I insist. |
| 1313 |
if ($r > 90 && $char != '@') /* do nothing */; |
| 1314 |
else if ($r < 45) $chars[$key] = '&#x'.dechex($ord).';'; |
| 1315 |
else $chars[$key] = '&#'.$ord.';'; |
| 1316 |
} |
| 1317 |
} |
| 1318 |
|
| 1319 |
$addr = implode('', $chars); |
| 1320 |
$text = implode('', array_slice($chars, 7)); # text without `mailto:` |
| 1321 |
$addr = "<a href=\"$addr\">$text</a>"; |
| 1322 |
|
| 1323 |
return $addr; |
| 1324 |
} |
| 1325 |
|
| 1326 |
|
| 1327 |
function parseSpan($str) { |
| 1328 |
# |
| 1329 |
# Take the string $str and parse it into tokens, hashing embeded HTML, |
| 1330 |
# escaped characters and handling code spans. |
| 1331 |
# |
| 1332 |
$output = ''; |
| 1333 |
|
| 1334 |
$span_re = '{ |
| 1335 |
( |
| 1336 |
\\\\'.$this->escape_chars_re.' |
| 1337 |
| |
| 1338 |
(?<![`\\\\]) |
| 1339 |
`+ # code span marker |
| 1340 |
'.( $this->no_markup ? '' : ' |
| 1341 |
| |
| 1342 |
<!-- .*? --> # comment |
| 1343 |
| |
| 1344 |
<\?.*?\?> | <%.*?%> # processing instruction |
| 1345 |
| |
| 1346 |
<[/!$]?[-a-zA-Z0-9:]+ # regular tags |
| 1347 |
(?> |
| 1348 |
\s |
| 1349 |
(?>[^"\'>]+|"[^"]*"|\'[^\']*\')* |
| 1350 |
)? |
| 1351 |
> |
| 1352 |
').' |
| 1353 |
) |
| 1354 |
}xs'; |
| 1355 |
|
| 1356 |
while (1) { |
| 1357 |
# |
| 1358 |
# Each loop iteration seach for either the next tag, the next |
| 1359 |
# openning code span marker, or the next escaped character. |
| 1360 |
# Each token is then passed to handleSpanToken. |
| 1361 |
# |
| 1362 |
$parts = preg_split($span_re, $str, 2, PREG_SPLIT_DELIM_CAPTURE); |
| 1363 |
|
| 1364 |
# Create token from text preceding tag. |
| 1365 |
if ($parts[0] != "") { |
| 1366 |
$output .= $parts[0]; |
| 1367 |
} |
| 1368 |
|
| 1369 |
# Check if we reach the end. |
| 1370 |
if (isset($parts[1])) { |
| 1371 |
$output .= $this->handleSpanToken($parts[1], $parts[2]); |
| 1372 |
$str = $parts[2]; |
| 1373 |
} |
| 1374 |
else { |
| 1375 |
break; |
| 1376 |
} |
| 1377 |
} |
| 1378 |
|
| 1379 |
return $output; |
| 1380 |
} |
| 1381 |
|
| 1382 |
|
| 1383 |
function handleSpanToken($token, &$str) { |
| 1384 |
# |
| 1385 |
# Handle $token provided by parseSpan by determining its nature and |
| 1386 |
# returning the corresponding value that should replace it. |
| 1387 |
# |
| 1388 |
switch ($token{0}) { |
| 1389 |
case "\\": |
| 1390 |
return $this->hashPart("&#". ord($token{1}). ";"); |
| 1391 |
case "`": |
| 1392 |
# Search for end marker in remaining text. |
| 1393 |
if (preg_match('/^(.*?[^`])'.preg_quote($token).'(?!`)(.*)$/sm', |
| 1394 |
$str, $matches)) |
| 1395 |
{ |
| 1396 |
$str = $matches[2]; |
| 1397 |
$codespan = $this->makeCodeSpan($matches[1]); |
| 1398 |
return $this->hashPart($codespan); |
| 1399 |
} |
| 1400 |
return $token; // return as text since no ending marker found. |
| 1401 |
default: |
| 1402 |
return $this->hashPart($token); |
| 1403 |
} |
| 1404 |
} |
| 1405 |
|
| 1406 |
|
| 1407 |
function outdent($text) { |
| 1408 |
# |
| 1409 |
# Remove one level of line-leading tabs or spaces |
| 1410 |
# |
| 1411 |
return preg_replace('/^(\t|[ ]{1,'.$this->tab_width.'})/m', '', $text); |
| 1412 |
} |
| 1413 |
|
| 1414 |
|
| 1415 |
# String length function for detab. `_initDetab` will create a function to |
| 1416 |
# hanlde UTF-8 if the default function does not exist. |
| 1417 |
var $utf8_strlen = 'mb_strlen'; |
| 1418 |
|
| 1419 |
function detab($text) { |
| 1420 |
# |
| 1421 |
# Replace tabs with the appropriate amount of space. |
| 1422 |
# |
| 1423 |
# For each line we separate the line in blocks delemited by |
| 1424 |
# tab characters. Then we reconstruct every line by adding the |
| 1425 |
# appropriate number of space between each blocks. |
| 1426 |
|
| 1427 |
$text = preg_replace_callback('/^.*\t.*$/m', |
| 1428 |
array(&$this, '_detab_callback'), $text); |
| 1429 |
|
| 1430 |
return $text; |
| 1431 |
} |
| 1432 |
function _detab_callback($matches) { |
| 1433 |
$line = $matches[0]; |
| 1434 |
$strlen = $this->utf8_strlen; # strlen function for UTF-8. |
| 1435 |
|
| 1436 |
# Split in blocks. |
| 1437 |
$blocks = explode("\t", $line); |
| 1438 |
# Add each blocks to the line. |
| 1439 |
$line = $blocks[0]; |
| 1440 |
unset($blocks[0]); # Do not add first block twice. |
| 1441 |
foreach ($blocks as $block) { |
| 1442 |
# Calculate amount of space, insert spaces, insert block. |
| 1443 |
$amount = $this->tab_width - |
| 1444 |
$strlen($line, 'UTF-8') % $this->tab_width; |
| 1445 |
$line .= str_repeat(" ", $amount) . $block; |
| 1446 |
} |
| 1447 |
return $line; |
| 1448 |
} |
| 1449 |
function _initDetab() { |
| 1450 |
# |
| 1451 |
# Check for the availability of the function in the `utf8_strlen` property |
| 1452 |
# (initially `mb_strlen`). If the function is not available, create a |
| 1453 |
# function that will loosely count the number of UTF-8 characters with a |
| 1454 |
# regular expression. |
| 1455 |
# |
| 1456 |
if (function_exists($this->utf8_strlen)) return; |
| 1457 |
$this->utf8_strlen = create_function('$text', 'return preg_match_all( |
| 1458 |
"/[\\\\x00-\\\\xBF]|[\\\\xC0-\\\\xFF][\\\\x80-\\\\xBF]*/", |
| 1459 |
$text, $m);'); |
| 1460 |
} |
| 1461 |
|
| 1462 |
|
| 1463 |
function unhash($text) { |
| 1464 |
# |
| 1465 |
# Swap back in all the tags hashed by _HashHTMLBlocks. |
| 1466 |
# |
| 1467 |
return preg_replace_callback('/(.)\x1A[0-9]+\1/', |
| 1468 |
array(&$this, '_unhash_callback'), $text); |
| 1469 |
} |
| 1470 |
function _unhash_callback($matches) { |
| 1471 |
return $this->html_hashes[$matches[0]]; |
| 1472 |
} |
| 1473 |
|
| 1474 |
} |
| 1475 |
?> |
