| 1 |
<?php |
| 2 |
|
| 3 |
// ======================== |
| 4 |
// Prefixed Markdown Reader |
| 5 |
// ======================== |
| 6 |
// for Plugin Panel Loader 1.3.1+ |
| 7 |
// ------------------------------ |
| 8 |
|
| 9 |
// === Markdown === |
| 10 |
// - Markdown Function |
| 11 |
// - Markdown Parser Class |
| 12 |
// - Markdown Extra Parser Class |
| 13 |
// - WordPress Readme Parser |
| 14 |
// - Github Flavoured Markdown |
| 15 |
|
| 16 |
// === Usage === |
| 17 |
// Simply replace all occurrences of radio_station_ in this file with the plugin namespace prefix eg. my_plugin_ |
| 18 |
// that matches your usage in loader.php - and remember to repeat this process if/when updating! |
| 19 |
|
| 20 |
|
| 21 |
# |
| 22 |
# Markdown Extra - A text-to-HTML conversion tool for web writers |
| 23 |
# |
| 24 |
# PHP Markdown & Extra |
| 25 |
# Copyright (c) 2004-2013 Michel Fortin |
| 26 |
# <http://michelf.ca/projects/php-markdown/> |
| 27 |
# |
| 28 |
# Original Markdown |
| 29 |
# Copyright (c) 2004-2006 John Gruber |
| 30 |
# <http://daringfireball.net/projects/markdown/> |
| 31 |
# |
| 32 |
# Tweaked to remove WordPress interface |
| 33 |
|
| 34 |
if ( !defined( 'ABSPATH' ) ) exit; |
| 35 |
|
| 36 |
// if (!defined('MARKDOWN_VERSION')) { |
| 37 |
// define( 'MARKDOWN_VERSION', "1.0.2" ); # 29 Nov 2013 |
| 38 |
// } |
| 39 |
// if (!defined('MARKDOWNEXTRA_VERSION')) { |
| 40 |
// define( 'MARKDOWNEXTRA_VERSION', "1.2.8" ); # 29 Nov 2013 |
| 41 |
// } |
| 42 |
|
| 43 |
|
| 44 |
# |
| 45 |
# Global default settings: |
| 46 |
# |
| 47 |
|
| 48 |
# Change to ">" for HTML output |
| 49 |
// if (!defined('MARKDOWN_EMPTY_ELEMENT_SUFFIX')) { |
| 50 |
// @define( 'MARKDOWN_EMPTY_ELEMENT_SUFFIX', " />"); |
| 51 |
// } |
| 52 |
|
| 53 |
# Define the width of a tab for code blocks. |
| 54 |
// if (!defined('MARKDOWN_TAB_WIDTH')) { |
| 55 |
// @define( 'MARKDOWN_TAB_WIDTH', 4 ); |
| 56 |
// } |
| 57 |
|
| 58 |
# Optional title attribute for footnote links and backlinks. |
| 59 |
// if (!defined('MARKDOWN_FN_LINK_TITLE')) { |
| 60 |
// @define( 'MARKDOWN_FN_LINK_TITLE', "" ); |
| 61 |
// } |
| 62 |
// if (!defined('MARKDOWN_FN_BACKLINK_TITLE')) { |
| 63 |
// @define( 'MARKDOWN_FN_BACKLINK_TITLE', "" ); |
| 64 |
// } |
| 65 |
|
| 66 |
# Optional class attribute for footnote links and backlinks. |
| 67 |
// if (!defined('MARKDOWN_FN_LINK_CLASS')) { |
| 68 |
// @define( 'MARKDOWN_FN_LINK_CLASS', "" ); |
| 69 |
// } |
| 70 |
// if (!defined('MARKDOWN_FN_BACKLINK_CLASS')) { |
| 71 |
// @define( 'MARKDOWN_FN_BACKLINK_CLASS', "" ); |
| 72 |
// } |
| 73 |
|
| 74 |
# Optional class prefix for fenced code block. |
| 75 |
// if (!defined('MARKDOWN_CODE_CLASS_PREFIX')) { |
| 76 |
// @define( 'MARKDOWN_CODE_CLASS_PREFIX', "" ); |
| 77 |
// } |
| 78 |
|
| 79 |
# Class attribute for code blocks goes on the `code` tag; |
| 80 |
# setting this to true will put attributes on the `pre` tag instead. |
| 81 |
// if (!defined('MARKDOWN_CODE_ATTR_ON_PRE')) { |
| 82 |
// @define( 'MARKDOWN_CODE_ATTR_ON_PRE', false ); |
| 83 |
// } |
| 84 |
|
| 85 |
|
| 86 |
# |
| 87 |
# WordPress settings: |
| 88 |
# |
| 89 |
|
| 90 |
# Change to false to remove Markdown from posts and/or comments. |
| 91 |
// if (!defined('MARKDOWN_WP_POSTS')) { |
| 92 |
// @define( 'MARKDOWN_WP_POSTS', false ); |
| 93 |
// } |
| 94 |
// if (!defined('MARKDOWN_WP_COMMENTS')) { |
| 95 |
// @define( 'MARKDOWN_WP_COMMENTS', false ); |
| 96 |
// } |
| 97 |
|
| 98 |
|
| 99 |
### Standard Function Interface ### |
| 100 |
// if (!defined('MARKDOWN_PARSER_CLASS')) { |
| 101 |
// @define( 'MARKDOWN_PARSER_CLASS', 'MarkdownExtra_Parser' ); |
| 102 |
// } |
| 103 |
|
| 104 |
if ( !function_exists( 'radio_station_markdown' ) ) { |
| 105 |
|
| 106 |
function radio_station_markdown( $text ) { |
| 107 |
# |
| 108 |
# Initialize the parser and return the result of its transform method. |
| 109 |
# |
| 110 |
# Setup static parser variable. |
| 111 |
static $parser; |
| 112 |
if ( !isset( $parser ) ) { |
| 113 |
$parser_class = 'radio_station_markdown_extra_parser'; // MARKDOWN_PARSER_CLASS |
| 114 |
$parser = new $parser_class; |
| 115 |
} |
| 116 |
|
| 117 |
# Transform text using parser. |
| 118 |
return $parser->transform( $text ); |
| 119 |
} |
| 120 |
} |
| 121 |
|
| 122 |
// moved to internal function method utf8_strlen |
| 123 |
/** |
| 124 |
* Returns the length of $text loosely counting the number of UTF-8 characters with regular expression. |
| 125 |
* Used by the Markdown_Parser class when mb_strlen is not available. |
| 126 |
* |
| 127 |
* @since 5.9 |
| 128 |
* |
| 129 |
* @return string Length of the multibyte string |
| 130 |
* |
| 131 |
*/ |
| 132 |
// if (!function_exists('markdown_extra_utf8_strlen')) { |
| 133 |
// function markdown_extra_utf8_strlen( $text ) { |
| 134 |
// return preg_match_all( "/[\\x00-\\xBF]|[\\xC0-\\xFF][\\x80-\\xBF]*/", $text, $m ); |
| 135 |
// } |
| 136 |
// } |
| 137 |
|
| 138 |
# |
| 139 |
# Markdown Parser Class |
| 140 |
# |
| 141 |
|
| 142 |
if ( !class_exists( 'radio_station_markdown_parser' ) ) { |
| 143 |
class radio_station_markdown_parser { |
| 144 |
|
| 145 |
### Configuration Variables ### |
| 146 |
|
| 147 |
# Change to ">" for HTML output. |
| 148 |
public $empty_element_suffix = " />"; // MARKDOWN_EMPTY_ELEMENT_SUFFIX |
| 149 |
public $tab_width = 4; // MARKDOWN_TAB_WIDTH; |
| 150 |
|
| 151 |
# Change to `true` to disallow markup or entities. |
| 152 |
public $no_markup = false; |
| 153 |
public $no_entities = false; |
| 154 |
|
| 155 |
# Predefined urls and titles for reference links and images. |
| 156 |
public $predef_urls = array(); |
| 157 |
public $predef_titles = array(); |
| 158 |
|
| 159 |
|
| 160 |
### Parser Implementation ### |
| 161 |
|
| 162 |
# Regex to match balanced [brackets]. |
| 163 |
# Needed to insert a maximum bracked depth while converting to PHP. |
| 164 |
public $nested_brackets_depth = 6; |
| 165 |
public $nested_brackets_re; |
| 166 |
|
| 167 |
public $nested_url_parenthesis_depth = 4; |
| 168 |
public $nested_url_parenthesis_re; |
| 169 |
|
| 170 |
# Table of hash values for escaped characters: |
| 171 |
public $escape_chars = '\`*_{}[]()>#+-.!'; |
| 172 |
public $escape_chars_re; |
| 173 |
|
| 174 |
# Constructor function. Initialize appropriate member variables. |
| 175 |
function __construct() { |
| 176 |
|
| 177 |
// _initDetab removed due to internal utf8_strlen function |
| 178 |
// $this->_initDetab(); |
| 179 |
$this->prepareItalicsAndBold(); |
| 180 |
|
| 181 |
$this->nested_brackets_re = |
| 182 |
str_repeat( '(?>[^\[\]]+|\[', $this->nested_brackets_depth ) . |
| 183 |
str_repeat( '\])*', $this->nested_brackets_depth ); |
| 184 |
|
| 185 |
$this->nested_url_parenthesis_re = |
| 186 |
str_repeat( '(?>[^()\s]+|\(', $this->nested_url_parenthesis_depth ) . |
| 187 |
str_repeat( '(?>\)))*', $this->nested_url_parenthesis_depth ); |
| 188 |
|
| 189 |
$this->escape_chars_re = '[' . preg_quote( $this->escape_chars ) . ']'; |
| 190 |
|
| 191 |
# Sort document, block, and span gamut in ascendent priority order. |
| 192 |
asort( $this->document_gamut ); |
| 193 |
asort( $this->block_gamut ); |
| 194 |
asort( $this->span_gamut ); |
| 195 |
} |
| 196 |
|
| 197 |
|
| 198 |
# Internal hashes used during transformation. |
| 199 |
public $urls = array(); |
| 200 |
public $titles = array(); |
| 201 |
public $html_hashes = array(); |
| 202 |
|
| 203 |
# Status flag to avoid invalid nesting. |
| 204 |
public $in_anchor = false; |
| 205 |
|
| 206 |
# Called before the transformation process starts to setup parser |
| 207 |
# states. |
| 208 |
function setup() { |
| 209 |
# Clear global hashes. |
| 210 |
$this->urls = $this->predef_urls; |
| 211 |
$this->titles = $this->predef_titles; |
| 212 |
$this->html_hashes = array(); |
| 213 |
$this->in_anchor = false; |
| 214 |
} |
| 215 |
|
| 216 |
# Called after the transformation process to clear any variable |
| 217 |
# which may be taking up memory unnecessarly. |
| 218 |
function teardown() { |
| 219 |
$this->urls = array(); |
| 220 |
$this->titles = array(); |
| 221 |
$this->html_hashes = array(); |
| 222 |
} |
| 223 |
|
| 224 |
# Main function. Performs some preprocessing on the input text |
| 225 |
# and pass it through the document gamut. |
| 226 |
function transform($text) { |
| 227 |
|
| 228 |
$this->setup(); |
| 229 |
|
| 230 |
# Remove UTF-8 BOM and marker character in input, if present. |
| 231 |
$text = preg_replace( '{^\xEF\xBB\xBF|\x1A}', '', $text ); |
| 232 |
|
| 233 |
# Standardize line endings: |
| 234 |
# DOS to Unix and Mac to Unix |
| 235 |
$text = preg_replace( '{\r\n?}', "\n", $text ); |
| 236 |
|
| 237 |
# Make sure $text ends with a couple of newlines: |
| 238 |
$text .= "\n\n"; |
| 239 |
|
| 240 |
# Convert all tabs to spaces. |
| 241 |
$text = $this->detab( $text ); |
| 242 |
|
| 243 |
# Turn block-level HTML blocks into hash entries |
| 244 |
$text = $this->hashHTMLBlocks( $text ); |
| 245 |
|
| 246 |
# Strip any lines consisting only of spaces and tabs. |
| 247 |
# This makes subsequent regexen easier to write, because we can |
| 248 |
# match consecutive blank lines with /\n+/ instead of something |
| 249 |
# contorted like /[ ]*\n+/ . |
| 250 |
$text = preg_replace( '/^[ ]+$/m', '', $text ); |
| 251 |
|
| 252 |
# Run document gamut methods. |
| 253 |
foreach ( $this->document_gamut as $method => $priority ) { |
| 254 |
$text = $this->$method($text); |
| 255 |
} |
| 256 |
|
| 257 |
$this->teardown(); |
| 258 |
|
| 259 |
return $text . "\n"; |
| 260 |
} |
| 261 |
|
| 262 |
public $document_gamut = array( |
| 263 |
# Strip link definitions, store in hashes. |
| 264 |
"stripLinkDefinitions" => 20, |
| 265 |
"runBasicBlockGamut" => 30, |
| 266 |
); |
| 267 |
|
| 268 |
# Strips link definitions from text, stores the URLs and titles in |
| 269 |
# hash references. |
| 270 |
function stripLinkDefinitions( $text ) { |
| 271 |
|
| 272 |
$less_than_tab = $this->tab_width - 1; |
| 273 |
|
| 274 |
# Link defs are in the form: ^[id]: url "optional title" |
| 275 |
$text = preg_replace_callback( '{ |
| 276 |
^[ ]{0,' . $less_than_tab . '}\[(.+)\][ ]?: # id = $1 |
| 277 |
[ ]* |
| 278 |
\n? # maybe *one* newline |
| 279 |
[ ]* |
| 280 |
(?: |
| 281 |
<(.+?)> # url = $2 |
| 282 |
| |
| 283 |
(\S+?) # url = $3 |
| 284 |
) |
| 285 |
[ ]* |
| 286 |
\n? # maybe one newline |
| 287 |
[ ]* |
| 288 |
(?: |
| 289 |
(?<=\s) # lookbehind for whitespace |
| 290 |
["(] |
| 291 |
(.*?) # title = $4 |
| 292 |
[")] |
| 293 |
[ ]* |
| 294 |
)? # title is optional |
| 295 |
(?:\n+|\Z) |
| 296 |
}xm', |
| 297 |
array( &$this, '_stripLinkDefinitions_callback' ), |
| 298 |
$text |
| 299 |
); |
| 300 |
return $text; |
| 301 |
} |
| 302 |
function _stripLinkDefinitions_callback( $matches ) { |
| 303 |
$link_id = strtolower( $matches[1] ); |
| 304 |
$url = $matches[2] == '' ? $matches[3] : $matches[2]; |
| 305 |
$this->urls[$link_id] = $url; |
| 306 |
$this->titles[$link_id] =& $matches[4]; |
| 307 |
return ''; # String that will replace the block |
| 308 |
} |
| 309 |
|
| 310 |
|
| 311 |
function hashHTMLBlocks( $text ) { |
| 312 |
if ( $this->no_markup ) { |
| 313 |
return $text; |
| 314 |
} |
| 315 |
|
| 316 |
$less_than_tab = $this->tab_width - 1; |
| 317 |
|
| 318 |
# Hashify HTML blocks: |
| 319 |
# We only want to do this for block-level HTML tags, such as headers, |
| 320 |
# lists, and tables. That's because we still want to wrap <p>s around |
| 321 |
# "paragraphs" that are wrapped in non-block-level tags, such as anchors, |
| 322 |
# phrase emphasis, and spans. The list of tags we're looking for is |
| 323 |
# hard-coded: |
| 324 |
# |
| 325 |
# * List "a" is made of tags which can be both inline or block-level. |
| 326 |
# These will be treated block-level when the start tag is alone on |
| 327 |
# its line, otherwise they're not matched here and will be taken as |
| 328 |
# inline later. |
| 329 |
# * List "b" is made of tags which are always block-level; |
| 330 |
# |
| 331 |
$block_tags_a_re = 'ins|del'; |
| 332 |
$block_tags_b_re = 'p|div|h[1-6]|blockquote|pre|table|dl|ol|ul|address|'. |
| 333 |
'script|noscript|form|fieldset|iframe|math|svg|'. |
| 334 |
'article|section|nav|aside|hgroup|header|footer|'. |
| 335 |
'figure'; |
| 336 |
|
| 337 |
# Regular expression for the content of a block tag. |
| 338 |
$nested_tags_level = 4; |
| 339 |
$attr = ' |
| 340 |
(?> # optional tag attributes |
| 341 |
\s # starts with whitespace |
| 342 |
(?> |
| 343 |
[^>"/]+ # text outside quotes |
| 344 |
| |
| 345 |
/+(?!>) # slash not followed by ">" |
| 346 |
| |
| 347 |
"[^"]*" # text inside double quotes (tolerate ">") |
| 348 |
| |
| 349 |
\'[^\']*\' # text inside single quotes (tolerate ">") |
| 350 |
)* |
| 351 |
)? |
| 352 |
'; |
| 353 |
$content = |
| 354 |
str_repeat( ' |
| 355 |
(?> |
| 356 |
[^<]+ # content without tag |
| 357 |
| |
| 358 |
<\2 # nested opening tag |
| 359 |
' . $attr . ' # attributes |
| 360 |
(?> |
| 361 |
/> |
| 362 |
| |
| 363 |
>', $nested_tags_level ) . # end of opening tag |
| 364 |
'.*?'. # last level nested tag content |
| 365 |
str_repeat( ' |
| 366 |
</\2\s*> # closing nested tag |
| 367 |
) |
| 368 |
| |
| 369 |
<(?!/\2\s*> # other tags with a different name |
| 370 |
) |
| 371 |
)*', |
| 372 |
$nested_tags_level ); |
| 373 |
$content2 = str_replace( '\2', '\3', $content ); |
| 374 |
|
| 375 |
# First, look for nested blocks, e.g.: |
| 376 |
# <div> |
| 377 |
# <div> |
| 378 |
# tags for inner block must be indented. |
| 379 |
# </div> |
| 380 |
# </div> |
| 381 |
# |
| 382 |
# The outermost tags must start at the left margin for this to match, and |
| 383 |
# the inner nested divs must be indented. |
| 384 |
# We need to do this before the next, more liberal match, because the next |
| 385 |
# match will start at the first `<div>` and stop at the first `</div>`. |
| 386 |
$text = preg_replace_callback( '{(?> |
| 387 |
(?> |
| 388 |
(?<=\n\n) # Starting after a blank line |
| 389 |
| # or |
| 390 |
\A\n? # the beginning of the doc |
| 391 |
) |
| 392 |
( # save in $1 |
| 393 |
|
| 394 |
# Match from `\n<tag>` to `</tag>\n`, handling nested tags |
| 395 |
# in between. |
| 396 |
|
| 397 |
[ ]{0,' . $less_than_tab . '} |
| 398 |
<(' . $block_tags_b_re . ')# start tag = $2 |
| 399 |
' . $attr . '> # attributes followed by > and \n |
| 400 |
' . $content . ' # content, support nesting |
| 401 |
</\2> # the matching end tag |
| 402 |
[ ]* # trailing spaces/tabs |
| 403 |
(?=\n+|\Z) # followed by a newline or end of document |
| 404 |
|
| 405 |
| # Special version for tags of group a. |
| 406 |
|
| 407 |
[ ]{0,' . $less_than_tab . '} |
| 408 |
<(' . $block_tags_a_re . ')# start tag = $3 |
| 409 |
' . $attr . '>[ ]*\n # attributes followed by > |
| 410 |
' . $content2 . ' # content, support nesting |
| 411 |
</\3> # the matching end tag |
| 412 |
[ ]* # trailing spaces/tabs |
| 413 |
(?=\n+|\Z) # followed by a newline or end of document |
| 414 |
|
| 415 |
| # Special case just for <hr />. It was easier to make a special |
| 416 |
# case than to make the other regex more complicated. |
| 417 |
|
| 418 |
[ ]{0,' . $less_than_tab . '} |
| 419 |
<(hr) # start tag = $2 |
| 420 |
' . $attr . ' # attributes |
| 421 |
/?> # the matching end tag |
| 422 |
[ ]* |
| 423 |
(?=\n{2,}|\Z) # followed by a blank line or end of document |
| 424 |
|
| 425 |
| # Special case for standalone HTML comments: |
| 426 |
|
| 427 |
[ ]{0,' . $less_than_tab . '} |
| 428 |
(?s: |
| 429 |
<!-- .*? --> |
| 430 |
) |
| 431 |
[ ]* |
| 432 |
(?=\n{2,}|\Z) # followed by a blank line or end of document |
| 433 |
|
| 434 |
| # PHP and ASP-style processor instructions |
| 435 |
|
| 436 |
[ ]{0,' . $less_than_tab . '} |
| 437 |
(?s: |
| 438 |
<([?%]) # $2 |
| 439 |
.*? |
| 440 |
\2> |
| 441 |
) |
| 442 |
[ ]* |
| 443 |
(?=\n{2,}|\Z) # followed by a blank line or end of document |
| 444 |
|
| 445 |
) |
| 446 |
)}Sxmi', |
| 447 |
array( &$this, '_hashHTMLBlocks_callback' ), |
| 448 |
$text |
| 449 |
); |
| 450 |
|
| 451 |
return $text; |
| 452 |
} |
| 453 |
|
| 454 |
function _hashHTMLBlocks_callback( $matches ) { |
| 455 |
$text = $matches[1]; |
| 456 |
$key = $this->hashBlock( $text ); |
| 457 |
return "\n\n" . $key . "\n\n"; |
| 458 |
} |
| 459 |
|
| 460 |
# Called whenever a tag must be hashed when a function insert an atomic |
| 461 |
# element in the text stream. Passing $text to through this function gives |
| 462 |
# a unique text-token which will be reverted back when calling unhash. |
| 463 |
# |
| 464 |
# The $boundary argument specify what character should be used to surround |
| 465 |
# the token. By convension, "B" is used for block elements that needs not |
| 466 |
# to be wrapped into paragraph tags at the end, ":" is used for elements |
| 467 |
# that are word separators and "X" is used in the general case. |
| 468 |
function hashPart( $text, $boundary = 'X' ) { |
| 469 |
|
| 470 |
# Swap back any tag hash found in $text so we do not have to `unhash` |
| 471 |
# multiple times at the end. |
| 472 |
$text = $this->unhash( $text ); |
| 473 |
|
| 474 |
# Then hash the block. |
| 475 |
static $i = 0; |
| 476 |
$key = $boundary . "\x1A" . ++$i . $boundary; |
| 477 |
$this->html_hashes[$key] = $text; |
| 478 |
return $key; # String that will replace the tag. |
| 479 |
} |
| 480 |
|
| 481 |
# Shortcut function for hashPart with block-level boundaries. |
| 482 |
function hashBlock( $text ) { |
| 483 |
return $this->hashPart( $text, 'B' ); |
| 484 |
} |
| 485 |
|
| 486 |
# These are all the transformations that form block-level |
| 487 |
# tags like paragraphs, headers, and list items. |
| 488 |
public $block_gamut = array( |
| 489 |
"doHeaders" => 10, |
| 490 |
"doHorizontalRules" => 20, |
| 491 |
|
| 492 |
"doLists" => 40, |
| 493 |
"doCodeBlocks" => 50, |
| 494 |
"doBlockQuotes" => 60, |
| 495 |
); |
| 496 |
|
| 497 |
# Run block gamut tranformations. |
| 498 |
function runBlockGamut( $text ) { |
| 499 |
# We need to escape raw HTML in Markdown source before doing anything |
| 500 |
# else. This need to be done for each block, and not only at the |
| 501 |
# beginning in the Markdown function since hashed blocks can be part of |
| 502 |
# list items and could have been indented. Indented blocks would have |
| 503 |
# been seen as a code block in a previous pass of hashHTMLBlocks. |
| 504 |
$text = $this->hashHTMLBlocks( $text ); |
| 505 |
|
| 506 |
return $this->runBasicBlockGamut( $text ); |
| 507 |
} |
| 508 |
|
| 509 |
# Run block gamut tranformations, without hashing HTML blocks. This is |
| 510 |
# useful when HTML blocks are known to be already hashed, like in the first |
| 511 |
# whole-document pass. |
| 512 |
function runBasicBlockGamut( $text ) { |
| 513 |
foreach ( $this->block_gamut as $method => $priority ) { |
| 514 |
$text = $this->$method($text); |
| 515 |
} |
| 516 |
|
| 517 |
# Finally form paragraph and restore hashed blocks. |
| 518 |
$text = $this->formParagraphs( $text ); |
| 519 |
|
| 520 |
return $text; |
| 521 |
} |
| 522 |
|
| 523 |
function doHorizontalRules( $text ) { |
| 524 |
# Do Horizontal Rules: |
| 525 |
return preg_replace( |
| 526 |
'{ |
| 527 |
^[ ]{0,3} # Leading space |
| 528 |
([-*_]) # $1: First marker |
| 529 |
(?> # Repeated marker group |
| 530 |
[ ]{0,2} # Zero, one, or two spaces. |
| 531 |
\1 # Marker character |
| 532 |
){2,} # Group repeated at least twice |
| 533 |
[ ]* # Tailing spaces |
| 534 |
$ # End of line. |
| 535 |
}mx', |
| 536 |
"\n" . $this->hashBlock( "<hr$this->empty_element_suffix" ) . "\n", |
| 537 |
$text |
| 538 |
); |
| 539 |
} |
| 540 |
|
| 541 |
# These are all the transformations that occur *within* block-level |
| 542 |
# tags like paragraphs, headers, and list items. |
| 543 |
public $span_gamut = array( |
| 544 |
# Process character escapes, code spans, and inline HTML |
| 545 |
# in one shot. |
| 546 |
"parseSpan" => -30, |
| 547 |
|
| 548 |
# Process anchor and image tags. Images must come first, |
| 549 |
# because ![foo][f] looks like an anchor. |
| 550 |
"doImages" => 10, |
| 551 |
"doAnchors" => 20, |
| 552 |
|
| 553 |
# Make links out of things like `<http://example.com/>` |
| 554 |
# Must come after doAnchors, because you can use < and > |
| 555 |
# delimiters in inline links like [this](<url>). |
| 556 |
"doAutoLinks" => 30, |
| 557 |
"encodeAmpsAndAngles" => 40, |
| 558 |
|
| 559 |
"doItalicsAndBold" => 50, |
| 560 |
"doHardBreaks" => 60, |
| 561 |
); |
| 562 |
|
| 563 |
# Run span gamut tranformations. |
| 564 |
function runSpanGamut( $text ) { |
| 565 |
|
| 566 |
foreach ( $this->span_gamut as $method => $priority ) { |
| 567 |
$text = $this->$method($text); |
| 568 |
} |
| 569 |
|
| 570 |
return $text; |
| 571 |
} |
| 572 |
|
| 573 |
# Do hard breaks: |
| 574 |
function doHardBreaks( $text ) { |
| 575 |
return preg_replace_callback( |
| 576 |
'/ {2,}\n/', |
| 577 |
array( &$this, '_doHardBreaks_callback' ), |
| 578 |
$text |
| 579 |
); |
| 580 |
} |
| 581 |
|
| 582 |
function _doHardBreaks_callback( $matches ) { |
| 583 |
return $this->hashPart( "<br$this->empty_element_suffix\n" ); |
| 584 |
} |
| 585 |
|
| 586 |
# Turn Markdown link shortcuts into XHTML <a> tags. |
| 587 |
function doAnchors( $text ) { |
| 588 |
|
| 589 |
if ( $this->in_anchor ) { |
| 590 |
return $text; |
| 591 |
} |
| 592 |
$this->in_anchor = true; |
| 593 |
|
| 594 |
# First, handle reference-style links: [link text] [id] |
| 595 |
$text = preg_replace_callback( '{ |
| 596 |
( # wrap whole match in $1 |
| 597 |
\[ |
| 598 |
('.$this->nested_brackets_re.') # link text = $2 |
| 599 |
\] |
| 600 |
|
| 601 |
[ ]? # one optional space |
| 602 |
(?:\n[ ]*)? # one optional newline followed by spaces |
| 603 |
|
| 604 |
\[ |
| 605 |
(.*?) # id = $3 |
| 606 |
\] |
| 607 |
) |
| 608 |
}xs', |
| 609 |
array( &$this, '_doAnchors_reference_callback' ), |
| 610 |
$text |
| 611 |
); |
| 612 |
|
| 613 |
# Next, inline-style links: [link text](url "optional title") |
| 614 |
$text = preg_replace_callback( '{ |
| 615 |
( # wrap whole match in $1 |
| 616 |
\[ |
| 617 |
(' . $this->nested_brackets_re . ') # link text = $2 |
| 618 |
\] |
| 619 |
\( # literal paren |
| 620 |
[ \n]* |
| 621 |
(?: |
| 622 |
<(.+?)> # href = $3 |
| 623 |
| |
| 624 |
(' . $this->nested_url_parenthesis_re . ') # href = $4 |
| 625 |
) |
| 626 |
[ \n]* |
| 627 |
( # $5 |
| 628 |
([\'"]) # quote char = $6 |
| 629 |
(.*?) # Title = $7 |
| 630 |
\6 # matching quote |
| 631 |
[ \n]* # ignore any spaces/tabs between closing quote and ) |
| 632 |
)? # title is optional |
| 633 |
\) |
| 634 |
) |
| 635 |
}xs', |
| 636 |
array( &$this, '_doAnchors_inline_callback' ), |
| 637 |
$text |
| 638 |
); |
| 639 |
|
| 640 |
# Last, handle reference-style shortcuts: [link text] |
| 641 |
# These must come last in case you've also got [link text][1] |
| 642 |
# or [link text](/foo) |
| 643 |
$text = preg_replace_callback( '{ |
| 644 |
( # wrap whole match in $1 |
| 645 |
\[ |
| 646 |
([^\[\]]+) # link text = $2; can\'t contain [ or ] |
| 647 |
\] |
| 648 |
) |
| 649 |
}xs', |
| 650 |
array( &$this, '_doAnchors_reference_callback' ), |
| 651 |
$text |
| 652 |
); |
| 653 |
|
| 654 |
$this->in_anchor = false; |
| 655 |
return $text; |
| 656 |
} |
| 657 |
|
| 658 |
function _doAnchors_reference_callback( $matches ) { |
| 659 |
|
| 660 |
$whole_match = $matches[1]; |
| 661 |
$link_text = $matches[2]; |
| 662 |
$link_id =& $matches[3]; |
| 663 |
|
| 664 |
if ( "" == $link_id ) { |
| 665 |
# for shortcut links like [this][] or [this]. |
| 666 |
$link_id = $link_text; |
| 667 |
} |
| 668 |
|
| 669 |
# lower-case and turn embedded newlines into spaces |
| 670 |
$link_id = strtolower( $link_id ); |
| 671 |
$link_id = preg_replace( '{[ ]?\n}', ' ', $link_id ); |
| 672 |
|
| 673 |
if ( isset( $this->urls[$link_id] ) ) { |
| 674 |
$url = $this->urls[$link_id]; |
| 675 |
$url = $this->encodeAttribute( $url ); |
| 676 |
|
| 677 |
$result = "<a href=\"" . $url . "\""; |
| 678 |
if ( isset( $this->titles[$link_id] ) ) { |
| 679 |
$title = $this->titles[$link_id]; |
| 680 |
$title = $this->encodeAttribute( $title ); |
| 681 |
$result .= " title=\"" . $title . "\""; |
| 682 |
} |
| 683 |
|
| 684 |
$link_text = $this->runSpanGamut( $link_text ); |
| 685 |
$result .= ">" . $link_text . "</a>"; |
| 686 |
$result = $this->hashPart( $result ); |
| 687 |
} |
| 688 |
else { |
| 689 |
$result = $whole_match; |
| 690 |
} |
| 691 |
return $result; |
| 692 |
} |
| 693 |
|
| 694 |
function _doAnchors_inline_callback( $matches ) { |
| 695 |
$whole_match = $matches[1]; |
| 696 |
$link_text = $this->runSpanGamut( $matches[2] ); |
| 697 |
$url = $matches[3] == '' ? $matches[4] : $matches[3]; |
| 698 |
$title =& $matches[7]; |
| 699 |
|
| 700 |
$url = $this->encodeAttribute( $url ); |
| 701 |
|
| 702 |
$result = "<a href=\"" . $url . "\""; |
| 703 |
if ( isset( $title ) ) { |
| 704 |
$title = $this->encodeAttribute( $title ); |
| 705 |
$result .= " title=\"$title\""; |
| 706 |
} |
| 707 |
|
| 708 |
$link_text = $this->runSpanGamut( $link_text ); |
| 709 |
$result .= ">" . $link_text . "</a>"; |
| 710 |
|
| 711 |
return $this->hashPart( $result ); |
| 712 |
} |
| 713 |
|
| 714 |
# Turn Markdown image shortcuts into <img> tags. |
| 715 |
function doImages( $text ) { |
| 716 |
|
| 717 |
# First, handle reference-style labeled images: ![alt text][id] |
| 718 |
$text = preg_replace_callback( '{ |
| 719 |
( # wrap whole match in $1 |
| 720 |
!\[ |
| 721 |
(' . $this->nested_brackets_re . ') # alt text = $2 |
| 722 |
\] |
| 723 |
|
| 724 |
[ ]? # one optional space |
| 725 |
(?:\n[ ]*)? # one optional newline followed by spaces |
| 726 |
|
| 727 |
\[ |
| 728 |
(.*?) # id = $3 |
| 729 |
\] |
| 730 |
|
| 731 |
) |
| 732 |
}xs', |
| 733 |
array( &$this, '_doImages_reference_callback' ), |
| 734 |
$text |
| 735 |
); |
| 736 |
|
| 737 |
# Next, handle inline images:  |
| 738 |
# Don't forget: encode * and _ |
| 739 |
$text = preg_replace_callback( '{ |
| 740 |
( # wrap whole match in $1 |
| 741 |
!\[ |
| 742 |
(' . $this->nested_brackets_re . ') # alt text = $2 |
| 743 |
\] |
| 744 |
\s? # One optional whitespace character |
| 745 |
\( # literal paren |
| 746 |
[ \n]* |
| 747 |
(?: |
| 748 |
<(\S*)> # src url = $3 |
| 749 |
| |
| 750 |
(' . $this->nested_url_parenthesis_re . ') # src url = $4 |
| 751 |
) |
| 752 |
[ \n]* |
| 753 |
( # $5 |
| 754 |
([\'"]) # quote char = $6 |
| 755 |
(.*?) # title = $7 |
| 756 |
\6 # matching quote |
| 757 |
[ \n]* |
| 758 |
)? # title is optional |
| 759 |
\) |
| 760 |
) |
| 761 |
}xs', |
| 762 |
array( &$this, '_doImages_inline_callback' ), |
| 763 |
$text |
| 764 |
); |
| 765 |
|
| 766 |
return $text; |
| 767 |
} |
| 768 |
|
| 769 |
function _doImages_reference_callback( $matches ) { |
| 770 |
|
| 771 |
$whole_match = $matches[1]; |
| 772 |
$alt_text = $matches[2]; |
| 773 |
$link_id = strtolower($matches[3]); |
| 774 |
|
| 775 |
if ( "" == $link_id ) { |
| 776 |
$link_id = strtolower( $alt_text ); # for shortcut links like ![this][]. |
| 777 |
} |
| 778 |
|
| 779 |
$alt_text = $this->encodeAttribute( $alt_text ); |
| 780 |
if ( isset( $this->urls[$link_id] ) ) { |
| 781 |
$url = $this->encodeAttribute( $this->urls[$link_id] ); |
| 782 |
$result = "<img src=\"" . $url . "\" alt=\"" . $alt_text . "\""; |
| 783 |
if ( isset( $this->titles[$link_id] ) ) { |
| 784 |
$title = $this->titles[$link_id]; |
| 785 |
$title = $this->encodeAttribute( $title ); |
| 786 |
$result .= " title=\"" . $title . "\""; |
| 787 |
} |
| 788 |
$result .= $this->empty_element_suffix; |
| 789 |
$result = $this->hashPart( $result ); |
| 790 |
} else { |
| 791 |
# If there's no such link ID, leave intact: |
| 792 |
$result = $whole_match; |
| 793 |
} |
| 794 |
|
| 795 |
return $result; |
| 796 |
} |
| 797 |
|
| 798 |
function _doImages_inline_callback( $matches ) { |
| 799 |
$whole_match = $matches[1]; |
| 800 |
$alt_text = $matches[2]; |
| 801 |
$url = $matches[3] == '' ? $matches[4] : $matches[3]; |
| 802 |
$title =& $matches[7]; |
| 803 |
|
| 804 |
$alt_text = $this->encodeAttribute( $alt_text ); |
| 805 |
$url = $this->encodeAttribute( $url ); |
| 806 |
$result = "<img src=\"" . $url . "\" alt=\"" . $alt_text . "\""; |
| 807 |
if ( isset( $title ) ) { |
| 808 |
$title = $this->encodeAttribute( $title ); |
| 809 |
$result .= " title=\"" . $title . "\""; # $title already quoted |
| 810 |
} |
| 811 |
$result .= $this->empty_element_suffix; |
| 812 |
|
| 813 |
return $this->hashPart( $result ); |
| 814 |
} |
| 815 |
|
| 816 |
function doHeaders( $text ) { |
| 817 |
|
| 818 |
# Setext-style headers: |
| 819 |
# Header 1 |
| 820 |
# ======== |
| 821 |
# |
| 822 |
# Header 2 |
| 823 |
# -------- |
| 824 |
$text = preg_replace_callback( |
| 825 |
'{ ^(.+?)[ ]*\n(=+|-+)[ ]*\n+ }mx', |
| 826 |
array( &$this, '_doHeaders_callback_setext' ), |
| 827 |
$text |
| 828 |
); |
| 829 |
|
| 830 |
# atx-style headers: |
| 831 |
# # Header 1 |
| 832 |
# ## Header 2 |
| 833 |
# ## Header 2 with closing hashes ## |
| 834 |
# ... |
| 835 |
# ###### Header 6 |
| 836 |
# |
| 837 |
$text = preg_replace_callback( '{ |
| 838 |
^(\#{1,6}) # $1 = string of #\'s |
| 839 |
[ ]* |
| 840 |
(.+?) # $2 = Header text |
| 841 |
[ ]* |
| 842 |
\#* # optional closing #\'s (not counted) |
| 843 |
\n+ |
| 844 |
}xm', |
| 845 |
array( &$this, '_doHeaders_callback_atx' ), |
| 846 |
$text |
| 847 |
); |
| 848 |
|
| 849 |
return $text; |
| 850 |
} |
| 851 |
|
| 852 |
function _doHeaders_callback_setext( $matches ) { |
| 853 |
# Terrible hack to check we haven't found an empty list item. |
| 854 |
if ( $matches[2] == '-' && preg_match( '{^-(?: |$)}', $matches[1] ) ) |
| 855 |
return $matches[0]; |
| 856 |
|
| 857 |
$level = $matches[2][0] == '=' ? 1 : 2; |
| 858 |
$block = "<h" . $level . ">" . $this->runSpanGamut( $matches[1] ) . "</h" . $level . ">"; |
| 859 |
return "\n" . $this->hashBlock( $block ) . "\n\n"; |
| 860 |
} |
| 861 |
|
| 862 |
function _doHeaders_callback_atx( $matches ) { |
| 863 |
$level = strlen( $matches[1] ); |
| 864 |
$block = "<h" . $level . ">" . $this->runSpanGamut( $matches[2] ) . "</h" . $level . ">"; |
| 865 |
return "\n" . $this->hashBlock( $block ) . "\n\n"; |
| 866 |
} |
| 867 |
|
| 868 |
# Form HTML ordered (numbered) and unordered (bulleted) lists. |
| 869 |
function doLists( $text ) { |
| 870 |
|
| 871 |
$less_than_tab = $this->tab_width - 1; |
| 872 |
|
| 873 |
# Re-usable patterns to match list item bullets and number markers: |
| 874 |
$marker_ul_re = '[*+-]'; |
| 875 |
$marker_ol_re = '\d+[\.]'; |
| 876 |
$marker_any_re = "(?:$marker_ul_re|$marker_ol_re)"; |
| 877 |
|
| 878 |
$markers_relist = array( |
| 879 |
$marker_ul_re => $marker_ol_re, |
| 880 |
$marker_ol_re => $marker_ul_re, |
| 881 |
); |
| 882 |
|
| 883 |
foreach ( $markers_relist as $marker_re => $other_marker_re ) { |
| 884 |
# Re-usable pattern to match any entirel ul or ol list: |
| 885 |
$whole_list_re = ' |
| 886 |
( # $1 = whole list |
| 887 |
( # $2 |
| 888 |
([ ]{0,' . $less_than_tab . '}) # $3 = number of spaces |
| 889 |
(' . $marker_re . ') # $4 = first list item marker |
| 890 |
[ ]+ |
| 891 |
) |
| 892 |
(?s:.+?) |
| 893 |
( # $5 |
| 894 |
\z |
| 895 |
| |
| 896 |
\n{2,} |
| 897 |
(?=\S) |
| 898 |
(?! # Negative lookahead for another list item marker |
| 899 |
[ ]* |
| 900 |
' . $marker_re . '[ ]+ |
| 901 |
) |
| 902 |
| |
| 903 |
(?= # Lookahead for another kind of list |
| 904 |
\n |
| 905 |
\3 # Must have the same indentation |
| 906 |
' . $other_marker_re . '[ ]+ |
| 907 |
) |
| 908 |
) |
| 909 |
) |
| 910 |
'; // mx |
| 911 |
|
| 912 |
# We use a different prefix before nested lists than top-level lists. |
| 913 |
# See extended comment in _ProcessListItems(). |
| 914 |
if ( $this->list_level ) { |
| 915 |
$text = preg_replace_callback( '{ |
| 916 |
^ |
| 917 |
' . $whole_list_re . ' |
| 918 |
}mx', |
| 919 |
array( &$this, '_doLists_callback' ), |
| 920 |
$text |
| 921 |
); |
| 922 |
} else { |
| 923 |
$text = preg_replace_callback( '{ |
| 924 |
(?:(?<=\n)\n|\A\n?) # Must eat the newline |
| 925 |
'.$whole_list_re.' |
| 926 |
}mx', |
| 927 |
array( &$this, '_doLists_callback' ), |
| 928 |
$text |
| 929 |
); |
| 930 |
} |
| 931 |
} |
| 932 |
|
| 933 |
return $text; |
| 934 |
} |
| 935 |
|
| 936 |
function _doLists_callback( $matches ) { |
| 937 |
# Re-usable patterns to match list item bullets and number markers: |
| 938 |
$marker_ul_re = '[*+-]'; |
| 939 |
$marker_ol_re = '\d+[\.]'; |
| 940 |
$marker_any_re = "(?:$marker_ul_re|$marker_ol_re)"; |
| 941 |
|
| 942 |
$list = $matches[1]; |
| 943 |
$list_type = preg_match( "/$marker_ul_re/", $matches[4]) ? "ul" : "ol"; |
| 944 |
|
| 945 |
$marker_any_re = ( $list_type == "ul" ? $marker_ul_re : $marker_ol_re ); |
| 946 |
|
| 947 |
$list .= "\n"; |
| 948 |
$result = $this->processListItems( $list, $marker_any_re ); |
| 949 |
|
| 950 |
$result = $this->hashBlock("<" . $list_type . ">\n" . $result . "</" . $list_type . ">"); |
| 951 |
return "\n" . $result . "\n\n"; |
| 952 |
} |
| 953 |
|
| 954 |
public $list_level = 0; |
| 955 |
|
| 956 |
# Process the contents of a single ordered or unordered list, splitting it |
| 957 |
# into individual list items. |
| 958 |
function processListItems( $list_str, $marker_any_re ) { |
| 959 |
|
| 960 |
# The $this->list_level global keeps track of when we're inside a list. |
| 961 |
# Each time we enter a list, we increment it; when we leave a list, |
| 962 |
# we decrement. If it's zero, we're not in a list anymore. |
| 963 |
# |
| 964 |
# We do this because when we're not inside a list, we want to treat |
| 965 |
# something like this: |
| 966 |
# |
| 967 |
# I recommend upgrading to version |
| 968 |
# 8. Oops, now this line is treated |
| 969 |
# as a sub-list. |
| 970 |
# |
| 971 |
# As a single paragraph, despite the fact that the second line starts |
| 972 |
# with a digit-period-space sequence. |
| 973 |
# |
| 974 |
# Whereas when we're inside a list (or sub-list), that line will be |
| 975 |
# treated as the start of a sub-list. What a kludge, huh? This is |
| 976 |
# an aspect of Markdown's syntax that's hard to parse perfectly |
| 977 |
# without resorting to mind-reading. Perhaps the solution is to |
| 978 |
# change the syntax rules such that sub-lists must start with a |
| 979 |
# starting cardinal number; e.g. "1." or "a.". |
| 980 |
|
| 981 |
$this->list_level++; |
| 982 |
|
| 983 |
# trim trailing blank lines: |
| 984 |
$list_str = preg_replace( "/\n{2,}\\z/", "\n", $list_str ); |
| 985 |
|
| 986 |
$list_str = preg_replace_callback('{ |
| 987 |
(\n)? # leading line = $1 |
| 988 |
(^[ ]*) # leading whitespace = $2 |
| 989 |
(' . $marker_any_re . ' # list marker and space = $3 |
| 990 |
(?:[ ]+|(?=\n)) # space only required if item is not empty |
| 991 |
) |
| 992 |
((?s:.*?)) # list item text = $4 |
| 993 |
(?:(\n+(?=\n))|\n) # tailing blank line = $5 |
| 994 |
(?= \n* (\z | \2 (' . $marker_any_re . ') (?:[ ]+|(?=\n)))) |
| 995 |
}xm', |
| 996 |
array( &$this, '_processListItems_callback'), $list_str ); |
| 997 |
|
| 998 |
$this->list_level--; |
| 999 |
return $list_str; |
| 1000 |
} |
| 1001 |
|
| 1002 |
function _processListItems_callback( $matches ) { |
| 1003 |
$item = $matches[4]; |
| 1004 |
$leading_line =& $matches[1]; |
| 1005 |
$leading_space =& $matches[2]; |
| 1006 |
$marker_space = $matches[3]; |
| 1007 |
$tailing_blank_line =& $matches[5]; |
| 1008 |
|
| 1009 |
if ( $leading_line || $tailing_blank_line || preg_match( '/\n{2,}/', $item ) ) { |
| 1010 |
# Replace marker with the appropriate whitespace indentation |
| 1011 |
$item = $leading_space . str_repeat( ' ', strlen( $marker_space ) ) . $item; |
| 1012 |
$item = $this->runBlockGamut( $this->outdent( $item ) . "\n"); |
| 1013 |
} else { |
| 1014 |
# Recursion for sub-lists: |
| 1015 |
$item = $this->doLists( $this->outdent( $item ) ); |
| 1016 |
$item = preg_replace( '/\n+$/', '', $item ); |
| 1017 |
$item = $this->runSpanGamut( $item ); |
| 1018 |
} |
| 1019 |
|
| 1020 |
return "<li>" . $item . "</li>\n"; |
| 1021 |
} |
| 1022 |
|
| 1023 |
# Process Markdown `<pre><code>` blocks. |
| 1024 |
function doCodeBlocks( $text ) { |
| 1025 |
|
| 1026 |
$text = preg_replace_callback( '{ |
| 1027 |
(?:\n\n|\A\n?) |
| 1028 |
( # $1 = the code block -- one or more lines, starting with a space/tab |
| 1029 |
(?> |
| 1030 |
[ ]{'.$this->tab_width.'} # Lines must start with a tab or a tab-width of spaces |
| 1031 |
.*\n+ |
| 1032 |
)+ |
| 1033 |
) |
| 1034 |
((?=^[ ]{0,'.$this->tab_width.'}\S)|\Z) # Lookahead for non-space at line-start, or end of doc |
| 1035 |
}xm', |
| 1036 |
array( &$this, '_doCodeBlocks_callback'), |
| 1037 |
$text |
| 1038 |
); |
| 1039 |
|
| 1040 |
return $text; |
| 1041 |
} |
| 1042 |
|
| 1043 |
function _doCodeBlocks_callback( $matches ) { |
| 1044 |
$codeblock = $matches[1]; |
| 1045 |
|
| 1046 |
$codeblock = $this->outdent( $codeblock ); |
| 1047 |
$codeblock = htmlspecialchars( $codeblock, ENT_NOQUOTES ); |
| 1048 |
|
| 1049 |
# trim leading newlines and trailing newlines |
| 1050 |
$codeblock = preg_replace( '/\A\n+|\n+\z/', '', $codeblock ); |
| 1051 |
|
| 1052 |
$codeblock = "<pre><code>" . $codeblock . "\n</code></pre>"; |
| 1053 |
return "\n\n" . $this->hashBlock( $codeblock ) . "\n\n"; |
| 1054 |
} |
| 1055 |
|
| 1056 |
function makeCodeSpan( $code ) { |
| 1057 |
# |
| 1058 |
# Create a code span markup for $code. Called from handleSpanToken. |
| 1059 |
# |
| 1060 |
$code = htmlspecialchars( trim( $code ), ENT_NOQUOTES ); |
| 1061 |
return $this->hashPart( "<code>" . $code . "</code>" ); |
| 1062 |
} |
| 1063 |
|
| 1064 |
public $em_relist = array( |
| 1065 |
'' => '(?:(?<!\*)\*(?!\*)|(?<!_)_(?!_))(?=\S|$)(?![\.,:;]\s)', |
| 1066 |
'*' => '(?<=\S|^)(?<!\*)\*(?!\*)', |
| 1067 |
'_' => '(?<=\S|^)(?<!_)_(?!_)', |
| 1068 |
); |
| 1069 |
public $strong_relist = array( |
| 1070 |
'' => '(?:(?<!\*)\*\*(?!\*)|(?<!_)__(?!_))(?=\S|$)(?![\.,:;]\s)', |
| 1071 |
'**' => '(?<=\S|^)(?<!\*)\*\*(?!\*)', |
| 1072 |
'__' => '(?<=\S|^)(?<!_)__(?!_)', |
| 1073 |
); |
| 1074 |
public $em_strong_relist = array( |
| 1075 |
'' => '(?:(?<!\*)\*\*\*(?!\*)|(?<!_)___(?!_))(?=\S|$)(?![\.,:;]\s)', |
| 1076 |
'***' => '(?<=\S|^)(?<!\*)\*\*\*(?!\*)', |
| 1077 |
'___' => '(?<=\S|^)(?<!_)___(?!_)', |
| 1078 |
); |
| 1079 |
public $em_strong_prepared_relist; |
| 1080 |
|
| 1081 |
# Prepare regular expressions for searching emphasis tokens in any |
| 1082 |
# context. |
| 1083 |
function prepareItalicsAndBold() { |
| 1084 |
|
| 1085 |
foreach ( $this->em_relist as $em => $em_re ) { |
| 1086 |
foreach ( $this->strong_relist as $strong => $strong_re ) { |
| 1087 |
# Construct list of allowed token expressions. |
| 1088 |
$token_relist = array(); |
| 1089 |
if ( isset( $this->em_strong_relist["$em$strong"] ) ) { |
| 1090 |
$token_relist[] = $this->em_strong_relist["$em$strong"]; |
| 1091 |
} |
| 1092 |
$token_relist[] = $em_re; |
| 1093 |
$token_relist[] = $strong_re; |
| 1094 |
|
| 1095 |
# Construct master expression from list. |
| 1096 |
$token_re = '{(' . implode( '|', $token_relist ) . ')}'; |
| 1097 |
$this->em_strong_prepared_relist["$em$strong"] = $token_re; |
| 1098 |
} |
| 1099 |
} |
| 1100 |
} |
| 1101 |
|
| 1102 |
function doItalicsAndBold( $text ) { |
| 1103 |
$token_stack = array( '' ); |
| 1104 |
$text_stack = array( '' ); |
| 1105 |
$em = ''; |
| 1106 |
$strong = ''; |
| 1107 |
$tree_char_em = false; |
| 1108 |
|
| 1109 |
while ( 1 ) { |
| 1110 |
|
| 1111 |
# Get prepared regular expression for seraching emphasis tokens |
| 1112 |
# in current context. |
| 1113 |
$token_re = $this->em_strong_prepared_relist["$em$strong"]; |
| 1114 |
|
| 1115 |
# Each loop iteration search for the next emphasis token. |
| 1116 |
# Each token is then passed to handleSpanToken. |
| 1117 |
$parts = preg_split( $token_re, $text, 2, PREG_SPLIT_DELIM_CAPTURE ); |
| 1118 |
$text_stack[0] .= $parts[0]; |
| 1119 |
$token =& $parts[1]; |
| 1120 |
$text =& $parts[2]; |
| 1121 |
|
| 1122 |
if ( empty( $token ) ) { |
| 1123 |
# Reached end of text span: empty stack without emitting. |
| 1124 |
# any more emphasis. |
| 1125 |
while ( $token_stack[0] ) { |
| 1126 |
$text_stack[1] .= array_shift( $token_stack ); |
| 1127 |
$text_stack[0] .= array_shift( $text_stack ); |
| 1128 |
} |
| 1129 |
break; |
| 1130 |
} |
| 1131 |
|
| 1132 |
$token_len = strlen( $token ); |
| 1133 |
if ( $tree_char_em ) { |
| 1134 |
# Reached closing marker while inside a three-char emphasis. |
| 1135 |
if ( $token_len == 3 ) { |
| 1136 |
# Three-char closing marker, close em and strong. |
| 1137 |
array_shift( $token_stack ); |
| 1138 |
$span = array_shift( $text_stack ); |
| 1139 |
$span = $this->runSpanGamut( $span ); |
| 1140 |
$span = "<strong><em>" . $span . "</em></strong>"; |
| 1141 |
$text_stack[0] .= $this->hashPart( $span ); |
| 1142 |
$em = ''; |
| 1143 |
$strong = ''; |
| 1144 |
} else { |
| 1145 |
# Other closing marker: close one em or strong and |
| 1146 |
# change current token state to match the other |
| 1147 |
$token_stack[0] = str_repeat( $token[0], 3-$token_len ); |
| 1148 |
$tag = $token_len == 2 ? "strong" : "em"; |
| 1149 |
$span = $text_stack[0]; |
| 1150 |
$span = $this->runSpanGamut( $span ); |
| 1151 |
$span = "<" . $tag . ">" . $span . "</" . $tag . ">"; |
| 1152 |
$text_stack[0] = $this->hashPart( $span ); |
| 1153 |
$$tag = ''; # $$tag stands for $em or $strong |
| 1154 |
} |
| 1155 |
$tree_char_em = false; |
| 1156 |
} else if ( $token_len == 3 ) { |
| 1157 |
if ( $em ) { |
| 1158 |
# Reached closing marker for both em and strong. |
| 1159 |
# Closing strong marker: |
| 1160 |
for ( $i = 0; $i < 2; ++$i ) { |
| 1161 |
$shifted_token = array_shift( $token_stack ); |
| 1162 |
$tag = strlen( $shifted_token ) == 2 ? "strong" : "em"; |
| 1163 |
$span = array_shift( $text_stack ); |
| 1164 |
$span = $this->runSpanGamut( $span ); |
| 1165 |
$span = "<" . $tag . ">" . $span . "</" . $tag . ">"; |
| 1166 |
$text_stack[0] .= $this->hashPart( $span ); |
| 1167 |
$$tag = ''; # $$tag stands for $em or $strong |
| 1168 |
} |
| 1169 |
} else { |
| 1170 |
# Reached opening three-char emphasis marker. Push on token |
| 1171 |
# stack; will be handled by the special condition above. |
| 1172 |
$em = $token[0]; |
| 1173 |
$strong = "$em$em"; |
| 1174 |
array_unshift( $token_stack, $token ); |
| 1175 |
array_unshift( $text_stack, '' ); |
| 1176 |
$tree_char_em = true; |
| 1177 |
} |
| 1178 |
} else if ( $token_len == 2 ) { |
| 1179 |
if ( $strong ) { |
| 1180 |
# Unwind any dangling emphasis marker: |
| 1181 |
if ( strlen( $token_stack[0] ) == 1 ) { |
| 1182 |
$text_stack[1] .= array_shift( $token_stack ); |
| 1183 |
$text_stack[0] .= array_shift( $text_stack ); |
| 1184 |
} |
| 1185 |
# Closing strong marker: |
| 1186 |
array_shift( $token_stack ); |
| 1187 |
$span = array_shift( $text_stack ); |
| 1188 |
$span = $this->runSpanGamut( $span ); |
| 1189 |
$span = "<strong>" . $span . "</strong>"; |
| 1190 |
$text_stack[0] .= $this->hashPart( $span ); |
| 1191 |
$strong = ''; |
| 1192 |
} else { |
| 1193 |
array_unshift( $token_stack, $token ); |
| 1194 |
array_unshift( $text_stack, '' ); |
| 1195 |
$strong = $token; |
| 1196 |
} |
| 1197 |
} else { |
| 1198 |
# Here $token_len == 1 |
| 1199 |
if ( $em ) { |
| 1200 |
if ( strlen( $token_stack[0] ) == 1 ) { |
| 1201 |
# Closing emphasis marker: |
| 1202 |
array_shift( $token_stack ); |
| 1203 |
$span = array_shift( $text_stack ); |
| 1204 |
$span = $this->runSpanGamut( $span ); |
| 1205 |
$span = "<em>" . $span . "</em>"; |
| 1206 |
$text_stack[0] .= $this->hashPart( $span ); |
| 1207 |
$em = ''; |
| 1208 |
} else { |
| 1209 |
$text_stack[0] .= $token; |
| 1210 |
} |
| 1211 |
} else { |
| 1212 |
array_unshift( $token_stack, $token ); |
| 1213 |
array_unshift( $text_stack, '' ); |
| 1214 |
$em = $token; |
| 1215 |
} |
| 1216 |
} |
| 1217 |
} |
| 1218 |
return $text_stack[0]; |
| 1219 |
} |
| 1220 |
|
| 1221 |
function doBlockQuotes( $text ) { |
| 1222 |
$text = preg_replace_callback( '/ |
| 1223 |
( # Wrap whole match in $1 |
| 1224 |
(?> |
| 1225 |
^[ ]*>[ ]? # ">" at the start of a line |
| 1226 |
.+\n # rest of the first line |
| 1227 |
(.+\n)* # subsequent consecutive lines |
| 1228 |
\n* # blanks |
| 1229 |
)+ |
| 1230 |
) |
| 1231 |
/xm', |
| 1232 |
array( &$this, '_doBlockQuotes_callback'), |
| 1233 |
$text |
| 1234 |
); |
| 1235 |
|
| 1236 |
return $text; |
| 1237 |
} |
| 1238 |
|
| 1239 |
function _doBlockQuotes_callback( $matches ) { |
| 1240 |
$bq = $matches[1]; |
| 1241 |
# trim one level of quoting - trim whitespace-only lines |
| 1242 |
$bq = preg_replace( '/^[ ]*>[ ]?|^[ ]+$/m', '', $bq ); |
| 1243 |
$bq = $this->runBlockGamut($bq ); # recurse |
| 1244 |
|
| 1245 |
$bq = preg_replace( '/^/m', " ", $bq ); |
| 1246 |
# These leading spaces cause problem with <pre> content, |
| 1247 |
# so we need to fix that: |
| 1248 |
$bq = preg_replace_callback( '{(\s*<pre>.+?</pre>)}sx', |
| 1249 |
array( &$this, '_doBlockQuotes_callback2' ), |
| 1250 |
$bq |
| 1251 |
); |
| 1252 |
|
| 1253 |
return "\n" . $this->hashBlock( "<blockquote>\n" . $bq . "\n</blockquote>" ) . "\n\n"; |
| 1254 |
} |
| 1255 |
|
| 1256 |
function _doBlockQuotes_callback2( $matches ) { |
| 1257 |
$pre = $matches[1]; |
| 1258 |
$pre = preg_replace( '/^ /m', '', $pre ); |
| 1259 |
return $pre; |
| 1260 |
} |
| 1261 |
|
| 1262 |
function formParagraphs( $text ) { |
| 1263 |
# |
| 1264 |
# Params: |
| 1265 |
# $text - string to process with html <p> tags |
| 1266 |
# |
| 1267 |
# Strip leading and trailing lines: |
| 1268 |
$text = preg_replace( '/\A\n+|\n+\z/', '', $text ); |
| 1269 |
|
| 1270 |
$grafs = preg_split( '/\n{2,}/', $text, -1, PREG_SPLIT_NO_EMPTY ); |
| 1271 |
|
| 1272 |
# |
| 1273 |
# Wrap <p> tags and unhashify HTML blocks |
| 1274 |
# |
| 1275 |
foreach ( $grafs as $key => $value ) { |
| 1276 |
if ( !preg_match( '/^B\x1A[0-9]+B$/', $value ) ) { |
| 1277 |
# Is a paragraph. |
| 1278 |
$value = $this->runSpanGamut( $value ); |
| 1279 |
$value = preg_replace( '/^([ ]*)/', "<p>", $value ); |
| 1280 |
$value .= "</p>"; |
| 1281 |
$grafs[$key] = $this->unhash( $value ); |
| 1282 |
} else { |
| 1283 |
# Is a block. |
| 1284 |
# Modify elements of @grafs in-place... |
| 1285 |
$graf = $value; |
| 1286 |
$block = $this->html_hashes[$graf]; |
| 1287 |
$graf = $block; |
| 1288 |
// if (preg_match('{ |
| 1289 |
// \A |
| 1290 |
// ( # $1 = <div> tag |
| 1291 |
// <div \s+ |
| 1292 |
// [^>]* |
| 1293 |
// \b |
| 1294 |
// markdown\s*=\s* ([\'"]) # $2 = attr quote char |
| 1295 |
// 1 |
| 1296 |
// \2 |
| 1297 |
// [^>]* |
| 1298 |
// > |
| 1299 |
// ) |
| 1300 |
// ( # $3 = contents |
| 1301 |
// .* |
| 1302 |
// ) |
| 1303 |
// (</div>) # $4 = closing tag |
| 1304 |
// \z |
| 1305 |
// }xs', $block, $matches)) |
| 1306 |
// { |
| 1307 |
// list(, $div_open, , $div_content, $div_close) = $matches; |
| 1308 |
// |
| 1309 |
// # We can't call Markdown(), because that resets the hash; |
| 1310 |
// # that initialization code should be pulled into its own sub, though. |
| 1311 |
// $div_content = $this->hashHTMLBlocks($div_content); |
| 1312 |
// |
| 1313 |
// # Run document gamut methods on the content. |
| 1314 |
// foreach ($this->document_gamut as $method => $priority) { |
| 1315 |
// $div_content = $this->$method($div_content); |
| 1316 |
// } |
| 1317 |
// |
| 1318 |
// $div_open = preg_replace( |
| 1319 |
// '{\smarkdown\s*=\s*([\'"]).+?\1}', '', $div_open); |
| 1320 |
// |
| 1321 |
// $graf = $div_open . "\n" . $div_content . "\n" . $div_close; |
| 1322 |
// } |
| 1323 |
$grafs[$key] = $graf; |
| 1324 |
} |
| 1325 |
} |
| 1326 |
|
| 1327 |
return implode( "\n\n", $grafs ); |
| 1328 |
} |
| 1329 |
|
| 1330 |
# Encode text for a double-quoted HTML attribute. This function |
| 1331 |
# is *not* suitable for attributes enclosed in single quotes. |
| 1332 |
function encodeAttribute( $text ) { |
| 1333 |
|
| 1334 |
$text = $this->encodeAmpsAndAngles( $text ); |
| 1335 |
$text = str_replace( '"', '"', $text ); |
| 1336 |
return $text; |
| 1337 |
} |
| 1338 |
|
| 1339 |
# Smart processing for ampersands and angle brackets that need to |
| 1340 |
# be encoded. Valid character entities are left alone unless the |
| 1341 |
# no-entities mode is set. |
| 1342 |
function encodeAmpsAndAngles( $text ) { |
| 1343 |
|
| 1344 |
if ( $this->no_entities ) { |
| 1345 |
$text = str_replace( '&', '&', $text ); |
| 1346 |
} else { |
| 1347 |
# Ampersand-encoding based entirely on Nat Irons's Amputator |
| 1348 |
# MT plugin: <http://bumppo.net/projects/amputator/> |
| 1349 |
$text = preg_replace( |
| 1350 |
'/&(?!#?[xX]?(?:[0-9a-fA-F]+|\w+);)/', |
| 1351 |
'&', |
| 1352 |
$text |
| 1353 |
); |
| 1354 |
} |
| 1355 |
# Encode remaining <'s |
| 1356 |
$text = str_replace( '<', '<', $text ); |
| 1357 |
|
| 1358 |
return $text; |
| 1359 |
} |
| 1360 |
|
| 1361 |
function doAutoLinks( $text ) { |
| 1362 |
|
| 1363 |
$text = preg_replace_callback( |
| 1364 |
'{<((https?|ftp|dict):[^\'">\s]+)>}i', |
| 1365 |
array( &$this, '_doAutoLinks_url_callback' ), |
| 1366 |
$text |
| 1367 |
); |
| 1368 |
|
| 1369 |
# Email addresses: <address@domain.foo> |
| 1370 |
$text = preg_replace_callback('{ |
| 1371 |
< |
| 1372 |
(?:mailto:)? |
| 1373 |
( |
| 1374 |
(?: |
| 1375 |
[-!#$%&\'*+/=?^_`.{|}~\w\x80-\xFF]+ |
| 1376 |
| |
| 1377 |
".*?" |
| 1378 |
) |
| 1379 |
\@ |
| 1380 |
(?: |
| 1381 |
[-a-z0-9\x80-\xFF]+(\.[-a-z0-9\x80-\xFF]+)*\.[a-z]+ |
| 1382 |
| |
| 1383 |
\[[\d.a-fA-F:]+\] # IPv4 & IPv6 |
| 1384 |
) |
| 1385 |
) |
| 1386 |
> |
| 1387 |
}xi', |
| 1388 |
array( &$this, '_doAutoLinks_email_callback' ), |
| 1389 |
$text |
| 1390 |
); |
| 1391 |
$text = preg_replace_callback( |
| 1392 |
'{<(tel:([^\'">\s]+))>}i', |
| 1393 |
array( &$this, '_doAutoLinks_tel_callback' ), |
| 1394 |
$text |
| 1395 |
); |
| 1396 |
|
| 1397 |
return $text; |
| 1398 |
} |
| 1399 |
|
| 1400 |
function _doAutoLinks_tel_callback( $matches ) { |
| 1401 |
$url = $this->encodeAttribute( $matches[1] ); |
| 1402 |
$tel = $this->encodeAttribute( $matches[2] ); |
| 1403 |
$link = "<a href=\"" . $url . "\">" . $tel . "</a>"; |
| 1404 |
return $this->hashPart( $link ); |
| 1405 |
} |
| 1406 |
|
| 1407 |
function _doAutoLinks_url_callback( $matches ) { |
| 1408 |
$url = $this->encodeAttribute( $matches[1] ); |
| 1409 |
$link = "<a href=\"" . $url . "\">" . $url . "</a>"; |
| 1410 |
return $this->hashPart( $link ); |
| 1411 |
} |
| 1412 |
|
| 1413 |
function _doAutoLinks_email_callback( $matches ) { |
| 1414 |
$address = $matches[1]; |
| 1415 |
$link = $this->encodeEmailAddress( $address ); |
| 1416 |
return $this->hashPart( $link ); |
| 1417 |
} |
| 1418 |
|
| 1419 |
function encodeEmailAddress( $addr ) { |
| 1420 |
# |
| 1421 |
# Input: an email address, e.g. "foo@example.com" |
| 1422 |
# |
| 1423 |
# Output: the email address as a mailto link, with each character |
| 1424 |
# of the address encoded as either a decimal or hex entity, in |
| 1425 |
# the hopes of foiling most address harvesting spam bots. E.g.: |
| 1426 |
# |
| 1427 |
# <p><a href="mailto:foo |
| 1428 |
# @example.co |
| 1429 |
# m">foo@exampl |
| 1430 |
# e.com</a></p> |
| 1431 |
# |
| 1432 |
# Based by a filter by Matthew Wickline, posted to BBEdit-Talk. |
| 1433 |
# With some optimizations by Milian Wolff. |
| 1434 |
# |
| 1435 |
$addr = "mailto:" . $addr; |
| 1436 |
$chars = preg_split( '/(?<!^)(?!$)/', $addr ); |
| 1437 |
$seed = (int) abs( crc32( $addr ) / strlen( $addr ) ); # Deterministic seed. |
| 1438 |
|
| 1439 |
foreach ( $chars as $key => $char ) { |
| 1440 |
$ord = ord( $char ); |
| 1441 |
# Ignore non-ascii chars. |
| 1442 |
if ( $ord < 128 ) { |
| 1443 |
$r = ( $seed * ( 1 + $key ) ) % 100; # Pseudo-random function. |
| 1444 |
# roughly 10% raw, 45% hex, 45% dec |
| 1445 |
# '@' *must* be encoded. I insist. |
| 1446 |
if ( $r > 90 && $char != '@' ) /* do nothing */; |
| 1447 |
else if ( $r < 45 ) $chars[$key] = '&#x' . dechex( $ord ) . ';'; |
| 1448 |
else $chars[$key] = '&#' . $ord . ';'; |
| 1449 |
} |
| 1450 |
} |
| 1451 |
|
| 1452 |
$addr = implode( '', $chars ); |
| 1453 |
$text = implode( '', array_slice( $chars, 7 ) ); # text without `mailto:` |
| 1454 |
$addr = "<a href=\"" . $addr . "\">" . $text . "</a>"; |
| 1455 |
|
| 1456 |
return $addr; |
| 1457 |
} |
| 1458 |
|
| 1459 |
function parseSpan( $str ) { |
| 1460 |
# |
| 1461 |
# Take the string $str and parse it into tokens, hashing embedded HTML, |
| 1462 |
# escaped characters and handling code spans. |
| 1463 |
# |
| 1464 |
$output = ''; |
| 1465 |
|
| 1466 |
$span_re = '{ |
| 1467 |
( |
| 1468 |
\\\\' . $this->escape_chars_re . ' |
| 1469 |
| |
| 1470 |
(?<![`\\\\]) |
| 1471 |
`+ # code span marker |
| 1472 |
' . ( $this->no_markup ? '' : ' |
| 1473 |
| |
| 1474 |
<!-- .*? --> # comment |
| 1475 |
| |
| 1476 |
<\?.*?\?> | <%.*?%> # processing instruction |
| 1477 |
| |
| 1478 |
<[!$]?[-a-zA-Z0-9:_]+ # regular tags |
| 1479 |
(?> |
| 1480 |
\s |
| 1481 |
(?>[^"\'>]+|"[^"]*"|\'[^\']*\')* |
| 1482 |
)? |
| 1483 |
> |
| 1484 |
| |
| 1485 |
<[-a-zA-Z0-9:_]+\s*/> # xml-style empty tag |
| 1486 |
| |
| 1487 |
</[-a-zA-Z0-9:_]+\s*> # closing tag |
| 1488 |
' ) . ' |
| 1489 |
) |
| 1490 |
}xs'; |
| 1491 |
|
| 1492 |
# Each loop iteration search for either the next tag, the next |
| 1493 |
# openning code span marker, or the next escaped character. |
| 1494 |
# Each token is then passed to handleSpanToken. |
| 1495 |
while ( 1 ) { |
| 1496 |
|
| 1497 |
$parts = preg_split( $span_re, $str, 2, PREG_SPLIT_DELIM_CAPTURE ); |
| 1498 |
|
| 1499 |
# Create token from text preceding tag. |
| 1500 |
if ( $parts[0] != "" ) { |
| 1501 |
$output .= $parts[0]; |
| 1502 |
} |
| 1503 |
|
| 1504 |
# Check if we reach the end. |
| 1505 |
if ( isset( $parts[1] ) ) { |
| 1506 |
$output .= $this->handleSpanToken( $parts[1], $parts[2] ); |
| 1507 |
$str = $parts[2]; |
| 1508 |
} else { |
| 1509 |
break; |
| 1510 |
} |
| 1511 |
} |
| 1512 |
|
| 1513 |
return $output; |
| 1514 |
} |
| 1515 |
|
| 1516 |
# Handle $token provided by parseSpan by determining its nature and |
| 1517 |
# returning the corresponding value that should replace it. |
| 1518 |
function handleSpanToken( $token, &$str ) { |
| 1519 |
|
| 1520 |
switch ( $token[0] ) { |
| 1521 |
case "\\": |
| 1522 |
return $this->hashPart( "&#" . ord($token[1]) . ";" ); |
| 1523 |
case "`": |
| 1524 |
# Search for end marker in remaining text. |
| 1525 |
if ( preg_match( '/^(.*?[^`])' . preg_quote( $token ) . '(?!`)(.*)$/sm', $str, $matches ) ) { |
| 1526 |
$str = $matches[2]; |
| 1527 |
$codespan = $this->makeCodeSpan( $matches[1] ); |
| 1528 |
return $this->hashPart( $codespan ); |
| 1529 |
} |
| 1530 |
return $token; // return as text since no ending marker found. |
| 1531 |
default: |
| 1532 |
return $this->hashPart( $token ); |
| 1533 |
} |
| 1534 |
} |
| 1535 |
|
| 1536 |
# Remove one level of line-leading tabs or spaces |
| 1537 |
function outdent( $text ) { |
| 1538 |
return preg_replace( '/^(\t|[ ]{1,' . $this->tab_width . '})/m', '', $text ); |
| 1539 |
} |
| 1540 |
|
| 1541 |
# String length function for detab. ``` will create a function to |
| 1542 |
# hanlde UTF-8 if the default function does not exist. |
| 1543 |
// note: removed in favour of utf8_strlen method |
| 1544 |
// public $utf8_strlen = 'mb_strlen'; |
| 1545 |
|
| 1546 |
# Replace tabs with the appropriate amount of space. |
| 1547 |
function detab( $text ) { |
| 1548 |
|
| 1549 |
# For each line we separate the line in blocks delemited by |
| 1550 |
# tab characters. Then we reconstruct every line by adding the |
| 1551 |
# appropriate number of space between each blocks. |
| 1552 |
|
| 1553 |
$text = preg_replace_callback( |
| 1554 |
'/^.*\t.*$/m', |
| 1555 |
array( &$this, '_detab_callback' ), |
| 1556 |
$text |
| 1557 |
); |
| 1558 |
|
| 1559 |
return $text; |
| 1560 |
} |
| 1561 |
|
| 1562 |
function _detab_callback( $matches ) { |
| 1563 |
$line = $matches[0]; |
| 1564 |
// $strlen = $this->utf8_strlen; # strlen function for UTF-8. |
| 1565 |
|
| 1566 |
# Split in blocks. |
| 1567 |
$blocks = explode( "\t", $line ); |
| 1568 |
# Add each blocks to the line. |
| 1569 |
$line = $blocks[0]; |
| 1570 |
unset( $blocks[0] ); # Do not add first block twice. |
| 1571 |
foreach ( $blocks as $block ) { |
| 1572 |
# Calculate amount of space, insert spaces, insert block. |
| 1573 |
$amount = $this->tab_width - $this->utf8_strlen( $line, 'UTF-8' ) % $this->tab_width; |
| 1574 |
$line .= str_repeat( " ", $amount) . $block; |
| 1575 |
} |
| 1576 |
return $line; |
| 1577 |
} |
| 1578 |
|
| 1579 |
# Check for the availability of the function in the `utf8_strlen` property |
| 1580 |
# (initially `mb_strlen`). If the function is not available, use markdown_extra_utf8_strlen |
| 1581 |
# that will loosely count the number of UTF-8 characters with a |
| 1582 |
# regular expression. |
| 1583 |
// note: removed in favour of utf8_strlen method |
| 1584 |
// function _initDetab() { |
| 1585 |
// if ( function_exists( $this->utf8_strlen ) ) { |
| 1586 |
// return; |
| 1587 |
// } |
| 1588 |
// $this->utf8_strlen = 'markdown_extra_utf8_strlen'; |
| 1589 |
// } |
| 1590 |
|
| 1591 |
// note: replacement for _initDetab and markdown_extra_utf8_strlen function |
| 1592 |
function utf8_strlen( $text ) { |
| 1593 |
if ( function_exists( 'mb_strlen' ) ) { |
| 1594 |
return mb_strlen( $text ); |
| 1595 |
} |
| 1596 |
return preg_match_all( "/[\\x00-\\xBF]|[\\xC0-\\xFF][\\x80-\\xBF]*/", $text, $m ); |
| 1597 |
} |
| 1598 |
|
| 1599 |
# Swap back in all the tags hashed by _HashHTMLBlocks. |
| 1600 |
function unhash( $text ) { |
| 1601 |
return preg_replace_callback( |
| 1602 |
'/(.)\x1A[0-9]+\1/', |
| 1603 |
array( &$this, '_unhash_callback' ), |
| 1604 |
$text |
| 1605 |
); |
| 1606 |
} |
| 1607 |
|
| 1608 |
function _unhash_callback( $matches ) { |
| 1609 |
return $this->html_hashes[$matches[0]]; |
| 1610 |
} |
| 1611 |
|
| 1612 |
} |
| 1613 |
} |
| 1614 |
|
| 1615 |
|
| 1616 |
# |
| 1617 |
# Markdown Extra Parser Class |
| 1618 |
# |
| 1619 |
if ( !class_exists( 'radio_station_markdown_extra_parser' ) ) { |
| 1620 |
class radio_station_markdown_extra_parser extends radio_station_markdown_parser { |
| 1621 |
|
| 1622 |
### Configuration Variables ### |
| 1623 |
|
| 1624 |
# Prefix for footnote ids. |
| 1625 |
public $fn_id_prefix = ""; |
| 1626 |
|
| 1627 |
# Optional title attribute for footnote links and backlinks. |
| 1628 |
public $fn_link_title = ""; // MARKDOWN_FN_LINK_TITLE |
| 1629 |
public $fn_backlink_title = ""; // MARKDOWN_FN_BACKLINK_TITLE |
| 1630 |
|
| 1631 |
# Optional class attribute for footnote links and backlinks. |
| 1632 |
public $fn_link_class = ""; // MARKDOWN_FN_LINK_CLASS |
| 1633 |
public $fn_backlink_class = ""; // MARKDOWN_FN_BACKLINK_CLASS |
| 1634 |
|
| 1635 |
# Optional class prefix for fenced code block. |
| 1636 |
public $code_class_prefix = ""; // MARKDOWN_CODE_CLASS_PREFIX |
| 1637 |
|
| 1638 |
# Class attribute for code blocks goes on the `code` tag; |
| 1639 |
# setting this to true will put attributes on the `pre` tag instead. |
| 1640 |
public $code_attr_on_pre = ""; // MARKDOWN_CODE_ATTR_ON_PRE |
| 1641 |
|
| 1642 |
# Predefined abbreviations. |
| 1643 |
public $predef_abbr = array(); |
| 1644 |
|
| 1645 |
|
| 1646 |
### Parser Implementation ### |
| 1647 |
|
| 1648 |
# Constructor function. Initialize the parser object. |
| 1649 |
function __construct() { |
| 1650 |
|
| 1651 |
# Add extra escapable characters before parent constructor |
| 1652 |
# initialize the table. |
| 1653 |
$this->escape_chars .= ':|'; |
| 1654 |
|
| 1655 |
# Insert extra document, block, and span transformations. |
| 1656 |
# Parent constructor will do the sorting. |
| 1657 |
$this->document_gamut += array( |
| 1658 |
"doFencedCodeBlocks" => 5, |
| 1659 |
"stripFootnotes" => 15, |
| 1660 |
"stripAbbreviations" => 25, |
| 1661 |
"appendFootnotes" => 50, |
| 1662 |
); |
| 1663 |
$this->block_gamut += array( |
| 1664 |
"doFencedCodeBlocks" => 5, |
| 1665 |
"doTables" => 15, |
| 1666 |
"doDefLists" => 45, |
| 1667 |
); |
| 1668 |
$this->span_gamut += array( |
| 1669 |
"doFootnotes" => 5, |
| 1670 |
"doAbbreviations" => 70, |
| 1671 |
); |
| 1672 |
|
| 1673 |
parent::__construct(); |
| 1674 |
} |
| 1675 |
|
| 1676 |
# Extra variables used during extra transformations. |
| 1677 |
public $footnotes = array(); |
| 1678 |
public $footnotes_ordered = array(); |
| 1679 |
public $footnotes_ref_count = array(); |
| 1680 |
public $footnotes_numbers = array(); |
| 1681 |
public $abbr_desciptions = array(); |
| 1682 |
public $abbr_word_re = ''; |
| 1683 |
|
| 1684 |
# Give the current footnote number. |
| 1685 |
public $footnote_counter = 1; |
| 1686 |
|
| 1687 |
# Setting up Extra-specific variables. |
| 1688 |
function setup() { |
| 1689 |
|
| 1690 |
parent::setup(); |
| 1691 |
|
| 1692 |
$this->footnotes = array(); |
| 1693 |
$this->footnotes_ordered = array(); |
| 1694 |
$this->footnotes_ref_count = array(); |
| 1695 |
$this->footnotes_numbers = array(); |
| 1696 |
$this->abbr_desciptions = array(); |
| 1697 |
$this->abbr_word_re = ''; |
| 1698 |
$this->footnote_counter = 1; |
| 1699 |
|
| 1700 |
foreach ( $this->predef_abbr as $abbr_word => $abbr_desc ) { |
| 1701 |
if ( $this->abbr_word_re ) |
| 1702 |
$this->abbr_word_re .= '|'; |
| 1703 |
$this->abbr_word_re .= preg_quote( $abbr_word ); |
| 1704 |
$this->abbr_desciptions[$abbr_word] = trim( $abbr_desc ); |
| 1705 |
} |
| 1706 |
} |
| 1707 |
|
| 1708 |
# Clearing Extra-specific variables. |
| 1709 |
function teardown() { |
| 1710 |
|
| 1711 |
$this->footnotes = array(); |
| 1712 |
$this->footnotes_ordered = array(); |
| 1713 |
$this->footnotes_ref_count = array(); |
| 1714 |
$this->footnotes_numbers = array(); |
| 1715 |
$this->abbr_desciptions = array(); |
| 1716 |
$this->abbr_word_re = ''; |
| 1717 |
|
| 1718 |
parent::teardown(); |
| 1719 |
} |
| 1720 |
|
| 1721 |
|
| 1722 |
### Extra Attribute Parser ### |
| 1723 |
|
| 1724 |
# Expression to use to catch attributes (includes the braces) |
| 1725 |
public $id_class_attr_catch_re = '\{((?:[ ]*[#.][-_:a-zA-Z0-9]+){1,})[ ]*\}'; |
| 1726 |
# Expression to use when parsing in a context when no capture is desired |
| 1727 |
public $id_class_attr_nocatch_re = '\{(?:[ ]*[#.][-_:a-zA-Z0-9]+){1,}[ ]*\}'; |
| 1728 |
|
| 1729 |
# Parse attributes caught by the $this->id_class_attr_catch_re expression |
| 1730 |
# and return the HTML-formatted list of attributes. |
| 1731 |
# Currently supported attributes are .class and #id. |
| 1732 |
function doExtraAttributes( $tag_name, $attr ) { |
| 1733 |
|
| 1734 |
if ( empty( $attr ) ) { |
| 1735 |
return ""; |
| 1736 |
} |
| 1737 |
|
| 1738 |
# Split on components |
| 1739 |
preg_match_all( '/[#.][-_:a-zA-Z0-9]+/', $attr, $matches ); |
| 1740 |
$elements = $matches[0]; |
| 1741 |
|
| 1742 |
# handle classes and ids (only first id taken into account) |
| 1743 |
$classes = array(); |
| 1744 |
$id = false; |
| 1745 |
foreach ( $elements as $element ) { |
| 1746 |
if ( $element[0] == '.' ) { |
| 1747 |
$classes[] = substr( $element, 1 ); |
| 1748 |
} elseif ( $element[0] == '#' ) { |
| 1749 |
if ( $id === false ) { |
| 1750 |
$id = substr( $element, 1 ); |
| 1751 |
} |
| 1752 |
} |
| 1753 |
} |
| 1754 |
|
| 1755 |
# compose attributes as string |
| 1756 |
$attr_str = ""; |
| 1757 |
if ( !empty( $id ) ) { |
| 1758 |
$attr_str .= ' id="' . $id . '"'; |
| 1759 |
} |
| 1760 |
if ( !empty( $classes ) ) { |
| 1761 |
$attr_str .= ' class="' . implode( " ", $classes ) . '"'; |
| 1762 |
} |
| 1763 |
return $attr_str; |
| 1764 |
} |
| 1765 |
|
| 1766 |
# Strips link definitions from text, stores the URLs and titles in |
| 1767 |
# hash references. |
| 1768 |
function stripLinkDefinitions( $text ) { |
| 1769 |
|
| 1770 |
$less_than_tab = $this->tab_width - 1; |
| 1771 |
|
| 1772 |
# Link defs are in the form: ^[id]: url "optional title" |
| 1773 |
$text = preg_replace_callback( '{ |
| 1774 |
^[ ]{0,' . $less_than_tab . '}\[(.+)\][ ]?: # id = $1 |
| 1775 |
[ ]* |
| 1776 |
\n? # maybe *one* newline |
| 1777 |
[ ]* |
| 1778 |
(?: |
| 1779 |
<(.+?)> # url = $2 |
| 1780 |
| |
| 1781 |
(\S+?) # url = $3 |
| 1782 |
) |
| 1783 |
[ ]* |
| 1784 |
\n? # maybe one newline |
| 1785 |
[ ]* |
| 1786 |
(?: |
| 1787 |
(?<=\s) # lookbehind for whitespace |
| 1788 |
["(] |
| 1789 |
(.*?) # title = $4 |
| 1790 |
[")] |
| 1791 |
[ ]* |
| 1792 |
)? # title is optional |
| 1793 |
(?:[ ]* ' . $this->id_class_attr_catch_re . ' )? # $5 = extra id & class attr |
| 1794 |
(?:\n+|\Z) |
| 1795 |
}xm', |
| 1796 |
array( &$this, '_stripLinkDefinitions_callback' ), |
| 1797 |
$text |
| 1798 |
); |
| 1799 |
return $text; |
| 1800 |
} |
| 1801 |
|
| 1802 |
function _stripLinkDefinitions_callback( $matches ) { |
| 1803 |
$link_id = strtolower( $matches[1] ); |
| 1804 |
$url = $matches[2] == '' ? $matches[3] : $matches[2]; |
| 1805 |
$this->urls[$link_id] = $url; |
| 1806 |
$this->titles[$link_id] =& $matches[4]; |
| 1807 |
$this->ref_attr[$link_id] = $this->doExtraAttributes( "", $dummy =& $matches[5] ); |
| 1808 |
return ''; # String that will replace the block |
| 1809 |
} |
| 1810 |
|
| 1811 |
### HTML Block Parser ### |
| 1812 |
|
| 1813 |
# Tags that are always treated as block tags: |
| 1814 |
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'; |
| 1815 |
|
| 1816 |
# Tags treated as block tags only if the opening tag is alone on its line: |
| 1817 |
public $context_block_tags_re = 'script|noscript|ins|del|iframe|object|source|track|param|math|svg|canvas|audio|video'; |
| 1818 |
|
| 1819 |
# Tags where markdown="1" default to span mode: |
| 1820 |
public $contain_span_tags_re = 'p|h[1-6]|li|dd|dt|td|th|legend|address'; |
| 1821 |
|
| 1822 |
# Tags which must not have their contents modified, no matter where |
| 1823 |
# they appear: |
| 1824 |
public $clean_tags_re = 'script|math|svg'; |
| 1825 |
|
| 1826 |
# Tags that do not need to be closed. |
| 1827 |
public $auto_close_tags_re = 'hr|img|param|source|track'; |
| 1828 |
|
| 1829 |
|
| 1830 |
# Hashify HTML Blocks and "clean tags". |
| 1831 |
# |
| 1832 |
# We only want to do this for block-level HTML tags, such as headers, |
| 1833 |
# lists, and tables. That's because we still want to wrap <p>s around |
| 1834 |
# "paragraphs" that are wrapped in non-block-level tags, such as anchors, |
| 1835 |
# phrase emphasis, and spans. The list of tags we're looking for is |
| 1836 |
# hard-coded. |
| 1837 |
# |
| 1838 |
# This works by calling _HashHTMLBlocks_InMarkdown, which then calls |
| 1839 |
# _HashHTMLBlocks_InHTML when it encounter block tags. When the markdown="1" |
| 1840 |
# attribute is found within a tag, _HashHTMLBlocks_InHTML calls back |
| 1841 |
# _HashHTMLBlocks_InMarkdown to handle the Markdown syntax within the tag. |
| 1842 |
# These two functions are calling each other. It's recursive! |
| 1843 |
function hashHTMLBlocks( $text ) { |
| 1844 |
|
| 1845 |
if ( $this->no_markup ) { |
| 1846 |
return $text; |
| 1847 |
} |
| 1848 |
|
| 1849 |
# Call the HTML-in-Markdown hasher. |
| 1850 |
list( $text, ) = $this->_hashHTMLBlocks_inMarkdown( $text ); |
| 1851 |
|
| 1852 |
return $text; |
| 1853 |
} |
| 1854 |
|
| 1855 |
# Parse markdown text, calling _HashHTMLBlocks_InHTML for block tags. |
| 1856 |
# |
| 1857 |
# * $indent is the number of space to be ignored when checking for code |
| 1858 |
# blocks. This is important because if we don't take the indent into |
| 1859 |
# account, something like this (which looks right) won't work as expected: |
| 1860 |
# |
| 1861 |
# <div> |
| 1862 |
# <div markdown="1"> |
| 1863 |
# Hello World. <-- Is this a Markdown code block or text? |
| 1864 |
# </div> <-- Is this a Markdown code block or a real tag? |
| 1865 |
# <div> |
| 1866 |
# |
| 1867 |
# If you don't like this, just don't indent the tag on which |
| 1868 |
# you apply the markdown="1" attribute. |
| 1869 |
# |
| 1870 |
# * If $enclosing_tag_re is not empty, stops at the first unmatched closing |
| 1871 |
# tag with that name. Nested tags supported. |
| 1872 |
# |
| 1873 |
# * If $span is true, text inside must treated as span. So any double |
| 1874 |
# newline will be replaced by a single newline so that it does not create |
| 1875 |
# paragraphs. |
| 1876 |
# |
| 1877 |
# Returns an array of that form: ( processed text , remaining text ) |
| 1878 |
function _hashHTMLBlocks_inMarkdown( $text, $indent = 0, $enclosing_tag_re = '', $span = false ) { |
| 1879 |
|
| 1880 |
if ( $text === '' ) { |
| 1881 |
return array( '', '' ); |
| 1882 |
} |
| 1883 |
|
| 1884 |
# Regex to check for the presence of newlines around a block tag. |
| 1885 |
$newline_before_re = '/(?:^\n?|\n\n)*$/'; |
| 1886 |
$newline_after_re = |
| 1887 |
'{ |
| 1888 |
^ # Start of text following the tag. |
| 1889 |
(?>[ ]*<!--.*?-->)? # Optional comment. |
| 1890 |
[ ]*\n # Must be followed by newline. |
| 1891 |
}xs'; |
| 1892 |
|
| 1893 |
# Regex to match any tag. |
| 1894 |
$block_tag_re = |
| 1895 |
'{ |
| 1896 |
( # $2: Capture whole tag. |
| 1897 |
</? # Any opening or closing tag. |
| 1898 |
(?> # Tag name. |
| 1899 |
' . $this->block_tags_re . ' | |
| 1900 |
' . $this->context_block_tags_re . ' | |
| 1901 |
' . $this->clean_tags_re . ' | |
| 1902 |
(?!\s)' . $enclosing_tag_re . ' |
| 1903 |
) |
| 1904 |
(?: |
| 1905 |
(?=[\s"\'/a-zA-Z0-9]) # Allowed characters after tag name. |
| 1906 |
(?> |
| 1907 |
".*?" | # Double quotes (can contain `>`) |
| 1908 |
\'.*?\' | # Single quotes (can contain `>`) |
| 1909 |
.+? # Anything but quotes and `>`. |
| 1910 |
)*? |
| 1911 |
)? |
| 1912 |
> # End of tag. |
| 1913 |
| |
| 1914 |
<!-- .*? --> # HTML Comment |
| 1915 |
| |
| 1916 |
<\?.*?\?> | <%.*?%> # Processing instruction |
| 1917 |
| |
| 1918 |
<!\[CDATA\[.*?\]\]> # CData Block |
| 1919 |
'. ( !$span ? ' # If not in span. |
| 1920 |
| |
| 1921 |
# Indented code block |
| 1922 |
(?: ^[ ]*\n | ^ | \n[ ]*\n ) |
| 1923 |
[ ]{' . ( $indent + 4 ) .'}[^\n]* \n |
| 1924 |
(?> |
| 1925 |
(?: [ ]{' . ( $indent + 4 ) . '}[^\n]* | [ ]* ) \n |
| 1926 |
)* |
| 1927 |
| |
| 1928 |
# Fenced code block marker |
| 1929 |
(?<= ^ | \n ) |
| 1930 |
[ ]{0,' . ( $indent + 3 ) .'}(?:~{3,}|`{3,}) |
| 1931 |
[ ]* |
| 1932 |
(?: |
| 1933 |
\.?[-_:a-zA-Z0-9]+ # standalone class name |
| 1934 |
| |
| 1935 |
' . $this->id_class_attr_nocatch_re . ' # extra attributes |
| 1936 |
)? |
| 1937 |
[ ]* |
| 1938 |
(?= \n ) |
| 1939 |
' : '' ) . ' # End (if not is span). |
| 1940 |
| |
| 1941 |
# Code span marker |
| 1942 |
# Note, this regex needs to go after backtick fenced |
| 1943 |
# code blocks but it should also be kept outside of the |
| 1944 |
# "if not in span" condition adding backticks to the parser |
| 1945 |
`+ |
| 1946 |
) |
| 1947 |
}xs'; |
| 1948 |
|
| 1949 |
|
| 1950 |
$depth = 0; # Current depth inside the tag tree. |
| 1951 |
$parsed = ""; # Parsed text that will be returned. |
| 1952 |
|
| 1953 |
# Loop through every tag until we find the closing tag of the parent |
| 1954 |
# or loop until reaching the end of text if no parent tag specified. |
| 1955 |
do { |
| 1956 |
|
| 1957 |
# Split the text using the first $tag_match pattern found. |
| 1958 |
# Text before pattern will be first in the array, text after |
| 1959 |
# pattern will be at the end, and between will be any catches made |
| 1960 |
# by the pattern. |
| 1961 |
$parts = preg_split( $block_tag_re, $text, 2, PREG_SPLIT_DELIM_CAPTURE ); |
| 1962 |
|
| 1963 |
# If in Markdown span mode, add a empty-string span-level hash |
| 1964 |
# after each newline to prevent triggering any block element. |
| 1965 |
if ( $span ) { |
| 1966 |
$void = $this->hashPart( "", ':' ); |
| 1967 |
$newline = "$void\n"; |
| 1968 |
$parts[0] = $void . str_replace( "\n", $newline, $parts[0] ) . $void; |
| 1969 |
} |
| 1970 |
|
| 1971 |
$parsed .= $parts[0]; # Text before current tag. |
| 1972 |
|
| 1973 |
# If end of $text has been reached. Stop loop. |
| 1974 |
if ( count( $parts ) < 3 ) { |
| 1975 |
$text = ""; |
| 1976 |
break; |
| 1977 |
} |
| 1978 |
|
| 1979 |
$tag = $parts[1]; # Tag to handle. |
| 1980 |
$text = $parts[2]; # Remaining text after current tag. |
| 1981 |
$tag_re = preg_quote( $tag ); # For use in a regular expression. |
| 1982 |
|
| 1983 |
# Check for: Fenced code block marker. |
| 1984 |
# Note: need to recheck the whole tag to disambiguate backtick |
| 1985 |
# fences from code spans |
| 1986 |
if ( preg_match( '{^\n?([ ]{0,'.($indent+3).'})(~{3,}|`{3,})[ ]*(?:\.?[-_:a-zA-Z0-9]+|' . $this->id_class_attr_nocatch_re . ')?[ ]*\n?$}', $tag, $capture ) ) { |
| 1987 |
# Fenced code block marker: find matching end marker. |
| 1988 |
$fence_indent = strlen( $capture[1] ); # use captured indent in re |
| 1989 |
$fence_re = $capture[2]; # use captured fence in re |
| 1990 |
if ( preg_match( '{^(?>.*\n)*?[ ]{' . ( $fence_indent ) . '}' . $fence_re . '[ ]*(?:\n|$)}', $text, $matches ) ) { |
| 1991 |
# End marker found: pass text unchanged until marker. |
| 1992 |
$parsed .= $tag . $matches[0]; |
| 1993 |
$text = substr( $text, strlen( $matches[0] ) ); |
| 1994 |
} else { |
| 1995 |
# No end marker: just skip it. |
| 1996 |
$parsed .= $tag; |
| 1997 |
} |
| 1998 |
} |
| 1999 |
|
| 2000 |
# Check for: Indented code block. |
| 2001 |
elseif ( $tag[0] == "\n" || $tag[0] == " " ) { |
| 2002 |
# Indented code block: pass it unchanged, will be handled |
| 2003 |
# later. |
| 2004 |
$parsed .= $tag; |
| 2005 |
} |
| 2006 |
|
| 2007 |
# Check for: Code span marker |
| 2008 |
# Note: need to check this after backtick fenced code blocks |
| 2009 |
elseif ( $tag[0] == "`" ) { |
| 2010 |
# Find corresponding end marker. |
| 2011 |
$tag_re = preg_quote( $tag ); |
| 2012 |
if ( preg_match('{^(?>.+?|\n(?!\n))*?(?<!`)' . $tag_re . '(?!`)}', $text, $matches ) ) { |
| 2013 |
# End marker found: pass text unchanged until marker. |
| 2014 |
$parsed .= $tag . $matches[0]; |
| 2015 |
$text = substr( $text, strlen( $matches[0] ) ); |
| 2016 |
} else { |
| 2017 |
# Unmatched marker: just skip it. |
| 2018 |
$parsed .= $tag; |
| 2019 |
} |
| 2020 |
} |
| 2021 |
|
| 2022 |
# Check for: Opening Block level tag or |
| 2023 |
# Opening Context Block tag (like ins and del) |
| 2024 |
# used as a block tag (tag is alone on it's line). |
| 2025 |
elseif ( preg_match( '{^<(?:' . $this->block_tags_re . ')\b}', $tag ) || |
| 2026 |
( preg_match('{^<(?:' . $this->context_block_tags_re.')\b}', $tag ) && |
| 2027 |
preg_match( $newline_before_re, $parsed ) && |
| 2028 |
preg_match( $newline_after_re, $text ) ) |
| 2029 |
) { |
| 2030 |
# Need to parse tag and following text using the HTML parser. |
| 2031 |
list( $block_text, $text ) = $this->_hashHTMLBlocks_inHTML( $tag . $text, "hashBlock", true ); |
| 2032 |
|
| 2033 |
# Make sure it stays outside of any paragraph by adding newlines. |
| 2034 |
$parsed .= "\n\n" . $block_text . "\n\n"; |
| 2035 |
} |
| 2036 |
|
| 2037 |
# Check for: Clean tag (like script, math) |
| 2038 |
# HTML Comments, processing instructions. |
| 2039 |
elseif ( preg_match( '{^<(?:'.$this->clean_tags_re.')\b}', $tag ) || $tag[1] == '!' || $tag[1] == '?' ) { |
| 2040 |
# Need to parse tag and following text using the HTML parser. |
| 2041 |
# (don't check for markdown attribute) |
| 2042 |
list( $block_text, $text ) = $this->_hashHTMLBlocks_inHTML( $tag . $text, "hashClean", false ); |
| 2043 |
|
| 2044 |
$parsed .= $block_text; |
| 2045 |
} |
| 2046 |
|
| 2047 |
# Check for: Tag with same name as enclosing tag. |
| 2048 |
elseif ( $enclosing_tag_re !== '' && |
| 2049 |
# Same name as enclosing tag. |
| 2050 |
preg_match( '{^</?(?:' . $enclosing_tag_re . ')\b}', $tag ) ) { |
| 2051 |
|
| 2052 |
# |
| 2053 |
# Increase/decrease nested tag count. |
| 2054 |
# |
| 2055 |
if ( $tag[1] == '/' ) $depth--; |
| 2056 |
else if ( $tag[strlen($tag)-2] != '/' ) $depth++; |
| 2057 |
|
| 2058 |
if ( $depth < 0 ) { |
| 2059 |
# |
| 2060 |
# Going out of parent element. Clean up and break so we |
| 2061 |
# return to the calling function. |
| 2062 |
# |
| 2063 |
$text = $tag . $text; |
| 2064 |
break; |
| 2065 |
} |
| 2066 |
|
| 2067 |
$parsed .= $tag; |
| 2068 |
} else { |
| 2069 |
$parsed .= $tag; |
| 2070 |
} |
| 2071 |
} while ( $depth >= 0 ); |
| 2072 |
|
| 2073 |
return array( $parsed, $text ); |
| 2074 |
} |
| 2075 |
|
| 2076 |
# Parse HTML, calling _HashHTMLBlocks_InMarkdown for block tags. |
| 2077 |
# |
| 2078 |
# * Calls $hash_method to convert any blocks. |
| 2079 |
# * Stops when the first opening tag closes. |
| 2080 |
# * $md_attr indicate if the use of the `markdown="1"` attribute is allowed. |
| 2081 |
# (it is not inside clean tags) |
| 2082 |
# |
| 2083 |
# Returns an array of that form: ( processed text , remaining text ) |
| 2084 |
function _hashHTMLBlocks_inHTML( $text, $hash_method, $md_attr ) { |
| 2085 |
|
| 2086 |
if ( $text === '' ) { |
| 2087 |
return array( '', '' ); |
| 2088 |
} |
| 2089 |
|
| 2090 |
# Regex to match `markdown` attribute inside of a tag. |
| 2091 |
$markdown_attr_re = ' |
| 2092 |
{ |
| 2093 |
\s* # Eat whitespace before the `markdown` attribute |
| 2094 |
markdown |
| 2095 |
\s*=\s* |
| 2096 |
(?> |
| 2097 |
(["\']) # $1: quote delimiter |
| 2098 |
(.*?) # $2: attribute value |
| 2099 |
\1 # matching delimiter |
| 2100 |
| |
| 2101 |
([^\s>]*) # $3: unquoted attribute value |
| 2102 |
) |
| 2103 |
() # $4: make $3 always defined (avoid warnings) |
| 2104 |
}xs'; |
| 2105 |
|
| 2106 |
# Regex to match any tag. |
| 2107 |
$tag_re = '{ |
| 2108 |
( # $2: Capture whole tag. |
| 2109 |
</? # Any opening or closing tag. |
| 2110 |
[\w:$]+ # Tag name. |
| 2111 |
(?: |
| 2112 |
(?=[\s"\'/a-zA-Z0-9]) # Allowed characters after tag name. |
| 2113 |
(?> |
| 2114 |
".*?" | # Double quotes (can contain `>`) |
| 2115 |
\'.*?\' | # Single quotes (can contain `>`) |
| 2116 |
.+? # Anything but quotes and `>`. |
| 2117 |
)*? |
| 2118 |
)? |
| 2119 |
> # End of tag. |
| 2120 |
| |
| 2121 |
<!-- .*? --> # HTML Comment |
| 2122 |
| |
| 2123 |
<\?.*?\?> | <%.*?%> # Processing instruction |
| 2124 |
| |
| 2125 |
<!\[CDATA\[.*?\]\]> # CData Block |
| 2126 |
) |
| 2127 |
}xs'; |
| 2128 |
|
| 2129 |
$original_text = $text; # Save original text in case of faliure. |
| 2130 |
|
| 2131 |
$depth = 0; # Current depth inside the tag tree. |
| 2132 |
$block_text = ""; # Temporary text holder for current text. |
| 2133 |
$parsed = ""; # Parsed text that will be returned. |
| 2134 |
|
| 2135 |
# Get the name of the starting tag. |
| 2136 |
# (This pattern makes $base_tag_name_re safe without quoting.) |
| 2137 |
if ( preg_match( '/^<([\w:$]*)\b/', $text, $matches ) ) { |
| 2138 |
$base_tag_name_re = $matches[1]; |
| 2139 |
} |
| 2140 |
|
| 2141 |
# Loop through every tag until we find the corresponding closing tag. |
| 2142 |
do { |
| 2143 |
|
| 2144 |
# Split the text using the first $tag_match pattern found. |
| 2145 |
# Text before pattern will be first in the array, text after |
| 2146 |
# pattern will be at the end, and between will be any catches made |
| 2147 |
# by the pattern. |
| 2148 |
$parts = preg_split( $tag_re, $text, 2, PREG_SPLIT_DELIM_CAPTURE ); |
| 2149 |
|
| 2150 |
if ( count( $parts ) < 3 ) { |
| 2151 |
|
| 2152 |
# End of $text reached with unbalenced tag(s). |
| 2153 |
# In that case, we return original text unchanged and pass the |
| 2154 |
# first character as filtered to prevent an infinite loop in the |
| 2155 |
# parent function. |
| 2156 |
return array( $original_text[0], substr( $original_text, 1 ) ); |
| 2157 |
} |
| 2158 |
|
| 2159 |
$block_text .= $parts[0]; # Text before current tag. |
| 2160 |
$tag = $parts[1]; # Tag to handle. |
| 2161 |
$text = $parts[2]; # Remaining text after current tag. |
| 2162 |
|
| 2163 |
# Check for: Auto-close tag (like <hr/>) |
| 2164 |
# Comments and Processing Instructions. |
| 2165 |
if ( preg_match('{^</?(?:' . $this->auto_close_tags_re.')\b}', $tag ) || $tag[1] == '!' || $tag[1] == '?' ) { |
| 2166 |
# Just add the tag to the block as if it was text. |
| 2167 |
$block_text .= $tag; |
| 2168 |
} else { |
| 2169 |
# Increase/decrease nested tag count. Only do so if |
| 2170 |
# the tag's name match base tag's. |
| 2171 |
if ( preg_match( '{^</?' . $base_tag_name_re . '\b}', $tag ) ) { |
| 2172 |
if ( $tag[1] == '/' ) { |
| 2173 |
$depth--; |
| 2174 |
} elseif ( $tag[strlen($tag)-2] != '/' ) { |
| 2175 |
$depth++; |
| 2176 |
} |
| 2177 |
} |
| 2178 |
|
| 2179 |
# Check for `markdown="1"` attribute and handle it. |
| 2180 |
if ( $md_attr && |
| 2181 |
preg_match( $markdown_attr_re, $tag, $attr_m ) && |
| 2182 |
preg_match( '/^1|block|span$/', $attr_m[2] . $attr_m[3] ) ) { |
| 2183 |
|
| 2184 |
# Remove `markdown` attribute from opening tag. |
| 2185 |
$tag = preg_replace( $markdown_attr_re, '', $tag ); |
| 2186 |
|
| 2187 |
# Check if text inside this tag must be parsed in span mode. |
| 2188 |
$this->mode = $attr_m[2] . $attr_m[3]; |
| 2189 |
$span_mode = $this->mode == 'span' || $this->mode != 'block' && |
| 2190 |
preg_match( '{^<(?:'.$this->contain_span_tags_re.')\b}', $tag ); |
| 2191 |
|
| 2192 |
# Calculate indent before tag. |
| 2193 |
if ( preg_match( '/(?:^|\n)( *?)(?! ).*?$/', $block_text, $matches ) ) { |
| 2194 |
// $strlen = $this->utf8_strlen; |
| 2195 |
$indent = $this->utf8_strlen( $matches[1], 'UTF-8' ); |
| 2196 |
} else { |
| 2197 |
$indent = 0; |
| 2198 |
} |
| 2199 |
|
| 2200 |
# End preceding block with this tag. |
| 2201 |
$block_text .= $tag; |
| 2202 |
$parsed .= $this->$hash_method( $block_text ); |
| 2203 |
|
| 2204 |
# Get enclosing tag name for the ParseMarkdown function. |
| 2205 |
# (This pattern makes $tag_name_re safe without quoting.) |
| 2206 |
preg_match( '/^<([\w:$]*)\b/', $tag, $matches ); |
| 2207 |
$tag_name_re = $matches[1]; |
| 2208 |
|
| 2209 |
# Parse the content using the HTML-in-Markdown parser. |
| 2210 |
list( $block_text, $text ) = $this->_hashHTMLBlocks_inMarkdown($text, $indent, $tag_name_re, $span_mode); |
| 2211 |
|
| 2212 |
# Outdent markdown text. |
| 2213 |
if ( $indent > 0 ) { |
| 2214 |
$block_text = preg_replace( "/^[ ]{1,$indent}/m", "", $block_text ); |
| 2215 |
} |
| 2216 |
|
| 2217 |
# Append tag content to parsed text. |
| 2218 |
if ( !$span_mode ) { |
| 2219 |
$parsed .= "\n\n" . $block_text . "\n\n"; |
| 2220 |
} else { |
| 2221 |
$parsed .= $block_text; |
| 2222 |
} |
| 2223 |
|
| 2224 |
# Start over with a new block. |
| 2225 |
$block_text = ""; |
| 2226 |
} else { |
| 2227 |
$block_text .= $tag; |
| 2228 |
} |
| 2229 |
} |
| 2230 |
|
| 2231 |
} while ( $depth > 0 ); |
| 2232 |
|
| 2233 |
# Hash last block text that wasn't processed inside the loop. |
| 2234 |
$parsed .= $this->$hash_method( $block_text ); |
| 2235 |
|
| 2236 |
return array( $parsed, $text ); |
| 2237 |
} |
| 2238 |
|
| 2239 |
# Called whenever a tag must be hashed when a function inserts a "clean" tag |
| 2240 |
# in $text, it passes through this function and is automaticaly escaped, |
| 2241 |
# blocking invalid nested overlap. |
| 2242 |
function hashClean( $text ) { |
| 2243 |
return $this->hashPart( $text, 'C' ); |
| 2244 |
} |
| 2245 |
|
| 2246 |
# Turn Markdown link shortcuts into XHTML <a> tags. |
| 2247 |
function doAnchors( $text ) { |
| 2248 |
|
| 2249 |
if ( $this->in_anchor ) { |
| 2250 |
return $text; |
| 2251 |
} |
| 2252 |
$this->in_anchor = true; |
| 2253 |
|
| 2254 |
# First, handle reference-style links: [link text] [id] |
| 2255 |
$text = preg_replace_callback( '{ |
| 2256 |
( # wrap whole match in $1 |
| 2257 |
\[ |
| 2258 |
(' . $this->nested_brackets_re . ') # link text = $2 |
| 2259 |
\] |
| 2260 |
|
| 2261 |
[ ]? # one optional space |
| 2262 |
(?:\n[ ]*)? # one optional newline followed by spaces |
| 2263 |
|
| 2264 |
\[ |
| 2265 |
(.*?) # id = $3 |
| 2266 |
\] |
| 2267 |
) |
| 2268 |
}xs', |
| 2269 |
array( &$this, '_doAnchors_reference_callback' ), $text ); |
| 2270 |
|
| 2271 |
# |
| 2272 |
# Next, inline-style links: [link text](url "optional title") |
| 2273 |
# |
| 2274 |
$text = preg_replace_callback( '{ |
| 2275 |
( # wrap whole match in $1 |
| 2276 |
\[ |
| 2277 |
(' . $this->nested_brackets_re . ') # link text = $2 |
| 2278 |
\] |
| 2279 |
\( # literal paren |
| 2280 |
[ \n]* |
| 2281 |
(?: |
| 2282 |
<(.+?)> # href = $3 |
| 2283 |
| |
| 2284 |
(' . $this->nested_url_parenthesis_re . ') # href = $4 |
| 2285 |
) |
| 2286 |
[ \n]* |
| 2287 |
( # $5 |
| 2288 |
([\'"]) # quote char = $6 |
| 2289 |
(.*?) # Title = $7 |
| 2290 |
\6 # matching quote |
| 2291 |
[ \n]* # ignore any spaces/tabs between closing quote and ) |
| 2292 |
)? # title is optional |
| 2293 |
\) |
| 2294 |
(?:[ ]? ' . $this->id_class_attr_catch_re . ' )? # $8 = id/class attributes |
| 2295 |
) |
| 2296 |
}xs', |
| 2297 |
array( &$this, '_doAnchors_inline_callback' ), |
| 2298 |
$text |
| 2299 |
); |
| 2300 |
|
| 2301 |
# Last, handle reference-style shortcuts: [link text] |
| 2302 |
# These must come last in case you've also got [link text][1] |
| 2303 |
# or [link text](/foo) |
| 2304 |
$text = preg_replace_callback( '{ |
| 2305 |
( # wrap whole match in $1 |
| 2306 |
\[ |
| 2307 |
([^\[\]]+) # link text = $2; can\'t contain [ or ] |
| 2308 |
\] |
| 2309 |
) |
| 2310 |
}xs', |
| 2311 |
array( &$this, '_doAnchors_reference_callback'), |
| 2312 |
$text |
| 2313 |
); |
| 2314 |
|
| 2315 |
$this->in_anchor = false; |
| 2316 |
return $text; |
| 2317 |
} |
| 2318 |
|
| 2319 |
function _doAnchors_reference_callback( $matches ) { |
| 2320 |
$whole_match = $matches[1]; |
| 2321 |
$link_text = $matches[2]; |
| 2322 |
$link_id =& $matches[3]; |
| 2323 |
|
| 2324 |
if ( $link_id == "" ) { |
| 2325 |
# for shortcut links like [this][] or [this]. |
| 2326 |
$link_id = $link_text; |
| 2327 |
} |
| 2328 |
|
| 2329 |
# lower-case and turn embedded newlines into spaces |
| 2330 |
$link_id = strtolower( $link_id ); |
| 2331 |
$link_id = preg_replace( '{[ ]?\n}', ' ', $link_id ); |
| 2332 |
|
| 2333 |
if ( isset( $this->urls[$link_id] ) ) { |
| 2334 |
$url = $this->urls[$link_id]; |
| 2335 |
$url = $this->encodeAttribute( $url ); |
| 2336 |
|
| 2337 |
$result = "<a href=\"" . $url . "\""; |
| 2338 |
if ( isset( $this->titles[$link_id] ) ) { |
| 2339 |
$title = $this->titles[$link_id]; |
| 2340 |
$title = $this->encodeAttribute( $title ); |
| 2341 |
$result .= " title=\"" . $title . "\""; |
| 2342 |
} |
| 2343 |
if ( isset( $this->ref_attr[$link_id] ) ) |
| 2344 |
$result .= $this->ref_attr[$link_id]; |
| 2345 |
|
| 2346 |
$link_text = $this->runSpanGamut( $link_text ); |
| 2347 |
$result .= ">" . $link_text . "</a>"; |
| 2348 |
$result = $this->hashPart( $result ); |
| 2349 |
} else { |
| 2350 |
$result = $whole_match; |
| 2351 |
} |
| 2352 |
return $result; |
| 2353 |
} |
| 2354 |
|
| 2355 |
function _doAnchors_inline_callback( $matches ) { |
| 2356 |
$whole_match = $matches[1]; |
| 2357 |
$link_text = $this->runSpanGamut( $matches[2] ); |
| 2358 |
$url = $matches[3] == '' ? $matches[4] : $matches[3]; |
| 2359 |
$title =& $matches[7]; |
| 2360 |
$attr = $this->doExtraAttributes( "a", $dummy =& $matches[8] ); |
| 2361 |
|
| 2362 |
$url = $this->encodeAttribute( $url ); |
| 2363 |
|
| 2364 |
$result = "<a href=\"" . $url . "\""; |
| 2365 |
if ( isset( $title ) ) { |
| 2366 |
$title = $this->encodeAttribute( $title ); |
| 2367 |
$result .= " title=\"" . $title . "\""; |
| 2368 |
} |
| 2369 |
$result .= $attr; |
| 2370 |
|
| 2371 |
$link_text = $this->runSpanGamut( $link_text ); |
| 2372 |
$result .= ">" . $link_text . "</a>"; |
| 2373 |
|
| 2374 |
return $this->hashPart( $result ); |
| 2375 |
} |
| 2376 |
|
| 2377 |
# Turn Markdown image shortcuts into <img> tags. |
| 2378 |
function doImages( $text ) { |
| 2379 |
|
| 2380 |
# First, handle reference-style labeled images: ![alt text][id] |
| 2381 |
$text = preg_replace_callback( '{ |
| 2382 |
( # wrap whole match in $1 |
| 2383 |
!\[ |
| 2384 |
(' . $this->nested_brackets_re . ') # alt text = $2 |
| 2385 |
\] |
| 2386 |
|
| 2387 |
[ ]? # one optional space |
| 2388 |
(?:\n[ ]*)? # one optional newline followed by spaces |
| 2389 |
|
| 2390 |
\[ |
| 2391 |
(.*?) # id = $3 |
| 2392 |
\] |
| 2393 |
|
| 2394 |
) |
| 2395 |
}xs', |
| 2396 |
array( &$this, '_doImages_reference_callback' ), |
| 2397 |
$text |
| 2398 |
); |
| 2399 |
|
| 2400 |
# Next, handle inline images:  |
| 2401 |
# Don't forget: encode * and _ |
| 2402 |
$text = preg_replace_callback( '{ |
| 2403 |
( # wrap whole match in $1 |
| 2404 |
!\[ |
| 2405 |
(' . $this->nested_brackets_re . ') # alt text = $2 |
| 2406 |
\] |
| 2407 |
\s? # One optional whitespace character |
| 2408 |
\( # literal paren |
| 2409 |
[ \n]* |
| 2410 |
(?: |
| 2411 |
<(\S*)> # src url = $3 |
| 2412 |
| |
| 2413 |
(' . $this->nested_url_parenthesis_re . ') # src url = $4 |
| 2414 |
) |
| 2415 |
[ \n]* |
| 2416 |
( # $5 |
| 2417 |
([\'"]) # quote char = $6 |
| 2418 |
(.*?) # title = $7 |
| 2419 |
\6 # matching quote |
| 2420 |
[ \n]* |
| 2421 |
)? # title is optional |
| 2422 |
\) |
| 2423 |
(?:[ ]? ' . $this->id_class_attr_catch_re . ' )? # $8 = id/class attributes |
| 2424 |
) |
| 2425 |
}xs', |
| 2426 |
array( &$this, '_doImages_inline_callback' ), |
| 2427 |
$text |
| 2428 |
); |
| 2429 |
|
| 2430 |
return $text; |
| 2431 |
} |
| 2432 |
|
| 2433 |
function _doImages_reference_callback( $matches ) { |
| 2434 |
$whole_match = $matches[1]; |
| 2435 |
$alt_text = $matches[2]; |
| 2436 |
$link_id = strtolower( $matches[3] ); |
| 2437 |
|
| 2438 |
if ( $link_id == "" ) { |
| 2439 |
$link_id = strtolower( $alt_text ); # for shortcut links like ![this][]. |
| 2440 |
} |
| 2441 |
|
| 2442 |
$alt_text = $this->encodeAttribute( $alt_text ); |
| 2443 |
if ( isset($this->urls[$link_id] ) ) { |
| 2444 |
$url = $this->encodeAttribute( $this->urls[$link_id] ); |
| 2445 |
$result = "<img src=\"" . $url . "\" alt=\"" . $alt_text . "\""; |
| 2446 |
if ( isset( $this->titles[$link_id] ) ) { |
| 2447 |
$title = $this->titles[$link_id]; |
| 2448 |
$title = $this->encodeAttribute( $title ); |
| 2449 |
$result .= " title=\"" . $title . "\""; |
| 2450 |
} |
| 2451 |
if ( isset( $this->ref_attr[$link_id] ) ) |
| 2452 |
$result .= $this->ref_attr[$link_id]; |
| 2453 |
$result .= $this->empty_element_suffix; |
| 2454 |
$result = $this->hashPart( $result ); |
| 2455 |
} else { |
| 2456 |
# If there's no such link ID, leave intact: |
| 2457 |
$result = $whole_match; |
| 2458 |
} |
| 2459 |
|
| 2460 |
return $result; |
| 2461 |
} |
| 2462 |
|
| 2463 |
function _doImages_inline_callback( $matches ) { |
| 2464 |
$whole_match = $matches[1]; |
| 2465 |
$alt_text = $matches[2]; |
| 2466 |
$url = $matches[3] == '' ? $matches[4] : $matches[3]; |
| 2467 |
$title =& $matches[7]; |
| 2468 |
$attr = $this->doExtraAttributes( "img", $dummy =& $matches[8] ); |
| 2469 |
|
| 2470 |
$alt_text = $this->encodeAttribute( $alt_text ); |
| 2471 |
$url = $this->encodeAttribute( $url ); |
| 2472 |
$result = "<img src=\"" . $url . "\" alt=\"" . $alt_text . "\""; |
| 2473 |
if ( isset( $title ) ) { |
| 2474 |
$title = $this->encodeAttribute( $title ); |
| 2475 |
$result .= " title=\"" . $title . "\""; # $title already quoted |
| 2476 |
} |
| 2477 |
$result .= $attr; |
| 2478 |
$result .= $this->empty_element_suffix; |
| 2479 |
|
| 2480 |
return $this->hashPart( $result ); |
| 2481 |
} |
| 2482 |
|
| 2483 |
# Redefined to add id and class attribute support. |
| 2484 |
function doHeaders( $text ) { |
| 2485 |
|
| 2486 |
# Setext-style headers: |
| 2487 |
# Header 1 {#header1} |
| 2488 |
# ======== |
| 2489 |
# |
| 2490 |
# Header 2 {#header2 .class1 .class2} |
| 2491 |
# -------- |
| 2492 |
# |
| 2493 |
$text = preg_replace_callback( |
| 2494 |
'{ |
| 2495 |
(^.+?) # $1: Header text |
| 2496 |
(?:[ ]+ ' . $this->id_class_attr_catch_re . ' )? # $3 = id/class attributes |
| 2497 |
[ ]*\n(=+|-+)[ ]*\n+ # $3: Header footer |
| 2498 |
}mx', |
| 2499 |
array( &$this, '_doHeaders_callback_setext'), |
| 2500 |
$text |
| 2501 |
); |
| 2502 |
|
| 2503 |
# atx-style headers: |
| 2504 |
# # Header 1 {#header1} |
| 2505 |
# ## Header 2 {#header2} |
| 2506 |
# ## Header 2 with closing hashes ## {#header3.class1.class2} |
| 2507 |
# ... |
| 2508 |
# ###### Header 6 {.class2} |
| 2509 |
# |
| 2510 |
$text = preg_replace_callback( '{ |
| 2511 |
^(\#{1,6}) # $1 = string of #\'s |
| 2512 |
[ ]* |
| 2513 |
(.+?) # $2 = Header text |
| 2514 |
[ ]* |
| 2515 |
\#* # optional closing #\'s (not counted) |
| 2516 |
(?:[ ]+ ' . $this->id_class_attr_catch_re . ' )? # $3 = id/class attributes |
| 2517 |
[ ]* |
| 2518 |
\n+ |
| 2519 |
}xm', |
| 2520 |
array( &$this, '_doHeaders_callback_atx'), |
| 2521 |
$text |
| 2522 |
); |
| 2523 |
|
| 2524 |
return $text; |
| 2525 |
} |
| 2526 |
|
| 2527 |
function _doHeaders_callback_setext( $matches ) { |
| 2528 |
if ( $matches[3] == '-' && preg_match( '{^- }', $matches[1] ) ) |
| 2529 |
return $matches[0]; |
| 2530 |
$level = $matches[3][0] == '=' ? 1 : 2; |
| 2531 |
$attr = $this->doExtraAttributes( "h" . $level, $dummy =& $matches[2] ); |
| 2532 |
$block = "<h" . $level . $attr . ">" . $this->runSpanGamut( $matches[1] ) . "</h" . $level . ">"; |
| 2533 |
return "\n" . $this->hashBlock( $block ) . "\n\n"; |
| 2534 |
} |
| 2535 |
|
| 2536 |
function _doHeaders_callback_atx( $matches ) { |
| 2537 |
$level = strlen( $matches[1] ); |
| 2538 |
$attr = $this->doExtraAttributes("h" . $level, $dummy =& $matches[3] ); |
| 2539 |
$block = "<h" . $level . $attr . ">" . $this->runSpanGamut( $matches[2] ) . "</h" . $level . ">"; |
| 2540 |
return "\n" . $this->hashBlock( $block ) . "\n\n"; |
| 2541 |
} |
| 2542 |
|
| 2543 |
# Form HTML tables. |
| 2544 |
function doTables( $text ) { |
| 2545 |
|
| 2546 |
$less_than_tab = $this->tab_width - 1; |
| 2547 |
|
| 2548 |
# Find tables with leading pipe. |
| 2549 |
# |
| 2550 |
# | Header 1 | Header 2 |
| 2551 |
# | -------- | -------- |
| 2552 |
# | Cell 1 | Cell 2 |
| 2553 |
# | Cell 3 | Cell 4 |
| 2554 |
$text = preg_replace_callback(' |
| 2555 |
{ |
| 2556 |
^ # Start of a line |
| 2557 |
[ ]{0,' . $less_than_tab . '} # Allowed whitespace. |
| 2558 |
[|] # Optional leading pipe (present) |
| 2559 |
(.+) \n # $1: Header row (at least one pipe) |
| 2560 |
|
| 2561 |
[ ]{0,' . $less_than_tab . '} # Allowed whitespace. |
| 2562 |
[|] ([ ]*[-:]+[-| :]*) \n # $2: Header underline |
| 2563 |
|
| 2564 |
( # $3: Cells |
| 2565 |
(?> |
| 2566 |
[ ]* # Allowed whitespace. |
| 2567 |
[|] .* \n # Row content. |
| 2568 |
)* |
| 2569 |
) |
| 2570 |
(?=\n|\Z) # Stop at final double newline. |
| 2571 |
}xm', |
| 2572 |
array( &$this, '_doTable_leadingPipe_callback'), |
| 2573 |
$text |
| 2574 |
); |
| 2575 |
|
| 2576 |
# Find tables without leading pipe. |
| 2577 |
# |
| 2578 |
# Header 1 | Header 2 |
| 2579 |
# -------- | -------- |
| 2580 |
# Cell 1 | Cell 2 |
| 2581 |
# Cell 3 | Cell 4 |
| 2582 |
$text = preg_replace_callback( ' |
| 2583 |
{ |
| 2584 |
^ # Start of a line |
| 2585 |
[ ]{0,' . $less_than_tab . '} # Allowed whitespace. |
| 2586 |
(\S.*[|].*) \n # $1: Header row (at least one pipe) |
| 2587 |
|
| 2588 |
[ ]{0,' . $less_than_tab . '} # Allowed whitespace. |
| 2589 |
([-:]+[ ]*[|][-| :]*) \n # $2: Header underline |
| 2590 |
|
| 2591 |
( # $3: Cells |
| 2592 |
(?> |
| 2593 |
.* [|] .* \n # Row content |
| 2594 |
)* |
| 2595 |
) |
| 2596 |
(?=\n|\Z) # Stop at final double newline. |
| 2597 |
}xm', |
| 2598 |
array( &$this, '_DoTable_callback' ), |
| 2599 |
$text |
| 2600 |
); |
| 2601 |
|
| 2602 |
return $text; |
| 2603 |
} |
| 2604 |
|
| 2605 |
function _doTable_leadingPipe_callback( $matches ) { |
| 2606 |
$head = $matches[1]; |
| 2607 |
$underline = $matches[2]; |
| 2608 |
$content = $matches[3]; |
| 2609 |
|
| 2610 |
# Remove leading pipe for each row. |
| 2611 |
$content = preg_replace( '/^ *[|]/m', '', $content ); |
| 2612 |
|
| 2613 |
return $this->_doTable_callback( array( $matches[0], $head, $underline, $content ) ); |
| 2614 |
} |
| 2615 |
|
| 2616 |
function _doTable_callback( $matches ) { |
| 2617 |
$head = $matches[1]; |
| 2618 |
$underline = $matches[2]; |
| 2619 |
$content = $matches[3]; |
| 2620 |
|
| 2621 |
# Remove any tailing pipes for each line. |
| 2622 |
$head = preg_replace( '/[|] *$/m', '', $head ); |
| 2623 |
$underline = preg_replace( '/[|] *$/m', '', $underline ); |
| 2624 |
$content = preg_replace( '/[|] *$/m', '', $content ); |
| 2625 |
|
| 2626 |
# Reading alignement from header underline. |
| 2627 |
$separators = preg_split( '/ *[|] */', $underline ); |
| 2628 |
foreach ( $separators as $n => $s ) { |
| 2629 |
if ( preg_match('/^ *-+: *$/', $s) ) { |
| 2630 |
$attr[$n] = ' align="right"'; |
| 2631 |
} else if ( preg_match('/^ *:-+: *$/', $s ) ) { |
| 2632 |
$attr[$n] = ' align="center"'; |
| 2633 |
} else if ( preg_match('/^ *:-+ *$/', $s ) ) { |
| 2634 |
$attr[$n] = ' align="left"'; |
| 2635 |
} else { |
| 2636 |
$attr[$n] = ''; |
| 2637 |
} |
| 2638 |
} |
| 2639 |
|
| 2640 |
# Parsing span elements, including code spans, character escapes, |
| 2641 |
# and inline HTML tags, so that pipes inside those gets ignored. |
| 2642 |
$head = $this->parseSpan( $head ); |
| 2643 |
$headers = preg_split( '/ *[|] */', $head ); |
| 2644 |
$col_count = count( $headers ); |
| 2645 |
$attr = array_pad( $attr, $col_count, '' ); |
| 2646 |
|
| 2647 |
# Write column headers. |
| 2648 |
$text = "<table>\n"; |
| 2649 |
$text .= "<thead>\n"; |
| 2650 |
$text .= "<tr>\n"; |
| 2651 |
foreach ( $headers as $n => $header ) { |
| 2652 |
$text .= " <th" . $attr[$n] . ">" . $this->runSpanGamut( trim( $header ) ) . "</th>\n"; |
| 2653 |
} |
| 2654 |
$text .= "</tr>\n"; |
| 2655 |
$text .= "</thead>\n"; |
| 2656 |
|
| 2657 |
# Split content by row. |
| 2658 |
$rows = explode( "\n", trim( $content, "\n" ) ); |
| 2659 |
|
| 2660 |
$text .= "<tbody>\n"; |
| 2661 |
foreach ( $rows as $row ) { |
| 2662 |
# Parsing span elements, including code spans, character escapes, |
| 2663 |
# and inline HTML tags, so that pipes inside those gets ignored. |
| 2664 |
$row = $this->parseSpan( $row ); |
| 2665 |
|
| 2666 |
# Split row by cell. |
| 2667 |
$row_cells = preg_split( '/ *[|] */', $row, $col_count ); |
| 2668 |
$row_cells = array_pad( $row_cells, $col_count, '' ); |
| 2669 |
|
| 2670 |
$text .= "<tr>\n"; |
| 2671 |
foreach ( $row_cells as $n => $cell ) { |
| 2672 |
$text .= " <td" . $attr[$n] . ">" . $this->runSpanGamut( trim( $cell ) ) . "</td>\n"; |
| 2673 |
} |
| 2674 |
$text .= "</tr>\n"; |
| 2675 |
} |
| 2676 |
$text .= "</tbody>\n"; |
| 2677 |
$text .= "</table>"; |
| 2678 |
|
| 2679 |
return $this->hashBlock( $text ) . "\n"; |
| 2680 |
} |
| 2681 |
|
| 2682 |
# Form HTML definition lists. |
| 2683 |
function doDefLists( $text ) { |
| 2684 |
|
| 2685 |
$less_than_tab = $this->tab_width - 1; |
| 2686 |
|
| 2687 |
# Re-usable pattern to match any entire dl list: |
| 2688 |
$whole_list_re = '(?> |
| 2689 |
( # $1 = whole list |
| 2690 |
( # $2 |
| 2691 |
[ ]{0,' . $less_than_tab . '} |
| 2692 |
((?>.*\S.*\n)+) # $3 = defined term |
| 2693 |
\n? |
| 2694 |
[ ]{0,' . $less_than_tab . '}:[ ]+ # colon starting definition |
| 2695 |
) |
| 2696 |
(?s:.+?) |
| 2697 |
( # $4 |
| 2698 |
\z |
| 2699 |
| |
| 2700 |
\n{2,} |
| 2701 |
(?=\S) |
| 2702 |
(?! # Negative lookahead for another term |
| 2703 |
[ ]{0,' . $less_than_tab . '} |
| 2704 |
(?: \S.*\n )+? # defined term |
| 2705 |
\n? |
| 2706 |
[ ]{0,' . $less_than_tab . '}:[ ]+ # colon starting definition |
| 2707 |
) |
| 2708 |
(?! # Negative lookahead for another definition |
| 2709 |
[ ]{0,' . $less_than_tab . '}:[ ]+ # colon starting definition |
| 2710 |
) |
| 2711 |
) |
| 2712 |
) |
| 2713 |
)'; // mx |
| 2714 |
|
| 2715 |
$text = preg_replace_callback( '{ |
| 2716 |
(?>\A\n?|(?<=\n\n)) |
| 2717 |
' . $whole_list_re . ' |
| 2718 |
}mx', |
| 2719 |
array( &$this, '_doDefLists_callback' ), |
| 2720 |
$text |
| 2721 |
); |
| 2722 |
|
| 2723 |
return $text; |
| 2724 |
} |
| 2725 |
|
| 2726 |
function _doDefLists_callback( $matches ) { |
| 2727 |
# Re-usable patterns to match list item bullets and number markers: |
| 2728 |
$list = $matches[1]; |
| 2729 |
|
| 2730 |
# Turn double returns into triple returns, so that we can make a |
| 2731 |
# paragraph for the last item in a list, if necessary: |
| 2732 |
$result = trim( $this->processDefListItems( $list ) ); |
| 2733 |
$result = "<dl>\n" . $result . "\n</dl>"; |
| 2734 |
return $this->hashBlock( $result ) . "\n\n"; |
| 2735 |
} |
| 2736 |
|
| 2737 |
# Process the contents of a single definition list, splitting it |
| 2738 |
# into individual term and definition list items. |
| 2739 |
function processDefListItems( $list_str ) { |
| 2740 |
|
| 2741 |
$less_than_tab = $this->tab_width - 1; |
| 2742 |
|
| 2743 |
# trim trailing blank lines: |
| 2744 |
$list_str = preg_replace( "/\n{2,}\\z/", "\n", $list_str ); |
| 2745 |
|
| 2746 |
# Process definition terms. |
| 2747 |
$list_str = preg_replace_callback( '{ |
| 2748 |
(?>\A\n?|\n\n+) # leading line |
| 2749 |
( # definition terms = $1 |
| 2750 |
[ ]{0,' . $less_than_tab . '} # leading whitespace |
| 2751 |
(?!\:[ ]|[ ]) # negative lookahead for a definition |
| 2752 |
# mark (colon) or more whitespace. |
| 2753 |
(?> \S.* \n)+? # actual term (not whitespace). |
| 2754 |
) |
| 2755 |
(?=\n?[ ]{0,3}:[ ]) # lookahead for following line feed |
| 2756 |
# with a definition mark. |
| 2757 |
}xm', |
| 2758 |
array( &$this, '_processDefListItems_callback_dt' ), |
| 2759 |
$list_str |
| 2760 |
); |
| 2761 |
|
| 2762 |
# Process actual definitions. |
| 2763 |
$list_str = preg_replace_callback( '{ |
| 2764 |
\n(\n+)? # leading line = $1 |
| 2765 |
( # marker space = $2 |
| 2766 |
[ ]{0,' . $less_than_tab . '} # whitespace before colon |
| 2767 |
\:[ ]+ # definition mark (colon) |
| 2768 |
) |
| 2769 |
((?s:.+?)) # definition text = $3 |
| 2770 |
(?= \n+ # stop at next definition mark, |
| 2771 |
(?: # next term or end of text |
| 2772 |
[ ]{0,' . $less_than_tab . '} \:[ ] | |
| 2773 |
<dt> | \z |
| 2774 |
) |
| 2775 |
) |
| 2776 |
}xm', |
| 2777 |
array( &$this, '_processDefListItems_callback_dd' ), |
| 2778 |
$list_str |
| 2779 |
); |
| 2780 |
|
| 2781 |
return $list_str; |
| 2782 |
} |
| 2783 |
|
| 2784 |
function _processDefListItems_callback_dt( $matches ) { |
| 2785 |
$terms = explode( "\n", trim( $matches[1] ) ); |
| 2786 |
$text = ''; |
| 2787 |
foreach ( $terms as $term ) { |
| 2788 |
$term = $this->runSpanGamut( trim( $term ) ); |
| 2789 |
$text .= "\n<dt>" . $term . "</dt>"; |
| 2790 |
} |
| 2791 |
return $text . "\n"; |
| 2792 |
} |
| 2793 |
|
| 2794 |
function _processDefListItems_callback_dd( $matches ) { |
| 2795 |
$leading_line = $matches[1]; |
| 2796 |
$marker_space = $matches[2]; |
| 2797 |
$def = $matches[3]; |
| 2798 |
|
| 2799 |
if ( $leading_line || preg_match( '/\n{2,}/', $def ) ) { |
| 2800 |
# Replace marker with the appropriate whitespace indentation |
| 2801 |
$def = str_repeat( ' ', strlen( $marker_space)) . $def; |
| 2802 |
$def = $this->runBlockGamut( $this->outdent( $def . "\n\n" ) ); |
| 2803 |
$def = "\n" . $def ."\n"; |
| 2804 |
} else { |
| 2805 |
$def = rtrim( $def ); |
| 2806 |
$def = $this->runSpanGamut( $this->outdent( $def ) ); |
| 2807 |
} |
| 2808 |
|
| 2809 |
return "\n<dd>" . $def . "</dd>\n"; |
| 2810 |
} |
| 2811 |
|
| 2812 |
# Adding the fenced code block syntax to regular Markdown: |
| 2813 |
function doFencedCodeBlocks( $text ) { |
| 2814 |
|
| 2815 |
# ~~~ |
| 2816 |
# Code block |
| 2817 |
# ~~~ |
| 2818 |
$less_than_tab = $this->tab_width; |
| 2819 |
|
| 2820 |
$text = preg_replace_callback( '{ |
| 2821 |
(?:\n|\A) |
| 2822 |
# 1: Opening marker |
| 2823 |
( |
| 2824 |
(?:~{3,}|`{3,}) # 3 or more tildes/backticks. |
| 2825 |
) |
| 2826 |
[ ]* |
| 2827 |
(?: |
| 2828 |
\.?([-_:a-zA-Z0-9]+) # 2: standalone class name |
| 2829 |
| |
| 2830 |
' . $this->id_class_attr_catch_re . ' # 3: Extra attributes |
| 2831 |
)? |
| 2832 |
[ ]* \n # Whitespace and newline following marker. |
| 2833 |
|
| 2834 |
# 4: Content |
| 2835 |
( |
| 2836 |
(?> |
| 2837 |
(?!\1 [ ]* \n) # Not a closing marker. |
| 2838 |
.*\n+ |
| 2839 |
)+ |
| 2840 |
) |
| 2841 |
|
| 2842 |
# Closing marker. |
| 2843 |
\1 [ ]* (?= \n ) |
| 2844 |
}xm', |
| 2845 |
array( &$this, '_doFencedCodeBlocks_callback' ), |
| 2846 |
$text |
| 2847 |
); |
| 2848 |
|
| 2849 |
return $text; |
| 2850 |
} |
| 2851 |
|
| 2852 |
function _doFencedCodeBlocks_callback( $matches ) { |
| 2853 |
$classname =& $matches[2]; |
| 2854 |
$attrs =& $matches[3]; |
| 2855 |
$codeblock = $matches[4]; |
| 2856 |
$codeblock = htmlspecialchars( $codeblock, ENT_NOQUOTES ); |
| 2857 |
$codeblock = preg_replace_callback( '/^\n+/', array( &$this, '_doFencedCodeBlocks_newlines' ), $codeblock ); |
| 2858 |
|
| 2859 |
if ( $classname != "" ) { |
| 2860 |
if ( $classname[0] == '.' ) { |
| 2861 |
$classname = substr($classname, 1 ); |
| 2862 |
} |
| 2863 |
$attr_str = ' class="' . $this->code_class_prefix.$classname . '"'; |
| 2864 |
} else { |
| 2865 |
$attr_str = $this->doExtraAttributes( $this->code_attr_on_pre ? "pre" : "code", $attrs ); |
| 2866 |
} |
| 2867 |
$pre_attr_str = $this->code_attr_on_pre ? $attr_str : ''; |
| 2868 |
$code_attr_str = $this->code_attr_on_pre ? '' : $attr_str; |
| 2869 |
$codeblock = "<pre" . $pre_attr_str . "><code" . $code_attr_str . ">" . $codeblock . "</code></pre>"; |
| 2870 |
|
| 2871 |
return "\n\n".$this->hashBlock( $codeblock ) . "\n\n"; |
| 2872 |
} |
| 2873 |
|
| 2874 |
function _doFencedCodeBlocks_newlines( $matches ) { |
| 2875 |
return str_repeat( "<br" . $this->empty_element_suffix, strlen( $matches[0] ) ); |
| 2876 |
} |
| 2877 |
|
| 2878 |
# Redefining emphasis markers so that emphasis by underscore does not |
| 2879 |
# work in the middle of a word. |
| 2880 |
public $em_relist = array( |
| 2881 |
'' => '(?:(?<!\*)\*(?!\*)|(?<![a-zA-Z0-9_])_(?!_))(?=\S|$)(?![\.,:;]\s)', |
| 2882 |
'*' => '(?<=\S|^)(?<!\*)\*(?!\*)', |
| 2883 |
'_' => '(?<=\S|^)(?<!_)_(?![a-zA-Z0-9_])', |
| 2884 |
); |
| 2885 |
public $strong_relist = array( |
| 2886 |
'' => '(?:(?<!\*)\*\*(?!\*)|(?<![a-zA-Z0-9_])__(?!_))(?=\S|$)(?![\.,:;]\s)', |
| 2887 |
'**' => '(?<=\S|^)(?<!\*)\*\*(?!\*)', |
| 2888 |
'__' => '(?<=\S|^)(?<!_)__(?![a-zA-Z0-9_])', |
| 2889 |
); |
| 2890 |
public $em_strong_relist = array( |
| 2891 |
'' => '(?:(?<!\*)\*\*\*(?!\*)|(?<![a-zA-Z0-9_])___(?!_))(?=\S|$)(?![\.,:;]\s)', |
| 2892 |
'***' => '(?<=\S|^)(?<!\*)\*\*\*(?!\*)', |
| 2893 |
'___' => '(?<=\S|^)(?<!_)___(?![a-zA-Z0-9_])', |
| 2894 |
); |
| 2895 |
|
| 2896 |
# Params: $text - string to process with html <p> tags |
| 2897 |
function formParagraphs( $text ) { |
| 2898 |
|
| 2899 |
# Strip leading and trailing lines: |
| 2900 |
$text = preg_replace( '/\A\n+|\n+\z/', '', $text ); |
| 2901 |
|
| 2902 |
$grafs = preg_split( '/\n{2,}/', $text, -1, PREG_SPLIT_NO_EMPTY ); |
| 2903 |
|
| 2904 |
# Wrap <p> tags and unhashify HTML blocks |
| 2905 |
foreach ( $grafs as $key => $value ) { |
| 2906 |
$value = trim( $this->runSpanGamut( $value ) ); |
| 2907 |
|
| 2908 |
# Check if this should be enclosed in a paragraph. |
| 2909 |
# Clean tag hashes & block tag hashes are left alone. |
| 2910 |
$is_p = !preg_match( '/^B\x1A[0-9]+B|^C\x1A[0-9]+C$/', $value ); |
| 2911 |
|
| 2912 |
if ( $is_p ) { |
| 2913 |
$value = "<p>$value</p>"; |
| 2914 |
} |
| 2915 |
$grafs[$key] = $value; |
| 2916 |
} |
| 2917 |
|
| 2918 |
# Join grafs in one text, then unhash HTML tags. |
| 2919 |
$text = implode( "\n\n", $grafs ); |
| 2920 |
|
| 2921 |
# Finish by removing any tag hashes still present in $text. |
| 2922 |
$text = $this->unhash( $text ); |
| 2923 |
|
| 2924 |
return $text; |
| 2925 |
} |
| 2926 |
|
| 2927 |
### Footnotes |
| 2928 |
|
| 2929 |
# Strips link definitions from text, stores the URLs and titles in |
| 2930 |
# hash references. |
| 2931 |
function stripFootnotes( $text ) { |
| 2932 |
|
| 2933 |
$less_than_tab = $this->tab_width - 1; |
| 2934 |
|
| 2935 |
# Link defs are in the form: [^id]: url "optional title" |
| 2936 |
$text = preg_replace_callback( '{ |
| 2937 |
^[ ]{0,' . $less_than_tab . '}\[\^(.+?)\][ ]?: # note_id = $1 |
| 2938 |
[ ]* |
| 2939 |
\n? # maybe *one* newline |
| 2940 |
( # text = $2 (no blank lines allowed) |
| 2941 |
(?: |
| 2942 |
.+ # actual text |
| 2943 |
| |
| 2944 |
\n # newlines but |
| 2945 |
(?!\[\^.+?\]:\s)# negative lookahead for footnote marker. |
| 2946 |
(?!\n+[ ]{0,3}\S)# ensure line is not blank and followed |
| 2947 |
# by non-indented content |
| 2948 |
)* |
| 2949 |
) |
| 2950 |
}xm', |
| 2951 |
array( &$this, '_stripFootnotes_callback' ), |
| 2952 |
$text |
| 2953 |
); |
| 2954 |
return $text; |
| 2955 |
} |
| 2956 |
|
| 2957 |
function _stripFootnotes_callback( $matches ) { |
| 2958 |
$note_id = $this->fn_id_prefix . $matches[1]; |
| 2959 |
$this->footnotes[$note_id] = $this->outdent( $matches[2] ); |
| 2960 |
return ''; # String that will replace the block |
| 2961 |
} |
| 2962 |
|
| 2963 |
# Replace footnote references in $text [^id] with a special text-token |
| 2964 |
# which will be replaced by the actual footnote marker in appendFootnotes. |
| 2965 |
function doFootnotes( $text ) { |
| 2966 |
|
| 2967 |
if ( !$this->in_anchor ) { |
| 2968 |
$text = preg_replace( '{\[\^(.+?)\]}', "F\x1Afn:\\1\x1A:", $text ); |
| 2969 |
} |
| 2970 |
return $text; |
| 2971 |
} |
| 2972 |
|
| 2973 |
|
| 2974 |
# Append footnote list to text. |
| 2975 |
function appendFootnotes( $text ) { |
| 2976 |
|
| 2977 |
$text = preg_replace_callback( '{F\x1Afn:(.*?)\x1A:}', array( &$this, '_appendFootnotes_callback' ), $text ); |
| 2978 |
|
| 2979 |
if ( !empty( $this->footnotes_ordered ) ) { |
| 2980 |
$text .= "\n\n"; |
| 2981 |
$text .= "<div class=\"footnotes\">\n"; |
| 2982 |
$text .= "<hr" . $this->empty_element_suffix ."\n"; |
| 2983 |
$text .= "<ol>\n\n"; |
| 2984 |
|
| 2985 |
$attr = ""; |
| 2986 |
if ( $this->fn_backlink_class != "" ) { |
| 2987 |
$class = $this->fn_backlink_class; |
| 2988 |
$class = $this->encodeAttribute( $class ); |
| 2989 |
$attr .= " class=\"" . $class . "\""; |
| 2990 |
} |
| 2991 |
if ( $this->fn_backlink_title != "" ) { |
| 2992 |
$title = $this->fn_backlink_title; |
| 2993 |
$title = $this->encodeAttribute( $title ); |
| 2994 |
$attr .= " title=\"" . $title . "\""; |
| 2995 |
} |
| 2996 |
$num = 0; |
| 2997 |
|
| 2998 |
while ( !empty( $this->footnotes_ordered ) ) { |
| 2999 |
$footnote = reset( $this->footnotes_ordered ); |
| 3000 |
$note_id = key( $this->footnotes_ordered ); |
| 3001 |
unset( $this->footnotes_ordered[$note_id] ); |
| 3002 |
$ref_count = $this->footnotes_ref_count[$note_id]; |
| 3003 |
unset( $this->footnotes_ref_count[$note_id] ); |
| 3004 |
unset( $this->footnotes[$note_id] ); |
| 3005 |
|
| 3006 |
$footnote .= "\n"; # Need to append newline before parsing. |
| 3007 |
$footnote = $this->runBlockGamut( $footnote . "\n"); |
| 3008 |
$footnote = preg_replace_callback( '{F\x1Afn:(.*?)\x1A:}', array( &$this, '_appendFootnotes_callback' ), $footnote ); |
| 3009 |
|
| 3010 |
$attr = str_replace( "%%", ++$num, $attr ); |
| 3011 |
$note_id = $this->encodeAttribute( $note_id ); |
| 3012 |
|
| 3013 |
# Prepare backlink, multiple backlinks if multiple references |
| 3014 |
$backlink = "<a href=\"#fnref:" . $note_id . "\"" . $attr . ">↩</a>"; |
| 3015 |
for ( $ref_num = 2; $ref_num <= $ref_count; ++$ref_num ) { |
| 3016 |
$backlink .= " <a href=\"#fnref" . $ref_num . ":" . $note_id . "\"" . $attr .">↩</a>"; |
| 3017 |
} |
| 3018 |
# Add backlink to last paragraph; create new paragraph if needed. |
| 3019 |
if ( preg_match( '{</p>$}', $footnote ) ) { |
| 3020 |
$footnote = substr( $footnote, 0, -4 ) . " " . $backlink . "</p>"; |
| 3021 |
} else { |
| 3022 |
$footnote .= "\n\n<p>" . $backlink . "</p>"; |
| 3023 |
} |
| 3024 |
|
| 3025 |
$text .= "<li id=\"fn:" . $note_id . "\">\n"; |
| 3026 |
$text .= $footnote . "\n"; |
| 3027 |
$text .= "</li>\n\n"; |
| 3028 |
} |
| 3029 |
|
| 3030 |
$text .= "</ol>\n"; |
| 3031 |
$text .= "</div>"; |
| 3032 |
} |
| 3033 |
return $text; |
| 3034 |
} |
| 3035 |
|
| 3036 |
function _appendFootnotes_callback( $matches ) { |
| 3037 |
$node_id = $this->fn_id_prefix . $matches[1]; |
| 3038 |
|
| 3039 |
# Create footnote marker only if it has a corresponding footnote *and* |
| 3040 |
# the footnote hasn't been used by another marker. |
| 3041 |
if ( isset( $this->footnotes[$node_id] ) ) { |
| 3042 |
$num =& $this->footnotes_numbers[$node_id]; |
| 3043 |
if ( !isset( $num ) ) { |
| 3044 |
# Transfer footnote content to the ordered list and give it its |
| 3045 |
# number |
| 3046 |
$this->footnotes_ordered[$node_id] = $this->footnotes[$node_id]; |
| 3047 |
$this->footnotes_ref_count[$node_id] = 1; |
| 3048 |
$num = $this->footnote_counter++; |
| 3049 |
$ref_count_mark = ''; |
| 3050 |
} else { |
| 3051 |
$ref_count_mark = $this->footnotes_ref_count[$node_id] += 1; |
| 3052 |
} |
| 3053 |
|
| 3054 |
$attr = ""; |
| 3055 |
if ( $this->fn_link_class != "" ) { |
| 3056 |
$class = $this->fn_link_class; |
| 3057 |
$class = $this->encodeAttribute( $class ); |
| 3058 |
$attr .= " class=\"" . $class . "\""; |
| 3059 |
} |
| 3060 |
if ( $this->fn_link_title != "" ) { |
| 3061 |
$title = $this->fn_link_title; |
| 3062 |
$title = $this->encodeAttribute( $title ); |
| 3063 |
$attr .= " title=\"" . $title . "\""; |
| 3064 |
} |
| 3065 |
|
| 3066 |
$attr = str_replace( "%%", $num, $attr ); |
| 3067 |
$node_id = $this->encodeAttribute( $node_id ); |
| 3068 |
|
| 3069 |
return |
| 3070 |
"<sup id=\"fnref" . $ref_count_mark . ":" . $node_id . "\">" . |
| 3071 |
"<a href=\"#fn:" . $node_id . "\"" . $attr . ">" . $num . "</a>" . |
| 3072 |
"</sup>"; |
| 3073 |
} |
| 3074 |
|
| 3075 |
return "[^" . $matches[1] . "]"; |
| 3076 |
} |
| 3077 |
|
| 3078 |
### Abbreviations ### |
| 3079 |
|
| 3080 |
# Strips abbreviations from text, stores titles in hash references. |
| 3081 |
function stripAbbreviations( $text ) { |
| 3082 |
|
| 3083 |
$less_than_tab = $this->tab_width - 1; |
| 3084 |
|
| 3085 |
# Link defs are in the form: [id]*: url "optional title" |
| 3086 |
$text = preg_replace_callback( '{ |
| 3087 |
^[ ]{0,' . $less_than_tab . '}\*\[(.+?)\][ ]?: # abbr_id = $1 |
| 3088 |
(.*) # text = $2 (no blank lines allowed) |
| 3089 |
}xm', |
| 3090 |
array( &$this, '_stripAbbreviations_callback' ), |
| 3091 |
$text |
| 3092 |
); |
| 3093 |
return $text; |
| 3094 |
} |
| 3095 |
|
| 3096 |
function _stripAbbreviations_callback( $matches ) { |
| 3097 |
$abbr_word = $matches[1]; |
| 3098 |
$abbr_desc = $matches[2]; |
| 3099 |
if ( $this->abbr_word_re ) { |
| 3100 |
$this->abbr_word_re .= '|'; |
| 3101 |
} |
| 3102 |
$this->abbr_word_re .= preg_quote( $abbr_word ); |
| 3103 |
$this->abbr_desciptions[$abbr_word] = trim( $abbr_desc ); |
| 3104 |
return ''; # String that will replace the block |
| 3105 |
} |
| 3106 |
|
| 3107 |
# Find defined abbreviations in text and wrap them in <abbr> elements. |
| 3108 |
function doAbbreviations( $text ) { |
| 3109 |
|
| 3110 |
if ( $this->abbr_word_re ) { |
| 3111 |
// cannot use the /x modifier because abbr_word_re may |
| 3112 |
// contain significant spaces: |
| 3113 |
$text = preg_replace_callback( '{'. |
| 3114 |
'(?<![\w\x1A])'. |
| 3115 |
'(?:' . $this->abbr_word_re . ')'. |
| 3116 |
'(?![\w\x1A])'. |
| 3117 |
'}', |
| 3118 |
array( &$this, '_doAbbreviations_callback' ), |
| 3119 |
$text |
| 3120 |
); |
| 3121 |
} |
| 3122 |
return $text; |
| 3123 |
} |
| 3124 |
|
| 3125 |
function _doAbbreviations_callback( $matches ) { |
| 3126 |
$abbr = $matches[0]; |
| 3127 |
if ( isset( $this->abbr_desciptions[$abbr] ) ) { |
| 3128 |
$desc = $this->abbr_desciptions[$abbr]; |
| 3129 |
if ( empty( $desc ) ) { |
| 3130 |
return $this->hashPart( "<abbr>" . $abbr . "</abbr>"); |
| 3131 |
} else { |
| 3132 |
$desc = $this->encodeAttribute( $desc ); |
| 3133 |
return $this->hashPart( "<abbr title=\"" . $desc . "\">" . $abbr . "</abbr>" ); |
| 3134 |
} |
| 3135 |
} else { |
| 3136 |
return $matches[0]; |
| 3137 |
} |
| 3138 |
} |
| 3139 |
|
| 3140 |
} |
| 3141 |
} |
| 3142 |
|
| 3143 |
|
| 3144 |
/* |
| 3145 |
|
| 3146 |
PHP Markdown Extra |
| 3147 |
================== |
| 3148 |
|
| 3149 |
Description |
| 3150 |
----------- |
| 3151 |
|
| 3152 |
This is a PHP port of the original Markdown formatter written in Perl |
| 3153 |
by John Gruber. This special "Extra" version of PHP Markdown features |
| 3154 |
further enhancements to the syntax for making additional constructs |
| 3155 |
such as tables and definition list. |
| 3156 |
|
| 3157 |
Markdown is a text-to-HTML filter; it translates an easy-to-read / |
| 3158 |
easy-to-write structured text format into HTML. Markdown's text format |
| 3159 |
is mostly similar to that of plain text email, and supports features such |
| 3160 |
as headers, *emphasis*, code blocks, blockquotes, and links. |
| 3161 |
|
| 3162 |
Markdown's syntax is designed not as a generic markup language, but |
| 3163 |
specifically to serve as a front-end to (X)HTML. You can use span-level |
| 3164 |
HTML tags anywhere in a Markdown document, and you can use block level |
| 3165 |
HTML tags (like <div> and <table> as well). |
| 3166 |
|
| 3167 |
For more information about Markdown's syntax, see: |
| 3168 |
|
| 3169 |
<http://daringfireball.net/projects/markdown/> |
| 3170 |
|
| 3171 |
|
| 3172 |
Bugs |
| 3173 |
---- |
| 3174 |
|
| 3175 |
To file bug reports please send email to: |
| 3176 |
|
| 3177 |
<michel.fortin@michelf.ca> |
| 3178 |
|
| 3179 |
Please include with your report: (1) the example input; (2) the output you |
| 3180 |
expected; (3) the output Markdown actually produced. |
| 3181 |
|
| 3182 |
|
| 3183 |
Version History |
| 3184 |
--------------- |
| 3185 |
|
| 3186 |
See the readme file for detailed release notes for this version. |
| 3187 |
|
| 3188 |
|
| 3189 |
Copyright and License |
| 3190 |
--------------------- |
| 3191 |
|
| 3192 |
PHP Markdown & Extra |
| 3193 |
Copyright (c) 2004-2013 Michel Fortin |
| 3194 |
<http://michelf.ca/> |
| 3195 |
All rights reserved. |
| 3196 |
|
| 3197 |
Based on Markdown |
| 3198 |
Copyright (c) 2003-2006 John Gruber |
| 3199 |
<http://daringfireball.net/> |
| 3200 |
All rights reserved. |
| 3201 |
|
| 3202 |
Redistribution and use in source and binary forms, with or without |
| 3203 |
modification, are permitted provided that the following conditions are |
| 3204 |
met: |
| 3205 |
|
| 3206 |
* Redistributions of source code must retain the above copyright notice, |
| 3207 |
this list of conditions and the following disclaimer. |
| 3208 |
|
| 3209 |
* Redistributions in binary form must reproduce the above copyright |
| 3210 |
notice, this list of conditions and the following disclaimer in the |
| 3211 |
documentation and/or other materials provided with the distribution. |
| 3212 |
|
| 3213 |
* Neither the name "Markdown" nor the names of its contributors may |
| 3214 |
be used to endorse or promote products derived from this software |
| 3215 |
without specific prior written permission. |
| 3216 |
|
| 3217 |
This software is provided by the copyright holders and contributors "as |
| 3218 |
is" and any express or implied warranties, including, but not limited |
| 3219 |
to, the implied warranties of merchantability and fitness for a |
| 3220 |
particular purpose are disclaimed. In no event shall the copyright owner |
| 3221 |
or contributors be liable for any direct, indirect, incidental, special, |
| 3222 |
exemplary, or consequential damages (including, but not limited to, |
| 3223 |
procurement of substitute goods or services; loss of use, data, or |
| 3224 |
profits; or business interruption) however caused and on any theory of |
| 3225 |
liability, whether in contract, strict liability, or tort (including |
| 3226 |
negligence or otherwise) arising in any way out of the use of this |
| 3227 |
software, even if advised of the possibility of such damage. |
| 3228 |
|
| 3229 |
*/ |
| 3230 |
|
| 3231 |
|
| 3232 |
// ----------------------- |
| 3233 |
// WordPress Readme Parser |
| 3234 |
// ----------------------- |
| 3235 |
|
| 3236 |
// if ( !class_exists( 'WordPress_Readme_Parser' ) ) { |
| 3237 |
if ( !class_exists( 'radio_station_readme_parser' ) ) { |
| 3238 |
class radio_station_readme_parser { |
| 3239 |
|
| 3240 |
function __construct() { |
| 3241 |
// This space intentially blank |
| 3242 |
} |
| 3243 |
|
| 3244 |
function parse_readme( $file ) { |
| 3245 |
$file_contents = @implode( '', @file( $file ) ); |
| 3246 |
return $this->parse_readme_contents( $file_contents ); |
| 3247 |
} |
| 3248 |
|
| 3249 |
function parse_readme_contents( $file_contents ) { |
| 3250 |
$file_contents = str_replace( array( "\r\n", "\r" ), "\n", $file_contents ); |
| 3251 |
$file_contents = trim( $file_contents ); |
| 3252 |
if ( 0 === strpos( $file_contents, "\xEF\xBB\xBF" ) ) { |
| 3253 |
$file_contents = substr( $file_contents, 3 ); |
| 3254 |
} |
| 3255 |
|
| 3256 |
// Markdown transformations |
| 3257 |
$file_contents = preg_replace( "|^###([^#]+)#*?\s*?\n|im", '=$1=' . "\n", $file_contents ); |
| 3258 |
$file_contents = preg_replace( "|^##([^#]+)#*?\s*?\n|im", '==$1==' . "\n", $file_contents ); |
| 3259 |
$file_contents = preg_replace( "|^#([^#]+)#*?\s*?\n|im", '===$1===' . "\n", $file_contents ); |
| 3260 |
|
| 3261 |
// === Plugin Name === |
| 3262 |
// Must be the very first thing. |
| 3263 |
if ( !preg_match('|^===(.*)===|', $file_contents, $_name ) ) { |
| 3264 |
return array(); // require a name |
| 3265 |
} |
| 3266 |
$name = trim( $_name[1], '=' ); |
| 3267 |
$name = $this->sanitize_text( $name ); |
| 3268 |
|
| 3269 |
$file_contents = $this->chop_string( $file_contents, $_name[0] ); |
| 3270 |
|
| 3271 |
// Requires at least: 1.5 |
| 3272 |
if ( preg_match( '|Requires at least:(.*)|i', $file_contents, $_requires_at_least ) ) { |
| 3273 |
$requires_at_least = $this->sanitize_text($_requires_at_least[1]); |
| 3274 |
} else { |
| 3275 |
$requires_at_least = NULL; |
| 3276 |
} |
| 3277 |
|
| 3278 |
// Tested up to: 2.1 |
| 3279 |
if ( preg_match( '|Tested up to:(.*)|i', $file_contents, $_tested_up_to ) ) { |
| 3280 |
$tested_up_to = $this->sanitize_text( $_tested_up_to[1] ); |
| 3281 |
} else { |
| 3282 |
$tested_up_to = NULL; |
| 3283 |
} |
| 3284 |
|
| 3285 |
// Stable tag: 10.4-ride-the-fire-eagle-danger-day |
| 3286 |
if ( preg_match( '|Stable tag:(.*)|i', $file_contents, $_stable_tag ) ) { |
| 3287 |
$stable_tag = $this->sanitize_text( $_stable_tag[1] ); |
| 3288 |
} else { |
| 3289 |
$stable_tag = NULL; // we assume trunk, but don't set it here to tell the difference between specified trunk and default trunk |
| 3290 |
} |
| 3291 |
|
| 3292 |
// Tags: some tag, another tag, we like tags |
| 3293 |
if ( preg_match( '|Tags:(.*)|i', $file_contents, $_tags ) ) { |
| 3294 |
$tags = preg_split('|,[\s]*?|', trim( $_tags[1] ) ); |
| 3295 |
foreach ( array_keys( $tags ) as $t ) { |
| 3296 |
$tags[$t] = $this->sanitize_text( $tags[$t] ); |
| 3297 |
} |
| 3298 |
} else { |
| 3299 |
$tags = array(); |
| 3300 |
} |
| 3301 |
|
| 3302 |
// Contributors: markjaquith, mdawaffe, zefrank |
| 3303 |
$contributors = array(); |
| 3304 |
if ( preg_match( '|Contributors:(.*)|i', $file_contents, $_contributors ) ) { |
| 3305 |
$temp_contributors = preg_split( '|,[\s]*|', trim( $_contributors[1] ) ); |
| 3306 |
foreach ( array_keys( $temp_contributors ) as $c ) { |
| 3307 |
$tmp_sanitized = $this->user_sanitize( $temp_contributors[$c] ); |
| 3308 |
if ( strlen( trim( $tmp_sanitized ) ) > 0 ) { |
| 3309 |
$contributors[$c] = $tmp_sanitized; |
| 3310 |
} |
| 3311 |
unset( $tmp_sanitized ); |
| 3312 |
} |
| 3313 |
} |
| 3314 |
|
| 3315 |
// Donate Link: URL |
| 3316 |
if ( preg_match( '|Donate link:(.*)|i', $file_contents, $_donate_link ) ) { |
| 3317 |
$donate_link = esc_url( $_donate_link[1] ); |
| 3318 |
} else { |
| 3319 |
$donate_link = NULL; |
| 3320 |
} |
| 3321 |
|
| 3322 |
// togs, conts, etc are optional and order shouldn't matter. So we chop them only after we've grabbed their values. |
| 3323 |
foreach ( array( 'tags', 'contributors', 'requires_at_least', 'tested_up_to', 'stable_tag', 'donate_link') as $chop ) { |
| 3324 |
if ( $$chop ) { |
| 3325 |
$_chop = '_' . $chop; |
| 3326 |
$file_contents = $this->chop_string( $file_contents, ${$_chop}[0] ); |
| 3327 |
} |
| 3328 |
} |
| 3329 |
|
| 3330 |
$file_contents = trim( $file_contents ); |
| 3331 |
|
| 3332 |
// short-description fu |
| 3333 |
if ( !preg_match( '/(^(.*?))^[\s]*=+?[\s]*.+?[\s]*=+?/ms', $file_contents, $_short_description ) ) { |
| 3334 |
$_short_description = array( 1 => &$file_contents, 2 => &$file_contents ); |
| 3335 |
} |
| 3336 |
$short_desc_filtered = $this->sanitize_text( $_short_description[2] ); |
| 3337 |
$short_desc_length = strlen( $short_desc_filtered ); |
| 3338 |
$short_description = substr( $short_desc_filtered, 0, 150 ); |
| 3339 |
if ( $short_desc_length > strlen( $short_description ) ) { |
| 3340 |
$truncated = true; |
| 3341 |
} else { |
| 3342 |
$truncated = false; |
| 3343 |
} |
| 3344 |
if ( $_short_description[1] ) { |
| 3345 |
$file_contents = $this->chop_string( $file_contents, $_short_description[1] ); // yes, the [1] is intentional |
| 3346 |
} |
| 3347 |
|
| 3348 |
// == Section == |
| 3349 |
// Break into sections |
| 3350 |
// $_sections[0] will be the title of the first section, $_sections[1] will be the content of the first section |
| 3351 |
// the array alternates from there: title2, content2, title3, content3... and so forth |
| 3352 |
$_sections = preg_split( '/^[\s]*==[\s]*(.+?)[\s]*==/m', $file_contents, -1, PREG_SPLIT_DELIM_CAPTURE|PREG_SPLIT_NO_EMPTY ); |
| 3353 |
|
| 3354 |
$sections = array(); |
| 3355 |
for ( $i = 1; $i <= count($_sections); $i +=2 ) { |
| 3356 |
$_sections[$i] = preg_replace( '/^[\s]*=[\s]+(.+?)[\s]+=/m', '<h4>$1</h4>', $_sections[$i] ); |
| 3357 |
$_sections[$i] = $this->filter_text( $_sections[$i], true ); |
| 3358 |
$title = $this->sanitize_text( $_sections[$i-1] ); |
| 3359 |
$sections[str_replace( ' ', '_', strtolower( $title ) )] = array( 'title' => $title, 'content' => $_sections[$i] ); |
| 3360 |
} |
| 3361 |
|
| 3362 |
|
| 3363 |
// Special sections |
| 3364 |
// This is where we nab our special sections, so we can enforce their order and treat them differently, if needed |
| 3365 |
// upgrade_notice is not a section, but parse it like it is for now |
| 3366 |
$final_sections = array(); |
| 3367 |
foreach ( array( 'description', 'installation', 'frequently_asked_questions', 'screenshots', 'changelog', 'change_log', 'upgrade_notice', 'extra_notes') as $special_section ) { |
| 3368 |
if ( isset( $sections[$special_section] ) ) { |
| 3369 |
$final_sections[$special_section] = $sections[$special_section]['content']; |
| 3370 |
unset( $sections[$special_section] ); |
| 3371 |
} |
| 3372 |
} |
| 3373 |
if ( isset( $final_sections['change_log'] ) && empty( $final_sections['changelog'] ) ) { |
| 3374 |
$final_sections['changelog'] = $final_sections['change_log']; |
| 3375 |
} |
| 3376 |
|
| 3377 |
$final_screenshots = array(); |
| 3378 |
if ( isset( $final_sections['screenshots'] ) ) { |
| 3379 |
preg_match_all( '|<li>(.*?)</li>|s', $final_sections['screenshots'], $screenshots, PREG_SET_ORDER ); |
| 3380 |
if ( $screenshots ) { |
| 3381 |
foreach ( (array) $screenshots as $ss ) { |
| 3382 |
$final_screenshots[] = $ss[1]; |
| 3383 |
} |
| 3384 |
} |
| 3385 |
} |
| 3386 |
|
| 3387 |
// Parse the upgrade_notice section specially: |
| 3388 |
// 1.0 => blah, 1.1 => fnord |
| 3389 |
if ( isset( $final_sections['upgrade_notice'] ) ) { |
| 3390 |
$upgrade_notice = array(); |
| 3391 |
$split = preg_split( '#<h4>(.*?)</h4>#', $final_sections['upgrade_notice'], -1, PREG_SPLIT_DELIM_CAPTURE | PREG_SPLIT_NO_EMPTY ); |
| 3392 |
for ( $i = 0; $i < count( $split ); $i += 2 ) { |
| 3393 |
// modified: use filter_text instead of sanitize text to maintain markup |
| 3394 |
$upgrade_notice[$this->sanitize_text( $split[$i] )] = substr( $this->filter_text( $split[$i + 1] ), 0, 1000 ); |
| 3395 |
} |
| 3396 |
unset( $final_sections['upgrade_notice'] ); |
| 3397 |
} |
| 3398 |
|
| 3399 |
// No description? |
| 3400 |
// No problem... we'll just fall back to the old style of description |
| 3401 |
// We'll even let you use markup this time! |
| 3402 |
$excerpt = false; |
| 3403 |
if ( !isset( $final_sections['description'] ) ) { |
| 3404 |
$final_sections = array_merge( array( 'description' => $this->filter_text( $_short_description[2], true ) ), $final_sections ); |
| 3405 |
$excerpt = true; |
| 3406 |
} |
| 3407 |
|
| 3408 |
// dump the non-special sections into $remaining_content |
| 3409 |
// their order will be determined by their original order in the readme.txt |
| 3410 |
$remaining_content = ''; |
| 3411 |
foreach ( $sections as $s_name => $s_data ) { |
| 3412 |
$remaining_content .= "\n<h3>" . $s_data['title'] . "</h3>\n" . $s_data['content']; |
| 3413 |
} |
| 3414 |
$remaining_content = trim( $remaining_content ); |
| 3415 |
|
| 3416 |
// All done! |
| 3417 |
// $r['tags'] and $r['contributors'] are simple arrays |
| 3418 |
// $r['sections'] is an array with named elements |
| 3419 |
$r = array( |
| 3420 |
'name' => $name, |
| 3421 |
'tags' => $tags, |
| 3422 |
'requires_at_least' => $requires_at_least, |
| 3423 |
'tested_up_to' => $tested_up_to, |
| 3424 |
'stable_tag' => $stable_tag, |
| 3425 |
'contributors' => $contributors, |
| 3426 |
'donate_link' => $donate_link, |
| 3427 |
'short_description' => $short_description, |
| 3428 |
'screenshots' => $final_screenshots, |
| 3429 |
'is_excerpt' => $excerpt, |
| 3430 |
'is_truncated' => $truncated, |
| 3431 |
'sections' => $final_sections, |
| 3432 |
'remaining_content' => $remaining_content, |
| 3433 |
'upgrade_notice' => $upgrade_notice |
| 3434 |
); |
| 3435 |
|
| 3436 |
return $r; |
| 3437 |
} |
| 3438 |
|
| 3439 |
function chop_string( $string, $chop ) { // chop a "prefix" from a string: Agressive! uses strstr not 0 === strpos |
| 3440 |
if ( $_string = strstr( $string, $chop ) ) { |
| 3441 |
$_string = substr( $_string, strlen( $chop ) ); |
| 3442 |
return trim( $_string ); |
| 3443 |
} else { |
| 3444 |
return trim( $string ); |
| 3445 |
} |
| 3446 |
} |
| 3447 |
|
| 3448 |
function user_sanitize( $text, $strict = false ) { // whitelisted chars |
| 3449 |
// if ( function_exists( 'user_sanitize' ) ) { // bbPress native |
| 3450 |
// return user_sanitize( $text, $strict ); |
| 3451 |
// } |
| 3452 |
|
| 3453 |
if ( $strict ) { |
| 3454 |
$text = preg_replace('/[^a-z0-9-]/i', '', $text ); |
| 3455 |
$text = preg_replace('|-+|', '-', $text ); |
| 3456 |
} else { |
| 3457 |
$text = preg_replace('/[^a-z0-9_-]/i', '', $text ); |
| 3458 |
} |
| 3459 |
return $text; |
| 3460 |
} |
| 3461 |
|
| 3462 |
function sanitize_text( $text ) { // not fancy |
| 3463 |
// $text = strip_tags( $text ); |
| 3464 |
$text = wp_strip_all_tags( $text ); |
| 3465 |
$text = esc_html( $text ); |
| 3466 |
$text = trim( $text ); |
| 3467 |
return $text; |
| 3468 |
} |
| 3469 |
|
| 3470 |
function filter_text( $text, $markdown = false ) { // fancy, Markdown |
| 3471 |
$text = trim( $text ); |
| 3472 |
|
| 3473 |
$text = call_user_func( array( __CLASS__, 'code_trick' ), $text, $markdown ); // A better parser than Markdown's for: backticks -> CODE |
| 3474 |
|
| 3475 |
if ( $markdown ) { // Parse markdown. |
| 3476 |
// if ( !function_exists( 'Markdown' ) ) |
| 3477 |
// require WORDPRESS_README_MARKDOWN; |
| 3478 |
// $text = Markdown( $text ); |
| 3479 |
$text = radio_station_markdown( $text ); |
| 3480 |
} |
| 3481 |
|
| 3482 |
$allowed = array( |
| 3483 |
'a' => array( |
| 3484 |
'href' => array(), |
| 3485 |
'title' => array(), |
| 3486 |
'rel' => array()), |
| 3487 |
'blockquote' => array('cite' => array()), |
| 3488 |
'br' => array(), |
| 3489 |
'p' => array(), |
| 3490 |
'code' => array(), |
| 3491 |
'pre' => array(), |
| 3492 |
'em' => array(), |
| 3493 |
'strong' => array(), |
| 3494 |
'ul' => array(), |
| 3495 |
'ol' => array(), |
| 3496 |
'li' => array(), |
| 3497 |
'h3' => array(), |
| 3498 |
'h4' => array() |
| 3499 |
); |
| 3500 |
|
| 3501 |
$text = balanceTags( $text ); |
| 3502 |
|
| 3503 |
$text = wp_kses( $text, $allowed ); |
| 3504 |
$text = trim( $text ); |
| 3505 |
return $text; |
| 3506 |
} |
| 3507 |
|
| 3508 |
function code_trick( $text, $markdown ) { |
| 3509 |
// Don't use bbPress native function - it's incompatible with Markdown |
| 3510 |
// If doing markdown, first take any user formatted code blocks and turn them into backticks so that |
| 3511 |
// markdown will preserve things like underscores in code blocks |
| 3512 |
if ( $markdown ) { |
| 3513 |
$text = preg_replace_callback( "!(<pre><code>|<code>)(.*?)(</code></pre>|</code>)!s", array( __CLASS__,'decodeit' ), $text ); |
| 3514 |
} |
| 3515 |
|
| 3516 |
$text = str_replace(array("\r\n", "\r"), "\n", $text); |
| 3517 |
if ( !$markdown ) { |
| 3518 |
// This gets the "inline" code blocks, but can't be used with Markdown. |
| 3519 |
$text = preg_replace_callback("|(`)(.*?)`|", array( __CLASS__, 'encodeit'), $text); |
| 3520 |
// This gets the "block level" code blocks and converts them to PRE CODE |
| 3521 |
$text = preg_replace_callback("!(^|\n)`(.*?)`!s", array( __CLASS__, 'encodeit'), $text); |
| 3522 |
} else { |
| 3523 |
// Markdown can do inline code, we convert bbPress style block level code to Markdown style |
| 3524 |
$text = preg_replace_callback("!(^|\n)([ \t]*?)`(.*?)`!s", array( __CLASS__, 'indent'), $text); |
| 3525 |
} |
| 3526 |
return $text; |
| 3527 |
} |
| 3528 |
|
| 3529 |
function indent( $matches ) { |
| 3530 |
$text = $matches[3]; |
| 3531 |
$text = preg_replace('|^|m', $matches[2] . ' ', $text); |
| 3532 |
return $matches[1] . $text; |
| 3533 |
} |
| 3534 |
|
| 3535 |
function encodeit( $matches ) { |
| 3536 |
// if ( function_exists('encodeit') ) { // bbPress native |
| 3537 |
// return encodeit( $matches ); |
| 3538 |
// } |
| 3539 |
|
| 3540 |
$text = trim( $matches[2] ); |
| 3541 |
$text = htmlspecialchars( $text, ENT_QUOTES ); |
| 3542 |
$text = str_replace(array( "\r\n", "\r"), "\n", $text ); |
| 3543 |
$text = preg_replace("|\n\n\n+|", "\n\n", $text ); |
| 3544 |
$text = str_replace( '&lt;', '<', $text ); |
| 3545 |
$text = str_replace( '&gt;', '>', $text ); |
| 3546 |
$text = "<code>" . $text . "</code>"; |
| 3547 |
if ( "`" != $matches[1] ) { |
| 3548 |
$text = "<pre>" . $text . "</pre>"; |
| 3549 |
} |
| 3550 |
return $text; |
| 3551 |
} |
| 3552 |
|
| 3553 |
function decodeit( $matches ) { |
| 3554 |
// if ( function_exists( 'decodeit' ) ) { // bbPress native |
| 3555 |
// return decodeit( $matches ); |
| 3556 |
// } |
| 3557 |
|
| 3558 |
$text = $matches[2]; |
| 3559 |
$trans_table = array_flip( get_html_translation_table( HTML_ENTITIES ) ); |
| 3560 |
$text = strtr( $text, $trans_table ); |
| 3561 |
$text = str_replace( '<br />', '', $text ); |
| 3562 |
$text = str_replace( '&', '&', $text ); |
| 3563 |
$text = str_replace( ''', "'", $text ); |
| 3564 |
if ( '<pre><code>' == $matches[1] ) { |
| 3565 |
$text = "\n" . $text . "\n"; |
| 3566 |
} |
| 3567 |
return "`" . $text . "`"; |
| 3568 |
} |
| 3569 |
|
| 3570 |
} // end class |
| 3571 |
} |
| 3572 |
|
| 3573 |
/** |
| 3574 |
* GitHub-Flavoured Markdown. Inspired by Evan's plugin, but modified. |
| 3575 |
* |
| 3576 |
* @author Evan Solomon |
| 3577 |
* @author Matt Wiebe <wiebe@automattic.com> |
| 3578 |
* @link https://github.com/evansolomon/wp-github-flavored-markdown-comments |
| 3579 |
* |
| 3580 |
* Add a few extras from GitHub's Markdown implementation. Must be used in a WordPress environment. |
| 3581 |
*/ |
| 3582 |
|
| 3583 |
// if ( !class_exists( 'GHF_Markdown_Parser' ) ) { |
| 3584 |
if ( !class_exists( 'radio_station_github_markdown_parser' ) ) { |
| 3585 |
class radio_station_github_markdown_parser extends radio_station_markdown_extra_parser { |
| 3586 |
|
| 3587 |
/** |
| 3588 |
* Hooray somewhat arbitrary numbers that are fearful of 1.0.x. |
| 3589 |
*/ |
| 3590 |
// const GHF_MARDOWN_VERSION = '0.9.0'; |
| 3591 |
|
| 3592 |
/** |
| 3593 |
* Use a [code] shortcode when encountering a fenced code block |
| 3594 |
* @var boolean |
| 3595 |
*/ |
| 3596 |
public $use_code_shortcode = true; |
| 3597 |
|
| 3598 |
/** |
| 3599 |
* Preserve shortcodes, untouched by Markdown. |
| 3600 |
* This requires use within a WordPress installation. |
| 3601 |
* @var boolean |
| 3602 |
*/ |
| 3603 |
public $preserve_shortcodes = true; |
| 3604 |
|
| 3605 |
/** |
| 3606 |
* Preserve the legacy $latex your-latex-code-here$ style |
| 3607 |
* LaTeX markup |
| 3608 |
*/ |
| 3609 |
public $preserve_latex = true; |
| 3610 |
|
| 3611 |
/** |
| 3612 |
* Preserve single-line <code> blocks. |
| 3613 |
* @var boolean |
| 3614 |
*/ |
| 3615 |
public $preserve_inline_code_blocks = true; |
| 3616 |
|
| 3617 |
/** |
| 3618 |
* Strip paragraphs from the output. This is the right default for WordPress, |
| 3619 |
* which generally wants to create its own paragraphs with `wpautop` |
| 3620 |
* @var boolean |
| 3621 |
*/ |
| 3622 |
public $strip_paras = true; |
| 3623 |
|
| 3624 |
// Will run through sprintf - you can supply your own syntax if you want |
| 3625 |
public $shortcode_start = '[code lang=%s]'; |
| 3626 |
public $shortcode_end = '[/code]'; |
| 3627 |
|
| 3628 |
// Stores shortcodes we remove and then replace |
| 3629 |
protected $preserve_text_hash = array(); |
| 3630 |
|
| 3631 |
/** |
| 3632 |
* Set environment defaults based on presence of key functions/classes. |
| 3633 |
*/ |
| 3634 |
public function __construct() { |
| 3635 |
$this->use_code_shortcode = class_exists( 'SyntaxHighlighter' ); |
| 3636 |
/** |
| 3637 |
* Allow processing shortcode contents. |
| 3638 |
* |
| 3639 |
* @module markdown |
| 3640 |
* |
| 3641 |
* @since 4.4.0 |
| 3642 |
* |
| 3643 |
* @param boolean $preserve_shortcodes Defaults to $this->preserve_shortcodes. |
| 3644 |
*/ |
| 3645 |
$this->preserve_shortcodes = apply_filters( 'jetpack_markdown_preserve_shortcodes', $this->preserve_shortcodes ) && function_exists( 'get_shortcode_regex' ); |
| 3646 |
$this->preserve_latex = function_exists( 'latex_markup' ); |
| 3647 |
$this->strip_paras = function_exists( 'wpautop' ); |
| 3648 |
|
| 3649 |
parent::__construct(); |
| 3650 |
} |
| 3651 |
|
| 3652 |
/** |
| 3653 |
* Overload to specify heading styles only if the hash has space(s) after it. This is actually in keeping with |
| 3654 |
* the documentation and eases the semantic overload of the hash character. |
| 3655 |
* #Will Not Produce a Heading 1 |
| 3656 |
* # This Will Produce a Heading 1 |
| 3657 |
* |
| 3658 |
* @param string $text Markdown text |
| 3659 |
* @return string HTML-transformed text |
| 3660 |
*/ |
| 3661 |
public function transform( $text ) { |
| 3662 |
// Preserve anything inside a single-line <code> element |
| 3663 |
if ( $this->preserve_inline_code_blocks ) { |
| 3664 |
$text = $this->single_line_code_preserve( $text ); |
| 3665 |
} |
| 3666 |
// Remove all shortcodes so their interiors are left intact |
| 3667 |
if ( $this->preserve_shortcodes ) { |
| 3668 |
$text = $this->shortcode_preserve( $text ); |
| 3669 |
} |
| 3670 |
// Remove legacy LaTeX so it's left intact |
| 3671 |
if ( $this->preserve_latex ) { |
| 3672 |
$text = $this->latex_preserve( $text ); |
| 3673 |
} |
| 3674 |
|
| 3675 |
// Do not process characters inside URLs. |
| 3676 |
$text = $this->urls_preserve( $text ); |
| 3677 |
|
| 3678 |
// escape line-beginning # chars that do not have a space after them. |
| 3679 |
$text = preg_replace_callback( '|^#{1,6}( )?|um', array( $this, '_doEscapeForHashWithoutSpacing' ), $text ); |
| 3680 |
|
| 3681 |
/** |
| 3682 |
* Allow third-party plugins to define custom patterns that won't be processed by Markdown. |
| 3683 |
* |
| 3684 |
* @module markdown |
| 3685 |
* |
| 3686 |
* @since 3.9.2 |
| 3687 |
* |
| 3688 |
* @param array $custom_patterns Array of custom patterns to be ignored by Markdown. |
| 3689 |
*/ |
| 3690 |
$custom_patterns = apply_filters( 'jetpack_markdown_preserve_pattern', array() ); |
| 3691 |
if ( is_array( $custom_patterns ) && ! empty( $custom_patterns ) ) { |
| 3692 |
foreach ( $custom_patterns as $pattern ) { |
| 3693 |
$text = preg_replace_callback( $pattern, array( $this, '_doRemoveText'), $text ); |
| 3694 |
} |
| 3695 |
} |
| 3696 |
|
| 3697 |
// run through core Markdown |
| 3698 |
$text = parent::transform( $text ); |
| 3699 |
|
| 3700 |
// Occasionally Markdown Extra chokes on a para structure, producing odd paragraphs. |
| 3701 |
$text = str_replace( "<p><</p>\n\n<p>p>", '<p>', $text ); |
| 3702 |
|
| 3703 |
// put start-of-line # chars back in place |
| 3704 |
$text = $this->restore_leading_hash( $text ); |
| 3705 |
|
| 3706 |
// Strip paras if set |
| 3707 |
if ( $this->strip_paras ) { |
| 3708 |
$text = $this->unp( $text ); |
| 3709 |
} |
| 3710 |
|
| 3711 |
// Restore preserved things like shortcodes/LaTeX |
| 3712 |
$text = $this->do_restore( $text ); |
| 3713 |
|
| 3714 |
return $text; |
| 3715 |
} |
| 3716 |
|
| 3717 |
/** |
| 3718 |
* Prevents blocks like <code>__this__</code> from turning into <code><strong>this</strong></code> |
| 3719 |
* @param string $text Text that may need preserving |
| 3720 |
* @return string Text that was preserved if needed |
| 3721 |
*/ |
| 3722 |
public function single_line_code_preserve( $text ) { |
| 3723 |
return preg_replace_callback( '|<code\b[^>]*>(.*?)</code>|', array( $this, 'do_single_line_code_preserve' ), $text ); |
| 3724 |
} |
| 3725 |
|
| 3726 |
/** |
| 3727 |
* Regex callback for inline code presevation |
| 3728 |
* @param array $matches Regex matches |
| 3729 |
* @return string Hashed content for later restoration |
| 3730 |
*/ |
| 3731 |
public function do_single_line_code_preserve( $matches ) { |
| 3732 |
return '<code>' . $this->hash_block( $matches[1] ) . '</code>'; |
| 3733 |
} |
| 3734 |
|
| 3735 |
/** |
| 3736 |
* Preserve code block contents by HTML encoding them. Useful before getting to KSES stripping. |
| 3737 |
* @param string $text Markdown/HTML content |
| 3738 |
* @return string Markdown/HTML content with escaped code blocks |
| 3739 |
*/ |
| 3740 |
public function codeblock_preserve( $text ) { |
| 3741 |
return preg_replace_callback( "/^([`~]{3})([^`\n]+)?\n([^`~]+)(\\1)/m", array( $this, 'do_codeblock_preserve' ), $text ); |
| 3742 |
} |
| 3743 |
|
| 3744 |
/** |
| 3745 |
* Regex callback for code block preservation. |
| 3746 |
* @param array $matches Regex matches |
| 3747 |
* @return string Codeblock with escaped interior |
| 3748 |
*/ |
| 3749 |
public function do_codeblock_preserve( $matches ) { |
| 3750 |
$block = stripslashes( $matches[3] ); |
| 3751 |
$block = esc_html( $block ); |
| 3752 |
$block = str_replace( '\\', '\\\\', $block ); |
| 3753 |
$open = $matches[1] . $matches[2] . "\n"; |
| 3754 |
return $open . $block . $matches[4]; |
| 3755 |
} |
| 3756 |
|
| 3757 |
/** |
| 3758 |
* Restore previously preserved (i.e. escaped) code block contents. |
| 3759 |
* @param string $text Markdown/HTML content with escaped code blocks |
| 3760 |
* @return string Markdown/HTML content |
| 3761 |
*/ |
| 3762 |
public function codeblock_restore( $text ) { |
| 3763 |
return preg_replace_callback( "/^([`~]{3})([^`\n]+)?\n([^`~]+)(\\1)/m", array( $this, 'do_codeblock_restore' ), $text ); |
| 3764 |
} |
| 3765 |
|
| 3766 |
/** |
| 3767 |
* Regex callback for code block restoration (unescaping). |
| 3768 |
* @param array $matches Regex matches |
| 3769 |
* @return string Codeblock with unescaped interior |
| 3770 |
*/ |
| 3771 |
public function do_codeblock_restore( $matches ) { |
| 3772 |
$block = html_entity_decode( $matches[3], ENT_QUOTES ); |
| 3773 |
$open = $matches[1] . $matches[2] . "\n"; |
| 3774 |
return $open . $block . $matches[4]; |
| 3775 |
} |
| 3776 |
|
| 3777 |
/** |
| 3778 |
* Called to preserve legacy LaTeX like $latex some-latex-text $ |
| 3779 |
* @param string $text Text in which to preserve LaTeX |
| 3780 |
* @return string Text with LaTeX replaced by a hash that will be restored later |
| 3781 |
*/ |
| 3782 |
protected function latex_preserve( $text ) { |
| 3783 |
// regex from latex_remove() |
| 3784 |
$regex = '% |
| 3785 |
\$latex(?:=\s*|\s+) |
| 3786 |
((?: |
| 3787 |
[^$]+ # Not a dollar |
| 3788 |
| |
| 3789 |
(?<=(?<!\\\\)\\\\)\$ # Dollar preceded by exactly one slash |
| 3790 |
)+) |
| 3791 |
(?<!\\\\)\$ # Dollar preceded by zero slashes |
| 3792 |
%ix'; |
| 3793 |
$text = preg_replace_callback( $regex, array( $this, '_doRemoveText'), $text ); |
| 3794 |
return $text; |
| 3795 |
} |
| 3796 |
|
| 3797 |
/** |
| 3798 |
* Called to preserve WP shortcodes from being formatted by Markdown in any way. |
| 3799 |
* @param string $text Text in which to preserve shortcodes |
| 3800 |
* @return string Text with shortcodes replaced by a hash that will be restored later |
| 3801 |
*/ |
| 3802 |
protected function shortcode_preserve( $text ) { |
| 3803 |
$text = preg_replace_callback( $this->get_shortcode_regex(), array( $this, '_doRemoveText' ), $text ); |
| 3804 |
return $text; |
| 3805 |
} |
| 3806 |
|
| 3807 |
/** |
| 3808 |
* Avoid characters inside URLs from being formatted by Markdown in any way. |
| 3809 |
* |
| 3810 |
* @param string $text Text in which to preserve URLs. |
| 3811 |
* |
| 3812 |
* @return string Text with URLs replaced by a hash that will be restored later. |
| 3813 |
*/ |
| 3814 |
protected function urls_preserve( $text ) { |
| 3815 |
$text = preg_replace_callback( |
| 3816 |
'#(?<!<)(?:https?|ftp)://([^\s<>"\'\[\]()]+|\[(?1)*+\]|\((?1)*+\))+(?<![_*.?])#i', |
| 3817 |
array( $this, '_doRemoveText' ), |
| 3818 |
$text |
| 3819 |
); |
| 3820 |
return $text; |
| 3821 |
} |
| 3822 |
|
| 3823 |
/** |
| 3824 |
* Restores any text preserved by $this->hash_block() |
| 3825 |
* @param string $text Text that may have hashed preservation placeholders |
| 3826 |
* @return string Text with hashed preseravtion placeholders replaced by original text |
| 3827 |
*/ |
| 3828 |
protected function do_restore( $text ) { |
| 3829 |
// Reverse hashes to ensure nested blocks are restored. |
| 3830 |
$hashes = array_reverse( $this->preserve_text_hash, true ); |
| 3831 |
foreach( $hashes as $hash => $value ) { |
| 3832 |
$placeholder = $this->hash_maker( $hash ); |
| 3833 |
$text = str_replace( $placeholder, $value, $text ); |
| 3834 |
} |
| 3835 |
// reset the hash |
| 3836 |
$this->preserve_text_hash = array(); |
| 3837 |
return $text; |
| 3838 |
} |
| 3839 |
|
| 3840 |
/** |
| 3841 |
* Regex callback for text preservation |
| 3842 |
* @param array $m Regex $matches array |
| 3843 |
* @return string A placeholder that will later be replaced by the original text |
| 3844 |
*/ |
| 3845 |
protected function _doRemoveText( $m ) { |
| 3846 |
return $this->hash_block( $m[0] ); |
| 3847 |
} |
| 3848 |
|
| 3849 |
/** |
| 3850 |
* Call this to store a text block for later restoration. |
| 3851 |
* @param string $text Text to preserve for later |
| 3852 |
* @return string Placeholder that will be swapped out later for the original text |
| 3853 |
*/ |
| 3854 |
protected function hash_block( $text ) { |
| 3855 |
$hash = md5( $text ); |
| 3856 |
$this->preserve_text_hash[ $hash ] = $text; |
| 3857 |
$placeholder = $this->hash_maker( $hash ); |
| 3858 |
return $placeholder; |
| 3859 |
} |
| 3860 |
|
| 3861 |
/** |
| 3862 |
* Less glamorous than the Keymaker |
| 3863 |
* @param string $hash An md5 hash |
| 3864 |
* @return string A placeholder hash |
| 3865 |
*/ |
| 3866 |
protected function hash_maker( $hash ) { |
| 3867 |
return 'MARKDOWN_HASH' . $hash . 'MARKDOWN_HASH'; |
| 3868 |
} |
| 3869 |
|
| 3870 |
/** |
| 3871 |
* Remove bare <p> elements. <p>s with attributes will be preserved. |
| 3872 |
* @param string $text HTML content |
| 3873 |
* @return string <p>-less content |
| 3874 |
*/ |
| 3875 |
public function unp( $text ) { |
| 3876 |
return preg_replace( "#<p>(.*?)</p>(\n|$)#ums", '$1$2', $text ); |
| 3877 |
} |
| 3878 |
|
| 3879 |
/** |
| 3880 |
* A regex of all shortcodes currently registered by the current |
| 3881 |
* WordPress installation |
| 3882 |
* @uses get_shortcode_regex() |
| 3883 |
* @return string A regex for grabbing shortcodes. |
| 3884 |
*/ |
| 3885 |
protected function get_shortcode_regex() { |
| 3886 |
$pattern = get_shortcode_regex(); |
| 3887 |
|
| 3888 |
// don't match markdown link anchors that could be mistaken for shortcodes. |
| 3889 |
$pattern .= '(?!\()'; |
| 3890 |
|
| 3891 |
return "/" . $pattern . "/s"; |
| 3892 |
} |
| 3893 |
|
| 3894 |
/** |
| 3895 |
* Since we escape unspaced #Headings, put things back later. |
| 3896 |
* @param string $text text with a leading escaped hash |
| 3897 |
* @return string text with leading hashes unescaped |
| 3898 |
*/ |
| 3899 |
protected function restore_leading_hash( $text ) { |
| 3900 |
return preg_replace( "/^(<p>)?(#|\\\\#)/um", "$1#", $text ); |
| 3901 |
} |
| 3902 |
|
| 3903 |
/** |
| 3904 |
* Overload to support ```-fenced code blocks for pre-Markdown Extra 1.2.8 |
| 3905 |
* https://help.github.com/articles/github-flavored-markdown#fenced-code-blocks |
| 3906 |
*/ |
| 3907 |
public function doFencedCodeBlocks( $text ) { |
| 3908 |
// If we're at least at 1.2.8, native fenced code blocks are in. |
| 3909 |
// Below is just copied from it in case we somehow got loaded on |
| 3910 |
// top of someone else's Markdown Extra |
| 3911 |
|
| 3912 |
// if ( version_compare( MARKDOWNEXTRA_VERSION, '1.2.8', '>=' ) ) |
| 3913 |
return parent::doFencedCodeBlocks( $text ); |
| 3914 |
|
| 3915 |
# |
| 3916 |
# Adding the fenced code block syntax to regular Markdown: |
| 3917 |
# |
| 3918 |
# ~~~ |
| 3919 |
# Code block |
| 3920 |
# ~~~ |
| 3921 |
# |
| 3922 |
$less_than_tab = $this->tab_width; |
| 3923 |
|
| 3924 |
$text = preg_replace_callback( '{ |
| 3925 |
(?:\n|\A) |
| 3926 |
# 1: Opening marker |
| 3927 |
( |
| 3928 |
(?:~{3,}|`{3,}) # 3 or more tildes/backticks. |
| 3929 |
) |
| 3930 |
[ ]* |
| 3931 |
(?: |
| 3932 |
\.?([-_:a-zA-Z0-9]+) # 2: standalone class name |
| 3933 |
| |
| 3934 |
' . $this->id_class_attr_catch_re . ' # 3: Extra attributes |
| 3935 |
)? |
| 3936 |
[ ]* \n # Whitespace and newline following marker. |
| 3937 |
|
| 3938 |
# 4: Content |
| 3939 |
( |
| 3940 |
(?> |
| 3941 |
(?!\1 [ ]* \n) # Not a closing marker. |
| 3942 |
.*\n+ |
| 3943 |
)+ |
| 3944 |
) |
| 3945 |
|
| 3946 |
# Closing marker. |
| 3947 |
\1 [ ]* (?= \n ) |
| 3948 |
}xm', |
| 3949 |
array( $this, '_doFencedCodeBlocks_callback' ), |
| 3950 |
$text |
| 3951 |
); |
| 3952 |
|
| 3953 |
return $text; |
| 3954 |
} |
| 3955 |
|
| 3956 |
/** |
| 3957 |
* Callback for pre-processing start of line hashes to slyly escape headings that don't |
| 3958 |
* have a leading space |
| 3959 |
* @param array $m preg_match matches |
| 3960 |
* @return string possibly escaped start of line hash |
| 3961 |
*/ |
| 3962 |
public function _doEscapeForHashWithoutSpacing( $m ) { |
| 3963 |
if ( !isset( $m[1] ) ) { |
| 3964 |
$m[0] = '\\' . $m[0]; |
| 3965 |
} |
| 3966 |
return $m[0]; |
| 3967 |
} |
| 3968 |
|
| 3969 |
/** |
| 3970 |
* Overload to support Viper's [code] shortcode. Because awesome. |
| 3971 |
*/ |
| 3972 |
public function _doFencedCodeBlocks_callback( $matches ) { |
| 3973 |
// in case we have some escaped leading hashes right at the start of the block |
| 3974 |
$matches[4] = $this->restore_leading_hash( $matches[4] ); |
| 3975 |
// just MarkdownExtra_Parser if we're not going ultra-deluxe |
| 3976 |
if ( !$this->use_code_shortcode ) { |
| 3977 |
return parent::_doFencedCodeBlocks_callback( $matches ); |
| 3978 |
} |
| 3979 |
|
| 3980 |
// default to a "text" class if one wasn't passed. Helps with encoding issues later. |
| 3981 |
if ( empty( $matches[2] ) ) { |
| 3982 |
$matches[2] = 'text'; |
| 3983 |
} |
| 3984 |
|
| 3985 |
$classname =& $matches[2]; |
| 3986 |
$codeblock = preg_replace_callback('/^\n+/', array( $this, '_doFencedCodeBlocks_newlines' ), $matches[4] ); |
| 3987 |
|
| 3988 |
if ( $classname[0] == '.' ) { |
| 3989 |
$classname = substr( $classname, 1 ); |
| 3990 |
} |
| 3991 |
|
| 3992 |
$codeblock = esc_html( $codeblock ); |
| 3993 |
$codeblock = sprintf( $this->shortcode_start, $classname ) . "\n{$codeblock}" . $this->shortcode_end; |
| 3994 |
return "\n\n" . $this->hashBlock( $codeblock ). "\n\n"; |
| 3995 |
} |
| 3996 |
|
| 3997 |
} |
| 3998 |
} |
| 3999 |
|
| 4000 |
|
| 4001 |
// ========= |
| 4002 |
// CHANGELOG |
| 4003 |
// ========= |
| 4004 |
|
| 4005 |
// == 1.3.1 == |
| 4006 |
// - Updated: Prefixed all classes and functions |
| 4007 |
// - Cleaned: Coding (WPCS) and comment formatting |
| 4008 |
|
| 4009 |
// == 1.1.8 == |
| 4010 |
// - Added: Github Flavoured Reademe Parser |
| 4011 |
// - Added: class_exists wrapper checks |
| 4012 |
|
| 4013 |
// == 1.0,7 == |
| 4014 |
// - Added: function_exists and already defined checks |
| 4015 |
// - Changed: filename from readme.php to reader.PHP |
| 4016 |
|
| 4017 |
|
| 4018 |
|
| 4019 |
|
