| blob | b51ef38a724d631ee08b0a8ba3493022a48f8c14 |
1 <?php
2 /**
3 * PHP parser that converts wiki markup to HTML.
4 *
5 * This program is free software; you can redistribute it and/or modify
6 * it under the terms of the GNU General Public License as published by
7 * the Free Software Foundation; either version 2 of the License, or
8 * (at your option) any later version.
9 *
10 * This program is distributed in the hope that it will be useful,
11 * but WITHOUT ANY WARRANTY; without even the implied warranty of
12 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
13 * GNU General Public License for more details.
14 *
15 * You should have received a copy of the GNU General Public License along
16 * with this program; if not, write to the Free Software Foundation, Inc.,
17 * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
18 * http://www.gnu.org/copyleft/gpl.html
19 *
20 * @file
21 * @ingroup Parser
22 */
24 /**
25 * @defgroup Parser Parser
26 */
28 /**
29 * PHP Parser - Processes wiki markup (which uses a more user-friendly
30 * syntax, such as "[[link]]" for making links), and provides a one-way
31 * transformation of that wiki markup it into XHTML output / markup
32 * (which in turn the browser understands, and can display).
33 *
34 * There are seven main entry points into the Parser class:
35 *
36 * - Parser::parse()
37 * produces HTML output
38 * - Parser::preSaveTransform().
39 * produces altered wiki markup.
40 * - Parser::preprocess()
41 * removes HTML comments and expands templates
42 * - Parser::cleanSig() and Parser::cleanSigInSig()
43 * Cleans a signature before saving it to preferences
44 * - Parser::getSection()
45 * Return the content of a section from an article for section editing
46 * - Parser::replaceSection()
47 * Replaces a section by number inside an article
48 * - Parser::getPreloadText()
49 * Removes <noinclude> sections, and <includeonly> tags.
50 *
51 * Globals used:
52 * object: $wgContLang
53 *
54 * @warning $wgUser or $wgTitle or $wgRequest or $wgLang. Keep them away!
55 *
56 * @par Settings:
57 * $wgLocaltimezone
58 * $wgNamespacesWithSubpages
59 *
60 * @par Settings only within ParserOptions:
61 * $wgAllowExternalImages
62 * $wgAllowSpecialInclusion
63 * $wgInterwikiMagic
64 * $wgMaxArticleSize
65 *
66 * @ingroup Parser
67 */
69 /**
70 * Update this version number when the ParserOutput format
71 * changes in an incompatible way, so the parser cache
72 * can automatically discard old data.
73 */
76 /**
77 * Update this version number when the output of serialiseHalfParsedText()
78 * changes in an incompatible way
79 */
82 # Flags for Parser::setFunctionHook
83 # Also available as global constants from Defines.php
87 # Constants needed for external link processing
88 # Everything except bracket, space, or control characters
89 # \p{Zs} is unicode 'separator, space' category. It covers the space 0x20
90 # as well as U+3000 is IDEOGRAPHIC SPACE for bug 19052
95 # State constants for the definition list colon extraction
105 # Flags for preprocessToDom
108 # Allowed values for $this->mOutputType
109 # Parameter to startExternalParse().
116 # Marker Suffix needs to be accessible staticly.
119 # Persistent:
133 # Initialised by initialiseVariables()
135 /**
136 * @var MagicWordArray
137 */
140 /**
141 * @var MagicWordArray
142 */
144 var $mConf, $mPreprocessor, $mExtLinkBracketedRegex, $mUrlProtocols; # Initialised in constructor
146 # Cleared with clearState():
147 /**
148 * @var ParserOutput
149 */
153 /**
154 * @var StripState
155 */
159 /**
160 * @var LinkHolderArray
161 */
172 /**
173 * @var User
174 */
177 # Temporary
178 # These are variables reset at least once per parse regardless of $clearState
180 /**
181 * @var ParserOptions
182 */
185 /**
186 * @var Title
187 */
197 /**
198 * @var string
199 */
202 /**
203 * @var Array with the language name of each language link (i.e. the
204 * interwiki prefix) in the key, value arbitrary. Used to avoid sending
205 * duplicate language links to the ParserOutput.
206 */
209 /**
210 * Constructor
211 *
212 * @param $conf array
213 */
222 # Preprocessor_Hash is much faster than Preprocessor_DOM in compiled mode
225 # PECL extension that conflicts with the core DOM extension (bug 13770)
232 }
234 }
236 /**
237 * Reduce memory usage to reduce the impact of circular references
238 */
242 }
245 }
246 }
248 /**
249 * Allow extensions to clean up when the parser is cloned
250 */
253 }
255 /**
256 * Do various kinds of initialisation on the first call of the parser
257 */
261 }
272 }
274 /**
275 * Clear Parser state
276 *
277 * @private
278 */
283 }
300 /**
301 * Prefix for temporary replacement strings for the multipass parser.
302 * \x07 should never appear in input as it's disallowed in XML.
303 * Using it at the front also gives us a little extra robustness
304 * since it shouldn't match when butted up against identifier-like
305 * string constructs.
306 *
307 * Must not consist of all title characters, or else it will change
308 * the behavior of <nowiki> in a link.
309 */
313 # Clear these on every parse, bug 4549
321 );
330 # Fix cloning
333 }
337 }
339 /**
340 * Convert wikitext to HTML
341 * Do not call this function recursively.
342 *
343 * @param string $text text we want to parse
344 * @param $title Title object
345 * @param $options ParserOptions
346 * @param $linestart boolean
347 * @param $clearState boolean
348 * @param int $revid number to pass in {{REVISIONID}}
349 * @return ParserOutput a ParserOutput
350 */
351 public function parse( $text, Title $title, ParserOptions $options, $linestart = true, $clearState = true, $revid = null ) {
352 /**
353 * First pass--just handle <nowiki> sections, pass the rest off
354 * to internalParse() which does all the real work.
355 */
364 # Remove the strip marker tag prefix from the input, if present.
367 }
378 }
381 # No more strip!
388 # Clean up special characters, only run once, next-to-last before doBlockLevels
390 # french spaces, last one Guillemet-left
391 # only if there is something before the space
393 # french spaces, Guillemet-right
396 );
403 /**
404 * The input doesn't get language converted if
405 * a) It's disabled
406 * b) Content isn't converted
407 * c) It's a conversion table
408 * d) it is an interface message (which is in the user language)
409 */
412 {
414 # The position of the convert() call should not be changed. it
415 # assumes that the links are all replaced and the only thing left
416 # is the <nowiki> mark.
418 }
419 }
421 /**
422 * A converted title will be provided in the output object if title and
423 * content conversion are enabled, the article text does not contain
424 * a conversion-suppressing double-underscore tag, and no
425 * {{DISPLAYTITLE:...}} is present. DISPLAYTITLE takes precedence over
426 * automatic link conversion.
427 */
432 {
439 }
440 }
454 # attempt to sanitize at least some nesting problems
455 # (bug #2702 and quite a few others)
457 # ''Something [http://www.cool.com cool''] -->
458 # <i>Something</i><a href="http://www.cool.com"..><i>cool></i></a>
461 # fix up an anchor inside another anchor, only
462 # at least for a single single nested link (bug 3695)
465 # fix div inside inline elements- doBlockLevels won't wrap a line which
466 # contains a div, so fix it up here; replace
467 # div with escaped text
470 # remove empty italic or bold tag pairs, some
471 # introduced by rules above
473 );
479 }
485 );
486 }
490 # Information on include size limits, for the benefit of users who try to skirt them
493 $PFreport = "Expensive parser function count: {$this->mExpensiveFunctionCount}/{$this->mOptions->getExpensiveParserFunctionLimit()}\n";
496 "Preprocessor visited node count: {$this->mPPNodeCount}/{$this->mOptions->getMaxPPNodeCount()}\n" .
501 "Highest expansion depth: {$this->mHighestExpansionDepth}/{$this->mOptions->getMaxPPExpandDepth()}\n" .
505 // Sanitize for comment. Note '‐' in the replacement is U+2010,
506 // which looks much like the problematic '-'.
514 }
515 }
526 }
528 /**
529 * Recursive parser entry point that can be called from an extension tag
530 * hook.
531 *
532 * If $frame is not provided, then template variables (e.g., {{{1}}}) within $text are not expanded
533 *
534 * @param string $text text extension wants to have parsed
535 * @param $frame PPFrame: The frame to use for expanding any template variables
536 *
537 * @return string
538 */
546 }
548 /**
549 * Expand templates and variables in the text, producing valid, static wikitext.
550 * Also removes comments.
551 * @return mixed|string
552 */
558 }
565 }
567 /**
568 * Recursive parser entry point that can be called from an extension tag
569 * hook.
570 *
571 * @param string $text text to be expanded
572 * @param $frame PPFrame: The frame to use for expanding any template variables
573 * @return String
574 * @since 1.19
575 */
582 }
584 /**
585 * Process the wikitext for the "?preload=" feature. (bug 5210)
586 *
587 * "<noinclude>", "<includeonly>" etc. are parsed as for template
588 * transclusion, comments, templates, arguments, tags hooks and parser
589 * functions are untouched.
590 *
591 * @param $text String
592 * @param $title Title
593 * @param $options ParserOptions
594 * @return String
595 */
597 # Parser (re)initialisation
605 }
607 /**
608 * Get a random string
609 *
610 * @return string
611 */
614 }
616 /**
617 * Set the current user.
618 * Should only be used when doing pre-save transform.
619 *
620 * @param $user Mixed: User object or null (to reset)
621 */
624 }
626 /**
627 * Accessor for mUniqPrefix.
628 *
629 * @return String
630 */
633 # @todo FIXME: This is probably *horribly wrong*
634 # LanguageConverter seems to want $wgParser's uniqPrefix, however
635 # if this is called for a parser cache hit, the parser may not
636 # have ever been initialized in the first place.
637 # Not really sure what the heck is supposed to be going on here.
639 # throw new MWException( "Accessing uninitialized mUniqPrefix" );
640 }
642 }
644 /**
645 * Set the context title
646 *
647 * @param $t Title
648 */
652 }
655 # Strip the fragment to avoid various odd effects
660 }
661 }
663 /**
664 * Accessor for the Title object
665 *
666 * @return Title object
667 */
670 }
672 /**
673 * Accessor/mutator for the Title object
674 *
675 * @param $x Title object or null to just get the current one
676 * @return Title object
677 */
680 }
682 /**
683 * Set the output type
684 *
685 * @param $ot Integer: new value
686 */
689 # Shortcut alias
695 );
696 }
698 /**
699 * Accessor/mutator for the output type
700 *
701 * @param int|null $x New value or null to just get the current one
702 * @return Integer
703 */
706 }
708 /**
709 * Get the ParserOutput object
710 *
711 * @return ParserOutput object
712 */
715 }
717 /**
718 * Get the ParserOptions object
719 *
720 * @return ParserOptions object
721 */
724 }
726 /**
727 * Accessor/mutator for the ParserOptions object
728 *
729 * @param $x ParserOptions New value or null to just get the current one
730 * @return ParserOptions Current ParserOptions object
731 */
734 }
736 /**
737 * @return int
738 */
741 }
743 /**
744 * @param $id int
745 */
748 }
750 /**
751 * Get a language object for use in parser functions such as {{FORMATNUM:}}
752 * @return Language
753 */
756 }
758 /**
759 * Get the target language for the content being parsed. This is usually the
760 * language that the content is in.
761 *
762 * @since 1.19
763 *
764 * @throws MWException
765 * @return Language|null
766 */
776 }
779 }
781 /**
782 * Get the language object for language conversion
783 */
786 }
788 /**
789 * Get a User object either from $this->mUser, if set, or from the
790 * ParserOptions object otherwise
791 *
792 * @return User object
793 */
797 }
799 }
801 /**
802 * Get a preprocessor object
803 *
804 * @return Preprocessor instance
805 */
810 }
812 }
814 /**
815 * Replaces all occurrences of HTML-style comments and the given tags
816 * in the text with a random marker and returns the next text. The output
817 * parameter $matches will be an associative array filled with data in
818 * the form:
819 *
820 * @code
821 * 'UNIQ-xxxxx' => array(
822 * 'element',
823 * 'tag content',
824 * array( 'param' => 'x' ),
825 * '<element param="x">tag content</element>' ) )
826 * @endcode
827 *
828 * @param array $elements list of element names. Comments are always extracted.
829 * @param string $text Source text string.
830 * @param array $matches Out parameter, Array: extracted tags
831 * @param $uniq_prefix string
832 * @return String: stripped text
833 */
847 }
849 # comment
855 # tag
860 }
866 # Empty element tag, <tag />
875 }
879 # No end tag -- let it run out to the end of the text.
885 }
886 }
892 }
894 }
896 /**
897 * Get a list of strippable XML-like elements
898 *
899 * @return array
900 */
903 }
905 /**
906 * Add an item to the strip state
907 * Returns the unique tag which must be inserted into the stripped text
908 * The tag will be replaced with the original text in unstrip()
909 *
910 * @param $text string
911 *
912 * @return string
913 */
919 }
921 /**
922 * parse the wiki syntax used to render tables
923 *
924 * @private
925 * @return string
926 */
945 }
951 # First check if we are starting a new table
964 # Don't do any of the following
968 # We are ending a table
974 }
978 }
982 }
986 # Now we have a table row
989 # Whats after the tag is now only attributes
1002 }
1006 }
1012 } elseif ( $first_character === '|' || $first_character === '!' || substr( $line, 0, 2 ) === '|+' ) {
1013 # This might be cell elements, td, th or captions
1017 }
1023 }
1025 # Split up multiple cells on the same line.
1026 # FIXME : This can result in improper nesting of tags processed
1027 # by earlier parser steps, but should avoid splitting up eg
1028 # attribute values containing literal "||".
1033 # Loop through each table cell
1040 }
1045 }
1051 }
1061 }
1065 # A cell could contain both parameters and data
1068 # Bug 553: Note that a '|' inside an invalid link should not
1069 # be mistaken as delimiting cell parameters
1078 }
1082 }
1083 }
1085 }
1087 # Closing open td, tr && table
1091 }
1094 }
1097 }
1100 }
1102 # Remove trailing line-ending (b/c)
1105 }
1107 # special case: don't return empty table
1110 }
1115 }
1117 /**
1118 * Helper function for parse() that transforms wiki markup into
1119 * HTML. Only called for $mOutputType == self::OT_HTML.
1120 *
1121 * @private
1122 *
1123 * @param $text string
1124 * @param $isMain bool
1125 * @param $frame bool
1126 *
1127 * @return string
1128 */
1134 # Hook to suspend the parser in this state
1135 if ( !wfRunHooks( 'ParserBeforeInternalParse', array( &$this, &$text, &$this->mStripState ) ) ) {
1138 }
1140 # if $frame is provided, then use $frame for replacing any variables
1142 # use frame depth to infer how include/noinclude tags should be handled
1143 # depth=0 means this is the top-level document; otherwise it's an included document
1148 }
1152 # if $frame is not provided, then use old-style replaceVariables
1154 }
1157 $text = Sanitizer::removeHTMLtags( $text, array( &$this, 'attributeStripCallback' ), false, array_keys( $this->mTransparentTagHooks ) );
1160 # Tables need to come after variable replacement for things to work
1161 # properly; putting them before other transformations should keep
1162 # exciting things like link expansions from showing up in surprising
1163 # places.
1175 # replaceInternalLinks may sometimes leave behind
1176 # absolute URLs, which have to be masked to hide them from replaceExternalLinks
1184 }
1186 /**
1187 * Replace special strings like "ISBN xxx" and "RFC xxx" with
1188 * magic external links.
1189 *
1190 * DML
1191 * @private
1192 *
1193 * @param $text string
1194 *
1195 * @return string
1196 */
1202 '!(?: # Start cases
1206 (?:RFC|PMID)\s+([0-9]+) | # m[4]: RFC or PMID, capture number
1208 (?: 97[89] [\ \-]? )? # optional 13-digit ISBN prefix
1209 (?: [0-9] [\ \-]? ){9} # 9 digits with opt. delimiters
1210 [0-9Xx] # check digit
1215 }
1217 /**
1218 * @throws MWException
1219 * @param $m array
1220 * @return HTML|string
1221 */
1224 # Skip anchor
1227 # Skip HTML element
1230 # Free external link
1233 # RFC or PMID
1247 }
1251 # ISBN
1257 ));
1264 }
1265 }
1267 /**
1268 * Make a free external link, given a user-supplied URL
1269 *
1270 * @param $url string
1271 *
1272 * @return string HTML
1273 * @private
1274 */
1280 # The characters '<' and '>' (which were escaped by
1281 # removeHTMLtags()) should not be included in
1282 # URLs, per RFC 2396.
1287 }
1289 # Move trailing punctuation to $trail
1291 # If there is no left bracket, then consider right brackets fair game too
1294 }
1300 }
1304 # Is this an external image?
1307 # Not an image, make a link
1312 # Register it in the output object...
1313 # Replace unnecessary URL escape codes with their equivalent characters
1316 }
1319 }
1321 /**
1322 * Parse headers and return html
1323 *
1324 * @private
1325 *
1326 * @param $text string
1327 *
1328 * @return string
1329 */
1335 }
1338 }
1340 /**
1341 * Replace single quotes with HTML markup
1342 * @private
1343 *
1344 * @param $text string
1345 *
1346 * @return string the altered text
1347 */
1354 }
1358 }
1360 /**
1361 * Helper function for doAllQuotes()
1362 *
1363 * @param $text string
1364 *
1365 * @return string
1366 */
1372 # First, do some preliminary work. This may shift some apostrophes from
1373 # being mark-up to being text. It also counts the number of occurrences
1374 # of bold and italics mark-ups.
1379 # If there are ever four apostrophes, assume the first is supposed to
1380 # be text, and the remaining three constitute mark-up for bold text.
1385 # If there are more than 5 apostrophes in a row, assume they're all
1386 # text except for the last 5.
1389 }
1390 # Count the number of occurrences of bold and italics mark-ups.
1391 # We are not counting sequences of five apostrophes.
1399 }
1400 }
1401 }
1403 # If there is an odd number of both bold and italics, it is likely
1404 # that one of the bold ones was meant to be an apostrophe followed
1405 # by italics. Which one we cannot know for certain, but it is more
1406 # likely to be one that has a single-letter word before it.
1419 }
1423 }
1427 }
1428 }
1429 }
1431 }
1433 # If there is a single-letter word, use it!
1438 # If not, but there's a multi-letter word, use that one.
1442 # ... otherwise use the first one that has neither.
1443 # (notice that it is possible for all three to be -1 if, for example,
1444 # there is only one pentuple-apostrophe in the line)
1447 }
1448 }
1450 # Now let's actually convert our apostrophic mush to HTML!
1461 }
1479 }
1496 }
1516 }
1517 }
1518 }
1520 }
1521 # Now close all remaining tags. Notice that the order is important.
1524 }
1527 }
1530 }
1531 # There might be lonely ''''', so make sure we have a buffer
1534 }
1536 }
1537 }
1539 /**
1540 * Replace external links (REL)
1541 *
1542 * Note: this is all very hackish and the order of execution matters a lot.
1543 * Make sure to run maintenance/parserTests.php if you change this code.
1544 *
1545 * @private
1546 *
1547 * @param $text string
1548 *
1549 * @throws MWException
1550 * @return string
1551 */
1558 throw new MWException( "PCRE needs to be compiled with --enable-unicode-properties in order for MediaWiki to function" );
1559 }
1569 # The characters '<' and '>' (which were escaped by
1570 # removeHTMLtags()) should not be included in
1571 # URLs, per RFC 2396.
1576 }
1578 # If the link text is an image URL, replace it with an <img> tag
1579 # This happened by accident in the original parser, but some people used it extensively
1583 }
1587 # Set linktype for CSS - if URL==text, link is essentially free
1590 # No link text, e.g. [http://domain.tld/some.link]
1592 # Autonumber
1597 # Have link text, e.g. [http://domain.tld/some.link text]s
1598 # Check for trail
1600 }
1606 # Use the encoded URL
1607 # This means that users can paste URLs directly into the text
1608 # Funny characters like ö aren't valid in URLs anyway
1609 # This was changed in August 2004
1613 # Register link in the output object.
1614 # Replace unnecessary URL escape codes with the referenced character
1615 # This prevents spammers from hiding links from the filters
1618 }
1622 }
1623 /**
1624 * Get the rel attribute for a particular external link.
1625 *
1626 * @since 1.21
1627 * @param string|bool $url optional URL, to extract the domain from for rel =>
1628 * nofollow if appropriate
1629 * @param $title Title optional Title, for wgNoFollowNsExceptions lookups
1630 * @return string|null rel attribute for $url
1631 */
1637 {
1639 }
1641 }
1642 /**
1643 * Get an associative array of additional HTML attributes appropriate for a
1644 * particular external link. This currently may include rel => nofollow
1645 * (depending on configuration, namespace, and the URL's domain) and/or a
1646 * target attribute (depending on configuration).
1647 *
1648 * @param string|bool $url optional URL, to extract the domain from for rel =>
1649 * nofollow if appropriate
1650 * @return Array associative array of HTML attributes
1651 */
1658 }
1660 }
1662 /**
1663 * Replace unusual URL escape codes with their equivalent characters
1664 *
1665 * @param $url String
1666 * @return String
1667 *
1668 * @todo This can merge genuinely required bits in the path or query string,
1669 * breaking legit URLs. A proper fix would treat the various parts of
1670 * the URL differently; as a workaround, just use the output for
1671 * statistical records, not for actual linking/output.
1672 */
1676 }
1678 /**
1679 * Callback function used in replaceUnusualEscapes().
1680 * Replaces unusual URL escape codes with their equivalent character
1681 *
1682 * @param $matches array
1683 *
1684 * @return string
1685 */
1689 # Is it an unsafe or HTTP reserved character according to RFC 1738?
1691 # No, shouldn't be escaped
1694 # Yes, leave it escaped
1696 }
1697 }
1699 /**
1700 * make an image if it's allowed, either through the global
1701 * option, through the exception, or through the on-wiki whitelist
1702 * @private
1703 *
1704 * $param $url string
1705 *
1706 * @return string
1707 */
1712 # $imagesfrom could be either a single string or an array of strings, parse out the latter
1719 }
1720 }
1725 }
1729 # Image found
1731 }
1732 }
1735 $whitelist = explode( "\n", wfMessage( 'external_image_whitelist' )->inContentLanguage()->text() );
1737 # Sanitize the regex fragment, make it case-insensitive, ignore blank entries/comments
1740 }
1742 # Image matches a whitelist entry
1745 }
1746 }
1747 }
1749 }
1751 /**
1752 * Process [[ ]] wikilinks
1753 *
1754 * @param $s string
1755 *
1756 * @return String: processed text
1757 *
1758 * @private
1759 */
1763 }
1765 /**
1766 * Process [[ ]] wikilinks (RIL)
1767 * @param $s
1768 * @throws MWException
1769 * @return LinkHolderArray
1770 *
1771 * @private
1772 */
1778 # the % is needed to support urlencoded titles as well
1781 # Match a link having the form [[namespace:link|alternate]]trail
1783 # Match cases where there is no "]]", which might still be images
1785 }
1789 # split the entire text string on occurrences of [[
1791 # get the first element (all text up to first [[), and remove the space we added
1800 # Match the end of a line for a word that's not followed by whitespace,
1801 # e.g. in the case of 'The Arab al[[Razi]]', 'al' will be matched
1803 }
1809 }
1818 }
1821 }
1826 # Loop for each link
1828 # Check for excessive memory usage
1830 # Too big
1831 # Do the existence check, replace the link holders and clear the array
1834 }
1843 }
1844 # first link
1848 }
1850 }
1857 # If we get a ] at the beginning of $m[3] that means we have a link that's something like:
1858 # [[Image:Foo.jpg|[http://example.com desc]]] <- having three ] in a row fucks up,
1859 # the real problem is with the $e1 regex
1860 # See bug 1300.
1861 #
1862 # Still some problems for cases where the ] is meant to be outside punctuation,
1863 # and no image is in sight. See bug 2095.
1864 #
1868 )
1869 {
1872 }
1873 # fix up urlencoded title texts
1875 # Should anchors '#' also be rejected?
1877 }
1879 } elseif ( preg_match( $e1_img, $line, $m ) ) { # Invalid, but might be an image with a link in its caption
1884 }
1890 }
1894 # Don't allow internal links to pages containing
1895 # PROTO: where PROTO is a valid URL protocol; these
1896 # should be external links.
1901 }
1903 # Make subpage if necessary
1908 }
1912 # Strip off leading ':'
1914 }
1923 }
1934 # look at the next 'line' to see if we can close it there
1939 }
1942 # the first ]] closes the inner link, the second the image
1948 # if there's exactly one ]] that's fine, we'll keep looking
1951 # if $next_line is invalid too, we need look no further
1954 }
1955 }
1957 # we couldn't find the end of this imageLink, so output it raw
1958 # but don't ignore what might be perfectly normal links in the text we've examined
1961 # note: no $trail, because without an end, there *is* no trail
1964 }
1967 # note: no $trail, because without an end, there *is* no trail
1970 }
1972 }
1978 # Bug 4598 madness. Handle the quotes only if they come from the alternate part
1979 # [[Lista d''e paise d''o munno]] -> <a href="...">Lista d''e paise d''o munno</a>
1980 # [[Criticism of Harry Potter|Criticism of ''Harry Potter'']]
1981 # -> <a href="Criticism of Harry Potter">Criticism of <i>Harry Potter</i></a>
1983 }
1985 # Link not escaped by : , create the various objects
1987 # Interwikis
1989 if ( $iw && $this->mOptions->getInterwikiMagic() && $nottalk && Language::fetchLanguageName( $iw, null, 'mw' ) ) {
1990 // XXX: the above check prevents links to sites with identifiers that are not language codes
1992 # Bug 24502: filter duplicates
1996 }
2002 }
2009 # if no parameters were passed, $text
2010 # becomes something like "File:Foo.png",
2011 # which we don't want to pass on to the
2012 # image generator
2015 # recursively parse links inside the image caption
2016 # actually, this will parse them in any other parameters, too,
2017 # but it might be hard to fix that, and it doesn't matter ATM
2020 }
2021 # cloak any absolute URLs inside the image markup, so replaceExternalLinks() won't touch them
2026 }
2029 }
2039 }
2045 /**
2046 * Strip the whitespace Category links produce, see bug 87
2047 * @todo We might want to use trim($tmp, "\n") here.
2048 */
2053 }
2054 }
2056 # Self-link checking
2061 true
2062 ) ) ) {
2065 }
2066 }
2068 # NS_MEDIA is a pseudo-namespace for linking directly to a file
2069 # @todo FIXME: Should do batch file existence checks, see comment below
2072 # Give extensions a chance to select the file revision for us
2077 # Fetch and register the file (file title may be different via hooks)
2079 # Cloak with NOPARSE to avoid replacement in replaceExternalLinks
2084 }
2087 # Some titles, such as valid special pages or files in foreign repos, should
2088 # be shown as bluelinks even though they're not included in the page table
2089 #
2090 # @todo FIXME: isAlwaysKnown() can be expensive for file links; we should really do
2091 # batch file existence checks for NS_FILE and NS_MEDIA
2096 # Links will be added to the output link list after checking
2098 }
2100 }
2103 }
2105 /**
2106 * Render a forced-blue link inline; protect against double expansion of
2107 * URLs if we're in a mode that prepends full URL prefixes to internal links.
2108 * Since this little disaster has to split off the trail text to avoid
2109 * breaking URLs in the following text without breaking trails on the
2110 * wiki links, it's been made into a horrible function.
2111 *
2112 * @param $nt Title
2113 * @param $text String
2114 * @param array $query or String
2115 * @param $trail String
2116 * @param $prefix String
2117 * @return String: HTML-wikitext mix oh yuck
2118 */
2124 }
2127 }
2132 }
2134 /**
2135 * Insert a NOPARSE hacky thing into any inline links in a chunk that's
2136 * going to go through further parsing steps before inline URL expansion.
2137 *
2138 * Not needed quite as much as it used to be since free links are a bit
2139 * more sensible these days. But bracketed links are still an issue.
2140 *
2141 * @param string $text more-or-less HTML
2142 * @return String: less-or-more HTML with NOPARSE bits
2143 */
2147 }
2149 /**
2150 * Return true if subpage links should be expanded on this page.
2151 * @return Boolean
2152 */
2154 # Some namespaces don't allow subpages
2156 }
2158 /**
2159 * Handle link to subpage if necessary
2160 *
2161 * @param string $target the source of the link
2162 * @param &$text String: the link text, modified as necessary
2163 * @return string the full name of the link
2164 * @private
2165 */
2168 }
2170 /**#@+
2171 * Used by doBlockLevels()
2172 * @private
2173 *
2174 * @return string
2175 */
2180 }
2184 }
2186 /**
2187 * getCommon() returns the length of the longest common substring
2188 * of both arguments, starting at the beginning of both.
2189 * @private
2190 *
2191 * @param $st1 string
2192 * @param $st2 string
2193 *
2194 * @return int
2195 */
2201 }
2206 }
2207 }
2209 }
2211 /**
2212 * These next three functions open, continue, and close the list
2213 * element appropriate to the prefix character passed into them.
2214 * @private
2215 *
2216 * @param $char string
2217 *
2218 * @return string
2219 */
2234 }
2237 }
2239 /**
2240 * TODO: document
2241 * @param $char String
2242 * @private
2243 *
2244 * @return string
2245 */
2253 }
2260 }
2261 }
2263 }
2265 /**
2266 * TODO: document
2267 * @param $char String
2268 * @private
2269 *
2270 * @return string
2271 */
2283 }
2286 }
2288 }
2289 /**#@-*/
2291 /**
2292 * Make lists from lines starting with ':', '*', '#', etc. (DBL)
2293 *
2294 * @param $text String
2295 * @param $linestart Boolean: whether or not this is at the start of a line.
2296 * @private
2297 * @return string the lists rendered as HTML
2298 */
2302 # Parsing through the text line by line. The main thing
2303 # happening here is handling of block-level elements p, pre,
2304 # and making lists from lines starting with * # : etc.
2305 #
2314 # Fix up $linestart
2319 }
2320 # * = ul
2321 # # = ol
2322 # ; = dt
2323 # : = dd
2328 # If not in a <pre> element, scan for and figure out what prefixes are there.
2330 # Multiple prefixes may abut each other for nested lists.
2334 # eh?
2335 # ; and : are both from definition-lists, so they're equivalent
2336 # for the purposes of determining whether or not we need to open/close
2337 # elements.
2342 # Don't interpret any other prefixes in preformatted text
2346 }
2348 # List generation
2350 # Same as the last item, so no need to deal with nesting or opening stuff
2355 # The one nasty exception: definition lists work like this:
2356 # ; title : definition text
2357 # So we check for : in the remainder text to split up the
2358 # title and definition, without b0rking links.
2363 }
2364 }
2366 # We need to open or close prefixes, or both.
2368 # Either open or close a level...
2372 # Close all the prefixes which aren't shared.
2376 }
2378 # Continue the current prefix if appropriate.
2381 }
2383 # Open prefixes where appropriate.
2389 # @todo FIXME: This is dupe of code above
2393 }
2394 }
2396 }
2398 }
2400 # If we have no prefixes, go to paragraph mode.
2403 # No prefix (not in list)--go to paragraph mode
2404 # XXX: use a stack for nestable elements like span, table and div
2405 $openmatch = preg_match( '/(?:<table|<blockquote|<h1|<h2|<h3|<h4|<h5|<h6|<pre|<tr|<p|<ul|<ol|<dl|<li|<\\/tr|<\\/td|<\\/th)/iS', $t );
2408 '<td|<th|<\\/?div|<hr|<\\/pre|<\\/p|' . $this->mUniqPrefix . '-pre|<\\/li|<\\/ul|<\\/ol|<\\/dl|<\\/?center)/iS', $t );
2411 # TODO bug 5718: paragraph closed
2415 }
2419 # pre
2424 }
2427 # paragraph
2440 }
2441 }
2450 }
2451 }
2452 }
2453 }
2455 }
2456 # somewhere above we forget to get out of pre block (bug 785)
2459 }
2462 }
2463 }
2467 }
2471 }
2475 }
2477 /**
2478 * Split up a string on ':', ignoring any occurrences inside tags
2479 * to prevent illegal overlapping.
2480 *
2481 * @param string $str the string to split
2482 * @param &$before String set to everything before the ':'
2483 * @param &$after String set to everything after the ':'
2484 * @throws MWException
2485 * @return String the position of the ':', or false if none found
2486 */
2492 # Nothing to find!
2495 }
2499 # Easy; no tag nesting to worry about
2504 }
2506 # Ugly state machine to walk through avoiding tags.
2514 # (Using the number is a performance hack for common cases)
2518 # Could be either a <start> tag or an </end> tag
2523 # We found it!
2528 }
2529 # Embedded in a tag; don't break it.
2532 # Skip ahead looking for something interesting
2535 # Nothing else interesting
2538 }
2542 # We found it!
2547 }
2548 }
2550 # Nothing else interesting to find; abort!
2551 # We're nested, but there's no close tags left. Abort!
2553 }
2554 # Skip ahead to next tag start
2557 }
2560 # In a <tag>
2567 # Slash may be followed by >?
2571 # ignore
2572 }
2583 # Illegal early close? This shouldn't happen D:
2588 }
2591 # In a </tag>
2598 }
2600 }
2604 # Yes, a self-closed tag <blah/>
2607 # Probably we're jumping the gun, and this is an attribute
2609 }
2614 }
2621 }
2628 }
2633 }
2634 }
2636 wfDebug( __METHOD__ . ": Invalid input; not enough close tags (stack $stack, state $state)\n" );
2639 }
2642 }
2644 /**
2645 * Return value of a magic variable (like PAGENAME)
2646 *
2647 * @private
2648 *
2649 * @param $index integer
2650 * @param bool|\PPFrame $frame
2651 *
2652 * @throws MWException
2653 * @return string
2654 */
2660 // If no title set, bad things are going to happen
2661 // later. Title should always be set since this
2662 // should only be called in the middle of a parse
2663 // operation (but the unit-tests do funky stuff)
2666 }
2668 /**
2669 * Some of these require message or data lookups and can be
2670 * expensive to check many times.
2671 */
2675 }
2676 }
2681 # Use the time zone
2686 }
2700 }
2769 $value = wfEscapeWikiText( wfUrlEncode( str_replace( ' ', '_', $this->mTitle->getBaseText() ) ) );
2777 }
2785 }
2798 # 0 means the page doesn't exist in the database,
2799 # which means the user is previewing a new page.
2800 # The vary-revision flag must be set, because the magic word
2801 # will have a different value once the page is saved.
2804 }
2808 # Let the edit saving system know we should parse the page
2809 # *after* a revision ID has been assigned.
2815 # Let the edit saving system know we should parse the page
2816 # *after* a revision ID has been assigned. This is for null edits.
2822 # Let the edit saving system know we should parse the page
2823 # *after* a revision ID has been assigned. This is for null edits.
2829 # Let the edit saving system know we should parse the page
2830 # *after* a revision ID has been assigned. This is for null edits.
2836 # Let the edit saving system know we should parse the page
2837 # *after* a revision ID has been assigned. This is for null edits.
2843 # Let the edit saving system know we should parse the page
2844 # *after* a revision ID has been assigned. This is for null edits.
2850 # Let the edit saving system know we should parse the page
2851 # *after* a revision ID has been assigned. This is for null edits.
2857 # Let the edit saving system know we should parse the page
2858 # *after* a revision ID has been assigned. This is for null edits.
2873 $value = $this->mTitle->canTalk() ? str_replace( '_', ' ', $this->mTitle->getTalkNsText() ) : '';
2897 # @bug 4594 PHP5 has it zero padded, PHP4 does not, cast to
2898 # int to remove the padding
2917 # @bug 4594 PHP5 has it zero padded, PHP4 does not, cast to
2918 # int to remove the padding
2978 if ( wfRunHooks( 'ParserGetVariableValueSwitch', array( &$this, &$this->mVarCache, &$index, &$ret, &$frame ) ) ) {
2982 }
2983 }
2987 }
2990 }
2992 /**
2993 * initialise the magic variables (like CURRENTMONTHNAME) and substitution modifiers
2994 *
2995 * @private
2996 */
3005 }
3007 /**
3008 * Preprocess some wikitext and return the document tree.
3009 * This is the ghost of replace_variables().
3010 *
3011 * @param string $text The text to parse
3012 * @param $flags Integer: bitwise combination of:
3013 * self::PTD_FOR_INCLUSION Handle "<noinclude>" and "<includeonly>" as if the text is being
3014 * included. Default is to assume a direct page view.
3015 *
3016 * The generated DOM tree must depend only on the input text and the flags.
3017 * The DOM tree must be the same in OT_HTML and OT_WIKI mode, to avoid a regression of bug 4899.
3018 *
3019 * Any flag added to the $flags parameter here, or any other parameter liable to cause a
3020 * change in the DOM tree for a given text, must be passed through the section identifier
3021 * in the section edit link and thus back to extractSections().
3022 *
3023 * The output of this function is currently only cached in process memory, but a persistent
3024 * cache may be implemented at a later date which takes further advantage of these strict
3025 * dependency requirements.
3026 *
3027 * @private
3028 *
3029 * @return PPNode
3030 */
3034 }
3036 /**
3037 * Return a three-element array: leading whitespace, string contents, trailing whitespace
3038 *
3039 * @param $s string
3040 *
3041 * @return array
3042 */
3052 }
3054 }
3056 /**
3057 * Replace magic variables, templates, and template arguments
3058 * with the appropriate text. Templates are substituted recursively,
3059 * taking care to avoid infinite loops.
3060 *
3061 * Note that the substitution depends on value of $mOutputType:
3062 * self::OT_WIKI: only {{subst:}} templates
3063 * self::OT_PREPROCESS: templates but not extension tags
3064 * self::OT_HTML: all templates and extension tags
3065 *
3066 * @param string $text the text to transform
3067 * @param $frame PPFrame Object describing the arguments passed to the template.
3068 * Arguments may also be provided as an associative array, as was the usual case before MW1.12.
3069 * Providing arguments this way may be useful for extensions wishing to perform variable replacement explicitly.
3070 * @param $argsOnly Boolean only do argument (triple-brace) expansion, not double-brace expansion
3071 * @private
3072 *
3073 * @return string
3074 */
3076 # Is there any text? Also, Prevent too big inclusions!
3079 }
3085 wfDebug( __METHOD__ . " called using plain parameters instead of a PPFrame instance. Creating custom frame.\n" );
3087 }
3095 }
3097 /**
3098 * Clean up argument array - refactored in 1.9 so parserfunctions can use it, too.
3099 *
3100 * @param $args array
3101 *
3102 * @return array
3103 */
3116 }
3119 }
3120 }
3121 }
3124 }
3126 /**
3127 * Warn the user when a parser limitation is reached
3128 * Will warn at most once the user per limitation type
3129 *
3130 * @param string $limitationType should be one of:
3131 * 'expensive-parserfunction' (corresponding messages:
3132 * 'expensive-parserfunction-warning',
3133 * 'expensive-parserfunction-category')
3134 * 'post-expand-template-argument' (corresponding messages:
3135 * 'post-expand-template-argument-warning',
3136 * 'post-expand-template-argument-category')
3137 * 'post-expand-template-inclusion' (corresponding messages:
3138 * 'post-expand-template-inclusion-warning',
3139 * 'post-expand-template-inclusion-category')
3140 * @param int|null $current Current value
3141 * @param int|null $max Maximum allowed, when an explicit limit has been
3142 * exceeded, provide the values (optional)
3143 */
3145 # does no harm if $current and $max are present but are unnecessary for the message
3150 }
3152 /**
3153 * Return the text of a template, after recursively
3154 * replacing any variables or templates within the template.
3155 *
3156 * @param array $piece the parts of the template
3157 * $piece['title']: the title, i.e. the part before the |
3158 * $piece['parts']: the parameter array
3159 * $piece['lineStart']: whether the brace was at the start of a line
3160 * @param $frame PPFrame The current frame, contains template arguments
3161 * @throws MWException
3162 * @return String: the text of the template
3163 * @private
3164 */
3169 # Flags
3177 # Title object, where $text came from
3180 # $part1 is the bit before the first |, and must contain only title characters.
3181 # Various prefixes will be stripped from it later.
3186 # Original title text preserved for various purposes
3189 # $args is a list of argument nodes, starting from index 0, not including $part1
3190 # @todo FIXME: If piece['parts'] is null then the call to getLength() below won't work b/c this $args isn't an object
3196 # SUBST
3202 # Possibilities for substMatch: "subst", "safesubst" or FALSE
3203 # Decide whether to expand template or keep wikitext as-is.
3209 }
3215 }
3216 }
3221 }
3222 }
3224 # Variables
3231 }
3233 }
3234 }
3236 # MSG, MSGNW and RAW
3238 # Check for MSGNW:
3243 # Remove obsolete MSG:
3246 }
3248 # Check for RAW:
3252 }
3253 }
3256 # Parser functions
3266 }
3273 }
3275 # The interface for parser functions allows for extracting
3276 # flags into the local scope. Extract any forwarded flags
3277 # here.
3279 }
3281 }
3283 # Finish mangling title and then check for loops.
3284 # Set $title to a Title object and $titleText to the PDBK
3287 # Split the title into page and subpage
3292 }
3296 # Check for language variants if the template is not found
3299 }
3300 # Do recursion depth check
3308 }
3309 }
3310 }
3312 # Load from database
3315 # Too many unique items can kill profiling DBs/collectors
3318 }
3324 {
3325 // Pass the template arguments as URL parameters.
3326 // "uselang" will have no effect since the Language object
3327 // is forced to the one defined in ParserOptions.
3335 }
3336 }
3338 // Create a new context to execute the special page
3351 }
3360 }
3361 }
3363 # If the title is valid but undisplayable, make a link to it
3367 }
3369 # Interwiki transclusion
3375 # Preprocess it like a template
3378 }
3380 }
3382 # Do infinite loop check
3383 # This has to be done after redirect resolution to avoid infinite loops via redirects
3390 }
3392 }
3394 # If we haven't found text to substitute by now, we're done
3395 # Recover the source wikitext and return it
3400 }
3403 }
3405 # Expand DOM-style return values in a child frame
3407 # Clean up argument array
3413 # Expansion is eligible for the empty-frame cache
3419 }
3421 # Uncached expansion
3423 }
3424 }
3428 }
3432 }
3434 # Replace raw HTML by a placeholder
3438 # Escape nowiki-style return values
3443 {
3444 # Bug 529: if the template begins with a table or block-level
3445 # element, it should be treated as beginning a new line.
3446 # This behavior is somewhat controversial.
3448 }
3451 # Error, oversize inclusion
3453 # Make a working, properly escaped link if possible (bug 23588)
3456 # This will probably not be a working link, but at least it may
3457 # provide some hint of where the problem is
3460 }
3461 $text .= $this->insertStripItem( '<!-- WARNING: template omitted, post-expand include size too large -->' );
3463 }
3469 }
3473 }
3475 /**
3476 * Call a parser function and return an array with text and flags.
3477 *
3478 * The returned array will always contain a boolean 'found', indicating
3479 * whether the parser function was found or not. It may also contain the
3480 * following:
3481 * text: string|object, resulting wikitext or PP DOM object
3482 * isHTML: bool, $text is HTML, armour it against wikitext transformation
3483 * isChildObj: bool, $text is a DOM node needing expansion in a child frame
3484 * isLocalObj: bool, $text is a DOM node needing expansion in the current frame
3485 * nowiki: bool, wiki markup in $text should be escaped
3486 *
3487 * @since 1.21
3488 * @param $frame PPFrame The current frame, contains template arguments
3489 * @param $function string Function name
3490 * @param $args array Arguments to the function
3491 * @return array
3492 */
3498 # Case sensitive functions
3502 # Case insensitive functions
3509 }
3510 }
3515 # Workaround for PHP bug 35229 and similar
3520 }
3524 # Convert arguments to PPNodes and collect for appending to $allArgs
3531 }
3532 }
3534 # Add a frame parameter, and pass the arguments as an array
3538 # Convert arguments to plain text and append to $allArgs
3546 }
3547 }
3548 }
3552 # The interface for function hooks allows them to return a wikitext
3553 # string or an array containing the string and any flags. This mungs
3554 # things around to match what this method should return.
3559 );
3563 }
3567 );
3568 }
3574 }
3577 }
3582 }
3587 }
3589 /**
3590 * Get the semi-parsed DOM representation of a template with a given title,
3591 * and its redirect destination title. Cached.
3592 *
3593 * @param $title Title
3594 *
3595 * @return array
3596 */
3605 }
3608 }
3610 # Cache miss, go to the database
3616 }
3624 }
3627 }
3629 /**
3630 * Fetch the unparsed text of a template and register a reference to it.
3631 * @param Title $title
3632 * @return Array ( string or false, Title )
3633 */
3635 $templateCb = $this->mOptions->getTemplateCallback(); # Defaults to Parser::statelessFetchTemplate()
3642 }
3643 }
3645 }
3647 /**
3648 * Fetch the unparsed text of a template and register a reference to it.
3649 * @param Title $title
3650 * @return mixed string or false
3651 */
3655 }
3657 /**
3658 * Static function to get a template
3659 * Can be overridden via ParserOptions::setTemplateCallback().
3660 *
3661 * @param $title Title
3662 * @param $parser Parser
3663 *
3664 * @return array
3665 */
3671 # Loop to fetch the article, with up to 1 redirect
3673 # Give extensions a chance to select the revision instead
3684 );
3686 }
3687 # Get the revision
3692 # If there is no current revision, there is no page
3696 }
3703 # We fetched a rev from a different title; register it too...
3708 }
3717 }
3724 }
3729 }
3732 }
3733 # Redirect?
3736 }
3741 }
3743 /**
3744 * Fetch a file and its title and register a reference to it.
3745 * If 'broken' is a key in $options then the file will appear as a broken thumbnail.
3746 * @param Title $title
3747 * @param array $options Array of options to RepoGroup::findFile
3748 * @return File|bool
3749 */
3753 }
3755 /**
3756 * Fetch a file and its title and register a reference to it.
3757 * If 'broken' is a key in $options then the file will appear as a broken thumbnail.
3758 * @param Title $title
3759 * @param array $options Array of options to RepoGroup::findFile
3760 * @return Array ( File or false, Title of file )
3761 */
3769 }
3772 # Register the file as a dependency...
3775 # Update fetched file title
3778 # This file was not a redirect, but the title does not match.
3779 # Register under the new name because otherwise the link will
3780 # get lost.
3782 }
3783 }
3785 }
3787 /**
3788 * Transclude an interwiki link.
3789 *
3790 * @param $title Title
3791 * @param $action
3792 *
3793 * @return string
3794 */
3800 }
3806 }
3808 }
3810 /**
3811 * @param $url string
3812 * @return Mixed|String
3813 */
3822 }
3828 } elseif ( $req->getStatus() != 200 ) { // Though we failed to fetch the content, this status is useless.
3829 return wfMessage( 'scarytranscludefailed-httpstatus', $url, $req->getStatus() /* HTTP status */ )->inContentLanguage()->text();
3832 }
3839 ) );
3841 }
3843 /**
3844 * Triple brace replacement -- used for template arguments
3845 * @private
3846 *
3847 * @param $piece array
3848 * @param $frame PPFrame
3849 *
3850 * @return array
3851 */
3862 && (
3866 )
3867 ) {
3868 # No match in frame, use the supplied default
3870 }
3874 }
3877 # No match anywhere
3879 }
3882 }
3887 }
3891 }
3893 /**
3894 * Return the text to be used for a given extension tag.
3895 * This is the ghost of strip().
3896 *
3897 * @param array $params Associative array of parameters:
3898 * name PPNode for the tag name
3899 * attr PPNode for unparsed text where tag attributes are thought to be
3900 * attributes Optional associative array of parsed attributes
3901 * inner Contents of extension element
3902 * noClose Original text did not have a close tag
3903 * @param $frame PPFrame
3904 *
3905 * @throws MWException
3906 * @return string
3907 */
3912 $marker = "{$this->mUniqPrefix}-$name-" . sprintf( '%08X', $this->mMarkerIndex++ ) . self::MARKER_SUFFIX;
3920 }
3926 }
3929 # Workaround for PHP bug 35229 and similar
3932 }
3939 }
3945 }
3948 # Extract flags to local scope (to override $markerType)
3953 }
3957 }
3962 }
3963 }
3969 }
3970 }
3980 }
3982 }
3984 /**
3985 * Increment an include size counter
3986 *
3987 * @param string $type the type of expansion
3988 * @param $size Integer: the size of the text
3989 * @return Boolean: false if this inclusion would take it over the maximum, true otherwise
3990 */
3997 }
3998 }
4000 /**
4001 * Increment the expensive function count
4002 *
4003 * @return Boolean: false if the limit has been exceeded
4004 */
4008 }
4010 /**
4011 * Strip double-underscore items like __NOGALLERY__ and __NOTOC__
4012 * Fills $this->mDoubleUnderscores, returns the modified text
4013 *
4014 * @param $text string
4015 *
4016 * @return string
4017 */
4021 # The position of __TOC__ needs to be recorded
4027 # Set a placeholder. At the end we'll fill it in with the TOC.
4030 # Only keep the first one.
4032 }
4034 # Now match and remove the rest of them
4040 }
4043 }
4044 if ( isset( $this->mDoubleUnderscores['hiddencat'] ) && $this->mTitle->getNamespace() == NS_CATEGORY ) {
4046 }
4047 # (bug 8068) Allow control over whether robots index a page.
4048 #
4049 # @todo FIXME: Bug 14899: __INDEX__ always overrides __NOINDEX__ here! This
4050 # is not desirable, the last one on the page should win.
4054 }
4058 }
4060 # Cache all double underscores in the database
4063 }
4067 }
4069 /**
4070 * Add a tracking category, getting the title from a system message,
4071 * or print a debug message if the title is invalid.
4072 *
4073 * @param string $msg message key
4074 * @return Boolean: whether the addition was successful
4075 */
4080 }
4081 // Important to parse with correct title (bug 31469)
4087 # Allow tracking categories to be disabled by setting them to "-"
4090 }
4099 }
4100 }
4102 /**
4103 * This function accomplishes several tasks:
4104 * 1) Auto-number headings if that option is enabled
4105 * 2) Add an [edit] link to sections for users who have enabled the option and can edit the page
4106 * 3) Add a Table of contents on the top for users who have enabled the option
4107 * 4) Auto-anchor headings
4108 *
4109 * It loops through all headlines, collects the necessary data, then splits up the
4110 * string and re-inserts the newly formatted headlines.
4111 *
4112 * @param $text String
4113 * @param string $origText original, untouched wikitext
4114 * @param $isMain Boolean
4115 * @return mixed|string
4116 * @private
4117 */
4121 # Inhibit editsection links if requested in the page
4127 }
4130 }
4132 # Get all headlines for numbering them and adding funky stuff like [edit]
4133 # links - this is for later, but we need the number of headlines right now
4135 $numMatches = preg_match_all( '/<H(?P<level>[1-6])(?P<attrib>.*?' . '>)\s*(?P<header>[\s\S]*?)\s*<\/H[1-6] *>/i', $text, $matches );
4137 # if there are fewer than 4 headlines in the article, do not show TOC
4138 # unless it's been explicitly enabled.
4142 # Allow user to stipulate that a page should have a "new section"
4143 # link added via __NEWSECTIONLINK__
4146 }
4148 # Allow user to remove the "new section"
4149 # link via __NONEWSECTIONLINK__
4152 }
4154 # if the string __FORCETOC__ (not case-sensitive) occurs in the HTML,
4155 # override above conditions and always show TOC above first header
4159 }
4161 # headline counter
4165 # Ugh .. the TOC should have neat indentation levels which can be
4166 # passed to the skin functions. These are determined here
4198 }
4202 }
4206 # Increase TOC level
4213 }
4215 # Decrease TOC level, find level to jump to
4219 # Found last matching level
4223 # Found first matching level below current level
4226 }
4227 }
4230 }
4233 # Unindent only if the previous toc level was shown :p
4238 }
4239 }
4241 # No change in level, end TOC line
4244 }
4245 }
4249 # count number of headlines for each level
4256 }
4259 }
4260 }
4262 # The safe header is a version of the header text safe to use for links
4264 # Remove link placeholders by the link text.
4265 # <!--LINK number-->
4266 # turns into
4267 # link text with suffix
4268 # Do this before unstrip since link text can contain strip markers
4271 # Avoid insertion of weird stuff like <math> by expanding the relevant sections
4274 # Strip out HTML (first regex removes any tag not allowed)
4275 # Allowed tags are:
4276 # * <sup> and <sub> (bug 8393)
4277 # * <i> (bug 26375)
4278 # * <b> (r105284)
4279 # * <span dir="rtl"> and <span dir="ltr"> (bug 35167)
4280 #
4281 # We strip any parameter from accepted tags (second regex), except dir="rtl|ltr" from <span>,
4282 # to allow setting directionality in toc items.
4284 array( '#<(?!/?(span|sup|sub|i|b)(?: [^>]*)?>).*?' . '>#', '#<(/?(?:span(?: dir="(?:rtl|ltr)")?|sup|sub|i|b))(?: .*?)?' . '>#' ),
4286 $safeHeadline
4287 );
4290 # For the anchor, strip out HTML-y stuff period
4294 # Save headline for section edit hint before it's escaped
4298 # For reverse compatibility, provide an id that's
4299 # HTML4-compatible, like we used to.
4300 #
4301 # It may be worth noting, academically, that it's possible for
4302 # the legacy anchor to conflict with a non-legacy headline
4303 # anchor on the page. In this case likely the "correct" thing
4304 # would be to either drop the legacy anchors or make sure
4305 # they're numbered first. However, this would require people
4306 # to type in section names like "abc_.D7.93.D7.90.D7.A4"
4307 # manually, so let's not bother worrying about it.
4313 # No reason to have both (in fact, we can't)
4315 }
4320 }
4322 # HTML names must be case-insensitively unique (bug 10721).
4323 # This does not apply to Unicode characters per
4324 # http://dev.w3.org/html5/spec/infrastructure.html#case-sensitivity-and-string-comparison
4325 # @todo FIXME: We may be changing them depending on the current locale.
4331 }
4333 # count how many in assoc. array so we can track dupes in anchors
4338 }
4343 }
4345 # Don't number the heading if it is the only one (looks silly)
4347 # the two are different if the line contains a link
4348 $headline = Html::element( 'span', array( 'class' => 'mw-headline-number' ), $numbering ) . ' ' . $headline;
4349 }
4351 # Create the anchor for linking from the TOC to the section
4356 }
4359 }
4363 }
4365 # Add the section to the section tree
4366 # Find the DOM node for this header
4372 }
4373 }
4377 }
4387 );
4389 # give headline the correct <h#> tag
4391 // Output edit section links as markers with styles that can be customized by skins
4393 # Put a T flag in the section identifier, to indicate to extractSections()
4394 # that sections inside <includeonly> should be counted.
4398 }
4399 // We use a bit of pesudo-xml for editsection markers. The language converter is run later on
4400 // Using a UNIQ style marker leads to the converter screwing up the tokens when it converts stuff
4401 // And trying to insert strip tags fails too. At this point all real inputted tags have already been escaped
4402 // so we don't have to worry about a user trying to input one of these markers directly.
4403 // We use a page and section attribute to stop the language converter from converting these important bits
4404 // of data, but put the headline hint inside a content block because the language converter is supposed to
4405 // be able to convert that piece of data.
4412 }
4415 }
4421 }
4425 # Never ever show TOC if no headers
4428 }
4433 }
4436 }
4440 }
4442 # split up and insert constructed headlines
4446 // build an array of document sections
4449 // $head is zero-based, sections aren't.
4454 }
4456 /**
4457 * Send a hook, one per section.
4458 * The idea here is to be able to make section-level DIVs, but to do so in a
4459 * lower-impact, more correct way than r50769
4460 *
4461 * $this : caller
4462 * $section : the section number
4463 * &$sectionContent : ref to the content of the section
4464 * $showEditLinks : boolean describing whether this section has an edit link
4465 */
4469 }
4472 // append the TOC at the beginning
4473 // Top anchor now in skin
4475 }
4483 }
4484 }
4486 /**
4487 * Transform wiki markup when saving a page by doing "\r\n" -> "\n"
4488 * conversion, substitting signatures, {{subst:}} templates, etc.
4489 *
4490 * @param string $text the text to transform
4491 * @param $title Title: the Title object for the current article
4492 * @param $user User: the User object describing the current user
4493 * @param $options ParserOptions: parsing options
4494 * @param $clearState Boolean: whether to clear the parser state first
4495 * @return String: the altered wiki markup
4496 */
4497 public function preSaveTransform( $text, Title $title, User $user, ParserOptions $options, $clearState = true ) {
4503 );
4507 }
4513 }
4515 /**
4516 * Pre-save transform helper function
4517 * @private
4518 *
4519 * @param $text string
4520 * @param $user User
4521 *
4522 * @return string
4523 */
4527 # Note: This is the timestamp saved as hardcoded wikitext to
4528 # the database, we use $wgContLang here in order to give
4529 # everyone the same signature and use the default one rather
4530 # than the one selected in each user's preferences.
4531 # (see also bug 12815)
4537 }
4545 # Allow translation of timezones through wiki. date() can return
4546 # whatever crap the system uses, localised or not, so we cannot
4547 # ship premade translations.
4552 }
4558 # Variable replacement
4559 # Because mOutputType is OT_WIKI, this will only process {{subst:xxx}} type tags
4562 # This works almost by chance, as the replaceVariables are done before the getUserSig(),
4563 # which may corrupt this parser instance via its wfMessage()->text() call-
4565 # Signatures
4571 ) );
4573 # Context links ("pipe tricks"): [[|name]] and [[name (context)|]]
4578 $p4 = "/\[\[(:?$nc+:|:|)($tc+?)( ?（$tc+）)\\|]]/"; # [[ns:page（context）|]] (double-width brackets, added in r40257)
4579 $p3 = "/\[\[(:?$nc+:|:|)($tc+?)( ?\\($tc+\\)|)((?:, |，)$tc+|)\\|]]/"; # [[ns:page (context), context|]] (using either single or double-width comma)
4582 # try $p1 first, to turn "[[A, B (C)|]]" into "[[A, B (C)|A, B]]"
4594 # if there's no context, don't bother duplicating the title
4596 }
4598 # Trim trailing whitespace
4602 }
4604 /**
4605 * Fetch the user's signature text, if any, and normalize to
4606 * validated, ready-to-insert wikitext.
4607 * If you have pre-fetched the nickname or the fancySig option, you can
4608 * specify them here to save a database query.
4609 * Do not reuse this parser instance after calling getUserSig(),
4610 * as it may have changed if it's the $wgParser.
4611 *
4612 * @param $user User
4613 * @param string|bool $nickname nickname to use or false to use user's default nickname
4614 * @param $fancySig Boolean|null whether the nicknname is the complete signature
4615 * or null to use default value
4616 * @return string
4617 */
4623 # If not given, retrieve from the user object.
4626 }
4630 }
4638 # Sig. might contain markup; validate this
4640 # Validated; clean up (if needed) and return it
4643 # Failed to validate; fall back to the default
4646 }
4647 }
4649 # Make sure nickname doesnt get a sig in a sig
4652 # If we're still here, make it a link to the user page
4657 return wfMessage( $msgName, $userText, $nickText )->inContentLanguage()->title( $this->getTitle() )->text();
4658 }
4660 /**
4661 * Check that the user's signature contains no bad XML
4662 *
4663 * @param $text String
4664 * @return mixed An expanded string, or false if invalid.
4665 */
4668 }
4670 /**
4671 * Clean up signature text
4672 *
4673 * 1) Strip ~~~, ~~~~ and ~~~~~ out of signatures @see cleanSigInSig
4674 * 2) Substitute all transclusions
4675 *
4676 * @param $text String
4677 * @param bool $parsing Whether we're cleaning (preferences save) or parsing
4678 * @return String: signature text
4679 */
4684 }
4686 # Option to disable this feature
4689 }
4691 # @todo FIXME: Regex doesn't respect extension tags or nowiki
4692 # => Move this logic to braceSubstitution()
4694 $substRegex = '/\{\{(?!(?:' . $substWord->getBaseRegex() . '))/x' . $substWord->getRegexCase();
4705 }
4708 }
4710 /**
4711 * Strip ~~~, ~~~~ and ~~~~~ out of signatures
4712 *
4713 * @param $text String
4714 * @return String: signature text with /~{3,5}/ removed
4715 */
4719 }
4721 /**
4722 * Set up some variables which are usually set up in parse()
4723 * so that an external function can call some class members with confidence
4724 *
4725 * @param $title Title|null
4726 * @param $options ParserOptions
4727 * @param $outputType
4728 * @param $clearState bool
4729 */
4730 public function startExternalParse( Title $title = null, ParserOptions $options, $outputType, $clearState = true ) {
4732 }
4734 /**
4735 * @param $title Title|null
4736 * @param $options ParserOptions
4737 * @param $outputType
4738 * @param $clearState bool
4739 */
4740 private function startParse( Title $title = null, ParserOptions $options, $outputType, $clearState = true ) {
4746 }
4747 }
4749 /**
4750 * Wrapper for preprocess()
4751 *
4752 * @param string $text the text to preprocess
4753 * @param $options ParserOptions: options
4754 * @param $title Title object or null to use $wgTitle
4755 * @return String
4756 */
4760 # Guard against infinite recursion
4763 }
4770 }
4777 }
4779 /**
4780 * Create an HTML-style tag, e.g. "<yourtag>special text</yourtag>"
4781 * The callback should have the following form:
4782 * function myParserHook( $text, $params, $parser, $frame ) { ... }
4783 *
4784 * Transform and return $text. Use $parser for any required context, e.g. use
4785 * $parser->getTitle() and $parser->getOptions() not $wgTitle or $wgOut->mParserOptions
4786 *
4787 * Hooks may return extended information by returning an array, of which the
4788 * first numbered element (index 0) must be the return string, and all other
4789 * entries are extracted into local variables within an internal function
4790 * in the Parser class.
4791 *
4792 * This interface (introduced r61913) appears to be undocumented, but
4793 * 'markerName' is used by some core tag hooks to override which strip
4794 * array their results are placed in. **Use great caution if attempting
4795 * this interface, as it is not documented and injudicious use could smash
4796 * private variables.**
4797 *
4798 * @param $tag Mixed: the tag to use, e.g. 'hook' for "<hook>"
4799 * @param $callback Mixed: the callback function (and object) to use for the tag
4800 * @throws MWException
4801 * @return Mixed|null The old value of the mTagHooks array associated with the hook
4802 */
4807 }
4812 }
4815 }
4817 /**
4818 * As setHook(), but letting the contents be parsed.
4819 *
4820 * Transparent tag hooks are like regular XML-style tag hooks, except they
4821 * operate late in the transformation sequence, on HTML instead of wikitext.
4822 *
4823 * This is probably obsoleted by things dealing with parser frames?
4824 * The only extension currently using it is geoserver.
4825 *
4826 * @since 1.10
4827 * @todo better document or deprecate this
4828 *
4829 * @param $tag Mixed: the tag to use, e.g. 'hook' for "<hook>"
4830 * @param $callback Mixed: the callback function (and object) to use for the tag
4831 * @throws MWException
4832 * @return Mixed|null The old value of the mTagHooks array associated with the hook
4833 */
4838 }
4839 $oldVal = isset( $this->mTransparentTagHooks[$tag] ) ? $this->mTransparentTagHooks[$tag] : null;
4843 }
4845 /**
4846 * Remove all tag hooks
4847 */
4852 }
4854 /**
4855 * Create a function, e.g. {{sum:1|2|3}}
4856 * The callback function should have the form:
4857 * function myParserFunction( &$parser, $arg1, $arg2, $arg3 ) { ... }
4858 *
4859 * Or with SFH_OBJECT_ARGS:
4860 * function myParserFunction( $parser, $frame, $args ) { ... }
4861 *
4862 * The callback may either return the text result of the function, or an array with the text
4863 * in element 0, and a number of flags in the other elements. The names of the flags are
4864 * specified in the keys. Valid flags are:
4865 * found The text returned is valid, stop processing the template. This
4866 * is on by default.
4867 * nowiki Wiki markup in the return value should be escaped
4868 * isHTML The returned text is HTML, armour it against wikitext transformation
4869 *
4870 * @param string $id The magic word ID
4871 * @param $callback Mixed: the callback function (and object) to use
4872 * @param $flags Integer: a combination of the following flags:
4873 * SFH_NO_HASH No leading hash, i.e. {{plural:...}} instead of {{#if:...}}
4874 *
4875 * SFH_OBJECT_ARGS Pass the template arguments as PPNode objects instead of text. This
4876 * allows for conditional expansion of the parse tree, allowing you to eliminate dead
4877 * branches and thus speed up parsing. It is also possible to analyse the parse tree of
4878 * the arguments, and to control the way they are expanded.
4879 *
4880 * The $frame parameter is a PPFrame. This can be used to produce expanded text from the
4881 * arguments, for instance:
4882 * $text = isset( $args[0] ) ? $frame->expand( $args[0] ) : '';
4883 *
4884 * For technical reasons, $args[0] is pre-expanded and will be a string. This may change in
4885 * future versions. Please call $frame->expand() on it anyway so that your code keeps
4886 * working if/when this is changed.
4887 *
4888 * If you want whitespace to be trimmed from $args, you need to do it yourself, post-
4889 * expansion.
4890 *
4891 * Please read the documentation in includes/parser/Preprocessor.php for more information
4892 * about the methods available in PPFrame and PPNode.
4893 *
4894 * @throws MWException
4895 * @return string|callback The old callback function for this name, if any
4896 */
4903 # Add to function cache
4907 }
4913 # Case
4916 }
4917 # Add leading hash
4920 }
4921 # Remove trailing colon
4924 }
4926 }
4928 }
4930 /**
4931 * Get all registered function hook identifiers
4932 *
4933 * @return Array
4934 */
4937 }
4939 /**
4940 * Create a tag function, e.g. "<test>some stuff</test>".
4941 * Unlike tag hooks, tag functions are parsed at preprocessor level.
4942 * Unlike parser functions, their content is not preprocessed.
4943 * @param $tag
4944 * @param $callback
4945 * @param $flags
4946 * @throws MWException
4947 * @return null
4948 */
4953 }
4960 }
4963 }
4965 /**
4966 * @todo FIXME: Update documentation. makeLinkObj() is deprecated.
4967 * Replace "<!--LINK-->" link placeholders with actual links, in the buffer
4968 * Placeholders created in Skin::makeLinkObj()
4969 *
4970 * @param $text string
4971 * @param $options int
4972 *
4973 * @return array of link CSS classes, indexed by PDBK.
4974 */
4977 }
4979 /**
4980 * Replace "<!--LINK-->" link placeholders with plain text of links
4981 * (not HTML-formatted).
4982 *
4983 * @param $text String
4984 * @return String
4985 */
4988 }
4990 /**
4991 * Renders an image gallery from a text with one line per image.
4992 * text labels may be given by using |-style alternative text. E.g.
4993 * Image:one.jpg|The number "1"
4994 * Image:tree.jpg|A tree
4995 * given as text will return the HTML of a gallery with two images,
4996 * labeled 'The number "1"' and
4997 * 'A tree'.
4998 *
4999 * @param string $text
5000 * @param array $params
5001 * @return string HTML
5002 */
5016 }
5022 }
5025 }
5028 }
5031 }
5037 # match lines like these:
5038 # Image:someimage.jpg|This is some image
5041 # Skip empty lines
5044 }
5048 }
5051 # Bogus title. Ignore these so we don't bomb out later.
5053 }
5059 // look for an |alt= definition while trying not to break existing
5060 // captions with multiple pipes (|) in it, until a more sensible grammar
5061 // is defined for images in galleries
5071 }
5076 //check to see if link matches an absolute url, if not then it must be a wiki link.
5083 }
5084 }
5085 }
5087 // concatenate all other pipes
5089 }
5090 }
5091 // remove the first pipe
5093 }
5096 }
5098 }
5100 /**
5101 * @param $handler
5102 * @return array
5103 */
5109 }
5111 # Initialise static lists
5118 );
5126 }
5127 }
5128 }
5130 # Add handler params
5136 }
5137 }
5140 }
5141 return array( $this->mImageParams[$handlerClass], $this->mImageParamsMagicArray[$handlerClass] );
5142 }
5144 /**
5145 * Parse image options text and use it to make an image
5146 *
5147 * @param $title Title
5148 * @param $options String
5149 * @param $holders LinkHolderArray|bool
5150 * @return string HTML
5151 */
5153 # Check if the options text is of the form "options|alt text"
5154 # Options are:
5155 # * thumbnail make a thumbnail with enlarge-icon and caption, alignment depends on lang
5156 # * left no resizing, just left align. label is used for alt= only
5157 # * right same, but right aligned
5158 # * none same, but not aligned
5159 # * ___px scale to ___ pixels width, no aligning. e.g. use in taxobox
5160 # * center center the image
5161 # * frame Keep original image size, no magnify-button.
5162 # * framed Same as "frame"
5163 # * frameless like 'thumb' but without a frame. Keeps user preferences for width
5164 # * upright reduce width for upright images, rounded to full __0 px
5165 # * border draw a 1px border around the image
5166 # * alt Text for HTML alt attribute (defaults to empty)
5167 # * class Set a class for img node
5168 # * link Set the target of the image link. Can be external, interwiki, or local
5169 # vertical-align values (no % or length right now):
5170 # * baseline
5171 # * sub
5172 # * super
5173 # * top
5174 # * text-top
5175 # * middle
5176 # * bottom
5177 # * text-bottom
5181 # Give extensions a chance to select the file revision for us
5186 # Fetch and register the file (file title may be different via hooks)
5189 # Get parameter map
5196 }
5198 # Process the input parameters
5209 # Special case; width and height come in one variable together
5217 }
5218 }
5224 }
5225 }
5226 # else no validation -- bug 13436
5229 # Validate handler parameter
5232 # Validate internal parameters
5237 # @todo FIXME: Possibly check validity here for
5238 # manualthumb? downstream behavior seems odd with
5239 # missing manual thumbs.
5256 }
5258 }
5266 }
5267 }
5270 # Most other things appear to be empty or numeric...
5272 }
5273 }
5277 }
5278 }
5279 }
5282 }
5283 }
5285 # Process alignment parameters
5288 }
5291 }
5295 # Will the image be presented in a frame, with the caption below?
5301 # In the old days, [[Image:Foo|text...]] would set alt text. Later it
5302 # came to also set the caption, ordinary text after the image -- which
5303 # makes no sense, because that just repeats the text multiple times in
5304 # screen readers. It *also* came to set the title attribute.
5305 #
5306 # Now that we have an alt attribute, we should not set the alt text to
5307 # equal the caption: that's worse than useless, it just repeats the
5308 # text. This is the framed/thumbnail case. If there's no caption, we
5309 # use the unnamed parameter for alt text as well, just for the time be-
5310 # ing, if the unnamed param is set and the alt param is not.
5311 #
5312 # For the future, we need to figure out if we want to tweak this more,
5313 # e.g., introducing a title= parameter for the title; ignoring the un-
5314 # named parameter entirely for images without a caption; adding an ex-
5315 # plicit caption= parameter and preserving the old magic unnamed para-
5316 # meter for BC; ...
5319 # No caption or alt text, add the filename as the alt text so
5320 # that screen readers at least get some description of the image
5322 }
5323 # Do not set $params['frame']['title'] because tooltips don't make sense
5324 # for framed images
5327 # No alt text, use the "caption" for the alt text
5331 # No caption, fall back to using the filename for the
5332 # alt text
5334 }
5335 }
5336 # Use the "caption" for the tooltip text
5338 }
5342 # Linker does the rest
5347 # Give the handler a chance to modify the parser object
5350 }
5353 }
5355 /**
5356 * @param $caption
5357 * @param $holders LinkHolderArray
5358 * @return mixed|String
5359 */
5361 # Strip bad stuff out of the title (tooltip). We can't just use
5362 # replaceLinkHoldersText() here, because if this function is called
5363 # from replaceInternalLinks2(), mLinkHolders won't be up-to-date.
5368 }
5370 # make sure there are no placeholders in thumbnail attributes
5371 # that are later expanded to html- so expand them now and
5372 # remove the tags
5377 }
5379 /**
5380 * Set a flag in the output object indicating that the content is dynamic and
5381 * shouldn't be cached.
5382 */
5388 }
5391 }
5393 /**
5394 * Callback from the Sanitizer for expanding items found in HTML attribute
5395 * values, so they can be safely tested and escaped.
5396 *
5397 * @param $text String
5398 * @param $frame PPFrame
5399 * @return String
5400 */
5405 }
5407 /**
5408 * Accessor
5409 *
5410 * @return array
5411 */
5413 return array_merge( array_keys( $this->mTransparentTagHooks ), array_keys( $this->mTagHooks ), array_keys( $this->mFunctionTagHooks ) );
5414 }
5416 /**
5417 * Replace transparent tags in $text with the values given by the callbacks.
5418 *
5419 * Transparent tag hooks are like regular XML-style tag hooks, except they
5420 * operate late in the transformation sequence, on HTML instead of wikitext.
5421 *
5422 * @param $text string
5423 *
5424 * @return string
5425 */
5436 $output = call_user_func_array( $this->mTransparentTagHooks[$tagName], array( $content, $params, $this ) );
5439 }
5441 }
5443 }
5445 /**
5446 * Break wikitext input into sections, and either pull or replace
5447 * some particular section's text.
5448 *
5449 * External callers should use the getSection and replaceSection methods.
5450 *
5451 * @param string $text Page wikitext
5452 * @param string $section a section identifier string of the form:
5453 * "<flag1> - <flag2> - ... - <section number>"
5454 *
5455 * Currently the only recognised flag is "T", which means the target section number
5456 * was derived during a template inclusion parse, in other words this is a template
5457 * section edit link. If no flags are given, it was an ordinary section edit link.
5458 * This flag is required to avoid a section numbering mismatch when a section is
5459 * enclosed by "<includeonly>" (bug 6563).
5460 *
5461 * The section number 0 pulls the text before the first heading; other numbers will
5462 * pull the given section along with its lower-level subsections. If the section is
5463 * not found, $mode=get will return $newtext, and $mode=replace will return $text.
5464 *
5465 * Section 0 is always considered to exist, even if it only contains the empty
5466 * string. If $text is the empty string and section 0 is replaced, $newText is
5467 * returned.
5468 *
5469 * @param string $mode one of "get" or "replace"
5470 * @param string $newText replacement text for section data.
5471 * @return String: for "get", the extracted section text.
5472 * for "replace", the whole page with the section replaced.
5473 */
5480 # Process section extraction flags
5487 }
5488 }
5490 # Check for empty input
5492 # Only sections 0 and T-0 exist in an empty document
5498 }
5504 }
5505 }
5506 }
5508 # Preprocess the text
5511 # <h> nodes indicate section breaks
5512 # They can only occur at the top level, so we can find them by iterating the root's children
5515 # Find the target section
5517 # Section zero doesn't nest, level=big
5526 }
5527 }
5530 }
5532 }
5533 }
5536 # Not found
5541 }
5542 }
5544 # Find the end of the section, including nested sections
5551 }
5552 }
5555 }
5559 # Write out the remainder (in replace mode only)
5561 # Output the replacement text
5562 # Add two newlines on -- trailing whitespace in $newText is conventionally
5563 # stripped by the editor, so we need both newlines to restore the paragraph gap
5564 # Only add trailing whitespace if there is newText
5567 }
5572 }
5573 }
5576 # Re-insert stripped tags
5578 }
5581 }
5583 /**
5584 * This function returns the text of a section, specified by a number ($section).
5585 * A section is text under a heading like == Heading == or \<h1\>Heading\</h1\>, or
5586 * the first section before any such heading (section 0).
5587 *
5588 * If a section contains subsections, these are also returned.
5589 *
5590 * @param string $text text to look in
5591 * @param string $section section identifier
5592 * @param string $deftext default to return if section is not found
5593 * @return string text of the requested section
5594 */
5597 }
5599 /**
5600 * This function returns $oldtext after the content of the section
5601 * specified by $section has been replaced with $text. If the target
5602 * section does not exist, $oldtext is returned unchanged.
5603 *
5604 * @param string $oldtext former text of the article
5605 * @param int $section section identifier
5606 * @param string $text replacing text
5607 * @return String: modified text
5608 */
5611 }
5613 /**
5614 * Get the ID of the revision we are parsing
5615 *
5616 * @return Mixed: integer or null
5617 */
5620 }
5622 /**
5623 * Get the revision object for $this->mRevisionId
5624 *
5625 * @return Revision|null either a Revision object or null
5626 */
5630 }
5633 }
5637 }
5639 /**
5640 * Get the timestamp associated with the current revision, adjusted for
5641 * the default server-local timestamp
5642 */
5652 # The cryptic '' timezone parameter tells to use the site-default
5653 # timezone offset instead of the user settings.
5654 #
5655 # Since this value will be saved into the parser cache, served
5656 # to other users, and potentially even used inside links and such,
5657 # it needs to be consistent for all visitors.
5661 }
5663 }
5665 /**
5666 * Get the name of the user that edited the last revision
5667 *
5668 * @return String: user name
5669 */
5674 # if this template is subst: the revision id will be blank,
5675 # so just use the current user's name
5680 }
5681 }
5683 }
5685 /**
5686 * Mutator for $mDefaultSort
5687 *
5688 * @param string $sort New value
5689 */
5693 }
5695 /**
5696 * Accessor for $mDefaultSort
5697 * Will use the empty string if none is set.
5698 *
5699 * This value is treated as a prefix, so the
5700 * empty string is equivalent to sorting by
5701 * page name.
5702 *
5703 * @return string
5704 */
5710 }
5711 }
5713 /**
5714 * Accessor for $mDefaultSort
5715 * Unlike getDefaultSort(), will return false if none is set
5716 *
5717 * @return string or false
5718 */
5721 }
5723 /**
5724 * Try to guess the section anchor name based on a wikitext fragment
5725 * presumably extracted from a heading, for example "Header" from
5726 * "== Header ==".
5727 *
5728 * @param $text string
5729 *
5730 * @return string
5731 */
5733 # Strip out wikitext links(they break the anchor)
5737 }
5739 /**
5740 * Same as guessSectionNameFromWikiText(), but produces legacy anchors
5741 * instead. For use in redirects, since IE6 interprets Redirect: headers
5742 * as something other than UTF-8 (apparently?), resulting in breakage.
5743 *
5744 * @param string $text The section name
5745 * @return string An anchor
5746 */
5748 # Strip out wikitext links(they break the anchor)
5752 }
5754 /**
5755 * Strips a text string of wikitext for use in a section anchor
5756 *
5757 * Accepts a text string and then removes all wikitext from the
5758 * string and leaves only the resultant text (i.e. the result of
5759 * [[User:WikiSysop|Sysop]] would be "Sysop" and the result of
5760 * [[User:WikiSysop]] would be "User:WikiSysop") - this is intended
5761 * to create valid section anchors by mimicing the output of the
5762 * parser when headings are parsed.
5763 *
5764 * @param string $text text string to be stripped of wikitext
5765 * for use in a Section anchor
5766 * @return string Filtered text string
5767 */
5769 # Strip internal link markup
5773 # Strip external link markup
5774 # @todo FIXME: Not tolerant to blank link text
5775 # I.E. [http://www.mediawiki.org] will render as [1] or something depending
5776 # on how many empty links there are on the page - need to figure that out.
5777 $text = preg_replace( '/\[(?i:' . $this->mUrlProtocols . ')([^ ]+?) ([^[]+)\]/', '$2', $text );
5779 # Parse wikitext quotes (italics & bold)
5782 # Strip HTML tags
5785 }
5787 /**
5788 * strip/replaceVariables/unstrip for preprocessor regression testing
5789 *
5790 * @param $text string
5791 * @param $title Title
5792 * @param $options ParserOptions
5793 * @param $outputType int
5794 *
5795 * @return string
5796 */
5797 function testSrvus( $text, Title $title, ParserOptions $options, $outputType = self::OT_HTML ) {
5804 }
5806 /**
5807 * @param $text string
5808 * @param $title Title
5809 * @param $options ParserOptions
5810 * @return string
5811 */
5814 }
5816 /**
5817 * @param $text
5818 * @param $title Title
5819 * @param $options ParserOptions
5820 * @return string
5821 */
5824 }
5826 /**
5827 * Call a callback function on all regions of the given text that are not
5828 * inside strip markers, and replace those regions with the return value
5829 * of the callback. For example, with input:
5830 *
5831 * aaa<MARKER>bbb
5832 *
5833 * This will call the callback function twice, with 'aaa' and 'bbb'. Those
5834 * two strings will be replaced with the value returned by the callback in
5835 * each case.
5836 *
5837 * @param $s string
5838 * @param $callback
5839 *
5840 * @return string
5841 */
5860 }
5861 }
5862 }
5864 }
5866 /**
5867 * Remove any strip markers found in the given text.
5868 *
5869 * @param $text Input string
5870 * @return string
5871 */
5874 }
5876 /**
5877 * Save the parser state required to convert the given half-parsed text to
5878 * HTML. "Half-parsed" in this context means the output of
5879 * recursiveTagParse() or internalParse(). This output has strip markers
5880 * from replaceVariables (extensionSubstitution() etc.), and link
5881 * placeholders from replaceLinkHolders().
5882 *
5883 * Returns an array which can be serialized and stored persistently. This
5884 * array can later be loaded into another parser instance with
5885 * unserializeHalfParsedText(). The text can then be safely incorporated into
5886 * the return value of a parser hook.
5887 *
5888 * @param $text string
5889 *
5890 * @return array
5891 */
5899 );
5902 }
5904 /**
5905 * Load the parser state given in the $data array, which is assumed to
5906 * have been generated by serializeHalfParsedText(). The text contents is
5907 * extracted from the array, and its markers are transformed into markers
5908 * appropriate for the current Parser instance. This transformed text is
5909 * returned, and can be safely included in the return value of a parser
5910 * hook.
5911 *
5912 * If the $data array has been stored persistently, the caller should first
5913 * check whether it is still valid, by calling isValidHalfParsedText().
5914 *
5915 * @param array $data Serialized data
5916 * @throws MWException
5917 * @return String
5918 */
5922 }
5924 # First, extract the strip state.
5928 # Now renumber links
5931 # Should be good to go.
5933 }
5935 /**
5936 * Returns true if the given array, presumed to be generated by
5937 * serializeHalfParsedText(), is compatible with the current version of the
5938 * parser.
5939 *
5940 * @param $data Array
5941 *
5942 * @return bool
5943 */
5946 }
5948 /**
5949 * Parsed a width param of imagelink like 300px or 200x300px
5950 *
5951 * @param $value String
5952 *
5953 * @return array
5954 * @since 1.20
5955 */
5960 }
5962 # (bug 13500) In both cases (width/height and width only),
5963 # permit trailing "px" for backward compatibility.
5972 }
5974 }
5975 }