| blob | 813aacad4dcbd75afc3c45e35fd26714e55e4540 |
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 (X)HTML 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 * $wgNamespacesWithSubpages
58 *
59 * @par Settings only within ParserOptions:
60 * $wgAllowExternalImages
61 * $wgAllowSpecialInclusion
62 * $wgInterwikiMagic
63 * $wgMaxArticleSize
64 *
65 * @ingroup Parser
66 */
68 /**
69 * Update this version number when the ParserOutput format
70 * changes in an incompatible way, so the parser cache
71 * can automatically discard old data.
72 */
75 /**
76 * Update this version number when the output of serialiseHalfParsedText()
77 * changes in an incompatible way
78 */
81 # Flags for Parser::setFunctionHook
82 # Also available as global constants from Defines.php
86 # Constants needed for external link processing
87 # Everything except bracket, space, or control characters
88 # \p{Zs} is unicode 'separator, space' category. It covers the space 0x20
89 # as well as U+3000 is IDEOGRAPHIC SPACE for bug 19052
94 # State constants for the definition list colon extraction
104 # Flags for preprocessToDom
107 # Allowed values for $this->mOutputType
108 # Parameter to startExternalParse().
115 # Marker Suffix needs to be accessible staticly.
118 # Persistent:
132 # Initialised by initialiseVariables()
134 /**
135 * @var MagicWordArray
136 */
139 /**
140 * @var MagicWordArray
141 */
143 var $mConf, $mPreprocessor, $mExtLinkBracketedRegex, $mUrlProtocols; # Initialised in constructor
145 # Cleared with clearState():
146 /**
147 * @var ParserOutput
148 */
152 /**
153 * @var StripState
154 */
158 /**
159 * @var LinkHolderArray
160 */
171 /**
172 * @var User
173 */
176 # Temporary
177 # These are variables reset at least once per parse regardless of $clearState
179 /**
180 * @var ParserOptions
181 */
184 /**
185 * @var Title
186 */
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 under HipHop
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 */
366 # Remove the strip marker tag prefix from the input, if present.
369 }
380 }
383 # No more strip!
390 # Clean up special characters, only run once, next-to-last before doBlockLevels
392 # french spaces, last one Guillemet-left
393 # only if there is something before the space
395 # french spaces, Guillemet-right
398 );
405 /**
406 * The input doesn't get language converted if
407 * a) It's disabled
408 * b) Content isn't converted
409 * c) It's a conversion table
410 * d) it is an interface message (which is in the user language)
411 */
414 {
416 # The position of the convert() call should not be changed. it
417 # assumes that the links are all replaced and the only thing left
418 # is the <nowiki> mark.
420 }
421 }
423 /**
424 * A converted title will be provided in the output object if title and
425 * content conversion are enabled, the article text does not contain
426 * a conversion-suppressing double-underscore tag, and no
427 * {{DISPLAYTITLE:...}} is present. DISPLAYTITLE takes precedence over
428 * automatic link conversion.
429 */
434 {
441 }
442 }
456 # attempt to sanitize at least some nesting problems
457 # (bug #2702 and quite a few others)
459 # ''Something [http://www.cool.com cool''] -->
460 # <i>Something</i><a href="http://www.cool.com"..><i>cool></i></a>
463 # fix up an anchor inside another anchor, only
464 # at least for a single single nested link (bug 3695)
467 # fix div inside inline elements- doBlockLevels won't wrap a line which
468 # contains a div, so fix it up here; replace
469 # div with escaped text
472 # remove empty italic or bold tag pairs, some
473 # introduced by rules above
475 );
481 }
487 );
488 }
492 # Information on include size limits, for the benefit of users who try to skirt them
495 $PFreport = "Expensive parser function count: {$this->mExpensiveFunctionCount}/{$this->mOptions->getExpensiveParserFunctionLimit()}\n";
498 "Preprocessor visited node count: {$this->mPPNodeCount}/{$this->mOptions->getMaxPPNodeCount()}\n" .
503 "Highest expansion depth: {$this->mHighestExpansionDepth}/{$this->mOptions->getMaxPPExpandDepth()}\n" .
507 // Sanitize for comment. Note '‐' in the replacement is U+2010,
508 // which looks much like the problematic '-'.
516 }
517 }
529 }
531 /**
532 * Recursive parser entry point that can be called from an extension tag
533 * hook.
534 *
535 * If $frame is not provided, then template variables (e.g., {{{1}}}) within $text are not expanded
536 *
537 * @param string $text text extension wants to have parsed
538 * @param $frame PPFrame: The frame to use for expanding any template variables
539 *
540 * @return string
541 */
549 }
551 /**
552 * Expand templates and variables in the text, producing valid, static wikitext.
553 * Also removes comments.
554 * @return mixed|string
555 */
561 }
568 }
570 /**
571 * Recursive parser entry point that can be called from an extension tag
572 * hook.
573 *
574 * @param string $text text to be expanded
575 * @param $frame PPFrame: The frame to use for expanding any template variables
576 * @return String
577 * @since 1.19
578 */
585 }
587 /**
588 * Process the wikitext for the "?preload=" feature. (bug 5210)
589 *
590 * "<noinclude>", "<includeonly>" etc. are parsed as for template
591 * transclusion, comments, templates, arguments, tags hooks and parser
592 * functions are untouched.
593 *
594 * @param $text String
595 * @param $title Title
596 * @param $options ParserOptions
597 * @return String
598 */
600 # Parser (re)initialisation
608 }
610 /**
611 * Get a random string
612 *
613 * @return string
614 */
617 }
619 /**
620 * Set the current user.
621 * Should only be used when doing pre-save transform.
622 *
623 * @param $user Mixed: User object or null (to reset)
624 */
627 }
629 /**
630 * Accessor for mUniqPrefix.
631 *
632 * @return String
633 */
636 # @todo FIXME: This is probably *horribly wrong*
637 # LanguageConverter seems to want $wgParser's uniqPrefix, however
638 # if this is called for a parser cache hit, the parser may not
639 # have ever been initialized in the first place.
640 # Not really sure what the heck is supposed to be going on here.
642 # throw new MWException( "Accessing uninitialized mUniqPrefix" );
643 }
645 }
647 /**
648 * Set the context title
649 *
650 * @param $t Title
651 */
655 }
658 # Strip the fragment to avoid various odd effects
663 }
664 }
666 /**
667 * Accessor for the Title object
668 *
669 * @return Title object
670 */
673 }
675 /**
676 * Accessor/mutator for the Title object
677 *
678 * @param $x Title object or null to just get the current one
679 * @return Title object
680 */
683 }
685 /**
686 * Set the output type
687 *
688 * @param $ot Integer: new value
689 */
692 # Shortcut alias
698 );
699 }
701 /**
702 * Accessor/mutator for the output type
703 *
704 * @param int|null $x New value or null to just get the current one
705 * @return Integer
706 */
709 }
711 /**
712 * Get the ParserOutput object
713 *
714 * @return ParserOutput object
715 */
718 }
720 /**
721 * Get the ParserOptions object
722 *
723 * @return ParserOptions object
724 */
727 }
729 /**
730 * Accessor/mutator for the ParserOptions object
731 *
732 * @param $x ParserOptions New value or null to just get the current one
733 * @return ParserOptions Current ParserOptions object
734 */
737 }
739 /**
740 * @return int
741 */
744 }
746 /**
747 * @param $id int
748 */
751 }
753 /**
754 * Get a language object for use in parser functions such as {{FORMATNUM:}}
755 * @return Language
756 */
759 }
761 /**
762 * Get the target language for the content being parsed. This is usually the
763 * language that the content is in.
764 *
765 * @since 1.19
766 *
767 * @throws MWException
768 * @return Language|null
769 */
779 }
782 }
784 /**
785 * Get the language object for language conversion
786 */
789 }
791 /**
792 * Get a User object either from $this->mUser, if set, or from the
793 * ParserOptions object otherwise
794 *
795 * @return User object
796 */
800 }
802 }
804 /**
805 * Get a preprocessor object
806 *
807 * @return Preprocessor instance
808 */
813 }
815 }
817 /**
818 * Replaces all occurrences of HTML-style comments and the given tags
819 * in the text with a random marker and returns the next text. The output
820 * parameter $matches will be an associative array filled with data in
821 * the form:
822 *
823 * @code
824 * 'UNIQ-xxxxx' => array(
825 * 'element',
826 * 'tag content',
827 * array( 'param' => 'x' ),
828 * '<element param="x">tag content</element>' ) )
829 * @endcode
830 *
831 * @param array $elements list of element names. Comments are always extracted.
832 * @param string $text Source text string.
833 * @param array $matches Out parameter, Array: extracted tags
834 * @param $uniq_prefix string
835 * @return String: stripped text
836 */
850 }
852 # comment
858 # tag
863 }
869 # Empty element tag, <tag />
878 }
882 # No end tag -- let it run out to the end of the text.
888 }
889 }
895 }
897 }
899 /**
900 * Get a list of strippable XML-like elements
901 *
902 * @return array
903 */
906 }
908 /**
909 * Add an item to the strip state
910 * Returns the unique tag which must be inserted into the stripped text
911 * The tag will be replaced with the original text in unstrip()
912 *
913 * @param $text string
914 *
915 * @return string
916 */
922 }
924 /**
925 * parse the wiki syntax used to render tables
926 *
927 * @private
928 * @return string
929 */
948 }
954 # First check if we are starting a new table
967 # Don't do any of the following
971 # We are ending a table
977 }
981 }
985 }
989 # Now we have a table row
992 # Whats after the tag is now only attributes
1005 }
1009 }
1015 } elseif ( $first_character === '|' || $first_character === '!' || substr( $line, 0, 2 ) === '|+' ) {
1016 # This might be cell elements, td, th or captions
1020 }
1026 }
1028 # Split up multiple cells on the same line.
1029 # FIXME : This can result in improper nesting of tags processed
1030 # by earlier parser steps, but should avoid splitting up eg
1031 # attribute values containing literal "||".
1036 # Loop through each table cell
1043 }
1048 }
1054 }
1064 }
1068 # A cell could contain both parameters and data
1071 # Bug 553: Note that a '|' inside an invalid link should not
1072 # be mistaken as delimiting cell parameters
1081 }
1085 }
1086 }
1088 }
1090 # Closing open td, tr && table
1094 }
1097 }
1100 }
1103 }
1105 # Remove trailing line-ending (b/c)
1108 }
1110 # special case: don't return empty table
1113 }
1118 }
1120 /**
1121 * Helper function for parse() that transforms wiki markup into
1122 * HTML. Only called for $mOutputType == self::OT_HTML.
1123 *
1124 * @private
1125 *
1126 * @param $text string
1127 * @param $isMain bool
1128 * @param $frame bool
1129 *
1130 * @return string
1131 */
1137 # Hook to suspend the parser in this state
1138 if ( !wfRunHooks( 'ParserBeforeInternalParse', array( &$this, &$text, &$this->mStripState ) ) ) {
1141 }
1143 # if $frame is provided, then use $frame for replacing any variables
1145 # use frame depth to infer how include/noinclude tags should be handled
1146 # depth=0 means this is the top-level document; otherwise it's an included document
1151 }
1155 # if $frame is not provided, then use old-style replaceVariables
1157 }
1160 $text = Sanitizer::removeHTMLtags( $text, array( &$this, 'attributeStripCallback' ), false, array_keys( $this->mTransparentTagHooks ) );
1163 # Tables need to come after variable replacement for things to work
1164 # properly; putting them before other transformations should keep
1165 # exciting things like link expansions from showing up in surprising
1166 # places.
1178 # replaceInternalLinks may sometimes leave behind
1179 # absolute URLs, which have to be masked to hide them from replaceExternalLinks
1187 }
1189 /**
1190 * Replace special strings like "ISBN xxx" and "RFC xxx" with
1191 * magic external links.
1192 *
1193 * DML
1194 * @private
1195 *
1196 * @param $text string
1197 *
1198 * @return string
1199 */
1205 '!(?: # Start cases
1209 (?:RFC|PMID)\s+([0-9]+) | # m[4]: RFC or PMID, capture number
1211 (?: 97[89] [\ \-]? )? # optional 13-digit ISBN prefix
1212 (?: [0-9] [\ \-]? ){9} # 9 digits with opt. delimiters
1213 [0-9Xx] # check digit
1218 }
1220 /**
1221 * @throws MWException
1222 * @param $m array
1223 * @return HTML|string
1224 */
1227 # Skip anchor
1230 # Skip HTML element
1233 # Free external link
1236 # RFC or PMID
1250 }
1254 # ISBN
1260 ));
1267 }
1268 }
1270 /**
1271 * Make a free external link, given a user-supplied URL
1272 *
1273 * @param $url string
1274 *
1275 * @return string HTML
1276 * @private
1277 */
1283 # The characters '<' and '>' (which were escaped by
1284 # removeHTMLtags()) should not be included in
1285 # URLs, per RFC 2396.
1290 }
1292 # Move trailing punctuation to $trail
1294 # If there is no left bracket, then consider right brackets fair game too
1297 }
1303 }
1307 # Is this an external image?
1310 # Not an image, make a link
1315 # Register it in the output object...
1316 # Replace unnecessary URL escape codes with their equivalent characters
1319 }
1322 }
1324 /**
1325 * Parse headers and return html
1326 *
1327 * @private
1328 *
1329 * @param $text string
1330 *
1331 * @return string
1332 */
1338 }
1341 }
1343 /**
1344 * Replace single quotes with HTML markup
1345 * @private
1346 *
1347 * @param $text string
1348 *
1349 * @return string the altered text
1350 */
1357 }
1361 }
1363 /**
1364 * Helper function for doAllQuotes()
1365 *
1366 * @param $text string
1367 *
1368 * @return string
1369 */
1375 # First, do some preliminary work. This may shift some apostrophes from
1376 # being mark-up to being text. It also counts the number of occurrences
1377 # of bold and italics mark-ups.
1382 # If there are ever four apostrophes, assume the first is supposed to
1383 # be text, and the remaining three constitute mark-up for bold text.
1388 # If there are more than 5 apostrophes in a row, assume they're all
1389 # text except for the last 5.
1392 }
1393 # Count the number of occurrences of bold and italics mark-ups.
1394 # We are not counting sequences of five apostrophes.
1402 }
1403 }
1404 }
1406 # If there is an odd number of both bold and italics, it is likely
1407 # that one of the bold ones was meant to be an apostrophe followed
1408 # by italics. Which one we cannot know for certain, but it is more
1409 # likely to be one that has a single-letter word before it.
1422 }
1426 }
1430 }
1431 }
1432 }
1434 }
1436 # If there is a single-letter word, use it!
1441 # If not, but there's a multi-letter word, use that one.
1445 # ... otherwise use the first one that has neither.
1446 # (notice that it is possible for all three to be -1 if, for example,
1447 # there is only one pentuple-apostrophe in the line)
1450 }
1451 }
1453 # Now let's actually convert our apostrophic mush to HTML!
1464 }
1482 }
1499 }
1519 }
1520 }
1521 }
1523 }
1524 # Now close all remaining tags. Notice that the order is important.
1527 }
1530 }
1533 }
1534 # There might be lonely ''''', so make sure we have a buffer
1537 }
1539 }
1540 }
1542 /**
1543 * Replace external links (REL)
1544 *
1545 * Note: this is all very hackish and the order of execution matters a lot.
1546 * Make sure to run maintenance/parserTests.php if you change this code.
1547 *
1548 * @private
1549 *
1550 * @param $text string
1551 *
1552 * @throws MWException
1553 * @return string
1554 */
1561 throw new MWException( "PCRE needs to be compiled with --enable-unicode-properties in order for MediaWiki to function" );
1562 }
1572 # The characters '<' and '>' (which were escaped by
1573 # removeHTMLtags()) should not be included in
1574 # URLs, per RFC 2396.
1579 }
1581 # If the link text is an image URL, replace it with an <img> tag
1582 # This happened by accident in the original parser, but some people used it extensively
1586 }
1590 # Set linktype for CSS - if URL==text, link is essentially free
1593 # No link text, e.g. [http://domain.tld/some.link]
1595 # Autonumber
1600 # Have link text, e.g. [http://domain.tld/some.link text]s
1601 # Check for trail
1603 }
1609 # Use the encoded URL
1610 # This means that users can paste URLs directly into the text
1611 # Funny characters like ö aren't valid in URLs anyway
1612 # This was changed in August 2004
1616 # Register link in the output object.
1617 # Replace unnecessary URL escape codes with the referenced character
1618 # This prevents spammers from hiding links from the filters
1621 }
1625 }
1626 /**
1627 * Get the rel attribute for a particular external link.
1628 *
1629 * @since 1.21
1630 * @param string|bool $url optional URL, to extract the domain from for rel =>
1631 * nofollow if appropriate
1632 * @param $title Title optional Title, for wgNoFollowNsExceptions lookups
1633 * @return string|null rel attribute for $url
1634 */
1640 {
1642 }
1644 }
1645 /**
1646 * Get an associative array of additional HTML attributes appropriate for a
1647 * particular external link. This currently may include rel => nofollow
1648 * (depending on configuration, namespace, and the URL's domain) and/or a
1649 * target attribute (depending on configuration).
1650 *
1651 * @param string|bool $url optional URL, to extract the domain from for rel =>
1652 * nofollow if appropriate
1653 * @return Array associative array of HTML attributes
1654 */
1661 }
1663 }
1665 /**
1666 * Replace unusual URL escape codes with their equivalent characters
1667 *
1668 * @param $url String
1669 * @return String
1670 *
1671 * @todo This can merge genuinely required bits in the path or query string,
1672 * breaking legit URLs. A proper fix would treat the various parts of
1673 * the URL differently; as a workaround, just use the output for
1674 * statistical records, not for actual linking/output.
1675 */
1679 }
1681 /**
1682 * Callback function used in replaceUnusualEscapes().
1683 * Replaces unusual URL escape codes with their equivalent character
1684 *
1685 * @param $matches array
1686 *
1687 * @return string
1688 */
1692 # Is it an unsafe or HTTP reserved character according to RFC 1738?
1694 # No, shouldn't be escaped
1697 # Yes, leave it escaped
1699 }
1700 }
1702 /**
1703 * make an image if it's allowed, either through the global
1704 * option, through the exception, or through the on-wiki whitelist
1705 * @private
1706 *
1707 * $param $url string
1708 *
1709 * @return string
1710 */
1715 # $imagesfrom could be either a single string or an array of strings, parse out the latter
1722 }
1723 }
1728 }
1732 # Image found
1734 }
1735 }
1738 $whitelist = explode( "\n", wfMessage( 'external_image_whitelist' )->inContentLanguage()->text() );
1740 # Sanitize the regex fragment, make it case-insensitive, ignore blank entries/comments
1743 }
1745 # Image matches a whitelist entry
1748 }
1749 }
1750 }
1752 }
1754 /**
1755 * Process [[ ]] wikilinks
1756 *
1757 * @param $s string
1758 *
1759 * @return String: processed text
1760 *
1761 * @private
1762 */
1766 }
1768 /**
1769 * Process [[ ]] wikilinks (RIL)
1770 * @param $s
1771 * @throws MWException
1772 * @return LinkHolderArray
1773 *
1774 * @private
1775 */
1781 # the % is needed to support urlencoded titles as well
1784 # Match a link having the form [[namespace:link|alternate]]trail
1786 # Match cases where there is no "]]", which might still be images
1788 }
1792 # split the entire text string on occurrences of [[
1794 # get the first element (all text up to first [[), and remove the space we added
1803 # Match the end of a line for a word that's not followed by whitespace,
1804 # e.g. in the case of 'The Arab al[[Razi]]', 'al' will be matched
1806 }
1812 }
1821 }
1824 }
1829 # Loop for each link
1831 # Check for excessive memory usage
1833 # Too big
1834 # Do the existence check, replace the link holders and clear the array
1837 }
1846 }
1847 # first link
1851 }
1853 }
1860 # If we get a ] at the beginning of $m[3] that means we have a link that's something like:
1861 # [[Image:Foo.jpg|[http://example.com desc]]] <- having three ] in a row fucks up,
1862 # the real problem is with the $e1 regex
1863 # See bug 1300.
1864 #
1865 # Still some problems for cases where the ] is meant to be outside punctuation,
1866 # and no image is in sight. See bug 2095.
1867 #
1871 )
1872 {
1875 }
1876 # fix up urlencoded title texts
1878 # Should anchors '#' also be rejected?
1880 }
1882 } elseif ( preg_match( $e1_img, $line, $m ) ) { # Invalid, but might be an image with a link in its caption
1887 }
1893 }
1897 # Don't allow internal links to pages containing
1898 # PROTO: where PROTO is a valid URL protocol; these
1899 # should be external links.
1904 }
1906 # Make subpage if necessary
1911 }
1915 # Strip off leading ':'
1917 }
1926 }
1937 # look at the next 'line' to see if we can close it there
1942 }
1945 # the first ]] closes the inner link, the second the image
1951 # if there's exactly one ]] that's fine, we'll keep looking
1954 # if $next_line is invalid too, we need look no further
1957 }
1958 }
1960 # we couldn't find the end of this imageLink, so output it raw
1961 # but don't ignore what might be perfectly normal links in the text we've examined
1964 # note: no $trail, because without an end, there *is* no trail
1967 }
1970 # note: no $trail, because without an end, there *is* no trail
1973 }
1975 }
1981 # Bug 4598 madness. Handle the quotes only if they come from the alternate part
1982 # [[Lista d''e paise d''o munno]] -> <a href="...">Lista d''e paise d''o munno</a>
1983 # [[Criticism of Harry Potter|Criticism of ''Harry Potter'']]
1984 # -> <a href="Criticism of Harry Potter">Criticism of <i>Harry Potter</i></a>
1986 }
1988 # Link not escaped by : , create the various objects
1990 # Interwikis
1992 if ( $iw && $this->mOptions->getInterwikiMagic() && $nottalk && Language::fetchLanguageName( $iw, null, 'mw' ) ) {
1993 // XXX: the above check prevents links to sites with identifiers that are not language codes
1995 # Bug 24502: filter duplicates
1999 }
2005 }
2012 # if no parameters were passed, $text
2013 # becomes something like "File:Foo.png",
2014 # which we don't want to pass on to the
2015 # image generator
2018 # recursively parse links inside the image caption
2019 # actually, this will parse them in any other parameters, too,
2020 # but it might be hard to fix that, and it doesn't matter ATM
2023 }
2024 # cloak any absolute URLs inside the image markup, so replaceExternalLinks() won't touch them
2029 }
2032 }
2042 }
2048 /**
2049 * Strip the whitespace Category links produce, see bug 87
2050 * @todo We might want to use trim($tmp, "\n") here.
2051 */
2056 }
2057 }
2059 # Self-link checking
2064 true
2065 ) ) ) {
2068 }
2069 }
2071 # NS_MEDIA is a pseudo-namespace for linking directly to a file
2072 # @todo FIXME: Should do batch file existence checks, see comment below
2075 # Give extensions a chance to select the file revision for us
2080 # Fetch and register the file (file title may be different via hooks)
2082 # Cloak with NOPARSE to avoid replacement in replaceExternalLinks
2087 }
2090 # Some titles, such as valid special pages or files in foreign repos, should
2091 # be shown as bluelinks even though they're not included in the page table
2092 #
2093 # @todo FIXME: isAlwaysKnown() can be expensive for file links; we should really do
2094 # batch file existence checks for NS_FILE and NS_MEDIA
2099 # Links will be added to the output link list after checking
2101 }
2103 }
2106 }
2108 /**
2109 * Render a forced-blue link inline; protect against double expansion of
2110 * URLs if we're in a mode that prepends full URL prefixes to internal links.
2111 * Since this little disaster has to split off the trail text to avoid
2112 * breaking URLs in the following text without breaking trails on the
2113 * wiki links, it's been made into a horrible function.
2114 *
2115 * @param $nt Title
2116 * @param $text String
2117 * @param array $query or String
2118 * @param $trail String
2119 * @param $prefix String
2120 * @return String: HTML-wikitext mix oh yuck
2121 */
2127 }
2130 }
2135 }
2137 /**
2138 * Insert a NOPARSE hacky thing into any inline links in a chunk that's
2139 * going to go through further parsing steps before inline URL expansion.
2140 *
2141 * Not needed quite as much as it used to be since free links are a bit
2142 * more sensible these days. But bracketed links are still an issue.
2143 *
2144 * @param string $text more-or-less HTML
2145 * @return String: less-or-more HTML with NOPARSE bits
2146 */
2150 }
2152 /**
2153 * Return true if subpage links should be expanded on this page.
2154 * @return Boolean
2155 */
2157 # Some namespaces don't allow subpages
2159 }
2161 /**
2162 * Handle link to subpage if necessary
2163 *
2164 * @param string $target the source of the link
2165 * @param &$text String: the link text, modified as necessary
2166 * @return string the full name of the link
2167 * @private
2168 */
2171 }
2173 /**#@+
2174 * Used by doBlockLevels()
2175 * @private
2176 *
2177 * @return string
2178 */
2183 }
2187 }
2189 /**
2190 * getCommon() returns the length of the longest common substring
2191 * of both arguments, starting at the beginning of both.
2192 * @private
2193 *
2194 * @param $st1 string
2195 * @param $st2 string
2196 *
2197 * @return int
2198 */
2204 }
2209 }
2210 }
2212 }
2214 /**
2215 * These next three functions open, continue, and close the list
2216 * element appropriate to the prefix character passed into them.
2217 * @private
2218 *
2219 * @param $char string
2220 *
2221 * @return string
2222 */
2237 }
2240 }
2242 /**
2243 * TODO: document
2244 * @param $char String
2245 * @private
2246 *
2247 * @return string
2248 */
2256 }
2263 }
2264 }
2266 }
2268 /**
2269 * TODO: document
2270 * @param $char String
2271 * @private
2272 *
2273 * @return string
2274 */
2286 }
2289 }
2291 }
2292 /**#@-*/
2294 /**
2295 * Make lists from lines starting with ':', '*', '#', etc. (DBL)
2296 *
2297 * @param $text String
2298 * @param $linestart Boolean: whether or not this is at the start of a line.
2299 * @private
2300 * @return string the lists rendered as HTML
2301 */
2305 # Parsing through the text line by line. The main thing
2306 # happening here is handling of block-level elements p, pre,
2307 # and making lists from lines starting with * # : etc.
2308 #
2317 # Fix up $linestart
2322 }
2323 # * = ul
2324 # # = ol
2325 # ; = dt
2326 # : = dd
2331 # If not in a <pre> element, scan for and figure out what prefixes are there.
2333 # Multiple prefixes may abut each other for nested lists.
2337 # eh?
2338 # ; and : are both from definition-lists, so they're equivalent
2339 # for the purposes of determining whether or not we need to open/close
2340 # elements.
2345 # Don't interpret any other prefixes in preformatted text
2349 }
2351 # List generation
2353 # Same as the last item, so no need to deal with nesting or opening stuff
2358 # The one nasty exception: definition lists work like this:
2359 # ; title : definition text
2360 # So we check for : in the remainder text to split up the
2361 # title and definition, without b0rking links.
2366 }
2367 }
2369 # We need to open or close prefixes, or both.
2371 # Either open or close a level...
2375 # Close all the prefixes which aren't shared.
2379 }
2381 # Continue the current prefix if appropriate.
2384 }
2386 # Open prefixes where appropriate.
2392 # @todo FIXME: This is dupe of code above
2396 }
2397 }
2399 }
2401 }
2403 # If we have no prefixes, go to paragraph mode.
2406 # No prefix (not in list)--go to paragraph mode
2407 # XXX: use a stack for nestable elements like span, table and div
2408 $openmatch = preg_match( '/(?:<table|<blockquote|<h1|<h2|<h3|<h4|<h5|<h6|<pre|<tr|<p|<ul|<ol|<dl|<li|<\\/tr|<\\/td|<\\/th)/iS', $t );
2411 '<td|<th|<\\/?div|<hr|<\\/pre|<\\/p|' . $this->mUniqPrefix . '-pre|<\\/li|<\\/ul|<\\/ol|<\\/dl|<\\/?center)/iS', $t );
2414 # TODO bug 5718: paragraph closed
2418 }
2422 # pre
2427 }
2430 # paragraph
2443 }
2444 }
2453 }
2454 }
2455 }
2456 }
2458 }
2459 # somewhere above we forget to get out of pre block (bug 785)
2462 }
2465 }
2466 }
2470 }
2474 }
2478 }
2480 /**
2481 * Split up a string on ':', ignoring any occurrences inside tags
2482 * to prevent illegal overlapping.
2483 *
2484 * @param string $str the string to split
2485 * @param &$before String set to everything before the ':'
2486 * @param &$after String set to everything after the ':'
2487 * @throws MWException
2488 * @return String the position of the ':', or false if none found
2489 */
2495 # Nothing to find!
2498 }
2502 # Easy; no tag nesting to worry about
2507 }
2509 # Ugly state machine to walk through avoiding tags.
2517 # (Using the number is a performance hack for common cases)
2521 # Could be either a <start> tag or an </end> tag
2526 # We found it!
2531 }
2532 # Embedded in a tag; don't break it.
2535 # Skip ahead looking for something interesting
2538 # Nothing else interesting
2541 }
2545 # We found it!
2550 }
2551 }
2553 # Nothing else interesting to find; abort!
2554 # We're nested, but there's no close tags left. Abort!
2556 }
2557 # Skip ahead to next tag start
2560 }
2563 # In a <tag>
2570 # Slash may be followed by >?
2574 # ignore
2575 }
2586 # Illegal early close? This shouldn't happen D:
2591 }
2594 # In a </tag>
2601 }
2603 }
2607 # Yes, a self-closed tag <blah/>
2610 # Probably we're jumping the gun, and this is an attribute
2612 }
2617 }
2624 }
2631 }
2636 }
2637 }
2639 wfDebug( __METHOD__ . ": Invalid input; not enough close tags (stack $stack, state $state)\n" );
2642 }
2645 }
2647 /**
2648 * Return value of a magic variable (like PAGENAME)
2649 *
2650 * @private
2651 *
2652 * @param $index integer
2653 * @param bool|\PPFrame $frame
2654 *
2655 * @throws MWException
2656 * @return string
2657 */
2663 // If no title set, bad things are going to happen
2664 // later. Title should always be set since this
2665 // should only be called in the middle of a parse
2666 // operation (but the unit-tests do funky stuff)
2669 }
2671 /**
2672 * Some of these require message or data lookups and can be
2673 * expensive to check many times.
2674 */
2678 }
2679 }
2721 $value = $pageLang->getMonthAbbreviation( MWTimestamp::getLocalInstance( $ts )->format( 'n' ) );
2751 $value = wfEscapeWikiText( wfUrlEncode( str_replace( ' ', '_', $this->mTitle->getRootText() ) ) );
2757 $value = wfEscapeWikiText( wfUrlEncode( str_replace( ' ', '_', $this->mTitle->getBaseText() ) ) );
2765 }
2773 }
2786 # 0 means the page doesn't exist in the database,
2787 # which means the user is previewing a new page.
2788 # The vary-revision flag must be set, because the magic word
2789 # will have a different value once the page is saved.
2792 }
2796 # Let the edit saving system know we should parse the page
2797 # *after* a revision ID has been assigned.
2803 # Let the edit saving system know we should parse the page
2804 # *after* a revision ID has been assigned. This is for null edits.
2810 # Let the edit saving system know we should parse the page
2811 # *after* a revision ID has been assigned. This is for null edits.
2817 # Let the edit saving system know we should parse the page
2818 # *after* a revision ID has been assigned. This is for null edits.
2824 # Let the edit saving system know we should parse the page
2825 # *after* a revision ID has been assigned. This is for null edits.
2831 # Let the edit saving system know we should parse the page
2832 # *after* a revision ID has been assigned. This is for null edits.
2838 # Let the edit saving system know we should parse the page
2839 # *after* a revision ID has been assigned. This is for null edits.
2845 # Let the edit saving system know we should parse the page
2846 # *after* a revision ID has been assigned. This is for null edits.
2861 $value = $this->mTitle->canTalk() ? str_replace( '_', ' ', $this->mTitle->getTalkNsText() ) : '';
2885 # @bug 4594 PHP5 has it zero padded, PHP4 does not, cast to
2886 # int to remove the padding
2899 $value = $pageLang->time( MWTimestamp::getLocalInstance( $ts )->format( 'YmdHis' ), false, false );
2905 # @bug 4594 PHP5 has it zero padded, PHP4 does not, cast to
2906 # int to remove the padding
2966 if ( wfRunHooks( 'ParserGetVariableValueSwitch', array( &$this, &$this->mVarCache, &$index, &$ret, &$frame ) ) ) {
2970 }
2971 }
2975 }
2978 }
2980 /**
2981 * initialise the magic variables (like CURRENTMONTHNAME) and substitution modifiers
2982 *
2983 * @private
2984 */
2993 }
2995 /**
2996 * Preprocess some wikitext and return the document tree.
2997 * This is the ghost of replace_variables().
2998 *
2999 * @param string $text The text to parse
3000 * @param $flags Integer: bitwise combination of:
3001 * self::PTD_FOR_INCLUSION Handle "<noinclude>" and "<includeonly>" as if the text is being
3002 * included. Default is to assume a direct page view.
3003 *
3004 * The generated DOM tree must depend only on the input text and the flags.
3005 * The DOM tree must be the same in OT_HTML and OT_WIKI mode, to avoid a regression of bug 4899.
3006 *
3007 * Any flag added to the $flags parameter here, or any other parameter liable to cause a
3008 * change in the DOM tree for a given text, must be passed through the section identifier
3009 * in the section edit link and thus back to extractSections().
3010 *
3011 * The output of this function is currently only cached in process memory, but a persistent
3012 * cache may be implemented at a later date which takes further advantage of these strict
3013 * dependency requirements.
3014 *
3015 * @private
3016 *
3017 * @return PPNode
3018 */
3022 }
3024 /**
3025 * Return a three-element array: leading whitespace, string contents, trailing whitespace
3026 *
3027 * @param $s string
3028 *
3029 * @return array
3030 */
3040 }
3042 }
3044 /**
3045 * Replace magic variables, templates, and template arguments
3046 * with the appropriate text. Templates are substituted recursively,
3047 * taking care to avoid infinite loops.
3048 *
3049 * Note that the substitution depends on value of $mOutputType:
3050 * self::OT_WIKI: only {{subst:}} templates
3051 * self::OT_PREPROCESS: templates but not extension tags
3052 * self::OT_HTML: all templates and extension tags
3053 *
3054 * @param string $text the text to transform
3055 * @param $frame PPFrame Object describing the arguments passed to the template.
3056 * Arguments may also be provided as an associative array, as was the usual case before MW1.12.
3057 * Providing arguments this way may be useful for extensions wishing to perform variable replacement explicitly.
3058 * @param $argsOnly Boolean only do argument (triple-brace) expansion, not double-brace expansion
3059 * @private
3060 *
3061 * @return string
3062 */
3064 # Is there any text? Also, Prevent too big inclusions!
3067 }
3073 wfDebug( __METHOD__ . " called using plain parameters instead of a PPFrame instance. Creating custom frame.\n" );
3075 }
3083 }
3085 /**
3086 * Clean up argument array - refactored in 1.9 so parserfunctions can use it, too.
3087 *
3088 * @param $args array
3089 *
3090 * @return array
3091 */
3104 }
3107 }
3108 }
3109 }
3112 }
3114 /**
3115 * Warn the user when a parser limitation is reached
3116 * Will warn at most once the user per limitation type
3117 *
3118 * @param string $limitationType should be one of:
3119 * 'expensive-parserfunction' (corresponding messages:
3120 * 'expensive-parserfunction-warning',
3121 * 'expensive-parserfunction-category')
3122 * 'post-expand-template-argument' (corresponding messages:
3123 * 'post-expand-template-argument-warning',
3124 * 'post-expand-template-argument-category')
3125 * 'post-expand-template-inclusion' (corresponding messages:
3126 * 'post-expand-template-inclusion-warning',
3127 * 'post-expand-template-inclusion-category')
3128 * @param int|null $current Current value
3129 * @param int|null $max Maximum allowed, when an explicit limit has been
3130 * exceeded, provide the values (optional)
3131 */
3133 # does no harm if $current and $max are present but are unnecessary for the message
3138 }
3140 /**
3141 * Return the text of a template, after recursively
3142 * replacing any variables or templates within the template.
3143 *
3144 * @param array $piece the parts of the template
3145 * $piece['title']: the title, i.e. the part before the |
3146 * $piece['parts']: the parameter array
3147 * $piece['lineStart']: whether the brace was at the start of a line
3148 * @param $frame PPFrame The current frame, contains template arguments
3149 * @throws MWException
3150 * @return String: the text of the template
3151 * @private
3152 */
3157 # Flags
3165 # Title object, where $text came from
3168 # $part1 is the bit before the first |, and must contain only title characters.
3169 # Various prefixes will be stripped from it later.
3174 # Original title text preserved for various purposes
3177 # $args is a list of argument nodes, starting from index 0, not including $part1
3178 # @todo FIXME: If piece['parts'] is null then the call to getLength() below won't work b/c this $args isn't an object
3184 # SUBST
3190 # Possibilities for substMatch: "subst", "safesubst" or FALSE
3191 # Decide whether to expand template or keep wikitext as-is.
3197 }
3203 }
3204 }
3209 }
3210 }
3212 # Variables
3219 }
3221 }
3222 }
3224 # MSG, MSGNW and RAW
3226 # Check for MSGNW:
3231 # Remove obsolete MSG:
3234 }
3236 # Check for RAW:
3240 }
3241 }
3244 # Parser functions
3254 }
3261 }
3263 # The interface for parser functions allows for extracting
3264 # flags into the local scope. Extract any forwarded flags
3265 # here.
3267 }
3269 }
3271 # Finish mangling title and then check for loops.
3272 # Set $title to a Title object and $titleText to the PDBK
3275 # Split the title into page and subpage
3281 }
3285 # Check for language variants if the template is not found
3288 }
3289 # Do recursion depth check
3297 }
3298 }
3299 }
3301 # Load from database
3304 # Too many unique items can kill profiling DBs/collectors
3307 }
3313 {
3314 // Pass the template arguments as URL parameters.
3315 // "uselang" will have no effect since the Language object
3316 // is forced to the one defined in ParserOptions.
3324 }
3325 }
3327 // Create a new context to execute the special page
3340 }
3349 }
3350 }
3352 # If the title is valid but undisplayable, make a link to it
3356 }
3358 # Interwiki transclusion
3364 # Preprocess it like a template
3367 }
3369 }
3371 # Do infinite loop check
3372 # This has to be done after redirect resolution to avoid infinite loops via redirects
3379 }
3381 }
3383 # If we haven't found text to substitute by now, we're done
3384 # Recover the source wikitext and return it
3389 }
3392 }
3394 # Expand DOM-style return values in a child frame
3396 # Clean up argument array
3402 # Expansion is eligible for the empty-frame cache
3408 }
3410 # Uncached expansion
3412 }
3413 }
3417 }
3421 }
3423 # Replace raw HTML by a placeholder
3427 # Escape nowiki-style return values
3432 {
3433 # Bug 529: if the template begins with a table or block-level
3434 # element, it should be treated as beginning a new line.
3435 # This behavior is somewhat controversial.
3437 }
3440 # Error, oversize inclusion
3442 # Make a working, properly escaped link if possible (bug 23588)
3445 # This will probably not be a working link, but at least it may
3446 # provide some hint of where the problem is
3449 }
3450 $text .= $this->insertStripItem( '<!-- WARNING: template omitted, post-expand include size too large -->' );
3452 }
3458 }
3462 }
3464 /**
3465 * Call a parser function and return an array with text and flags.
3466 *
3467 * The returned array will always contain a boolean 'found', indicating
3468 * whether the parser function was found or not. It may also contain the
3469 * following:
3470 * text: string|object, resulting wikitext or PP DOM object
3471 * isHTML: bool, $text is HTML, armour it against wikitext transformation
3472 * isChildObj: bool, $text is a DOM node needing expansion in a child frame
3473 * isLocalObj: bool, $text is a DOM node needing expansion in the current frame
3474 * nowiki: bool, wiki markup in $text should be escaped
3475 *
3476 * @since 1.21
3477 * @param $frame PPFrame The current frame, contains template arguments
3478 * @param $function string Function name
3479 * @param $args array Arguments to the function
3480 * @return array
3481 */
3487 # Case sensitive functions
3491 # Case insensitive functions
3498 }
3499 }
3504 # Workaround for PHP bug 35229 and similar
3509 }
3513 # Convert arguments to PPNodes and collect for appending to $allArgs
3520 }
3521 }
3523 # Add a frame parameter, and pass the arguments as an array
3527 # Convert arguments to plain text and append to $allArgs
3535 }
3536 }
3537 }
3541 # The interface for function hooks allows them to return a wikitext
3542 # string or an array containing the string and any flags. This mungs
3543 # things around to match what this method should return.
3548 );
3552 }
3556 );
3557 }
3563 }
3566 }
3571 }
3576 }
3578 /**
3579 * Get the semi-parsed DOM representation of a template with a given title,
3580 * and its redirect destination title. Cached.
3581 *
3582 * @param $title Title
3583 *
3584 * @return array
3585 */
3594 }
3597 }
3599 # Cache miss, go to the database
3605 }
3613 }
3616 }
3618 /**
3619 * Fetch the unparsed text of a template and register a reference to it.
3620 * @param Title $title
3621 * @return Array ( string or false, Title )
3622 */
3624 $templateCb = $this->mOptions->getTemplateCallback(); # Defaults to Parser::statelessFetchTemplate()
3632 // If we transclude ourselves, the final result
3633 // will change based on the new version of the page
3635 }
3636 }
3637 }
3639 }
3641 /**
3642 * Fetch the unparsed text of a template and register a reference to it.
3643 * @param Title $title
3644 * @return mixed string or false
3645 */
3649 }
3651 /**
3652 * Static function to get a template
3653 * Can be overridden via ParserOptions::setTemplateCallback().
3654 *
3655 * @param $title Title
3656 * @param $parser Parser
3657 *
3658 * @return array
3659 */
3665 # Loop to fetch the article, with up to 1 redirect
3667 # Give extensions a chance to select the revision instead
3678 );
3680 }
3681 # Get the revision
3686 # If there is no current revision, there is no page
3690 }
3697 # We fetched a rev from a different title; register it too...
3702 }
3711 }
3718 }
3723 }
3726 }
3727 # Redirect?
3730 }
3735 }
3737 /**
3738 * Fetch a file and its title and register a reference to it.
3739 * If 'broken' is a key in $options then the file will appear as a broken thumbnail.
3740 * @param Title $title
3741 * @param array $options Array of options to RepoGroup::findFile
3742 * @return File|bool
3743 */
3747 }
3749 /**
3750 * Fetch a file and its title and register a reference to it.
3751 * If 'broken' is a key in $options then the file will appear as a broken thumbnail.
3752 * @param Title $title
3753 * @param array $options Array of options to RepoGroup::findFile
3754 * @return Array ( File or false, Title of file )
3755 */
3761 # Register the file as a dependency...
3764 # Update fetched file title
3767 # This file was not a redirect, but the title does not match.
3768 # Register under the new name because otherwise the link will
3769 # get lost.
3771 }
3772 }
3774 }
3776 /**
3777 * Helper function for fetchFileAndTitle.
3778 *
3779 * Also useful if you need to fetch a file but not use it yet,
3780 * for example to get the file's handler.
3781 *
3782 * @param Title $title
3783 * @param array $options Array of options to RepoGroup::findFile
3784 * @return File or false
3785 */
3793 }
3795 }
3797 /**
3798 * Transclude an interwiki link.
3799 *
3800 * @param $title Title
3801 * @param $action
3802 *
3803 * @return string
3804 */
3810 }
3816 }
3818 }
3820 /**
3821 * @param $url string
3822 * @return Mixed|String
3823 */
3832 }
3838 } elseif ( $req->getStatus() != 200 ) { // Though we failed to fetch the content, this status is useless.
3839 return wfMessage( 'scarytranscludefailed-httpstatus', $url, $req->getStatus() /* HTTP status */ )->inContentLanguage()->text();
3842 }
3849 ) );
3851 }
3853 /**
3854 * Triple brace replacement -- used for template arguments
3855 * @private
3856 *
3857 * @param $piece array
3858 * @param $frame PPFrame
3859 *
3860 * @return array
3861 */
3872 && (
3876 )
3877 ) {
3878 # No match in frame, use the supplied default
3880 }
3884 }
3887 # No match anywhere
3889 }
3892 }
3897 }
3901 }
3903 /**
3904 * Return the text to be used for a given extension tag.
3905 * This is the ghost of strip().
3906 *
3907 * @param array $params Associative array of parameters:
3908 * name PPNode for the tag name
3909 * attr PPNode for unparsed text where tag attributes are thought to be
3910 * attributes Optional associative array of parsed attributes
3911 * inner Contents of extension element
3912 * noClose Original text did not have a close tag
3913 * @param $frame PPFrame
3914 *
3915 * @throws MWException
3916 * @return string
3917 */
3922 $marker = "{$this->mUniqPrefix}-$name-" . sprintf( '%08X', $this->mMarkerIndex++ ) . self::MARKER_SUFFIX;
3930 }
3936 }
3939 # Workaround for PHP bug 35229 and similar
3942 }
3949 }
3955 }
3958 # Extract flags to local scope (to override $markerType)
3963 }
3967 }
3972 }
3973 }
3979 }
3980 }
3990 }
3992 }
3994 /**
3995 * Increment an include size counter
3996 *
3997 * @param string $type the type of expansion
3998 * @param $size Integer: the size of the text
3999 * @return Boolean: false if this inclusion would take it over the maximum, true otherwise
4000 */
4007 }
4008 }
4010 /**
4011 * Increment the expensive function count
4012 *
4013 * @return Boolean: false if the limit has been exceeded
4014 */
4018 }
4020 /**
4021 * Strip double-underscore items like __NOGALLERY__ and __NOTOC__
4022 * Fills $this->mDoubleUnderscores, returns the modified text
4023 *
4024 * @param $text string
4025 *
4026 * @return string
4027 */
4031 # The position of __TOC__ needs to be recorded
4037 # Set a placeholder. At the end we'll fill it in with the TOC.
4040 # Only keep the first one.
4042 }
4044 # Now match and remove the rest of them
4050 }
4053 }
4054 if ( isset( $this->mDoubleUnderscores['hiddencat'] ) && $this->mTitle->getNamespace() == NS_CATEGORY ) {
4056 }
4057 # (bug 8068) Allow control over whether robots index a page.
4058 #
4059 # @todo FIXME: Bug 14899: __INDEX__ always overrides __NOINDEX__ here! This
4060 # is not desirable, the last one on the page should win.
4064 }
4068 }
4070 # Cache all double underscores in the database
4073 }
4077 }
4079 /**
4080 * Add a tracking category, getting the title from a system message,
4081 * or print a debug message if the title is invalid.
4082 *
4083 * @param string $msg message key
4084 * @return Boolean: whether the addition was successful
4085 */
4090 }
4091 // Important to parse with correct title (bug 31469)
4097 # Allow tracking categories to be disabled by setting them to "-"
4100 }
4109 }
4110 }
4112 /**
4113 * This function accomplishes several tasks:
4114 * 1) Auto-number headings if that option is enabled
4115 * 2) Add an [edit] link to sections for users who have enabled the option and can edit the page
4116 * 3) Add a Table of contents on the top for users who have enabled the option
4117 * 4) Auto-anchor headings
4118 *
4119 * It loops through all headlines, collects the necessary data, then splits up the
4120 * string and re-inserts the newly formatted headlines.
4121 *
4122 * @param $text String
4123 * @param string $origText original, untouched wikitext
4124 * @param $isMain Boolean
4125 * @return mixed|string
4126 * @private
4127 */
4131 # Inhibit editsection links if requested in the page
4137 }
4140 }
4142 # Get all headlines for numbering them and adding funky stuff like [edit]
4143 # links - this is for later, but we need the number of headlines right now
4145 $numMatches = preg_match_all( '/<H(?P<level>[1-6])(?P<attrib>.*?' . '>)\s*(?P<header>[\s\S]*?)\s*<\/H[1-6] *>/i', $text, $matches );
4147 # if there are fewer than 4 headlines in the article, do not show TOC
4148 # unless it's been explicitly enabled.
4152 # Allow user to stipulate that a page should have a "new section"
4153 # link added via __NEWSECTIONLINK__
4156 }
4158 # Allow user to remove the "new section"
4159 # link via __NONEWSECTIONLINK__
4162 }
4164 # if the string __FORCETOC__ (not case-sensitive) occurs in the HTML,
4165 # override above conditions and always show TOC above first header
4169 }
4171 # headline counter
4175 # Ugh .. the TOC should have neat indentation levels which can be
4176 # passed to the skin functions. These are determined here
4208 }
4212 }
4216 # Increase TOC level
4223 }
4225 # Decrease TOC level, find level to jump to
4229 # Found last matching level
4233 # Found first matching level below current level
4236 }
4237 }
4240 }
4243 # Unindent only if the previous toc level was shown :p
4248 }
4249 }
4251 # No change in level, end TOC line
4254 }
4255 }
4259 # count number of headlines for each level
4266 }
4269 }
4270 }
4272 # The safe header is a version of the header text safe to use for links
4274 # Remove link placeholders by the link text.
4275 # <!--LINK number-->
4276 # turns into
4277 # link text with suffix
4278 # Do this before unstrip since link text can contain strip markers
4281 # Avoid insertion of weird stuff like <math> by expanding the relevant sections
4284 # Strip out HTML (first regex removes any tag not allowed)
4285 # Allowed tags are:
4286 # * <sup> and <sub> (bug 8393)
4287 # * <i> (bug 26375)
4288 # * <b> (r105284)
4289 # * <span dir="rtl"> and <span dir="ltr"> (bug 35167)
4290 #
4291 # We strip any parameter from accepted tags (second regex), except dir="rtl|ltr" from <span>,
4292 # to allow setting directionality in toc items.
4294 array( '#<(?!/?(span|sup|sub|i|b)(?: [^>]*)?>).*?' . '>#', '#<(/?(?:span(?: dir="(?:rtl|ltr)")?|sup|sub|i|b))(?: .*?)?' . '>#' ),
4296 $safeHeadline
4297 );
4300 # For the anchor, strip out HTML-y stuff period
4304 # Save headline for section edit hint before it's escaped
4308 # For reverse compatibility, provide an id that's
4309 # HTML4-compatible, like we used to.
4310 #
4311 # It may be worth noting, academically, that it's possible for
4312 # the legacy anchor to conflict with a non-legacy headline
4313 # anchor on the page. In this case likely the "correct" thing
4314 # would be to either drop the legacy anchors or make sure
4315 # they're numbered first. However, this would require people
4316 # to type in section names like "abc_.D7.93.D7.90.D7.A4"
4317 # manually, so let's not bother worrying about it.
4323 # No reason to have both (in fact, we can't)
4325 }
4330 }
4332 # HTML names must be case-insensitively unique (bug 10721).
4333 # This does not apply to Unicode characters per
4334 # http://dev.w3.org/html5/spec/infrastructure.html#case-sensitivity-and-string-comparison
4335 # @todo FIXME: We may be changing them depending on the current locale.
4341 }
4343 # count how many in assoc. array so we can track dupes in anchors
4348 }
4353 }
4355 # Don't number the heading if it is the only one (looks silly)
4357 # the two are different if the line contains a link
4358 $headline = Html::element( 'span', array( 'class' => 'mw-headline-number' ), $numbering ) . ' ' . $headline;
4359 }
4361 # Create the anchor for linking from the TOC to the section
4366 }
4369 }
4373 }
4375 # Add the section to the section tree
4376 # Find the DOM node for this header
4382 }
4383 }
4387 }
4397 );
4399 # give headline the correct <h#> tag
4401 // Output edit section links as markers with styles that can be customized by skins
4403 # Put a T flag in the section identifier, to indicate to extractSections()
4404 # that sections inside <includeonly> should be counted.
4408 }
4409 // We use a bit of pesudo-xml for editsection markers. The language converter is run later on
4410 // Using a UNIQ style marker leads to the converter screwing up the tokens when it converts stuff
4411 // And trying to insert strip tags fails too. At this point all real inputted tags have already been escaped
4412 // so we don't have to worry about a user trying to input one of these markers directly.
4413 // We use a page and section attribute to stop the language converter from converting these important bits
4414 // of data, but put the headline hint inside a content block because the language converter is supposed to
4415 // be able to convert that piece of data.
4422 }
4425 }
4431 }
4435 # Never ever show TOC if no headers
4438 }
4443 }
4446 }
4450 }
4452 # split up and insert constructed headlines
4456 // build an array of document sections
4459 // $head is zero-based, sections aren't.
4464 }
4466 /**
4467 * Send a hook, one per section.
4468 * The idea here is to be able to make section-level DIVs, but to do so in a
4469 * lower-impact, more correct way than r50769
4470 *
4471 * $this : caller
4472 * $section : the section number
4473 * &$sectionContent : ref to the content of the section
4474 * $showEditLinks : boolean describing whether this section has an edit link
4475 */
4479 }
4482 // append the TOC at the beginning
4483 // Top anchor now in skin
4485 }
4493 }
4494 }
4496 /**
4497 * Transform wiki markup when saving a page by doing "\r\n" -> "\n"
4498 * conversion, substitting signatures, {{subst:}} templates, etc.
4499 *
4500 * @param string $text the text to transform
4501 * @param $title Title: the Title object for the current article
4502 * @param $user User: the User object describing the current user
4503 * @param $options ParserOptions: parsing options
4504 * @param $clearState Boolean: whether to clear the parser state first
4505 * @return String: the altered wiki markup
4506 */
4507 public function preSaveTransform( $text, Title $title, User $user, ParserOptions $options, $clearState = true ) {
4513 );
4517 }
4523 }
4525 /**
4526 * Pre-save transform helper function
4527 * @private
4528 *
4529 * @param $text string
4530 * @param $user User
4531 *
4532 * @return string
4533 */
4537 # Note: This is the timestamp saved as hardcoded wikitext to
4538 # the database, we use $wgContLang here in order to give
4539 # everyone the same signature and use the default one rather
4540 # than the one selected in each user's preferences.
4541 # (see also bug 12815)
4547 # Allow translation of timezones through wiki. format() can return
4548 # whatever crap the system uses, localised or not, so we cannot
4549 # ship premade translations.
4554 }
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 */
5017 }
5023 }
5026 }
5029 }
5032 }
5038 # match lines like these:
5039 # Image:someimage.jpg|This is some image
5042 # Skip empty lines
5045 }
5049 }
5052 # Bogus title. Ignore these so we don't bomb out later.
5054 }
5056 # We need to get what handler the file uses, to figure out parameters.
5057 # Note, a hook can overide the file name, and chose an entirely different
5058 # file (which potentially could be of a different type and have different handler).
5063 # Don't register it now, as ImageGallery does that later.
5071 );
5074 // We don't want people to specify per-image widths.
5075 // Additionally the width parameter would need special casing anyhow.
5077 }
5087 // look for an |alt= definition while trying not to break existing
5088 // captions with multiple pipes (|) in it, until a more sensible grammar
5089 // is defined for images in galleries
5091 // FIXME: Doing recursiveTagParse at this stage, and the trim before
5092 // splitting on '|' is a bit odd, and different from makeImage.
5109 //check to see if link matches an absolute url, if not then it must be a wiki link.
5116 }
5117 }
5120 // Must be a handler specific parameter.
5124 // Guess not. Append it to the caption.
5127 }
5128 }
5131 // concatenate all other pipes
5133 }
5134 }
5135 // remove the first pipe
5137 }
5140 }
5144 }
5146 /**
5147 * @param $handler
5148 * @return array
5149 */
5155 }
5157 # Initialise static lists
5164 );
5172 }
5173 }
5174 }
5176 # Add handler params
5182 }
5183 }
5186 }
5187 return array( $this->mImageParams[$handlerClass], $this->mImageParamsMagicArray[$handlerClass] );
5188 }
5190 /**
5191 * Parse image options text and use it to make an image
5192 *
5193 * @param $title Title
5194 * @param $options String
5195 * @param $holders LinkHolderArray|bool
5196 * @return string HTML
5197 */
5199 # Check if the options text is of the form "options|alt text"
5200 # Options are:
5201 # * thumbnail make a thumbnail with enlarge-icon and caption, alignment depends on lang
5202 # * left no resizing, just left align. label is used for alt= only
5203 # * right same, but right aligned
5204 # * none same, but not aligned
5205 # * ___px scale to ___ pixels width, no aligning. e.g. use in taxobox
5206 # * center center the image
5207 # * frame Keep original image size, no magnify-button.
5208 # * framed Same as "frame"
5209 # * frameless like 'thumb' but without a frame. Keeps user preferences for width
5210 # * upright reduce width for upright images, rounded to full __0 px
5211 # * border draw a 1px border around the image
5212 # * alt Text for HTML alt attribute (defaults to empty)
5213 # * class Set a class for img node
5214 # * link Set the target of the image link. Can be external, interwiki, or local
5215 # vertical-align values (no % or length right now):
5216 # * baseline
5217 # * sub
5218 # * super
5219 # * top
5220 # * text-top
5221 # * middle
5222 # * bottom
5223 # * text-bottom
5227 # Give extensions a chance to select the file revision for us
5232 # Fetch and register the file (file title may be different via hooks)
5235 # Get parameter map
5242 }
5244 # Process the input parameters
5255 # Special case; width and height come in one variable together
5263 }
5264 }
5270 }
5271 }
5272 # else no validation -- bug 13436
5275 # Validate handler parameter
5278 # Validate internal parameters
5283 # @todo FIXME: Possibly check validity here for
5284 # manualthumb? downstream behavior seems odd with
5285 # missing manual thumbs.
5302 }
5304 }
5312 }
5313 }
5316 # Most other things appear to be empty or numeric...
5318 }
5319 }
5323 }
5324 }
5325 }
5328 }
5329 }
5331 # Process alignment parameters
5334 }
5337 }
5341 # Will the image be presented in a frame, with the caption below?
5347 # In the old days, [[Image:Foo|text...]] would set alt text. Later it
5348 # came to also set the caption, ordinary text after the image -- which
5349 # makes no sense, because that just repeats the text multiple times in
5350 # screen readers. It *also* came to set the title attribute.
5351 #
5352 # Now that we have an alt attribute, we should not set the alt text to
5353 # equal the caption: that's worse than useless, it just repeats the
5354 # text. This is the framed/thumbnail case. If there's no caption, we
5355 # use the unnamed parameter for alt text as well, just for the time be-
5356 # ing, if the unnamed param is set and the alt param is not.
5357 #
5358 # For the future, we need to figure out if we want to tweak this more,
5359 # e.g., introducing a title= parameter for the title; ignoring the un-
5360 # named parameter entirely for images without a caption; adding an ex-
5361 # plicit caption= parameter and preserving the old magic unnamed para-
5362 # meter for BC; ...
5365 # No caption or alt text, add the filename as the alt text so
5366 # that screen readers at least get some description of the image
5368 }
5369 # Do not set $params['frame']['title'] because tooltips don't make sense
5370 # for framed images
5373 # No alt text, use the "caption" for the alt text
5377 # No caption, fall back to using the filename for the
5378 # alt text
5380 }
5381 }
5382 # Use the "caption" for the tooltip text
5384 }
5388 # Linker does the rest
5393 # Give the handler a chance to modify the parser object
5396 }
5399 }
5401 /**
5402 * @param $caption
5403 * @param $holders LinkHolderArray
5404 * @return mixed|String
5405 */
5407 # Strip bad stuff out of the title (tooltip). We can't just use
5408 # replaceLinkHoldersText() here, because if this function is called
5409 # from replaceInternalLinks2(), mLinkHolders won't be up-to-date.
5414 }
5416 # make sure there are no placeholders in thumbnail attributes
5417 # that are later expanded to html- so expand them now and
5418 # remove the tags
5423 }
5425 /**
5426 * Set a flag in the output object indicating that the content is dynamic and
5427 * shouldn't be cached.
5428 */
5434 }
5437 }
5439 /**
5440 * Callback from the Sanitizer for expanding items found in HTML attribute
5441 * values, so they can be safely tested and escaped.
5442 *
5443 * @param $text String
5444 * @param $frame PPFrame
5445 * @return String
5446 */
5451 }
5453 /**
5454 * Accessor
5455 *
5456 * @return array
5457 */
5459 return array_merge( array_keys( $this->mTransparentTagHooks ), array_keys( $this->mTagHooks ), array_keys( $this->mFunctionTagHooks ) );
5460 }
5462 /**
5463 * Replace transparent tags in $text with the values given by the callbacks.
5464 *
5465 * Transparent tag hooks are like regular XML-style tag hooks, except they
5466 * operate late in the transformation sequence, on HTML instead of wikitext.
5467 *
5468 * @param $text string
5469 *
5470 * @return string
5471 */
5482 $output = call_user_func_array( $this->mTransparentTagHooks[$tagName], array( $content, $params, $this ) );
5485 }
5487 }
5489 }
5491 /**
5492 * Break wikitext input into sections, and either pull or replace
5493 * some particular section's text.
5494 *
5495 * External callers should use the getSection and replaceSection methods.
5496 *
5497 * @param string $text Page wikitext
5498 * @param string $section a section identifier string of the form:
5499 * "<flag1> - <flag2> - ... - <section number>"
5500 *
5501 * Currently the only recognised flag is "T", which means the target section number
5502 * was derived during a template inclusion parse, in other words this is a template
5503 * section edit link. If no flags are given, it was an ordinary section edit link.
5504 * This flag is required to avoid a section numbering mismatch when a section is
5505 * enclosed by "<includeonly>" (bug 6563).
5506 *
5507 * The section number 0 pulls the text before the first heading; other numbers will
5508 * pull the given section along with its lower-level subsections. If the section is
5509 * not found, $mode=get will return $newtext, and $mode=replace will return $text.
5510 *
5511 * Section 0 is always considered to exist, even if it only contains the empty
5512 * string. If $text is the empty string and section 0 is replaced, $newText is
5513 * returned.
5514 *
5515 * @param string $mode one of "get" or "replace"
5516 * @param string $newText replacement text for section data.
5517 * @return String: for "get", the extracted section text.
5518 * for "replace", the whole page with the section replaced.
5519 */
5526 # Process section extraction flags
5533 }
5534 }
5536 # Check for empty input
5538 # Only sections 0 and T-0 exist in an empty document
5544 }
5550 }
5551 }
5552 }
5554 # Preprocess the text
5557 # <h> nodes indicate section breaks
5558 # They can only occur at the top level, so we can find them by iterating the root's children
5561 # Find the target section
5563 # Section zero doesn't nest, level=big
5572 }
5573 }
5576 }
5578 }
5579 }
5582 # Not found
5587 }
5588 }
5590 # Find the end of the section, including nested sections
5597 }
5598 }
5601 }
5605 # Write out the remainder (in replace mode only)
5607 # Output the replacement text
5608 # Add two newlines on -- trailing whitespace in $newText is conventionally
5609 # stripped by the editor, so we need both newlines to restore the paragraph gap
5610 # Only add trailing whitespace if there is newText
5613 }
5618 }
5619 }
5622 # Re-insert stripped tags
5624 }
5627 }
5629 /**
5630 * This function returns the text of a section, specified by a number ($section).
5631 * A section is text under a heading like == Heading == or \<h1\>Heading\</h1\>, or
5632 * the first section before any such heading (section 0).
5633 *
5634 * If a section contains subsections, these are also returned.
5635 *
5636 * @param string $text text to look in
5637 * @param string $section section identifier
5638 * @param string $deftext default to return if section is not found
5639 * @return string text of the requested section
5640 */
5643 }
5645 /**
5646 * This function returns $oldtext after the content of the section
5647 * specified by $section has been replaced with $text. If the target
5648 * section does not exist, $oldtext is returned unchanged.
5649 *
5650 * @param string $oldtext former text of the article
5651 * @param int $section section identifier
5652 * @param string $text replacing text
5653 * @return String: modified text
5654 */
5657 }
5659 /**
5660 * Get the ID of the revision we are parsing
5661 *
5662 * @return Mixed: integer or null
5663 */
5666 }
5668 /**
5669 * Get the revision object for $this->mRevisionId
5670 *
5671 * @return Revision|null either a Revision object or null
5672 */
5676 }
5679 }
5683 }
5685 /**
5686 * Get the timestamp associated with the current revision, adjusted for
5687 * the default server-local timestamp
5688 */
5698 # The cryptic '' timezone parameter tells to use the site-default
5699 # timezone offset instead of the user settings.
5700 #
5701 # Since this value will be saved into the parser cache, served
5702 # to other users, and potentially even used inside links and such,
5703 # it needs to be consistent for all visitors.
5707 }
5709 }
5711 /**
5712 * Get the name of the user that edited the last revision
5713 *
5714 * @return String: user name
5715 */
5720 # if this template is subst: the revision id will be blank,
5721 # so just use the current user's name
5726 }
5727 }
5729 }
5731 /**
5732 * Mutator for $mDefaultSort
5733 *
5734 * @param string $sort New value
5735 */
5739 }
5741 /**
5742 * Accessor for $mDefaultSort
5743 * Will use the empty string if none is set.
5744 *
5745 * This value is treated as a prefix, so the
5746 * empty string is equivalent to sorting by
5747 * page name.
5748 *
5749 * @return string
5750 */
5756 }
5757 }
5759 /**
5760 * Accessor for $mDefaultSort
5761 * Unlike getDefaultSort(), will return false if none is set
5762 *
5763 * @return string or false
5764 */
5767 }
5769 /**
5770 * Try to guess the section anchor name based on a wikitext fragment
5771 * presumably extracted from a heading, for example "Header" from
5772 * "== Header ==".
5773 *
5774 * @param $text string
5775 *
5776 * @return string
5777 */
5779 # Strip out wikitext links(they break the anchor)
5783 }
5785 /**
5786 * Same as guessSectionNameFromWikiText(), but produces legacy anchors
5787 * instead. For use in redirects, since IE6 interprets Redirect: headers
5788 * as something other than UTF-8 (apparently?), resulting in breakage.
5789 *
5790 * @param string $text The section name
5791 * @return string An anchor
5792 */
5794 # Strip out wikitext links(they break the anchor)
5798 }
5800 /**
5801 * Strips a text string of wikitext for use in a section anchor
5802 *
5803 * Accepts a text string and then removes all wikitext from the
5804 * string and leaves only the resultant text (i.e. the result of
5805 * [[User:WikiSysop|Sysop]] would be "Sysop" and the result of
5806 * [[User:WikiSysop]] would be "User:WikiSysop") - this is intended
5807 * to create valid section anchors by mimicing the output of the
5808 * parser when headings are parsed.
5809 *
5810 * @param string $text text string to be stripped of wikitext
5811 * for use in a Section anchor
5812 * @return string Filtered text string
5813 */
5815 # Strip internal link markup
5819 # Strip external link markup
5820 # @todo FIXME: Not tolerant to blank link text
5821 # I.E. [http://www.mediawiki.org] will render as [1] or something depending
5822 # on how many empty links there are on the page - need to figure that out.
5823 $text = preg_replace( '/\[(?i:' . $this->mUrlProtocols . ')([^ ]+?) ([^[]+)\]/', '$2', $text );
5825 # Parse wikitext quotes (italics & bold)
5828 # Strip HTML tags
5831 }
5833 /**
5834 * strip/replaceVariables/unstrip for preprocessor regression testing
5835 *
5836 * @param $text string
5837 * @param $title Title
5838 * @param $options ParserOptions
5839 * @param $outputType int
5840 *
5841 * @return string
5842 */
5843 function testSrvus( $text, Title $title, ParserOptions $options, $outputType = self::OT_HTML ) {
5850 }
5852 /**
5853 * @param $text string
5854 * @param $title Title
5855 * @param $options ParserOptions
5856 * @return string
5857 */
5860 }
5862 /**
5863 * @param $text
5864 * @param $title Title
5865 * @param $options ParserOptions
5866 * @return string
5867 */
5870 }
5872 /**
5873 * Call a callback function on all regions of the given text that are not
5874 * inside strip markers, and replace those regions with the return value
5875 * of the callback. For example, with input:
5876 *
5877 * aaa<MARKER>bbb
5878 *
5879 * This will call the callback function twice, with 'aaa' and 'bbb'. Those
5880 * two strings will be replaced with the value returned by the callback in
5881 * each case.
5882 *
5883 * @param $s string
5884 * @param $callback
5885 *
5886 * @return string
5887 */
5906 }
5907 }
5908 }
5910 }
5912 /**
5913 * Remove any strip markers found in the given text.
5914 *
5915 * @param $text Input string
5916 * @return string
5917 */
5920 }
5922 /**
5923 * Save the parser state required to convert the given half-parsed text to
5924 * HTML. "Half-parsed" in this context means the output of
5925 * recursiveTagParse() or internalParse(). This output has strip markers
5926 * from replaceVariables (extensionSubstitution() etc.), and link
5927 * placeholders from replaceLinkHolders().
5928 *
5929 * Returns an array which can be serialized and stored persistently. This
5930 * array can later be loaded into another parser instance with
5931 * unserializeHalfParsedText(). The text can then be safely incorporated into
5932 * the return value of a parser hook.
5933 *
5934 * @param $text string
5935 *
5936 * @return array
5937 */
5945 );
5948 }
5950 /**
5951 * Load the parser state given in the $data array, which is assumed to
5952 * have been generated by serializeHalfParsedText(). The text contents is
5953 * extracted from the array, and its markers are transformed into markers
5954 * appropriate for the current Parser instance. This transformed text is
5955 * returned, and can be safely included in the return value of a parser
5956 * hook.
5957 *
5958 * If the $data array has been stored persistently, the caller should first
5959 * check whether it is still valid, by calling isValidHalfParsedText().
5960 *
5961 * @param array $data Serialized data
5962 * @throws MWException
5963 * @return String
5964 */
5968 }
5970 # First, extract the strip state.
5974 # Now renumber links
5977 # Should be good to go.
5979 }
5981 /**
5982 * Returns true if the given array, presumed to be generated by
5983 * serializeHalfParsedText(), is compatible with the current version of the
5984 * parser.
5985 *
5986 * @param $data Array
5987 *
5988 * @return bool
5989 */
5992 }
5994 /**
5995 * Parsed a width param of imagelink like 300px or 200x300px
5996 *
5997 * @param $value String
5998 *
5999 * @return array
6000 * @since 1.20
6001 */
6006 }
6008 # (bug 13500) In both cases (width/height and width only),
6009 # permit trailing "px" for backward compatibility.
6018 }
6020 }
6021 }