| blob | abc017bc9d1c5f9290c667ef3181c882cbbbe44c |
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 * $wgUseDynamicDates
66 *
67 * @ingroup Parser
68 */
70 /**
71 * Update this version number when the ParserOutput format
72 * changes in an incompatible way, so the parser cache
73 * can automatically discard old data.
74 */
77 /**
78 * Update this version number when the output of serialiseHalfParsedText()
79 * changes in an incompatible way
80 */
83 # Flags for Parser::setFunctionHook
84 # Also available as global constants from Defines.php
88 # Constants needed for external link processing
89 # Everything except bracket, space, or control characters
90 # \p{Zs} is unicode 'separator, space' category. It covers the space 0x20
91 # as well as U+3000 is IDEOGRAPHIC SPACE for bug 19052
96 # State constants for the definition list colon extraction
106 # Flags for preprocessToDom
109 # Allowed values for $this->mOutputType
110 # Parameter to startExternalParse().
117 # Marker Suffix needs to be accessible staticly.
120 # Persistent:
134 # Initialised by initialiseVariables()
136 /**
137 * @var MagicWordArray
138 */
141 /**
142 * @var MagicWordArray
143 */
145 var $mConf, $mPreprocessor, $mExtLinkBracketedRegex, $mUrlProtocols; # Initialised in constructor
147 # Cleared with clearState():
148 /**
149 * @var ParserOutput
150 */
154 /**
155 * @var StripState
156 */
160 /**
161 * @var LinkHolderArray
162 */
173 /**
174 * @var User
175 */
178 # Temporary
179 # These are variables reset at least once per parse regardless of $clearState
181 /**
182 * @var ParserOptions
183 */
186 /**
187 * @var Title
188 */
198 /**
199 * @var string
200 */
203 /**
204 * @var Array with the language name of each language link (i.e. the
205 * interwiki prefix) in the key, value arbitrary. Used to avoid sending
206 * duplicate language links to the ParserOutput.
207 */
210 /**
211 * Constructor
212 *
213 * @param $conf array
214 */
223 # Preprocessor_Hash is much faster than Preprocessor_DOM in compiled mode
226 # PECL extension that conflicts with the core DOM extension (bug 13770)
233 }
235 }
237 /**
238 * Reduce memory usage to reduce the impact of circular references
239 */
243 }
246 }
247 }
249 /**
250 * Do various kinds of initialisation on the first call of the parser
251 */
255 }
266 }
268 /**
269 * Clear Parser state
270 *
271 * @private
272 */
277 }
294 /**
295 * Prefix for temporary replacement strings for the multipass parser.
296 * \x07 should never appear in input as it's disallowed in XML.
297 * Using it at the front also gives us a little extra robustness
298 * since it shouldn't match when butted up against identifier-like
299 * string constructs.
300 *
301 * Must not consist of all title characters, or else it will change
302 * the behaviour of <nowiki> in a link.
303 */
308 # Clear these on every parse, bug 4549
316 );
325 # Fix cloning
328 }
332 }
334 /**
335 * Convert wikitext to HTML
336 * Do not call this function recursively.
337 *
338 * @param $text String: text we want to parse
339 * @param $title Title object
340 * @param $options ParserOptions
341 * @param $linestart boolean
342 * @param $clearState boolean
343 * @param $revid Int: number to pass in {{REVISIONID}}
344 * @return ParserOutput a ParserOutput
345 */
346 public function parse( $text, Title $title, ParserOptions $options, $linestart = true, $clearState = true, $revid = null ) {
347 /**
348 * First pass--just handle <nowiki> sections, pass the rest off
349 * to internalParse() which does all the real work.
350 */
359 # Remove the strip marker tag prefix from the input, if present.
362 }
373 }
376 # No more strip!
383 # Clean up special characters, only run once, next-to-last before doBlockLevels
385 # french spaces, last one Guillemet-left
386 # only if there is something before the space
388 # french spaces, Guillemet-right
391 );
398 /**
399 * The input doesn't get language converted if
400 * a) It's disabled
401 * b) Content isn't converted
402 * c) It's a conversion table
403 * d) it is an interface message (which is in the user language)
404 */
407 {
408 # Run convert unconditionally in 1.18-compatible mode
411 # The position of the convert() call should not be changed. it
412 # assumes that the links are all replaced and the only thing left
413 # is the <nowiki> mark.
415 }
416 }
418 /**
419 * A converted title will be provided in the output object if title and
420 * content conversion are enabled, the article text does not contain
421 * a conversion-suppressing double-underscore tag, and no
422 * {{DISPLAYTITLE:...}} is present. DISPLAYTITLE takes precedence over
423 * automatic link conversion.
424 */
429 {
436 }
437 }
451 # attempt to sanitize at least some nesting problems
452 # (bug #2702 and quite a few others)
454 # ''Something [http://www.cool.com cool''] -->
455 # <i>Something</i><a href="http://www.cool.com"..><i>cool></i></a>
458 # fix up an anchor inside another anchor, only
459 # at least for a single single nested link (bug 3695)
462 # fix div inside inline elements- doBlockLevels won't wrap a line which
463 # contains a div, so fix it up here; replace
464 # div with escaped text
467 # remove empty italic or bold tag pairs, some
468 # introduced by rules above
470 );
476 }
482 );
483 }
487 # Information on include size limits, for the benefit of users who try to skirt them
490 $PFreport = "Expensive parser function count: {$this->mExpensiveFunctionCount}/{$this->mOptions->getExpensiveParserFunctionLimit()}\n";
493 "Preprocessor visited node count: {$this->mPPNodeCount}/{$this->mOptions->getMaxPPNodeCount()}\n" .
498 "Highest expansion depth: {$this->mHighestExpansionDepth}/{$this->mOptions->getMaxPPExpandDepth()}\n".
506 }
507 }
518 }
520 /**
521 * Recursive parser entry point that can be called from an extension tag
522 * hook.
523 *
524 * If $frame is not provided, then template variables (e.g., {{{1}}}) within $text are not expanded
525 *
526 * @param $text String: text extension wants to have parsed
527 * @param $frame PPFrame: The frame to use for expanding any template variables
528 *
529 * @return string
530 */
538 }
540 /**
541 * Expand templates and variables in the text, producing valid, static wikitext.
542 * Also removes comments.
543 * @return mixed|string
544 */
550 }
557 }
559 /**
560 * Recursive parser entry point that can be called from an extension tag
561 * hook.
562 *
563 * @param $text String: text to be expanded
564 * @param $frame PPFrame: The frame to use for expanding any template variables
565 * @return String
566 * @since 1.19
567 */
574 }
576 /**
577 * Process the wikitext for the "?preload=" feature. (bug 5210)
578 *
579 * "<noinclude>", "<includeonly>" etc. are parsed as for template
580 * transclusion, comments, templates, arguments, tags hooks and parser
581 * functions are untouched.
582 *
583 * @param $text String
584 * @param $title Title
585 * @param $options ParserOptions
586 * @return String
587 */
589 # Parser (re)initialisation
597 }
599 /**
600 * Get a random string
601 *
602 * @return string
603 */
606 }
608 /**
609 * Set the current user.
610 * Should only be used when doing pre-save transform.
611 *
612 * @param $user Mixed: User object or null (to reset)
613 */
616 }
618 /**
619 * Accessor for mUniqPrefix.
620 *
621 * @return String
622 */
625 # @todo FIXME: This is probably *horribly wrong*
626 # LanguageConverter seems to want $wgParser's uniqPrefix, however
627 # if this is called for a parser cache hit, the parser may not
628 # have ever been initialized in the first place.
629 # Not really sure what the heck is supposed to be going on here.
631 # throw new MWException( "Accessing uninitialized mUniqPrefix" );
632 }
634 }
636 /**
637 * Set the context title
638 *
639 * @param $t Title
640 */
644 }
647 # Strip the fragment to avoid various odd effects
652 }
653 }
655 /**
656 * Accessor for the Title object
657 *
658 * @return Title object
659 */
662 }
664 /**
665 * Accessor/mutator for the Title object
666 *
667 * @param $x Title object or null to just get the current one
668 * @return Title object
669 */
672 }
674 /**
675 * Set the output type
676 *
677 * @param $ot Integer: new value
678 */
681 # Shortcut alias
687 );
688 }
690 /**
691 * Accessor/mutator for the output type
692 *
693 * @param $x int|null New value or null to just get the current one
694 * @return Integer
695 */
698 }
700 /**
701 * Get the ParserOutput object
702 *
703 * @return ParserOutput object
704 */
707 }
709 /**
710 * Get the ParserOptions object
711 *
712 * @return ParserOptions object
713 */
716 }
718 /**
719 * Accessor/mutator for the ParserOptions object
720 *
721 * @param $x ParserOptions New value or null to just get the current one
722 * @return ParserOptions Current ParserOptions object
723 */
726 }
728 /**
729 * @return int
730 */
733 }
735 /**
736 * @param $id int
737 */
740 }
742 /**
743 * Get a language object for use in parser functions such as {{FORMATNUM:}}
744 * @return Language
745 */
748 }
750 /**
751 * Get the target language for the content being parsed. This is usually the
752 * language that the content is in.
753 *
754 * @since 1.19
755 *
756 * @return Language|null
757 */
767 }
770 }
772 /**
773 * Get the language object for language conversion
774 */
781 }
782 }
784 /**
785 * Get a User object either from $this->mUser, if set, or from the
786 * ParserOptions object otherwise
787 *
788 * @return User object
789 */
793 }
795 }
797 /**
798 * Get a preprocessor object
799 *
800 * @return Preprocessor instance
801 */
806 }
808 }
810 /**
811 * Replaces all occurrences of HTML-style comments and the given tags
812 * in the text with a random marker and returns the next text. The output
813 * parameter $matches will be an associative array filled with data in
814 * the form:
815 *
816 * @code
817 * 'UNIQ-xxxxx' => array(
818 * 'element',
819 * 'tag content',
820 * array( 'param' => 'x' ),
821 * '<element param="x">tag content</element>' ) )
822 * @endcode
823 *
824 * @param $elements array list of element names. Comments are always extracted.
825 * @param $text string Source text string.
826 * @param $matches array Out parameter, Array: extracted tags
827 * @param $uniq_prefix string
828 * @return String: stripped text
829 */
843 }
845 # comment
851 # tag
856 }
862 # Empty element tag, <tag />
871 }
875 # No end tag -- let it run out to the end of the text.
881 }
882 }
888 }
890 }
892 /**
893 * Get a list of strippable XML-like elements
894 *
895 * @return array
896 */
899 }
901 /**
902 * Add an item to the strip state
903 * Returns the unique tag which must be inserted into the stripped text
904 * The tag will be replaced with the original text in unstrip()
905 *
906 * @param $text string
907 *
908 * @return string
909 */
915 }
917 /**
918 * parse the wiki syntax used to render tables
919 *
920 * @private
921 * @return string
922 */
941 }
947 # First check if we are starting a new table
960 # Don't do any of the following
964 # We are ending a table
970 }
974 }
978 }
982 # Now we have a table row
985 # Whats after the tag is now only attributes
998 }
1002 }
1008 } elseif ( $first_character === '|' || $first_character === '!' || substr( $line , 0 , 2 ) === '|+' ) {
1009 # This might be cell elements, td, th or captions
1013 }
1019 }
1021 # Split up multiple cells on the same line.
1022 # FIXME : This can result in improper nesting of tags processed
1023 # by earlier parser steps, but should avoid splitting up eg
1024 # attribute values containing literal "||".
1029 # Loop through each table cell
1036 }
1041 }
1047 }
1057 }
1061 # A cell could contain both parameters and data
1064 # Bug 553: Note that a '|' inside an invalid link should not
1065 # be mistaken as delimiting cell parameters
1074 }
1078 }
1079 }
1081 }
1083 # Closing open td, tr && table
1087 }
1090 }
1093 }
1096 }
1098 # Remove trailing line-ending (b/c)
1101 }
1103 # special case: don't return empty table
1106 }
1111 }
1113 /**
1114 * Helper function for parse() that transforms wiki markup into
1115 * HTML. Only called for $mOutputType == self::OT_HTML.
1116 *
1117 * @private
1118 *
1119 * @param $text string
1120 * @param $isMain bool
1121 * @param $frame bool
1122 *
1123 * @return string
1124 */
1130 # Hook to suspend the parser in this state
1131 if ( !wfRunHooks( 'ParserBeforeInternalParse', array( &$this, &$text, &$this->mStripState ) ) ) {
1134 }
1136 # if $frame is provided, then use $frame for replacing any variables
1138 # use frame depth to infer how include/noinclude tags should be handled
1139 # depth=0 means this is the top-level document; otherwise it's an included document
1144 }
1148 # if $frame is not provided, then use old-style replaceVariables
1150 }
1153 $text = Sanitizer::removeHTMLtags( $text, array( &$this, 'attributeStripCallback' ), false, array_keys( $this->mTransparentTagHooks ) );
1156 # Tables need to come after variable replacement for things to work
1157 # properly; putting them before other transformations should keep
1158 # exciting things like link expansions from showing up in surprising
1159 # places.
1170 }
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
1311 # Register it in the output object...
1312 # Replace unnecessary URL escape codes with their equivalent characters
1315 }
1318 }
1321 /**
1322 * Parse headers and return html
1323 *
1324 * @private
1325 *
1326 * @param $text string
1327 *
1328 * @return string
1329 */
1336 }
1339 }
1341 /**
1342 * Replace single quotes with HTML markup
1343 * @private
1344 *
1345 * @param $text string
1346 *
1347 * @return string the altered text
1348 */
1355 }
1359 }
1361 /**
1362 * Helper function for doAllQuotes()
1363 *
1364 * @param $text string
1365 *
1366 * @return string
1367 */
1373 # First, do some preliminary work. This may shift some apostrophes from
1374 # being mark-up to being text. It also counts the number of occurrences
1375 # of bold and italics mark-ups.
1380 # If there are ever four apostrophes, assume the first is supposed to
1381 # be text, and the remaining three constitute mark-up for bold text.
1386 # If there are more than 5 apostrophes in a row, assume they're all
1387 # text except for the last 5.
1390 }
1391 # Count the number of occurrences of bold and italics mark-ups.
1392 # We are not counting sequences of five apostrophes.
1400 }
1401 }
1402 }
1404 # If there is an odd number of both bold and italics, it is likely
1405 # that one of the bold ones was meant to be an apostrophe followed
1406 # by italics. Which one we cannot know for certain, but it is more
1407 # likely to be one that has a single-letter word before it.
1420 }
1424 }
1428 }
1429 }
1430 }
1432 }
1434 # If there is a single-letter word, use it!
1439 # If not, but there's a multi-letter word, use that one.
1443 # ... otherwise use the first one that has neither.
1444 # (notice that it is possible for all three to be -1 if, for example,
1445 # there is only one pentuple-apostrophe in the line)
1448 }
1449 }
1451 # Now let's actually convert our apostrophic mush to HTML!
1462 }
1475 }
1487 }
1501 }
1502 }
1503 }
1505 }
1506 # Now close all remaining tags. Notice that the order is important.
1509 }
1512 }
1515 }
1516 # There might be lonely ''''', so make sure we have a buffer
1519 }
1521 }
1522 }
1524 /**
1525 * Replace external links (REL)
1526 *
1527 * Note: this is all very hackish and the order of execution matters a lot.
1528 * Make sure to run maintenance/parserTests.php if you change this code.
1529 *
1530 * @private
1531 *
1532 * @param $text string
1533 *
1534 * @return string
1535 */
1541 throw new MWException( "PCRE needs to be compiled with --enable-unicode-properties in order for MediaWiki to function" );
1542 }
1552 # The characters '<' and '>' (which were escaped by
1553 # removeHTMLtags()) should not be included in
1554 # URLs, per RFC 2396.
1559 }
1561 # If the link text is an image URL, replace it with an <img> tag
1562 # This happened by accident in the original parser, but some people used it extensively
1566 }
1570 # Set linktype for CSS - if URL==text, link is essentially free
1573 # No link text, e.g. [http://domain.tld/some.link]
1575 # Autonumber
1580 # Have link text, e.g. [http://domain.tld/some.link text]s
1581 # Check for trail
1583 }
1589 # Use the encoded URL
1590 # This means that users can paste URLs directly into the text
1591 # Funny characters like ö aren't valid in URLs anyway
1592 # This was changed in August 2004
1596 # Register link in the output object.
1597 # Replace unnecessary URL escape codes with the referenced character
1598 # This prevents spammers from hiding links from the filters
1601 }
1605 }
1607 /**
1608 * Get an associative array of additional HTML attributes appropriate for a
1609 * particular external link. This currently may include rel => nofollow
1610 * (depending on configuration, namespace, and the URL's domain) and/or a
1611 * target attribute (depending on configuration).
1612 *
1613 * @param $url String|bool optional URL, to extract the domain from for rel =>
1614 * nofollow if appropriate
1615 * @return Array associative array of HTML attributes
1616 */
1623 {
1625 }
1628 }
1630 }
1632 /**
1633 * Replace unusual URL escape codes with their equivalent characters
1634 *
1635 * @param $url String
1636 * @return String
1637 *
1638 * @todo This can merge genuinely required bits in the path or query string,
1639 * breaking legit URLs. A proper fix would treat the various parts of
1640 * the URL differently; as a workaround, just use the output for
1641 * statistical records, not for actual linking/output.
1642 */
1646 }
1648 /**
1649 * Callback function used in replaceUnusualEscapes().
1650 * Replaces unusual URL escape codes with their equivalent character
1651 *
1652 * @param $matches array
1653 *
1654 * @return string
1655 */
1659 # Is it an unsafe or HTTP reserved character according to RFC 1738?
1661 # No, shouldn't be escaped
1664 # Yes, leave it escaped
1666 }
1667 }
1669 /**
1670 * make an image if it's allowed, either through the global
1671 * option, through the exception, or through the on-wiki whitelist
1672 * @private
1673 *
1674 * $param $url string
1675 *
1676 * @return string
1677 */
1682 # $imagesfrom could be either a single string or an array of strings, parse out the latter
1689 }
1690 }
1695 }
1699 # Image found
1701 }
1702 }
1705 $whitelist = explode( "\n", wfMessage( 'external_image_whitelist' )->inContentLanguage()->text() );
1707 # Sanitize the regex fragment, make it case-insensitive, ignore blank entries/comments
1710 }
1712 # Image matches a whitelist entry
1715 }
1716 }
1717 }
1719 }
1721 /**
1722 * Process [[ ]] wikilinks
1723 *
1724 * @param $s string
1725 *
1726 * @return String: processed text
1727 *
1728 * @private
1729 */
1733 }
1735 /**
1736 * Process [[ ]] wikilinks (RIL)
1737 * @return LinkHolderArray
1738 *
1739 * @private
1740 */
1746 # the % is needed to support urlencoded titles as well
1749 # Match a link having the form [[namespace:link|alternate]]trail
1751 # Match cases where there is no "]]", which might still be images
1753 }
1757 # split the entire text string on occurrences of [[
1759 # get the first element (all text up to first [[), and remove the space we added
1768 # Match the end of a line for a word that's not followed by whitespace,
1769 # e.g. in the case of 'The Arab al[[Razi]]', 'al' will be matched
1771 }
1777 }
1786 }
1789 }
1796 }
1800 # Loop for each link
1802 # Check for excessive memory usage
1804 # Too big
1805 # Do the existence check, replace the link holders and clear the array
1808 }
1817 }
1818 # first link
1822 }
1824 }
1831 # If we get a ] at the beginning of $m[3] that means we have a link that's something like:
1832 # [[Image:Foo.jpg|[http://example.com desc]]] <- having three ] in a row fucks up,
1833 # the real problem is with the $e1 regex
1834 # See bug 1300.
1835 #
1836 # Still some problems for cases where the ] is meant to be outside punctuation,
1837 # and no image is in sight. See bug 2095.
1838 #
1842 )
1843 {
1846 }
1847 # fix up urlencoded title texts
1849 # Should anchors '#' also be rejected?
1851 }
1853 } elseif ( preg_match( $e1_img, $line, $m ) ) { # Invalid, but might be an image with a link in its caption
1858 }
1864 }
1868 # Don't allow internal links to pages containing
1869 # PROTO: where PROTO is a valid URL protocol; these
1870 # should be external links.
1875 }
1877 # Make subpage if necessary
1882 }
1886 # Strip off leading ':'
1888 }
1897 }
1908 # look at the next 'line' to see if we can close it there
1913 }
1916 # the first ]] closes the inner link, the second the image
1922 # if there's exactly one ]] that's fine, we'll keep looking
1925 # if $next_line is invalid too, we need look no further
1928 }
1929 }
1931 # we couldn't find the end of this imageLink, so output it raw
1932 # but don't ignore what might be perfectly normal links in the text we've examined
1935 # note: no $trail, because without an end, there *is* no trail
1938 }
1941 # note: no $trail, because without an end, there *is* no trail
1944 }
1946 }
1952 # Bug 4598 madness. Handle the quotes only if they come from the alternate part
1953 # [[Lista d''e paise d''o munno]] -> <a href="...">Lista d''e paise d''o munno</a>
1954 # [[Criticism of Harry Potter|Criticism of ''Harry Potter'']]
1955 # -> <a href="Criticism of Harry Potter">Criticism of <i>Harry Potter</i></a>
1957 }
1959 # Link not escaped by : , create the various objects
1961 # Interwikis
1963 if ( $iw && $this->mOptions->getInterwikiMagic() && $nottalk && Language::fetchLanguageName( $iw, null, 'mw' ) ) {
1964 # Bug 24502: filter duplicates
1968 }
1973 }
1980 # if no parameters were passed, $text
1981 # becomes something like "File:Foo.png",
1982 # which we don't want to pass on to the
1983 # image generator
1986 # recursively parse links inside the image caption
1987 # actually, this will parse them in any other parameters, too,
1988 # but it might be hard to fix that, and it doesn't matter ATM
1991 }
1992 # cloak any absolute URLs inside the image markup, so replaceExternalLinks() won't touch them
1997 }
2000 }
2010 }
2016 /**
2017 * Strip the whitespace Category links produce, see bug 87
2018 * @todo We might want to use trim($tmp, "\n") here.
2019 */
2024 }
2025 }
2027 # Self-link checking
2032 }
2033 }
2035 # NS_MEDIA is a pseudo-namespace for linking directly to a file
2036 # @todo FIXME: Should do batch file existence checks, see comment below
2039 # Give extensions a chance to select the file revision for us
2044 # Fetch and register the file (file title may be different via hooks)
2046 # Cloak with NOPARSE to avoid replacement in replaceExternalLinks
2051 }
2054 # Some titles, such as valid special pages or files in foreign repos, should
2055 # be shown as bluelinks even though they're not included in the page table
2056 #
2057 # @todo FIXME: isAlwaysKnown() can be expensive for file links; we should really do
2058 # batch file existence checks for NS_FILE and NS_MEDIA
2063 # Links will be added to the output link list after checking
2065 }
2067 }
2070 }
2072 /**
2073 * Render a forced-blue link inline; protect against double expansion of
2074 * URLs if we're in a mode that prepends full URL prefixes to internal links.
2075 * Since this little disaster has to split off the trail text to avoid
2076 * breaking URLs in the following text without breaking trails on the
2077 * wiki links, it's been made into a horrible function.
2078 *
2079 * @param $nt Title
2080 * @param $text String
2081 * @param $query Array or String
2082 * @param $trail String
2083 * @param $prefix String
2084 * @return String: HTML-wikitext mix oh yuck
2085 */
2091 }
2094 }
2099 }
2101 /**
2102 * Insert a NOPARSE hacky thing into any inline links in a chunk that's
2103 * going to go through further parsing steps before inline URL expansion.
2104 *
2105 * Not needed quite as much as it used to be since free links are a bit
2106 * more sensible these days. But bracketed links are still an issue.
2107 *
2108 * @param $text String: more-or-less HTML
2109 * @return String: less-or-more HTML with NOPARSE bits
2110 */
2114 }
2116 /**
2117 * Return true if subpage links should be expanded on this page.
2118 * @return Boolean
2119 */
2121 # Some namespaces don't allow subpages
2123 }
2125 /**
2126 * Handle link to subpage if necessary
2127 *
2128 * @param $target String: the source of the link
2129 * @param &$text String: the link text, modified as necessary
2130 * @return string the full name of the link
2131 * @private
2132 */
2135 }
2137 /**#@+
2138 * Used by doBlockLevels()
2139 * @private
2140 *
2141 * @return string
2142 */
2147 }
2151 }
2153 /**
2154 * getCommon() returns the length of the longest common substring
2155 * of both arguments, starting at the beginning of both.
2156 * @private
2157 *
2158 * @param $st1 string
2159 * @param $st2 string
2160 *
2161 * @return int
2162 */
2168 }
2173 }
2174 }
2176 }
2178 /**
2179 * These next three functions open, continue, and close the list
2180 * element appropriate to the prefix character passed into them.
2181 * @private
2182 *
2183 * @param $char string
2184 *
2185 * @return string
2186 */
2201 }
2204 }
2206 /**
2207 * TODO: document
2208 * @param $char String
2209 * @private
2210 *
2211 * @return string
2212 */
2220 }
2227 }
2228 }
2230 }
2232 /**
2233 * TODO: document
2234 * @param $char String
2235 * @private
2236 *
2237 * @return string
2238 */
2250 }
2253 }
2255 }
2256 /**#@-*/
2258 /**
2259 * Make lists from lines starting with ':', '*', '#', etc. (DBL)
2260 *
2261 * @param $text String
2262 * @param $linestart Boolean: whether or not this is at the start of a line.
2263 * @private
2264 * @return string the lists rendered as HTML
2265 */
2269 # Parsing through the text line by line. The main thing
2270 # happening here is handling of block-level elements p, pre,
2271 # and making lists from lines starting with * # : etc.
2272 #
2281 # Fix up $linestart
2286 }
2287 # * = ul
2288 # # = ol
2289 # ; = dt
2290 # : = dd
2295 # If not in a <pre> element, scan for and figure out what prefixes are there.
2297 # Multiple prefixes may abut each other for nested lists.
2301 # eh?
2302 # ; and : are both from definition-lists, so they're equivalent
2303 # for the purposes of determining whether or not we need to open/close
2304 # elements.
2309 # Don't interpret any other prefixes in preformatted text
2313 }
2315 # List generation
2317 # Same as the last item, so no need to deal with nesting or opening stuff
2322 # The one nasty exception: definition lists work like this:
2323 # ; title : definition text
2324 # So we check for : in the remainder text to split up the
2325 # title and definition, without b0rking links.
2330 }
2331 }
2333 # We need to open or close prefixes, or both.
2335 # Either open or close a level...
2339 # Close all the prefixes which aren't shared.
2343 }
2345 # Continue the current prefix if appropriate.
2348 }
2350 # Open prefixes where appropriate.
2356 # @todo FIXME: This is dupe of code above
2360 }
2361 }
2363 }
2365 }
2367 # If we have no prefixes, go to paragraph mode.
2370 # No prefix (not in list)--go to paragraph mode
2371 # XXX: use a stack for nestable elements like span, table and div
2372 $openmatch = preg_match('/(?:<table|<blockquote|<h1|<h2|<h3|<h4|<h5|<h6|<pre|<tr|<p|<ul|<ol|<li|<\\/tr|<\\/td|<\\/th)/iS', $t );
2375 '<td|<th|<\\/?div|<hr|<\\/pre|<\\/p|'.$this->mUniqPrefix.'-pre|<\\/li|<\\/ul|<\\/ol|<\\/?center)/iS', $t );
2378 # TODO bug 5718: paragraph closed
2382 }
2386 # pre
2391 }
2394 # paragraph
2407 }
2408 }
2417 }
2418 }
2419 }
2420 }
2422 }
2423 # somewhere above we forget to get out of pre block (bug 785)
2426 }
2429 }
2430 }
2434 }
2438 }
2442 }
2444 /**
2445 * Split up a string on ':', ignoring any occurrences inside tags
2446 * to prevent illegal overlapping.
2447 *
2448 * @param $str String the string to split
2449 * @param &$before String set to everything before the ':'
2450 * @param &$after String set to everything after the ':'
2451 * @return String the position of the ':', or false if none found
2452 */
2458 # Nothing to find!
2461 }
2465 # Easy; no tag nesting to worry about
2470 }
2472 # Ugly state machine to walk through avoiding tags.
2480 # (Using the number is a performance hack for common cases)
2484 # Could be either a <start> tag or an </end> tag
2489 # We found it!
2494 }
2495 # Embedded in a tag; don't break it.
2498 # Skip ahead looking for something interesting
2501 # Nothing else interesting
2504 }
2508 # We found it!
2513 }
2514 }
2516 # Nothing else interesting to find; abort!
2517 # We're nested, but there's no close tags left. Abort!
2519 }
2520 # Skip ahead to next tag start
2523 }
2526 # In a <tag>
2533 # Slash may be followed by >?
2537 # ignore
2538 }
2549 # Illegal early close? This shouldn't happen D:
2554 }
2557 # In a </tag>
2564 }
2566 }
2570 # Yes, a self-closed tag <blah/>
2573 # Probably we're jumping the gun, and this is an attribute
2575 }
2580 }
2587 }
2594 }
2598 }
2599 }
2604 }
2607 }
2609 /**
2610 * Return value of a magic variable (like PAGENAME)
2611 *
2612 * @private
2613 *
2614 * @param $index integer
2615 * @param $frame PPFrame
2616 *
2617 * @return string
2618 */
2624 // If no title set, bad things are going to happen
2625 // later. Title should always be set since this
2626 // should only be called in the middle of a parse
2627 // operation (but the unit-tests do funky stuff)
2630 }
2632 /**
2633 * Some of these require message or data lookups and can be
2634 * expensive to check many times.
2635 */
2639 }
2640 }
2645 # Use the time zone
2650 }
2664 }
2733 $value = wfEscapeWikiText( wfUrlEncode( str_replace( ' ', '_', $this->mTitle->getBaseText() ) ) );
2741 }
2749 }
2762 # 0 means the page doesn't exist in the database,
2763 # which means the user is previewing a new page.
2764 # The vary-revision flag must be set, because the magic word
2765 # will have a different value once the page is saved.
2768 }
2772 # Let the edit saving system know we should parse the page
2773 # *after* a revision ID has been assigned.
2779 # Let the edit saving system know we should parse the page
2780 # *after* a revision ID has been assigned. This is for null edits.
2786 # Let the edit saving system know we should parse the page
2787 # *after* a revision ID has been assigned. This is for null edits.
2793 # Let the edit saving system know we should parse the page
2794 # *after* a revision ID has been assigned. This is for null edits.
2800 # Let the edit saving system know we should parse the page
2801 # *after* a revision ID has been assigned. This is for null edits.
2807 # Let the edit saving system know we should parse the page
2808 # *after* a revision ID has been assigned. This is for null edits.
2814 # Let the edit saving system know we should parse the page
2815 # *after* a revision ID has been assigned. This is for null edits.
2821 # Let the edit saving system know we should parse the page
2822 # *after* a revision ID has been assigned. This is for null edits.
2837 $value = $this->mTitle->canTalk() ? str_replace( '_',' ',$this->mTitle->getTalkNsText() ) : '';
2861 # @bug 4594 PHP5 has it zero padded, PHP4 does not, cast to
2862 # int to remove the padding
2881 # @bug 4594 PHP5 has it zero padded, PHP4 does not, cast to
2882 # int to remove the padding
2942 if ( wfRunHooks( 'ParserGetVariableValueSwitch', array( &$this, &$this->mVarCache, &$index, &$ret, &$frame ) ) ) {
2946 }
2947 }
2951 }
2954 }
2956 /**
2957 * initialise the magic variables (like CURRENTMONTHNAME) and substitution modifiers
2958 *
2959 * @private
2960 */
2969 }
2971 /**
2972 * Preprocess some wikitext and return the document tree.
2973 * This is the ghost of replace_variables().
2974 *
2975 * @param $text String: The text to parse
2976 * @param $flags Integer: bitwise combination of:
2977 * self::PTD_FOR_INCLUSION Handle "<noinclude>" and "<includeonly>" as if the text is being
2978 * included. Default is to assume a direct page view.
2979 *
2980 * The generated DOM tree must depend only on the input text and the flags.
2981 * The DOM tree must be the same in OT_HTML and OT_WIKI mode, to avoid a regression of bug 4899.
2982 *
2983 * Any flag added to the $flags parameter here, or any other parameter liable to cause a
2984 * change in the DOM tree for a given text, must be passed through the section identifier
2985 * in the section edit link and thus back to extractSections().
2986 *
2987 * The output of this function is currently only cached in process memory, but a persistent
2988 * cache may be implemented at a later date which takes further advantage of these strict
2989 * dependency requirements.
2990 *
2991 * @private
2992 *
2993 * @return PPNode
2994 */
2998 }
3000 /**
3001 * Return a three-element array: leading whitespace, string contents, trailing whitespace
3002 *
3003 * @param $s string
3004 *
3005 * @return array
3006 */
3016 }
3018 }
3020 /**
3021 * Replace magic variables, templates, and template arguments
3022 * with the appropriate text. Templates are substituted recursively,
3023 * taking care to avoid infinite loops.
3024 *
3025 * Note that the substitution depends on value of $mOutputType:
3026 * self::OT_WIKI: only {{subst:}} templates
3027 * self::OT_PREPROCESS: templates but not extension tags
3028 * self::OT_HTML: all templates and extension tags
3029 *
3030 * @param $text String the text to transform
3031 * @param $frame PPFrame Object describing the arguments passed to the template.
3032 * Arguments may also be provided as an associative array, as was the usual case before MW1.12.
3033 * Providing arguments this way may be useful for extensions wishing to perform variable replacement explicitly.
3034 * @param $argsOnly Boolean only do argument (triple-brace) expansion, not double-brace expansion
3035 * @private
3036 *
3037 * @return string
3038 */
3040 # Is there any text? Also, Prevent too big inclusions!
3043 }
3049 wfDebug( __METHOD__." called using plain parameters instead of a PPFrame instance. Creating custom frame.\n" );
3051 }
3059 }
3061 /**
3062 * Clean up argument array - refactored in 1.9 so parserfunctions can use it, too.
3063 *
3064 * @param $args array
3065 *
3066 * @return array
3067 */
3080 }
3083 }
3084 }
3085 }
3088 }
3090 /**
3091 * Warn the user when a parser limitation is reached
3092 * Will warn at most once the user per limitation type
3093 *
3094 * @param $limitationType String: should be one of:
3095 * 'expensive-parserfunction' (corresponding messages:
3096 * 'expensive-parserfunction-warning',
3097 * 'expensive-parserfunction-category')
3098 * 'post-expand-template-argument' (corresponding messages:
3099 * 'post-expand-template-argument-warning',
3100 * 'post-expand-template-argument-category')
3101 * 'post-expand-template-inclusion' (corresponding messages:
3102 * 'post-expand-template-inclusion-warning',
3103 * 'post-expand-template-inclusion-category')
3104 * @param $current int|null Current value
3105 * @param $max int|null Maximum allowed, when an explicit limit has been
3106 * exceeded, provide the values (optional)
3107 */
3109 # does no harm if $current and $max are present but are unnecessary for the message
3114 }
3116 /**
3117 * Return the text of a template, after recursively
3118 * replacing any variables or templates within the template.
3119 *
3120 * @param $piece Array: the parts of the template
3121 * $piece['title']: the title, i.e. the part before the |
3122 * $piece['parts']: the parameter array
3123 * $piece['lineStart']: whether the brace was at the start of a line
3124 * @param $frame PPFrame The current frame, contains template arguments
3125 * @return String: the text of the template
3126 * @private
3127 */
3133 # Flags
3141 # Title object, where $text came from
3144 # $part1 is the bit before the first |, and must contain only title characters.
3145 # Various prefixes will be stripped from it later.
3150 # Original title text preserved for various purposes
3153 # $args is a list of argument nodes, starting from index 0, not including $part1
3154 # @todo FIXME: If piece['parts'] is null then the call to getLength() below won't work b/c this $args isn't an object
3160 # SUBST
3166 # Possibilities for substMatch: "subst", "safesubst" or FALSE
3167 # Decide whether to expand template or keep wikitext as-is.
3173 }
3179 }
3180 }
3185 }
3186 }
3188 # Variables
3195 }
3197 }
3198 }
3200 # MSG, MSGNW and RAW
3202 # Check for MSGNW:
3207 # Remove obsolete MSG:
3210 }
3212 # Check for RAW:
3216 }
3217 }
3220 # Parser functions
3226 # Case sensitive functions
3231 # Case insensitive functions
3237 }
3238 }
3245 # Add a frame parameter, and pass the arguments as an array
3250 }
3253 # Convert arguments to plain text
3256 }
3258 }
3260 # Workaround for PHP bug 35229 and similar
3266 }
3276 }
3278 # Extract flags into the local scope
3279 # This allows callers to set flags such as nowiki, found, etc.
3283 }
3287 }
3289 }
3290 }
3292 }
3294 # Finish mangling title and then check for loops.
3295 # Set $title to a Title object and $titleText to the PDBK
3298 # Split the title into page and subpage
3303 }
3307 # Check for language variants if the template is not found
3310 }
3311 # Do recursion depth check
3319 }
3320 }
3321 }
3323 # Load from database
3326 # Too many unique items can kill profiling DBs/collectors
3329 }
3335 {
3336 // Pass the template arguments as URL parameters.
3337 // "uselang" will have no effect since the Language object
3338 // is forced to the one defined in ParserOptions.
3346 }
3347 }
3349 // Create a new context to execute the special page
3362 }
3371 }
3372 }
3374 # If the title is valid but undisplayable, make a link to it
3378 }
3380 # Interwiki transclusion
3386 # Preprocess it like a template
3389 }
3391 }
3393 # Do infinite loop check
3394 # This has to be done after redirect resolution to avoid infinite loops via redirects
3401 }
3403 }
3405 # If we haven't found text to substitute by now, we're done
3406 # Recover the source wikitext and return it
3411 }
3414 }
3416 # Expand DOM-style return values in a child frame
3418 # Clean up argument array
3424 # Expansion is eligible for the empty-frame cache
3430 }
3432 # Uncached expansion
3434 }
3435 }
3439 }
3443 }
3445 # Replace raw HTML by a placeholder
3449 # Escape nowiki-style return values
3454 {
3455 # Bug 529: if the template begins with a table or block-level
3456 # element, it should be treated as beginning a new line.
3457 # This behaviour is somewhat controversial.
3459 }
3462 # Error, oversize inclusion
3464 # Make a working, properly escaped link if possible (bug 23588)
3467 # This will probably not be a working link, but at least it may
3468 # provide some hint of where the problem is
3471 }
3472 $text .= $this->insertStripItem( '<!-- WARNING: template omitted, post-expand include size too large -->' );
3474 }
3480 }
3484 }
3486 /**
3487 * Get the semi-parsed DOM representation of a template with a given title,
3488 * and its redirect destination title. Cached.
3489 *
3490 * @param $title Title
3491 *
3492 * @return array
3493 */
3502 }
3505 }
3507 # Cache miss, go to the database
3513 }
3521 }
3524 }
3526 /**
3527 * Fetch the unparsed text of a template and register a reference to it.
3528 * @param Title $title
3529 * @return Array ( string or false, Title )
3530 */
3532 $templateCb = $this->mOptions->getTemplateCallback(); # Defaults to Parser::statelessFetchTemplate()
3539 }
3540 }
3542 }
3544 /**
3545 * Fetch the unparsed text of a template and register a reference to it.
3546 * @param Title $title
3547 * @return mixed string or false
3548 */
3552 }
3554 /**
3555 * Static function to get a template
3556 * Can be overridden via ParserOptions::setTemplateCallback().
3557 *
3558 * @param $title Title
3559 * @param $parser Parser
3560 *
3561 * @return array
3562 */
3568 # Loop to fetch the article, with up to 1 redirect
3570 # Give extensions a chance to select the revision instead
3581 );
3583 }
3584 # Get the revision
3589 # If there is no current revision, there is no page
3593 }
3600 # We fetched a rev from a different title; register it too...
3605 }
3615 }
3619 }
3622 }
3623 # Redirect?
3626 }
3631 }
3633 /**
3634 * Fetch a file and its title and register a reference to it.
3635 * If 'broken' is a key in $options then the file will appear as a broken thumbnail.
3636 * @param Title $title
3637 * @param Array $options Array of options to RepoGroup::findFile
3638 * @return File|bool
3639 */
3643 }
3645 /**
3646 * Fetch a file and its title and register a reference to it.
3647 * If 'broken' is a key in $options then the file will appear as a broken thumbnail.
3648 * @param Title $title
3649 * @param Array $options Array of options to RepoGroup::findFile
3650 * @return Array ( File or false, Title of file )
3651 */
3659 }
3662 # Register the file as a dependency...
3665 # Update fetched file title
3668 # This file was not a redirect, but the title does not match.
3669 # Register under the new name because otherwise the link will
3670 # get lost.
3672 }
3673 }
3675 }
3677 /**
3678 * Transclude an interwiki link.
3679 *
3680 * @param $title Title
3681 * @param $action
3682 *
3683 * @return string
3684 */
3690 }
3696 }
3698 }
3700 /**
3701 * @param $url string
3702 * @return Mixed|String
3703 */
3712 }
3717 }
3724 );
3726 }
3728 /**
3729 * Triple brace replacement -- used for template arguments
3730 * @private
3731 *
3732 * @param $piece array
3733 * @param $frame PPFrame
3734 *
3735 * @return array
3736 */
3747 && (
3751 )
3752 ) {
3753 # No match in frame, use the supplied default
3755 }
3759 }
3762 # No match anywhere
3764 }
3767 }
3772 }
3776 }
3778 /**
3779 * Return the text to be used for a given extension tag.
3780 * This is the ghost of strip().
3781 *
3782 * @param $params array Associative array of parameters:
3783 * name PPNode for the tag name
3784 * attr PPNode for unparsed text where tag attributes are thought to be
3785 * attributes Optional associative array of parsed attributes
3786 * inner Contents of extension element
3787 * noClose Original text did not have a close tag
3788 * @param $frame PPFrame
3789 *
3790 * @return string
3791 */
3796 $marker = "{$this->mUniqPrefix}-$name-" . sprintf( '%08X', $this->mMarkerIndex++ ) . self::MARKER_SUFFIX;
3804 }
3810 }
3813 # Workaround for PHP bug 35229 and similar
3816 }
3823 }
3829 }
3832 # Extract flags to local scope (to override $markerType)
3837 }
3841 }
3846 }
3847 }
3853 }
3854 }
3864 }
3866 }
3868 /**
3869 * Increment an include size counter
3870 *
3871 * @param $type String: the type of expansion
3872 * @param $size Integer: the size of the text
3873 * @return Boolean: false if this inclusion would take it over the maximum, true otherwise
3874 */
3881 }
3882 }
3884 /**
3885 * Increment the expensive function count
3886 *
3887 * @return Boolean: false if the limit has been exceeded
3888 */
3892 }
3894 /**
3895 * Strip double-underscore items like __NOGALLERY__ and __NOTOC__
3896 * Fills $this->mDoubleUnderscores, returns the modified text
3897 *
3898 * @param $text string
3899 *
3900 * @return string
3901 */
3905 # The position of __TOC__ needs to be recorded
3911 # Set a placeholder. At the end we'll fill it in with the TOC.
3914 # Only keep the first one.
3916 }
3918 # Now match and remove the rest of them
3924 }
3927 }
3928 if ( isset( $this->mDoubleUnderscores['hiddencat'] ) && $this->mTitle->getNamespace() == NS_CATEGORY ) {
3930 }
3931 # (bug 8068) Allow control over whether robots index a page.
3932 #
3933 # @todo FIXME: Bug 14899: __INDEX__ always overrides __NOINDEX__ here! This
3934 # is not desirable, the last one on the page should win.
3938 }
3942 }
3944 # Cache all double underscores in the database
3947 }
3951 }
3953 /**
3954 * Add a tracking category, getting the title from a system message,
3955 * or print a debug message if the title is invalid.
3956 *
3957 * @param $msg String: message key
3958 * @return Boolean: whether the addition was successful
3959 */
3964 }
3965 // Important to parse with correct title (bug 31469)
3971 # Allow tracking categories to be disabled by setting them to "-"
3974 }
3983 }
3984 }
3986 /**
3987 * This function accomplishes several tasks:
3988 * 1) Auto-number headings if that option is enabled
3989 * 2) Add an [edit] link to sections for users who have enabled the option and can edit the page
3990 * 3) Add a Table of contents on the top for users who have enabled the option
3991 * 4) Auto-anchor headings
3992 *
3993 * It loops through all headlines, collects the necessary data, then splits up the
3994 * string and re-inserts the newly formatted headlines.
3995 *
3996 * @param $text String
3997 * @param $origText String: original, untouched wikitext
3998 * @param $isMain Boolean
3999 * @return mixed|string
4000 * @private
4001 */
4005 # Inhibit editsection links if requested in the page
4011 }
4014 }
4016 # Get all headlines for numbering them and adding funky stuff like [edit]
4017 # links - this is for later, but we need the number of headlines right now
4019 $numMatches = preg_match_all( '/<H(?P<level>[1-6])(?P<attrib>.*?'.'>)(?P<header>.*?)<\/H[1-6] *>/i', $text, $matches );
4021 # if there are fewer than 4 headlines in the article, do not show TOC
4022 # unless it's been explicitly enabled.
4026 # Allow user to stipulate that a page should have a "new section"
4027 # link added via __NEWSECTIONLINK__
4030 }
4032 # Allow user to remove the "new section"
4033 # link via __NONEWSECTIONLINK__
4036 }
4038 # if the string __FORCETOC__ (not case-sensitive) occurs in the HTML,
4039 # override above conditions and always show TOC above first header
4043 }
4045 # headline counter
4049 # Ugh .. the TOC should have neat indentation levels which can be
4050 # passed to the skin functions. These are determined here
4082 }
4086 }
4090 # Increase TOC level
4097 }
4099 # Decrease TOC level, find level to jump to
4103 # Found last matching level
4107 # Found first matching level below current level
4110 }
4111 }
4114 }
4117 # Unindent only if the previous toc level was shown :p
4122 }
4123 }
4125 # No change in level, end TOC line
4128 }
4129 }
4133 # count number of headlines for each level
4140 }
4143 }
4144 }
4146 # The safe header is a version of the header text safe to use for links
4148 # Remove link placeholders by the link text.
4149 # <!--LINK number-->
4150 # turns into
4151 # link text with suffix
4152 # Do this before unstrip since link text can contain strip markers
4155 # Avoid insertion of weird stuff like <math> by expanding the relevant sections
4158 # Strip out HTML (first regex removes any tag not allowed)
4159 # Allowed tags are <sup> and <sub> (bug 8393), <i> (bug 26375) and <b> (r105284)
4160 # We strip any parameter from accepted tags (second regex)
4164 $safeHeadline
4165 );
4168 # For the anchor, strip out HTML-y stuff period
4172 # Save headline for section edit hint before it's escaped
4176 # For reverse compatibility, provide an id that's
4177 # HTML4-compatible, like we used to.
4178 #
4179 # It may be worth noting, academically, that it's possible for
4180 # the legacy anchor to conflict with a non-legacy headline
4181 # anchor on the page. In this case likely the "correct" thing
4182 # would be to either drop the legacy anchors or make sure
4183 # they're numbered first. However, this would require people
4184 # to type in section names like "abc_.D7.93.D7.90.D7.A4"
4185 # manually, so let's not bother worrying about it.
4191 # No reason to have both (in fact, we can't)
4193 }
4198 }
4200 # HTML names must be case-insensitively unique (bug 10721).
4201 # This does not apply to Unicode characters per
4202 # http://dev.w3.org/html5/spec/infrastructure.html#case-sensitivity-and-string-comparison
4203 # @todo FIXME: We may be changing them depending on the current locale.
4209 }
4211 # count how many in assoc. array so we can track dupes in anchors
4216 }
4221 }
4223 # Don't number the heading if it is the only one (looks silly)
4225 # the two are different if the line contains a link
4226 $headline = Html::element( 'span', array( 'class' => 'mw-headline-number' ), $numbering ) . ' ' . $headline;
4227 }
4229 # Create the anchor for linking from the TOC to the section
4234 }
4237 }
4241 }
4243 # Add the section to the section tree
4244 # Find the DOM node for this header
4250 }
4251 }
4255 }
4265 );
4267 # give headline the correct <h#> tag
4269 // Output edit section links as markers with styles that can be customized by skins
4271 # Put a T flag in the section identifier, to indicate to extractSections()
4272 # that sections inside <includeonly> should be counted.
4276 }
4277 // We use a bit of pesudo-xml for editsection markers. The language converter is run later on
4278 // Using a UNIQ style marker leads to the converter screwing up the tokens when it converts stuff
4279 // And trying to insert strip tags fails too. At this point all real inputted tags have already been escaped
4280 // so we don't have to worry about a user trying to input one of these markers directly.
4281 // We use a page and section attribute to stop the language converter from converting these important bits
4282 // of data, but put the headline hint inside a content block because the language converter is supposed to
4283 // be able to convert that piece of data.
4290 }
4293 }
4299 }
4303 # Never ever show TOC if no headers
4306 }
4311 }
4314 }
4318 }
4320 # split up and insert constructed headlines
4324 // build an array of document sections
4327 // $head is zero-based, sections aren't.
4332 }
4334 /**
4335 * Send a hook, one per section.
4336 * The idea here is to be able to make section-level DIVs, but to do so in a
4337 * lower-impact, more correct way than r50769
4338 *
4339 * $this : caller
4340 * $section : the section number
4341 * &$sectionContent : ref to the content of the section
4342 * $showEditLinks : boolean describing whether this section has an edit link
4343 */
4347 }
4350 // append the TOC at the beginning
4351 // Top anchor now in skin
4353 }
4361 }
4362 }
4364 /**
4365 * Transform wiki markup when saving a page by doing "\r\n" -> "\n"
4366 * conversion, substitting signatures, {{subst:}} templates, etc.
4367 *
4368 * @param $text String: the text to transform
4369 * @param $title Title: the Title object for the current article
4370 * @param $user User: the User object describing the current user
4371 * @param $options ParserOptions: parsing options
4372 * @param $clearState Boolean: whether to clear the parser state first
4373 * @return String: the altered wiki markup
4374 */
4375 public function preSaveTransform( $text, Title $title, User $user, ParserOptions $options, $clearState = true ) {
4381 );
4385 }
4391 }
4393 /**
4394 * Pre-save transform helper function
4395 * @private
4396 *
4397 * @param $text string
4398 * @param $user User
4399 *
4400 * @return string
4401 */
4405 # Note: This is the timestamp saved as hardcoded wikitext to
4406 # the database, we use $wgContLang here in order to give
4407 # everyone the same signature and use the default one rather
4408 # than the one selected in each user's preferences.
4409 # (see also bug 12815)
4415 }
4423 # Allow translation of timezones through wiki. date() can return
4424 # whatever crap the system uses, localised or not, so we cannot
4425 # ship premade translations.
4430 }
4436 # Variable replacement
4437 # Because mOutputType is OT_WIKI, this will only process {{subst:xxx}} type tags
4440 # This works almost by chance, as the replaceVariables are done before the getUserSig(),
4441 # which may corrupt this parser instance via its wfMessage()->text() call-
4443 # Signatures
4449 ) );
4451 # Context links: [[|name]] and [[name (context)|]]
4457 $p3 = "/\[\[(:?$nc+:|:|)($tc+?)( ?\\($tc+\\)|)((?:, |，)$tc+|)\\|]]/"; # [[ns:page (context), context|]]
4460 # try $p1 first, to turn "[[A, B (C)|]]" into "[[A, B (C)|A, B]]"
4472 # if there's no context, don't bother duplicating the title
4474 }
4476 # Trim trailing whitespace
4480 }
4482 /**
4483 * Fetch the user's signature text, if any, and normalize to
4484 * validated, ready-to-insert wikitext.
4485 * If you have pre-fetched the nickname or the fancySig option, you can
4486 * specify them here to save a database query.
4487 * Do not reuse this parser instance after calling getUserSig(),
4488 * as it may have changed if it's the $wgParser.
4489 *
4490 * @param $user User
4491 * @param $nickname String|bool nickname to use or false to use user's default nickname
4492 * @param $fancySig Boolean|null whether the nicknname is the complete signature
4493 * or null to use default value
4494 * @return string
4495 */
4501 # If not given, retrieve from the user object.
4507 }
4515 # Sig. might contain markup; validate this
4517 # Validated; clean up (if needed) and return it
4520 # Failed to validate; fall back to the default
4523 }
4524 }
4526 # Make sure nickname doesnt get a sig in a sig
4529 # If we're still here, make it a link to the user page
4534 return wfMessage( $msgName, $userText, $nickText )->inContentLanguage()->title( $this->getTitle() )->text();
4535 }
4537 /**
4538 * Check that the user's signature contains no bad XML
4539 *
4540 * @param $text String
4541 * @return mixed An expanded string, or false if invalid.
4542 */
4545 }
4547 /**
4548 * Clean up signature text
4549 *
4550 * 1) Strip ~~~, ~~~~ and ~~~~~ out of signatures @see cleanSigInSig
4551 * 2) Substitute all transclusions
4552 *
4553 * @param $text String
4554 * @param $parsing bool Whether we're cleaning (preferences save) or parsing
4555 * @return String: signature text
4556 */
4561 }
4563 # Option to disable this feature
4566 }
4568 # @todo FIXME: Regex doesn't respect extension tags or nowiki
4569 # => Move this logic to braceSubstitution()
4571 $substRegex = '/\{\{(?!(?:' . $substWord->getBaseRegex() . '))/x' . $substWord->getRegexCase();
4582 }
4585 }
4587 /**
4588 * Strip ~~~, ~~~~ and ~~~~~ out of signatures
4589 *
4590 * @param $text String
4591 * @return String: signature text with /~{3,5}/ removed
4592 */
4596 }
4598 /**
4599 * Set up some variables which are usually set up in parse()
4600 * so that an external function can call some class members with confidence
4601 *
4602 * @param $title Title|null
4603 * @param $options ParserOptions
4604 * @param $outputType
4605 * @param $clearState bool
4606 */
4607 public function startExternalParse( Title $title = null, ParserOptions $options, $outputType, $clearState = true ) {
4609 }
4611 /**
4612 * @param $title Title|null
4613 * @param $options ParserOptions
4614 * @param $outputType
4615 * @param $clearState bool
4616 */
4617 private function startParse( Title $title = null, ParserOptions $options, $outputType, $clearState = true ) {
4623 }
4624 }
4626 /**
4627 * Wrapper for preprocess()
4628 *
4629 * @param $text String: the text to preprocess
4630 * @param $options ParserOptions: options
4631 * @param $title Title object or null to use $wgTitle
4632 * @return String
4633 */
4637 # Guard against infinite recursion
4640 }
4647 }
4649 # It's not uncommon having a null $wgTitle in scripts. See r80898
4650 # Create a ghost title in such case
4652 }
4658 }
4660 /**
4661 * Create an HTML-style tag, e.g. "<yourtag>special text</yourtag>"
4662 * The callback should have the following form:
4663 * function myParserHook( $text, $params, $parser, $frame ) { ... }
4664 *
4665 * Transform and return $text. Use $parser for any required context, e.g. use
4666 * $parser->getTitle() and $parser->getOptions() not $wgTitle or $wgOut->mParserOptions
4667 *
4668 * Hooks may return extended information by returning an array, of which the
4669 * first numbered element (index 0) must be the return string, and all other
4670 * entries are extracted into local variables within an internal function
4671 * in the Parser class.
4672 *
4673 * This interface (introduced r61913) appears to be undocumented, but
4674 * 'markerName' is used by some core tag hooks to override which strip
4675 * array their results are placed in. **Use great caution if attempting
4676 * this interface, as it is not documented and injudicious use could smash
4677 * private variables.**
4678 *
4679 * @param $tag Mixed: the tag to use, e.g. 'hook' for "<hook>"
4680 * @param $callback Mixed: the callback function (and object) to use for the tag
4681 * @return Mixed|null The old value of the mTagHooks array associated with the hook
4682 */
4687 }
4692 }
4695 }
4697 /**
4698 * As setHook(), but letting the contents be parsed.
4699 *
4700 * Transparent tag hooks are like regular XML-style tag hooks, except they
4701 * operate late in the transformation sequence, on HTML instead of wikitext.
4702 *
4703 * This is probably obsoleted by things dealing with parser frames?
4704 * The only extension currently using it is geoserver.
4705 *
4706 * @since 1.10
4707 * @todo better document or deprecate this
4708 *
4709 * @param $tag Mixed: the tag to use, e.g. 'hook' for "<hook>"
4710 * @param $callback Mixed: the callback function (and object) to use for the tag
4711 * @return Mixed|null The old value of the mTagHooks array associated with the hook
4712 */
4717 }
4718 $oldVal = isset( $this->mTransparentTagHooks[$tag] ) ? $this->mTransparentTagHooks[$tag] : null;
4722 }
4724 /**
4725 * Remove all tag hooks
4726 */
4731 }
4733 /**
4734 * Create a function, e.g. {{sum:1|2|3}}
4735 * The callback function should have the form:
4736 * function myParserFunction( &$parser, $arg1, $arg2, $arg3 ) { ... }
4737 *
4738 * Or with SFH_OBJECT_ARGS:
4739 * function myParserFunction( $parser, $frame, $args ) { ... }
4740 *
4741 * The callback may either return the text result of the function, or an array with the text
4742 * in element 0, and a number of flags in the other elements. The names of the flags are
4743 * specified in the keys. Valid flags are:
4744 * found The text returned is valid, stop processing the template. This
4745 * is on by default.
4746 * nowiki Wiki markup in the return value should be escaped
4747 * isHTML The returned text is HTML, armour it against wikitext transformation
4748 *
4749 * @param $id String: The magic word ID
4750 * @param $callback Mixed: the callback function (and object) to use
4751 * @param $flags Integer: a combination of the following flags:
4752 * SFH_NO_HASH No leading hash, i.e. {{plural:...}} instead of {{#if:...}}
4753 *
4754 * SFH_OBJECT_ARGS Pass the template arguments as PPNode objects instead of text. This
4755 * allows for conditional expansion of the parse tree, allowing you to eliminate dead
4756 * branches and thus speed up parsing. It is also possible to analyse the parse tree of
4757 * the arguments, and to control the way they are expanded.
4758 *
4759 * The $frame parameter is a PPFrame. This can be used to produce expanded text from the
4760 * arguments, for instance:
4761 * $text = isset( $args[0] ) ? $frame->expand( $args[0] ) : '';
4762 *
4763 * For technical reasons, $args[0] is pre-expanded and will be a string. This may change in
4764 * future versions. Please call $frame->expand() on it anyway so that your code keeps
4765 * working if/when this is changed.
4766 *
4767 * If you want whitespace to be trimmed from $args, you need to do it yourself, post-
4768 * expansion.
4769 *
4770 * Please read the documentation in includes/parser/Preprocessor.php for more information
4771 * about the methods available in PPFrame and PPNode.
4772 *
4773 * @return string|callback The old callback function for this name, if any
4774 */
4781 # Add to function cache
4790 # Case
4793 }
4794 # Add leading hash
4797 }
4798 # Remove trailing colon
4801 }
4803 }
4805 }
4807 /**
4808 * Get all registered function hook identifiers
4809 *
4810 * @return Array
4811 */
4814 }
4816 /**
4817 * Create a tag function, e.g. "<test>some stuff</test>".
4818 * Unlike tag hooks, tag functions are parsed at preprocessor level.
4819 * Unlike parser functions, their content is not preprocessed.
4820 * @return null
4821 */
4824 if ( preg_match( '/[<>\r\n]/', $tag, $m ) ) throw new MWException( "Invalid character {$m[0]} in setFunctionTagHook('$tag', ...) call" );
4831 }
4834 }
4836 /**
4837 * @todo FIXME: Update documentation. makeLinkObj() is deprecated.
4838 * Replace "<!--LINK-->" link placeholders with actual links, in the buffer
4839 * Placeholders created in Skin::makeLinkObj()
4840 *
4841 * @param $text string
4842 * @param $options int
4843 *
4844 * @return array of link CSS classes, indexed by PDBK.
4845 */
4848 }
4850 /**
4851 * Replace "<!--LINK-->" link placeholders with plain text of links
4852 * (not HTML-formatted).
4853 *
4854 * @param $text String
4855 * @return String
4856 */
4859 }
4861 /**
4862 * Renders an image gallery from a text with one line per image.
4863 * text labels may be given by using |-style alternative text. E.g.
4864 * Image:one.jpg|The number "1"
4865 * Image:tree.jpg|A tree
4866 * given as text will return the HTML of a gallery with two images,
4867 * labeled 'The number "1"' and
4868 * 'A tree'.
4869 *
4870 * @param string $text
4871 * @param array $params
4872 * @return string HTML
4873 */
4887 }
4893 }
4896 }
4899 }
4902 }
4908 # match lines like these:
4909 # Image:someimage.jpg|This is some image
4912 # Skip empty lines
4915 }
4919 }
4922 # Bogus title. Ignore these so we don't bomb out later.
4924 }
4930 // look for an |alt= definition while trying not to break existing
4931 // captions with multiple pipes (|) in it, until a more sensible grammar
4932 // is defined for images in galleries
4942 }
4947 //check to see if link matches an absolute url, if not then it must be a wiki link.
4951 }
4952 }
4954 // concatenate all other pipes
4956 }
4957 }
4958 // remove the first pipe
4960 }
4963 }
4965 }
4967 /**
4968 * @param $handler
4969 * @return array
4970 */
4976 }
4978 # Initialise static lists
4985 );
4993 }
4994 }
4995 }
4997 # Add handler params
5003 }
5004 }
5007 }
5008 return array( $this->mImageParams[$handlerClass], $this->mImageParamsMagicArray[$handlerClass] );
5009 }
5011 /**
5012 * Parse image options text and use it to make an image
5013 *
5014 * @param $title Title
5015 * @param $options String
5016 * @param $holders LinkHolderArray|bool
5017 * @return string HTML
5018 */
5020 # Check if the options text is of the form "options|alt text"
5021 # Options are:
5022 # * thumbnail make a thumbnail with enlarge-icon and caption, alignment depends on lang
5023 # * left no resizing, just left align. label is used for alt= only
5024 # * right same, but right aligned
5025 # * none same, but not aligned
5026 # * ___px scale to ___ pixels width, no aligning. e.g. use in taxobox
5027 # * center center the image
5028 # * frame Keep original image size, no magnify-button.
5029 # * framed Same as "frame"
5030 # * frameless like 'thumb' but without a frame. Keeps user preferences for width
5031 # * upright reduce width for upright images, rounded to full __0 px
5032 # * border draw a 1px border around the image
5033 # * alt Text for HTML alt attribute (defaults to empty)
5034 # * class Set a class for img node
5035 # * link Set the target of the image link. Can be external, interwiki, or local
5036 # vertical-align values (no % or length right now):
5037 # * baseline
5038 # * sub
5039 # * super
5040 # * top
5041 # * text-top
5042 # * middle
5043 # * bottom
5044 # * text-bottom
5048 # Give extensions a chance to select the file revision for us
5053 # Fetch and register the file (file title may be different via hooks)
5056 # Get parameter map
5063 }
5065 # Process the input parameters
5076 # Special case; width and height come in one variable together
5084 }
5085 }
5091 }
5092 }
5093 # else no validation -- bug 13436
5096 # Validate handler parameter
5099 # Validate internal parameters
5104 # @todo FIXME: Possibly check validity here for
5105 # manualthumb? downstream behavior seems odd with
5106 # missing manual thumbs.
5123 }
5125 }
5133 }
5134 }
5137 # Most other things appear to be empty or numeric...
5139 }
5140 }
5144 }
5145 }
5146 }
5149 }
5150 }
5152 # Process alignment parameters
5155 }
5158 }
5162 # Will the image be presented in a frame, with the caption below?
5168 # In the old days, [[Image:Foo|text...]] would set alt text. Later it
5169 # came to also set the caption, ordinary text after the image -- which
5170 # makes no sense, because that just repeats the text multiple times in
5171 # screen readers. It *also* came to set the title attribute.
5172 #
5173 # Now that we have an alt attribute, we should not set the alt text to
5174 # equal the caption: that's worse than useless, it just repeats the
5175 # text. This is the framed/thumbnail case. If there's no caption, we
5176 # use the unnamed parameter for alt text as well, just for the time be-
5177 # ing, if the unnamed param is set and the alt param is not.
5178 #
5179 # For the future, we need to figure out if we want to tweak this more,
5180 # e.g., introducing a title= parameter for the title; ignoring the un-
5181 # named parameter entirely for images without a caption; adding an ex-
5182 # plicit caption= parameter and preserving the old magic unnamed para-
5183 # meter for BC; ...
5186 # No caption or alt text, add the filename as the alt text so
5187 # that screen readers at least get some description of the image
5189 }
5190 # Do not set $params['frame']['title'] because tooltips don't make sense
5191 # for framed images
5194 # No alt text, use the "caption" for the alt text
5198 # No caption, fall back to using the filename for the
5199 # alt text
5201 }
5202 }
5203 # Use the "caption" for the tooltip text
5205 }
5209 # Linker does the rest
5214 # Give the handler a chance to modify the parser object
5217 }
5220 }
5222 /**
5223 * @param $caption
5224 * @param $holders LinkHolderArray
5225 * @return mixed|String
5226 */
5228 # Strip bad stuff out of the title (tooltip). We can't just use
5229 # replaceLinkHoldersText() here, because if this function is called
5230 # from replaceInternalLinks2(), mLinkHolders won't be up-to-date.
5235 }
5237 # make sure there are no placeholders in thumbnail attributes
5238 # that are later expanded to html- so expand them now and
5239 # remove the tags
5244 }
5246 /**
5247 * Set a flag in the output object indicating that the content is dynamic and
5248 * shouldn't be cached.
5249 */
5255 }
5258 }
5260 /**
5261 * Callback from the Sanitizer for expanding items found in HTML attribute
5262 * values, so they can be safely tested and escaped.
5263 *
5264 * @param $text String
5265 * @param $frame PPFrame
5266 * @return String
5267 */
5272 }
5274 /**
5275 * Accessor
5276 *
5277 * @return array
5278 */
5280 return array_merge( array_keys( $this->mTransparentTagHooks ), array_keys( $this->mTagHooks ), array_keys( $this->mFunctionTagHooks ) );
5281 }
5283 /**
5284 * Replace transparent tags in $text with the values given by the callbacks.
5285 *
5286 * Transparent tag hooks are like regular XML-style tag hooks, except they
5287 * operate late in the transformation sequence, on HTML instead of wikitext.
5288 *
5289 * @param $text string
5290 *
5291 * @return string
5292 */
5303 $output = call_user_func_array( $this->mTransparentTagHooks[$tagName], array( $content, $params, $this ) );
5306 }
5308 }
5310 }
5312 /**
5313 * Break wikitext input into sections, and either pull or replace
5314 * some particular section's text.
5315 *
5316 * External callers should use the getSection and replaceSection methods.
5317 *
5318 * @param $text String: Page wikitext
5319 * @param $section String: a section identifier string of the form:
5320 * "<flag1> - <flag2> - ... - <section number>"
5321 *
5322 * Currently the only recognised flag is "T", which means the target section number
5323 * was derived during a template inclusion parse, in other words this is a template
5324 * section edit link. If no flags are given, it was an ordinary section edit link.
5325 * This flag is required to avoid a section numbering mismatch when a section is
5326 * enclosed by "<includeonly>" (bug 6563).
5327 *
5328 * The section number 0 pulls the text before the first heading; other numbers will
5329 * pull the given section along with its lower-level subsections. If the section is
5330 * not found, $mode=get will return $newtext, and $mode=replace will return $text.
5331 *
5332 * Section 0 is always considered to exist, even if it only contains the empty
5333 * string. If $text is the empty string and section 0 is replaced, $newText is
5334 * returned.
5335 *
5336 * @param $mode String: one of "get" or "replace"
5337 * @param $newText String: replacement text for section data.
5338 * @return String: for "get", the extracted section text.
5339 * for "replace", the whole page with the section replaced.
5340 */
5347 # Process section extraction flags
5354 }
5355 }
5357 # Check for empty input
5359 # Only sections 0 and T-0 exist in an empty document
5365 }
5371 }
5372 }
5373 }
5375 # Preprocess the text
5378 # <h> nodes indicate section breaks
5379 # They can only occur at the top level, so we can find them by iterating the root's children
5382 # Find the target section
5384 # Section zero doesn't nest, level=big
5393 }
5394 }
5397 }
5399 }
5400 }
5403 # Not found
5408 }
5409 }
5411 # Find the end of the section, including nested sections
5418 }
5419 }
5422 }
5426 # Write out the remainder (in replace mode only)
5428 # Output the replacement text
5429 # Add two newlines on -- trailing whitespace in $newText is conventionally
5430 # stripped by the editor, so we need both newlines to restore the paragraph gap
5431 # Only add trailing whitespace if there is newText
5434 }
5439 }
5440 }
5443 # Re-insert stripped tags
5445 }
5448 }
5450 /**
5451 * This function returns the text of a section, specified by a number ($section).
5452 * A section is text under a heading like == Heading == or \<h1\>Heading\</h1\>, or
5453 * the first section before any such heading (section 0).
5454 *
5455 * If a section contains subsections, these are also returned.
5456 *
5457 * @param $text String: text to look in
5458 * @param $section String: section identifier
5459 * @param $deftext String: default to return if section is not found
5460 * @return string text of the requested section
5461 */
5464 }
5466 /**
5467 * This function returns $oldtext after the content of the section
5468 * specified by $section has been replaced with $text. If the target
5469 * section does not exist, $oldtext is returned unchanged.
5470 *
5471 * @param $oldtext String: former text of the article
5472 * @param $section int section identifier
5473 * @param $text String: replacing text
5474 * @return String: modified text
5475 */
5478 }
5480 /**
5481 * Get the ID of the revision we are parsing
5482 *
5483 * @return Mixed: integer or null
5484 */
5487 }
5489 /**
5490 * Get the revision object for $this->mRevisionId
5491 *
5492 * @return Revision|null either a Revision object or null
5493 */
5497 }
5500 }
5504 }
5506 /**
5507 * Get the timestamp associated with the current revision, adjusted for
5508 * the default server-local timestamp
5509 */
5519 # The cryptic '' timezone parameter tells to use the site-default
5520 # timezone offset instead of the user settings.
5521 #
5522 # Since this value will be saved into the parser cache, served
5523 # to other users, and potentially even used inside links and such,
5524 # it needs to be consistent for all visitors.
5528 }
5530 }
5532 /**
5533 * Get the name of the user that edited the last revision
5534 *
5535 * @return String: user name
5536 */
5541 # if this template is subst: the revision id will be blank,
5542 # so just use the current user's name
5547 }
5548 }
5550 }
5552 /**
5553 * Mutator for $mDefaultSort
5554 *
5555 * @param $sort string New value
5556 */
5560 }
5562 /**
5563 * Accessor for $mDefaultSort
5564 * Will use the empty string if none is set.
5565 *
5566 * This value is treated as a prefix, so the
5567 * empty string is equivalent to sorting by
5568 * page name.
5569 *
5570 * @return string
5571 */
5577 }
5578 }
5580 /**
5581 * Accessor for $mDefaultSort
5582 * Unlike getDefaultSort(), will return false if none is set
5583 *
5584 * @return string or false
5585 */
5588 }
5590 /**
5591 * Try to guess the section anchor name based on a wikitext fragment
5592 * presumably extracted from a heading, for example "Header" from
5593 * "== Header ==".
5594 *
5595 * @param $text string
5596 *
5597 * @return string
5598 */
5600 # Strip out wikitext links(they break the anchor)
5604 }
5606 /**
5607 * Same as guessSectionNameFromWikiText(), but produces legacy anchors
5608 * instead. For use in redirects, since IE6 interprets Redirect: headers
5609 * as something other than UTF-8 (apparently?), resulting in breakage.
5610 *
5611 * @param $text String: The section name
5612 * @return string An anchor
5613 */
5615 # Strip out wikitext links(they break the anchor)
5619 }
5621 /**
5622 * Strips a text string of wikitext for use in a section anchor
5623 *
5624 * Accepts a text string and then removes all wikitext from the
5625 * string and leaves only the resultant text (i.e. the result of
5626 * [[User:WikiSysop|Sysop]] would be "Sysop" and the result of
5627 * [[User:WikiSysop]] would be "User:WikiSysop") - this is intended
5628 * to create valid section anchors by mimicing the output of the
5629 * parser when headings are parsed.
5630 *
5631 * @param $text String: text string to be stripped of wikitext
5632 * for use in a Section anchor
5633 * @return string Filtered text string
5634 */
5636 # Strip internal link markup
5640 # Strip external link markup
5641 # @todo FIXME: Not tolerant to blank link text
5642 # I.E. [http://www.mediawiki.org] will render as [1] or something depending
5643 # on how many empty links there are on the page - need to figure that out.
5644 $text = preg_replace( '/\[(?i:' . $this->mUrlProtocols . ')([^ ]+?) ([^[]+)\]/', '$2', $text );
5646 # Parse wikitext quotes (italics & bold)
5649 # Strip HTML tags
5652 }
5654 /**
5655 * strip/replaceVariables/unstrip for preprocessor regression testing
5656 *
5657 * @param $text string
5658 * @param $title Title
5659 * @param $options ParserOptions
5660 * @param $outputType int
5661 *
5662 * @return string
5663 */
5664 function testSrvus( $text, Title $title, ParserOptions $options, $outputType = self::OT_HTML ) {
5671 }
5673 /**
5674 * @param $text string
5675 * @param $title Title
5676 * @param $options ParserOptions
5677 * @return string
5678 */
5681 }
5683 /**
5684 * @param $text
5685 * @param $title Title
5686 * @param $options ParserOptions
5687 * @return string
5688 */
5691 }
5693 /**
5694 * Call a callback function on all regions of the given text that are not
5695 * inside strip markers, and replace those regions with the return value
5696 * of the callback. For example, with input:
5697 *
5698 * aaa<MARKER>bbb
5699 *
5700 * This will call the callback function twice, with 'aaa' and 'bbb'. Those
5701 * two strings will be replaced with the value returned by the callback in
5702 * each case.
5703 *
5704 * @param $s string
5705 * @param $callback
5706 *
5707 * @return string
5708 */
5727 }
5728 }
5729 }
5731 }
5733 /**
5734 * Remove any strip markers found in the given text.
5735 *
5736 * @param $text Input string
5737 * @return string
5738 */
5741 }
5743 /**
5744 * Save the parser state required to convert the given half-parsed text to
5745 * HTML. "Half-parsed" in this context means the output of
5746 * recursiveTagParse() or internalParse(). This output has strip markers
5747 * from replaceVariables (extensionSubstitution() etc.), and link
5748 * placeholders from replaceLinkHolders().
5749 *
5750 * Returns an array which can be serialized and stored persistently. This
5751 * array can later be loaded into another parser instance with
5752 * unserializeHalfParsedText(). The text can then be safely incorporated into
5753 * the return value of a parser hook.
5754 *
5755 * @param $text string
5756 *
5757 * @return array
5758 */
5766 );
5769 }
5771 /**
5772 * Load the parser state given in the $data array, which is assumed to
5773 * have been generated by serializeHalfParsedText(). The text contents is
5774 * extracted from the array, and its markers are transformed into markers
5775 * appropriate for the current Parser instance. This transformed text is
5776 * returned, and can be safely included in the return value of a parser
5777 * hook.
5778 *
5779 * If the $data array has been stored persistently, the caller should first
5780 * check whether it is still valid, by calling isValidHalfParsedText().
5781 *
5782 * @param $data array Serialized data
5783 * @return String
5784 */
5788 }
5790 # First, extract the strip state.
5794 # Now renumber links
5797 # Should be good to go.
5799 }
5801 /**
5802 * Returns true if the given array, presumed to be generated by
5803 * serializeHalfParsedText(), is compatible with the current version of the
5804 * parser.
5805 *
5806 * @param $data Array
5807 *
5808 * @return bool
5809 */
5812 }
5814 /**
5815 * Parsed a width param of imagelink like 300px or 200x300px
5816 *
5817 * @param $value String
5818 *
5819 * @return array
5820 * @since 1.20
5821 */
5826 }
5828 # (bug 13500) In both cases (width/height and width only),
5829 # permit trailing "px" for backward compatibility.
5838 }
5840 }
5841 }