| blob | 28cde7f09f4311617ca5b95b7917a26708bd96b9 |
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 * $wgLocaltimezone
58 * $wgNamespacesWithSubpages
59 *
60 * @par Settings only within ParserOptions:
61 * $wgAllowExternalImages
62 * $wgAllowSpecialInclusion
63 * $wgInterwikiMagic
64 * $wgMaxArticleSize
65 *
66 * @ingroup Parser
67 */
69 /**
70 * Update this version number when the ParserOutput format
71 * changes in an incompatible way, so the parser cache
72 * can automatically discard old data.
73 */
76 /**
77 * Update this version number when the output of serialiseHalfParsedText()
78 * changes in an incompatible way
79 */
82 # Flags for Parser::setFunctionHook
83 # Also available as global constants from Defines.php
87 # Constants needed for external link processing
88 # Everything except bracket, space, or control characters
89 # \p{Zs} is unicode 'separator, space' category. It covers the space 0x20
90 # as well as U+3000 is IDEOGRAPHIC SPACE for bug 19052
95 # State constants for the definition list colon extraction
105 # Flags for preprocessToDom
108 # Allowed values for $this->mOutputType
109 # Parameter to startExternalParse().
116 # Marker Suffix needs to be accessible staticly.
119 # Persistent:
133 # Initialised by initialiseVariables()
135 /**
136 * @var MagicWordArray
137 */
140 /**
141 * @var MagicWordArray
142 */
144 var $mConf, $mPreprocessor, $mExtLinkBracketedRegex, $mUrlProtocols; # Initialised in constructor
146 # Cleared with clearState():
147 /**
148 * @var ParserOutput
149 */
153 /**
154 * @var StripState
155 */
159 /**
160 * @var LinkHolderArray
161 */
172 /**
173 * @var User
174 */
177 # Temporary
178 # These are variables reset at least once per parse regardless of $clearState
180 /**
181 * @var ParserOptions
182 */
185 /**
186 * @var Title
187 */
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 under HipHop
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 * Allow extensions to clean up when the parser is cloned
251 */
254 }
256 /**
257 * Do various kinds of initialisation on the first call of the parser
258 */
262 }
273 }
275 /**
276 * Clear Parser state
277 *
278 * @private
279 */
284 }
301 /**
302 * Prefix for temporary replacement strings for the multipass parser.
303 * \x07 should never appear in input as it's disallowed in XML.
304 * Using it at the front also gives us a little extra robustness
305 * since it shouldn't match when butted up against identifier-like
306 * string constructs.
307 *
308 * Must not consist of all title characters, or else it will change
309 * the behavior of <nowiki> in a link.
310 */
314 # Clear these on every parse, bug 4549
322 );
331 # Fix cloning
334 }
338 }
340 /**
341 * Convert wikitext to HTML
342 * Do not call this function recursively.
343 *
344 * @param string $text text we want to parse
345 * @param $title Title object
346 * @param $options ParserOptions
347 * @param $linestart boolean
348 * @param $clearState boolean
349 * @param int $revid number to pass in {{REVISIONID}}
350 * @return ParserOutput a ParserOutput
351 */
352 public function parse( $text, Title $title, ParserOptions $options, $linestart = true, $clearState = true, $revid = null ) {
353 /**
354 * First pass--just handle <nowiki> sections, pass the rest off
355 * to internalParse() which does all the real work.
356 */
367 # Remove the strip marker tag prefix from the input, if present.
370 }
381 }
384 # No more strip!
391 # Clean up special characters, only run once, next-to-last before doBlockLevels
393 # french spaces, last one Guillemet-left
394 # only if there is something before the space
396 # french spaces, Guillemet-right
399 );
406 /**
407 * The input doesn't get language converted if
408 * a) It's disabled
409 * b) Content isn't converted
410 * c) It's a conversion table
411 * d) it is an interface message (which is in the user language)
412 */
415 {
417 # The position of the convert() call should not be changed. it
418 # assumes that the links are all replaced and the only thing left
419 # is the <nowiki> mark.
421 }
422 }
424 /**
425 * A converted title will be provided in the output object if title and
426 * content conversion are enabled, the article text does not contain
427 * a conversion-suppressing double-underscore tag, and no
428 * {{DISPLAYTITLE:...}} is present. DISPLAYTITLE takes precedence over
429 * automatic link conversion.
430 */
435 {
442 }
443 }
457 # attempt to sanitize at least some nesting problems
458 # (bug #2702 and quite a few others)
460 # ''Something [http://www.cool.com cool''] -->
461 # <i>Something</i><a href="http://www.cool.com"..><i>cool></i></a>
464 # fix up an anchor inside another anchor, only
465 # at least for a single single nested link (bug 3695)
468 # fix div inside inline elements- doBlockLevels won't wrap a line which
469 # contains a div, so fix it up here; replace
470 # div with escaped text
473 # remove empty italic or bold tag pairs, some
474 # introduced by rules above
476 );
482 }
488 );
489 }
493 # Information on include size limits, for the benefit of users who try to skirt them
496 $PFreport = "Expensive parser function count: {$this->mExpensiveFunctionCount}/{$this->mOptions->getExpensiveParserFunctionLimit()}\n";
499 "Preprocessor visited node count: {$this->mPPNodeCount}/{$this->mOptions->getMaxPPNodeCount()}\n" .
504 "Highest expansion depth: {$this->mHighestExpansionDepth}/{$this->mOptions->getMaxPPExpandDepth()}\n" .
508 // Sanitize for comment. Note '‐' in the replacement is U+2010,
509 // which looks much like the problematic '-'.
517 }
518 }
530 }
532 /**
533 * Recursive parser entry point that can be called from an extension tag
534 * hook.
535 *
536 * If $frame is not provided, then template variables (e.g., {{{1}}}) within $text are not expanded
537 *
538 * @param string $text text extension wants to have parsed
539 * @param $frame PPFrame: The frame to use for expanding any template variables
540 *
541 * @return string
542 */
550 }
552 /**
553 * Expand templates and variables in the text, producing valid, static wikitext.
554 * Also removes comments.
555 * @return mixed|string
556 */
562 }
569 }
571 /**
572 * Recursive parser entry point that can be called from an extension tag
573 * hook.
574 *
575 * @param string $text text to be expanded
576 * @param $frame PPFrame: The frame to use for expanding any template variables
577 * @return String
578 * @since 1.19
579 */
586 }
588 /**
589 * Process the wikitext for the "?preload=" feature. (bug 5210)
590 *
591 * "<noinclude>", "<includeonly>" etc. are parsed as for template
592 * transclusion, comments, templates, arguments, tags hooks and parser
593 * functions are untouched.
594 *
595 * @param $text String
596 * @param $title Title
597 * @param $options ParserOptions
598 * @return String
599 */
601 # Parser (re)initialisation
609 }
611 /**
612 * Get a random string
613 *
614 * @return string
615 */
618 }
620 /**
621 * Set the current user.
622 * Should only be used when doing pre-save transform.
623 *
624 * @param $user Mixed: User object or null (to reset)
625 */
628 }
630 /**
631 * Accessor for mUniqPrefix.
632 *
633 * @return String
634 */
637 # @todo FIXME: This is probably *horribly wrong*
638 # LanguageConverter seems to want $wgParser's uniqPrefix, however
639 # if this is called for a parser cache hit, the parser may not
640 # have ever been initialized in the first place.
641 # Not really sure what the heck is supposed to be going on here.
643 # throw new MWException( "Accessing uninitialized mUniqPrefix" );
644 }
646 }
648 /**
649 * Set the context title
650 *
651 * @param $t Title
652 */
656 }
659 # Strip the fragment to avoid various odd effects
664 }
665 }
667 /**
668 * Accessor for the Title object
669 *
670 * @return Title object
671 */
674 }
676 /**
677 * Accessor/mutator for the Title object
678 *
679 * @param $x Title object or null to just get the current one
680 * @return Title object
681 */
684 }
686 /**
687 * Set the output type
688 *
689 * @param $ot Integer: new value
690 */
693 # Shortcut alias
699 );
700 }
702 /**
703 * Accessor/mutator for the output type
704 *
705 * @param int|null $x New value or null to just get the current one
706 * @return Integer
707 */
710 }
712 /**
713 * Get the ParserOutput object
714 *
715 * @return ParserOutput object
716 */
719 }
721 /**
722 * Get the ParserOptions object
723 *
724 * @return ParserOptions object
725 */
728 }
730 /**
731 * Accessor/mutator for the ParserOptions object
732 *
733 * @param $x ParserOptions New value or null to just get the current one
734 * @return ParserOptions Current ParserOptions object
735 */
738 }
740 /**
741 * @return int
742 */
745 }
747 /**
748 * @param $id int
749 */
752 }
754 /**
755 * Get a language object for use in parser functions such as {{FORMATNUM:}}
756 * @return Language
757 */
760 }
762 /**
763 * Get the target language for the content being parsed. This is usually the
764 * language that the content is in.
765 *
766 * @since 1.19
767 *
768 * @throws MWException
769 * @return Language|null
770 */
780 }
783 }
785 /**
786 * Get the language object for language conversion
787 */
790 }
792 /**
793 * Get a User object either from $this->mUser, if set, or from the
794 * ParserOptions object otherwise
795 *
796 * @return User object
797 */
801 }
803 }
805 /**
806 * Get a preprocessor object
807 *
808 * @return Preprocessor instance
809 */
814 }
816 }
818 /**
819 * Replaces all occurrences of HTML-style comments and the given tags
820 * in the text with a random marker and returns the next text. The output
821 * parameter $matches will be an associative array filled with data in
822 * the form:
823 *
824 * @code
825 * 'UNIQ-xxxxx' => array(
826 * 'element',
827 * 'tag content',
828 * array( 'param' => 'x' ),
829 * '<element param="x">tag content</element>' ) )
830 * @endcode
831 *
832 * @param array $elements list of element names. Comments are always extracted.
833 * @param string $text Source text string.
834 * @param array $matches Out parameter, Array: extracted tags
835 * @param $uniq_prefix string
836 * @return String: stripped text
837 */
851 }
853 # comment
859 # tag
864 }
870 # Empty element tag, <tag />
879 }
883 # No end tag -- let it run out to the end of the text.
889 }
890 }
896 }
898 }
900 /**
901 * Get a list of strippable XML-like elements
902 *
903 * @return array
904 */
907 }
909 /**
910 * Add an item to the strip state
911 * Returns the unique tag which must be inserted into the stripped text
912 * The tag will be replaced with the original text in unstrip()
913 *
914 * @param $text string
915 *
916 * @return string
917 */
923 }
925 /**
926 * parse the wiki syntax used to render tables
927 *
928 * @private
929 * @return string
930 */
949 }
955 # First check if we are starting a new table
968 # Don't do any of the following
972 # We are ending a table
978 }
982 }
986 }
990 # Now we have a table row
993 # Whats after the tag is now only attributes
1006 }
1010 }
1016 } elseif ( $first_character === '|' || $first_character === '!' || substr( $line, 0, 2 ) === '|+' ) {
1017 # This might be cell elements, td, th or captions
1021 }
1027 }
1029 # Split up multiple cells on the same line.
1030 # FIXME : This can result in improper nesting of tags processed
1031 # by earlier parser steps, but should avoid splitting up eg
1032 # attribute values containing literal "||".
1037 # Loop through each table cell
1044 }
1049 }
1055 }
1065 }
1069 # A cell could contain both parameters and data
1072 # Bug 553: Note that a '|' inside an invalid link should not
1073 # be mistaken as delimiting cell parameters
1082 }
1086 }
1087 }
1089 }
1091 # Closing open td, tr && table
1095 }
1098 }
1101 }
1104 }
1106 # Remove trailing line-ending (b/c)
1109 }
1111 # special case: don't return empty table
1114 }
1119 }
1121 /**
1122 * Helper function for parse() that transforms wiki markup into
1123 * HTML. Only called for $mOutputType == self::OT_HTML.
1124 *
1125 * @private
1126 *
1127 * @param $text string
1128 * @param $isMain bool
1129 * @param $frame bool
1130 *
1131 * @return string
1132 */
1138 # Hook to suspend the parser in this state
1139 if ( !wfRunHooks( 'ParserBeforeInternalParse', array( &$this, &$text, &$this->mStripState ) ) ) {
1142 }
1144 # if $frame is provided, then use $frame for replacing any variables
1146 # use frame depth to infer how include/noinclude tags should be handled
1147 # depth=0 means this is the top-level document; otherwise it's an included document
1152 }
1156 # if $frame is not provided, then use old-style replaceVariables
1158 }
1161 $text = Sanitizer::removeHTMLtags( $text, array( &$this, 'attributeStripCallback' ), false, array_keys( $this->mTransparentTagHooks ) );
1164 # Tables need to come after variable replacement for things to work
1165 # properly; putting them before other transformations should keep
1166 # exciting things like link expansions from showing up in surprising
1167 # places.
1179 # replaceInternalLinks may sometimes leave behind
1180 # absolute URLs, which have to be masked to hide them from replaceExternalLinks
1188 }
1190 /**
1191 * Replace special strings like "ISBN xxx" and "RFC xxx" with
1192 * magic external links.
1193 *
1194 * DML
1195 * @private
1196 *
1197 * @param $text string
1198 *
1199 * @return string
1200 */
1206 '!(?: # Start cases
1210 (?:RFC|PMID)\s+([0-9]+) | # m[4]: RFC or PMID, capture number
1212 (?: 97[89] [\ \-]? )? # optional 13-digit ISBN prefix
1213 (?: [0-9] [\ \-]? ){9} # 9 digits with opt. delimiters
1214 [0-9Xx] # check digit
1219 }
1221 /**
1222 * @throws MWException
1223 * @param $m array
1224 * @return HTML|string
1225 */
1228 # Skip anchor
1231 # Skip HTML element
1234 # Free external link
1237 # RFC or PMID
1251 }
1255 # ISBN
1261 ));
1268 }
1269 }
1271 /**
1272 * Make a free external link, given a user-supplied URL
1273 *
1274 * @param $url string
1275 *
1276 * @return string HTML
1277 * @private
1278 */
1284 # The characters '<' and '>' (which were escaped by
1285 # removeHTMLtags()) should not be included in
1286 # URLs, per RFC 2396.
1291 }
1293 # Move trailing punctuation to $trail
1295 # If there is no left bracket, then consider right brackets fair game too
1298 }
1304 }
1308 # Is this an external image?
1311 # Not an image, make a link
1316 # Register it in the output object...
1317 # Replace unnecessary URL escape codes with their equivalent characters
1320 }
1323 }
1325 /**
1326 * Parse headers and return html
1327 *
1328 * @private
1329 *
1330 * @param $text string
1331 *
1332 * @return string
1333 */
1339 }
1342 }
1344 /**
1345 * Replace single quotes with HTML markup
1346 * @private
1347 *
1348 * @param $text string
1349 *
1350 * @return string the altered text
1351 */
1358 }
1362 }
1364 /**
1365 * Helper function for doAllQuotes()
1366 *
1367 * @param $text string
1368 *
1369 * @return string
1370 */
1376 # First, do some preliminary work. This may shift some apostrophes from
1377 # being mark-up to being text. It also counts the number of occurrences
1378 # of bold and italics mark-ups.
1383 # If there are ever four apostrophes, assume the first is supposed to
1384 # be text, and the remaining three constitute mark-up for bold text.
1389 # If there are more than 5 apostrophes in a row, assume they're all
1390 # text except for the last 5.
1393 }
1394 # Count the number of occurrences of bold and italics mark-ups.
1395 # We are not counting sequences of five apostrophes.
1403 }
1404 }
1405 }
1407 # If there is an odd number of both bold and italics, it is likely
1408 # that one of the bold ones was meant to be an apostrophe followed
1409 # by italics. Which one we cannot know for certain, but it is more
1410 # likely to be one that has a single-letter word before it.
1423 }
1427 }
1431 }
1432 }
1433 }
1435 }
1437 # If there is a single-letter word, use it!
1442 # If not, but there's a multi-letter word, use that one.
1446 # ... otherwise use the first one that has neither.
1447 # (notice that it is possible for all three to be -1 if, for example,
1448 # there is only one pentuple-apostrophe in the line)
1451 }
1452 }
1454 # Now let's actually convert our apostrophic mush to HTML!
1465 }
1483 }
1500 }
1520 }
1521 }
1522 }
1524 }
1525 # Now close all remaining tags. Notice that the order is important.
1528 }
1531 }
1534 }
1535 # There might be lonely ''''', so make sure we have a buffer
1538 }
1540 }
1541 }
1543 /**
1544 * Replace external links (REL)
1545 *
1546 * Note: this is all very hackish and the order of execution matters a lot.
1547 * Make sure to run maintenance/parserTests.php if you change this code.
1548 *
1549 * @private
1550 *
1551 * @param $text string
1552 *
1553 * @throws MWException
1554 * @return string
1555 */
1562 throw new MWException( "PCRE needs to be compiled with --enable-unicode-properties in order for MediaWiki to function" );
1563 }
1573 # The characters '<' and '>' (which were escaped by
1574 # removeHTMLtags()) should not be included in
1575 # URLs, per RFC 2396.
1580 }
1582 # If the link text is an image URL, replace it with an <img> tag
1583 # This happened by accident in the original parser, but some people used it extensively
1587 }
1591 # Set linktype for CSS - if URL==text, link is essentially free
1594 # No link text, e.g. [http://domain.tld/some.link]
1596 # Autonumber
1601 # Have link text, e.g. [http://domain.tld/some.link text]s
1602 # Check for trail
1604 }
1610 # Use the encoded URL
1611 # This means that users can paste URLs directly into the text
1612 # Funny characters like ö aren't valid in URLs anyway
1613 # This was changed in August 2004
1617 # Register link in the output object.
1618 # Replace unnecessary URL escape codes with the referenced character
1619 # This prevents spammers from hiding links from the filters
1622 }
1626 }
1627 /**
1628 * Get the rel attribute for a particular external link.
1629 *
1630 * @since 1.21
1631 * @param string|bool $url optional URL, to extract the domain from for rel =>
1632 * nofollow if appropriate
1633 * @param $title Title optional Title, for wgNoFollowNsExceptions lookups
1634 * @return string|null rel attribute for $url
1635 */
1641 {
1643 }
1645 }
1646 /**
1647 * Get an associative array of additional HTML attributes appropriate for a
1648 * particular external link. This currently may include rel => nofollow
1649 * (depending on configuration, namespace, and the URL's domain) and/or a
1650 * target attribute (depending on configuration).
1651 *
1652 * @param string|bool $url optional URL, to extract the domain from for rel =>
1653 * nofollow if appropriate
1654 * @return Array associative array of HTML attributes
1655 */
1662 }
1664 }
1666 /**
1667 * Replace unusual URL escape codes with their equivalent characters
1668 *
1669 * @param $url String
1670 * @return String
1671 *
1672 * @todo This can merge genuinely required bits in the path or query string,
1673 * breaking legit URLs. A proper fix would treat the various parts of
1674 * the URL differently; as a workaround, just use the output for
1675 * statistical records, not for actual linking/output.
1676 */
1680 }
1682 /**
1683 * Callback function used in replaceUnusualEscapes().
1684 * Replaces unusual URL escape codes with their equivalent character
1685 *
1686 * @param $matches array
1687 *
1688 * @return string
1689 */
1693 # Is it an unsafe or HTTP reserved character according to RFC 1738?
1695 # No, shouldn't be escaped
1698 # Yes, leave it escaped
1700 }
1701 }
1703 /**
1704 * make an image if it's allowed, either through the global
1705 * option, through the exception, or through the on-wiki whitelist
1706 * @private
1707 *
1708 * $param $url string
1709 *
1710 * @return string
1711 */
1716 # $imagesfrom could be either a single string or an array of strings, parse out the latter
1723 }
1724 }
1729 }
1733 # Image found
1735 }
1736 }
1739 $whitelist = explode( "\n", wfMessage( 'external_image_whitelist' )->inContentLanguage()->text() );
1741 # Sanitize the regex fragment, make it case-insensitive, ignore blank entries/comments
1744 }
1746 # Image matches a whitelist entry
1749 }
1750 }
1751 }
1753 }
1755 /**
1756 * Process [[ ]] wikilinks
1757 *
1758 * @param $s string
1759 *
1760 * @return String: processed text
1761 *
1762 * @private
1763 */
1767 }
1769 /**
1770 * Process [[ ]] wikilinks (RIL)
1771 * @param $s
1772 * @throws MWException
1773 * @return LinkHolderArray
1774 *
1775 * @private
1776 */
1782 # the % is needed to support urlencoded titles as well
1785 # Match a link having the form [[namespace:link|alternate]]trail
1787 # Match cases where there is no "]]", which might still be images
1789 }
1793 # split the entire text string on occurrences of [[
1795 # get the first element (all text up to first [[), and remove the space we added
1804 # Match the end of a line for a word that's not followed by whitespace,
1805 # e.g. in the case of 'The Arab al[[Razi]]', 'al' will be matched
1807 }
1813 }
1822 }
1825 }
1830 # Loop for each link
1832 # Check for excessive memory usage
1834 # Too big
1835 # Do the existence check, replace the link holders and clear the array
1838 }
1847 }
1848 # first link
1852 }
1854 }
1861 # If we get a ] at the beginning of $m[3] that means we have a link that's something like:
1862 # [[Image:Foo.jpg|[http://example.com desc]]] <- having three ] in a row fucks up,
1863 # the real problem is with the $e1 regex
1864 # See bug 1300.
1865 #
1866 # Still some problems for cases where the ] is meant to be outside punctuation,
1867 # and no image is in sight. See bug 2095.
1868 #
1872 )
1873 {
1876 }
1877 # fix up urlencoded title texts
1879 # Should anchors '#' also be rejected?
1881 }
1883 } elseif ( preg_match( $e1_img, $line, $m ) ) { # Invalid, but might be an image with a link in its caption
1888 }
1894 }
1898 # Don't allow internal links to pages containing
1899 # PROTO: where PROTO is a valid URL protocol; these
1900 # should be external links.
1905 }
1907 # Make subpage if necessary
1912 }
1916 # Strip off leading ':'
1918 }
1927 }
1938 # look at the next 'line' to see if we can close it there
1943 }
1946 # the first ]] closes the inner link, the second the image
1952 # if there's exactly one ]] that's fine, we'll keep looking
1955 # if $next_line is invalid too, we need look no further
1958 }
1959 }
1961 # we couldn't find the end of this imageLink, so output it raw
1962 # but don't ignore what might be perfectly normal links in the text we've examined
1965 # note: no $trail, because without an end, there *is* no trail
1968 }
1971 # note: no $trail, because without an end, there *is* no trail
1974 }
1976 }
1982 # Bug 4598 madness. Handle the quotes only if they come from the alternate part
1983 # [[Lista d''e paise d''o munno]] -> <a href="...">Lista d''e paise d''o munno</a>
1984 # [[Criticism of Harry Potter|Criticism of ''Harry Potter'']]
1985 # -> <a href="Criticism of Harry Potter">Criticism of <i>Harry Potter</i></a>
1987 }
1989 # Link not escaped by : , create the various objects
1991 # Interwikis
1993 if ( $iw && $this->mOptions->getInterwikiMagic() && $nottalk && Language::fetchLanguageName( $iw, null, 'mw' ) ) {
1994 // XXX: the above check prevents links to sites with identifiers that are not language codes
1996 # Bug 24502: filter duplicates
2000 }
2006 }
2013 # if no parameters were passed, $text
2014 # becomes something like "File:Foo.png",
2015 # which we don't want to pass on to the
2016 # image generator
2019 # recursively parse links inside the image caption
2020 # actually, this will parse them in any other parameters, too,
2021 # but it might be hard to fix that, and it doesn't matter ATM
2024 }
2025 # cloak any absolute URLs inside the image markup, so replaceExternalLinks() won't touch them
2030 }
2033 }
2043 }
2049 /**
2050 * Strip the whitespace Category links produce, see bug 87
2051 * @todo We might want to use trim($tmp, "\n") here.
2052 */
2057 }
2058 }
2060 # Self-link checking
2065 true
2066 ) ) ) {
2069 }
2070 }
2072 # NS_MEDIA is a pseudo-namespace for linking directly to a file
2073 # @todo FIXME: Should do batch file existence checks, see comment below
2076 # Give extensions a chance to select the file revision for us
2081 # Fetch and register the file (file title may be different via hooks)
2083 # Cloak with NOPARSE to avoid replacement in replaceExternalLinks
2088 }
2091 # Some titles, such as valid special pages or files in foreign repos, should
2092 # be shown as bluelinks even though they're not included in the page table
2093 #
2094 # @todo FIXME: isAlwaysKnown() can be expensive for file links; we should really do
2095 # batch file existence checks for NS_FILE and NS_MEDIA
2100 # Links will be added to the output link list after checking
2102 }
2104 }
2107 }
2109 /**
2110 * Render a forced-blue link inline; protect against double expansion of
2111 * URLs if we're in a mode that prepends full URL prefixes to internal links.
2112 * Since this little disaster has to split off the trail text to avoid
2113 * breaking URLs in the following text without breaking trails on the
2114 * wiki links, it's been made into a horrible function.
2115 *
2116 * @param $nt Title
2117 * @param $text String
2118 * @param array $query or String
2119 * @param $trail String
2120 * @param $prefix String
2121 * @return String: HTML-wikitext mix oh yuck
2122 */
2128 }
2131 }
2136 }
2138 /**
2139 * Insert a NOPARSE hacky thing into any inline links in a chunk that's
2140 * going to go through further parsing steps before inline URL expansion.
2141 *
2142 * Not needed quite as much as it used to be since free links are a bit
2143 * more sensible these days. But bracketed links are still an issue.
2144 *
2145 * @param string $text more-or-less HTML
2146 * @return String: less-or-more HTML with NOPARSE bits
2147 */
2151 }
2153 /**
2154 * Return true if subpage links should be expanded on this page.
2155 * @return Boolean
2156 */
2158 # Some namespaces don't allow subpages
2160 }
2162 /**
2163 * Handle link to subpage if necessary
2164 *
2165 * @param string $target the source of the link
2166 * @param &$text String: the link text, modified as necessary
2167 * @return string the full name of the link
2168 * @private
2169 */
2172 }
2174 /**#@+
2175 * Used by doBlockLevels()
2176 * @private
2177 *
2178 * @return string
2179 */
2184 }
2188 }
2190 /**
2191 * getCommon() returns the length of the longest common substring
2192 * of both arguments, starting at the beginning of both.
2193 * @private
2194 *
2195 * @param $st1 string
2196 * @param $st2 string
2197 *
2198 * @return int
2199 */
2205 }
2210 }
2211 }
2213 }
2215 /**
2216 * These next three functions open, continue, and close the list
2217 * element appropriate to the prefix character passed into them.
2218 * @private
2219 *
2220 * @param $char string
2221 *
2222 * @return string
2223 */
2238 }
2241 }
2243 /**
2244 * TODO: document
2245 * @param $char String
2246 * @private
2247 *
2248 * @return string
2249 */
2257 }
2264 }
2265 }
2267 }
2269 /**
2270 * TODO: document
2271 * @param $char String
2272 * @private
2273 *
2274 * @return string
2275 */
2287 }
2290 }
2292 }
2293 /**#@-*/
2295 /**
2296 * Make lists from lines starting with ':', '*', '#', etc. (DBL)
2297 *
2298 * @param $text String
2299 * @param $linestart Boolean: whether or not this is at the start of a line.
2300 * @private
2301 * @return string the lists rendered as HTML
2302 */
2306 # Parsing through the text line by line. The main thing
2307 # happening here is handling of block-level elements p, pre,
2308 # and making lists from lines starting with * # : etc.
2309 #
2318 # Fix up $linestart
2323 }
2324 # * = ul
2325 # # = ol
2326 # ; = dt
2327 # : = dd
2332 # If not in a <pre> element, scan for and figure out what prefixes are there.
2334 # Multiple prefixes may abut each other for nested lists.
2338 # eh?
2339 # ; and : are both from definition-lists, so they're equivalent
2340 # for the purposes of determining whether or not we need to open/close
2341 # elements.
2346 # Don't interpret any other prefixes in preformatted text
2350 }
2352 # List generation
2354 # Same as the last item, so no need to deal with nesting or opening stuff
2359 # The one nasty exception: definition lists work like this:
2360 # ; title : definition text
2361 # So we check for : in the remainder text to split up the
2362 # title and definition, without b0rking links.
2367 }
2368 }
2370 # We need to open or close prefixes, or both.
2372 # Either open or close a level...
2376 # Close all the prefixes which aren't shared.
2380 }
2382 # Continue the current prefix if appropriate.
2385 }
2387 # Open prefixes where appropriate.
2393 # @todo FIXME: This is dupe of code above
2397 }
2398 }
2400 }
2402 }
2404 # If we have no prefixes, go to paragraph mode.
2407 # No prefix (not in list)--go to paragraph mode
2408 # XXX: use a stack for nestable elements like span, table and div
2409 $openmatch = preg_match( '/(?:<table|<blockquote|<h1|<h2|<h3|<h4|<h5|<h6|<pre|<tr|<p|<ul|<ol|<dl|<li|<\\/tr|<\\/td|<\\/th)/iS', $t );
2412 '<td|<th|<\\/?div|<hr|<\\/pre|<\\/p|' . $this->mUniqPrefix . '-pre|<\\/li|<\\/ul|<\\/ol|<\\/dl|<\\/?center)/iS', $t );
2415 # TODO bug 5718: paragraph closed
2419 }
2423 # pre
2428 }
2431 # paragraph
2444 }
2445 }
2454 }
2455 }
2456 }
2457 }
2459 }
2460 # somewhere above we forget to get out of pre block (bug 785)
2463 }
2466 }
2467 }
2471 }
2475 }
2479 }
2481 /**
2482 * Split up a string on ':', ignoring any occurrences inside tags
2483 * to prevent illegal overlapping.
2484 *
2485 * @param string $str the string to split
2486 * @param &$before String set to everything before the ':'
2487 * @param &$after String set to everything after the ':'
2488 * @throws MWException
2489 * @return String the position of the ':', or false if none found
2490 */
2496 # Nothing to find!
2499 }
2503 # Easy; no tag nesting to worry about
2508 }
2510 # Ugly state machine to walk through avoiding tags.
2518 # (Using the number is a performance hack for common cases)
2522 # Could be either a <start> tag or an </end> tag
2527 # We found it!
2532 }
2533 # Embedded in a tag; don't break it.
2536 # Skip ahead looking for something interesting
2539 # Nothing else interesting
2542 }
2546 # We found it!
2551 }
2552 }
2554 # Nothing else interesting to find; abort!
2555 # We're nested, but there's no close tags left. Abort!
2557 }
2558 # Skip ahead to next tag start
2561 }
2564 # In a <tag>
2571 # Slash may be followed by >?
2575 # ignore
2576 }
2587 # Illegal early close? This shouldn't happen D:
2592 }
2595 # In a </tag>
2602 }
2604 }
2608 # Yes, a self-closed tag <blah/>
2611 # Probably we're jumping the gun, and this is an attribute
2613 }
2618 }
2625 }
2632 }
2637 }
2638 }
2640 wfDebug( __METHOD__ . ": Invalid input; not enough close tags (stack $stack, state $state)\n" );
2643 }
2646 }
2648 /**
2649 * Return value of a magic variable (like PAGENAME)
2650 *
2651 * @private
2652 *
2653 * @param $index integer
2654 * @param bool|\PPFrame $frame
2655 *
2656 * @throws MWException
2657 * @return string
2658 */
2664 // If no title set, bad things are going to happen
2665 // later. Title should always be set since this
2666 // should only be called in the middle of a parse
2667 // operation (but the unit-tests do funky stuff)
2670 }
2672 /**
2673 * Some of these require message or data lookups and can be
2674 * expensive to check many times.
2675 */
2679 }
2680 }
2685 # Use the time zone
2690 }
2704 }
2773 $value = wfEscapeWikiText( wfUrlEncode( str_replace( ' ', '_', $this->mTitle->getRootText() ) ) );
2779 $value = wfEscapeWikiText( wfUrlEncode( str_replace( ' ', '_', $this->mTitle->getBaseText() ) ) );
2787 }
2795 }
2808 # 0 means the page doesn't exist in the database,
2809 # which means the user is previewing a new page.
2810 # The vary-revision flag must be set, because the magic word
2811 # will have a different value once the page is saved.
2814 }
2818 # Let the edit saving system know we should parse the page
2819 # *after* a revision ID has been assigned.
2825 # Let the edit saving system know we should parse the page
2826 # *after* a revision ID has been assigned. This is for null edits.
2832 # Let the edit saving system know we should parse the page
2833 # *after* a revision ID has been assigned. This is for null edits.
2839 # Let the edit saving system know we should parse the page
2840 # *after* a revision ID has been assigned. This is for null edits.
2846 # Let the edit saving system know we should parse the page
2847 # *after* a revision ID has been assigned. This is for null edits.
2853 # Let the edit saving system know we should parse the page
2854 # *after* a revision ID has been assigned. This is for null edits.
2860 # Let the edit saving system know we should parse the page
2861 # *after* a revision ID has been assigned. This is for null edits.
2867 # Let the edit saving system know we should parse the page
2868 # *after* a revision ID has been assigned. This is for null edits.
2883 $value = $this->mTitle->canTalk() ? str_replace( '_', ' ', $this->mTitle->getTalkNsText() ) : '';
2907 # @bug 4594 PHP5 has it zero padded, PHP4 does not, cast to
2908 # int to remove the padding
2927 # @bug 4594 PHP5 has it zero padded, PHP4 does not, cast to
2928 # int to remove the padding
2988 if ( wfRunHooks( 'ParserGetVariableValueSwitch', array( &$this, &$this->mVarCache, &$index, &$ret, &$frame ) ) ) {
2992 }
2993 }
2997 }
3000 }
3002 /**
3003 * initialise the magic variables (like CURRENTMONTHNAME) and substitution modifiers
3004 *
3005 * @private
3006 */
3015 }
3017 /**
3018 * Preprocess some wikitext and return the document tree.
3019 * This is the ghost of replace_variables().
3020 *
3021 * @param string $text The text to parse
3022 * @param $flags Integer: bitwise combination of:
3023 * self::PTD_FOR_INCLUSION Handle "<noinclude>" and "<includeonly>" as if the text is being
3024 * included. Default is to assume a direct page view.
3025 *
3026 * The generated DOM tree must depend only on the input text and the flags.
3027 * The DOM tree must be the same in OT_HTML and OT_WIKI mode, to avoid a regression of bug 4899.
3028 *
3029 * Any flag added to the $flags parameter here, or any other parameter liable to cause a
3030 * change in the DOM tree for a given text, must be passed through the section identifier
3031 * in the section edit link and thus back to extractSections().
3032 *
3033 * The output of this function is currently only cached in process memory, but a persistent
3034 * cache may be implemented at a later date which takes further advantage of these strict
3035 * dependency requirements.
3036 *
3037 * @private
3038 *
3039 * @return PPNode
3040 */
3044 }
3046 /**
3047 * Return a three-element array: leading whitespace, string contents, trailing whitespace
3048 *
3049 * @param $s string
3050 *
3051 * @return array
3052 */
3062 }
3064 }
3066 /**
3067 * Replace magic variables, templates, and template arguments
3068 * with the appropriate text. Templates are substituted recursively,
3069 * taking care to avoid infinite loops.
3070 *
3071 * Note that the substitution depends on value of $mOutputType:
3072 * self::OT_WIKI: only {{subst:}} templates
3073 * self::OT_PREPROCESS: templates but not extension tags
3074 * self::OT_HTML: all templates and extension tags
3075 *
3076 * @param string $text the text to transform
3077 * @param $frame PPFrame Object describing the arguments passed to the template.
3078 * Arguments may also be provided as an associative array, as was the usual case before MW1.12.
3079 * Providing arguments this way may be useful for extensions wishing to perform variable replacement explicitly.
3080 * @param $argsOnly Boolean only do argument (triple-brace) expansion, not double-brace expansion
3081 * @private
3082 *
3083 * @return string
3084 */
3086 # Is there any text? Also, Prevent too big inclusions!
3089 }
3095 wfDebug( __METHOD__ . " called using plain parameters instead of a PPFrame instance. Creating custom frame.\n" );
3097 }
3105 }
3107 /**
3108 * Clean up argument array - refactored in 1.9 so parserfunctions can use it, too.
3109 *
3110 * @param $args array
3111 *
3112 * @return array
3113 */
3126 }
3129 }
3130 }
3131 }
3134 }
3136 /**
3137 * Warn the user when a parser limitation is reached
3138 * Will warn at most once the user per limitation type
3139 *
3140 * @param string $limitationType should be one of:
3141 * 'expensive-parserfunction' (corresponding messages:
3142 * 'expensive-parserfunction-warning',
3143 * 'expensive-parserfunction-category')
3144 * 'post-expand-template-argument' (corresponding messages:
3145 * 'post-expand-template-argument-warning',
3146 * 'post-expand-template-argument-category')
3147 * 'post-expand-template-inclusion' (corresponding messages:
3148 * 'post-expand-template-inclusion-warning',
3149 * 'post-expand-template-inclusion-category')
3150 * @param int|null $current Current value
3151 * @param int|null $max Maximum allowed, when an explicit limit has been
3152 * exceeded, provide the values (optional)
3153 */
3155 # does no harm if $current and $max are present but are unnecessary for the message
3160 }
3162 /**
3163 * Return the text of a template, after recursively
3164 * replacing any variables or templates within the template.
3165 *
3166 * @param array $piece the parts of the template
3167 * $piece['title']: the title, i.e. the part before the |
3168 * $piece['parts']: the parameter array
3169 * $piece['lineStart']: whether the brace was at the start of a line
3170 * @param $frame PPFrame The current frame, contains template arguments
3171 * @throws MWException
3172 * @return String: the text of the template
3173 * @private
3174 */
3179 # Flags
3187 # Title object, where $text came from
3190 # $part1 is the bit before the first |, and must contain only title characters.
3191 # Various prefixes will be stripped from it later.
3196 # Original title text preserved for various purposes
3199 # $args is a list of argument nodes, starting from index 0, not including $part1
3200 # @todo FIXME: If piece['parts'] is null then the call to getLength() below won't work b/c this $args isn't an object
3206 # SUBST
3212 # Possibilities for substMatch: "subst", "safesubst" or FALSE
3213 # Decide whether to expand template or keep wikitext as-is.
3219 }
3225 }
3226 }
3231 }
3232 }
3234 # Variables
3241 }
3243 }
3244 }
3246 # MSG, MSGNW and RAW
3248 # Check for MSGNW:
3253 # Remove obsolete MSG:
3256 }
3258 # Check for RAW:
3262 }
3263 }
3266 # Parser functions
3276 }
3283 }
3285 # The interface for parser functions allows for extracting
3286 # flags into the local scope. Extract any forwarded flags
3287 # here.
3289 }
3291 }
3293 # Finish mangling title and then check for loops.
3294 # Set $title to a Title object and $titleText to the PDBK
3297 # Split the title into page and subpage
3302 }
3306 # Check for language variants if the template is not found
3309 }
3310 # Do recursion depth check
3318 }
3319 }
3320 }
3322 # Load from database
3325 # Too many unique items can kill profiling DBs/collectors
3328 }
3334 {
3335 // Pass the template arguments as URL parameters.
3336 // "uselang" will have no effect since the Language object
3337 // is forced to the one defined in ParserOptions.
3345 }
3346 }
3348 // Create a new context to execute the special page
3361 }
3370 }
3371 }
3373 # If the title is valid but undisplayable, make a link to it
3377 }
3379 # Interwiki transclusion
3385 # Preprocess it like a template
3388 }
3390 }
3392 # Do infinite loop check
3393 # This has to be done after redirect resolution to avoid infinite loops via redirects
3400 }
3402 }
3404 # If we haven't found text to substitute by now, we're done
3405 # Recover the source wikitext and return it
3410 }
3413 }
3415 # Expand DOM-style return values in a child frame
3417 # Clean up argument array
3423 # Expansion is eligible for the empty-frame cache
3429 }
3431 # Uncached expansion
3433 }
3434 }
3438 }
3442 }
3444 # Replace raw HTML by a placeholder
3448 # Escape nowiki-style return values
3453 {
3454 # Bug 529: if the template begins with a table or block-level
3455 # element, it should be treated as beginning a new line.
3456 # This behavior is somewhat controversial.
3458 }
3461 # Error, oversize inclusion
3463 # Make a working, properly escaped link if possible (bug 23588)
3466 # This will probably not be a working link, but at least it may
3467 # provide some hint of where the problem is
3470 }
3471 $text .= $this->insertStripItem( '<!-- WARNING: template omitted, post-expand include size too large -->' );
3473 }
3479 }
3483 }
3485 /**
3486 * Call a parser function and return an array with text and flags.
3487 *
3488 * The returned array will always contain a boolean 'found', indicating
3489 * whether the parser function was found or not. It may also contain the
3490 * following:
3491 * text: string|object, resulting wikitext or PP DOM object
3492 * isHTML: bool, $text is HTML, armour it against wikitext transformation
3493 * isChildObj: bool, $text is a DOM node needing expansion in a child frame
3494 * isLocalObj: bool, $text is a DOM node needing expansion in the current frame
3495 * nowiki: bool, wiki markup in $text should be escaped
3496 *
3497 * @since 1.21
3498 * @param $frame PPFrame The current frame, contains template arguments
3499 * @param $function string Function name
3500 * @param $args array Arguments to the function
3501 * @return array
3502 */
3508 # Case sensitive functions
3512 # Case insensitive functions
3519 }
3520 }
3525 # Workaround for PHP bug 35229 and similar
3530 }
3534 # Convert arguments to PPNodes and collect for appending to $allArgs
3541 }
3542 }
3544 # Add a frame parameter, and pass the arguments as an array
3548 # Convert arguments to plain text and append to $allArgs
3556 }
3557 }
3558 }
3562 # The interface for function hooks allows them to return a wikitext
3563 # string or an array containing the string and any flags. This mungs
3564 # things around to match what this method should return.
3569 );
3573 }
3577 );
3578 }
3584 }
3587 }
3592 }
3597 }
3599 /**
3600 * Get the semi-parsed DOM representation of a template with a given title,
3601 * and its redirect destination title. Cached.
3602 *
3603 * @param $title Title
3604 *
3605 * @return array
3606 */
3615 }
3618 }
3620 # Cache miss, go to the database
3626 }
3634 }
3637 }
3639 /**
3640 * Fetch the unparsed text of a template and register a reference to it.
3641 * @param Title $title
3642 * @return Array ( string or false, Title )
3643 */
3645 $templateCb = $this->mOptions->getTemplateCallback(); # Defaults to Parser::statelessFetchTemplate()
3653 // If we transclude ourselves, the final result
3654 // will change based on the new version of the page
3656 }
3657 }
3658 }
3660 }
3662 /**
3663 * Fetch the unparsed text of a template and register a reference to it.
3664 * @param Title $title
3665 * @return mixed string or false
3666 */
3670 }
3672 /**
3673 * Static function to get a template
3674 * Can be overridden via ParserOptions::setTemplateCallback().
3675 *
3676 * @param $title Title
3677 * @param $parser Parser
3678 *
3679 * @return array
3680 */
3686 # Loop to fetch the article, with up to 1 redirect
3688 # Give extensions a chance to select the revision instead
3699 );
3701 }
3702 # Get the revision
3707 # If there is no current revision, there is no page
3711 }
3718 # We fetched a rev from a different title; register it too...
3723 }
3732 }
3739 }
3744 }
3747 }
3748 # Redirect?
3751 }
3756 }
3758 /**
3759 * Fetch a file and its title and register a reference to it.
3760 * If 'broken' is a key in $options then the file will appear as a broken thumbnail.
3761 * @param Title $title
3762 * @param array $options Array of options to RepoGroup::findFile
3763 * @return File|bool
3764 */
3768 }
3770 /**
3771 * Fetch a file and its title and register a reference to it.
3772 * If 'broken' is a key in $options then the file will appear as a broken thumbnail.
3773 * @param Title $title
3774 * @param array $options Array of options to RepoGroup::findFile
3775 * @return Array ( File or false, Title of file )
3776 */
3784 }
3787 # Register the file as a dependency...
3790 # Update fetched file title
3793 # This file was not a redirect, but the title does not match.
3794 # Register under the new name because otherwise the link will
3795 # get lost.
3797 }
3798 }
3800 }
3802 /**
3803 * Transclude an interwiki link.
3804 *
3805 * @param $title Title
3806 * @param $action
3807 *
3808 * @return string
3809 */
3815 }
3821 }
3823 }
3825 /**
3826 * @param $url string
3827 * @return Mixed|String
3828 */
3837 }
3843 } elseif ( $req->getStatus() != 200 ) { // Though we failed to fetch the content, this status is useless.
3844 return wfMessage( 'scarytranscludefailed-httpstatus', $url, $req->getStatus() /* HTTP status */ )->inContentLanguage()->text();
3847 }
3854 ) );
3856 }
3858 /**
3859 * Triple brace replacement -- used for template arguments
3860 * @private
3861 *
3862 * @param $piece array
3863 * @param $frame PPFrame
3864 *
3865 * @return array
3866 */
3877 && (
3881 )
3882 ) {
3883 # No match in frame, use the supplied default
3885 }
3889 }
3892 # No match anywhere
3894 }
3897 }
3902 }
3906 }
3908 /**
3909 * Return the text to be used for a given extension tag.
3910 * This is the ghost of strip().
3911 *
3912 * @param array $params Associative array of parameters:
3913 * name PPNode for the tag name
3914 * attr PPNode for unparsed text where tag attributes are thought to be
3915 * attributes Optional associative array of parsed attributes
3916 * inner Contents of extension element
3917 * noClose Original text did not have a close tag
3918 * @param $frame PPFrame
3919 *
3920 * @throws MWException
3921 * @return string
3922 */
3927 $marker = "{$this->mUniqPrefix}-$name-" . sprintf( '%08X', $this->mMarkerIndex++ ) . self::MARKER_SUFFIX;
3935 }
3941 }
3944 # Workaround for PHP bug 35229 and similar
3947 }
3954 }
3960 }
3963 # Extract flags to local scope (to override $markerType)
3968 }
3972 }
3977 }
3978 }
3984 }
3985 }
3995 }
3997 }
3999 /**
4000 * Increment an include size counter
4001 *
4002 * @param string $type the type of expansion
4003 * @param $size Integer: the size of the text
4004 * @return Boolean: false if this inclusion would take it over the maximum, true otherwise
4005 */
4012 }
4013 }
4015 /**
4016 * Increment the expensive function count
4017 *
4018 * @return Boolean: false if the limit has been exceeded
4019 */
4023 }
4025 /**
4026 * Strip double-underscore items like __NOGALLERY__ and __NOTOC__
4027 * Fills $this->mDoubleUnderscores, returns the modified text
4028 *
4029 * @param $text string
4030 *
4031 * @return string
4032 */
4036 # The position of __TOC__ needs to be recorded
4042 # Set a placeholder. At the end we'll fill it in with the TOC.
4045 # Only keep the first one.
4047 }
4049 # Now match and remove the rest of them
4055 }
4058 }
4059 if ( isset( $this->mDoubleUnderscores['hiddencat'] ) && $this->mTitle->getNamespace() == NS_CATEGORY ) {
4061 }
4062 # (bug 8068) Allow control over whether robots index a page.
4063 #
4064 # @todo FIXME: Bug 14899: __INDEX__ always overrides __NOINDEX__ here! This
4065 # is not desirable, the last one on the page should win.
4069 }
4073 }
4075 # Cache all double underscores in the database
4078 }
4082 }
4084 /**
4085 * Add a tracking category, getting the title from a system message,
4086 * or print a debug message if the title is invalid.
4087 *
4088 * @param string $msg message key
4089 * @return Boolean: whether the addition was successful
4090 */
4095 }
4096 // Important to parse with correct title (bug 31469)
4102 # Allow tracking categories to be disabled by setting them to "-"
4105 }
4114 }
4115 }
4117 /**
4118 * This function accomplishes several tasks:
4119 * 1) Auto-number headings if that option is enabled
4120 * 2) Add an [edit] link to sections for users who have enabled the option and can edit the page
4121 * 3) Add a Table of contents on the top for users who have enabled the option
4122 * 4) Auto-anchor headings
4123 *
4124 * It loops through all headlines, collects the necessary data, then splits up the
4125 * string and re-inserts the newly formatted headlines.
4126 *
4127 * @param $text String
4128 * @param string $origText original, untouched wikitext
4129 * @param $isMain Boolean
4130 * @return mixed|string
4131 * @private
4132 */
4136 # Inhibit editsection links if requested in the page
4142 }
4145 }
4147 # Get all headlines for numbering them and adding funky stuff like [edit]
4148 # links - this is for later, but we need the number of headlines right now
4150 $numMatches = preg_match_all( '/<H(?P<level>[1-6])(?P<attrib>.*?' . '>)\s*(?P<header>[\s\S]*?)\s*<\/H[1-6] *>/i', $text, $matches );
4152 # if there are fewer than 4 headlines in the article, do not show TOC
4153 # unless it's been explicitly enabled.
4157 # Allow user to stipulate that a page should have a "new section"
4158 # link added via __NEWSECTIONLINK__
4161 }
4163 # Allow user to remove the "new section"
4164 # link via __NONEWSECTIONLINK__
4167 }
4169 # if the string __FORCETOC__ (not case-sensitive) occurs in the HTML,
4170 # override above conditions and always show TOC above first header
4174 }
4176 # headline counter
4180 # Ugh .. the TOC should have neat indentation levels which can be
4181 # passed to the skin functions. These are determined here
4213 }
4217 }
4221 # Increase TOC level
4228 }
4230 # Decrease TOC level, find level to jump to
4234 # Found last matching level
4238 # Found first matching level below current level
4241 }
4242 }
4245 }
4248 # Unindent only if the previous toc level was shown :p
4253 }
4254 }
4256 # No change in level, end TOC line
4259 }
4260 }
4264 # count number of headlines for each level
4271 }
4274 }
4275 }
4277 # The safe header is a version of the header text safe to use for links
4279 # Remove link placeholders by the link text.
4280 # <!--LINK number-->
4281 # turns into
4282 # link text with suffix
4283 # Do this before unstrip since link text can contain strip markers
4286 # Avoid insertion of weird stuff like <math> by expanding the relevant sections
4289 # Strip out HTML (first regex removes any tag not allowed)
4290 # Allowed tags are:
4291 # * <sup> and <sub> (bug 8393)
4292 # * <i> (bug 26375)
4293 # * <b> (r105284)
4294 # * <span dir="rtl"> and <span dir="ltr"> (bug 35167)
4295 #
4296 # We strip any parameter from accepted tags (second regex), except dir="rtl|ltr" from <span>,
4297 # to allow setting directionality in toc items.
4299 array( '#<(?!/?(span|sup|sub|i|b)(?: [^>]*)?>).*?' . '>#', '#<(/?(?:span(?: dir="(?:rtl|ltr)")?|sup|sub|i|b))(?: .*?)?' . '>#' ),
4301 $safeHeadline
4302 );
4305 # For the anchor, strip out HTML-y stuff period
4309 # Save headline for section edit hint before it's escaped
4313 # For reverse compatibility, provide an id that's
4314 # HTML4-compatible, like we used to.
4315 #
4316 # It may be worth noting, academically, that it's possible for
4317 # the legacy anchor to conflict with a non-legacy headline
4318 # anchor on the page. In this case likely the "correct" thing
4319 # would be to either drop the legacy anchors or make sure
4320 # they're numbered first. However, this would require people
4321 # to type in section names like "abc_.D7.93.D7.90.D7.A4"
4322 # manually, so let's not bother worrying about it.
4328 # No reason to have both (in fact, we can't)
4330 }
4335 }
4337 # HTML names must be case-insensitively unique (bug 10721).
4338 # This does not apply to Unicode characters per
4339 # http://dev.w3.org/html5/spec/infrastructure.html#case-sensitivity-and-string-comparison
4340 # @todo FIXME: We may be changing them depending on the current locale.
4346 }
4348 # count how many in assoc. array so we can track dupes in anchors
4353 }
4358 }
4360 # Don't number the heading if it is the only one (looks silly)
4362 # the two are different if the line contains a link
4363 $headline = Html::element( 'span', array( 'class' => 'mw-headline-number' ), $numbering ) . ' ' . $headline;
4364 }
4366 # Create the anchor for linking from the TOC to the section
4371 }
4374 }
4378 }
4380 # Add the section to the section tree
4381 # Find the DOM node for this header
4387 }
4388 }
4392 }
4402 );
4404 # give headline the correct <h#> tag
4406 // Output edit section links as markers with styles that can be customized by skins
4408 # Put a T flag in the section identifier, to indicate to extractSections()
4409 # that sections inside <includeonly> should be counted.
4413 }
4414 // We use a bit of pesudo-xml for editsection markers. The language converter is run later on
4415 // Using a UNIQ style marker leads to the converter screwing up the tokens when it converts stuff
4416 // And trying to insert strip tags fails too. At this point all real inputted tags have already been escaped
4417 // so we don't have to worry about a user trying to input one of these markers directly.
4418 // We use a page and section attribute to stop the language converter from converting these important bits
4419 // of data, but put the headline hint inside a content block because the language converter is supposed to
4420 // be able to convert that piece of data.
4427 }
4430 }
4436 }
4440 # Never ever show TOC if no headers
4443 }
4448 }
4451 }
4455 }
4457 # split up and insert constructed headlines
4461 // build an array of document sections
4464 // $head is zero-based, sections aren't.
4469 }
4471 /**
4472 * Send a hook, one per section.
4473 * The idea here is to be able to make section-level DIVs, but to do so in a
4474 * lower-impact, more correct way than r50769
4475 *
4476 * $this : caller
4477 * $section : the section number
4478 * &$sectionContent : ref to the content of the section
4479 * $showEditLinks : boolean describing whether this section has an edit link
4480 */
4484 }
4487 // append the TOC at the beginning
4488 // Top anchor now in skin
4490 }
4498 }
4499 }
4501 /**
4502 * Transform wiki markup when saving a page by doing "\r\n" -> "\n"
4503 * conversion, substitting signatures, {{subst:}} templates, etc.
4504 *
4505 * @param string $text the text to transform
4506 * @param $title Title: the Title object for the current article
4507 * @param $user User: the User object describing the current user
4508 * @param $options ParserOptions: parsing options
4509 * @param $clearState Boolean: whether to clear the parser state first
4510 * @return String: the altered wiki markup
4511 */
4512 public function preSaveTransform( $text, Title $title, User $user, ParserOptions $options, $clearState = true ) {
4518 );
4522 }
4528 }
4530 /**
4531 * Pre-save transform helper function
4532 * @private
4533 *
4534 * @param $text string
4535 * @param $user User
4536 *
4537 * @return string
4538 */
4542 # Note: This is the timestamp saved as hardcoded wikitext to
4543 # the database, we use $wgContLang here in order to give
4544 # everyone the same signature and use the default one rather
4545 # than the one selected in each user's preferences.
4546 # (see also bug 12815)
4552 }
4560 # Allow translation of timezones through wiki. date() can return
4561 # whatever crap the system uses, localised or not, so we cannot
4562 # ship premade translations.
4567 }
4573 # Variable replacement
4574 # Because mOutputType is OT_WIKI, this will only process {{subst:xxx}} type tags
4577 # This works almost by chance, as the replaceVariables are done before the getUserSig(),
4578 # which may corrupt this parser instance via its wfMessage()->text() call-
4580 # Signatures
4586 ) );
4588 # Context links ("pipe tricks"): [[|name]] and [[name (context)|]]
4593 $p4 = "/\[\[(:?$nc+:|:|)($tc+?)( ?（$tc+）)\\|]]/"; # [[ns:page（context）|]] (double-width brackets, added in r40257)
4594 $p3 = "/\[\[(:?$nc+:|:|)($tc+?)( ?\\($tc+\\)|)((?:, |，)$tc+|)\\|]]/"; # [[ns:page (context), context|]] (using either single or double-width comma)
4597 # try $p1 first, to turn "[[A, B (C)|]]" into "[[A, B (C)|A, B]]"
4609 # if there's no context, don't bother duplicating the title
4611 }
4613 # Trim trailing whitespace
4617 }
4619 /**
4620 * Fetch the user's signature text, if any, and normalize to
4621 * validated, ready-to-insert wikitext.
4622 * If you have pre-fetched the nickname or the fancySig option, you can
4623 * specify them here to save a database query.
4624 * Do not reuse this parser instance after calling getUserSig(),
4625 * as it may have changed if it's the $wgParser.
4626 *
4627 * @param $user User
4628 * @param string|bool $nickname nickname to use or false to use user's default nickname
4629 * @param $fancySig Boolean|null whether the nicknname is the complete signature
4630 * or null to use default value
4631 * @return string
4632 */
4638 # If not given, retrieve from the user object.
4641 }
4645 }
4653 # Sig. might contain markup; validate this
4655 # Validated; clean up (if needed) and return it
4658 # Failed to validate; fall back to the default
4661 }
4662 }
4664 # Make sure nickname doesnt get a sig in a sig
4667 # If we're still here, make it a link to the user page
4672 return wfMessage( $msgName, $userText, $nickText )->inContentLanguage()->title( $this->getTitle() )->text();
4673 }
4675 /**
4676 * Check that the user's signature contains no bad XML
4677 *
4678 * @param $text String
4679 * @return mixed An expanded string, or false if invalid.
4680 */
4683 }
4685 /**
4686 * Clean up signature text
4687 *
4688 * 1) Strip ~~~, ~~~~ and ~~~~~ out of signatures @see cleanSigInSig
4689 * 2) Substitute all transclusions
4690 *
4691 * @param $text String
4692 * @param bool $parsing Whether we're cleaning (preferences save) or parsing
4693 * @return String: signature text
4694 */
4699 }
4701 # Option to disable this feature
4704 }
4706 # @todo FIXME: Regex doesn't respect extension tags or nowiki
4707 # => Move this logic to braceSubstitution()
4709 $substRegex = '/\{\{(?!(?:' . $substWord->getBaseRegex() . '))/x' . $substWord->getRegexCase();
4720 }
4723 }
4725 /**
4726 * Strip ~~~, ~~~~ and ~~~~~ out of signatures
4727 *
4728 * @param $text String
4729 * @return String: signature text with /~{3,5}/ removed
4730 */
4734 }
4736 /**
4737 * Set up some variables which are usually set up in parse()
4738 * so that an external function can call some class members with confidence
4739 *
4740 * @param $title Title|null
4741 * @param $options ParserOptions
4742 * @param $outputType
4743 * @param $clearState bool
4744 */
4745 public function startExternalParse( Title $title = null, ParserOptions $options, $outputType, $clearState = true ) {
4747 }
4749 /**
4750 * @param $title Title|null
4751 * @param $options ParserOptions
4752 * @param $outputType
4753 * @param $clearState bool
4754 */
4755 private function startParse( Title $title = null, ParserOptions $options, $outputType, $clearState = true ) {
4761 }
4762 }
4764 /**
4765 * Wrapper for preprocess()
4766 *
4767 * @param string $text the text to preprocess
4768 * @param $options ParserOptions: options
4769 * @param $title Title object or null to use $wgTitle
4770 * @return String
4771 */
4775 # Guard against infinite recursion
4778 }
4785 }
4792 }
4794 /**
4795 * Create an HTML-style tag, e.g. "<yourtag>special text</yourtag>"
4796 * The callback should have the following form:
4797 * function myParserHook( $text, $params, $parser, $frame ) { ... }
4798 *
4799 * Transform and return $text. Use $parser for any required context, e.g. use
4800 * $parser->getTitle() and $parser->getOptions() not $wgTitle or $wgOut->mParserOptions
4801 *
4802 * Hooks may return extended information by returning an array, of which the
4803 * first numbered element (index 0) must be the return string, and all other
4804 * entries are extracted into local variables within an internal function
4805 * in the Parser class.
4806 *
4807 * This interface (introduced r61913) appears to be undocumented, but
4808 * 'markerName' is used by some core tag hooks to override which strip
4809 * array their results are placed in. **Use great caution if attempting
4810 * this interface, as it is not documented and injudicious use could smash
4811 * private variables.**
4812 *
4813 * @param $tag Mixed: the tag to use, e.g. 'hook' for "<hook>"
4814 * @param $callback Mixed: the callback function (and object) to use for the tag
4815 * @throws MWException
4816 * @return Mixed|null The old value of the mTagHooks array associated with the hook
4817 */
4822 }
4827 }
4830 }
4832 /**
4833 * As setHook(), but letting the contents be parsed.
4834 *
4835 * Transparent tag hooks are like regular XML-style tag hooks, except they
4836 * operate late in the transformation sequence, on HTML instead of wikitext.
4837 *
4838 * This is probably obsoleted by things dealing with parser frames?
4839 * The only extension currently using it is geoserver.
4840 *
4841 * @since 1.10
4842 * @todo better document or deprecate this
4843 *
4844 * @param $tag Mixed: the tag to use, e.g. 'hook' for "<hook>"
4845 * @param $callback Mixed: the callback function (and object) to use for the tag
4846 * @throws MWException
4847 * @return Mixed|null The old value of the mTagHooks array associated with the hook
4848 */
4853 }
4854 $oldVal = isset( $this->mTransparentTagHooks[$tag] ) ? $this->mTransparentTagHooks[$tag] : null;
4858 }
4860 /**
4861 * Remove all tag hooks
4862 */
4867 }
4869 /**
4870 * Create a function, e.g. {{sum:1|2|3}}
4871 * The callback function should have the form:
4872 * function myParserFunction( &$parser, $arg1, $arg2, $arg3 ) { ... }
4873 *
4874 * Or with SFH_OBJECT_ARGS:
4875 * function myParserFunction( $parser, $frame, $args ) { ... }
4876 *
4877 * The callback may either return the text result of the function, or an array with the text
4878 * in element 0, and a number of flags in the other elements. The names of the flags are
4879 * specified in the keys. Valid flags are:
4880 * found The text returned is valid, stop processing the template. This
4881 * is on by default.
4882 * nowiki Wiki markup in the return value should be escaped
4883 * isHTML The returned text is HTML, armour it against wikitext transformation
4884 *
4885 * @param string $id The magic word ID
4886 * @param $callback Mixed: the callback function (and object) to use
4887 * @param $flags Integer: a combination of the following flags:
4888 * SFH_NO_HASH No leading hash, i.e. {{plural:...}} instead of {{#if:...}}
4889 *
4890 * SFH_OBJECT_ARGS Pass the template arguments as PPNode objects instead of text. This
4891 * allows for conditional expansion of the parse tree, allowing you to eliminate dead
4892 * branches and thus speed up parsing. It is also possible to analyse the parse tree of
4893 * the arguments, and to control the way they are expanded.
4894 *
4895 * The $frame parameter is a PPFrame. This can be used to produce expanded text from the
4896 * arguments, for instance:
4897 * $text = isset( $args[0] ) ? $frame->expand( $args[0] ) : '';
4898 *
4899 * For technical reasons, $args[0] is pre-expanded and will be a string. This may change in
4900 * future versions. Please call $frame->expand() on it anyway so that your code keeps
4901 * working if/when this is changed.
4902 *
4903 * If you want whitespace to be trimmed from $args, you need to do it yourself, post-
4904 * expansion.
4905 *
4906 * Please read the documentation in includes/parser/Preprocessor.php for more information
4907 * about the methods available in PPFrame and PPNode.
4908 *
4909 * @throws MWException
4910 * @return string|callback The old callback function for this name, if any
4911 */
4918 # Add to function cache
4922 }
4928 # Case
4931 }
4932 # Add leading hash
4935 }
4936 # Remove trailing colon
4939 }
4941 }
4943 }
4945 /**
4946 * Get all registered function hook identifiers
4947 *
4948 * @return Array
4949 */
4952 }
4954 /**
4955 * Create a tag function, e.g. "<test>some stuff</test>".
4956 * Unlike tag hooks, tag functions are parsed at preprocessor level.
4957 * Unlike parser functions, their content is not preprocessed.
4958 * @param $tag
4959 * @param $callback
4960 * @param $flags
4961 * @throws MWException
4962 * @return null
4963 */
4968 }
4975 }
4978 }
4980 /**
4981 * @todo FIXME: Update documentation. makeLinkObj() is deprecated.
4982 * Replace "<!--LINK-->" link placeholders with actual links, in the buffer
4983 * Placeholders created in Skin::makeLinkObj()
4984 *
4985 * @param $text string
4986 * @param $options int
4987 *
4988 * @return array of link CSS classes, indexed by PDBK.
4989 */
4992 }
4994 /**
4995 * Replace "<!--LINK-->" link placeholders with plain text of links
4996 * (not HTML-formatted).
4997 *
4998 * @param $text String
4999 * @return String
5000 */
5003 }
5005 /**
5006 * Renders an image gallery from a text with one line per image.
5007 * text labels may be given by using |-style alternative text. E.g.
5008 * Image:one.jpg|The number "1"
5009 * Image:tree.jpg|A tree
5010 * given as text will return the HTML of a gallery with two images,
5011 * labeled 'The number "1"' and
5012 * 'A tree'.
5013 *
5014 * @param string $text
5015 * @param array $params
5016 * @return string HTML
5017 */
5031 }
5037 }
5040 }
5043 }
5046 }
5052 # match lines like these:
5053 # Image:someimage.jpg|This is some image
5056 # Skip empty lines
5059 }
5063 }
5066 # Bogus title. Ignore these so we don't bomb out later.
5068 }
5074 // look for an |alt= definition while trying not to break existing
5075 // captions with multiple pipes (|) in it, until a more sensible grammar
5076 // is defined for images in galleries
5086 }
5091 //check to see if link matches an absolute url, if not then it must be a wiki link.
5098 }
5099 }
5100 }
5102 // concatenate all other pipes
5104 }
5105 }
5106 // remove the first pipe
5108 }
5111 }
5113 }
5115 /**
5116 * @param $handler
5117 * @return array
5118 */
5124 }
5126 # Initialise static lists
5133 );
5141 }
5142 }
5143 }
5145 # Add handler params
5151 }
5152 }
5155 }
5156 return array( $this->mImageParams[$handlerClass], $this->mImageParamsMagicArray[$handlerClass] );
5157 }
5159 /**
5160 * Parse image options text and use it to make an image
5161 *
5162 * @param $title Title
5163 * @param $options String
5164 * @param $holders LinkHolderArray|bool
5165 * @return string HTML
5166 */
5168 # Check if the options text is of the form "options|alt text"
5169 # Options are:
5170 # * thumbnail make a thumbnail with enlarge-icon and caption, alignment depends on lang
5171 # * left no resizing, just left align. label is used for alt= only
5172 # * right same, but right aligned
5173 # * none same, but not aligned
5174 # * ___px scale to ___ pixels width, no aligning. e.g. use in taxobox
5175 # * center center the image
5176 # * frame Keep original image size, no magnify-button.
5177 # * framed Same as "frame"
5178 # * frameless like 'thumb' but without a frame. Keeps user preferences for width
5179 # * upright reduce width for upright images, rounded to full __0 px
5180 # * border draw a 1px border around the image
5181 # * alt Text for HTML alt attribute (defaults to empty)
5182 # * class Set a class for img node
5183 # * link Set the target of the image link. Can be external, interwiki, or local
5184 # vertical-align values (no % or length right now):
5185 # * baseline
5186 # * sub
5187 # * super
5188 # * top
5189 # * text-top
5190 # * middle
5191 # * bottom
5192 # * text-bottom
5196 # Give extensions a chance to select the file revision for us
5201 # Fetch and register the file (file title may be different via hooks)
5204 # Get parameter map
5211 }
5213 # Process the input parameters
5224 # Special case; width and height come in one variable together
5232 }
5233 }
5239 }
5240 }
5241 # else no validation -- bug 13436
5244 # Validate handler parameter
5247 # Validate internal parameters
5252 # @todo FIXME: Possibly check validity here for
5253 # manualthumb? downstream behavior seems odd with
5254 # missing manual thumbs.
5271 }
5273 }
5281 }
5282 }
5285 # Most other things appear to be empty or numeric...
5287 }
5288 }
5292 }
5293 }
5294 }
5297 }
5298 }
5300 # Process alignment parameters
5303 }
5306 }
5310 # Will the image be presented in a frame, with the caption below?
5316 # In the old days, [[Image:Foo|text...]] would set alt text. Later it
5317 # came to also set the caption, ordinary text after the image -- which
5318 # makes no sense, because that just repeats the text multiple times in
5319 # screen readers. It *also* came to set the title attribute.
5320 #
5321 # Now that we have an alt attribute, we should not set the alt text to
5322 # equal the caption: that's worse than useless, it just repeats the
5323 # text. This is the framed/thumbnail case. If there's no caption, we
5324 # use the unnamed parameter for alt text as well, just for the time be-
5325 # ing, if the unnamed param is set and the alt param is not.
5326 #
5327 # For the future, we need to figure out if we want to tweak this more,
5328 # e.g., introducing a title= parameter for the title; ignoring the un-
5329 # named parameter entirely for images without a caption; adding an ex-
5330 # plicit caption= parameter and preserving the old magic unnamed para-
5331 # meter for BC; ...
5334 # No caption or alt text, add the filename as the alt text so
5335 # that screen readers at least get some description of the image
5337 }
5338 # Do not set $params['frame']['title'] because tooltips don't make sense
5339 # for framed images
5342 # No alt text, use the "caption" for the alt text
5346 # No caption, fall back to using the filename for the
5347 # alt text
5349 }
5350 }
5351 # Use the "caption" for the tooltip text
5353 }
5357 # Linker does the rest
5362 # Give the handler a chance to modify the parser object
5365 }
5368 }
5370 /**
5371 * @param $caption
5372 * @param $holders LinkHolderArray
5373 * @return mixed|String
5374 */
5376 # Strip bad stuff out of the title (tooltip). We can't just use
5377 # replaceLinkHoldersText() here, because if this function is called
5378 # from replaceInternalLinks2(), mLinkHolders won't be up-to-date.
5383 }
5385 # make sure there are no placeholders in thumbnail attributes
5386 # that are later expanded to html- so expand them now and
5387 # remove the tags
5392 }
5394 /**
5395 * Set a flag in the output object indicating that the content is dynamic and
5396 * shouldn't be cached.
5397 */
5403 }
5406 }
5408 /**
5409 * Callback from the Sanitizer for expanding items found in HTML attribute
5410 * values, so they can be safely tested and escaped.
5411 *
5412 * @param $text String
5413 * @param $frame PPFrame
5414 * @return String
5415 */
5420 }
5422 /**
5423 * Accessor
5424 *
5425 * @return array
5426 */
5428 return array_merge( array_keys( $this->mTransparentTagHooks ), array_keys( $this->mTagHooks ), array_keys( $this->mFunctionTagHooks ) );
5429 }
5431 /**
5432 * Replace transparent tags in $text with the values given by the callbacks.
5433 *
5434 * Transparent tag hooks are like regular XML-style tag hooks, except they
5435 * operate late in the transformation sequence, on HTML instead of wikitext.
5436 *
5437 * @param $text string
5438 *
5439 * @return string
5440 */
5451 $output = call_user_func_array( $this->mTransparentTagHooks[$tagName], array( $content, $params, $this ) );
5454 }
5456 }
5458 }
5460 /**
5461 * Break wikitext input into sections, and either pull or replace
5462 * some particular section's text.
5463 *
5464 * External callers should use the getSection and replaceSection methods.
5465 *
5466 * @param string $text Page wikitext
5467 * @param string $section a section identifier string of the form:
5468 * "<flag1> - <flag2> - ... - <section number>"
5469 *
5470 * Currently the only recognised flag is "T", which means the target section number
5471 * was derived during a template inclusion parse, in other words this is a template
5472 * section edit link. If no flags are given, it was an ordinary section edit link.
5473 * This flag is required to avoid a section numbering mismatch when a section is
5474 * enclosed by "<includeonly>" (bug 6563).
5475 *
5476 * The section number 0 pulls the text before the first heading; other numbers will
5477 * pull the given section along with its lower-level subsections. If the section is
5478 * not found, $mode=get will return $newtext, and $mode=replace will return $text.
5479 *
5480 * Section 0 is always considered to exist, even if it only contains the empty
5481 * string. If $text is the empty string and section 0 is replaced, $newText is
5482 * returned.
5483 *
5484 * @param string $mode one of "get" or "replace"
5485 * @param string $newText replacement text for section data.
5486 * @return String: for "get", the extracted section text.
5487 * for "replace", the whole page with the section replaced.
5488 */
5495 # Process section extraction flags
5502 }
5503 }
5505 # Check for empty input
5507 # Only sections 0 and T-0 exist in an empty document
5513 }
5519 }
5520 }
5521 }
5523 # Preprocess the text
5526 # <h> nodes indicate section breaks
5527 # They can only occur at the top level, so we can find them by iterating the root's children
5530 # Find the target section
5532 # Section zero doesn't nest, level=big
5541 }
5542 }
5545 }
5547 }
5548 }
5551 # Not found
5556 }
5557 }
5559 # Find the end of the section, including nested sections
5566 }
5567 }
5570 }
5574 # Write out the remainder (in replace mode only)
5576 # Output the replacement text
5577 # Add two newlines on -- trailing whitespace in $newText is conventionally
5578 # stripped by the editor, so we need both newlines to restore the paragraph gap
5579 # Only add trailing whitespace if there is newText
5582 }
5587 }
5588 }
5591 # Re-insert stripped tags
5593 }
5596 }
5598 /**
5599 * This function returns the text of a section, specified by a number ($section).
5600 * A section is text under a heading like == Heading == or \<h1\>Heading\</h1\>, or
5601 * the first section before any such heading (section 0).
5602 *
5603 * If a section contains subsections, these are also returned.
5604 *
5605 * @param string $text text to look in
5606 * @param string $section section identifier
5607 * @param string $deftext default to return if section is not found
5608 * @return string text of the requested section
5609 */
5612 }
5614 /**
5615 * This function returns $oldtext after the content of the section
5616 * specified by $section has been replaced with $text. If the target
5617 * section does not exist, $oldtext is returned unchanged.
5618 *
5619 * @param string $oldtext former text of the article
5620 * @param int $section section identifier
5621 * @param string $text replacing text
5622 * @return String: modified text
5623 */
5626 }
5628 /**
5629 * Get the ID of the revision we are parsing
5630 *
5631 * @return Mixed: integer or null
5632 */
5635 }
5637 /**
5638 * Get the revision object for $this->mRevisionId
5639 *
5640 * @return Revision|null either a Revision object or null
5641 */
5645 }
5648 }
5652 }
5654 /**
5655 * Get the timestamp associated with the current revision, adjusted for
5656 * the default server-local timestamp
5657 */
5667 # The cryptic '' timezone parameter tells to use the site-default
5668 # timezone offset instead of the user settings.
5669 #
5670 # Since this value will be saved into the parser cache, served
5671 # to other users, and potentially even used inside links and such,
5672 # it needs to be consistent for all visitors.
5676 }
5678 }
5680 /**
5681 * Get the name of the user that edited the last revision
5682 *
5683 * @return String: user name
5684 */
5689 # if this template is subst: the revision id will be blank,
5690 # so just use the current user's name
5695 }
5696 }
5698 }
5700 /**
5701 * Mutator for $mDefaultSort
5702 *
5703 * @param string $sort New value
5704 */
5708 }
5710 /**
5711 * Accessor for $mDefaultSort
5712 * Will use the empty string if none is set.
5713 *
5714 * This value is treated as a prefix, so the
5715 * empty string is equivalent to sorting by
5716 * page name.
5717 *
5718 * @return string
5719 */
5725 }
5726 }
5728 /**
5729 * Accessor for $mDefaultSort
5730 * Unlike getDefaultSort(), will return false if none is set
5731 *
5732 * @return string or false
5733 */
5736 }
5738 /**
5739 * Try to guess the section anchor name based on a wikitext fragment
5740 * presumably extracted from a heading, for example "Header" from
5741 * "== Header ==".
5742 *
5743 * @param $text string
5744 *
5745 * @return string
5746 */
5748 # Strip out wikitext links(they break the anchor)
5752 }
5754 /**
5755 * Same as guessSectionNameFromWikiText(), but produces legacy anchors
5756 * instead. For use in redirects, since IE6 interprets Redirect: headers
5757 * as something other than UTF-8 (apparently?), resulting in breakage.
5758 *
5759 * @param string $text The section name
5760 * @return string An anchor
5761 */
5763 # Strip out wikitext links(they break the anchor)
5767 }
5769 /**
5770 * Strips a text string of wikitext for use in a section anchor
5771 *
5772 * Accepts a text string and then removes all wikitext from the
5773 * string and leaves only the resultant text (i.e. the result of
5774 * [[User:WikiSysop|Sysop]] would be "Sysop" and the result of
5775 * [[User:WikiSysop]] would be "User:WikiSysop") - this is intended
5776 * to create valid section anchors by mimicing the output of the
5777 * parser when headings are parsed.
5778 *
5779 * @param string $text text string to be stripped of wikitext
5780 * for use in a Section anchor
5781 * @return string Filtered text string
5782 */
5784 # Strip internal link markup
5788 # Strip external link markup
5789 # @todo FIXME: Not tolerant to blank link text
5790 # I.E. [http://www.mediawiki.org] will render as [1] or something depending
5791 # on how many empty links there are on the page - need to figure that out.
5792 $text = preg_replace( '/\[(?i:' . $this->mUrlProtocols . ')([^ ]+?) ([^[]+)\]/', '$2', $text );
5794 # Parse wikitext quotes (italics & bold)
5797 # Strip HTML tags
5800 }
5802 /**
5803 * strip/replaceVariables/unstrip for preprocessor regression testing
5804 *
5805 * @param $text string
5806 * @param $title Title
5807 * @param $options ParserOptions
5808 * @param $outputType int
5809 *
5810 * @return string
5811 */
5812 function testSrvus( $text, Title $title, ParserOptions $options, $outputType = self::OT_HTML ) {
5819 }
5821 /**
5822 * @param $text string
5823 * @param $title Title
5824 * @param $options ParserOptions
5825 * @return string
5826 */
5829 }
5831 /**
5832 * @param $text
5833 * @param $title Title
5834 * @param $options ParserOptions
5835 * @return string
5836 */
5839 }
5841 /**
5842 * Call a callback function on all regions of the given text that are not
5843 * inside strip markers, and replace those regions with the return value
5844 * of the callback. For example, with input:
5845 *
5846 * aaa<MARKER>bbb
5847 *
5848 * This will call the callback function twice, with 'aaa' and 'bbb'. Those
5849 * two strings will be replaced with the value returned by the callback in
5850 * each case.
5851 *
5852 * @param $s string
5853 * @param $callback
5854 *
5855 * @return string
5856 */
5875 }
5876 }
5877 }
5879 }
5881 /**
5882 * Remove any strip markers found in the given text.
5883 *
5884 * @param $text Input string
5885 * @return string
5886 */
5889 }
5891 /**
5892 * Save the parser state required to convert the given half-parsed text to
5893 * HTML. "Half-parsed" in this context means the output of
5894 * recursiveTagParse() or internalParse(). This output has strip markers
5895 * from replaceVariables (extensionSubstitution() etc.), and link
5896 * placeholders from replaceLinkHolders().
5897 *
5898 * Returns an array which can be serialized and stored persistently. This
5899 * array can later be loaded into another parser instance with
5900 * unserializeHalfParsedText(). The text can then be safely incorporated into
5901 * the return value of a parser hook.
5902 *
5903 * @param $text string
5904 *
5905 * @return array
5906 */
5914 );
5917 }
5919 /**
5920 * Load the parser state given in the $data array, which is assumed to
5921 * have been generated by serializeHalfParsedText(). The text contents is
5922 * extracted from the array, and its markers are transformed into markers
5923 * appropriate for the current Parser instance. This transformed text is
5924 * returned, and can be safely included in the return value of a parser
5925 * hook.
5926 *
5927 * If the $data array has been stored persistently, the caller should first
5928 * check whether it is still valid, by calling isValidHalfParsedText().
5929 *
5930 * @param array $data Serialized data
5931 * @throws MWException
5932 * @return String
5933 */
5937 }
5939 # First, extract the strip state.
5943 # Now renumber links
5946 # Should be good to go.
5948 }
5950 /**
5951 * Returns true if the given array, presumed to be generated by
5952 * serializeHalfParsedText(), is compatible with the current version of the
5953 * parser.
5954 *
5955 * @param $data Array
5956 *
5957 * @return bool
5958 */
5961 }
5963 /**
5964 * Parsed a width param of imagelink like 300px or 200x300px
5965 *
5966 * @param $value String
5967 *
5968 * @return array
5969 * @since 1.20
5970 */
5975 }
5977 # (bug 13500) In both cases (width/height and width only),
5978 # permit trailing "px" for backward compatibility.
5987 }
5989 }
5990 }