| blob | 66cfd550bf45255f640afb1d7bdddfd8e9f10a0f |
1 <?php
2 /**
3 * PHP parser that converts wiki markup to HTML.
4 *
5 * This program is free software; you can redistribute it and/or modify
6 * it under the terms of the GNU General Public License as published by
7 * the Free Software Foundation; either version 2 of the License, or
8 * (at your option) any later version.
9 *
10 * This program is distributed in the hope that it will be useful,
11 * but WITHOUT ANY WARRANTY; without even the implied warranty of
12 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
13 * GNU General Public License for more details.
14 *
15 * You should have received a copy of the GNU General Public License along
16 * with this program; if not, write to the Free Software Foundation, Inc.,
17 * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
18 * http://www.gnu.org/copyleft/gpl.html
19 *
20 * @file
21 * @ingroup Parser
22 */
24 /**
25 * @defgroup Parser Parser
26 */
28 /**
29 * PHP Parser - Processes wiki markup (which uses a more user-friendly
30 * syntax, such as "[[link]]" for making links), and provides a one-way
31 * transformation of that wiki markup it into (X)HTML output / markup
32 * (which in turn the browser understands, and can display).
33 *
34 * There are seven main entry points into the Parser class:
35 *
36 * - Parser::parse()
37 * produces HTML output
38 * - Parser::preSaveTransform().
39 * produces altered wiki markup.
40 * - Parser::preprocess()
41 * removes HTML comments and expands templates
42 * - Parser::cleanSig() and Parser::cleanSigInSig()
43 * Cleans a signature before saving it to preferences
44 * - Parser::getSection()
45 * Return the content of a section from an article for section editing
46 * - Parser::replaceSection()
47 * Replaces a section by number inside an article
48 * - Parser::getPreloadText()
49 * Removes <noinclude> sections, and <includeonly> tags.
50 *
51 * Globals used:
52 * object: $wgContLang
53 *
54 * @warning $wgUser or $wgTitle or $wgRequest or $wgLang. Keep them away!
55 *
56 * @par Settings:
57 * $wgNamespacesWithSubpages
58 *
59 * @par Settings only within ParserOptions:
60 * $wgAllowExternalImages
61 * $wgAllowSpecialInclusion
62 * $wgInterwikiMagic
63 * $wgMaxArticleSize
64 *
65 * @ingroup Parser
66 */
68 /**
69 * Update this version number when the ParserOutput format
70 * changes in an incompatible way, so the parser cache
71 * can automatically discard old data.
72 */
75 /**
76 * Update this version number when the output of serialiseHalfParsedText()
77 * changes in an incompatible way
78 */
81 # Flags for Parser::setFunctionHook
82 # Also available as global constants from Defines.php
86 # Constants needed for external link processing
87 # Everything except bracket, space, or control characters
88 # \p{Zs} is unicode 'separator, space' category. It covers the space 0x20
89 # as well as U+3000 is IDEOGRAPHIC SPACE for bug 19052
94 # State constants for the definition list colon extraction
104 # Flags for preprocessToDom
107 # Allowed values for $this->mOutputType
108 # Parameter to startExternalParse().
115 # Marker Suffix needs to be accessible staticly.
118 # Markers used for wrapping the table of contents
122 # Persistent:
136 # Initialised by initialiseVariables()
138 /**
139 * @var MagicWordArray
140 */
143 /**
144 * @var MagicWordArray
145 */
147 var $mConf, $mPreprocessor, $mExtLinkBracketedRegex, $mUrlProtocols; # Initialised in constructor
149 # Cleared with clearState():
150 /**
151 * @var ParserOutput
152 */
156 /**
157 * @var StripState
158 */
162 /**
163 * @var LinkHolderArray
164 */
174 /**
175 * @var User
176 */
179 # Temporary
180 # These are variables reset at least once per parse regardless of $clearState
182 /**
183 * @var ParserOptions
184 */
187 /**
188 * @var Title
189 */
201 /**
202 * @var string
203 */
206 /**
207 * @var array Array with the language name of each language link (i.e. the
208 * interwiki prefix) in the key, value arbitrary. Used to avoid sending
209 * duplicate language links to the ParserOutput.
210 */
213 /**
214 * @var boolean Recursive call protection.
215 * This variable should be treated as if it were private.
216 */
219 /**
220 * @param array $conf
221 */
230 # Preprocessor_Hash is much faster than Preprocessor_DOM under HipHop
233 # PECL extension that conflicts with the core DOM extension (bug 13770)
240 }
242 }
244 /**
245 * Reduce memory usage to reduce the impact of circular references
246 */
250 }
253 }
254 }
256 /**
257 * Allow extensions to clean up when the parser is cloned
258 */
262 }
264 /**
265 * Do various kinds of initialisation on the first call of the parser
266 */
270 }
281 }
283 /**
284 * Clear Parser state
285 *
286 * @private
287 */
292 }
309 /**
310 * Prefix for temporary replacement strings for the multipass parser.
311 * \x07 should never appear in input as it's disallowed in XML.
312 * Using it at the front also gives us a little extra robustness
313 * since it shouldn't match when butted up against identifier-like
314 * string constructs.
315 *
316 * Must not consist of all title characters, or else it will change
317 * the behavior of <nowiki> in a link.
318 */
322 # Clear these on every parse, bug 4549
330 );
339 # Fix cloning
342 }
346 }
348 /**
349 * Convert wikitext to HTML
350 * Do not call this function recursively.
351 *
352 * @param string $text text we want to parse
353 * @param Title $title
354 * @param ParserOptions $options
355 * @param bool $linestart
356 * @param bool $clearState
357 * @param int $revid Number to pass in {{REVISIONID}}
358 * @return ParserOutput A ParserOutput
359 */
362 ) {
363 /**
364 * First pass--just handle <nowiki> sections, pass the rest off
365 * to internalParse() which does all the real work.
366 */
375 }
382 }
384 # Remove the strip marker tag prefix from the input, if present.
387 }
400 }
403 # No more strip!
410 # Clean up special characters, only run once, next-to-last before doBlockLevels
412 # french spaces, last one Guillemet-left
413 # only if there is something before the space
415 # french spaces, Guillemet-right
418 );
425 /**
426 * The input doesn't get language converted if
427 * a) It's disabled
428 * b) Content isn't converted
429 * c) It's a conversion table
430 * d) it is an interface message (which is in the user language)
431 */
434 ) {
436 # The position of the convert() call should not be changed. it
437 # assumes that the links are all replaced and the only thing left
438 # is the <nowiki> mark.
440 }
441 }
443 /**
444 * A converted title will be provided in the output object if title and
445 * content conversion are enabled, the article text does not contain
446 * a conversion-suppressing double-underscore tag, and no
447 * {{DISPLAYTITLE:...}} is present. DISPLAYTITLE takes precedence over
448 * automatic link conversion.
449 */
454 ) {
461 }
462 }
476 # attempt to sanitize at least some nesting problems
477 # (bug #2702 and quite a few others)
479 # ''Something [http://www.cool.com cool''] -->
480 # <i>Something</i><a href="http://www.cool.com"..><i>cool></i></a>
483 # fix up an anchor inside another anchor, only
484 # at least for a single single nested link (bug 3695)
487 # fix div inside inline elements- doBlockLevels won't wrap a line which
488 # contains a div, so fix it up here; replace
489 # div with escaped text
492 # remove empty italic or bold tag pairs, some
493 # introduced by rules above
495 );
501 }
507 );
508 }
512 # Information on include size limits, for the benefit of users who try to skirt them
520 );
521 }
526 );
530 );
533 );
536 );
539 );
542 );
545 );
551 }
555 ) ) {
561 }
565 }
566 }
567 }
568 // Since we're not really outputting HTML, decode the entities and
569 // then re-encode the things that need hiding inside HTML comments.
573 // Sanitize for comment. Note '‐' in the replacement is U+2010,
574 // which looks much like the problematic '-'.
581 }
582 }
595 }
597 /**
598 * Recursive parser entry point that can be called from an extension tag
599 * hook.
600 *
601 * If $frame is not provided, then template variables (e.g., {{{1}}}) within $text are not expanded
602 *
603 * @param string $text Text extension wants to have parsed
604 * @param bool|PPFrame $frame The frame to use for expanding any template variables
605 *
606 * @return string
607 */
615 }
617 /**
618 * Expand templates and variables in the text, producing valid, static wikitext.
619 * Also removes comments.
620 * Do not call this function recursively.
621 * @param string $text
622 * @param Title $title
623 * @param ParserOptions $options
624 * @param int|null $revid
625 * @param bool|PPFrame $frame
626 * @return mixed|string
627 */
628 function preprocess( $text, Title $title = null, ParserOptions $options, $revid = null, $frame = false ) {
634 }
641 }
643 /**
644 * Recursive parser entry point that can be called from an extension tag
645 * hook.
646 *
647 * @param string $text Text to be expanded
648 * @param bool|PPFrame $frame The frame to use for expanding any template variables
649 * @return string
650 * @since 1.19
651 */
658 }
660 /**
661 * Process the wikitext for the "?preload=" feature. (bug 5210)
662 *
663 * "<noinclude>", "<includeonly>" etc. are parsed as for template
664 * transclusion, comments, templates, arguments, tags hooks and parser
665 * functions are untouched.
666 *
667 * @param string $text
668 * @param Title $title
669 * @param ParserOptions $options
670 * @param array $params
671 * @return string
672 */
673 public function getPreloadText( $text, Title $title, ParserOptions $options, $params = array() ) {
677 # Parser (re)initialisation
686 }
688 /**
689 * Get a random string
690 *
691 * @return string
692 */
695 }
697 /**
698 * Set the current user.
699 * Should only be used when doing pre-save transform.
700 *
701 * @param User|null $user User object or null (to reset)
702 */
705 }
707 /**
708 * Accessor for mUniqPrefix.
709 *
710 * @return string
711 */
714 # @todo FIXME: This is probably *horribly wrong*
715 # LanguageConverter seems to want $wgParser's uniqPrefix, however
716 # if this is called for a parser cache hit, the parser may not
717 # have ever been initialized in the first place.
718 # Not really sure what the heck is supposed to be going on here.
720 # throw new MWException( "Accessing uninitialized mUniqPrefix" );
721 }
723 }
725 /**
726 * Set the context title
727 *
728 * @param Title $t
729 */
733 }
736 # Strip the fragment to avoid various odd effects
741 }
742 }
744 /**
745 * Accessor for the Title object
746 *
747 * @return Title
748 */
751 }
753 /**
754 * Accessor/mutator for the Title object
755 *
756 * @param Title $x Title object or null to just get the current one
757 * @return Title
758 */
761 }
763 /**
764 * Set the output type
765 *
766 * @param int $ot New value
767 */
770 # Shortcut alias
776 );
777 }
779 /**
780 * Accessor/mutator for the output type
781 *
782 * @param int|null $x New value or null to just get the current one
783 * @return int
784 */
787 }
789 /**
790 * Get the ParserOutput object
791 *
792 * @return ParserOutput
793 */
796 }
798 /**
799 * Get the ParserOptions object
800 *
801 * @return ParserOptions object
802 */
805 }
807 /**
808 * Accessor/mutator for the ParserOptions object
809 *
810 * @param ParserOptions $x New value or null to just get the current one
811 * @return ParserOptions Current ParserOptions object
812 */
815 }
817 /**
818 * @return int
819 */
822 }
824 /**
825 * @param int $id
826 */
829 }
831 /**
832 * Get a language object for use in parser functions such as {{FORMATNUM:}}
833 * @return Language
834 */
837 }
839 /**
840 * Get the target language for the content being parsed. This is usually the
841 * language that the content is in.
842 *
843 * @since 1.19
844 *
845 * @throws MWException
846 * @return Language|null
847 */
857 }
860 }
862 /**
863 * Get the language object for language conversion
864 * @return Language|null
865 */
868 }
870 /**
871 * Get a User object either from $this->mUser, if set, or from the
872 * ParserOptions object otherwise
873 *
874 * @return User
875 */
879 }
881 }
883 /**
884 * Get a preprocessor object
885 *
886 * @return Preprocessor
887 */
892 }
894 }
896 /**
897 * Replaces all occurrences of HTML-style comments and the given tags
898 * in the text with a random marker and returns the next text. The output
899 * parameter $matches will be an associative array filled with data in
900 * the form:
901 *
902 * @code
903 * 'UNIQ-xxxxx' => array(
904 * 'element',
905 * 'tag content',
906 * array( 'param' => 'x' ),
907 * '<element param="x">tag content</element>' ) )
908 * @endcode
909 *
910 * @param array $elements List of element names. Comments are always extracted.
911 * @param string $text Source text string.
912 * @param array $matches Out parameter, Array: extracted tags
913 * @param string $uniq_prefix
914 * @return string Stripped text
915 */
929 }
931 # comment
937 # tag
942 }
948 # Empty element tag, <tag />
957 }
961 # No end tag -- let it run out to the end of the text.
967 }
968 }
974 }
976 }
978 /**
979 * Get a list of strippable XML-like elements
980 *
981 * @return array
982 */
985 }
987 /**
988 * Add an item to the strip state
989 * Returns the unique tag which must be inserted into the stripped text
990 * The tag will be replaced with the original text in unstrip()
991 *
992 * @param string $text
993 *
994 * @return string
995 */
1001 }
1003 /**
1004 * parse the wiki syntax used to render tables
1005 *
1006 * @private
1007 * @param string $text
1008 * @return string
1009 */
1028 }
1034 # First check if we are starting a new table
1047 # Don't do any of the following
1051 # We are ending a table
1057 }
1061 }
1065 }
1069 # Now we have a table row
1072 # Whats after the tag is now only attributes
1085 }
1089 }
1098 ) {
1099 # This might be cell elements, td, th or captions
1103 }
1109 }
1111 # Split up multiple cells on the same line.
1112 # FIXME : This can result in improper nesting of tags processed
1113 # by earlier parser steps, but should avoid splitting up eg
1114 # attribute values containing literal "||".
1119 # Loop through each table cell
1126 }
1131 }
1137 }
1147 }
1151 # A cell could contain both parameters and data
1154 # Bug 553: Note that a '|' inside an invalid link should not
1155 # be mistaken as delimiting cell parameters
1164 }
1168 }
1169 }
1171 }
1173 # Closing open td, tr && table
1177 }
1180 }
1183 }
1186 }
1188 # Remove trailing line-ending (b/c)
1191 }
1193 # special case: don't return empty table
1196 }
1201 }
1203 /**
1204 * Helper function for parse() that transforms wiki markup into
1205 * HTML. Only called for $mOutputType == self::OT_HTML.
1206 *
1207 * @private
1208 *
1209 * @param string $text
1210 * @param bool $isMain
1211 * @param bool $frame
1212 *
1213 * @return string
1214 */
1220 # Hook to suspend the parser in this state
1221 if ( !wfRunHooks( 'ParserBeforeInternalParse', array( &$this, &$text, &$this->mStripState ) ) ) {
1224 }
1226 # if $frame is provided, then use $frame for replacing any variables
1228 # use frame depth to infer how include/noinclude tags should be handled
1229 # depth=0 means this is the top-level document; otherwise it's an included document
1234 }
1238 # if $frame is not provided, then use old-style replaceVariables
1240 }
1248 );
1251 # Tables need to come after variable replacement for things to work
1252 # properly; putting them before other transformations should keep
1253 # exciting things like link expansions from showing up in surprising
1254 # places.
1266 # replaceInternalLinks may sometimes leave behind
1267 # absolute URLs, which have to be masked to hide them from replaceExternalLinks
1275 }
1277 /**
1278 * Replace special strings like "ISBN xxx" and "RFC xxx" with
1279 * magic external links.
1280 *
1281 * DML
1282 * @private
1283 *
1284 * @param string $text
1285 *
1286 * @return string
1287 */
1293 '!(?: # Start cases
1297 (?:RFC|PMID)\s+([0-9]+) | # m[4]: RFC or PMID, capture number
1299 (?: 97[89] [\ \-]? )? # optional 13-digit ISBN prefix
1300 (?: [0-9] [\ \-]? ){9} # 9 digits with opt. delimiters
1301 [0-9Xx] # check digit
1306 }
1308 /**
1309 * @throws MWException
1310 * @param array $m
1311 * @return HTML|string
1312 */
1315 # Skip anchor
1318 # Skip HTML element
1321 # Free external link
1324 # RFC or PMID
1338 }
1342 # ISBN
1348 ));
1355 }
1356 }
1358 /**
1359 * Make a free external link, given a user-supplied URL
1360 *
1361 * @param string $url
1362 *
1363 * @return string HTML
1364 * @private
1365 */
1371 # The characters '<' and '>' (which were escaped by
1372 # removeHTMLtags()) should not be included in
1373 # URLs, per RFC 2396.
1378 }
1380 # Move trailing punctuation to $trail
1382 # If there is no left bracket, then consider right brackets fair game too
1385 }
1391 }
1395 # Is this an external image?
1398 # Not an image, make a link
1403 # Register it in the output object...
1404 # Replace unnecessary URL escape codes with their equivalent characters
1407 }
1410 }
1412 /**
1413 * Parse headers and return html
1414 *
1415 * @private
1416 *
1417 * @param string $text
1418 *
1419 * @return string
1420 */
1426 }
1429 }
1431 /**
1432 * Replace single quotes with HTML markup
1433 * @private
1434 *
1435 * @param string $text
1436 *
1437 * @return string the altered text
1438 */
1445 }
1449 }
1451 /**
1452 * Helper function for doAllQuotes()
1453 *
1454 * @param string $text
1455 *
1456 * @return string
1457 */
1463 }
1465 // First, do some preliminary work. This may shift some apostrophes from
1466 // being mark-up to being text. It also counts the number of occurrences
1467 // of bold and italics mark-ups.
1472 // If there are ever four apostrophes, assume the first is supposed to
1473 // be text, and the remaining three constitute mark-up for bold text.
1474 // (bug 13227: ''''foo'''' turns into ' ''' foo ' ''')
1480 // If there are more than 5 apostrophes in a row, assume they're all
1481 // text except for the last 5.
1482 // (bug 13227: ''''''foo'''''' turns into ' ''''' foo ' ''''')
1486 }
1487 // Count the number of occurrences of bold and italics mark-ups.
1495 }
1496 }
1498 // If there is an odd number of both bold and italics, it is likely
1499 // that one of the bold ones was meant to be an apostrophe followed
1500 // by italics. Which one we cannot know for certain, but it is more
1501 // likely to be one that has a single-letter word before it.
1513 }
1517 // if $firstsingleletterword is set, we don't
1518 // look at the other options, so we can bail early.
1520 }
1524 }
1525 }
1526 }
1527 }
1529 // If there is a single-letter word, use it!
1534 // If not, but there's a multi-letter word, use that one.
1538 // ... otherwise use the first one that has neither.
1539 // (notice that it is possible for all three to be -1 if, for example,
1540 // there is only one pentuple-apostrophe in the line)
1543 }
1544 }
1546 // Now let's actually convert our apostrophic mush to HTML!
1557 }
1576 }
1593 }
1613 }
1614 }
1615 }
1617 }
1618 // Now close all remaining tags. Notice that the order is important.
1621 }
1624 }
1627 }
1628 // There might be lonely ''''', so make sure we have a buffer
1631 }
1633 }
1635 /**
1636 * Replace external links (REL)
1637 *
1638 * Note: this is all very hackish and the order of execution matters a lot.
1639 * Make sure to run tests/parserTests.php if you change this code.
1640 *
1641 * @private
1642 *
1643 * @param string $text
1644 *
1645 * @throws MWException
1646 * @return string
1647 */
1656 }
1666 # The characters '<' and '>' (which were escaped by
1667 # removeHTMLtags()) should not be included in
1668 # URLs, per RFC 2396.
1673 }
1675 # If the link text is an image URL, replace it with an <img> tag
1676 # This happened by accident in the original parser, but some people used it extensively
1680 }
1684 # Set linktype for CSS - if URL==text, link is essentially free
1687 # No link text, e.g. [http://domain.tld/some.link]
1689 # Autonumber
1694 # Have link text, e.g. [http://domain.tld/some.link text]s
1695 # Check for trail
1697 }
1703 # Use the encoded URL
1704 # This means that users can paste URLs directly into the text
1705 # Funny characters like ö aren't valid in URLs anyway
1706 # This was changed in August 2004
1710 # Register link in the output object.
1711 # Replace unnecessary URL escape codes with the referenced character
1712 # This prevents spammers from hiding links from the filters
1715 }
1719 }
1721 /**
1722 * Get the rel attribute for a particular external link.
1723 *
1724 * @since 1.21
1725 * @param string|bool $url Optional URL, to extract the domain from for rel =>
1726 * nofollow if appropriate
1727 * @param Title $title Optional Title, for wgNoFollowNsExceptions lookups
1728 * @return string|null Rel attribute for $url
1729 */
1735 ) {
1737 }
1739 }
1741 /**
1742 * Get an associative array of additional HTML attributes appropriate for a
1743 * particular external link. This currently may include rel => nofollow
1744 * (depending on configuration, namespace, and the URL's domain) and/or a
1745 * target attribute (depending on configuration).
1746 *
1747 * @param string|bool $url Optional URL, to extract the domain from for rel =>
1748 * nofollow if appropriate
1749 * @return array Associative array of HTML attributes
1750 */
1757 }
1759 }
1761 /**
1762 * Replace unusual URL escape codes with their equivalent characters
1763 *
1764 * @param string $url
1765 * @return string
1766 *
1767 * @todo This can merge genuinely required bits in the path or query string,
1768 * breaking legit URLs. A proper fix would treat the various parts of
1769 * the URL differently; as a workaround, just use the output for
1770 * statistical records, not for actual linking/output.
1771 */
1775 }
1777 /**
1778 * Callback function used in replaceUnusualEscapes().
1779 * Replaces unusual URL escape codes with their equivalent character
1780 *
1781 * @param array $matches
1782 *
1783 * @return string
1784 */
1788 # Is it an unsafe or HTTP reserved character according to RFC 1738?
1790 # No, shouldn't be escaped
1793 # Yes, leave it escaped
1795 }
1796 }
1798 /**
1799 * make an image if it's allowed, either through the global
1800 * option, through the exception, or through the on-wiki whitelist
1801 *
1802 * @param string $url
1803 *
1804 * @return string
1805 */
1810 # $imagesfrom could be either a single string or an array of strings, parse out the latter
1817 }
1818 }
1823 }
1827 ) {
1829 # Image found
1831 }
1832 }
1835 ) {
1839 );
1842 # Sanitize the regex fragment, make it case-insensitive, ignore blank entries/comments
1845 }
1847 # Image matches a whitelist entry
1850 }
1851 }
1852 }
1854 }
1856 /**
1857 * Process [[ ]] wikilinks
1858 *
1859 * @param string $s
1860 *
1861 * @return string Processed text
1862 *
1863 * @private
1864 */
1868 }
1870 /**
1871 * Process [[ ]] wikilinks (RIL)
1872 * @param string $s
1873 * @throws MWException
1874 * @return LinkHolderArray
1875 *
1876 * @private
1877 */
1883 # the % is needed to support urlencoded titles as well
1886 # Match a link having the form [[namespace:link|alternate]]trail
1888 # Match cases where there is no "]]", which might still be images
1890 }
1894 # split the entire text string on occurrences of [[
1896 # get the first element (all text up to first [[), and remove the space we added
1905 # Match the end of a line for a word that's not followed by whitespace,
1906 # e.g. in the case of 'The Arab al[[Razi]]', 'al' will be matched
1910 }
1916 }
1925 }
1928 }
1933 // @codingStandardsIgnoreStart Squiz.WhiteSpace.SemicolonSpacing.Incorrect
1934 # Loop for each link
1936 // @codingStandardsIgnoreStart
1938 # Check for excessive memory usage
1940 # Too big
1941 # Do the existence check, replace the link holders and clear the array
1944 }
1953 }
1954 # first link
1958 }
1960 }
1967 # If we get a ] at the beginning of $m[3] that means we have a link that's something like:
1968 # [[Image:Foo.jpg|[http://example.com desc]]] <- having three ] in a row fucks up,
1969 # the real problem is with the $e1 regex
1970 # See bug 1300.
1971 #
1972 # Still some problems for cases where the ] is meant to be outside punctuation,
1973 # and no image is in sight. See bug 2095.
1974 #
1978 ) {
1981 }
1982 # fix up urlencoded title texts
1984 # Should anchors '#' also be rejected?
1986 }
1989 # Invalid, but might be an image with a link in its caption
1994 }
2000 }
2004 # Don't allow internal links to pages containing
2005 # PROTO: where PROTO is a valid URL protocol; these
2006 # should be external links.
2011 }
2013 # Make subpage if necessary
2018 }
2022 # Strip off leading ':'
2024 }
2033 }
2044 # look at the next 'line' to see if we can close it there
2049 }
2052 # the first ]] closes the inner link, the second the image
2058 # if there's exactly one ]] that's fine, we'll keep looking
2061 # if $next_line is invalid too, we need look no further
2064 }
2065 }
2067 # we couldn't find the end of this imageLink, so output it raw
2068 # but don't ignore what might be perfectly normal links in the text we've examined
2071 # note: no $trail, because without an end, there *is* no trail
2074 }
2077 # note: no $trail, because without an end, there *is* no trail
2080 }
2082 }
2088 # Bug 4598 madness. Handle the quotes only if they come from the alternate part
2089 # [[Lista d''e paise d''o munno]] -> <a href="...">Lista d''e paise d''o munno</a>
2090 # [[Criticism of Harry Potter|Criticism of ''Harry Potter'']]
2091 # -> <a href="Criticism of Harry Potter">Criticism of <i>Harry Potter</i></a>
2093 }
2095 # Link not escaped by : , create the various objects
2097 # Interwikis
2101 ) {
2102 // XXX: the above check prevents links to sites with identifiers that are not language codes
2104 # Bug 24502: filter duplicates
2108 }
2114 }
2121 # if no parameters were passed, $text
2122 # becomes something like "File:Foo.png",
2123 # which we don't want to pass on to the
2124 # image generator
2127 # recursively parse links inside the image caption
2128 # actually, this will parse them in any other parameters, too,
2129 # but it might be hard to fix that, and it doesn't matter ATM
2132 }
2133 # cloak any absolute URLs inside the image markup, so replaceExternalLinks() won't touch them
2138 }
2141 }
2151 }
2157 /**
2158 * Strip the whitespace Category links produce, see bug 87
2159 */
2164 }
2165 }
2167 # Self-link checking. For some languages, variants of the title are checked in
2168 # LinkHolderArray::doVariants() to allow batching the existence checks necessary
2169 # for linking to a different variant.
2173 }
2175 # NS_MEDIA is a pseudo-namespace for linking directly to a file
2176 # @todo FIXME: Should do batch file existence checks, see comment below
2179 # Give extensions a chance to select the file revision for us
2184 # Fetch and register the file (file title may be different via hooks)
2186 # Cloak with NOPARSE to avoid replacement in replaceExternalLinks
2191 }
2194 # Some titles, such as valid special pages or files in foreign repos, should
2195 # be shown as bluelinks even though they're not included in the page table
2196 #
2197 # @todo FIXME: isAlwaysKnown() can be expensive for file links; we should really do
2198 # batch file existence checks for NS_FILE and NS_MEDIA
2203 # Links will be added to the output link list after checking
2205 }
2207 }
2210 }
2212 /**
2213 * Render a forced-blue link inline; protect against double expansion of
2214 * URLs if we're in a mode that prepends full URL prefixes to internal links.
2215 * Since this little disaster has to split off the trail text to avoid
2216 * breaking URLs in the following text without breaking trails on the
2217 * wiki links, it's been made into a horrible function.
2218 *
2219 * @param Title $nt
2220 * @param string $text
2221 * @param array|string $query
2222 * @param string $trail
2223 * @param string $prefix
2224 * @return string HTML-wikitext mix oh yuck
2225 */
2231 }
2234 }
2239 }
2241 /**
2242 * Insert a NOPARSE hacky thing into any inline links in a chunk that's
2243 * going to go through further parsing steps before inline URL expansion.
2244 *
2245 * Not needed quite as much as it used to be since free links are a bit
2246 * more sensible these days. But bracketed links are still an issue.
2247 *
2248 * @param string $text More-or-less HTML
2249 * @return string Less-or-more HTML with NOPARSE bits
2250 */
2254 }
2256 /**
2257 * Return true if subpage links should be expanded on this page.
2258 * @return bool
2259 */
2261 # Some namespaces don't allow subpages
2263 }
2265 /**
2266 * Handle link to subpage if necessary
2267 *
2268 * @param string $target The source of the link
2269 * @param string &$text The link text, modified as necessary
2270 * @return string The full name of the link
2271 * @private
2272 */
2275 }
2277 /**#@+
2278 * Used by doBlockLevels()
2279 * @private
2280 *
2281 * @return string
2282 */
2287 }
2291 }
2293 /**
2294 * getCommon() returns the length of the longest common substring
2295 * of both arguments, starting at the beginning of both.
2296 * @private
2297 *
2298 * @param string $st1
2299 * @param string $st2
2300 *
2301 * @return int
2302 */
2308 }
2313 }
2314 }
2316 }
2318 /**
2319 * These next three functions open, continue, and close the list
2320 * element appropriate to the prefix character passed into them.
2321 * @private
2322 *
2323 * @param string $char
2324 *
2325 * @return string
2326 */
2341 }
2344 }
2346 /**
2347 * TODO: document
2348 * @param string $char
2349 * @private
2350 *
2351 * @return string
2352 */
2360 }
2367 }
2368 }
2370 }
2372 /**
2373 * @todo Document
2374 * @param string $char
2375 * @private
2376 *
2377 * @return string
2378 */
2390 }
2393 }
2395 }
2396 /**#@-*/
2398 /**
2399 * Make lists from lines starting with ':', '*', '#', etc. (DBL)
2400 *
2401 * @param string $text
2402 * @param bool $linestart Whether or not this is at the start of a line.
2403 * @private
2404 * @return string The lists rendered as HTML
2405 */
2409 # Parsing through the text line by line. The main thing
2410 # happening here is handling of block-level elements p, pre,
2411 # and making lists from lines starting with * # : etc.
2412 #
2422 # Fix up $linestart
2427 }
2428 # * = ul
2429 # # = ol
2430 # ; = dt
2431 # : = dd
2436 # If not in a <pre> element, scan for and figure out what prefixes are there.
2438 # Multiple prefixes may abut each other for nested lists.
2442 # eh?
2443 # ; and : are both from definition-lists, so they're equivalent
2444 # for the purposes of determining whether or not we need to open/close
2445 # elements.
2450 # Don't interpret any other prefixes in preformatted text
2454 }
2456 # List generation
2458 # Same as the last item, so no need to deal with nesting or opening stuff
2463 # The one nasty exception: definition lists work like this:
2464 # ; title : definition text
2465 # So we check for : in the remainder text to split up the
2466 # title and definition, without b0rking links.
2471 }
2472 }
2474 # We need to open or close prefixes, or both.
2476 # Either open or close a level...
2480 # Close all the prefixes which aren't shared.
2484 }
2486 # Continue the current prefix if appropriate.
2489 }
2491 # Open prefixes where appropriate.
2494 }
2500 # @todo FIXME: This is dupe of code above
2504 }
2505 }
2507 }
2510 }
2512 }
2514 # If we have no prefixes, go to paragraph mode.
2517 # No prefix (not in list)--go to paragraph mode
2518 # XXX: use a stack for nestable elements like span, table and div
2520 '/(?:<table|<h1|<h2|<h3|<h4|<h5|<h6|<pre|<tr|'
2522 $t
2523 );
2529 $t
2530 );
2534 # @todo bug 5718: paragraph closed
2538 }
2540 while ( preg_match( '/<(\\/?)blockquote[\s>]/i', $t, $bqMatch, PREG_OFFSET_CAPTURE, $bqOffset ) ) {
2543 }
2549 ) {
2550 # pre
2555 }
2558 # paragraph
2571 }
2572 }
2581 }
2582 }
2583 }
2584 }
2586 }
2587 # somewhere above we forget to get out of pre block (bug 785)
2590 }
2595 }
2596 }
2597 }
2603 }
2604 }
2608 }
2612 }
2614 /**
2615 * Split up a string on ':', ignoring any occurrences inside tags
2616 * to prevent illegal overlapping.
2617 *
2618 * @param string $str The string to split
2619 * @param string &$before Set to everything before the ':'
2620 * @param string &$after Set to everything after the ':'
2621 * @throws MWException
2622 * @return string The position of the ':', or false if none found
2623 */
2629 # Nothing to find!
2632 }
2636 # Easy; no tag nesting to worry about
2641 }
2643 # Ugly state machine to walk through avoiding tags.
2651 # (Using the number is a performance hack for common cases)
2655 # Could be either a <start> tag or an </end> tag
2660 # We found it!
2665 }
2666 # Embedded in a tag; don't break it.
2669 # Skip ahead looking for something interesting
2672 # Nothing else interesting
2675 }
2679 # We found it!
2684 }
2685 }
2687 # Nothing else interesting to find; abort!
2688 # We're nested, but there's no close tags left. Abort!
2690 }
2691 # Skip ahead to next tag start
2694 }
2697 # In a <tag>
2704 # Slash may be followed by >?
2708 # ignore
2709 }
2720 # Illegal early close? This shouldn't happen D:
2725 }
2728 # In a </tag>
2735 }
2737 }
2741 # Yes, a self-closed tag <blah/>
2744 # Probably we're jumping the gun, and this is an attribute
2746 }
2751 }
2758 }
2765 }
2770 }
2771 }
2773 wfDebug( __METHOD__ . ": Invalid input; not enough close tags (stack $stack, state $state)\n" );
2776 }
2779 }
2781 /**
2782 * Return value of a magic variable (like PAGENAME)
2783 *
2784 * @private
2785 *
2786 * @param int $index
2787 * @param bool|PPFrame $frame
2788 *
2789 * @throws MWException
2790 * @return string
2791 */
2797 // If no title set, bad things are going to happen
2798 // later. Title should always be set since this
2799 // should only be called in the middle of a parse
2800 // operation (but the unit-tests do funky stuff)
2803 }
2805 /**
2806 * Some of these require message or data lookups and can be
2807 * expensive to check many times.
2808 */
2812 }
2813 }
2855 $value = $pageLang->getMonthAbbreviation( MWTimestamp::getLocalInstance( $ts )->format( 'n' ) );
2889 ) ) );
2899 ) ) );
2907 }
2915 }
2928 # 0 means the page doesn't exist in the database,
2929 # which means the user is previewing a new page.
2930 # The vary-revision flag must be set, because the magic word
2931 # will have a different value once the page is saved.
2934 }
2938 # Let the edit saving system know we should parse the page
2939 # *after* a revision ID has been assigned.
2945 # Let the edit saving system know we should parse the page
2946 # *after* a revision ID has been assigned. This is for null edits.
2952 # Let the edit saving system know we should parse the page
2953 # *after* a revision ID has been assigned. This is for null edits.
2959 # Let the edit saving system know we should parse the page
2960 # *after* a revision ID has been assigned. This is for null edits.
2966 # Let the edit saving system know we should parse the page
2967 # *after* a revision ID has been assigned. This is for null edits.
2973 # Let the edit saving system know we should parse the page
2974 # *after* a revision ID has been assigned. This is for null edits.
2980 # Let the edit saving system know we should parse the page
2981 # *after* a revision ID has been assigned. This is for null edits.
2987 # Let the edit saving system know we should parse the page
2988 # *after* a revision ID has been assigned. This is for null edits.
2994 # Let the edit saving system know we should parse the page
2995 # *after* a revision ID has been assigned. This is for null edits.
3036 # @bug 4594 PHP5 has it zero padded, PHP4 does not, cast to
3037 # int to remove the padding
3046 );
3055 false
3056 );
3062 # @bug 4594 PHP5 has it zero padded, PHP4 does not, cast to
3063 # int to remove the padding
3128 );
3131 }
3135 }
3138 }
3140 /**
3141 * initialise the magic variables (like CURRENTMONTHNAME) and substitution modifiers
3142 *
3143 * @private
3144 */
3153 }
3155 /**
3156 * Preprocess some wikitext and return the document tree.
3157 * This is the ghost of replace_variables().
3158 *
3159 * @param string $text The text to parse
3160 * @param int $flags Bitwise combination of:
3161 * - self::PTD_FOR_INCLUSION: Handle "<noinclude>" and "<includeonly>" as if the text is being
3162 * included. Default is to assume a direct page view.
3163 *
3164 * The generated DOM tree must depend only on the input text and the flags.
3165 * The DOM tree must be the same in OT_HTML and OT_WIKI mode, to avoid a regression of bug 4899.
3166 *
3167 * Any flag added to the $flags parameter here, or any other parameter liable to cause a
3168 * change in the DOM tree for a given text, must be passed through the section identifier
3169 * in the section edit link and thus back to extractSections().
3170 *
3171 * The output of this function is currently only cached in process memory, but a persistent
3172 * cache may be implemented at a later date which takes further advantage of these strict
3173 * dependency requirements.
3174 *
3175 * @return PPNode
3176 */
3180 }
3182 /**
3183 * Return a three-element array: leading whitespace, string contents, trailing whitespace
3184 *
3185 * @param string $s
3186 *
3187 * @return array
3188 */
3198 }
3200 }
3202 /**
3203 * Replace magic variables, templates, and template arguments
3204 * with the appropriate text. Templates are substituted recursively,
3205 * taking care to avoid infinite loops.
3206 *
3207 * Note that the substitution depends on value of $mOutputType:
3208 * self::OT_WIKI: only {{subst:}} templates
3209 * self::OT_PREPROCESS: templates but not extension tags
3210 * self::OT_HTML: all templates and extension tags
3211 *
3212 * @param string $text The text to transform
3213 * @param bool|PPFrame $frame Object describing the arguments passed to the
3214 * template. Arguments may also be provided as an associative array, as
3215 * was the usual case before MW1.12. Providing arguments this way may be
3216 * useful for extensions wishing to perform variable replacement
3217 * explicitly.
3218 * @param bool $argsOnly Only do argument (triple-brace) expansion, not
3219 * double-brace expansion.
3220 * @return string
3221 */
3223 # Is there any text? Also, Prevent too big inclusions!
3226 }
3235 }
3243 }
3245 /**
3246 * Clean up argument array - refactored in 1.9 so parserfunctions can use it, too.
3247 *
3248 * @param array $args
3249 *
3250 * @return array
3251 */
3264 }
3267 }
3268 }
3269 }
3272 }
3274 /**
3275 * Warn the user when a parser limitation is reached
3276 * Will warn at most once the user per limitation type
3277 *
3278 * @param string $limitationType Should be one of:
3279 * 'expensive-parserfunction' (corresponding messages:
3280 * 'expensive-parserfunction-warning',
3281 * 'expensive-parserfunction-category')
3282 * 'post-expand-template-argument' (corresponding messages:
3283 * 'post-expand-template-argument-warning',
3284 * 'post-expand-template-argument-category')
3285 * 'post-expand-template-inclusion' (corresponding messages:
3286 * 'post-expand-template-inclusion-warning',
3287 * 'post-expand-template-inclusion-category')
3288 * 'node-count-exceeded' (corresponding messages:
3289 * 'node-count-exceeded-warning',
3290 * 'node-count-exceeded-category')
3291 * 'expansion-depth-exceeded' (corresponding messages:
3292 * 'expansion-depth-exceeded-warning',
3293 * 'expansion-depth-exceeded-category')
3294 * @param string|int|null $current Current value
3295 * @param string|int|null $max Maximum allowed, when an explicit limit has been
3296 * exceeded, provide the values (optional)
3297 */
3299 # does no harm if $current and $max are present but are unnecessary for the message
3304 }
3306 /**
3307 * Return the text of a template, after recursively
3308 * replacing any variables or templates within the template.
3309 *
3310 * @param array $piece The parts of the template
3311 * $piece['title']: the title, i.e. the part before the |
3312 * $piece['parts']: the parameter array
3313 * $piece['lineStart']: whether the brace was at the start of a line
3314 * @param PPFrame $frame The current frame, contains template arguments
3315 * @throws Exception
3316 * @return string The text of the template
3317 */
3322 // Flags
3324 // $text has been filled
3326 // wiki markup in $text should be escaped
3328 // $text is HTML, armour it against wikitext transformation
3330 // Force interwiki transclusion to be done in raw mode not rendered
3332 // $text is a DOM node needing expansion in a child frame
3334 // $text is a DOM node needing expansion in the current frame
3337 # Title object, where $text came from
3340 # $part1 is the bit before the first |, and must contain only title characters.
3341 # Various prefixes will be stripped from it later.
3346 # Original title text preserved for various purposes
3349 # $args is a list of argument nodes, starting from index 0, not including $part1
3350 # @todo FIXME: If piece['parts'] is null then the call to getLength()
3351 # below won't work b/c this $args isn't an object
3357 # SUBST
3363 # Possibilities for substMatch: "subst", "safesubst" or FALSE
3364 # Decide whether to expand template or keep wikitext as-is.
3370 }
3376 }
3377 }
3382 }
3383 }
3385 # Variables
3392 }
3394 }
3395 }
3397 # MSG, MSGNW and RAW
3399 # Check for MSGNW:
3404 # Remove obsolete MSG:
3407 }
3409 # Check for RAW:
3413 }
3414 }
3417 # Parser functions
3427 }
3434 }
3436 # The interface for parser functions allows for extracting
3437 # flags into the local scope. Extract any forwarded flags
3438 # here.
3440 }
3442 }
3444 # Finish mangling title and then check for loops.
3445 # Set $title to a Title object and $titleText to the PDBK
3448 # Split the title into page and subpage
3454 }
3458 # Check for language variants if the template is not found
3461 }
3462 # Do recursion depth check
3470 }
3471 }
3472 }
3474 # Load from database
3477 # Too many unique items can kill profiling DBs/collectors
3480 }
3486 ) {
3487 // Pass the template arguments as URL parameters.
3488 // "uselang" will have no effect since the Language object
3489 // is forced to the one defined in ParserOptions.
3498 }
3499 }
3501 // Create a new context to execute the special page
3514 }
3524 }
3525 }
3527 # If the title is valid but undisplayable, make a link to it
3531 }
3533 # Interwiki transclusion
3539 # Preprocess it like a template
3542 }
3544 }
3546 # Do infinite loop check
3547 # This has to be done after redirect resolution to avoid infinite loops via redirects
3554 }
3556 }
3558 # If we haven't found text to substitute by now, we're done
3559 # Recover the source wikitext and return it
3564 }
3567 }
3569 # Expand DOM-style return values in a child frame
3571 # Clean up argument array
3577 # Expansion is eligible for the empty-frame cache
3580 # Uncached expansion
3582 }
3583 }
3587 }
3591 }
3593 # Replace raw HTML by a placeholder
3597 # Escape nowiki-style return values
3602 ) {
3603 # Bug 529: if the template begins with a table or block-level
3604 # element, it should be treated as beginning a new line.
3605 # This behavior is somewhat controversial.
3607 }
3610 # Error, oversize inclusion
3612 # Make a working, properly escaped link if possible (bug 23588)
3615 # This will probably not be a working link, but at least it may
3616 # provide some hint of where the problem is
3619 }
3623 }
3629 }
3633 }
3635 /**
3636 * Call a parser function and return an array with text and flags.
3637 *
3638 * The returned array will always contain a boolean 'found', indicating
3639 * whether the parser function was found or not. It may also contain the
3640 * following:
3641 * text: string|object, resulting wikitext or PP DOM object
3642 * isHTML: bool, $text is HTML, armour it against wikitext transformation
3643 * isChildObj: bool, $text is a DOM node needing expansion in a child frame
3644 * isLocalObj: bool, $text is a DOM node needing expansion in the current frame
3645 * nowiki: bool, wiki markup in $text should be escaped
3646 *
3647 * @since 1.21
3648 * @param PPFrame $frame The current frame, contains template arguments
3649 * @param string $function Function name
3650 * @param array $args Arguments to the function
3651 * @throws MWException
3652 * @return array
3653 */
3659 # Case sensitive functions
3663 # Case insensitive functions
3670 }
3671 }
3676 # Workaround for PHP bug 35229 and similar
3681 }
3685 # Convert arguments to PPNodes and collect for appending to $allArgs
3692 }
3693 }
3695 # Add a frame parameter, and pass the arguments as an array
3699 # Convert arguments to plain text and append to $allArgs
3707 }
3708 }
3709 }
3713 # The interface for function hooks allows them to return a wikitext
3714 # string or an array containing the string and any flags. This mungs
3715 # things around to match what this method should return.
3720 );
3724 }
3728 );
3729 }
3735 }
3738 }
3743 }
3748 }
3750 /**
3751 * Get the semi-parsed DOM representation of a template with a given title,
3752 * and its redirect destination title. Cached.
3753 *
3754 * @param Title $title
3755 *
3756 * @return array
3757 */
3766 }
3769 }
3771 # Cache miss, go to the database
3777 }
3785 }
3788 }
3790 /**
3791 * Fetch the unparsed text of a template and register a reference to it.
3792 * @param Title $title
3793 * @return array ( string or false, Title )
3794 */
3796 // Defaults to Parser::statelessFetchTemplate()
3805 // If we transclude ourselves, the final result
3806 // will change based on the new version of the page
3808 }
3809 }
3810 }
3812 }
3814 /**
3815 * Fetch the unparsed text of a template and register a reference to it.
3816 * @param Title $title
3817 * @return string|bool
3818 */
3822 }
3824 /**
3825 * Static function to get a template
3826 * Can be overridden via ParserOptions::setTemplateCallback().
3827 *
3828 * @param Title $title
3829 * @param bool|Parser $parser
3830 *
3831 * @return array
3832 */
3838 # Loop to fetch the article, with up to 1 redirect
3840 # Give extensions a chance to select the revision instead
3851 );
3853 }
3854 # Get the revision
3859 # If there is no current revision, there is no page
3863 }
3870 # We fetched a rev from a different title; register it too...
3875 }
3884 }
3891 }
3896 }
3899 }
3900 # Redirect?
3903 }
3908 }
3910 /**
3911 * Fetch a file and its title and register a reference to it.
3912 * If 'broken' is a key in $options then the file will appear as a broken thumbnail.
3913 * @param Title $title
3914 * @param array $options Array of options to RepoGroup::findFile
3915 * @return File|bool
3916 */
3920 }
3922 /**
3923 * Fetch a file and its title and register a reference to it.
3924 * If 'broken' is a key in $options then the file will appear as a broken thumbnail.
3925 * @param Title $title
3926 * @param array $options Array of options to RepoGroup::findFile
3927 * @return array ( File or false, Title of file )
3928 */
3934 # Register the file as a dependency...
3937 # Update fetched file title
3940 }
3942 }
3944 /**
3945 * Helper function for fetchFileAndTitle.
3946 *
3947 * Also useful if you need to fetch a file but not use it yet,
3948 * for example to get the file's handler.
3949 *
3950 * @param Title $title
3951 * @param array $options Array of options to RepoGroup::findFile
3952 * @return File|bool
3953 */
3961 }
3963 }
3965 /**
3966 * Transclude an interwiki link.
3967 *
3968 * @param Title $title
3969 * @param string $action
3970 *
3971 * @return string
3972 */
3978 }
3984 }
3986 }
3988 /**
3989 * @param string $url
3990 * @return mixed|string
3991 */
4000 }
4007 // Though we failed to fetch the content, this status is useless.
4012 }
4019 ) );
4021 }
4023 /**
4024 * Triple brace replacement -- used for template arguments
4025 * @private
4026 *
4027 * @param array $piece
4028 * @param PPFrame $frame
4029 *
4030 * @return array
4031 */
4045 )
4046 ) {
4047 # No match in frame, use the supplied default
4049 }
4053 }
4056 # No match anywhere
4058 }
4061 }
4066 }
4070 }
4072 /**
4073 * Return the text to be used for a given extension tag.
4074 * This is the ghost of strip().
4075 *
4076 * @param array $params Associative array of parameters:
4077 * name PPNode for the tag name
4078 * attr PPNode for unparsed text where tag attributes are thought to be
4079 * attributes Optional associative array of parsed attributes
4080 * inner Contents of extension element
4081 * noClose Original text did not have a close tag
4082 * @param PPFrame $frame
4083 *
4084 * @throws MWException
4085 * @return string
4086 */
4100 }
4106 }
4109 # Workaround for PHP bug 35229 and similar
4112 }
4119 }
4125 }
4128 # Extract flags to local scope (to override $markerType)
4133 }
4137 }
4142 }
4143 }
4149 }
4150 }
4160 }
4162 }
4164 /**
4165 * Increment an include size counter
4166 *
4167 * @param string $type The type of expansion
4168 * @param int $size The size of the text
4169 * @return bool False if this inclusion would take it over the maximum, true otherwise
4170 */
4177 }
4178 }
4180 /**
4181 * Increment the expensive function count
4182 *
4183 * @return bool False if the limit has been exceeded
4184 */
4188 }
4190 /**
4191 * Strip double-underscore items like __NOGALLERY__ and __NOTOC__
4192 * Fills $this->mDoubleUnderscores, returns the modified text
4193 *
4194 * @param string $text
4195 *
4196 * @return string
4197 */
4201 # The position of __TOC__ needs to be recorded
4207 # Set a placeholder. At the end we'll fill it in with the TOC.
4210 # Only keep the first one.
4212 }
4214 # Now match and remove the rest of them
4220 }
4223 }
4226 ) {
4228 }
4229 # (bug 8068) Allow control over whether robots index a page.
4230 #
4231 # @todo FIXME: Bug 14899: __INDEX__ always overrides __NOINDEX__ here! This
4232 # is not desirable, the last one on the page should win.
4236 }
4240 }
4242 # Cache all double underscores in the database
4245 }
4249 }
4251 /**
4252 * Add a tracking category, getting the title from a system message,
4253 * or print a debug message if the title is invalid.
4254 *
4255 * Please add any message that you use with this function to
4256 * $wgTrackingCategories. That way they will be listed on
4257 * Special:TrackingCategories.
4258 *
4259 * @param string $msg Message key
4260 * @return bool Whether the addition was successful
4261 */
4266 }
4267 // Important to parse with correct title (bug 31469)
4273 # Allow tracking categories to be disabled by setting them to "-"
4276 }
4285 }
4286 }
4288 /**
4289 * This function accomplishes several tasks:
4290 * 1) Auto-number headings if that option is enabled
4291 * 2) Add an [edit] link to sections for users who have enabled the option and can edit the page
4292 * 3) Add a Table of contents on the top for users who have enabled the option
4293 * 4) Auto-anchor headings
4294 *
4295 * It loops through all headlines, collects the necessary data, then splits up the
4296 * string and re-inserts the newly formatted headlines.
4297 *
4298 * @param string $text
4299 * @param string $origText Original, untouched wikitext
4300 * @param bool $isMain
4301 * @return mixed|string
4302 * @private
4303 */
4307 # Inhibit editsection links if requested in the page
4313 }
4316 }
4318 # Get all headlines for numbering them and adding funky stuff like [edit]
4319 # links - this is for later, but we need the number of headlines right now
4324 $matches
4325 );
4327 # if there are fewer than 4 headlines in the article, do not show TOC
4328 # unless it's been explicitly enabled.
4332 # Allow user to stipulate that a page should have a "new section"
4333 # link added via __NEWSECTIONLINK__
4336 }
4338 # Allow user to remove the "new section"
4339 # link via __NONEWSECTIONLINK__
4342 }
4344 # if the string __FORCETOC__ (not case-sensitive) occurs in the HTML,
4345 # override above conditions and always show TOC above first header
4349 }
4351 # headline counter
4355 # Ugh .. the TOC should have neat indentation levels which can be
4356 # passed to the skin functions. These are determined here
4388 }
4392 }
4396 # Increase TOC level
4403 }
4405 # Decrease TOC level, find level to jump to
4409 # Found last matching level
4413 # Found first matching level below current level
4416 }
4417 }
4420 }
4423 # Unindent only if the previous toc level was shown :p
4428 }
4429 }
4431 # No change in level, end TOC line
4434 }
4435 }
4439 # count number of headlines for each level
4446 }
4449 }
4450 }
4452 # The safe header is a version of the header text safe to use for links
4454 # Remove link placeholders by the link text.
4455 # <!--LINK number-->
4456 # turns into
4457 # link text with suffix
4458 # Do this before unstrip since link text can contain strip markers
4461 # Avoid insertion of weird stuff like <math> by expanding the relevant sections
4464 # Strip out HTML (first regex removes any tag not allowed)
4465 # Allowed tags are:
4466 # * <sup> and <sub> (bug 8393)
4467 # * <i> (bug 26375)
4468 # * <b> (r105284)
4469 # * <span dir="rtl"> and <span dir="ltr"> (bug 35167)
4470 #
4471 # We strip any parameter from accepted tags (second regex), except dir="rtl|ltr" from <span>,
4472 # to allow setting directionality in toc items.
4477 ),
4479 $safeHeadline
4480 );
4483 # For the anchor, strip out HTML-y stuff period
4487 # Save headline for section edit hint before it's escaped
4491 # For reverse compatibility, provide an id that's
4492 # HTML4-compatible, like we used to.
4493 #
4494 # It may be worth noting, academically, that it's possible for
4495 # the legacy anchor to conflict with a non-legacy headline
4496 # anchor on the page. In this case likely the "correct" thing
4497 # would be to either drop the legacy anchors or make sure
4498 # they're numbered first. However, this would require people
4499 # to type in section names like "abc_.D7.93.D7.90.D7.A4"
4500 # manually, so let's not bother worrying about it.
4506 # No reason to have both (in fact, we can't)
4508 }
4513 }
4515 # HTML names must be case-insensitively unique (bug 10721).
4516 # This does not apply to Unicode characters per
4517 # http://dev.w3.org/html5/spec/infrastructure.html#case-sensitivity-and-string-comparison
4518 # @todo FIXME: We may be changing them depending on the current locale.
4524 }
4526 # count how many in assoc. array so we can track dupes in anchors
4531 }
4536 }
4538 # Don't number the heading if it is the only one (looks silly)
4540 # the two are different if the line contains a link
4544 $numbering
4546 }
4548 # Create the anchor for linking from the TOC to the section
4553 }
4556 }
4560 }
4562 # Add the section to the section tree
4563 # Find the DOM node for this header
4570 }
4571 }
4575 }
4585 );
4587 # give headline the correct <h#> tag
4589 // Output edit section links as markers with styles that can be customized by skins
4591 # Put a T flag in the section identifier, to indicate to extractSections()
4592 # that sections inside <includeonly> should be counted.
4598 $headlineHint
4599 );
4600 }
4601 // We use a bit of pesudo-xml for editsection markers. The
4602 // language converter is run later on. Using a UNIQ style marker
4603 // leads to the converter screwing up the tokens when it
4604 // converts stuff. And trying to insert strip tags fails too. At
4605 // this point all real inputted tags have already been escaped,
4606 // so we don't have to worry about a user trying to input one of
4607 // these markers directly. We use a page and section attribute
4608 // to stop the language converter from converting these
4609 // important bits of data, but put the headline hint inside a
4610 // content block because the language converter is supposed to
4611 // be able to convert that piece of data.
4618 }
4621 }
4627 }
4631 # Never ever show TOC if no headers
4634 }
4639 }
4644 }
4648 }
4650 # split up and insert constructed headlines
4654 // build an array of document sections
4657 // $head is zero-based, sections aren't.
4662 }
4664 /**
4665 * Send a hook, one per section.
4666 * The idea here is to be able to make section-level DIVs, but to do so in a
4667 * lower-impact, more correct way than r50769
4668 *
4669 * $this : caller
4670 * $section : the section number
4671 * &$sectionContent : ref to the content of the section
4672 * $showEditLinks : boolean describing whether this section has an edit link
4673 */
4677 }
4680 // append the TOC at the beginning
4681 // Top anchor now in skin
4683 }
4691 }
4692 }
4694 /**
4695 * Transform wiki markup when saving a page by doing "\r\n" -> "\n"
4696 * conversion, substitting signatures, {{subst:}} templates, etc.
4697 *
4698 * @param string $text The text to transform
4699 * @param Title $title The Title object for the current article
4700 * @param User $user The User object describing the current user
4701 * @param ParserOptions $options Parsing options
4702 * @param bool $clearState Whether to clear the parser state first
4703 * @return string The altered wiki markup
4704 */
4707 ) {
4710 }
4716 );
4720 }
4726 }
4728 /**
4729 * Pre-save transform helper function
4730 *
4731 * @param string $text
4732 * @param User $user
4733 *
4734 * @return string
4735 */
4739 # Note: This is the timestamp saved as hardcoded wikitext to
4740 # the database, we use $wgContLang here in order to give
4741 # everyone the same signature and use the default one rather
4742 # than the one selected in each user's preferences.
4743 # (see also bug 12815)
4749 # Allow translation of timezones through wiki. format() can return
4750 # whatever crap the system uses, localised or not, so we cannot
4751 # ship premade translations.
4756 }
4760 # Variable replacement
4761 # Because mOutputType is OT_WIKI, this will only process {{subst:xxx}} type tags
4764 # This works almost by chance, as the replaceVariables are done before the getUserSig(),
4765 # which may corrupt this parser instance via its wfMessage()->text() call-
4767 # Signatures
4773 ) );
4775 # Context links ("pipe tricks"): [[|name]] and [[name (context)|]]
4779 // [[ns:page (context)|]]
4781 // [[ns:page（context）|]] (double-width brackets, added in r40257)
4783 // [[ns:page (context), context|]] (using either single or double-width comma)
4785 // [[|page]] (reverse pipe trick: add context from page title)
4788 # try $p1 first, to turn "[[A, B (C)|]]" into "[[A, B (C)|A, B]]"
4800 # if there's no context, don't bother duplicating the title
4802 }
4804 # Trim trailing whitespace
4808 }
4810 /**
4811 * Fetch the user's signature text, if any, and normalize to
4812 * validated, ready-to-insert wikitext.
4813 * If you have pre-fetched the nickname or the fancySig option, you can
4814 * specify them here to save a database query.
4815 * Do not reuse this parser instance after calling getUserSig(),
4816 * as it may have changed if it's the $wgParser.
4817 *
4818 * @param User $user
4819 * @param string|bool $nickname Nickname to use or false to use user's default nickname
4820 * @param bool|null $fancySig whether the nicknname is the complete signature
4821 * or null to use default value
4822 * @return string
4823 */
4829 # If not given, retrieve from the user object.
4832 }
4836 }
4844 # Sig. might contain markup; validate this
4846 # Validated; clean up (if needed) and return it
4849 # Failed to validate; fall back to the default
4852 }
4853 }
4855 # Make sure nickname doesnt get a sig in a sig
4858 # If we're still here, make it a link to the user page
4865 }
4867 /**
4868 * Check that the user's signature contains no bad XML
4869 *
4870 * @param string $text
4871 * @return string|bool An expanded string, or false if invalid.
4872 */
4875 }
4877 /**
4878 * Clean up signature text
4879 *
4880 * 1) Strip ~~~, ~~~~ and ~~~~~ out of signatures @see cleanSigInSig
4881 * 2) Substitute all transclusions
4882 *
4883 * @param string $text
4884 * @param bool $parsing Whether we're cleaning (preferences save) or parsing
4885 * @return string Signature text
4886 */
4892 }
4894 # Option to disable this feature
4897 }
4899 # @todo FIXME: Regex doesn't respect extension tags or nowiki
4900 # => Move this logic to braceSubstitution()
4902 $substRegex = '/\{\{(?!(?:' . $substWord->getBaseRegex() . '))/x' . $substWord->getRegexCase();
4913 }
4916 }
4918 /**
4919 * Strip ~~~, ~~~~ and ~~~~~ out of signatures
4920 *
4921 * @param string $text
4922 * @return string Signature text with /~{3,5}/ removed
4923 */
4927 }
4929 /**
4930 * Set up some variables which are usually set up in parse()
4931 * so that an external function can call some class members with confidence
4932 *
4933 * @param Title|null $title
4934 * @param ParserOptions $options
4935 * @param int $outputType
4936 * @param bool $clearState
4937 */
4940 ) {
4942 }
4944 /**
4945 * @param Title|null $title
4946 * @param ParserOptions $options
4947 * @param int $outputType
4948 * @param bool $clearState
4949 */
4952 ) {
4958 }
4959 }
4961 /**
4962 * Wrapper for preprocess()
4963 *
4964 * @param string $text The text to preprocess
4965 * @param ParserOptions $options Options
4966 * @param Title|null $title Title object or null to use $wgTitle
4967 * @return string
4968 */
4972 # Guard against infinite recursion
4975 }
4982 }
4989 }
4991 /**
4992 * Create an HTML-style tag, e.g. "<yourtag>special text</yourtag>"
4993 * The callback should have the following form:
4994 * function myParserHook( $text, $params, $parser, $frame ) { ... }
4995 *
4996 * Transform and return $text. Use $parser for any required context, e.g. use
4997 * $parser->getTitle() and $parser->getOptions() not $wgTitle or $wgOut->mParserOptions
4998 *
4999 * Hooks may return extended information by returning an array, of which the
5000 * first numbered element (index 0) must be the return string, and all other
5001 * entries are extracted into local variables within an internal function
5002 * in the Parser class.
5003 *
5004 * This interface (introduced r61913) appears to be undocumented, but
5005 * 'markerName' is used by some core tag hooks to override which strip
5006 * array their results are placed in. **Use great caution if attempting
5007 * this interface, as it is not documented and injudicious use could smash
5008 * private variables.**
5009 *
5010 * @param string $tag The tag to use, e.g. 'hook' for "<hook>"
5011 * @param mixed $callback The callback function (and object) to use for the tag
5012 * @throws MWException
5013 * @return mixed|null The old value of the mTagHooks array associated with the hook
5014 */
5019 }
5024 }
5027 }
5029 /**
5030 * As setHook(), but letting the contents be parsed.
5031 *
5032 * Transparent tag hooks are like regular XML-style tag hooks, except they
5033 * operate late in the transformation sequence, on HTML instead of wikitext.
5034 *
5035 * This is probably obsoleted by things dealing with parser frames?
5036 * The only extension currently using it is geoserver.
5037 *
5038 * @since 1.10
5039 * @todo better document or deprecate this
5040 *
5041 * @param string $tag The tag to use, e.g. 'hook' for "<hook>"
5042 * @param mixed $callback The callback function (and object) to use for the tag
5043 * @throws MWException
5044 * @return mixed|null The old value of the mTagHooks array associated with the hook
5045 */
5050 }
5051 $oldVal = isset( $this->mTransparentTagHooks[$tag] ) ? $this->mTransparentTagHooks[$tag] : null;
5055 }
5057 /**
5058 * Remove all tag hooks
5059 */
5064 }
5066 /**
5067 * Create a function, e.g. {{sum:1|2|3}}
5068 * The callback function should have the form:
5069 * function myParserFunction( &$parser, $arg1, $arg2, $arg3 ) { ... }
5070 *
5071 * Or with SFH_OBJECT_ARGS:
5072 * function myParserFunction( $parser, $frame, $args ) { ... }
5073 *
5074 * The callback may either return the text result of the function, or an array with the text
5075 * in element 0, and a number of flags in the other elements. The names of the flags are
5076 * specified in the keys. Valid flags are:
5077 * found The text returned is valid, stop processing the template. This
5078 * is on by default.
5079 * nowiki Wiki markup in the return value should be escaped
5080 * isHTML The returned text is HTML, armour it against wikitext transformation
5081 *
5082 * @param string $id The magic word ID
5083 * @param mixed $callback The callback function (and object) to use
5084 * @param int $flags A combination of the following flags:
5085 * SFH_NO_HASH No leading hash, i.e. {{plural:...}} instead of {{#if:...}}
5086 *
5087 * SFH_OBJECT_ARGS Pass the template arguments as PPNode objects instead of text. This
5088 * allows for conditional expansion of the parse tree, allowing you to eliminate dead
5089 * branches and thus speed up parsing. It is also possible to analyse the parse tree of
5090 * the arguments, and to control the way they are expanded.
5091 *
5092 * The $frame parameter is a PPFrame. This can be used to produce expanded text from the
5093 * arguments, for instance:
5094 * $text = isset( $args[0] ) ? $frame->expand( $args[0] ) : '';
5095 *
5096 * For technical reasons, $args[0] is pre-expanded and will be a string. This may change in
5097 * future versions. Please call $frame->expand() on it anyway so that your code keeps
5098 * working if/when this is changed.
5099 *
5100 * If you want whitespace to be trimmed from $args, you need to do it yourself, post-
5101 * expansion.
5102 *
5103 * Please read the documentation in includes/parser/Preprocessor.php for more information
5104 * about the methods available in PPFrame and PPNode.
5105 *
5106 * @throws MWException
5107 * @return string|callable The old callback function for this name, if any
5108 */
5115 # Add to function cache
5119 }
5125 # Case
5128 }
5129 # Add leading hash
5132 }
5133 # Remove trailing colon
5136 }
5138 }
5140 }
5142 /**
5143 * Get all registered function hook identifiers
5144 *
5145 * @return array
5146 */
5149 }
5151 /**
5152 * Create a tag function, e.g. "<test>some stuff</test>".
5153 * Unlike tag hooks, tag functions are parsed at preprocessor level.
5154 * Unlike parser functions, their content is not preprocessed.
5155 * @param string $tag
5156 * @param callable $callback
5157 * @param int $flags
5158 * @throws MWException
5159 * @return null
5160 */
5165 }
5172 }
5175 }
5177 /**
5178 * @todo FIXME: Update documentation. makeLinkObj() is deprecated.
5179 * Replace "<!--LINK-->" link placeholders with actual links, in the buffer
5180 * Placeholders created in Skin::makeLinkObj()
5181 *
5182 * @param string $text
5183 * @param int $options
5184 *
5185 * @return array Array of link CSS classes, indexed by PDBK.
5186 */
5189 }
5191 /**
5192 * Replace "<!--LINK-->" link placeholders with plain text of links
5193 * (not HTML-formatted).
5194 *
5195 * @param string $text
5196 * @return string
5197 */
5200 }
5202 /**
5203 * Renders an image gallery from a text with one line per image.
5204 * text labels may be given by using |-style alternative text. E.g.
5205 * Image:one.jpg|The number "1"
5206 * Image:tree.jpg|A tree
5207 * given as text will return the HTML of a gallery with two images,
5208 * labeled 'The number "1"' and
5209 * 'A tree'.
5210 *
5211 * @param string $text
5212 * @param array $params
5213 * @return string HTML
5214 */
5221 }
5226 // If invalid type set, fallback to default.
5228 }
5241 }
5247 }
5250 }
5253 }
5256 }
5263 # match lines like these:
5264 # Image:someimage.jpg|This is some image
5267 # Skip empty lines
5270 }
5274 }
5277 # Bogus title. Ignore these so we don't bomb out later.
5279 }
5281 # We need to get what handler the file uses, to figure out parameters.
5282 # Note, a hook can overide the file name, and chose an entirely different
5283 # file (which potentially could be of a different type and have different handler).
5288 # Don't register it now, as ImageGallery does that later.
5296 );
5299 // We don't want people to specify per-image widths.
5300 // Additionally the width parameter would need special casing anyhow.
5302 }
5312 // look for an |alt= definition while trying not to break existing
5313 // captions with multiple pipes (|) in it, until a more sensible grammar
5314 // is defined for images in galleries
5316 // FIXME: Doing recursiveTagParse at this stage, and the trim before
5317 // splitting on '|' is a bit odd, and different from makeImage.
5334 //check to see if link matches an absolute url, if not then it must be a wiki link.
5341 }
5342 }
5345 // Must be a handler specific parameter.
5349 // Guess not. Append it to the caption.
5352 }
5353 }
5356 // concatenate all other pipes
5358 }
5359 }
5360 // remove the first pipe
5362 }
5365 }
5369 }
5371 /**
5372 * @param string $handler
5373 * @return array
5374 */
5380 }
5382 # Initialise static lists
5389 );
5397 }
5398 }
5399 }
5401 # Add handler params
5407 }
5408 }
5411 }
5412 return array( $this->mImageParams[$handlerClass], $this->mImageParamsMagicArray[$handlerClass] );
5413 }
5415 /**
5416 * Parse image options text and use it to make an image
5417 *
5418 * @param Title $title
5419 * @param string $options
5420 * @param LinkHolderArray|bool $holders
5421 * @return string HTML
5422 */
5424 # Check if the options text is of the form "options|alt text"
5425 # Options are:
5426 # * thumbnail make a thumbnail with enlarge-icon and caption, alignment depends on lang
5427 # * left no resizing, just left align. label is used for alt= only
5428 # * right same, but right aligned
5429 # * none same, but not aligned
5430 # * ___px scale to ___ pixels width, no aligning. e.g. use in taxobox
5431 # * center center the image
5432 # * frame Keep original image size, no magnify-button.
5433 # * framed Same as "frame"
5434 # * frameless like 'thumb' but without a frame. Keeps user preferences for width
5435 # * upright reduce width for upright images, rounded to full __0 px
5436 # * border draw a 1px border around the image
5437 # * alt Text for HTML alt attribute (defaults to empty)
5438 # * class Set a class for img node
5439 # * link Set the target of the image link. Can be external, interwiki, or local
5440 # vertical-align values (no % or length right now):
5441 # * baseline
5442 # * sub
5443 # * super
5444 # * top
5445 # * text-top
5446 # * middle
5447 # * bottom
5448 # * text-bottom
5452 # Give extensions a chance to select the file revision for us
5457 # Fetch and register the file (file title may be different via hooks)
5460 # Get parameter map
5467 }
5469 # Process the input parameters
5481 # Special case; width and height come in one variable together
5489 }
5490 }
5496 }
5497 }
5498 # else no validation -- bug 13436
5501 # Validate handler parameter
5504 # Validate internal parameters
5509 # @todo FIXME: Possibly check validity here for
5510 # manualthumb? downstream behavior seems odd with
5511 # missing manual thumbs.
5528 }
5530 }
5538 }
5539 }
5544 // use first appearing option, discard others.
5549 # Most other things appear to be empty or numeric...
5551 }
5552 }
5556 }
5557 }
5558 }
5561 }
5562 }
5564 # Process alignment parameters
5567 }
5570 }
5574 # Will the image be presented in a frame, with the caption below?
5580 # In the old days, [[Image:Foo|text...]] would set alt text. Later it
5581 # came to also set the caption, ordinary text after the image -- which
5582 # makes no sense, because that just repeats the text multiple times in
5583 # screen readers. It *also* came to set the title attribute.
5584 #
5585 # Now that we have an alt attribute, we should not set the alt text to
5586 # equal the caption: that's worse than useless, it just repeats the
5587 # text. This is the framed/thumbnail case. If there's no caption, we
5588 # use the unnamed parameter for alt text as well, just for the time be-
5589 # ing, if the unnamed param is set and the alt param is not.
5590 #
5591 # For the future, we need to figure out if we want to tweak this more,
5592 # e.g., introducing a title= parameter for the title; ignoring the un-
5593 # named parameter entirely for images without a caption; adding an ex-
5594 # plicit caption= parameter and preserving the old magic unnamed para-
5595 # meter for BC; ...
5598 # No caption or alt text, add the filename as the alt text so
5599 # that screen readers at least get some description of the image
5601 }
5602 # Do not set $params['frame']['title'] because tooltips don't make sense
5603 # for framed images
5606 # No alt text, use the "caption" for the alt text
5610 # No caption, fall back to using the filename for the
5611 # alt text
5613 }
5614 }
5615 # Use the "caption" for the tooltip text
5617 }
5621 # Linker does the rest
5626 # Give the handler a chance to modify the parser object
5629 }
5632 }
5634 /**
5635 * @param string $caption
5636 * @param LinkHolderArray|bool $holders
5637 * @return mixed|string
5638 */
5640 # Strip bad stuff out of the title (tooltip). We can't just use
5641 # replaceLinkHoldersText() here, because if this function is called
5642 # from replaceInternalLinks2(), mLinkHolders won't be up-to-date.
5647 }
5649 # make sure there are no placeholders in thumbnail attributes
5650 # that are later expanded to html- so expand them now and
5651 # remove the tags
5656 }
5658 /**
5659 * Set a flag in the output object indicating that the content is dynamic and
5660 * shouldn't be cached.
5661 */
5667 }
5670 }
5672 /**
5673 * Callback from the Sanitizer for expanding items found in HTML attribute
5674 * values, so they can be safely tested and escaped.
5675 *
5676 * @param string $text
5677 * @param bool|PPFrame $frame
5678 * @return string
5679 */
5684 }
5686 /**
5687 * Accessor
5688 *
5689 * @return array
5690 */
5696 );
5697 }
5699 /**
5700 * Replace transparent tags in $text with the values given by the callbacks.
5701 *
5702 * Transparent tag hooks are like regular XML-style tag hooks, except they
5703 * operate late in the transformation sequence, on HTML instead of wikitext.
5704 *
5705 * @param string $text
5706 *
5707 * @return string
5708 */
5722 );
5725 }
5727 }
5729 }
5731 /**
5732 * Break wikitext input into sections, and either pull or replace
5733 * some particular section's text.
5734 *
5735 * External callers should use the getSection and replaceSection methods.
5736 *
5737 * @param string $text Page wikitext
5738 * @param string $section A section identifier string of the form:
5739 * "<flag1> - <flag2> - ... - <section number>"
5740 *
5741 * Currently the only recognised flag is "T", which means the target section number
5742 * was derived during a template inclusion parse, in other words this is a template
5743 * section edit link. If no flags are given, it was an ordinary section edit link.
5744 * This flag is required to avoid a section numbering mismatch when a section is
5745 * enclosed by "<includeonly>" (bug 6563).
5746 *
5747 * The section number 0 pulls the text before the first heading; other numbers will
5748 * pull the given section along with its lower-level subsections. If the section is
5749 * not found, $mode=get will return $newtext, and $mode=replace will return $text.
5750 *
5751 * Section 0 is always considered to exist, even if it only contains the empty
5752 * string. If $text is the empty string and section 0 is replaced, $newText is
5753 * returned.
5754 *
5755 * @param string $mode One of "get" or "replace"
5756 * @param string $newText Replacement text for section data.
5757 * @return string For "get", the extracted section text.
5758 * for "replace", the whole page with the section replaced.
5759 */
5768 # Process section extraction flags
5775 }
5776 }
5778 # Check for empty input
5780 # Only sections 0 and T-0 exist in an empty document
5786 }
5792 }
5793 }
5794 }
5796 # Preprocess the text
5799 # <h> nodes indicate section breaks
5800 # They can only occur at the top level, so we can find them by iterating the root's children
5803 # Find the target section
5805 # Section zero doesn't nest, level=big
5814 }
5815 }
5818 }
5820 }
5821 }
5824 # Not found
5829 }
5830 }
5832 # Find the end of the section, including nested sections
5839 }
5840 }
5843 }
5847 # Write out the remainder (in replace mode only)
5849 # Output the replacement text
5850 # Add two newlines on -- trailing whitespace in $newText is conventionally
5851 # stripped by the editor, so we need both newlines to restore the paragraph gap
5852 # Only add trailing whitespace if there is newText
5855 }
5860 }
5861 }
5864 # Re-insert stripped tags
5866 }
5869 }
5871 /**
5872 * This function returns the text of a section, specified by a number ($section).
5873 * A section is text under a heading like == Heading == or \<h1\>Heading\</h1\>, or
5874 * the first section before any such heading (section 0).
5875 *
5876 * If a section contains subsections, these are also returned.
5877 *
5878 * @param string $text Text to look in
5879 * @param string $section Section identifier
5880 * @param string $deftext Default to return if section is not found
5881 * @return string Text of the requested section
5882 */
5885 }
5887 /**
5888 * This function returns $oldtext after the content of the section
5889 * specified by $section has been replaced with $text. If the target
5890 * section does not exist, $oldtext is returned unchanged.
5891 *
5892 * @param string $oldtext Former text of the article
5893 * @param int $section Section identifier
5894 * @param string $text Replacing text
5895 * @return string Modified text
5896 */
5899 }
5901 /**
5902 * Get the ID of the revision we are parsing
5903 *
5904 * @return int|null
5905 */
5908 }
5910 /**
5911 * Get the revision object for $this->mRevisionId
5912 *
5913 * @return Revision|null Either a Revision object or null
5914 * @since 1.23 (public since 1.23)
5915 */
5919 }
5922 }
5926 }
5928 /**
5929 * Get the timestamp associated with the current revision, adjusted for
5930 * the default server-local timestamp
5931 * @return string
5932 */
5942 # The cryptic '' timezone parameter tells to use the site-default
5943 # timezone offset instead of the user settings.
5944 #
5945 # Since this value will be saved into the parser cache, served
5946 # to other users, and potentially even used inside links and such,
5947 # it needs to be consistent for all visitors.
5951 }
5953 }
5955 /**
5956 * Get the name of the user that edited the last revision
5957 *
5958 * @return string User name
5959 */
5964 # if this template is subst: the revision id will be blank,
5965 # so just use the current user's name
5970 }
5971 }
5973 }
5975 /**
5976 * Get the size of the revision
5977 *
5978 * @return int|null Revision size
5979 */
5984 # if this variable is subst: the revision id will be blank,
5985 # so just use the parser input size, because the own substituation
5986 # will change the size.
5991 }
5992 }
5994 }
5996 /**
5997 * Mutator for $mDefaultSort
5998 *
5999 * @param string $sort New value
6000 */
6004 }
6006 /**
6007 * Accessor for $mDefaultSort
6008 * Will use the empty string if none is set.
6009 *
6010 * This value is treated as a prefix, so the
6011 * empty string is equivalent to sorting by
6012 * page name.
6013 *
6014 * @return string
6015 */
6021 }
6022 }
6024 /**
6025 * Accessor for $mDefaultSort
6026 * Unlike getDefaultSort(), will return false if none is set
6027 *
6028 * @return string|bool
6029 */
6032 }
6034 /**
6035 * Try to guess the section anchor name based on a wikitext fragment
6036 * presumably extracted from a heading, for example "Header" from
6037 * "== Header ==".
6038 *
6039 * @param string $text
6040 *
6041 * @return string
6042 */
6044 # Strip out wikitext links(they break the anchor)
6048 }
6050 /**
6051 * Same as guessSectionNameFromWikiText(), but produces legacy anchors
6052 * instead. For use in redirects, since IE6 interprets Redirect: headers
6053 * as something other than UTF-8 (apparently?), resulting in breakage.
6054 *
6055 * @param string $text The section name
6056 * @return string An anchor
6057 */
6059 # Strip out wikitext links(they break the anchor)
6063 }
6065 /**
6066 * Strips a text string of wikitext for use in a section anchor
6067 *
6068 * Accepts a text string and then removes all wikitext from the
6069 * string and leaves only the resultant text (i.e. the result of
6070 * [[User:WikiSysop|Sysop]] would be "Sysop" and the result of
6071 * [[User:WikiSysop]] would be "User:WikiSysop") - this is intended
6072 * to create valid section anchors by mimicing the output of the
6073 * parser when headings are parsed.
6074 *
6075 * @param string $text Text string to be stripped of wikitext
6076 * for use in a Section anchor
6077 * @return string Filtered text string
6078 */
6080 # Strip internal link markup
6084 # Strip external link markup
6085 # @todo FIXME: Not tolerant to blank link text
6086 # I.E. [https://www.mediawiki.org] will render as [1] or something depending
6087 # on how many empty links there are on the page - need to figure that out.
6088 $text = preg_replace( '/\[(?i:' . $this->mUrlProtocols . ')([^ ]+?) ([^[]+)\]/', '$2', $text );
6090 # Parse wikitext quotes (italics & bold)
6093 # Strip HTML tags
6096 }
6098 /**
6099 * strip/replaceVariables/unstrip for preprocessor regression testing
6100 *
6101 * @param string $text
6102 * @param Title $title
6103 * @param ParserOptions $options
6104 * @param int $outputType
6105 *
6106 * @return string
6107 */
6108 function testSrvus( $text, Title $title, ParserOptions $options, $outputType = self::OT_HTML ) {
6116 }
6118 /**
6119 * @param string $text
6120 * @param Title $title
6121 * @param ParserOptions $options
6122 * @return string
6123 */
6126 }
6128 /**
6129 * @param string $text
6130 * @param Title $title
6131 * @param ParserOptions $options
6132 * @return string
6133 */
6136 }
6138 /**
6139 * Call a callback function on all regions of the given text that are not
6140 * inside strip markers, and replace those regions with the return value
6141 * of the callback. For example, with input:
6142 *
6143 * aaa<MARKER>bbb
6144 *
6145 * This will call the callback function twice, with 'aaa' and 'bbb'. Those
6146 * two strings will be replaced with the value returned by the callback in
6147 * each case.
6148 *
6149 * @param string $s
6150 * @param callable $callback
6151 *
6152 * @return string
6153 */
6172 }
6173 }
6174 }
6176 }
6178 /**
6179 * Remove any strip markers found in the given text.
6180 *
6181 * @param string $text Input string
6182 * @return string
6183 */
6186 }
6188 /**
6189 * Save the parser state required to convert the given half-parsed text to
6190 * HTML. "Half-parsed" in this context means the output of
6191 * recursiveTagParse() or internalParse(). This output has strip markers
6192 * from replaceVariables (extensionSubstitution() etc.), and link
6193 * placeholders from replaceLinkHolders().
6194 *
6195 * Returns an array which can be serialized and stored persistently. This
6196 * array can later be loaded into another parser instance with
6197 * unserializeHalfParsedText(). The text can then be safely incorporated into
6198 * the return value of a parser hook.
6199 *
6200 * @param string $text
6201 *
6202 * @return array
6203 */
6211 );
6214 }
6216 /**
6217 * Load the parser state given in the $data array, which is assumed to
6218 * have been generated by serializeHalfParsedText(). The text contents is
6219 * extracted from the array, and its markers are transformed into markers
6220 * appropriate for the current Parser instance. This transformed text is
6221 * returned, and can be safely included in the return value of a parser
6222 * hook.
6223 *
6224 * If the $data array has been stored persistently, the caller should first
6225 * check whether it is still valid, by calling isValidHalfParsedText().
6226 *
6227 * @param array $data Serialized data
6228 * @throws MWException
6229 * @return string
6230 */
6234 }
6236 # First, extract the strip state.
6240 # Now renumber links
6243 # Should be good to go.
6245 }
6247 /**
6248 * Returns true if the given array, presumed to be generated by
6249 * serializeHalfParsedText(), is compatible with the current version of the
6250 * parser.
6251 *
6252 * @param array $data
6253 *
6254 * @return bool
6255 */
6258 }
6260 /**
6261 * Parsed a width param of imagelink like 300px or 200x300px
6262 *
6263 * @param string $value
6264 *
6265 * @return array
6266 * @since 1.20
6267 */
6272 }
6274 # (bug 13500) In both cases (width/height and width only),
6275 # permit trailing "px" for backward compatibility.
6284 }
6286 }
6288 /**
6289 * Lock the current instance of the parser.
6290 *
6291 * This is meant to stop someone from calling the parser
6292 * recursively and messing up all the strip state.
6293 *
6294 * @throws MWException If parser is in a parse
6295 * @return ScopedCallback The lock will be released once the return value goes out of scope.
6296 */
6301 }
6307 } );
6310 }
6312 /**
6313 * Strip outer <p></p> tag from the HTML source of a single paragraph.
6314 *
6315 * Returns original HTML if the <p/> tag has any attributes, if there's no wrapping <p/> tag,
6316 * or if there is more than one <p/> tag in the input HTML.
6317 *
6318 * @param string $html
6319 * @return string
6320 * @since 1.24
6321 */
6327 }
6328 }
6331 }
6332 }