| blob | 2066580f6c2824ec197351951345067395db9531 |
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
124 /** @var array */
127 /** @var array */
130 /** @var array */
133 /** @var array */
136 /** @var array */
139 /** @var array */
142 /**
143 * @var array
144 * @todo Unused?
145 */
148 /** @var array */
151 /** @var array */
154 /** @var array */
157 /** @var int */
160 /** @var bool */
163 # Initialised by initialiseVariables()
165 /** @var MagicWordArray */
168 /** @var MagicWordArray */
171 # Initialised in constructor
173 /** @var array */
176 /** @var Parser */
179 /** @var string */
182 /** @var string */
185 # Cleared with clearState():
187 /** @var ParserOutput */
190 /** @var int */
193 /** @var bool */
196 /** @var StripState */
199 /**
200 * @var array
201 * @todo Unused?
202 */
205 /**
206 * @var bool
207 * @todo Unused?
208 */
211 /** @var string */
214 /** @var bool */
217 /** @var LinkHolderArray */
220 /** @var int */
223 /** @var array */
226 /** @var int */
229 /** @var int */
232 /** @var int */
235 /** @var bool|string */
238 /** @var array Empty-frame expansion cache */
241 /** @var array */
244 /** @var array */
247 /** @var array */
250 /** @var array */
253 /** @var int Number of expensive parser function calls */
256 /** @var bool */
259 /** @var bool */
262 /** @var User User object; only used when doing pre-save transform */
265 # Temporary
266 # These are variables reset at least once per parse regardless of $clearState
268 /** @var ParserOptions */
271 /** @var Title Title context, used for self-link rendering and similar things */
274 /** @var array Shortcut alias, see setOutputType() */
277 /** @var string The timestamp of the specified revision ID */
280 /** @var string */
283 /**
284 * @var boolean Recursive call protection.
285 * This variable should be treated as if it were private.
286 */
289 /** @var int Output type, one of the OT_xxx constants */
292 /** @var Revision The revision object of the specified revision ID */
295 /** @var int ID to display in {{REVISIONID}} tags */
298 /** @var string User to display in {{REVISIONUSER}} tag */
301 /** @var int Size to display in {{REVISIONSIZE}} variable */
304 /** @var bool|int For {{PAGESIZE}} on current page. */
307 /**
308 * @var array Array with the language name of each language link (i.e. the
309 * interwiki prefix) in the key, value arbitrary. Used to avoid sending
310 * duplicate language links to the ParserOutput.
311 */
314 /**
315 * @var int The revision ID which was used to fetch the timestamp
316 * @todo Unused?
317 */
320 /**
321 * @param array $conf
322 */
331 # Preprocessor_Hash is much faster than Preprocessor_DOM under HipHop
334 # PECL extension that conflicts with the core DOM extension (bug 13770)
341 }
343 }
345 /**
346 * Reduce memory usage to reduce the impact of circular references
347 */
351 }
354 }
355 }
357 /**
358 * Allow extensions to clean up when the parser is cloned
359 */
363 }
365 /**
366 * Do various kinds of initialisation on the first call of the parser
367 */
371 }
382 }
384 /**
385 * Clear Parser state
386 *
387 * @private
388 */
393 }
410 /**
411 * Prefix for temporary replacement strings for the multipass parser.
412 * \x07 should never appear in input as it's disallowed in XML.
413 * Using it at the front also gives us a little extra robustness
414 * since it shouldn't match when butted up against identifier-like
415 * string constructs.
416 *
417 * Must not consist of all title characters, or else it will change
418 * the behavior of <nowiki> in a link.
419 */
423 # Clear these on every parse, bug 4549
431 );
440 # Fix cloning
443 }
447 }
449 /**
450 * Convert wikitext to HTML
451 * Do not call this function recursively.
452 *
453 * @param string $text text we want to parse
454 * @param Title $title
455 * @param ParserOptions $options
456 * @param bool $linestart
457 * @param bool $clearState
458 * @param int $revid Number to pass in {{REVISIONID}}
459 * @return ParserOutput A ParserOutput
460 */
463 ) {
464 /**
465 * First pass--just handle <nowiki> sections, pass the rest off
466 * to internalParse() which does all the real work.
467 */
476 }
483 }
485 # Remove the strip marker tag prefix from the input, if present.
488 }
501 }
504 # No more strip!
511 # Clean up special characters, only run once, next-to-last before doBlockLevels
513 # french spaces, last one Guillemet-left
514 # only if there is something before the space
516 # french spaces, Guillemet-right
519 );
526 /**
527 * The input doesn't get language converted if
528 * a) It's disabled
529 * b) Content isn't converted
530 * c) It's a conversion table
531 * d) it is an interface message (which is in the user language)
532 */
535 ) {
537 # The position of the convert() call should not be changed. it
538 # assumes that the links are all replaced and the only thing left
539 # is the <nowiki> mark.
541 }
542 }
544 /**
545 * A converted title will be provided in the output object if title and
546 * content conversion are enabled, the article text does not contain
547 * a conversion-suppressing double-underscore tag, and no
548 * {{DISPLAYTITLE:...}} is present. DISPLAYTITLE takes precedence over
549 * automatic link conversion.
550 */
555 ) {
562 }
563 }
577 # attempt to sanitize at least some nesting problems
578 # (bug #2702 and quite a few others)
580 # ''Something [http://www.cool.com cool''] -->
581 # <i>Something</i><a href="http://www.cool.com"..><i>cool></i></a>
584 # fix up an anchor inside another anchor, only
585 # at least for a single single nested link (bug 3695)
588 # fix div inside inline elements- doBlockLevels won't wrap a line which
589 # contains a div, so fix it up here; replace
590 # div with escaped text
593 # remove empty italic or bold tag pairs, some
594 # introduced by rules above
596 );
602 }
608 );
609 }
613 # Information on include size limits, for the benefit of users who try to skirt them
621 );
622 }
627 );
631 );
634 );
637 );
640 );
643 );
646 );
652 }
656 ) ) {
662 }
666 }
667 }
668 }
669 // Since we're not really outputting HTML, decode the entities and
670 // then re-encode the things that need hiding inside HTML comments.
674 // Sanitize for comment. Note '‐' in the replacement is U+2010,
675 // which looks much like the problematic '-'.
682 }
683 }
696 }
698 /**
699 * Recursive parser entry point that can be called from an extension tag
700 * hook.
701 *
702 * If $frame is not provided, then template variables (e.g., {{{1}}}) within $text are not expanded
703 *
704 * @param string $text Text extension wants to have parsed
705 * @param bool|PPFrame $frame The frame to use for expanding any template variables
706 *
707 * @return string
708 */
716 }
718 /**
719 * Expand templates and variables in the text, producing valid, static wikitext.
720 * Also removes comments.
721 * @param string $text
722 * @param Title $title
723 * @param ParserOptions $options
724 * @param int|null $revid
725 * @return mixed|string
726 */
733 }
740 }
742 /**
743 * Recursive parser entry point that can be called from an extension tag
744 * hook.
745 *
746 * @param string $text Text to be expanded
747 * @param bool|PPFrame $frame The frame to use for expanding any template variables
748 * @return string
749 * @since 1.19
750 */
757 }
759 /**
760 * Process the wikitext for the "?preload=" feature. (bug 5210)
761 *
762 * "<noinclude>", "<includeonly>" etc. are parsed as for template
763 * transclusion, comments, templates, arguments, tags hooks and parser
764 * functions are untouched.
765 *
766 * @param string $text
767 * @param Title $title
768 * @param ParserOptions $options
769 * @param array $params
770 * @return string
771 */
772 public function getPreloadText( $text, Title $title, ParserOptions $options, $params = array() ) {
776 # Parser (re)initialisation
785 }
787 /**
788 * Get a random string
789 *
790 * @return string
791 */
794 }
796 /**
797 * Set the current user.
798 * Should only be used when doing pre-save transform.
799 *
800 * @param User|null $user User object or null (to reset)
801 */
804 }
806 /**
807 * Accessor for mUniqPrefix.
808 *
809 * @return string
810 */
813 # @todo FIXME: This is probably *horribly wrong*
814 # LanguageConverter seems to want $wgParser's uniqPrefix, however
815 # if this is called for a parser cache hit, the parser may not
816 # have ever been initialized in the first place.
817 # Not really sure what the heck is supposed to be going on here.
819 # throw new MWException( "Accessing uninitialized mUniqPrefix" );
820 }
822 }
824 /**
825 * Set the context title
826 *
827 * @param Title $t
828 */
832 }
835 # Strip the fragment to avoid various odd effects
840 }
841 }
843 /**
844 * Accessor for the Title object
845 *
846 * @return Title
847 */
850 }
852 /**
853 * Accessor/mutator for the Title object
854 *
855 * @param Title $x Title object or null to just get the current one
856 * @return Title
857 */
860 }
862 /**
863 * Set the output type
864 *
865 * @param int $ot New value
866 */
869 # Shortcut alias
875 );
876 }
878 /**
879 * Accessor/mutator for the output type
880 *
881 * @param int|null $x New value or null to just get the current one
882 * @return int
883 */
886 }
888 /**
889 * Get the ParserOutput object
890 *
891 * @return ParserOutput
892 */
895 }
897 /**
898 * Get the ParserOptions object
899 *
900 * @return ParserOptions object
901 */
904 }
906 /**
907 * Accessor/mutator for the ParserOptions object
908 *
909 * @param ParserOptions $x New value or null to just get the current one
910 * @return ParserOptions Current ParserOptions object
911 */
914 }
916 /**
917 * @return int
918 */
921 }
923 /**
924 * @param int $id
925 */
928 }
930 /**
931 * Get a language object for use in parser functions such as {{FORMATNUM:}}
932 * @return Language
933 */
936 }
938 /**
939 * Get the target language for the content being parsed. This is usually the
940 * language that the content is in.
941 *
942 * @since 1.19
943 *
944 * @throws MWException
945 * @return Language|null
946 */
956 }
959 }
961 /**
962 * Get the language object for language conversion
963 * @return Language|null
964 */
967 }
969 /**
970 * Get a User object either from $this->mUser, if set, or from the
971 * ParserOptions object otherwise
972 *
973 * @return User
974 */
978 }
980 }
982 /**
983 * Get a preprocessor object
984 *
985 * @return Preprocessor
986 */
991 }
993 }
995 /**
996 * Replaces all occurrences of HTML-style comments and the given tags
997 * in the text with a random marker and returns the next text. The output
998 * parameter $matches will be an associative array filled with data in
999 * the form:
1000 *
1001 * @code
1002 * 'UNIQ-xxxxx' => array(
1003 * 'element',
1004 * 'tag content',
1005 * array( 'param' => 'x' ),
1006 * '<element param="x">tag content</element>' ) )
1007 * @endcode
1008 *
1009 * @param array $elements List of element names. Comments are always extracted.
1010 * @param string $text Source text string.
1011 * @param array $matches Out parameter, Array: extracted tags
1012 * @param string $uniq_prefix
1013 * @return string Stripped text
1014 */
1015 public static function extractTagsAndParams( $elements, $text, &$matches, $uniq_prefix = '' ) {
1028 }
1030 # comment
1036 # tag
1041 }
1047 # Empty element tag, <tag />
1056 }
1060 # No end tag -- let it run out to the end of the text.
1066 }
1067 }
1073 }
1075 }
1077 /**
1078 * Get a list of strippable XML-like elements
1079 *
1080 * @return array
1081 */
1084 }
1086 /**
1087 * Add an item to the strip state
1088 * Returns the unique tag which must be inserted into the stripped text
1089 * The tag will be replaced with the original text in unstrip()
1090 *
1091 * @param string $text
1092 *
1093 * @return string
1094 */
1100 }
1102 /**
1103 * parse the wiki syntax used to render tables
1104 *
1105 * @private
1106 * @param string $text
1107 * @return string
1108 */
1127 }
1133 # First check if we are starting a new table
1146 # Don't do any of the following
1150 # We are ending a table
1156 }
1160 }
1164 }
1168 # Now we have a table row
1171 # Whats after the tag is now only attributes
1184 }
1188 }
1197 ) {
1198 # This might be cell elements, td, th or captions
1202 }
1208 }
1210 # Split up multiple cells on the same line.
1211 # FIXME : This can result in improper nesting of tags processed
1212 # by earlier parser steps, but should avoid splitting up eg
1213 # attribute values containing literal "||".
1218 # Loop through each table cell
1225 }
1230 }
1236 }
1246 }
1250 # A cell could contain both parameters and data
1253 # Bug 553: Note that a '|' inside an invalid link should not
1254 # be mistaken as delimiting cell parameters
1263 }
1267 }
1268 }
1270 }
1272 # Closing open td, tr && table
1276 }
1279 }
1282 }
1285 }
1287 # Remove trailing line-ending (b/c)
1290 }
1292 # special case: don't return empty table
1295 }
1300 }
1302 /**
1303 * Helper function for parse() that transforms wiki markup into
1304 * HTML. Only called for $mOutputType == self::OT_HTML.
1305 *
1306 * @private
1307 *
1308 * @param string $text
1309 * @param bool $isMain
1310 * @param bool $frame
1311 *
1312 * @return string
1313 */
1319 # Hook to suspend the parser in this state
1320 if ( !wfRunHooks( 'ParserBeforeInternalParse', array( &$this, &$text, &$this->mStripState ) ) ) {
1323 }
1325 # if $frame is provided, then use $frame for replacing any variables
1327 # use frame depth to infer how include/noinclude tags should be handled
1328 # depth=0 means this is the top-level document; otherwise it's an included document
1333 }
1337 # if $frame is not provided, then use old-style replaceVariables
1339 }
1347 );
1350 # Tables need to come after variable replacement for things to work
1351 # properly; putting them before other transformations should keep
1352 # exciting things like link expansions from showing up in surprising
1353 # places.
1365 # replaceInternalLinks may sometimes leave behind
1366 # absolute URLs, which have to be masked to hide them from replaceExternalLinks
1374 }
1376 /**
1377 * Replace special strings like "ISBN xxx" and "RFC xxx" with
1378 * magic external links.
1379 *
1380 * DML
1381 * @private
1382 *
1383 * @param string $text
1384 *
1385 * @return string
1386 */
1392 '!(?: # Start cases
1396 (?:RFC|PMID)\s+([0-9]+) | # m[4]: RFC or PMID, capture number
1398 (?: 97[89] [\ \-]? )? # optional 13-digit ISBN prefix
1399 (?: [0-9] [\ \-]? ){9} # 9 digits with opt. delimiters
1400 [0-9Xx] # check digit
1405 }
1407 /**
1408 * @throws MWException
1409 * @param array $m
1410 * @return HTML|string
1411 */
1414 # Skip anchor
1417 # Skip HTML element
1420 # Free external link
1423 # RFC or PMID
1437 }
1441 # ISBN
1447 ));
1454 }
1455 }
1457 /**
1458 * Make a free external link, given a user-supplied URL
1459 *
1460 * @param string $url
1461 *
1462 * @return string HTML
1463 * @private
1464 */
1470 # The characters '<' and '>' (which were escaped by
1471 # removeHTMLtags()) should not be included in
1472 # URLs, per RFC 2396.
1477 }
1479 # Move trailing punctuation to $trail
1481 # If there is no left bracket, then consider right brackets fair game too
1484 }
1490 }
1494 # Is this an external image?
1497 # Not an image, make a link
1502 # Register it in the output object...
1503 # Replace unnecessary URL escape codes with their equivalent characters
1506 }
1509 }
1511 /**
1512 * Parse headers and return html
1513 *
1514 * @private
1515 *
1516 * @param string $text
1517 *
1518 * @return string
1519 */
1525 }
1528 }
1530 /**
1531 * Replace single quotes with HTML markup
1532 * @private
1533 *
1534 * @param string $text
1535 *
1536 * @return string the altered text
1537 */
1544 }
1548 }
1550 /**
1551 * Helper function for doAllQuotes()
1552 *
1553 * @param string $text
1554 *
1555 * @return string
1556 */
1562 }
1564 // First, do some preliminary work. This may shift some apostrophes from
1565 // being mark-up to being text. It also counts the number of occurrences
1566 // of bold and italics mark-ups.
1571 // If there are ever four apostrophes, assume the first is supposed to
1572 // be text, and the remaining three constitute mark-up for bold text.
1573 // (bug 13227: ''''foo'''' turns into ' ''' foo ' ''')
1579 // If there are more than 5 apostrophes in a row, assume they're all
1580 // text except for the last 5.
1581 // (bug 13227: ''''''foo'''''' turns into ' ''''' foo ' ''''')
1585 }
1586 // Count the number of occurrences of bold and italics mark-ups.
1594 }
1595 }
1597 // If there is an odd number of both bold and italics, it is likely
1598 // that one of the bold ones was meant to be an apostrophe followed
1599 // by italics. Which one we cannot know for certain, but it is more
1600 // likely to be one that has a single-letter word before it.
1612 }
1616 // if $firstsingleletterword is set, we don't
1617 // look at the other options, so we can bail early.
1619 }
1623 }
1624 }
1625 }
1626 }
1628 // If there is a single-letter word, use it!
1633 // If not, but there's a multi-letter word, use that one.
1637 // ... otherwise use the first one that has neither.
1638 // (notice that it is possible for all three to be -1 if, for example,
1639 // there is only one pentuple-apostrophe in the line)
1642 }
1643 }
1645 // Now let's actually convert our apostrophic mush to HTML!
1656 }
1675 }
1692 }
1712 }
1713 }
1714 }
1716 }
1717 // Now close all remaining tags. Notice that the order is important.
1720 }
1723 }
1726 }
1727 // There might be lonely ''''', so make sure we have a buffer
1730 }
1732 }
1734 /**
1735 * Replace external links (REL)
1736 *
1737 * Note: this is all very hackish and the order of execution matters a lot.
1738 * Make sure to run tests/parserTests.php if you change this code.
1739 *
1740 * @private
1741 *
1742 * @param string $text
1743 *
1744 * @throws MWException
1745 * @return string
1746 */
1755 }
1765 # The characters '<' and '>' (which were escaped by
1766 # removeHTMLtags()) should not be included in
1767 # URLs, per RFC 2396.
1772 }
1774 # If the link text is an image URL, replace it with an <img> tag
1775 # This happened by accident in the original parser, but some people used it extensively
1779 }
1783 # Set linktype for CSS - if URL==text, link is essentially free
1786 # No link text, e.g. [http://domain.tld/some.link]
1788 # Autonumber
1793 # Have link text, e.g. [http://domain.tld/some.link text]s
1794 # Check for trail
1796 }
1802 # Use the encoded URL
1803 # This means that users can paste URLs directly into the text
1804 # Funny characters like ö aren't valid in URLs anyway
1805 # This was changed in August 2004
1809 # Register link in the output object.
1810 # Replace unnecessary URL escape codes with the referenced character
1811 # This prevents spammers from hiding links from the filters
1814 }
1818 }
1820 /**
1821 * Get the rel attribute for a particular external link.
1822 *
1823 * @since 1.21
1824 * @param string|bool $url Optional URL, to extract the domain from for rel =>
1825 * nofollow if appropriate
1826 * @param Title $title Optional Title, for wgNoFollowNsExceptions lookups
1827 * @return string|null Rel attribute for $url
1828 */
1834 ) {
1836 }
1838 }
1840 /**
1841 * Get an associative array of additional HTML attributes appropriate for a
1842 * particular external link. This currently may include rel => nofollow
1843 * (depending on configuration, namespace, and the URL's domain) and/or a
1844 * target attribute (depending on configuration).
1845 *
1846 * @param string|bool $url Optional URL, to extract the domain from for rel =>
1847 * nofollow if appropriate
1848 * @return array Associative array of HTML attributes
1849 */
1856 }
1858 }
1860 /**
1861 * Replace unusual URL escape codes with their equivalent characters
1862 *
1863 * @param string $url
1864 * @return string
1865 *
1866 * @todo This can merge genuinely required bits in the path or query string,
1867 * breaking legit URLs. A proper fix would treat the various parts of
1868 * the URL differently; as a workaround, just use the output for
1869 * statistical records, not for actual linking/output.
1870 */
1874 }
1876 /**
1877 * Callback function used in replaceUnusualEscapes().
1878 * Replaces unusual URL escape codes with their equivalent character
1879 *
1880 * @param array $matches
1881 *
1882 * @return string
1883 */
1887 # Is it an unsafe or HTTP reserved character according to RFC 1738?
1889 # No, shouldn't be escaped
1892 # Yes, leave it escaped
1894 }
1895 }
1897 /**
1898 * make an image if it's allowed, either through the global
1899 * option, through the exception, or through the on-wiki whitelist
1900 *
1901 * @param string $url
1902 *
1903 * @return string
1904 */
1909 # $imagesfrom could be either a single string or an array of strings, parse out the latter
1916 }
1917 }
1922 }
1926 ) {
1928 # Image found
1930 }
1931 }
1934 ) {
1938 );
1941 # Sanitize the regex fragment, make it case-insensitive, ignore blank entries/comments
1944 }
1946 # Image matches a whitelist entry
1949 }
1950 }
1951 }
1953 }
1955 /**
1956 * Process [[ ]] wikilinks
1957 *
1958 * @param string $s
1959 *
1960 * @return string Processed text
1961 *
1962 * @private
1963 */
1967 }
1969 /**
1970 * Process [[ ]] wikilinks (RIL)
1971 * @param string $s
1972 * @throws MWException
1973 * @return LinkHolderArray
1974 *
1975 * @private
1976 */
1982 # the % is needed to support urlencoded titles as well
1985 # Match a link having the form [[namespace:link|alternate]]trail
1987 # Match cases where there is no "]]", which might still be images
1989 }
1993 # split the entire text string on occurrences of [[
1995 # get the first element (all text up to first [[), and remove the space we added
2004 # Match the end of a line for a word that's not followed by whitespace,
2005 # e.g. in the case of 'The Arab al[[Razi]]', 'al' will be matched
2009 }
2015 }
2024 }
2027 }
2032 // @codingStandardsIgnoreStart Squiz.WhiteSpace.SemicolonSpacing.Incorrect
2033 # Loop for each link
2035 // @codingStandardsIgnoreStart
2037 # Check for excessive memory usage
2039 # Too big
2040 # Do the existence check, replace the link holders and clear the array
2043 }
2052 }
2053 # first link
2057 }
2059 }
2066 # If we get a ] at the beginning of $m[3] that means we have a link that's something like:
2067 # [[Image:Foo.jpg|[http://example.com desc]]] <- having three ] in a row fucks up,
2068 # the real problem is with the $e1 regex
2069 # See bug 1300.
2070 #
2071 # Still some problems for cases where the ] is meant to be outside punctuation,
2072 # and no image is in sight. See bug 2095.
2073 #
2077 ) {
2080 }
2081 # fix up urlencoded title texts
2083 # Should anchors '#' also be rejected?
2085 }
2088 # Invalid, but might be an image with a link in its caption
2093 }
2099 }
2103 # Don't allow internal links to pages containing
2104 # PROTO: where PROTO is a valid URL protocol; these
2105 # should be external links.
2110 }
2112 # Make subpage if necessary
2117 }
2121 # Strip off leading ':'
2123 }
2132 }
2143 # look at the next 'line' to see if we can close it there
2148 }
2151 # the first ]] closes the inner link, the second the image
2157 # if there's exactly one ]] that's fine, we'll keep looking
2160 # if $next_line is invalid too, we need look no further
2163 }
2164 }
2166 # we couldn't find the end of this imageLink, so output it raw
2167 # but don't ignore what might be perfectly normal links in the text we've examined
2170 # note: no $trail, because without an end, there *is* no trail
2173 }
2176 # note: no $trail, because without an end, there *is* no trail
2179 }
2181 }
2187 # Bug 4598 madness. Handle the quotes only if they come from the alternate part
2188 # [[Lista d''e paise d''o munno]] -> <a href="...">Lista d''e paise d''o munno</a>
2189 # [[Criticism of Harry Potter|Criticism of ''Harry Potter'']]
2190 # -> <a href="Criticism of Harry Potter">Criticism of <i>Harry Potter</i></a>
2192 }
2194 # Link not escaped by : , create the various objects
2196 # Interwikis
2200 ) {
2201 // XXX: the above check prevents links to sites with identifiers that are not language codes
2203 # Bug 24502: filter duplicates
2207 }
2213 }
2220 # if no parameters were passed, $text
2221 # becomes something like "File:Foo.png",
2222 # which we don't want to pass on to the
2223 # image generator
2226 # recursively parse links inside the image caption
2227 # actually, this will parse them in any other parameters, too,
2228 # but it might be hard to fix that, and it doesn't matter ATM
2231 }
2232 # cloak any absolute URLs inside the image markup, so replaceExternalLinks() won't touch them
2237 }
2240 }
2250 }
2256 /**
2257 * Strip the whitespace Category links produce, see bug 87
2258 */
2263 }
2264 }
2266 # Self-link checking. For some languages, variants of the title are checked in
2267 # LinkHolderArray::doVariants() to allow batching the existence checks necessary
2268 # for linking to a different variant.
2272 }
2274 # NS_MEDIA is a pseudo-namespace for linking directly to a file
2275 # @todo FIXME: Should do batch file existence checks, see comment below
2278 # Give extensions a chance to select the file revision for us
2283 # Fetch and register the file (file title may be different via hooks)
2285 # Cloak with NOPARSE to avoid replacement in replaceExternalLinks
2290 }
2293 # Some titles, such as valid special pages or files in foreign repos, should
2294 # be shown as bluelinks even though they're not included in the page table
2295 #
2296 # @todo FIXME: isAlwaysKnown() can be expensive for file links; we should really do
2297 # batch file existence checks for NS_FILE and NS_MEDIA
2302 # Links will be added to the output link list after checking
2304 }
2306 }
2309 }
2311 /**
2312 * Render a forced-blue link inline; protect against double expansion of
2313 * URLs if we're in a mode that prepends full URL prefixes to internal links.
2314 * Since this little disaster has to split off the trail text to avoid
2315 * breaking URLs in the following text without breaking trails on the
2316 * wiki links, it's been made into a horrible function.
2317 *
2318 * @param Title $nt
2319 * @param string $text
2320 * @param array|string $query
2321 * @param string $trail
2322 * @param string $prefix
2323 * @return string HTML-wikitext mix oh yuck
2324 */
2330 }
2333 }
2338 }
2340 /**
2341 * Insert a NOPARSE hacky thing into any inline links in a chunk that's
2342 * going to go through further parsing steps before inline URL expansion.
2343 *
2344 * Not needed quite as much as it used to be since free links are a bit
2345 * more sensible these days. But bracketed links are still an issue.
2346 *
2347 * @param string $text More-or-less HTML
2348 * @return string Less-or-more HTML with NOPARSE bits
2349 */
2353 }
2355 /**
2356 * Return true if subpage links should be expanded on this page.
2357 * @return bool
2358 */
2360 # Some namespaces don't allow subpages
2362 }
2364 /**
2365 * Handle link to subpage if necessary
2366 *
2367 * @param string $target The source of the link
2368 * @param string &$text The link text, modified as necessary
2369 * @return string The full name of the link
2370 * @private
2371 */
2374 }
2376 /**#@+
2377 * Used by doBlockLevels()
2378 * @private
2379 *
2380 * @return string
2381 */
2386 }
2390 }
2392 /**
2393 * getCommon() returns the length of the longest common substring
2394 * of both arguments, starting at the beginning of both.
2395 * @private
2396 *
2397 * @param string $st1
2398 * @param string $st2
2399 *
2400 * @return int
2401 */
2407 }
2412 }
2413 }
2415 }
2417 /**
2418 * These next three functions open, continue, and close the list
2419 * element appropriate to the prefix character passed into them.
2420 * @private
2421 *
2422 * @param string $char
2423 *
2424 * @return string
2425 */
2440 }
2443 }
2445 /**
2446 * TODO: document
2447 * @param string $char
2448 * @private
2449 *
2450 * @return string
2451 */
2459 }
2466 }
2467 }
2469 }
2471 /**
2472 * @todo Document
2473 * @param string $char
2474 * @private
2475 *
2476 * @return string
2477 */
2489 }
2492 }
2494 }
2495 /**#@-*/
2497 /**
2498 * Make lists from lines starting with ':', '*', '#', etc. (DBL)
2499 *
2500 * @param string $text
2501 * @param bool $linestart Whether or not this is at the start of a line.
2502 * @private
2503 * @return string The lists rendered as HTML
2504 */
2508 # Parsing through the text line by line. The main thing
2509 # happening here is handling of block-level elements p, pre,
2510 # and making lists from lines starting with * # : etc.
2511 #
2521 # Fix up $linestart
2526 }
2527 # * = ul
2528 # # = ol
2529 # ; = dt
2530 # : = dd
2535 # If not in a <pre> element, scan for and figure out what prefixes are there.
2537 # Multiple prefixes may abut each other for nested lists.
2541 # eh?
2542 # ; and : are both from definition-lists, so they're equivalent
2543 # for the purposes of determining whether or not we need to open/close
2544 # elements.
2549 # Don't interpret any other prefixes in preformatted text
2553 }
2555 # List generation
2557 # Same as the last item, so no need to deal with nesting or opening stuff
2562 # The one nasty exception: definition lists work like this:
2563 # ; title : definition text
2564 # So we check for : in the remainder text to split up the
2565 # title and definition, without b0rking links.
2570 }
2571 }
2573 # We need to open or close prefixes, or both.
2575 # Either open or close a level...
2579 # Close all the prefixes which aren't shared.
2583 }
2585 # Continue the current prefix if appropriate.
2588 }
2590 # Open prefixes where appropriate.
2596 # @todo FIXME: This is dupe of code above
2600 }
2601 }
2603 }
2605 }
2607 # If we have no prefixes, go to paragraph mode.
2610 # No prefix (not in list)--go to paragraph mode
2611 # XXX: use a stack for nestable elements like span, table and div
2613 '/(?:<table|<h1|<h2|<h3|<h4|<h5|<h6|<pre|<tr|'
2615 $t
2616 );
2622 $t
2623 );
2627 # @todo bug 5718: paragraph closed
2631 }
2633 while ( preg_match( '/<(\\/?)blockquote[\s>]/i', $t, $bqMatch, PREG_OFFSET_CAPTURE, $bqOffset ) ) {
2636 }
2642 ) {
2643 # pre
2648 }
2651 # paragraph
2664 }
2665 }
2674 }
2675 }
2676 }
2677 }
2679 }
2680 # somewhere above we forget to get out of pre block (bug 785)
2683 }
2686 }
2687 }
2691 }
2695 }
2699 }
2701 /**
2702 * Split up a string on ':', ignoring any occurrences inside tags
2703 * to prevent illegal overlapping.
2704 *
2705 * @param string $str The string to split
2706 * @param string &$before Set to everything before the ':'
2707 * @param string &$after Set to everything after the ':'
2708 * @throws MWException
2709 * @return string The position of the ':', or false if none found
2710 */
2716 # Nothing to find!
2719 }
2723 # Easy; no tag nesting to worry about
2728 }
2730 # Ugly state machine to walk through avoiding tags.
2738 # (Using the number is a performance hack for common cases)
2742 # Could be either a <start> tag or an </end> tag
2747 # We found it!
2752 }
2753 # Embedded in a tag; don't break it.
2756 # Skip ahead looking for something interesting
2759 # Nothing else interesting
2762 }
2766 # We found it!
2771 }
2772 }
2774 # Nothing else interesting to find; abort!
2775 # We're nested, but there's no close tags left. Abort!
2777 }
2778 # Skip ahead to next tag start
2781 }
2784 # In a <tag>
2791 # Slash may be followed by >?
2795 # ignore
2796 }
2807 # Illegal early close? This shouldn't happen D:
2812 }
2815 # In a </tag>
2822 }
2824 }
2828 # Yes, a self-closed tag <blah/>
2831 # Probably we're jumping the gun, and this is an attribute
2833 }
2838 }
2845 }
2852 }
2857 }
2858 }
2860 wfDebug( __METHOD__ . ": Invalid input; not enough close tags (stack $stack, state $state)\n" );
2863 }
2866 }
2868 /**
2869 * Return value of a magic variable (like PAGENAME)
2870 *
2871 * @private
2872 *
2873 * @param int $index
2874 * @param bool|PPFrame $frame
2875 *
2876 * @throws MWException
2877 * @return string
2878 */
2884 // If no title set, bad things are going to happen
2885 // later. Title should always be set since this
2886 // should only be called in the middle of a parse
2887 // operation (but the unit-tests do funky stuff)
2890 }
2892 /**
2893 * Some of these require message or data lookups and can be
2894 * expensive to check many times.
2895 */
2899 }
2900 }
2942 $value = $pageLang->getMonthAbbreviation( MWTimestamp::getLocalInstance( $ts )->format( 'n' ) );
2976 ) ) );
2986 ) ) );
2994 }
3002 }
3015 # 0 means the page doesn't exist in the database,
3016 # which means the user is previewing a new page.
3017 # The vary-revision flag must be set, because the magic word
3018 # will have a different value once the page is saved.
3021 }
3025 # Let the edit saving system know we should parse the page
3026 # *after* a revision ID has been assigned.
3032 # Let the edit saving system know we should parse the page
3033 # *after* a revision ID has been assigned. This is for null edits.
3039 # Let the edit saving system know we should parse the page
3040 # *after* a revision ID has been assigned. This is for null edits.
3046 # Let the edit saving system know we should parse the page
3047 # *after* a revision ID has been assigned. This is for null edits.
3053 # Let the edit saving system know we should parse the page
3054 # *after* a revision ID has been assigned. This is for null edits.
3060 # Let the edit saving system know we should parse the page
3061 # *after* a revision ID has been assigned. This is for null edits.
3067 # Let the edit saving system know we should parse the page
3068 # *after* a revision ID has been assigned. This is for null edits.
3074 # Let the edit saving system know we should parse the page
3075 # *after* a revision ID has been assigned. This is for null edits.
3081 # Let the edit saving system know we should parse the page
3082 # *after* a revision ID has been assigned. This is for null edits.
3123 # @bug 4594 PHP5 has it zero padded, PHP4 does not, cast to
3124 # int to remove the padding
3133 );
3142 false
3143 );
3149 # @bug 4594 PHP5 has it zero padded, PHP4 does not, cast to
3150 # int to remove the padding
3215 );
3218 }
3222 }
3225 }
3227 /**
3228 * initialise the magic variables (like CURRENTMONTHNAME) and substitution modifiers
3229 *
3230 * @private
3231 */
3240 }
3242 /**
3243 * Preprocess some wikitext and return the document tree.
3244 * This is the ghost of replace_variables().
3245 *
3246 * @param string $text The text to parse
3247 * @param int $flags Bitwise combination of:
3248 * - self::PTD_FOR_INCLUSION: Handle "<noinclude>" and "<includeonly>" as if the text is being
3249 * included. Default is to assume a direct page view.
3250 *
3251 * The generated DOM tree must depend only on the input text and the flags.
3252 * The DOM tree must be the same in OT_HTML and OT_WIKI mode, to avoid a regression of bug 4899.
3253 *
3254 * Any flag added to the $flags parameter here, or any other parameter liable to cause a
3255 * change in the DOM tree for a given text, must be passed through the section identifier
3256 * in the section edit link and thus back to extractSections().
3257 *
3258 * The output of this function is currently only cached in process memory, but a persistent
3259 * cache may be implemented at a later date which takes further advantage of these strict
3260 * dependency requirements.
3261 *
3262 * @return PPNode
3263 */
3267 }
3269 /**
3270 * Return a three-element array: leading whitespace, string contents, trailing whitespace
3271 *
3272 * @param string $s
3273 *
3274 * @return array
3275 */
3285 }
3287 }
3289 /**
3290 * Replace magic variables, templates, and template arguments
3291 * with the appropriate text. Templates are substituted recursively,
3292 * taking care to avoid infinite loops.
3293 *
3294 * Note that the substitution depends on value of $mOutputType:
3295 * self::OT_WIKI: only {{subst:}} templates
3296 * self::OT_PREPROCESS: templates but not extension tags
3297 * self::OT_HTML: all templates and extension tags
3298 *
3299 * @param string $text The text to transform
3300 * @param bool|PPFrame $frame Object describing the arguments passed to the
3301 * template. Arguments may also be provided as an associative array, as
3302 * was the usual case before MW1.12. Providing arguments this way may be
3303 * useful for extensions wishing to perform variable replacement
3304 * explicitly.
3305 * @param bool $argsOnly Only do argument (triple-brace) expansion, not
3306 * double-brace expansion.
3307 * @return string
3308 */
3310 # Is there any text? Also, Prevent too big inclusions!
3313 }
3322 }
3330 }
3332 /**
3333 * Clean up argument array - refactored in 1.9 so parserfunctions can use it, too.
3334 *
3335 * @param array $args
3336 *
3337 * @return array
3338 */
3351 }
3354 }
3355 }
3356 }
3359 }
3361 /**
3362 * Warn the user when a parser limitation is reached
3363 * Will warn at most once the user per limitation type
3364 *
3365 * @param string $limitationType Should be one of:
3366 * 'expensive-parserfunction' (corresponding messages:
3367 * 'expensive-parserfunction-warning',
3368 * 'expensive-parserfunction-category')
3369 * 'post-expand-template-argument' (corresponding messages:
3370 * 'post-expand-template-argument-warning',
3371 * 'post-expand-template-argument-category')
3372 * 'post-expand-template-inclusion' (corresponding messages:
3373 * 'post-expand-template-inclusion-warning',
3374 * 'post-expand-template-inclusion-category')
3375 * 'node-count-exceeded' (corresponding messages:
3376 * 'node-count-exceeded-warning',
3377 * 'node-count-exceeded-category')
3378 * 'expansion-depth-exceeded' (corresponding messages:
3379 * 'expansion-depth-exceeded-warning',
3380 * 'expansion-depth-exceeded-category')
3381 * @param string|int|null $current Current value
3382 * @param string|int|null $max Maximum allowed, when an explicit limit has been
3383 * exceeded, provide the values (optional)
3384 */
3386 # does no harm if $current and $max are present but are unnecessary for the message
3391 }
3393 /**
3394 * Return the text of a template, after recursively
3395 * replacing any variables or templates within the template.
3396 *
3397 * @param array $piece The parts of the template
3398 * $piece['title']: the title, i.e. the part before the |
3399 * $piece['parts']: the parameter array
3400 * $piece['lineStart']: whether the brace was at the start of a line
3401 * @param PPFrame $frame The current frame, contains template arguments
3402 * @throws Exception
3403 * @return string The text of the template
3404 */
3409 // Flags
3411 // $text has been filled
3413 // wiki markup in $text should be escaped
3415 // $text is HTML, armour it against wikitext transformation
3417 // Force interwiki transclusion to be done in raw mode not rendered
3419 // $text is a DOM node needing expansion in a child frame
3421 // $text is a DOM node needing expansion in the current frame
3424 # Title object, where $text came from
3427 # $part1 is the bit before the first |, and must contain only title characters.
3428 # Various prefixes will be stripped from it later.
3433 # Original title text preserved for various purposes
3436 # $args is a list of argument nodes, starting from index 0, not including $part1
3437 # @todo FIXME: If piece['parts'] is null then the call to getLength()
3438 # below won't work b/c this $args isn't an object
3444 # SUBST
3450 # Possibilities for substMatch: "subst", "safesubst" or FALSE
3451 # Decide whether to expand template or keep wikitext as-is.
3457 }
3463 }
3464 }
3469 }
3470 }
3472 # Variables
3479 }
3481 }
3482 }
3484 # MSG, MSGNW and RAW
3486 # Check for MSGNW:
3491 # Remove obsolete MSG:
3494 }
3496 # Check for RAW:
3500 }
3501 }
3504 # Parser functions
3514 }
3521 }
3523 # The interface for parser functions allows for extracting
3524 # flags into the local scope. Extract any forwarded flags
3525 # here.
3527 }
3529 }
3531 # Finish mangling title and then check for loops.
3532 # Set $title to a Title object and $titleText to the PDBK
3535 # Split the title into page and subpage
3541 }
3545 # Check for language variants if the template is not found
3548 }
3549 # Do recursion depth check
3557 }
3558 }
3559 }
3561 # Load from database
3564 # Too many unique items can kill profiling DBs/collectors
3567 }
3573 ) {
3574 // Pass the template arguments as URL parameters.
3575 // "uselang" will have no effect since the Language object
3576 // is forced to the one defined in ParserOptions.
3585 }
3586 }
3588 // Create a new context to execute the special page
3601 }
3611 }
3612 }
3614 # If the title is valid but undisplayable, make a link to it
3618 }
3620 # Interwiki transclusion
3626 # Preprocess it like a template
3629 }
3631 }
3633 # Do infinite loop check
3634 # This has to be done after redirect resolution to avoid infinite loops via redirects
3641 }
3643 }
3645 # If we haven't found text to substitute by now, we're done
3646 # Recover the source wikitext and return it
3651 }
3654 }
3656 # Expand DOM-style return values in a child frame
3658 # Clean up argument array
3664 # Expansion is eligible for the empty-frame cache
3670 }
3672 # Uncached expansion
3674 }
3675 }
3679 }
3683 }
3685 # Replace raw HTML by a placeholder
3689 # Escape nowiki-style return values
3694 ) {
3695 # Bug 529: if the template begins with a table or block-level
3696 # element, it should be treated as beginning a new line.
3697 # This behavior is somewhat controversial.
3699 }
3702 # Error, oversize inclusion
3704 # Make a working, properly escaped link if possible (bug 23588)
3707 # This will probably not be a working link, but at least it may
3708 # provide some hint of where the problem is
3711 }
3715 }
3721 }
3725 }
3727 /**
3728 * Call a parser function and return an array with text and flags.
3729 *
3730 * The returned array will always contain a boolean 'found', indicating
3731 * whether the parser function was found or not. It may also contain the
3732 * following:
3733 * text: string|object, resulting wikitext or PP DOM object
3734 * isHTML: bool, $text is HTML, armour it against wikitext transformation
3735 * isChildObj: bool, $text is a DOM node needing expansion in a child frame
3736 * isLocalObj: bool, $text is a DOM node needing expansion in the current frame
3737 * nowiki: bool, wiki markup in $text should be escaped
3738 *
3739 * @since 1.21
3740 * @param PPFrame $frame The current frame, contains template arguments
3741 * @param string $function Function name
3742 * @param array $args Arguments to the function
3743 * @throws MWException
3744 * @return array
3745 */
3751 # Case sensitive functions
3755 # Case insensitive functions
3762 }
3763 }
3768 # Workaround for PHP bug 35229 and similar
3773 }
3777 # Convert arguments to PPNodes and collect for appending to $allArgs
3784 }
3785 }
3787 # Add a frame parameter, and pass the arguments as an array
3791 # Convert arguments to plain text and append to $allArgs
3799 }
3800 }
3801 }
3805 # The interface for function hooks allows them to return a wikitext
3806 # string or an array containing the string and any flags. This mungs
3807 # things around to match what this method should return.
3812 );
3816 }
3820 );
3821 }
3827 }
3830 }
3835 }
3840 }
3842 /**
3843 * Get the semi-parsed DOM representation of a template with a given title,
3844 * and its redirect destination title. Cached.
3845 *
3846 * @param Title $title
3847 *
3848 * @return array
3849 */
3858 }
3861 }
3863 # Cache miss, go to the database
3869 }
3877 }
3880 }
3882 /**
3883 * Fetch the unparsed text of a template and register a reference to it.
3884 * @param Title $title
3885 * @return array ( string or false, Title )
3886 */
3888 // Defaults to Parser::statelessFetchTemplate()
3897 // If we transclude ourselves, the final result
3898 // will change based on the new version of the page
3900 }
3901 }
3902 }
3904 }
3906 /**
3907 * Fetch the unparsed text of a template and register a reference to it.
3908 * @param Title $title
3909 * @return string|bool
3910 */
3914 }
3916 /**
3917 * Static function to get a template
3918 * Can be overridden via ParserOptions::setTemplateCallback().
3919 *
3920 * @param Title $title
3921 * @param bool|Parser $parser
3922 *
3923 * @return array
3924 */
3930 # Loop to fetch the article, with up to 1 redirect
3932 # Give extensions a chance to select the revision instead
3943 );
3945 }
3946 # Get the revision
3951 # If there is no current revision, there is no page
3955 }
3962 # We fetched a rev from a different title; register it too...
3967 }
3976 }
3983 }
3988 }
3991 }
3992 # Redirect?
3995 }
4000 }
4002 /**
4003 * Fetch a file and its title and register a reference to it.
4004 * If 'broken' is a key in $options then the file will appear as a broken thumbnail.
4005 * @param Title $title
4006 * @param array $options Array of options to RepoGroup::findFile
4007 * @return File|bool
4008 */
4012 }
4014 /**
4015 * Fetch a file and its title and register a reference to it.
4016 * If 'broken' is a key in $options then the file will appear as a broken thumbnail.
4017 * @param Title $title
4018 * @param array $options Array of options to RepoGroup::findFile
4019 * @return array ( File or false, Title of file )
4020 */
4026 # Register the file as a dependency...
4029 # Update fetched file title
4032 }
4034 }
4036 /**
4037 * Helper function for fetchFileAndTitle.
4038 *
4039 * Also useful if you need to fetch a file but not use it yet,
4040 * for example to get the file's handler.
4041 *
4042 * @param Title $title
4043 * @param array $options Array of options to RepoGroup::findFile
4044 * @return File|bool
4045 */
4053 }
4055 }
4057 /**
4058 * Transclude an interwiki link.
4059 *
4060 * @param Title $title
4061 * @param string $action
4062 *
4063 * @return string
4064 */
4070 }
4076 }
4078 }
4080 /**
4081 * @param string $url
4082 * @return mixed|string
4083 */
4092 }
4099 // Though we failed to fetch the content, this status is useless.
4104 }
4111 ) );
4113 }
4115 /**
4116 * Triple brace replacement -- used for template arguments
4117 * @private
4118 *
4119 * @param array $piece
4120 * @param PPFrame $frame
4121 *
4122 * @return array
4123 */
4137 )
4138 ) {
4139 # No match in frame, use the supplied default
4141 }
4145 }
4148 # No match anywhere
4150 }
4153 }
4158 }
4162 }
4164 /**
4165 * Return the text to be used for a given extension tag.
4166 * This is the ghost of strip().
4167 *
4168 * @param array $params Associative array of parameters:
4169 * name PPNode for the tag name
4170 * attr PPNode for unparsed text where tag attributes are thought to be
4171 * attributes Optional associative array of parsed attributes
4172 * inner Contents of extension element
4173 * noClose Original text did not have a close tag
4174 * @param PPFrame $frame
4175 *
4176 * @throws MWException
4177 * @return string
4178 */
4192 }
4198 }
4201 # Workaround for PHP bug 35229 and similar
4204 }
4211 }
4217 }
4220 # Extract flags to local scope (to override $markerType)
4225 }
4229 }
4234 }
4235 }
4241 }
4242 }
4252 }
4254 }
4256 /**
4257 * Increment an include size counter
4258 *
4259 * @param string $type The type of expansion
4260 * @param int $size The size of the text
4261 * @return bool False if this inclusion would take it over the maximum, true otherwise
4262 */
4269 }
4270 }
4272 /**
4273 * Increment the expensive function count
4274 *
4275 * @return bool False if the limit has been exceeded
4276 */
4280 }
4282 /**
4283 * Strip double-underscore items like __NOGALLERY__ and __NOTOC__
4284 * Fills $this->mDoubleUnderscores, returns the modified text
4285 *
4286 * @param string $text
4287 *
4288 * @return string
4289 */
4293 # The position of __TOC__ needs to be recorded
4299 # Set a placeholder. At the end we'll fill it in with the TOC.
4302 # Only keep the first one.
4304 }
4306 # Now match and remove the rest of them
4312 }
4315 }
4318 ) {
4320 }
4321 # (bug 8068) Allow control over whether robots index a page.
4322 #
4323 # @todo FIXME: Bug 14899: __INDEX__ always overrides __NOINDEX__ here! This
4324 # is not desirable, the last one on the page should win.
4328 }
4332 }
4334 # Cache all double underscores in the database
4337 }
4341 }
4343 /**
4344 * Add a tracking category, getting the title from a system message,
4345 * or print a debug message if the title is invalid.
4346 *
4347 * Please add any message that you use with this function to
4348 * $wgTrackingCategories. That way they will be listed on
4349 * Special:TrackingCategories.
4350 *
4351 * @param string $msg Message key
4352 * @return bool Whether the addition was successful
4353 */
4358 }
4359 // Important to parse with correct title (bug 31469)
4365 # Allow tracking categories to be disabled by setting them to "-"
4368 }
4377 }
4378 }
4380 /**
4381 * This function accomplishes several tasks:
4382 * 1) Auto-number headings if that option is enabled
4383 * 2) Add an [edit] link to sections for users who have enabled the option and can edit the page
4384 * 3) Add a Table of contents on the top for users who have enabled the option
4385 * 4) Auto-anchor headings
4386 *
4387 * It loops through all headlines, collects the necessary data, then splits up the
4388 * string and re-inserts the newly formatted headlines.
4389 *
4390 * @param string $text
4391 * @param string $origText Original, untouched wikitext
4392 * @param bool $isMain
4393 * @return mixed|string
4394 * @private
4395 */
4399 # Inhibit editsection links if requested in the page
4405 }
4408 }
4410 # Get all headlines for numbering them and adding funky stuff like [edit]
4411 # links - this is for later, but we need the number of headlines right now
4416 $matches
4417 );
4419 # if there are fewer than 4 headlines in the article, do not show TOC
4420 # unless it's been explicitly enabled.
4424 # Allow user to stipulate that a page should have a "new section"
4425 # link added via __NEWSECTIONLINK__
4428 }
4430 # Allow user to remove the "new section"
4431 # link via __NONEWSECTIONLINK__
4434 }
4436 # if the string __FORCETOC__ (not case-sensitive) occurs in the HTML,
4437 # override above conditions and always show TOC above first header
4441 }
4443 # headline counter
4447 # Ugh .. the TOC should have neat indentation levels which can be
4448 # passed to the skin functions. These are determined here
4480 }
4484 }
4488 # Increase TOC level
4495 }
4497 # Decrease TOC level, find level to jump to
4501 # Found last matching level
4505 # Found first matching level below current level
4508 }
4509 }
4512 }
4515 # Unindent only if the previous toc level was shown :p
4520 }
4521 }
4523 # No change in level, end TOC line
4526 }
4527 }
4531 # count number of headlines for each level
4538 }
4541 }
4542 }
4544 # The safe header is a version of the header text safe to use for links
4546 # Remove link placeholders by the link text.
4547 # <!--LINK number-->
4548 # turns into
4549 # link text with suffix
4550 # Do this before unstrip since link text can contain strip markers
4553 # Avoid insertion of weird stuff like <math> by expanding the relevant sections
4556 # Strip out HTML (first regex removes any tag not allowed)
4557 # Allowed tags are:
4558 # * <sup> and <sub> (bug 8393)
4559 # * <i> (bug 26375)
4560 # * <b> (r105284)
4561 # * <span dir="rtl"> and <span dir="ltr"> (bug 35167)
4562 #
4563 # We strip any parameter from accepted tags (second regex), except dir="rtl|ltr" from <span>,
4564 # to allow setting directionality in toc items.
4569 ),
4571 $safeHeadline
4572 );
4575 # For the anchor, strip out HTML-y stuff period
4579 # Save headline for section edit hint before it's escaped
4583 # For reverse compatibility, provide an id that's
4584 # HTML4-compatible, like we used to.
4585 #
4586 # It may be worth noting, academically, that it's possible for
4587 # the legacy anchor to conflict with a non-legacy headline
4588 # anchor on the page. In this case likely the "correct" thing
4589 # would be to either drop the legacy anchors or make sure
4590 # they're numbered first. However, this would require people
4591 # to type in section names like "abc_.D7.93.D7.90.D7.A4"
4592 # manually, so let's not bother worrying about it.
4598 # No reason to have both (in fact, we can't)
4600 }
4605 }
4607 # HTML names must be case-insensitively unique (bug 10721).
4608 # This does not apply to Unicode characters per
4609 # http://dev.w3.org/html5/spec/infrastructure.html#case-sensitivity-and-string-comparison
4610 # @todo FIXME: We may be changing them depending on the current locale.
4616 }
4618 # count how many in assoc. array so we can track dupes in anchors
4623 }
4628 }
4630 # Don't number the heading if it is the only one (looks silly)
4632 # the two are different if the line contains a link
4636 $numbering
4638 }
4640 # Create the anchor for linking from the TOC to the section
4645 }
4648 }
4652 }
4654 # Add the section to the section tree
4655 # Find the DOM node for this header
4662 }
4663 }
4667 }
4677 );
4679 # give headline the correct <h#> tag
4681 // Output edit section links as markers with styles that can be customized by skins
4683 # Put a T flag in the section identifier, to indicate to extractSections()
4684 # that sections inside <includeonly> should be counted.
4690 $headlineHint
4691 );
4692 }
4693 // We use a bit of pesudo-xml for editsection markers. The
4694 // language converter is run later on. Using a UNIQ style marker
4695 // leads to the converter screwing up the tokens when it
4696 // converts stuff. And trying to insert strip tags fails too. At
4697 // this point all real inputted tags have already been escaped,
4698 // so we don't have to worry about a user trying to input one of
4699 // these markers directly. We use a page and section attribute
4700 // to stop the language converter from converting these
4701 // important bits of data, but put the headline hint inside a
4702 // content block because the language converter is supposed to
4703 // be able to convert that piece of data.
4710 }
4713 }
4719 }
4723 # Never ever show TOC if no headers
4726 }
4731 }
4736 }
4740 }
4742 # split up and insert constructed headlines
4746 // build an array of document sections
4749 // $head is zero-based, sections aren't.
4754 }
4756 /**
4757 * Send a hook, one per section.
4758 * The idea here is to be able to make section-level DIVs, but to do so in a
4759 * lower-impact, more correct way than r50769
4760 *
4761 * $this : caller
4762 * $section : the section number
4763 * &$sectionContent : ref to the content of the section
4764 * $showEditLinks : boolean describing whether this section has an edit link
4765 */
4769 }
4772 // append the TOC at the beginning
4773 // Top anchor now in skin
4775 }
4783 }
4784 }
4786 /**
4787 * Transform wiki markup when saving a page by doing "\r\n" -> "\n"
4788 * conversion, substitting signatures, {{subst:}} templates, etc.
4789 *
4790 * @param string $text The text to transform
4791 * @param Title $title The Title object for the current article
4792 * @param User $user The User object describing the current user
4793 * @param ParserOptions $options Parsing options
4794 * @param bool $clearState Whether to clear the parser state first
4795 * @return string The altered wiki markup
4796 */
4799 ) {
4802 }
4808 );
4812 }
4818 }
4820 /**
4821 * Pre-save transform helper function
4822 *
4823 * @param string $text
4824 * @param User $user
4825 *
4826 * @return string
4827 */
4831 # Note: This is the timestamp saved as hardcoded wikitext to
4832 # the database, we use $wgContLang here in order to give
4833 # everyone the same signature and use the default one rather
4834 # than the one selected in each user's preferences.
4835 # (see also bug 12815)
4841 # Allow translation of timezones through wiki. format() can return
4842 # whatever crap the system uses, localised or not, so we cannot
4843 # ship premade translations.
4848 }
4852 # Variable replacement
4853 # Because mOutputType is OT_WIKI, this will only process {{subst:xxx}} type tags
4856 # This works almost by chance, as the replaceVariables are done before the getUserSig(),
4857 # which may corrupt this parser instance via its wfMessage()->text() call-
4859 # Signatures
4865 ) );
4867 # Context links ("pipe tricks"): [[|name]] and [[name (context)|]]
4871 // [[ns:page (context)|]]
4873 // [[ns:page（context）|]] (double-width brackets, added in r40257)
4875 // [[ns:page (context), context|]] (using either single or double-width comma)
4877 // [[|page]] (reverse pipe trick: add context from page title)
4880 # try $p1 first, to turn "[[A, B (C)|]]" into "[[A, B (C)|A, B]]"
4892 # if there's no context, don't bother duplicating the title
4894 }
4896 # Trim trailing whitespace
4900 }
4902 /**
4903 * Fetch the user's signature text, if any, and normalize to
4904 * validated, ready-to-insert wikitext.
4905 * If you have pre-fetched the nickname or the fancySig option, you can
4906 * specify them here to save a database query.
4907 * Do not reuse this parser instance after calling getUserSig(),
4908 * as it may have changed if it's the $wgParser.
4909 *
4910 * @param User $user
4911 * @param string|bool $nickname Nickname to use or false to use user's default nickname
4912 * @param bool|null $fancySig whether the nicknname is the complete signature
4913 * or null to use default value
4914 * @return string
4915 */
4921 # If not given, retrieve from the user object.
4924 }
4928 }
4936 # Sig. might contain markup; validate this
4938 # Validated; clean up (if needed) and return it
4941 # Failed to validate; fall back to the default
4944 }
4945 }
4947 # Make sure nickname doesnt get a sig in a sig
4950 # If we're still here, make it a link to the user page
4957 }
4959 /**
4960 * Check that the user's signature contains no bad XML
4961 *
4962 * @param string $text
4963 * @return string|bool An expanded string, or false if invalid.
4964 */
4967 }
4969 /**
4970 * Clean up signature text
4971 *
4972 * 1) Strip ~~~, ~~~~ and ~~~~~ out of signatures @see cleanSigInSig
4973 * 2) Substitute all transclusions
4974 *
4975 * @param string $text
4976 * @param bool $parsing Whether we're cleaning (preferences save) or parsing
4977 * @return string Signature text
4978 */
4984 }
4986 # Option to disable this feature
4989 }
4991 # @todo FIXME: Regex doesn't respect extension tags or nowiki
4992 # => Move this logic to braceSubstitution()
4994 $substRegex = '/\{\{(?!(?:' . $substWord->getBaseRegex() . '))/x' . $substWord->getRegexCase();
5005 }
5008 }
5010 /**
5011 * Strip ~~~, ~~~~ and ~~~~~ out of signatures
5012 *
5013 * @param string $text
5014 * @return string Signature text with /~{3,5}/ removed
5015 */
5019 }
5021 /**
5022 * Set up some variables which are usually set up in parse()
5023 * so that an external function can call some class members with confidence
5024 *
5025 * @param Title|null $title
5026 * @param ParserOptions $options
5027 * @param int $outputType
5028 * @param bool $clearState
5029 */
5032 ) {
5034 }
5036 /**
5037 * @param Title|null $title
5038 * @param ParserOptions $options
5039 * @param int $outputType
5040 * @param bool $clearState
5041 */
5044 ) {
5050 }
5051 }
5053 /**
5054 * Wrapper for preprocess()
5055 *
5056 * @param string $text The text to preprocess
5057 * @param ParserOptions $options Options
5058 * @param Title|null $title Title object or null to use $wgTitle
5059 * @return string
5060 */
5064 # Guard against infinite recursion
5067 }
5074 }
5081 }
5083 /**
5084 * Create an HTML-style tag, e.g. "<yourtag>special text</yourtag>"
5085 * The callback should have the following form:
5086 * function myParserHook( $text, $params, $parser, $frame ) { ... }
5087 *
5088 * Transform and return $text. Use $parser for any required context, e.g. use
5089 * $parser->getTitle() and $parser->getOptions() not $wgTitle or $wgOut->mParserOptions
5090 *
5091 * Hooks may return extended information by returning an array, of which the
5092 * first numbered element (index 0) must be the return string, and all other
5093 * entries are extracted into local variables within an internal function
5094 * in the Parser class.
5095 *
5096 * This interface (introduced r61913) appears to be undocumented, but
5097 * 'markerName' is used by some core tag hooks to override which strip
5098 * array their results are placed in. **Use great caution if attempting
5099 * this interface, as it is not documented and injudicious use could smash
5100 * private variables.**
5101 *
5102 * @param string $tag The tag to use, e.g. 'hook' for "<hook>"
5103 * @param mixed $callback The callback function (and object) to use for the tag
5104 * @throws MWException
5105 * @return mixed|null The old value of the mTagHooks array associated with the hook
5106 */
5111 }
5116 }
5119 }
5121 /**
5122 * As setHook(), but letting the contents be parsed.
5123 *
5124 * Transparent tag hooks are like regular XML-style tag hooks, except they
5125 * operate late in the transformation sequence, on HTML instead of wikitext.
5126 *
5127 * This is probably obsoleted by things dealing with parser frames?
5128 * The only extension currently using it is geoserver.
5129 *
5130 * @since 1.10
5131 * @todo better document or deprecate this
5132 *
5133 * @param string $tag The tag to use, e.g. 'hook' for "<hook>"
5134 * @param mixed $callback The callback function (and object) to use for the tag
5135 * @throws MWException
5136 * @return mixed|null The old value of the mTagHooks array associated with the hook
5137 */
5142 }
5143 $oldVal = isset( $this->mTransparentTagHooks[$tag] ) ? $this->mTransparentTagHooks[$tag] : null;
5147 }
5149 /**
5150 * Remove all tag hooks
5151 */
5156 }
5158 /**
5159 * Create a function, e.g. {{sum:1|2|3}}
5160 * The callback function should have the form:
5161 * function myParserFunction( &$parser, $arg1, $arg2, $arg3 ) { ... }
5162 *
5163 * Or with SFH_OBJECT_ARGS:
5164 * function myParserFunction( $parser, $frame, $args ) { ... }
5165 *
5166 * The callback may either return the text result of the function, or an array with the text
5167 * in element 0, and a number of flags in the other elements. The names of the flags are
5168 * specified in the keys. Valid flags are:
5169 * found The text returned is valid, stop processing the template. This
5170 * is on by default.
5171 * nowiki Wiki markup in the return value should be escaped
5172 * isHTML The returned text is HTML, armour it against wikitext transformation
5173 *
5174 * @param string $id The magic word ID
5175 * @param mixed $callback The callback function (and object) to use
5176 * @param int $flags A combination of the following flags:
5177 * SFH_NO_HASH No leading hash, i.e. {{plural:...}} instead of {{#if:...}}
5178 *
5179 * SFH_OBJECT_ARGS Pass the template arguments as PPNode objects instead of text. This
5180 * allows for conditional expansion of the parse tree, allowing you to eliminate dead
5181 * branches and thus speed up parsing. It is also possible to analyse the parse tree of
5182 * the arguments, and to control the way they are expanded.
5183 *
5184 * The $frame parameter is a PPFrame. This can be used to produce expanded text from the
5185 * arguments, for instance:
5186 * $text = isset( $args[0] ) ? $frame->expand( $args[0] ) : '';
5187 *
5188 * For technical reasons, $args[0] is pre-expanded and will be a string. This may change in
5189 * future versions. Please call $frame->expand() on it anyway so that your code keeps
5190 * working if/when this is changed.
5191 *
5192 * If you want whitespace to be trimmed from $args, you need to do it yourself, post-
5193 * expansion.
5194 *
5195 * Please read the documentation in includes/parser/Preprocessor.php for more information
5196 * about the methods available in PPFrame and PPNode.
5197 *
5198 * @throws MWException
5199 * @return string|callable The old callback function for this name, if any
5200 */
5207 # Add to function cache
5211 }
5217 # Case
5220 }
5221 # Add leading hash
5224 }
5225 # Remove trailing colon
5228 }
5230 }
5232 }
5234 /**
5235 * Get all registered function hook identifiers
5236 *
5237 * @return array
5238 */
5241 }
5243 /**
5244 * Create a tag function, e.g. "<test>some stuff</test>".
5245 * Unlike tag hooks, tag functions are parsed at preprocessor level.
5246 * Unlike parser functions, their content is not preprocessed.
5247 * @param string $tag
5248 * @param callable $callback
5249 * @param int $flags
5250 * @throws MWException
5251 * @return null
5252 */
5257 }
5264 }
5267 }
5269 /**
5270 * @todo FIXME: Update documentation. makeLinkObj() is deprecated.
5271 * Replace "<!--LINK-->" link placeholders with actual links, in the buffer
5272 * Placeholders created in Skin::makeLinkObj()
5273 *
5274 * @param string $text
5275 * @param int $options
5276 *
5277 * @return array Array of link CSS classes, indexed by PDBK.
5278 */
5281 }
5283 /**
5284 * Replace "<!--LINK-->" link placeholders with plain text of links
5285 * (not HTML-formatted).
5286 *
5287 * @param string $text
5288 * @return string
5289 */
5292 }
5294 /**
5295 * Renders an image gallery from a text with one line per image.
5296 * text labels may be given by using |-style alternative text. E.g.
5297 * Image:one.jpg|The number "1"
5298 * Image:tree.jpg|A tree
5299 * given as text will return the HTML of a gallery with two images,
5300 * labeled 'The number "1"' and
5301 * 'A tree'.
5302 *
5303 * @param string $text
5304 * @param array $params
5305 * @return string HTML
5306 */
5313 }
5318 // If invalid type set, fallback to default.
5320 }
5333 }
5339 }
5342 }
5345 }
5348 }
5355 # match lines like these:
5356 # Image:someimage.jpg|This is some image
5359 # Skip empty lines
5362 }
5366 }
5369 # Bogus title. Ignore these so we don't bomb out later.
5371 }
5373 # We need to get what handler the file uses, to figure out parameters.
5374 # Note, a hook can overide the file name, and chose an entirely different
5375 # file (which potentially could be of a different type and have different handler).
5380 # Don't register it now, as ImageGallery does that later.
5388 );
5391 // We don't want people to specify per-image widths.
5392 // Additionally the width parameter would need special casing anyhow.
5394 }
5404 // look for an |alt= definition while trying not to break existing
5405 // captions with multiple pipes (|) in it, until a more sensible grammar
5406 // is defined for images in galleries
5408 // FIXME: Doing recursiveTagParse at this stage, and the trim before
5409 // splitting on '|' is a bit odd, and different from makeImage.
5426 //check to see if link matches an absolute url, if not then it must be a wiki link.
5433 }
5434 }
5437 // Must be a handler specific parameter.
5441 // Guess not. Append it to the caption.
5444 }
5445 }
5448 // concatenate all other pipes
5450 }
5451 }
5452 // remove the first pipe
5454 }
5457 }
5461 }
5463 /**
5464 * @param string $handler
5465 * @return array
5466 */
5472 }
5474 # Initialise static lists
5481 );
5489 }
5490 }
5491 }
5493 # Add handler params
5499 }
5500 }
5503 }
5504 return array( $this->mImageParams[$handlerClass], $this->mImageParamsMagicArray[$handlerClass] );
5505 }
5507 /**
5508 * Parse image options text and use it to make an image
5509 *
5510 * @param Title $title
5511 * @param string $options
5512 * @param LinkHolderArray|bool $holders
5513 * @return string HTML
5514 */
5516 # Check if the options text is of the form "options|alt text"
5517 # Options are:
5518 # * thumbnail make a thumbnail with enlarge-icon and caption, alignment depends on lang
5519 # * left no resizing, just left align. label is used for alt= only
5520 # * right same, but right aligned
5521 # * none same, but not aligned
5522 # * ___px scale to ___ pixels width, no aligning. e.g. use in taxobox
5523 # * center center the image
5524 # * frame Keep original image size, no magnify-button.
5525 # * framed Same as "frame"
5526 # * frameless like 'thumb' but without a frame. Keeps user preferences for width
5527 # * upright reduce width for upright images, rounded to full __0 px
5528 # * border draw a 1px border around the image
5529 # * alt Text for HTML alt attribute (defaults to empty)
5530 # * class Set a class for img node
5531 # * link Set the target of the image link. Can be external, interwiki, or local
5532 # vertical-align values (no % or length right now):
5533 # * baseline
5534 # * sub
5535 # * super
5536 # * top
5537 # * text-top
5538 # * middle
5539 # * bottom
5540 # * text-bottom
5544 # Give extensions a chance to select the file revision for us
5549 # Fetch and register the file (file title may be different via hooks)
5552 # Get parameter map
5559 }
5561 # Process the input parameters
5573 # Special case; width and height come in one variable together
5581 }
5582 }
5588 }
5589 }
5590 # else no validation -- bug 13436
5593 # Validate handler parameter
5596 # Validate internal parameters
5601 # @todo FIXME: Possibly check validity here for
5602 # manualthumb? downstream behavior seems odd with
5603 # missing manual thumbs.
5620 }
5622 }
5630 }
5631 }
5636 // use first appearing option, discard others.
5641 # Most other things appear to be empty or numeric...
5643 }
5644 }
5648 }
5649 }
5650 }
5653 }
5654 }
5656 # Process alignment parameters
5659 }
5662 }
5666 # Will the image be presented in a frame, with the caption below?
5672 # In the old days, [[Image:Foo|text...]] would set alt text. Later it
5673 # came to also set the caption, ordinary text after the image -- which
5674 # makes no sense, because that just repeats the text multiple times in
5675 # screen readers. It *also* came to set the title attribute.
5676 #
5677 # Now that we have an alt attribute, we should not set the alt text to
5678 # equal the caption: that's worse than useless, it just repeats the
5679 # text. This is the framed/thumbnail case. If there's no caption, we
5680 # use the unnamed parameter for alt text as well, just for the time be-
5681 # ing, if the unnamed param is set and the alt param is not.
5682 #
5683 # For the future, we need to figure out if we want to tweak this more,
5684 # e.g., introducing a title= parameter for the title; ignoring the un-
5685 # named parameter entirely for images without a caption; adding an ex-
5686 # plicit caption= parameter and preserving the old magic unnamed para-
5687 # meter for BC; ...
5690 # No caption or alt text, add the filename as the alt text so
5691 # that screen readers at least get some description of the image
5693 }
5694 # Do not set $params['frame']['title'] because tooltips don't make sense
5695 # for framed images
5698 # No alt text, use the "caption" for the alt text
5702 # No caption, fall back to using the filename for the
5703 # alt text
5705 }
5706 }
5707 # Use the "caption" for the tooltip text
5709 }
5713 # Linker does the rest
5718 # Give the handler a chance to modify the parser object
5721 }
5724 }
5726 /**
5727 * @param string $caption
5728 * @param LinkHolderArray|bool $holders
5729 * @return mixed|string
5730 */
5732 # Strip bad stuff out of the title (tooltip). We can't just use
5733 # replaceLinkHoldersText() here, because if this function is called
5734 # from replaceInternalLinks2(), mLinkHolders won't be up-to-date.
5739 }
5741 # make sure there are no placeholders in thumbnail attributes
5742 # that are later expanded to html- so expand them now and
5743 # remove the tags
5748 }
5750 /**
5751 * Set a flag in the output object indicating that the content is dynamic and
5752 * shouldn't be cached.
5753 */
5759 }
5762 }
5764 /**
5765 * Callback from the Sanitizer for expanding items found in HTML attribute
5766 * values, so they can be safely tested and escaped.
5767 *
5768 * @param string $text
5769 * @param bool|PPFrame $frame
5770 * @return string
5771 */
5776 }
5778 /**
5779 * Accessor
5780 *
5781 * @return array
5782 */
5788 );
5789 }
5791 /**
5792 * Replace transparent tags in $text with the values given by the callbacks.
5793 *
5794 * Transparent tag hooks are like regular XML-style tag hooks, except they
5795 * operate late in the transformation sequence, on HTML instead of wikitext.
5796 *
5797 * @param string $text
5798 *
5799 * @return string
5800 */
5814 );
5817 }
5819 }
5821 }
5823 /**
5824 * Break wikitext input into sections, and either pull or replace
5825 * some particular section's text.
5826 *
5827 * External callers should use the getSection and replaceSection methods.
5828 *
5829 * @param string $text Page wikitext
5830 * @param string $section A section identifier string of the form:
5831 * "<flag1> - <flag2> - ... - <section number>"
5832 *
5833 * Currently the only recognised flag is "T", which means the target section number
5834 * was derived during a template inclusion parse, in other words this is a template
5835 * section edit link. If no flags are given, it was an ordinary section edit link.
5836 * This flag is required to avoid a section numbering mismatch when a section is
5837 * enclosed by "<includeonly>" (bug 6563).
5838 *
5839 * The section number 0 pulls the text before the first heading; other numbers will
5840 * pull the given section along with its lower-level subsections. If the section is
5841 * not found, $mode=get will return $newtext, and $mode=replace will return $text.
5842 *
5843 * Section 0 is always considered to exist, even if it only contains the empty
5844 * string. If $text is the empty string and section 0 is replaced, $newText is
5845 * returned.
5846 *
5847 * @param string $mode One of "get" or "replace"
5848 * @param string $newText Replacement text for section data.
5849 * @return string For "get", the extracted section text.
5850 * for "replace", the whole page with the section replaced.
5851 */
5860 # Process section extraction flags
5867 }
5868 }
5870 # Check for empty input
5872 # Only sections 0 and T-0 exist in an empty document
5878 }
5884 }
5885 }
5886 }
5888 # Preprocess the text
5891 # <h> nodes indicate section breaks
5892 # They can only occur at the top level, so we can find them by iterating the root's children
5895 # Find the target section
5897 # Section zero doesn't nest, level=big
5906 }
5907 }
5910 }
5912 }
5913 }
5916 # Not found
5921 }
5922 }
5924 # Find the end of the section, including nested sections
5931 }
5932 }
5935 }
5939 # Write out the remainder (in replace mode only)
5941 # Output the replacement text
5942 # Add two newlines on -- trailing whitespace in $newText is conventionally
5943 # stripped by the editor, so we need both newlines to restore the paragraph gap
5944 # Only add trailing whitespace if there is newText
5947 }
5952 }
5953 }
5956 # Re-insert stripped tags
5958 }
5961 }
5963 /**
5964 * This function returns the text of a section, specified by a number ($section).
5965 * A section is text under a heading like == Heading == or \<h1\>Heading\</h1\>, or
5966 * the first section before any such heading (section 0).
5967 *
5968 * If a section contains subsections, these are also returned.
5969 *
5970 * @param string $text Text to look in
5971 * @param string $section Section identifier
5972 * @param string $deftext Default to return if section is not found
5973 * @return string Text of the requested section
5974 */
5977 }
5979 /**
5980 * This function returns $oldtext after the content of the section
5981 * specified by $section has been replaced with $text. If the target
5982 * section does not exist, $oldtext is returned unchanged.
5983 *
5984 * @param string $oldtext Former text of the article
5985 * @param int $section Section identifier
5986 * @param string $text Replacing text
5987 * @return string Modified text
5988 */
5991 }
5993 /**
5994 * Get the ID of the revision we are parsing
5995 *
5996 * @return int|null
5997 */
6000 }
6002 /**
6003 * Get the revision object for $this->mRevisionId
6004 *
6005 * @return Revision|null Either a Revision object or null
6006 * @since 1.23 (public since 1.23)
6007 */
6011 }
6014 }
6018 }
6020 /**
6021 * Get the timestamp associated with the current revision, adjusted for
6022 * the default server-local timestamp
6023 * @return string
6024 */
6034 # The cryptic '' timezone parameter tells to use the site-default
6035 # timezone offset instead of the user settings.
6036 #
6037 # Since this value will be saved into the parser cache, served
6038 # to other users, and potentially even used inside links and such,
6039 # it needs to be consistent for all visitors.
6043 }
6045 }
6047 /**
6048 * Get the name of the user that edited the last revision
6049 *
6050 * @return string User name
6051 */
6056 # if this template is subst: the revision id will be blank,
6057 # so just use the current user's name
6062 }
6063 }
6065 }
6067 /**
6068 * Get the size of the revision
6069 *
6070 * @return int|null Revision size
6071 */
6076 # if this variable is subst: the revision id will be blank,
6077 # so just use the parser input size, because the own substituation
6078 # will change the size.
6083 }
6084 }
6086 }
6088 /**
6089 * Mutator for $mDefaultSort
6090 *
6091 * @param string $sort New value
6092 */
6096 }
6098 /**
6099 * Accessor for $mDefaultSort
6100 * Will use the empty string if none is set.
6101 *
6102 * This value is treated as a prefix, so the
6103 * empty string is equivalent to sorting by
6104 * page name.
6105 *
6106 * @return string
6107 */
6113 }
6114 }
6116 /**
6117 * Accessor for $mDefaultSort
6118 * Unlike getDefaultSort(), will return false if none is set
6119 *
6120 * @return string|bool
6121 */
6124 }
6126 /**
6127 * Try to guess the section anchor name based on a wikitext fragment
6128 * presumably extracted from a heading, for example "Header" from
6129 * "== Header ==".
6130 *
6131 * @param string $text
6132 *
6133 * @return string
6134 */
6136 # Strip out wikitext links(they break the anchor)
6140 }
6142 /**
6143 * Same as guessSectionNameFromWikiText(), but produces legacy anchors
6144 * instead. For use in redirects, since IE6 interprets Redirect: headers
6145 * as something other than UTF-8 (apparently?), resulting in breakage.
6146 *
6147 * @param string $text The section name
6148 * @return string An anchor
6149 */
6151 # Strip out wikitext links(they break the anchor)
6155 }
6157 /**
6158 * Strips a text string of wikitext for use in a section anchor
6159 *
6160 * Accepts a text string and then removes all wikitext from the
6161 * string and leaves only the resultant text (i.e. the result of
6162 * [[User:WikiSysop|Sysop]] would be "Sysop" and the result of
6163 * [[User:WikiSysop]] would be "User:WikiSysop") - this is intended
6164 * to create valid section anchors by mimicing the output of the
6165 * parser when headings are parsed.
6166 *
6167 * @param string $text Text string to be stripped of wikitext
6168 * for use in a Section anchor
6169 * @return string Filtered text string
6170 */
6172 # Strip internal link markup
6176 # Strip external link markup
6177 # @todo FIXME: Not tolerant to blank link text
6178 # I.E. [https://www.mediawiki.org] will render as [1] or something depending
6179 # on how many empty links there are on the page - need to figure that out.
6180 $text = preg_replace( '/\[(?i:' . $this->mUrlProtocols . ')([^ ]+?) ([^[]+)\]/', '$2', $text );
6182 # Parse wikitext quotes (italics & bold)
6185 # Strip HTML tags
6188 }
6190 /**
6191 * strip/replaceVariables/unstrip for preprocessor regression testing
6192 *
6193 * @param string $text
6194 * @param Title $title
6195 * @param ParserOptions $options
6196 * @param int $outputType
6197 *
6198 * @return string
6199 */
6200 function testSrvus( $text, Title $title, ParserOptions $options, $outputType = self::OT_HTML ) {
6208 }
6210 /**
6211 * @param string $text
6212 * @param Title $title
6213 * @param ParserOptions $options
6214 * @return string
6215 */
6218 }
6220 /**
6221 * @param string $text
6222 * @param Title $title
6223 * @param ParserOptions $options
6224 * @return string
6225 */
6228 }
6230 /**
6231 * Call a callback function on all regions of the given text that are not
6232 * inside strip markers, and replace those regions with the return value
6233 * of the callback. For example, with input:
6234 *
6235 * aaa<MARKER>bbb
6236 *
6237 * This will call the callback function twice, with 'aaa' and 'bbb'. Those
6238 * two strings will be replaced with the value returned by the callback in
6239 * each case.
6240 *
6241 * @param string $s
6242 * @param callable $callback
6243 *
6244 * @return string
6245 */
6264 }
6265 }
6266 }
6268 }
6270 /**
6271 * Remove any strip markers found in the given text.
6272 *
6273 * @param string $text Input string
6274 * @return string
6275 */
6278 }
6280 /**
6281 * Save the parser state required to convert the given half-parsed text to
6282 * HTML. "Half-parsed" in this context means the output of
6283 * recursiveTagParse() or internalParse(). This output has strip markers
6284 * from replaceVariables (extensionSubstitution() etc.), and link
6285 * placeholders from replaceLinkHolders().
6286 *
6287 * Returns an array which can be serialized and stored persistently. This
6288 * array can later be loaded into another parser instance with
6289 * unserializeHalfParsedText(). The text can then be safely incorporated into
6290 * the return value of a parser hook.
6291 *
6292 * @param string $text
6293 *
6294 * @return array
6295 */
6303 );
6306 }
6308 /**
6309 * Load the parser state given in the $data array, which is assumed to
6310 * have been generated by serializeHalfParsedText(). The text contents is
6311 * extracted from the array, and its markers are transformed into markers
6312 * appropriate for the current Parser instance. This transformed text is
6313 * returned, and can be safely included in the return value of a parser
6314 * hook.
6315 *
6316 * If the $data array has been stored persistently, the caller should first
6317 * check whether it is still valid, by calling isValidHalfParsedText().
6318 *
6319 * @param array $data Serialized data
6320 * @throws MWException
6321 * @return string
6322 */
6326 }
6328 # First, extract the strip state.
6332 # Now renumber links
6335 # Should be good to go.
6337 }
6339 /**
6340 * Returns true if the given array, presumed to be generated by
6341 * serializeHalfParsedText(), is compatible with the current version of the
6342 * parser.
6343 *
6344 * @param array $data
6345 *
6346 * @return bool
6347 */
6350 }
6352 /**
6353 * Parsed a width param of imagelink like 300px or 200x300px
6354 *
6355 * @param string $value
6356 *
6357 * @return array
6358 * @since 1.20
6359 */
6364 }
6366 # (bug 13500) In both cases (width/height and width only),
6367 # permit trailing "px" for backward compatibility.
6376 }
6378 }
6380 /**
6381 * Lock the current instance of the parser.
6382 *
6383 * This is meant to stop someone from calling the parser
6384 * recursively and messing up all the strip state.
6385 *
6386 * @throws MWException If parser is in a parse
6387 * @return ScopedCallback The lock will be released once the return value goes out of scope.
6388 */
6393 }
6399 } );
6402 }
6403 }