| blob | 95cef5d1e9469f8b6e00f5952b58414e9163c7d0 |
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 /**
116 * Prefix for temporary replacement strings generated by the preprocessor
117 * ("strip markers"). Using \x7f at the front gives us a little extra
118 * robustness since it shouldn't match when butted up against
119 * identifier-like string constructs.
120 *
121 * Must not consist of all title characters, or else it will change
122 * the behavior of <nowiki> in a link.
123 */
125 /** Suffix for strip markers */
127 /** Regex which matches the state ID part of strip markers */
130 # Markers used for wrapping the table of contents
134 # Persistent:
148 # Initialised by initialiseVariables()
150 /**
151 * @var MagicWordArray
152 */
155 /**
156 * @var MagicWordArray
157 */
159 public $mConf, $mPreprocessor, $mExtLinkBracketedRegex, $mUrlProtocols; # Initialised in constructor
161 # Cleared with clearState():
162 /**
163 * @var ParserOutput
164 */
168 /**
169 * @var StripState
170 */
174 /**
175 * @var LinkHolderArray
176 */
186 /**
187 * @var User
188 */
191 # Temporary
192 # These are variables reset at least once per parse regardless of $clearState
194 /**
195 * @var ParserOptions
196 */
199 /**
200 * @var Title
201 */
213 /**
214 * @var string
215 */
218 /**
219 * @var array Array with the language name of each language link (i.e. the
220 * interwiki prefix) in the key, value arbitrary. Used to avoid sending
221 * duplicate language links to the ParserOutput.
222 */
225 /**
226 * @var MapCacheLRU|null
227 * @since 1.24
228 *
229 * A cache of the current revisions of titles. Keys are $title->getPrefixedDbKey()
230 */
233 /**
234 * @var bool Recursive call protection.
235 * This variable should be treated as if it were private.
236 */
239 /**
240 * @param array $conf
241 */
250 # Preprocessor_Hash is much faster than Preprocessor_DOM under HipHop
253 # PECL extension that conflicts with the core DOM extension (bug 13770)
260 }
262 }
264 /**
265 * Reduce memory usage to reduce the impact of circular references
266 */
270 }
273 }
274 }
276 /**
277 * Allow extensions to clean up when the parser is cloned
278 */
282 // Bug 56226: When you create a reference "to" an object field, that
283 // makes the object field itself be a reference too (until the other
284 // reference goes out of scope). When cloning, any field that's a
285 // reference is copied as a reference in the new object. Both of these
286 // are defined PHP5 behaviors, as inconvenient as it is for us when old
287 // hooks from PHP4 days are passing fields by reference.
289 // Make a non-reference copy of the field, then rebind the field to
290 // reference the new copy.
294 }
297 }
299 /**
300 * Do various kinds of initialisation on the first call of the parser
301 */
305 }
316 }
318 /**
319 * Clear Parser state
320 *
321 * @private
322 */
327 }
349 # Clear these on every parse, bug 4549
357 );
366 # Fix cloning
369 }
373 }
375 /**
376 * Convert wikitext to HTML
377 * Do not call this function recursively.
378 *
379 * @param string $text Text we want to parse
380 * @param Title $title
381 * @param ParserOptions $options
382 * @param bool $linestart
383 * @param bool $clearState
384 * @param int $revid Number to pass in {{REVISIONID}}
385 * @return ParserOutput A ParserOutput
386 */
389 ) {
390 /**
391 * First pass--just handle <nowiki> sections, pass the rest off
392 * to internalParse() which does all the real work.
393 */
402 }
410 }
412 # Remove the strip marker tag prefix from the input, if present.
415 }
428 }
431 # No more strip!
438 # Clean up special characters, only run once, next-to-last before doBlockLevels
440 # french spaces, last one Guillemet-left
441 # only if there is something before the space
443 # french spaces, Guillemet-right
446 );
453 /**
454 * The input doesn't get language converted if
455 * a) It's disabled
456 * b) Content isn't converted
457 * c) It's a conversion table
458 * d) it is an interface message (which is in the user language)
459 */
462 ) {
464 # The position of the convert() call should not be changed. it
465 # assumes that the links are all replaced and the only thing left
466 # is the <nowiki> mark.
468 }
469 }
471 /**
472 * A converted title will be provided in the output object if title and
473 * content conversion are enabled, the article text does not contain
474 * a conversion-suppressing double-underscore tag, and no
475 * {{DISPLAYTITLE:...}} is present. DISPLAYTITLE takes precedence over
476 * automatic link conversion.
477 */
482 ) {
489 }
490 }
504 # attempt to sanitize at least some nesting problems
505 # (bug #2702 and quite a few others)
507 # ''Something [http://www.cool.com cool''] -->
508 # <i>Something</i><a href="http://www.cool.com"..><i>cool></i></a>
511 # fix up an anchor inside another anchor, only
512 # at least for a single single nested link (bug 3695)
515 # fix div inside inline elements- doBlockLevels won't wrap a line which
516 # contains a div, so fix it up here; replace
517 # div with escaped text
520 # remove empty italic or bold tag pairs, some
521 # introduced by rules above
523 );
529 }
535 );
536 }
540 # Information on include size limits, for the benefit of users who try to skirt them
548 );
549 }
554 );
558 );
561 );
564 );
567 );
570 );
573 );
579 }
583 ) ) {
589 }
593 }
594 }
595 }
596 // Since we're not really outputting HTML, decode the entities and
597 // then re-encode the things that need hiding inside HTML comments.
601 // Sanitize for comment. Note '‐' in the replacement is U+2010,
602 // which looks much like the problematic '-'.
609 }
610 }
624 }
626 /**
627 * Recursive parser entry point that can be called from an extension tag
628 * hook.
629 *
630 * If $frame is not provided, then template variables (e.g., {{{1}}}) within $text are not expanded
631 *
632 * @param string $text Text extension wants to have parsed
633 * @param bool|PPFrame $frame The frame to use for expanding any template variables
634 *
635 * @return string
636 */
644 }
646 /**
647 * Expand templates and variables in the text, producing valid, static wikitext.
648 * Also removes comments.
649 * Do not call this function recursively.
650 * @param string $text
651 * @param Title $title
652 * @param ParserOptions $options
653 * @param int|null $revid
654 * @param bool|PPFrame $frame
655 * @return mixed|string
656 */
657 public function preprocess( $text, Title $title = null, ParserOptions $options, $revid = null, $frame = false ) {
663 }
670 }
672 /**
673 * Recursive parser entry point that can be called from an extension tag
674 * hook.
675 *
676 * @param string $text Text to be expanded
677 * @param bool|PPFrame $frame The frame to use for expanding any template variables
678 * @return string
679 * @since 1.19
680 */
687 }
689 /**
690 * Process the wikitext for the "?preload=" feature. (bug 5210)
691 *
692 * "<noinclude>", "<includeonly>" etc. are parsed as for template
693 * transclusion, comments, templates, arguments, tags hooks and parser
694 * functions are untouched.
695 *
696 * @param string $text
697 * @param Title $title
698 * @param ParserOptions $options
699 * @param array $params
700 * @return string
701 */
702 public function getPreloadText( $text, Title $title, ParserOptions $options, $params = array() ) {
706 # Parser (re)initialisation
715 }
717 /**
718 * Get a random string
719 *
720 * @return string
721 */
724 }
726 /**
727 * Set the current user.
728 * Should only be used when doing pre-save transform.
729 *
730 * @param User|null $user User object or null (to reset)
731 */
734 }
736 /**
737 * Accessor for mUniqPrefix.
738 *
739 * @return string
740 */
743 # @todo FIXME: This is probably *horribly wrong*
744 # LanguageConverter seems to want $wgParser's uniqPrefix, however
745 # if this is called for a parser cache hit, the parser may not
746 # have ever been initialized in the first place.
747 # Not really sure what the heck is supposed to be going on here.
749 # throw new MWException( "Accessing uninitialized mUniqPrefix" );
750 }
752 }
754 /**
755 * Set the context title
756 *
757 * @param Title $t
758 */
762 }
765 # Strip the fragment to avoid various odd effects
770 }
771 }
773 /**
774 * Accessor for the Title object
775 *
776 * @return Title
777 */
780 }
782 /**
783 * Accessor/mutator for the Title object
784 *
785 * @param Title $x Title object or null to just get the current one
786 * @return Title
787 */
790 }
792 /**
793 * Set the output type
794 *
795 * @param int $ot New value
796 */
799 # Shortcut alias
805 );
806 }
808 /**
809 * Accessor/mutator for the output type
810 *
811 * @param int|null $x New value or null to just get the current one
812 * @return int
813 */
816 }
818 /**
819 * Get the ParserOutput object
820 *
821 * @return ParserOutput
822 */
825 }
827 /**
828 * Get the ParserOptions object
829 *
830 * @return ParserOptions
831 */
834 }
836 /**
837 * Accessor/mutator for the ParserOptions object
838 *
839 * @param ParserOptions $x New value or null to just get the current one
840 * @return ParserOptions Current ParserOptions object
841 */
844 }
846 /**
847 * @return int
848 */
851 }
853 /**
854 * @param int $id
855 */
858 }
860 /**
861 * Get a language object for use in parser functions such as {{FORMATNUM:}}
862 * @return Language
863 */
866 }
868 /**
869 * Get the target language for the content being parsed. This is usually the
870 * language that the content is in.
871 *
872 * @since 1.19
873 *
874 * @throws MWException
875 * @return Language
876 */
886 }
889 }
891 /**
892 * Get the language object for language conversion
893 * @return Language|null
894 */
897 }
899 /**
900 * Get a User object either from $this->mUser, if set, or from the
901 * ParserOptions object otherwise
902 *
903 * @return User
904 */
908 }
910 }
912 /**
913 * Get a preprocessor object
914 *
915 * @return Preprocessor
916 */
921 }
923 }
925 /**
926 * Replaces all occurrences of HTML-style comments and the given tags
927 * in the text with a random marker and returns the next text. The output
928 * parameter $matches will be an associative array filled with data in
929 * the form:
930 *
931 * @code
932 * 'UNIQ-xxxxx' => array(
933 * 'element',
934 * 'tag content',
935 * array( 'param' => 'x' ),
936 * '<element param="x">tag content</element>' ) )
937 * @endcode
938 *
939 * @param array $elements List of element names. Comments are always extracted.
940 * @param string $text Source text string.
941 * @param array $matches Out parameter, Array: extracted tags
942 * @param string $uniq_prefix
943 * @return string Stripped text
944 */
958 }
960 # comment
966 # tag
971 }
977 # Empty element tag, <tag />
986 }
990 # No end tag -- let it run out to the end of the text.
996 }
997 }
1003 }
1005 }
1007 /**
1008 * Get a list of strippable XML-like elements
1009 *
1010 * @return array
1011 */
1014 }
1016 /**
1017 * Add an item to the strip state
1018 * Returns the unique tag which must be inserted into the stripped text
1019 * The tag will be replaced with the original text in unstrip()
1020 *
1021 * @param string $text
1022 *
1023 * @return string
1024 */
1030 }
1032 /**
1033 * parse the wiki syntax used to render tables
1034 *
1035 * @private
1036 * @param string $text
1037 * @return string
1038 */
1057 }
1063 # First check if we are starting a new table
1076 # Don't do any of the following
1080 # We are ending a table
1086 }
1090 }
1094 }
1098 # Now we have a table row
1101 # Whats after the tag is now only attributes
1114 }
1118 }
1127 ) {
1128 # This might be cell elements, td, th or captions
1132 }
1138 }
1140 # Split up multiple cells on the same line.
1141 # FIXME : This can result in improper nesting of tags processed
1142 # by earlier parser steps, but should avoid splitting up eg
1143 # attribute values containing literal "||".
1148 # Loop through each table cell
1155 }
1160 }
1166 }
1176 }
1180 # A cell could contain both parameters and data
1183 # Bug 553: Note that a '|' inside an invalid link should not
1184 # be mistaken as delimiting cell parameters
1193 }
1197 }
1198 }
1200 }
1202 # Closing open td, tr && table
1206 }
1209 }
1212 }
1215 }
1217 # Remove trailing line-ending (b/c)
1220 }
1222 # special case: don't return empty table
1225 }
1230 }
1232 /**
1233 * Helper function for parse() that transforms wiki markup into
1234 * HTML. Only called for $mOutputType == self::OT_HTML.
1235 *
1236 * @private
1237 *
1238 * @param string $text
1239 * @param bool $isMain
1240 * @param bool $frame
1241 *
1242 * @return string
1243 */
1249 # Hook to suspend the parser in this state
1250 if ( !wfRunHooks( 'ParserBeforeInternalParse', array( &$this, &$text, &$this->mStripState ) ) ) {
1253 }
1255 # if $frame is provided, then use $frame for replacing any variables
1257 # use frame depth to infer how include/noinclude tags should be handled
1258 # depth=0 means this is the top-level document; otherwise it's an included document
1263 }
1267 # if $frame is not provided, then use old-style replaceVariables
1269 }
1277 );
1280 # Tables need to come after variable replacement for things to work
1281 # properly; putting them before other transformations should keep
1282 # exciting things like link expansions from showing up in surprising
1283 # places.
1295 # replaceInternalLinks may sometimes leave behind
1296 # absolute URLs, which have to be masked to hide them from replaceExternalLinks
1304 }
1306 /**
1307 * Replace special strings like "ISBN xxx" and "RFC xxx" with
1308 * magic external links.
1309 *
1310 * DML
1311 * @private
1312 *
1313 * @param string $text
1314 *
1315 * @return string
1316 */
1322 '!(?: # Start cases
1326 (?:RFC|PMID)\s+([0-9]+) | # m[4]: RFC or PMID, capture number
1328 (?: 97[89] [\ \-]? )? # optional 13-digit ISBN prefix
1329 (?: [0-9] [\ \-]? ){9} # 9 digits with opt. delimiters
1330 [0-9Xx] # check digit
1335 }
1337 /**
1338 * @throws MWException
1339 * @param array $m
1340 * @return HTML|string
1341 */
1344 # Skip anchor
1347 # Skip HTML element
1350 # Free external link
1353 # RFC or PMID
1367 }
1371 # ISBN
1377 ));
1384 }
1385 }
1387 /**
1388 * Make a free external link, given a user-supplied URL
1389 *
1390 * @param string $url
1391 *
1392 * @return string HTML
1393 * @private
1394 */
1400 # The characters '<' and '>' (which were escaped by
1401 # removeHTMLtags()) should not be included in
1402 # URLs, per RFC 2396.
1407 }
1409 # Move trailing punctuation to $trail
1411 # If there is no left bracket, then consider right brackets fair game too
1414 }
1420 }
1424 # Is this an external image?
1427 # Not an image, make a link
1432 # Register it in the output object...
1433 # Replace unnecessary URL escape codes with their equivalent characters
1436 }
1439 }
1441 /**
1442 * Parse headers and return html
1443 *
1444 * @private
1445 *
1446 * @param string $text
1447 *
1448 * @return string
1449 */
1455 }
1458 }
1460 /**
1461 * Replace single quotes with HTML markup
1462 * @private
1463 *
1464 * @param string $text
1465 *
1466 * @return string The altered text
1467 */
1474 }
1478 }
1480 /**
1481 * Helper function for doAllQuotes()
1482 *
1483 * @param string $text
1484 *
1485 * @return string
1486 */
1492 }
1494 // First, do some preliminary work. This may shift some apostrophes from
1495 // being mark-up to being text. It also counts the number of occurrences
1496 // of bold and italics mark-ups.
1501 // If there are ever four apostrophes, assume the first is supposed to
1502 // be text, and the remaining three constitute mark-up for bold text.
1503 // (bug 13227: ''''foo'''' turns into ' ''' foo ' ''')
1509 // If there are more than 5 apostrophes in a row, assume they're all
1510 // text except for the last 5.
1511 // (bug 13227: ''''''foo'''''' turns into ' ''''' foo ' ''''')
1515 }
1516 // Count the number of occurrences of bold and italics mark-ups.
1524 }
1525 }
1527 // If there is an odd number of both bold and italics, it is likely
1528 // that one of the bold ones was meant to be an apostrophe followed
1529 // by italics. Which one we cannot know for certain, but it is more
1530 // likely to be one that has a single-letter word before it.
1542 }
1546 // if $firstsingleletterword is set, we don't
1547 // look at the other options, so we can bail early.
1549 }
1553 }
1554 }
1555 }
1556 }
1558 // If there is a single-letter word, use it!
1563 // If not, but there's a multi-letter word, use that one.
1567 // ... otherwise use the first one that has neither.
1568 // (notice that it is possible for all three to be -1 if, for example,
1569 // there is only one pentuple-apostrophe in the line)
1572 }
1573 }
1575 // Now let's actually convert our apostrophic mush to HTML!
1586 }
1605 }
1622 }
1642 }
1643 }
1644 }
1646 }
1647 // Now close all remaining tags. Notice that the order is important.
1650 }
1653 }
1656 }
1657 // There might be lonely ''''', so make sure we have a buffer
1660 }
1662 }
1664 /**
1665 * Replace external links (REL)
1666 *
1667 * Note: this is all very hackish and the order of execution matters a lot.
1668 * Make sure to run tests/parserTests.php if you change this code.
1669 *
1670 * @private
1671 *
1672 * @param string $text
1673 *
1674 * @throws MWException
1675 * @return string
1676 */
1685 }
1695 # The characters '<' and '>' (which were escaped by
1696 # removeHTMLtags()) should not be included in
1697 # URLs, per RFC 2396.
1702 }
1704 # If the link text is an image URL, replace it with an <img> tag
1705 # This happened by accident in the original parser, but some people used it extensively
1709 }
1713 # Set linktype for CSS - if URL==text, link is essentially free
1716 # No link text, e.g. [http://domain.tld/some.link]
1718 # Autonumber
1723 # Have link text, e.g. [http://domain.tld/some.link text]s
1724 # Check for trail
1726 }
1732 # Use the encoded URL
1733 # This means that users can paste URLs directly into the text
1734 # Funny characters like ö aren't valid in URLs anyway
1735 # This was changed in August 2004
1739 # Register link in the output object.
1740 # Replace unnecessary URL escape codes with the referenced character
1741 # This prevents spammers from hiding links from the filters
1744 }
1748 }
1750 /**
1751 * Get the rel attribute for a particular external link.
1752 *
1753 * @since 1.21
1754 * @param string|bool $url Optional URL, to extract the domain from for rel =>
1755 * nofollow if appropriate
1756 * @param Title $title Optional Title, for wgNoFollowNsExceptions lookups
1757 * @return string|null Rel attribute for $url
1758 */
1764 ) {
1766 }
1768 }
1770 /**
1771 * Get an associative array of additional HTML attributes appropriate for a
1772 * particular external link. This currently may include rel => nofollow
1773 * (depending on configuration, namespace, and the URL's domain) and/or a
1774 * target attribute (depending on configuration).
1775 *
1776 * @param string|bool $url Optional URL, to extract the domain from for rel =>
1777 * nofollow if appropriate
1778 * @return array Associative array of HTML attributes
1779 */
1786 }
1788 }
1790 /**
1791 * Replace unusual escape codes in a URL with their equivalent characters
1792 *
1793 * @deprecated since 1.24, use normalizeLinkUrl
1794 * @param string $url
1795 * @return string
1796 */
1800 }
1802 /**
1803 * Replace unusual escape codes in a URL with their equivalent characters
1804 *
1805 * This generally follows the syntax defined in RFC 3986, with special
1806 * consideration for HTTP query strings.
1807 *
1808 * @param string $url
1809 * @return string
1810 */
1812 # First, make sure unsafe characters are encoded
1816 },
1817 $url
1818 );
1823 # Fragment part - 'fragment'
1829 }
1831 # Query part - 'query' minus &=+;
1837 }
1839 # Scheme and path part - 'pchar'
1840 # (we assume no userinfo or encoded colons in the host)
1845 }
1852 # Unescape it
1855 # Leave it escaped, but use uppercase for a-f
1857 }
1858 };
1860 }
1862 /**
1863 * make an image if it's allowed, either through the global
1864 * option, through the exception, or through the on-wiki whitelist
1865 *
1866 * @param string $url
1867 *
1868 * @return string
1869 */
1874 # $imagesfrom could be either a single string or an array of strings, parse out the latter
1881 }
1882 }
1887 }
1891 ) {
1893 # Image found
1895 }
1896 }
1899 ) {
1903 );
1906 # Sanitize the regex fragment, make it case-insensitive, ignore blank entries/comments
1909 }
1911 # Image matches a whitelist entry
1914 }
1915 }
1916 }
1918 }
1920 /**
1921 * Process [[ ]] wikilinks
1922 *
1923 * @param string $s
1924 *
1925 * @return string Processed text
1926 *
1927 * @private
1928 */
1932 }
1934 /**
1935 * Process [[ ]] wikilinks (RIL)
1936 * @param string $s
1937 * @throws MWException
1938 * @return LinkHolderArray
1939 *
1940 * @private
1941 */
1948 # the % is needed to support urlencoded titles as well
1951 # Match a link having the form [[namespace:link|alternate]]trail
1953 # Match cases where there is no "]]", which might still be images
1955 }
1959 # split the entire text string on occurrences of [[
1961 # get the first element (all text up to first [[), and remove the space we added
1970 # Match the end of a line for a word that's not followed by whitespace,
1971 # e.g. in the case of 'The Arab al[[Razi]]', 'al' will be matched
1975 }
1981 }
1990 }
1993 }
1998 // @codingStandardsIgnoreStart Squiz.WhiteSpace.SemicolonSpacing.Incorrect
1999 # Loop for each link
2001 // @codingStandardsIgnoreStart
2003 # Check for excessive memory usage
2005 # Too big
2006 # Do the existence check, replace the link holders and clear the array
2009 }
2018 }
2019 # first link
2023 }
2025 }
2032 # If we get a ] at the beginning of $m[3] that means we have a link that's something like:
2033 # [[Image:Foo.jpg|[http://example.com desc]]] <- having three ] in a row fucks up,
2034 # the real problem is with the $e1 regex
2035 # See bug 1300.
2036 #
2037 # Still some problems for cases where the ] is meant to be outside punctuation,
2038 # and no image is in sight. See bug 2095.
2039 #
2043 ) {
2046 }
2047 # fix up urlencoded title texts
2049 # Should anchors '#' also be rejected?
2051 }
2054 # Invalid, but might be an image with a link in its caption
2059 }
2065 }
2071 # Don't allow internal links to pages containing
2072 # PROTO: where PROTO is a valid URL protocol; these
2073 # should be external links.
2078 }
2080 # Make subpage if necessary
2085 }
2089 # Strip off leading ':'
2091 }
2100 }
2111 # look at the next 'line' to see if we can close it there
2116 }
2119 # the first ]] closes the inner link, the second the image
2125 # if there's exactly one ]] that's fine, we'll keep looking
2128 # if $next_line is invalid too, we need look no further
2131 }
2132 }
2134 # we couldn't find the end of this imageLink, so output it raw
2135 # but don't ignore what might be perfectly normal links in the text we've examined
2138 # note: no $trail, because without an end, there *is* no trail
2141 }
2144 # note: no $trail, because without an end, there *is* no trail
2147 }
2149 }
2155 # Bug 4598 madness. Handle the quotes only if they come from the alternate part
2156 # [[Lista d''e paise d''o munno]] -> <a href="...">Lista d''e paise d''o munno</a>
2157 # [[Criticism of Harry Potter|Criticism of ''Harry Potter'']]
2158 # -> <a href="Criticism of Harry Potter">Criticism of <i>Harry Potter</i></a>
2160 }
2162 # Link not escaped by : , create the various objects
2164 # Interwikis
2170 )
2171 ) {
2172 # Bug 24502: filter duplicates
2176 }
2182 }
2189 # if no parameters were passed, $text
2190 # becomes something like "File:Foo.png",
2191 # which we don't want to pass on to the
2192 # image generator
2195 # recursively parse links inside the image caption
2196 # actually, this will parse them in any other parameters, too,
2197 # but it might be hard to fix that, and it doesn't matter ATM
2200 }
2201 # cloak any absolute URLs inside the image markup, so replaceExternalLinks() won't touch them
2206 }
2209 }
2219 }
2225 /**
2226 * Strip the whitespace Category links produce, see bug 87
2227 */
2232 }
2233 }
2235 # Self-link checking. For some languages, variants of the title are checked in
2236 # LinkHolderArray::doVariants() to allow batching the existence checks necessary
2237 # for linking to a different variant.
2241 }
2243 # NS_MEDIA is a pseudo-namespace for linking directly to a file
2244 # @todo FIXME: Should do batch file existence checks, see comment below
2247 # Give extensions a chance to select the file revision for us
2252 # Fetch and register the file (file title may be different via hooks)
2254 # Cloak with NOPARSE to avoid replacement in replaceExternalLinks
2259 }
2262 # Some titles, such as valid special pages or files in foreign repos, should
2263 # be shown as bluelinks even though they're not included in the page table
2264 #
2265 # @todo FIXME: isAlwaysKnown() can be expensive for file links; we should really do
2266 # batch file existence checks for NS_FILE and NS_MEDIA
2271 # Links will be added to the output link list after checking
2273 }
2275 }
2278 }
2280 /**
2281 * Render a forced-blue link inline; protect against double expansion of
2282 * URLs if we're in a mode that prepends full URL prefixes to internal links.
2283 * Since this little disaster has to split off the trail text to avoid
2284 * breaking URLs in the following text without breaking trails on the
2285 * wiki links, it's been made into a horrible function.
2286 *
2287 * @param Title $nt
2288 * @param string $text
2289 * @param array|string $query
2290 * @param string $trail
2291 * @param string $prefix
2292 * @return string HTML-wikitext mix oh yuck
2293 */
2294 public function makeKnownLinkHolder( $nt, $text = '', $query = array(), $trail = '', $prefix = '' ) {
2299 }
2302 }
2307 }
2309 /**
2310 * Insert a NOPARSE hacky thing into any inline links in a chunk that's
2311 * going to go through further parsing steps before inline URL expansion.
2312 *
2313 * Not needed quite as much as it used to be since free links are a bit
2314 * more sensible these days. But bracketed links are still an issue.
2315 *
2316 * @param string $text More-or-less HTML
2317 * @return string Less-or-more HTML with NOPARSE bits
2318 */
2322 }
2324 /**
2325 * Return true if subpage links should be expanded on this page.
2326 * @return bool
2327 */
2329 # Some namespaces don't allow subpages
2331 }
2333 /**
2334 * Handle link to subpage if necessary
2335 *
2336 * @param string $target The source of the link
2337 * @param string &$text The link text, modified as necessary
2338 * @return string The full name of the link
2339 * @private
2340 */
2343 }
2345 /**#@+
2346 * Used by doBlockLevels()
2347 * @private
2348 *
2349 * @return string
2350 */
2355 }
2359 }
2361 /**
2362 * getCommon() returns the length of the longest common substring
2363 * of both arguments, starting at the beginning of both.
2364 * @private
2365 *
2366 * @param string $st1
2367 * @param string $st2
2368 *
2369 * @return int
2370 */
2376 }
2381 }
2382 }
2384 }
2386 /**
2387 * These next three functions open, continue, and close the list
2388 * element appropriate to the prefix character passed into them.
2389 * @private
2390 *
2391 * @param string $char
2392 *
2393 * @return string
2394 */
2409 }
2412 }
2414 /**
2415 * TODO: document
2416 * @param string $char
2417 * @private
2418 *
2419 * @return string
2420 */
2428 }
2435 }
2436 }
2438 }
2440 /**
2441 * @todo Document
2442 * @param string $char
2443 * @private
2444 *
2445 * @return string
2446 */
2458 }
2461 }
2463 }
2464 /**#@-*/
2466 /**
2467 * Make lists from lines starting with ':', '*', '#', etc. (DBL)
2468 *
2469 * @param string $text
2470 * @param bool $linestart Whether or not this is at the start of a line.
2471 * @private
2472 * @return string The lists rendered as HTML
2473 */
2477 # Parsing through the text line by line. The main thing
2478 # happening here is handling of block-level elements p, pre,
2479 # and making lists from lines starting with * # : etc.
2480 #
2490 # Fix up $linestart
2495 }
2496 # * = ul
2497 # # = ol
2498 # ; = dt
2499 # : = dd
2504 # If not in a <pre> element, scan for and figure out what prefixes are there.
2506 # Multiple prefixes may abut each other for nested lists.
2510 # eh?
2511 # ; and : are both from definition-lists, so they're equivalent
2512 # for the purposes of determining whether or not we need to open/close
2513 # elements.
2518 # Don't interpret any other prefixes in preformatted text
2522 }
2524 # List generation
2526 # Same as the last item, so no need to deal with nesting or opening stuff
2531 # The one nasty exception: definition lists work like this:
2532 # ; title : definition text
2533 # So we check for : in the remainder text to split up the
2534 # title and definition, without b0rking links.
2539 }
2540 }
2542 # We need to open or close prefixes, or both.
2544 # Either open or close a level...
2548 # Close all the prefixes which aren't shared.
2552 }
2554 # Continue the current prefix if appropriate.
2557 }
2559 # Open prefixes where appropriate.
2562 }
2568 # @todo FIXME: This is dupe of code above
2572 }
2573 }
2575 }
2578 }
2580 }
2582 # If we have no prefixes, go to paragraph mode.
2585 # No prefix (not in list)--go to paragraph mode
2586 # XXX: use a stack for nestable elements like span, table and div
2588 '/(?:<table|<h1|<h2|<h3|<h4|<h5|<h6|<pre|<tr|'
2590 $t
2591 );
2597 $t
2598 );
2602 # @todo bug 5718: paragraph closed
2606 }
2608 while ( preg_match( '/<(\\/?)blockquote[\s>]/i', $t, $bqMatch, PREG_OFFSET_CAPTURE, $bqOffset ) ) {
2611 }
2617 ) {
2618 # pre
2623 }
2626 # paragraph
2639 }
2640 }
2649 }
2650 }
2651 }
2652 }
2654 }
2655 # somewhere above we forget to get out of pre block (bug 785)
2658 }
2663 }
2664 }
2665 }
2671 }
2672 }
2676 }
2680 }
2682 /**
2683 * Split up a string on ':', ignoring any occurrences inside tags
2684 * to prevent illegal overlapping.
2685 *
2686 * @param string $str The string to split
2687 * @param string &$before Set to everything before the ':'
2688 * @param string &$after Set to everything after the ':'
2689 * @throws MWException
2690 * @return string The position of the ':', or false if none found
2691 */
2697 # Nothing to find!
2700 }
2704 # Easy; no tag nesting to worry about
2709 }
2711 # Ugly state machine to walk through avoiding tags.
2719 # (Using the number is a performance hack for common cases)
2723 # Could be either a <start> tag or an </end> tag
2728 # We found it!
2733 }
2734 # Embedded in a tag; don't break it.
2737 # Skip ahead looking for something interesting
2740 # Nothing else interesting
2743 }
2747 # We found it!
2752 }
2753 }
2755 # Nothing else interesting to find; abort!
2756 # We're nested, but there's no close tags left. Abort!
2758 }
2759 # Skip ahead to next tag start
2762 }
2765 # In a <tag>
2772 # Slash may be followed by >?
2776 # ignore
2777 }
2788 # Illegal early close? This shouldn't happen D:
2793 }
2796 # In a </tag>
2803 }
2805 }
2809 # Yes, a self-closed tag <blah/>
2812 # Probably we're jumping the gun, and this is an attribute
2814 }
2819 }
2826 }
2833 }
2838 }
2839 }
2841 wfDebug( __METHOD__ . ": Invalid input; not enough close tags (stack $stack, state $state)\n" );
2844 }
2847 }
2849 /**
2850 * Return value of a magic variable (like PAGENAME)
2851 *
2852 * @private
2853 *
2854 * @param int $index
2855 * @param bool|PPFrame $frame
2856 *
2857 * @throws MWException
2858 * @return string
2859 */
2865 // If no title set, bad things are going to happen
2866 // later. Title should always be set since this
2867 // should only be called in the middle of a parse
2868 // operation (but the unit-tests do funky stuff)
2871 }
2873 /**
2874 * Some of these require message or data lookups and can be
2875 * expensive to check many times.
2876 */
2880 }
2881 }
2926 $value = $pageLang->getMonthAbbreviation( MWTimestamp::getLocalInstance( $ts )->format( 'n' ) );
2960 ) ) );
2970 ) ) );
2978 }
2986 }
2999 # 0 means the page doesn't exist in the database,
3000 # which means the user is previewing a new page.
3001 # The vary-revision flag must be set, because the magic word
3002 # will have a different value once the page is saved.
3005 }
3009 # Let the edit saving system know we should parse the page
3010 # *after* a revision ID has been assigned.
3016 # Let the edit saving system know we should parse the page
3017 # *after* a revision ID has been assigned. This is for null edits.
3023 # Let the edit saving system know we should parse the page
3024 # *after* a revision ID has been assigned. This is for null edits.
3030 # Let the edit saving system know we should parse the page
3031 # *after* a revision ID has been assigned. This is for null edits.
3037 # Let the edit saving system know we should parse the page
3038 # *after* a revision ID has been assigned. This is for null edits.
3044 # Let the edit saving system know we should parse the page
3045 # *after* a revision ID has been assigned. This is for null edits.
3051 # Let the edit saving system know we should parse the page
3052 # *after* a revision ID has been assigned. This is for null edits.
3058 # Let the edit saving system know we should parse the page
3059 # *after* a revision ID has been assigned. This is for null edits.
3065 # Let the edit saving system know we should parse the page
3066 # *after* a revision ID has been assigned. This is for null edits.
3107 # @bug 4594 PHP5 has it zero padded, PHP4 does not, cast to
3108 # int to remove the padding
3117 );
3126 false
3127 );
3133 # @bug 4594 PHP5 has it zero padded, PHP4 does not, cast to
3134 # int to remove the padding
3195 );
3198 }
3202 }
3205 }
3207 /**
3208 * initialise the magic variables (like CURRENTMONTHNAME) and substitution modifiers
3209 *
3210 * @private
3211 */
3220 }
3222 /**
3223 * Preprocess some wikitext and return the document tree.
3224 * This is the ghost of replace_variables().
3225 *
3226 * @param string $text The text to parse
3227 * @param int $flags Bitwise combination of:
3228 * - self::PTD_FOR_INCLUSION: Handle "<noinclude>" and "<includeonly>" as if the text is being
3229 * included. Default is to assume a direct page view.
3230 *
3231 * The generated DOM tree must depend only on the input text and the flags.
3232 * The DOM tree must be the same in OT_HTML and OT_WIKI mode, to avoid a regression of bug 4899.
3233 *
3234 * Any flag added to the $flags parameter here, or any other parameter liable to cause a
3235 * change in the DOM tree for a given text, must be passed through the section identifier
3236 * in the section edit link and thus back to extractSections().
3237 *
3238 * The output of this function is currently only cached in process memory, but a persistent
3239 * cache may be implemented at a later date which takes further advantage of these strict
3240 * dependency requirements.
3241 *
3242 * @return PPNode
3243 */
3247 }
3249 /**
3250 * Return a three-element array: leading whitespace, string contents, trailing whitespace
3251 *
3252 * @param string $s
3253 *
3254 * @return array
3255 */
3265 }
3267 }
3269 /**
3270 * Replace magic variables, templates, and template arguments
3271 * with the appropriate text. Templates are substituted recursively,
3272 * taking care to avoid infinite loops.
3273 *
3274 * Note that the substitution depends on value of $mOutputType:
3275 * self::OT_WIKI: only {{subst:}} templates
3276 * self::OT_PREPROCESS: templates but not extension tags
3277 * self::OT_HTML: all templates and extension tags
3278 *
3279 * @param string $text The text to transform
3280 * @param bool|PPFrame $frame Object describing the arguments passed to the
3281 * template. Arguments may also be provided as an associative array, as
3282 * was the usual case before MW1.12. Providing arguments this way may be
3283 * useful for extensions wishing to perform variable replacement
3284 * explicitly.
3285 * @param bool $argsOnly Only do argument (triple-brace) expansion, not
3286 * double-brace expansion.
3287 * @return string
3288 */
3290 # Is there any text? Also, Prevent too big inclusions!
3293 }
3302 }
3310 }
3312 /**
3313 * Clean up argument array - refactored in 1.9 so parserfunctions can use it, too.
3314 *
3315 * @param array $args
3316 *
3317 * @return array
3318 */
3331 }
3334 }
3335 }
3336 }
3339 }
3341 /**
3342 * Warn the user when a parser limitation is reached
3343 * Will warn at most once the user per limitation type
3344 *
3345 * @param string $limitationType Should be one of:
3346 * 'expensive-parserfunction' (corresponding messages:
3347 * 'expensive-parserfunction-warning',
3348 * 'expensive-parserfunction-category')
3349 * 'post-expand-template-argument' (corresponding messages:
3350 * 'post-expand-template-argument-warning',
3351 * 'post-expand-template-argument-category')
3352 * 'post-expand-template-inclusion' (corresponding messages:
3353 * 'post-expand-template-inclusion-warning',
3354 * 'post-expand-template-inclusion-category')
3355 * 'node-count-exceeded' (corresponding messages:
3356 * 'node-count-exceeded-warning',
3357 * 'node-count-exceeded-category')
3358 * 'expansion-depth-exceeded' (corresponding messages:
3359 * 'expansion-depth-exceeded-warning',
3360 * 'expansion-depth-exceeded-category')
3361 * @param string|int|null $current Current value
3362 * @param string|int|null $max Maximum allowed, when an explicit limit has been
3363 * exceeded, provide the values (optional)
3364 */
3366 # does no harm if $current and $max are present but are unnecessary for the message
3371 }
3373 /**
3374 * Return the text of a template, after recursively
3375 * replacing any variables or templates within the template.
3376 *
3377 * @param array $piece The parts of the template
3378 * $piece['title']: the title, i.e. the part before the |
3379 * $piece['parts']: the parameter array
3380 * $piece['lineStart']: whether the brace was at the start of a line
3381 * @param PPFrame $frame The current frame, contains template arguments
3382 * @throws Exception
3383 * @return string The text of the template
3384 */
3389 // Flags
3391 // $text has been filled
3393 // wiki markup in $text should be escaped
3395 // $text is HTML, armour it against wikitext transformation
3397 // Force interwiki transclusion to be done in raw mode not rendered
3399 // $text is a DOM node needing expansion in a child frame
3401 // $text is a DOM node needing expansion in the current frame
3404 # Title object, where $text came from
3407 # $part1 is the bit before the first |, and must contain only title characters.
3408 # Various prefixes will be stripped from it later.
3413 # Original title text preserved for various purposes
3416 # $args is a list of argument nodes, starting from index 0, not including $part1
3417 # @todo FIXME: If piece['parts'] is null then the call to getLength()
3418 # below won't work b/c this $args isn't an object
3424 # SUBST
3430 # Possibilities for substMatch: "subst", "safesubst" or FALSE
3431 # Decide whether to expand template or keep wikitext as-is.
3437 }
3443 }
3444 }
3449 }
3450 }
3452 # Variables
3459 }
3461 }
3462 }
3464 # MSG, MSGNW and RAW
3466 # Check for MSGNW:
3471 # Remove obsolete MSG:
3474 }
3476 # Check for RAW:
3480 }
3481 }
3484 # Parser functions
3494 }
3501 }
3503 # The interface for parser functions allows for extracting
3504 # flags into the local scope. Extract any forwarded flags
3505 # here.
3507 }
3509 }
3511 # Finish mangling title and then check for loops.
3512 # Set $title to a Title object and $titleText to the PDBK
3515 # Split the title into page and subpage
3521 }
3525 # Check for language variants if the template is not found
3528 }
3529 # Do recursion depth check
3537 }
3538 }
3539 }
3541 # Load from database
3544 # Too many unique items can kill profiling DBs/collectors
3547 }
3553 ) {
3554 // Pass the template arguments as URL parameters.
3555 // "uselang" will have no effect since the Language object
3556 // is forced to the one defined in ParserOptions.
3565 }
3566 }
3568 // Create a new context to execute the special page
3581 }
3591 }
3592 }
3594 # If the title is valid but undisplayable, make a link to it
3598 }
3600 # Interwiki transclusion
3606 # Preprocess it like a template
3609 }
3611 }
3613 # Do infinite loop check
3614 # This has to be done after redirect resolution to avoid infinite loops via redirects
3621 }
3623 }
3625 # If we haven't found text to substitute by now, we're done
3626 # Recover the source wikitext and return it
3631 }
3634 }
3636 # Expand DOM-style return values in a child frame
3638 # Clean up argument array
3644 # Expansion is eligible for the empty-frame cache
3647 # Uncached expansion
3649 }
3650 }
3654 }
3658 }
3660 # Replace raw HTML by a placeholder
3664 # Escape nowiki-style return values
3669 ) {
3670 # Bug 529: if the template begins with a table or block-level
3671 # element, it should be treated as beginning a new line.
3672 # This behavior is somewhat controversial.
3674 }
3677 # Error, oversize inclusion
3679 # Make a working, properly escaped link if possible (bug 23588)
3682 # This will probably not be a working link, but at least it may
3683 # provide some hint of where the problem is
3686 }
3690 }
3696 }
3700 }
3702 /**
3703 * Call a parser function and return an array with text and flags.
3704 *
3705 * The returned array will always contain a boolean 'found', indicating
3706 * whether the parser function was found or not. It may also contain the
3707 * following:
3708 * text: string|object, resulting wikitext or PP DOM object
3709 * isHTML: bool, $text is HTML, armour it against wikitext transformation
3710 * isChildObj: bool, $text is a DOM node needing expansion in a child frame
3711 * isLocalObj: bool, $text is a DOM node needing expansion in the current frame
3712 * nowiki: bool, wiki markup in $text should be escaped
3713 *
3714 * @since 1.21
3715 * @param PPFrame $frame The current frame, contains template arguments
3716 * @param string $function Function name
3717 * @param array $args Arguments to the function
3718 * @throws MWException
3719 * @return array
3720 */
3726 # Case sensitive functions
3730 # Case insensitive functions
3737 }
3738 }
3743 # Workaround for PHP bug 35229 and similar
3748 }
3752 # Convert arguments to PPNodes and collect for appending to $allArgs
3759 }
3760 }
3762 # Add a frame parameter, and pass the arguments as an array
3766 # Convert arguments to plain text and append to $allArgs
3774 }
3775 }
3776 }
3780 # The interface for function hooks allows them to return a wikitext
3781 # string or an array containing the string and any flags. This mungs
3782 # things around to match what this method should return.
3787 );
3791 }
3795 );
3796 }
3802 }
3805 }
3810 }
3815 }
3817 /**
3818 * Get the semi-parsed DOM representation of a template with a given title,
3819 * and its redirect destination title. Cached.
3820 *
3821 * @param Title $title
3822 *
3823 * @return array
3824 */
3833 }
3836 }
3838 # Cache miss, go to the database
3844 }
3852 }
3855 }
3857 /**
3858 * Fetch the current revision of a given title. Note that the revision
3859 * (and even the title) may not exist in the database, so everything
3860 * contributing to the output of the parser should use this method
3861 * where possible, rather than getting the revisions themselves. This
3862 * method also caches its results, so using it benefits performance.
3863 *
3864 * @since 1.24
3865 * @param Title $title
3866 * @return Revision
3867 */
3872 }
3875 // Defaults to Parser::statelessFetchRevision()
3877 );
3878 }
3880 }
3882 /**
3883 * Wrapper around Revision::newFromTitle to allow passing additional parameters
3884 * without passing them on to it.
3885 *
3886 * @since 1.24
3887 * @param Title $title
3888 * @param Parser|bool $parser
3889 * @return Revision
3890 */
3893 }
3895 /**
3896 * Fetch the unparsed text of a template and register a reference to it.
3897 * @param Title $title
3898 * @return array ( string or false, Title )
3899 */
3901 // Defaults to Parser::statelessFetchTemplate()
3910 // If we transclude ourselves, the final result
3911 // will change based on the new version of the page
3913 }
3914 }
3915 }
3917 }
3919 /**
3920 * Fetch the unparsed text of a template and register a reference to it.
3921 * @param Title $title
3922 * @return string|bool
3923 */
3927 }
3929 /**
3930 * Static function to get a template
3931 * Can be overridden via ParserOptions::setTemplateCallback().
3932 *
3933 * @param Title $title
3934 * @param bool|Parser $parser
3935 *
3936 * @return array
3937 */
3943 # Loop to fetch the article, with up to 1 redirect
3945 # Give extensions a chance to select the revision instead
3956 );
3958 }
3959 # Get the revision
3966 }
3968 # If there is no current revision, there is no page
3972 }
3979 # We fetched a rev from a different title; register it too...
3984 }
3993 }
4000 }
4005 }
4008 }
4009 # Redirect?
4012 }
4017 }
4019 /**
4020 * Fetch a file and its title and register a reference to it.
4021 * If 'broken' is a key in $options then the file will appear as a broken thumbnail.
4022 * @param Title $title
4023 * @param array $options Array of options to RepoGroup::findFile
4024 * @return File|bool
4025 */
4029 }
4031 /**
4032 * Fetch a file and its title and register a reference to it.
4033 * If 'broken' is a key in $options then the file will appear as a broken thumbnail.
4034 * @param Title $title
4035 * @param array $options Array of options to RepoGroup::findFile
4036 * @return array ( File or false, Title of file )
4037 */
4043 # Register the file as a dependency...
4046 # Update fetched file title
4049 }
4051 }
4053 /**
4054 * Helper function for fetchFileAndTitle.
4055 *
4056 * Also useful if you need to fetch a file but not use it yet,
4057 * for example to get the file's handler.
4058 *
4059 * @param Title $title
4060 * @param array $options Array of options to RepoGroup::findFile
4061 * @return File|bool
4062 */
4070 }
4072 }
4074 /**
4075 * Transclude an interwiki link.
4076 *
4077 * @param Title $title
4078 * @param string $action
4079 *
4080 * @return string
4081 */
4087 }
4093 }
4095 }
4097 /**
4098 * @param string $url
4099 * @return mixed|string
4100 */
4109 }
4116 // Though we failed to fetch the content, this status is useless.
4121 }
4128 ) );
4130 }
4132 /**
4133 * Triple brace replacement -- used for template arguments
4134 * @private
4135 *
4136 * @param array $piece
4137 * @param PPFrame $frame
4138 *
4139 * @return array
4140 */
4154 )
4155 ) {
4156 # No match in frame, use the supplied default
4158 }
4162 }
4165 # No match anywhere
4167 }
4170 }
4175 }
4179 }
4181 /**
4182 * Return the text to be used for a given extension tag.
4183 * This is the ghost of strip().
4184 *
4185 * @param array $params Associative array of parameters:
4186 * name PPNode for the tag name
4187 * attr PPNode for unparsed text where tag attributes are thought to be
4188 * attributes Optional associative array of parsed attributes
4189 * inner Contents of extension element
4190 * noClose Original text did not have a close tag
4191 * @param PPFrame $frame
4192 *
4193 * @throws MWException
4194 * @return string
4195 */
4209 }
4215 }
4218 # Workaround for PHP bug 35229 and similar
4221 }
4228 }
4234 }
4237 # Extract flags to local scope (to override $markerType)
4242 }
4246 }
4251 }
4252 }
4258 }
4259 }
4269 }
4271 }
4273 /**
4274 * Increment an include size counter
4275 *
4276 * @param string $type The type of expansion
4277 * @param int $size The size of the text
4278 * @return bool False if this inclusion would take it over the maximum, true otherwise
4279 */
4286 }
4287 }
4289 /**
4290 * Increment the expensive function count
4291 *
4292 * @return bool False if the limit has been exceeded
4293 */
4297 }
4299 /**
4300 * Strip double-underscore items like __NOGALLERY__ and __NOTOC__
4301 * Fills $this->mDoubleUnderscores, returns the modified text
4302 *
4303 * @param string $text
4304 *
4305 * @return string
4306 */
4310 # The position of __TOC__ needs to be recorded
4316 # Set a placeholder. At the end we'll fill it in with the TOC.
4319 # Only keep the first one.
4321 }
4323 # Now match and remove the rest of them
4329 }
4332 }
4335 ) {
4337 }
4338 # (bug 8068) Allow control over whether robots index a page.
4339 #
4340 # @todo FIXME: Bug 14899: __INDEX__ always overrides __NOINDEX__ here! This
4341 # is not desirable, the last one on the page should win.
4345 }
4349 }
4351 # Cache all double underscores in the database
4354 }
4358 }
4360 /**
4361 * @see ParserOutput::addTrackingCategory()
4362 * @param string $msg Message key
4363 * @return bool Whether the addition was successful
4364 */
4367 }
4369 /**
4370 * This function accomplishes several tasks:
4371 * 1) Auto-number headings if that option is enabled
4372 * 2) Add an [edit] link to sections for users who have enabled the option and can edit the page
4373 * 3) Add a Table of contents on the top for users who have enabled the option
4374 * 4) Auto-anchor headings
4375 *
4376 * It loops through all headlines, collects the necessary data, then splits up the
4377 * string and re-inserts the newly formatted headlines.
4378 *
4379 * @param string $text
4380 * @param string $origText Original, untouched wikitext
4381 * @param bool $isMain
4382 * @return mixed|string
4383 * @private
4384 */
4388 # Inhibit editsection links if requested in the page
4394 }
4397 }
4399 # Get all headlines for numbering them and adding funky stuff like [edit]
4400 # links - this is for later, but we need the number of headlines right now
4405 $matches
4406 );
4408 # if there are fewer than 4 headlines in the article, do not show TOC
4409 # unless it's been explicitly enabled.
4413 # Allow user to stipulate that a page should have a "new section"
4414 # link added via __NEWSECTIONLINK__
4417 }
4419 # Allow user to remove the "new section"
4420 # link via __NONEWSECTIONLINK__
4423 }
4425 # if the string __FORCETOC__ (not case-sensitive) occurs in the HTML,
4426 # override above conditions and always show TOC above first header
4430 }
4432 # headline counter
4436 # Ugh .. the TOC should have neat indentation levels which can be
4437 # passed to the skin functions. These are determined here
4469 }
4473 }
4477 # Increase TOC level
4484 }
4486 # Decrease TOC level, find level to jump to
4490 # Found last matching level
4494 # Found first matching level below current level
4497 }
4498 }
4501 }
4504 # Unindent only if the previous toc level was shown :p
4509 }
4510 }
4512 # No change in level, end TOC line
4515 }
4516 }
4520 # count number of headlines for each level
4527 }
4530 }
4531 }
4533 # The safe header is a version of the header text safe to use for links
4535 # Remove link placeholders by the link text.
4536 # <!--LINK number-->
4537 # turns into
4538 # link text with suffix
4539 # Do this before unstrip since link text can contain strip markers
4542 # Avoid insertion of weird stuff like <math> by expanding the relevant sections
4545 # Strip out HTML (first regex removes any tag not allowed)
4546 # Allowed tags are:
4547 # * <sup> and <sub> (bug 8393)
4548 # * <i> (bug 26375)
4549 # * <b> (r105284)
4550 # * <span dir="rtl"> and <span dir="ltr"> (bug 35167)
4551 #
4552 # We strip any parameter from accepted tags (second regex), except dir="rtl|ltr" from <span>,
4553 # to allow setting directionality in toc items.
4558 ),
4560 $safeHeadline
4561 );
4564 # For the anchor, strip out HTML-y stuff period
4568 # Save headline for section edit hint before it's escaped
4572 # For reverse compatibility, provide an id that's
4573 # HTML4-compatible, like we used to.
4574 #
4575 # It may be worth noting, academically, that it's possible for
4576 # the legacy anchor to conflict with a non-legacy headline
4577 # anchor on the page. In this case likely the "correct" thing
4578 # would be to either drop the legacy anchors or make sure
4579 # they're numbered first. However, this would require people
4580 # to type in section names like "abc_.D7.93.D7.90.D7.A4"
4581 # manually, so let's not bother worrying about it.
4587 # No reason to have both (in fact, we can't)
4589 }
4594 }
4596 # HTML names must be case-insensitively unique (bug 10721).
4597 # This does not apply to Unicode characters per
4598 # http://dev.w3.org/html5/spec/infrastructure.html#case-sensitivity-and-string-comparison
4599 # @todo FIXME: We may be changing them depending on the current locale.
4605 }
4607 # count how many in assoc. array so we can track dupes in anchors
4612 }
4617 }
4619 # Don't number the heading if it is the only one (looks silly)
4621 # the two are different if the line contains a link
4625 $numbering
4627 }
4629 # Create the anchor for linking from the TOC to the section
4634 }
4637 }
4641 }
4643 # Add the section to the section tree
4644 # Find the DOM node for this header
4651 }
4652 }
4656 }
4666 );
4668 # give headline the correct <h#> tag
4670 // Output edit section links as markers with styles that can be customized by skins
4672 # Put a T flag in the section identifier, to indicate to extractSections()
4673 # that sections inside <includeonly> should be counted.
4681 }
4682 // We use a bit of pesudo-xml for editsection markers. The
4683 // language converter is run later on. Using a UNIQ style marker
4684 // leads to the converter screwing up the tokens when it
4685 // converts stuff. And trying to insert strip tags fails too. At
4686 // this point all real inputted tags have already been escaped,
4687 // so we don't have to worry about a user trying to input one of
4688 // these markers directly. We use a page and section attribute
4689 // to stop the language converter from converting these
4690 // important bits of data, but put the headline hint inside a
4691 // content block because the language converter is supposed to
4692 // be able to convert that piece of data.
4693 // Gets replaced with html in ParserOutput::getText
4700 }
4703 }
4709 }
4713 # Never ever show TOC if no headers
4716 }
4721 }
4726 }
4730 }
4732 # split up and insert constructed headlines
4736 // build an array of document sections
4739 // $head is zero-based, sections aren't.
4744 }
4746 /**
4747 * Send a hook, one per section.
4748 * The idea here is to be able to make section-level DIVs, but to do so in a
4749 * lower-impact, more correct way than r50769
4750 *
4751 * $this : caller
4752 * $section : the section number
4753 * &$sectionContent : ref to the content of the section
4754 * $showEditLinks : boolean describing whether this section has an edit link
4755 */
4759 }
4762 // append the TOC at the beginning
4763 // Top anchor now in skin
4765 }
4773 }
4774 }
4776 /**
4777 * Transform wiki markup when saving a page by doing "\r\n" -> "\n"
4778 * conversion, substitting signatures, {{subst:}} templates, etc.
4779 *
4780 * @param string $text The text to transform
4781 * @param Title $title The Title object for the current article
4782 * @param User $user The User object describing the current user
4783 * @param ParserOptions $options Parsing options
4784 * @param bool $clearState Whether to clear the parser state first
4785 * @return string The altered wiki markup
4786 */
4789 ) {
4792 }
4798 );
4802 }
4808 }
4810 /**
4811 * Pre-save transform helper function
4812 *
4813 * @param string $text
4814 * @param User $user
4815 *
4816 * @return string
4817 */
4821 # Note: This is the timestamp saved as hardcoded wikitext to
4822 # the database, we use $wgContLang here in order to give
4823 # everyone the same signature and use the default one rather
4824 # than the one selected in each user's preferences.
4825 # (see also bug 12815)
4831 # Allow translation of timezones through wiki. format() can return
4832 # whatever crap the system uses, localised or not, so we cannot
4833 # ship premade translations.
4838 }
4842 # Variable replacement
4843 # Because mOutputType is OT_WIKI, this will only process {{subst:xxx}} type tags
4846 # This works almost by chance, as the replaceVariables are done before the getUserSig(),
4847 # which may corrupt this parser instance via its wfMessage()->text() call-
4849 # Signatures
4855 ) );
4857 # Context links ("pipe tricks"): [[|name]] and [[name (context)|]]
4861 // [[ns:page (context)|]]
4863 // [[ns:page（context）|]] (double-width brackets, added in r40257)
4865 // [[ns:page (context), context|]] (using either single or double-width comma)
4867 // [[|page]] (reverse pipe trick: add context from page title)
4870 # try $p1 first, to turn "[[A, B (C)|]]" into "[[A, B (C)|A, B]]"
4882 # if there's no context, don't bother duplicating the title
4884 }
4886 # Trim trailing whitespace
4890 }
4892 /**
4893 * Fetch the user's signature text, if any, and normalize to
4894 * validated, ready-to-insert wikitext.
4895 * If you have pre-fetched the nickname or the fancySig option, you can
4896 * specify them here to save a database query.
4897 * Do not reuse this parser instance after calling getUserSig(),
4898 * as it may have changed if it's the $wgParser.
4899 *
4900 * @param User $user
4901 * @param string|bool $nickname Nickname to use or false to use user's default nickname
4902 * @param bool|null $fancySig whether the nicknname is the complete signature
4903 * or null to use default value
4904 * @return string
4905 */
4911 # If not given, retrieve from the user object.
4914 }
4918 }
4926 # Sig. might contain markup; validate this
4928 # Validated; clean up (if needed) and return it
4931 # Failed to validate; fall back to the default
4934 }
4935 }
4937 # Make sure nickname doesnt get a sig in a sig
4940 # If we're still here, make it a link to the user page
4947 }
4949 /**
4950 * Check that the user's signature contains no bad XML
4951 *
4952 * @param string $text
4953 * @return string|bool An expanded string, or false if invalid.
4954 */
4957 }
4959 /**
4960 * Clean up signature text
4961 *
4962 * 1) Strip ~~~, ~~~~ and ~~~~~ out of signatures @see cleanSigInSig
4963 * 2) Substitute all transclusions
4964 *
4965 * @param string $text
4966 * @param bool $parsing Whether we're cleaning (preferences save) or parsing
4967 * @return string Signature text
4968 */
4974 }
4976 # Option to disable this feature
4979 }
4981 # @todo FIXME: Regex doesn't respect extension tags or nowiki
4982 # => Move this logic to braceSubstitution()
4984 $substRegex = '/\{\{(?!(?:' . $substWord->getBaseRegex() . '))/x' . $substWord->getRegexCase();
4995 }
4998 }
5000 /**
5001 * Strip ~~~, ~~~~ and ~~~~~ out of signatures
5002 *
5003 * @param string $text
5004 * @return string Signature text with /~{3,5}/ removed
5005 */
5009 }
5011 /**
5012 * Set up some variables which are usually set up in parse()
5013 * so that an external function can call some class members with confidence
5014 *
5015 * @param Title|null $title
5016 * @param ParserOptions $options
5017 * @param int $outputType
5018 * @param bool $clearState
5019 */
5022 ) {
5024 }
5026 /**
5027 * @param Title|null $title
5028 * @param ParserOptions $options
5029 * @param int $outputType
5030 * @param bool $clearState
5031 */
5034 ) {
5040 }
5041 }
5043 /**
5044 * Wrapper for preprocess()
5045 *
5046 * @param string $text The text to preprocess
5047 * @param ParserOptions $options Options
5048 * @param Title|null $title Title object or null to use $wgTitle
5049 * @return string
5050 */
5054 # Guard against infinite recursion
5057 }
5064 }
5071 }
5073 /**
5074 * Create an HTML-style tag, e.g. "<yourtag>special text</yourtag>"
5075 * The callback should have the following form:
5076 * function myParserHook( $text, $params, $parser, $frame ) { ... }
5077 *
5078 * Transform and return $text. Use $parser for any required context, e.g. use
5079 * $parser->getTitle() and $parser->getOptions() not $wgTitle or $wgOut->mParserOptions
5080 *
5081 * Hooks may return extended information by returning an array, of which the
5082 * first numbered element (index 0) must be the return string, and all other
5083 * entries are extracted into local variables within an internal function
5084 * in the Parser class.
5085 *
5086 * This interface (introduced r61913) appears to be undocumented, but
5087 * 'markerName' is used by some core tag hooks to override which strip
5088 * array their results are placed in. **Use great caution if attempting
5089 * this interface, as it is not documented and injudicious use could smash
5090 * private variables.**
5091 *
5092 * @param string $tag The tag to use, e.g. 'hook' for "<hook>"
5093 * @param callable $callback The callback function (and object) to use for the tag
5094 * @throws MWException
5095 * @return callable|null The old value of the mTagHooks array associated with the hook
5096 */
5101 }
5106 }
5109 }
5111 /**
5112 * As setHook(), but letting the contents be parsed.
5113 *
5114 * Transparent tag hooks are like regular XML-style tag hooks, except they
5115 * operate late in the transformation sequence, on HTML instead of wikitext.
5116 *
5117 * This is probably obsoleted by things dealing with parser frames?
5118 * The only extension currently using it is geoserver.
5119 *
5120 * @since 1.10
5121 * @todo better document or deprecate this
5122 *
5123 * @param string $tag The tag to use, e.g. 'hook' for "<hook>"
5124 * @param callable $callback The callback function (and object) to use for the tag
5125 * @throws MWException
5126 * @return callable|null The old value of the mTagHooks array associated with the hook
5127 */
5132 }
5133 $oldVal = isset( $this->mTransparentTagHooks[$tag] ) ? $this->mTransparentTagHooks[$tag] : null;
5137 }
5139 /**
5140 * Remove all tag hooks
5141 */
5146 }
5148 /**
5149 * Create a function, e.g. {{sum:1|2|3}}
5150 * The callback function should have the form:
5151 * function myParserFunction( &$parser, $arg1, $arg2, $arg3 ) { ... }
5152 *
5153 * Or with SFH_OBJECT_ARGS:
5154 * function myParserFunction( $parser, $frame, $args ) { ... }
5155 *
5156 * The callback may either return the text result of the function, or an array with the text
5157 * in element 0, and a number of flags in the other elements. The names of the flags are
5158 * specified in the keys. Valid flags are:
5159 * found The text returned is valid, stop processing the template. This
5160 * is on by default.
5161 * nowiki Wiki markup in the return value should be escaped
5162 * isHTML The returned text is HTML, armour it against wikitext transformation
5163 *
5164 * @param string $id The magic word ID
5165 * @param callable $callback The callback function (and object) to use
5166 * @param int $flags A combination of the following flags:
5167 * SFH_NO_HASH No leading hash, i.e. {{plural:...}} instead of {{#if:...}}
5168 *
5169 * SFH_OBJECT_ARGS Pass the template arguments as PPNode objects instead of text. This
5170 * allows for conditional expansion of the parse tree, allowing you to eliminate dead
5171 * branches and thus speed up parsing. It is also possible to analyse the parse tree of
5172 * the arguments, and to control the way they are expanded.
5173 *
5174 * The $frame parameter is a PPFrame. This can be used to produce expanded text from the
5175 * arguments, for instance:
5176 * $text = isset( $args[0] ) ? $frame->expand( $args[0] ) : '';
5177 *
5178 * For technical reasons, $args[0] is pre-expanded and will be a string. This may change in
5179 * future versions. Please call $frame->expand() on it anyway so that your code keeps
5180 * working if/when this is changed.
5181 *
5182 * If you want whitespace to be trimmed from $args, you need to do it yourself, post-
5183 * expansion.
5184 *
5185 * Please read the documentation in includes/parser/Preprocessor.php for more information
5186 * about the methods available in PPFrame and PPNode.
5187 *
5188 * @throws MWException
5189 * @return string|callable The old callback function for this name, if any
5190 */
5197 # Add to function cache
5201 }
5207 # Case
5210 }
5211 # Add leading hash
5214 }
5215 # Remove trailing colon
5218 }
5220 }
5222 }
5224 /**
5225 * Get all registered function hook identifiers
5226 *
5227 * @return array
5228 */
5231 }
5233 /**
5234 * Create a tag function, e.g. "<test>some stuff</test>".
5235 * Unlike tag hooks, tag functions are parsed at preprocessor level.
5236 * Unlike parser functions, their content is not preprocessed.
5237 * @param string $tag
5238 * @param callable $callback
5239 * @param int $flags
5240 * @throws MWException
5241 * @return null
5242 */
5247 }
5254 }
5257 }
5259 /**
5260 * @todo FIXME: Update documentation. makeLinkObj() is deprecated.
5261 * Replace "<!--LINK-->" link placeholders with actual links, in the buffer
5262 * Placeholders created in Skin::makeLinkObj()
5263 *
5264 * @param string $text
5265 * @param int $options
5266 *
5267 * @return array Array of link CSS classes, indexed by PDBK.
5268 */
5271 }
5273 /**
5274 * Replace "<!--LINK-->" link placeholders with plain text of links
5275 * (not HTML-formatted).
5276 *
5277 * @param string $text
5278 * @return string
5279 */
5282 }
5284 /**
5285 * Renders an image gallery from a text with one line per image.
5286 * text labels may be given by using |-style alternative text. E.g.
5287 * Image:one.jpg|The number "1"
5288 * Image:tree.jpg|A tree
5289 * given as text will return the HTML of a gallery with two images,
5290 * labeled 'The number "1"' and
5291 * 'A tree'.
5292 *
5293 * @param string $text
5294 * @param array $params
5295 * @return string HTML
5296 */
5303 }
5308 // If invalid type set, fallback to default.
5310 }
5323 }
5329 }
5332 }
5335 }
5338 }
5345 # match lines like these:
5346 # Image:someimage.jpg|This is some image
5349 # Skip empty lines
5352 }
5356 }
5359 # Bogus title. Ignore these so we don't bomb out later.
5361 }
5363 # We need to get what handler the file uses, to figure out parameters.
5364 # Note, a hook can overide the file name, and chose an entirely different
5365 # file (which potentially could be of a different type and have different handler).
5370 # Don't register it now, as ImageGallery does that later.
5378 );
5381 // We don't want people to specify per-image widths.
5382 // Additionally the width parameter would need special casing anyhow.
5384 }
5394 // look for an |alt= definition while trying not to break existing
5395 // captions with multiple pipes (|) in it, until a more sensible grammar
5396 // is defined for images in galleries
5398 // FIXME: Doing recursiveTagParse at this stage, and the trim before
5399 // splitting on '|' is a bit odd, and different from makeImage.
5416 //check to see if link matches an absolute url, if not then it must be a wiki link.
5423 }
5424 }
5427 // Must be a handler specific parameter.
5431 // Guess not. Append it to the caption.
5434 }
5435 }
5438 // concatenate all other pipes
5440 }
5441 }
5442 // remove the first pipe
5444 }
5447 }
5452 }
5454 /**
5455 * @param string $handler
5456 * @return array
5457 */
5463 }
5465 # Initialise static lists
5472 );
5480 }
5481 }
5482 }
5484 # Add handler params
5490 }
5491 }
5494 }
5495 return array( $this->mImageParams[$handlerClass], $this->mImageParamsMagicArray[$handlerClass] );
5496 }
5498 /**
5499 * Parse image options text and use it to make an image
5500 *
5501 * @param Title $title
5502 * @param string $options
5503 * @param LinkHolderArray|bool $holders
5504 * @return string HTML
5505 */
5507 # Check if the options text is of the form "options|alt text"
5508 # Options are:
5509 # * thumbnail make a thumbnail with enlarge-icon and caption, alignment depends on lang
5510 # * left no resizing, just left align. label is used for alt= only
5511 # * right same, but right aligned
5512 # * none same, but not aligned
5513 # * ___px scale to ___ pixels width, no aligning. e.g. use in taxobox
5514 # * center center the image
5515 # * frame Keep original image size, no magnify-button.
5516 # * framed Same as "frame"
5517 # * frameless like 'thumb' but without a frame. Keeps user preferences for width
5518 # * upright reduce width for upright images, rounded to full __0 px
5519 # * border draw a 1px border around the image
5520 # * alt Text for HTML alt attribute (defaults to empty)
5521 # * class Set a class for img node
5522 # * link Set the target of the image link. Can be external, interwiki, or local
5523 # vertical-align values (no % or length right now):
5524 # * baseline
5525 # * sub
5526 # * super
5527 # * top
5528 # * text-top
5529 # * middle
5530 # * bottom
5531 # * text-bottom
5535 # Give extensions a chance to select the file revision for us
5540 # Fetch and register the file (file title may be different via hooks)
5543 # Get parameter map
5550 }
5552 # Process the input parameters
5564 # Special case; width and height come in one variable together
5572 }
5573 }
5579 }
5580 }
5581 # else no validation -- bug 13436
5584 # Validate handler parameter
5587 # Validate internal parameters
5592 # @todo FIXME: Possibly check validity here for
5593 # manualthumb? downstream behavior seems odd with
5594 # missing manual thumbs.
5611 }
5613 }
5621 }
5622 }
5627 // use first appearing option, discard others.
5632 # Most other things appear to be empty or numeric...
5634 }
5635 }
5639 }
5640 }
5641 }
5644 }
5645 }
5647 # Process alignment parameters
5650 }
5653 }
5657 # Will the image be presented in a frame, with the caption below?
5663 # In the old days, [[Image:Foo|text...]] would set alt text. Later it
5664 # came to also set the caption, ordinary text after the image -- which
5665 # makes no sense, because that just repeats the text multiple times in
5666 # screen readers. It *also* came to set the title attribute.
5667 #
5668 # Now that we have an alt attribute, we should not set the alt text to
5669 # equal the caption: that's worse than useless, it just repeats the
5670 # text. This is the framed/thumbnail case. If there's no caption, we
5671 # use the unnamed parameter for alt text as well, just for the time be-
5672 # ing, if the unnamed param is set and the alt param is not.
5673 #
5674 # For the future, we need to figure out if we want to tweak this more,
5675 # e.g., introducing a title= parameter for the title; ignoring the un-
5676 # named parameter entirely for images without a caption; adding an ex-
5677 # plicit caption= parameter and preserving the old magic unnamed para-
5678 # meter for BC; ...
5681 # No caption or alt text, add the filename as the alt text so
5682 # that screen readers at least get some description of the image
5684 }
5685 # Do not set $params['frame']['title'] because tooltips don't make sense
5686 # for framed images
5689 # No alt text, use the "caption" for the alt text
5693 # No caption, fall back to using the filename for the
5694 # alt text
5696 }
5697 }
5698 # Use the "caption" for the tooltip text
5700 }
5704 # Linker does the rest
5709 # Give the handler a chance to modify the parser object
5712 }
5715 }
5717 /**
5718 * @param string $caption
5719 * @param LinkHolderArray|bool $holders
5720 * @return mixed|string
5721 */
5723 # Strip bad stuff out of the title (tooltip). We can't just use
5724 # replaceLinkHoldersText() here, because if this function is called
5725 # from replaceInternalLinks2(), mLinkHolders won't be up-to-date.
5730 }
5732 # make sure there are no placeholders in thumbnail attributes
5733 # that are later expanded to html- so expand them now and
5734 # remove the tags
5739 }
5741 /**
5742 * Set a flag in the output object indicating that the content is dynamic and
5743 * shouldn't be cached.
5744 */
5750 }
5753 }
5755 /**
5756 * Callback from the Sanitizer for expanding items found in HTML attribute
5757 * values, so they can be safely tested and escaped.
5758 *
5759 * @param string $text
5760 * @param bool|PPFrame $frame
5761 * @return string
5762 */
5767 }
5769 /**
5770 * Accessor
5771 *
5772 * @return array
5773 */
5779 );
5780 }
5782 /**
5783 * Replace transparent tags in $text with the values given by the callbacks.
5784 *
5785 * Transparent tag hooks are like regular XML-style tag hooks, except they
5786 * operate late in the transformation sequence, on HTML instead of wikitext.
5787 *
5788 * @param string $text
5789 *
5790 * @return string
5791 */
5805 );
5808 }
5810 }
5812 }
5814 /**
5815 * Break wikitext input into sections, and either pull or replace
5816 * some particular section's text.
5817 *
5818 * External callers should use the getSection and replaceSection methods.
5819 *
5820 * @param string $text Page wikitext
5821 * @param string|number $sectionId A section identifier string of the form:
5822 * "<flag1> - <flag2> - ... - <section number>"
5823 *
5824 * Currently the only recognised flag is "T", which means the target section number
5825 * was derived during a template inclusion parse, in other words this is a template
5826 * section edit link. If no flags are given, it was an ordinary section edit link.
5827 * This flag is required to avoid a section numbering mismatch when a section is
5828 * enclosed by "<includeonly>" (bug 6563).
5829 *
5830 * The section number 0 pulls the text before the first heading; other numbers will
5831 * pull the given section along with its lower-level subsections. If the section is
5832 * not found, $mode=get will return $newtext, and $mode=replace will return $text.
5833 *
5834 * Section 0 is always considered to exist, even if it only contains the empty
5835 * string. If $text is the empty string and section 0 is replaced, $newText is
5836 * returned.
5837 *
5838 * @param string $mode One of "get" or "replace"
5839 * @param string $newText Replacement text for section data.
5840 * @return string For "get", the extracted section text.
5841 * for "replace", the whole page with the section replaced.
5842 */
5851 # Process section extraction flags
5858 }
5859 }
5861 # Check for empty input
5863 # Only sections 0 and T-0 exist in an empty document
5869 }
5875 }
5876 }
5877 }
5879 # Preprocess the text
5882 # <h> nodes indicate section breaks
5883 # They can only occur at the top level, so we can find them by iterating the root's children
5886 # Find the target section
5888 # Section zero doesn't nest, level=big
5897 }
5898 }
5901 }
5903 }
5904 }
5907 # Not found
5912 }
5913 }
5915 # Find the end of the section, including nested sections
5922 }
5923 }
5926 }
5930 # Write out the remainder (in replace mode only)
5932 # Output the replacement text
5933 # Add two newlines on -- trailing whitespace in $newText is conventionally
5934 # stripped by the editor, so we need both newlines to restore the paragraph gap
5935 # Only add trailing whitespace if there is newText
5938 }
5943 }
5944 }
5947 # Re-insert stripped tags
5949 }
5952 }
5954 /**
5955 * This function returns the text of a section, specified by a number ($section).
5956 * A section is text under a heading like == Heading == or \<h1\>Heading\</h1\>, or
5957 * the first section before any such heading (section 0).
5958 *
5959 * If a section contains subsections, these are also returned.
5960 *
5961 * @param string $text Text to look in
5962 * @param string|number $sectionId Section identifier as a number or string
5963 * (e.g. 0, 1 or 'T-1').
5964 * @param string $defaultText Default to return if section is not found
5965 *
5966 * @return string Text of the requested section
5967 */
5970 }
5972 /**
5973 * This function returns $oldtext after the content of the section
5974 * specified by $section has been replaced with $text. If the target
5975 * section does not exist, $oldtext is returned unchanged.
5976 *
5977 * @param string $oldText Former text of the article
5978 * @param string|number $sectionId Section identifier as a number or string
5979 * (e.g. 0, 1 or 'T-1').
5980 * @param string $newText Replacing text
5981 *
5982 * @return string Modified text
5983 */
5986 }
5988 /**
5989 * Get the ID of the revision we are parsing
5990 *
5991 * @return int|null
5992 */
5995 }
5997 /**
5998 * Get the revision object for $this->mRevisionId
5999 *
6000 * @return Revision|null Either a Revision object or null
6001 * @since 1.23 (public since 1.23)
6002 */
6006 }
6009 }
6013 }
6015 /**
6016 * Get the timestamp associated with the current revision, adjusted for
6017 * the default server-local timestamp
6018 * @return string
6019 */
6029 # The cryptic '' timezone parameter tells to use the site-default
6030 # timezone offset instead of the user settings.
6031 #
6032 # Since this value will be saved into the parser cache, served
6033 # to other users, and potentially even used inside links and such,
6034 # it needs to be consistent for all visitors.
6038 }
6040 }
6042 /**
6043 * Get the name of the user that edited the last revision
6044 *
6045 * @return string User name
6046 */
6051 # if this template is subst: the revision id will be blank,
6052 # so just use the current user's name
6057 }
6058 }
6060 }
6062 /**
6063 * Get the size of the revision
6064 *
6065 * @return int|null Revision size
6066 */
6071 # if this variable is subst: the revision id will be blank,
6072 # so just use the parser input size, because the own substituation
6073 # will change the size.
6078 }
6079 }
6081 }
6083 /**
6084 * Mutator for $mDefaultSort
6085 *
6086 * @param string $sort New value
6087 */
6091 }
6093 /**
6094 * Accessor for $mDefaultSort
6095 * Will use the empty string if none is set.
6096 *
6097 * This value is treated as a prefix, so the
6098 * empty string is equivalent to sorting by
6099 * page name.
6100 *
6101 * @return string
6102 */
6108 }
6109 }
6111 /**
6112 * Accessor for $mDefaultSort
6113 * Unlike getDefaultSort(), will return false if none is set
6114 *
6115 * @return string|bool
6116 */
6119 }
6121 /**
6122 * Try to guess the section anchor name based on a wikitext fragment
6123 * presumably extracted from a heading, for example "Header" from
6124 * "== Header ==".
6125 *
6126 * @param string $text
6127 *
6128 * @return string
6129 */
6131 # Strip out wikitext links(they break the anchor)
6135 }
6137 /**
6138 * Same as guessSectionNameFromWikiText(), but produces legacy anchors
6139 * instead. For use in redirects, since IE6 interprets Redirect: headers
6140 * as something other than UTF-8 (apparently?), resulting in breakage.
6141 *
6142 * @param string $text The section name
6143 * @return string An anchor
6144 */
6146 # Strip out wikitext links(they break the anchor)
6150 }
6152 /**
6153 * Strips a text string of wikitext for use in a section anchor
6154 *
6155 * Accepts a text string and then removes all wikitext from the
6156 * string and leaves only the resultant text (i.e. the result of
6157 * [[User:WikiSysop|Sysop]] would be "Sysop" and the result of
6158 * [[User:WikiSysop]] would be "User:WikiSysop") - this is intended
6159 * to create valid section anchors by mimicing the output of the
6160 * parser when headings are parsed.
6161 *
6162 * @param string $text Text string to be stripped of wikitext
6163 * for use in a Section anchor
6164 * @return string Filtered text string
6165 */
6167 # Strip internal link markup
6171 # Strip external link markup
6172 # @todo FIXME: Not tolerant to blank link text
6173 # I.E. [https://www.mediawiki.org] will render as [1] or something depending
6174 # on how many empty links there are on the page - need to figure that out.
6175 $text = preg_replace( '/\[(?i:' . $this->mUrlProtocols . ')([^ ]+?) ([^[]+)\]/', '$2', $text );
6177 # Parse wikitext quotes (italics & bold)
6180 # Strip HTML tags
6183 }
6185 /**
6186 * strip/replaceVariables/unstrip for preprocessor regression testing
6187 *
6188 * @param string $text
6189 * @param Title $title
6190 * @param ParserOptions $options
6191 * @param int $outputType
6192 *
6193 * @return string
6194 */
6195 public function testSrvus( $text, Title $title, ParserOptions $options, $outputType = self::OT_HTML ) {
6203 }
6205 /**
6206 * @param string $text
6207 * @param Title $title
6208 * @param ParserOptions $options
6209 * @return string
6210 */
6213 }
6215 /**
6216 * @param string $text
6217 * @param Title $title
6218 * @param ParserOptions $options
6219 * @return string
6220 */
6223 }
6225 /**
6226 * Call a callback function on all regions of the given text that are not
6227 * inside strip markers, and replace those regions with the return value
6228 * of the callback. For example, with input:
6229 *
6230 * aaa<MARKER>bbb
6231 *
6232 * This will call the callback function twice, with 'aaa' and 'bbb'. Those
6233 * two strings will be replaced with the value returned by the callback in
6234 * each case.
6235 *
6236 * @param string $s
6237 * @param callable $callback
6238 *
6239 * @return string
6240 */
6259 }
6260 }
6261 }
6263 }
6265 /**
6266 * Remove any strip markers found in the given text.
6267 *
6268 * @param string $text Input string
6269 * @return string
6270 */
6273 }
6275 /**
6276 * Save the parser state required to convert the given half-parsed text to
6277 * HTML. "Half-parsed" in this context means the output of
6278 * recursiveTagParse() or internalParse(). This output has strip markers
6279 * from replaceVariables (extensionSubstitution() etc.), and link
6280 * placeholders from replaceLinkHolders().
6281 *
6282 * Returns an array which can be serialized and stored persistently. This
6283 * array can later be loaded into another parser instance with
6284 * unserializeHalfParsedText(). The text can then be safely incorporated into
6285 * the return value of a parser hook.
6286 *
6287 * @param string $text
6288 *
6289 * @return array
6290 */
6298 );
6301 }
6303 /**
6304 * Load the parser state given in the $data array, which is assumed to
6305 * have been generated by serializeHalfParsedText(). The text contents is
6306 * extracted from the array, and its markers are transformed into markers
6307 * appropriate for the current Parser instance. This transformed text is
6308 * returned, and can be safely included in the return value of a parser
6309 * hook.
6310 *
6311 * If the $data array has been stored persistently, the caller should first
6312 * check whether it is still valid, by calling isValidHalfParsedText().
6313 *
6314 * @param array $data Serialized data
6315 * @throws MWException
6316 * @return string
6317 */
6321 }
6323 # First, extract the strip state.
6327 # Now renumber links
6330 # Should be good to go.
6332 }
6334 /**
6335 * Returns true if the given array, presumed to be generated by
6336 * serializeHalfParsedText(), is compatible with the current version of the
6337 * parser.
6338 *
6339 * @param array $data
6340 *
6341 * @return bool
6342 */
6345 }
6347 /**
6348 * Parsed a width param of imagelink like 300px or 200x300px
6349 *
6350 * @param string $value
6351 *
6352 * @return array
6353 * @since 1.20
6354 */
6359 }
6361 # (bug 13500) In both cases (width/height and width only),
6362 # permit trailing "px" for backward compatibility.
6371 }
6373 }
6375 /**
6376 * Lock the current instance of the parser.
6377 *
6378 * This is meant to stop someone from calling the parser
6379 * recursively and messing up all the strip state.
6380 *
6381 * @throws MWException If parser is in a parse
6382 * @return ScopedCallback The lock will be released once the return value goes out of scope.
6383 */
6388 }
6394 } );
6397 }
6399 /**
6400 * Strip outer <p></p> tag from the HTML source of a single paragraph.
6401 *
6402 * Returns original HTML if the <p/> tag has any attributes, if there's no wrapping <p/> tag,
6403 * or if there is more than one <p/> tag in the input HTML.
6404 *
6405 * @param string $html
6406 * @return string
6407 * @since 1.24
6408 */
6414 }
6415 }
6418 }
6420 /**
6421 * Return this parser if it is not doing anything, otherwise
6422 * get a fresh parser. You can use this method by doing
6423 * $myParser = $wgParser->getFreshParser(), or more simply
6424 * $wgParser->getFreshParser()->parse( ... );
6425 * if you're unsure if $wgParser is safe to use.
6426 *
6427 * @since 1.24
6428 * @return Parser A parser object that is not parsing anything
6429 */
6436 }
6437 }
6438 }