| blob | 702a4796355fea5f47b53bdb1aa21def85b70d72 |
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 */
27 /**
28 * @defgroup Parser Parser
29 */
31 /**
32 * PHP Parser - Processes wiki markup (which uses a more user-friendly
33 * syntax, such as "[[link]]" for making links), and provides a one-way
34 * transformation of that wiki markup it into (X)HTML output / markup
35 * (which in turn the browser understands, and can display).
36 *
37 * There are seven main entry points into the Parser class:
38 *
39 * - Parser::parse()
40 * produces HTML output
41 * - Parser::preSaveTransform()
42 * produces altered wiki markup
43 * - Parser::preprocess()
44 * removes HTML comments and expands templates
45 * - Parser::cleanSig() and Parser::cleanSigInSig()
46 * cleans a signature before saving it to preferences
47 * - Parser::getSection()
48 * return the content of a section from an article for section editing
49 * - Parser::replaceSection()
50 * replaces a section by number inside an article
51 * - Parser::getPreloadText()
52 * removes <noinclude> sections and <includeonly> tags
53 *
54 * Globals used:
55 * object: $wgContLang
56 *
57 * @warning $wgUser or $wgTitle or $wgRequest or $wgLang. Keep them away!
58 *
59 * @par Settings:
60 * $wgNamespacesWithSubpages
61 *
62 * @par Settings only within ParserOptions:
63 * $wgAllowExternalImages
64 * $wgAllowSpecialInclusion
65 * $wgInterwikiMagic
66 * $wgMaxArticleSize
67 *
68 * @ingroup Parser
69 */
71 /**
72 * Update this version number when the ParserOutput format
73 * changes in an incompatible way, so the parser cache
74 * can automatically discard old data.
75 */
78 /**
79 * Update this version number when the output of serialiseHalfParsedText()
80 * changes in an incompatible way
81 */
84 # Flags for Parser::setFunctionHook
88 # Constants needed for external link processing
89 # Everything except bracket, space, or control characters
90 # \p{Zs} is unicode 'separator, space' category. It covers the space 0x20
91 # as well as U+3000 is IDEOGRAPHIC SPACE for T21052
93 # Simplified expression to match an IPv4 or IPv6 address, or
94 # at least one character of a host name (embeds EXT_LINK_URL_CLASS)
96 # RegExp to make image URLs (embeds IPv6 part of EXT_LINK_ADDR)
97 // @codingStandardsIgnoreStart Generic.Files.LineLength
98 const EXT_IMAGE_REGEX = '/^(http:\/\/|https:\/\/)((?:\\[(?i:[0-9a-f:.]+)\\])?[^][<>"\\x00-\\x20\\x7F\p{Zs}]+)
100 // @codingStandardsIgnoreEnd
102 # Regular expression for a non-newline space
105 # Flags for preprocessToDom
108 # Allowed values for $this->mOutputType
109 # Parameter to startExternalParse().
116 /**
117 * @var string Prefix and suffix for temporary replacement strings
118 * for the multipass parser.
119 *
120 * \x7f should never appear in input as it's disallowed in XML.
121 * Using it at the front also gives us a little extra robustness
122 * since it shouldn't match when butted up against identifier-like
123 * string constructs.
124 *
125 * Must not consist of all title characters, or else it will change
126 * the behavior of <nowiki> in a link.
127 *
128 * Must have a character that needs escaping in attributes, otherwise
129 * someone could put a strip marker in an attribute, to get around
130 * escaping quote marks, and break out of the attribute. Thus we add
131 * `'".
132 */
136 # Markers used for wrapping the table of contents
140 # Persistent:
154 # Initialised by initialiseVariables()
156 /**
157 * @var MagicWordArray
158 */
161 /**
162 * @var MagicWordArray
163 */
165 # Initialised in constructor
168 # Initialized in getPreprocessor()
169 /** @var Preprocessor */
172 # Cleared with clearState():
173 /**
174 * @var ParserOutput
175 */
179 /**
180 * @var StripState
181 */
185 /**
186 * @var LinkHolderArray
187 */
197 /**
198 * @var User
199 */
202 # Temporary
203 # These are variables reset at least once per parse regardless of $clearState
205 /**
206 * @var ParserOptions
207 */
210 /**
211 * @var Title
212 */
224 /**
225 * @var string Deprecated accessor for the strip marker prefix.
226 * @deprecated since 1.26; use Parser::MARKER_PREFIX instead.
227 */
230 /**
231 * @var array Array with the language name of each language link (i.e. the
232 * interwiki prefix) in the key, value arbitrary. Used to avoid sending
233 * duplicate language links to the ParserOutput.
234 */
237 /**
238 * @var MapCacheLRU|null
239 * @since 1.24
240 *
241 * A cache of the current revisions of titles. Keys are $title->getPrefixedDbKey()
242 */
245 /**
246 * @var bool Recursive call protection.
247 * This variable should be treated as if it were private.
248 */
251 /** @var SectionProfiler */
254 /**
255 * @var LinkRenderer
256 */
259 /**
260 * @param array $conf
261 */
271 # Preprocessor_Hash is much faster than Preprocessor_DOM under HipHop
274 # PECL extension that conflicts with the core DOM extension (T15770)
281 }
283 }
285 /**
286 * Reduce memory usage to reduce the impact of circular references
287 */
291 }
294 }
295 }
297 /**
298 * Allow extensions to clean up when the parser is cloned
299 */
303 // T58226: When you create a reference "to" an object field, that
304 // makes the object field itself be a reference too (until the other
305 // reference goes out of scope). When cloning, any field that's a
306 // reference is copied as a reference in the new object. Both of these
307 // are defined PHP5 behaviors, as inconvenient as it is for us when old
308 // hooks from PHP4 days are passing fields by reference.
310 // Make a non-reference copy of the field, then rebind the field to
311 // reference the new copy.
315 }
318 }
320 /**
321 * Do various kinds of initialisation on the first call of the parser
322 */
326 }
334 }
336 /**
337 * Clear Parser state
338 *
339 * @private
340 */
344 }
360 # Clear these on every parse, T6549
368 ];
377 # Fix cloning
380 }
385 }
387 /**
388 * Convert wikitext to HTML
389 * Do not call this function recursively.
390 *
391 * @param string $text Text we want to parse
392 * @param Title $title
393 * @param ParserOptions $options
394 * @param bool $linestart
395 * @param bool $clearState
396 * @param int $revid Number to pass in {{REVISIONID}}
397 * @return ParserOutput A ParserOutput
398 */
402 ) {
403 /**
404 * First pass--just handle <nowiki> sections, pass the rest off
405 * to internalParse() which does all the real work.
406 */
411 // We use U+007F DELETE to construct strip markers, so we have to make
412 // sure that this character does not occur in the input text.
415 }
423 }
436 }
439 # No more strip!
446 /**
447 * A converted title will be provided in the output object if title and
448 * content conversion are enabled, the article text does not contain
449 * a conversion-suppressing double-underscore tag, and no
450 * {{DISPLAYTITLE:...}} is present. DISPLAYTITLE takes precedence over
451 * automatic link conversion.
452 */
457 ) {
464 }
465 }
467 # Done parsing! Compute runtime adaptive expiry if set
470 # Warn if too many heavyweight parser functions were used
475 );
476 }
478 # Information on include size limits, for the benefit of users who try to skirt them
486 );
487 }
492 );
496 );
499 );
502 );
505 );
508 );
511 );
517 }
527 ) ) {
533 }
537 }
538 }
539 }
540 // Since we're not really outputting HTML, decode the entities and
541 // then re-encode the things that need hiding inside HTML comments.
545 // Sanitize for comment. Note '‐' in the replacement is U+2010,
546 // which looks much like the problematic '-'.
550 // Add on template profiling data in human/machine readable way
554 } );
560 }
566 // Add other cache related metadata
569 }
580 }
581 }
593 }
595 /**
596 * Half-parse wikitext to half-parsed HTML. This recursive parser entry point
597 * can be called from an extension tag hook.
598 *
599 * The output of this function IS NOT SAFE PARSED HTML; it is "half-parsed"
600 * instead, which means that lists and links have not been fully parsed yet,
601 * and strip markers are still present.
602 *
603 * Use recursiveTagParseFully() to fully parse wikitext to output-safe HTML.
604 *
605 * Use this function if you're a parser tag hook and you want to parse
606 * wikitext before or after applying additional transformations, and you
607 * intend to *return the result as hook output*, which will cause it to go
608 * through the rest of parsing process automatically.
609 *
610 * If $frame is not provided, then template variables (e.g., {{{1}}}) within
611 * $text are not expanded
612 *
613 * @param string $text Text extension wants to have parsed
614 * @param bool|PPFrame $frame The frame to use for expanding any template variables
615 * @return string UNSAFE half-parsed HTML
616 */
622 }
624 /**
625 * Fully parse wikitext to fully parsed HTML. This recursive parser entry
626 * point can be called from an extension tag hook.
627 *
628 * The output of this function is fully-parsed HTML that is safe for output.
629 * If you're a parser tag hook, you might want to use recursiveTagParse()
630 * instead.
631 *
632 * If $frame is not provided, then template variables (e.g., {{{1}}}) within
633 * $text are not expanded
634 *
635 * @since 1.25
636 *
637 * @param string $text Text extension wants to have parsed
638 * @param bool|PPFrame $frame The frame to use for expanding any template variables
639 * @return string Fully parsed HTML
640 */
645 }
647 /**
648 * Expand templates and variables in the text, producing valid, static wikitext.
649 * Also removes comments.
650 * Do not call this function recursively.
651 * @param string $text
652 * @param Title $title
653 * @param ParserOptions $options
654 * @param int|null $revid
655 * @param bool|PPFrame $frame
656 * @return mixed|string
657 */
660 ) {
665 }
671 }
673 /**
674 * Recursive parser entry point that can be called from an extension tag
675 * hook.
676 *
677 * @param string $text Text to be expanded
678 * @param bool|PPFrame $frame The frame to use for expanding any template variables
679 * @return string
680 * @since 1.19
681 */
686 }
688 /**
689 * Process the wikitext for the "?preload=" feature. (T7210)
690 *
691 * "<noinclude>", "<includeonly>" etc. are parsed as for template
692 * transclusion, comments, templates, arguments, tags hooks and parser
693 * functions are untouched.
694 *
695 * @param string $text
696 * @param Title $title
697 * @param ParserOptions $options
698 * @param array $params
699 * @return string
700 */
705 # Parser (re)initialisation
714 }
716 /**
717 * Get a random string
718 *
719 * @return string
720 * @deprecated since 1.26; use wfRandomString() instead.
721 */
725 }
727 /**
728 * Set the current user.
729 * Should only be used when doing pre-save transform.
730 *
731 * @param User|null $user User object or null (to reset)
732 */
735 }
737 /**
738 * Accessor for mUniqPrefix.
739 *
740 * @return string
741 * @deprecated since 1.26; use Parser::MARKER_PREFIX instead.
742 */
746 }
748 /**
749 * Set the context title
750 *
751 * @param Title $t
752 */
756 }
759 # Strip the fragment to avoid various odd effects
763 }
764 }
766 /**
767 * Accessor for the Title object
768 *
769 * @return Title
770 */
773 }
775 /**
776 * Accessor/mutator for the Title object
777 *
778 * @param Title $x Title object or null to just get the current one
779 * @return Title
780 */
783 }
785 /**
786 * Set the output type
787 *
788 * @param int $ot New value
789 */
792 # Shortcut alias
798 ];
799 }
801 /**
802 * Accessor/mutator for the output type
803 *
804 * @param int|null $x New value or null to just get the current one
805 * @return int
806 */
809 }
811 /**
812 * Get the ParserOutput object
813 *
814 * @return ParserOutput
815 */
818 }
820 /**
821 * Get the ParserOptions object
822 *
823 * @return ParserOptions
824 */
827 }
829 /**
830 * Accessor/mutator for the ParserOptions object
831 *
832 * @param ParserOptions $x New value or null to just get the current one
833 * @return ParserOptions Current ParserOptions object
834 */
837 }
839 /**
840 * @return int
841 */
844 }
846 /**
847 * @param int $id
848 */
851 }
853 /**
854 * Get a language object for use in parser functions such as {{FORMATNUM:}}
855 * @return Language
856 */
859 }
861 /**
862 * Get the target language for the content being parsed. This is usually the
863 * language that the content is in.
864 *
865 * @since 1.19
866 *
867 * @throws MWException
868 * @return Language
869 */
879 }
882 }
884 /**
885 * Get the language object for language conversion
886 * @return Language|null
887 */
890 }
892 /**
893 * Get a User object either from $this->mUser, if set, or from the
894 * ParserOptions object otherwise
895 *
896 * @return User
897 */
901 }
903 }
905 /**
906 * Get a preprocessor object
907 *
908 * @return Preprocessor
909 */
914 }
916 }
918 /**
919 * Get a LinkRenderer instance to make links with
920 *
921 * @since 1.28
922 * @return LinkRenderer
923 */
930 );
931 }
934 }
936 /**
937 * Replaces all occurrences of HTML-style comments and the given tags
938 * in the text with a random marker and returns the next text. The output
939 * parameter $matches will be an associative array filled with data in
940 * the form:
941 *
942 * @code
943 * 'UNIQ-xxxxx' => [
944 * 'element',
945 * 'tag content',
946 * [ 'param' => 'x' ],
947 * '<element param="x">tag content</element>' ]
948 * @endcode
949 *
950 * @param array $elements List of element names. Comments are always extracted.
951 * @param string $text Source text string.
952 * @param array $matches Out parameter, Array: extracted tags
953 * @param string|null $uniq_prefix
954 * @return string Stripped text
955 * @since 1.26 The uniq_prefix argument is deprecated.
956 */
957 public static function extractTagsAndParams( $elements, $text, &$matches, $uniq_prefix = null ) {
960 }
973 }
975 # comment
981 # tag
986 }
992 # Empty element tag, <tag />
1001 }
1005 # No end tag -- let it run out to the end of the text.
1011 }
1012 }
1018 }
1020 }
1022 /**
1023 * Get a list of strippable XML-like elements
1024 *
1025 * @return array
1026 */
1029 }
1031 /**
1032 * Add an item to the strip state
1033 * Returns the unique tag which must be inserted into the stripped text
1034 * The tag will be replaced with the original text in unstrip()
1035 *
1036 * @param string $text
1037 *
1038 * @return string
1039 */
1045 }
1047 /**
1048 * parse the wiki syntax used to render tables
1049 *
1050 * @private
1051 * @param string $text
1052 * @return string
1053 */
1071 }
1078 # First check if we are starting a new table
1091 # Don't do any of the following
1095 # We are ending a table
1101 }
1105 }
1109 }
1113 # Now we have a table row
1116 # Whats after the tag is now only attributes
1129 }
1133 }
1142 ) {
1143 # This might be cell elements, td, th or captions
1149 }
1151 // Implies both are valid for table headings.
1154 }
1156 # Split up multiple cells on the same line.
1157 # FIXME : This can result in improper nesting of tags processed
1158 # by earlier parser steps.
1163 # Loop through each table cell
1170 }
1175 }
1181 }
1191 }
1195 # A cell could contain both parameters and data
1198 # T2553: Note that a '|' inside an invalid link should not
1199 # be mistaken as delimiting cell parameters
1200 # Bug T153140: Neither should language converter markup.
1209 }
1213 }
1214 }
1216 }
1218 # Closing open td, tr && table
1222 }
1225 }
1228 }
1231 }
1233 # Remove trailing line-ending (b/c)
1236 }
1238 # special case: don't return empty table
1241 }
1244 }
1246 /**
1247 * Helper function for parse() that transforms wiki markup into half-parsed
1248 * HTML. Only called for $mOutputType == self::OT_HTML.
1249 *
1250 * @private
1251 *
1252 * @param string $text The text to parse
1253 * @param bool $isMain Whether this is being called from the main parse() function
1254 * @param PPFrame|bool $frame A pre-processor frame
1255 *
1256 * @return string
1257 */
1262 # Hook to suspend the parser in this state
1265 }
1267 # if $frame is provided, then use $frame for replacing any variables
1269 # use frame depth to infer how include/noinclude tags should be handled
1270 # depth=0 means this is the top-level document; otherwise it's an included document
1275 }
1279 # if $frame is not provided, then use old-style replaceVariables
1281 }
1289 [],
1291 );
1294 # Tables need to come after variable replacement for things to work
1295 # properly; putting them before other transformations should keep
1296 # exciting things like link expansions from showing up in surprising
1297 # places.
1309 # replaceInternalLinks may sometimes leave behind
1310 # absolute URLs, which have to be masked to hide them from replaceExternalLinks
1317 }
1319 /**
1320 * Helper function for parse() that transforms half-parsed HTML into fully
1321 * parsed HTML.
1322 *
1323 * @param string $text
1324 * @param bool $isMain
1325 * @param bool $linestart
1326 * @return string
1327 */
1333 }
1335 # Clean up special characters, only run once, next-to-last before doBlockLevels
1337 # French spaces, last one Guillemet-left
1338 # only if there is something before the space
1340 # french spaces, Guillemet-right
1343 ];
1350 /**
1351 * The input doesn't get language converted if
1352 * a) It's disabled
1353 * b) Content isn't converted
1354 * c) It's a conversion table
1355 * d) it is an interface message (which is in the user language)
1356 */
1359 ) {
1361 # The position of the convert() call should not be changed. it
1362 # assumes that the links are all replaced and the only thing left
1363 # is the <nowiki> mark.
1365 }
1366 }
1372 }
1382 }
1384 # attempt to sanitize at least some nesting problems
1385 # (T4702 and quite a few others)
1387 # ''Something [http://www.cool.com cool''] -->
1388 # <i>Something</i><a href="http://www.cool.com"..><i>cool></i></a>
1391 # fix up an anchor inside another anchor, only
1392 # at least for a single single nested link (T5695)
1395 # fix div inside inline elements- doBlockLevels won't wrap a line which
1396 # contains a div, so fix it up here; replace
1397 # div with escaped text
1400 # remove empty italic or bold tag pairs, some
1401 # introduced by rules above
1403 ];
1409 }
1413 }
1416 }
1418 /**
1419 * Replace special strings like "ISBN xxx" and "RFC xxx" with
1420 * magic external links.
1421 *
1422 * DML
1423 * @private
1424 *
1425 * @param string $text
1426 *
1427 * @return string
1428 */
1437 '!(?: # Start cases
1439 (<.*?>) | # m[2]: Skip stuff inside
1442 # m[4]: Post-protocol path
1448 [0-9Xx] # check digit
1452 }
1454 /**
1455 * @throws MWException
1456 * @param array $m
1457 * @return HTML|string
1458 */
1461 # Skip anchor
1464 # Skip HTML element
1467 # Free external link
1470 # RFC or PMID
1474 }
1483 }
1492 }
1495 return Linker::makeExternalLink( $url, "{$keyword} {$id}", true, $cssClass, [], $this->mTitle );
1498 ) {
1499 # ISBN
1507 ] );
1512 [
1515 ]
1516 );
1519 }
1520 }
1522 /**
1523 * Make a free external link, given a user-supplied URL
1524 *
1525 * @param string $url
1526 * @param int $numPostProto
1527 * The number of characters after the protocol.
1528 * @return string HTML
1529 * @private
1530 */
1534 # The characters '<' and '>' (which were escaped by
1535 # removeHTMLtags()) should not be included in
1536 # URLs, per RFC 2396.
1537 # Make terminate a URL as well (bug T84937)
1543 PREG_OFFSET_CAPTURE
1544 ) ) {
1547 }
1549 # Move trailing punctuation to $trail
1551 # If there is no left bracket, then consider right brackets fair game too
1554 }
1558 # Don't break a trailing HTML entity by moving the ; into $trail
1559 # This is in hot code, so use substr_compare to avoid having to
1560 # create a new string object for the comparison
1562 # more optimization: instead of running preg_match with a $
1563 # anchor, which can be slow, do the match on the reversed
1564 # string starting at the desired offset.
1565 # un-reversed regexp is: /&([a-z]+|#x[\da-f]+|#\d+)$/i
1568 }
1569 }
1573 }
1575 # Verify that we still have a real URL after trail removal, and
1576 # not just lone protocol
1579 }
1583 # Is this an external image?
1586 # Not an image, make a link
1591 # Register it in the output object...
1592 # Replace unnecessary URL escape codes with their equivalent characters
1595 }
1597 }
1599 /**
1600 * Parse headers and return html
1601 *
1602 * @private
1603 *
1604 * @param string $text
1605 *
1606 * @return string
1607 */
1612 }
1614 }
1616 /**
1617 * Replace single quotes with HTML markup
1618 * @private
1619 *
1620 * @param string $text
1621 *
1622 * @return string The altered text
1623 */
1629 }
1632 }
1634 /**
1635 * Helper function for doAllQuotes()
1636 *
1637 * @param string $text
1638 *
1639 * @return string
1640 */
1646 }
1648 // First, do some preliminary work. This may shift some apostrophes from
1649 // being mark-up to being text. It also counts the number of occurrences
1650 // of bold and italics mark-ups.
1655 // If there are ever four apostrophes, assume the first is supposed to
1656 // be text, and the remaining three constitute mark-up for bold text.
1657 // (T15227: ''''foo'''' turns into ' ''' foo ' ''')
1663 // If there are more than 5 apostrophes in a row, assume they're all
1664 // text except for the last 5.
1665 // (T15227: ''''''foo'''''' turns into ' ''''' foo ' ''''')
1669 }
1670 // Count the number of occurrences of bold and italics mark-ups.
1678 }
1679 }
1681 // If there is an odd number of both bold and italics, it is likely
1682 // that one of the bold ones was meant to be an apostrophe followed
1683 // by italics. Which one we cannot know for certain, but it is more
1684 // likely to be one that has a single-letter word before it.
1696 }
1699 // if $firstsingleletterword is set, we don't
1700 // look at the other options, so we can bail early.
1705 }
1706 }
1707 }
1708 }
1710 // If there is a single-letter word, use it!
1715 // If not, but there's a multi-letter word, use that one.
1719 // ... otherwise use the first one that has neither.
1720 // (notice that it is possible for all three to be -1 if, for example,
1721 // there is only one pentuple-apostrophe in the line)
1724 }
1725 }
1727 // Now let's actually convert our apostrophic mush to HTML!
1738 }
1757 }
1774 }
1794 }
1795 }
1796 }
1798 }
1799 // Now close all remaining tags. Notice that the order is important.
1802 }
1805 }
1808 }
1809 // There might be lonely ''''', so make sure we have a buffer
1812 }
1814 }
1816 /**
1817 * Replace external links (REL)
1818 *
1819 * Note: this is all very hackish and the order of execution matters a lot.
1820 * Make sure to run tests/parser/parserTests.php if you change this code.
1821 *
1822 * @private
1823 *
1824 * @param string $text
1825 *
1826 * @throws MWException
1827 * @return string
1828 */
1835 }
1845 # The characters '<' and '>' (which were escaped by
1846 # removeHTMLtags()) should not be included in
1847 # URLs, per RFC 2396.
1852 }
1854 # If the link text is an image URL, replace it with an <img> tag
1855 # This happened by accident in the original parser, but some people used it extensively
1859 }
1863 # Set linktype for CSS - if URL==text, link is essentially free
1866 # No link text, e.g. [http://domain.tld/some.link]
1868 # Autonumber
1873 # Have link text, e.g. [http://domain.tld/some.link text]s
1874 # Check for trail
1876 }
1882 # Use the encoded URL
1883 # This means that users can paste URLs directly into the text
1884 # Funny characters like ö aren't valid in URLs anyway
1885 # This was changed in August 2004
1889 # Register link in the output object.
1890 # Replace unnecessary URL escape codes with the referenced character
1891 # This prevents spammers from hiding links from the filters
1894 }
1897 }
1899 /**
1900 * Get the rel attribute for a particular external link.
1901 *
1902 * @since 1.21
1903 * @param string|bool $url Optional URL, to extract the domain from for rel =>
1904 * nofollow if appropriate
1905 * @param Title $title Optional Title, for wgNoFollowNsExceptions lookups
1906 * @return string|null Rel attribute for $url
1907 */
1913 ) {
1915 }
1917 }
1919 /**
1920 * Get an associative array of additional HTML attributes appropriate for a
1921 * particular external link. This currently may include rel => nofollow
1922 * (depending on configuration, namespace, and the URL's domain) and/or a
1923 * target attribute (depending on configuration).
1924 *
1925 * @param string $url URL to extract the domain from for rel =>
1926 * nofollow if appropriate
1927 * @return array Associative array of HTML attributes
1928 */
1937 // T133507. New windows can navigate parent cross-origin.
1938 // Including noreferrer due to lacking browser
1939 // support of noopener. Eventually noreferrer should be removed.
1942 }
1944 }
1945 }
1948 }
1950 /**
1951 * Replace unusual escape codes in a URL with their equivalent characters
1952 *
1953 * This generally follows the syntax defined in RFC 3986, with special
1954 * consideration for HTTP query strings.
1955 *
1956 * @param string $url
1957 * @return string
1958 */
1960 # First, make sure unsafe characters are encoded
1964 },
1965 $url
1966 );
1971 # Fragment part - 'fragment'
1977 }
1979 # Query part - 'query' minus &=+;
1985 }
1987 # Scheme and path part - 'pchar'
1988 # (we assume no userinfo or encoded colons in the host)
1993 }
2000 # Unescape it
2003 # Leave it escaped, but use uppercase for a-f
2005 }
2006 };
2008 }
2010 /**
2011 * make an image if it's allowed, either through the global
2012 * option, through the exception, or through the on-wiki whitelist
2013 *
2014 * @param string $url
2015 *
2016 * @return string
2017 */
2022 # $imagesfrom could be either a single string or an array of strings, parse out the latter
2029 }
2030 }
2035 }
2039 ) {
2041 # Image found
2043 }
2044 }
2047 ) {
2051 );
2054 # Sanitize the regex fragment, make it case-insensitive, ignore blank entries/comments
2057 }
2059 # Image matches a whitelist entry
2062 }
2063 }
2064 }
2066 }
2068 /**
2069 * Process [[ ]] wikilinks
2070 *
2071 * @param string $s
2072 *
2073 * @return string Processed text
2074 *
2075 * @private
2076 */
2080 }
2082 /**
2083 * Process [[ ]] wikilinks (RIL)
2084 * @param string $s
2085 * @throws MWException
2086 * @return LinkHolderArray
2087 *
2088 * @private
2089 */
2094 # the % is needed to support urlencoded titles as well
2097 # Match a link having the form [[namespace:link|alternate]]trail
2099 # Match cases where there is no "]]", which might still be images
2101 }
2105 # split the entire text string on occurrences of [[
2107 # get the first element (all text up to first [[), and remove the space we added
2116 # Match the end of a line for a word that's not followed by whitespace,
2117 # e.g. in the case of 'The Arab al[[Razi]]', 'al' will be matched
2121 }
2125 }
2134 }
2137 }
2141 // @codingStandardsIgnoreStart Squiz.WhiteSpace.SemicolonSpacing.Incorrect
2142 # Loop for each link
2144 // @codingStandardsIgnoreEnd
2146 # Check for excessive memory usage
2148 # Too big
2149 # Do the existence check, replace the link holders and clear the array
2152 }
2160 }
2161 # first link
2165 }
2166 }
2172 # If we get a ] at the beginning of $m[3] that means we have a link that's something like:
2173 # [[Image:Foo.jpg|[http://example.com desc]]] <- having three ] in a row fucks up,
2174 # the real problem is with the $e1 regex
2175 # See T1500.
2176 # Still some problems for cases where the ] is meant to be outside punctuation,
2177 # and no image is in sight. See T4095.
2181 ) {
2184 }
2185 # fix up urlencoded title texts
2187 # Should anchors '#' also be rejected?
2189 }
2192 # Invalid, but might be an image with a link in its caption
2197 }
2202 }
2206 # Don't allow internal links to pages containing
2207 # PROTO: where PROTO is a valid URL protocol; these
2208 # should be external links.
2212 }
2214 # Make subpage if necessary
2219 }
2223 # Strip off leading ':'
2225 }
2232 }
2241 # look at the next 'line' to see if we can close it there
2246 }
2249 # the first ]] closes the inner link, the second the image
2255 # if there's exactly one ]] that's fine, we'll keep looking
2258 # if $next_line is invalid too, we need look no further
2261 }
2262 }
2264 # we couldn't find the end of this imageLink, so output it raw
2265 # but don't ignore what might be perfectly normal links in the text we've examined
2268 # note: no $trail, because without an end, there *is* no trail
2270 }
2273 # note: no $trail, because without an end, there *is* no trail
2275 }
2276 }
2282 # T6598 madness. Handle the quotes only if they come from the alternate part
2283 # [[Lista d''e paise d''o munno]] -> <a href="...">Lista d''e paise d''o munno</a>
2284 # [[Criticism of Harry Potter|Criticism of ''Harry Potter'']]
2285 # -> <a href="Criticism of Harry Potter">Criticism of <i>Harry Potter</i></a>
2287 }
2289 # Link not escaped by : , create the various objects
2291 # Interwikis
2296 )
2297 ) {
2298 # T26502: filter duplicates
2302 }
2307 }
2312 # if no parameters were passed, $text
2313 # becomes something like "File:Foo.png",
2314 # which we don't want to pass on to the
2315 # image generator
2318 # recursively parse links inside the image caption
2319 # actually, this will parse them in any other parameters, too,
2320 # but it might be hard to fix that, and it doesn't matter ATM
2323 }
2324 # cloak any absolute URLs inside the image markup, so replaceExternalLinks() won't touch them
2328 }
2336 }
2342 /**
2343 * Strip the whitespace Category links produce, see T2087
2344 */
2348 }
2349 }
2351 # Self-link checking. For some languages, variants of the title are checked in
2352 # LinkHolderArray::doVariants() to allow batching the existence checks necessary
2353 # for linking to a different variant.
2357 }
2359 # NS_MEDIA is a pseudo-namespace for linking directly to a file
2360 # @todo FIXME: Should do batch file existence checks, see comment below
2362 # Give extensions a chance to select the file revision for us
2367 # Fetch and register the file (file title may be different via hooks)
2369 # Cloak with NOPARSE to avoid replacement in replaceExternalLinks
2373 }
2375 # Some titles, such as valid special pages or files in foreign repos, should
2376 # be shown as bluelinks even though they're not included in the page table
2377 # @todo FIXME: isAlwaysKnown() can be expensive for file links; we should really do
2378 # batch file existence checks for NS_FILE and NS_MEDIA
2383 # Links will be added to the output link list after checking
2385 }
2386 }
2388 }
2390 /**
2391 * Render a forced-blue link inline; protect against double expansion of
2392 * URLs if we're in a mode that prepends full URL prefixes to internal links.
2393 * Since this little disaster has to split off the trail text to avoid
2394 * breaking URLs in the following text without breaking trails on the
2395 * wiki links, it's been made into a horrible function.
2396 *
2397 * @param Title $nt
2398 * @param string $text
2399 * @param string $trail
2400 * @param string $prefix
2401 * @return string HTML-wikitext mix oh yuck
2402 */
2408 }
2412 );
2415 }
2417 /**
2418 * Insert a NOPARSE hacky thing into any inline links in a chunk that's
2419 * going to go through further parsing steps before inline URL expansion.
2420 *
2421 * Not needed quite as much as it used to be since free links are a bit
2422 * more sensible these days. But bracketed links are still an issue.
2423 *
2424 * @param string $text More-or-less HTML
2425 * @return string Less-or-more HTML with NOPARSE bits
2426 */
2430 }
2432 /**
2433 * Return true if subpage links should be expanded on this page.
2434 * @return bool
2435 */
2437 # Some namespaces don't allow subpages
2439 }
2441 /**
2442 * Handle link to subpage if necessary
2443 *
2444 * @param string $target The source of the link
2445 * @param string &$text The link text, modified as necessary
2446 * @return string The full name of the link
2447 * @private
2448 */
2451 }
2453 /**
2454 * Make lists from lines starting with ':', '*', '#', etc. (DBL)
2455 *
2456 * @param string $text
2457 * @param bool $linestart Whether or not this is at the start of a line.
2458 * @private
2459 * @return string The lists rendered as HTML
2460 */
2463 }
2465 /**
2466 * Return value of a magic variable (like PAGENAME)
2467 *
2468 * @private
2469 *
2470 * @param int $index
2471 * @param bool|PPFrame $frame
2472 *
2473 * @throws MWException
2474 * @return string
2475 */
2481 // If no title set, bad things are going to happen
2482 // later. Title should always be set since this
2483 // should only be called in the middle of a parse
2484 // operation (but the unit-tests do funky stuff)
2487 }
2489 /**
2490 * Some of these require message or data lookups and can be
2491 * expensive to check many times.
2492 */
2496 }
2497 }
2542 $value = $pageLang->getMonthAbbreviation( MWTimestamp::getLocalInstance( $ts )->format( 'n' ) );
2576 ) ) );
2586 ) ) );
2594 }
2602 }
2615 # 0 means the page doesn't exist in the database,
2616 # which means the user is previewing a new page.
2617 # The vary-revision flag must be set, because the magic word
2618 # will have a different value once the page is saved.
2621 }
2625 # Let the edit saving system know we should parse the page
2626 # *after* a revision ID has been assigned.
2633 }
2636 # Let the edit saving system know we should parse the page
2637 # *after* a revision ID has been assigned. This is for null edits.
2643 # Let the edit saving system know we should parse the page
2644 # *after* a revision ID has been assigned. This is for null edits.
2650 # Let the edit saving system know we should parse the page
2651 # *after* a revision ID has been assigned. This is for null edits.
2657 # Let the edit saving system know we should parse the page
2658 # *after* a revision ID has been assigned. This is for null edits.
2664 # Let the edit saving system know we should parse the page
2665 # *after* a revision ID has been assigned. This is for null edits.
2671 # Let the edit saving system know we should parse the page
2672 # *after* a revision ID has been assigned. This is for null edits.
2678 # Let the edit saving system know we should parse the page
2679 # *after* a revision ID has been assigned for null edits.
2723 # @bug T6594 PHP5 has it zero padded, PHP4 does not, cast to
2724 # int to remove the padding
2733 );
2742 false
2743 );
2749 # @bug T6594 PHP5 has it zero padded, PHP4 does not, cast to
2750 # int to remove the padding
2814 );
2817 }
2821 }
2824 }
2826 /**
2827 * initialise the magic variables (like CURRENTMONTHNAME) and substitution modifiers
2828 *
2829 * @private
2830 */
2837 }
2839 /**
2840 * Preprocess some wikitext and return the document tree.
2841 * This is the ghost of replace_variables().
2842 *
2843 * @param string $text The text to parse
2844 * @param int $flags Bitwise combination of:
2845 * - self::PTD_FOR_INCLUSION: Handle "<noinclude>" and "<includeonly>" as if the text is being
2846 * included. Default is to assume a direct page view.
2847 *
2848 * The generated DOM tree must depend only on the input text and the flags.
2849 * The DOM tree must be the same in OT_HTML and OT_WIKI mode, to avoid a regression of T6899.
2850 *
2851 * Any flag added to the $flags parameter here, or any other parameter liable to cause a
2852 * change in the DOM tree for a given text, must be passed through the section identifier
2853 * in the section edit link and thus back to extractSections().
2854 *
2855 * The output of this function is currently only cached in process memory, but a persistent
2856 * cache may be implemented at a later date which takes further advantage of these strict
2857 * dependency requirements.
2858 *
2859 * @return PPNode
2860 */
2864 }
2866 /**
2867 * Return a three-element array: leading whitespace, string contents, trailing whitespace
2868 *
2869 * @param string $s
2870 *
2871 * @return array
2872 */
2882 }
2884 }
2886 /**
2887 * Replace magic variables, templates, and template arguments
2888 * with the appropriate text. Templates are substituted recursively,
2889 * taking care to avoid infinite loops.
2890 *
2891 * Note that the substitution depends on value of $mOutputType:
2892 * self::OT_WIKI: only {{subst:}} templates
2893 * self::OT_PREPROCESS: templates but not extension tags
2894 * self::OT_HTML: all templates and extension tags
2895 *
2896 * @param string $text The text to transform
2897 * @param bool|PPFrame $frame Object describing the arguments passed to the
2898 * template. Arguments may also be provided as an associative array, as
2899 * was the usual case before MW1.12. Providing arguments this way may be
2900 * useful for extensions wishing to perform variable replacement
2901 * explicitly.
2902 * @param bool $argsOnly Only do argument (triple-brace) expansion, not
2903 * double-brace expansion.
2904 * @return string
2905 */
2907 # Is there any text? Also, Prevent too big inclusions!
2911 }
2919 }
2926 }
2928 /**
2929 * Clean up argument array - refactored in 1.9 so parserfunctions can use it, too.
2930 *
2931 * @param array $args
2932 *
2933 * @return array
2934 */
2947 }
2950 }
2951 }
2952 }
2955 }
2957 /**
2958 * Warn the user when a parser limitation is reached
2959 * Will warn at most once the user per limitation type
2960 *
2961 * The results are shown during preview and run through the Parser (See EditPage.php)
2962 *
2963 * @param string $limitationType Should be one of:
2964 * 'expensive-parserfunction' (corresponding messages:
2965 * 'expensive-parserfunction-warning',
2966 * 'expensive-parserfunction-category')
2967 * 'post-expand-template-argument' (corresponding messages:
2968 * 'post-expand-template-argument-warning',
2969 * 'post-expand-template-argument-category')
2970 * 'post-expand-template-inclusion' (corresponding messages:
2971 * 'post-expand-template-inclusion-warning',
2972 * 'post-expand-template-inclusion-category')
2973 * 'node-count-exceeded' (corresponding messages:
2974 * 'node-count-exceeded-warning',
2975 * 'node-count-exceeded-category')
2976 * 'expansion-depth-exceeded' (corresponding messages:
2977 * 'expansion-depth-exceeded-warning',
2978 * 'expansion-depth-exceeded-category')
2979 * @param string|int|null $current Current value
2980 * @param string|int|null $max Maximum allowed, when an explicit limit has been
2981 * exceeded, provide the values (optional)
2982 */
2984 # does no harm if $current and $max are present but are unnecessary for the message
2985 # Not doing ->inLanguage( $this->mOptions->getUserLangObj() ), since this is shown
2986 # only during preview, and that would split the parser cache unnecessarily.
2991 }
2993 /**
2994 * Return the text of a template, after recursively
2995 * replacing any variables or templates within the template.
2996 *
2997 * @param array $piece The parts of the template
2998 * $piece['title']: the title, i.e. the part before the |
2999 * $piece['parts']: the parameter array
3000 * $piece['lineStart']: whether the brace was at the start of a line
3001 * @param PPFrame $frame The current frame, contains template arguments
3002 * @throws Exception
3003 * @return string The text of the template
3004 */
3007 // Flags
3009 // $text has been filled
3011 // wiki markup in $text should be escaped
3013 // $text is HTML, armour it against wikitext transformation
3015 // Force interwiki transclusion to be done in raw mode not rendered
3017 // $text is a DOM node needing expansion in a child frame
3019 // $text is a DOM node needing expansion in the current frame
3022 # Title object, where $text came from
3025 # $part1 is the bit before the first |, and must contain only title characters.
3026 # Various prefixes will be stripped from it later.
3031 # Original title text preserved for various purposes
3034 # $args is a list of argument nodes, starting from index 0, not including $part1
3035 # @todo FIXME: If piece['parts'] is null then the call to getLength()
3036 # below won't work b/c this $args isn't an object
3041 # SUBST
3045 # Possibilities for substMatch: "subst", "safesubst" or FALSE
3046 # Decide whether to expand template or keep wikitext as-is.
3052 }
3058 }
3059 }
3064 }
3065 }
3067 # Variables
3074 }
3076 }
3077 }
3079 # MSG, MSGNW and RAW
3081 # Check for MSGNW:
3086 # Remove obsolete MSG:
3089 }
3091 # Check for RAW:
3095 }
3096 }
3098 # Parser functions
3107 }
3112 }
3114 # The interface for parser functions allows for extracting
3115 # flags into the local scope. Extract any forwarded flags
3116 # here.
3118 }
3119 }
3121 # Finish mangling title and then check for loops.
3122 # Set $title to a Title object and $titleText to the PDBK
3125 # Split the title into page and subpage
3131 }
3135 # Check for language variants if the template is not found
3138 }
3139 # Do recursion depth check
3147 }
3148 }
3149 }
3151 # Load from database
3158 ) {
3160 // Pass the template arguments as URL parameters.
3161 // "uselang" will have no effect since the Language object
3162 // is forced to the one defined in ParserOptions.
3171 }
3172 }
3174 // Create a new context to execute the special page
3181 // If this page is cached, then we better not be per user.
3183 }
3195 );
3196 }
3197 }
3207 }
3208 }
3210 # If the title is valid but undisplayable, make a link to it
3214 }
3216 # Interwiki transclusion
3222 # Preprocess it like a template
3225 }
3227 }
3229 # Do infinite loop check
3230 # This has to be done after redirect resolution to avoid infinite loops via redirects
3237 }
3238 }
3240 # If we haven't found text to substitute by now, we're done
3241 # Recover the source wikitext and return it
3246 }
3248 }
3250 # Expand DOM-style return values in a child frame
3252 # Clean up argument array
3258 # Expansion is eligible for the empty-frame cache
3261 # Uncached expansion
3263 }
3264 }
3268 }
3272 }
3274 # Replace raw HTML by a placeholder
3278 # Escape nowiki-style return values
3283 ) {
3284 # T2529: if the template begins with a table or block-level
3285 # element, it should be treated as beginning a new line.
3286 # This behavior is somewhat controversial.
3288 }
3291 # Error, oversize inclusion
3293 # Make a working, properly escaped link if possible (T25588)
3296 # This will probably not be a working link, but at least it may
3297 # provide some hint of where the problem is
3300 }
3304 }
3310 }
3313 }
3315 /**
3316 * Call a parser function and return an array with text and flags.
3317 *
3318 * The returned array will always contain a boolean 'found', indicating
3319 * whether the parser function was found or not. It may also contain the
3320 * following:
3321 * text: string|object, resulting wikitext or PP DOM object
3322 * isHTML: bool, $text is HTML, armour it against wikitext transformation
3323 * isChildObj: bool, $text is a DOM node needing expansion in a child frame
3324 * isLocalObj: bool, $text is a DOM node needing expansion in the current frame
3325 * nowiki: bool, wiki markup in $text should be escaped
3326 *
3327 * @since 1.21
3328 * @param PPFrame $frame The current frame, contains template arguments
3329 * @param string $function Function name
3330 * @param array $args Arguments to the function
3331 * @throws MWException
3332 * @return array
3333 */
3337 # Case sensitive functions
3341 # Case insensitive functions
3347 }
3348 }
3352 # Workaround for PHP bug 35229 and similar
3355 }
3359 # Convert arguments to PPNodes and collect for appending to $allArgs
3366 }
3367 }
3369 # Add a frame parameter, and pass the arguments as an array
3373 # Convert arguments to plain text and append to $allArgs
3381 }
3382 }
3383 }
3387 # The interface for function hooks allows them to return a wikitext
3388 # string or an array containing the string and any flags. This mungs
3389 # things around to match what this method should return.
3394 ];
3398 }
3402 ];
3403 }
3409 }
3412 }
3417 }
3420 }
3422 /**
3423 * Get the semi-parsed DOM representation of a template with a given title,
3424 * and its redirect destination title. Cached.
3425 *
3426 * @param Title $title
3427 *
3428 * @return array
3429 */
3438 }
3441 }
3443 # Cache miss, go to the database
3449 }
3457 }
3460 }
3462 /**
3463 * Fetch the current revision of a given title. Note that the revision
3464 * (and even the title) may not exist in the database, so everything
3465 * contributing to the output of the parser should use this method
3466 * where possible, rather than getting the revisions themselves. This
3467 * method also caches its results, so using it benefits performance.
3468 *
3469 * @since 1.24
3470 * @param Title $title
3471 * @return Revision
3472 */
3477 }
3480 // Defaults to Parser::statelessFetchRevision()
3482 );
3483 }
3485 }
3487 /**
3488 * Wrapper around Revision::newFromTitle to allow passing additional parameters
3489 * without passing them on to it.
3490 *
3491 * @since 1.24
3492 * @param Title $title
3493 * @param Parser|bool $parser
3494 * @return Revision|bool False if missing
3495 */
3503 }
3506 }
3508 /**
3509 * Fetch the unparsed text of a template and register a reference to it.
3510 * @param Title $title
3511 * @return array ( string or false, Title )
3512 */
3514 // Defaults to Parser::statelessFetchTemplate()
3517 // We use U+007F DELETE to distinguish strip markers from regular text.
3521 }
3527 // If we transclude ourselves, the final result
3528 // will change based on the new version of the page
3530 }
3531 }
3532 }
3534 }
3536 /**
3537 * Fetch the unparsed text of a template and register a reference to it.
3538 * @param Title $title
3539 * @return string|bool
3540 */
3543 }
3545 /**
3546 * Static function to get a template
3547 * Can be overridden via ParserOptions::setTemplateCallback().
3548 *
3549 * @param Title $title
3550 * @param bool|Parser $parser
3551 *
3552 * @return array
3553 */
3559 # Loop to fetch the article, with up to 1 redirect
3560 // @codingStandardsIgnoreStart Generic.CodeAnalysis.ForLoopWithTestFunctionCall.NotAllowed
3562 // @codingStandardsIgnoreEnd
3563 # Give extensions a chance to select the revision instead
3574 ];
3576 }
3577 # Get the revision
3584 }
3586 # If there is no current revision, there is no page
3590 }
3597 # We fetched a rev from a different title; register it too...
3602 }
3614 }
3621 }
3626 }
3629 }
3630 # Redirect?
3633 }
3638 }
3640 /**
3641 * Fetch a file and its title and register a reference to it.
3642 * If 'broken' is a key in $options then the file will appear as a broken thumbnail.
3643 * @param Title $title
3644 * @param array $options Array of options to RepoGroup::findFile
3645 * @return File|bool
3646 */
3649 }
3651 /**
3652 * Fetch a file and its title and register a reference to it.
3653 * If 'broken' is a key in $options then the file will appear as a broken thumbnail.
3654 * @param Title $title
3655 * @param array $options Array of options to RepoGroup::findFile
3656 * @return array ( File or false, Title of file )
3657 */
3663 # Register the file as a dependency...
3666 # Update fetched file title
3669 }
3671 }
3673 /**
3674 * Helper function for fetchFileAndTitle.
3675 *
3676 * Also useful if you need to fetch a file but not use it yet,
3677 * for example to get the file's handler.
3678 *
3679 * @param Title $title
3680 * @param array $options Array of options to RepoGroup::findFile
3681 * @return File|bool
3682 */
3690 }
3692 }
3694 /**
3695 * Transclude an interwiki link.
3696 *
3697 * @param Title $title
3698 * @param string $action
3699 *
3700 * @return string
3701 */
3707 }
3713 }
3715 }
3717 /**
3718 * @param string $url
3719 * @return mixed|string
3720 */
3729 }
3736 // Though we failed to fetch the content, this status is useless.
3741 }
3748 ] );
3750 }
3752 /**
3753 * Triple brace replacement -- used for template arguments
3754 * @private
3755 *
3756 * @param array $piece
3757 * @param PPFrame $frame
3758 *
3759 * @return array
3760 */
3773 )
3774 ) {
3775 # No match in frame, use the supplied default
3777 }
3781 }
3784 # No match anywhere
3786 }
3789 }
3794 }
3797 }
3799 /**
3800 * Return the text to be used for a given extension tag.
3801 * This is the ghost of strip().
3802 *
3803 * @param array $params Associative array of parameters:
3804 * name PPNode for the tag name
3805 * attr PPNode for unparsed text where tag attributes are thought to be
3806 * attributes Optional associative array of parsed attributes
3807 * inner Contents of extension element
3808 * noClose Original text did not have a close tag
3809 * @param PPFrame $frame
3810 *
3811 * @throws MWException
3812 * @return string
3813 */
3820 // Probably expansion depth or node count exceeded. Just punt the
3821 // error up.
3823 }
3827 // See above
3829 }
3831 // We can't safely check if the expansion for $content resulted in an
3832 // error, because the content could happen to be the error string
3833 // (T149622).
3845 }
3851 }
3854 # Workaround for PHP bug 35229 and similar
3857 }
3864 }
3870 }
3873 # Extract flags to local scope (to override $markerType)
3878 }
3882 }
3887 }
3888 }
3894 // See above
3896 }
3898 }
3899 }
3909 }
3911 }
3913 /**
3914 * Increment an include size counter
3915 *
3916 * @param string $type The type of expansion
3917 * @param int $size The size of the text
3918 * @return bool False if this inclusion would take it over the maximum, true otherwise
3919 */
3926 }
3927 }
3929 /**
3930 * Increment the expensive function count
3931 *
3932 * @return bool False if the limit has been exceeded
3933 */
3937 }
3939 /**
3940 * Strip double-underscore items like __NOGALLERY__ and __NOTOC__
3941 * Fills $this->mDoubleUnderscores, returns the modified text
3942 *
3943 * @param string $text
3944 *
3945 * @return string
3946 */
3949 # The position of __TOC__ needs to be recorded
3955 # Set a placeholder. At the end we'll fill it in with the TOC.
3958 # Only keep the first one.
3960 }
3962 # Now match and remove the rest of them
3968 }
3971 }
3974 ) {
3976 }
3977 # (T10068) Allow control over whether robots index a page.
3978 # __INDEX__ always overrides __NOINDEX__, see T16899
3982 }
3986 }
3988 # Cache all double underscores in the database
3991 }
3994 }
3996 /**
3997 * @see ParserOutput::addTrackingCategory()
3998 * @param string $msg Message key
3999 * @return bool Whether the addition was successful
4000 */
4003 }
4005 /**
4006 * This function accomplishes several tasks:
4007 * 1) Auto-number headings if that option is enabled
4008 * 2) Add an [edit] link to sections for users who have enabled the option and can edit the page
4009 * 3) Add a Table of contents on the top for users who have enabled the option
4010 * 4) Auto-anchor headings
4011 *
4012 * It loops through all headlines, collects the necessary data, then splits up the
4013 * string and re-inserts the newly formatted headlines.
4014 *
4015 * @param string $text
4016 * @param string $origText Original, untouched wikitext
4017 * @param bool $isMain
4018 * @return mixed|string
4019 * @private
4020 */
4024 # Inhibit editsection links if requested in the page
4030 }
4033 }
4035 # Get all headlines for numbering them and adding funky stuff like [edit]
4036 # links - this is for later, but we need the number of headlines right now
4041 $matches
4042 );
4044 # if there are fewer than 4 headlines in the article, do not show TOC
4045 # unless it's been explicitly enabled.
4049 # Allow user to stipulate that a page should have a "new section"
4050 # link added via __NEWSECTIONLINK__
4053 }
4055 # Allow user to remove the "new section"
4056 # link via __NONEWSECTIONLINK__
4059 }
4061 # if the string __FORCETOC__ (not case-sensitive) occurs in the HTML,
4062 # override above conditions and always show TOC above first header
4066 }
4068 # headline counter
4072 # Ugh .. the TOC should have neat indentation levels which can be
4073 # passed to the skin functions. These are determined here
4107 }
4111 }
4115 # Increase TOC level
4122 }
4124 # Decrease TOC level, find level to jump to
4128 # Found last matching level
4132 # Found first matching level below current level
4135 }
4136 }
4139 }
4142 # Unindent only if the previous toc level was shown :p
4147 }
4148 }
4150 # No change in level, end TOC line
4153 }
4154 }
4158 # count number of headlines for each level
4165 }
4168 }
4169 }
4171 # The safe header is a version of the header text safe to use for links
4173 # Remove link placeholders by the link text.
4174 # <!--LINK number-->
4175 # turns into
4176 # link text with suffix
4177 # Do this before unstrip since link text can contain strip markers
4180 # Avoid insertion of weird stuff like <math> by expanding the relevant sections
4183 # Strip out HTML (first regex removes any tag not allowed)
4184 # Allowed tags are:
4185 # * <sup> and <sub> (T10393)
4186 # * <i> (T28375)
4187 # * <b> (r105284)
4188 # * <bdi> (T74884)
4189 # * <span dir="rtl"> and <span dir="ltr"> (T37167)
4190 # * <s> and <strike> (T35715)
4191 # We strip any parameter from accepted tags (second regex), except dir="rtl|ltr" from <span>,
4192 # to allow setting directionality in toc items.
4194 [
4196 '#<(/?(?:span(?: dir="(?:rtl|ltr)")?|sup|sub|bdi|i|b|s|strike))(?: .*?)?>#'
4197 ],
4199 $safeHeadline
4200 );
4202 # Strip '<span></span>', which is the result from the above if
4203 # <span id="foo"></span> is used to produce an additional anchor
4204 # for a section.
4209 # For the anchor, strip out HTML-y stuff period
4213 # Save headline for section edit hint before it's escaped
4217 # For reverse compatibility, provide an id that's
4218 # HTML4-compatible, like we used to.
4219 # It may be worth noting, academically, that it's possible for
4220 # the legacy anchor to conflict with a non-legacy headline
4221 # anchor on the page. In this case likely the "correct" thing
4222 # would be to either drop the legacy anchors or make sure
4223 # they're numbered first. However, this would require people
4224 # to type in section names like "abc_.D7.93.D7.90.D7.A4"
4225 # manually, so let's not bother worrying about it.
4231 # No reason to have both (in fact, we can't)
4233 }
4238 }
4240 # HTML names must be case-insensitively unique (T12721).
4241 # This does not apply to Unicode characters per
4242 # https://www.w3.org/TR/html5/infrastructure.html#case-sensitivity-and-string-comparison
4243 # @todo FIXME: We may be changing them depending on the current locale.
4249 }
4251 # Create the anchor for linking from the TOC to the section
4255 // @codingStandardsIgnoreStart
4257 // @codingStandardsIgnoreEnd
4262 }
4264 // @codingStandardsIgnoreStart
4266 // @codingStandardsIgnoreEnd
4271 }
4273 # Don't number the heading if it is the only one (looks silly)
4275 # the two are different if the line contains a link
4279 $numbering
4281 }
4286 }
4288 # Add the section to the section tree
4289 # Find the DOM node for this header
4296 }
4297 }
4301 }
4311 ];
4313 # give headline the correct <h#> tag
4315 // Output edit section links as markers with styles that can be customized by skins
4317 # Put a T flag in the section identifier, to indicate to extractSections()
4318 # that sections inside <includeonly> should be counted.
4326 }
4327 // We use a bit of pesudo-xml for editsection markers. The
4328 // language converter is run later on. Using a UNIQ style marker
4329 // leads to the converter screwing up the tokens when it
4330 // converts stuff. And trying to insert strip tags fails too. At
4331 // this point all real inputted tags have already been escaped,
4332 // so we don't have to worry about a user trying to input one of
4333 // these markers directly. We use a page and section attribute
4334 // to stop the language converter from converting these
4335 // important bits of data, but put the headline hint inside a
4336 // content block because the language converter is supposed to
4337 // be able to convert that piece of data.
4338 // Gets replaced with html in ParserOutput::getText
4345 }
4348 }
4354 }
4358 # Never ever show TOC if no headers
4361 }
4366 }
4371 }
4375 }
4377 # split up and insert constructed headlines
4381 // build an array of document sections
4384 // $head is zero-based, sections aren't.
4389 }
4391 /**
4392 * Send a hook, one per section.
4393 * The idea here is to be able to make section-level DIVs, but to do so in a
4394 * lower-impact, more correct way than r50769
4395 *
4396 * $this : caller
4397 * $section : the section number
4398 * &$sectionContent : ref to the content of the section
4399 * $showEditLinks : boolean describing whether this section has an edit link
4400 */
4404 }
4407 // append the TOC at the beginning
4408 // Top anchor now in skin
4410 }
4418 }
4419 }
4421 /**
4422 * Transform wiki markup when saving a page by doing "\r\n" -> "\n"
4423 * conversion, substituting signatures, {{subst:}} templates, etc.
4424 *
4425 * @param string $text The text to transform
4426 * @param Title $title The Title object for the current article
4427 * @param User $user The User object describing the current user
4428 * @param ParserOptions $options Parsing options
4429 * @param bool $clearState Whether to clear the parser state first
4430 * @return string The altered wiki markup
4431 */
4434 ) {
4437 }
4441 // We still normalize line endings for backwards-compatibility
4442 // with other code that just calls PST, but this should already
4443 // be handled in TextContent subclasses
4448 }
4454 }
4456 /**
4457 * Pre-save transform helper function
4458 *
4459 * @param string $text
4460 * @param User $user
4461 *
4462 * @return string
4463 */
4467 # Note: This is the timestamp saved as hardcoded wikitext to
4468 # the database, we use $wgContLang here in order to give
4469 # everyone the same signature and use the default one rather
4470 # than the one selected in each user's preferences.
4471 # (see also T14815)
4479 # Variable replacement
4480 # Because mOutputType is OT_WIKI, this will only process {{subst:xxx}} type tags
4483 # This works almost by chance, as the replaceVariables are done before the getUserSig(),
4484 # which may corrupt this parser instance via its wfMessage()->text() call-
4486 # Signatures
4492 ] );
4494 # Context links ("pipe tricks"): [[|name]] and [[name (context)|]]
4498 // [[ns:page (context)|]]
4500 // [[ns:page（context）|]] (double-width brackets, added in r40257)
4502 // [[ns:page (context), context|]] (using either single or double-width comma)
4504 // [[|page]] (reverse pipe trick: add context from page title)
4507 # try $p1 first, to turn "[[A, B (C)|]]" into "[[A, B (C)|A, B]]"
4519 # if there's no context, don't bother duplicating the title
4521 }
4524 }
4526 /**
4527 * Fetch the user's signature text, if any, and normalize to
4528 * validated, ready-to-insert wikitext.
4529 * If you have pre-fetched the nickname or the fancySig option, you can
4530 * specify them here to save a database query.
4531 * Do not reuse this parser instance after calling getUserSig(),
4532 * as it may have changed if it's the $wgParser.
4533 *
4534 * @param User $user
4535 * @param string|bool $nickname Nickname to use or false to use user's default nickname
4536 * @param bool|null $fancySig whether the nicknname is the complete signature
4537 * or null to use default value
4538 * @return string
4539 */
4545 # If not given, retrieve from the user object.
4548 }
4552 }
4560 # Sig. might contain markup; validate this
4562 # Validated; clean up (if needed) and return it
4565 # Failed to validate; fall back to the default
4568 }
4569 }
4571 # Make sure nickname doesnt get a sig in a sig
4574 # If we're still here, make it a link to the user page
4581 }
4583 /**
4584 * Check that the user's signature contains no bad XML
4585 *
4586 * @param string $text
4587 * @return string|bool An expanded string, or false if invalid.
4588 */
4591 }
4593 /**
4594 * Clean up signature text
4595 *
4596 * 1) Strip 3, 4 or 5 tildes out of signatures @see cleanSigInSig
4597 * 2) Substitute all transclusions
4598 *
4599 * @param string $text
4600 * @param bool $parsing Whether we're cleaning (preferences save) or parsing
4601 * @return string Signature text
4602 */
4608 }
4610 # Option to disable this feature
4613 }
4615 # @todo FIXME: Regex doesn't respect extension tags or nowiki
4616 # => Move this logic to braceSubstitution()
4618 $substRegex = '/\{\{(?!(?:' . $substWord->getBaseRegex() . '))/x' . $substWord->getRegexCase();
4629 }
4632 }
4634 /**
4635 * Strip 3, 4 or 5 tildes out of signatures.
4636 *
4637 * @param string $text
4638 * @return string Signature text with /~{3,5}/ removed
4639 */
4643 }
4645 /**
4646 * Set up some variables which are usually set up in parse()
4647 * so that an external function can call some class members with confidence
4648 *
4649 * @param Title|null $title
4650 * @param ParserOptions $options
4651 * @param int $outputType
4652 * @param bool $clearState
4653 */
4656 ) {
4658 }
4660 /**
4661 * @param Title|null $title
4662 * @param ParserOptions $options
4663 * @param int $outputType
4664 * @param bool $clearState
4665 */
4668 ) {
4674 }
4675 }
4677 /**
4678 * Wrapper for preprocess()
4679 *
4680 * @param string $text The text to preprocess
4681 * @param ParserOptions $options Options
4682 * @param Title|null $title Title object or null to use $wgTitle
4683 * @return string
4684 */
4688 # Guard against infinite recursion
4691 }
4697 }
4703 }
4705 /**
4706 * Create an HTML-style tag, e.g. "<yourtag>special text</yourtag>"
4707 * The callback should have the following form:
4708 * function myParserHook( $text, $params, $parser, $frame ) { ... }
4709 *
4710 * Transform and return $text. Use $parser for any required context, e.g. use
4711 * $parser->getTitle() and $parser->getOptions() not $wgTitle or $wgOut->mParserOptions
4712 *
4713 * Hooks may return extended information by returning an array, of which the
4714 * first numbered element (index 0) must be the return string, and all other
4715 * entries are extracted into local variables within an internal function
4716 * in the Parser class.
4717 *
4718 * This interface (introduced r61913) appears to be undocumented, but
4719 * 'markerType' is used by some core tag hooks to override which strip
4720 * array their results are placed in. **Use great caution if attempting
4721 * this interface, as it is not documented and injudicious use could smash
4722 * private variables.**
4723 *
4724 * @param string $tag The tag to use, e.g. 'hook' for "<hook>"
4725 * @param callable $callback The callback function (and object) to use for the tag
4726 * @throws MWException
4727 * @return callable|null The old value of the mTagHooks array associated with the hook
4728 */
4733 }
4738 }
4741 }
4743 /**
4744 * As setHook(), but letting the contents be parsed.
4745 *
4746 * Transparent tag hooks are like regular XML-style tag hooks, except they
4747 * operate late in the transformation sequence, on HTML instead of wikitext.
4748 *
4749 * This is probably obsoleted by things dealing with parser frames?
4750 * The only extension currently using it is geoserver.
4751 *
4752 * @since 1.10
4753 * @todo better document or deprecate this
4754 *
4755 * @param string $tag The tag to use, e.g. 'hook' for "<hook>"
4756 * @param callable $callback The callback function (and object) to use for the tag
4757 * @throws MWException
4758 * @return callable|null The old value of the mTagHooks array associated with the hook
4759 */
4764 }
4765 $oldVal = isset( $this->mTransparentTagHooks[$tag] ) ? $this->mTransparentTagHooks[$tag] : null;
4769 }
4771 /**
4772 * Remove all tag hooks
4773 */
4778 }
4780 /**
4781 * Create a function, e.g. {{sum:1|2|3}}
4782 * The callback function should have the form:
4783 * function myParserFunction( &$parser, $arg1, $arg2, $arg3 ) { ... }
4784 *
4785 * Or with Parser::SFH_OBJECT_ARGS:
4786 * function myParserFunction( $parser, $frame, $args ) { ... }
4787 *
4788 * The callback may either return the text result of the function, or an array with the text
4789 * in element 0, and a number of flags in the other elements. The names of the flags are
4790 * specified in the keys. Valid flags are:
4791 * found The text returned is valid, stop processing the template. This
4792 * is on by default.
4793 * nowiki Wiki markup in the return value should be escaped
4794 * isHTML The returned text is HTML, armour it against wikitext transformation
4795 *
4796 * @param string $id The magic word ID
4797 * @param callable $callback The callback function (and object) to use
4798 * @param int $flags A combination of the following flags:
4799 * Parser::SFH_NO_HASH No leading hash, i.e. {{plural:...}} instead of {{#if:...}}
4800 *
4801 * Parser::SFH_OBJECT_ARGS Pass the template arguments as PPNode objects instead of text.
4802 * This allows for conditional expansion of the parse tree, allowing you to eliminate dead
4803 * branches and thus speed up parsing. It is also possible to analyse the parse tree of
4804 * the arguments, and to control the way they are expanded.
4805 *
4806 * The $frame parameter is a PPFrame. This can be used to produce expanded text from the
4807 * arguments, for instance:
4808 * $text = isset( $args[0] ) ? $frame->expand( $args[0] ) : '';
4809 *
4810 * For technical reasons, $args[0] is pre-expanded and will be a string. This may change in
4811 * future versions. Please call $frame->expand() on it anyway so that your code keeps
4812 * working if/when this is changed.
4813 *
4814 * If you want whitespace to be trimmed from $args, you need to do it yourself, post-
4815 * expansion.
4816 *
4817 * Please read the documentation in includes/parser/Preprocessor.php for more information
4818 * about the methods available in PPFrame and PPNode.
4819 *
4820 * @throws MWException
4821 * @return string|callable The old callback function for this name, if any
4822 */
4829 # Add to function cache
4833 }
4839 # Case
4842 }
4843 # Add leading hash
4846 }
4847 # Remove trailing colon
4850 }
4852 }
4854 }
4856 /**
4857 * Get all registered function hook identifiers
4858 *
4859 * @return array
4860 */
4863 }
4865 /**
4866 * Create a tag function, e.g. "<test>some stuff</test>".
4867 * Unlike tag hooks, tag functions are parsed at preprocessor level.
4868 * Unlike parser functions, their content is not preprocessed.
4869 * @param string $tag
4870 * @param callable $callback
4871 * @param int $flags
4872 * @throws MWException
4873 * @return null
4874 */
4879 }
4886 }
4889 }
4891 /**
4892 * Replace "<!--LINK-->" link placeholders with actual links, in the buffer
4893 * Placeholders created in Linker::link()
4894 *
4895 * @param string $text
4896 * @param int $options
4897 */
4900 }
4902 /**
4903 * Replace "<!--LINK-->" link placeholders with plain text of links
4904 * (not HTML-formatted).
4905 *
4906 * @param string $text
4907 * @return string
4908 */
4911 }
4913 /**
4914 * Renders an image gallery from a text with one line per image.
4915 * text labels may be given by using |-style alternative text. E.g.
4916 * Image:one.jpg|The number "1"
4917 * Image:tree.jpg|A tree
4918 * given as text will return the HTML of a gallery with two images,
4919 * labeled 'The number "1"' and
4920 * 'A tree'.
4921 *
4922 * @param string $text
4923 * @param array $params
4924 * @return string HTML
4925 */
4931 }
4936 // If invalid type set, fallback to default.
4938 }
4951 }
4957 }
4960 }
4963 }
4966 }
4973 # match lines like these:
4974 # Image:someimage.jpg|This is some image
4977 # Skip empty lines
4980 }
4984 }
4987 # Bogus title. Ignore these so we don't bomb out later.
4989 }
4991 # We need to get what handler the file uses, to figure out parameters.
4992 # Note, a hook can overide the file name, and chose an entirely different
4993 # file (which potentially could be of a different type and have different handler).
4998 # Don't register it now, as TraditionalImageGallery does that later.
5005 ];
5008 // We don't want people to specify per-image widths.
5009 // Additionally the width parameter would need special casing anyhow.
5011 }
5020 // look for an |alt= definition while trying not to break existing
5021 // captions with multiple pipes (|) in it, until a more sensible grammar
5022 // is defined for images in galleries
5024 // FIXME: Doing recursiveTagParse at this stage, and the trim before
5025 // splitting on '|' is a bit odd, and different from makeImage.
5027 // Protect LanguageConverter markup
5030 );
5046 // check to see if link matches an absolute url, if not then it must be a wiki link.
5048 // Result of LanguageConverter::markNoConversion
5049 // invoked on an external link.
5051 }
5058 }
5059 }
5062 // Must be a handler specific parameter.
5066 // Guess not, consider it as caption.
5069 }
5070 }
5073 // Last pipe wins.
5075 }
5076 }
5077 // Remove the pipe.
5079 }
5082 }
5086 }
5088 /**
5089 * @param MediaHandler $handler
5090 * @return array
5091 */
5097 }
5099 # Initialise static lists
5106 ];
5114 }
5115 }
5116 }
5118 # Add handler params
5124 }
5125 }
5128 }
5130 }
5132 /**
5133 * Parse image options text and use it to make an image
5134 *
5135 * @param Title $title
5136 * @param string $options
5137 * @param LinkHolderArray|bool $holders
5138 * @return string HTML
5139 */
5141 # Check if the options text is of the form "options|alt text"
5142 # Options are:
5143 # * thumbnail make a thumbnail with enlarge-icon and caption, alignment depends on lang
5144 # * left no resizing, just left align. label is used for alt= only
5145 # * right same, but right aligned
5146 # * none same, but not aligned
5147 # * ___px scale to ___ pixels width, no aligning. e.g. use in taxobox
5148 # * center center the image
5149 # * frame Keep original image size, no magnify-button.
5150 # * framed Same as "frame"
5151 # * frameless like 'thumb' but without a frame. Keeps user preferences for width
5152 # * upright reduce width for upright images, rounded to full __0 px
5153 # * border draw a 1px border around the image
5154 # * alt Text for HTML alt attribute (defaults to empty)
5155 # * class Set a class for img node
5156 # * link Set the target of the image link. Can be external, interwiki, or local
5157 # vertical-align values (no % or length right now):
5158 # * baseline
5159 # * sub
5160 # * super
5161 # * top
5162 # * text-top
5163 # * middle
5164 # * bottom
5165 # * text-bottom
5167 # Protect LanguageConverter markup when splitting into parts
5170 );
5172 # Give extensions a chance to select the file revision for us
5177 # Fetch and register the file (file title may be different via hooks)
5180 # Get parameter map
5187 }
5189 # Process the input parameters
5201 # Special case; width and height come in one variable together
5209 }
5210 }
5216 }
5217 }
5218 # else no validation -- T15436
5221 # Validate handler parameter
5224 # Validate internal parameters
5229 # @todo FIXME: Possibly check validity here for
5230 # manualthumb? downstream behavior seems odd with
5231 # missing manual thumbs.
5249 }
5251 }
5259 }
5260 }
5265 // use first appearing option, discard others.
5270 # Most other things appear to be empty or numeric...
5272 }
5273 }
5277 }
5278 }
5279 }
5282 }
5283 }
5285 # Process alignment parameters
5288 }
5291 }
5295 # Will the image be presented in a frame, with the caption below?
5301 # In the old days, [[Image:Foo|text...]] would set alt text. Later it
5302 # came to also set the caption, ordinary text after the image -- which
5303 # makes no sense, because that just repeats the text multiple times in
5304 # screen readers. It *also* came to set the title attribute.
5305 # Now that we have an alt attribute, we should not set the alt text to
5306 # equal the caption: that's worse than useless, it just repeats the
5307 # text. This is the framed/thumbnail case. If there's no caption, we
5308 # use the unnamed parameter for alt text as well, just for the time be-
5309 # ing, if the unnamed param is set and the alt param is not.
5310 # For the future, we need to figure out if we want to tweak this more,
5311 # e.g., introducing a title= parameter for the title; ignoring the un-
5312 # named parameter entirely for images without a caption; adding an ex-
5313 # plicit caption= parameter and preserving the old magic unnamed para-
5314 # meter for BC; ...
5317 # No caption or alt text, add the filename as the alt text so
5318 # that screen readers at least get some description of the image
5320 }
5321 # Do not set $params['frame']['title'] because tooltips don't make sense
5322 # for framed images
5325 # No alt text, use the "caption" for the alt text
5329 # No caption, fall back to using the filename for the
5330 # alt text
5332 }
5333 }
5334 # Use the "caption" for the tooltip text
5336 }
5340 # Linker does the rest
5345 # Give the handler a chance to modify the parser object
5348 }
5351 }
5353 /**
5354 * @param string $caption
5355 * @param LinkHolderArray|bool $holders
5356 * @return mixed|string
5357 */
5359 # Strip bad stuff out of the title (tooltip). We can't just use
5360 # replaceLinkHoldersText() here, because if this function is called
5361 # from replaceInternalLinks2(), mLinkHolders won't be up-to-date.
5366 }
5368 # make sure there are no placeholders in thumbnail attributes
5369 # that are later expanded to html- so expand them now and
5370 # remove the tags
5375 }
5377 /**
5378 * Set a flag in the output object indicating that the content is dynamic and
5379 * shouldn't be cached.
5380 * @deprecated since 1.28; use getOutput()->updateCacheExpiry()
5381 */
5387 }
5389 }
5391 /**
5392 * Callback from the Sanitizer for expanding items found in HTML attribute
5393 * values, so they can be safely tested and escaped.
5394 *
5395 * @param string $text
5396 * @param bool|PPFrame $frame
5397 * @return string
5398 */
5403 }
5405 /**
5406 * Accessor
5407 *
5408 * @return array
5409 */
5415 );
5416 }
5418 /**
5419 * Replace transparent tags in $text with the values given by the callbacks.
5420 *
5421 * Transparent tag hooks are like regular XML-style tag hooks, except they
5422 * operate late in the transformation sequence, on HTML instead of wikitext.
5423 *
5424 * @param string $text
5425 *
5426 * @return string
5427 */
5441 );
5444 }
5446 }
5448 }
5450 /**
5451 * Break wikitext input into sections, and either pull or replace
5452 * some particular section's text.
5453 *
5454 * External callers should use the getSection and replaceSection methods.
5455 *
5456 * @param string $text Page wikitext
5457 * @param string|int $sectionId A section identifier string of the form:
5458 * "<flag1> - <flag2> - ... - <section number>"
5459 *
5460 * Currently the only recognised flag is "T", which means the target section number
5461 * was derived during a template inclusion parse, in other words this is a template
5462 * section edit link. If no flags are given, it was an ordinary section edit link.
5463 * This flag is required to avoid a section numbering mismatch when a section is
5464 * enclosed by "<includeonly>" (T8563).
5465 *
5466 * The section number 0 pulls the text before the first heading; other numbers will
5467 * pull the given section along with its lower-level subsections. If the section is
5468 * not found, $mode=get will return $newtext, and $mode=replace will return $text.
5469 *
5470 * Section 0 is always considered to exist, even if it only contains the empty
5471 * string. If $text is the empty string and section 0 is replaced, $newText is
5472 * returned.
5473 *
5474 * @param string $mode One of "get" or "replace"
5475 * @param string $newText Replacement text for section data.
5476 * @return string For "get", the extracted section text.
5477 * for "replace", the whole page with the section replaced.
5478 */
5487 # Process section extraction flags
5494 }
5495 }
5497 # Check for empty input
5499 # Only sections 0 and T-0 exist in an empty document
5505 }
5511 }
5512 }
5513 }
5515 # Preprocess the text
5518 # <h> nodes indicate section breaks
5519 # They can only occur at the top level, so we can find them by iterating the root's children
5522 # Find the target section
5524 # Section zero doesn't nest, level=big
5533 }
5534 }
5537 }
5539 }
5540 }
5543 # Not found
5548 }
5549 }
5551 # Find the end of the section, including nested sections
5558 }
5559 }
5562 }
5566 # Write out the remainder (in replace mode only)
5568 # Output the replacement text
5569 # Add two newlines on -- trailing whitespace in $newText is conventionally
5570 # stripped by the editor, so we need both newlines to restore the paragraph gap
5571 # Only add trailing whitespace if there is newText
5574 }
5579 }
5580 }
5583 # Re-insert stripped tags
5585 }
5588 }
5590 /**
5591 * This function returns the text of a section, specified by a number ($section).
5592 * A section is text under a heading like == Heading == or \<h1\>Heading\</h1\>, or
5593 * the first section before any such heading (section 0).
5594 *
5595 * If a section contains subsections, these are also returned.
5596 *
5597 * @param string $text Text to look in
5598 * @param string|int $sectionId Section identifier as a number or string
5599 * (e.g. 0, 1 or 'T-1').
5600 * @param string $defaultText Default to return if section is not found
5601 *
5602 * @return string Text of the requested section
5603 */
5606 }
5608 /**
5609 * This function returns $oldtext after the content of the section
5610 * specified by $section has been replaced with $text. If the target
5611 * section does not exist, $oldtext is returned unchanged.
5612 *
5613 * @param string $oldText Former text of the article
5614 * @param string|int $sectionId Section identifier as a number or string
5615 * (e.g. 0, 1 or 'T-1').
5616 * @param string $newText Replacing text
5617 *
5618 * @return string Modified text
5619 */
5622 }
5624 /**
5625 * Get the ID of the revision we are parsing
5626 *
5627 * @return int|null
5628 */
5631 }
5633 /**
5634 * Get the revision object for $this->mRevisionId
5635 *
5636 * @return Revision|null Either a Revision object or null
5637 * @since 1.23 (public since 1.23)
5638 */
5642 }
5645 }
5649 );
5651 # If the parse is for a new revision, then the callback should have
5652 # already been set to force the object and should match mRevisionId.
5653 # If not, try to fetch by mRevisionId for sanity.
5656 }
5661 }
5663 /**
5664 * Get the timestamp associated with the current revision, adjusted for
5665 * the default server-local timestamp
5666 * @return string
5667 */
5675 # The cryptic '' timezone parameter tells to use the site-default
5676 # timezone offset instead of the user settings.
5677 # Since this value will be saved into the parser cache, served
5678 # to other users, and potentially even used inside links and such,
5679 # it needs to be consistent for all visitors.
5682 }
5684 }
5686 /**
5687 * Get the name of the user that edited the last revision
5688 *
5689 * @return string User name
5690 */
5695 # if this template is subst: the revision id will be blank,
5696 # so just use the current user's name
5701 }
5702 }
5704 }
5706 /**
5707 * Get the size of the revision
5708 *
5709 * @return int|null Revision size
5710 */
5715 # if this variable is subst: the revision id will be blank,
5716 # so just use the parser input size, because the own substituation
5717 # will change the size.
5722 }
5723 }
5725 }
5727 /**
5728 * Mutator for $mDefaultSort
5729 *
5730 * @param string $sort New value
5731 */
5735 }
5737 /**
5738 * Accessor for $mDefaultSort
5739 * Will use the empty string if none is set.
5740 *
5741 * This value is treated as a prefix, so the
5742 * empty string is equivalent to sorting by
5743 * page name.
5744 *
5745 * @return string
5746 */
5752 }
5753 }
5755 /**
5756 * Accessor for $mDefaultSort
5757 * Unlike getDefaultSort(), will return false if none is set
5758 *
5759 * @return string|bool
5760 */
5763 }
5765 /**
5766 * Try to guess the section anchor name based on a wikitext fragment
5767 * presumably extracted from a heading, for example "Header" from
5768 * "== Header ==".
5769 *
5770 * @param string $text
5771 *
5772 * @return string
5773 */
5775 # Strip out wikitext links(they break the anchor)
5779 }
5781 /**
5782 * Same as guessSectionNameFromWikiText(), but produces legacy anchors
5783 * instead. For use in redirects, since IE6 interprets Redirect: headers
5784 * as something other than UTF-8 (apparently?), resulting in breakage.
5785 *
5786 * @param string $text The section name
5787 * @return string An anchor
5788 */
5790 # Strip out wikitext links(they break the anchor)
5794 }
5796 /**
5797 * Strips a text string of wikitext for use in a section anchor
5798 *
5799 * Accepts a text string and then removes all wikitext from the
5800 * string and leaves only the resultant text (i.e. the result of
5801 * [[User:WikiSysop|Sysop]] would be "Sysop" and the result of
5802 * [[User:WikiSysop]] would be "User:WikiSysop") - this is intended
5803 * to create valid section anchors by mimicing the output of the
5804 * parser when headings are parsed.
5805 *
5806 * @param string $text Text string to be stripped of wikitext
5807 * for use in a Section anchor
5808 * @return string Filtered text string
5809 */
5811 # Strip internal link markup
5815 # Strip external link markup
5816 # @todo FIXME: Not tolerant to blank link text
5817 # I.E. [https://www.mediawiki.org] will render as [1] or something depending
5818 # on how many empty links there are on the page - need to figure that out.
5819 $text = preg_replace( '/\[(?i:' . $this->mUrlProtocols . ')([^ ]+?) ([^[]+)\]/', '$2', $text );
5821 # Parse wikitext quotes (italics & bold)
5824 # Strip HTML tags
5827 }
5829 /**
5830 * strip/replaceVariables/unstrip for preprocessor regression testing
5831 *
5832 * @param string $text
5833 * @param Title $title
5834 * @param ParserOptions $options
5835 * @param int $outputType
5836 *
5837 * @return string
5838 */
5841 ) {
5849 }
5851 /**
5852 * @param string $text
5853 * @param Title $title
5854 * @param ParserOptions $options
5855 * @return string
5856 */
5859 }
5861 /**
5862 * @param string $text
5863 * @param Title $title
5864 * @param ParserOptions $options
5865 * @return string
5866 */
5869 }
5871 /**
5872 * Call a callback function on all regions of the given text that are not
5873 * inside strip markers, and replace those regions with the return value
5874 * of the callback. For example, with input:
5875 *
5876 * aaa<MARKER>bbb
5877 *
5878 * This will call the callback function twice, with 'aaa' and 'bbb'. Those
5879 * two strings will be replaced with the value returned by the callback in
5880 * each case.
5881 *
5882 * @param string $s
5883 * @param callable $callback
5884 *
5885 * @return string
5886 */
5905 }
5906 }
5907 }
5909 }
5911 /**
5912 * Remove any strip markers found in the given text.
5913 *
5914 * @param string $text Input string
5915 * @return string
5916 */
5919 }
5921 /**
5922 * Save the parser state required to convert the given half-parsed text to
5923 * HTML. "Half-parsed" in this context means the output of
5924 * recursiveTagParse() or internalParse(). This output has strip markers
5925 * from replaceVariables (extensionSubstitution() etc.), and link
5926 * placeholders from replaceLinkHolders().
5927 *
5928 * Returns an array which can be serialized and stored persistently. This
5929 * array can later be loaded into another parser instance with
5930 * unserializeHalfParsedText(). The text can then be safely incorporated into
5931 * the return value of a parser hook.
5932 *
5933 * @param string $text
5934 *
5935 * @return array
5936 */
5943 ];
5945 }
5947 /**
5948 * Load the parser state given in the $data array, which is assumed to
5949 * have been generated by serializeHalfParsedText(). The text contents is
5950 * extracted from the array, and its markers are transformed into markers
5951 * appropriate for the current Parser instance. This transformed text is
5952 * returned, and can be safely included in the return value of a parser
5953 * hook.
5954 *
5955 * If the $data array has been stored persistently, the caller should first
5956 * check whether it is still valid, by calling isValidHalfParsedText().
5957 *
5958 * @param array $data Serialized data
5959 * @throws MWException
5960 * @return string
5961 */
5965 }
5967 # First, extract the strip state.
5971 # Now renumber links
5974 # Should be good to go.
5976 }
5978 /**
5979 * Returns true if the given array, presumed to be generated by
5980 * serializeHalfParsedText(), is compatible with the current version of the
5981 * parser.
5982 *
5983 * @param array $data
5984 *
5985 * @return bool
5986 */
5989 }
5991 /**
5992 * Parsed a width param of imagelink like 300px or 200x300px
5993 *
5994 * @param string $value
5995 *
5996 * @return array
5997 * @since 1.20
5998 */
6003 }
6005 # (T15500) In both cases (width/height and width only),
6006 # permit trailing "px" for backward compatibility.
6015 }
6017 }
6019 /**
6020 * Lock the current instance of the parser.
6021 *
6022 * This is meant to stop someone from calling the parser
6023 * recursively and messing up all the strip state.
6024 *
6025 * @throws MWException If parser is in a parse
6026 * @return ScopedCallback The lock will be released once the return value goes out of scope.
6027 */
6032 }
6037 } );
6040 }
6042 /**
6043 * Strip outer <p></p> tag from the HTML source of a single paragraph.
6044 *
6045 * Returns original HTML if the <p/> tag has any attributes, if there's no wrapping <p/> tag,
6046 * or if there is more than one <p/> tag in the input HTML.
6047 *
6048 * @param string $html
6049 * @return string
6050 * @since 1.24
6051 */
6057 }
6058 }
6061 }
6063 /**
6064 * Return this parser if it is not doing anything, otherwise
6065 * get a fresh parser. You can use this method by doing
6066 * $myParser = $wgParser->getFreshParser(), or more simply
6067 * $wgParser->getFreshParser()->parse( ... );
6068 * if you're unsure if $wgParser is safe to use.
6069 *
6070 * @since 1.24
6071 * @return Parser A parser object that is not parsing anything
6072 */
6079 }
6080 }
6082 /**
6083 * Set's up the PHP implementation of OOUI for use in this request
6084 * and instructs OutputPage to enable OOUI for itself.
6085 *
6086 * @since 1.26
6087 */
6091 }
6092 }