MediaWiki  1.28.1
Parser.php
Go to the documentation of this file.
1 <?php
26 
70 class Parser {
76  const VERSION = '1.6.4';
77 
83 
84  # Flags for Parser::setFunctionHook
85  const SFH_NO_HASH = 1;
86  const SFH_OBJECT_ARGS = 2;
87 
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 bug 19052
92  const EXT_LINK_URL_CLASS = '[^][<>"\\x00-\\x20\\x7F\p{Zs}]';
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)
95  const EXT_LINK_ADDR = '(?:[0-9.]+|\\[(?i:[0-9a-f:.]+)\\]|[^][<>"\\x00-\\x20\\x7F\p{Zs}])';
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}]+)
99  \\/([A-Za-z0-9_.,~%\\-+&;#*?!=()@\\x80-\\xFF]+)\\.((?i)gif|png|jpg|jpeg)$/Sxu';
100  // @codingStandardsIgnoreEnd
101 
102  # Regular expression for a non-newline space
103  const SPACE_NOT_NL = '(?:\t|&nbsp;|&\#0*160;|&\#[Xx]0*[Aa]0;|\p{Zs})';
104 
105  # Flags for preprocessToDom
106  const PTD_FOR_INCLUSION = 1;
107 
108  # Allowed values for $this->mOutputType
109  # Parameter to startExternalParse().
110  const OT_HTML = 1; # like parse()
111  const OT_WIKI = 2; # like preSaveTransform()
113  const OT_MSG = 3;
114  const OT_PLAIN = 4; # like extractSections() - portions of the original are returned unchanged.
115 
133  const MARKER_SUFFIX = "-QINU`\"'\x7f";
134  const MARKER_PREFIX = "\x7f'\"`UNIQ-";
135 
136  # Markers used for wrapping the table of contents
137  const TOC_START = '<mw:toc>';
138  const TOC_END = '</mw:toc>';
139 
140  # Persistent:
141  public $mTagHooks = [];
143  public $mFunctionHooks = [];
144  public $mFunctionSynonyms = [ 0 => [], 1 => [] ];
145  public $mFunctionTagHooks = [];
146  public $mStripList = [];
147  public $mDefaultStripList = [];
148  public $mVarCache = [];
149  public $mImageParams = [];
151  public $mMarkerIndex = 0;
152  public $mFirstCall = true;
153 
154  # Initialised by initialiseVariables()
155 
159  public $mVariables;
160 
164  public $mSubstWords;
165  # Initialised in constructor
167 
168  # Initialized in getPreprocessor()
169 
171 
172  # Cleared with clearState():
173 
176  public $mOutput;
177  public $mAutonumber;
178 
182  public $mStripState;
183 
189 
190  public $mLinkID;
194  public $mExpensiveFunctionCount; # number of expensive parser function calls
196 
200  public $mUser; # User object; only used when doing pre-save transform
201 
202  # Temporary
203  # These are variables reset at least once per parse regardless of $clearState
204 
208  public $mOptions;
209 
213  public $mTitle; # Title context, used for self-link rendering and similar things
214  public $mOutputType; # Output type, one of the OT_xxx constants
215  public $ot; # Shortcut alias, see setOutputType()
216  public $mRevisionObject; # The revision object of the specified revision ID
217  public $mRevisionId; # ID to display in {{REVISIONID}} tags
218  public $mRevisionTimestamp; # The timestamp of the specified revision ID
219  public $mRevisionUser; # User to display in {{REVISIONUSER}} tag
220  public $mRevisionSize; # Size to display in {{REVISIONSIZE}} variable
221  public $mRevIdForTs; # The revision ID which was used to fetch the timestamp
222  public $mInputSize = false; # For {{PAGESIZE}} on current page.
223 
228  public $mUniqPrefix = Parser::MARKER_PREFIX;
229 
236 
244 
249  public $mInParse = false;
250 
252  protected $mProfiler;
253 
257  protected $mLinkRenderer;
258 
262  public function __construct( $conf = [] ) {
263  $this->mConf = $conf;
264  $this->mUrlProtocols = wfUrlProtocols();
265  $this->mExtLinkBracketedRegex = '/\[(((?i)' . $this->mUrlProtocols . ')' .
266  self::EXT_LINK_ADDR .
267  self::EXT_LINK_URL_CLASS . '*)\p{Zs}*([^\]\\x00-\\x08\\x0a-\\x1F]*?)\]/Su';
268  if ( isset( $conf['preprocessorClass'] ) ) {
269  $this->mPreprocessorClass = $conf['preprocessorClass'];
270  } elseif ( defined( 'HPHP_VERSION' ) ) {
271  # Preprocessor_Hash is much faster than Preprocessor_DOM under HipHop
272  $this->mPreprocessorClass = 'Preprocessor_Hash';
273  } elseif ( extension_loaded( 'domxml' ) ) {
274  # PECL extension that conflicts with the core DOM extension (bug 13770)
275  wfDebug( "Warning: you have the obsolete domxml extension for PHP. Please remove it!\n" );
276  $this->mPreprocessorClass = 'Preprocessor_Hash';
277  } elseif ( extension_loaded( 'dom' ) ) {
278  $this->mPreprocessorClass = 'Preprocessor_DOM';
279  } else {
280  $this->mPreprocessorClass = 'Preprocessor_Hash';
281  }
282  wfDebug( __CLASS__ . ": using preprocessor: {$this->mPreprocessorClass}\n" );
283  }
284 
288  public function __destruct() {
289  if ( isset( $this->mLinkHolders ) ) {
290  unset( $this->mLinkHolders );
291  }
292  foreach ( $this as $name => $value ) {
293  unset( $this->$name );
294  }
295  }
296 
300  public function __clone() {
301  $this->mInParse = false;
302 
303  // Bug 56226: 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.
309  foreach ( [ 'mStripState', 'mVarCache' ] as $k ) {
310  // Make a non-reference copy of the field, then rebind the field to
311  // reference the new copy.
312  $tmp = $this->$k;
313  $this->$k =& $tmp;
314  unset( $tmp );
315  }
316 
317  Hooks::run( 'ParserCloned', [ $this ] );
318  }
319 
323  public function firstCallInit() {
324  if ( !$this->mFirstCall ) {
325  return;
326  }
327  $this->mFirstCall = false;
328 
330  CoreTagHooks::register( $this );
331  $this->initialiseVariables();
332 
333  Hooks::run( 'ParserFirstCallInit', [ &$this ] );
334  }
335 
341  public function clearState() {
342  if ( $this->mFirstCall ) {
343  $this->firstCallInit();
344  }
345  $this->mOutput = new ParserOutput;
346  $this->mOptions->registerWatcher( [ $this->mOutput, 'recordOption' ] );
347  $this->mAutonumber = 0;
348  $this->mIncludeCount = [];
349  $this->mLinkHolders = new LinkHolderArray( $this );
350  $this->mLinkID = 0;
351  $this->mRevisionObject = $this->mRevisionTimestamp =
352  $this->mRevisionId = $this->mRevisionUser = $this->mRevisionSize = null;
353  $this->mVarCache = [];
354  $this->mUser = null;
355  $this->mLangLinkLanguages = [];
356  $this->currentRevisionCache = null;
357 
358  $this->mStripState = new StripState;
359 
360  # Clear these on every parse, bug 4549
361  $this->mTplRedirCache = $this->mTplDomCache = [];
362 
363  $this->mShowToc = true;
364  $this->mForceTocPosition = false;
365  $this->mIncludeSizes = [
366  'post-expand' => 0,
367  'arg' => 0,
368  ];
369  $this->mPPNodeCount = 0;
370  $this->mGeneratedPPNodeCount = 0;
371  $this->mHighestExpansionDepth = 0;
372  $this->mDefaultSort = false;
373  $this->mHeadings = [];
374  $this->mDoubleUnderscores = [];
375  $this->mExpensiveFunctionCount = 0;
376 
377  # Fix cloning
378  if ( isset( $this->mPreprocessor ) && $this->mPreprocessor->parser !== $this ) {
379  $this->mPreprocessor = null;
380  }
381 
382  $this->mProfiler = new SectionProfiler();
383 
384  Hooks::run( 'ParserClearState', [ &$this ] );
385  }
386 
399  public function parse(
401  $linestart = true, $clearState = true, $revid = null
402  ) {
408  global $wgShowHostnames;
409 
410  if ( $clearState ) {
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.
413  $text = strtr( $text, "\x7f", "?" );
414  $magicScopeVariable = $this->lock();
415  }
416 
417  $this->startParse( $title, $options, self::OT_HTML, $clearState );
418 
419  $this->currentRevisionCache = null;
420  $this->mInputSize = strlen( $text );
421  if ( $this->mOptions->getEnableLimitReport() ) {
422  $this->mOutput->resetParseStartTime();
423  }
424 
425  $oldRevisionId = $this->mRevisionId;
426  $oldRevisionObject = $this->mRevisionObject;
427  $oldRevisionTimestamp = $this->mRevisionTimestamp;
428  $oldRevisionUser = $this->mRevisionUser;
429  $oldRevisionSize = $this->mRevisionSize;
430  if ( $revid !== null ) {
431  $this->mRevisionId = $revid;
432  $this->mRevisionObject = null;
433  $this->mRevisionTimestamp = null;
434  $this->mRevisionUser = null;
435  $this->mRevisionSize = null;
436  }
437 
438  Hooks::run( 'ParserBeforeStrip', [ &$this, &$text, &$this->mStripState ] );
439  # No more strip!
440  Hooks::run( 'ParserAfterStrip', [ &$this, &$text, &$this->mStripState ] );
441  $text = $this->internalParse( $text );
442  Hooks::run( 'ParserAfterParse', [ &$this, &$text, &$this->mStripState ] );
443 
444  $text = $this->internalParseHalfParsed( $text, true, $linestart );
445 
453  if ( !( $options->getDisableTitleConversion()
454  || isset( $this->mDoubleUnderscores['nocontentconvert'] )
455  || isset( $this->mDoubleUnderscores['notitleconvert'] )
456  || $this->mOutput->getDisplayTitle() !== false )
457  ) {
458  $convruletitle = $this->getConverterLanguage()->getConvRuleTitle();
459  if ( $convruletitle ) {
460  $this->mOutput->setTitleText( $convruletitle );
461  } else {
462  $titleText = $this->getConverterLanguage()->convertTitle( $title );
463  $this->mOutput->setTitleText( $titleText );
464  }
465  }
466 
467  # Done parsing! Compute runtime adaptive expiry if set
468  $this->mOutput->finalizeAdaptiveCacheExpiry();
469 
470  # Warn if too many heavyweight parser functions were used
471  if ( $this->mExpensiveFunctionCount > $this->mOptions->getExpensiveParserFunctionLimit() ) {
472  $this->limitationWarn( 'expensive-parserfunction',
473  $this->mExpensiveFunctionCount,
474  $this->mOptions->getExpensiveParserFunctionLimit()
475  );
476  }
477 
478  # Information on include size limits, for the benefit of users who try to skirt them
479  if ( $this->mOptions->getEnableLimitReport() ) {
480  $max = $this->mOptions->getMaxIncludeSize();
481 
482  $cpuTime = $this->mOutput->getTimeSinceStart( 'cpu' );
483  if ( $cpuTime !== null ) {
484  $this->mOutput->setLimitReportData( 'limitreport-cputime',
485  sprintf( "%.3f", $cpuTime )
486  );
487  }
488 
489  $wallTime = $this->mOutput->getTimeSinceStart( 'wall' );
490  $this->mOutput->setLimitReportData( 'limitreport-walltime',
491  sprintf( "%.3f", $wallTime )
492  );
493 
494  $this->mOutput->setLimitReportData( 'limitreport-ppvisitednodes',
495  [ $this->mPPNodeCount, $this->mOptions->getMaxPPNodeCount() ]
496  );
497  $this->mOutput->setLimitReportData( 'limitreport-ppgeneratednodes',
498  [ $this->mGeneratedPPNodeCount, $this->mOptions->getMaxGeneratedPPNodeCount() ]
499  );
500  $this->mOutput->setLimitReportData( 'limitreport-postexpandincludesize',
501  [ $this->mIncludeSizes['post-expand'], $max ]
502  );
503  $this->mOutput->setLimitReportData( 'limitreport-templateargumentsize',
504  [ $this->mIncludeSizes['arg'], $max ]
505  );
506  $this->mOutput->setLimitReportData( 'limitreport-expansiondepth',
507  [ $this->mHighestExpansionDepth, $this->mOptions->getMaxPPExpandDepth() ]
508  );
509  $this->mOutput->setLimitReportData( 'limitreport-expensivefunctioncount',
510  [ $this->mExpensiveFunctionCount, $this->mOptions->getExpensiveParserFunctionLimit() ]
511  );
512  Hooks::run( 'ParserLimitReportPrepare', [ $this, $this->mOutput ] );
513 
514  $limitReport = "NewPP limit report\n";
515  if ( $wgShowHostnames ) {
516  $limitReport .= 'Parsed by ' . wfHostname() . "\n";
517  }
518  $limitReport .= 'Cached time: ' . $this->mOutput->getCacheTime() . "\n";
519  $limitReport .= 'Cache expiry: ' . $this->mOutput->getCacheExpiry() . "\n";
520  $limitReport .= 'Dynamic content: ' .
521  ( $this->mOutput->hasDynamicContent() ? 'true' : 'false' ) .
522  "\n";
523 
524  foreach ( $this->mOutput->getLimitReportData() as $key => $value ) {
525  if ( Hooks::run( 'ParserLimitReportFormat',
526  [ $key, &$value, &$limitReport, false, false ]
527  ) ) {
528  $keyMsg = wfMessage( $key )->inLanguage( 'en' )->useDatabase( false );
529  $valueMsg = wfMessage( [ "$key-value-text", "$key-value" ] )
530  ->inLanguage( 'en' )->useDatabase( false );
531  if ( !$valueMsg->exists() ) {
532  $valueMsg = new RawMessage( '$1' );
533  }
534  if ( !$keyMsg->isDisabled() && !$valueMsg->isDisabled() ) {
535  $valueMsg->params( $value );
536  $limitReport .= "{$keyMsg->text()}: {$valueMsg->text()}\n";
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.
542  $limitReport = htmlspecialchars_decode( $limitReport );
543  Hooks::run( 'ParserLimitReport', [ $this, &$limitReport ] );
544 
545  // Sanitize for comment. Note '‐' in the replacement is U+2010,
546  // which looks much like the problematic '-'.
547  $limitReport = str_replace( [ '-', '&' ], [ '‐', '&amp;' ], $limitReport );
548  $text .= "\n<!-- \n$limitReport-->\n";
549 
550  // Add on template profiling data
551  $dataByFunc = $this->mProfiler->getFunctionStats();
552  uasort( $dataByFunc, function ( $a, $b ) {
553  return $a['real'] < $b['real']; // descending order
554  } );
555  $profileReport = "Transclusion expansion time report (%,ms,calls,template)\n";
556  foreach ( array_slice( $dataByFunc, 0, 10 ) as $item ) {
557  $profileReport .= sprintf( "%6.2f%% %8.3f %6d - %s\n",
558  $item['%real'], $item['real'], $item['calls'],
559  htmlspecialchars( $item['name'] ) );
560  }
561  $text .= "\n<!-- \n$profileReport-->\n";
562 
563  if ( $this->mGeneratedPPNodeCount > $this->mOptions->getMaxGeneratedPPNodeCount() / 10 ) {
564  wfDebugLog( 'generated-pp-node-count', $this->mGeneratedPPNodeCount . ' ' .
565  $this->mTitle->getPrefixedDBkey() );
566  }
567  }
568  $this->mOutput->setText( $text );
569 
570  $this->mRevisionId = $oldRevisionId;
571  $this->mRevisionObject = $oldRevisionObject;
572  $this->mRevisionTimestamp = $oldRevisionTimestamp;
573  $this->mRevisionUser = $oldRevisionUser;
574  $this->mRevisionSize = $oldRevisionSize;
575  $this->mInputSize = false;
576  $this->currentRevisionCache = null;
577 
578  return $this->mOutput;
579  }
580 
603  public function recursiveTagParse( $text, $frame = false ) {
604  Hooks::run( 'ParserBeforeStrip', [ &$this, &$text, &$this->mStripState ] );
605  Hooks::run( 'ParserAfterStrip', [ &$this, &$text, &$this->mStripState ] );
606  $text = $this->internalParse( $text, false, $frame );
607  return $text;
608  }
609 
627  public function recursiveTagParseFully( $text, $frame = false ) {
628  $text = $this->recursiveTagParse( $text, $frame );
629  $text = $this->internalParseHalfParsed( $text, false );
630  return $text;
631  }
632 
644  public function preprocess( $text, Title $title = null,
645  ParserOptions $options, $revid = null, $frame = false
646  ) {
647  $magicScopeVariable = $this->lock();
648  $this->startParse( $title, $options, self::OT_PREPROCESS, true );
649  if ( $revid !== null ) {
650  $this->mRevisionId = $revid;
651  }
652  Hooks::run( 'ParserBeforeStrip', [ &$this, &$text, &$this->mStripState ] );
653  Hooks::run( 'ParserAfterStrip', [ &$this, &$text, &$this->mStripState ] );
654  $text = $this->replaceVariables( $text, $frame );
655  $text = $this->mStripState->unstripBoth( $text );
656  return $text;
657  }
658 
668  public function recursivePreprocess( $text, $frame = false ) {
669  $text = $this->replaceVariables( $text, $frame );
670  $text = $this->mStripState->unstripBoth( $text );
671  return $text;
672  }
673 
687  public function getPreloadText( $text, Title $title, ParserOptions $options, $params = [] ) {
688  $msg = new RawMessage( $text );
689  $text = $msg->params( $params )->plain();
690 
691  # Parser (re)initialisation
692  $magicScopeVariable = $this->lock();
693  $this->startParse( $title, $options, self::OT_PLAIN, true );
694 
696  $dom = $this->preprocessToDom( $text, self::PTD_FOR_INCLUSION );
697  $text = $this->getPreprocessor()->newFrame()->expand( $dom, $flags );
698  $text = $this->mStripState->unstripBoth( $text );
699  return $text;
700  }
701 
708  public static function getRandomString() {
709  wfDeprecated( __METHOD__, '1.26' );
710  return wfRandomString( 16 );
711  }
712 
719  public function setUser( $user ) {
720  $this->mUser = $user;
721  }
722 
729  public function uniqPrefix() {
730  wfDeprecated( __METHOD__, '1.26' );
731  return self::MARKER_PREFIX;
732  }
733 
739  public function setTitle( $t ) {
740  if ( !$t ) {
741  $t = Title::newFromText( 'NO TITLE' );
742  }
743 
744  if ( $t->hasFragment() ) {
745  # Strip the fragment to avoid various odd effects
746  $this->mTitle = $t->createFragmentTarget( '' );
747  } else {
748  $this->mTitle = $t;
749  }
750  }
751 
757  public function getTitle() {
758  return $this->mTitle;
759  }
760 
767  public function Title( $x = null ) {
768  return wfSetVar( $this->mTitle, $x );
769  }
770 
776  public function setOutputType( $ot ) {
777  $this->mOutputType = $ot;
778  # Shortcut alias
779  $this->ot = [
780  'html' => $ot == self::OT_HTML,
781  'wiki' => $ot == self::OT_WIKI,
782  'pre' => $ot == self::OT_PREPROCESS,
783  'plain' => $ot == self::OT_PLAIN,
784  ];
785  }
786 
793  public function OutputType( $x = null ) {
794  return wfSetVar( $this->mOutputType, $x );
795  }
796 
802  public function getOutput() {
803  return $this->mOutput;
804  }
805 
811  public function getOptions() {
812  return $this->mOptions;
813  }
814 
821  public function Options( $x = null ) {
822  return wfSetVar( $this->mOptions, $x );
823  }
824 
828  public function nextLinkID() {
829  return $this->mLinkID++;
830  }
831 
835  public function setLinkID( $id ) {
836  $this->mLinkID = $id;
837  }
838 
843  public function getFunctionLang() {
844  return $this->getTargetLanguage();
845  }
846 
856  public function getTargetLanguage() {
857  $target = $this->mOptions->getTargetLanguage();
858 
859  if ( $target !== null ) {
860  return $target;
861  } elseif ( $this->mOptions->getInterfaceMessage() ) {
862  return $this->mOptions->getUserLangObj();
863  } elseif ( is_null( $this->mTitle ) ) {
864  throw new MWException( __METHOD__ . ': $this->mTitle is null' );
865  }
866 
867  return $this->mTitle->getPageLanguage();
868  }
869 
874  public function getConverterLanguage() {
875  return $this->getTargetLanguage();
876  }
877 
884  public function getUser() {
885  if ( !is_null( $this->mUser ) ) {
886  return $this->mUser;
887  }
888  return $this->mOptions->getUser();
889  }
890 
896  public function getPreprocessor() {
897  if ( !isset( $this->mPreprocessor ) ) {
898  $class = $this->mPreprocessorClass;
899  $this->mPreprocessor = new $class( $this );
900  }
901  return $this->mPreprocessor;
902  }
903 
910  public function getLinkRenderer() {
911  if ( !$this->mLinkRenderer ) {
912  $this->mLinkRenderer = MediaWikiServices::getInstance()
913  ->getLinkRendererFactory()->create();
914  $this->mLinkRenderer->setStubThreshold(
915  $this->getOptions()->getStubThreshold()
916  );
917  }
918 
919  return $this->mLinkRenderer;
920  }
921 
943  public static function extractTagsAndParams( $elements, $text, &$matches, $uniq_prefix = null ) {
944  if ( $uniq_prefix !== null ) {
945  wfDeprecated( __METHOD__ . ' called with $prefix argument', '1.26' );
946  }
947  static $n = 1;
948  $stripped = '';
949  $matches = [];
950 
951  $taglist = implode( '|', $elements );
952  $start = "/<($taglist)(\\s+[^>]*?|\\s*?)(\/?" . ">)|<(!--)/i";
953 
954  while ( $text != '' ) {
955  $p = preg_split( $start, $text, 2, PREG_SPLIT_DELIM_CAPTURE );
956  $stripped .= $p[0];
957  if ( count( $p ) < 5 ) {
958  break;
959  }
960  if ( count( $p ) > 5 ) {
961  # comment
962  $element = $p[4];
963  $attributes = '';
964  $close = '';
965  $inside = $p[5];
966  } else {
967  # tag
968  $element = $p[1];
969  $attributes = $p[2];
970  $close = $p[3];
971  $inside = $p[4];
972  }
973 
974  $marker = self::MARKER_PREFIX . "-$element-" . sprintf( '%08X', $n++ ) . self::MARKER_SUFFIX;
975  $stripped .= $marker;
976 
977  if ( $close === '/>' ) {
978  # Empty element tag, <tag />
979  $content = null;
980  $text = $inside;
981  $tail = null;
982  } else {
983  if ( $element === '!--' ) {
984  $end = '/(-->)/';
985  } else {
986  $end = "/(<\\/$element\\s*>)/i";
987  }
988  $q = preg_split( $end, $inside, 2, PREG_SPLIT_DELIM_CAPTURE );
989  $content = $q[0];
990  if ( count( $q ) < 3 ) {
991  # No end tag -- let it run out to the end of the text.
992  $tail = '';
993  $text = '';
994  } else {
995  $tail = $q[1];
996  $text = $q[2];
997  }
998  }
999 
1000  $matches[$marker] = [ $element,
1001  $content,
1002  Sanitizer::decodeTagAttributes( $attributes ),
1003  "<$element$attributes$close$content$tail" ];
1004  }
1005  return $stripped;
1006  }
1007 
1013  public function getStripList() {
1014  return $this->mStripList;
1015  }
1016 
1026  public function insertStripItem( $text ) {
1027  $marker = self::MARKER_PREFIX . "-item-{$this->mMarkerIndex}-" . self::MARKER_SUFFIX;
1028  $this->mMarkerIndex++;
1029  $this->mStripState->addGeneral( $marker, $text );
1030  return $marker;
1031  }
1032 
1040  public function doTableStuff( $text ) {
1041 
1042  $lines = StringUtils::explode( "\n", $text );
1043  $out = '';
1044  $td_history = []; # Is currently a td tag open?
1045  $last_tag_history = []; # Save history of last lag activated (td, th or caption)
1046  $tr_history = []; # Is currently a tr tag open?
1047  $tr_attributes = []; # history of tr attributes
1048  $has_opened_tr = []; # Did this table open a <tr> element?
1049  $indent_level = 0; # indent level of the table
1050 
1051  foreach ( $lines as $outLine ) {
1052  $line = trim( $outLine );
1053 
1054  if ( $line === '' ) { # empty line, go to next line
1055  $out .= $outLine . "\n";
1056  continue;
1057  }
1058 
1059  $first_character = $line[0];
1060  $first_two = substr( $line, 0, 2 );
1061  $matches = [];
1062 
1063  if ( preg_match( '/^(:*)\s*\{\|(.*)$/', $line, $matches ) ) {
1064  # First check if we are starting a new table
1065  $indent_level = strlen( $matches[1] );
1066 
1067  $attributes = $this->mStripState->unstripBoth( $matches[2] );
1068  $attributes = Sanitizer::fixTagAttributes( $attributes, 'table' );
1069 
1070  $outLine = str_repeat( '<dl><dd>', $indent_level ) . "<table{$attributes}>";
1071  array_push( $td_history, false );
1072  array_push( $last_tag_history, '' );
1073  array_push( $tr_history, false );
1074  array_push( $tr_attributes, '' );
1075  array_push( $has_opened_tr, false );
1076  } elseif ( count( $td_history ) == 0 ) {
1077  # Don't do any of the following
1078  $out .= $outLine . "\n";
1079  continue;
1080  } elseif ( $first_two === '|}' ) {
1081  # We are ending a table
1082  $line = '</table>' . substr( $line, 2 );
1083  $last_tag = array_pop( $last_tag_history );
1084 
1085  if ( !array_pop( $has_opened_tr ) ) {
1086  $line = "<tr><td></td></tr>{$line}";
1087  }
1088 
1089  if ( array_pop( $tr_history ) ) {
1090  $line = "</tr>{$line}";
1091  }
1092 
1093  if ( array_pop( $td_history ) ) {
1094  $line = "</{$last_tag}>{$line}";
1095  }
1096  array_pop( $tr_attributes );
1097  $outLine = $line . str_repeat( '</dd></dl>', $indent_level );
1098  } elseif ( $first_two === '|-' ) {
1099  # Now we have a table row
1100  $line = preg_replace( '#^\|-+#', '', $line );
1101 
1102  # Whats after the tag is now only attributes
1103  $attributes = $this->mStripState->unstripBoth( $line );
1104  $attributes = Sanitizer::fixTagAttributes( $attributes, 'tr' );
1105  array_pop( $tr_attributes );
1106  array_push( $tr_attributes, $attributes );
1107 
1108  $line = '';
1109  $last_tag = array_pop( $last_tag_history );
1110  array_pop( $has_opened_tr );
1111  array_push( $has_opened_tr, true );
1112 
1113  if ( array_pop( $tr_history ) ) {
1114  $line = '</tr>';
1115  }
1116 
1117  if ( array_pop( $td_history ) ) {
1118  $line = "</{$last_tag}>{$line}";
1119  }
1120 
1121  $outLine = $line;
1122  array_push( $tr_history, false );
1123  array_push( $td_history, false );
1124  array_push( $last_tag_history, '' );
1125  } elseif ( $first_character === '|'
1126  || $first_character === '!'
1127  || $first_two === '|+'
1128  ) {
1129  # This might be cell elements, td, th or captions
1130  if ( $first_two === '|+' ) {
1131  $first_character = '+';
1132  $line = substr( $line, 2 );
1133  } else {
1134  $line = substr( $line, 1 );
1135  }
1136 
1137  // Implies both are valid for table headings.
1138  if ( $first_character === '!' ) {
1139  $line = StringUtils::replaceMarkup( '!!', '||', $line );
1140  }
1141 
1142  # Split up multiple cells on the same line.
1143  # FIXME : This can result in improper nesting of tags processed
1144  # by earlier parser steps.
1145  $cells = explode( '||', $line );
1146 
1147  $outLine = '';
1148 
1149  # Loop through each table cell
1150  foreach ( $cells as $cell ) {
1151  $previous = '';
1152  if ( $first_character !== '+' ) {
1153  $tr_after = array_pop( $tr_attributes );
1154  if ( !array_pop( $tr_history ) ) {
1155  $previous = "<tr{$tr_after}>\n";
1156  }
1157  array_push( $tr_history, true );
1158  array_push( $tr_attributes, '' );
1159  array_pop( $has_opened_tr );
1160  array_push( $has_opened_tr, true );
1161  }
1162 
1163  $last_tag = array_pop( $last_tag_history );
1164 
1165  if ( array_pop( $td_history ) ) {
1166  $previous = "</{$last_tag}>\n{$previous}";
1167  }
1168 
1169  if ( $first_character === '|' ) {
1170  $last_tag = 'td';
1171  } elseif ( $first_character === '!' ) {
1172  $last_tag = 'th';
1173  } elseif ( $first_character === '+' ) {
1174  $last_tag = 'caption';
1175  } else {
1176  $last_tag = '';
1177  }
1178 
1179  array_push( $last_tag_history, $last_tag );
1180 
1181  # A cell could contain both parameters and data
1182  $cell_data = explode( '|', $cell, 2 );
1183 
1184  # Bug 553: Note that a '|' inside an invalid link should not
1185  # be mistaken as delimiting cell parameters
1186  if ( strpos( $cell_data[0], '[[' ) !== false ) {
1187  $cell = "{$previous}<{$last_tag}>{$cell}";
1188  } elseif ( count( $cell_data ) == 1 ) {
1189  $cell = "{$previous}<{$last_tag}>{$cell_data[0]}";
1190  } else {
1191  $attributes = $this->mStripState->unstripBoth( $cell_data[0] );
1192  $attributes = Sanitizer::fixTagAttributes( $attributes, $last_tag );
1193  $cell = "{$previous}<{$last_tag}{$attributes}>{$cell_data[1]}";
1194  }
1195 
1196  $outLine .= $cell;
1197  array_push( $td_history, true );
1198  }
1199  }
1200  $out .= $outLine . "\n";
1201  }
1202 
1203  # Closing open td, tr && table
1204  while ( count( $td_history ) > 0 ) {
1205  if ( array_pop( $td_history ) ) {
1206  $out .= "</td>\n";
1207  }
1208  if ( array_pop( $tr_history ) ) {
1209  $out .= "</tr>\n";
1210  }
1211  if ( !array_pop( $has_opened_tr ) ) {
1212  $out .= "<tr><td></td></tr>\n";
1213  }
1214 
1215  $out .= "</table>\n";
1216  }
1217 
1218  # Remove trailing line-ending (b/c)
1219  if ( substr( $out, -1 ) === "\n" ) {
1220  $out = substr( $out, 0, -1 );
1221  }
1222 
1223  # special case: don't return empty table
1224  if ( $out === "<table>\n<tr><td></td></tr>\n</table>" ) {
1225  $out = '';
1226  }
1227 
1228  return $out;
1229  }
1230 
1243  public function internalParse( $text, $isMain = true, $frame = false ) {
1244 
1245  $origText = $text;
1246 
1247  # Hook to suspend the parser in this state
1248  if ( !Hooks::run( 'ParserBeforeInternalParse', [ &$this, &$text, &$this->mStripState ] ) ) {
1249  return $text;
1250  }
1251 
1252  # if $frame is provided, then use $frame for replacing any variables
1253  if ( $frame ) {
1254  # use frame depth to infer how include/noinclude tags should be handled
1255  # depth=0 means this is the top-level document; otherwise it's an included document
1256  if ( !$frame->depth ) {
1257  $flag = 0;
1258  } else {
1259  $flag = Parser::PTD_FOR_INCLUSION;
1260  }
1261  $dom = $this->preprocessToDom( $text, $flag );
1262  $text = $frame->expand( $dom );
1263  } else {
1264  # if $frame is not provided, then use old-style replaceVariables
1265  $text = $this->replaceVariables( $text );
1266  }
1267 
1268  Hooks::run( 'InternalParseBeforeSanitize', [ &$this, &$text, &$this->mStripState ] );
1269  $text = Sanitizer::removeHTMLtags(
1270  $text,
1271  [ &$this, 'attributeStripCallback' ],
1272  false,
1273  array_keys( $this->mTransparentTagHooks ),
1274  [],
1275  [ &$this, 'addTrackingCategory' ]
1276  );
1277  Hooks::run( 'InternalParseBeforeLinks', [ &$this, &$text, &$this->mStripState ] );
1278 
1279  # Tables need to come after variable replacement for things to work
1280  # properly; putting them before other transformations should keep
1281  # exciting things like link expansions from showing up in surprising
1282  # places.
1283  $text = $this->doTableStuff( $text );
1284 
1285  $text = preg_replace( '/(^|\n)-----*/', '\\1<hr />', $text );
1286 
1287  $text = $this->doDoubleUnderscore( $text );
1288 
1289  $text = $this->doHeadings( $text );
1290  $text = $this->replaceInternalLinks( $text );
1291  $text = $this->doAllQuotes( $text );
1292  $text = $this->replaceExternalLinks( $text );
1293 
1294  # replaceInternalLinks may sometimes leave behind
1295  # absolute URLs, which have to be masked to hide them from replaceExternalLinks
1296  $text = str_replace( self::MARKER_PREFIX . 'NOPARSE', '', $text );
1297 
1298  $text = $this->doMagicLinks( $text );
1299  $text = $this->formatHeadings( $text, $origText, $isMain );
1300 
1301  return $text;
1302  }
1303 
1313  private function internalParseHalfParsed( $text, $isMain = true, $linestart = true ) {
1314  $text = $this->mStripState->unstripGeneral( $text );
1315 
1316  if ( $isMain ) {
1317  Hooks::run( 'ParserAfterUnstrip', [ &$this, &$text ] );
1318  }
1319 
1320  # Clean up special characters, only run once, next-to-last before doBlockLevels
1321  $fixtags = [
1322  # french spaces, last one Guillemet-left
1323  # only if there is something before the space
1324  '/(.) (?=\\?|:|;|!|%|\\302\\273)/' => '\\1&#160;',
1325  # french spaces, Guillemet-right
1326  '/(\\302\\253) /' => '\\1&#160;',
1327  '/&#160;(!\s*important)/' => ' \\1', # Beware of CSS magic word !important, bug #11874.
1328  ];
1329  $text = preg_replace( array_keys( $fixtags ), array_values( $fixtags ), $text );
1330 
1331  $text = $this->doBlockLevels( $text, $linestart );
1332 
1333  $this->replaceLinkHolders( $text );
1334 
1342  if ( !( $this->mOptions->getDisableContentConversion()
1343  || isset( $this->mDoubleUnderscores['nocontentconvert'] ) )
1344  ) {
1345  if ( !$this->mOptions->getInterfaceMessage() ) {
1346  # The position of the convert() call should not be changed. it
1347  # assumes that the links are all replaced and the only thing left
1348  # is the <nowiki> mark.
1349  $text = $this->getConverterLanguage()->convert( $text );
1350  }
1351  }
1352 
1353  $text = $this->mStripState->unstripNoWiki( $text );
1354 
1355  if ( $isMain ) {
1356  Hooks::run( 'ParserBeforeTidy', [ &$this, &$text ] );
1357  }
1358 
1359  $text = $this->replaceTransparentTags( $text );
1360  $text = $this->mStripState->unstripGeneral( $text );
1361 
1362  $text = Sanitizer::normalizeCharReferences( $text );
1363 
1364  if ( MWTidy::isEnabled() ) {
1365  if ( $this->mOptions->getTidy() ) {
1366  $text = MWTidy::tidy( $text );
1367  }
1368  } else {
1369  # attempt to sanitize at least some nesting problems
1370  # (bug #2702 and quite a few others)
1371  $tidyregs = [
1372  # ''Something [http://www.cool.com cool''] -->
1373  # <i>Something</i><a href="http://www.cool.com"..><i>cool></i></a>
1374  '/(<([bi])>)(<([bi])>)?([^<]*)(<\/?a[^<]*>)([^<]*)(<\/\\4>)?(<\/\\2>)/' =>
1375  '\\1\\3\\5\\8\\9\\6\\1\\3\\7\\8\\9',
1376  # fix up an anchor inside another anchor, only
1377  # at least for a single single nested link (bug 3695)
1378  '/(<a[^>]+>)([^<]*)(<a[^>]+>[^<]*)<\/a>(.*)<\/a>/' =>
1379  '\\1\\2</a>\\3</a>\\1\\4</a>',
1380  # fix div inside inline elements- doBlockLevels won't wrap a line which
1381  # contains a div, so fix it up here; replace
1382  # div with escaped text
1383  '/(<([aib]) [^>]+>)([^<]*)(<div([^>]*)>)(.*)(<\/div>)([^<]*)(<\/\\2>)/' =>
1384  '\\1\\3&lt;div\\5&gt;\\6&lt;/div&gt;\\8\\9',
1385  # remove empty italic or bold tag pairs, some
1386  # introduced by rules above
1387  '/<([bi])><\/\\1>/' => '',
1388  ];
1389 
1390  $text = preg_replace(
1391  array_keys( $tidyregs ),
1392  array_values( $tidyregs ),
1393  $text );
1394  }
1395 
1396  if ( $isMain ) {
1397  Hooks::run( 'ParserAfterTidy', [ &$this, &$text ] );
1398  }
1399 
1400  return $text;
1401  }
1402 
1414  public function doMagicLinks( $text ) {
1415  $prots = wfUrlProtocolsWithoutProtRel();
1416  $urlChar = self::EXT_LINK_URL_CLASS;
1417  $addr = self::EXT_LINK_ADDR;
1418  $space = self::SPACE_NOT_NL; # non-newline space
1419  $spdash = "(?:-|$space)"; # a dash or a non-newline space
1420  $spaces = "$space++"; # possessive match of 1 or more spaces
1421  $text = preg_replace_callback(
1422  '!(?: # Start cases
1423  (<a[ \t\r\n>].*?</a>) | # m[1]: Skip link text
1424  (<.*?>) | # m[2]: Skip stuff inside
1425  # HTML elements' . "
1426  (\b(?i:$prots)($addr$urlChar*)) | # m[3]: Free external links
1427  # m[4]: Post-protocol path
1428  \b(?:RFC|PMID) $spaces # m[5]: RFC or PMID, capture number
1429  ([0-9]+)\b |
1430  \bISBN $spaces ( # m[6]: ISBN, capture number
1431  (?: 97[89] $spdash? )? # optional 13-digit ISBN prefix
1432  (?: [0-9] $spdash? ){9} # 9 digits with opt. delimiters
1433  [0-9Xx] # check digit
1434  )\b
1435  )!xu", [ &$this, 'magicLinkCallback' ], $text );
1436  return $text;
1437  }
1438 
1444  public function magicLinkCallback( $m ) {
1445  if ( isset( $m[1] ) && $m[1] !== '' ) {
1446  # Skip anchor
1447  return $m[0];
1448  } elseif ( isset( $m[2] ) && $m[2] !== '' ) {
1449  # Skip HTML element
1450  return $m[0];
1451  } elseif ( isset( $m[3] ) && $m[3] !== '' ) {
1452  # Free external link
1453  return $this->makeFreeExternalLink( $m[0], strlen( $m[4] ) );
1454  } elseif ( isset( $m[5] ) && $m[5] !== '' ) {
1455  # RFC or PMID
1456  if ( substr( $m[0], 0, 3 ) === 'RFC' ) {
1457  if ( !$this->mOptions->getMagicRFCLinks() ) {
1458  return $m[0];
1459  }
1460  $keyword = 'RFC';
1461  $urlmsg = 'rfcurl';
1462  $cssClass = 'mw-magiclink-rfc';
1463  $trackingCat = 'magiclink-tracking-rfc';
1464  $id = $m[5];
1465  } elseif ( substr( $m[0], 0, 4 ) === 'PMID' ) {
1466  if ( !$this->mOptions->getMagicPMIDLinks() ) {
1467  return $m[0];
1468  }
1469  $keyword = 'PMID';
1470  $urlmsg = 'pubmedurl';
1471  $cssClass = 'mw-magiclink-pmid';
1472  $trackingCat = 'magiclink-tracking-pmid';
1473  $id = $m[5];
1474  } else {
1475  throw new MWException( __METHOD__ . ': unrecognised match type "' .
1476  substr( $m[0], 0, 20 ) . '"' );
1477  }
1478  $url = wfMessage( $urlmsg, $id )->inContentLanguage()->text();
1479  $this->addTrackingCategory( $trackingCat );
1480  return Linker::makeExternalLink( $url, "{$keyword} {$id}", true, $cssClass, [], $this->mTitle );
1481  } elseif ( isset( $m[6] ) && $m[6] !== ''
1482  && $this->mOptions->getMagicISBNLinks()
1483  ) {
1484  # ISBN
1485  $isbn = $m[6];
1486  $space = self::SPACE_NOT_NL; # non-newline space
1487  $isbn = preg_replace( "/$space/", ' ', $isbn );
1488  $num = strtr( $isbn, [
1489  '-' => '',
1490  ' ' => '',
1491  'x' => 'X',
1492  ] );
1493  $this->addTrackingCategory( 'magiclink-tracking-isbn' );
1494  return $this->getLinkRenderer()->makeKnownLink(
1495  SpecialPage::getTitleFor( 'Booksources', $num ),
1496  "ISBN $isbn",
1497  [
1498  'class' => 'internal mw-magiclink-isbn',
1499  'title' => false // suppress title attribute
1500  ]
1501  );
1502  } else {
1503  return $m[0];
1504  }
1505  }
1506 
1516  public function makeFreeExternalLink( $url, $numPostProto ) {
1517  $trail = '';
1518 
1519  # The characters '<' and '>' (which were escaped by
1520  # removeHTMLtags()) should not be included in
1521  # URLs, per RFC 2396.
1522  # Make &nbsp; terminate a URL as well (bug T84937)
1523  $m2 = [];
1524  if ( preg_match(
1525  '/&(lt|gt|nbsp|#x0*(3[CcEe]|[Aa]0)|#0*(60|62|160));/',
1526  $url,
1527  $m2,
1528  PREG_OFFSET_CAPTURE
1529  ) ) {
1530  $trail = substr( $url, $m2[0][1] ) . $trail;
1531  $url = substr( $url, 0, $m2[0][1] );
1532  }
1533 
1534  # Move trailing punctuation to $trail
1535  $sep = ',;\.:!?';
1536  # If there is no left bracket, then consider right brackets fair game too
1537  if ( strpos( $url, '(' ) === false ) {
1538  $sep .= ')';
1539  }
1540 
1541  $urlRev = strrev( $url );
1542  $numSepChars = strspn( $urlRev, $sep );
1543  # Don't break a trailing HTML entity by moving the ; into $trail
1544  # This is in hot code, so use substr_compare to avoid having to
1545  # create a new string object for the comparison
1546  if ( $numSepChars && substr_compare( $url, ";", -$numSepChars, 1 ) === 0 ) {
1547  # more optimization: instead of running preg_match with a $
1548  # anchor, which can be slow, do the match on the reversed
1549  # string starting at the desired offset.
1550  # un-reversed regexp is: /&([a-z]+|#x[\da-f]+|#\d+)$/i
1551  if ( preg_match( '/\G([a-z]+|[\da-f]+x#|\d+#)&/i', $urlRev, $m2, 0, $numSepChars ) ) {
1552  $numSepChars--;
1553  }
1554  }
1555  if ( $numSepChars ) {
1556  $trail = substr( $url, -$numSepChars ) . $trail;
1557  $url = substr( $url, 0, -$numSepChars );
1558  }
1559 
1560  # Verify that we still have a real URL after trail removal, and
1561  # not just lone protocol
1562  if ( strlen( $trail ) >= $numPostProto ) {
1563  return $url . $trail;
1564  }
1565 
1566  $url = Sanitizer::cleanUrl( $url );
1567 
1568  # Is this an external image?
1569  $text = $this->maybeMakeExternalImage( $url );
1570  if ( $text === false ) {
1571  # Not an image, make a link
1572  $text = Linker::makeExternalLink( $url,
1573  $this->getConverterLanguage()->markNoConversion( $url, true ),
1574  true, 'free',
1575  $this->getExternalLinkAttribs( $url ), $this->mTitle );
1576  # Register it in the output object...
1577  $this->mOutput->addExternalLink( $url );
1578  }
1579  return $text . $trail;
1580  }
1581 
1591  public function doHeadings( $text ) {
1592  for ( $i = 6; $i >= 1; --$i ) {
1593  $h = str_repeat( '=', $i );
1594  $text = preg_replace( "/^$h(.+)$h\\s*$/m", "<h$i>\\1</h$i>", $text );
1595  }
1596  return $text;
1597  }
1598 
1607  public function doAllQuotes( $text ) {
1608  $outtext = '';
1609  $lines = StringUtils::explode( "\n", $text );
1610  foreach ( $lines as $line ) {
1611  $outtext .= $this->doQuotes( $line ) . "\n";
1612  }
1613  $outtext = substr( $outtext, 0, -1 );
1614  return $outtext;
1615  }
1616 
1624  public function doQuotes( $text ) {
1625  $arr = preg_split( "/(''+)/", $text, -1, PREG_SPLIT_DELIM_CAPTURE );
1626  $countarr = count( $arr );
1627  if ( $countarr == 1 ) {
1628  return $text;
1629  }
1630 
1631  // First, do some preliminary work. This may shift some apostrophes from
1632  // being mark-up to being text. It also counts the number of occurrences
1633  // of bold and italics mark-ups.
1634  $numbold = 0;
1635  $numitalics = 0;
1636  for ( $i = 1; $i < $countarr; $i += 2 ) {
1637  $thislen = strlen( $arr[$i] );
1638  // If there are ever four apostrophes, assume the first is supposed to
1639  // be text, and the remaining three constitute mark-up for bold text.
1640  // (bug 13227: ''''foo'''' turns into ' ''' foo ' ''')
1641  if ( $thislen == 4 ) {
1642  $arr[$i - 1] .= "'";
1643  $arr[$i] = "'''";
1644  $thislen = 3;
1645  } elseif ( $thislen > 5 ) {
1646  // If there are more than 5 apostrophes in a row, assume they're all
1647  // text except for the last 5.
1648  // (bug 13227: ''''''foo'''''' turns into ' ''''' foo ' ''''')
1649  $arr[$i - 1] .= str_repeat( "'", $thislen - 5 );
1650  $arr[$i] = "'''''";
1651  $thislen = 5;
1652  }
1653  // Count the number of occurrences of bold and italics mark-ups.
1654  if ( $thislen == 2 ) {
1655  $numitalics++;
1656  } elseif ( $thislen == 3 ) {
1657  $numbold++;
1658  } elseif ( $thislen == 5 ) {
1659  $numitalics++;
1660  $numbold++;
1661  }
1662  }
1663 
1664  // If there is an odd number of both bold and italics, it is likely
1665  // that one of the bold ones was meant to be an apostrophe followed
1666  // by italics. Which one we cannot know for certain, but it is more
1667  // likely to be one that has a single-letter word before it.
1668  if ( ( $numbold % 2 == 1 ) && ( $numitalics % 2 == 1 ) ) {
1669  $firstsingleletterword = -1;
1670  $firstmultiletterword = -1;
1671  $firstspace = -1;
1672  for ( $i = 1; $i < $countarr; $i += 2 ) {
1673  if ( strlen( $arr[$i] ) == 3 ) {
1674  $x1 = substr( $arr[$i - 1], -1 );
1675  $x2 = substr( $arr[$i - 1], -2, 1 );
1676  if ( $x1 === ' ' ) {
1677  if ( $firstspace == -1 ) {
1678  $firstspace = $i;
1679  }
1680  } elseif ( $x2 === ' ' ) {
1681  $firstsingleletterword = $i;
1682  // if $firstsingleletterword is set, we don't
1683  // look at the other options, so we can bail early.
1684  break;
1685  } else {
1686  if ( $firstmultiletterword == -1 ) {
1687  $firstmultiletterword = $i;
1688  }
1689  }
1690  }
1691  }
1692 
1693  // If there is a single-letter word, use it!
1694  if ( $firstsingleletterword > -1 ) {
1695  $arr[$firstsingleletterword] = "''";
1696  $arr[$firstsingleletterword - 1] .= "'";
1697  } elseif ( $firstmultiletterword > -1 ) {
1698  // If not, but there's a multi-letter word, use that one.
1699  $arr[$firstmultiletterword] = "''";
1700  $arr[$firstmultiletterword - 1] .= "'";
1701  } elseif ( $firstspace > -1 ) {
1702  // ... otherwise use the first one that has neither.
1703  // (notice that it is possible for all three to be -1 if, for example,
1704  // there is only one pentuple-apostrophe in the line)
1705  $arr[$firstspace] = "''";
1706  $arr[$firstspace - 1] .= "'";
1707  }
1708  }
1709 
1710  // Now let's actually convert our apostrophic mush to HTML!
1711  $output = '';
1712  $buffer = '';
1713  $state = '';
1714  $i = 0;
1715  foreach ( $arr as $r ) {
1716  if ( ( $i % 2 ) == 0 ) {
1717  if ( $state === 'both' ) {
1718  $buffer .= $r;
1719  } else {
1720  $output .= $r;
1721  }
1722  } else {
1723  $thislen = strlen( $r );
1724  if ( $thislen == 2 ) {
1725  if ( $state === 'i' ) {
1726  $output .= '</i>';
1727  $state = '';
1728  } elseif ( $state === 'bi' ) {
1729  $output .= '</i>';
1730  $state = 'b';
1731  } elseif ( $state === 'ib' ) {
1732  $output .= '</b></i><b>';
1733  $state = 'b';
1734  } elseif ( $state === 'both' ) {
1735  $output .= '<b><i>' . $buffer . '</i>';
1736  $state = 'b';
1737  } else { // $state can be 'b' or ''
1738  $output .= '<i>';
1739  $state .= 'i';
1740  }
1741  } elseif ( $thislen == 3 ) {
1742  if ( $state === 'b' ) {
1743  $output .= '</b>';
1744  $state = '';
1745  } elseif ( $state === 'bi' ) {
1746  $output .= '</i></b><i>';
1747  $state = 'i';
1748  } elseif ( $state === 'ib' ) {
1749  $output .= '</b>';
1750  $state = 'i';
1751  } elseif ( $state === 'both' ) {
1752  $output .= '<i><b>' . $buffer . '</b>';
1753  $state = 'i';
1754  } else { // $state can be 'i' or ''
1755  $output .= '<b>';
1756  $state .= 'b';
1757  }
1758  } elseif ( $thislen == 5 ) {
1759  if ( $state === 'b' ) {
1760  $output .= '</b><i>';
1761  $state = 'i';
1762  } elseif ( $state === 'i' ) {
1763  $output .= '</i><b>';
1764  $state = 'b';
1765  } elseif ( $state === 'bi' ) {
1766  $output .= '</i></b>';
1767  $state = '';
1768  } elseif ( $state === 'ib' ) {
1769  $output .= '</b></i>';
1770  $state = '';
1771  } elseif ( $state === 'both' ) {
1772  $output .= '<i><b>' . $buffer . '</b></i>';
1773  $state = '';
1774  } else { // ($state == '')
1775  $buffer = '';
1776  $state = 'both';
1777  }
1778  }
1779  }
1780  $i++;
1781  }
1782  // Now close all remaining tags. Notice that the order is important.
1783  if ( $state === 'b' || $state === 'ib' ) {
1784  $output .= '</b>';
1785  }
1786  if ( $state === 'i' || $state === 'bi' || $state === 'ib' ) {
1787  $output .= '</i>';
1788  }
1789  if ( $state === 'bi' ) {
1790  $output .= '</b>';
1791  }
1792  // There might be lonely ''''', so make sure we have a buffer
1793  if ( $state === 'both' && $buffer ) {
1794  $output .= '<b><i>' . $buffer . '</i></b>';
1795  }
1796  return $output;
1797  }
1798 
1812  public function replaceExternalLinks( $text ) {
1813 
1814  $bits = preg_split( $this->mExtLinkBracketedRegex, $text, -1, PREG_SPLIT_DELIM_CAPTURE );
1815  if ( $bits === false ) {
1816  throw new MWException( "PCRE needs to be compiled with "
1817  . "--enable-unicode-properties in order for MediaWiki to function" );
1818  }
1819  $s = array_shift( $bits );
1820 
1821  $i = 0;
1822  while ( $i < count( $bits ) ) {
1823  $url = $bits[$i++];
1824  $i++; // protocol
1825  $text = $bits[$i++];
1826  $trail = $bits[$i++];
1827 
1828  # The characters '<' and '>' (which were escaped by
1829  # removeHTMLtags()) should not be included in
1830  # URLs, per RFC 2396.
1831  $m2 = [];
1832  if ( preg_match( '/&(lt|gt);/', $url, $m2, PREG_OFFSET_CAPTURE ) ) {
1833  $text = substr( $url, $m2[0][1] ) . ' ' . $text;
1834  $url = substr( $url, 0, $m2[0][1] );
1835  }
1836 
1837  # If the link text is an image URL, replace it with an <img> tag
1838  # This happened by accident in the original parser, but some people used it extensively
1839  $img = $this->maybeMakeExternalImage( $text );
1840  if ( $img !== false ) {
1841  $text = $img;
1842  }
1843 
1844  $dtrail = '';
1845 
1846  # Set linktype for CSS - if URL==text, link is essentially free
1847  $linktype = ( $text === $url ) ? 'free' : 'text';
1848 
1849  # No link text, e.g. [http://domain.tld/some.link]
1850  if ( $text == '' ) {
1851  # Autonumber
1852  $langObj = $this->getTargetLanguage();
1853  $text = '[' . $langObj->formatNum( ++$this->mAutonumber ) . ']';
1854  $linktype = 'autonumber';
1855  } else {
1856  # Have link text, e.g. [http://domain.tld/some.link text]s
1857  # Check for trail
1858  list( $dtrail, $trail ) = Linker::splitTrail( $trail );
1859  }
1860 
1861  $text = $this->getConverterLanguage()->markNoConversion( $text );
1862 
1863  $url = Sanitizer::cleanUrl( $url );
1864 
1865  # Use the encoded URL
1866  # This means that users can paste URLs directly into the text
1867  # Funny characters like ö aren't valid in URLs anyway
1868  # This was changed in August 2004
1869  $s .= Linker::makeExternalLink( $url, $text, false, $linktype,
1870  $this->getExternalLinkAttribs( $url ), $this->mTitle ) . $dtrail . $trail;
1871 
1872  # Register link in the output object.
1873  $this->mOutput->addExternalLink( $url );
1874  }
1875 
1876  return $s;
1877  }
1878 
1888  public static function getExternalLinkRel( $url = false, $title = null ) {
1889  global $wgNoFollowLinks, $wgNoFollowNsExceptions, $wgNoFollowDomainExceptions;
1890  $ns = $title ? $title->getNamespace() : false;
1891  if ( $wgNoFollowLinks && !in_array( $ns, $wgNoFollowNsExceptions )
1892  && !wfMatchesDomainList( $url, $wgNoFollowDomainExceptions )
1893  ) {
1894  return 'nofollow';
1895  }
1896  return null;
1897  }
1898 
1909  public function getExternalLinkAttribs( $url ) {
1910  $attribs = [];
1911  $rel = self::getExternalLinkRel( $url, $this->mTitle );
1912 
1913  $target = $this->mOptions->getExternalLinkTarget();
1914  if ( $target ) {
1915  $attribs['target'] = $target;
1916  if ( !in_array( $target, [ '_self', '_parent', '_top' ] ) ) {
1917  // T133507. New windows can navigate parent cross-origin.
1918  // Including noreferrer due to lacking browser
1919  // support of noopener. Eventually noreferrer should be removed.
1920  if ( $rel !== '' ) {
1921  $rel .= ' ';
1922  }
1923  $rel .= 'noreferrer noopener';
1924  }
1925  }
1926  $attribs['rel'] = $rel;
1927  return $attribs;
1928  }
1929 
1937  public static function replaceUnusualEscapes( $url ) {
1938  wfDeprecated( __METHOD__, '1.24' );
1939  return self::normalizeLinkUrl( $url );
1940  }
1941 
1951  public static function normalizeLinkUrl( $url ) {
1952  # First, make sure unsafe characters are encoded
1953  $url = preg_replace_callback( '/[\x00-\x20"<>\[\\\\\]^`{|}\x7F-\xFF]/',
1954  function ( $m ) {
1955  return rawurlencode( $m[0] );
1956  },
1957  $url
1958  );
1959 
1960  $ret = '';
1961  $end = strlen( $url );
1962 
1963  # Fragment part - 'fragment'
1964  $start = strpos( $url, '#' );
1965  if ( $start !== false && $start < $end ) {
1966  $ret = self::normalizeUrlComponent(
1967  substr( $url, $start, $end - $start ), '"#%<>[\]^`{|}' ) . $ret;
1968  $end = $start;
1969  }
1970 
1971  # Query part - 'query' minus &=+;
1972  $start = strpos( $url, '?' );
1973  if ( $start !== false && $start < $end ) {
1974  $ret = self::normalizeUrlComponent(
1975  substr( $url, $start, $end - $start ), '"#%<>[\]^`{|}&=+;' ) . $ret;
1976  $end = $start;
1977  }
1978 
1979  # Scheme and path part - 'pchar'
1980  # (we assume no userinfo or encoded colons in the host)
1981  $ret = self::normalizeUrlComponent(
1982  substr( $url, 0, $end ), '"#%<>[\]^`{|}/?' ) . $ret;
1983 
1984  return $ret;
1985  }
1986 
1987  private static function normalizeUrlComponent( $component, $unsafe ) {
1988  $callback = function ( $matches ) use ( $unsafe ) {
1989  $char = urldecode( $matches[0] );
1990  $ord = ord( $char );
1991  if ( $ord > 32 && $ord < 127 && strpos( $unsafe, $char ) === false ) {
1992  # Unescape it
1993  return $char;
1994  } else {
1995  # Leave it escaped, but use uppercase for a-f
1996  return strtoupper( $matches[0] );
1997  }
1998  };
1999  return preg_replace_callback( '/%[0-9A-Fa-f]{2}/', $callback, $component );
2000  }
2001 
2010  private function maybeMakeExternalImage( $url ) {
2011  $imagesfrom = $this->mOptions->getAllowExternalImagesFrom();
2012  $imagesexception = !empty( $imagesfrom );
2013  $text = false;
2014  # $imagesfrom could be either a single string or an array of strings, parse out the latter
2015  if ( $imagesexception && is_array( $imagesfrom ) ) {
2016  $imagematch = false;
2017  foreach ( $imagesfrom as $match ) {
2018  if ( strpos( $url, $match ) === 0 ) {
2019  $imagematch = true;
2020  break;
2021  }
2022  }
2023  } elseif ( $imagesexception ) {
2024  $imagematch = ( strpos( $url, $imagesfrom ) === 0 );
2025  } else {
2026  $imagematch = false;
2027  }
2028 
2029  if ( $this->mOptions->getAllowExternalImages()
2030  || ( $imagesexception && $imagematch )
2031  ) {
2032  if ( preg_match( self::EXT_IMAGE_REGEX, $url ) ) {
2033  # Image found
2034  $text = Linker::makeExternalImage( $url );
2035  }
2036  }
2037  if ( !$text && $this->mOptions->getEnableImageWhitelist()
2038  && preg_match( self::EXT_IMAGE_REGEX, $url )
2039  ) {
2040  $whitelist = explode(
2041  "\n",
2042  wfMessage( 'external_image_whitelist' )->inContentLanguage()->text()
2043  );
2044 
2045  foreach ( $whitelist as $entry ) {
2046  # Sanitize the regex fragment, make it case-insensitive, ignore blank entries/comments
2047  if ( strpos( $entry, '#' ) === 0 || $entry === '' ) {
2048  continue;
2049  }
2050  if ( preg_match( '/' . str_replace( '/', '\\/', $entry ) . '/i', $url ) ) {
2051  # Image matches a whitelist entry
2052  $text = Linker::makeExternalImage( $url );
2053  break;
2054  }
2055  }
2056  }
2057  return $text;
2058  }
2059 
2069  public function replaceInternalLinks( $s ) {
2070  $this->mLinkHolders->merge( $this->replaceInternalLinks2( $s ) );
2071  return $s;
2072  }
2073 
2082  public function replaceInternalLinks2( &$s ) {
2084 
2085  static $tc = false, $e1, $e1_img;
2086  # the % is needed to support urlencoded titles as well
2087  if ( !$tc ) {
2088  $tc = Title::legalChars() . '#%';
2089  # Match a link having the form [[namespace:link|alternate]]trail
2090  $e1 = "/^([{$tc}]+)(?:\\|(.+?))?]](.*)\$/sD";
2091  # Match cases where there is no "]]", which might still be images
2092  $e1_img = "/^([{$tc}]+)\\|(.*)\$/sD";
2093  }
2094 
2095  $holders = new LinkHolderArray( $this );
2096 
2097  # split the entire text string on occurrences of [[
2098  $a = StringUtils::explode( '[[', ' ' . $s );
2099  # get the first element (all text up to first [[), and remove the space we added
2100  $s = $a->current();
2101  $a->next();
2102  $line = $a->current(); # Workaround for broken ArrayIterator::next() that returns "void"
2103  $s = substr( $s, 1 );
2104 
2105  $useLinkPrefixExtension = $this->getTargetLanguage()->linkPrefixExtension();
2106  $e2 = null;
2107  if ( $useLinkPrefixExtension ) {
2108  # Match the end of a line for a word that's not followed by whitespace,
2109  # e.g. in the case of 'The Arab al[[Razi]]', 'al' will be matched
2111  $charset = $wgContLang->linkPrefixCharset();
2112  $e2 = "/^((?>.*[^$charset]|))(.+)$/sDu";
2113  }
2114 
2115  if ( is_null( $this->mTitle ) ) {
2116  throw new MWException( __METHOD__ . ": \$this->mTitle is null\n" );
2117  }
2118  $nottalk = !$this->mTitle->isTalkPage();
2119 
2120  if ( $useLinkPrefixExtension ) {
2121  $m = [];
2122  if ( preg_match( $e2, $s, $m ) ) {
2123  $first_prefix = $m[2];
2124  } else {
2125  $first_prefix = false;
2126  }
2127  } else {
2128  $prefix = '';
2129  }
2130 
2131  $useSubpages = $this->areSubpagesAllowed();
2132 
2133  // @codingStandardsIgnoreStart Squiz.WhiteSpace.SemicolonSpacing.Incorrect
2134  # Loop for each link
2135  for ( ; $line !== false && $line !== null; $a->next(), $line = $a->current() ) {
2136  // @codingStandardsIgnoreEnd
2137 
2138  # Check for excessive memory usage
2139  if ( $holders->isBig() ) {
2140  # Too big
2141  # Do the existence check, replace the link holders and clear the array
2142  $holders->replace( $s );
2143  $holders->clear();
2144  }
2145 
2146  if ( $useLinkPrefixExtension ) {
2147  if ( preg_match( $e2, $s, $m ) ) {
2148  $prefix = $m[2];
2149  $s = $m[1];
2150  } else {
2151  $prefix = '';
2152  }
2153  # first link
2154  if ( $first_prefix ) {
2155  $prefix = $first_prefix;
2156  $first_prefix = false;
2157  }
2158  }
2159 
2160  $might_be_img = false;
2161 
2162  if ( preg_match( $e1, $line, $m ) ) { # page with normal text or alt
2163  $text = $m[2];
2164  # If we get a ] at the beginning of $m[3] that means we have a link that's something like:
2165  # [[Image:Foo.jpg|[http://example.com desc]]] <- having three ] in a row fucks up,
2166  # the real problem is with the $e1 regex
2167  # See bug 1300.
2168  # Still some problems for cases where the ] is meant to be outside punctuation,
2169  # and no image is in sight. See bug 2095.
2170  if ( $text !== ''
2171  && substr( $m[3], 0, 1 ) === ']'
2172  && strpos( $text, '[' ) !== false
2173  ) {
2174  $text .= ']'; # so that replaceExternalLinks($text) works later
2175  $m[3] = substr( $m[3], 1 );
2176  }
2177  # fix up urlencoded title texts
2178  if ( strpos( $m[1], '%' ) !== false ) {
2179  # Should anchors '#' also be rejected?
2180  $m[1] = str_replace( [ '<', '>' ], [ '&lt;', '&gt;' ], rawurldecode( $m[1] ) );
2181  }
2182  $trail = $m[3];
2183  } elseif ( preg_match( $e1_img, $line, $m ) ) {
2184  # Invalid, but might be an image with a link in its caption
2185  $might_be_img = true;
2186  $text = $m[2];
2187  if ( strpos( $m[1], '%' ) !== false ) {
2188  $m[1] = str_replace( [ '<', '>' ], [ '&lt;', '&gt;' ], rawurldecode( $m[1] ) );
2189  }
2190  $trail = "";
2191  } else { # Invalid form; output directly
2192  $s .= $prefix . '[[' . $line;
2193  continue;
2194  }
2195 
2196  $origLink = $m[1];
2197 
2198  # Don't allow internal links to pages containing
2199  # PROTO: where PROTO is a valid URL protocol; these
2200  # should be external links.
2201  if ( preg_match( '/^(?i:' . $this->mUrlProtocols . ')/', $origLink ) ) {
2202  $s .= $prefix . '[[' . $line;
2203  continue;
2204  }
2205 
2206  # Make subpage if necessary
2207  if ( $useSubpages ) {
2208  $link = $this->maybeDoSubpageLink( $origLink, $text );
2209  } else {
2210  $link = $origLink;
2211  }
2212 
2213  $noforce = ( substr( $origLink, 0, 1 ) !== ':' );
2214  if ( !$noforce ) {
2215  # Strip off leading ':'
2216  $link = substr( $link, 1 );
2217  }
2218 
2219  $unstrip = $this->mStripState->unstripNoWiki( $link );
2220  $nt = is_string( $unstrip ) ? Title::newFromText( $unstrip ) : null;
2221  if ( $nt === null ) {
2222  $s .= $prefix . '[[' . $line;
2223  continue;
2224  }
2225 
2226  $ns = $nt->getNamespace();
2227  $iw = $nt->getInterwiki();
2228 
2229  if ( $might_be_img ) { # if this is actually an invalid link
2230  if ( $ns == NS_FILE && $noforce ) { # but might be an image
2231  $found = false;
2232  while ( true ) {
2233  # look at the next 'line' to see if we can close it there
2234  $a->next();
2235  $next_line = $a->current();
2236  if ( $next_line === false || $next_line === null ) {
2237  break;
2238  }
2239  $m = explode( ']]', $next_line, 3 );
2240  if ( count( $m ) == 3 ) {
2241  # the first ]] closes the inner link, the second the image
2242  $found = true;
2243  $text .= "[[{$m[0]}]]{$m[1]}";
2244  $trail = $m[2];
2245  break;
2246  } elseif ( count( $m ) == 2 ) {
2247  # if there's exactly one ]] that's fine, we'll keep looking
2248  $text .= "[[{$m[0]}]]{$m[1]}";
2249  } else {
2250  # if $next_line is invalid too, we need look no further
2251  $text .= '[[' . $next_line;
2252  break;
2253  }
2254  }
2255  if ( !$found ) {
2256  # we couldn't find the end of this imageLink, so output it raw
2257  # but don't ignore what might be perfectly normal links in the text we've examined
2258  $holders->merge( $this->replaceInternalLinks2( $text ) );
2259  $s .= "{$prefix}[[$link|$text";
2260  # note: no $trail, because without an end, there *is* no trail
2261  continue;
2262  }
2263  } else { # it's not an image, so output it raw
2264  $s .= "{$prefix}[[$link|$text";
2265  # note: no $trail, because without an end, there *is* no trail
2266  continue;
2267  }
2268  }
2269 
2270  $wasblank = ( $text == '' );
2271  if ( $wasblank ) {
2272  $text = $link;
2273  } else {
2274  # Bug 4598 madness. Handle the quotes only if they come from the alternate part
2275  # [[Lista d''e paise d''o munno]] -> <a href="...">Lista d''e paise d''o munno</a>
2276  # [[Criticism of Harry Potter|Criticism of ''Harry Potter'']]
2277  # -> <a href="Criticism of Harry Potter">Criticism of <i>Harry Potter</i></a>
2278  $text = $this->doQuotes( $text );
2279  }
2280 
2281  # Link not escaped by : , create the various objects
2282  if ( $noforce && !$nt->wasLocalInterwiki() ) {
2283  # Interwikis
2284  if (
2285  $iw && $this->mOptions->getInterwikiMagic() && $nottalk && (
2286  Language::fetchLanguageName( $iw, null, 'mw' ) ||
2287  in_array( $iw, $wgExtraInterlanguageLinkPrefixes )
2288  )
2289  ) {
2290  # Bug 24502: filter duplicates
2291  if ( !isset( $this->mLangLinkLanguages[$iw] ) ) {
2292  $this->mLangLinkLanguages[$iw] = true;
2293  $this->mOutput->addLanguageLink( $nt->getFullText() );
2294  }
2295 
2296  $s = rtrim( $s . $prefix );
2297  $s .= trim( $trail, "\n" ) == '' ? '': $prefix . $trail;
2298  continue;
2299  }
2300 
2301  if ( $ns == NS_FILE ) {
2302  if ( !wfIsBadImage( $nt->getDBkey(), $this->mTitle ) ) {
2303  if ( $wasblank ) {
2304  # if no parameters were passed, $text
2305  # becomes something like "File:Foo.png",
2306  # which we don't want to pass on to the
2307  # image generator
2308  $text = '';
2309  } else {
2310  # recursively parse links inside the image caption
2311  # actually, this will parse them in any other parameters, too,
2312  # but it might be hard to fix that, and it doesn't matter ATM
2313  $text = $this->replaceExternalLinks( $text );
2314  $holders->merge( $this->replaceInternalLinks2( $text ) );
2315  }
2316  # cloak any absolute URLs inside the image markup, so replaceExternalLinks() won't touch them
2317  $s .= $prefix . $this->armorLinks(
2318  $this->makeImage( $nt, $text, $holders ) ) . $trail;
2319  continue;
2320  }
2321  } elseif ( $ns == NS_CATEGORY ) {
2322  $s = rtrim( $s . "\n" ); # bug 87
2323 
2324  if ( $wasblank ) {
2325  $sortkey = $this->getDefaultSort();
2326  } else {
2327  $sortkey = $text;
2328  }
2329  $sortkey = Sanitizer::decodeCharReferences( $sortkey );
2330  $sortkey = str_replace( "\n", '', $sortkey );
2331  $sortkey = $this->getConverterLanguage()->convertCategoryKey( $sortkey );
2332  $this->mOutput->addCategory( $nt->getDBkey(), $sortkey );
2333 
2337  $s .= trim( $prefix . $trail, "\n" ) == '' ? '' : $prefix . $trail;
2338 
2339  continue;
2340  }
2341  }
2342 
2343  # Self-link checking. For some languages, variants of the title are checked in
2344  # LinkHolderArray::doVariants() to allow batching the existence checks necessary
2345  # for linking to a different variant.
2346  if ( $ns != NS_SPECIAL && $nt->equals( $this->mTitle ) && !$nt->hasFragment() ) {
2347  $s .= $prefix . Linker::makeSelfLinkObj( $nt, $text, '', $trail );
2348  continue;
2349  }
2350 
2351  # NS_MEDIA is a pseudo-namespace for linking directly to a file
2352  # @todo FIXME: Should do batch file existence checks, see comment below
2353  if ( $ns == NS_MEDIA ) {
2354  # Give extensions a chance to select the file revision for us
2355  $options = [];
2356  $descQuery = false;
2357  Hooks::run( 'BeforeParserFetchFileAndTitle',
2358  [ $this, $nt, &$options, &$descQuery ] );
2359  # Fetch and register the file (file title may be different via hooks)
2360  list( $file, $nt ) = $this->fetchFileAndTitle( $nt, $options );
2361  # Cloak with NOPARSE to avoid replacement in replaceExternalLinks
2362  $s .= $prefix . $this->armorLinks(
2363  Linker::makeMediaLinkFile( $nt, $file, $text ) ) . $trail;
2364  continue;
2365  }
2366 
2367  # Some titles, such as valid special pages or files in foreign repos, should
2368  # be shown as bluelinks even though they're not included in the page table
2369  # @todo FIXME: isAlwaysKnown() can be expensive for file links; we should really do
2370  # batch file existence checks for NS_FILE and NS_MEDIA
2371  if ( $iw == '' && $nt->isAlwaysKnown() ) {
2372  $this->mOutput->addLink( $nt );
2373  $s .= $this->makeKnownLinkHolder( $nt, $text, $trail, $prefix );
2374  } else {
2375  # Links will be added to the output link list after checking
2376  $s .= $holders->makeHolder( $nt, $text, [], $trail, $prefix );
2377  }
2378  }
2379  return $holders;
2380  }
2381 
2395  protected function makeKnownLinkHolder( $nt, $text = '', $trail = '', $prefix = '' ) {
2396  list( $inside, $trail ) = Linker::splitTrail( $trail );
2397 
2398  if ( $text == '' ) {
2399  $text = htmlspecialchars( $nt->getPrefixedText() );
2400  }
2401 
2402  $link = $this->getLinkRenderer()->makeKnownLink(
2403  $nt, new HtmlArmor( "$prefix$text$inside" )
2404  );
2405 
2406  return $this->armorLinks( $link ) . $trail;
2407  }
2408 
2419  public function armorLinks( $text ) {
2420  return preg_replace( '/\b((?i)' . $this->mUrlProtocols . ')/',
2421  self::MARKER_PREFIX . "NOPARSE$1", $text );
2422  }
2423 
2428  public function areSubpagesAllowed() {
2429  # Some namespaces don't allow subpages
2430  return MWNamespace::hasSubpages( $this->mTitle->getNamespace() );
2431  }
2432 
2441  public function maybeDoSubpageLink( $target, &$text ) {
2442  return Linker::normalizeSubpageLink( $this->mTitle, $target, $text );
2443  }
2444 
2453  public function doBlockLevels( $text, $linestart ) {
2454  return BlockLevelPass::doBlockLevels( $text, $linestart );
2455  }
2456 
2468  public function getVariableValue( $index, $frame = false ) {
2471 
2472  if ( is_null( $this->mTitle ) ) {
2473  // If no title set, bad things are going to happen
2474  // later. Title should always be set since this
2475  // should only be called in the middle of a parse
2476  // operation (but the unit-tests do funky stuff)
2477  throw new MWException( __METHOD__ . ' Should only be '
2478  . ' called while parsing (no title set)' );
2479  }
2480 
2485  if ( Hooks::run( 'ParserGetVariableValueVarCache', [ &$this, &$this->mVarCache ] ) ) {
2486  if ( isset( $this->mVarCache[$index] ) ) {
2487  return $this->mVarCache[$index];
2488  }
2489  }
2490 
2491  $ts = wfTimestamp( TS_UNIX, $this->mOptions->getTimestamp() );
2492  Hooks::run( 'ParserGetVariableValueTs', [ &$this, &$ts ] );
2493 
2494  $pageLang = $this->getFunctionLang();
2495 
2496  switch ( $index ) {
2497  case '!':
2498  $value = '|';
2499  break;
2500  case 'currentmonth':
2501  $value = $pageLang->formatNum( MWTimestamp::getInstance( $ts )->format( 'm' ) );
2502  break;
2503  case 'currentmonth1':
2504  $value = $pageLang->formatNum( MWTimestamp::getInstance( $ts )->format( 'n' ) );
2505  break;
2506  case 'currentmonthname':
2507  $value = $pageLang->getMonthName( MWTimestamp::getInstance( $ts )->format( 'n' ) );
2508  break;
2509  case 'currentmonthnamegen':
2510  $value = $pageLang->getMonthNameGen( MWTimestamp::getInstance( $ts )->format( 'n' ) );
2511  break;
2512  case 'currentmonthabbrev':
2513  $value = $pageLang->getMonthAbbreviation( MWTimestamp::getInstance( $ts )->format( 'n' ) );
2514  break;
2515  case 'currentday':
2516  $value = $pageLang->formatNum( MWTimestamp::getInstance( $ts )->format( 'j' ) );
2517  break;
2518  case 'currentday2':
2519  $value = $pageLang->formatNum( MWTimestamp::getInstance( $ts )->format( 'd' ) );
2520  break;
2521  case 'localmonth':
2522  $value = $pageLang->formatNum( MWTimestamp::getLocalInstance( $ts )->format( 'm' ) );
2523  break;
2524  case 'localmonth1':
2525  $value = $pageLang->formatNum( MWTimestamp::getLocalInstance( $ts )->format( 'n' ) );
2526  break;
2527  case 'localmonthname':
2528  $value = $pageLang->getMonthName( MWTimestamp::getLocalInstance( $ts )->format( 'n' ) );
2529  break;
2530  case 'localmonthnamegen':
2531  $value = $pageLang->getMonthNameGen( MWTimestamp::getLocalInstance( $ts )->format( 'n' ) );
2532  break;
2533  case 'localmonthabbrev':
2534  $value = $pageLang->getMonthAbbreviation( MWTimestamp::getLocalInstance( $ts )->format( 'n' ) );
2535  break;
2536  case 'localday':
2537  $value = $pageLang->formatNum( MWTimestamp::getLocalInstance( $ts )->format( 'j' ) );
2538  break;
2539  case 'localday2':
2540  $value = $pageLang->formatNum( MWTimestamp::getLocalInstance( $ts )->format( 'd' ) );
2541  break;
2542  case 'pagename':
2543  $value = wfEscapeWikiText( $this->mTitle->getText() );
2544  break;
2545  case 'pagenamee':
2546  $value = wfEscapeWikiText( $this->mTitle->getPartialURL() );
2547  break;
2548  case 'fullpagename':
2549  $value = wfEscapeWikiText( $this->mTitle->getPrefixedText() );
2550  break;
2551  case 'fullpagenamee':
2552  $value = wfEscapeWikiText( $this->mTitle->getPrefixedURL() );
2553  break;
2554  case 'subpagename':
2555  $value = wfEscapeWikiText( $this->mTitle->getSubpageText() );
2556  break;
2557  case 'subpagenamee':
2558  $value = wfEscapeWikiText( $this->mTitle->getSubpageUrlForm() );
2559  break;
2560  case 'rootpagename':
2561  $value = wfEscapeWikiText( $this->mTitle->getRootText() );
2562  break;
2563  case 'rootpagenamee':
2564  $value = wfEscapeWikiText( wfUrlencode( str_replace(
2565  ' ',
2566  '_',
2567  $this->mTitle->getRootText()
2568  ) ) );
2569  break;
2570  case 'basepagename':
2571  $value = wfEscapeWikiText( $this->mTitle->getBaseText() );
2572  break;
2573  case 'basepagenamee':
2574  $value = wfEscapeWikiText( wfUrlencode( str_replace(
2575  ' ',
2576  '_',
2577  $this->mTitle->getBaseText()
2578  ) ) );
2579  break;
2580  case 'talkpagename':
2581  if ( $this->mTitle->canTalk() ) {
2582  $talkPage = $this->mTitle->getTalkPage();
2583  $value = wfEscapeWikiText( $talkPage->getPrefixedText() );
2584  } else {
2585  $value = '';
2586  }
2587  break;
2588  case 'talkpagenamee':
2589  if ( $this->mTitle->canTalk() ) {
2590  $talkPage = $this->mTitle->getTalkPage();
2591  $value = wfEscapeWikiText( $talkPage->getPrefixedURL() );
2592  } else {
2593  $value = '';
2594  }
2595  break;
2596  case 'subjectpagename':
2597  $subjPage = $this->mTitle->getSubjectPage();
2598  $value = wfEscapeWikiText( $subjPage->getPrefixedText() );
2599  break;
2600  case 'subjectpagenamee':
2601  $subjPage = $this->mTitle->getSubjectPage();
2602  $value = wfEscapeWikiText( $subjPage->getPrefixedURL() );
2603  break;
2604  case 'pageid': // requested in bug 23427
2605  $pageid = $this->getTitle()->getArticleID();
2606  if ( $pageid == 0 ) {
2607  # 0 means the page doesn't exist in the database,
2608  # which means the user is previewing a new page.
2609  # The vary-revision flag must be set, because the magic word
2610  # will have a different value once the page is saved.
2611  $this->mOutput->setFlag( 'vary-revision' );
2612  wfDebug( __METHOD__ . ": {{PAGEID}} used in a new page, setting vary-revision...\n" );
2613  }
2614  $value = $pageid ? $pageid : null;
2615  break;
2616  case 'revisionid':
2617  # Let the edit saving system know we should parse the page
2618  # *after* a revision ID has been assigned.
2619  $this->mOutput->setFlag( 'vary-revision-id' );
2620  wfDebug( __METHOD__ . ": {{REVISIONID}} used, setting vary-revision-id...\n" );
2621  $value = $this->mRevisionId;
2622  if ( !$value && $this->mOptions->getSpeculativeRevIdCallback() ) {
2623  $value = call_user_func( $this->mOptions->getSpeculativeRevIdCallback() );
2624  $this->mOutput->setSpeculativeRevIdUsed( $value );
2625  }
2626  break;
2627  case 'revisionday':
2628  # Let the edit saving system know we should parse the page
2629  # *after* a revision ID has been assigned. This is for null edits.
2630  $this->mOutput->setFlag( 'vary-revision' );
2631  wfDebug( __METHOD__ . ": {{REVISIONDAY}} used, setting vary-revision...\n" );
2632  $value = intval( substr( $this->getRevisionTimestamp(), 6, 2 ) );
2633  break;
2634  case 'revisionday2':
2635  # Let the edit saving system know we should parse the page
2636  # *after* a revision ID has been assigned. This is for null edits.
2637  $this->mOutput->setFlag( 'vary-revision' );
2638  wfDebug( __METHOD__ . ": {{REVISIONDAY2}} used, setting vary-revision...\n" );
2639  $value = substr( $this->getRevisionTimestamp(), 6, 2 );
2640  break;
2641  case 'revisionmonth':
2642  # Let the edit saving system know we should parse the page
2643  # *after* a revision ID has been assigned. This is for null edits.
2644  $this->mOutput->setFlag( 'vary-revision' );
2645  wfDebug( __METHOD__ . ": {{REVISIONMONTH}} used, setting vary-revision...\n" );
2646  $value = substr( $this->getRevisionTimestamp(), 4, 2 );
2647  break;
2648  case 'revisionmonth1':
2649  # Let the edit saving system know we should parse the page
2650  # *after* a revision ID has been assigned. This is for null edits.
2651  $this->mOutput->setFlag( 'vary-revision' );
2652  wfDebug( __METHOD__ . ": {{REVISIONMONTH1}} used, setting vary-revision...\n" );
2653  $value = intval( substr( $this->getRevisionTimestamp(), 4, 2 ) );
2654  break;
2655  case 'revisionyear':
2656  # Let the edit saving system know we should parse the page
2657  # *after* a revision ID has been assigned. This is for null edits.
2658  $this->mOutput->setFlag( 'vary-revision' );
2659  wfDebug( __METHOD__ . ": {{REVISIONYEAR}} used, setting vary-revision...\n" );
2660  $value = substr( $this->getRevisionTimestamp(), 0, 4 );
2661  break;
2662  case 'revisiontimestamp':
2663  # Let the edit saving system know we should parse the page
2664  # *after* a revision ID has been assigned. This is for null edits.
2665  $this->mOutput->setFlag( 'vary-revision' );
2666  wfDebug( __METHOD__ . ": {{REVISIONTIMESTAMP}} used, setting vary-revision...\n" );
2667  $value = $this->getRevisionTimestamp();
2668  break;
2669  case 'revisionuser':
2670  # Let the edit saving system know we should parse the page
2671  # *after* a revision ID has been assigned for null edits.
2672  $this->mOutput->setFlag( 'vary-user' );
2673  wfDebug( __METHOD__ . ": {{REVISIONUSER}} used, setting vary-user...\n" );
2674  $value = $this->getRevisionUser();
2675  break;
2676  case 'revisionsize':
2677  $value = $this->getRevisionSize();
2678  break;
2679  case 'namespace':
2680  $value = str_replace( '_', ' ', $wgContLang->getNsText( $this->mTitle->getNamespace() ) );
2681  break;
2682  case 'namespacee':
2683  $value = wfUrlencode( $wgContLang->getNsText( $this->mTitle->getNamespace() ) );
2684  break;
2685  case 'namespacenumber':
2686  $value = $this->mTitle->getNamespace();
2687  break;
2688  case 'talkspace':
2689  $value = $this->mTitle->canTalk()
2690  ? str_replace( '_', ' ', $this->mTitle->getTalkNsText() )
2691  : '';
2692  break;
2693  case 'talkspacee':
2694  $value = $this->mTitle->canTalk() ? wfUrlencode( $this->mTitle->getTalkNsText() ) : '';
2695  break;
2696  case 'subjectspace':
2697  $value = str_replace( '_', ' ', $this->mTitle->getSubjectNsText() );
2698  break;
2699  case 'subjectspacee':
2700  $value = ( wfUrlencode( $this->mTitle->getSubjectNsText() ) );
2701  break;
2702  case 'currentdayname':
2703  $value = $pageLang->getWeekdayName( (int)MWTimestamp::getInstance( $ts )->format( 'w' ) + 1 );
2704  break;
2705  case 'currentyear':
2706  $value = $pageLang->formatNum( MWTimestamp::getInstance( $ts )->format( 'Y' ), true );
2707  break;
2708  case 'currenttime':
2709  $value = $pageLang->time( wfTimestamp( TS_MW, $ts ), false, false );
2710  break;
2711  case 'currenthour':
2712  $value = $pageLang->formatNum( MWTimestamp::getInstance( $ts )->format( 'H' ), true );
2713  break;
2714  case 'currentweek':
2715  # @bug 4594 PHP5 has it zero padded, PHP4 does not, cast to
2716  # int to remove the padding
2717  $value = $pageLang->formatNum( (int)MWTimestamp::getInstance( $ts )->format( 'W' ) );
2718  break;
2719  case 'currentdow':
2720  $value = $pageLang->formatNum( MWTimestamp::getInstance( $ts )->format( 'w' ) );
2721  break;
2722  case 'localdayname':
2723  $value = $pageLang->getWeekdayName(
2724  (int)MWTimestamp::getLocalInstance( $ts )->format( 'w' ) + 1
2725  );
2726  break;
2727  case 'localyear':
2728  $value = $pageLang->formatNum( MWTimestamp::getLocalInstance( $ts )->format( 'Y' ), true );
2729  break;
2730  case 'localtime':
2731  $value = $pageLang->time(
2732  MWTimestamp::getLocalInstance( $ts )->format( 'YmdHis' ),
2733  false,
2734  false
2735  );
2736  break;
2737  case 'localhour':
2738  $value = $pageLang->formatNum( MWTimestamp::getLocalInstance( $ts )->format( 'H' ), true );
2739  break;
2740  case 'localweek':
2741  # @bug 4594 PHP5 has it zero padded, PHP4 does not, cast to
2742  # int to remove the padding
2743  $value = $pageLang->formatNum( (int)MWTimestamp::getLocalInstance( $ts )->format( 'W' ) );
2744  break;
2745  case 'localdow':
2746  $value = $pageLang->formatNum( MWTimestamp::getLocalInstance( $ts )->format( 'w' ) );
2747  break;
2748  case 'numberofarticles':
2749  $value = $pageLang->formatNum( SiteStats::articles() );
2750  break;
2751  case 'numberoffiles':
2752  $value = $pageLang->formatNum( SiteStats::images() );
2753  break;
2754  case 'numberofusers':
2755  $value = $pageLang->formatNum( SiteStats::users() );
2756  break;
2757  case 'numberofactiveusers':
2758  $value = $pageLang->formatNum( SiteStats::activeUsers() );
2759  break;
2760  case 'numberofpages':
2761  $value = $pageLang->formatNum( SiteStats::pages() );
2762  break;
2763  case 'numberofadmins':
2764  $value = $pageLang->formatNum( SiteStats::numberingroup( 'sysop' ) );
2765  break;
2766  case 'numberofedits':
2767  $value = $pageLang->formatNum( SiteStats::edits() );
2768  break;
2769  case 'currenttimestamp':
2770  $value = wfTimestamp( TS_MW, $ts );
2771  break;
2772  case 'localtimestamp':
2773  $value = MWTimestamp::getLocalInstance( $ts )->format( 'YmdHis' );
2774  break;
2775  case 'currentversion':
2777  break;
2778  case 'articlepath':
2779  return $wgArticlePath;
2780  case 'sitename':
2781  return $wgSitename;
2782  case 'server':
2783  return $wgServer;
2784  case 'servername':
2785  return $wgServerName;
2786  case 'scriptpath':
2787  return $wgScriptPath;
2788  case 'stylepath':
2789  return $wgStylePath;
2790  case 'directionmark':
2791  return $pageLang->getDirMark();
2792  case 'contentlanguage':
2794  return $wgLanguageCode;
2795  case 'cascadingsources':
2797  break;
2798  default:
2799  $ret = null;
2800  Hooks::run(
2801  'ParserGetVariableValueSwitch',
2802  [ &$this, &$this->mVarCache, &$index, &$ret, &$frame ]
2803  );
2804 
2805  return $ret;
2806  }
2807 
2808  if ( $index ) {
2809  $this->mVarCache[$index] = $value;
2810  }
2811 
2812  return $value;
2813  }
2814 
2820  public function initialiseVariables() {
2821  $variableIDs = MagicWord::getVariableIDs();
2822  $substIDs = MagicWord::getSubstIDs();
2823 
2824  $this->mVariables = new MagicWordArray( $variableIDs );
2825  $this->mSubstWords = new MagicWordArray( $substIDs );
2826  }
2827 
2850  public function preprocessToDom( $text, $flags = 0 ) {
2851  $dom = $this->getPreprocessor()->preprocessToObj( $text, $flags );
2852  return $dom;
2853  }
2854 
2862  public static function splitWhitespace( $s ) {
2863  $ltrimmed = ltrim( $s );
2864  $w1 = substr( $s, 0, strlen( $s ) - strlen( $ltrimmed ) );
2865  $trimmed = rtrim( $ltrimmed );
2866  $diff = strlen( $ltrimmed ) - strlen( $trimmed );
2867  if ( $diff > 0 ) {
2868  $w2 = substr( $ltrimmed, -$diff );
2869  } else {
2870  $w2 = '';
2871  }
2872  return [ $w1, $trimmed, $w2 ];
2873  }
2874 
2895  public function replaceVariables( $text, $frame = false, $argsOnly = false ) {
2896  # Is there any text? Also, Prevent too big inclusions!
2897  $textSize = strlen( $text );
2898  if ( $textSize < 1 || $textSize > $this->mOptions->getMaxIncludeSize() ) {
2899  return $text;
2900  }
2901 
2902  if ( $frame === false ) {
2903  $frame = $this->getPreprocessor()->newFrame();
2904  } elseif ( !( $frame instanceof PPFrame ) ) {
2905  wfDebug( __METHOD__ . " called using plain parameters instead of "
2906  . "a PPFrame instance. Creating custom frame.\n" );
2907  $frame = $this->getPreprocessor()->newCustomFrame( $frame );
2908  }
2909 
2910  $dom = $this->preprocessToDom( $text );
2911  $flags = $argsOnly ? PPFrame::NO_TEMPLATES : 0;
2912  $text = $frame->expand( $dom, $flags );
2913 
2914  return $text;
2915  }
2916 
2924  public static function createAssocArgs( $args ) {
2925  $assocArgs = [];
2926  $index = 1;
2927  foreach ( $args as $arg ) {
2928  $eqpos = strpos( $arg, '=' );
2929  if ( $eqpos === false ) {
2930  $assocArgs[$index++] = $arg;
2931  } else {
2932  $name = trim( substr( $arg, 0, $eqpos ) );
2933  $value = trim( substr( $arg, $eqpos + 1 ) );
2934  if ( $value === false ) {
2935  $value = '';
2936  }
2937  if ( $name !== false ) {
2938  $assocArgs[$name] = $value;
2939  }
2940  }
2941  }
2942 
2943  return $assocArgs;
2944  }
2945 
2972  public function limitationWarn( $limitationType, $current = '', $max = '' ) {
2973  # does no harm if $current and $max are present but are unnecessary for the message
2974  # Not doing ->inLanguage( $this->mOptions->getUserLangObj() ), since this is shown
2975  # only during preview, and that would split the parser cache unnecessarily.
2976  $warning = wfMessage( "$limitationType-warning" )->numParams( $current, $max )
2977  ->text();
2978  $this->mOutput->addWarning( $warning );
2979  $this->addTrackingCategory( "$limitationType-category" );
2980  }
2981 
2994  public function braceSubstitution( $piece, $frame ) {
2995 
2996  // Flags
2997 
2998  // $text has been filled
2999  $found = false;
3000  // wiki markup in $text should be escaped
3001  $nowiki = false;
3002  // $text is HTML, armour it against wikitext transformation
3003  $isHTML = false;
3004  // Force interwiki transclusion to be done in raw mode not rendered
3005  $forceRawInterwiki = false;
3006  // $text is a DOM node needing expansion in a child frame
3007  $isChildObj = false;
3008  // $text is a DOM node needing expansion in the current frame
3009  $isLocalObj = false;
3010 
3011  # Title object, where $text came from
3012  $title = false;
3013 
3014  # $part1 is the bit before the first |, and must contain only title characters.
3015  # Various prefixes will be stripped from it later.
3016  $titleWithSpaces = $frame->expand( $piece['title'] );
3017  $part1 = trim( $titleWithSpaces );
3018  $titleText = false;
3019 
3020  # Original title text preserved for various purposes
3021  $originalTitle = $part1;
3022 
3023  # $args is a list of argument nodes, starting from index 0, not including $part1
3024  # @todo FIXME: If piece['parts'] is null then the call to getLength()
3025  # below won't work b/c this $args isn't an object
3026  $args = ( null == $piece['parts'] ) ? [] : $piece['parts'];
3027 
3028  $profileSection = null; // profile templates
3029 
3030  # SUBST
3031  if ( !$found ) {
3032  $substMatch = $this->mSubstWords->matchStartAndRemove( $part1 );
3033 
3034  # Possibilities for substMatch: "subst", "safesubst" or FALSE
3035  # Decide whether to expand template or keep wikitext as-is.
3036  if ( $this->ot['wiki'] ) {
3037  if ( $substMatch === false ) {
3038  $literal = true; # literal when in PST with no prefix
3039  } else {
3040  $literal = false; # expand when in PST with subst: or safesubst:
3041  }
3042  } else {
3043  if ( $substMatch == 'subst' ) {
3044  $literal = true; # literal when not in PST with plain subst:
3045  } else {
3046  $literal = false; # expand when not in PST with safesubst: or no prefix
3047  }
3048  }
3049  if ( $literal ) {
3050  $text = $frame->virtualBracketedImplode( '{{', '|', '}}', $titleWithSpaces, $args );
3051  $isLocalObj = true;
3052  $found = true;
3053  }
3054  }
3055 
3056  # Variables
3057  if ( !$found && $args->getLength() == 0 ) {
3058  $id = $this->mVariables->matchStartToEnd( $part1 );
3059  if ( $id !== false ) {
3060  $text = $this->getVariableValue( $id, $frame );
3061  if ( MagicWord::getCacheTTL( $id ) > -1 ) {
3062  $this->mOutput->updateCacheExpiry( MagicWord::getCacheTTL( $id ) );
3063  }
3064  $found = true;
3065  }
3066  }
3067 
3068  # MSG, MSGNW and RAW
3069  if ( !$found ) {
3070  # Check for MSGNW:
3071  $mwMsgnw = MagicWord::get( 'msgnw' );
3072  if ( $mwMsgnw->matchStartAndRemove( $part1 ) ) {
3073  $nowiki = true;
3074  } else {
3075  # Remove obsolete MSG:
3076  $mwMsg = MagicWord::get( 'msg' );
3077  $mwMsg->matchStartAndRemove( $part1 );
3078  }
3079 
3080  # Check for RAW:
3081  $mwRaw = MagicWord::get( 'raw' );
3082  if ( $mwRaw->matchStartAndRemove( $part1 ) ) {
3083  $forceRawInterwiki = true;
3084  }
3085  }
3086 
3087  # Parser functions
3088  if ( !$found ) {
3089  $colonPos = strpos( $part1, ':' );
3090  if ( $colonPos !== false ) {
3091  $func = substr( $part1, 0, $colonPos );
3092  $funcArgs = [ trim( substr( $part1, $colonPos + 1 ) ) ];
3093  $argsLength = $args->getLength();
3094  for ( $i = 0; $i < $argsLength; $i++ ) {
3095  $funcArgs[] = $args->item( $i );
3096  }
3097  try {
3098  $result = $this->callParserFunction( $frame, $func, $funcArgs );
3099  } catch ( Exception $ex ) {
3100  throw $ex;
3101  }
3102 
3103  # The interface for parser functions allows for extracting
3104  # flags into the local scope. Extract any forwarded flags
3105  # here.
3106  extract( $result );
3107  }
3108  }
3109 
3110  # Finish mangling title and then check for loops.
3111  # Set $title to a Title object and $titleText to the PDBK
3112  if ( !$found ) {
3113  $ns = NS_TEMPLATE;
3114  # Split the title into page and subpage
3115  $subpage = '';
3116  $relative = $this->maybeDoSubpageLink( $part1, $subpage );
3117  if ( $part1 !== $relative ) {
3118  $part1 = $relative;
3119  $ns = $this->mTitle->getNamespace();
3120  }
3121  $title = Title::newFromText( $part1, $ns );
3122  if ( $title ) {
3123  $titleText = $title->getPrefixedText();
3124  # Check for language variants if the template is not found
3125  if ( $this->getConverterLanguage()->hasVariants() && $title->getArticleID() == 0 ) {
3126  $this->getConverterLanguage()->findVariantLink( $part1, $title, true );
3127  }
3128  # Do recursion depth check
3129  $limit = $this->mOptions->getMaxTemplateDepth();
3130  if ( $frame->depth >= $limit ) {
3131  $found = true;
3132  $text = '<span class="error">'
3133  . wfMessage( 'parser-template-recursion-depth-warning' )
3134  ->numParams( $limit )->inContentLanguage()->text()
3135  . '</span>';
3136  }
3137  }
3138  }
3139 
3140  # Load from database
3141  if ( !$found && $title ) {
3142  $profileSection = $this->mProfiler->scopedProfileIn( $title->getPrefixedDBkey() );
3143  if ( !$title->isExternal() ) {
3144  if ( $title->isSpecialPage()
3145  && $this->mOptions->getAllowSpecialInclusion()
3146  && $this->ot['html']
3147  ) {
3148  $specialPage = SpecialPageFactory::getPage( $title->getDBkey() );
3149  // Pass the template arguments as URL parameters.
3150  // "uselang" will have no effect since the Language object
3151  // is forced to the one defined in ParserOptions.
3152  $pageArgs = [];
3153  $argsLength = $args->getLength();
3154  for ( $i = 0; $i < $argsLength; $i++ ) {
3155  $bits = $args->item( $i )->splitArg();
3156  if ( strval( $bits['index'] ) === '' ) {
3157  $name = trim( $frame->expand( $bits['name'], PPFrame::STRIP_COMMENTS ) );
3158  $value = trim( $frame->expand( $bits['value'] ) );
3159  $pageArgs[$name] = $value;
3160  }
3161  }
3162 
3163  // Create a new context to execute the special page
3164  $context = new RequestContext;
3165  $context->setTitle( $title );
3166  $context->setRequest( new FauxRequest( $pageArgs ) );
3167  if ( $specialPage && $specialPage->maxIncludeCacheTime() === 0 ) {
3168  $context->setUser( $this->getUser() );
3169  } else {
3170  // If this page is cached, then we better not be per user.
3171  $context->setUser( User::newFromName( '127.0.0.1', false ) );
3172  }
3173  $context->setLanguage( $this->mOptions->getUserLangObj() );
3175  $title, $context, $this->getLinkRenderer() );
3176  if ( $ret ) {
3177  $text = $context->getOutput()->getHTML();
3178  $this->mOutput->addOutputPageMetadata( $context->getOutput() );
3179  $found = true;
3180  $isHTML = true;
3181  if ( $specialPage && $specialPage->maxIncludeCacheTime() !== false ) {
3182  $this->mOutput->updateRuntimeAdaptiveExpiry(
3183  $specialPage->maxIncludeCacheTime()
3184  );
3185  }
3186  }
3187  } elseif ( MWNamespace::isNonincludable( $title->getNamespace() ) ) {
3188  $found = false; # access denied
3189  wfDebug( __METHOD__ . ": template inclusion denied for " .
3190  $title->getPrefixedDBkey() . "\n" );
3191  } else {
3192  list( $text, $title ) = $this->getTemplateDom( $title );
3193  if ( $text !== false ) {
3194  $found = true;
3195  $isChildObj = true;
3196  }
3197  }
3198 
3199  # If the title is valid but undisplayable, make a link to it
3200  if ( !$found && ( $this->ot['html'] || $this->ot['pre'] ) ) {
3201  $text = "[[:$titleText]]";
3202  $found = true;
3203  }
3204  } elseif ( $title->isTrans() ) {
3205  # Interwiki transclusion
3206  if ( $this->ot['html'] && !$forceRawInterwiki ) {
3207  $text = $this->interwikiTransclude( $title, 'render' );
3208  $isHTML = true;
3209  } else {
3210  $text = $this->interwikiTransclude( $title, 'raw' );
3211  # Preprocess it like a template
3212  $text = $this->preprocessToDom( $text, self::PTD_FOR_INCLUSION );
3213  $isChildObj = true;
3214  }
3215  $found = true;
3216  }
3217 
3218  # Do infinite loop check
3219  # This has to be done after redirect resolution to avoid infinite loops via redirects
3220  if ( !$frame->loopCheck( $title ) ) {
3221  $found = true;
3222  $text = '<span class="error">'
3223  . wfMessage( 'parser-template-loop-warning', $titleText )->inContentLanguage()->text()
3224  . '</span>';
3225  wfDebug( __METHOD__ . ": template loop broken at '$titleText'\n" );
3226  }
3227  }
3228 
3229  # If we haven't found text to substitute by now, we're done
3230  # Recover the source wikitext and return it
3231  if ( !$found ) {
3232  $text = $frame->virtualBracketedImplode( '{{', '|', '}}', $titleWithSpaces, $args );
3233  if ( $profileSection ) {
3234  $this->mProfiler->scopedProfileOut( $profileSection );
3235  }
3236  return [ 'object' => $text ];
3237  }
3238 
3239  # Expand DOM-style return values in a child frame
3240  if ( $isChildObj ) {
3241  # Clean up argument array
3242  $newFrame = $frame->newChild( $args, $title );
3243 
3244  if ( $nowiki ) {
3245  $text = $newFrame->expand( $text, PPFrame::RECOVER_ORIG );
3246  } elseif ( $titleText !== false && $newFrame->isEmpty() ) {
3247  # Expansion is eligible for the empty-frame cache
3248  $text = $newFrame->cachedExpand( $titleText, $text );
3249  } else {
3250  # Uncached expansion
3251  $text = $newFrame->expand( $text );
3252  }
3253  }
3254  if ( $isLocalObj && $nowiki ) {
3255  $text = $frame->expand( $text, PPFrame::RECOVER_ORIG );
3256  $isLocalObj = false;
3257  }
3258 
3259  if ( $profileSection ) {
3260  $this->mProfiler->scopedProfileOut( $profileSection );
3261  }
3262 
3263  # Replace raw HTML by a placeholder
3264  if ( $isHTML ) {
3265  $text = $this->insertStripItem( $text );
3266  } elseif ( $nowiki && ( $this->ot['html'] || $this->ot['pre'] ) ) {
3267  # Escape nowiki-style return values
3268  $text = wfEscapeWikiText( $text );
3269  } elseif ( is_string( $text )
3270  && !$piece['lineStart']
3271  && preg_match( '/^(?:{\\||:|;|#|\*)/', $text )
3272  ) {
3273  # Bug 529: if the template begins with a table or block-level
3274  # element, it should be treated as beginning a new line.
3275  # This behavior is somewhat controversial.
3276  $text = "\n" . $text;
3277  }
3278 
3279  if ( is_string( $text ) && !$this->incrementIncludeSize( 'post-expand', strlen( $text ) ) ) {
3280  # Error, oversize inclusion
3281  if ( $titleText !== false ) {
3282  # Make a working, properly escaped link if possible (bug 23588)
3283  $text = "[[:$titleText]]";
3284  } else {
3285  # This will probably not be a working link, but at least it may
3286  # provide some hint of where the problem is
3287  preg_replace( '/^:/', '', $originalTitle );
3288  $text = "[[:$originalTitle]]";
3289  }
3290  $text .= $this->insertStripItem( '<!-- WARNING: template omitted, '
3291  . 'post-expand include size too large -->' );
3292  $this->limitationWarn( 'post-expand-template-inclusion' );
3293  }
3294 
3295  if ( $isLocalObj ) {
3296  $ret = [ 'object' => $text ];
3297  } else {
3298  $ret = [ 'text' => $text ];
3299  }
3300 
3301  return $ret;
3302  }
3303 
3323  public function callParserFunction( $frame, $function, array $args = [] ) {
3325 
3326  # Case sensitive functions
3327  if ( isset( $this->mFunctionSynonyms[1][$function] ) ) {
3328  $function = $this->mFunctionSynonyms[1][$function];
3329  } else {
3330  # Case insensitive functions
3331  $function = $wgContLang->lc( $function );
3332  if ( isset( $this->mFunctionSynonyms[0][$function] ) ) {
3333  $function = $this->mFunctionSynonyms[0][$function];
3334  } else {
3335  return [ 'found' => false ];
3336  }
3337  }
3338 
3339  list( $callback, $flags ) = $this->mFunctionHooks[$function];
3340 
3341  # Workaround for PHP bug 35229 and similar
3342  if ( !is_callable( $callback ) ) {
3343  throw new MWException( "Tag hook for $function is not callable\n" );
3344  }
3345 
3346  $allArgs = [ &$this ];
3347  if ( $flags & self::SFH_OBJECT_ARGS ) {
3348  # Convert arguments to PPNodes and collect for appending to $allArgs
3349  $funcArgs = [];
3350  foreach ( $args as $k => $v ) {
3351  if ( $v instanceof PPNode || $k === 0 ) {
3352  $funcArgs[] = $v;
3353  } else {
3354  $funcArgs[] = $this->mPreprocessor->newPartNodeArray( [ $k => $v ] )->item( 0 );
3355  }
3356  }
3357 
3358  # Add a frame parameter, and pass the arguments as an array
3359  $allArgs[] = $frame;
3360  $allArgs[] = $funcArgs;
3361  } else {
3362  # Convert arguments to plain text and append to $allArgs
3363  foreach ( $args as $k => $v ) {
3364  if ( $v instanceof PPNode ) {
3365  $allArgs[] = trim( $frame->expand( $v ) );
3366  } elseif ( is_int( $k ) && $k >= 0 ) {
3367  $allArgs[] = trim( $v );
3368  } else {
3369  $allArgs[] = trim( "$k=$v" );
3370  }
3371  }
3372  }
3373 
3374  $result = call_user_func_array( $callback, $allArgs );
3375 
3376  # The interface for function hooks allows them to return a wikitext
3377  # string or an array containing the string and any flags. This mungs
3378  # things around to match what this method should return.
3379  if ( !is_array( $result ) ) {
3380  $result =[
3381  'found' => true,
3382  'text' => $result,
3383  ];
3384  } else {
3385  if ( isset( $result[0] ) && !isset( $result['text'] ) ) {
3386  $result['text'] = $result[0];
3387  }
3388  unset( $result[0] );
3389  $result += [
3390  'found' => true,
3391  ];
3392  }
3393 
3394  $noparse = true;
3395  $preprocessFlags = 0;
3396  if ( isset( $result['noparse'] ) ) {
3397  $noparse = $result['noparse'];
3398  }
3399  if ( isset( $result['preprocessFlags'] ) ) {
3400  $preprocessFlags = $result['preprocessFlags'];
3401  }
3402 
3403  if ( !$noparse ) {
3404  $result['text'] = $this->preprocessToDom( $result['text'], $preprocessFlags );
3405  $result['isChildObj'] = true;
3406  }
3407 
3408  return $result;
3409  }
3410 
3419  public function getTemplateDom( $title ) {
3420  $cacheTitle = $title;
3421  $titleText = $title->getPrefixedDBkey();
3422 
3423  if ( isset( $this->mTplRedirCache[$titleText] ) ) {
3424  list( $ns, $dbk ) = $this->mTplRedirCache[$titleText];
3425  $title = Title::makeTitle( $ns, $dbk );
3426  $titleText = $title->getPrefixedDBkey();
3427  }
3428  if ( isset( $this->mTplDomCache[$titleText] ) ) {
3429  return [ $this->mTplDomCache[$titleText], $title ];
3430  }
3431 
3432  # Cache miss, go to the database
3433  list( $text, $title ) = $this->fetchTemplateAndTitle( $title );
3434 
3435  if ( $text === false ) {
3436  $this->mTplDomCache[$titleText] = false;
3437  return [ false, $title ];
3438  }
3439 
3440  $dom = $this->preprocessToDom( $text, self::PTD_FOR_INCLUSION );
3441  $this->mTplDomCache[$titleText] = $dom;
3442 
3443  if ( !$title->equals( $cacheTitle ) ) {
3444  $this->mTplRedirCache[$cacheTitle->getPrefixedDBkey()] =
3445  [ $title->getNamespace(), $cdb = $title->getDBkey() ];
3446  }
3447 
3448  return [ $dom, $title ];
3449  }
3450 
3463  $cacheKey = $title->getPrefixedDBkey();
3464  if ( !$this->currentRevisionCache ) {
3465  $this->currentRevisionCache = new MapCacheLRU( 100 );
3466  }
3467  if ( !$this->currentRevisionCache->has( $cacheKey ) ) {
3468  $this->currentRevisionCache->set( $cacheKey,
3469  // Defaults to Parser::statelessFetchRevision()
3470  call_user_func( $this->mOptions->getCurrentRevisionCallback(), $title, $this )
3471  );
3472  }
3473  return $this->currentRevisionCache->get( $cacheKey );
3474  }
3475 
3485  public static function statelessFetchRevision( Title $title, $parser = false ) {
3486  $pageId = $title->getArticleID();
3487  $revId = $title->getLatestRevID();
3488 
3490  if ( $rev ) {
3491  $rev->setTitle( $title );
3492  }
3493 
3494  return $rev;
3495  }
3496 
3502  public function fetchTemplateAndTitle( $title ) {
3503  // Defaults to Parser::statelessFetchTemplate()
3504  $templateCb = $this->mOptions->getTemplateCallback();
3505  $stuff = call_user_func( $templateCb, $title, $this );
3506  // We use U+007F DELETE to distinguish strip markers from regular text.
3507  $text = $stuff['text'];
3508  if ( is_string( $stuff['text'] ) ) {
3509  $text = strtr( $text, "\x7f", "?" );
3510  }
3511  $finalTitle = isset( $stuff['finalTitle'] ) ? $stuff['finalTitle'] : $title;
3512  if ( isset( $stuff['deps'] ) ) {
3513  foreach ( $stuff['deps'] as $dep ) {
3514  $this->mOutput->addTemplate( $dep['title'], $dep['page_id'], $dep['rev_id'] );
3515  if ( $dep['title']->equals( $this->getTitle() ) ) {
3516  // If we transclude ourselves, the final result
3517  // will change based on the new version of the page
3518  $this->mOutput->setFlag( 'vary-revision' );
3519  }
3520  }
3521  }
3522  return [ $text, $finalTitle ];
3523  }
3524 
3530  public function fetchTemplate( $title ) {
3531  return $this->fetchTemplateAndTitle( $title )[0];
3532  }
3533 
3543  public static function statelessFetchTemplate( $title, $parser = false ) {
3544  $text = $skip = false;
3545  $finalTitle = $title;
3546  $deps = [];
3547 
3548  # Loop to fetch the article, with up to 1 redirect
3549  // @codingStandardsIgnoreStart Generic.CodeAnalysis.ForLoopWithTestFunctionCall.NotAllowed
3550  for ( $i = 0; $i < 2 && is_object( $title ); $i++ ) {
3551  // @codingStandardsIgnoreEnd
3552  # Give extensions a chance to select the revision instead
3553  $id = false; # Assume current
3554  Hooks::run( 'BeforeParserFetchTemplateAndtitle',
3555  [ $parser, $title, &$skip, &$id ] );
3556 
3557  if ( $skip ) {
3558  $text = false;
3559  $deps[] = [
3560  'title' => $title,
3561  'page_id' => $title->getArticleID(),
3562  'rev_id' => null
3563  ];
3564  break;
3565  }
3566  # Get the revision
3567  if ( $id ) {
3568  $rev = Revision::newFromId( $id );
3569  } elseif ( $parser ) {
3570  $rev = $parser->fetchCurrentRevisionOfTitle( $title );
3571  } else {
3573  }
3574  $rev_id = $rev ? $rev->getId() : 0;
3575  # If there is no current revision, there is no page
3576  if ( $id === false && !$rev ) {
3577  $linkCache = LinkCache::singleton();
3578  $linkCache->addBadLinkObj( $title );
3579  }
3580 
3581  $deps[] = [
3582  'title' => $title,
3583  'page_id' => $title->getArticleID(),
3584  'rev_id' => $rev_id ];
3585  if ( $rev && !$title->equals( $rev->getTitle() ) ) {
3586  # We fetched a rev from a different title; register it too...
3587  $deps[] = [
3588  'title' => $rev->getTitle(),
3589  'page_id' => $rev->getPage(),
3590  'rev_id' => $rev_id ];
3591  }
3592 
3593  if ( $rev ) {
3594  $content = $rev->getContent();
3595  $text = $content ? $content->getWikitextForTransclusion() : null;
3596 
3597  if ( $text === false || $text === null ) {
3598  $text = false;
3599  break;
3600  }
3601  } elseif ( $title->getNamespace() == NS_MEDIAWIKI ) {
3603  $message = wfMessage( $wgContLang->lcfirst( $title->getText() ) )->inContentLanguage();
3604  if ( !$message->exists() ) {
3605  $text = false;
3606  break;
3607  }
3608  $content = $message->content();
3609  $text = $message->plain();
3610  } else {
3611  break;
3612  }
3613  if ( !$content ) {
3614  break;
3615  }
3616  # Redirect?
3617  $finalTitle = $title;
3618  $title = $content->getRedirectTarget();
3619  }
3620  return [
3621  'text' => $text,
3622  'finalTitle' => $finalTitle,
3623  'deps' => $deps ];
3624  }
3625 
3633  public function fetchFile( $title, $options = [] ) {
3634  return $this->fetchFileAndTitle( $title, $options )[0];
3635  }
3636 
3644  public function fetchFileAndTitle( $title, $options = [] ) {
3645  $file = $this->fetchFileNoRegister( $title, $options );
3646 
3647  $time = $file ? $file->getTimestamp() : false;
3648  $sha1 = $file ? $file->getSha1() : false;
3649  # Register the file as a dependency...
3650  $this->mOutput->addImage( $title->getDBkey(), $time, $sha1 );
3651  if ( $file && !$title->equals( $file->getTitle() ) ) {
3652  # Update fetched file title
3653  $title = $file->getTitle();
3654  $this->mOutput->addImage( $title->getDBkey(), $time, $sha1 );
3655  }
3656  return [ $file, $title ];
3657  }
3658 
3669  protected function fetchFileNoRegister( $title, $options = [] ) {
3670  if ( isset( $options['broken'] ) ) {
3671  $file = false; // broken thumbnail forced by hook
3672  } elseif ( isset( $options['sha1'] ) ) { // get by (sha1,timestamp)
3673  $file = RepoGroup::singleton()->findFileFromKey( $options['sha1'], $options );
3674  } else { // get by (name,timestamp)
3675  $file = wfFindFile( $title, $options );
3676  }
3677  return $file;
3678  }
3679 
3688  public function interwikiTransclude( $title, $action ) {
3689  global $wgEnableScaryTranscluding;
3690 
3691  if ( !$wgEnableScaryTranscluding ) {
3692  return wfMessage( 'scarytranscludedisabled' )->inContentLanguage()->text();
3693  }
3694 
3695  $url = $title->getFullURL( [ 'action' => $action ] );
3696 
3697  if ( strlen( $url ) > 255 ) {
3698  return wfMessage( 'scarytranscludetoolong' )->inContentLanguage()->text();
3699  }
3700  return $this->fetchScaryTemplateMaybeFromCache( $url );
3701  }
3702 
3707  public function fetchScaryTemplateMaybeFromCache( $url ) {
3708  global $wgTranscludeCacheExpiry;
3709  $dbr = wfGetDB( DB_REPLICA );
3710  $tsCond = $dbr->timestamp( time() - $wgTranscludeCacheExpiry );
3711  $obj = $dbr->selectRow( 'transcache', [ 'tc_time', 'tc_contents' ],
3712  [ 'tc_url' => $url, "tc_time >= " . $dbr->addQuotes( $tsCond ) ] );
3713  if ( $obj ) {
3714  return $obj->tc_contents;
3715  }
3716 
3717  $req = MWHttpRequest::factory( $url, [], __METHOD__ );
3718  $status = $req->execute(); // Status object
3719  if ( $status->isOK() ) {
3720  $text = $req->getContent();
3721  } elseif ( $req->getStatus() != 200 ) {
3722  // Though we failed to fetch the content, this status is useless.
3723  return wfMessage( 'scarytranscludefailed-httpstatus' )
3724  ->params( $url, $req->getStatus() /* HTTP status */ )->inContentLanguage()->text();
3725  } else {
3726  return wfMessage( 'scarytranscludefailed', $url )->inContentLanguage()->text();
3727  }
3728 
3729  $dbw = wfGetDB( DB_MASTER );
3730  $dbw->replace( 'transcache', [ 'tc_url' ], [
3731  'tc_url' => $url,
3732  'tc_time' => $dbw->timestamp( time() ),
3733  'tc_contents' => $text
3734  ] );
3735  return $text;
3736  }
3737 
3747  public function argSubstitution( $piece, $frame ) {
3748 
3749  $error = false;
3750  $parts = $piece['parts'];
3751  $nameWithSpaces = $frame->expand( $piece['title'] );
3752  $argName = trim( $nameWithSpaces );
3753  $object = false;
3754  $text = $frame->getArgument( $argName );
3755  if ( $text === false && $parts->getLength() > 0
3756  && ( $this->ot['html']
3757  || $this->ot['pre']
3758  || ( $this->ot['wiki'] && $frame->isTemplate() )
3759  )
3760  ) {
3761  # No match in frame, use the supplied default
3762  $object = $parts->item( 0 )->getChildren();
3763  }
3764  if ( !$this->incrementIncludeSize( 'arg', strlen( $text ) ) ) {
3765  $error = '<!-- WARNING: argument omitted, expansion size too large -->';
3766  $this->limitationWarn( 'post-expand-template-argument' );
3767  }
3768 
3769  if ( $text === false && $object === false ) {
3770  # No match anywhere
3771  $object = $frame->virtualBracketedImplode( '{{{', '|', '}}}', $nameWithSpaces, $parts );
3772  }
3773  if ( $error !== false ) {
3774  $text .= $error;
3775  }
3776  if ( $object !== false ) {
3777  $ret = [ 'object' => $object ];
3778  } else {
3779  $ret = [ 'text' => $text ];
3780  }
3781 
3782  return $ret;
3783  }
3784 
3800  public function extensionSubstitution( $params, $frame ) {
3801  static $errorStr = '<span class="error">';
3802  static $errorLen = 20;
3803 
3804  $name = $frame->expand( $params['name'] );
3805  if ( substr( $name, 0, $errorLen ) === $errorStr ) {
3806  // Probably expansion depth or node count exceeded. Just punt the
3807  // error up.
3808  return $name;
3809  }
3810 
3811  $attrText = !isset( $params['attr'] ) ? null : $frame->expand( $params['attr'] );
3812  if ( substr( $attrText, 0, $errorLen ) === $errorStr ) {
3813  // See above
3814  return $attrText;
3815  }
3816 
3817  $content = !isset( $params['inner'] ) ? null : $frame->expand( $params['inner'] );
3818  if ( substr( $content, 0, $errorLen ) === $errorStr ) {
3819  // See above
3820  return $content;
3821  }
3822 
3823  $marker = self::MARKER_PREFIX . "-$name-"
3824  . sprintf( '%08X', $this->mMarkerIndex++ ) . self::MARKER_SUFFIX;
3825 
3826  $isFunctionTag = isset( $this->mFunctionTagHooks[strtolower( $name )] ) &&
3827  ( $this->ot['html'] || $this->ot['pre'] );
3828  if ( $isFunctionTag ) {
3829  $markerType = 'none';
3830  } else {
3831  $markerType = 'general';
3832  }
3833  if ( $this->ot['html'] || $isFunctionTag ) {
3834  $name = strtolower( $name );
3835  $attributes = Sanitizer::decodeTagAttributes( $attrText );
3836  if ( isset( $params['attributes'] ) ) {
3837  $attributes = $attributes + $params['attributes'];
3838  }
3839 
3840  if ( isset( $this->mTagHooks[$name] ) ) {
3841  # Workaround for PHP bug 35229 and similar
3842  if ( !is_callable( $this->mTagHooks[$name] ) ) {
3843  throw new MWException( "Tag hook for $name is not callable\n" );
3844  }
3845  $output = call_user_func_array( $this->mTagHooks[$name],
3846  [ $content, $attributes, $this, $frame ] );
3847  } elseif ( isset( $this->mFunctionTagHooks[$name] ) ) {
3848  list( $callback, ) = $this->mFunctionTagHooks[$name];
3849  if ( !is_callable( $callback ) ) {
3850  throw new MWException( "Tag hook for $name is not callable\n" );
3851  }
3852 
3853  $output = call_user_func_array( $callback, [ &$this, $frame, $content, $attributes ] );
3854  } else {
3855  $output = '<span class="error">Invalid tag extension name: ' .
3856  htmlspecialchars( $name ) . '</span>';
3857  }
3858 
3859  if ( is_array( $output ) ) {
3860  # Extract flags to local scope (to override $markerType)
3861  $flags = $output;
3862  $output = $flags[0];
3863  unset( $flags[0] );
3864  extract( $flags );
3865  }
3866  } else {
3867  if ( is_null( $attrText ) ) {
3868  $attrText = '';
3869  }
3870  if ( isset( $params['attributes'] ) ) {
3871  foreach ( $params['attributes'] as $attrName => $attrValue ) {
3872  $attrText .= ' ' . htmlspecialchars( $attrName ) . '="' .
3873  htmlspecialchars( $attrValue ) . '"';
3874  }
3875  }
3876  if ( $content === null ) {
3877  $output = "<$name$attrText/>";
3878  } else {
3879  $close = is_null( $params['close'] ) ? '' : $frame->expand( $params['close'] );
3880  if ( substr( $close, 0, $errorLen ) === $errorStr ) {
3881  // See above
3882  return $close;
3883  }
3884  $output = "<$name$attrText>$content$close";
3885  }
3886  }
3887 
3888  if ( $markerType === 'none' ) {
3889  return $output;
3890  } elseif ( $markerType === 'nowiki' ) {
3891  $this->mStripState->addNoWiki( $marker, $output );
3892  } elseif ( $markerType === 'general' ) {
3893  $this->mStripState->addGeneral( $marker, $output );
3894  } else {
3895  throw new MWException( __METHOD__ . ': invalid marker type' );
3896  }
3897  return $marker;
3898  }
3899 
3907  public function incrementIncludeSize( $type, $size ) {
3908  if ( $this->mIncludeSizes[$type] + $size > $this->mOptions->getMaxIncludeSize() ) {
3909  return false;
3910  } else {
3911  $this->mIncludeSizes[$type] += $size;
3912  return true;
3913  }
3914  }
3915 
3922  $this->mExpensiveFunctionCount++;
3923  return $this->mExpensiveFunctionCount <= $this->mOptions->getExpensiveParserFunctionLimit();
3924  }
3925 
3934  public function doDoubleUnderscore( $text ) {
3935 
3936  # The position of __TOC__ needs to be recorded
3937  $mw = MagicWord::get( 'toc' );
3938  if ( $mw->match( $text ) ) {
3939  $this->mShowToc = true;
3940  $this->mForceTocPosition = true;
3941 
3942  # Set a placeholder. At the end we'll fill it in with the TOC.
3943  $text = $mw->replace( '<!--MWTOC-->', $text, 1 );
3944 
3945  # Only keep the first one.
3946  $text = $mw->replace( '', $text );
3947  }
3948 
3949  # Now match and remove the rest of them
3951  $this->mDoubleUnderscores = $mwa->matchAndRemove( $text );
3952 
3953  if ( isset( $this->mDoubleUnderscores['nogallery'] ) ) {
3954  $this->mOutput->mNoGallery = true;
3955  }
3956  if ( isset( $this->mDoubleUnderscores['notoc'] ) && !$this->mForceTocPosition ) {
3957  $this->mShowToc = false;
3958  }
3959  if ( isset( $this->mDoubleUnderscores['hiddencat'] )
3960  && $this->mTitle->getNamespace() == NS_CATEGORY
3961  ) {
3962  $this->addTrackingCategory( 'hidden-category-category' );
3963  }
3964  # (bug 8068) Allow control over whether robots index a page.
3965  # @todo FIXME: Bug 14899: __INDEX__ always overrides __NOINDEX__ here! This
3966  # is not desirable, the last one on the page should win.
3967  if ( isset( $this->mDoubleUnderscores['noindex'] ) && $this->mTitle->canUseNoindex() ) {
3968  $this->mOutput->setIndexPolicy( 'noindex' );
3969  $this->addTrackingCategory( 'noindex-category' );
3970  }
3971  if ( isset( $this->mDoubleUnderscores['index'] ) && $this->mTitle->canUseNoindex() ) {
3972  $this->mOutput->setIndexPolicy( 'index' );
3973  $this->addTrackingCategory( 'index-category' );
3974  }
3975 
3976  # Cache all double underscores in the database
3977  foreach ( $this->mDoubleUnderscores as $key => $val ) {
3978  $this->mOutput->setProperty( $key, '' );
3979  }
3980 
3981  return $text;
3982  }
3983 
3989  public function addTrackingCategory( $msg ) {
3990  return $this->mOutput->addTrackingCategory( $msg, $this->mTitle );
3991  }
3992 
4009  public function formatHeadings( $text, $origText, $isMain = true ) {
4010  global $wgMaxTocLevel, $wgExperimentalHtmlIds;
4011 
4012  # Inhibit editsection links if requested in the page
4013  if ( isset( $this->mDoubleUnderscores['noeditsection'] ) ) {
4014  $maybeShowEditLink = $showEditLink = false;
4015  } else {
4016  $maybeShowEditLink = true; /* Actual presence will depend on ParserOptions option */
4017  $showEditLink = $this->mOptions->getEditSection();
4018  }
4019  if ( $showEditLink ) {
4020  $this->mOutput->setEditSectionTokens( true );
4021  }
4022 
4023  # Get all headlines for numbering them and adding funky stuff like [edit]
4024  # links - this is for later, but we need the number of headlines right now
4025  $matches = [];
4026  $numMatches = preg_match_all(
4027  '/<H(?P<level>[1-6])(?P<attrib>.*?>)\s*(?P<header>[\s\S]*?)\s*<\/H[1-6] *>/i',
4028  $text,
4029  $matches
4030  );
4031 
4032  # if there are fewer than 4 headlines in the article, do not show TOC
4033  # unless it's been explicitly enabled.
4034  $enoughToc = $this->mShowToc &&
4035  ( ( $numMatches >= 4 ) || $this->mForceTocPosition );
4036 
4037  # Allow user to stipulate that a page should have a "new section"
4038  # link added via __NEWSECTIONLINK__
4039  if ( isset( $this->mDoubleUnderscores['newsectionlink'] ) ) {
4040  $this->mOutput->setNewSection( true );
4041  }
4042 
4043  # Allow user to remove the "new section"
4044  # link via __NONEWSECTIONLINK__
4045  if ( isset( $this->mDoubleUnderscores['nonewsectionlink'] ) ) {
4046  $this->mOutput->hideNewSection( true );
4047  }
4048 
4049  # if the string __FORCETOC__ (not case-sensitive) occurs in the HTML,
4050  # override above conditions and always show TOC above first header
4051  if ( isset( $this->mDoubleUnderscores['forcetoc'] ) ) {
4052  $this->mShowToc = true;
4053  $enoughToc = true;
4054  }
4055 
4056  # headline counter
4057  $headlineCount = 0;
4058  $numVisible = 0;
4059 
4060  # Ugh .. the TOC should have neat indentation levels which can be
4061  # passed to the skin functions. These are determined here
4062  $toc = '';
4063  $full = '';
4064  $head = [];
4065  $sublevelCount = [];
4066  $levelCount = [];
4067  $level = 0;
4068  $prevlevel = 0;
4069  $toclevel = 0;
4070  $prevtoclevel = 0;
4071  $markerRegex = self::MARKER_PREFIX . "-h-(\d+)-" . self::MARKER_SUFFIX;
4072  $baseTitleText = $this->mTitle->getPrefixedDBkey();
4073  $oldType = $this->mOutputType;
4074  $this->setOutputType( self::OT_WIKI );
4075  $frame = $this->getPreprocessor()->newFrame();
4076  $root = $this->preprocessToDom( $origText );
4077  $node = $root->getFirstChild();
4078  $byteOffset = 0;
4079  $tocraw = [];
4080  $refers = [];
4081 
4082  $headlines = $numMatches !== false ? $matches[3] : [];
4083 
4084  foreach ( $headlines as $headline ) {
4085  $isTemplate = false;
4086  $titleText = false;
4087  $sectionIndex = false;
4088  $numbering = '';
4089  $markerMatches = [];
4090  if ( preg_match( "/^$markerRegex/", $headline, $markerMatches ) ) {
4091  $serial = $markerMatches[1];
4092  list( $titleText, $sectionIndex ) = $this->mHeadings[$serial];
4093  $isTemplate = ( $titleText != $baseTitleText );
4094  $headline = preg_replace( "/^$markerRegex\\s*/", "", $headline );
4095  }
4096 
4097  if ( $toclevel ) {
4098  $prevlevel = $level;
4099  }
4100  $level = $matches[1][$headlineCount];
4101 
4102  if ( $level > $prevlevel ) {
4103  # Increase TOC level
4104  $toclevel++;
4105  $sublevelCount[$toclevel] = 0;
4106  if ( $toclevel < $wgMaxTocLevel ) {
4107  $prevtoclevel = $toclevel;
4108  $toc .= Linker::tocIndent();
4109  $numVisible++;
4110  }
4111  } elseif ( $level < $prevlevel && $toclevel > 1 ) {
4112  # Decrease TOC level, find level to jump to
4113 
4114  for ( $i = $toclevel; $i > 0; $i-- ) {
4115  if ( $levelCount[$i] == $level ) {
4116  # Found last matching level
4117  $toclevel = $i;
4118  break;
4119  } elseif ( $levelCount[$i] < $level ) {
4120  # Found first matching level below current level
4121  $toclevel = $i + 1;
4122  break;
4123  }
4124  }
4125  if ( $i == 0 ) {
4126  $toclevel = 1;
4127  }
4128  if ( $toclevel < $wgMaxTocLevel ) {
4129  if ( $prevtoclevel < $wgMaxTocLevel ) {
4130  # Unindent only if the previous toc level was shown :p
4131  $toc .= Linker::tocUnindent( $prevtoclevel - $toclevel );
4132  $prevtoclevel = $toclevel;
4133  } else {
4134  $toc .= Linker::tocLineEnd();
4135  }
4136  }
4137  } else {
4138  # No change in level, end TOC line
4139  if ( $toclevel < $wgMaxTocLevel ) {
4140  $toc .= Linker::tocLineEnd();
4141  }
4142  }
4143 
4144  $levelCount[$toclevel] = $level;
4145 
4146  # count number of headlines for each level
4147  $sublevelCount[$toclevel]++;
4148  $dot = 0;
4149  for ( $i = 1; $i <= $toclevel; $i++ ) {
4150  if ( !empty( $sublevelCount[$i] ) ) {
4151  if ( $dot ) {
4152  $numbering .= '.';
4153  }
4154  $numbering .= $this->getTargetLanguage()->formatNum( $sublevelCount[$i] );
4155  $dot = 1;
4156  }
4157  }
4158 
4159  # The safe header is a version of the header text safe to use for links
4160 
4161  # Remove link placeholders by the link text.
4162  # <!--LINK number-->
4163  # turns into
4164  # link text with suffix
4165  # Do this before unstrip since link text can contain strip markers
4166  $safeHeadline = $this->replaceLinkHoldersText( $headline );
4167 
4168  # Avoid insertion of weird stuff like <math> by expanding the relevant sections
4169  $safeHeadline = $this->mStripState->unstripBoth( $safeHeadline );
4170 
4171  # Strip out HTML (first regex removes any tag not allowed)
4172  # Allowed tags are:
4173  # * <sup> and <sub> (bug 8393)
4174  # * <i> (bug 26375)
4175  # * <b> (r105284)
4176  # * <bdi> (bug 72884)
4177  # * <span dir="rtl"> and <span dir="ltr"> (bug 35167)
4178  # * <s> and <strike> (T35715)
4179  # We strip any parameter from accepted tags (second regex), except dir="rtl|ltr" from <span>,
4180  # to allow setting directionality in toc items.
4181  $tocline = preg_replace(
4182  [
4183  '#<(?!/?(span|sup|sub|bdi|i|b|s|strike)(?: [^>]*)?>).*?>#',
4184  '#<(/?(?:span(?: dir="(?:rtl|ltr)")?|sup|sub|bdi|i|b|s|strike))(?: .*?)?>#'
4185  ],
4186  [ '', '<$1>' ],
4187  $safeHeadline
4188  );
4189 
4190  # Strip '<span></span>', which is the result from the above if
4191  # <span id="foo"></span> is used to produce an additional anchor
4192  # for a section.
4193  $tocline = str_replace( '<span></span>', '', $tocline );
4194 
4195  $tocline = trim( $tocline );
4196 
4197  # For the anchor, strip out HTML-y stuff period
4198  $safeHeadline = preg_replace( '/<.*?>/', '', $safeHeadline );
4199  $safeHeadline = Sanitizer::normalizeSectionNameWhitespace( $safeHeadline );
4200 
4201  # Save headline for section edit hint before it's escaped
4202  $headlineHint = $safeHeadline;
4203 
4204  if ( $wgExperimentalHtmlIds ) {
4205  # For reverse compatibility, provide an id that's
4206  # HTML4-compatible, like we used to.
4207  # It may be worth noting, academically, that it's possible for
4208  # the legacy anchor to conflict with a non-legacy headline
4209  # anchor on the page. In this case likely the "correct" thing
4210  # would be to either drop the legacy anchors or make sure
4211  # they're numbered first. However, this would require people
4212  # to type in section names like "abc_.D7.93.D7.90.D7.A4"
4213  # manually, so let's not bother worrying about it.
4214  $legacyHeadline = Sanitizer::escapeId( $safeHeadline,
4215  [ 'noninitial', 'legacy' ] );
4216  $safeHeadline = Sanitizer::escapeId( $safeHeadline );
4217 
4218  if ( $legacyHeadline == $safeHeadline ) {
4219  # No reason to have both (in fact, we can't)
4220  $legacyHeadline = false;
4221  }
4222  } else {
4223  $legacyHeadline = false;
4224  $safeHeadline = Sanitizer::escapeId( $safeHeadline,
4225  'noninitial' );
4226  }
4227 
4228  # HTML names must be case-insensitively unique (bug 10721).
4229  # This does not apply to Unicode characters per
4230  # http://www.w3.org/TR/html5/infrastructure.html#case-sensitivity-and-string-comparison
4231  # @todo FIXME: We may be changing them depending on the current locale.
4232  $arrayKey = strtolower( $safeHeadline );
4233  if ( $legacyHeadline === false ) {
4234  $legacyArrayKey = false;
4235  } else {
4236  $legacyArrayKey = strtolower( $legacyHeadline );
4237  }
4238 
4239  # Create the anchor for linking from the TOC to the section
4240  $anchor = $safeHeadline;
4241  $legacyAnchor = $legacyHeadline;
4242  if ( isset( $refers[$arrayKey] ) ) {
4243  // @codingStandardsIgnoreStart
4244  for ( $i = 2; isset( $refers["${arrayKey}_$i"] ); ++$i );
4245  // @codingStandardsIgnoreEnd
4246  $anchor .= "_$i";
4247  $refers["${arrayKey}_$i"] = true;
4248  } else {
4249  $refers[$arrayKey] = true;
4250  }
4251  if ( $legacyHeadline !== false && isset( $refers[$legacyArrayKey] ) ) {
4252  // @codingStandardsIgnoreStart
4253  for ( $i = 2; isset( $refers["${legacyArrayKey}_$i"] ); ++$i );
4254  // @codingStandardsIgnoreEnd
4255  $legacyAnchor .= "_$i";
4256  $refers["${legacyArrayKey}_$i"] = true;
4257  } else {
4258  $refers[$legacyArrayKey] = true;
4259  }
4260 
4261  # Don't number the heading if it is the only one (looks silly)
4262  if ( count( $matches[3] ) > 1 && $this->mOptions->getNumberHeadings() ) {
4263  # the two are different if the line contains a link
4264  $headline = Html::element(
4265  'span',
4266  [ 'class' => 'mw-headline-number' ],
4267  $numbering
4268  ) . ' ' . $headline;
4269  }
4270 
4271  if ( $enoughToc && ( !isset( $wgMaxTocLevel ) || $toclevel < $wgMaxTocLevel ) ) {
4272  $toc .= Linker::tocLine( $anchor, $tocline,
4273  $numbering, $toclevel, ( $isTemplate ? false : $sectionIndex ) );
4274  }
4275 
4276  # Add the section to the section tree
4277  # Find the DOM node for this header
4278  $noOffset = ( $isTemplate || $sectionIndex === false );
4279  while ( $node && !$noOffset ) {
4280  if ( $node->getName() === 'h' ) {
4281  $bits = $node->splitHeading();
4282  if ( $bits['i'] == $sectionIndex ) {
4283  break;
4284  }
4285  }
4286  $byteOffset += mb_strlen( $this->mStripState->unstripBoth(
4287  $frame->expand( $node, PPFrame::RECOVER_ORIG ) ) );
4288  $node = $node->getNextSibling();
4289  }
4290  $tocraw[] = [
4291  'toclevel' => $toclevel,
4292  'level' => $level,
4293  'line' => $tocline,
4294  'number' => $numbering,
4295  'index' => ( $isTemplate ? 'T-' : '' ) . $sectionIndex,
4296  'fromtitle' => $titleText,
4297  'byteoffset' => ( $noOffset ? null : $byteOffset ),
4298  'anchor' => $anchor,
4299  ];
4300 
4301  # give headline the correct <h#> tag
4302  if ( $maybeShowEditLink && $sectionIndex !== false ) {
4303  // Output edit section links as markers with styles that can be customized by skins
4304  if ( $isTemplate ) {
4305  # Put a T flag in the section identifier, to indicate to extractSections()
4306  # that sections inside <includeonly> should be counted.
4307  $editsectionPage = $titleText;
4308  $editsectionSection = "T-$sectionIndex";
4309  $editsectionContent = null;
4310  } else {
4311  $editsectionPage = $this->mTitle->getPrefixedText();
4312  $editsectionSection = $sectionIndex;
4313  $editsectionContent = $headlineHint;
4314  }
4315  // We use a bit of pesudo-xml for editsection markers. The
4316  // language converter is run later on. Using a UNIQ style marker
4317  // leads to the converter screwing up the tokens when it
4318  // converts stuff. And trying to insert strip tags fails too. At
4319  // this point all real inputted tags have already been escaped,
4320  // so we don't have to worry about a user trying to input one of
4321  // these markers directly. We use a page and section attribute
4322  // to stop the language converter from converting these
4323  // important bits of data, but put the headline hint inside a
4324  // content block because the language converter is supposed to
4325  // be able to convert that piece of data.
4326  // Gets replaced with html in ParserOutput::getText
4327  $editlink = '<mw:editsection page="' . htmlspecialchars( $editsectionPage );
4328  $editlink .= '" section="' . htmlspecialchars( $editsectionSection ) . '"';
4329  if ( $editsectionContent !== null ) {
4330  $editlink .= '>' . $editsectionContent . '</mw:editsection>';
4331  } else {
4332  $editlink .= '/>';
4333  }
4334  } else {
4335  $editlink = '';
4336  }
4337  $head[$headlineCount] = Linker::makeHeadline( $level,
4338  $matches['attrib'][$headlineCount], $anchor, $headline,
4339  $editlink, $legacyAnchor );
4340 
4341  $headlineCount++;
4342  }
4343 
4344  $this->setOutputType( $oldType );
4345 
4346  # Never ever show TOC if no headers
4347  if ( $numVisible < 1 ) {
4348  $enoughToc = false;
4349  }
4350 
4351  if ( $enoughToc ) {
4352  if ( $prevtoclevel > 0 && $prevtoclevel < $wgMaxTocLevel ) {
4353  $toc .= Linker::tocUnindent( $prevtoclevel - 1 );
4354  }
4355  $toc = Linker::tocList( $toc, $this->mOptions->getUserLangObj() );
4356  $this->mOutput->setTOCHTML( $toc );
4357  $toc = self::TOC_START . $toc . self::TOC_END;
4358  $this->mOutput->addModules( 'mediawiki.toc' );
4359  }
4360 
4361  if ( $isMain ) {
4362  $this->mOutput->setSections( $tocraw );
4363  }
4364 
4365  # split up and insert constructed headlines
4366  $blocks = preg_split( '/<H[1-6].*?>[\s\S]*?<\/H[1-6]>/i', $text );
4367  $i = 0;
4368 
4369  // build an array of document sections
4370  $sections = [];
4371  foreach ( $blocks as $block ) {
4372  // $head is zero-based, sections aren't.
4373  if ( empty( $head[$i - 1] ) ) {
4374  $sections[$i] = $block;
4375  } else {
4376  $sections[$i] = $head[$i - 1] . $block;
4377  }
4378 
4389  Hooks::run( 'ParserSectionCreate', [ $this, $i, &$sections[$i], $showEditLink ] );
4390 
4391  $i++;
4392  }
4393 
4394  if ( $enoughToc && $isMain && !$this->mForceTocPosition ) {
4395  // append the TOC at the beginning
4396  // Top anchor now in skin
4397  $sections[0] = $sections[0] . $toc . "\n";
4398  }
4399 
4400  $full .= implode( '', $sections );
4401 
4402  if ( $this->mForceTocPosition ) {
4403  return str_replace( '<!--MWTOC-->', $toc, $full );
4404  } else {
4405  return $full;
4406  }
4407  }
4408 
4420  public function preSaveTransform( $text, Title $title, User $user,
4421  ParserOptions $options, $clearState = true
4422  ) {
4423  if ( $clearState ) {
4424  $magicScopeVariable = $this->lock();
4425  }
4426  $this->startParse( $title, $options, self::OT_WIKI, $clearState );
4427  $this->setUser( $user );
4428 
4429  // We still normalize line endings for backwards-compatibility
4430  // with other code that just calls PST, but this should already
4431  // be handled in TextContent subclasses
4432  $text = TextContent::normalizeLineEndings( $text );
4433 
4434  if ( $options->getPreSaveTransform() ) {
4435  $text = $this->pstPass2( $text, $user );
4436  }
4437  $text = $this->mStripState->unstripBoth( $text );
4438 
4439  $this->setUser( null ); # Reset
4440 
4441  return $text;
4442  }
4443 
4452  private function pstPass2( $text, $user ) {
4454 
4455  # Note: This is the timestamp saved as hardcoded wikitext to
4456  # the database, we use $wgContLang here in order to give
4457  # everyone the same signature and use the default one rather
4458  # than the one selected in each user's preferences.
4459  # (see also bug 12815)
4460  $ts = $this->mOptions->getTimestamp();
4462  $ts = $timestamp->format( 'YmdHis' );
4463  $tzMsg = $timestamp->getTimezoneMessage()->inContentLanguage()->text();
4464 
4465  $d = $wgContLang->timeanddate( $ts, false, false ) . " ($tzMsg)";
4466 
4467  # Variable replacement
4468  # Because mOutputType is OT_WIKI, this will only process {{subst:xxx}} type tags
4469  $text = $this->replaceVariables( $text );
4470 
4471  # This works almost by chance, as the replaceVariables are done before the getUserSig(),
4472  # which may corrupt this parser instance via its wfMessage()->text() call-
4473 
4474  # Signatures
4475  $sigText = $this->getUserSig( $user );
4476  $text = strtr( $text, [
4477  '~~~~~' => $d,
4478  '~~~~' => "$sigText $d",
4479  '~~~' => $sigText
4480  ] );
4481 
4482  # Context links ("pipe tricks"): [[|name]] and [[name (context)|]]
4483  $tc = '[' . Title::legalChars() . ']';
4484  $nc = '[ _0-9A-Za-z\x80-\xff-]'; # Namespaces can use non-ascii!
4485 
4486  // [[ns:page (context)|]]
4487  $p1 = "/\[\[(:?$nc+:|:|)($tc+?)( ?\\($tc+\\))\\|]]/";
4488  // [[ns:page(context)|]] (double-width brackets, added in r40257)
4489  $p4 = "/\[\[(:?$nc+:|:|)($tc+?)( ?($tc+))\\|]]/";
4490  // [[ns:page (context), context|]] (using either single or double-width comma)
4491  $p3 = "/\[\[(:?$nc+:|:|)($tc+?)( ?\\($tc+\\)|)((?:, |,)$tc+|)\\|]]/";
4492  // [[|page]] (reverse pipe trick: add context from page title)
4493  $p2 = "/\[\[\\|($tc+)]]/";
4494 
4495  # try $p1 first, to turn "[[A, B (C)|]]" into "[[A, B (C)|A, B]]"
4496  $text = preg_replace( $p1, '[[\\1\\2\\3|\\2]]', $text );
4497  $text = preg_replace( $p4, '[[\\1\\2\\3|\\2]]', $text );
4498  $text = preg_replace( $p3, '[[\\1\\2\\3\\4|\\2]]', $text );
4499 
4500  $t = $this->mTitle->getText();
4501  $m = [];
4502  if ( preg_match( "/^($nc+:|)$tc+?( \\($tc+\\))$/", $t, $m ) ) {
4503  $text = preg_replace( $p2, "[[$m[1]\\1$m[2]|\\1]]", $text );
4504  } elseif ( preg_match( "/^($nc+:|)$tc+?(, $tc+|)$/", $t, $m ) && "$m[1]$m[2]" != '' ) {
4505  $text = preg_replace( $p2, "[[$m[1]\\1$m[2]|\\1]]", $text );
4506  } else {
4507  # if there's no context, don't bother duplicating the title
4508  $text = preg_replace( $p2, '[[\\1]]', $text );
4509  }
4510 
4511  return $text;
4512  }
4513 
4528  public function getUserSig( &$user, $nickname = false, $fancySig = null ) {
4529  global $wgMaxSigChars;
4530 
4531  $username = $user->getName();
4532 
4533  # If not given, retrieve from the user object.
4534  if ( $nickname === false ) {
4535  $nickname = $user->getOption( 'nickname' );
4536  }
4537 
4538  if ( is_null( $fancySig ) ) {
4539  $fancySig = $user->getBoolOption( 'fancysig' );
4540  }
4541 
4542  $nickname = $nickname == null ? $username : $nickname;
4543 
4544  if ( mb_strlen( $nickname ) > $wgMaxSigChars ) {
4545  $nickname = $username;
4546  wfDebug( __METHOD__ . ": $username has overlong signature.\n" );
4547  } elseif ( $fancySig !== false ) {
4548  # Sig. might contain markup; validate this
4549  if ( $this->validateSig( $nickname ) !== false ) {
4550  # Validated; clean up (if needed) and return it
4551  return $this->cleanSig( $nickname, true );
4552  } else {
4553  # Failed to validate; fall back to the default
4554  $nickname = $username;
4555  wfDebug( __METHOD__ . ": $username has bad XML tags in signature.\n" );
4556  }
4557  }
4558 
4559  # Make sure nickname doesnt get a sig in a sig
4560  $nickname = self::cleanSigInSig( $nickname );
4561 
4562  # If we're still here, make it a link to the user page
4563  $userText = wfEscapeWikiText( $username );
4564  $nickText = wfEscapeWikiText( $nickname );
4565  $msgName = $user->isAnon() ? 'signature-anon' : 'signature';
4566 
4567  return wfMessage( $msgName, $userText, $nickText )->inContentLanguage()
4568  ->title( $this->getTitle() )->text();
4569  }
4570 
4577  public function validateSig( $text ) {
4578  return Xml::isWellFormedXmlFragment( $text ) ? $text : false;
4579  }
4580 
4591  public function cleanSig( $text, $parsing = false ) {
4592  if ( !$parsing ) {
4593  global $wgTitle;
4594  $magicScopeVariable = $this->lock();
4595  $this->startParse( $wgTitle, new ParserOptions, self::OT_PREPROCESS, true );
4596  }
4597 
4598  # Option to disable this feature
4599  if ( !$this->mOptions->getCleanSignatures() ) {
4600  return $text;
4601  }
4602 
4603  # @todo FIXME: Regex doesn't respect extension tags or nowiki
4604  # => Move this logic to braceSubstitution()
4605  $substWord = MagicWord::get( 'subst' );
4606  $substRegex = '/\{\{(?!(?:' . $substWord->getBaseRegex() . '))/x' . $substWord->getRegexCase();
4607  $substText = '{{' . $substWord->getSynonym( 0 );
4608 
4609  $text = preg_replace( $substRegex, $substText, $text );
4610  $text = self::cleanSigInSig( $text );
4611  $dom = $this->preprocessToDom( $text );
4612  $frame = $this->getPreprocessor()->newFrame();
4613  $text = $frame->expand( $dom );
4614 
4615  if ( !$parsing ) {
4616  $text = $this->mStripState->unstripBoth( $text );
4617  }
4618 
4619  return $text;
4620  }
4621 
4628  public static function cleanSigInSig( $text ) {
4629  $text = preg_replace( '/~{3,5}/', '', $text );
4630  return $text;
4631  }
4632 
4643  $outputType, $clearState = true
4644  ) {
4645  $this->startParse( $title, $options, $outputType, $clearState );
4646  }
4647 
4654  private function startParse( Title $title = null, ParserOptions $options,
4655  $outputType, $clearState = true
4656  ) {
4657  $this->setTitle( $title );
4658  $this->mOptions = $options;
4659  $this->setOutputType( $outputType );
4660  if ( $clearState ) {
4661  $this->clearState();
4662  }
4663  }
4664 
4673  public function transformMsg( $text, $options, $title = null ) {
4674  static $executing = false;
4675 
4676  # Guard against infinite recursion
4677  if ( $executing ) {
4678  return $text;
4679  }
4680  $executing = true;
4681 
4682  if ( !$title ) {
4683  global $wgTitle;
4684  $title = $wgTitle;
4685  }
4686 
4687  $text = $this->preprocess( $text, $title, $options );
4688 
4689  $executing = false;
4690  return $text;
4691  }
4692 
4717  public function setHook( $tag, $callback ) {
4718  $tag = strtolower( $tag );
4719  if ( preg_match( '/[<>\r\n]/', $tag, $m ) ) {
4720  throw new MWException( "Invalid character {$m[0]} in setHook('$tag', ...) call" );
4721  }
4722  $oldVal = isset( $this->mTagHooks[$tag] ) ? $this->mTagHooks[$tag] : null;
4723  $this->mTagHooks[$tag] = $callback;
4724  if ( !in_array( $tag, $this->mStripList ) ) {
4725  $this->mStripList[] = $tag;
4726  }
4727 
4728  return $oldVal;
4729  }
4730 
4748  public function setTransparentTagHook( $tag, $callback ) {
4749  $tag = strtolower( $tag );
4750  if ( preg_match( '/[<>\r\n]/', $tag, $m ) ) {
4751  throw new MWException( "Invalid character {$m[0]} in setTransparentHook('$tag', ...) call" );
4752  }
4753  $oldVal = isset( $this->mTransparentTagHooks[$tag] ) ? $this->mTransparentTagHooks[$tag] : null;
4754  $this->mTransparentTagHooks[$tag] = $callback;
4755 
4756  return $oldVal;
4757  }
4758 
4762  public function clearTagHooks() {
4763  $this->mTagHooks = [];
4764  $this->mFunctionTagHooks = [];
4765  $this->mStripList = $this->mDefaultStripList;
4766  }
4767 
4811  public function setFunctionHook( $id, $callback, $flags = 0 ) {
4813 
4814  $oldVal = isset( $this->mFunctionHooks[$id] ) ? $this->mFunctionHooks[$id][0] : null;
4815  $this->mFunctionHooks[$id] = [ $callback, $flags ];
4816 
4817  # Add to function cache
4818  $mw = MagicWord::get( $id );
4819  if ( !$mw ) {
4820  throw new MWException( __METHOD__ . '() expecting a magic word identifier.' );
4821  }
4822 
4823  $synonyms = $mw->getSynonyms();
4824  $sensitive = intval( $mw->isCaseSensitive() );
4825 
4826  foreach ( $synonyms as $syn ) {
4827  # Case
4828  if ( !$sensitive ) {
4829  $syn = $wgContLang->lc( $syn );
4830  }
4831  # Add leading hash
4832  if ( !( $flags & self::SFH_NO_HASH ) ) {
4833  $syn = '#' . $syn;
4834  }
4835  # Remove trailing colon
4836  if ( substr( $syn, -1, 1 ) === ':' ) {
4837  $syn = substr( $syn, 0, -1 );
4838  }
4839  $this->mFunctionSynonyms[$sensitive][$syn] = $id;
4840  }
4841  return $oldVal;
4842  }
4843 
4849  public function getFunctionHooks() {
4850  return array_keys( $this->mFunctionHooks );
4851  }
4852 
4863  public function setFunctionTagHook( $tag, $callback, $flags ) {
4864  $tag = strtolower( $tag );
4865  if ( preg_match( '/[<>\r\n]/', $tag, $m ) ) {
4866  throw new MWException( "Invalid character {$m[0]} in setFunctionTagHook('$tag', ...) call" );
4867  }
4868  $old = isset( $this->mFunctionTagHooks[$tag] ) ?
4869  $this->mFunctionTagHooks[$tag] : null;
4870  $this->mFunctionTagHooks[$tag] = [ $callback, $flags ];
4871 
4872  if ( !in_array( $tag, $this->mStripList ) ) {
4873  $this->mStripList[] = $tag;
4874  }
4875 
4876  return $old;
4877  }
4878 
4886  public function replaceLinkHolders( &$text, $options = 0 ) {
4887  $this->mLinkHolders->replace( $text );
4888  }
4889 
4897  public function replaceLinkHoldersText( $text ) {
4898  return $this->mLinkHolders->replaceText( $text );
4899  }
4900 
4914  public function renderImageGallery( $text, $params ) {
4915 
4916  $mode = false;
4917  if ( isset( $params['mode'] ) ) {
4918  $mode = $params['mode'];
4919  }
4920 
4921  try {
4922  $ig = ImageGalleryBase::factory( $mode );
4923  } catch ( Exception $e ) {
4924  // If invalid type set, fallback to default.
4925  $ig = ImageGalleryBase::factory( false );
4926  }
4927 
4928  $ig->setContextTitle( $this->mTitle );
4929  $ig->setShowBytes( false );
4930  $ig->setShowFilename( false );
4931  $ig->setParser( $this );
4932  $ig->setHideBadImages();
4933  $ig->setAttributes( Sanitizer::validateTagAttributes( $params, 'table' ) );
4934 
4935  if ( isset( $params['showfilename'] ) ) {
4936  $ig->setShowFilename( true );
4937  } else {
4938  $ig->setShowFilename( false );
4939  }
4940  if ( isset( $params['caption'] ) ) {
4941  $caption = $params['caption'];
4942  $caption = htmlspecialchars( $caption );
4943  $caption = $this->replaceInternalLinks( $caption );
4944  $ig->setCaptionHtml( $caption );
4945  }
4946  if ( isset( $params['perrow'] ) ) {
4947  $ig->setPerRow( $params['perrow'] );
4948  }
4949  if ( isset( $params['widths'] ) ) {
4950  $ig->setWidths( $params['widths'] );
4951  }
4952  if ( isset( $params['heights'] ) ) {
4953  $ig->setHeights( $params['heights'] );
4954  }
4955  $ig->setAdditionalOptions( $params );
4956 
4957  Hooks::run( 'BeforeParserrenderImageGallery', [ &$this, &$ig ] );
4958 
4959  $lines = StringUtils::explode( "\n", $text );
4960  foreach ( $lines as $line ) {
4961  # match lines like these:
4962  # Image:someimage.jpg|This is some image
4963  $matches = [];
4964  preg_match( "/^([^|]+)(\\|(.*))?$/", $line, $matches );
4965  # Skip empty lines
4966  if ( count( $matches ) == 0 ) {
4967  continue;
4968  }
4969 
4970  if ( strpos( $matches[0], '%' ) !== false ) {
4971  $matches[1] = rawurldecode( $matches[1] );
4972  }
4974  if ( is_null( $title ) ) {
4975  # Bogus title. Ignore these so we don't bomb out later.
4976  continue;
4977  }
4978 
4979  # We need to get what handler the file uses, to figure out parameters.
4980  # Note, a hook can overide the file name, and chose an entirely different
4981  # file (which potentially could be of a different type and have different handler).
4982  $options = [];
4983  $descQuery = false;
4984  Hooks::run( 'BeforeParserFetchFileAndTitle',
4985  [ $this, $title, &$options, &$descQuery ] );
4986  # Don't register it now, as ImageGallery does that later.
4987  $file = $this->fetchFileNoRegister( $title, $options );
4988  $handler = $file ? $file->getHandler() : false;
4989 
4990  $paramMap = [
4991  'img_alt' => 'gallery-internal-alt',
4992  'img_link' => 'gallery-internal-link',
4993  ];
4994  if ( $handler ) {
4995  $paramMap = $paramMap + $handler->getParamMap();
4996  // We don't want people to specify per-image widths.
4997  // Additionally the width parameter would need special casing anyhow.
4998  unset( $paramMap['img_width'] );
4999  }
5000 
5001  $mwArray = new MagicWordArray( array_keys( $paramMap ) );
5002 
5003  $label = '';
5004  $alt = '';
5005  $link = '';
5006  $handlerOptions = [];
5007  if ( isset( $matches[3] ) ) {
5008  // look for an |alt= definition while trying not to break existing
5009  // captions with multiple pipes (|) in it, until a more sensible grammar
5010  // is defined for images in galleries
5011 
5012  // FIXME: Doing recursiveTagParse at this stage, and the trim before
5013  // splitting on '|' is a bit odd, and different from makeImage.
5014  $matches[3] = $this->recursiveTagParse( trim( $matches[3] ) );
5015  $parameterMatches = StringUtils::explode( '|', $matches[3] );
5016 
5017  foreach ( $parameterMatches as $parameterMatch ) {
5018  list( $magicName, $match ) = $mwArray->matchVariableStartToEnd( $parameterMatch );
5019  if ( $magicName ) {
5020  $paramName = $paramMap[$magicName];
5021 
5022  switch ( $paramName ) {
5023  case 'gallery-internal-alt':
5024  $alt = $this->stripAltText( $match, false );
5025  break;
5026  case 'gallery-internal-link':
5027  $linkValue = strip_tags( $this->replaceLinkHoldersText( $match ) );
5028  $chars = self::EXT_LINK_URL_CLASS;
5029  $addr = self::EXT_LINK_ADDR;
5030  $prots = $this->mUrlProtocols;
5031  // check to see if link matches an absolute url, if not then it must be a wiki link.
5032  if ( preg_match( "/^($prots)$addr$chars*$/u", $linkValue ) ) {
5033  $link = $linkValue;
5034  $this->mOutput->addExternalLink( $link );
5035  } else {
5036  $localLinkTitle = Title::newFromText( $linkValue );
5037  if ( $localLinkTitle !== null ) {
5038  $this->mOutput->addLink( $localLinkTitle );
5039  $link = $localLinkTitle->getLinkURL();
5040  }
5041  }
5042  break;
5043  default:
5044  // Must be a handler specific parameter.
5045  if ( $handler->validateParam( $paramName, $match ) ) {
5046  $handlerOptions[$paramName] = $match;
5047  } else {
5048  // Guess not, consider it as caption.
5049  wfDebug( "$parameterMatch failed parameter validation\n" );
5050  $label = '|' . $parameterMatch;
5051  }
5052  }
5053 
5054  } else {
5055  // Last pipe wins.
5056  $label = '|' . $parameterMatch;
5057  }
5058  }
5059  // Remove the pipe.
5060  $label = substr( $label, 1 );
5061  }
5062 
5063  $ig->add( $title, $label, $alt, $link, $handlerOptions );
5064  }
5065  $html = $ig->toHTML();
5066  Hooks::run( 'AfterParserFetchFileAndTitle', [ $this, $ig, &$html ] );
5067  return $html;
5068  }
5069 
5074  public function getImageParams( $handler ) {
5075  if ( $handler ) {
5076  $handlerClass = get_class( $handler );
5077  } else {
5078  $handlerClass = '';
5079  }
5080  if ( !isset( $this->mImageParams[$handlerClass] ) ) {
5081  # Initialise static lists
5082  static $internalParamNames = [
5083  'horizAlign' => [ 'left', 'right', 'center', 'none' ],
5084  'vertAlign' => [ 'baseline', 'sub', 'super', 'top', 'text-top', 'middle',
5085  'bottom', 'text-bottom' ],
5086  'frame' => [ 'thumbnail', 'manualthumb', 'framed', 'frameless',
5087  'upright', 'border', 'link', 'alt', 'class' ],
5088  ];
5089  static $internalParamMap;
5090  if ( !$internalParamMap ) {
5091  $internalParamMap = [];
5092  foreach ( $internalParamNames as $type => $names ) {
5093  foreach ( $names as $name ) {
5094  $magicName = str_replace( '-', '_', "img_$name" );
5095  $internalParamMap[$magicName] = [ $type, $name ];
5096  }
5097  }
5098  }
5099 
5100  # Add handler params
5101  $paramMap = $internalParamMap;
5102  if ( $handler ) {
5103  $handlerParamMap = $handler->getParamMap();
5104  foreach ( $handlerParamMap as $magic => $paramName ) {
5105  $paramMap[$magic] = [ 'handler', $paramName ];
5106  }
5107  }
5108  $this->mImageParams[$handlerClass] = $paramMap;
5109  $this->mImageParamsMagicArray[$handlerClass] = new MagicWordArray( array_keys( $paramMap ) );
5110  }
5111  return [ $this->mImageParams[$handlerClass], $this->mImageParamsMagicArray[$handlerClass] ];
5112  }
5113 
5122  public function makeImage( $title, $options, $holders = false ) {
5123  # Check if the options text is of the form "options|alt text"
5124  # Options are:
5125  # * thumbnail make a thumbnail with enlarge-icon and caption, alignment depends on lang
5126  # * left no resizing, just left align. label is used for alt= only
5127  # * right same, but right aligned
5128  # * none same, but not aligned
5129  # * ___px scale to ___ pixels width, no aligning. e.g. use in taxobox
5130  # * center center the image
5131  # * frame Keep original image size, no magnify-button.
5132  # * framed Same as "frame"
5133  # * frameless like 'thumb' but without a frame. Keeps user preferences for width
5134  # * upright reduce width for upright images, rounded to full __0 px
5135  # * border draw a 1px border around the image
5136  # * alt Text for HTML alt attribute (defaults to empty)
5137  # * class Set a class for img node
5138  # * link Set the target of the image link. Can be external, interwiki, or local
5139  # vertical-align values (no % or length right now):
5140  # * baseline
5141  # * sub
5142  # * super
5143  # * top
5144  # * text-top
5145  # * middle
5146  # * bottom
5147  # * text-bottom
5148 
5149  $parts = StringUtils::explode( "|", $options );
5150 
5151  # Give extensions a chance to select the file revision for us
5152  $options = [];
5153  $descQuery = false;
5154  Hooks::run( 'BeforeParserFetchFileAndTitle',
5155  [ $this, $title, &$options, &$descQuery ] );
5156  # Fetch and register the file (file title may be different via hooks)
5157  list( $file, $title ) = $this->fetchFileAndTitle( $title, $options );
5158 
5159  # Get parameter map
5160  $handler = $file ? $file->getHandler() : false;
5161 
5162  list( $paramMap, $mwArray ) = $this->getImageParams( $handler );
5163 
5164  if ( !$file ) {
5165  $this->addTrackingCategory( 'broken-file-category' );
5166  }
5167 
5168  # Process the input parameters
5169  $caption = '';
5170  $params = [ 'frame' => [], 'handler' => [],
5171  'horizAlign' => [], 'vertAlign' => [] ];
5172  $seenformat = false;
5173  foreach ( $parts as $part ) {
5174  $part = trim( $part );
5175  list( $magicName, $value ) = $mwArray->matchVariableStartToEnd( $part );
5176  $validated = false;
5177  if ( isset( $paramMap[$magicName] ) ) {
5178  list( $type, $paramName ) = $paramMap[$magicName];
5179 
5180  # Special case; width and height come in one variable together
5181  if ( $type === 'handler' && $paramName === 'width' ) {
5182  $parsedWidthParam = $this->parseWidthParam( $value );
5183  if ( isset( $parsedWidthParam['width'] ) ) {
5184  $width = $parsedWidthParam['width'];
5185  if ( $handler->validateParam( 'width', $width ) ) {
5186  $params[$type]['width'] = $width;
5187  $validated = true;
5188  }
5189  }
5190  if ( isset( $parsedWidthParam['height'] ) ) {
5191  $height = $parsedWidthParam['height'];
5192  if ( $handler->validateParam( 'height', $height ) ) {
5193  $params[$type]['height'] = $height;
5194  $validated = true;
5195  }
5196  }
5197  # else no validation -- bug 13436
5198  } else {
5199  if ( $type === 'handler' ) {
5200  # Validate handler parameter
5201  $validated = $handler->validateParam( $paramName, $value );
5202  } else {
5203  # Validate internal parameters
5204  switch ( $paramName ) {
5205  case 'manualthumb':
5206  case 'alt':
5207  case 'class':
5208  # @todo FIXME: Possibly check validity here for
5209  # manualthumb? downstream behavior seems odd with
5210  # missing manual thumbs.
5211  $validated = true;
5212  $value = $this->stripAltText( $value, $holders );
5213  break;
5214  case 'link':
5215  $chars = self::EXT_LINK_URL_CLASS;
5216  $addr = self::EXT_LINK_ADDR;
5217  $prots = $this->mUrlProtocols;
5218  if ( $value === '' ) {
5219  $paramName = 'no-link';
5220  $value = true;
5221  $validated = true;
5222  } elseif ( preg_match( "/^((?i)$prots)/", $value ) ) {
5223  if ( preg_match( "/^((?i)$prots)$addr$chars*$/u", $value, $m ) ) {
5224  $paramName = 'link-url';
5225  $this->mOutput->addExternalLink( $value );
5226  if ( $this->mOptions->getExternalLinkTarget() ) {
5227  $params[$type]['link-target'] = $this->mOptions->getExternalLinkTarget();
5228  }
5229  $validated = true;
5230  }
5231  } else {
5232  $linkTitle = Title::newFromText( $value );
5233  if ( $linkTitle ) {
5234  $paramName = 'link-title';
5235  $value = $linkTitle;
5236  $this->mOutput->addLink( $linkTitle );
5237  $validated = true;
5238  }
5239  }
5240  break;
5241  case 'frameless':
5242  case 'framed':
5243  case 'thumbnail':
5244  // use first appearing option, discard others.
5245  $validated = ! $seenformat;
5246  $seenformat = true;
5247  break;
5248  default:
5249  # Most other things appear to be empty or numeric...
5250  $validated = ( $value === false || is_numeric( trim( $value ) ) );
5251  }
5252  }
5253 
5254  if ( $validated ) {
5255  $params[$type][$paramName] = $value;
5256  }
5257  }
5258  }
5259  if ( !$validated ) {
5260  $caption = $part;
5261  }
5262  }
5263 
5264  # Process alignment parameters
5265  if ( $params['horizAlign'] ) {
5266  $params['frame']['align'] = key( $params['horizAlign'] );
5267  }
5268  if ( $params['vertAlign'] ) {
5269  $params['frame']['valign'] = key( $params['vertAlign'] );
5270  }
5271 
5272  $params['frame']['caption'] = $caption;
5273 
5274  # Will the image be presented in a frame, with the caption below?
5275  $imageIsFramed = isset( $params['frame']['frame'] )
5276  || isset( $params['frame']['framed'] )
5277  || isset( $params['frame']['thumbnail'] )
5278  || isset( $params['frame']['manualthumb'] );
5279 
5280  # In the old days, [[Image:Foo|text...]] would set alt text. Later it
5281  # came to also set the caption, ordinary text after the image -- which
5282  # makes no sense, because that just repeats the text multiple times in
5283  # screen readers. It *also* came to set the title attribute.
5284  # Now that we have an alt attribute, we should not set the alt text to
5285  # equal the caption: that's worse than useless, it just repeats the
5286  # text. This is the framed/thumbnail case. If there's no caption, we
5287  # use the unnamed parameter for alt text as well, just for the time be-
5288  # ing, if the unnamed param is set and the alt param is not.
5289  # For the future, we need to figure out if we want to tweak this more,
5290  # e.g., introducing a title= parameter for the title; ignoring the un-
5291  # named parameter entirely for images without a caption; adding an ex-
5292  # plicit caption= parameter and preserving the old magic unnamed para-
5293  # meter for BC; ...
5294  if ( $imageIsFramed ) { # Framed image
5295  if ( $caption === '' && !isset( $params['frame']['alt'] ) ) {
5296  # No caption or alt text, add the filename as the alt text so
5297  # that screen readers at least get some description of the image
5298  $params['frame']['alt'] = $title->getText();
5299  }
5300  # Do not set $params['frame']['title'] because tooltips don't make sense
5301  # for framed images
5302  } else { # Inline image
5303  if ( !isset( $params['frame']['alt'] ) ) {
5304  # No alt text, use the "caption" for the alt text
5305  if ( $caption !== '' ) {
5306  $params['frame']['alt'] = $this->stripAltText( $caption, $holders );
5307  } else {
5308  # No caption, fall back to using the filename for the
5309  # alt text
5310  $params['frame']['alt'] = $title->getText();
5311  }
5312  }
5313  # Use the "caption" for the tooltip text
5314  $params['frame']['title'] = $this->stripAltText( $caption, $holders );
5315  }
5316 
5317  Hooks::run( 'ParserMakeImageParams', [ $title, $file, &$params, $this ] );
5318 
5319  # Linker does the rest
5320  $time = isset( $options['time'] ) ? $options['time'] : false;
5321  $ret = Linker::makeImageLink( $this, $title, $file, $params['frame'], $params['handler'],
5322  $time, $descQuery, $this->mOptions->getThumbSize() );
5323 
5324  # Give the handler a chance to modify the parser object
5325  if ( $handler ) {
5326  $handler->parserTransformHook( $this, $file );
5327  }
5328 
5329  return $ret;
5330  }
5331 
5337  protected function stripAltText( $caption, $holders ) {
5338  # Strip bad stuff out of the title (tooltip). We can't just use
5339  # replaceLinkHoldersText() here, because if this function is called
5340  # from replaceInternalLinks2(), mLinkHolders won't be up-to-date.
5341  if ( $holders ) {
5342  $tooltip = $holders->replaceText( $caption );
5343  } else {
5344  $tooltip = $this->replaceLinkHoldersText( $caption );
5345  }
5346 
5347  # make sure there are no placeholders in thumbnail attributes
5348  # that are later expanded to html- so expand them now and
5349  # remove the tags
5350  $tooltip = $this->mStripState->unstripBoth( $tooltip );
5351  $tooltip = Sanitizer::stripAllTags( $tooltip );
5352 
5353  return $tooltip;
5354  }
5355 
5361  public function disableCache() {
5362  wfDebug( "Parser output marked as uncacheable.\n" );
5363  if ( !$this->mOutput ) {
5364  throw new MWException( __METHOD__ .
5365  " can only be called when actually parsing something" );
5366  }
5367  $this->mOutput->updateCacheExpiry( 0 ); // new style, for consistency
5368  }
5369 
5378  public function attributeStripCallback( &$text, $frame = false ) {
5379  $text = $this->replaceVariables( $text, $frame );
5380  $text = $this->mStripState->unstripBoth( $text );
5381  return $text;
5382  }
5383 
5389  public function getTags() {
5390  return array_merge(
5391  array_keys( $this->mTransparentTagHooks ),
5392  array_keys( $this->mTagHooks ),
5393  array_keys( $this->mFunctionTagHooks )
5394  );
5395  }
5396 
5407  public function replaceTransparentTags( $text ) {
5408  $matches = [];
5409  $elements = array_keys( $this->mTransparentTagHooks );
5410  $text = self::extractTagsAndParams( $elements, $text, $matches );
5411  $replacements = [];
5412 
5413  foreach ( $matches as $marker => $data ) {
5414  list( $element, $content, $params, $tag ) = $data;
5415  $tagName = strtolower( $element );
5416  if ( isset( $this->mTransparentTagHooks[$tagName] ) ) {
5417  $output = call_user_func_array(
5418  $this->mTransparentTagHooks[$tagName],
5419  [ $content, $params, $this ]
5420  );
5421  } else {
5422  $output = $tag;
5423  }
5424  $replacements[$marker] = $output;
5425  }
5426  return strtr( $text, $replacements );
5427  }
5428 
5458  private function extractSections( $text, $sectionId, $mode, $newText = '' ) {
5459  global $wgTitle; # not generally used but removes an ugly failure mode
5460 
5461  $magicScopeVariable = $this->lock();
5462  $this->startParse( $wgTitle, new ParserOptions, self::OT_PLAIN, true );
5463  $outText = '';
5464  $frame = $this->getPreprocessor()->newFrame();
5465 
5466  # Process section extraction flags
5467  $flags = 0;
5468  $sectionParts = explode( '-', $sectionId );
5469  $sectionIndex = array_pop( $sectionParts );
5470  foreach ( $sectionParts as $part ) {
5471  if ( $part === 'T' ) {
5472  $flags |= self::PTD_FOR_INCLUSION;
5473  }
5474  }
5475 
5476  # Check for empty input
5477  if ( strval( $text ) === '' ) {
5478  # Only sections 0 and T-0 exist in an empty document
5479  if ( $sectionIndex == 0 ) {
5480  if ( $mode === 'get' ) {
5481  return '';
5482  } else {
5483  return $newText;
5484  }
5485  } else {
5486  if ( $mode === 'get' ) {
5487  return $newText;
5488  } else {
5489  return $text;
5490  }
5491  }
5492  }
5493 
5494  # Preprocess the text
5495  $root = $this->preprocessToDom( $text, $flags );
5496 
5497  # <h> nodes indicate section breaks
5498  # They can only occur at the top level, so we can find them by iterating the root's children
5499  $node = $root->getFirstChild();
5500 
5501  # Find the target section
5502  if ( $sectionIndex == 0 ) {
5503  # Section zero doesn't nest, level=big
5504  $targetLevel = 1000;
5505  } else {
5506  while ( $node ) {
5507  if ( $node->getName() === 'h' ) {
5508  $bits = $node->splitHeading();
5509  if ( $bits['i'] == $sectionIndex ) {
5510  $targetLevel = $bits['level'];
5511  break;
5512  }
5513  }
5514  if ( $mode === 'replace' ) {
5515  $outText .= $frame->expand( $node, PPFrame::RECOVER_ORIG );
5516  }
5517  $node = $node->getNextSibling();
5518  }
5519  }
5520 
5521  if ( !$node ) {
5522  # Not found
5523  if ( $mode === 'get' ) {
5524  return $newText;
5525  } else {
5526  return $text;
5527  }
5528  }
5529 
5530  # Find the end of the section, including nested sections
5531  do {
5532  if ( $node->getName() === 'h' ) {
5533  $bits = $node->splitHeading();
5534  $curLevel = $bits['level'];
5535  if ( $bits['i'] != $sectionIndex && $curLevel <= $targetLevel ) {
5536  break;
5537  }
5538  }
5539  if ( $mode === 'get' ) {
5540  $outText .= $frame->expand( $node, PPFrame::RECOVER_ORIG );
5541  }
5542  $node = $node->getNextSibling();
5543  } while ( $node );
5544 
5545  # Write out the remainder (in replace mode only)
5546  if ( $mode === 'replace' ) {
5547  # Output the replacement text
5548  # Add two newlines on -- trailing whitespace in $newText is conventionally
5549  # stripped by the editor, so we need both newlines to restore the paragraph gap
5550  # Only add trailing whitespace if there is newText
5551  if ( $newText != "" ) {
5552  $outText .= $newText . "\n\n";
5553  }
5554 
5555  while ( $node ) {
5556  $outText .= $frame->expand( $node, PPFrame::RECOVER_ORIG );
5557  $node = $node->getNextSibling();
5558  }
5559  }
5560 
5561  if ( is_string( $outText ) ) {
5562  # Re-insert stripped tags
5563  $outText = rtrim( $this->mStripState->unstripBoth( $outText ) );
5564  }
5565 
5566  return $outText;
5567  }
5568 
5583  public function getSection( $text, $sectionId, $defaultText = '' ) {
5584  return $this->extractSections( $text, $sectionId, 'get', $defaultText );
5585  }
5586 
5599  public function replaceSection( $oldText, $sectionId, $newText ) {
5600  return $this->extractSections( $oldText, $sectionId, 'replace', $newText );
5601  }
5602 
5608  public function getRevisionId() {
5609  return $this->mRevisionId;
5610  }
5611 
5618  public function getRevisionObject() {
5619  if ( !is_null( $this->mRevisionObject ) ) {
5620  return $this->mRevisionObject;
5621  }
5622  if ( is_null( $this->mRevisionId ) ) {
5623  return null;
5624  }
5625 
5626  $rev = call_user_func(
5627  $this->mOptions->getCurrentRevisionCallback(), $this->getTitle(), $this
5628  );
5629 
5630  # If the parse is for a new revision, then the callback should have
5631  # already been set to force the object and should match mRevisionId.
5632  # If not, try to fetch by mRevisionId for sanity.
5633  if ( $rev && $rev->getId() != $this->mRevisionId ) {
5634  $rev = Revision::newFromId( $this->mRevisionId );
5635  }
5636 
5637  $this->mRevisionObject = $rev;
5638 
5639  return $this->mRevisionObject;
5640  }
5641 
5647  public function getRevisionTimestamp() {
5648  if ( is_null( $this->mRevisionTimestamp ) ) {
5650 
5651  $revObject = $this->getRevisionObject();
5652  $timestamp = $revObject ? $revObject->getTimestamp() : wfTimestampNow();
5653 
5654  # The cryptic '' timezone parameter tells to use the site-default
5655  # timezone offset instead of the user settings.
5656  # Since this value will be saved into the parser cache, served
5657  # to other users, and potentially even used inside links and such,
5658  # it needs to be consistent for all visitors.
5659  $this->mRevisionTimestamp = $wgContLang->userAdjust( $timestamp, '' );
5660 
5661  }
5662  return $this->mRevisionTimestamp;
5663  }
5664 
5670  public function getRevisionUser() {
5671  if ( is_null( $this->mRevisionUser ) ) {
5672  $revObject = $this->getRevisionObject();
5673 
5674  # if this template is subst: the revision id will be blank,
5675  # so just use the current user's name
5676  if ( $revObject ) {
5677  $this->mRevisionUser = $revObject->getUserText();
5678  } elseif ( $this->ot['wiki'] || $this->mOptions->getIsPreview() ) {
5679  $this->mRevisionUser = $this->getUser()->getName();
5680  }
5681  }
5682  return $this->mRevisionUser;
5683  }
5684 
5690  public function getRevisionSize() {
5691  if ( is_null( $this->mRevisionSize ) ) {
5692  $revObject = $this->getRevisionObject();
5693 
5694  # if this variable is subst: the revision id will be blank,
5695  # so just use the parser input size, because the own substituation
5696  # will change the size.
5697  if ( $revObject ) {
5698  $this->mRevisionSize = $revObject->getSize();
5699  } else {
5700  $this->mRevisionSize = $this->mInputSize;
5701  }
5702  }
5703  return $this->mRevisionSize;
5704  }
5705 
5711  public function setDefaultSort( $sort ) {
5712  $this->mDefaultSort = $sort;
5713  $this->mOutput->setProperty( 'defaultsort', $sort );
5714  }
5715 
5726  public function getDefaultSort() {
5727  if ( $this->mDefaultSort !== false ) {
5728  return $this->mDefaultSort;
5729  } else {
5730  return '';
5731  }
5732  }
5733 
5740  public function getCustomDefaultSort() {
5741  return $this->mDefaultSort;
5742  }
5743 
5753  public function guessSectionNameFromWikiText( $text ) {
5754  # Strip out wikitext links(they break the anchor)
5755  $text = $this->stripSectionName( $text );
5757  return '#' . Sanitizer::escapeId( $text, 'noninitial' );
5758  }
5759 
5768  public function guessLegacySectionNameFromWikiText( $text ) {
5769  # Strip out wikitext links(they break the anchor)
5770  $text = $this->stripSectionName( $text );
5772  return '#' . Sanitizer::escapeId( $text, [ 'noninitial', 'legacy' ] );
5773  }
5774 
5789  public function stripSectionName( $text ) {
5790  # Strip internal link markup
5791  $text = preg_replace( '/\[\[:?([^[|]+)\|([^[]+)\]\]/', '$2', $text );
5792  $text = preg_replace( '/\[\[:?([^[]+)\|?\]\]/', '$1', $text );
5793 
5794  # Strip external link markup
5795  # @todo FIXME: Not tolerant to blank link text
5796  # I.E. [https://www.mediawiki.org] will render as [1] or something depending
5797  # on how many empty links there are on the page - need to figure that out.
5798  $text = preg_replace( '/\[(?i:' . $this->mUrlProtocols . ')([^ ]+?) ([^[]+)\]/', '$2', $text );
5799 
5800  # Parse wikitext quotes (italics & bold)
5801  $text = $this->doQuotes( $text );
5802 
5803  # Strip HTML tags
5804  $text = StringUtils::delimiterReplace( '<', '>', '', $text );
5805  return $text;
5806  }
5807 
5818  public function testSrvus( $text, Title $title, ParserOptions $options,
5819  $outputType = self::OT_HTML
5820  ) {
5821  $magicScopeVariable = $this->lock();
5822  $this->startParse( $title, $options, $outputType, true );
5823 
5824  $text = $this->replaceVariables( $text );
5825  $text = $this->mStripState->unstripBoth( $text );
5826  $text = Sanitizer::removeHTMLtags( $text );
5827  return $text;
5828  }
5829 
5836  public function testPst( $text, Title $title, ParserOptions $options ) {
5837  return $this->preSaveTransform( $text, $title, $options->getUser(), $options );
5838  }
5839 
5846  public function testPreprocess( $text, Title $title, ParserOptions $options ) {
5847  return $this->testSrvus( $text, $title, $options, self::OT_PREPROCESS );
5848  }
5849 
5866  public function markerSkipCallback( $s, $callback ) {
5867  $i = 0;
5868  $out = '';
5869  while ( $i < strlen( $s ) ) {
5870  $markerStart = strpos( $s, self::MARKER_PREFIX, $i );
5871  if ( $markerStart === false ) {
5872  $out .= call_user_func( $callback, substr( $s, $i ) );
5873  break;
5874  } else {
5875  $out .= call_user_func( $callback, substr( $s, $i, $markerStart - $i ) );
5876  $markerEnd = strpos( $s, self::MARKER_SUFFIX, $markerStart );
5877  if ( $markerEnd === false ) {
5878  $out .= substr( $s, $markerStart );
5879  break;
5880  } else {
5881  $markerEnd += strlen( self::MARKER_SUFFIX );
5882  $out .= substr( $s, $markerStart, $markerEnd - $markerStart );
5883  $i = $markerEnd;
5884  }
5885  }
5886  }
5887  return $out;
5888  }
5889 
5896  public function killMarkers( $text ) {
5897  return $this->mStripState->killMarkers( $text );
5898  }
5899 
5916  public function serializeHalfParsedText( $text ) {
5917  $data = [
5918  'text' => $text,
5919  'version' => self::HALF_PARSED_VERSION,
5920  'stripState' => $this->mStripState->getSubState( $text ),
5921  'linkHolders' => $this->mLinkHolders->getSubArray( $text )
5922  ];
5923  return $data;
5924  }
5925 
5941  public function unserializeHalfParsedText( $data ) {
5942  if ( !isset( $data['version'] ) || $data['version'] != self::HALF_PARSED_VERSION ) {
5943  throw new MWException( __METHOD__ . ': invalid version' );
5944  }
5945 
5946  # First, extract the strip state.
5947  $texts = [ $data['text'] ];
5948  $texts = $this->mStripState->merge( $data['stripState'], $texts );
5949 
5950  # Now renumber links
5951  $texts = $this->mLinkHolders->mergeForeign( $data['linkHolders'], $texts );
5952 
5953  # Should be good to go.
5954  return $texts[0];
5955  }
5956 
5966  public function isValidHalfParsedText( $data ) {
5967  return isset( $data['version'] ) && $data['version'] == self::HALF_PARSED_VERSION;
5968  }
5969 
5978  public function parseWidthParam( $value ) {
5979  $parsedWidthParam = [];
5980  if ( $value === '' ) {
5981  return $parsedWidthParam;
5982  }
5983  $m = [];
5984  # (bug 13500) In both cases (width/height and width only),
5985  # permit trailing "px" for backward compatibility.
5986  if ( preg_match( '/^([0-9]*)x([0-9]*)\s*(?:px)?\s*$/', $value, $m ) ) {
5987  $width = intval( $m[1] );
5988  $height = intval( $m[2] );
5989  $parsedWidthParam['width'] = $width;
5990  $parsedWidthParam['height'] = $height;
5991  } elseif ( preg_match( '/^[0-9]*\s*(?:px)?\s*$/', $value ) ) {
5992  $width = intval( $value );
5993  $parsedWidthParam['width'] = $width;
5994  }
5995  return $parsedWidthParam;
5996  }
5997 
6007  protected function lock() {
6008  if ( $this->mInParse ) {
6009  throw new MWException( "Parser state cleared while parsing. "
6010  . "Did you call Parser::parse recursively?" );
6011  }
6012  $this->mInParse = true;
6013 
6014  $recursiveCheck = new ScopedCallback( function() {
6015  $this->mInParse = false;
6016  } );
6017 
6018  return $recursiveCheck;
6019  }
6020 
6031  public static function stripOuterParagraph( $html ) {
6032  $m = [];
6033  if ( preg_match( '/^<p>(.*)\n?<\/p>\n?$/sU', $html, $m ) ) {
6034  if ( strpos( $m[1], '</p>' ) === false ) {
6035  $html = $m[1];
6036  }
6037  }
6038 
6039  return $html;
6040  }
6041 
6052  public function getFreshParser() {
6053  global $wgParserConf;
6054  if ( $this->mInParse ) {
6055  return new $wgParserConf['class']( $wgParserConf );
6056  } else {
6057  return $this;
6058  }
6059  }
6060 
6067  public function enableOOUI() {
6069  $this->mOutput->setEnableOOUI( true );
6070  }
6071 }
getRevisionObject()
Get the revision object for $this->mRevisionId.
Definition: Parser.php:5618
static newFromName($name, $validate= 'valid')
Static factory method for creation from username.
Definition: User.php:525
setTitle($t)
Set the context title.
Definition: Parser.php:739
$mAutonumber
Definition: Parser.php:177
getLatestRevID($flags=0)
What is the page_latest field for this page?
Definition: Title.php:3298
markerSkipCallback($s, $callback)
Call a callback function on all regions of the given text that are not inside strip markers...
Definition: Parser.php:5866
#define the
table suitable for use with IDatabase::select()
$mPPNodeCount
Definition: Parser.php:191
replaceInternalLinks2(&$s)
Process [[ ]] wikilinks (RIL)
Definition: Parser.php:2082
static getVariableIDs()
Get an array of parser variable IDs.
Definition: MagicWord.php:271
you don t have to do a grep find to see where the $wgReverseTitle variable is used
Definition: hooks.txt:117
getExternalLinkAttribs($url)
Get an associative array of additional HTML attributes appropriate for a particular external link...
Definition: Parser.php:1909
const MARKER_PREFIX
Definition: Parser.php:134
deferred txt A few of the database updates required by various functions here can be deferred until after the result page is displayed to the user For updating the view updating the linked to tables after a etc PHP does not yet have any way to tell the server to actually return and disconnect while still running these but it might have such a feature in the future We handle these by creating a deferred update object and putting those objects on a global then executing the whole list after the page is displayed We don t do anything smart like collating updates to the same table or such because the list is almost always going to have just one item on if that
Definition: deferred.txt:11
isValidHalfParsedText($data)
Returns true if the given array, presumed to be generated by serializeHalfParsedText(), is compatible with the current version of the parser.
Definition: Parser.php:5966
null means default in associative array form
Definition: hooks.txt:1936
null means default in associative array with keys and values unescaped Should be merged with default with a value of false meaning to suppress the attribute in associative array with keys and values unescaped noclasses just before the function returns a value If you return an< a > element with HTML attributes $attribs and contents $html will be returned If you return $ret will be returned and may include noclasses & $html
Definition: hooks.txt:1936
static tocLineEnd()
End a Table Of Contents line.
Definition: Linker.php:1633
deferred txt A few of the database updates required by various functions here can be deferred until after the result page is displayed to the user For updating the view updating the linked to tables after a etc PHP does not yet have any way to tell the server to actually return and disconnect while still running these but it might have such a feature in the future We handle these by creating a deferred update object and putting those objects on a global list
Definition: deferred.txt:11
getSection($text, $sectionId, $defaultText= '')
This function returns the text of a section, specified by a number ($section).
Definition: Parser.php:5583
static decodeTagAttributes($text)
Return an associative array of attribute names and values from a partial tag string.
Definition: Sanitizer.php:1287
$mTplRedirCache
Definition: Parser.php:193
killMarkers($text)
Remove any strip markers found in the given text.
Definition: Parser.php:5896
wfGetDB($db, $groups=[], $wiki=false)
Get a Database object.
static tocList($toc, $lang=false)
Wraps the TOC in a table and provides the hide/collapse javascript.
Definition: Linker.php:1645
LinkRenderer $mLinkRenderer
Definition: Parser.php:257
fetchTemplateAndTitle($title)
Fetch the unparsed text of a template and register a reference to it.
Definition: Parser.php:3502
this hook is for auditing only or null if authentication failed before getting that far or null if we can t even determine that probably a stub it is not rendered in wiki pages or galleries in category pages allow injecting custom HTML after the section Any uses of the hook need to handle escaping see BaseTemplate::getToolbox and BaseTemplate::makeListItem for details on the format of individual items inside of this array or by returning and letting standard HTTP rendering take place modifiable or by returning false and taking over the output $out
Definition: hooks.txt:802
getRevisionUser()
Get the name of the user that edited the last revision.
Definition: Parser.php:5670
setFunctionTagHook($tag, $callback, $flags)
Create a tag function, e.g.
Definition: Parser.php:4863
the array() calling protocol came about after MediaWiki 1.4rc1.
stripSectionName($text)
Strips a text string of wikitext for use in a section anchor.
Definition: Parser.php:5789
const OT_PREPROCESS
Definition: Defines.php:190
either a plain
Definition: hooks.txt:1987
$mDoubleUnderscores
Definition: Parser.php:193
Group all the pieces relevant to the context of a request into one instance.
getPreloadText($text, Title $title, ParserOptions $options, $params=[])
Process the wikitext for the "?preload=" feature.
Definition: Parser.php:687
$context
Definition: load.php:50
validateSig($text)
Check that the user's signature contains no bad XML.
Definition: Parser.php:4577
MapCacheLRU null $currentRevisionCache
Definition: Parser.php:243
getArticleID($flags=0)
Get the article ID for this Title from the link cache, adding it if necessary.
Definition: Title.php:3209
$wgSitename
Name of the site.
renderImageGallery($text, $params)
Renders an image gallery from a text with one line per image.
Definition: Parser.php:4914
recursivePreprocess($text, $frame=false)
Recursive parser entry point that can be called from an extension tag hook.
Definition: Parser.php:668
replaceExternalLinks($text)
Replace external links (REL)
Definition: Parser.php:1812
processing should stop and the error should be shown to the user * false
Definition: hooks.txt:189
static isNonincludable($index)
It is not possible to use pages from this namespace as template?
nextLinkID()
Definition: Parser.php:828
const SPACE_NOT_NL
Definition: Parser.php:103
null means default in associative array with keys and values unescaped Should be merged with default with a value of false meaning to suppress the attribute in associative array with keys and values unescaped noclasses & $ret
Definition: hooks.txt:1936
static replaceUnusualEscapes($url)
Replace unusual escape codes in a URL with their equivalent characters.
Definition: Parser.php:1937
getImageParams($handler)
Definition: Parser.php:5074
Apache License January AND DISTRIBUTION Definitions License shall mean the terms and conditions for use
doHeadings($text)
Parse headers and return html.
Definition: Parser.php:1591
static getTitleFor($name, $subpage=false, $fragment= '')
Get a localised Title object for a specified special page name If you don't need a full Title object...
Definition: SpecialPage.php:82
const OT_PLAIN
Definition: Parser.php:114
getTags()
Accessor.
Definition: Parser.php:5389
static isWellFormedXmlFragment($text)
Check if a string is a well-formed XML fragment.
Definition: Xml.php:735
const OT_WIKI
Definition: Parser.php:111
div flags Integer display flags(NO_ACTION_LINK, NO_EXTRA_USER_LINKS) 'LogException'returning false will NOT prevent logging $e
Definition: hooks.txt:2102
fetchFileAndTitle($title, $options=[])
Fetch a file and its title and register a reference to it.
Definition: Parser.php:3644
User $mUser
Definition: Parser.php:200
We use the convention $dbr for read and $dbw for write to help you keep track of whether the database object is a the world will explode Or to be a subsequent write query which succeeded on the master may fail when replicated to the slave due to a unique key collision Replication on the slave will stop and it may take hours to repair the database and get it back online Setting read_only in my cnf on the slave will avoid this but given the dire we prefer to have as many checks as possible We provide a but the wrapper functions like please read the documentation for except in special pages derived from QueryPage It s a common pitfall for new developers to submit code containing SQL queries which examine huge numbers of rows Remember that COUNT * is(N), counting rows in atable is like counting beans in a bucket.------------------------------------------------------------------------Replication------------------------------------------------------------------------The largest installation of MediaWiki, Wikimedia, uses a large set ofslave MySQL servers replicating writes made to a master MySQL server.Itis important to understand the issues associated with this setup if youwant to write code destined for Wikipedia.It's often the case that the best algorithm to use for a given taskdepends on whether or not replication is in use.Due to our unabashedWikipedia-centrism, we often just use the replication-friendly version, but if you like, you can use wfGetLB() ->getServerCount() > 1 tocheck to see if replication is in use.===Lag===Lag primarily occurs when large write queries are sent to the master.Writes on the master are executed in parallel, but they are executed inserial when they are replicated to the slaves.The master writes thequery to the binlog when the transaction is committed.The slaves pollthe binlog and start executing the query as soon as it appears.They canservice reads while they are performing a write query, but will not readanything more from the binlog and thus will perform no more writes.Thismeans that if the write query runs for a long time, the slaves will lagbehind the master for the time it takes for the write query to complete.Lag can be exacerbated by high read load.MediaWiki's load balancer willstop sending reads to a slave when it is lagged by more than 30 seconds.If the load ratios are set incorrectly, or if there is too much loadgenerally, this may lead to a slave permanently hovering around 30seconds lag.If all slaves are lagged by more than 30 seconds, MediaWiki will stopwriting to the database.All edits and other write operations will berefused, with an error returned to the user.This gives the slaves achance to catch up.Before we had this mechanism, the slaves wouldregularly lag by several minutes, making review of recent editsdifficult.In addition to this, MediaWiki attempts to ensure that the user seesevents occurring on the wiki in chronological order.A few seconds of lagcan be tolerated, as long as the user sees a consistent picture fromsubsequent requests.This is done by saving the master binlog positionin the session, and then at the start of each request, waiting for theslave to catch up to that position before doing any reads from it.Ifthis wait times out, reads are allowed anyway, but the request isconsidered to be in"lagged slave mode".Lagged slave mode can bechecked by calling wfGetLB() ->getLaggedSlaveMode().The onlypractical consequence at present is a warning displayed in the pagefooter.===Lag avoidance===To avoid excessive lag, queries which write large numbers of rows shouldbe split up, generally to write one row at a time.Multi-row INSERT...SELECT queries are the worst offenders should be avoided altogether.Instead do the select first and then the insert.===Working with lag===Despite our best efforts, it's not practical to guarantee a low-lagenvironment.Lag will usually be less than one second, but mayoccasionally be up to 30 seconds.For scalability, it's very importantto keep load on the master low, so simply sending all your queries tothe master is not the answer.So when you have a genuine need forup-to-date data, the following approach is advised:1) Do a quick query to the master for a sequence number or timestamp 2) Run the full query on the slave and check if it matches the data you gotfrom the master 3) If it doesn't, run the full query on the masterTo avoid swamping the master every time the slaves lag, use of thisapproach should be kept to a minimum.In most cases you should just readfrom the slave and let the user deal with the delay.------------------------------------------------------------------------Lock contention------------------------------------------------------------------------Due to the high write rate on Wikipedia(and some other wikis), MediaWiki developers need to be very careful to structure their writesto avoid long-lasting locks.By default, MediaWiki opens a transactionat the first query, and commits it before the output is sent.Locks willbe held from the time when the query is done until the commit.So youcan reduce lock time by doing as much processing as possible before youdo your write queries.Often this approach is not good enough, and it becomes necessary toenclose small groups of queries in their own transaction.Use thefollowing syntax:$dbw=wfGetDB(DB_MASTER
initialiseVariables()
initialise the magic variables (like CURRENTMONTHNAME) and substitution modifiers ...
Definition: Parser.php:2820
static isEnabled()
Definition: MWTidy.php:79
Set options of the Parser.
static tidy($text)
Interface with html tidy.
Definition: MWTidy.php:46
getFunctionHooks()
Get all registered function hook identifiers.
Definition: Parser.php:4849
static fixTagAttributes($text, $element, $sorted=false)
Take a tag soup fragment listing an HTML element's attributes and normalize it to well-formed XML...
Definition: Sanitizer.php:1071
globals txt Globals are evil The original MediaWiki code relied on globals for processing context far too often MediaWiki development since then has been a story of slowly moving context out of global variables and into objects Storing processing context in object member variables allows those objects to be reused in a much more flexible way Consider the elegance of
database rows
Definition: globals.txt:10
wfHostname()
Fetch server name for use in error reporting etc.
getFunctionLang()
Get a language object for use in parser functions such as {{FORMATNUM:}}.
Definition: Parser.php:843
argSubstitution($piece, $frame)
Triple brace replacement – used for template arguments.
Definition: Parser.php:3747
testSrvus($text, Title $title, ParserOptions $options, $outputType=self::OT_HTML)
strip/replaceVariables/unstrip for preprocessor regression testing
Definition: Parser.php:5818
uniqPrefix()
Accessor for mUniqPrefix.
Definition: Parser.php:729
const TOC_START
Definition: Parser.php:137
Title($x=null)
Accessor/mutator for the Title object.
Definition: Parser.php:767
SectionProfiler $mProfiler
Definition: Parser.php:252
$sort
fetchFileNoRegister($title, $options=[])
Helper function for fetchFileAndTitle.
Definition: Parser.php:3669
null for the local wiki Added in
Definition: hooks.txt:1555
There are three types of nodes:
$mHeadings
Definition: Parser.php:193
$value
clearTagHooks()
Remove all tag hooks.
Definition: Parser.php:4762
static makeSelfLinkObj($nt, $html= '', $query= '', $trail= '', $prefix= '')
Make appropriate markup for a link to the current article.
Definition: Linker.php:277
const NS_SPECIAL
Definition: Defines.php:45
clearState()
Clear Parser state.
Definition: Parser.php:341
this hook is for auditing only RecentChangesLinked and Watchlist RecentChangesLinked and Watchlist e g Watchlist removed from all revisions and log entries to which it was applied This gives extensions a chance to take it off their books as the deletion has already been partly carried out by this point or something similar the user will be unable to create the tag set and then return false from the hook function Ensure you consume the ChangeTagAfterDelete hook to carry out custom deletion actions as context called by AbstractContent::getParserOutput May be used to override the normal model specific rendering of page content as context $revId
Definition: hooks.txt:1046
__construct($conf=[])
Definition: Parser.php:262
const EXT_LINK_ADDR
Definition: Parser.php:95
$mFirstCall
Definition: Parser.php:152
interwikiTransclude($title, $action)
Transclude an interwiki link.
Definition: Parser.php:3688
pstPass2($text, $user)
Pre-save transform helper function.
Definition: Parser.php:4452
guessLegacySectionNameFromWikiText($text)
Same as guessSectionNameFromWikiText(), but produces legacy anchors instead.
Definition: Parser.php:5768
injection txt This is an overview of how MediaWiki makes use of dependency injection The design described here grew from the discussion of RFC T384 The term dependency this means that anything an object needs to operate should be injected from the the object itself should only know narrow no concrete implementation of the logic it relies on The requirement to inject everything typically results in an architecture that based on two main types of and essentially stateless service objects that use other service objects to operate on the value objects As of the beginning MediaWiki is only starting to use the DI approach Much of the code still relies on global state or direct resulting in a highly cyclical dependency MediaWikiServices
Definition: injection.txt:23
wfUrlProtocolsWithoutProtRel()
Like wfUrlProtocols(), but excludes '//' from the protocol list.
Options($x=null)
Accessor/mutator for the ParserOptions object.
Definition: Parser.php:821
it s the revision text itself In either if gzip is the revision text is gzipped $flags
Definition: hooks.txt:2703
serializeHalfParsedText($text)
Save the parser state required to convert the given half-parsed text to HTML.
Definition: Parser.php:5916
replaceLinkHolders(&$text, $options=0)
Replace "" link placeholders with actual links, in the buffer Placeholders created in Link...
Definition: Parser.php:4886
static statelessFetchRevision(Title $title, $parser=false)
Wrapper around Revision::newFromTitle to allow passing additional parameters without passing them on ...
Definition: Parser.php:3485
static activeUsers()
Definition: SiteStats.php:165
$mLinkID
Definition: Parser.php:190
doQuotes($text)
Helper function for doAllQuotes()
Definition: Parser.php:1624
preprocessToDom($text, $flags=0)
Preprocess some wikitext and return the document tree.
Definition: Parser.php:2850
limitationWarn($limitationType, $current= '', $max= '')
Warn the user when a parser limitation is reached Will warn at most once the user per limitation type...
Definition: Parser.php:2972
static cleanUrl($url)
Definition: Sanitizer.php:1856
wfUrlencode($s)
We want some things to be included as literal characters in our title URLs for prettiness, which urlencode encodes by default.
static newFromText($text, $defaultNamespace=NS_MAIN)
Create a new Title from text, such as what one would find in a link.
Definition: Title.php:262
$mGeneratedPPNodeCount
Definition: Parser.php:191
static getRandomString()
Get a random string.
Definition: Parser.php:708
$mRevisionId
Definition: Parser.php:217
static stripAllTags($text)
Take a fragment of (potentially invalid) HTML and return a version with any tags removed, encoded as plain text.
Definition: Sanitizer.php:1823
when a variable name is used in a it is silently declared as a new local masking the global
Definition: design.txt:93
doBlockLevels($text, $linestart)
Make lists from lines starting with ':', '*', '#', etc.
Definition: Parser.php:2453
$wgArticlePath
Definition: img_auth.php:45
OutputType($x=null)
Accessor/mutator for the output type.
Definition: Parser.php:793
getLinkRenderer()
Get a LinkRenderer instance to make links with.
Definition: Parser.php:910
const NS_TEMPLATE
Definition: Defines.php:66
static newFromTitle(LinkTarget $linkTarget, $id=0, $flags=0)
Load either the current, or a specified, revision that's attached to a given link target...
Definition: Revision.php:128
getVariableValue($index, $frame=false)
Return value of a magic variable (like PAGENAME)
Definition: Parser.php:2468
recursiveTagParse($text, $frame=false)
Half-parse wikitext to half-parsed HTML.
Definition: Parser.php:603
const NO_ARGS
magic word & $parser
Definition: hooks.txt:2487
MagicWordArray $mVariables
Definition: Parser.php:159
this hook is for auditing only RecentChangesLinked and Watchlist RecentChangesLinked and Watchlist e g Watchlist removed from all revisions and log entries to which it was applied This gives extensions a chance to take it off their books as the deletion has already been partly carried out by this point or something similar the user will be unable to create the tag set and then return false from the hook function Ensure you consume the ChangeTagAfterDelete hook to carry out custom deletion actions as context called by AbstractContent::getParserOutput May be used to override the normal model specific rendering of page content as context as context the output can only depend on parameters provided to this hook not on global state indicating whether full HTML should be generated If generation of HTML may be but other information should still be present in the ParserOutput object to manipulate or replace but no entry for that model exists in $wgContentHandlers if desired whether it is OK to use $contentModel on $title Handler functions that modify $ok should generally return false to prevent further hooks from further modifying $ok inclusive false for true for descending in case the handler function wants to provide a converted Content object Note that $result getContentModel() must return $toModel. 'CustomEditor'$rcid is used in generating this variable which contains information about the new revision
Definition: hooks.txt:1156
static validateTagAttributes($attribs, $element)
Take an array of attribute names and values and normalize or discard illegal values for the given ele...
Definition: Sanitizer.php:748
const SFH_NO_HASH
Definition: Parser.php:85
const DB_MASTER
Definition: defines.php:23
globals will be eliminated from MediaWiki replaced by an application object which would be passed to constructors Whether that would be an convenient solution remains to be but certainly PHP makes such object oriented programming models easier than they were in previous versions For the time being MediaWiki programmers will have to work in an environment with some global context At the time of globals were initialised on startup by MediaWiki of these were configuration which are documented in DefaultSettings php There is no comprehensive documentation for the remaining however some of the most important ones are listed below They are typically initialised either in index php or in Setup php For a description of the see design txt $wgTitle Title object created from the request URL $wgOut OutputPage object for HTTP response $wgUser User object for the user associated with the current request $wgLang Language object selected by user preferences $wgContLang Language object associated with the wiki being viewed $wgParser Parser object Parser extensions register their hooks here $wgRequest WebRequest object
Definition: globals.txt:25
wfRandomString($length=32)
Get a random string containing a number of pseudo-random hex characters.
$mForceTocPosition
Definition: Parser.php:195
preprocess($text, Title $title=null, ParserOptions $options, $revid=null, $frame=false)
Expand templates and variables in the text, producing valid, static wikitext.
Definition: Parser.php:644
static getCacheTTL($id)
Allow external reads of TTL array.
Definition: MagicWord.php:294
getRevisionId()
Get the ID of the revision we are parsing.
Definition: Parser.php:5608
const OT_PREPROCESS
Definition: Parser.php:112
maybeDoSubpageLink($target, &$text)
Handle link to subpage if necessary.
Definition: Parser.php:2441
$mFunctionSynonyms
Definition: Parser.php:144
If you want to remove the page from your watchlist later
replaceLinkHoldersText($text)
Replace "" link placeholders with plain text of links (not HTML-formatted).
Definition: Parser.php:4897
setLinkID($id)
Definition: Parser.php:835
$mOutputType
Definition: Parser.php:214
wfDebug($text, $dest= 'all', array $context=[])
Sends a line to the debug log if enabled or, optionally, to a comment in output.
$mDefaultStripList
Definition: Parser.php:147
static createAssocArgs($args)
Clean up argument array - refactored in 1.9 so parserfunctions can use it, too.
Definition: Parser.php:2924
$mExtLinkBracketedRegex
Definition: Parser.php:166
The index of the header message $result[1]=The index of the body text message $result[2 through n]=Parameters passed to body text message.Please note the header message cannot receive/use parameters. 'ImportHandleLogItemXMLTag':When parsing a XML tag in a log item.Return false to stop further processing of the tag $reader:XMLReader object $logInfo:Array of information 'ImportHandlePageXMLTag':When parsing a XML tag in a page.Return false to stop further processing of the tag $reader:XMLReader object &$pageInfo:Array of information 'ImportHandleRevisionXMLTag':When parsing a XML tag in a page revision.Return false to stop further processing of the tag $reader:XMLReader object $pageInfo:Array of page information $revisionInfo:Array of revision information 'ImportHandleToplevelXMLTag':When parsing a top level XML tag.Return false to stop further processing of the tag $reader:XMLReader object 'ImportHandleUploadXMLTag':When parsing a XML tag in a file upload.Return false to stop further processing of the tag $reader:XMLReader object $revisionInfo:Array of information 'ImportLogInterwikiLink':Hook to change the interwiki link used in log entries and edit summaries for transwiki imports.&$fullInterwikiPrefix:Interwiki prefix, may contain colons.&$pageTitle:String that contains page title. 'ImportSources':Called when reading from the $wgImportSources configuration variable.Can be used to lazy-load the import sources list.&$importSources:The value of $wgImportSources.Modify as necessary.See the comment in DefaultSettings.php for the detail of how to structure this array. 'InfoAction':When building information to display on the action=info page.$context:IContextSource object &$pageInfo:Array of information 'InitializeArticleMaybeRedirect':MediaWiki check to see if title is a redirect.&$title:Title object for the current page &$request:WebRequest &$ignoreRedirect:boolean to skip redirect check &$target:Title/string of redirect target &$article:Article object 'InternalParseBeforeLinks':during Parser's internalParse method before links but after nowiki/noinclude/includeonly/onlyinclude and other processings.&$parser:Parser object &$text:string containing partially parsed text &$stripState:Parser's internal StripState object 'InternalParseBeforeSanitize':during Parser's internalParse method just before the parser removes unwanted/dangerous HTML tags and after nowiki/noinclude/includeonly/onlyinclude and other processings.Ideal for syntax-extensions after template/parser function execution which respect nowiki and HTML-comments.&$parser:Parser object &$text:string containing partially parsed text &$stripState:Parser's internal StripState object 'InterwikiLoadPrefix':When resolving if a given prefix is an interwiki or not.Return true without providing an interwiki to continue interwiki search.$prefix:interwiki prefix we are looking for.&$iwData:output array describing the interwiki with keys iw_url, iw_local, iw_trans and optionally iw_api and iw_wikiid. 'InvalidateEmailComplete':Called after a user's email has been invalidated successfully.$user:user(object) whose email is being invalidated 'IRCLineURL':When constructing the URL to use in an IRC notification.Callee may modify $url and $query, URL will be constructed as $url.$query &$url:URL to index.php &$query:Query string $rc:RecentChange object that triggered url generation 'IsFileCacheable':Override the result of Article::isFileCacheable()(if true) &$article:article(object) being checked 'IsTrustedProxy':Override the result of IP::isTrustedProxy() &$ip:IP being check &$result:Change this value to override the result of IP::isTrustedProxy() 'IsUploadAllowedFromUrl':Override the result of UploadFromUrl::isAllowedUrl() $url:URL used to upload from &$allowed:Boolean indicating if uploading is allowed for given URL 'isValidEmailAddr':Override the result of Sanitizer::validateEmail(), for instance to return false if the domain name doesn't match your organization.$addr:The e-mail address entered by the user &$result:Set this and return false to override the internal checks 'isValidPassword':Override the result of User::isValidPassword() $password:The password entered by the user &$result:Set this and return false to override the internal checks $user:User the password is being validated for 'Language::getMessagesFileName':$code:The language code or the language we're looking for a messages file for &$file:The messages file path, you can override this to change the location. 'LanguageGetMagic':DEPRECATED!Use $magicWords in a file listed in $wgExtensionMessagesFiles instead.Use this to define synonyms of magic words depending of the language &$magicExtensions:associative array of magic words synonyms $lang:language code(string) 'LanguageGetNamespaces':Provide custom ordering for namespaces or remove namespaces.Do not use this hook to add namespaces.Use CanonicalNamespaces for that.&$namespaces:Array of namespaces indexed by their numbers 'LanguageGetSpecialPageAliases':DEPRECATED!Use $specialPageAliases in a file listed in $wgExtensionMessagesFiles instead.Use to define aliases of special pages names depending of the language &$specialPageAliases:associative array of magic words synonyms $lang:language code(string) 'LanguageGetTranslatedLanguageNames':Provide translated language names.&$names:array of language code=> language name $code:language of the preferred translations 'LanguageLinks':Manipulate a page's language links.This is called in various places to allow extensions to define the effective language links for a page.$title:The page's Title.&$links:Associative array mapping language codes to prefixed links of the form"language:title".&$linkFlags:Associative array mapping prefixed links to arrays of flags.Currently unused, but planned to provide support for marking individual language links in the UI, e.g.for featured articles. 'LanguageSelector':Hook to change the language selector available on a page.$out:The output page.$cssClassName:CSS class name of the language selector. 'LinkBegin':DEPRECATED!Use HtmlPageLinkRendererBegin instead.Used when generating internal and interwiki links in Linker::link(), before processing starts.Return false to skip default processing and return $ret.See documentation for Linker::link() for details on the expected meanings of parameters.$skin:the Skin object $target:the Title that the link is pointing to &$html:the contents that the< a > tag should have(raw HTML) $result
Definition: hooks.txt:1934
const TS_UNIX
Unix time - the number of seconds since 1970-01-01 00:00:00 UTC.
Definition: defines.php:6
if($line===false) $args
Definition: cdb.php:64
static getLocalInstance($ts=false)
Get a timestamp instance in the server local timezone ($wgLocaltimezone)
static getDoubleUnderscoreArray()
Get a MagicWordArray of double-underscore entities.
Definition: MagicWord.php:307
static splitTrail($trail)
Split a link trail, return the "inside" portion and the remainder of the trail as a two-element array...
Definition: Linker.php:1720
getTemplateDom($title)
Get the semi-parsed DOM representation of a template with a given title, and its redirect destination...
Definition: Parser.php:3419
usually copyright or history_copyright This message must be in HTML not wikitext & $link
Definition: hooks.txt:2889
static decodeCharReferences($text)
Decode any character references, numeric or named entities, in the text and return a UTF-8 string...
Definition: Sanitizer.php:1500
cleanSig($text, $parsing=false)
Clean up signature text.
Definition: Parser.php:4591
wfTimestamp($outputtype=TS_UNIX, $ts=0)
Get a timestamp string in one of various formats.
static factory($mode=false, IContextSource $context=null)
Get a new image gallery.
$wgLanguageCode
Site language code.
Custom PHP profiler for parser/DB type section names that xhprof/xdebug can't handle.
static getPage($name)
Find the object with a given name and return it (or NULL)
static edits()
Definition: SiteStats.php:133
$wgExtraInterlanguageLinkPrefixes
List of additional interwiki prefixes that should be treated as interlanguage links (i...
startExternalParse(Title $title=null, ParserOptions $options, $outputType, $clearState=true)
Set up some variables which are usually set up in parse() so that an external function can call some ...
Definition: Parser.php:4642
wfDebugLog($logGroup, $text, $dest= 'all', array $context=[])
Send a line to a supplementary debug log file, if configured, or main debug log if not...
const NO_TEMPLATES
addTrackingCategory($msg)
Definition: Parser.php:3989
replaceInternalLinks($s)
Process [[ ]] wikilinks.
Definition: Parser.php:2069
$mVarCache
Definition: Parser.php:148
$wgStylePath
The URL path of the skins directory.
disableCache()
Set a flag in the output object indicating that the content is dynamic and shouldn't be cached...
Definition: Parser.php:5361
$mRevisionObject
Definition: Parser.php:216
static normalizeSectionNameWhitespace($section)
Normalizes whitespace in a section name, such as might be returned by Parser::stripSectionName(), for use in the id's that are used for section links.
Definition: Sanitizer.php:1381
internalParse($text, $isMain=true, $frame=false)
Helper function for parse() that transforms wiki markup into half-parsed HTML.
Definition: Parser.php:1243
Title $mTitle
Definition: Parser.php:213
static delimiterReplace($startDelim, $endDelim, $replace, $subject, $flags= '')
Perform an operation equivalent to preg_replace() with flags.
__destruct()
Reduce memory usage to reduce the impact of circular references.
Definition: Parser.php:288
wfEscapeWikiText($text)
Escapes the given text so that it may be output using addWikiText() without any linking, formatting, etc.
getRevisionTimestamp()
Get the timestamp associated with the current revision, adjusted for the default server-local timesta...
Definition: Parser.php:5647
static stripOuterParagraph($html)
Strip outer.
Definition: Parser.php:6031
static register($parser)
$mRevIdForTs
Definition: Parser.php:221
static singleton()
Get an instance of this class.
Definition: LinkCache.php:64
design txt This is a brief overview of the new design More thorough and up to date information is available on the documentation wiki at etc Handles the details of getting and saving to the user table of the and dealing with sessions and cookies OutputPage Encapsulates the entire HTML page that will be sent in response to any server request It is used by calling its functions to add in any and then calling but I prefer the flexibility This should also do the output encoding The system allocates a global one in $wgOut Title Represents the title of an and does all the work of translating among various forms such as plain database key
Definition: design.txt:25
static normalizeSubpageLink($contextTitle, $target, &$text)
Definition: Linker.php:1439
parseWidthParam($value)
Parsed a width param of imagelink like 300px or 200x300px.
Definition: Parser.php:5978
$mStripList
Definition: Parser.php:146
$mFunctionTagHooks
Definition: Parser.php:145
fetchScaryTemplateMaybeFromCache($url)
Definition: Parser.php:3707
const OT_PLAIN
Definition: Defines.php:192
fetchCurrentRevisionOfTitle($title)
Fetch the current revision of a given title.
Definition: Parser.php:3462
$mRevisionTimestamp
Definition: Parser.php:218
$mImageParams
Definition: Parser.php:149
stripAltText($caption, $holders)
Definition: Parser.php:5337
doAllQuotes($text)
Replace single quotes with HTML markup.
Definition: Parser.php:1607
either a unescaped string or a HtmlArmor object after in associative array form externallinks including delete and has completed for all link tables whether this was an auto creation default is conds Array Extra conditions for the No matching items in log is displayed if loglist is empty msgKey Array If you want a nice box with a set this to the key of the message First element is the message additional optional elements are parameters for the key that are processed with wfMessage() -> params() ->parseAsBlock()-offset Set to overwrite offset parameter in $wgRequest set to ''to unsetoffset-wrap String Wrap the message in html(usually something like"&lt
static replaceMarkup($search, $replace, $text)
More or less "markup-safe" str_replace() Ignores any instances of the separator inside <...
static normalizeUrlComponent($component, $unsafe)
Definition: Parser.php:1987
if($limit) $timestamp
const VERSION
Update this version number when the ParserOutput format changes in an incompatible way...
Definition: Parser.php:76
this hook is for auditing only RecentChangesLinked and Watchlist RecentChangesLinked and Watchlist e g Watchlist removed from all revisions and log entries to which it was applied This gives extensions a chance to take it off their books as the deletion has already been partly carried out by this point or something similar the user will be unable to create the tag set and then return false from the hook function Ensure you consume the ChangeTagAfterDelete hook to carry out custom deletion actions as context called by AbstractContent::getParserOutput May be used to override the normal model specific rendering of page content as context as context $options
Definition: hooks.txt:1046
setHook($tag, $callback)
Create an HTML-style tag, e.g.
Definition: Parser.php:4717
const OT_WIKI
Definition: Defines.php:189
Preprocessor $mPreprocessor
Definition: Parser.php:170
getPreprocessor()
Get a preprocessor object.
Definition: Parser.php:896
This document is intended to provide useful advice for parties seeking to redistribute MediaWiki to end users It s targeted particularly at maintainers for Linux since it s been observed that distribution packages of MediaWiki often break We ve consistently had to recommend that users seeking support use official tarballs instead of their distribution s and this often solves whatever problem the user is having It would be nice if this could such and we might be restricted by PHP settings such as safe mode or open_basedir We cannot assume that the software even has read access anywhere useful Many shared hosts run all users web applications under the same so they can t rely on Unix and must forbid reads to even standard directories like tmp lest users read each others files We cannot assume that the user has the ability to install or run any programs not written as web accessible PHP scripts Since anything that works on cheap shared hosting will work if you have shell or root access MediaWiki s design is based around catering to the lowest common denominator Although we support higher end setups as the way many things work by default is tailored toward shared hosting These defaults are unconventional from the point of view of normal(non-web) applications--they might conflict with distributors'policies
static getInstance($ts=false)
Get a timestamp instance in GMT.
Definition: MWTimestamp.php:38
const NS_MEDIA
Definition: Defines.php:44
static singleton()
Get a RepoGroup instance.
Definition: RepoGroup.php:59
replaceVariables($text, $frame=false, $argsOnly=false)
Replace magic variables, templates, and template arguments with the appropriate text.
Definition: Parser.php:2895
const RECOVER_ORIG
wfMatchesDomainList($url, $domains)
Check whether a given URL has a domain that occurs in a given set of domains.
StripState $mStripState
Definition: Parser.php:182
$mDefaultSort
Definition: Parser.php:192
getUser()
Get a User object either from $this->mUser, if set, or from the ParserOptions object otherwise...
Definition: Parser.php:884
wfTimestampNow()
Convenience function; returns MediaWiki timestamp for the present time.
incrementIncludeSize($type, $size)
Increment an include size counter.
Definition: Parser.php:3907
getStripList()
Get a list of strippable XML-like elements.
Definition: Parser.php:1013
const EXT_IMAGE_REGEX
Definition: Parser.php:98
startParse(Title $title=null, ParserOptions $options, $outputType, $clearState=true)
Definition: Parser.php:4654
$params
const NS_CATEGORY
Definition: Defines.php:70
static makeHeadline($level, $attribs, $anchor, $html, $link, $legacyAnchor=false)
Create a headline for content.
Definition: Linker.php:1701
static extractTagsAndParams($elements, $text, &$matches, $uniq_prefix=null)
Replaces all occurrences of HTML-style comments and the given tags in the text with a random marker a...
Definition: Parser.php:943
and(b) You must cause any modified files to carry prominent notices stating that You changed the files
doTableStuff($text)
parse the wiki syntax used to render tables
Definition: Parser.php:1040
wfDeprecated($function, $version=false, $component=false, $callerOffset=2)
Throws a warning that $function is deprecated.
getRevisionSize()
Get the size of the revision.
Definition: Parser.php:5690
$mImageParamsMagicArray
Definition: Parser.php:150
LinkHolderArray $mLinkHolders
Definition: Parser.php:188
const TS_MW
MediaWiki concatenated string timestamp (YYYYMMDDHHMMSS)
Definition: defines.php:11
static register($parser)
null means default in associative array with keys and values unescaped Should be merged with default with a value of false meaning to suppress the attribute in associative array with keys and values unescaped noclasses just before the function returns a value If you return an< a > element with HTML attributes $attribs and contents $html will be returned If you return $ret will be returned and may include noclasses after processing & $attribs
Definition: hooks.txt:1936
deferred txt A few of the database updates required by various functions here can be deferred until after the result page is displayed to the user For updating the view updating the linked to tables after a save
Definition: deferred.txt:4
as see the revision history and available at free of to any person obtaining a copy of this software and associated documentation to deal in the Software without including without limitation the rights to and or sell copies of the and to permit persons to whom the Software is furnished to do so
Definition: LICENSE.txt:10
Some information about database access in MediaWiki By Tim January Database layout For information about the MediaWiki database such as a description of the tables and their please see
Definition: database.txt:2
preSaveTransform($text, Title $title, User $user, ParserOptions $options, $clearState=true)
Transform wiki markup when saving a page by doing "\\r\\n" -> "\\n" conversion, substituting signatur...
Definition: Parser.php:4420
static capturePath(Title $title, IContextSource $context, LinkRenderer $linkRenderer=null)
Just like executePath() but will override global variables and execute the page in "inclusion" mode...
getTargetLanguage()
Get the target language for the content being parsed.
Definition: Parser.php:856
$buffer
namespace and then decline to actually register it file or subcat img or subcat $title
Definition: hooks.txt:953
static newKnownCurrent(IDatabase $db, $pageId, $revId)
Load a revision based on a known page ID and current revision ID from the DB.
Definition: Revision.php:1905
static hasSubpages($index)
Does the namespace allow subpages?
formatHeadings($text, $origText, $isMain=true)
This function accomplishes several tasks: 1) Auto-number headings if that option is enabled 2) Add an...
Definition: Parser.php:4009
getConverterLanguage()
Get the language object for language conversion.
Definition: Parser.php:874
static tocUnindent($level)
Finish one or more sublevels on the Table of Contents.
Definition: Linker.php:1600
static run($event, array $args=[], $deprecatedVersion=null)
Call hook functions defined in Hooks::register and $wgHooks.
Definition: Hooks.php:131
static tocLine($anchor, $tocline, $tocnumber, $level, $sectionIndex=false)
parameter level defines if we are on an indentation level
Definition: Linker.php:1615
design txt This is a brief overview of the new design More thorough and up to date information is available on the documentation wiki at etc Handles the details of getting and saving to the user table of the and dealing with sessions and cookies OutputPage Encapsulates the entire HTML page that will be sent in response to any server request It is used by calling its functions to add text
Definition: design.txt:12
$mInputSize
Definition: Parser.php:222
magicword txt Magic Words are some phrases used in the wikitext They are used for two things
Definition: magicword.txt:4
this hook is for auditing only RecentChangesLinked and Watchlist RecentChangesLinked and Watchlist e g Watchlist removed from all revisions and log entries to which it was applied This gives extensions a chance to take it off their books $tag
Definition: hooks.txt:1007
getUserSig(&$user, $nickname=false, $fancySig=null)
Fetch the user's signature text, if any, and normalize to validated, ready-to-insert wikitext...
Definition: Parser.php:4528
const HALF_PARSED_VERSION
Update this version number when the output of serialiseHalfParsedText() changes in an incompatible wa...
Definition: Parser.php:82
const NS_FILE
Definition: Defines.php:62
firstCallInit()
Do various kinds of initialisation on the first call of the parser.
Definition: Parser.php:323
static makeImageLink(Parser $parser, Title $title, $file, $frameParams=[], $handlerParams=[], $time=false, $query="", $widthOption=null)
Given parameters derived from [[Image:Foo|options...]], generate the HTML that that syntax inserts in...
Definition: Linker.php:415
const PTD_FOR_INCLUSION
Definition: Parser.php:106
null means default in associative array with keys and values unescaped Should be merged with default with a value of false meaning to suppress the attribute in associative array with keys and values unescaped broken
Definition: hooks.txt:1936
armorLinks($text)
Insert a NOPARSE hacky thing into any inline links in a chunk that's going to go through further pars...
Definition: Parser.php:2419
presenting them properly to the user as errors is done by the caller return true use this to change the list i e etc $rev
Definition: hooks.txt:1721
static splitWhitespace($s)
Return a three-element array: leading whitespace, string contents, trailing whitespace.
Definition: Parser.php:2862
This document is intended to provide useful advice for parties seeking to redistribute MediaWiki to end users It s targeted particularly at maintainers for Linux since it s been observed that distribution packages of MediaWiki often break We ve consistently had to recommend that users seeking support use official tarballs instead of their distribution s and this often solves whatever problem the user is having It would be nice if this could such as
Definition: distributors.txt:9
setOutputType($ot)
Set the output type.
Definition: Parser.php:776
$mTagHooks
Definition: Parser.php:141
Class for handling an array of magic words.
const NS_MEDIAWIKI
Definition: Defines.php:64
static & get($id)
Factory: creates an object representing an ID.
Definition: MagicWord.php:257
enableOOUI()
Set's up the PHP implementation of OOUI for use in this request and instructs OutputPage to enable OO...
Definition: Parser.php:6067
please add to it if you re going to add events to the MediaWiki code where normally authentication against an external auth plugin would be creating a local account $user
Definition: hooks.txt:242
this hook is for auditing only RecentChangesLinked and Watchlist RecentChangesLinked and Watchlist e g Watchlist removed from all revisions and log entries to which it was applied This gives extensions a chance to take it off their books as the deletion has already been partly carried out by this point or something similar the user will be unable to create the tag set and then return false from the hook function Ensure you consume the ChangeTagAfterDelete hook to carry out custom deletion actions as context called by AbstractContent::getParserOutput May be used to override the normal model specific rendering of page content as context as context the output can only depend on parameters provided to this hook not on global state indicating whether full HTML should be generated If generation of HTML may be but other information should still be present in the ParserOutput object to manipulate or replace but no entry for that model exists in $wgContentHandlers if desired whether it is OK to use $contentModel on $title Handler functions that modify $ok should generally return false to prevent further hooks from further modifying $ok inclusive false for true for descending in case the handler function wants to provide a converted Content object Note that $result getContentModel() must return $toModel. 'CustomEditor'$rcid is used in generating this variable which contains information about the new such as the revision s whether the revision was marked as a minor edit or not
Definition: hooks.txt:1156
fetchTemplate($title)
Fetch the unparsed text of a template and register a reference to it.
Definition: Parser.php:3530
maybeMakeExternalImage($url)
make an image if it's allowed, either through the global option, through the exception, or through the on-wiki whitelist
Definition: Parser.php:2010
areSubpagesAllowed()
Return true if subpage links should be expanded on this page.
Definition: Parser.php:2428
const OT_HTML
Definition: Defines.php:188
static escapeId($id, $options=[])
Given a value, escape it so that it can be used in an id attribute and return it. ...
Definition: Sanitizer.php:1170
this hook is for auditing only RecentChangesLinked and Watchlist RecentChangesLinked and Watchlist e g Watchlist removed from all revisions and log entries to which it was applied This gives extensions a chance to take it off their books as the deletion has already been partly carried out by this point or something similar the user will be unable to create the tag set and then return false from the hook function Ensure you consume the ChangeTagAfterDelete hook to carry out custom deletion actions as context called by AbstractContent::getParserOutput May be used to override the normal model specific rendering of page content as context as context the output can only depend on parameters provided to this hook not on global state indicating whether full HTML should be generated If generation of HTML may be but other information should still be present in the ParserOutput object & $output
Definition: hooks.txt:1046
static getSubstIDs()
Get an array of parser substitution modifier IDs.
Definition: MagicWord.php:284
static images()
Definition: SiteStats.php:173
$mTransparentTagHooks
Definition: Parser.php:142
$mExpensiveFunctionCount
Definition: Parser.php:194
$mUrlProtocols
Definition: Parser.php:166
$mConf
Definition: Parser.php:166
transformMsg($text, $options, $title=null)
Wrapper for preprocess()
Definition: Parser.php:4673
static newFromId($id, $flags=0)
Load a page revision from a given revision ID number.
Definition: Revision.php:110
wfUrlProtocols($includeProtocolRelative=true)
Returns a regular expression of url protocols.
static makeExternalLink($url, $text, $escape=true, $linktype= '', $attribs=[], $title=null)
Make an external link.
Definition: Linker.php:934
__clone()
Allow extensions to clean up when the parser is cloned.
Definition: Parser.php:300
static getExternalLinkRel($url=false, $title=null)
Get the rel attribute for a particular external link.
Definition: Parser.php:1888
injection txt This is an overview of how MediaWiki makes use of dependency injection The design described here grew from the discussion of RFC T384 The term dependency this means that anything an object needs to operate should be injected from the the object itself should only know narrow no concrete implementation of the logic it relies on The requirement to inject everything typically results in an architecture that based on two main types of and essentially stateless service objects that use other service objects to operate on the value objects As of the beginning MediaWiki is only starting to use the DI approach Much of the code still relies on global state or direct resulting in a highly cyclical dependency which acts as the top level factory for services in MediaWiki which can be used to gain access to default instances of various services MediaWikiServices however also allows new services to be defined and default services to be redefined Services are defined or redefined by providing a callback the instantiator that will return a new instance of the service When it will create an instance of MediaWikiServices and populate it with the services defined in the files listed by thereby bootstrapping the DI framework Per $wgServiceWiringFiles lists includes ServiceWiring php
Definition: injection.txt:35
wfSetVar(&$dest, $source, $force=false)
Sets dest to source and returns the original value of dest If source is NULL, it just returns the val...
this hook is for auditing only $req
Definition: hooks.txt:1007
this hook is for auditing only or null if authentication failed before getting that far $username
Definition: hooks.txt:802
presenting them properly to the user as errors is done by the caller return true use this to change the list i e etc next in line in page history
Definition: hooks.txt:1721
array $mLangLinkLanguages
Array with the language name of each language link (i.e.
Definition: Parser.php:235
const OT_MSG
Definition: Parser.php:113
replaceTransparentTags($text)
Replace transparent tags in $text with the values given by the callbacks.
Definition: Parser.php:5407
This document describes the state of Postgres support in and is fairly well maintained The main code is very well while extensions are very hit and miss it is probably the most supported database after MySQL Much of the work in making MediaWiki database agnostic came about through the work of creating Postgres as and are nearing end of but without copying over all the usage comments General notes on the but these can almost always be programmed around *Although Postgres has a true BOOLEAN type
Definition: postgres.txt:22
replaceSection($oldText, $sectionId, $newText)
This function returns $oldtext after the content of the section specified by $section has been replac...
Definition: Parser.php:5599
doDoubleUnderscore($text)
Strip double-underscore items like NOGALLERY and NOTOC Fills $this->mDoubleUnderscores, returns the modified text.
Definition: Parser.php:3934
$mFunctionHooks
Definition: Parser.php:143
static removeHTMLtags($text, $processCallback=null, $args=[], $extratags=[], $removetags=[], $warnCallback=null)
Cleans up HTML, removes dangerous tags and attributes, and removes HTML comments. ...
Definition: Sanitizer.php:462
$lines
Definition: router.php:67
testPreprocess($text, Title $title, ParserOptions $options)
Definition: Parser.php:5846
deferred txt A few of the database updates required by various functions here can be deferred until after the result page is displayed to the user For updating the view updating the linked to tables after a etc PHP does not yet have any way to tell the server to actually return and disconnect while still running these but it might have such a feature in the future We handle these by creating a deferred update object and putting those objects on a global then executing the whole list after the page is displayed We don t do anything smart like collating updates to the same table or such because the list is almost always going to have just one item on if so it s not worth the trouble Since there is a job queue in the jobs table
Definition: deferred.txt:11
MagicWordArray $mSubstWords
Definition: Parser.php:164
const TOC_END
Definition: Parser.php:138
static normalizeCharReferences($text)
Ensure that any entities and character references are legal for XML and XHTML specifically.
Definition: Sanitizer.php:1400
callParserFunction($frame, $function, array $args=[])
Call a parser function and return an array with text and flags.
Definition: Parser.php:3323
$wgScriptPath
The path we should point to.
Variant of the Message class.
Definition: Message.php:1242
getFreshParser()
Return this parser if it is not doing anything, otherwise get a fresh parser.
Definition: Parser.php:6052
design txt This is a brief overview of the new design More thorough and up to date information is available on the documentation wiki at etc Handles the details of getting and saving to the user table of the and dealing with sessions and cookies OutputPage Encapsulates the entire HTML page that will be sent in response to any server request It is used by calling its functions to add in any and then calling but I prefer the flexibility This should also do the output encoding The system allocates a global one in $wgOut Title Represents the title of an and does all the work of translating among various forms such as plain database etc For and for historical it also represents a few features of articles that don t involve their such as access rights See also title txt Article Encapsulates access to the page table of the database The object represents a an and maintains state such as etc Revision Encapsulates individual page revision data and access to the revision text blobs storage system Higher level code should never touch text storage directly
Definition: design.txt:34
this hook is for auditing only RecentChangesLinked and Watchlist RecentChangesLinked and Watchlist e g Watchlist removed from all revisions and log entries to which it was applied This gives extensions a chance to take it off their books as the deletion has already been partly carried out by this point or something similar the user will be unable to create the tag set and then return false from the hook function Ensure you consume the ChangeTagAfterDelete hook to carry out custom deletion actions as context called by AbstractContent::getParserOutput May be used to override the normal model specific rendering of page content $content
Definition: hooks.txt:1046
static articles()
Definition: SiteStats.php:141
$mRevisionUser
Definition: Parser.php:219
lock()
Lock the current instance of the parser.
Definition: Parser.php:6007
static pages()
Definition: SiteStats.php:149
$line
Definition: cdb.php:59
const SFH_OBJECT_ARGS
Definition: Parser.php:86
makeKnownLinkHolder($nt, $text= '', $trail= '', $prefix= '')
Render a forced-blue link inline; protect against double expansion of URLs if we're in a mode that pr...
Definition: Parser.php:2395
static statelessFetchTemplate($title, $parser=false)
Static function to get a template Can be overridden via ParserOptions::setTemplateCallback().
Definition: Parser.php:3543
I won t presume to tell you how to I m just describing the methods I chose to use for myself If you do choose to follow these it will probably be easier for you to collaborate with others on the but if you want to contribute without by all means do which work well I also use K &R brace matching style I know that s a religious issue for so if you want to use a style that puts opening braces on the next line
Definition: design.txt:79
setFunctionHook($id, $callback, $flags=0)
Create a function, e.g.
Definition: Parser.php:4811
static setupOOUI($skinName= '', $dir= 'ltr')
Helper function to setup the PHP implementation of OOUI to use in this request.
static makeMediaLinkFile(Title $title, $file, $html= '')
Create a direct link to a given uploaded file.
Definition: Linker.php:874
$mIncludeCount
Definition: Parser.php:184
usually copyright or history_copyright This message must be in HTML not wikitext if the section is included from a template to be included in the link
Definition: hooks.txt:2889
$mMarkerIndex
Definition: Parser.php:151
this hook is for auditing only RecentChangesLinked and Watchlist RecentChangesLinked and Watchlist e g Watchlist removed from all revisions and log entries to which it was applied This gives extensions a chance to take it off their books as the deletion has already been partly carried out by this point or something similar the user will be unable to create the tag set and then return false from the hook function Ensure you consume the ChangeTagAfterDelete hook to carry out custom deletion actions as context called by AbstractContent::getParserOutput May be used to override the normal model specific rendering of page content as context as context the output can only depend on parameters provided to this hook not on global state indicating whether full HTML should be generated If generation of HTML may be but other information should still be present in the ParserOutput object to manipulate or replace but no entry for that model exists in $wgContentHandlers if desired whether it is OK to use $contentModel on $title Handler functions that modify $ok should generally return false to prevent further hooks from further modifying $ok inclusive $limit
Definition: hooks.txt:1046
getTitle()
Accessor for the Title object.
Definition: Parser.php:757
this class mediates it Skin Encapsulates a look and feel for the wiki All of the functions that render HTML and make choices about how to render it are here and are called from various other places when and is meant to be subclassed with other skins that may override some of its functions The User object contains a reference to a and so rather than having a global skin object we just rely on the global User and get the skin with $wgUser and also has some character encoding functions and other locale stuff The current user interface language is instantiated as and the local content language as $wgContLang
Definition: design.txt:56
extractSections($text, $sectionId, $mode, $newText= '')
Break wikitext input into sections, and either pull or replace some particular section's text...
Definition: Parser.php:5458
ParserOutput $mOutput
Definition: Parser.php:176
getOutput()
Get the ParserOutput object.
Definition: Parser.php:802
$wgExperimentalHtmlIds
Should we allow a broader set of characters in id attributes, per HTML5? If not, use only HTML 4-comp...
doMagicLinks($text)
Replace special strings like "ISBN xxx" and "RFC xxx" with magic external links.
Definition: Parser.php:1414
Apache License January AND DISTRIBUTION Definitions License shall mean the terms and conditions for and distribution as defined by Sections through of this document Licensor shall mean the copyright owner or entity authorized by the copyright owner that is granting the License Legal Entity shall mean the union of the acting entity and all other entities that control are controlled by or are under common control with that entity For the purposes of this definition control direct or to cause the direction or management of such whether by contract or including but not limited to software source documentation and configuration files Object form shall mean any form resulting from mechanical transformation or translation of a Source including but not limited to compiled object generated and conversions to other media types Work shall mean the work of whether in Source or Object made available under the as indicated by a copyright notice that is included in or attached to the whether in Source or Object that is based or other modifications as a an original work of authorship For the purposes of this Derivative Works shall not include works that remain separable or merely the Work and Derivative Works thereof Contribution shall mean any work of including the original version of the Work and any modifications or additions to that Work or Derivative Works that is intentionally submitted to Licensor for inclusion in the Work by the copyright owner or by an individual or Legal Entity authorized to submit on behalf of the copyright owner For the purposes of this submitted means any form of or written communication sent to the Licensor or its including but not limited to communication on electronic mailing source code control and issue tracking systems that are managed or on behalf the Licensor for the purpose of discussing and improving the but excluding communication that is conspicuously marked or otherwise designated in writing by the copyright owner as Not a Contribution Contributor shall mean Licensor and any individual or Legal Entity on behalf of whom a Contribution has been received by Licensor and subsequently incorporated within the Work Grant of Copyright License Subject to the terms and conditions of this each Contributor hereby grants to You a non no royalty irrevocable copyright license to prepare Derivative Works publicly display
this hook is for auditing only RecentChangesLinked and Watchlist RecentChangesLinked and Watchlist e g Watchlist removed from all revisions and log entries to which it was applied This gives extensions a chance to take it off their books as the deletion has already been partly carried out by this point or something similar the user will be unable to create the tag set $status
Definition: hooks.txt:1046
static cleanSigInSig($text)
Strip 3, 4 or 5 tildes out of signatures.
Definition: Parser.php:4628
setDefaultSort($sort)
Mutator for $mDefaultSort.
Definition: Parser.php:5711
fetchFile($title, $options=[])
Fetch a file and its title and register a reference to it.
Definition: Parser.php:3633
static tocIndent()
Add another level to the Table of Contents.
Definition: Linker.php:1589
static legalChars()
Get a regex character class describing the legal characters in a link.
Definition: Title.php:593
design txt This is a brief overview of the new design More thorough and up to date information is available on the documentation wiki at etc Handles the details of getting and saving to the user table of the and dealing with sessions and cookies OutputPage Encapsulates the entire HTML page that will be sent in response to any server request It is used by calling its functions to add in any and then calling output() to send it all.It could be easily changed to send incrementally if that becomes useful
static doBlockLevels($text, $lineStart)
Make lists from lines starting with ':', '*', '#', etc.
$wgServer
URL of the server.
We ve cleaned up the code here by removing clumps of infrequently used code and moving them off somewhere else It s much easier for someone working with this code to see what s _really_ going on
Definition: hooks.txt:86
incrementExpensiveFunctionCount()
Increment the expensive function count.
Definition: Parser.php:3921
$mShowToc
Definition: Parser.php:195
static normalizeLinkUrl($url)
Replace unusual escape codes in a URL with their equivalent characters.
Definition: Parser.php:1951
const DB_REPLICA
Definition: defines.php:22
magicLinkCallback($m)
Definition: Parser.php:1444
const EXT_LINK_URL_CLASS
Definition: Parser.php:92
insertStripItem($text)
Add an item to the strip state Returns the unique tag which must be inserted into the stripped text T...
Definition: Parser.php:1026
testPst($text, Title $title, ParserOptions $options)
Definition: Parser.php:5836
static factory($url, $options=null, $caller=__METHOD__)
Generate a new request object.
if(!$wgRequest->checkUrlExtension()) if(!$wgEnableAPI) $wgTitle
Definition: api.php:57
static explode($separator, $subject)
Workalike for explode() with limited memory usage.
ParserOptions $mOptions
Definition: Parser.php:208
parse($text, Title $title, ParserOptions $options, $linestart=true, $clearState=true, $revid=null)
Convert wikitext to HTML Do not call this function recursively.
Definition: Parser.php:399
this hook is for auditing only or null if authentication failed before getting that far or null if we can t even determine that probably a stub it is not rendered in wiki pages or galleries in category pages allow injecting custom HTML after the section Any uses of the hook need to handle escaping see BaseTemplate::getToolbox and BaseTemplate::makeListItem for details on the format of individual items inside of this array or by returning and letting standard HTTP rendering take place modifiable or by returning false and taking over the output modifiable modifiable after all normalizations have been except for the $wgMaxImageArea check set to true or false to override the $wgMaxImageArea check result gives extension the possibility to transform it themselves $handler
Definition: hooks.txt:802
static numberingroup($group)
Find the number of users in a given user group.
Definition: SiteStats.php:183
=Architecture==Two class hierarchies are used to provide the functionality associated with the different content models:*Content interface(and AbstractContent base class) define functionality that acts on the concrete content of a page, and *ContentHandler base class provides functionality specific to a content model, but not acting on concrete content.The most important function of ContentHandler is to act as a factory for the appropriate implementation of Content.These Content objects are to be used by MediaWiki everywhere, instead of passing page content around as text.All manipulation and analysis of page content must be done via the appropriate methods of the Content object.For each content model, a subclass of ContentHandler has to be registered with $wgContentHandlers.The ContentHandler object for a given content model can be obtained using ContentHandler::getForModelID($id).Also Title, WikiPage and Revision now have getContentHandler() methods for convenience.ContentHandler objects are singletons that provide functionality specific to the content type, but not directly acting on the content of some page.ContentHandler::makeEmptyContent() and ContentHandler::unserializeContent() can be used to create a Content object of the appropriate type.However, it is recommended to instead use WikiPage::getContent() resp.Revision::getContent() to get a page's content as a Content object.These two methods should be the ONLY way in which page content is accessed.Another important function of ContentHandler objects is to define custom action handlers for a content model, see ContentHandler::getActionOverrides().This is similar to what WikiPage::getActionOverrides() was already doing.==Serialization==With the ContentHandler facility, page content no longer has to be text based.Objects implementing the Content interface are used to represent and handle the content internally.For storage and data exchange, each content model supports at least one serialization format via ContentHandler::serializeContent($content).The list of supported formats for a given content model can be accessed using ContentHandler::getSupportedFormats().Content serialization formats are identified using MIME type like strings.The following formats are built in:*text/x-wiki-wikitext *text/javascript-for js pages *text/css-for css pages *text/plain-for future use, e.g.with plain text messages.*text/html-for future use, e.g.with plain html messages.*application/vnd.php.serialized-for future use with the api and for extensions *application/json-for future use with the api, and for use by extensions *application/xml-for future use with the api, and for use by extensions In PHP, use the corresponding CONTENT_FORMAT_XXX constant.Note that when using the API to access page content, especially action=edit, action=parse and action=query &prop=revisions, the model and format of the content should always be handled explicitly.Without that information, interpretation of the provided content is not reliable.The same applies to XML dumps generated via maintenance/dumpBackup.php or Special:Export.Also note that the API will provide encapsulated, serialized content-so if the API was called with format=json, and contentformat is also json(or rather, application/json), the page content is represented as a string containing an escaped json structure.Extensions that use JSON to serialize some types of page content may provide specialized API modules that allow access to that content in a more natural form.==Compatibility==The ContentHandler facility is introduced in a way that should allow all existing code to keep functioning at least for pages that contain wikitext or other text based content.However, a number of functions and hooks have been deprecated in favor of new versions that are aware of the page's content model, and will now generate warnings when used.Most importantly, the following functions have been deprecated:*Revisions::getText() is deprecated in favor Revisions::getContent()*WikiPage::getText() is deprecated in favor WikiPage::getContent() Also, the old Article::getContent()(which returns text) is superceded by Article::getContentObject().However, both methods should be avoided since they do not provide clean access to the page's actual content.For instance, they may return a system message for non-existing pages.Use WikiPage::getContent() instead.Code that relies on a textual representation of the page content should eventually be rewritten.However, ContentHandler::getContentText() provides a stop-gap that can be used to get text for a page.Its behavior is controlled by $wgContentHandlerTextFallback it
const STRIP_COMMENTS
static getVersion($flags= '', $lang=null)
Return a string of the MediaWiki version with Git revision if available.
braceSubstitution($piece, $frame)
Return the text of a template, after recursively replacing any variables or templates within the temp...
Definition: Parser.php:2994
setUser($user)
Set the current user.
Definition: Parser.php:719
$mHighestExpansionDepth
Definition: Parser.php:191
makeImage($title, $options, $holders=false)
Parse image options text and use it to make an image.
Definition: Parser.php:5122
attributeStripCallback(&$text, $frame=false)
Callback from the Sanitizer for expanding items found in HTML attribute values, so they can be safely...
Definition: Parser.php:5378
static cascadingsources($parser, $title= '')
Returns the sources of any cascading protection acting on a specified page.
getCustomDefaultSort()
Accessor for $mDefaultSort Unlike getDefaultSort(), will return false if none is set.
Definition: Parser.php:5740
extensionSubstitution($params, $frame)
Return the text to be used for a given extension tag.
Definition: Parser.php:3800
static normalizeLineEndings($text)
Do a "\\r\\n" -> "\\n" and "\\r" -> "\\n" transformation as well as trim trailing whitespace...
static makeExternalImage($url, $alt= '')
Return the code for images which were added via external links, via Parser::maybeMakeExternalImage()...
Definition: Linker.php:362
recursiveTagParseFully($text, $frame=false)
Fully parse wikitext to fully parsed HTML.
Definition: Parser.php:627
setTransparentTagHook($tag, $callback)
As setHook(), but letting the contents be parsed.
Definition: Parser.php:4748