const EXT_IMAGE_REGEX = '/^(http:\/\/|https:\/\/)([^][<>"\\x00-\\x20\\x7F\p{Zs}]+)
\\/([A-Za-z0-9_.,~%\\-+&;#*?!=()@\\x80-\\xFF]+)\\.((?i)gif|png|jpg|jpeg)$/Sxu';
+ # Regular expression for a non-newline space
+ const SPACE_NOT_NL = '(?:\t| |&\#0*160;|&\#[Xx]0*[Aa]0;|\p{Zs})';
+
# State constants for the definition list colon extraction
const COLON_STATE_TEXT = 0;
const COLON_STATE_TAG = 1;
* @var MagicWordArray
*/
public $mSubstWords;
- public $mConf, $mPreprocessor, $mExtLinkBracketedRegex, $mUrlProtocols; # Initialised in constructor
+ # Initialised in constructor
+ public $mConf, $mPreprocessor, $mExtLinkBracketedRegex, $mUrlProtocols;
# Cleared with clearState():
/**
}
$this->mFirstCall = false;
- wfProfileIn( __METHOD__ );
CoreParserFunctions::register( $this );
CoreTagHooks::register( $this );
$this->initialiseVariables();
Hooks::run( 'ParserFirstCallInit', array( &$this ) );
- wfProfileOut( __METHOD__ );
}
/**
* @private
*/
public function clearState() {
- wfProfileIn( __METHOD__ );
if ( $this->mFirstCall ) {
$this->firstCallInit();
}
$this->mProfiler = new SectionProfiler();
Hooks::run( 'ParserClearState', array( &$this ) );
- wfProfileOut( __METHOD__ );
}
/**
global $wgShowHostnames;
$fname = __METHOD__ . '-' . wfGetCaller();
- wfProfileIn( __METHOD__ );
wfProfileIn( $fname );
if ( $clearState ) {
$this->mInputSize = false;
$this->currentRevisionCache = null;
wfProfileOut( $fname );
- wfProfileOut( __METHOD__ );
return $this->mOutput;
}
* @return string UNSAFE half-parsed HTML
*/
public function recursiveTagParse( $text, $frame = false ) {
- wfProfileIn( __METHOD__ );
Hooks::run( 'ParserBeforeStrip', array( &$this, &$text, &$this->mStripState ) );
Hooks::run( 'ParserAfterStrip', array( &$this, &$text, &$this->mStripState ) );
$text = $this->internalParse( $text, false, $frame );
- wfProfileOut( __METHOD__ );
return $text;
}
* @return string Fully parsed HTML
*/
public function recursiveTagParseFully( $text, $frame = false ) {
- wfProfileIn( __METHOD__ );
$text = $this->recursiveTagParse( $text, $frame );
$text = $this->internalParseHalfParsed( $text, false );
- wfProfileOut( __METHOD__ );
return $text;
}
* @param bool|PPFrame $frame
* @return mixed|string
*/
- public function preprocess( $text, Title $title = null, ParserOptions $options, $revid = null, $frame = false ) {
- wfProfileIn( __METHOD__ );
+ public function preprocess( $text, Title $title = null,
+ ParserOptions $options, $revid = null, $frame = false
+ ) {
$magicScopeVariable = $this->lock();
$this->startParse( $title, $options, self::OT_PREPROCESS, true );
if ( $revid !== null ) {
Hooks::run( 'ParserAfterStrip', array( &$this, &$text, &$this->mStripState ) );
$text = $this->replaceVariables( $text, $frame );
$text = $this->mStripState->unstripBoth( $text );
- wfProfileOut( __METHOD__ );
return $text;
}
* @since 1.19
*/
public function recursivePreprocess( $text, $frame = false ) {
- wfProfileIn( __METHOD__ );
$text = $this->replaceVariables( $text, $frame );
$text = $this->mStripState->unstripBoth( $text );
- wfProfileOut( __METHOD__ );
return $text;
}
* @return string
*/
public function doTableStuff( $text ) {
- wfProfileIn( __METHOD__ );
$lines = StringUtils::explode( "\n", $text );
$out = '';
$out = '';
}
- wfProfileOut( __METHOD__ );
return $out;
}
* @return string
*/
public function internalParse( $text, $isMain = true, $frame = false ) {
- wfProfileIn( __METHOD__ );
$origText = $text;
# Hook to suspend the parser in this state
if ( !Hooks::run( 'ParserBeforeInternalParse', array( &$this, &$text, &$this->mStripState ) ) ) {
- wfProfileOut( __METHOD__ );
return $text;
}
$text = $this->doMagicLinks( $text );
$text = $this->formatHeadings( $text, $origText, $isMain );
- wfProfileOut( __METHOD__ );
return $text;
}
* @return string
*/
public function doMagicLinks( $text ) {
- wfProfileIn( __METHOD__ );
$prots = wfUrlProtocolsWithoutProtRel();
$urlChar = self::EXT_LINK_URL_CLASS;
+ $space = self::SPACE_NOT_NL; # non-newline space
+ $spdash = "(?:-|$space)"; # a dash or a non-newline space
+ $spaces = "$space++"; # possessive match of 1 or more spaces
$text = preg_replace_callback(
'!(?: # Start cases
(<a[ \t\r\n>].*?</a>) | # m[1]: Skip link text
(<.*?>) | # m[2]: Skip stuff inside HTML elements' . "
- (\b(?i:$prots)$urlChar+) | # m[3]: Free external links" . '
- \b(?:RFC|PMID)\s+([0-9]+)\b |# m[4]: RFC or PMID, capture number
- \bISBN\s+( # m[5]: ISBN, capture number
- (?: 97[89] [\ \-]? )? # optional 13-digit ISBN prefix
- (?: [0-9] [\ \-]? ){9} # 9 digits with opt. delimiters
+ (\b(?i:$prots)$urlChar+) | # m[3]: Free external links
+ \b(?:RFC|PMID) $spaces # m[4]: RFC or PMID, capture number
+ ([0-9]+)\b |
+ \bISBN $spaces ( # m[5]: ISBN, capture number
+ (?: 97[89] $spdash? )? # optional 13-digit ISBN prefix
+ (?: [0-9] $spdash? ){9} # 9 digits with opt. delimiters
[0-9Xx] # check digit
- )\b
- )!xu', array( &$this, 'magicLinkCallback' ), $text );
- wfProfileOut( __METHOD__ );
+ )\b
+ )!xu", array( &$this, 'magicLinkCallback' ), $text );
return $text;
}
} elseif ( isset( $m[5] ) && $m[5] !== '' ) {
# ISBN
$isbn = $m[5];
+ $space = self::SPACE_NOT_NL; # non-newline space
+ $isbn = preg_replace( "/$space/", ' ', $isbn );
$num = strtr( $isbn, array(
'-' => '',
' ' => '',
* @private
*/
public function makeFreeExternalLink( $url ) {
- wfProfileIn( __METHOD__ );
$trail = '';
$sep .= ')';
}
- $numSepChars = strspn( strrev( $url ), $sep );
+ $urlRev = strrev( $url );
+ $numSepChars = strspn( $urlRev, $sep );
+ # Don't break a trailing HTML entity by moving the ; into $trail
+ # This is in hot code, so use substr_compare to avoid having to
+ # create a new string object for the comparison
+ if ( $numSepChars && substr_compare( $url, ";", -$numSepChars, 1 ) === 0) {
+ # more optimization: instead of running preg_match with a $
+ # anchor, which can be slow, do the match on the reversed
+ # string starting at the desired offset.
+ # un-reversed regexp is: /&([a-z]+|#x[\da-f]+|#\d+)$/i
+ if ( preg_match( '/\G([a-z]+|[\da-f]+x#|\d+#)&/i', $urlRev, $m2, 0, $numSepChars ) ) {
+ $numSepChars--;
+ }
+ }
if ( $numSepChars ) {
$trail = substr( $url, -$numSepChars ) . $trail;
$url = substr( $url, 0, -$numSepChars );
$pasteurized = self::normalizeLinkUrl( $url );
$this->mOutput->addExternalLink( $pasteurized );
}
- wfProfileOut( __METHOD__ );
return $text . $trail;
}
* @return string
*/
public function doHeadings( $text ) {
- wfProfileIn( __METHOD__ );
for ( $i = 6; $i >= 1; --$i ) {
$h = str_repeat( '=', $i );
$text = preg_replace( "/^$h(.+)$h\\s*$/m", "<h$i>\\1</h$i>", $text );
}
- wfProfileOut( __METHOD__ );
return $text;
}
* @return string The altered text
*/
public function doAllQuotes( $text ) {
- wfProfileIn( __METHOD__ );
$outtext = '';
$lines = StringUtils::explode( "\n", $text );
foreach ( $lines as $line ) {
$outtext .= $this->doQuotes( $line ) . "\n";
}
$outtext = substr( $outtext, 0, -1 );
- wfProfileOut( __METHOD__ );
return $outtext;
}
* @return string
*/
public function replaceExternalLinks( $text ) {
- wfProfileIn( __METHOD__ );
$bits = preg_split( $this->mExtLinkBracketedRegex, $text, -1, PREG_SPLIT_DELIM_CAPTURE );
if ( $bits === false ) {
- wfProfileOut( __METHOD__ );
throw new MWException( "PCRE needs to be compiled with "
. "--enable-unicode-properties in order for MediaWiki to function" );
}
$this->mOutput->addExternalLink( $pasteurized );
}
- wfProfileOut( __METHOD__ );
return $s;
}
*/
public function replaceInternalLinks2( &$s ) {
global $wgExtraInterlanguageLinkPrefixes;
- wfProfileIn( __METHOD__ );
wfProfileIn( __METHOD__ . '-setup' );
static $tc = false, $e1, $e1_img;
if ( is_null( $this->mTitle ) ) {
wfProfileOut( __METHOD__ . '-setup' );
- wfProfileOut( __METHOD__ );
throw new MWException( __METHOD__ . ": \$this->mTitle is null\n" );
}
$nottalk = !$this->mTitle->isTalkPage();
}
wfProfileOut( __METHOD__ . "-always_known" );
}
- wfProfileOut( __METHOD__ );
return $holders;
}
* @return string The lists rendered as HTML
*/
public function doBlockLevels( $text, $linestart ) {
- wfProfileIn( __METHOD__ );
# Parsing through the text line by line. The main thing
# happening here is handling of block-level elements p, pre,
$this->mLastSection = '';
}
- wfProfileOut( __METHOD__ );
return $output;
}
* @return string The position of the ':', or false if none found
*/
public function findColonNoLinks( $str, &$before, &$after ) {
- wfProfileIn( __METHOD__ );
$pos = strpos( $str, ':' );
if ( $pos === false ) {
# Nothing to find!
- wfProfileOut( __METHOD__ );
return false;
}
# Easy; no tag nesting to worry about
$before = substr( $str, 0, $pos );
$after = substr( $str, $pos + 1 );
- wfProfileOut( __METHOD__ );
return $pos;
}
# We found it!
$before = substr( $str, 0, $i );
$after = substr( $str, $i + 1 );
- wfProfileOut( __METHOD__ );
return $i;
}
# Embedded in a tag; don't break it.
$colon = strpos( $str, ':', $i );
if ( $colon === false ) {
# Nothing else interesting
- wfProfileOut( __METHOD__ );
return false;
}
$lt = strpos( $str, '<', $i );
# We found it!
$before = substr( $str, 0, $colon );
$after = substr( $str, $colon + 1 );
- wfProfileOut( __METHOD__ );
return $i;
}
}
$stack--;
if ( $stack < 0 ) {
wfDebug( __METHOD__ . ": Invalid input; too many close tags\n" );
- wfProfileOut( __METHOD__ );
return false;
}
$state = self::COLON_STATE_TEXT;
}
break;
default:
- wfProfileOut( __METHOD__ );
throw new MWException( "State machine error in " . __METHOD__ );
}
}
if ( $stack > 0 ) {
wfDebug( __METHOD__ . ": Invalid input; not enough close tags (stack $stack, state $state)\n" );
- wfProfileOut( __METHOD__ );
return false;
}
- wfProfileOut( __METHOD__ );
return false;
}
* @private
*/
public function initialiseVariables() {
- wfProfileIn( __METHOD__ );
$variableIDs = MagicWord::getVariableIDs();
$substIDs = MagicWord::getSubstIDs();
$this->mVariables = new MagicWordArray( $variableIDs );
$this->mSubstWords = new MagicWordArray( $substIDs );
- wfProfileOut( __METHOD__ );
}
/**
if ( strlen( $text ) < 1 || strlen( $text ) > $this->mOptions->getMaxIncludeSize() ) {
return $text;
}
- wfProfileIn( __METHOD__ );
if ( $frame === false ) {
$frame = $this->getPreprocessor()->newFrame();
$flags = $argsOnly ? PPFrame::NO_TEMPLATES : 0;
$text = $frame->expand( $dom, $flags );
- wfProfileOut( __METHOD__ );
return $text;
}
* @return string The text of the template
*/
public function braceSubstitution( $piece, $frame ) {
- wfProfileIn( __METHOD__ );
wfProfileIn( __METHOD__ . '-setup' );
// Flags
$result = $this->callParserFunction( $frame, $func, $funcArgs );
} catch ( Exception $ex ) {
wfProfileOut( __METHOD__ . '-pfunc' );
- wfProfileOut( __METHOD__ );
throw $ex;
}
if ( $profileSection ) {
$this->mProfiler->scopedProfileOut( $profileSection );
}
- wfProfileOut( __METHOD__ );
return array( 'object' => $text );
}
$ret = array( 'text' => $text );
}
- wfProfileOut( __METHOD__ );
return $ret;
}
public function callParserFunction( $frame, $function, array $args = array() ) {
global $wgContLang;
- wfProfileIn( __METHOD__ );
# Case sensitive functions
if ( isset( $this->mFunctionSynonyms[1][$function] ) ) {
if ( isset( $this->mFunctionSynonyms[0][$function] ) ) {
$function = $this->mFunctionSynonyms[0][$function];
} else {
- wfProfileOut( __METHOD__ );
return array( 'found' => false );
}
}
# Workaround for PHP bug 35229 and similar
if ( !is_callable( $callback ) ) {
wfProfileOut( __METHOD__ . '-pfunc-' . $function );
- wfProfileOut( __METHOD__ );
throw new MWException( "Tag hook for $function is not callable\n" );
}
$result['isChildObj'] = true;
}
wfProfileOut( __METHOD__ . '-pfunc-' . $function );
- wfProfileOut( __METHOD__ );
return $result;
}
* @return array
*/
public function argSubstitution( $piece, $frame ) {
- wfProfileIn( __METHOD__ );
$error = false;
$parts = $piece['parts'];
$ret = array( 'text' => $text );
}
- wfProfileOut( __METHOD__ );
return $ret;
}
* @return string
*/
public function doDoubleUnderscore( $text ) {
- wfProfileIn( __METHOD__ );
# The position of __TOC__ needs to be recorded
$mw = MagicWord::get( 'toc' );
$this->mOutput->setProperty( $key, '' );
}
- wfProfileOut( __METHOD__ );
return $text;
}
# * <sup> and <sub> (bug 8393)
# * <i> (bug 26375)
# * <b> (r105284)
+ # * <bdi> (bug 72884)
# * <span dir="rtl"> and <span dir="ltr"> (bug 35167)
#
# We strip any parameter from accepted tags (second regex), except dir="rtl|ltr" from <span>,
# to allow setting directionality in toc items.
$tocline = preg_replace(
array(
- '#<(?!/?(span|sup|sub|i|b)(?: [^>]*)?>).*?' . '>#',
- '#<(/?(?:span(?: dir="(?:rtl|ltr)")?|sup|sub|i|b))(?: .*?)?' . '>#'
+ '#<(?!/?(span|sup|sub|bdi|i|b)(?: [^>]*)?>).*?' . '>#',
+ '#<(/?(?:span(?: dir="(?:rtl|ltr)")?|sup|sub|bdi|i|b))(?: .*?)?' . '>#'
),
array( '', '<$1>' ),
$safeHeadline
}
$executing = true;
- wfProfileIn( __METHOD__ );
if ( !$title ) {
global $wgTitle;
$title = $wgTitle;
$text = $this->preprocess( $text, $title, $options );
$executing = false;
- wfProfileOut( __METHOD__ );
return $text;
}
* @return string HTML
*/
public function renderImageGallery( $text, $params ) {
- wfProfileIn( __METHOD__ );
$mode = false;
if ( isset( $params['mode'] ) ) {
}
$html = $ig->toHTML();
Hooks::run( 'AfterParserFetchFileAndTitle', array( $this, $ig, &$html ) );
- wfProfileOut( __METHOD__ );
return $html;
}
*/
public function getRevisionTimestamp() {
if ( is_null( $this->mRevisionTimestamp ) ) {
- wfProfileIn( __METHOD__ );
global $wgContLang;
# it needs to be consistent for all visitors.
$this->mRevisionTimestamp = $wgContLang->userAdjust( $timestamp, '' );
- wfProfileOut( __METHOD__ );
}
return $this->mRevisionTimestamp;
}
* @return array
*/
public function serializeHalfParsedText( $text ) {
- wfProfileIn( __METHOD__ );
$data = array(
'text' => $text,
'version' => self::HALF_PARSED_VERSION,
'stripState' => $this->mStripState->getSubState( $text ),
'linkHolders' => $this->mLinkHolders->getSubArray( $text )
);
- wfProfileOut( __METHOD__ );
return $data;
}
dépôts / lhc / web / wiklou.git / blobdiff