X-Git-Url: http://git.cyclocoop.org/%22%20.%20generer_url_ecrire%28%22auteur_infos%22%2C%20%22id_auteur=%24id%22%29%20.%20%22?a=blobdiff_plain;f=includes%2FParser.php;h=a910afe259a1f134780687ea714c36b052986add;hb=798df26f95341980b8e19373193bf23b00797973;hp=9477b66f6f3ec3e205dfc4bcce5ecca3d25f5930;hpb=94ba835da9ad494a1b9d597a28650b66cb7fa8ea;p=lhc%2Fweb%2Fwiklou.git
diff --git a/includes/Parser.php b/includes/Parser.php
index 9477b66f6f..a910afe259 100644
--- a/includes/Parser.php
+++ b/includes/Parser.php
@@ -1,17 +1,20 @@
"\\x00-\\x20\\x7F]' );
define( 'EXT_LINK_TEXT_CLASS', '[^\]\\x00-\\x1F\\x7F]' );
define( 'EXT_IMAGE_FNAME_CLASS', '[A-Za-z0-9_.,~%\\-+&;#*?!=()@\\x80-\\xFF]' );
define( 'EXT_IMAGE_EXTENSIONS', 'gif|png|jpg|jpeg' );
-define( 'EXT_LINK_BRACKETED', '/\[(('.URL_PROTOCOLS.'):'.EXT_LINK_URL_CLASS.'+) *('.EXT_LINK_TEXT_CLASS.'*?)\]/S' );
+define( 'EXT_LINK_BRACKETED', '/\[(\b('.URL_PROTOCOLS.'):'.EXT_LINK_URL_CLASS.'+) *('.EXT_LINK_TEXT_CLASS.'*?)\]/S' );
define( 'EXT_IMAGE_REGEX',
'/^('.HTTP_PROTOCOLS.':)'. # Protocol
'('.EXT_LINK_URL_CLASS.'+)\\/'. # Hostname and path
@@ -73,14 +76,14 @@ define( 'EXT_IMAGE_REGEX',
* performs brace substitution on MediaWiki messages
*
* Globals used:
- * objects: $wgLang, $wgDateFormatter, $wgLinkCache
+ * objects: $wgLang, $wgLinkCache
*
* NOT $wgArticle, $wgUser or $wgTitle. Keep them away!
*
* settings:
* $wgUseTex*, $wgUseDynamicDates*, $wgInterwikiMagic*,
* $wgNamespacesWithSubpages, $wgAllowExternalImages*,
- * $wgLocaltimezone
+ * $wgLocaltimezone, $wgAllowSpecialInclusion*
*
* * only within ParserOptions
*
@@ -98,6 +101,7 @@ class Parser
# Cleared with clearState():
var $mOutput, $mAutonumber, $mDTopen, $mStripState = array();
var $mVariables, $mIncludeCount, $mArgStack, $mLastSection, $mInPre;
+ var $mInterwikiLinkHolders, $mLinkHolders;
# Temporary:
var $mOptions, $mTitle, $mOutputType,
@@ -114,6 +118,7 @@ class Parser
* @access public
*/
function Parser() {
+ global $wgContLang;
$this->mTemplates = array();
$this->mTemplatePath = array();
$this->mTagHooks = array();
@@ -128,13 +133,24 @@ class Parser
function clearState() {
$this->mOutput = new ParserOutput;
$this->mAutonumber = 0;
- $this->mLastSection = "";
+ $this->mLastSection = '';
$this->mDTopen = false;
$this->mVariables = false;
$this->mIncludeCount = array();
$this->mStripState = array();
$this->mArgStack = array();
$this->mInPre = false;
+ $this->mInterwikiLinkHolders = array(
+ 'texts' => array(),
+ 'titles' => array()
+ );
+ $this->mLinkHolders = array(
+ 'namespaces' => array(),
+ 'dbkeys' => array(),
+ 'queries' => array(),
+ 'texts' => array(),
+ 'titles' => array()
+ );
}
/**
@@ -142,6 +158,11 @@ class Parser
* to internalParse() which does all the real work.
*
* @access private
+ * @param string $text Text we want to parse
+ * @param Title &$title A title object
+ * @param array $options
+ * @param boolean $linestart
+ * @param boolean $clearState
* @return ParserOutput a ParserOutput
*/
function parse( $text, &$title, $options, $linestart = true, $clearState = true ) {
@@ -157,46 +178,51 @@ class Parser
$this->mTitle =& $title;
$this->mOutputType = OT_HTML;
- $stripState = NULL;
- $text = $this->strip( $text, $this->mStripState );
+ $this->mStripState = NULL;
+
+ //$text = $this->strip( $text, $this->mStripState );
+ // VOODOO MAGIC FIX! Sometimes the above segfaults in PHP5.
+ $x =& $this->mStripState;
+ $text = $this->strip( $text, $x );
+
+ $text = $this->internalParse( $text );
- $text = $this->internalParse( $text, $linestart );
+
$text = $this->unstrip( $text, $this->mStripState );
+
# Clean up special characters, only run once, next-to-last before doBlockLevels
- if(!$wgUseTidy) {
- $fixtags = array(
- # french spaces, last one Guillemet-left
- # only if there is something before the space
- '/(.) (?=\\?|:|;|!|\\302\\273)/i' => '\\1 \\2',
- # french spaces, Guillemet-right
- "/(\\302\\253) /i"=>"\\1 ",
- '/
/i' => '
',
- '/
/i' => '
',
- '//i' => '',
- '/<\\/center *>/i' => '
',
- # Clean up spare ampersands; note that we probably ought to be
- # more careful about named entities.
- '/&(?!:amp;|#[Xx][0-9A-fa-f]+;|#[0-9]+;|[a-zA-Z0-9]+;)/' => '&'
- );
- $text = preg_replace( array_keys($fixtags), array_values($fixtags), $text );
- } else {
- $fixtags = array(
- # french spaces, last one Guillemet-left
- '/ (\\?|:|;|!|\\302\\273)/i' => ' \\1',
- # french spaces, Guillemet-right
- '/(\\302\\253) /i' => '\\1 ',
- '//i' => '',
- '/<\\/center *>/i' => '
'
- );
- $text = preg_replace( array_keys($fixtags), array_values($fixtags), $text );
- }
+ $fixtags = array(
+ # french spaces, last one Guillemet-left
+ # only if there is something before the space
+ '/(.) (?=\\?|:|;|!|\\302\\273)/' => '\\1 \\2',
+ # french spaces, Guillemet-right
+ '/(\\302\\253) /' => '\\1 ',
+ '//i' => '',
+ '/<\\/center *>/i' => '
',
+ );
+ $text = preg_replace( array_keys($fixtags), array_values($fixtags), $text );
+
# only once and last
$text = $this->doBlockLevels( $text, $linestart );
$this->replaceLinkHolders( $text );
+
+ $dashReplace = array(
+ '/ - /' => " – ", # N dash
+ '/(?<=[\d])-(?=[\d])/' => "–", # N dash between numbers
+ '/ -- /' => " — " # M dash
+ );
+ $text = preg_replace( array_keys($dashReplace), array_values($dashReplace), $text );
+
+ # the position of the convert() call should not be changed. it
+ # assumes that the links are all replaces and the only thing left
+ # is the mark.
$text = $wgContLang->convert($text);
+ $this->mOutput->setTitleText($wgContLang->getParsedTitle());
$text = $this->unstripNoWiki( $text, $this->mStripState );
+
+ $text = Sanitizer::normalizeCharReferences( $text );
global $wgUseTidy;
if ($wgUseTidy) {
$text = Parser::tidy($text);
@@ -230,38 +256,76 @@ class Parser
* @access private
* @static
*/
- function extractTags($tag, $text, &$content, $uniq_prefix = ''){
+ function extractTagsAndParams($tag, $text, &$content, &$tags, &$params, $uniq_prefix = ''){
$rnd = $uniq_prefix . '-' . $tag . Parser::getRandomString();
if ( !$content ) {
$content = array( );
}
$n = 1;
$stripped = '';
+
+ if ( !$tags ) {
+ $tags = array( );
+ }
+
+ if ( !$params ) {
+ $params = array( );
+ }
+
+ // Hack to support short XML style tags
+ $text = preg_replace( "/<$tag(\\s+[^>]*|\\s*)\\/>/i", "<$tag\\1>$tag>", $text );
+
+ if( $tag == STRIP_COMMENTS ) {
+ $start = '//';
+ } else {
+ $start = "/<$tag(\\s+[^>]*|\\s*)>/i";
+ $end = "/<\\/$tag\\s*>/i";
+ }
while ( '' != $text ) {
- if($tag==STRIP_COMMENTS) {
- $p = preg_split( '//i', $p[1], 2 );
- } else {
- $q = preg_split( "/<\\/\\s*$tag\\s*>/i", $p[1], 2 );
- }
- $marker = $rnd . sprintf('%08X', $n++);
- $content[$marker] = $q[0];
- $stripped .= $marker;
$text = $q[1];
}
}
return $stripped;
}
+ /**
+ * Wrapper function for extractTagsAndParams
+ * for cases where $tags and $params isn't needed
+ * i.e. where tags will never have params, like
+ *
+ * @access private
+ * @static
+ */
+ function extractTags( $tag, $text, &$content, $uniq_prefix = '' ) {
+ $dummy_tags = array();
+ $dummy_params = array();
+
+ return Parser::extractTagsAndParams( $tag, $text, $content,
+ $dummy_tags, $dummy_params, $uniq_prefix );
+ }
+
/**
* Strips and renders nowiki, pre, math, hiero
* If $render is set, performs necessary rendering operations on plugins
@@ -283,6 +347,8 @@ class Parser
$pre_content = array();
$comment_content = array();
$ext_content = array();
+ $ext_tags = array();
+ $ext_params = array();
$gallery_content = array();
# Replace any instances of the placeholders
@@ -358,13 +424,16 @@ class Parser
# Extensions
foreach ( $this->mTagHooks as $tag => $callback ) {
- $ext_contents[$tag] = array();
- $text = Parser::extractTags( $tag, $text, $ext_content[$tag], $uniq_prefix );
+ $ext_content[$tag] = array();
+ $text = Parser::extractTagsAndParams( $tag, $text, $ext_content[$tag],
+ $ext_tags[$tag], $ext_params[$tag], $uniq_prefix );
foreach( $ext_content[$tag] as $marker => $content ) {
+ $full_tag = $ext_tags[$tag][$marker];
+ $params = $ext_params[$tag][$marker];
if ( $render ) {
- $ext_content[$tag][$marker] = $callback( $content );
+ $ext_content[$tag][$marker] = $callback( $content, $params );
} else {
- $ext_content[$tag][$marker] = "<$tag>$content$tag>";
+ $ext_content[$tag][$marker] = "$full_tag$content$tag>";
}
}
}
@@ -459,110 +528,108 @@ class Parser
}
/**
- * Return allowed HTML attributes
+ * Interface with html tidy, used if $wgUseTidy = true.
+ * If tidy isn't able to correct the markup, the original will be
+ * returned in all its glory with a warning comment appended.
*
- * @access private
- */
- function getHTMLattrs () {
- $htmlattrs = array( # Allowed attributes--no scripting, etc.
- 'title', 'align', 'lang', 'dir', 'width', 'height',
- 'bgcolor', 'clear', /* BR */ 'noshade', /* HR */
- 'cite', /* BLOCKQUOTE, Q */ 'size', 'face', 'color',
- /* FONT */ 'type', 'start', 'value', 'compact',
- /* For various lists, mostly deprecated but safe */
- 'summary', 'width', 'border', 'frame', 'rules',
- 'cellspacing', 'cellpadding', 'valign', 'char',
- 'charoff', 'colgroup', 'col', 'span', 'abbr', 'axis',
- 'headers', 'scope', 'rowspan', 'colspan', /* Tables */
- 'id', 'class', 'name', 'style' /* For CSS */
- );
- return $htmlattrs ;
- }
-
- /**
- * Remove non approved attributes and javascript in css
+ * Either the external tidy program or the in-process tidy extension
+ * will be used depending on availability. Override the default
+ * $wgTidyInternal setting to disable the internal if it's not working.
*
- * @access private
+ * @param string $text Hideous HTML input
+ * @return string Corrected HTML output
+ * @access public
+ * @static
*/
- function fixTagAttributes ( $t ) {
- if ( trim ( $t ) == '' ) return '' ; # Saves runtime ;-)
- $htmlattrs = $this->getHTMLattrs() ;
-
- # Strip non-approved attributes from the tag
- $t = preg_replace(
- '/(\\w+)(\\s*=\\s*([^\\s\">]+|\"[^\">]*\"))?/e',
- "(in_array(strtolower(\"\$1\"),\$htmlattrs)?(\"\$1\".((\"x\$3\" != \"x\")?\"=\$3\":'')):'')",
- $t);
-
- $t = str_replace ( '<>>' , '' , $t ) ; # This should fix bug 980557
-
- # Strip javascript "expression" from stylesheets. Brute force approach:
- # If anythin offensive is found, all attributes of the HTML tag are dropped
-
- if( preg_match(
- '/style\\s*=.*(expression|tps*:\/\/|url\\s*\().*/is',
- wfMungeToUtf8( $t ) ) )
- {
- $t='';
+ function tidy( $text ) {
+ global $wgTidyInternal;
+ $wrappedtext = ''.
+'test'.$text.'';
+ if( $wgTidyInternal ) {
+ $correctedtext = Parser::internalTidy( $wrappedtext );
+ } else {
+ $correctedtext = Parser::externalTidy( $wrappedtext );
}
-
- return trim ( $t ) ;
+ if( is_null( $correctedtext ) ) {
+ wfDebug( "Tidy error detected!\n" );
+ return $text . "\n\n";
+ }
+ return $correctedtext;
}
-
+
/**
- * interface with html tidy, used if $wgUseTidy = true
+ * Spawn an external HTML tidy process and get corrected markup back from it.
*
- * @access public
+ * @access private
* @static
*/
- function tidy ( $text ) {
+ function externalTidy( $text ) {
global $wgTidyConf, $wgTidyBin, $wgTidyOpts;
- global $wgInputEncoding, $wgOutputEncoding;
- $fname = 'Parser::tidy';
+ $fname = 'Parser::externalTidy';
wfProfileIn( $fname );
$cleansource = '';
- $opts = '';
- switch(strtoupper($wgOutputEncoding)) {
- case 'ISO-8859-1':
- $opts .= ($wgInputEncoding == $wgOutputEncoding)? ' -latin1':' -raw';
- break;
- case 'UTF-8':
- $opts .= ($wgInputEncoding == $wgOutputEncoding)? ' -utf8':' -raw';
- break;
- default:
- $opts .= ' -raw';
- }
+ $opts = ' -utf8';
- $wrappedtext = ''.
-'test'.$text.'';
$descriptorspec = array(
0 => array('pipe', 'r'),
1 => array('pipe', 'w'),
2 => array('file', '/dev/null', 'a')
);
+ $pipes = array();
$process = proc_open("$wgTidyBin -config $wgTidyConf $wgTidyOpts$opts", $descriptorspec, $pipes);
if (is_resource($process)) {
- fwrite($pipes[0], $wrappedtext);
+ fwrite($pipes[0], $text);
fclose($pipes[0]);
while (!feof($pipes[1])) {
$cleansource .= fgets($pipes[1], 1024);
}
fclose($pipes[1]);
- $return_value = proc_close($process);
+ proc_close($process);
}
wfProfileOut( $fname );
if( $cleansource == '' && $text != '') {
- wfDebug( "Tidy error detected!\n" );
- return $text . "\n\n";
+ // Some kind of error happened, so we couldn't get the corrected text.
+ // Just give up; we'll use the source text and append a warning.
+ return null;
} else {
return $cleansource;
}
}
+ /**
+ * Use the HTML tidy PECL extension to use the tidy library in-process,
+ * saving the overhead of spawning a new process. Currently written to
+ * the PHP 4.3.x version of the extension, may not work on PHP 5.
+ *
+ * 'pear install tidy' should be able to compile the extension module.
+ *
+ * @access private
+ * @static
+ */
+ function internalTidy( $text ) {
+ global $wgTidyConf;
+ $fname = 'Parser::internalTidy';
+ wfProfileIn( $fname );
+
+ tidy_load_config( $wgTidyConf );
+ tidy_set_encoding( 'utf8' );
+ tidy_parse_string( $text );
+ tidy_clean_repair();
+ if( tidy_get_status() == 2 ) {
+ // 2 is magic number for fatal error
+ // http://www.php.net/manual/en/function.tidy-get-status.php
+ $cleansource = null;
+ } else {
+ $cleansource = tidy_get_output();
+ }
+ wfProfileOut( $fname );
+ return $cleansource;
+ }
+
/**
* parse the wiki syntax used to render tables
*
@@ -584,9 +651,8 @@ class Parser
$fc = substr ( $x , 0 , 1 ) ;
if ( preg_match( '/^(:*)\{\|(.*)$/', $x, $matches ) ) {
$indent_level = strlen( $matches[1] );
- $t[$k] = "\n" .
- str_repeat( '- ', $indent_level ) .
- '
fixTagAttributes ( $matches[2] ) . '>' ;
+ $t[$k] = str_repeat( '- ', $indent_level ) .
+ '
' ;
array_push ( $td , false ) ;
array_push ( $ltd , '' ) ;
array_push ( $tr , false ) ;
@@ -594,7 +660,7 @@ class Parser
}
else if ( count ( $td ) == 0 ) { } # Don't do any of the following
else if ( '|}' == substr ( $x , 0 , 2 ) ) {
- $z = "
\n" ;
+ $z = "
" . substr ( $x , 2);
$l = array_pop ( $ltd ) ;
if ( array_pop ( $tr ) ) $z = '' . $z ;
if ( array_pop ( $td ) ) $z = ''.$l.'>' . $z ;
@@ -613,7 +679,7 @@ class Parser
array_push ( $tr , false ) ;
array_push ( $td , false ) ;
array_push ( $ltd , '' ) ;
- array_push ( $ltr , $this->fixTagAttributes ( $x ) ) ;
+ array_push ( $ltr , Sanitizer::fixTagAttributes ( $x, 'tr' ) ) ;
}
else if ( '|' == $fc || '!' == $fc || '|+' == substr ( $x , 0 , 2 ) ) { # Caption
# $x is a table row
@@ -633,7 +699,7 @@ class Parser
if ( $fc != '+' )
{
$tra = array_pop ( $ltr ) ;
- if ( !array_pop ( $tr ) ) $z = '\n" ;
+ if ( !array_pop ( $tr ) ) $z = '
\n" ;
array_push ( $tr , true ) ;
array_push ( $ltr , '' ) ;
}
@@ -655,7 +721,7 @@ class Parser
}
if ( count ( $y ) == 1 )
$y = "{$z}<{$l}>{$y[0]}" ;
- else $y = $y = "{$z}<{$l} ".$this->fixTagAttributes($y[0]).">{$y[1]}" ;
+ else $y = $y = "{$z}<{$l}".Sanitizer::fixTagAttributes($y[0], $l).">{$y[1]}" ;
$t[$k] .= $y ;
array_push ( $td , true ) ;
}
@@ -671,7 +737,6 @@ class Parser
}
$t = implode ( "\n" , $t ) ;
- # $t = $this->removeHTMLtags( $t );
wfProfileOut( $fname );
return $t ;
}
@@ -682,21 +747,22 @@ class Parser
*
* @access private
*/
- function internalParse( $text, $linestart, $args = array(), $isMain=true ) {
+ function internalParse( $text ) {
global $wgContLang;
-
+ $args = array();
+ $isMain = true;
$fname = 'Parser::internalParse';
wfProfileIn( $fname );
- $text = $this->removeHTMLtags( $text );
+ $text = Sanitizer::removeHTMLtags( $text, array( &$this, 'replaceVariables' ) );
$text = $this->replaceVariables( $text, $args );
$text = preg_replace( '/(^|\n)-----*/', '\\1
', $text );
$text = $this->doHeadings( $text );
if($this->mOptions->getUseDynamicDates()) {
- global $wgDateFormatter;
- $text = $wgDateFormatter->reformat( $this->mOptions->getDateFormat(), $text );
+ $df =& DateFormatter::getInstance();
+ $text = $df->reformat( $this->mOptions->getDateFormat(), $text );
}
$text = $this->doAllQuotes( $text );
$text = $this->replaceInternalLinks( $text );
@@ -709,8 +775,6 @@ class Parser
$text = $this->doMagicLinks( $text );
$text = $this->doTableStuff( $text );
$text = $this->formatHeadings( $text, $isMain );
- $sk =& $this->mOptions->getSkin();
- $text = $sk->transformContent( $text );
wfProfileOut( $fname );
return $text;
@@ -723,11 +787,7 @@ class Parser
* @access private
*/
function &doMagicLinks( &$text ) {
- global $wgUseGeoMode;
$text = $this->magicISBN( $text );
- if ( isset( $wgUseGeoMode ) && $wgUseGeoMode ) {
- $text = $this->magicGEO( $text );
- }
$text = $this->magicRFC( $text, 'RFC ', 'rfcurl' );
$text = $this->magicRFC( $text, 'PMID ', 'pubmedurl' );
return $text;
@@ -955,11 +1015,12 @@ class Parser
* @access private
*/
function replaceExternalLinks( $text ) {
+ global $wgContLang;
$fname = 'Parser::replaceExternalLinks';
wfProfileIn( $fname );
$sk =& $this->mOptions->getSkin();
- $linktrail = wfMsgForContent('linktrail');
+
$bits = preg_split( EXT_LINK_BRACKETED, $text, -1, PREG_SPLIT_DELIM_CAPTURE );
$s = $this->replaceFreeExternalLinks( array_shift( $bits ) );
@@ -981,53 +1042,50 @@ class Parser
# If the link text is an image URL, replace it with an tag
# This happened by accident in the original parser, but some people used it extensively
- $img = $this->maybeMakeImageLink( $text );
+ $img = $this->maybeMakeExternalImage( $text );
if ( $img !== false ) {
$text = $img;
}
$dtrail = '';
+ # Set linktype for CSS - if URL==text, link is essentially free
+ $linktype = ($text == $url) ? 'free' : 'text';
+
# No link text, e.g. [http://domain.tld/some.link]
if ( $text == '' ) {
# Autonumber if allowed
if ( strpos( HTTP_PROTOCOLS, $protocol ) !== false ) {
$text = '[' . ++$this->mAutonumber . ']';
+ $linktype = 'autonumber';
} else {
# Otherwise just use the URL
$text = htmlspecialchars( $url );
+ $linktype = 'free';
}
} else {
# Have link text, e.g. [http://domain.tld/some.link text]s
# Check for trail
- if ( preg_match( $linktrail, $trail, $m2 ) ) {
- $dtrail = $m2[1];
- $trail = $m2[2];
- }
+ list( $dtrail, $trail ) = Linker::splitTrail( $trail );
}
- $encUrl = htmlspecialchars( $url );
- # Bit in parentheses showing the URL for the printable version
- if( $url == $text || preg_match( "!$protocol://" . preg_quote( $text, '/' ) . "/?$!", $url ) ) {
- $paren = '';
- } else {
- # Expand the URL for printable version
- if ( ! $sk->suppressUrlExpansion() ) {
- $paren = " (" . htmlspecialchars ( $encUrl ) . ")";
- } else {
- $paren = '';
- }
- }
+ $text = $wgContLang->markNoConversion($text);
+
+ # Replace & from obsolete syntax with &.
+ # All HTML entities will be escaped by makeExternalLink()
+ # or maybeMakeExternalImage()
+ $url = str_replace( '&', '&', $url );
# Process the trail (i.e. everything after this link up until start of the next link),
# replacing any non-bracketed links
$trail = $this->replaceFreeExternalLinks( $trail );
+
# Use the encoded URL
# This means that users can paste URLs directly into the text
# Funny characters like ö aren't valid in URLs anyway
# This was changed in August 2004
- $s .= $sk->makeExternalLink( $url, $text, false ) . $dtrail. $paren . $trail;
+ $s .= $sk->makeExternalLink( $url, $text, false, $linktype ) . $dtrail . $trail;
}
wfProfileOut( $fname );
@@ -1039,10 +1097,11 @@ class Parser
* @access private
*/
function replaceFreeExternalLinks( $text ) {
+ global $wgContLang;
$fname = 'Parser::replaceFreeExternalLinks';
wfProfileIn( $fname );
- $bits = preg_split( '/((?:'.URL_PROTOCOLS.'):)/S', $text, -1, PREG_SPLIT_DELIM_CAPTURE );
+ $bits = preg_split( '/(\b(?:'.URL_PROTOCOLS.'):)/S', $text, -1, PREG_SPLIT_DELIM_CAPTURE );
$s = array_shift( $bits );
$i = 0;
@@ -1080,14 +1139,14 @@ class Parser
# Replace & from obsolete syntax with &.
# All HTML entities will be escaped by makeExternalLink()
- # or maybeMakeImageLink()
+ # or maybeMakeExternalImage()
$url = str_replace( '&', '&', $url );
# Is this an external image?
- $text = $this->maybeMakeImageLink( $url );
+ $text = $this->maybeMakeExternalImage( $url );
if ( $text === false ) {
# Not an image, make a link
- $text = $sk->makeExternalLink( $url, $url );
+ $text = $sk->makeExternalLink( $url, $wgContLang->markNoConversion($url), true, 'free' );
}
$s .= $text . $trail;
} else {
@@ -1102,13 +1161,13 @@ class Parser
* make an image if it's allowed
* @access private
*/
- function maybeMakeImageLink( $url ) {
+ function maybeMakeExternalImage( $url ) {
$sk =& $this->mOptions->getSkin();
$text = false;
if ( $this->mOptions->getAllowExternalImages() ) {
if ( preg_match( EXT_IMAGE_REGEX, $url ) ) {
# Image found
- $text = $sk->makeImage( htmlspecialchars( $url ) );
+ $text = $sk->makeExternalImage( htmlspecialchars( $url ) );
}
}
return $text;
@@ -1119,10 +1178,8 @@ class Parser
*
* @access private
*/
-
function replaceInternalLinks( $s ) {
- global $wgLang, $wgContLang, $wgLinkCache;
- global $wgDisableLangConversion;
+ global $wgContLang, $wgLinkCache;
static $fname = 'Parser::replaceInternalLinks' ;
wfProfileIn( $fname );
@@ -1133,12 +1190,6 @@ class Parser
if ( !$tc ) { $tc = Title::legalChars() . '#%'; }
$sk =& $this->mOptions->getSkin();
- global $wgUseOldExistenceCheck;
- # "Post-parse link colour check" works only on wiki text since it's now
- # in Parser. Enable it, then disable it when we're done.
- $saveParseColour = $sk->postParseLinkColour( !$wgUseOldExistenceCheck );
-
- $redirect = MagicWord::get ( MAG_REDIRECT ) ;
#split the entire text string on occurences of [[
$a = explode( '[[', ' ' . $s );
@@ -1148,7 +1199,7 @@ class Parser
# Match a link having the form [[namespace:link|alternate]]trail
static $e1 = FALSE;
- if ( !$e1 ) { $e1 = "/^([{$tc}]+)(?:\\|([^]]+))?]](.*)\$/sD"; }
+ if ( !$e1 ) { $e1 = "/^([{$tc}]+)(?:\\|(.+?))?]](.*)\$/sD"; }
# Match cases where there is no "]]", which might still be images
static $e1_img = FALSE;
if ( !$e1_img ) { $e1_img = "/^([{$tc}]+)\\|(.*)\$/sD"; }
@@ -1158,7 +1209,10 @@ class Parser
$useLinkPrefixExtension = $wgContLang->linkPrefixExtension();
- $nottalk = !Namespace::isTalk( $this->mTitle->getNamespace() );
+ if( is_null( $this->mTitle ) ) {
+ wfDebugDieBacktrace( 'nooo' );
+ }
+ $nottalk = !$this->mTitle->isTalkPage();
if ( $useLinkPrefixExtension ) {
if ( preg_match( $e2, $s, $m ) ) {
@@ -1200,6 +1254,18 @@ class Parser
if ( preg_match( $e1, $line, $m ) ) { # page with normal text or alt
$text = $m[2];
+ # If we get a ] at the beginning of $m[3] that means we have a link that's something like:
+ # [[Image:Foo.jpg|[http://example.com desc]]] <- having three ] in a row fucks up,
+ # the real problem is with the $e1 regex
+ # See bug 1300.
+ #
+ # Still some problems for cases where the ] is meant to be outside punctuation,
+ # and no image is in sight. See bug 2095.
+ #
+ if( $text !== '' && preg_match( "/^\](.*)/s", $m[3], $n ) ) {
+ $text .= ']'; # so that replaceExternalLinks($text) works later
+ $m[3] = $n[1];
+ }
# fix up urlencoded title texts
if(preg_match('/%/', $m[1] )) $m[1] = urldecode($m[1]);
$trail = $m[3];
@@ -1216,7 +1282,7 @@ class Parser
# Don't allow internal links to pages containing
# PROTO: where PROTO is a valid URL protocol; these
# should be external links.
- if (preg_match('/^((?:'.URL_PROTOCOLS.'):)/', $m[1])) {
+ if (preg_match('/^(\b(?:'.URL_PROTOCOLS.'):)/', $m[1])) {
$s .= $prefix . '[[' . $line ;
continue;
}
@@ -1296,37 +1362,38 @@ class Parser
# Interwikis
if( $iw && $this->mOptions->getInterwikiMagic() && $nottalk && $wgContLang->getLanguageName( $iw ) ) {
array_push( $this->mOutput->mLanguageLinks, $nt->getFullText() );
- $tmp = $prefix . $trail ;
- $s .= (trim($tmp) == '')? '': $tmp;
+ $s = rtrim($s . "\n");
+ $s .= trim($prefix . $trail, "\n") == '' ? '': $prefix . $trail;
continue;
}
if ( $ns == NS_IMAGE ) {
wfProfileIn( "$fname-image" );
-
- # recursively parse links inside the image caption
- # actually, this will parse them in any other parameters, too,
- # but it might be hard to fix that, and it doesn't matter ATM
- $text = $this->replaceExternalLinks($text);
- $text = $this->replaceInternalLinks($text);
-
- # replace the image with a link-holder so that replaceExternalLinks() can't mess with it
- $s .= $prefix . $this->insertStripItem( $sk->makeImageLinkObj( $nt, $text ), $this->mStripState ) . $trail;
- $wgLinkCache->addImageLinkObj( $nt );
-
+ if ( !wfIsBadImage( $nt->getDBkey() ) ) {
+ # recursively parse links inside the image caption
+ # actually, this will parse them in any other parameters, too,
+ # but it might be hard to fix that, and it doesn't matter ATM
+ $text = $this->replaceExternalLinks($text);
+ $text = $this->replaceInternalLinks($text);
+
+ # cloak any absolute URLs inside the image markup, so replaceExternalLinks() won't touch them
+ $s .= $prefix . str_replace('http://', 'http-noparse://', $this->makeImage( $nt, $text ) ) . $trail;
+ $wgLinkCache->addImageLinkObj( $nt );
+
+ wfProfileOut( "$fname-image" );
+ continue;
+ }
wfProfileOut( "$fname-image" );
- continue;
+
}
if ( $ns == NS_CATEGORY ) {
wfProfileIn( "$fname-category" );
- $t = $nt->getText();
+ $t = $wgContLang->convert($nt->getText());
+ $s = rtrim($s . "\n"); # bug 87
$wgLinkCache->suspend(); # Don't save in links/brokenlinks
- $pPLC=$sk->postParseLinkColour();
- $sk->postParseLinkColour( false );
$t = $sk->makeLinkObj( $nt, $t, '', '' , $prefix );
- $sk->postParseLinkColour( $pPLC );
$wgLinkCache->resume();
if ( $wasblank ) {
@@ -1338,9 +1405,15 @@ class Parser
} else {
$sortkey = $text;
}
+ $sortkey = $wgContLang->convertCategoryKey( $sortkey );
$wgLinkCache->addCategoryLinkObj( $nt, $sortkey );
$this->mOutput->addCategoryLink( $t );
- $s .= $prefix . $trail ;
+
+ /**
+ * Strip the whitespace Category links produce, see bug 87
+ * @todo We might want to use trim($tmp, "\n") here.
+ */
+ $s .= trim($prefix . $trail, "\n") == '' ? '': $prefix . $trail;
wfProfileOut( "$fname-category" );
continue;
@@ -1363,13 +1436,59 @@ class Parser
$s .= $prefix . $sk->makeKnownLinkObj( $nt, $text, '', $trail );
continue;
}
- $s .= $sk->makeLinkObj( $nt, $text, '', $trail, $prefix );
+ if( $nt->isLocal() && $nt->isAlwaysKnown() ) {
+ /**
+ * Skip lookups for special pages and self-links.
+ * External interwiki links are not included here because
+ * the HTTP urls would break output in the next parse step;
+ * they will have placeholders kept.
+ */
+ $s .= $sk->makeKnownLinkObj( $nt, $text, '', $trail, $prefix );
+ } else {
+ /**
+ * Add a link placeholder
+ * Later, this will be replaced by a real link, after the existence or
+ * non-existence of all the links is known
+ */
+ $s .= $this->makeLinkHolder( $nt, $text, '', $trail, $prefix );
+ }
}
- $sk->postParseLinkColour( $saveParseColour );
wfProfileOut( $fname );
return $s;
}
+ /**
+ * Make a link placeholder. The text returned can be later resolved to a real link with
+ * replaceLinkHolders(). This is done for two reasons: firstly to avoid further
+ * parsing of interwiki links, and secondly to allow all extistence checks and
+ * article length checks (for stub links) to be bundled into a single query.
+ *
+ */
+ function makeLinkHolder( &$nt, $text = '', $query = '', $trail = '', $prefix = '' ) {
+ if ( ! is_object($nt) ) {
+ # Fail gracefully
+ $retVal = "{$prefix}{$text}{$trail}";
+ } else {
+ # Separate the link trail from the rest of the link
+ list( $inside, $trail ) = Linker::splitTrail( $trail );
+
+ if ( $nt->isExternal() ) {
+ $nr = array_push( $this->mInterwikiLinkHolders['texts'], $prefix.$text.$inside );
+ $this->mInterwikiLinkHolders['titles'][] =& $nt;
+ $retVal = '{$trail}";
+ } else {
+ $nr = array_push( $this->mLinkHolders['namespaces'], $nt->getNamespace() );
+ $this->mLinkHolders['dbkeys'][] = $nt->getDBkey();
+ $this->mLinkHolders['queries'][] = $query;
+ $this->mLinkHolders['texts'][] = $prefix.$text.$inside;
+ $this->mLinkHolders['titles'][] =& $nt;
+
+ $retVal = '{$trail}";
+ }
+ }
+ return $retVal;
+ }
+
/**
* Return true if subpage links should be expanded on this page.
* @return bool
@@ -1541,7 +1660,7 @@ class Parser
#
$textLines = explode( "\n", $text );
- $lastPrefix = $output = $lastLine = '';
+ $lastPrefix = $output = '';
$this->mDTopen = $inBlockElem = false;
$prefixLength = 0;
$paragraphStack = false;
@@ -1580,6 +1699,7 @@ class Parser
# ; title : definition text
# So we check for : in the remainder text to split up the
# title and definition, without b0rking links.
+ $term = $t2 = '';
if ($this->findColonNoLinks($t, $term, $t2) !== false) {
$t = $t2;
$output .= $term . $this->nextItem( ':' );
@@ -1734,7 +1854,7 @@ class Parser
* @access private
*/
function getVariableValue( $index ) {
- global $wgContLang, $wgSitename, $wgServer;
+ global $wgContLang, $wgSitename, $wgServer, $wgArticle;
/**
* Some of these require message or data lookups and can be
@@ -1750,21 +1870,29 @@ class Parser
return $varCache[$index] = $wgContLang->getMonthName( date('n') );
case MAG_CURRENTMONTHNAMEGEN:
return $varCache[$index] = $wgContLang->getMonthNameGen( date('n') );
+ case MAG_CURRENTMONTHABBREV:
+ return $varCache[$index] = $wgContLang->getMonthAbbreviation( date('n') );
case MAG_CURRENTDAY:
return $varCache[$index] = $wgContLang->formatNum( date('j') );
case MAG_PAGENAME:
return $this->mTitle->getText();
case MAG_PAGENAMEE:
return $this->mTitle->getPartialURL();
+ case MAG_REVISIONID:
+ return $wgArticle->getRevIdFetched();
case MAG_NAMESPACE:
# return Namespace::getCanonicalName($this->mTitle->getNamespace());
return $wgContLang->getNsText($this->mTitle->getNamespace()); # Patch by Dori
case MAG_CURRENTDAYNAME:
return $varCache[$index] = $wgContLang->getWeekdayName( date('w')+1 );
case MAG_CURRENTYEAR:
- return $varCache[$index] = $wgContLang->formatNum( date( 'Y' ) );
+ return $varCache[$index] = $wgContLang->formatNum( date( 'Y' ), true );
case MAG_CURRENTTIME:
return $varCache[$index] = $wgContLang->time( wfTimestampNow(), false );
+ case MAG_CURRENTWEEK:
+ return $varCache[$index] = $wgContLang->formatNum( date('W') );
+ case MAG_CURRENTDOW:
+ return $varCache[$index] = $wgContLang->formatNum( date('w') );
case MAG_NUMBEROFARTICLES:
return $varCache[$index] = $wgContLang->formatNum( wfNumberOfArticles() );
case MAG_SITENAME:
@@ -1788,7 +1916,7 @@ class Parser
$this->mVariables = array();
foreach ( $wgVariableIDs as $id ) {
$mw =& MagicWord::get( $id );
- $mw->addToArray( $this->mVariables, $this->getVariableValue( $id ) );
+ $mw->addToArray( $this->mVariables, $id );
}
wfProfileOut( $fname );
}
@@ -1808,7 +1936,6 @@ class Parser
* @access private
*/
function replaceVariables( $text, $args = array() ) {
- global $wgLang, $wgScript, $wgArticlePath;
# Prevent too big inclusions
if( strlen( $text ) > MAX_INCLUDE_SIZE ) {
@@ -1846,6 +1973,7 @@ class Parser
*/
function variableSubstitution( $matches ) {
$fname = 'parser::variableSubstitution';
+ $varname = $matches[1];
wfProfileIn( $fname );
if ( !$this->mVariables ) {
$this->initialiseVariables();
@@ -1854,14 +1982,15 @@ class Parser
if ( $this->mOutputType == OT_WIKI ) {
# Do only magic variables prefixed by SUBST
$mwSubst =& MagicWord::get( MAG_SUBST );
- if (!$mwSubst->matchStartAndRemove( $matches[1] ))
+ if (!$mwSubst->matchStartAndRemove( $varname ))
$skip = true;
# Note that if we don't substitute the variable below,
# we don't remove the {{subst:}} magic word, in case
# it is a template rather than a magic variable.
}
- if ( !$skip && array_key_exists( $matches[1], $this->mVariables ) ) {
- $text = $this->mVariables[$matches[1]];
+ if ( !$skip && array_key_exists( $varname, $this->mVariables ) ) {
+ $id = $this->mVariables[$varname];
+ $text = $this->getVariableValue( $id );
$this->mOutput->mContainsOldMagic = true;
} else {
$text = $matches[0];
@@ -1882,7 +2011,7 @@ class Parser
# merged with the next arg because the '|' character between belongs
# to the link syntax and not the template parameter syntax.
$argc = count($args);
- $i = 0;
+
for ( $i = 0; $i < $argc-1; $i++ ) {
if ( substr_count ( $args[$i], '[[' ) != substr_count ( $args[$i], ']]' ) ) {
$args[$i] .= '|'.$args[$i+1];
@@ -2031,20 +2160,26 @@ class Parser
# Did we encounter this template already? If yes, it is in the cache
# and we need to check for loops.
if ( !$found && isset( $this->mTemplates[$part1] ) ) {
- # set $text to cached message.
- $text = $linestart . $this->mTemplates[$part1];
$found = true;
# Infinite loop test
if ( isset( $this->mTemplatePath[$part1] ) ) {
$noparse = true;
$found = true;
- $text .= '';
+ $text = $linestart .
+ "\{\{$part1}}" .
+ '';
+ wfDebug( "$fname: template loop broken at '$part1'\n" );
+ } else {
+ # set $text to cached message.
+ $text = $linestart . $this->mTemplates[$part1];
}
}
# Load from database
- $itcamefromthedatabase = false;
+ $replaceHeadings = false;
+ $isHTML = false;
+ $lastPathLevel = $this->mTemplatePath;
if ( !$found ) {
$ns = NS_TEMPLATE;
$part1 = $this->maybeDoSubpageLink( $part1, $subpage='' );
@@ -2056,24 +2191,37 @@ class Parser
# Check for excessive inclusion
$dbk = $title->getPrefixedDBkey();
if ( $this->incrementIncludeCount( $dbk ) ) {
- # This should never be reached.
- $article = new Article( $title );
- $articleContent = $article->getContentWithoutUsingSoManyDamnGlobals();
- if ( $articleContent !== false ) {
- $found = true;
- $text = $linestart . $articleContent;
- $itcamefromthedatabase = true;
+ if ( $title->getNamespace() == NS_SPECIAL && $this->mOptions->getAllowSpecialInclusion() ) {
+ # Capture special page output
+ $text = SpecialPage::capturePath( $title );
+ if ( $text && !is_object( $text ) ) {
+ $found = true;
+ $noparse = true;
+ $isHTML = true;
+ $this->mOutput->setCacheTime( -1 );
+ }
+ } else {
+ $article = new Article( $title );
+ $articleContent = $article->getContentWithoutUsingSoManyDamnGlobals();
+ if ( $articleContent !== false ) {
+ $found = true;
+ $text = $articleContent;
+ $replaceHeadings = true;
+ }
}
}
# If the title is valid but undisplayable, make a link to it
if ( $this->mOutputType == OT_HTML && !$found ) {
- $text = $linestart . '[['.$title->getPrefixedText().']]';
+ $text = '[['.$title->getPrefixedText().']]';
$found = true;
}
# Template cache array insertion
- $this->mTemplates[$part1] = $text;
+ if( $found ) {
+ $this->mTemplates[$part1] = $text;
+ $text = $linestart . $text;
+ }
}
}
@@ -2104,8 +2252,10 @@ class Parser
# Add a new element to the templace recursion path
$this->mTemplatePath[$part1] = 1;
- $text = $this->strip( $text, $this->mStripState );
- $text = $this->removeHTMLtags( $text );
+ if( $this->mOutputType == OT_HTML ) {
+ $text = $this->strip( $text, $this->mStripState );
+ $text = Sanitizer::removeHTMLtags( $text, array( &$this, 'replaceVariables' ), $assocArgs );
+ }
$text = $this->replaceVariables( $text, $assocArgs );
# Resume the link cache and register the inclusion as a link
@@ -2119,43 +2269,50 @@ class Parser
$text = "\n" . $text;
}
}
-
- # Empties the template path
- $this->mTemplatePath = array();
+ # Prune lower levels off the recursion check path
+ $this->mTemplatePath = $lastPathLevel;
+
if ( !$found ) {
wfProfileOut( $fname );
return $matches[0];
} else {
- # replace ==section headers==
- # XXX this needs to go away once we have a better parser.
- if ( $this->mOutputType != OT_WIKI && $itcamefromthedatabase ) {
- if( !is_null( $title ) )
- $encodedname = base64_encode($title->getPrefixedDBkey());
- else
- $encodedname = base64_encode("");
- $m = preg_split('/(^={1,6}.*?={1,6}\s*?$)/m', $text, -1,
- PREG_SPLIT_DELIM_CAPTURE);
- $text = '';
- $nsec = 0;
- for( $i = 0; $i < count($m); $i += 2 ) {
- $text .= $m[$i];
- if (!isset($m[$i + 1]) || $m[$i + 1] == "") continue;
- $hl = $m[$i + 1];
- if( strstr($hl, "" . $m2[3];
+
+ $nsec++;
}
- preg_match('/^(={1,6})(.*?)(={1,6})\s*?$/m', $hl, $m2);
- $text .= $m2[1] . $m2[2] . "" . $m2[3];
-
- $nsec++;
}
}
}
-
- # Empties the template path
- $this->mTemplatePath = array();
+
+ # Prune lower levels off the recursion check path
+ $this->mTemplatePath = $lastPathLevel;
if ( !$found ) {
wfProfileOut( $fname );
@@ -2197,171 +2354,6 @@ class Parser
}
}
-
- /**
- * Cleans up HTML, removes dangerous tags and attributes, and
- * removes HTML comments
- * @access private
- */
- function removeHTMLtags( $text ) {
- global $wgUseTidy, $wgUserHtml;
- $fname = 'Parser::removeHTMLtags';
- wfProfileIn( $fname );
-
- if( $wgUserHtml ) {
- $htmlpairs = array( # Tags that must be closed
- 'b', 'del', 'i', 'ins', 'u', 'font', 'big', 'small', 'sub', 'sup', 'h1',
- 'h2', 'h3', 'h4', 'h5', 'h6', 'cite', 'code', 'em', 's',
- 'strike', 'strong', 'tt', 'var', 'div', 'center',
- 'blockquote', 'ol', 'ul', 'dl', 'table', 'caption', 'pre',
- 'ruby', 'rt' , 'rb' , 'rp', 'p'
- );
- $htmlsingle = array(
- 'br', 'hr', 'li', 'dt', 'dd'
- );
- $htmlnest = array( # Tags that can be nested--??
- 'table', 'tr', 'td', 'th', 'div', 'blockquote', 'ol', 'ul',
- 'dl', 'font', 'big', 'small', 'sub', 'sup'
- );
- $tabletags = array( # Can only appear inside table
- 'td', 'th', 'tr'
- );
- } else {
- $htmlpairs = array();
- $htmlsingle = array();
- $htmlnest = array();
- $tabletags = array();
- }
-
- $htmlsingle = array_merge( $tabletags, $htmlsingle );
- $htmlelements = array_merge( $htmlsingle, $htmlpairs );
-
- $htmlattrs = $this->getHTMLattrs () ;
-
- # Remove HTML comments
- $text = $this->removeHTMLcomments( $text );
-
- $bits = explode( '<', $text );
- $text = array_shift( $bits );
- if(!$wgUseTidy) {
- $tagstack = array(); $tablestack = array();
- foreach ( $bits as $x ) {
- $prev = error_reporting( E_ALL & ~( E_NOTICE | E_WARNING ) );
- preg_match( '/^(\\/?)(\\w+)([^>]*)(\\/{0,1}>)([^<]*)$/',
- $x, $regs );
- list( $qbar, $slash, $t, $params, $brace, $rest ) = $regs;
- error_reporting( $prev );
-
- $badtag = 0 ;
- if ( in_array( $t = strtolower( $t ), $htmlelements ) ) {
- # Check our stack
- if ( $slash ) {
- # Closing a tag...
- if ( ! in_array( $t, $htmlsingle ) &&
- ( $ot = @array_pop( $tagstack ) ) != $t ) {
- @array_push( $tagstack, $ot );
- $badtag = 1;
- } else {
- if ( $t == 'table' ) {
- $tagstack = array_pop( $tablestack );
- }
- $newparams = '';
- }
- } else {
- # Keep track for later
- if ( in_array( $t, $tabletags ) &&
- ! in_array( 'table', $tagstack ) ) {
- $badtag = 1;
- } else if ( in_array( $t, $tagstack ) &&
- ! in_array ( $t , $htmlnest ) ) {
- $badtag = 1 ;
- } else if ( ! in_array( $t, $htmlsingle ) ) {
- if ( $t == 'table' ) {
- array_push( $tablestack, $tagstack );
- $tagstack = array();
- }
- array_push( $tagstack, $t );
- }
- # Strip non-approved attributes from the tag
- $newparams = $this->fixTagAttributes($params);
-
- }
- if ( ! $badtag ) {
- $rest = str_replace( '>', '>', $rest );
- $text .= "<$slash$t $newparams$brace$rest";
- continue;
- }
- }
- $text .= '<' . str_replace( '>', '>', $x);
- }
- # Close off any remaining tags
- while ( is_array( $tagstack ) && ($t = array_pop( $tagstack )) ) {
- $text .= "$t>\n";
- if ( $t == 'table' ) { $tagstack = array_pop( $tablestack ); }
- }
- } else {
- # this might be possible using tidy itself
- foreach ( $bits as $x ) {
- preg_match( '/^(\\/?)(\\w+)([^>]*)(\\/{0,1}>)([^<]*)$/',
- $x, $regs );
- @list( $qbar, $slash, $t, $params, $brace, $rest ) = $regs;
- if ( in_array( $t = strtolower( $t ), $htmlelements ) ) {
- $newparams = $this->fixTagAttributes($params);
- $rest = str_replace( '>', '>', $rest );
- $text .= "<$slash$t $newparams$brace$rest";
- } else {
- $text .= '<' . str_replace( '>', '>', $x);
- }
- }
- }
- wfProfileOut( $fname );
- return $text;
- }
-
- /**
- * Remove '', and everything between.
- * To avoid leaving blank lines, when a comment is both preceded
- * and followed by a newline (ignoring spaces), trim leading and
- * trailing spaces and one of the newlines.
- *
- * @access private
- */
- function removeHTMLcomments( $text ) {
- $fname='Parser::removeHTMLcomments';
- wfProfileIn( $fname );
- while (($start = strpos($text, '', $start + 4);
- if ($end === false) {
- # Unterminated comment; bail out
- break;
- }
-
- $end += 3;
-
- # Trim space and newline if the comment is both
- # preceded and followed by a newline
- $spaceStart = max($start - 1, 0);
- $spaceLen = $end - $spaceStart;
- while (substr($text, $spaceStart, 1) === ' ' && $spaceStart > 0) {
- $spaceStart--;
- $spaceLen++;
- }
- while (substr($text, $spaceStart + $spaceLen, 1) === ' ')
- $spaceLen++;
- if (substr($text, $spaceStart, 1) === "\n" and substr($text, $spaceStart + $spaceLen, 1) === "\n") {
- # Remove the comment, leading and trailing
- # spaces, and leave only one newline.
- $text = substr_replace($text, "\n", $spaceStart, $spaceLen + 1);
- }
- else {
- # Remove just the comment.
- $text = substr_replace($text, '', $start, $end - $start);
- }
- }
- wfProfileOut( $fname );
- return $text;
- }
-
/**
* This function accomplishes several tasks:
* 1) Auto-number headings if that option is enabled
@@ -2371,20 +2363,21 @@ class Parser
*
* It loops through all headlines, collects the necessary data, then splits up the
* string and re-inserts the newly formatted headlines.
+ *
+ * @param string $text
+ * @param boolean $isMain
* @access private
*/
- /* private */ function formatHeadings( $text, $isMain=true ) {
- global $wgInputEncoding, $wgMaxTocLevel, $wgContLang, $wgLinkHolders;
+ function formatHeadings( $text, $isMain=true ) {
+ global $wgMaxTocLevel, $wgContLang, $wgLinkHolders, $wgInterwikiLinkHolders;
$doNumberHeadings = $this->mOptions->getNumberHeadings();
- $doShowToc = $this->mOptions->getShowToc();
+ $doShowToc = true;
$forceTocHere = false;
if( !$this->mTitle->userCanEdit() ) {
$showEditLink = 0;
- $rightClickHack = 0;
} else {
$showEditLink = $this->mOptions->getEditSection();
- $rightClickHack = $this->mOptions->getEditSectionOnRightClick();
}
# Inhibit editsection links if requested in the page
@@ -2396,40 +2389,38 @@ class Parser
# do not add TOC
$mw =& MagicWord::get( MAG_NOTOC );
if( $mw->matchAndRemove( $text ) ) {
- $doShowToc = 0;
- }
-
- # never add the TOC to the Main Page. This is an entry page that should not
- # be more than 1-2 screens large anyway
- if( $this->mTitle->getPrefixedText() == wfMsg('mainpage') ) {
- $doShowToc = 0;
+ $doShowToc = false;
}
# Get all headlines for numbering them and adding funky stuff like [edit]
# links - this is for later, but we need the number of headlines right now
- $numMatches = preg_match_all( '/)(.*?)<\/H[1-6]>/i', $text, $matches );
+ $numMatches = preg_match_all( '/)(.*?)<\/H[1-6] *>/i', $text, $matches );
# if there are fewer than 4 headlines in the article, do not show TOC
if( $numMatches < 4 ) {
- $doShowToc = 0;
+ $doShowToc = false;
}
# if the string __TOC__ (not case-sensitive) occurs in the HTML,
# override above conditions and always show TOC at that place
+
$mw =& MagicWord::get( MAG_TOC );
- if ($mw->match( $text ) ) {
- $doShowToc = 1;
+ if($mw->match( $text ) ) {
+ $doShowToc = true;
$forceTocHere = true;
} else {
# if the string __FORCETOC__ (not case-sensitive) occurs in the HTML,
# override above conditions and always show TOC above first header
$mw =& MagicWord::get( MAG_FORCETOC );
if ($mw->matchAndRemove( $text ) ) {
- $doShowToc = 1;
+ $doShowToc = true;
}
}
-
+ # Never ever show TOC if no headers
+ if( $numMatches < 1 ) {
+ $doShowToc = false;
+ }
# We need this to perform operations on the HTML
$sk =& $this->mOptions->getSkin();
@@ -2440,17 +2431,22 @@ class Parser
# Ugh .. the TOC should have neat indentation levels which can be
# passed to the skin functions. These are determined here
- $toclevel = 0;
$toc = '';
$full = '';
$head = array();
$sublevelCount = array();
+ $levelCount = array();
+ $toclevel = 0;
$level = 0;
$prevlevel = 0;
+ $toclevel = 0;
+ $prevtoclevel = 0;
+
foreach( $matches[3] as $headline ) {
$istemplate = 0;
- $templatetitle = "";
+ $templatetitle = '';
$templatesection = 0;
+ $numbering = '';
if (preg_match("//", $headline, $mat)) {
$istemplate = 1;
@@ -2459,28 +2455,54 @@ class Parser
$headline = preg_replace("//", "", $headline);
}
- $numbering = '';
- if( $level ) {
+ if( $toclevel ) {
$prevlevel = $level;
+ $prevtoclevel = $toclevel;
}
$level = $matches[1][$headlineCount];
- if( ( $doNumberHeadings || $doShowToc ) && $prevlevel && $level > $prevlevel ) {
- # reset when we enter a new level
- $sublevelCount[$level] = 0;
- $toc .= $sk->tocIndent( $level - $prevlevel );
- $toclevel += $level - $prevlevel;
- }
- if( ( $doNumberHeadings || $doShowToc ) && $level < $prevlevel ) {
- # reset when we step back a level
- $sublevelCount[$level+1]=0;
- $toc .= $sk->tocUnindent( $prevlevel - $level );
- $toclevel -= $prevlevel - $level;
- }
- # count number of headlines for each level
- @$sublevelCount[$level]++;
+
if( $doNumberHeadings || $doShowToc ) {
+
+ if ( $level > $prevlevel ) {
+ # Increase TOC level
+ $toclevel++;
+ $sublevelCount[$toclevel] = 0;
+ $toc .= $sk->tocIndent();
+ }
+ elseif ( $level < $prevlevel && $toclevel > 1 ) {
+ # Decrease TOC level, find level to jump to
+
+ if ( $toclevel == 2 && $level <= $levelCount[1] ) {
+ # Can only go down to level 1
+ $toclevel = 1;
+ } else {
+ for ($i = $toclevel; $i > 0; $i--) {
+ if ( $levelCount[$i] == $level ) {
+ # Found last matching level
+ $toclevel = $i;
+ break;
+ }
+ elseif ( $levelCount[$i] < $level ) {
+ # Found first matching level below current level
+ $toclevel = $i + 1;
+ break;
+ }
+ }
+ }
+
+ $toc .= $sk->tocUnindent( $prevtoclevel - $toclevel );
+ }
+ else {
+ # No change in level, end TOC line
+ $toc .= $sk->tocLineEnd();
+ }
+
+ $levelCount[$toclevel] = $level;
+
+ # count number of headlines for each level
+ @$sublevelCount[$toclevel]++;
$dot = 0;
- for( $i = 1; $i <= $level; $i++ ) {
+ for( $i = 1; $i <= $toclevel; $i++ ) {
if( !empty( $sublevelCount[$i] ) ) {
if( $dot ) {
$numbering .= '.';
@@ -2494,41 +2516,38 @@ class Parser
# The canonized header is a version of the header text safe to use for links
# Avoid insertion of weird stuff like