From 8b9bedbad7e74849e420c26381d3e58f6cd712f9 Mon Sep 17 00:00:00 2001 From: Tim Starling Date: Wed, 27 Jan 2010 02:41:22 +0000 Subject: [PATCH] Revert r61528, r61527, r61526, r61525, r61519, r61515, r61053, r61052 (Parser::doQuotes() rewrite). Lots of issues to discuss, needs more review than I have time to give it pre-1.16. I'll split it out to a branch. --- RELEASE-NOTES | 2 - includes/StringUtils.php | 87 ------------ includes/parser/Parser.php | 273 ++++++++++++++++-------------------- maintenance/parserTests.txt | 90 ++---------- tests/preg_split_test.php | 24 ---- 5 files changed, 132 insertions(+), 344 deletions(-) delete mode 100644 tests/preg_split_test.php diff --git a/RELEASE-NOTES b/RELEASE-NOTES index 7daff4b4eb..3230382630 100644 --- a/RELEASE-NOTES +++ b/RELEASE-NOTES @@ -711,8 +711,6 @@ it from source control: http://www.mediawiki.org/wiki/Download_from_SVN * (bug 9794) User rights log entries for foreign user now links to the foreign user's page if possible * (bug 14717) Don't load nonexistent CSS fix files for non-Monobook skins -* (bug 18765) Increased consistency of bold-italic markup for unbalanced quotes. - Improved representation of six quotes (may break existing markup). * (bug 22034) Use wfClientAcceptsGzip() in wfGzipHandler instead of reimplementing it. * (bug 19226) First line renders differently on many UI messages. diff --git a/includes/StringUtils.php b/includes/StringUtils.php index bab9be4bbe..c437b3c19e 100644 --- a/includes/StringUtils.php +++ b/includes/StringUtils.php @@ -179,14 +179,6 @@ class StringUtils { return new ArrayIterator( explode( $separator, $subject ) ); } } - - /** - * Workalike for preg_split() with limited memory usage. - * Returns an Iterator - */ - static function preg_split( $pattern, $subject, $limit = -1, $flags = 0 ) { - return new PregSplitIterator( $pattern, $subject, $limit, $flags ); - } } /** @@ -417,82 +409,3 @@ class ExplodeIterator implements Iterator { } } - -/** - * An iterator which works exactly like: - * - * foreach ( preg_split( $pattern, $s, $limit, $flags ) as $element ) { - * ... - * } - * - * Except it doesn't use huge amounts of memory when $limit is -1 - * - * The flag PREG_SPLIT_OFFSET_CAPTURE isn't supported. - */ -class PregSplitIterator implements Iterator { - // The subject string - var $pattern, $subject, $originalLimit, $flags; - - // The last extracted group of items. - var $smallArray; - - // The position on the iterator. - var $curPos; - - const MAX_LIMIT = 100; - - /** - * Construct a PregSplitIterator - */ - function __construct( $pattern, $s, $limit, $flags) { - $this->pattern = $pattern; - $this->subject = $s; - $this->originalLimit = $limit; - $this->flags = $flags; - - $this->rewind(); - } - - private function effectiveLimit() { - if ($this->originalLimit == -1) { - return self::MAX_LIMIT + 1; - } else if ($this->limit > self::MAX_LIMIT) { - $this->limit -= self::MAX_LIMIT; - return self::MAX_LIMIT + 1; - } else { - $old = $this->limit; - $this->limit = 0; - return $old; - } - } - - function rewind() { - $this->curPos = 0; - $this->limit = $this->originalLimit; - if ($this->limit == -1) $this->limit = self::MAX_LIMIT; - $this->smallArray = preg_split( $this->pattern, $this->subject, $this->effectiveLimit(), $this->flags); - } - - function current() { - return $this->smallArray[$this->curPos % self::MAX_LIMIT]; - } - - function key() { - return $this->curPos; - } - - function next() { - $this->curPos++; - if ( $this->curPos % self::MAX_LIMIT == 0 ) { - # Last item contains the rest unsplitted. - if ($this->limit > 0) { - $this->smallArray = preg_split( $this->pattern, $this->smallArray[self::MAX_LIMIT], $this->effectiveLimit(), $this->flags); - } - } - return; - } - - function valid() { - return $this->curPos % self::MAX_LIMIT < count($this->smallArray); - } -} diff --git a/includes/parser/Parser.php b/includes/parser/Parser.php index 50e15dc4f8..ee7383dc2f 100644 --- a/includes/parser/Parser.php +++ b/includes/parser/Parser.php @@ -213,7 +213,7 @@ class Parser * Must not consist of all title characters, or else it will change * the behaviour of in a link. */ - # $this->mUniqPrefix = "\x07UNIQ" . Parser::getRandomString(); + #$this->mUniqPrefix = "\x07UNIQ" . Parser::getRandomString(); # Changed to \x7f to allow XML double-parsing -- TS $this->mUniqPrefix = "\x7fUNIQ" . self::getRandomString(); @@ -338,7 +338,7 @@ class Parser '/(.) (?=\\?|:|;|!|%|\\302\\273)/' => '\\1 \\2', # french spaces, Guillemet-right '/(\\302\\253) /' => '\\1 ', - '/ (!\s*important)/' => ' \\1', # Beware of CSS magic word !important, bug #11874. + '/ (!\s*important)/' => ' \\1', #Beware of CSS magic word !important, bug #11874. ); $text = preg_replace( array_keys($fixtags), array_values($fixtags), $text ); @@ -556,7 +556,7 @@ class Parser $taglist = implode( '|', $elements ); $start = "/<($taglist)(\\s+[^>]*?|\\s*?)(\/?" . ">)|<(!--)/i"; - while ( $text !== '' ) { + while ( $text != '' ) { $p = preg_split( $start, $text, 2, PREG_SPLIT_DELIM_CAPTURE ); $stripped .= $p[0]; if( count( $p ) < 5 ) { @@ -723,11 +723,11 @@ class Parser array_push( $tr_history , false ); array_push( $tr_attributes , '' ); array_push( $has_opened_tr , false ); - } elseif ( count ( $td_history ) == 0 ) { + } else if ( count ( $td_history ) == 0 ) { // Don't do any of the following $out .= $outLine."\n"; continue; - } elseif ( substr ( $line , 0 , 2 ) === '|}' ) { + } else if ( substr ( $line , 0 , 2 ) === '|}' ) { // We are ending a table $line = '' . substr ( $line , 2 ); $last_tag = array_pop ( $last_tag_history ); @@ -745,7 +745,7 @@ class Parser } array_pop ( $tr_attributes ); $outLine = $line . str_repeat( '' , $indent_level ); - } elseif ( substr ( $line , 0 , 2 ) === '|-' ) { + } else if ( substr ( $line , 0 , 2 ) === '|-' ) { // Now we have a table row $line = preg_replace( '#^\|-+#', '', $line ); @@ -773,7 +773,7 @@ class Parser array_push ( $td_history , false ); array_push ( $last_tag_history , '' ); } - elseif ( $first_character === '|' || $first_character === '!' || substr ( $line , 0 , 2 ) === '|+' ) { + else if ( $first_character === '|' || $first_character === '!' || substr ( $line , 0 , 2 ) === '|+' ) { // This might be cell elements, td, th or captions if ( substr ( $line , 0 , 2 ) === '|+' ) { $first_character = '+'; @@ -818,9 +818,9 @@ class Parser if ( $first_character === '|' ) { $last_tag = 'td'; - } elseif ( $first_character === '!' ) { + } else if ( $first_character === '!' ) { $last_tag = 'th'; - } elseif ( $first_character === '+' ) { + } else if ( $first_character === '+' ) { $last_tag = 'caption'; } else { $last_tag = ''; @@ -835,7 +835,7 @@ class Parser // be mistaken as delimiting cell parameters if ( strpos( $cell_data[0], '[[' ) !== false ) { $cell = "{$previous}<{$last_tag}>{$cell}"; - } elseif ( count ( $cell_data ) == 1 ) + } else if ( count ( $cell_data ) == 1 ) $cell = "{$previous}<{$last_tag}>{$cell_data[0]}"; else { $attributes = $this->mStripState->unstripBoth( $cell_data[0] ); @@ -1108,59 +1108,100 @@ class Parser } /** - * Processes bolds and italics on a single line. * Helper function for doAllQuotes() */ public function doQuotes( $text ) { - # Counts the number of occurrences of bold and italics mark-ups. - self::countBoldAndItalic($text, $numbold, $numitalics); - - if ( ( $numbold == 0 ) && ( $numitalics == 0 ) ) + $arr = preg_split( "/(''+)/", $text, -1, PREG_SPLIT_DELIM_CAPTURE ); + if ( count( $arr ) == 1 ) return $text; else { + # First, do some preliminary work. This may shift some apostrophes from + # being mark-up to being text. It also counts the number of occurrences + # of bold and italics mark-ups. + $i = 0; + $numbold = 0; + $numitalics = 0; + foreach ( $arr as $r ) + { + if ( ( $i % 2 ) == 1 ) + { + # If there are ever four apostrophes, assume the first is supposed to + # be text, and the remaining three constitute mark-up for bold text. + if ( strlen( $arr[$i] ) == 4 ) + { + $arr[$i-1] .= "'"; + $arr[$i] = "'''"; + } + # If there are more than 5 apostrophes in a row, assume they're all + # text except for the last 5. + else if ( strlen( $arr[$i] ) > 5 ) + { + $arr[$i-1] .= str_repeat( "'", strlen( $arr[$i] ) - 5 ); + $arr[$i] = "'''''"; + } + # Count the number of occurrences of bold and italics mark-ups. + # We are not counting sequences of five apostrophes. + if ( strlen( $arr[$i] ) == 2 ) { $numitalics++; } + else if ( strlen( $arr[$i] ) == 3 ) { $numbold++; } + else if ( strlen( $arr[$i] ) == 5 ) { $numitalics++; $numbold++; } + } + $i++; + } + # If there is an odd number of both bold and italics, it is likely # that one of the bold ones was meant to be an apostrophe followed # by italics. Which one we cannot know for certain, but it is more # likely to be one that has a single-letter word before it. if ( ( $numbold % 2 == 1 ) && ( $numitalics % 2 == 1 ) ) { + $i = 0; + $firstsingleletterword = -1; + $firstmultiletterword = -1; + $firstspace = -1; + foreach ( $arr as $r ) + { + if ( ( $i % 2 == 1 ) and ( strlen( $r ) == 3 ) ) + { + $x1 = substr ($arr[$i-1], -1); + $x2 = substr ($arr[$i-1], -2, 1); + if ($x1 === ' ') { + if ($firstspace == -1) $firstspace = $i; + } else if ($x2 === ' ') { + if ($firstsingleletterword == -1) $firstsingleletterword = $i; + } else { + if ($firstmultiletterword == -1) $firstmultiletterword = $i; + } + } + $i++; + } - # This algorithm moves the literal quote at the - # right of a single word, at the right of a - # multiletter word or at the right of a space. - # Otherwise, it does nothing. - # - # The original if-based version can be found at - # http://svn.wikimedia.org/viewvc/mediawiki/trunk/phase3/includes/parser/Parser.php?revision=61519&view=markup - # - # Unlike the original one, here we convert the - # texty quotes to ' which shouldn't matter. - - $quoteBalancerReplacements = array( - "/(?<= [^ ])'''(?!')/"=>"'''", - "/(?<=[^ '])'''(?!')/"=>"'''", - "/(^|(?<=[^'])) '''(?!')/"=>" '''"); - - foreach( $quoteBalancerReplacements as $k => $v) { - $text = preg_replace($k, $v, $text, 1, $count); - if ($count != 0) - break; + # If there is a single-letter word, use it! + if ($firstsingleletterword > -1) + { + $arr [ $firstsingleletterword ] = "''"; + $arr [ $firstsingleletterword-1 ] .= "'"; + } + # If not, but there's a multi-letter word, use that one. + else if ($firstmultiletterword > -1) + { + $arr [ $firstmultiletterword ] = "''"; + $arr [ $firstmultiletterword-1 ] .= "'"; + } + # ... otherwise use the first one that has neither. + # (notice that it is possible for all three to be -1 if, for example, + # there is only one pentuple-apostrophe in the line) + else if ($firstspace > -1) + { + $arr [ $firstspace ] = "''"; + $arr [ $firstspace-1 ] .= "'"; } } - # Split in groups of 2, 3, 5 or 6 apostrophes. - # If there are ever four apostrophes, assume the first is supposed to - # be text, and the remaining three constitute mark-up for bold text. - # If there are more than 6 apostrophes in a row, assume they're all - # text except for the last 6. - $arr = Stringutils::preg_split( "/('{2,3}(?:''')?)(?!')/", $text, -1, PREG_SPLIT_DELIM_CAPTURE ); - - # Now let's actually convert our apostrophic mush to HTML! - $output = ''; # Processed text - $buffer = ''; # Content if $state is 'both' - $state = ''; # Flags with the order of open tags: '|b|i|bi|ib|both' + $output = ''; + $buffer = ''; + $state = ''; $i = 0; foreach ($arr as $r) { @@ -1177,58 +1218,43 @@ class Parser { if ($state === 'i') { $output .= ''; $state = ''; } - elseif ($state === 'bi') + else if ($state === 'bi') { $output .= ''; $state = 'b'; } - elseif ($state === 'ib') + else if ($state === 'ib') { $output .= ''; $state = 'b'; } - elseif ($state === 'both') + else if ($state === 'both') { $output .= ''.$buffer.''; $state = 'b'; } else # $state can be 'b' or '' { $output .= ''; $state .= 'i'; } } - elseif (strlen ($r) == 3) + else if (strlen ($r) == 3) { if ($state === 'b') { $output .= ''; $state = ''; } - elseif ($state === 'bi') + else if ($state === 'bi') { $output .= ''; $state = 'i'; } - elseif ($state === 'ib') + else if ($state === 'ib') { $output .= ''; $state = 'i'; } - elseif ($state === 'both') + else if ($state === 'both') { $output .= ''.$buffer.''; $state = 'i'; } else # $state can be 'i' or '' { $output .= ''; $state .= 'b'; } } - elseif (strlen ($r) == 5) + else if (strlen ($r) == 5) { if ($state === 'b') { $output .= ''; $state = 'i'; } - elseif ($state === 'i') + else if ($state === 'i') { $output .= ''; $state = 'b'; } - elseif ($state === 'bi') + else if ($state === 'bi') { $output .= ''; $state = ''; } - elseif ($state === 'ib') + else if ($state === 'ib') { $output .= ''; $state = ''; } - elseif ($state === 'both') + else if ($state === 'both') { $output .= ''.$buffer.''; $state = ''; } else # ($state == '') { $buffer = ''; $state = 'both'; } } - elseif (strlen ($r) == 6) - { - if ($state === 'b') - { $output .= ''; $state = 'b'; } - elseif ($state === 'i') - { $output .= '\''; $state = 'b'; } - elseif ($state === 'bi') - { $output .= '\''; $state = ''; } - elseif ($state === 'ib') - { $output .= '\''; $state = ''; } - elseif ($state === 'both') - { $output .= ''.$buffer.''; $state = 'ib'; } - else # ($state == '') - { $buffer = ''; $state = ''; } - } } $i++; } @@ -1246,57 +1272,6 @@ class Parser } } - /** - * Counts the number of bold and italic items from a line of text. - * Helper function for doQuotes() - */ - private static function countBoldAndItalic($text, &$numBold, &$numItalics) { - $numBold = 0; - $numItalics = 0; - $offset = 0; - - do { - $offset = strpos($text, "'", $offset); - if ($offset === false) - return; - - $quoteLen = strspn($text, "'", $offset); - $offset += $quoteLen; - - switch ($quoteLen) { - case 0: - case 1: - break; - - case 2: - $numItalics++; - break; - - case 3: - $numBold++; - break; - - case 4: - # If there are ever four apostrophes, assume the first is supposed to - # be text, and the remaining three constitute mark-up for bold text. - $numBold++; - $numItalics++; - break; - - case 5: - $numItalics++; - $numBold++; - break; - - case 6: - default: - # If there are more than 6 apostrophes in a row, assume they're all - # text except for the last 6. - $numBold+=2; - } - } while (true); - } - /** * Replace external links (REL) * @@ -1538,9 +1513,9 @@ class Parser $sk = $this->mOptions->getSkin(); $holders = new LinkHolderArray( $this ); - # split the entire text string on occurences of [[ + #split the entire text string on occurences of [[ $a = StringUtils::explode( '[[', ' ' . $s ); - # get the first element (all text up to first [[), and remove the space we added + #get the first element (all text up to first [[), and remove the space we added $s = $a->current(); $a->next(); $line = $a->current(); # Workaround for broken ArrayIterator::next() that returns "void" @@ -1685,10 +1660,10 @@ class Parser if ( $might_be_img ) { # if this is actually an invalid link wfProfileIn( __METHOD__."-might_be_img" ); - if ( $ns == NS_FILE && $noforce ) { # but might be an image + if ( $ns == NS_FILE && $noforce ) { #but might be an image $found = false; while ( true ) { - # look at the next 'line' to see if we can close it there + #look at the next 'line' to see if we can close it there $a->next(); $next_line = $a->current(); if ( $next_line === false || $next_line === null ) { @@ -1702,24 +1677,24 @@ class Parser $trail = $m[2]; break; } elseif ( count( $m ) == 2 ) { - # if there's exactly one ]] that's fine, we'll keep looking + #if there's exactly one ]] that's fine, we'll keep looking $text .= "[[{$m[0]}]]{$m[1]}"; } else { - # if $next_line is invalid too, we need look no further + #if $next_line is invalid too, we need look no further $text .= '[[' . $next_line; break; } } if ( !$found ) { # we couldn't find the end of this imageLink, so output it raw - # but don't ignore what might be perfectly normal links in the text we've examined + #but don't ignore what might be perfectly normal links in the text we've examined $holders->merge( $this->replaceInternalLinks2( $text ) ); $s .= "{$prefix}[[$link|$text"; # note: no $trail, because without an end, there *is* no trail wfProfileOut( __METHOD__."-might_be_img" ); continue; } - } else { # it's not an image, so output it raw + } else { #it's not an image, so output it raw $s .= "{$prefix}[[$link|$text"; # note: no $trail, because without an end, there *is* no trail wfProfileOut( __METHOD__."-might_be_img" ); @@ -1796,7 +1771,7 @@ class Parser } # Self-link checking - if( $nt->getFragment() === '' && $ns !== NS_SPECIAL ) { + if( $nt->getFragment() === '' && $ns != NS_SPECIAL ) { if( in_array( $nt->getPrefixedText(), $selflink, true ) ) { $s .= $prefix . $sk->makeSelfLinkObj( $nt, $text, '', $trail ); continue; @@ -1916,7 +1891,7 @@ class Parser */ /* private */ function closeParagraph() { $result = ''; - if ( $this->mLastSection !== '' ) { + if ( $this->mLastSection != '' ) { $result = 'mLastSection . ">\n"; } $this->mInPre = false; @@ -1932,7 +1907,7 @@ class Parser if ( $fl < $shorter ) { $shorter = $fl; } for ( $i = 0; $i < $shorter; ++$i ) { - if ( $st1{$i} !== $st2{$i} ) { break; } + if ( $st1{$i} != $st2{$i} ) { break; } } return $i; } @@ -2105,7 +2080,7 @@ class Parser 'mUniqPrefix.'-pre|<\\/li|<\\/ul|<\\/ol|<\\/?center)/iS', $t ); if ( $openmatch or $closematch ) { $paragraphStack = false; - # TODO bug 5718: paragraph closed + # TODO bug 5718: paragraph closed $output .= $this->closeParagraph(); if ( $preOpenMatch and !$preCloseMatch ) { $this->mInPre = true; @@ -2115,8 +2090,8 @@ class Parser } else { $inBlockElem = true; } - } elseif ( !$inBlockElem && !$this->mInPre ) { - if ( ' ' == substr( $t, 0, 1 ) and ( $this->mLastSection === 'pre' or trim($t) !== '' ) ) { + } else if ( !$inBlockElem && !$this->mInPre ) { + if ( ' ' == substr( $t, 0, 1 ) and ( $this->mLastSection === 'pre' or trim($t) != '' ) ) { // pre if ($this->mLastSection !== 'pre') { $paragraphStack = false; @@ -2145,7 +2120,7 @@ class Parser $output .= $paragraphStack; $paragraphStack = false; $this->mLastSection = 'p'; - } elseif ($this->mLastSection !== 'p') { + } else if ($this->mLastSection !== 'p') { $output .= $this->closeParagraph().'

'; $this->mLastSection = 'p'; } @@ -2166,7 +2141,7 @@ class Parser $output .= $this->closeList( $prefix2[$prefixLength-1] ); --$prefixLength; } - if ( $this->mLastSection !== '' ) { + if ( $this->mLastSection != '' ) { $output .= 'mLastSection . '>'; $this->mLastSection = ''; } @@ -2972,7 +2947,7 @@ class Parser $isHTML = true; $this->disableCache(); } - } elseif ( $wgNonincludableNamespaces && in_array( $title->getNamespace(), $wgNonincludableNamespaces ) ) { + } else if ( $wgNonincludableNamespaces && in_array( $title->getNamespace(), $wgNonincludableNamespaces ) ) { $found = false; //access denied wfDebug( __METHOD__.": template inclusion denied for " . $title->getPrefixedDBkey() ); } else { @@ -3585,7 +3560,7 @@ class Parser if (preg_match("/^$markerRegex/", $headline, $markerMatches)) { $serial = $markerMatches[1]; list( $titleText, $sectionIndex ) = $this->mHeadings[$serial]; - $isTemplate = ($titleText !== $baseTitleText); + $isTemplate = ($titleText != $baseTitleText); $headline = preg_replace("/^$markerRegex/", "", $headline); } @@ -3701,7 +3676,7 @@ class Parser if ( $legacyHeadline == $safeHeadline ) { # No reason to have both (in fact, we can't) $legacyHeadline = false; - } elseif ( $legacyHeadline !== Sanitizer::escapeId( + } elseif ( $legacyHeadline != Sanitizer::escapeId( $legacyHeadline, 'xml' ) ) { # The legacy id is invalid XML. We used to allow this, but # there's no reason to do so anymore. Backward @@ -3875,8 +3850,8 @@ class Parser else continue; } - if ( $s['index'] !== $section || - $s['fromtitle'] !== $titletext ) { + if ( $s['index'] != $section || + $s['fromtitle'] != $titletext ) { self::incrementNumbering( $numbering, $s['toclevel'], $lastLevel ); @@ -3927,7 +3902,7 @@ class Parser private static function incrementNumbering( &$number, $level, $lastLevel ) { if ( $level > $lastLevel ) $number[$level - 1] = 1; - elseif ( $level < $lastLevel ) { + else if ( $level < $lastLevel ) { foreach ( $number as $key => $unused ) if ( $key >= $level ) unset( $number[$key] ); @@ -4037,7 +4012,7 @@ class Parser $m = array(); if ( preg_match( "/^($nc+:|)$tc+?( \\($tc+\\))$/", $t, $m ) ) { $text = preg_replace( $p2, "[[$m[1]\\1$m[2]|\\1]]", $text ); - } elseif ( preg_match( "/^($nc+:|)$tc+?(, $tc+|)$/", $t, $m ) && "$m[1]$m[2]" !== '' ) { + } elseif ( preg_match( "/^($nc+:|)$tc+?(, $tc+|)$/", $t, $m ) && "$m[1]$m[2]" != '' ) { $text = preg_replace( $p2, "[[$m[1]\\1$m[2]|\\1]]", $text ); } else { # if there's no context, don't bother duplicating the title @@ -4876,7 +4851,7 @@ class Parser if ( $node->getName() === 'h' ) { $bits = $node->splitHeading(); $curLevel = $bits['level']; - if ( $bits['i'] !== $sectionIndex && $curLevel <= $targetLevel ) { + if ( $bits['i'] != $sectionIndex && $curLevel <= $targetLevel ) { break; } } @@ -4892,7 +4867,7 @@ class Parser // Add two newlines on -- trailing whitespace in $newText is conventionally // stripped by the editor, so we need both newlines to restore the paragraph gap // Only add trailing whitespace if there is newText - if($newText !== "") { + if($newText != "") { $outText .= $newText . "\n\n"; } diff --git a/maintenance/parserTests.txt b/maintenance/parserTests.txt index fedbb85c1c..6695973adf 100644 --- a/maintenance/parserTests.txt +++ b/maintenance/parserTests.txt @@ -116,7 +116,7 @@ Italics and bold

  • plainbold-italicboldplain
  • plainitalicbold-italicplain
  • plainboldbold-italicplain -
  • plain l'italicplain +
  • plain l'italicplain
  • plain l'bold plain
    • @@ -5253,17 +5253,19 @@ New wiki paragraph

    !! end -# This was the original html, but it has also been -#

    'bold'boldbolditalics +# Original result was this: +#

    boldboldbolditalics #

    -# See bug 18765. +# While that might be marginally more intuitive, maybe, the six-apostrophe +# construct is clearly pathological and the result stated here (which is what +# the parser actually does) is about as reasonable as anything. !!test Mixing markup for italics and bold !! options !! input '''bold''''''bold''bolditalics''''' !! result -

    boldboldbolditalics +

    'bold'boldbolditalics

    !! end @@ -6415,7 +6417,7 @@ Another italics / bold test !! input ''' ''x' !! result -
    ' x'
+
    ' x'
 
    
 !!end
 
@@ -7558,82 +7560,6 @@ title=[[Main Page]]
 #section
 !! end
 
-!! test
-Bold/italic markup handled differently depending on leading whitespace (bug 18765)
-!!input
-'''Look at ''this edit'''s complicated bold/italic markup!'''
-
-'''Look at ''this edit'''s complicated bold/italic markup!'''
-
- '''Look at ''this edit'''s complicated bold/italic markup!'''
-
- '''Look at ''this edit'''s complicated bold/italic markup!'''
-
- '''Look at ''this edit'''s complicated bold/italic markup!'''
-
-{|
-| '''Look at ''this edit'''s complicated bold/italic markup!'''
-|}
-
-'''This was Italic'' this was plain''' and this was bold'''
-but '''This is bold'' this is bold italic''' and this is bold'''
-
-
-!! result
-
    Look at this edit's complicated bold/italic markup!
-
    Look at this edit's complicated bold/italic markup!
-
     Look at this edit's complicated bold/italic markup!
-
     Look at this edit's complicated bold/italic markup!
-
    
-
    Look at this edit's complicated bold/italic markup!
-
    
-
-
-
     Look at this edit's complicated bold/italic markup!
-
    
-
    This was Italic this was plain' and this was bold
-but This is bold this is bold italic' and this is bold
-

    
-
    
-!! end
-
-!! test
-Six quotes
-!!input
-''Italic''''''Bold
-
-'''Bold''BoldItalic''''''Normal
-
-''Italic'''BoldItalic''''''Normal'''''
-
-'''''BoldItalic''''''MoreBoldItalic''
-
-''''''Normal
-!!result
-
    Italic'Bold
-
    BoldBoldItalic'Normal
-
    ItalicBoldItalic'Normal
-
    BoldItalicMoreBoldItalic
-
    Normal
-
    
-!!end
-
-
-!! test
-Too many quotes
-!!input
-I '''like'''''quotes''''''''''' 
-!! result
-
    I likequotes'''''' 
-
    
-!! end
-
-
 Note: some elements used in these Microdata examples don't work, like 
 and