Cscott has uploaded a new change for review. https://gerrit.wikimedia.org/r/79822
Change subject: Optimize Parser::doQuotes(). ...................................................................... Optimize Parser::doQuotes(). Performance improvements to doQuotes(), since it is a hot function. Co-authored-by: Tyler Anthony Romeo <[email protected]> Change-Id: If78d4372a2acd78d58b020385da400978716cbf5 --- M includes/parser/Parser.php 1 file changed, 20 insertions(+), 12 deletions(-) git pull ssh://gerrit.wikimedia.org:29418/mediawiki/core refs/changes/22/79822/1 diff --git a/includes/parser/Parser.php b/includes/parser/Parser.php index 2943afa..4561351 100644 --- a/includes/parser/Parser.php +++ b/includes/parser/Parser.php @@ -1414,7 +1414,8 @@ */ public function doQuotes( $text ) { $arr = preg_split( "/(''+)/", $text, -1, PREG_SPLIT_DELIM_CAPTURE ); - if ( count( $arr ) == 1 ) { + $countarr = count( $arr ); + if ( $countarr == 1 ) { return $text; } @@ -1423,26 +1424,29 @@ // of bold and italics mark-ups. $numbold = 0; $numitalics = 0; - for ( $i = 1; $i < count( $arr ); $i+=2 ) { + for ( $i = 1; $i < $countarr; $i+=2 ) { + $thislen = strlen( $arr[$i] ); // If there are ever four apostrophes, assume the first is supposed to // be text, and the remaining three constitute mark-up for bold text. // (bug 13227: ''''foo'''' turns into ''' 'foo ''' ') - if ( strlen( $arr[$i] ) == 4 ) { + if ( $thislen == 4 ) { $arr[$i - 1] .= "'"; $arr[$i] = "'''"; - } elseif ( strlen( $arr[$i] ) > 5 ) { + $thislen = 3; + } elseif ( $thislen > 5 ) { // If there are more than 5 apostrophes in a row, assume they're all // text except for the last 5. // (bug 13227: ''''''foo'''''' turns into ''''' 'foo ''''' ') - $arr[$i - 1] .= str_repeat( "'", strlen( $arr[$i] ) - 5 ); + $arr[$i - 1] .= str_repeat( "'", $thislen - 5 ); $arr[$i] = "'''''"; + $thislen = 5; } // Count the number of occurrences of bold and italics mark-ups. - if ( strlen( $arr[$i] ) == 2 ) { + if ( $thislen == 2 ) { $numitalics++; - } elseif ( strlen( $arr[$i] ) == 3 ) { + } elseif ( $thislen == 3 ) { $numbold++; - } elseif ( strlen( $arr[$i] ) == 5 ) { + } elseif ( $thislen == 5 ) { $numitalics++; $numbold++; } @@ -1456,7 +1460,7 @@ $firstsingleletterword = -1; $firstmultiletterword = -1; $firstspace = -1; - for ( $i = 1; $i < count( $arr ); $i+=2 ) { + for ( $i = 1; $i < $countarr; $i+=2 ) { if ( strlen( $arr[$i] ) == 3 ) { $x1 = substr( $arr[$i - 1], -1 ); $x2 = substr( $arr[$i - 1], -2, 1 ); @@ -1467,6 +1471,9 @@ } elseif ( $x2 === ' ' ) { if ( $firstsingleletterword == -1 ) { $firstsingleletterword = $i; + // if $firstsingleletterword is set, we don't + // look at the other options, so we can bail early. + break; } } else { if ( $firstmultiletterword == -1 ) { @@ -1506,7 +1513,8 @@ $output .= $r; } } else { - if ( strlen( $r ) == 2 ) { + $thislen = strlen( $r ); + if ( $thislen == 2 ) { if ( $state === 'i' ) { $output .= '</i>'; $state = ''; @@ -1523,7 +1531,7 @@ $output .= '<i>'; $state .= 'i'; } - } elseif ( strlen( $r ) == 3 ) { + } elseif ( $thislen == 3 ) { if ( $state === 'b' ) { $output .= '</b>'; $state = ''; @@ -1540,7 +1548,7 @@ $output .= '<b>'; $state .= 'b'; } - } elseif ( strlen( $r ) == 5 ) { + } elseif ( $thislen == 5 ) { if ( $state === 'b' ) { $output .= '</b><i>'; $state = 'i'; -- To view, visit https://gerrit.wikimedia.org/r/79822 To unsubscribe, visit https://gerrit.wikimedia.org/r/settings Gerrit-MessageType: newchange Gerrit-Change-Id: If78d4372a2acd78d58b020385da400978716cbf5 Gerrit-PatchSet: 1 Gerrit-Project: mediawiki/core Gerrit-Branch: master Gerrit-Owner: Cscott <[email protected]> _______________________________________________ MediaWiki-commits mailing list [email protected] https://lists.wikimedia.org/mailman/listinfo/mediawiki-commits
