| Index: trunk/phase3/includes/Parser.php |
| — | — | @@ -822,9 +822,6 @@ |
| 823 | 823 | $fname = 'Parser::internalParse'; |
| 824 | 824 | wfProfileIn( $fname ); |
| 825 | 825 | |
| 826 | | - global $fixLbug ; |
| 827 | | - if ( $fixLbug ) $text = preg_replace ( '/(l|L)\'/' , '\\1'' , $text ) ; |
| 828 | | - |
| 829 | 826 | $text = $this->removeHTMLtags( $text ); |
| 830 | 827 | $text = $this->replaceVariables( $text, $args ); |
| 831 | 828 | |
| — | — | @@ -836,12 +833,10 @@ |
| 837 | 834 | $text = $wgDateFormatter->reformat( $this->mOptions->getDateFormat(), $text ); |
| 838 | 835 | } |
| 839 | 836 | $text = $this->doAllQuotes( $text ); |
| 840 | | - // $text = $this->doExponent( $text ); |
| 841 | 837 | $text = $this->replaceExternalLinks( $text ); |
| 842 | 838 | $text = $this->doMagicLinks( $text ); |
| 843 | 839 | $text = $this->replaceInternalLinks ( $text ); |
| 844 | 840 | $text = $this->replaceInternalLinks ( $text ); |
| 845 | | - //$text = $this->doTokenizedParser ( $text ); |
| 846 | 841 | $text = $this->doTableStuff( $text ); |
| 847 | 842 | $text = $this->formatHeadings( $text, $isMain ); |
| 848 | 843 | $sk =& $this->mOptions->getSkin(); |
| — | — | @@ -892,57 +887,167 @@ |
| 893 | 888 | $outtext = ''; |
| 894 | 889 | $lines = explode( "\n", $text ); |
| 895 | 890 | foreach ( $lines as $line ) { |
| 896 | | - $outtext .= $this->doQuotes ( '', $line, '' ) . "\n"; |
| | 891 | + $outtext .= $this->doQuotes ( $line ) . "\n"; |
| 897 | 892 | } |
| 898 | 893 | $outtext = substr($outtext, 0,-1); |
| 899 | 894 | wfProfileOut( $fname ); |
| 900 | 895 | return $outtext; |
| 901 | 896 | } |
| 902 | 897 | |
| 903 | | - /* private */ function doQuotes( $pre, $text, $mode ) { |
| 904 | | - if ( preg_match( "/^(.*)''(.*)$/sU", $text, $m ) ) { |
| 905 | | - $m1_strong = ($m[1] == "") ? "" : "<strong>{$m[1]}</strong>"; |
| 906 | | - $m1_em = ($m[1] == "") ? "" : "<em>{$m[1]}</em>"; |
| 907 | | - if ( substr ($m[2], 0, 1) == '\'' ) { |
| 908 | | - $m[2] = substr ($m[2], 1); |
| 909 | | - if ($mode == 'em') { |
| 910 | | - return $this->doQuotes ( $m[1], $m[2], ($m[1] == '') ? 'both' : 'emstrong' ); |
| 911 | | - } else if ($mode == 'strong') { |
| 912 | | - return $m1_strong . $this->doQuotes ( '', $m[2], '' ); |
| 913 | | - } else if (($mode == 'emstrong') || ($mode == 'both')) { |
| 914 | | - return $this->doQuotes ( '', $pre.$m1_strong.$m[2], 'em' ); |
| 915 | | - } else if ($mode == 'strongem') { |
| 916 | | - return "<strong>{$pre}{$m1_em}</strong>" . $this->doQuotes ( '', $m[2], 'em' ); |
| 917 | | - } else { |
| 918 | | - return $m[1] . $this->doQuotes ( '', $m[2], 'strong' ); |
| | 898 | + /* private */ function doQuotes( $text ) { |
| | 899 | + $arr = preg_split ("/(''+)/", $text, -1, PREG_SPLIT_DELIM_CAPTURE); |
| | 900 | + if (count ($arr) == 1) |
| | 901 | + return $text; |
| | 902 | + else |
| | 903 | + { |
| | 904 | + $i = 0; |
| | 905 | + foreach ($arr as $r) |
| | 906 | + { |
| | 907 | + if (($i % 2) == 1) |
| | 908 | + { |
| | 909 | + # If there are ever four apostrophes, assume the first is supposed to |
| | 910 | + # be text, and the remaining three constitute mark-up for bold text. |
| | 911 | + if (strlen ($arr[$i]) == 4) |
| | 912 | + { |
| | 913 | + $arr[$i-1] .= "'"; |
| | 914 | + $arr[$i] = "'''"; |
| | 915 | + } |
| | 916 | + # If there are more than 5 apostrophes in a row, assume they're all |
| | 917 | + # text except for the last 5. |
| | 918 | + else if (strlen ($arr[$i]) > 5) |
| | 919 | + { |
| | 920 | + $arr[$i-1] .= str_repeat ("'", strlen ($arr[$i]) - 5); |
| | 921 | + $arr[$i] = "'''''"; |
| | 922 | + } |
| | 923 | + |
| 919 | 924 | } |
| 920 | | - } else { |
| 921 | | - if ($mode == 'strong') { |
| 922 | | - return $this->doQuotes ( $m[1], $m[2], ($m[1] == '') ? 'both' : 'strongem' ); |
| 923 | | - } else if ($mode == 'em') { |
| 924 | | - return $m1_em . $this->doQuotes ( '', $m[2], '' ); |
| 925 | | - } else if ($mode == 'emstrong') { |
| 926 | | - return "<em>{$pre}{$m1_strong}</em>" . $this->doQuotes ( '', $m[2], 'strong' ); |
| 927 | | - } else if (($mode == 'strongem') || ($mode == 'both')) { |
| 928 | | - return $this->doQuotes ( '', $pre.$m1_em.$m[2], 'strong' ); |
| 929 | | - } else { |
| 930 | | - return $m[1] . $this->doQuotes ( '', $m[2], 'em' ); |
| | 925 | + $i++; |
| | 926 | + } |
| | 927 | + |
| | 928 | + # Now see if there's an odd or even number of "bold" and "italic" |
| | 929 | + # mark-up. There should normally be an even number of both. |
| | 930 | + $i = 0; |
| | 931 | + $numbold = 0; |
| | 932 | + $numitalics = 0; |
| | 933 | + foreach ($arr as $r) |
| | 934 | + { |
| | 935 | + if (($i % 2) == 1) |
| | 936 | + { |
| | 937 | + if (strlen ($r) == 2) $numitalics++; else |
| | 938 | + if (strlen ($r) == 3) $numbold++; else |
| | 939 | + if (strlen ($r) == 5) { $numitalics++; $numbold++; } |
| 931 | 940 | } |
| | 941 | + $i++; |
| 932 | 942 | } |
| 933 | | - } else { |
| 934 | | - $text_strong = ($text == '') ? '' : "<strong>{$text}</strong>"; |
| 935 | | - $text_em = ($text == '') ? '' : "<em>{$text}</em>"; |
| 936 | | - if ($mode == '') { |
| 937 | | - return $pre . $text; |
| 938 | | - } else if ($mode == 'em') { |
| 939 | | - return $pre . $text_em; |
| 940 | | - } else if ($mode == 'strong') { |
| 941 | | - return $pre . $text_strong; |
| 942 | | - } else if ($mode == 'strongem') { |
| 943 | | - return (($pre == '') && ($text == '')) ? '' : "<strong>{$pre}{$text_em}</strong>"; |
| 944 | | - } else { |
| 945 | | - return (($pre == '') && ($text == '')) ? '' : "<em>{$pre}{$text_strong}</em>"; |
| | 943 | + |
| | 944 | + # If there is an odd number of both bold and italics, it is likely |
| | 945 | + # that one of the bold ones was meant to be an apostrophe followed |
| | 946 | + # by italics. Which one we cannot know for certain, but it is more |
| | 947 | + # likely to be one that has a single-letter word before it. |
| | 948 | + if (($numbold % 2 == 1) && ($numitalics % 2 == 1)) |
| | 949 | + { |
| | 950 | + $i = 0; |
| | 951 | + $firstsingleletterword = -1; |
| | 952 | + $firstmultiletterword = -1; |
| | 953 | + $firstspace = -1; |
| | 954 | + foreach ($arr as $r) |
| | 955 | + { |
| | 956 | + if (($i % 2 == 1) and (strlen ($r) == 3)) |
| | 957 | + { |
| | 958 | + $x1 = substr ($arr[$i-1], -1); |
| | 959 | + $x2 = substr ($arr[$i-1], -2, 1); |
| | 960 | + if ($x1 == " ") { |
| | 961 | + if ($firstspace == -1) $firstspace = $i; |
| | 962 | + } else if ($x2 == " ") { |
| | 963 | + if ($firstsingleletterword == -1) $firstsingleletterword = $i; |
| | 964 | + } else { |
| | 965 | + if ($firstmultiletterword == -1) $firstmultiletterword = $i; |
| | 966 | + } |
| | 967 | + } |
| | 968 | + $i++; |
| | 969 | + } |
| | 970 | + |
| | 971 | + # If there is a single-letter word, use it! |
| | 972 | + if ($firstsingleletterword > -1) |
| | 973 | + { |
| | 974 | + $arr [ $firstsingleletterword ] = "''"; |
| | 975 | + $arr [ $firstsingleletterword-1 ] .= "'"; |
| | 976 | + } |
| | 977 | + # If not, but there's a multi-letter word, use that one. |
| | 978 | + else if ($firstmultiletterword > -1) |
| | 979 | + { |
| | 980 | + $arr [ $firstmultiletterword ] = "''"; |
| | 981 | + $arr [ $firstmultiletterword-1 ] .= "'"; |
| | 982 | + } |
| | 983 | + # ... otherwise use the first one that has neither. |
| | 984 | + else |
| | 985 | + { |
| | 986 | + $arr [ $firstspace ] = "''"; |
| | 987 | + $arr [ $firstspace-1 ] .= "'"; |
| | 988 | + } |
| 946 | 989 | } |
| | 990 | + |
| | 991 | + # Now let's actually convert our apostrophic mush to HTML! |
| | 992 | + $output = ''; |
| | 993 | + $buffer = ''; |
| | 994 | + $state = ''; |
| | 995 | + $i = 0; |
| | 996 | + foreach ($arr as $r) |
| | 997 | + { |
| | 998 | + if (($i % 2) == 0) |
| | 999 | + { |
| | 1000 | + if ($state == 'both') |
| | 1001 | + $buffer .= $r; |
| | 1002 | + else |
| | 1003 | + $output .= $r; |
| | 1004 | + } |
| | 1005 | + else |
| | 1006 | + { |
| | 1007 | + if (strlen ($r) == 2) |
| | 1008 | + { |
| | 1009 | + if ($state == 'em') |
| | 1010 | + { $output .= "</em>"; $state = ''; } |
| | 1011 | + else if ($state == 'strongem') |
| | 1012 | + { $output .= "</em>"; $state = 'strong'; } |
| | 1013 | + else if ($state == 'emstrong') |
| | 1014 | + { $output .= "</strong></em><strong>"; $state = 'strong'; } |
| | 1015 | + else if ($state == 'both') |
| | 1016 | + { $output .= "<strong><em>{$buffer}</em>"; $state = 'strong'; } |
| | 1017 | + else # $state can be 'strong' or '' |
| | 1018 | + { $output .= "<em>"; $state .= 'em'; } |
| | 1019 | + } |
| | 1020 | + else if (strlen ($r) == 3) |
| | 1021 | + { |
| | 1022 | + if ($state == 'strong') |
| | 1023 | + { $output .= "</strong>"; $state = ''; } |
| | 1024 | + else if ($state == 'strongem') |
| | 1025 | + { $output .= "</em></strong><em>"; $state = 'em'; } |
| | 1026 | + else if ($state == 'emstrong') |
| | 1027 | + { $output .= "</strong>"; $state = 'em'; } |
| | 1028 | + else if ($state == 'both') |
| | 1029 | + { $output .= "<em><strong>{$buffer}</strong>"; $state = 'em'; } |
| | 1030 | + else # $state can be 'em' or '' |
| | 1031 | + { $output .= "<strong>"; $state .= 'strong'; } |
| | 1032 | + } |
| | 1033 | + else if (strlen ($r) == 5) |
| | 1034 | + { |
| | 1035 | + if ($state == 'strong') |
| | 1036 | + { $output .= "</strong><em>"; $state = 'em'; } |
| | 1037 | + else if ($state == 'em') |
| | 1038 | + { $output .= "</em><strong>"; $state = 'strong'; } |
| | 1039 | + else if ($state == 'strongem') |
| | 1040 | + { $output .= "</em></strong>"; $state = ''; } |
| | 1041 | + else if ($state == 'emstrong') |
| | 1042 | + { $output .= "</strong></em>"; $state = ''; } |
| | 1043 | + else if ($state == 'both') |
| | 1044 | + { $output .= "<em><strong>{$buffer}</strong></em>"; $state = ''; } |
| | 1045 | + else # ($state == '') |
| | 1046 | + { $buffer = ''; $state = 'both'; } |
| | 1047 | + } |
| | 1048 | + } |
| | 1049 | + $i++; |
| | 1050 | + } |
| | 1051 | + return $output; |
| 947 | 1052 | } |
| 948 | 1053 | } |
| 949 | 1054 | |