From f3f0262c480d7e509b008d37c90aed884532bba8 Mon Sep 17 00:00:00 2001 From: andi Date: Wed, 12 Jan 2005 21:24:54 +0100 Subject: Initial revision. darcs-hash:20050112202454-9977f-60936f24fe2092a30223627e0683de2df61d0c4a.gz --- inc/parser.php | 763 +++++++++++++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 763 insertions(+) create mode 100644 inc/parser.php (limited to 'inc/parser.php') diff --git a/inc/parser.php b/inc/parser.php new file mode 100644 index 000000000..5b8a1323e --- /dev/null +++ b/inc/parser.php @@ -0,0 +1,763 @@ +(.*?)#se","preformat('\\1','nowiki')"); + firstpass($table,$text,"#%%(.*?)%%#se","preformat('\\1','nowiki')"); + firstpass($table,$text,"#(.*?)#se","preformat('\\3','code','\\2')"); + firstpass($table,$text,"#(.*?)#se","preformat('\\1','file')"); + + // html and php includes + firstpass($table,$text,"#(.*?)#se","preformat('\\1','html')"); + firstpass($table,$text,"#(.*?)#se","preformat('\\1','php')"); + + // codeblocks + firstpass($table,$text,"/(\n( {2,}|\t)[^\*\-\n ][^\n]+)(\n( {2,}|\t)[^\n]*)*/se","preformat('\\0','block')","\n"); + + //check if toc is wanted + if(!isset($parser['toc'])){ + if(strpos($text,'~~NOTOC~~')!== false){ + $text = str_replace('~~NOTOC~~','',$text); + $parser['toc'] = false; + }else{ + $parser['toc'] = true; + } + } + + //check if this file may be cached + if(!isset($parser['cache'])){ + if(strpos($text,'~~NOCACHE~~')!=false){ + $text = str_replace('~~NOCACHE~~','',$text); + $parser['cache'] = false; + }else{ + $parser['cache'] = true; + } + } + + //headlines + format_headlines($table,$hltable,$text); + + //links + firstpass($table,$text,"#\[\[([^\]]+?)\]\]#ie","linkformat('\\1')"); + + //media + firstpass($table,$text,"/\{\{([^\}]+)\}\}/se","mediaformat('\\1')"); + + //match full URLs (adapted from Perl cookbook) + firstpass($table,$text,"#(\b)($urls://[$any]+?)([$punc]*[^$any])#ie","linkformat('\\2')",'\1','\4'); + + //short www URLs + firstpass($table,$text,"#(\b)(www\.[$host]+?\.[$host]+?[$any]+?)([$punc]*[^$any])#ie","linkformat('http://\\2|\\2')",'\1','\3'); + + //windows shares + firstpass($table,$text,"#([$gunk$punc\s])(\\\\\\\\[$host]+?\\\\[$any]+?)([$punc]*[^$any])#ie","linkformat('\\2')",'\1','\3'); + + //short ftp URLs + firstpass($table,$text,"#(\b)(ftp\.[$host]+?\.[$host]+?[$any]+?)([$punc]*[^$any])#ie","linkformat('ftp://\\2')",'\1','\3'); + + // email@domain.tld + firstpass($table,$text,"#<([\w0-9\-_.]+?)@([\w\-]+\.([\w\-\.]+\.)*[\w]+)>#ie", "linkformat('\\1@\\2')"); + + //CamelCase if wanted + if($conf['camelcase']){ + firstpass($table,$text,"#(\b)([A-Z]+[a-z]+[A-Z][A-Za-z]*)(\b)#se","linkformat('\\2')",'\1','\3'); + } + + $text = htmlspecialchars($text); + + //smileys + smileys($table,$text); + + //acronyms + acronyms($table,$text); + + /* second pass for simple formating */ + $text = simpleformat($text); + + /* third pass - insert the matches from 1st pass */ + reset($table); + while (list($key, $val) = each($table)) { + $text = str_replace($key,$val,$text); + } + + /* remove empty paragraphs */ + $text = preg_replace('"

\n*

"','',$text); + + /* remove padding */ + $text = trim($text); + return $text; +} + +/** + * This preparses the text by walking it line by line. This + * is the only place where linenumbers are still available (needed + * for section edit. Some precautions have to be taken to not change + * any noparse block. + */ +function preparse($text,&$table,&$hltable){ + $lines = split("\n",$text); + + //prepare a tokens for paragraphs + $po = mkToken(); + $table[$po] = "

"; + $pc = mkToken(); + $table[$pc] = "

"; + + for ($l=0; $l(.*?)#","",$line); + $line = preg_replace("#%%(.*?)%%#","",$line); + $line = preg_replace("#(.*?)#","",$line); + $line = preg_replace("#(.*?)#","",$line); + $line = preg_replace("#(.*?)#","",$line); + $line = preg_replace("#(.*?)#","",$line); + //check for start of multiline noparse areas + if(preg_match('#^.*?<(nowiki|code|php|html|file)( (\w+))?>#',$line,$matches)){ + list($noparse) = split(" ",$matches[1]); //remove options + $noparse = ''; + continue; + }elseif(preg_match('#^.*?%%#',$line)){ + $noparse = '%%'; + continue; + } + } + + //handle headlines + if(preg_match('/^(\s)*(==+)(.+?)(==+)(\s*)$/',$lines[$l],$matches)){ + //get token + $tk = tokenize_headline($hltable,$matches[2],$matches[3],$l); + //replace line with token + $lines[$l] = $tk; + } + + //handle paragraphs + if(empty($lines[$l])){ + $lines[$l] = "$pc\n$po"; + } + } + + //reassemble full text + $text = join("\n",$lines); + //open first and close last paragraph + $text = "$po\n$text\n$pc"; + + return $text; +} + +/** + * This function adds some information about the given headline + * to a lookuptable to be processed later. Returns a unique token + * that idetifies the headline later + */ +function tokenize_headline(&$hltable,$pre,$hline,$lno){ + switch (strlen($pre)){ + case 2: + $lvl = 5; + break; + case 3: + $lvl = 4; + break; + case 4: + $lvl = 3; + break; + case 5: + $lvl = 2; + break; + default: + $lvl = 1; + break; + } + $token = mkToken(); + $hltable[] = array( 'name' => htmlspecialchars(trim($hline)), + 'level' => $lvl, + 'line' => $lno, + 'token' => $token ); + return $token; +} + +function format_headlines(&$table,&$hltable,&$text){ + global $parser; + global $conf; + global $lang; + global $ID; + + // walk the headline table prepared in preparsing + $last = 0; + $cnt = 0; + $hashs = array(); + foreach($hltable as $hl){ + $cnt++; + + //make unique headlinehash + $hash = cleanID($hl['name']); + $i=2; + while(in_array($hash,$hashs)) + $hash = cleanID($hl['name']).$i++; + $hashs[] = $hash; + + // build headline + $headline = "

\n"; //close paragraph + if($cnt - 1) $headline .= ''; //no close on first HL + $headline .= ''; + $headline .= ''; + $headline .= $hl['name']; + $headline .= ''; + $headline .= '
'; + $headline .= "\n

"; //open new paragraph + + //remember for autoTOC + if($hl['level'] <= $conf['maxtoclevel']){ + $content[] = array('id' => $hash, + 'name' => $hl['name'], + 'level' => $hl['level']); + } + + //add link for section edit for HLs 1, and 3 + if( ($hl['level'] <= $conf['maxseclevel']) && + ($hl['line'] - $last > 1)){ + $secedit = ''; + $headline = $secedit.$headline; + $last = $hl['line']; + } + + //put headline into firstpasstable + $table[$hl['token']] = $headline; + } + + //add link for editing the last section + if($last){ + $secedit = ''; + $token = mktoken(); + $text .= $token; + $table[$token] = $secedit; + } + + //close last div + if ($cnt){ + $token = mktoken(); + $text .= $token; + $table[$token] = '

'; + } + + //prepend toc + if ($parser['toc'] && count($content) > 2){ + $token = mktoken(); + $text = $token.$text; + $table[$token] = html_toc($content); + } +} + +/** + * Formats various link types using the functions from format.php + */ +function linkformat($match){ + global $conf; + //unescape + $match = str_replace('\\"','"',$match); + + //prepare variables for the formaters + $link = array(); + list($link['url'],$link['name']) = split('\|',$match,2); + $link['url'] = trim($link['url']); + $link['name'] = trim($link['name']); + $link['class'] = ''; + $link['target'] = ''; + $link['style'] = ''; + $link['pre'] = ''; + $link['suf'] = ''; + $link['more'] = ''; + + //save real name for image check + $realname = $link['name']; + + /* put it into the right formater */ + if(strpos($link['url'],'>')){ + // InterWiki + $link = format_link_interwiki($link); + }elseif(preg_match('#^([a-z0-9]+?){1}://#i',$link['url'])){ + // external URL + $link = format_link_externalurl($link); + }elseif(preg_match("/^\\\\\\\\([a-z0-9\-_.]+)\\\\(.+)$/",$link['url'])){ + // windows shares + $link = format_link_windows($link); + }elseif(preg_match('#([a-z0-9\-_.]+?)@([\w\-]+\.([\w\-\.]+\.)*[\w]+)#i',$link['url'])){ + // email + $link = format_link_email($link); + }else{ + // wiki link + $link = format_link_wiki($link); + } + + //is realname an image? use media formater + if(preg_match('#^{{.*?\.(gif|png|jpe?g)(\?.*?)?\s*(\|.*?)?}}$#',$realname)){ + $link['name'] = substr($realname,2,-2); + $link = format_link_media($link); + } + + // build the replacement with the variables set by the formaters + return format_link_build($link); +} + +/** + * Simple text formating and typography is done here + */ +function simpleformat($text){ + global $conf; + + $text = preg_replace('/__(.+?)__/s','\1',$text); //underline + $text = preg_replace('/\/\/(.+?)\/\//s','\1',$text); //emphasize + $text = preg_replace('/\*\*(.+?)\*\*/s','\1',$text); //bold + $text = preg_replace('/\'\'(.+?)\'\'/s','\1',$text); //code + $text = preg_replace('#<del>(.*?)</del>#is','\1',$text); //deleted + $text = preg_replace('/^(\s)*----+(\s*)$/m',"

\n
\n

",$text); //hr + + //sub and superscript + $text = preg_replace('#<sub>(.*?)</sub>#is','\1',$text); + $text = preg_replace('#<sup>(.*?)</sup>#is','\1',$text); + + //do quoting + $text = preg_replace("/\n((>)[^\n]*?\n)+/se","'\n'.quoteformat('\\0').'\n'",$text); + + // Typography + if($conf['typography']){ + $text = preg_replace('/([^-])--([^-])/s','\1–\2',$text); //endash + $text = preg_replace('/([^-])---([^-])/s','\1—\2',$text); //emdash + $text = preg_replace('/"([^\"]+?)"/s','“\1”',$text); //curly quotes + $text = preg_replace('/(\s)\'(\S)/m','\1‘\2',$text); //single open quote + $text = preg_replace('/(\S)\'/','\1’',$text); //single closing quote or apostroph + $text = preg_replace('/\.\.\./','\1…\2',$text); //ellipse + $text = preg_replace('/(\d+)x(\d+)/i','\1×\2',$text); //640x480 + + $text = preg_replace('/>>/i','»',$text); // >> + $text = preg_replace('/<</i','«',$text); // << + + $text = preg_replace('/<->/i','↔',$text); // <-> + $text = preg_replace('/<-/i','←',$text); // <- + $text = preg_replace('/->/i','→',$text); // -> + + $text = preg_replace('/<=>/i','⇔',$text); // <=> + $text = preg_replace('/<=/i','⇐',$text); // <= + $text = preg_replace('/=>/i','⇒',$text); // => + + $text = preg_replace('/\(c\)/i','©',$text); // copyrigtht + $text = preg_replace('/\(r\)/i','®',$text); // registered + $text = preg_replace('/\(tm\)/i','™',$text); // trademark + } + + //forced linebreaks + $text = preg_replace('#\\\\\\\\(\s)#',"
\\1",$text); + + // lists (blocks leftover after blockformat) + $text = preg_replace("/(\n( {2,}|\t)[\*\-][^\n]+)(\n( {2,}|\t)[^\n]*)*/se","\"\\n\".listformat('\\0')",$text); + + // tables + $text = preg_replace("/\n(([\|\^][^\n]*?)+[\|\^] *\n)+/se","\"\\n\".tableformat('\\0')",$text); + + // footnotes + $text = footnotes($text); + + // run custom text replacements + $text = customs($text); + + return $text; +} + +/** + * Does the footnote formating + */ +function footnotes($text){ + $num = 0; + while (preg_match('/\(\((.+?)\)\)/s',$text,$match)){ + $num++; + $fn = $match[1]; + $linkt = ''.$num.')'; + $linkb = ''.$num.')'; + + $text = preg_replace('/ ?\(\((.+?)\)\)/s',$linkt,$text,1); + if($num == 1) $text .= '

'; + $text .= '
'.$linkb.' '.$fn.'
'; + } + + if($num) $text .= '
'; + return $text; +} + +/** + * Replaces smileys with their graphic equivalents + */ +function smileys(&$table,&$text){ + $smileys = file('conf/smileys.conf'); + foreach($smileys as $smiley){ + $smiley = preg_replace('/#.*$/','',$smiley); //ignore comments + $smiley = trim($smiley); + if(empty($smiley)) continue; + $sm = preg_split('/\s+/',$smiley,2); + $sm[1] = ''.$sm[0].''; + $sm[0] = preg_quote($sm[0],'/'); + firstpass($table,$text,'/(\W)'.$sm[0].'(\W)/s',$sm[1],"\\1","\\2"); + } +} + +/** + * Adds acronym tags to known acronyms + */ +function acronyms(&$table,&$text){ + $acronyms = file('conf/acronyms.conf'); + foreach($acronyms as $acro){ + $acro = preg_replace('/#.*$/','',$acro); //ignore comments + $acro = trim($acro); + if(empty($acro)) continue; + list($ac,$desc) = preg_split('/\s+/',$acro,2); + $ac = preg_quote($ac,'/'); + firstpass($table,$text,'/(\b)('.$ac.')(\b)/s',"\\2","\\1","\\3"); + } +} + + +/** + * Applies custom text replacements + */ +function customs($text){ + $reps = file ('conf/custom.conf'); + foreach($reps as $rep){ + //strip comments only outside a regexp + $rep = preg_replace('/#[^\/]*$/','',$rep); //ignore comments + $rep = trim($rep); + if(empty($rep)) continue; + if(preg_match('#^(/.+/\w*)\s+\'(.*)\'$#',$rep,$matches)){ + $text = preg_replace($matches[1],$matches[2],$text); + } + } + return $text; +} + +function firstpass(&$table,&$text,$regexp,$replace,$lpad='',$rpad=''){ + //extended regexps have to be disabled for inserting the token + //and later reenabled when handling the actual code: + $ext=''; + if(substr($regexp,-1) == 'e'){ + $ext='e'; + $regexp = substr($regexp,0,-1); + } + + while(preg_match($regexp,$text,$matches)){ + $token = mkToken(); + $match = $matches[0]; + $text = preg_replace($regexp,$lpad.$token.$rpad,$text,1); + $table[$token] = preg_replace($regexp.$ext,$replace,$match); + } +} + +function mkToken(){ + return '~'.md5(uniqid(rand(), true)).'~'; +} + +/** + * Do quote blocks + * + * FIXME fix paragraphs + */ +function quoteformat($block){ + $block = trim($block); + $lines = split("\n",$block); + + $lvl = 0; + $ret = ""; + foreach ($lines as $line){ + //remove '>' and count them + $cnt = 0; + while(substr($line,0,4) == '>'){ + $line = substr($line,4); + $cnt++; + } + //compare to last level and open or close new divs if needed + if($cnt > $lvl){ + $ret .= "

\n"; + for ($i=0; $i< $cnt - $lvl; $i++){ + $ret .= '
'; + } + $ret .= "\n

"; + }elseif($cnt < $lvl){ + $ret .= "\n

"; + for ($i=0; $i< $lvl - $cnt; $i++){ + $ret .= "
\n"; + } + $ret .= "

\n"; + }elseif(empty($line)){ + $ret .= "

\n

"; + } + //keep rest of line but trim left whitespaces + $ret .= ltrim($line)."\n"; + //remember level + $lvl = $cnt; + } + + //close remaining divs + $ret .= "

\n"; + for ($i=0; $i< $lvl; $i++){ + $ret .= "\n"; + } + $ret .= "

\n"; + + return "$ret"; +} + +function tableformat($block) { + $block = trim($block); + $lines = split("\n",$block); + $ret = ""; + //build a row array + $rows = array(); + for($r=0; $r < count($lines); $r++){ + $line = $lines[$r]; + //remove last seperator and trailing whitespace + $line = preg_replace('/[\|\^]\s*$/', '', $line); + $c = -1; //prepare colcounter) + for($chr=0; $chr < strlen($line); $chr++){ + if($line[$chr] == '^'){ + $c++; + $rows[$r][$c]['head'] = true; + $rows[$r][$c]['data'] = ''; + }elseif($line[$chr] == '|'){ + $c++; + $rows[$r][$c]['head'] = false; + $rows[$r][$c]['data'] = ''; + }else{ + $rows[$r][$c]['data'].= $line[$chr]; + } + } + } + + //build table + $ret .= "

\n\n"; + for($r=0; $r < count($rows); $r++){ + $ret .= " \n"; + + for ($c=0; $c < count($rows[$r]); $c++){ + $cspan=1; + $data = trim($rows[$r][$c]['data']); + $head = $rows[$r][$c]['head']; + + //join cells if next is empty + while($c < count($rows[$r])-1 && $rows[$r][$c+1]['data'] == ''){ + $c++; + $cspan++; + } + if($cspan > 1){ + $cspan = 'colspan="'.$cspan.'"'; + }else{ + $cspan = ''; + } + + if ($head) { + $ret .= " \n"; + } else { + $ret .= " \n"; + } + } + $ret .= " \n"; + } + $ret .= "
$data$data
\n

"; + + return $ret; +} + +function listformat($block){ + //remove 1st newline + $block = substr($block,1); + //unescape + $block = str_replace('\\"','"',$block); + +//dbg($block); + + //walk line by line + $ret=''; + $lst=0; + $lvl=0; + $enc=0; + $lines = split("\n",$block); + + //build an item array + $cnt=0; + $items = array(); + foreach ($lines as $line){ + //get intendion level + $lvl = 0; + $lvl += floor(strspn($line,' ')/2); + $lvl += strspn($line,"\t"); + //remove indents + $line = preg_replace('/^[ \t]+/','',$line); + //get type of list + (substr($line,0,1) == '-') ? $type='ol' : $type='ul'; + // remove bullet and following spaces + $line = preg_replace('/^[*\-]\s*/','',$line); + //add item to the list + $items[$cnt]['level'] = $lvl; + $items[$cnt]['type'] = $type; + $items[$cnt]['text'] = $line; + //increase counter + $cnt++; + } + + $level = 0; + $opens = array(); + + foreach ($items as $item){ + if( $item['level'] > $level ){ + //open new list + $ret .= "\n<".$item['type'].">\n"; + array_push($opens,$item['type']); + }elseif( $item['level'] < $level ){ + //close last item + $ret .= "\n"; + for ($i=0; $i<($level - $item['level']); $i++){ + //close higher lists + $ret .= '\n\n"; + } + }elseif($item['type'] != $opens[count($opens)-1]){ + //close last list and open new + $ret .= '\n\n"; + $ret .= "\n<".$item['type'].">\n"; + array_push($opens,$item['type']); + }else{ + //close last item + $ret .= "\n"; + } + + //remember current level and type + $level = $item['level']; + + //print item + $ret .= '

  • '; + $ret .= ''.$item['text'].''; + } + + //close remaining items and lists + while ($open = array_pop($opens)){ + $ret .= "
    • \n"; + $ret .= '\n"; + } + return "

    \n".$ret."\n

    "; +} + +function preformat($text,$type,$option=''){ + global $conf; + //unescape + $text = str_replace('\\"','"',$text); + + if($type == 'php' && !$conf['phpok']) $type='file'; + if($type == 'html' && !$conf['htmlok']) $type='file'; + + switch ($type){ + case 'php': + ob_start(); + eval($text); + $text = ob_get_contents(); + ob_end_clean(); + break; + case 'html': + break; + case 'nowiki': + $text = htmlspecialchars($text); + break; + case 'file': + $text = htmlspecialchars($text); + $text = "

    \n
    ".$text."
    \n

    "; + break; + case 'code': + if(empty($option)){ + $text = htmlspecialchars($text); + $text = '

    '.$text.'
    '; + }else{ + //strip leading blank line + $text = preg_replace('/^\s*?\n/','',$text); + //use geshi for highlighting + require_once("inc/geshi.php"); + $geshi = new GeSHi($text, strtolower($option), "inc/geshi"); + $geshi->enable_classes(); + $geshi->set_header_type(GESHI_HEADER_PRE); + $geshi->set_overall_class('code'); + $geshi->set_link_target($conf['target']['extern']); + $text = $geshi->parse_code(); + } + $text = "

    \n".$text."\n

    "; + break; + case 'block': + $text = substr($text,1); //remove 1st newline + $lines = split("\n",$text); //break into lines + $text = ''; + foreach($lines as $line){ + $text .= substr($line,2)."\n"; //remove indents + } + $text = htmlspecialchars($text); + $text = "

    \n
    ".$text."
    \n

    "; + break; + } + return $text; +} + +function mediaformat($text){ + global $conf; + + //unescape + $text = str_replace('\\"','"',$text); + + // format RSS + if(substr($text,0,4) == 'rss>'){ + return format_rss(substr($text,4)); + } + + //handle normal media stuff + $link = array(); + $link['name'] = $text; + $link = format_link_media($link); + return format_link_build($link); +} + +?> -- cgit v1.2.3