Duplicate code is one of the most pungent code smells. A rule that is often used is to re-structure code once it is duplicated in three or more places.
Common duplication problems, and corresponding solutions are:
Complex classes like htmLawed often do a lot of different things. To break such a class down, we need to identify a cohesive component within that class. A common approach to find such a component is to look for fields/methods that share the same prefixes, or suffixes. You can also have a look at the cohesion graph to spot any un-connected, or weakly-connected components.
Once you have determined the fields that belong together, you can apply the Extract Class refactoring. If the component makes sense as a sub-class, Extract Subclass is also a candidate, and is often faster.
While breaking up the class, it is a good idea to analyze how other classes use htmLawed, and based on these observations, apply Extract Interface, too.
| 1 | <?php |
||
| 13 | class htmLawed{ |
||
| 14 | // begin class |
||
| 15 | 1 | public static function hl($t, $C=1, $S=array()){ |
|
|
|
|||
| 16 | 1 | $C = is_array($C) ? $C : array(); |
|
| 17 | 1 | if(!empty($C['valid_xhtml'])){ |
|
| 18 | $C['elements'] = empty($C['elements']) ? '*-center-dir-font-isindex-menu-s-strike-u' : $C['elements']; |
||
| 19 | $C['make_tag_strict'] = isset($C['make_tag_strict']) ? $C['make_tag_strict'] : 2; |
||
| 20 | $C['xml:lang'] = isset($C['xml:lang']) ? $C['xml:lang'] : 2; |
||
| 21 | } |
||
| 22 | // config eles |
||
| 23 | 1 | $e = array('a'=>1, 'abbr'=>1, 'acronym'=>1, 'address'=>1, 'applet'=>1, 'area'=>1, 'b'=>1, 'bdo'=>1, 'big'=>1, 'blockquote'=>1, 'br'=>1, 'button'=>1, 'caption'=>1, 'center'=>1, 'cite'=>1, 'code'=>1, 'col'=>1, 'colgroup'=>1, 'dd'=>1, 'del'=>1, 'dfn'=>1, 'dir'=>1, 'div'=>1, 'dl'=>1, 'dt'=>1, 'em'=>1, 'embed'=>1, 'fieldset'=>1, 'font'=>1, 'form'=>1, 'h1'=>1, 'h2'=>1, 'h3'=>1, 'h4'=>1, 'h5'=>1, 'h6'=>1, 'hr'=>1, 'i'=>1, 'iframe'=>1, 'img'=>1, 'input'=>1, 'ins'=>1, 'isindex'=>1, 'kbd'=>1, 'label'=>1, 'legend'=>1, 'li'=>1, 'map'=>1, 'menu'=>1, 'noscript'=>1, 'object'=>1, 'ol'=>1, 'optgroup'=>1, 'option'=>1, 'p'=>1, 'param'=>1, 'pre'=>1, 'q'=>1, 'rb'=>1, 'rbc'=>1, 'rp'=>1, 'rt'=>1, 'rtc'=>1, 'ruby'=>1, 's'=>1, 'samp'=>1, 'script'=>1, 'select'=>1, 'small'=>1, 'span'=>1, 'strike'=>1, 'strong'=>1, 'sub'=>1, 'sup'=>1, 'table'=>1, 'tbody'=>1, 'td'=>1, 'textarea'=>1, 'tfoot'=>1, 'th'=>1, 'thead'=>1, 'tr'=>1, 'tt'=>1, 'u'=>1, 'ul'=>1, 'var'=>1); // 86/deprecated+embed+ruby |
|
| 24 | 1 | if(!empty($C['safe'])){ |
|
| 25 | 1 | unset($e['applet'], $e['embed'], $e['iframe'], $e['object'], $e['script']); |
|
| 26 | } |
||
| 27 | 1 | $x = !empty($C['elements']) ? str_replace(array("\n", "\r", "\t", ' '), '', $C['elements']) : '*'; |
|
| 28 | 1 | if($x == '-*'){$e = array();} |
|
| 29 | 1 | elseif(strpos($x, '*') === false){$e = array_flip(explode(',', $x));} |
|
| 30 | else{ |
||
| 31 | 1 | if(isset($x[1])){ |
|
| 32 | preg_match_all('`(?:^|-|\+)[^\-+]+?(?=-|\+|$)`', $x, $m, PREG_SET_ORDER); |
||
| 33 | for($i=count($m); --$i>=0;){$m[$i] = $m[$i][0];} |
||
| 34 | foreach($m as $v){ |
||
| 35 | if($v[0] == '+'){$e[substr($v, 1)] = 1;} |
||
| 36 | if($v[0] == '-' && isset($e[($v = substr($v, 1))]) && !in_array('+'. $v, $m)){unset($e[$v]);} |
||
| 37 | } |
||
| 38 | } |
||
| 39 | } |
||
| 40 | 1 | $C['elements'] =& $e; |
|
| 41 | // config attrs |
||
| 42 | 1 | $x = !empty($C['deny_attribute']) ? str_replace(array("\n", "\r", "\t", ' '), '', $C['deny_attribute']) : ''; |
|
| 43 | 1 | $x = array_flip((isset($x[0]) && $x[0] == '*') ? explode('-', $x) : explode(',', $x. (!empty($C['safe']) ? ',on*' : ''))); |
|
| 44 | 1 | if(isset($x['on*'])){ |
|
| 45 | 1 | unset($x['on*']); |
|
| 46 | 1 | $x += array('onblur'=>1, 'onchange'=>1, 'onclick'=>1, 'ondblclick'=>1, 'onfocus'=>1, 'onkeydown'=>1, 'onkeypress'=>1, 'onkeyup'=>1, 'onmousedown'=>1, 'onmousemove'=>1, 'onmouseout'=>1, 'onmouseover'=>1, 'onmouseup'=>1, 'onreset'=>1, 'onselect'=>1, 'onsubmit'=>1); |
|
| 47 | } |
||
| 48 | 1 | $C['deny_attribute'] = $x; |
|
| 49 | // config URL |
||
| 50 | 1 | $x = (isset($C['schemes'][2]) && strpos($C['schemes'], ':')) ? strtolower($C['schemes']) : 'href: aim, feed, file, ftp, gopher, http, https, irc, mailto, news, nntp, sftp, ssh, telnet; *:file, http, https'; |
|
| 51 | 1 | $C['schemes'] = array(); |
|
| 52 | 1 | foreach(explode(';', str_replace(array(' ', "\t", "\r", "\n"), '', $x)) as $v){ |
|
| 53 | 1 | $x = $x2 = null; list($x, $x2) = explode(':', $v, 2); |
|
| 54 | 1 | if($x2){$C['schemes'][$x] = array_flip(explode(',', $x2));} |
|
| 55 | } |
||
| 56 | 1 | if(!isset($C['schemes']['*'])){$C['schemes']['*'] = array('file'=>1, 'http'=>1, 'https'=>1,);} |
|
| 57 | 1 | if(!empty($C['safe']) && empty($C['schemes']['style'])){$C['schemes']['style'] = array('!'=>1);} |
|
| 58 | 1 | $C['abs_url'] = isset($C['abs_url']) ? $C['abs_url'] : 0; |
|
| 59 | 1 | if(!isset($C['base_url']) or !preg_match('`^[a-zA-Z\d.+\-]+://[^/]+/(.+?/)?$`', $C['base_url'])){ |
|
| 60 | 1 | $C['base_url'] = $C['abs_url'] = 0; |
|
| 61 | } |
||
| 62 | // config rest |
||
| 63 | 1 | $C['and_mark'] = empty($C['and_mark']) ? 0 : 1; |
|
| 64 | 1 | $C['anti_link_spam'] = (isset($C['anti_link_spam']) && is_array($C['anti_link_spam']) && count($C['anti_link_spam']) == 2 && (empty($C['anti_link_spam'][0]) or htmLawed::hl_regex($C['anti_link_spam'][0])) && (empty($C['anti_link_spam'][1]) or htmLawed::hl_regex($C['anti_link_spam'][1]))) ? $C['anti_link_spam'] : 0; |
|
| 65 | 1 | $C['anti_mail_spam'] = isset($C['anti_mail_spam']) ? $C['anti_mail_spam'] : 0; |
|
| 66 | 1 | $C['balance'] = isset($C['balance']) ? (bool)$C['balance'] : 1; |
|
| 67 | 1 | $C['cdata'] = isset($C['cdata']) ? $C['cdata'] : (empty($C['safe']) ? 3 : 0); |
|
| 68 | 1 | $C['clean_ms_char'] = empty($C['clean_ms_char']) ? 0 : $C['clean_ms_char']; |
|
| 69 | 1 | $C['comment'] = isset($C['comment']) ? $C['comment'] : (empty($C['safe']) ? 3 : 0); |
|
| 70 | 1 | $C['css_expression'] = empty($C['css_expression']) ? 0 : 1; |
|
| 71 | 1 | $C['direct_list_nest'] = empty($C['direct_list_nest']) ? 0 : 1; |
|
| 72 | 1 | $C['hexdec_entity'] = isset($C['hexdec_entity']) ? $C['hexdec_entity'] : 1; |
|
| 73 | 1 | $C['hook'] = (!empty($C['hook']) && function_exists($C['hook'])) ? $C['hook'] : 0; |
|
| 74 | 1 | $C['hook_tag'] = (!empty($C['hook_tag']) && function_exists($C['hook_tag'])) ? $C['hook_tag'] : 0; |
|
| 75 | 1 | $C['keep_bad'] = isset($C['keep_bad']) ? $C['keep_bad'] : 6; |
|
| 76 | 1 | $C['lc_std_val'] = isset($C['lc_std_val']) ? (bool)$C['lc_std_val'] : 1; |
|
| 77 | 1 | $C['make_tag_strict'] = isset($C['make_tag_strict']) ? $C['make_tag_strict'] : 1; |
|
| 78 | 1 | $C['named_entity'] = isset($C['named_entity']) ? (bool)$C['named_entity'] : 1; |
|
| 79 | 1 | $C['no_deprecated_attr'] = isset($C['no_deprecated_attr']) ? $C['no_deprecated_attr'] : 1; |
|
| 80 | 1 | $C['parent'] = isset($C['parent'][0]) ? strtolower($C['parent']) : 'body'; |
|
| 81 | 1 | $C['show_setting'] = !empty($C['show_setting']) ? $C['show_setting'] : 0; |
|
| 82 | 1 | $C['style_pass'] = empty($C['style_pass']) ? 0 : 1; |
|
| 83 | 1 | $C['tidy'] = empty($C['tidy']) ? 0 : $C['tidy']; |
|
| 84 | 1 | $C['unique_ids'] = isset($C['unique_ids']) ? $C['unique_ids'] : 1; |
|
| 85 | 1 | $C['xml:lang'] = isset($C['xml:lang']) ? $C['xml:lang'] : 0; |
|
| 86 | |||
| 87 | 1 | if(isset($GLOBALS['C'])){$reC = $GLOBALS['C'];} |
|
| 88 | 1 | $GLOBALS['C'] = $C; |
|
| 89 | 1 | $S = is_array($S) ? $S : htmLawed::hl_spec($S); |
|
| 90 | 1 | if(isset($GLOBALS['S'])){$reS = $GLOBALS['S'];} |
|
| 91 | 1 | $GLOBALS['S'] = $S; |
|
| 92 | |||
| 93 | 1 | $t = preg_replace('`[\x00-\x08\x0b-\x0c\x0e-\x1f]`', '', $t); |
|
| 94 | 1 | if($C['clean_ms_char']){ |
|
| 95 | $x = array("\x7f"=>'', "\x80"=>'€', "\x81"=>'', "\x83"=>'ƒ', "\x85"=>'…', "\x86"=>'†', "\x87"=>'‡', "\x88"=>'ˆ', "\x89"=>'‰', "\x8a"=>'Š', "\x8b"=>'‹', "\x8c"=>'Œ', "\x8d"=>'', "\x8e"=>'Ž', "\x8f"=>'', "\x90"=>'', "\x95"=>'•', "\x96"=>'–', "\x97"=>'—', "\x98"=>'˜', "\x99"=>'™', "\x9a"=>'š', "\x9b"=>'›', "\x9c"=>'œ', "\x9d"=>'', "\x9e"=>'ž', "\x9f"=>'Ÿ'); |
||
| 96 | $x = $x + ($C['clean_ms_char'] == 1 ? array("\x82"=>'‚', "\x84"=>'„', "\x91"=>'‘', "\x92"=>'’', "\x93"=>'“', "\x94"=>'”') : array("\x82"=>'\'', "\x84"=>'"', "\x91"=>'\'', "\x92"=>'\'', "\x93"=>'"', "\x94"=>'"')); |
||
| 97 | $t = strtr($t, $x); |
||
| 98 | } |
||
| 99 | 1 | if($C['cdata'] or $C['comment']){$t = preg_replace_callback('`<!(?:(?:--.*?--)|(?:\[CDATA\[.*?\]\]))>`sm', 'htmLawed::hl_cmtcd', $t);} |
|
| 100 | 1 | $t = preg_replace_callback('`&([A-Za-z][A-Za-z0-9]{1,30}|#(?:[0-9]{1,8}|[Xx][0-9A-Fa-f]{1,7}));`', 'htmLawed::hl_ent', str_replace('&', '&', $t)); |
|
| 101 | 1 | if($C['unique_ids'] && !isset($GLOBALS['hl_Ids'])){$GLOBALS['hl_Ids'] = array();} |
|
| 102 | 1 | if($C['hook']){$t = $C['hook']($t, $C, $S);} |
|
| 103 | 1 | if($C['show_setting'] && preg_match('`^[a-z][a-z0-9_]*$`i', $C['show_setting'])){ |
|
| 104 | $GLOBALS[$C['show_setting']] = array('config'=>$C, 'spec'=>$S, 'time'=>microtime()); |
||
| 105 | } |
||
| 106 | // main |
||
| 107 | 1 | $t = preg_replace_callback('`<(?:(?:\s|$)|(?:[^>]*(?:>|$)))|>`m', 'htmLawed::hl_tag', $t); |
|
| 108 | 1 | $t = $C['balance'] ? htmLawed::hl_bal($t, $C['keep_bad'], $C['parent']) : $t; |
|
| 109 | 1 | $t = (($C['cdata'] or $C['comment']) && strpos($t, "\x01") !== false) ? str_replace(array("\x01", "\x02", "\x03", "\x04", "\x05"), array('', '', '&', '<', '>'), $t) : $t; |
|
| 110 | 1 | $t = $C['tidy'] ? htmLawed::hl_tidy($t, $C['tidy'], $C['parent']) : $t; |
|
| 111 | 1 | unset($C, $e); |
|
| 112 | 1 | if(isset($reC)){$GLOBALS['C'] = $reC;} |
|
| 113 | 1 | if(isset($reS)){$GLOBALS['S'] = $reS;} |
|
| 114 | 1 | return $t; |
|
| 115 | // eof |
||
| 116 | } |
||
| 117 | |||
| 118 | public static function hl_attrval($t, $p){ |
||
| 119 | // check attr val against $S |
||
| 120 | $o = 1; $l = strlen($t); |
||
| 121 | foreach($p as $k=>$v){ |
||
| 122 | switch($k){ |
||
| 123 | case 'maxlen':if($l > $v){$o = 0;} |
||
| 124 | break; case 'minlen': if($l < $v){$o = 0;} |
||
| 125 | break; case 'maxval': if((float)($t) > $v){$o = 0;} |
||
| 126 | break; case 'minval': if((float)($t) < $v){$o = 0;} |
||
| 127 | break; case 'match': if(!preg_match($v, $t)){$o = 0;} |
||
| 128 | break; case 'nomatch': if(preg_match($v, $t)){$o = 0;} |
||
| 129 | break; case 'oneof': |
||
| 130 | $m = 0; |
||
| 131 | foreach(explode('|', $v) as $n){if($t == $n){$m = 1; break;}} |
||
| 132 | $o = $m; |
||
| 133 | break; case 'noneof': |
||
| 134 | $m = 1; |
||
| 135 | foreach(explode('|', $v) as $n){if($t == $n){$m = 0; break;}} |
||
| 136 | $o = $m; |
||
| 137 | break; default: |
||
| 138 | break; |
||
| 139 | } |
||
| 140 | if(!$o){break;} |
||
| 141 | } |
||
| 142 | return ($o ? $t : (isset($p['default']) ? $p['default'] : 0)); |
||
| 143 | // eof |
||
| 144 | } |
||
| 145 | |||
| 146 | 1 | public static function hl_bal($t, $do=1, $in='div'){ |
|
| 147 | // balance tags |
||
| 148 | // by content |
||
| 149 | 1 | $cB = array('blockquote'=>1, 'form'=>1, 'map'=>1, 'noscript'=>1); // Block |
|
| 150 | 1 | $cE = array('area'=>1, 'br'=>1, 'col'=>1, 'embed'=>1, 'hr'=>1, 'img'=>1, 'input'=>1, 'isindex'=>1, 'param'=>1); // Empty |
|
| 151 | 1 | $cF = array('button'=>1, 'del'=>1, 'div'=>1, 'dd'=>1, 'fieldset'=>1, 'iframe'=>1, 'ins'=>1, 'li'=>1, 'noscript'=>1, 'object'=>1, 'td'=>1, 'th'=>1); // Flow; later context-wise dynamic move of ins & del to $cI |
|
| 152 | 1 | $cI = array('a'=>1, 'abbr'=>1, 'acronym'=>1, 'address'=>1, 'b'=>1, 'bdo'=>1, 'big'=>1, 'caption'=>1, 'cite'=>1, 'code'=>1, 'dfn'=>1, 'dt'=>1, 'em'=>1, 'font'=>1, 'h1'=>1, 'h2'=>1, 'h3'=>1, 'h4'=>1, 'h5'=>1, 'h6'=>1, 'i'=>1, 'kbd'=>1, 'label'=>1, 'legend'=>1, 'p'=>1, 'pre'=>1, 'q'=>1, 'rb'=>1, 'rt'=>1, 's'=>1, 'samp'=>1, 'small'=>1, 'span'=>1, 'strike'=>1, 'strong'=>1, 'sub'=>1, 'sup'=>1, 'tt'=>1, 'u'=>1, 'var'=>1); // Inline |
|
| 153 | 1 | $cN = array('a'=>array('a'=>1), 'button'=>array('a'=>1, 'button'=>1, 'fieldset'=>1, 'form'=>1, 'iframe'=>1, 'input'=>1, 'label'=>1, 'select'=>1, 'textarea'=>1), 'fieldset'=>array('fieldset'=>1), 'form'=>array('form'=>1), 'label'=>array('label'=>1), 'noscript'=>array('script'=>1), 'pre'=>array('big'=>1, 'font'=>1, 'img'=>1, 'object'=>1, 'script'=>1, 'small'=>1, 'sub'=>1, 'sup'=>1), 'rb'=>array('ruby'=>1), 'rt'=>array('ruby'=>1)); // Illegal |
|
| 154 | 1 | $cN2 = array_keys($cN); |
|
| 155 | 1 | $cR = array('blockquote'=>1, 'dir'=>1, 'dl'=>1, 'form'=>1, 'map'=>1, 'menu'=>1, 'noscript'=>1, 'ol'=>1, 'optgroup'=>1, 'rbc'=>1, 'rtc'=>1, 'ruby'=>1, 'select'=>1, 'table'=>1, 'tbody'=>1, 'tfoot'=>1, 'thead'=>1, 'tr'=>1, 'ul'=>1); |
|
| 156 | 1 | $cS = array('colgroup'=>array('col'=>1), 'dir'=>array('li'=>1), 'dl'=>array('dd'=>1, 'dt'=>1), 'menu'=>array('li'=>1), 'ol'=>array('li'=>1), 'optgroup'=>array('option'=>1), 'option'=>array('#pcdata'=>1), 'rbc'=>array('rb'=>1), 'rp'=>array('#pcdata'=>1), 'rtc'=>array('rt'=>1), 'ruby'=>array('rb'=>1, 'rbc'=>1, 'rp'=>1, 'rt'=>1, 'rtc'=>1), 'select'=>array('optgroup'=>1, 'option'=>1), 'script'=>array('#pcdata'=>1), 'table'=>array('caption'=>1, 'col'=>1, 'colgroup'=>1, 'tfoot'=>1, 'tbody'=>1, 'tr'=>1, 'thead'=>1), 'tbody'=>array('tr'=>1), 'tfoot'=>array('tr'=>1), 'textarea'=>array('#pcdata'=>1), 'thead'=>array('tr'=>1), 'tr'=>array('td'=>1, 'th'=>1), 'ul'=>array('li'=>1)); // Specific - immediate parent-child |
|
| 157 | 1 | if($GLOBALS['C']['direct_list_nest']){$cS['ol'] = $cS['ul'] += array('ol'=>1, 'ul'=>1);} |
|
| 158 | 1 | $cO = array('address'=>array('p'=>1), 'applet'=>array('param'=>1), 'blockquote'=>array('script'=>1), 'fieldset'=>array('legend'=>1, '#pcdata'=>1), 'form'=>array('script'=>1), 'map'=>array('area'=>1), 'object'=>array('param'=>1, 'embed'=>1)); // Other |
|
| 159 | 1 | $cT = array('colgroup'=>1, 'dd'=>1, 'dt'=>1, 'li'=>1, 'option'=>1, 'p'=>1, 'td'=>1, 'tfoot'=>1, 'th'=>1, 'thead'=>1, 'tr'=>1); // Omitable closing |
|
| 160 | // block/inline type; ins & del both type; #pcdata: text |
||
| 161 | 1 | $eB = array('address'=>1, 'blockquote'=>1, 'center'=>1, 'del'=>1, 'dir'=>1, 'dl'=>1, 'div'=>1, 'fieldset'=>1, 'form'=>1, 'ins'=>1, 'h1'=>1, 'h2'=>1, 'h3'=>1, 'h4'=>1, 'h5'=>1, 'h6'=>1, 'hr'=>1, 'isindex'=>1, 'menu'=>1, 'noscript'=>1, 'ol'=>1, 'p'=>1, 'pre'=>1, 'table'=>1, 'ul'=>1); |
|
| 162 | 1 | $eI = array('#pcdata'=>1, 'a'=>1, 'abbr'=>1, 'acronym'=>1, 'applet'=>1, 'b'=>1, 'bdo'=>1, 'big'=>1, 'br'=>1, 'button'=>1, 'cite'=>1, 'code'=>1, 'del'=>1, 'dfn'=>1, 'em'=>1, 'embed'=>1, 'font'=>1, 'i'=>1, 'iframe'=>1, 'img'=>1, 'input'=>1, 'ins'=>1, 'kbd'=>1, 'label'=>1, 'map'=>1, 'object'=>1, 'q'=>1, 'ruby'=>1, 's'=>1, 'samp'=>1, 'select'=>1, 'script'=>1, 'small'=>1, 'span'=>1, 'strike'=>1, 'strong'=>1, 'sub'=>1, 'sup'=>1, 'textarea'=>1, 'tt'=>1, 'u'=>1, 'var'=>1); |
|
| 163 | 1 | $eN = array('a'=>1, 'big'=>1, 'button'=>1, 'fieldset'=>1, 'font'=>1, 'form'=>1, 'iframe'=>1, 'img'=>1, 'input'=>1, 'label'=>1, 'object'=>1, 'ruby'=>1, 'script'=>1, 'select'=>1, 'small'=>1, 'sub'=>1, 'sup'=>1, 'textarea'=>1); // Exclude from specific ele; $cN values |
|
| 164 | 1 | $eO = array('area'=>1, 'caption'=>1, 'col'=>1, 'colgroup'=>1, 'dd'=>1, 'dt'=>1, 'legend'=>1, 'li'=>1, 'optgroup'=>1, 'option'=>1, 'param'=>1, 'rb'=>1, 'rbc'=>1, 'rp'=>1, 'rt'=>1, 'rtc'=>1, 'script'=>1, 'tbody'=>1, 'td'=>1, 'tfoot'=>1, 'thead'=>1, 'th'=>1, 'tr'=>1); // Missing in $eB & $eI |
|
| 165 | 1 | $eF = $eB + $eI; |
|
| 166 | |||
| 167 | // $in sets allowed child |
||
| 168 | 1 | $in = ((isset($eF[$in]) && $in != '#pcdata') or isset($eO[$in])) ? $in : 'div'; |
|
| 169 | 1 | if(isset($cE[$in])){ |
|
| 170 | return (!$do ? '' : str_replace(array('<', '>'), array('<', '>'), $t)); |
||
| 171 | } |
||
| 172 | 1 | if(isset($cS[$in])){$inOk = $cS[$in];} |
|
| 173 | 1 | elseif(isset($cI[$in])){$inOk = $eI; $cI['del'] = 1; $cI['ins'] = 1;} |
|
| 174 | 1 | elseif(isset($cF[$in])){$inOk = $eF; unset($cI['del'], $cI['ins']);} |
|
| 175 | elseif(isset($cB[$in])){$inOk = $eB; unset($cI['del'], $cI['ins']);} |
||
| 176 | 1 | if(isset($cO[$in])){$inOk = $inOk + $cO[$in];} |
|
| 177 | 1 | if(isset($cN[$in])){$inOk = array_diff_assoc($inOk, $cN[$in]);} |
|
| 178 | |||
| 179 | 1 | $t = explode('<', $t); |
|
| 180 | 1 | $ok = $q = array(); // $q seq list of open non-empty ele |
|
| 181 | 1 | ob_start(); |
|
| 182 | |||
| 183 | 1 | for($i=-1, $ci=count($t); ++$i<$ci;){ |
|
| 184 | // allowed $ok in parent $p |
||
| 185 | 1 | View Code Duplication | if($ql = count($q)){ |
| 186 | $p = array_pop($q); |
||
| 187 | $q[] = $p; |
||
| 188 | if(isset($cS[$p])){$ok = $cS[$p];} |
||
| 189 | elseif(isset($cI[$p])){$ok = $eI; $cI['del'] = 1; $cI['ins'] = 1;} |
||
| 190 | elseif(isset($cF[$p])){$ok = $eF; unset($cI['del'], $cI['ins']);} |
||
| 191 | elseif(isset($cB[$p])){$ok = $eB; unset($cI['del'], $cI['ins']);} |
||
| 192 | if(isset($cO[$p])){$ok = $ok + $cO[$p];} |
||
| 193 | if(isset($cN[$p])){$ok = array_diff_assoc($ok, $cN[$p]);} |
||
| 194 | 1 | }else{$ok = $inOk; unset($cI['del'], $cI['ins']);} |
|
| 195 | // bad tags, & ele content |
||
| 196 | 1 | View Code Duplication | if(isset($e) && ($do == 1 or (isset($ok['#pcdata']) && ($do == 3 or $do == 5)))){ |
| 197 | echo '<', $s, $e, $a, '>'; |
||
| 198 | } |
||
| 199 | 1 | View Code Duplication | if(isset($x[0])){ |
| 200 | if(strlen(trim($x)) && (($ql && isset($cB[$p])) or (isset($cB[$in]) && !$ql))){ |
||
| 201 | echo '<div>', $x, '</div>'; |
||
| 202 | } |
||
| 203 | elseif($do < 3 or isset($ok['#pcdata'])){echo $x;} |
||
| 204 | elseif(strpos($x, "\x02\x04")){ |
||
| 205 | foreach(preg_split('`(\x01\x02[^\x01\x02]+\x02\x01)`', $x, -1, PREG_SPLIT_DELIM_CAPTURE | PREG_SPLIT_NO_EMPTY) as $v){ |
||
| 206 | echo (substr($v, 0, 2) == "\x01\x02" ? $v : ($do > 4 ? preg_replace('`\S`', '', $v) : '')); |
||
| 207 | } |
||
| 208 | }elseif($do > 4){echo preg_replace('`\S`', '', $x);} |
||
| 209 | } |
||
| 210 | // get markup |
||
| 211 | 1 | if(!preg_match('`^(/?)([a-z1-6]+)([^>]*)>(.*)`sm', $t[$i], $r)){$x = $t[$i]; continue;} |
|
| 212 | $s = null; $e = null; $a = null; $x = null; list($all, $s, $e, $a, $x) = $r; |
||
| 213 | // close tag |
||
| 214 | if($s){ |
||
| 215 | if(isset($cE[$e]) or !in_array($e, $q)){continue;} // Empty/unopen |
||
| 216 | if($p == $e){array_pop($q); echo '</', $e, '>'; unset($e); continue;} // Last open |
||
| 217 | $add = ''; // Nesting - close open tags that need to be |
||
| 218 | for($j=-1, $cj=count($q); ++$j<$cj;){ |
||
| 219 | if(($d = array_pop($q)) == $e){break;} |
||
| 220 | else{$add .= "</{$d}>";} |
||
| 221 | } |
||
| 222 | echo $add, '</', $e, '>'; unset($e); continue; |
||
| 223 | } |
||
| 224 | // open tag |
||
| 225 | // $cB ele needs $eB ele as child |
||
| 226 | if(isset($cB[$e]) && strlen(trim($x))){ |
||
| 227 | $t[$i] = "{$e}{$a}>"; |
||
| 228 | array_splice($t, $i+1, 0, 'div>'. $x); unset($e, $x); ++$ci; --$i; continue; |
||
| 229 | } |
||
| 230 | if((($ql && isset($cB[$p])) or (isset($cB[$in]) && !$ql)) && !isset($eB[$e]) && !isset($ok[$e])){ |
||
| 231 | array_splice($t, $i, 0, 'div>'); unset($e, $x); ++$ci; --$i; continue; |
||
| 232 | } |
||
| 233 | // if no open ele, $in = parent; mostly immediate parent-child relation should hold |
||
| 234 | if(!$ql or !isset($eN[$e]) or !array_intersect($q, $cN2)){ |
||
| 235 | if(!isset($ok[$e])){ |
||
| 236 | if($ql && isset($cT[$p])){echo '</', array_pop($q), '>'; unset($e, $x); --$i;} |
||
| 237 | continue; |
||
| 238 | } |
||
| 239 | if(!isset($cE[$e])){$q[] = $e;} |
||
| 240 | echo '<', $e, $a, '>'; unset($e); continue; |
||
| 241 | } |
||
| 242 | // specific parent-child |
||
| 243 | if(isset($cS[$p][$e])){ |
||
| 244 | if(!isset($cE[$e])){$q[] = $e;} |
||
| 245 | echo '<', $e, $a, '>'; unset($e); continue; |
||
| 246 | } |
||
| 247 | // nesting |
||
| 248 | $add = ''; |
||
| 249 | $q2 = array(); |
||
| 250 | for($k=-1, $kc=count($q); ++$k<$kc;){ |
||
| 251 | $d = $q[$k]; |
||
| 252 | $ok2 = array(); |
||
| 253 | if(isset($cS[$d])){$q2[] = $d; continue;} |
||
| 254 | $ok2 = isset($cI[$d]) ? $eI : $eF; |
||
| 255 | if(isset($cO[$d])){$ok2 = $ok2 + $cO[$d];} |
||
| 256 | if(isset($cN[$d])){$ok2 = array_diff_assoc($ok2, $cN[$d]);} |
||
| 257 | if(!isset($ok2[$e])){ |
||
| 258 | if(!$k && !isset($inOk[$e])){continue 2;} |
||
| 259 | $add = "</{$d}>"; |
||
| 260 | for(;++$k<$kc;){$add = "</{$q[$k]}>{$add}";} |
||
| 261 | break; |
||
| 262 | } |
||
| 263 | else{$q2[] = $d;} |
||
| 264 | } |
||
| 265 | $q = $q2; |
||
| 266 | if(!isset($cE[$e])){$q[] = $e;} |
||
| 267 | echo $add, '<', $e, $a, '>'; unset($e); continue; |
||
| 268 | } |
||
| 269 | |||
| 270 | // end |
||
| 271 | 1 | View Code Duplication | if($ql = count($q)){ |
| 272 | $p = array_pop($q); |
||
| 273 | $q[] = $p; |
||
| 274 | if(isset($cS[$p])){$ok = $cS[$p];} |
||
| 275 | elseif(isset($cI[$p])){$ok = $eI; $cI['del'] = 1; $cI['ins'] = 1;} |
||
| 276 | elseif(isset($cF[$p])){$ok = $eF; unset($cI['del'], $cI['ins']);} |
||
| 277 | elseif(isset($cB[$p])){$ok = $eB; unset($cI['del'], $cI['ins']);} |
||
| 278 | if(isset($cO[$p])){$ok = $ok + $cO[$p];} |
||
| 279 | if(isset($cN[$p])){$ok = array_diff_assoc($ok, $cN[$p]);} |
||
| 280 | 1 | }else{$ok = $inOk; unset($cI['del'], $cI['ins']);} |
|
| 281 | 1 | View Code Duplication | if(isset($e) && ($do == 1 or (isset($ok['#pcdata']) && ($do == 3 or $do == 5)))){ |
| 282 | echo '<', $s, $e, $a, '>'; |
||
| 283 | } |
||
| 284 | 1 | View Code Duplication | if(isset($x[0])){ |
| 285 | if(strlen(trim($x)) && (($ql && isset($cB[$p])) or (isset($cB[$in]) && !$ql))){ |
||
| 286 | echo '<div>', $x, '</div>'; |
||
| 287 | } |
||
| 288 | elseif($do < 3 or isset($ok['#pcdata'])){echo $x;} |
||
| 289 | elseif(strpos($x, "\x02\x04")){ |
||
| 290 | foreach(preg_split('`(\x01\x02[^\x01\x02]+\x02\x01)`', $x, -1, PREG_SPLIT_DELIM_CAPTURE | PREG_SPLIT_NO_EMPTY) as $v){ |
||
| 291 | echo (substr($v, 0, 2) == "\x01\x02" ? $v : ($do > 4 ? preg_replace('`\S`', '', $v) : '')); |
||
| 292 | } |
||
| 293 | }elseif($do > 4){echo preg_replace('`\S`', '', $x);} |
||
| 294 | } |
||
| 295 | 1 | while(!empty($q) && ($e = array_pop($q))){echo '</', $e, '>';} |
|
| 296 | 1 | $o = ob_get_contents(); |
|
| 297 | 1 | ob_end_clean(); |
|
| 298 | 1 | return $o; |
|
| 299 | // eof |
||
| 300 | } |
||
| 301 | |||
| 302 | public static function hl_cmtcd($t){ |
||
| 303 | // comment/CDATA sec handler |
||
| 304 | $t = $t[0]; |
||
| 305 | global $C; |
||
| 306 | if(!($v = $C[$n = $t[3] == '-' ? 'comment' : 'cdata'])){return $t;} |
||
| 307 | if($v == 1){return '';} |
||
| 308 | if($n == 'comment'){ |
||
| 309 | if(substr(($t = preg_replace('`--+`', '-', substr($t, 4, -3))), -1) != ' '){$t .= ' ';} |
||
| 310 | } |
||
| 311 | else{$t = substr($t, 1, -1);} |
||
| 312 | $t = $v == 2 ? str_replace(array('&', '<', '>'), array('&', '<', '>'), $t) : $t; |
||
| 313 | return str_replace(array('&', '<', '>'), array("\x03", "\x04", "\x05"), ($n == 'comment' ? "\x01\x02\x04!--$t--\x05\x02\x01" : "\x01\x01\x04$t\x05\x01\x01")); |
||
| 314 | // eof |
||
| 315 | } |
||
| 316 | |||
| 317 | public static function hl_ent($t){ |
||
| 318 | // entitity handler |
||
| 319 | global $C; |
||
| 320 | $t = $t[1]; |
||
| 321 | static $U = array('quot'=>1,'amp'=>1,'lt'=>1,'gt'=>1); |
||
| 322 | static $N = array('fnof'=>'402', 'Alpha'=>'913', 'Beta'=>'914', 'Gamma'=>'915', 'Delta'=>'916', 'Epsilon'=>'917', 'Zeta'=>'918', 'Eta'=>'919', 'Theta'=>'920', 'Iota'=>'921', 'Kappa'=>'922', 'Lambda'=>'923', 'Mu'=>'924', 'Nu'=>'925', 'Xi'=>'926', 'Omicron'=>'927', 'Pi'=>'928', 'Rho'=>'929', 'Sigma'=>'931', 'Tau'=>'932', 'Upsilon'=>'933', 'Phi'=>'934', 'Chi'=>'935', 'Psi'=>'936', 'Omega'=>'937', 'alpha'=>'945', 'beta'=>'946', 'gamma'=>'947', 'delta'=>'948', 'epsilon'=>'949', 'zeta'=>'950', 'eta'=>'951', 'theta'=>'952', 'iota'=>'953', 'kappa'=>'954', 'lambda'=>'955', 'mu'=>'956', 'nu'=>'957', 'xi'=>'958', 'omicron'=>'959', 'pi'=>'960', 'rho'=>'961', 'sigmaf'=>'962', 'sigma'=>'963', 'tau'=>'964', 'upsilon'=>'965', 'phi'=>'966', 'chi'=>'967', 'psi'=>'968', 'omega'=>'969', 'thetasym'=>'977', 'upsih'=>'978', 'piv'=>'982', 'bull'=>'8226', 'hellip'=>'8230', 'prime'=>'8242', 'Prime'=>'8243', 'oline'=>'8254', 'frasl'=>'8260', 'weierp'=>'8472', 'image'=>'8465', 'real'=>'8476', 'trade'=>'8482', 'alefsym'=>'8501', 'larr'=>'8592', 'uarr'=>'8593', 'rarr'=>'8594', 'darr'=>'8595', 'harr'=>'8596', 'crarr'=>'8629', 'lArr'=>'8656', 'uArr'=>'8657', 'rArr'=>'8658', 'dArr'=>'8659', 'hArr'=>'8660', 'forall'=>'8704', 'part'=>'8706', 'exist'=>'8707', 'empty'=>'8709', 'nabla'=>'8711', 'isin'=>'8712', 'notin'=>'8713', 'ni'=>'8715', 'prod'=>'8719', 'sum'=>'8721', 'minus'=>'8722', 'lowast'=>'8727', 'radic'=>'8730', 'prop'=>'8733', 'infin'=>'8734', 'ang'=>'8736', 'and'=>'8743', 'or'=>'8744', 'cap'=>'8745', 'cup'=>'8746', 'int'=>'8747', 'there4'=>'8756', 'sim'=>'8764', 'cong'=>'8773', 'asymp'=>'8776', 'ne'=>'8800', 'equiv'=>'8801', 'le'=>'8804', 'ge'=>'8805', 'sub'=>'8834', 'sup'=>'8835', 'nsub'=>'8836', 'sube'=>'8838', 'supe'=>'8839', 'oplus'=>'8853', 'otimes'=>'8855', 'perp'=>'8869', 'sdot'=>'8901', 'lceil'=>'8968', 'rceil'=>'8969', 'lfloor'=>'8970', 'rfloor'=>'8971', 'lang'=>'9001', 'rang'=>'9002', 'loz'=>'9674', 'spades'=>'9824', 'clubs'=>'9827', 'hearts'=>'9829', 'diams'=>'9830', 'apos'=>'39', 'OElig'=>'338', 'oelig'=>'339', 'Scaron'=>'352', 'scaron'=>'353', 'Yuml'=>'376', 'circ'=>'710', 'tilde'=>'732', 'ensp'=>'8194', 'emsp'=>'8195', 'thinsp'=>'8201', 'zwnj'=>'8204', 'zwj'=>'8205', 'lrm'=>'8206', 'rlm'=>'8207', 'ndash'=>'8211', 'mdash'=>'8212', 'lsquo'=>'8216', 'rsquo'=>'8217', 'sbquo'=>'8218', 'ldquo'=>'8220', 'rdquo'=>'8221', 'bdquo'=>'8222', 'dagger'=>'8224', 'Dagger'=>'8225', 'permil'=>'8240', 'lsaquo'=>'8249', 'rsaquo'=>'8250', 'euro'=>'8364', 'nbsp'=>'160', 'iexcl'=>'161', 'cent'=>'162', 'pound'=>'163', 'curren'=>'164', 'yen'=>'165', 'brvbar'=>'166', 'sect'=>'167', 'uml'=>'168', 'copy'=>'169', 'ordf'=>'170', 'laquo'=>'171', 'not'=>'172', 'shy'=>'173', 'reg'=>'174', 'macr'=>'175', 'deg'=>'176', 'plusmn'=>'177', 'sup2'=>'178', 'sup3'=>'179', 'acute'=>'180', 'micro'=>'181', 'para'=>'182', 'middot'=>'183', 'cedil'=>'184', 'sup1'=>'185', 'ordm'=>'186', 'raquo'=>'187', 'frac14'=>'188', 'frac12'=>'189', 'frac34'=>'190', 'iquest'=>'191', 'Agrave'=>'192', 'Aacute'=>'193', 'Acirc'=>'194', 'Atilde'=>'195', 'Auml'=>'196', 'Aring'=>'197', 'AElig'=>'198', 'Ccedil'=>'199', 'Egrave'=>'200', 'Eacute'=>'201', 'Ecirc'=>'202', 'Euml'=>'203', 'Igrave'=>'204', 'Iacute'=>'205', 'Icirc'=>'206', 'Iuml'=>'207', 'ETH'=>'208', 'Ntilde'=>'209', 'Ograve'=>'210', 'Oacute'=>'211', 'Ocirc'=>'212', 'Otilde'=>'213', 'Ouml'=>'214', 'times'=>'215', 'Oslash'=>'216', 'Ugrave'=>'217', 'Uacute'=>'218', 'Ucirc'=>'219', 'Uuml'=>'220', 'Yacute'=>'221', 'THORN'=>'222', 'szlig'=>'223', 'agrave'=>'224', 'aacute'=>'225', 'acirc'=>'226', 'atilde'=>'227', 'auml'=>'228', 'aring'=>'229', 'aelig'=>'230', 'ccedil'=>'231', 'egrave'=>'232', 'eacute'=>'233', 'ecirc'=>'234', 'euml'=>'235', 'igrave'=>'236', 'iacute'=>'237', 'icirc'=>'238', 'iuml'=>'239', 'eth'=>'240', 'ntilde'=>'241', 'ograve'=>'242', 'oacute'=>'243', 'ocirc'=>'244', 'otilde'=>'245', 'ouml'=>'246', 'divide'=>'247', 'oslash'=>'248', 'ugrave'=>'249', 'uacute'=>'250', 'ucirc'=>'251', 'uuml'=>'252', 'yacute'=>'253', 'thorn'=>'254', 'yuml'=>'255'); |
||
| 323 | if($t[0] != '#'){ |
||
| 324 | return ($C['and_mark'] ? "\x06" : '&'). (isset($U[$t]) ? $t : (isset($N[$t]) ? (!$C['named_entity'] ? '#'. ($C['hexdec_entity'] > 1 ? 'x'. dechex($N[$t]) : $N[$t]) : $t) : 'amp;'. $t)). ';'; |
||
| 325 | } |
||
| 326 | if(($n = ctype_digit($t = substr($t, 1)) ? intval($t) : hexdec(substr($t, 1))) < 9 or ($n > 13 && $n < 32) or $n == 11 or $n == 12 or ($n > 126 && $n < 160 && $n != 133) or ($n > 55295 && ($n < 57344 or ($n > 64975 && $n < 64992) or $n == 65534 or $n == 65535 or $n > 1114111))){ |
||
| 327 | return ($C['and_mark'] ? "\x06" : '&'). "amp;#{$t};"; |
||
| 328 | } |
||
| 329 | return ($C['and_mark'] ? "\x06" : '&'). '#'. (((ctype_digit($t) && $C['hexdec_entity'] < 2) or !$C['hexdec_entity']) ? $n : 'x'. dechex($n)). ';'; |
||
| 330 | // eof |
||
| 331 | } |
||
| 332 | |||
| 333 | public static function hl_prot($p, $c=null){ |
||
| 334 | // check URL scheme |
||
| 335 | global $C; |
||
| 336 | $b = $a = ''; |
||
| 337 | if($c == null){$c = 'style'; $b = $p[1]; $a = $p[3]; $p = trim($p[2]);} |
||
| 338 | $c = isset($C['schemes'][$c]) ? $C['schemes'][$c] : $C['schemes']['*']; |
||
| 339 | static $d = 'denied:'; |
||
| 340 | if(isset($c['!']) && substr($p, 0, 7) != $d){$p = "$d$p";} |
||
| 341 | if(isset($c['*']) or !strcspn($p, '#?;') or (substr($p, 0, 7) == $d)){return "{$b}{$p}{$a}";} // All ok, frag, query, param |
||
| 342 | if(preg_match('`^([^:?[@!$()*,=/\'\]]+?)(:|&#(58|x3a);|%3a|\\\\0{0,4}3a).`i', $p, $m) && !isset($c[strtolower($m[1])])){ // Denied prot |
||
| 343 | return "{$b}{$d}{$p}{$a}"; |
||
| 344 | } |
||
| 345 | if($C['abs_url']){ |
||
| 346 | if($C['abs_url'] == -1 && strpos($p, $C['base_url']) === 0){ // Make url rel |
||
| 347 | $p = substr($p, strlen($C['base_url'])); |
||
| 348 | }elseif(empty($m[1])){ // Make URL abs |
||
| 349 | if(substr($p, 0, 2) == '//'){$p = substr($C['base_url'], 0, strpos($C['base_url'], ':')+1). $p;} |
||
| 350 | elseif($p[0] == '/'){$p = preg_replace('`(^.+?://[^/]+)(.*)`', '$1', $C['base_url']). $p;} |
||
| 351 | elseif(strcspn($p, './')){$p = $C['base_url']. $p;} |
||
| 352 | else{ |
||
| 353 | preg_match('`^([a-zA-Z\d\-+.]+://[^/]+)(.*)`', $C['base_url'], $m); |
||
| 354 | $p = preg_replace('`(?<=/)\./`', '', $m[2]. $p); |
||
| 355 | while(preg_match('`(?<=/)([^/]{3,}|[^/.]+?|\.[^/.]|[^/.]\.)/\.\./`', $p)){ |
||
| 356 | $p = preg_replace('`(?<=/)([^/]{3,}|[^/.]+?|\.[^/.]|[^/.]\.)/\.\./`', '', $p); |
||
| 357 | } |
||
| 358 | $p = $m[1]. $p; |
||
| 359 | } |
||
| 360 | } |
||
| 361 | } |
||
| 362 | return "{$b}{$p}{$a}"; |
||
| 363 | // eof |
||
| 364 | } |
||
| 365 | |||
| 366 | public static function hl_regex($p){ |
||
| 367 | // ?regex |
||
| 368 | if(empty($p)){return 0;} |
||
| 369 | if($t = ini_get('track_errors')){$o = isset($php_errormsg) ? $php_errormsg : null;} |
||
| 370 | else{ini_set('track_errors', 1);} |
||
| 371 | unset($php_errormsg); |
||
| 372 | if(($d = ini_get('display_errors'))){ini_set('display_errors', 0);} |
||
| 373 | preg_match($p, ''); |
||
| 374 | if($d){ini_set('display_errors', 1);} |
||
| 375 | $r = isset($php_errormsg) ? 0 : 1; |
||
| 376 | if($t){$php_errormsg = isset($o) ? $o : null;} |
||
| 377 | else{ini_set('track_errors', 0);} |
||
| 378 | return $r; |
||
| 379 | // eof |
||
| 380 | } |
||
| 381 | |||
| 382 | public static function hl_spec($t){ |
||
| 383 | // final $spec |
||
| 384 | $s = array(); |
||
| 385 | $t = str_replace(array("\t", "\r", "\n", ' '), '', preg_replace_callback('/"(?>(`.|[^"])*)"/sm', create_function('$m', 'return substr(str_replace(array(";", "|", "~", " ", ",", "/", "(", ")", \'`"\'), array("\x01", "\x02", "\x03", "\x04", "\x05", "\x06", "\x07", "\x08", "\""), $m[0]), 1, -1);'), trim($t))); |
||
| 386 | for($i = count(($t = explode(';', $t))); --$i>=0;){ |
||
| 387 | $w = $t[$i]; |
||
| 388 | if(empty($w) or ($e = strpos($w, '=')) === false or !strlen(($a = substr($w, $e+1)))){continue;} |
||
| 389 | $y = $n = array(); |
||
| 390 | foreach(explode(',', $a) as $v){ |
||
| 391 | if(!preg_match('`^([a-z:\-\*]+)(?:\((.*?)\))?`i', $v, $m)){continue;} |
||
| 392 | if(($x = strtolower($m[1])) == '-*'){$n['*'] = 1; continue;} |
||
| 393 | if($x[0] == '-'){$n[substr($x, 1)] = 1; continue;} |
||
| 394 | if(!isset($m[2])){$y[$x] = 1; continue;} |
||
| 395 | foreach(explode('/', $m[2]) as $m){ |
||
| 396 | if(empty($m) or ($p = strpos($m, '=')) == 0 or $p < 5){$y[$x] = 1; continue;} |
||
| 397 | $y[$x][strtolower(substr($m, 0, $p))] = str_replace(array("\x01", "\x02", "\x03", "\x04", "\x05", "\x06", "\x07", "\x08"), array(";", "|", "~", " ", ",", "/", "(", ")"), substr($m, $p+1)); |
||
| 398 | } |
||
| 399 | View Code Duplication | if(isset($y[$x]['match']) && !htmLawed::hl_regex($y[$x]['match'])){unset($y[$x]['match']);} |
|
| 400 | View Code Duplication | if(isset($y[$x]['nomatch']) && !htmLawed::hl_regex($y[$x]['nomatch'])){unset($y[$x]['nomatch']);} |
|
| 401 | } |
||
| 402 | if(!count($y) && !count($n)){continue;} |
||
| 403 | foreach(explode(',', substr($w, 0, $e)) as $v){ |
||
| 404 | if(!strlen(($v = strtolower($v)))){continue;} |
||
| 405 | if(count($y)){$s[$v] = $y;} |
||
| 406 | if(count($n)){$s[$v]['n'] = $n;} |
||
| 407 | } |
||
| 408 | } |
||
| 409 | return $s; |
||
| 410 | // eof |
||
| 411 | } |
||
| 412 | |||
| 413 | public static function hl_tag($t){ |
||
| 621 | |||
| 622 | public static function hl_tag2(&$e, &$a, $t=1){ |
||
| 623 | // transform tag |
||
| 624 | if($e == 'center'){$e = 'div'; return 'text-align: center;';} |
||
| 625 | if($e == 'dir' or $e == 'menu'){$e = 'ul'; return '';} |
||
| 626 | if($e == 's' or $e == 'strike'){$e = 'span'; return 'text-decoration: line-through;';} |
||
| 627 | if($e == 'u'){$e = 'span'; return 'text-decoration: underline;';} |
||
| 628 | static $fs = array('0'=>'xx-small', '1'=>'xx-small', '2'=>'small', '3'=>'medium', '4'=>'large', '5'=>'x-large', '6'=>'xx-large', '7'=>'300%', '-1'=>'smaller', '-2'=>'60%', '+1'=>'larger', '+2'=>'150%', '+3'=>'200%', '+4'=>'300%'); |
||
| 629 | if($e == 'font'){ |
||
| 630 | $a2 = ''; |
||
| 631 | View Code Duplication | if(preg_match('`face\s*=\s*(\'|")([^=]+?)\\1`i', $a, $m) or preg_match('`face\s*=(\s*)(\S+)`i', $a, $m)){ |
|
| 632 | $a2 .= ' font-family: '. str_replace('"', '\'', trim($m[2])). ';'; |
||
| 633 | } |
||
| 634 | View Code Duplication | if(preg_match('`color\s*=\s*(\'|")?(.+?)(\\1|\s|$)`i', $a, $m)){ |
|
| 635 | $a2 .= ' color: '. str_replace('"', '\'', trim($m[2])). ';'; |
||
| 636 | } |
||
| 637 | if(preg_match('`size\s*=\s*(\'|")?(.+?)(\\1|\s|$)`i', $a, $m) && isset($fs[($m = trim($m[2]))])){ |
||
| 638 | $a2 .= ' font-size: '. str_replace('"', '\'', $fs[$m]). ';'; |
||
| 639 | } |
||
| 640 | $e = 'span'; return ltrim($a2); |
||
| 641 | } |
||
| 642 | if($t == 2){$e = 0; return 0;} |
||
| 643 | return ''; |
||
| 644 | // eof |
||
| 645 | } |
||
| 646 | |||
| 647 | public static function hl_tidy($t, $w, $p){ |
||
| 648 | // Tidy/compact HTM |
||
| 649 | if(strpos(' pre,script,textarea', "$p,")){return $t;} |
||
| 650 | $t = preg_replace('`\s+`', ' ', preg_replace_callback(array('`(<(!\[CDATA\[))(.+?)(\]\]>)`sm', '`(<(!--))(.+?)(-->)`sm', '`(<(pre|script|textarea)[^>]*?>)(.+?)(</\2>)`sm'), create_function('$m', 'return $m[1]. str_replace(array("<", ">", "\n", "\r", "\t", " "), array("\x01", "\x02", "\x03", "\x04", "\x05", "\x07"), $m[3]). $m[4];'), $t)); |
||
| 651 | if(($w = strtolower($w)) == -1){ |
||
| 652 | return str_replace(array("\x01", "\x02", "\x03", "\x04", "\x05", "\x07"), array('<', '>', "\n", "\r", "\t", ' '), $t); |
||
| 653 | } |
||
| 654 | $s = strpos(" $w", 't') ? "\t" : ' '; |
||
| 655 | $s = preg_match('`\d`', $w, $m) ? str_repeat($s, $m[0]) : str_repeat($s, ($s == "\t" ? 1 : 2)); |
||
| 656 | $N = preg_match('`[ts]([1-9])`', $w, $m) ? $m[1] : 0; |
||
| 657 | $a = array('br'=>1); |
||
| 658 | $b = array('button'=>1, 'input'=>1, 'option'=>1, 'param'=>1); |
||
| 659 | $c = array('caption'=>1, 'dd'=>1, 'dt'=>1, 'h1'=>1, 'h2'=>1, 'h3'=>1, 'h4'=>1, 'h5'=>1, 'h6'=>1, 'isindex'=>1, 'label'=>1, 'legend'=>1, 'li'=>1, 'object'=>1, 'p'=>1, 'pre'=>1, 'td'=>1, 'textarea'=>1, 'th'=>1); |
||
| 660 | $d = array('address'=>1, 'blockquote'=>1, 'center'=>1, 'colgroup'=>1, 'dir'=>1, 'div'=>1, 'dl'=>1, 'fieldset'=>1, 'form'=>1, 'hr'=>1, 'iframe'=>1, 'map'=>1, 'menu'=>1, 'noscript'=>1, 'ol'=>1, 'optgroup'=>1, 'rbc'=>1, 'rtc'=>1, 'ruby'=>1, 'script'=>1, 'select'=>1, 'table'=>1, 'tbody'=>1, 'tfoot'=>1, 'thead'=>1, 'tr'=>1, 'ul'=>1); |
||
| 661 | $T = explode('<', $t); |
||
| 662 | $X = 1; |
||
| 663 | while($X){ |
||
| 664 | $n = $N; |
||
| 665 | $t = $T; |
||
| 666 | ob_start(); |
||
| 667 | if(isset($d[$p])){echo str_repeat($s, ++$n);} |
||
| 668 | echo ltrim(array_shift($t)); |
||
| 669 | for($i=-1, $j=count($t); ++$i<$j;){ |
||
| 670 | $r = ''; list($e, $r) = explode('>', $t[$i]); |
||
| 671 | $x = $e[0] == '/' ? 0 : (substr($e, -1) == '/' ? 1 : ($e[0] != '!' ? 2 : -1)); |
||
| 672 | $y = !$x ? ltrim($e, '/') : ($x > 0 ? substr($e, 0, strcspn($e, ' ')) : 0); |
||
| 673 | $e = "<$e>"; |
||
| 674 | if(isset($d[$y])){ |
||
| 675 | if(!$x){ |
||
| 676 | if($n){echo "\n", str_repeat($s, --$n), "$e\n", str_repeat($s, $n);} |
||
| 677 | else{++$N; ob_end_clean(); continue 2;} |
||
| 678 | } |
||
| 679 | else{echo "\n", str_repeat($s, $n), "$e\n", str_repeat($s, ($x != 1 ? ++$n : $n));} |
||
| 680 | echo $r; continue; |
||
| 681 | } |
||
| 682 | $f = "\n". str_repeat($s, $n); |
||
| 683 | if(isset($c[$y])){ |
||
| 684 | if(!$x){echo $e, $f, $r;} |
||
| 685 | else{echo $f, $e, $r;} |
||
| 686 | }elseif(isset($b[$y])){echo $f, $e, $r; |
||
| 687 | }elseif(isset($a[$y])){echo $e, $f, $r; |
||
| 688 | }elseif(!$y){echo $f, $e, $f, $r; |
||
| 689 | }else{echo $e, $r;} |
||
| 690 | } |
||
| 691 | $X = 0; |
||
| 692 | } |
||
| 693 | $t = str_replace(array("\n ", " \n"), "\n", preg_replace('`[\n]\s*?[\n]+`', "\n", ob_get_contents())); |
||
| 694 | ob_end_clean(); |
||
| 695 | if(($l = strpos(" $w", 'r') ? (strpos(" $w", 'n') ? "\r\n" : "\r") : 0)){ |
||
| 696 | $t = str_replace("\n", $l, $t); |
||
| 697 | } |
||
| 698 | return str_replace(array("\x01", "\x02", "\x03", "\x04", "\x05", "\x07"), array('<', '>', "\n", "\r", "\t", ' '), $t); |
||
| 699 | // eof |
||
| 700 | } |
||
| 701 | |||
| 702 | public static function hl_version(){ |
||
| 707 | |||
| 708 | public static function kses($t, $h, $p=array('http', 'https', 'ftp', 'news', 'nntp', 'telnet', 'gopher', 'mailto')){ |
||
| 709 | // kses compat |
||
| 710 | foreach($h as $k=>$v){ |
||
| 711 | $h[$k]['n']['*'] = 1; |
||
| 712 | } |
||
| 713 | $C['cdata'] = $C['comment'] = $C['make_tag_strict'] = $C['no_deprecated_attr'] = $C['unique_ids'] = 0; |
||
| 714 | $C['keep_bad'] = 1; |
||
| 715 | $C['elements'] = count($h) ? strtolower(implode(',', array_keys($h))) : '-*'; |
||
| 716 | $C['hook'] = 'htmLawed::kses_hook'; |
||
| 717 | $C['schemes'] = '*:'. implode(',', $p); |
||
| 718 | return htmLawed::hl($t, $C, $h); |
||
| 719 | // eof |
||
| 720 | } |
||
| 721 | |||
| 722 | public static function kses_hook($t, &$C, &$S){ |
||
| 727 | // end class |
||
| 728 | } |
||
| 729 |
Instead of super-globals, we recommend to explicitly inject the dependencies of your class. This makes your code less dependent on global state and it becomes generally more testable: