]*)>(.*?)<\/title>!is', $content, $buff); return trim($buff[2]); } /** * @brief Extract header script **/ function getHeadScript($content) { // remove the title tag $content = preg_replace('!]*)>(.*?)<\/title>!is','', $content); // Remove meta tags $content = preg_replace('!<(\/){0,1}meta([^>]*)>!is','', $content); // Extract information such as ]*)>!is', $content, $link_buff); for($i=0;$i!is', $content, $script_buff); for($i=0;$i]*)>(.*?)<\/body>!is', $content, $body_buff); $body_script = $body_buff[2]; // Remove link, style, script, etc. $body_script = preg_replace('!]*)>!is', '', $body_script); $body_script = preg_replace('!<(style|script)(.*?)<\/(style|script)>!is', '', $body_script); return $body_script; } /** * @brief Change the value of src, href in the content **/ function replaceSrc($content, $path) { $url_info = parse_url($path); $host = sprintf("%s://%s%s",$url_info['scheme'],$url_info['host'],$url_info['port']?':'.$url_info['port']:''); $this->host = $host.'/'; $path = $url_info['path']; if(substr($path,-1)=='/') $path = substr($path,-1); $t = explode('/',$path); $_t = array(); for($i=0,$c=count($t)-1;$i<$c;$i++) { $v = trim($t[$i]); if(!$v) continue; $_t[] = $v; } $path = $host.'/'.implode('/',$_t); if(substr($path,-1)!='/') $path .= '/'; $this->path = $path; $content = preg_replace_callback('/(src=|href=|url\()("|\')?([^"\'\)]+)("|\'\))?/is',array($this,'_replacePath'),$content); return $content; } function _replacePath($matches) { $val = trim($matches[3]); if(preg_match('/^(http|https|ftp|telnet|mms|mailto)/i',$val)) return $matches[0]; if(substr($val,0,2)=='./') { $path = $this->path.substr($val,2); } elseif(substr($val,0,1)=='/') { $path = $this->host.substr($val,1); } else { $path = $this->path.$val; } return sprintf("%s%s%s%s", $matches[1], $matches[2], $path, $matches[4]); } } ?>