1）新增遠程主機判斷，節約伺服器資源。避免遠程主機不存在的時候仍舊fsockopen，導致的死機佔用伺服器CPU （2）新增響應401的判斷和支援。 （3）增加對json返迴文本的支援。 （4）日誌的產生，如果設定了DEBUG_LEVEL且為TRUE則會對每次遠程下載做日誌。 （5）擷取位元組的限定dataLimit，節約伺服器資源。（6）修改日期：2013-1-17 如果還有更好的方法建議，可以隨時聯絡我本人。admin@zbphp.com [php] m_url = $url; if(is_array($urls)) { $this->m_host = $urls["host"]; if(!emptyempty($urls["scheme"])) { $this->m_scheme = $urls["scheme"]; } if(!emptyempty($urls["user"])) { $this->m_user = $urls["user"]; } if(!emptyempty($urls["pass"])) { $this->m_pass = $urls["pass"]; } if(!emptyempty($urls["port"])) { $this->m_port = $urls["port"]; } if(!emptyempty($urls["path"])) { $this->m_path = $urls["path"]; } $this->m_urlpath = $this->m_path; if(!emptyempty($urls["query"])) { $this->m_query = $urls["query"]; $this->m_urlpath .= "?".$this->m_query; } $this->HomeUrl = $urls["host"]; $this->BaseUrlPath = $this->HomeUrl.$urls["path"]; $this->BaseUrlPath = preg_replace("/\/([^\/]*)\.(.*)$/","/",$this->BaseUrlPath); $this->BaseUrlPath = preg_replace("/\/$/","",$this->BaseUrlPath); } } /** * 重設各參數 * * @access public * @return void */ function ResetAny() { $this->m_url = ""; $this->m_urlpath = ""; $this->m_scheme = "http"; $this->m_host = ""; $this->m_port = "80"; $this->m_user = ""; $this->m_pass = ""; $this->m_path = "/"; $this->m_query = ""; $this->m_error = ""; } /** * 開啟指定網址 * * @access public * @param string $url 地址 * @param string $requestType 請求類型 * @return string */ function OpenUrl($url,$requestType="GET") { $this->ResetAny(); $this->JumpCount = 0; $this->m_httphead = Array() ; $this->m_html = ''; $this->dataLimit = 0; $this->reTry = 0; $this->Close(); //初始化系統 $this->PrivateInit($url); $this->PrivateStartSession($requestType); } /** * 轉到303重新導向網址 * * @access public * @param string $url 地址 * @return string */ function JumpOpenUrl($url) { $this->ResetAny(); $this->JumpCount++; $this->m_httphead = Array() ; $this->m_html = ""; $this->Close(); //初始化系統 $this->PrivateInit($url); $this->PrivateStartSession('GET'); } /** * 獲得某操作錯誤的原因 * * @access public * @return void */ function printError() { echo "錯誤資訊：".$this->m_error; echo "
具體返回頭：
"; foreach($this->m_httphead as $k=>$v){ echo "$k => $v
\r\n"; } } /** * 判別用Get方法發送的頭的應答結果是否正確 * * @access public * @return bool */ function IsGetOK() { if( preg_match("/^2/",$this->GetHead("http-state")) ) { return TRUE; } else { $this->m_error .= $this->GetHead("http-state")." - ".$this->GetHead("http-describe")."
"; return FALSE; } } /** * 看看返回的網頁是否是text類型 * * @access public * @return bool */ function IsText() { if( preg_match("/^(2|401)/",$this->GetHead("http-state")) && preg_match("/text|xml|json/i",$this->GetHead("content-type")) ) { return TRUE; } else { $this->m_error .= "內容為非文本類型或網址重新導向
"; return FALSE; } } /** * 判斷返回的網頁是否是特定的類型 * * @access public * @param string $ctype 內容類型 * @return string */ function IsContentType($ctype) { if(preg_match("/^2/",$this->GetHead("http-state")) && $this->GetHead("content-type")==strtolower($ctype)) { return TRUE; } else { $this->m_error .= "類型不對 ".$this->GetHead("content-type")."
"; return FALSE; } } /** * 用Http協議下載檔案 * * @access public * @param string $savefilename 儲存檔案名稱 * @return string */ function SaveToBin($savefilename) { if(!$this->IsGetOK()) { return FALSE; } if(@feof($this->m_fp)) { $this->m_error = "串連已經關閉！"; return FALSE; } $fp = fopen($savefilename,"w"); while(!feof($this->m_fp)) { fwrite($fp, fread($this->m_fp, 1024)); } fclose($this->m_fp); fclose($fp); return TRUE; } /** * 儲存網頁內容為Text檔案 * * @access public * @param string $savefilename 儲存檔案名稱 * @return string */ function SaveToText($savefilename) { if($this->IsText()) { $this->SaveBinFile($savefilename); } else { return ""; } } /** * 用Http協議獲得一個網頁的內容 * * @access public * @return string */ function GetHtml() { $tm1 = microtime(true); if(!$this->IsText()) { return ''; } if($this->m_html!='') { return $this->m_html; } if(!$this->m_fp||@feof($this->m_fp)) { return ''; } while(!feof($this->m_fp)) { $this->m_html .= fgets($this->m_fp,256); if($this->dataLimit > 0 && strlen($this->m_html) > $this->dataLimit) break; } @fclose($this->m_fp); $tm2 = microtime(true); $log = "\ntm2-tm1 = ".($tm2-$tm1); $log.= "\n".$this->m_html; $this->log_write('GetHtml',$log); return $this->m_html; } /** * 開始HTTP會話 * * @access public * @param string $requestType 請求類型 * @return string */ function PrivateStartSession($requestType="GET") { if(!$this->PrivateOpenHost()) { $this->m_error .= "開啟遠程主機出錯!"; return FALSE; } $this->reTry++; if($this->GetHead("http-edition")=="HTTP/1.1") { $httpv = "HTTP/1.1"; } else { $httpv = "HTTP/1.0"; } $ps = explode('?',$this->m_urlpath); $headString = ''; //發送固定的起始要求標頭GET、Host資訊 if($requestType=="GET") { $headString .= "GET ".$this->m_urlpath." $httpv\r\n"; } else { $headString .= "POST ".$ps[0]." $httpv\r\n"; } if($this->m_user || $this->m_pass) { $headString .= "Authorization: Basic ".base64_encode($this->m_user.":".$this->m_pass)."\r\n"; } $this->m_puthead["Host"] = $this->m_host; //發送使用者自訂的要求標頭 if(!isset($this->m_puthead["User-Agent"])) { $this->m_puthead["User-Agent"] = "Mozilla/4.0 (compatible; MSIE 6.0; Windows NT 5.2)"; } if(!isset($this->m_puthead["Refer"])) { $this->m_puthead["Refer"] = "http://".$this->m_puthead["Host"]; } /* add on 2012-12-19 */ $headString.="Connection:keep-alive\r\n"; $headString.="Accept-Language:zh-cn,zh;q=0.8,en-us;q=0.5,en;q=0.3\r\n"; $headString.="Accept:text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8\r\n"; foreach($this->m_puthead as $k=>$v) { $k = trim($k); $v = trim($v); if($k!=""&&$v!="") { $headString .= "$k: $v\r\n"; } } fputs($this->m_fp, $headString); if($requestType=="POST") { $postdata = ""; if(count($ps)>1) { for($i=1;$im_fp,"Content-Type: application/x-www-form-urlencoded\r\n"); fputs($this->m_fp,"Content-Length: $plen\r\n"); } //發送固定的結束要求標頭 //HTTP1.1協議必須指定文檔結束後關閉連結,否則讀取文檔時無法使用feof判斷結束 if($httpv=="HTTP/1.1") { fputs($this->m_fp,"Connection: Close\r\n\r\n"); } else { fputs($this->m_fp,"\r\n"); } if($requestType=="POST") { fputs($this->m_fp,$postdata); } //擷取應答頭狀態資訊 $httpstas = explode(" ",fgets($this->m_fp,256)); $this->m_httphead["http-edition"] = trim($httpstas[0]); $this->m_httphead["http-state"] = trim($httpstas[1]); $this->m_httphead["http-describe"] = ""; for($i=2;$im_httphead["http-describe"] .= " ".trim($httpstas[$i]); } //擷取詳細應答頭 while(!feof($this->m_fp)) { $line = trim(fgets($this->m_fp,256)); if($line == "") { break; } $hkey = ""; $hvalue = ""; $v = 0; for($i=0;$im_httphead[strtolower($hkey)] = trim($hvalue); } } //如果串連被不正常關閉，重試 if(feof($this->m_fp)) { if($this->reTry > 10) { return FALSE; } $this->PrivateStartSession($requestType); } //判斷是否是3xx開頭的應答 if(preg_match("/^3/",$this->m_httphead["http-state"])) { if($this->JumpCount > 3) { return; } if(isset($this->m_httphead["location"])) { $newurl = $this->m_httphead["location"]; if(preg_match("/^http/i",$newurl)) { $this->JumpOpenUrl($newurl); } else { $newurl = $this->FillUrl($newurl); $this->JumpOpenUrl($newurl); } } else { $this->m_error = "無法識別的回覆！"; } } } /** * 獲得一個Http頭的值 * * @access public * @param string $headname 標頭檔名稱 * @return string */ function GetHead($headname) { $headname = strtolower($headname); return isset($this->m_httphead[$headname]) ? $this->m_httphead[$headname] : ''; } /** * 設定Http頭的值 * * @access public * @param string $skey 鍵 * @param string $svalue 值 * @return string */ function SetHead($skey,$svalue) { $this->m_puthead[$skey] = $svalue; } /** * 開啟串連 * * @access public * @return bool */ function PrivateOpenHost() { if($this->m_host=="") { return FALSE; } if(function_exists('checkdnsrr') && !checkdnsrr($this->m_host,'A') && !checkdnsrr($this->m_host,'CNAME')) { $this->m_error = '遠程主機'.$this->m_host.'不存在!checkdnsrr !'; return FALSE; } $errno = ""; $errstr = ""; $this->m_fp = @fsockopen($this->m_host, $this->m_port, $errno, $errstr,10); if(!$this->m_fp) { $this->m_error = $errstr; return FALSE; } else { return TRUE; } } /** * 關閉串連 * * @access public * @return void */ function Close() { @fclose($this->m_fp); } /** * 補全相對網址 * * @access public * @param string $surl 需要不全的地址 * @return string */ function FillUrl($surl) { $i = 0; $dstr = ""; $pstr = ""; $okurl = ""; $pathStep = 0; $surl = trim($surl); if($surl=="") { return ""; } $pos = strpos($surl,"#"); if($pos>0) { $surl = substr($surl,0,$pos); } if($surl[0]=="/") { $okurl = "http://".$this->HomeUrl.$surl; } else if($surl[0]==".") { if(strlen($surl)<=1) { return ""; } else if($surl[1]=="/") { $okurl = "http://".$this->BaseUrlPath."/".substr($surl,2,strlen($surl)-2); } else { $urls = explode("/",$surl); foreach($urls as $u) { if($u=="..") { $pathStep++; } else if($iBaseUrlPath); if(count($urls) <= $pathStep) { return ""; } else { $pstr = "http://"; for($i=0;$iBaseUrlPath."/".$surl; } else if(strtolower(substr($surl,0,7))=="http://") { $okurl = $surl; } else { $okurl = "http://".$this->BaseUrlPath."/".$surl; } } $okurl = preg_replace("/^(http:\/\/)/i","",$okurl); $okurl = preg_replace("/\/{1,}/", "/", $okurl); return "http://".$okurl; } function log_write($funcname,$message) { if(!(defined('DEBUG_LEVEL') && DEBUG_LEVEL == TRUE)) return ; $log = "\n".date("Y-M-d H:i:s ").get_current_user()."[".getmypid()."]"; $log.= "\n".$this->m_url."\n".str_repeat('------', 10)."\n".$message; $path = $funcname.' '.date('Y m d H i s ').preg_replace('/([\W]+|\s+)/i', ' ', $this->m_url); if(strlen($path) > 250) $path = substr($path,0,250); $dir = DEDEDATA.'/httpdownlog'; if(!is_dir($dir) && !mkdir($dir)) exit('Can not make dir '.$dir); $path = $dir.'/'.$path; if(!file_exists($path)) touch($path); $fp = fopen($path,'a+'); flock($fp, LOCK_EX); fputs($fp, "PATH:".$path."\nREAL:".realpath($path)."\nMSSG:".$log); fclose($fp); return TRUE; } }//End Class 使用方法： [php] $dhd = new DeDeHttpDown(); $dhd->OpenUrl($Rs['wurl']); $dhd->dataLimit = 5120; $dhd->m_puthead["Refer"] = $Rs['wurl']; $filecnt = trim($dhd->GetHtml()); 存在未解決的問題： （1）假如網域名稱是綁定了A記錄和CNAME，有ip指向，但是IP地址是不存在的或者虛假的，程式仍舊會繼續擷取。 （2）PHP的fsockopen 裡面的time out貌似根本就沒有起作用。設定了10s逾時，但是根本就是執行到程式time out 為止。 如果好的方法建議，可以隨時聯絡我本人。admin@zbphp.com
http://www.bkjia.com/PHPjc/477811.htmlwww.bkjia.comtruehttp://www.bkjia.com/PHPjc/477811.htmlTechArticle1）新增遠程主機判斷，節約伺服器資源。避免遠程主機不存在的時候仍舊fsockopen，導致的死機佔用伺服器CPU （2）新增響應401的判斷和支援...
