如果支持打开远程内容的选项的话，实际上php用fopen或file_get_contents都能获得一个网页的内容，但是默认的函数有个不足的地方就是无法获取http头，这在一些特殊的应用中很不方便，如，有一个链接：

http://www.abc.com/showvd.asp?id=18 

假如它返回的是一个图片，用默认函数就很难识别，但如果通过http应答头来判断就简单多了，此外如果对方通过 refer 来防盗链的话，也是无法获取的，用http类就能完美解决这些问题，而且速度也相差无几。


使用方法：

$hd = new dedehttpdown();
$hd->openurl("http://www.dedecms.com");
echo $hd->gethtml();
//如果保存为文件则用 $hd->savebin("dede.html");
$hd->close();

获得http请求头用
$hd->gethead("key")
设置请求头
$hd->sethead(key,value); (必须在调用 openurl 之前设定)


代码如下:

<?
/* ---------------------------------------------------------------------
//织梦http下载类v1.0
//出自：织梦之旅 http://www.dedecms.com
//作者: it柏拉图
//时间: 2005-11-13 12:39
//声明: 首发在落伍者网站，转载请保留版权信息
--------------------------------------------------------------------- */
class dedehttpdown
{
        var $m_url = "";
        var $m_urlpath = "";
        var $m_scheme = "http";
        var $m_host = "";
        var $m_port = "80";
        var $m_user = "";
        var $m_pass = "";
        var $m_path = "/";
        var $m_query = "";
        var $m_fp = "";
        var $m_error = "";
        var $m_httphead = "" ;
        var $m_html = "";
        var $m_puthead = "";
        var $baseurlpath = "";
        var $homeurl = "";
        var $jumpcount = 0;//防止多重重定向陷入死循环
        //
        //初始化系统
        //
        function privateinit($url)
        {
                        if($url=="") return ;
                        $urls = "";
                        $urls = @parse_url($url);
                        $this->m_url = $url;
            if(is_array($urls))
            {
                          $this->m_host = $urls["host"];
                          if(!empty($urls["scheme"])) $this->m_scheme = $urls["scheme"];

                          if(!empty($urls["user"])){
                                        $this->m_user = $urls["user"];
                          }

                          if(!empty($urls["pass"])){
                                        $this->m_pass = $urls["pass"];
                          }

                          if(!empty($urls["port"])){
                                        $this->m_port = $urls["port"];
                          }

                          if(!empty($urls["path"])) $this->m_path = $urls["path"];
                          $this->m_urlpath = $this->m_path;

                          if(!empty($urls["query"])){
                                        $this->m_query = $urls["query"];
                                        $this->m_urlpath .= "?".$this->m_query;
                          }
                          $this->homeurl = $urls["host"];
                          $this->baseurlpath = $this->homeurl.$urls["path"];
                          $this->baseurlpath = ereg_replace("/([^/]*)\.(.*)$","/",$this->baseurlpath);
                          $this->baseurlpath = ereg_replace("/$","",$this->baseurlpath);
                 }
        }
        //
        //打开指定网址
        //
        function openurl($url)
        {
                //重设各参数
                $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 = "";
                $this->jumpcount = 0;
                $this->m_httphead = array() ;
                //$this->m_puthead = "";
                $this->m_html = "";
                $this->close();
                //初始化系统
                $this->privateinit($url);
                $this->privatestartsession();
        }
        //
        //打开303重定向网址
        //
        function jumpopenurl($url)
        {
                //重设各参数
                $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 = "";
                $this->jumpcount++;
                $this->m_httphead = array() ;
                $this->m_html = "";
                $this->close();
                //初始化系统
                $this->privateinit($url);
                $this->privatestartsession();
        }
        //
        //获得某操作错误的原因
        //
        function printerror()
        {
                echo "错误信息：".$this->m_error;
                echo "具体返回头：<br>";
                foreach($this->m_httphead as $k=>$v)
                { echo "$k => $v <br>\r\n"; }
        }
        //
        //判别用get方法发送的头的应答结果是否正确
        //
        function isgetok()
        {
                if( ereg("^2",$this->gethead("http-state")) )
                {        return true; }
                else
                {
                        $this->m_error .= $this->gethead("http-state")." - ".$this->gethead("http-describe")."<br>";
                        return false;
                }
        }
        //
        //看看返回的网页是否是text类型
        //
        function istext()
        {
                if(ereg("^2",$this->gethead("http-state"))
                        && eregi("^text",$this->gethead("content-type")))
                {        return true; }
                else
                {
                        $this->m_error .= "内容为非文本类型或网址重定向<br>";
                        return false;
                }
        }
        //
        //判断返回的网页是否是特定的类型
        //
        function iscontenttype($ctype)
        {
                if(ereg("^2",$this->gethead("http-state"))
                        && $this->gethead("content-type")==strtolower($ctype))
                {        return true; }
                else
                {
                        $this->m_error .= "类型不对 ".$this->gethead("content-type")."<br>";
                        return false;
                }
        }
        //
        //用http协议下载文件
        //
        function savetobin($savefilename)
        {
                if(!$this->isgetok()) return false;
                if(@feof($this->m_fp)) { $this->m_error = "连接已经关闭！"; return false; }
                $fp = fopen($savefilename,"w");
                while(!feof($this->m_fp)){
                        fwrite($fp,fread($this->m_fp,1024));
                }
                fclose($this->m_fp);

                fclose($fp);
                return true;
        }
        //
        //保存网页内容为text文件
        //
        function savetotext($savefilename)
        {
                if($this->istext()) $this->savebinfile($savefilename);
                else return "";
        }
        //
        //用http协议获得一个网页的内容
        //
        function gethtml()
        {
                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);
                }
                @fclose($this->m_fp);
                return $this->m_html;
        }
        //
        //开始http会话
        //
        function privatestartsession()
        {

                if(!$this->privateopenhost()){
                        $this->m_error .= "打开远程主机出错!";
                        return false;
                }

                if($this->gethead("http-edition")=="http/1.1") $httpv = "http/1.1";
                else $httpv = "http/1.0";

                //发送固定的起始请求头get、host信息
                fputs($this->m_fp,"get ".$this->m_urlpath." $httpv\r\n");
                $this->m_puthead["host"] = $this->m_host;

                //发送用户自定义的请求头
                if(!isset($this->m_puthead["accept"])) { $this->m_puthead["accept"] = "*/*"; }
                if(!isset($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"]; }
                foreach($this->m_puthead as $k=>$v){
                        $k = trim($k);
                        $v = trim($v);
                        if($k!=""&&$v!=""){
                                fputs($this->m_fp,"$k: $v\r\n");
                        }
                }

                //发送固定的结束请求头
                //http1.1协议必须指定文档结束后关闭链接,否则读取文档时无法使用feof判断结束
                if($httpv=="http/1.1") fputs($this->m_fp,"connection: close\r\n\r\n");
                else fputs($this->m_fp,"\r\n");

                //获取应答头状态信息
                $httpstas = explode(" ",fgets($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;$i<count($httpstas);$i++){
                        $this->m_httphead["http-describe"] .= " ".trim($httpstas[$i]);
                }
                //获取详细应答头
                while(!feof($this->m_fp)){
                        $line = trim(fgets($this->m_fp,256));
                        if($line == "") break;
                        $hkey = "";
                        $hvalue = "";
                        $v = 0;
                        for($i=0;$i<strlen($line);$i++){
                                if($v==1) $hvalue .= $line[$i];
                                if($line[$i]==":") $v = 1;
                                if($v==0) $hkey .= $line[$i];
                        }
                        $hkey = trim($hkey);
                        if($hkey!="") $this->m_httphead[strtolower($hkey)] = trim($hvalue);
                }
                //判断是否是3xx开头的应答
                if(ereg("^3",$this->m_httphead["http-state"]))
                {
                        if($this->jumpcount > 3) return;
                        if(isset($this->m_httphead["location"])){
                                $newurl = $this->m_httphead["location"];
                                if(eregi("^http",$newurl)){
                                        $this->jumpopenurl($newurl);
                                }
                                else{
                                        $newurl = $this->fillurl($newurl);
                                        $this->jumpopenurl($newurl);
                                }
                        }
                        else
                        {        $this->m_error = "无法识别的转移应答！"; }
                }//
        }
        //
        //获得一个http头的值
        //
        function gethead($headname)
        {
                $headname = strtolower($headname);
                if(isset($this->m_httphead[$headname]))
                        return $this->m_httphead[$headname];
                else
                        return "";
        }
        //
        //设置http头的值
        //
        function sethead($skey,$svalue)
        {
                $this->m_puthead[$skey] = $svalue;
        }
        //
        //打开连接
        //
        function privateopenhost()
        {
                if($this->m_host=="") return false;
                $this->m_fp = @fsockopen($this->m_host, $this->m_port, &$errno, &$errstr,10);
                if(!$this->m_fp){
                        $this->m_error = $errstr;
                        return false;
                }
                else{
                        return true;
                }
        }
        //
        //关闭连接
        //
        function close(){
                @fclose($this->m_fp);
        }
        //
        //补全相对网址
        //
        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($surl)<=2) return "";
      else if($surl[0]=="/")
      {
              $okurl = "http://".$this->baseurlpath."/".substr($surl,2,strlen($surl)-2);
            }
      else{
        $urls = explode("/",$surl);
        foreach($urls as $u){
          if($u=="..") $pathstep++;
          else if($i<count($urls)-1) $dstr .= $urls[$i]."/";
          else $dstr .= $urls[$i];
          $i++;
        }
        $urls = explode("/",$this->baseurlpath);
        if(count($urls) <= $pathstep)
                return "";
        else{
          $pstr = "http://";
          for($i=0;$i<count($urls)-$pathstep;$i++)
          { $pstr .= $urls[$i]."/"; }
          $okurl = $pstr.$dstr;
        }
      }
    }
    else
    {
      if(strlen($surl)<7)
        $okurl = "http://".$this->baseurlpath."/".$surl;
      else if(strtolower(substr($surl,0,7))=="http://")
        $okurl = $surl;
      else
        $okurl = "http://".$this->baseurlpath."/".$surl;
    }
    $okurl = eregi_replace("^(http://)","",$okurl);
    $okurl = eregi_replace("/{1,}","/",$okurl);
    return "http://".$okurl;
  }
}
?>
申明：本教程内容由威凡网编辑整理并提供IT程序员分享学习，如文中有侵权行为，请与站长联系(QQ:254677821)！