发布一个用PHP fsockopen写的HTTP下载的类


Posted in PHP onFebruary 22, 2007

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;
  }
}
?>

PHP 相关文章推荐
使用php来实现网络服务
Sep 15 PHP
php绝对路径与相对路径之间关系的的分析
Mar 03 PHP
让PHP COOKIE立即生效,不用刷新就可以使用
Mar 09 PHP
fleaphp rolesNameField bug解决方法
Apr 23 PHP
php 文件缓存函数
Oct 08 PHP
PHP学习笔记之字符串编码的转换和判断
May 22 PHP
PHP通过内置函数memory_get_usage()获取内存使用情况
Nov 20 PHP
PHP中strpos、strstr和stripos、stristr函数分析
Jun 11 PHP
PHP 年月日的三级联动实例代码
May 24 PHP
关于Yii中模型场景的一些简单介绍
Sep 22 PHP
PHP大文件及断点续传下载实现代码
Aug 18 PHP
如何用PHP实现分布算法之一致性哈希算法
May 26 PHP
隐藏X-Space个人空间下方版权方法隐藏X-Space个人空间标题隐藏X-Space个人空间管理版权方法
Feb 22 #PHP
excellent!――ASCII Art(由目标图象生成ascii)
Feb 20 #PHP
珊瑚虫IP库浅析
Feb 15 #PHP
PHP中HTTP方式下的Gzip压缩传输方法举偶
Feb 15 #PHP
PHP+.htaccess实现全站静态HTML文件GZIP压缩传输(一)
Feb 15 #PHP
php调用mysql存储过程
Feb 14 #PHP
mysql中存储过程、函数的一些问题
Feb 14 #PHP
You might like
phpcms手机内容页面添加上一篇和下一篇
2015/06/05 PHP
如何在旧的PHP系统中使用PHP 5.3之后的库
2015/12/02 PHP
PHP实现移除数组中为空或为某值元素的方法
2017/01/07 PHP
深入理解PHP中mt_rand()随机数的安全
2017/10/12 PHP
js 文本滚动效果的实例代码
2013/08/17 Javascript
js获取对象为null的解决方法
2013/11/21 Javascript
jQuery实现tab标签自动切换的方法
2015/02/28 Javascript
js中substr,substring,indexOf,lastIndexOf,split,replace的用法详解
2015/11/09 Javascript
Node.js的Web模板引擎ejs的入门使用教程
2016/06/06 Javascript
动态JavaScript所造成一些你不知道的危害
2016/09/25 Javascript
AngularJS extend用法详解及实例代码
2016/11/15 Javascript
12306 刷票脚本及稳固刷票脚本(防挂)
2017/01/04 Javascript
Angular排序实例详解
2017/06/28 Javascript
vue与TypeScript集成配置最简教程(推荐)
2017/10/17 Javascript
vue.js input框之间赋值方法
2018/08/24 Javascript
vue自定义键盘信息、监听数据变化的方法示例【基于vm.$watch】
2019/03/16 Javascript
微信小程序生成海报分享朋友圈的实现方法
2019/05/06 Javascript
vue项目打包之开发环境和部署环境的实现
2020/04/23 Javascript
实例分析javascript中的异步
2020/06/02 Javascript
详解python之多进程和进程池(Processing库)
2017/06/09 Python
简单的python协同过滤程序实例代码
2018/01/31 Python
解决Python selenium get页面很慢时的问题
2019/01/30 Python
Python新建项目自动添加介绍和utf-8编码的方法
2020/12/26 Python
CSS3不透明度实例讲解
2016/04/26 HTML / CSS
canvas学习笔记之绘制简单路径
2019/01/28 HTML / CSS
英国最大的海报商店:GB Posters
2018/03/20 全球购物
软件工程师面试题
2012/06/25 面试题
历史专业个人求职信范文
2013/12/07 职场文书
学习张林森心得体会
2014/09/10 职场文书
会计人员演讲稿
2014/09/11 职场文书
工作试用期自我评价
2015/03/10 职场文书
一个都不能少观后感
2015/06/04 职场文书
最新农村养殖致富:资金投入较低的创业项目有哪些?
2019/09/26 职场文书
详解Redis主从复制实践
2021/05/19 Redis
MySQL 亿级数据导入导出及迁移笔记
2021/06/18 MySQL
go goroutine 怎样进行错误处理
2021/07/16 Golang