使用PHP下载文件

使用PHP脚本来下载文件，无非是通过两种方式，一种是使用system、exec等即有的函数调用系统自带的下载工具，比如 wget 之类的来下载文件，还有一种是使用php本身利用Socket来下载文件，我选择第二种方式。

使用Socket下载文件，首先如果是http协议的文件，必须明白HTTP协议的运行过程，如果是FTP协议的则要了解ftp协议运行过程，比较繁琐。比如HTTP协议访问一个文件的代码：（来自手册）

<?php
$fp = fsockopen("www.example.com", 80, $errno, $errstr, 30);
if (!$fp) {
echo "$errstr ($errno)<br />\n";
} else {
$out = "GET / HTTP/1.1\r\n";
$out .= "Host: www.example.com\r\n";
$out .= "Connection: Close\r\n\r\n";

fwrite($fp, $out);
while (!feof($fp)) {
echo fgets($fp, 128);
}
fclose($fp);
}
?>

我们为了简单起见，使用fopen直接访问远程文件来达到目的，同事又能够访问http，也能访问ftp，比较合适。当然，如果按照上面的思路来说，也可以使用ftp的函数库来实现。

我们使用fopen函数来完成我们的工作，实现了如下代码：

#! /usr/bin/php
<?php
error_reporting(0);
set_time_limit(0);

//无参数则给出提示
if ($argc < 2){
echo "Usage: ". $argv[0] ." URL [Destination]\n\n";
exit();
}

//设置获取基本变量
$url = $argv[1];
$save_path = $argv[2] ? $argv[2] : "./";
$file_name = array_pop(explode("/", $url));
$localfile = $save_path . $file_name;

//检查变量
if (!check_url($url)){
exit("Error: URL ". $url ." invalid.\n\n");
}
if (file_exists($localfile)){
exit("Error: local file ". $localfile ." exists.\n\n");
}

//打开远程文件
$fp = fopen($url, "rb");
if (!$fp){
exit("Error: Download ". $url ." failed.\n\n");
}

//打开本地文件
$sp = fopen($localfile, "wb");
if (!$sp){
exit("Error: Open local file ". $localfile ." failed.\n\n");
}

//下载远程文件
echo "Downloading, please waiting...\n\n";
while (!feof($fp)){
$tmpfile .= fread($fp, 1024);
}

//保存文件到本地
fwrite($sp, $tmpfile);
fclose($fp);
fclose($sp);
echo "Download file ". $file_name ." succeed!\n\n";

/* 检查URL合法性函数 */
function check_url($url){
return preg_match("/^(http|ftp)(:\/\/)([a-zA-Z0-9-_]+[\.\/]+[\w\-_\/]+.*)+$/i", $url);
}

?>

我们把以上代码保存为 download.php 文件，在Linux/Unix下记得要加上可执行属性：
chmod +x download.php

另外，PHP脚本引擎的路径必须是 /usr/bin/php ，如果不是，请自行修改第一行为实际的PHP引擎路径，比如：
#! /usr/local/php/bin/php

使用上面的脚本来下载文件：
download.php 远程文件 保存路径

如把Google Talk程序下载到我们的 /tmp 目录下：
download.php http://dl.google.com/googletalk/googletalk-setup.exe /tmp/

如果不出错，等待一会就能够在 /tmp/ 目录下看到 googletalk-setup.exe 文件。

能够改进的就是支持更多协议、需要验证的能够输入用户名密码、有下载进度条。至于断点续传和多线程对于PHP来说还不太现实，有兴趣的可以自己加深一步。

PS: 我另外发现一个更强的HTTP下载类，是dedeCMS的作者IT柏拉图写的:

<?
/*=======================================
// 织梦Http下载类
// 织梦之旅 www.dedecms.com
=======================================*/
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 $m_html = "";
//
//初始化系统
//
function PrivateInit($url)
{
$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;
}
}
}
//
//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->m_httphead = "" ;
$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
{
$this->m_error .= "内容为非文本类型<br>";
return false;
}
}
//
//判断返回的网页是否是特定的类型
//
function IsContentType($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") or die("写入文件 $savefilename 失败！");
while(!feof($this->m_fp)){
@fwrite($fp,fgets($this->m_fp,256));
}
@fclose($this->m_fp);
return true;
}
//
//保存网页内容为Text文件
//
function SaveToText($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 $httpv = "HTTP/1.0";
fputs($this->m_fp,"GET ".$this->m_urlpath." $httpv\r\n");
fputs($this->m_fp,"Host: ".$this->m_host."\r\n");
fputs($this->m_fp,"Accept: */*\r\n");
fputs($this->m_fp,"User-Agent: Mozilla/4.0+(compatible;+MSIE+6.0;+Windows+NT+5.2)\r\n");
//HTTP1.1协议必须指定文档结束后关闭链接,否则读取文档时无法使用feof判断结束
if($httpv=="HTTP/1.1") fputs($this->m_fp,"Connection: Close\r\n\r\n");
else fputs($this->m_fp,"\r\n");
$httpstas = fgets($this->m_fp,256);
$httpstas = split(" ",$httpstas);
$this->m_httphead["http-edition"] = trim($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[$i]);
}
while(!feof($this->m_fp)){
$line = str_replace("\"","",trim(fgets($this->m_fp,256)));
if($line == "") break;
if(ereg(":",$line)){
$lines = split(":",$line);
$this->m_httphead[strtolower(trim($lines[0]))] = trim($lines[1]);
}
}
}
//
//获得一个Http头的值
//
function GetHead($headname)
{
$headname = strtolower($headname);
if(isset($this->m_httphead[$headname]))
return $this->m_httphead[$headname];
else
return "";
}
//
//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);
}
}

?>

----------------------------------------------------------------------------------

这个类的使用方法：

下载网页
<?
$httpdown = new DedeHttpDown();
$httpdown->OpenUrl("http://www.dedecms.com");
echo $httpdown->GetHtml();
$httpdown->Close();
?>
如果下载图片并保存,可以用
<?
$httpdown = new DedeHttpDown();
$httpdown->OpenUrl("http://prato.bokele.com/0/0/399/bGluMi5qcGc=.jpg");
echo $httpdown->SaveBin("test.jpg");
$httpdown->Close();
echo "<img src='test.jpg'>";
?>