php分享十八：网页抓取

function grabImage($url, $filename = ‘‘) { 
if($url == ‘‘) { 
return false; //如果 $url 为空则返回 false; 
} 
$ext_name = strrchr($url, ‘.‘); //获取图片的扩展名 
if($ext_name != ‘.gif‘ && $ext_name != ‘.jpg‘ && $ext_name != ‘.bmp‘ && $ext_name != ‘.png‘) { 
return false; //格式不在允许的范围 
} 
if($filename == ‘‘) { 
$filename = time().$ext_name; //以时间戳另起名 
} 
//开始捕获 
ob_start(); 
readfile($url); 
$img_data = ob_get_contents(); 
ob_end_clean(); 
$size = strlen($img_data); 
$local_file = fopen($filename , ‘a‘); 
fwrite($local_file, $img_data); 
fclose($local_file); 
return $filename; 
} 

$a = "http://jb51.net/content/uploadfile/201106/thum-f3ccdd27d2000e3f9255a7e3e2c4880020110622095243.jpg"; 
$local = ‘socket1.gif‘; 
$aa = getImg($a,$local); 
/* 
*@ 完整的图片地址 
*@ 要存储的文件名 
*/ 
function getImg( $url = "", $filename = "" ) { 
if(is_dir(basename($filename))) { 
echo "The Dir was not exits"; 
Return false; 
} 
//去除URL连接上面可能的引号 
$url = preg_replace( ‘/(?:^[\‘"]+|[\‘"\/]+$)/‘, ‘‘, $url ); 
if (!extension_loaded(‘sockets‘)) return false; 
//获取url各相关信息 
preg_match( ‘/http:\/\/([^\/\:]+(\:\d{1,5})?)(.*)/i‘, $url, $matches ); 
if (!$matches) return false; 
$sock = socket_create( AF_INET, SOCK_STREAM, SOL_TCP ); 
if ( !@socket_connect( $sock, $matches[1], $matches[2] ? substr($matches[2], 1 ) : 80 ) ) { 
return false; 
} 
//图片的相对地址 
$msg = ‘GET ‘ . $matches[3] . " HTTP/1.1\r\n"; 
//主机名称 
$msg .= ‘Host: ‘ . $matches[1] . "\r\n"; 
$msg .= ‘Connection: Close‘ . "\r\n\r\n"; 
socket_write( $sock, $msg ); 
$bin = ‘‘; 
while ( $tmp = socket_read( $sock, 10 ) ) { 
$bin .= $tmp; 
$tmp = ‘‘; 
} 
$bin = explode("\r\n\r\n", $bin); 
$img = $bin[1]; 
$h = fopen( $filename, ‘wb‘ ); 
$res = fwrite( $h, $img ) === false ? false : true; 
@socket_close( $sock ); 
Return $res; 
} 

<?php 
$url = "http://jb51.net/content/uploadfile/201106/thum-f3ccdd27d2000e3f9255a7e3e2c4880020110622095243.jpg"; 
$filename = ‘curl.gif‘; 
//http://jb51.net 
getImg($url, $filename); 
/* 
*@通过curl方式获取制定的图片到本地 
*@ 完整的图片地址 
*@ 要存储的文件名 
*/ 
function getImg($url = "", $filename = "") { 
if(is_dir(basename($filename))) { 
echo "The Dir was not exits"; 
Return false; 
} 
//去除URL连接上面可能的引号 
$url = preg_replace( ‘/(?:^[\‘"]+|[\‘"\/]+$)/‘, ‘‘, $url ); 
$hander = curl_init(); 
$fp = fopen($filename,‘wb‘); 
curl_setopt($hander,CURLOPT_URL,$url); 
curl_setopt($hander,CURLOPT_FILE,$fp); 
curl_setopt($hander,CURLOPT_HEADER,0); 
curl_setopt($hander,CURLOPT_FOLLOWLOCATION,1); 
//curl_setopt($hander,CURLOPT_RETURNTRANSFER,false);//以数据流的方式返回数据,当为false是直接显示出来 
curl_setopt($hander,CURLOPT_TIMEOUT,60); 
/*$options = array( 
CURLOPT_URL=> ‘http://jb51.net/content/uploadfile/201106/thum-f3ccdd27d2000e3f9255a7e3e2c4880020110622095243.jpg‘, 
CURLOPT_FILE => $fp, 
CURLOPT_HEADER => 0, 
CURLOPT_FOLLOWLOCATION => 1, 
CURLOPT_TIMEOUT => 60 
); 
curl_setopt_array($hander, $options); 
*/ 
curl_exec($hander); 
curl_close($hander); 
fclose($fp); 
Return true; 
} 
?> 

<?php
$url = "图片绝对地址/thumbnail.jpg";
$filename = ‘curl.jpg‘;
getImg($url, $filename);
/*
 *@通过curl方式获取制定的图片到本地
 *@ 完整的图片地址
 *@ 要存储的文件名
 */
function getImg($url = "", $filename = "") {
 if(is_dir(basename($filename))) {
  echo "The Dir was not exits";
  return false;
 }
 //去除URL连接上面可能的引号
 $url = preg_replace( ‘/(?:^[‘"]+|[‘"/]+$)/‘, ‘‘, $url );
 $hander = curl_init();
 $fp = fopen($filename,‘wb‘);
 curl_setopt($hander,CURLOPT_URL,$url);
 curl_setopt($hander,CURLOPT_FILE,$fp);
 curl_setopt($hander,CURLOPT_HEADER,0);
 curl_setopt($hander,CURLOPT_FOLLOWLOCATION,1);
 //curl_setopt($hander,CURLOPT_RETURNTRANSFER,false);//以数据流的方式返回数据,当为false是直接显示出来
 curl_setopt($hander,CURLOPT_TIMEOUT,60);
 /*$options = array(
  CURLOPT_URL=> ‘/thum-f3ccdd27d2000e3f9255a7e3e2c4880020110622095243.jpg‘,
  CURLOPT_FILE => $fp,
  CURLOPT_HEADER => 0,
  CURLOPT_FOLLOWLOCATION => 1,
  CURLOPT_TIMEOUT => 60
 );
 curl_setopt_array($hander, $options);
 */
 curl_exec($hander);
 curl_close($hander);
 fclose($fp);
 return true;
}
?>