冯定波PHP技术分享博客

#!/usr/bin/php
<?php
// 需要抓取的网页地址
$url = 'http://www.fengdingbo.com/category/linux';
$content = file_get_contents($url);
preg_match_all('/<img\s+src="(.*?)"/', $content, $matches,PREG_SET_ORDER);
echo "已发现".count($matches)."张图片\n";
 
list($sm, $ss) = explode(" ", microtime());
foreach ($matches as $k => $val)
{
	$pid[$k] = pcntl_fork();
	if(!$pid[$k])
	{
		download($url, $val);
		// 子进程要exit否则会进行递归多进程，父进程不要exit否则终止多进程
		exit(0);
	}
 
	if ($pid[$k])
	{
// 			pcntl_waitpid($pid[$k], $status, WUNTRACED);
	}
 
}
echo "下载完成\n";
 
list($em, $es) = explode(" ", microtime());
 
echo "用时：",($es+$em) - ($ss + $sm),"\n";
/**
 * 抓取网页图片
 * 
 */
function download($url, $val)
{
	$pic_url = $val[1];
	if (strpos($val[1], '//') !== false)
	{
		;
	}
	elseif (preg_match('@^(.*?)/@', $val[1], $inner_matches) == 0)
	{
		$pic_url = $url.$val[1];
	}
	elseif (preg_match('@[:.]@', $inner_matches[1], $tmp_matches) == 0)
	{
		$pic_url = $url.$val[1];
	}
 
	$pic = file_get_contents($pic_url);
 
	if ($pic === false)
	{
		return;
	}
 
	preg_match('@/([^/]+)$@', $pic_url, $tmp_matches);
	// 可使用assert处理异常
	$pic_file_name = $tmp_matches[1];
	$f = fopen("tmp/".$pic_file_name, "wb"); #
	fwrite($f, $pic);
	fclose($f);
}
 
/* End of file pcntl_fork.php */