php 正则 抓取,php正则抓取整个域名下的图片

代码出处：jUnion

适用平台：Windows， Linux(Ubuntu)，php-5.2.5+，Apache

功能：抓取整个站点的图片，暂无借助php的curl插件开发， 后期完善

配置：config目录下

domain_name：域名(默认：bizhibar.com)

request_site：网站网址(默认：http://www.bizhibar.com/)

request_url：从网站的哪个页面开始(默认：http://www.bizhibar.com/)

accept_type: 图片类型(默认：gif, bmp, png, ico,  jpg, jpeg)

save_path：图片保存路径(默认：savefiles/)

partition_name：图片保存目录名称前缀(默认：img_)

dir_file_limit: 每个目录容许多少个文件(默认：100)

serialize_img_size: 当读取了多少个图片地址才缓存到cache目录下的accompImg文件当中，下次继续抓取的时候会忽略这些地址。(默认：30)

serialize_url_size：与serialize_url_size一样，已读取多少个链接地址才缓存到cache目录

下的overURL，下次继续抓取的时候忽略这些地址。(默认：10)

说明：欢迎诸君批评指教，有任何新问题或者需要改进的地方，请您反馈给我

set_time_limit(0);

require dirname(__FILE__).DIRECTORY_SEPARATOR.'include'.DIRECTORY_SEPARATOR.'Capture.const.php';

require __Home__.'include'.__Os__.'Capture.class.php';

$_cfg = array(

'site' => __Home__.'config'.__Os__.'capture.site.php',

'preg' => __Home__.'config'.__Os__.'capture.preg.php',

'accompImg' => __Home__.'cache'.__Os__.'accompImg',

'overURL' => __Home__.'cache'.__Os__.'overURL'

);

$_parse = new Capture( $_cfg );

$_parse->parseQuestUrl();

?>

/**

* The main class

* @author pankai<530911044@qq.com>

* @date 2013-08-10

*/

class Capture {

private static $_Config = array();

private static $_CapSite = NULL;

private static $_CapPreg = NULL;

private static $_overURL = array();

private $_mark = FALSE;

private static $_markTime = 1;

/**

* initialize the main class: Capture

* @param $_cfg array

*/

public function __construct( &$_cfg ) {

self::$_Config = &$_cfg;

self::$_CapSite = require $_cfg['site'];

self::$_CapPreg = require $_cfg['preg'];

foreach( self::$_CapPreg as $_key => $_value ) {

self::$_CapPreg[$_key] = str_replace( '_request_site', self::$_CapSite['request_site'], $_value );

}

self::import( 'file.OperateFile' );

if( file_exists( $_cfg['overURL'] ) && filesize( $_cfg['overURL'] ) > 0 ) {

$_contents = OperateFile::readText( $_cfg['overURL'], filesize( $_cfg['overURL'] ) );

self::$_overURL = unserialize( $_contents );

}

self::import('pivotal.Pivotal');

if( file_exists( $_cfg['accompImg'] ) && filesize( $_cfg['accompImg'] ) > 0 ) {

$_contents = OperateFile::readText( $_cfg['accompImg'], filesize( $_cfg['accompImg'] ) );

Pivotal::$_accompImg = unserialize( $_contents );

}

}

/**

* load class, follow Java pragrammer(package): import com.jUnion.Capture

* @param $_class

*/

public static function import( $_class ) {

require_once __Home__.'include'.__Os__.str_replace( '.', __Os__, $_class ).'.class.php';

}

/**

* create an instance of Pivotal class

* @param $_source

*/

private function getCapInstance( &$_source ) {

$this->_mark = FALSE;

$_Captal = new Pivotal( self::$_Config, $_source );

$_tagA = $_Captal->parseUrl();

$this->_mark = TRUE;

return $_tagA;

}

/**

* go forward one by one

* @param $_tagArr

*/

private function roundTagA( &$_tagArr ) {

if( $_tagArr == NULL ) {

return;

}

$_tagArrLength = count( $_tagArr );

for( $i = 0; $i < $_tagArrLength; $i ++ ) {

if( is_array( $_tagArr[ $i ] ) ) {

$this->roundTagA( $_tagArr[ $i ] );

}

else {

if( stripos( $_tagArr[$i], self::$_CapSite['domain_name'] )

=== FALSE ) {

continue;

}

if( in_array( $_tagArr[$i], self::$_overURL ) ) {

continue;

}

self::$_overURL[] = $_tagArr[$i];

if( count( self::$_overURL ) % self::$_CapSite['serialize_url_size'] == 0 ) {

OperateFile::setText( self::$_Config['overURL'], serialize( self::$_overURL ) );

}

do {

$_tagA = $this->getCapInstance( Http::get( $_tagArr[$i] ) );

sleep( self::$_CapSite['preform_page_time'] * self::$_markTime );

if( $this->_mark === TRUE ) {

self::$_markTime = self::$_CapSite['preform_page_time'];

break;

}

self::$_markTime *= 2;

} while( true );

/* parse the main page and return next page */

$this->roundTagA( $_tagA );

}

}

}

//www.php100.com

public function parseQuestUrl() {

self::import('http.Http');

$_round_Arr = $this->getCapInstance( Http::get( self::$_CapSite['request_url'] ) );

$this->roundTagA( $_round_Arr );

}

}

?>