自己写的一个php基于phpQuery的通用采集类

WBOY

Lepaskan： 2016-07-25 08:50:15

asal

1188 orang telah melayarinya

还是小菜，第一次分享代码哈，这是自己以前写的一个php的采集类，自己一直在用，自我感觉很简单很强大，只要懂一点点选择器的知识就可以采集任何页面了，也支持https页面，做简单的采集足够用了。

/**
*通用列表采集类
*版本V1.3
*作者:JAE
*博客:http://blog.jaekj.com
*/
require_once '../phpQuery/phpQuery/phpQuery.php';
class QueryList{
private $pageURL;
private $regArr = array();
public $jsonArr = array();
private $regRange;
private $html;
/************************************************
* 参数: 页面地址选择器数组块选择器
* 【选择器数组】说明：格式array("名称"=>array("选择器","类型"),.......)
* 【类型】说明：值 "text" ,"html" ,"属性"
*【块选择器】：指先按照规则选出几个大块，然后再分别再在块里面进行相关的选择
*************************************************/
function QueryList($pageURL,$regArr=array(),$regRange='')
{
$this->pageURL = $pageURL;
//为了能获取https://
$ch = curl_init();
curl_setopt($ch, CURLOPT_URL,$this->pageURL);
curl_setopt($ch, CURLOPT_SSL_VERIFYPEER, false);
curl_setopt($ch, CURLOPT_SSL_VERIFYHOST, false);
curl_setopt($ch,CURLOPT_RETURNTRANSFER,1);
$this->html = curl_exec($ch);
curl_close($ch);
if(!empty($regArr))
{
$this->regArr = $regArr;
$this->regRange = $regRange;
$this->getList();
}
}
function setQuery($regArr,$regRange='')
{
$this->jsonArr=array();
$this->regArr = $regArr;
$this->regRange = $regRange;
$this->getList();
}
private function getList()
{
$hobj = phpQuery::newDocumentHTML($this->html);
if(!empty($this->regRange))
{
$robj = pq($hobj)->find($this->regRange);
$i=0;
foreach($robj as $item)
{
while(list($key,$reg_value)=each($this->regArr))
{
$iobj = pq($item)->find($reg_value[0]);
switch($reg_value[1])
{
case 'text':
$this->jsonArr[$i][$key] = trim(pq($iobj)->text());
break;
case 'html':
$this->jsonArr[$i][$key] = trim(pq($iobj)->html());
break;
default:
$this->jsonArr[$i][$key] = pq($iobj)->attr($reg_value[1]);
break;
}
}
//重置数组指针
reset($this->regArr);
$i++;
}
}
else
{
while(list($key,$reg_value)=each($this->regArr))
{
$lobj = pq($hobj)->find($reg_value[0]);
$i=0;
foreach($lobj as $item)
{
switch($reg_value[1])
{
case 'text':
$this->jsonArr[$i++][$key] = trim(pq($item)->text());
break;
case 'html':
$this->jsonArr[$i++][$key] = trim(pq($item)->html());
break;
default:
$this->jsonArr[$i++][$key] = pq($item)->attr($reg_value[1]);
break;
}
}
}
}
}
function getJSON()
{
return json_encode($this->jsonArr);
}
}

复制代码

require 'Query/QueryList.class.php';
//采集OSC的代码分享列表，标题链接作者
$url = "http://www.oschina.net/code/list";
$reg = array("title"=>array(".code_title a:eq(0)","text"),"url"=>array(".code_title a:eq(0)","href"),"author"=>array("img","title"));
$rang = ".code_list li";
$hj = new QueryList($url,$reg,$rang);
$arr = $hj->jsonArr;
print_r($arr);
//如果还想采当前页面右边的 TOP40活跃贡献者图像，得到JSON数据,可以这样写
$reg = array("portrait"=>array(".hot_top img","src"));
$hj->setQuery($reg);
$json = $hj->getJSON();
echo $json . "
";
//采OSC内容页内容
$url = "http://www.oschina.net/code/snippet_186288_23816";
$reg = array("title"=>array(".QTitle h1","text"),"con"=>array(".Content","html"));
$hj = new QueryList($url,$reg);
$arr = $hj->jsonArr;
print_r($arr);
//就举这么多例子吧，是不是用来做采集很方便

复制代码

/**
*自己写的百度和谷歌搜索API
*版本V2.0
*作者:JAE
*博客:http://blog.jaekj.com
**/
require_once 'QueryList_class.php';
class Searcher
{
private $searcher;
private $key;
private $num;
private $page;
private $regArr ;
private $regRange ;
private $regZnum;
public $jsonArr;
//参数搜索引擎搜索关键字返回的结果条数第几页
function Searcher($searcher,$key,$num,$page)
{
if($searcher=='baidu')
{
$this->regArr = array("title"=>array("h3.t a,#ting_singlesong_box a","text"),"tCon"=>array("div.c-abstract,font:slice(0,2),div#weibo,table tr:eq(0),div.c-abstract-size p:eq(0),div.vd_sitcom_new_tinfo","text"),"url"=>array("h3.t a,#ting_singlesong_box a","href"));
$this->regRange = 'table.result,table.result-op';
$this->regZnum=array("zNum"=>array("span.nums","text"));
}
else if($searcher=='google')
{
$this->regArr = array("title"=>array("h3.r a","text"),"tCon"=>array("span.st","text"),"url"=>array("h3.r a","href"));
$this->regRange = 'li.g';
$this->regZnum=array("zNum"=>array("div#resultStats","text"));
}
$this->searcher = $searcher;
$this->key = $key;
$this->num = $num;
$this->page = $page-1;
$this->getList();
}
private function getList()
{
$s = urlencode($this->key);
$num = $this->num;
$start = $this->num*$this->page;
if($this->searcher=='baidu')
{
$url = "http://www.baidu.com/s?pn=$start&rn=$num&wd=$s";
$reg_znum='/[\d,]+/';
}
else if($this->searcher=='google')
{
$url="https://www.google.com.hk/search?filter=0&lr=&newwindow=1&safe=images&hl=en&as_qdr=all&num=$num&start=$start&q=$s";
$reg_znum='/([\d,]+) result(s)?/';
}
$searcherObj = new QueryList($url,$this->regArr,$this->regRange);
for($i=0;$ijsonArr);$i++)
{
if($this->searcher=='baidu')
{
$searcherObj->jsonArr[$i]['url'] = $this->getBaiduRealURL($searcherObj->jsonArr[$i]['url']);
}
else if($this->searcher=='google')
{
$searcherObj->jsonArr[$i]['url'] = $this->getGoogleRealURL($searcherObj->jsonArr[$i]['url']);
}
}
$this->jsonArr = $searcherObj->jsonArr ;
//获取总共结果条数
$searcherObj->setQuery($this->regZnum);
$zNum = $searcherObj->jsonArr[0]['zNum'];
preg_match($reg_znum,$zNum,$arr)?$zNum=$arr[0]:$zNum=0;
$zNum = (int)str_replace(',','',$zNum);
//计算总页数
$zPage = ceil($zNum/$this->num);
$this->jsonArr=array('num'=>$this->num,'page'=>((int)$this->page+1),'zNum'=>$zNum,'zPage'=>$zPage,"s"=>"$this->key",'other'=>array('author'=>'JAE','QQ'=>'734708094','blog'=>'http://blog.jaekj.com'),'data'=>$this->jsonArr);
}
function getJSON()
{
return json_encode($this->jsonArr);
}
private function getBaiduRealURL($url)
{
//得到百度跳转的真正地址
$header = get_headers($url,1);
if (strpos($header[0],'301') || strpos($header[0],'302'))
{
if(is_array($header['Location']))
{
//return $header['Location'][count($header['Location'])-1];
return $header['Location'][0];
}
else
{
return $header['Location'];
}
}
else
{
return $url;
}
}
private function getGoogleRealURL($url)
{
$reg_url = '/q=(.+)&/U';
return preg_match($reg_url,$url,$arr)?urldecode($arr[1]):$url;
}
}
// $hj = new Searcher('google','oschina',20,2);
// print_r( $hj->jsonArr);
//效果演示地址
//http://blog.jaekj.com//jae/demo/searcher/Searcher_class.php?searcher=baidu&s=jaekj&num=20&page=1

复制代码