采集

采集

关注
发信
关注(28)粉丝(399)

自己写的一个php基于phpQuery的通用采集类

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。
  1. /**
  2. *通用列表采集类
  3. *版本V1.3
  4. *作者:JAE
  5. *博客:http://blog.jaekj.com
  6. */
  7. require_once '../phpQuery/phpQuery/phpQuery.php';
  8. class QueryList{
  10. private $pageURL;
  11. private $regArr = array();
  12. public $jsonArr = array();
  13. private $regRange;
  14. private $html;
  15. /************************************************
  16. * 参数: 页面地址 选择器数组 块选择器
  17. * 【选择器数组】说明：格式array("名称"=>array("选择器","类型"),.......)
  18. * 【类型】说明：值 "text" ,"html" ,"属性"
  19. *【块选择器】：指 先按照规则 选出 几个大块 ，然后再分别再在块里面 进行相关的选择
  20. *************************************************/
  21. function QueryList($pageURL,$regArr=array(),$regRange='')
  22. {
  23. $this->pageURL = $pageURL;
  25. //为了能获取https://
  26. $ch = curl_init();
  27. curl_setopt($ch, CURLOPT_URL,$this->pageURL);
  28. curl_setopt($ch, CURLOPT_SSL_VERIFYPEER, false);
  29. curl_setopt($ch, CURLOPT_SSL_VERIFYHOST, false);
  30. curl_setopt($ch,CURLOPT_RETURNTRANSFER,1);
  31. $this->html = curl_exec($ch);
  32. curl_close($ch);
  34. if(!empty($regArr))
  35. {
  37. $this->regArr = $regArr;
  38. $this->regRange = $regRange;
  39. $this->getList();
  40. }
  42. }
  43. function setQuery($regArr,$regRange='')
  44. {
  45. $this->jsonArr=array();
  46. $this->regArr = $regArr;
  47. $this->regRange = $regRange;
  48. $this->getList();
  49. }
  50. private function getList()
  51. {
  53. $hobj = phpQuery::newDocumentHTML($this->html);
  54. if(!empty($this->regRange))
  55. {
  56. $robj = pq($hobj)->find($this->regRange);
  58. $i=0;
  59. foreach($robj as $item)
  60. {
  62. while(list($key,$reg_value)=each($this->regArr))
  63. {
  64. $iobj = pq($item)->find($reg_value[0]);
  66. switch($reg_value[1])
  67. {
  68. case 'text':
  69. $this->jsonArr[$i][$key] = trim(pq($iobj)->text());
  70. break;
  71. case 'html':
  72. $this->jsonArr[$i][$key] = trim(pq($iobj)->html());
  73. break;
  74. default:
  75. $this->jsonArr[$i][$key] = pq($iobj)->attr($reg_value[1]);
  76. break;
  78. }
  79. }
  80. //重置数组指针
  81. reset($this->regArr);
  82. $i++;
  83. }
  84. }
  85. else
  86. {
  87. while(list($key,$reg_value)=each($this->regArr))
  88. {
  89. $lobj = pq($hobj)->find($reg_value[0]);
  92. $i=0;
  93. foreach($lobj as $item)
  94. {
  95. switch($reg_value[1])
  96. {
  97. case 'text':
  98. $this->jsonArr[$i++][$key] = trim(pq($item)->text());
  99. break;
  100. case 'html':
  101. $this->jsonArr[$i++][$key] = trim(pq($item)->html());
  102. break;
  103. default:
  104. $this->jsonArr[$i++][$key] = pq($item)->attr($reg_value[1]);
  105. break;
  107. }
  110. }
  113. }
  114. }
  115. }
  116. function getJSON()
  117. {
  118. return json_encode($this->jsonArr);
  119. }
  121. }
复制代码
  3. //采集OSC的代码分享列表，标题 链接 作者
  4. $url = "http://www.oschina.net/code/list";
  5. $reg = array("title"=>array(".code_title a:eq(0)","text"),"url"=>array(".code_title a:eq(0)","href"),"author"=>array("img","title"));
  6. $rang = ".code_list li";
  7. $hj = new QueryList($url,$reg,$rang);
  8. $arr = $hj->jsonArr;
  9. print_r($arr);
  10. //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
  11. $reg = array("portrait"=>array(".hot_top img","src"));
  12. $hj->setQuery($reg);
  13. $json = $hj->getJSON();
  14. echo $json . "
    ";
  16. //采OSC内容页内容
  17. $url = "http://www.oschina.net/code/snippet_186288_23816";
  18. $reg = array("title"=>array(".QTitle h1","text"),"con"=>array(".Content","html"));
  19. $hj = new QueryList($url,$reg);
  20. $arr = $hj->jsonArr;
  21. print_r($arr);
  23. //就举这么多例子吧，是不是用来做采集很方便
复制代码
  1. /**
  2. *自己写的百度和谷歌搜索API
  3. *版本V2.0
  4. *作者:JAE
  5. *博客:http://blog.jaekj.com
  6. **/
  7. require_once 'QueryList_class.php';
  8. class Searcher
  9. {
  10. private $searcher;
  11. private $key;
  12. private $num;
  13. private $page;
  14. private $regArr ;
  15. private $regRange ;
  16. private $regZnum;
  17. public $jsonArr;
  18. //参数 搜索引擎 搜索关键字 返回的结果条数 第几页
  19. function Searcher($searcher,$key,$num,$page)
  20. {
  21. if($searcher=='baidu')
  22. {
  23. $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"));
  24. $this->regRange = 'table.result,table.result-op';
  25. $this->regZnum=array("zNum"=>array("span.nums","text"));
  26. }
  27. else if($searcher=='google')
  28. {
  29. $this->regArr = array("title"=>array("h3.r a","text"),"tCon"=>array("span.st","text"),"url"=>array("h3.r a","href"));
  30. $this->regRange = 'li.g';
  31. $this->regZnum=array("zNum"=>array("div#resultStats","text"));
  32. }
  33. $this->searcher = $searcher;
  34. $this->key = $key;
  35. $this->num = $num;
  36. $this->page = $page-1;
  37. $this->getList();
  38. }
  39. private function getList()
  40. {
  41. $s = urlencode($this->key);
  42. $num = $this->num;
  43. $start = $this->num*$this->page;
  44. if($this->searcher=='baidu')
  45. {
  46. $url = "http://www.baidu.com/s?pn=$start&rn=$num&wd=$s";
  47. $reg_znum='/[\d,]+/';
  48. }
  49. else if($this->searcher=='google')
  50. {
  51. $url="https://www.google.com.hk/search?filter=0&lr=&newwindow=1&safe=images&hl=en&as_qdr=all&num=$num&start=$start&q=$s";
  52. $reg_znum='/([\d,]+) result(s)?/';
  53. }
  54. $searcherObj = new QueryList($url,$this->regArr,$this->regRange);
  55. for($i=0;$ijsonArr);$i++)
  56. {
  57. if($this->searcher=='baidu')
  58. {
  59. $searcherObj->jsonArr[$i]['url'] = $this->getBaiduRealURL($searcherObj->jsonArr[$i]['url']);
  60. }
  61. else if($this->searcher=='google')
  62. {
  63. $searcherObj->jsonArr[$i]['url'] = $this->getGoogleRealURL($searcherObj->jsonArr[$i]['url']);
  64. }
  65. }
  66. $this->jsonArr = $searcherObj->jsonArr ;
  68. //获取总共结果条数
  70. $searcherObj->setQuery($this->regZnum);
  71. $zNum = $searcherObj->jsonArr[0]['zNum'];
  72. preg_match($reg_znum,$zNum,$arr)?$zNum=$arr[0]:$zNum=0;
  73. $zNum = (int)str_replace(',','',$zNum);
  74. //计算总页数
  75. $zPage = ceil($zNum/$this->num);
  76. $this->jsonArr=array('num'=>$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'=>$this->jsonArr);
  79. }
  80. function getJSON()
  81. {
  82. return json_encode($this->jsonArr);
  83. }
  84. private function getBaiduRealURL($url)
  85. {
  86. //得到百度跳转的真正地址
  87. $header = get_headers($url,1);
  88. if (strpos($header[0],'301') || strpos($header[0],'302'))
  89. {
  90. if(is_array($header['Location']))
  91. {
  92. //return $header['Location'][count($header['Location'])-1];
  93. return $header['Location'][0];
  94. }
  95. else
  96. {
  97. return $header['Location'];
  98. }
  99. }
  100. else
  101. {
  102. return $url;
  103. }
  104. }
  105. private function getGoogleRealURL($url)
  106. {
  107. $reg_url = '/q=(.+)&/U';
  108. return preg_match($reg_url,$url,$arr)?urldecode($arr[1]):$url;
  110. }
  111. }
  112. // $hj = new Searcher('google','oschina',20,2);
  113. // print_r( $hj->jsonArr);
  114. //效果演示地址
  115. //http://blog.jaekj.com//jae/demo/searcher/Searcher_class.php?searcher=baidu&s=jaekj&num=20&page=1
复制代码


09-02 20:01
Powered by LMLPHP ©2024 采集 0.001492