Write your own PHP Phpquery-based Universal Collection class

Or a side dish, the first time to share code ah, this is their own previous write a PHP collection class, I have been in use, self-feeling is very simple and powerful, as long as a little bit of knowledge of the selector can be collected any page, also support HTTPS page, do simple collection enough. /** * Universal List Acquisition Class * Version V1.3 * Author: JAE * Blog: http://blog.jaekj.com */ Require_once '. /phpquery/phpquery/phpquery.php '; Class querylist{ Private $pageURL; Private $REGARR = Array (); Public $JSONARR = Array (); Private $regRange; Private $html; /************************************************ * Parameters: Page Address selector array block selector * "Selector array" description: Format Array ("Name" =>array ("selector", "type"),.......) * "Type" description: Value "text", "HTML", "Properties" * "Block selector": refers to the rule to select a few chunks, and then separately in the block to make relevant choices *************************************************/ function Querylist ($pageURL, $regArr =array (), $regRange = ") { $this->pageurl = $pageURL; In order to get 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 ($reg _value[1]); Break } } Resetting an array pointer Reset ($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); } } Copy Code Require ' query/querylist.class.php '; Acquisition of the OSC code share list, Title link author $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); If you want to take the TOP40 active contributor image to the right of the current page, get the JSON data, you can write this $reg = Array ("Portrait" =>array (". Hot_top img", "src")); $hj->setquery ($reg); $json = $hj->getjson (); Echo $json. " "; Content page content of OSC $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); So many examples, is not used to do collection is very convenient Copy Code /** * Self-written Baidu and Google search API * Version V2.0 * Author: JAE * Blog: 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 $JSONARR; Parameter search engine Search keyword returns the number of result bars page function Searcher ($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; $i Jsonarr); $i + +) { if ($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; Get total number of result bars $SEARCHEROBJ->setquery ($this->regznum); $zNum = $searcherObj->jsonarr[0][' znum '); Preg_match ($reg _znum, $zNum, $arr)? $zNum = $arr [0]: $zNum = 0; $zNum = (int) str_replace (', ', ', $zNum); Calculate Total Pages $zPage = Ceil ($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) { Get Baidu jump to the real address $header = Get_headers ($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); Effect Demo Address Http://blog.jaekj.com//jae/demo/searcher/Searcher_class.php?searcher=baidu&s=jaekj&num=20&page=1 Copy Code

