爱站关键词挖掘

aizhan.php

@@ -0,0 +1,164 @@
+<?php
+require_once('vendor/autoload.php');
+
+use PhpOffice\PhpSpreadsheet\Spreadsheet;
+use PhpOffice\PhpSpreadsheet\Writer\Xlsx;
+
+$word = $_GET['t'];
+$js = new JS();
+$header = [[
+    '关键词',
+    'PC',
+    '移动'
+]];
+$html = $js->getPage($word);
+$ql = (new QL\QueryList)->html($html);
+$table = $ql->find('table');
+
+// 采集表头
+$tableHeader = $table->find('tr:eq(0)')->find('td')->texts();
+// 采集表的每行内容
+$tableRows = $table->find('tr:gt(0)')->map(function ($row) {
+    return $row->find('td')->texts()->all();
+});
+
+//print_r($tableHeader->all());
+$data = $tableRows->all();
+$rt = $ql->find('.pager li');
+$pages = $rt->count();
+for ($page = 2; $page <= $pages; $page++) {
+    $_data = $js->getPage($word, $page);
+    $_ql = (new QL\QueryList)->html($html);
+    $table = $_ql->find('table');
+// 采集表的每行内容
+    $tableRows = $table->find('tr:gt(0)')->map(function ($row) {
+        return $row->find('td')->texts()->all();
+    });
+    $_data = $tableRows->all();
+    if ($_data) {
+        $data = array_merge($data, $_data);
+    }
+}
+$new_data = array_map(function ($a) {
+    $pat = "#\d+#";
+    preg_match_all($pat, $a[2], $arr);
+    return [
+        'title' => $a[1],
+        'pc' => $arr[0][0],
+        'mobile' => $arr[0][1],
+    ];
+}, $data);
+$data=array_merge($header,$new_data);
+
+$spreadsheet = new Spreadsheet();
+try {
+    $title="关键词挖掘-" . $word;
+    $sheet = $spreadsheet->getActiveSheet();
+    $sheet->setTitle($title);
+    $sheet->fromArray($data);
+    $writer = new Xlsx($spreadsheet);
+    header("Pragma: public");
+    header("Expires: 0");
+    header("Cache-Control:must-revalidate, post-check=0, pre-check=0");
+    header("Content-Type:application/force-download");
+    header("Content-Type:application/vnd.ms-execl");
+    header("Content-Type:application/octet-stream");
+    header("Content-Type:application/download");
+    header('Content-Disposition:attachment;filename="'.$title.'.xlsx"');
+    header("Content-Transfer-Encoding:binary");
+    $writer->save('php://output');
+    $spreadsheet->disconnectWorksheets();
+} catch (\PhpOffice\PhpSpreadsheet\Exception $e) {
+}
+
+
+class JS
+{
+
+    private $cookie_file = "/www/cookie_wyt.txt";
+
+    public function getPage($word, $page = 1)
+    {
+        $word = $this->encode($word);
+        $curl = curl_init();
+        if ($page == 1) {
+            $url = "https://ci.aizhan.com/" . $word . "/";
+        } else {
+            $url = "https://ci.aizhan.com/" . $word . "/" . $page;
+        }
+
+        curl_setopt_array($curl, array(
+            CURLOPT_URL => $url,
+            CURLOPT_RETURNTRANSFER => true,
+            CURLOPT_ENCODING => "",
+            CURLOPT_MAXREDIRS => 10,
+            CURLOPT_TIMEOUT => 0,
+            CURLOPT_FOLLOWLOCATION => true,
+            CURLOPT_HTTP_VERSION => CURL_HTTP_VERSION_1_1,
+            CURLOPT_CUSTOMREQUEST => "GET",
+            CURLOPT_HTTPHEADER => array(
+                "Connection: keep-alive",
+                "Pragma: no-cache",
+                "Cache-Control: no-cache",
+                "Upgrade-Insecure-Requests: 1",
+                "DNT: 1",
+                "User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10_14_6) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/86.0.4240.80 Safari/537.36",
+                "Accept: text/html,application/xhtml+xml,application/xml;q=0.9,image/avif,image/webp,image/apng,*/*;q=0.8,application/signed-exchange;v=b3;q=0.9",
+                "Sec-Fetch-Site: same-origin",
+                "Sec-Fetch-Mode: navigate",
+                "Sec-Fetch-User: ?1",
+                "Sec-Fetch-Dest: document",
+                "Referer: https://ci.aizhan.com/",
+                "Accept-Language: zh-CN,zh;q=0.9,en;q=0.8",
+                "Cookie: _csrf=07a32f1f032d03f1d744d53696ae7add08ec7d0fe08202093af6bd0183354165a%3A2%3A%7Bi%3A0%3Bs%3A5%3A%22_csrf%22%3Bi%3A1%3Bs%3A32%3A%22meXqZAnQyb4P9Tp2mYSl8zuGXy-6uMbT%22%3B%7D; Hm_lvt_b37205f3f69d03924c5447d020c09192=1603156102; allWords=%E6%B5%8B%E8%AF%95%2C0; userId=508549; userName=824711626%40qq.com; userGroup=1; userSecure=6LJjF3kwjMXxZsIW%2BWciJlC40ros0bUvHZJE5v9IJRUsJoIQoJh1JD177E1B3fHCTMdQsv4I1vRADQgf; Hm_lpvt_b37205f3f69d03924c5447d020c09192=1603180587"
+            ),
+
+        ));
+        curl_setopt($curl, CURLOPT_COOKIEJAR, $this->cookie_file); //连接结束后保存cookie信息的文件。
+        curl_setopt($curl, CURLOPT_COOKIEFILE, $this->cookie_file); //包含cookie数据的文件名，cookie文件的格式可以是Netscape格式，或者只是纯HTTP头部信息存入文件。
+
+        $response = curl_exec($curl);
+
+        curl_close($curl);
+        return $response;
+    }
+
+    /**
+     * 将字符串转换为ascii码
+     * @param string $c 要编码的字符串
+     * @param string $prefix 前缀，默认：&#
+     * @return string
+     */
+    public function encode($c, $prefix = "&#")
+    {
+        $len = strlen($c);
+        $a = 0;
+        $scill = '';
+        while ($a < $len) {
+            $ud = 0;
+            if (ord($c{$a}) >= 0 && ord($c{$a}) <= 127) {
+                $ud = ord($c{$a});
+                $a += 1;
+            } else if (ord($c{$a}) >= 192 && ord($c{$a}) <= 223) {
+                $ud = (ord($c{$a}) - 192) * 64 + (ord($c{$a + 1}) - 128);
+                $a += 2;
+            } else if (ord($c{$a}) >= 224 && ord($c{$a}) <= 239) {
+                $ud = (ord($c{$a}) - 224) * 4096 + (ord($c{$a + 1}) - 128) * 64 + (ord($c{$a + 2}) - 128);
+                $a += 3;
+            } else if (ord($c{$a}) >= 240 && ord($c{$a}) <= 247) {
+                $ud = (ord($c{$a}) - 240) * 262144 + (ord($c{$a + 1}) - 128) * 4096 + (ord($c{$a + 2}) - 128) * 64 + (ord($c{$a + 3}) - 128);
+                $a += 4;
+            } else if (ord($c{$a}) >= 248 && ord($c{$a}) <= 251) {
+                $ud = (ord($c{$a}) - 248) * 16777216 + (ord($c{$a + 1}) - 128) * 262144 + (ord($c{$a + 2}) - 128) * 4096 + (ord($c{$a + 3}) - 128) * 64 + (ord($c{$a + 4}) - 128);
+                $a += 5;
+            } else if (ord($c{$a}) >= 252 && ord($c{$a}) <= 253) {
+                $ud = (ord($c{$a}) - 252) * 1073741824 + (ord($c{$a + 1}) - 128) * 16777216 + (ord($c{$a + 2}) - 128) * 262144 + (ord($c{$a + 3}) - 128) * 4096 + (ord($c{$a + 4}) - 128) * 64 + (ord($c{$a + 5}) - 128);
+                $a += 6;
+            } else if (ord($c{$a}) >= 254 && ord($c{$a}) <= 255) { //error
+                $ud = false;
+            }
+            $scill .= dechex($ud);
+        }
+        return $scill;
+    }
+}