tp5.0编写的采集微博热搜榜列表的代码

<?php
namespace app\index\controller;

use think\Controller;
use think\Db;

class Weibo extends Controller
{
    public function index()
    {
        // 设置请求头信息
        $headers = array(
            'User-Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/58.0.3029.110 Safari/537.36',
            'Referer: https://weibo.com/',
            'Cookie: your_cookie_here'
        );

        // 发送请求获取热搜榜列表
        $url = 'https://s.weibo.com/top/summary?cate=realtimehot';
        $ch = curl_init();
        curl_setopt($ch, CURLOPT_URL, $url);
        curl_setopt($ch, CURLOPT_RETURNTRANSFER, 1);
        curl_setopt($ch, CURLOPT_HTTPHEADER, $headers);
        $result = curl_exec($ch);
        curl_close($ch);

        // 解析HTML获取热搜榜列表
        $pattern = '/<td class="td-02">.*?<a.*?>(.*?)<\/a>.*?<\/td>.*?<td class="td-03">.*?<span>(.*?)<\/span>.*?<\/td>/s';
        preg_match_all($pattern, $result, $matches);

        // 将热搜榜列表存入数据库
        $data = array();
        foreach ($matches[1] as $key => $value) {
            $data[] = array(
                'title' => $value,
                'hot' => $matches[2][$key]
            );
        }
        Db::name('weibo_hot')->insertAll($data);

        // 输出结果
        echo '采集成功！';
    }
}

本文来自投稿，不代表站三界导航立场，如若转载，请注明出处：https://www.zhansanjie.com/article/details/43668.html