美文网首页
node 爬虫

node 爬虫

作者: shadow123 | 来源:发表于2019-08-03 15:11 被阅读0次
    const request = require('request');
    const cheerio = require("cheerio");
    const iconv = require('iconv-lite')
    
    let url = 'http://m.17dm.com/wugengji/manhua/124112.html'
    request(url,function (error, response, body) {
        if(response && response.statusCode == 200){
            var html = iconv.decode(body.toString(),'gb2312');
            var $ = cheerio.load(html)
        }
        
        console.log($('select[name="listNarImg"]').children("option:last-child").attr("value"))
    });
    
    
    // http://img.17dm.com/wugengji/manhua/2b21/17.jpg
    
    const request = require('request');
    const fs= require('fs')
    let page = 8;
    let setion = 21
    let chapter = 2
    
    async function downloadImage() {
        let url = `http://img.17dm.com/wugengji/manhua/${chapter}b${setion}/${page}.jpg`;
        if (!fs.existsSync(`wugeng/${chapter}`)) {//查看是否存在这个文件夹
            fs.mkdirSync(`wugeng/${chapter}`);
        }
    
        if (!fs.existsSync(`wugeng/${chapter}/${setion}`)) {//查看是否存在这个文件夹
            fs.mkdirSync(`wugeng/${chapter}/${setion}`);//不存在就建文件夹
            console.log(`wugeng/${chapter}/${setion} 文件夹创建成功`);
        } else {
            console.log(`wugeng/${chapter}/${setion} 文件夹已经存在`);
            
        }
    
        let headers ={
            'User-Agent': 'Mozilla/5.0 (Macintosh; Intel Mac OS X 10_10_1) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/39.0.2171.99 Safari/537.36',
            'Connection':'keep-alive'
        };
    
        await request({
            url: url,
            encoding: 'binary'
        },function (error, response, body) {
            if (!error && response.statusCode == 200) {
                console.log('开始下载')
                if(!body)  console.log("(╥╯^╰╥)哎呀没有内容。。。")
                fs.writeFile(`wugeng/${chapter}/${setion}/${page}.jpg`,body, 'binary', function (err) {
                    if (err) {console.log(err);}
                    console.log(`wugeng/${chapter}/${setion}/${page}.jpg 下载成功`);
            
                })
            }
    
            setTimeout(()=>{
                page++
                if(page == 34){
                    setion++ 
                    page = 1
                }
                if(setion == 64){
                    chapter = 3
                    setion = 1;
                    page = 1
                }
                downloadImage()
            },300)
        })
    }
    
    downloadImage()
    

    相关文章

      网友评论

          本文标题:node 爬虫

          本文链接:https://www.haomeiwen.com/subject/ipmjdctx.html