nodejs 寫爬蟲爬取最近上映電影數據

直接上代碼：這是http模塊

'use strict'
//引入內建和第三方模塊
const https = require("https")
const url = require("url")
const Promise = require("bluebird")

//創建啓動服務模塊
function start(url){
	return new Promise((resolve,reject)=>{
			https.get(url,(res)=>{
			const statusCode = res.statusCode
			const contentType = res.headers['content-type']

			let error

			if(statusCode != 200){
				error = new Error(`請求失敗.\n`+`code:${statusCode}`)
			}

			if(error){
				console.log(error)
				res.resume()
				return
			}

			res.setEncoding("utf-8")
			let getData = ""

			res.on("data",(datachunk)=>{
				getData += datachunk
			})

			res.on("end",()=>{
				resolve(getData)
			})
		}).on("error",(e)=>{
			reject(e)
			console.log("獲取數據出錯")
		})
	})
}


exports.start = start

主程序入口:

const crawler = require("./crawler")
const querystring = require("querystring")
const cheerio = require("cheerio")
const fs = require("fs")
//url
const douban_url = "https://movie.douban.com/nowplaying/chengdu/"

crawler.start(douban_url).then(res=>{
	var $ = cheerio.load(res)
	var data = []
	$('#nowplaying .list-item').each(function(i,e){
		var obj = "obj" + i
		obj = {}
		obj.title = $(e).attr("data-title")+"\n"
		obj.score = $(e).attr("data-score")+"\n"
		obj.src = $('#nowplaying .list-item img').attr("src")+"\n"
		data.push(obj)
	})
	fs.writeFileSync("./get_data/data.txt",JSON.stringify(data));
})

效果展示: