node：提取网页中的指定内容

代码

const http = require('http')
const path = require('path')
const fs = require('fs')
const cheerio = require('cheerio')

/* 
  url: 请求的网址
  dst: 存储的路径(html为`${dst}.html`, 提取的数据存储在`${dst}.json`)
  selectorPath: 查找的路径
  findSelector: 要查找的selector
 */
const getDataAndSave = (url, dst, selectorPath, findSelector) => {
  const aHtml = path.resolve(__dirname, `./datas/${dst}.html`)
  const aJson = path.resolve(__dirname, `./datas/${dst}.json`)

  if (fs.existsSync(aHtml)) {
    // 先删除旧的文件
    fs.rmSync(aHtml)
  }
  // 其实这里没有必要删除json, 因为writeFile会自动删除已存在的
  if (fs.existsSync(aJson)) {
    // 先删除旧的文件
    fs.rmSync(aJson)
  }

  // 创建写入流
  const writerStream = fs.createWriteStream(aHtml)

  const req = http
    .get(url, res => {
      res.on('data', chunk => {
        // 分段写入
        writerStream.write(chunk)
        console.log('请求')
      })
    })
    .on('error', e => {
      console.log('request-err', e.message)
    })
    .on('close', () => {
      // 关闭流
      writerStream.close()

      console.log('开始解析')

      // 读取html
      fs.readFile(aHtml, { encoding: 'utf-8' }, (readError, datas) => {
        const $ = cheerio.load(datas)
        const elements = $(selectorPath).find(findSelector)
        const length = elements.length
        if (length > 0) {
          const result = []
          elements.each((i, item) => {
            result[i] = $(item).text().trim()
          })

          fs.writeFile(aJson, JSON.stringify(result, null, 2), err => {
            console.log(`fs-err`, err)
          })
        }
      })
    })
  req.end()
}

getDataAndSave('http://www.lg.gov.cn/zwfw/zdfw/yl/fwsm/yljg/sqjkfwzx/', '社康', '.sk-box .sk-item', 'a')