简单好用的nodejs 爬虫框架分享

$ npm install crawl-pet -g --production

$ crawl-pet

> Set project dir: ./test-crawl-pet
> Create crawl-pet in ./test-crawl-pet [y/n]: y
> Set target url: http://foodshot.co/
> Set save rule [url/simple/group]: url
> Set file type limit: 
> The limit: not limit
> Set parser rule module:
> The module: use default crawl-pet.parser

$ crawl-pet -o ./test-crawl-pet

$ crawl-pet --create-parser ./test-crawl-pet/parser.js

// crawl-pet 支持使用 cheerio，来进行页面分析，如果你有这个需要
const cheerio = require("cheerio")

/*
 * header 函数是在请求发送前调用，可以配置请求的头信息，如果返回 false，则中断请求
 *
 * 参数：
 *  options:   详细设置请看 https://github.com/request/request
 *  crawler_handle: 与队列通信的对象，详情见下
 *
 * header 函数是可选的，可不写
 */
exports.header = function(options, crawler_handle) {  
}

/*
 * body 函数是在请求返回后调用，用来解析返回结果
 *
 * 参数:
 *  url:   请求的 url
 *  body:   请求返回结果, string 类型
 *  response:  请求的响应，详情请看： https://github.com/request/request
 *  crawler_handle: 与队列通信的对象，该对象包含以下方法
 *   .info    : crawl-pet 的配置信息
 *   .uri     : 当前请求的 uri 信息
 *   .addPage(url)  : 向队列里添加一个待解析页面
 *   .addDown(url)  : 向队列里添加一个待下载文件
 *   .save(content, ext) : 保存文本到本地，ext 设置保存文件的后缀名
 *   .over()    : 结束当前队列，取出下一条队列数据
 */

exports.body = function(url, body, response, crawler_handle) {
 const re = /\b(href|src)\s*=\s*["']([^'"#]+)/ig
 var m = null
 while (m = re.exec(body)){
  let href = m[2]
  if (/\.(png|gif|jpg|jpeg|mp4)\b/i.test(href)) {
    // 这理添加了一条下载
   crawler_handle.addDown(href)
  }else if(!/\.(css|js|json|xml|svg)/.test(href)){
    // 这理添加了一个待解析页面
   crawler_handle.addPage(href)
  }
 }
  // 记得在解析结束后一定要执行
 crawler_handle.over()
}

$ crawl-pet -f ./test-crawl-pet/photos.foodshot.co/*.jpg

$ crawl-pet -l queue

$ crawl-pet --help

 Crawl-pet options help:

 -u, --url  string    Destination address
 -o, --outdir string    Save the directory, Default use pwd
 -r, --restart      Reload all page
 --clear        Clear queue
 --save   string    Save file rules following options
          = url: Save the path consistent with url
          = simple: Save file in the project path
          = group: Save 500 files in one folder
 --types   array    Limit download file type
 --limit   number=5   Concurrency limit
 --sleep   number=200   Concurrent interval
 --timeout  number=180000  Queue timeout
 --proxy   string    Set up proxy
 --parser  string    Set crawl rule, it's a js file path!
          The default load the parser.js file in the project path
 --maxsize  number    Limit the maximum size of the download file
 --minwidth  number    Limit the minimum width of the download file
 --minheight  number    Limit the minimum height of the download file
 -i, --info       View the configuration file
 -l, --list  array    View the queue data 
          e.g. [page/down/queue],0,-1
 -f, --find  array    Find the download URL of the local file
 --json        Print result to json format
 -v, --version      View version
 -h, --help       View help

$ crawl-pet -u https://www.reddit.com/r/funny/ -o reddit --save group

{
 "url": "https://www.reddit.com/r/funny/",
 "outdir": ".",
 "save": "group",
 "types": "",
 "limit": "5",
 "parser": "my_parser.js",
 "sleep": "200",
 "timeout": "180000",
 "proxy": "",
 "maxsize": 0,
 "minwidth": 0,
 "minheight": 0,


 "cookie": "over18=1"
}

exports.body = function(url, body, response, crawler_handle) {
 const re = /\b(data-url|href|src)\s*=\s*["']([^'"#]+)/ig
 var m = null
 while (m = re.exec(body)){
  let href = m[2]
  if (/thumb|user|icon|\.(css|json|js|xml|svg)\b/i.test(href)) {
   continue
  }
  if (/\.(png|gif|jpg|jpeg|mp4)\b/i.test(href)) {
   crawler_handle.addDown(href)
   continue
  }
  if(/reddit\.com\/r\//i.test(href)){
   crawler_handle.addPage(href)
  }
 }
 crawler_handle.over()
}