比较爬虫用的语言Python与Go

Python是我比较喜欢的语言，莫名的喜欢，对Python的学习可能起初是敲错了网址开始的，哈哈哈~

工作的任务从一个网站后台做登录、爬取数据，写入服务器Redis中，同事认为我会用PHP来写，哼！让你猜到那该多没意思，于是乎有了如下Python的代码，你看50多行搞定了。

 #!/usr/bin/python3
 import requests
 import re
 import redis
 from pyquery import PyQuery as pq

 loginUrl = 'https://manage.xxx.com.cn/home/login'
 userName = 'xxx'
 passWord = 'xxx'

 redisServer = '192.168.0.2'
 redisPort = 6379
 redisPass = ''

 productList = {'椰油':'CL_Spot','咖啡':'COFFEE','工业铜':'COPPER'}
 volumeList = {'CL_Spot':[0, 0], 'COFFEE':[0, 0], 'COPPER':[0, 0]}

 def main():
     jsessionid = getCookie()
     doLogin(jsessionid)
     dataUrl = 'https://manage.xxx.cn/?pageNo=1&pageSize=100'
     cookies = {'JSESSIONID': jsessionid}
     r = requests.get(dataUrl, cookies = cookies)
     dom = pq(r.text)
     lines = dom('table').eq(1).find('tr').items()
     for line in lines:
         line = re.sub(r'<!--.*-->', '', str(line))
         pattern = re.compile(r'<td>(.*?)</td>')
         group = pattern.findall(line)
         if not group:
             continue
         productCode = productList[group[3]]
         if group[6] == '买':
             volumeList[productCode][0]+= int(group[7]) * int(group[8])
         if group[6] == '卖':
             volumeList[productCode][1]+= int(group[7]) * int(group[8])

     redisClient = redis.Redis(host=redisServer, port=redisPort, password=redisPass)
     for x in volumeList:
         keyUp = 'redis_order_count_u_%s' % x
         keyDown = 'redis_order_count_d_%s' % x
         redisClient.set(keyUp, int(volumeList[x][0]))
         redisClient.set(keyDown, int(volumeList[x][1]))

 def getCookie():
     ua = {'user-agent': 'Mozilla/5.0 (Macintosh; Intel Mac OS X 10_14_1) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/70.0.3538.110 Safari/537.36'}
     r = requests.get(loginUrl, headers = ua)
     return r.cookies['JSESSIONID']

 def doLogin(jsessionid):
     param = {'userName': userName, 'password': passWord}
     cookies = {'JSESSIONID': jsessionid}
     requests.post(loginUrl, data = param, cookies = cookies)

 if __name__ == '__main__':
     main()

另一个服务也需要这个需求，用了最近看的Golang来实现一次，瞧写了100多行

 package main

 import (
     "fmt"
     "net/http"
     "net/url"
     "os"
     "strings"
     "strconv"
     "gopkg.in/redis.v4"
     "github.com/PuerkitoBio/goquery"
 )

 var loginUrl string = "https://manage.xxx.com.cn/home/login"
 var dataUrl string = "https://manage.xxx.com.cn/?pageNo=1&pageSize=100"
 var userName string = "xxx"
 var passWord string = "xxx"
 var redisServer string = "192.168.1.2"
 var redisPort string = ""
 var redisPass string = ""
 var redisDB   int = 0

 func main() {
     productList := make(map[string] string)
     productList["椰油"] = "CL_Spot"
     productList["咖啡"] = "COFFEE"
     productList["工业铜"] = "COPPER"
     volumeList := make(map[string] int)
     volumeList["u_CL_Spot"] = 0
     volumeList["d_CL_Spot"] = 0
     volumeList["u_COFFEE"] = 0
     volumeList["d_COFFEE"] = 0
     volumeList["u_COPPER"] = 0
     volumeList["d_COPPER"] = 0
     jsessionid := getCookie()
     doLogin(jsessionid)

     request, err := http.NewRequest("GET", dataUrl, nil)
     request.AddCookie(&http.Cookie{Name: "JSESSIONID", Value: jsessionid})
     client := &http.Client{}
     response, err := client.Do(request)
     if err != nil {
         fmt.Println(err.Error())
         os.Exit(0)
     }
     defer response.Body.Close()
     doc, err := goquery.NewDocumentFromReader(response.Body)
     doc.Find("table").Eq(1).Find("tr").Each(func(i int, tr *goquery.Selection) {
         td := tr.Find("td")
         name := td.Eq(3).Text()
         dir := td.Eq(6).Text()
         if val, ok := productList[name]; ok {
             buyNum, _ := strconv.Atoi(td.Eq(7).Text())
             buyUnit, _ := strconv.Atoi(td.Eq(8).Text())
             num :=  buyNum * buyUnit
             cacheKey := ""
             if dir == "买" {
                 cacheKey = fmt.Sprintf("u_%s", val)
             } else if dir == "卖" {
                 cacheKey = fmt.Sprintf("d_%s", val)
             }
             volumeList[cacheKey] += num
         }
     })
     redisClient := redis.NewClient(&redis.Options{
         Addr:     fmt.Sprintf("%s:%s", redisServer, redisPort),
         Password: redisPass,
         DB:       redisDB,
     })
     for k, v := range volumeList {
         strKey := fmt.Sprintf("redis_order_count_%s", k)
         redisClient.Set(strKey, int(v), 0)
     }
     fmt.Println("puti volume get success")
 }

 func getCookie() string {
     jsessionid := ""
     response, err := http.Get(loginUrl)
     if err != nil {
         fmt.Println(err.Error())
         os.Exit(0)
     }
     defer response.Body.Close()
     for _, val := range response.Cookies() {
         if val.Name == "JSESSIONID" {
             jsessionid = val.Value
         }
     }
     return jsessionid
 }

 func doLogin(jsessionid string) bool {
     data := url.Values{}
     data.Set("userName", userName)
     data.Add("password", passWord)
     request, _ := http.NewRequest("POST", loginUrl, strings.NewReader(data.Encode()))
     request.Header.Add("Content-Type", "application/x-www-form-urlencoded")
     request.Header.Add("Content-Length", strconv.Itoa(len(data.Encode())))
     request.AddCookie(&http.Cookie{Name: "JSESSIONID", Value: jsessionid})
     client := &http.Client{}
     response, err := client.Do(request)
     if err != nil {
         fmt.Println(err.Error())
         os.Exit(0)
     }
     defer response.Body.Close()
     return true
 }

Python的实现到上线半天的功夫搞定了，Go足足搞了1整天，蹩脚的语法与不熟悉的语法让我学习了很多知识点，最后Mac编译到Linux上执行也给我上了一课。

觉得入门学习这两门语言挺好，一个是脚本语言另一个是编译语言，用处都很广泛。轩轩你准备好了吗？