|
|

楼主 |
发表于 2019-9-22 22:34:03
|
显示全部楼层
- import requests
- import csv
- import jsonpath
- kw = input('情输入想要查询的工作:')
- with open(f'{kw}.csv', 'a', newline='') as f:
- csvWriter = csv.writer(f, dialect='excel')
- csvWriter.writerow(['城市', '岗位', '薪资'])
- for i in range(1,11):
- session = requests.Session()
- headers = {'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/73.0.3683.86 Safari/537.36',
- 'Referer': 'https://www.lagou.com/jobs/list_Python%E7%88%AC%E8%99%AB?labelWords=&fromSearch=true&suginput='}
- session.get('https://www.lagou.com/jobs/list_Python%E7%88%AC%E8%99%AB?labelWords=&fromSearch=true&suginput=', headers=headers)
- cookies = session.cookies
- print(cookies)
- cookies = cookies.get_dict()
- print(cookies)
- data = {
- 'first': 'true',
- 'pn': i,
- 'kd': kw,}
- req = session.post('https://www.lagou.com/jobs/positionAjax.json?needAddtionalResult=false', headers=headers, data=data, cookies=cookies).json()
- print(req)
- req = req.get('content')
- print(req)
- req = req['positionResult']['result']
- print(req)
- cities = jsonpath.jsonpath(req, '$..city')
- positionNames = jsonpath.jsonpath(req, '$..positionName')
- salaries = jsonpath.jsonpath(req, '$..salary')
- print(cities)
- print(positionNames)
- print(salaries)
- for city, positionName, salary in zip(cities, positionNames, salaries):
- print(city, positionName, salary)
- with open(f'{kw}.csv', 'a', newline='') as f:
- csvWriter = csv.writer(f, dialect='excel')
- csvWriter.writerow([city, positionName, salary])
复制代码 |
|