python拉钩网数据爬取

spider.py

#encoding: utf-8


# pip install bs4
# pip install requests


# sudo pip install xxx

# 显示当前所有安装的库
# pip list

# sudo easy_install pip


import requests
from bs4 import BeautifulSoup
import json
import time


def crawl_detail(id):
    url = 'https://www.lagou.com/jobs/%s.html' % id
    headers = {
        'Host': 'www.lagou.com',
        'Referer': 'https://www.lagou.com/jobs/list_python?labelWords=&fromSearch=true&suginput=',
        'Upgrade-Insecure-Requests': '1',
        'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/59.0.3071.115 Safari/537.36'
    }
    req = requests.get(url,headers=headers)
    soup = BeautifulSoup(req.content,'lxml')
    job_bt = soup.find('dd',attrs={'class':'job_bt'})
    return job_bt.text

def main():
    headers = {
        'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/59.0.3071.115 Safari/537.36',
        'Host': 'www.lagou.com',
        'Referer': 'https://www.lagou.com/jobs/list_python?labelWords=&fromSearch=true&suginput=',
        'X-Anit-Forge-Code': '0',
        'X-Anit-Forge-Token': None,
        'X-Requested-With': 'XMLHttpRequest'
    }
    form_data = {
        'first': 'true',
        'pn': '1',
        'kd': 'python'
    }
    # result = requests.post('https://www.lagou.com/jobs/positionAjax.json?city=%E5%8C%97%E4%BA%AC&needAddtionalResult=false',headers=headers,data=form_data)
    # json_result =  result.json()
    # positions = json_result['content']['positionResult']['result']

    positions = []
    for x in range(1,5):
        form_data = {
            'first': 'true',
            'pn': x,
            'kd': 'python'
        }
        result = requests.post('https://www.lagou.com/jobs/positionAjax.json?city=%E5%8C%97%E4%BA%AC&needAddtionalResult=false',headers=headers,data=form_data)
        json_result =  result.json()
        print json_result
        print '='*50
        page_positions = json_result['content']['positionResult']['result']
        for position in page_positions:
            # print position
            # print '-'*30
            # 先把需要的信息拿到，不需要的就不要了
            position_dict = {
                'position_name': position['positionName'],
                'work_year': position['workYear'],
                'salary': position['salary'],
                'district': position['district'],
                'company_name': position['companyFullName'],
            }
            position_id = position['positionId']
            # 拿到这个position，然后再去爬这个职位的详情页面
            position_detail = crawl_detail(position_id)
            position_dict['position_detail'] = position_detail
            positions.append(position_dict)
        # 出现您操作太频繁，请稍后再试的解决办法。
        # 1. 要么把sleep时间改大一点
        # 2. 每次请求不要请求这么多，分多次请求
        time.sleep(5)

    line = json.dumps(positions,ensure_ascii=False)
    with open('lagou.json','w') as fp:
        fp.write(line.encode('utf-8'))

if __name__ == '__main__':
    main()
    # crawl_detail('3265286')


    # selenium+PhtonmJS/ChromeDriver

spider_selenium.py

#encoding: utf-8

from selenium import webdriver
from bs4 import BeautifulSoup

# chromedriver
# PhantomJS
driver = webdriver.PhantomJS(r'C:\Users\hynev\develop\phantomjs\bin\phantomjs.exe')

driver.get('https://www.lagou.com/jobs/list_python?labelWords=&fromSearch=true&suginput=')

print driver.find_element_by_class_name('item_con_list').click()

本文转自：https://www.tongpankt.com/7486