复习bs4，爬去腾讯招聘信息

复习一下另一种提取方式：Python根据xpath造的轮子，美味汤。/20170110

复习findAll和select两种提取方法。/20170110

# coding: utf-8
__author__ = '姜枫渔火'
import requests
from bs4 import BeautifulSoup
import json
import sys
from fake_useragent import UserAgent
from multiprocessing import Pool 

def getOnePage(num):
    headers = {'User-Agent' : UserAgent().random}
    res = requests.get('http://hr.tencent.com/position.php?&start=' + str(num), headers=headers)
    html = res.text
    return html

def prasePage(html):
    bsObj = BeautifulSoup(html, 'lxml')    # 招聘列表在tr标签下，分为odd和even两个子标签交替
#     print(bsObj)
#     odd = bsObj.select('.odd')
#     even = bsObj.select('.even')
#     recruit_list = odd + even
#     print(type(recruit_list))
#     for item in recruit_list:
#         print(type(item))
#         _item={}
#         _item['zhiweimingcheng']=item.select('td a')[0].get_text()
#         _item['zhiweilianjie']='http://hr.tencent.com/'+item.select('td a')[0].attrs['href']
#         _item['zhiweileibie']=item.select('td')[1].get_text()
#         _item['zhaopinrenshu']=item.select('td')[2].get_text()
#         _item['gongzuodidian']=item.select('td')[3].get_text()
#         _item['fabushijian']=item.select('td')[4].get_text()
    recruit_list = bsObj.findAll('tr')
#     print(type(recruit_list))    # 注意类型(<class 'bs4.element.ResultSet'>)
    for item in recruit_list[1:-2]:
#         print(type(item))    # 注意类型(<class 'bs4.element.Tag'>)，可以在列表中继续使用bs4而不必用正则
        yield {
            '职位名称' : item.select('td a')[0].get_text(),
            '职位链接' : 'http://hr.tencent.com/'+item.select('td a')[0].attrs['href'],
            '职位类别' : item.select('td')[1].get_text(),
            '招聘人数' : item.select('td')[2].get_text(),
            '工作地点' : item.select('td')[3].get_text(),
            '发布时间' : item.select('td')[4].get_text()
        }

def writeFile(content):
    with open('招聘信息1.txt', 'a') as f:
        f.write(json.dumps(content, ensure_ascii = False) + '\n')    # 解决中文乱码问题

def main(i):
    html = getOnePage(i)
    for items in prasePage(html):
        print(items)
        writeFile(items)

if __name__ == '__main__':
    pool = Pool()  # 开启进程池  
    pool.map(main,[i*10 for i in range(50)])  
    pool.close()  
    pool.join()