You can not select more than 25 topics
Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
16 lines
930 B
16 lines
930 B
import requests
|
|
from bs4 import BeautifulSoup
|
|
HEADERS = {
|
|
"X-Requested-With": "XMLHttpRequest",
|
|
"User-Agent": "Mozilla/5.0 (Windows NT 10.0; WOW64) AppleWebKit/537.36"
|
|
"(KHTML, like Gecko) Chrome/56.0.2924.87 Safari/537.36",
|
|
}
|
|
html=requests.get('https://jobs.51job.com/guangzhou/119982727.html?s=01&t=0',headers=HEADERS).content
|
|
bs_tmp = BeautifulSoup(html, 'lxml').select(
|
|
'body > div.tCompanyPage > div.tCompany_center.clearfix > div.tHeader.tHjob > div > div.cn')[0]
|
|
bs_tmp1 = bs_tmp.select('h1')[0]
|
|
bs_tmp2 = bs_tmp.select('strong')[0]
|
|
bs_tmp3 = bs_tmp.select('p.cname > a.catn')[0]
|
|
bs_tmp4 = bs_tmp.select(' p.msg.ltype')[0].text.replace(u'\xa0', '').split('|')
|
|
tmp = {"岗位": bs_tmp1.text, "公司": bs_tmp3['title'], "薪资": bs_tmp2.text, '位置': bs_tmp4[0],
|
|
'工作经验': bs_tmp4[1], '学历': bs_tmp4[2], '招聘人数': bs_tmp4[3], '发布时间': bs_tmp4[4]} |