最近看到requests 作者 kennethreitz 出了一个新库 requests-html，拿来练练手。该库旨在尽可能简单直观地解析html（例如，抓取网页）。

官方文档
http://html.python-requests.org/

来抓抓网易11选5的彩票的数据。
首先我们打开网站，打开开发者工具找到对应的html。

image

session = HTMLSession()

def getData():
    response = session.get('http://caipiao.163.com/award/11xuan5/')
    content = response.html.find('section.main', first=True)
    body = content.find('tbody')
    itemDicts = dict()
    for tr in body:
        list = tr.find('td.start')
        for td in list:
            try:
                period = td.attrs['data-period']
                award = td.attrs['data-award']
                print("序号：" + td.text + " 期号：" + period + " 开奖号码：" + award)
                itemDicts[period] = award
            except KeyError as e:
                print('except: ', e)
            finally:
                print('finally')

因为还有没有开出来的开奖号码 我们就try...except了。我们发现网页是表格的，我们需要按期号排列。

sortItemDict = sorted(itemDicts.keys(), reverse=False)
    # print(sortItemDict)
    for key in sortItemDict:
        print("期号：", key, " 开奖号码：", itemDicts[key])

最后结果：

image

完整代码(发现省了不少事，直接find元素s)

from requests_html import HTMLSession
import requests

session = HTMLSession()


def getData():
    response = session.get('http://caipiao.163.com/award/11xuan5/')
    content = response.html.find('section.main', first=True)
    body = content.find('tbody')
    itemDicts = dict()
    for tr in body:
        list = tr.find('td.start')
        for td in list:
            try:
                period = td.attrs['data-period']
                award = td.attrs['data-award']
                print("序号：" + td.text + " 期号：" + period + " 开奖号码：" + award)
                itemDicts[period] = award
            except KeyError as e:
                print('except: ', e)
            finally:
                print('finally')
    sortItemDict = sorted(itemDicts.keys(), reverse=False)
    # print(sortItemDict)
    for key in sortItemDict:
        print("期号：", key, " 开奖号码：", itemDicts[key])


if __name__ == '__main__':
    getData()