猫眼电影榜单

from requests_html import HTMLSession
from collections import namedtuple
import pandas as pd
import re


MovieRecord = namedtuple('MovieRecord', ['rank', 'title', 'star', 'releasetime', 'integer', 'fraction'])


def nth_page(n):
    url = f'https://maoyan.com/board/4?offset={10*(n-1)}'
    pat = re.compile(r'''
       <i\ class="board-index.*?>(?P<{}>.*?)</i>
       .*?title="(?P<{}>.*?)"
       .*?star">\s*主演：(?P<{}>.*?)\s*</p>
       .*?releasetime">上映时间：(?P<{}>.*?)</p>
       .*?integer">(?P<{}>.*?)</i>
       .*?fraction">(?P<{}>.*?)</i>'''.format(*MovieRecord._fields), re.X | re.S)
    req = HTMLSession().get(url)
    return [MovieRecord(*x) for x in pat.findall(req.text)]


if __name__ == '__main__':
    movie_list = []
    max_pages = 10
    for i in range(1, max_pages):
        movie_list.extend(nth_page(i))
        print(f'{i}/{max_pages} finished.')
    print(pd.DataFrame(movie_list, columns=MovieRecord._fields))