import requests
from fack user-agent import User-Agent
from lxml import etree
from random import randint
from time import sleep
def get_html(url):
headers= {
“User-Agent”:UserAgent().chrome
}
sleep(randint(3 ,10))
response = request.get(url,headers=headers)
response.encoding=‘utf-8’
if response.status_code == 200:
return response.text
else:
return None
type = e.xpath(‘//li[@class=“ellipsis”][1]/text()’)
actors = e.xpath(‘//li[@class=“celebrity actor”]/div[@class=“info”]/a/text()’)
actors = format_actors(actors)
return{
“name”:name
“types”:types
“actors”:actors
}
def format_actors(actors):
actor_set = set()
for actor in actors:
actor_set.add(actor.strip())
return actor_set