The python implementation of the Zhihu high-color image crawling

import time
import pydash
import base64
import requests
from lxml import etree
from aip import AipFace
from pathlib import Path

#The only information you have to fill in are these three lines #
APP_ID = "xxxxxxxx"
API_KEY = "xxxxxxxxxxxxxxxx"
SECRET_KEY = "xxxxxxxxxxxxxxxx"
# Filter the face value threshold, feel free if you have a lot of storage space
BEAUTY_THRESHOLD = 55
AUTHORIZATION = "oauth c3cef7c66a1843f8b3a9e6a1e3160e20"
# If the permissions are wrong, open Knowles in the browser and copy one in the developer tools without logging in
# It is advisable to change，Because I don't know the anti-crawler strategy of Knowledgeable，If too many people use the same，May affect program operation

# The length of the discussion list for each request, it is not recommended to set it too long.
LIMIT = 5
# This is the ID of the topic "Beauty", which is the parent topic of "Face Value" (20013528).
SOURCE = "19552207"

USER_AGENT = "Mozilla/5.0 (Windows NT 5.1) AppleWebKit/534.55.3 (KHTML, like Gecko) Version/5.1.5 Safari/534.55.3"
REFERER = "/topic/%s/newest" % SOURCE
# Discussion list request url for a topic
BASE_URL = "/api/v4/topics/%s/feeds/timeline_activity"
# Request parameters attached to the initial request url
URL_QUERY = "?include=data%5B%3F%%3Dtopic_sticky_module%29%%5B%3F%%3Danswer%29%%2Crelationship.is_authorized%2Cis_author%2Cvoting%2Cis_thanked%2Cis_nothelp%3Bdata%5B%3F%%3Dtopic_sticky_module%29%%5B%3F%%3Danswer%29%.is_normal%2Ccomment_count%2Cvoteup_count%2Ccontent%2Crelevant_info%%5B%3F%28type%3Dbest_answerer%29%%3Bdata%5B%3F%%3Dtopic_sticky_module%29%%5B%3F%%3Darticle%29%%2Cvoteup_count%2Ccomment_count%2Cvoting%%5B%3F%28type%3Dbest_answerer%29%%3Bdata%5B%3F%%3Dtopic_sticky_module%29%%5B%3F%%3Dpeople%29%.answer_count%2Carticles_count%2Cgender%2Cfollower_count%2Cis_followed%2Cis_following%2Cbadge%5B%3F%28type%3Dbest_answerer%29%%3Bdata%5B%3F%%3Danswer%29%%2Crelationship.is_authorized%2Cis_author%2Cvoting%2Cis_thanked%2Cis_nothelp%3Bdata%5B%3F%%3Danswer%29%%5B%3F%28type%3Dbest_answerer%29%%3Bdata%5B%3F%%3Darticle%29%%%5B%3F%28type%3Dbest_answerer%29%%3Bdata%5B%3F%%3Dquestion%29%.comment_count&limit=" + str(
  LIMIT)

HEADERS = {
  "User-Agent": USER_AGENT,
  "Referer": REFERER,
  "authorization": AUTHORIZATION

def fetch_image(url):
  try:
    response = (url, headers=HEADERS)
  except Exception as e:
    raise e
  return 

def fetch_activities(url):
  try:
    response = (url, headers=HEADERS)
  except Exception as e:
    raise e
  return ()

def parser_activities(datums, face_detective):
  for data in datums["data"]:
    target = data["target"]
    if "content" not in target or "question" not in target or "author" not in target:
      continue
    html = (target["content"])
    seq = 0
    title = target["question"]["title"]
    author = target["author"]["name"]
    images = ("//img/@src")
    for image in images:
      if not ("http"):
        continue
      image_data = fetch_image(image)
      score = face_detective(image_data)
      if not score:
        continue
      name = "{}--{}--{}--{}.jpg".format(score, author, title, seq)
      seq = seq + 1
      path = Path(__file__).("image").joinpath(name)
      try:
        f = open(path, "wb")
        (image_data)
        ()
        ()
        print(path)
        (2)
      except Exception as e:
        continue
  if not datums["paging"]["is_end"]:
    return datums["paging"]["next"]
  else:
    return None

def init_detective(app_id, api_key, secret_key):
  client = AipFace(app_id, api_key, secret_key)
  options = {"face_field": "age,gender,beauty,qualities"}
  def detective(image):
    image = str(base64.b64encode(image), "utf-8")
    response = (str(image), "BASE64", options)
    response = ("result")
    if not response:
      return
    if (not response) or (response["face_num"] == 0):
      return
    face_list = response["face_list"]
    if (face_list, "0.face_probability") < 0.6:
      return
    if (face_list, "") < BEAUTY_THRESHOLD:
      return
    if (face_list, "") != "female":
      return
    score = (face_list, "")
    return score
  return detective

def main():
  face_detective = init_detective(APP_ID, API_KEY, SECRET_KEY)
  url = BASE_URL % SOURCE + URL_QUERY
  while url is not None:
    datums = fetch_activities(url)
    url = parser_activities(datums, face_detective)
    (5)
if __name__ == '__main__':
  main()