技術標籤：自學心得python爬蟲

twitter圖片視訊批量下載

學python之前試用過一個twitter批量下載圖片的軟體，甚至沒有視訊下載，官方收費還是蠻貴的，學了python之後嘗試著自己寫了一個，把程式碼釋出出來，和大家交流一下，有不足的地方也請大家指正

import requests
import re
from urllib.request import urlretrieve
import os
import ssl
ssl._create_default_https_context = ssl._create_unverified_context
res=requests.
 
session()
prox={
#如有合適的代理可以放在這裡
      }
h={
   'authorization': ''# 需要抓包獲取authorization
   'x-csrf-token': '' ,#需要抓包獲取x-csrf-token
   'cookie': '' ,#需要抓包獲取cookie
​
   }
def download(file_link,file_name):
#下載媒體
    if os.path.exists(file_name) == False:
        urlretrieve(file_link, file_name)
        print
 
('下載完成')
    else:
        print('檔案已存在')
        pass
​
def getrestid(id):
#獲取rest_id函式
    if os.path.exists('./twitter/'+id)==False:
        os.makedirs('./twitter/'+id)
    idurl = 'https://api.twitter.com/graphql/ZRnOhhXPwue_JGILb9TNug/UserByScreenName?variables=%7B%22screen_name%22%3A%22' + id + '%22%2C%22withHighlightedLabel%22%3Atrue%7D'
 

    rest_id = res.get(idurl, headers=h，proxies=prox).json()['data']['user']['rest_id']
    return (rest_id)
​
​
def search(id):
    rest_id=getrestid(id)#獲取rest_id
    daurl = 'https://api.twitter.com/2/timeline/media/' + rest_id + '.json?include_profile_interstitial_type=1.txt&include_blocking=1.txt&include_blocked_by=1.txt&include_followed_by=1.txt&include_want_retweets=1.txt&include_mute_edge=1.txt&include_can_dm=1.txt&include_can_media_tag=1.txt&skip_status=1.txt&cards_platform=Web-12&include_cards=1.txt&include_composer_source=true&include_ext_alt_text=true&include_reply_count=1.txt&tweet_mode=extended&include_entities=true&include_user_entities=true&include_ext_media_color=true&include_ext_media_availability=true&send_error_codes=true&simple_quoted_tweets=true&count=3000&ext=mediaStats%2ChighlightedLabel%2CcameraMoment'
    dat = res.get(daurl, headers=h).json()['globalObjects']['tweets'].values()
    values = [i for i in dat]
    for i in values:
        try:
            try:
                full_text = i['full_text']#推文正文
                cretime = i['created_at']
                media_url_list = i['extended_entities']['media']#媒體list
                for j in media_url_list:
                    media_url = j['media_url_https']
                    media_type = j['type']
​
                    if media_type == 'video':#視訊
                        media_name = './twitter/' + id + '/' + re.sub('https://\S+', '-', full_text) + re.sub('\+.*', '-',cretime) + str(media_url_list.index(j)) + '.mp4'
                        videos = j['video_info']['variants']
                        video_size = 0
                        video_url = ""
                        live_url = ""
                        for k in videos:
                            video_type = k['content_type']
                            if video_type == 'application/x-mpegURL':
                                live_url=k['url']
                            else:
                                if k.get('bitrate') > video_size:
                                    video_size = k['bitrate']
                                    video_url = k['url']
​
                        download(video_url,media_name)
                    else:
                    #圖片
                        media_name = './twitter/' + id + '/' + re.sub('https://\S+', '-', full_text) + re.sub('\+.*', '-',cretime) + str(media_url_list.index(j)) + '.jpg'
                        download(media_url,media_name)
​
            except:
            #多個圖片
                media_url_list = i['entities']['media']
                for j in media_url_list:
                    media_url = j['media_url_https']
                    media_name = 'twitter/' + id + '/' + re.sub('https://\S+', '-', full_text) + re.sub('\+.*', '-',media_url_list.index(j)) + '.jpg'
                    download(media_url, media_name)
                    
        except:
            print('no media')
​
if __name__ == '__main__':
    id=input()#輸入使用者id
    search(id)

老魏的公眾號，大家喜歡可以關注一下，我會分享一些自己的爬蟲，大家可以一起交流