我无法通过请求登录 Instagram

Question

I've been trying to login to Instagram using the Requests library but I can't get it to work. The connection always get refused.

import requests

#Creating URL, usr/pass and user agent variables

BASE_URL = 'https://www.instagram.com/'
LOGIN_URL = BASE_URL + 'accounts/login/ajax/'
USERNAME = '******'
PASSWD = '******'
USER_AGENT = 'Mozilla/5.0 (Windows NT 10.0; WOW64) AppleWebKit/537.36 (KHTML, like Gecko)\
 Chrome/59.0.3071.115 Safari/537.36'

#Setting some headers and refers
session = requests.Session()
session.headers = {'user-agent': USER_AGENT}
session.headers.update({'Referer': BASE_URL})


try:
    #Requesting the base url. Grabbing and inserting the csrftoken

    req = session.get(BASE_URL)
    session.headers.update({'X-CSRFToken': req.cookies['csrftoken']})
    login_data = {'username': USERNAME, 'password': PASSWD}

    #Finally login in
    login = session.post(LOGIN_URL, data=login_data, allow_redirects=True)
    session.headers.update({'X-CSRFToken': login.cookies['csrftoken']})

    cookies = login.cookies

    #Print the html results after I've logged in
    print(login.text)

#In case of refused connection
except requests.exceptions.ConnectionError:
    print("Connection refused")

I don't know what I'm doing wrong. I would really appreciate if anyone posted any solutions. Please do not suggest API or Selenium (They're not an option for me at the moment)

Answer 1

Since requests doesn't execute JavaScript's you don't have the CSRFToken in your cookies.

If you have a look at the content you can find the csrf_token inside the html.

Using bs4 and json you can extract it and use it in your post.

from bs4 import BeautifulSoup
import json, random, re, requests

BASE_URL = 'https://www.instagram.com/accounts/login/'
LOGIN_URL = BASE_URL + 'ajax/'

headers_list = [
        "Mozilla/5.0 (Windows NT 5.1; rv:41.0) Gecko/20100101"\
        " Firefox/41.0",
        "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_11_2)"\
        " AppleWebKit/601.3.9 (KHTML, like Gecko) Version/9.0.2"\
        " Safari/601.3.9",
        "Mozilla/5.0 (X11; Ubuntu; Linux x86_64; rv:15.0)"\
        " Gecko/20100101 Firefox/15.0.1",
        "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36"\
        " (KHTML, like Gecko) Chrome/42.0.2311.135 Safari/537.36"\
        " Edge/12.246"
        ]


USERNAME = '****'
PASSWD = '*****'
USER_AGENT = headers_list[random.randrange(0,4)]

session = requests.Session()
session.headers = {'user-agent': USER_AGENT}
session.headers.update({'Referer': BASE_URL})    
req = session.get(BASE_URL)    
soup = BeautifulSoup(req.content, 'html.parser')    
body = soup.find('body')

pattern = re.compile('window._sharedData')
script = body.find("script", text=pattern)

script = script.get_text().replace('window._sharedData = ', '')[:-1]
data = json.loads(script)

csrf = data['config'].get('csrf_token')
login_data = {'username': USERNAME, 'password': PASSWD}
session.headers.update({'X-CSRFToken': csrf})
login = session.post(LOGIN_URL, data=login_data, allow_redirects=True)
login.content
# b'{"authenticated": true, "user": true, "userId": "*******", "oneTapPrompt": false, "status": "ok"}'

Have in mind that most of the data in instagram it's loaded with javascript, so you may have more troubles in future.

You can refer to this post on how to recover data : https://stackoverflow.com/a/49831347

Or you can use different library like dryscrape or spynner

Answer 2

它不是返回空的脚本，而是基于我的研究的.get_text()

Answer 3

About script returning empty on 2Pacho's answer, it's not Instagram that changed since his post, but rather the behavior of the get_text() method, altered in April 2020. From thebs4 Documentation :

As of Beautiful Soup version 4.9.0, when lxml or html.parser are in use, the contents of <script>, <style>, and <template> tags are not considered to be 'text', since those tags are not part of the human-visible content of the page.

Using .contents[0] instead will do the job as intended:

script = script.contents[0].replace('window._sharedData = ', '')[:-1]
data = json.loads(script)