erjemin/instaloader
erjemin/instaloader
1
0
Fork 0
Code Issues Wiki Activity

Fix 403 during login due to incorrect login endpoint and CSRF token location (#2133)

Browse Source
This commit is contained in:
LostXOR 2023-12-09 10:56:21 -06:00 committed by GitHub
parent 51a16a433e
commit b52547ea1c
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
1 changed files with 10 additions and 15 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

25
instaloader/instaloadercontext.py
View File

@ -3,7 +3,6 @@ import json
import os
import pickle
import random
import re
import shutil
import sys
import textwrap
@ -256,18 +255,24 @@ class InstaloaderContext:
# Override default timeout behavior.
# Need to silence mypy bug for this. See: https://github.com/python/mypy/issues/2427
session.request = partial(session.request, timeout=self.request_timeout) # type: ignore
csrf_json = self.get_json('accounts/login/', {}, session=session)
csrf_token = csrf_json['config']['csrf_token']
self.do_sleep()
# Make a request to Instagram's root URL, which will set the session's csrftoken cookie
# Not using self.get_json() here, because we need to access the cookie
session.get('https://www.instagram.com/')
# Add session's csrftoken cookie to session headers
csrf_token = session.cookies.get_dict()['csrftoken']
session.headers.update({'X-CSRFToken': csrf_token})
# Not using self.get_json() here, because we need to access csrftoken cookie
self.do_sleep()
# Workaround credits to pgrimaud.
# See: https://github.com/pgrimaud/instagram-user-feed/commit/96ad4cf54d1ad331b337f325c73e664999a6d066
enc_password = '#PWD_INSTAGRAM_BROWSER:0:{}:{}'.format(int(datetime.now().timestamp()), passwd)
login = session.post('https://www.instagram.com/accounts/login/ajax/',
login = session.post('https://www.instagram.com/api/v1/web/accounts/login/ajax/',
data={'enc_password': enc_password, 'username': user}, allow_redirects=True)
try:
resp_json = login.json()
except json.decoder.JSONDecodeError as err:
raise ConnectionException(
"Login error: JSON decode fail, {} - {}.".format(login.status_code, login.reason)
@ -403,16 +408,6 @@ class InstaloaderContext:
raise TooManyRequestsException("429 Too Many Requests")
if resp.status_code != 200:
raise ConnectionException("HTTP error code {}.".format(resp.status_code))
is_html_query = not is_graphql_query and not "__a" in params and host == "www.instagram.com"
if is_html_query:
# Extract JSON from HTML response
match = re.search('(?<={"raw":").*?(?<!\\\\)(?=")', resp.text)
if match is None:
raise QueryReturnedNotFoundException("Could not find JSON data in html response.")
# Unescape escaped JSON string
unescaped_string = match.group(0).encode("utf-8").decode("unicode_escape")
resp_json = json.loads(unescaped_string)
return resp_json
else:
resp_json = resp.json()
if 'status' in resp_json and resp_json['status'] != "ok":