attempt to fix 2FA

Fixes #1905.

README.rst

@@ -124,6 +124,7 @@ Supporters
 .. current-sponsors-start
 
 | Instaloader is proudly sponsored by
+|  `@rocketapi-io <https://github.com/rocketapi-io>`__
 |  `@socialmethod <https://github.com/socialmethod>`__
 
 See `Alex' GitHub Sponsors <https://github.com/sponsors/aandergr>`__ page for

instaloader/__init__.py

@@ -1,7 +1,7 @@
 """Download pictures (or videos) along with their captions and other metadata from Instagram."""
 
 
-__version__ = '4.10b1'
+__version__ = '4.10'
 
 
 try:

instaloader/instaloadercontext.py

@@ -253,6 +253,8 @@ class InstaloaderContext:
                                 'ig_vw': '1920', 'ig_cb': '1', 'csrftoken': '',
                                 's_network': '', 'ds_user_id': ''})
         session.headers.update(self._default_http_header())
+        # Getting mid cookie value
+        session.get("https://i.instagram.com/api/v1/web/login_page/")
         # 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
@@ -307,7 +309,8 @@ class InstaloaderContext:
                 # username is invalid.
                 raise InvalidArgumentException('Login error: User {} does not exist.'.format(user))
         # '{"authenticated": true, "user": true, "userId": ..., "oneTapPrompt": false, "status": "ok"}'
-        session.headers.update({'X-CSRFToken': login.cookies['csrftoken']})
+        if 'csrftoken' in login.cookies:
+            session.headers.update({'X-CSRFToken': login.cookies['csrftoken']})
         self._session = session
         self.username = user
         self.user_id = resp_json['userId']
@@ -324,8 +327,15 @@ class InstaloaderContext:
             raise InvalidArgumentException("No two-factor authentication pending.")
         (session, user, two_factor_id) = self.two_factor_auth_pending
 
-        login = session.post('https://www.instagram.com/accounts/login/ajax/two_factor/',
-                             data={'username': user, 'verificationCode': two_factor_code, 'identifier': two_factor_id},
+        login = session.post('https://www.instagram.com/api/v1/web/accounts/login/ajax/two_factor/',
+                             data={
+                                 'identifier': two_factor_id,
+                                 'queryParams': '{"next":"/"}',
+                                 'trust_signal': 'true',
+                                 'username': user,
+                                 'verification_method': '1',
+                                 'verificationCode': two_factor_code,
+                             },
                              allow_redirects=True)
         resp_json = login.json()
         if resp_json['status'] != 'ok':
@@ -333,7 +343,8 @@ class InstaloaderContext:
                 raise BadCredentialsException("2FA error: {}".format(resp_json['message']))
             else:
                 raise BadCredentialsException("2FA error: \"{}\" status.".format(resp_json['status']))
-        session.headers.update({'X-CSRFToken': login.cookies['csrftoken']})
+        if 'csrftoken' in login.cookies:
+            session.headers.update({'X-CSRFToken': login.cookies['csrftoken']})
         self._session = session
         self.username = user
         self.two_factor_auth_pending = None
@@ -372,8 +383,11 @@ class InstaloaderContext:
             resp = sess.get('https://{0}/{1}'.format(host, path), params=params, allow_redirects=False)
             if resp.status_code in self.fatal_status_codes:
                 redirect = " redirect to {}".format(resp.headers['location']) if 'location' in resp.headers else ""
-                raise AbortDownloadException("Query to https://{}/{} responded with \"{} {}\"{}".format(
-                    host, path, resp.status_code, resp.reason, redirect
+                body = ""
+                if resp.headers['Content-Type'].startswith('application/json'):
+                    body = ': ' + resp.text[:500] + ('…' if len(resp.text) > 501 else '')
+                raise AbortDownloadException("Query to https://{}/{} responded with \"{} {}\"{}{}".format(
+                    host, path, resp.status_code, resp.reason, redirect, body
                 ))
             while resp.is_redirect:
                 redirect_url = resp.headers['location']
@@ -402,20 +416,13 @@ class InstaloaderContext:
                 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:
-                match = re.search(r'window\._sharedData = (.*);</script>', resp.text)
+                # Extract JSON from HTML response
+                match = re.search('(?<={"raw":").*?(?<!\\\\)(?=")', resp.text)
                 if match is None:
-                    raise QueryReturnedNotFoundException("Could not find \"window._sharedData\" in html response.")
-                resp_json = json.loads(match.group(1))
-                entry_data = resp_json.get('entry_data')
-                post_or_profile_page = list(entry_data.values())[0] if entry_data is not None else None
-                if post_or_profile_page is None:
-                    raise ConnectionException("\"window._sharedData\" does not contain required keys.")
-                # If GraphQL data is missing in `window._sharedData`, search for it in `__additionalDataLoaded`.
-                if 'graphql' not in post_or_profile_page[0]:
-                    match = re.search(r'window\.__additionalDataLoaded\(.*?({.*"graphql":.*})\);</script>',
-                                      resp.text)
-                    if match is not None:
-                        post_or_profile_page[0]['graphql'] = json.loads(match.group(1))['graphql']
+                    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()

instaloader/lateststamps.py

@@ -27,7 +27,8 @@ class LatestStamps:
         self.data.read(latest_stamps_file)
 
     def _save(self):
-        makedirs(dirname(self.file), exist_ok=True)
+        if dn := dirname(self.file):
+            makedirs(dn, exist_ok=True)
         with open(self.file, 'w') as f:
             self.data.write(f)
 

instaloader/nodeiterator.py

@@ -137,7 +137,7 @@ class NodeIterator(Iterator[T]):
                 if self._first_node is None:
                     self._first_node = node
             return item
-        if self._data['page_info']['has_next_page']:
+        if self._data.get('page_info', {}).get('has_next_page'):
             query_response = self._query(self._data['page_info']['end_cursor'])
             if self._data['edges'] != query_response['edges'] and len(query_response['edges']) > 0:
                 page_index, data = self._page_index, self._data
@@ -289,7 +289,7 @@ def resumable_iteration(context: InstaloaderContext,
             is_resuming = True
             start_index = iterator.total_index
             context.log("Resuming from {}.".format(resume_file_path))
-        except (InvalidArgumentException, LZMAError, json.decoder.JSONDecodeError) as exc:
+        except (InvalidArgumentException, LZMAError, json.decoder.JSONDecodeError, EOFError) as exc:
             context.error("Warning: Not resuming from {}: {}".format(resume_file_path, exc))
     try:
         yield is_resuming, start_index

instaloader/structures.py

@@ -1097,6 +1097,27 @@ class Profile:
     def _make_is_newest_checker() -> Callable[[Post, Optional[Post]], bool]:
         return lambda post, first: first is None or post.date_local > first.date_local
 
+    def get_followed_hashtags(self) -> NodeIterator['Hashtag']:
+        """
+        Retrieve list of hashtags followed by given profile.
+        To use this, one needs to be logged in and private profiles has to be followed.
+
+        :rtype: NodeIterator[Hashtag]
+
+        .. versionadded:: 4.10
+        """
+        if not self._context.is_logged_in:
+            raise LoginRequiredException("--login required to get a profile's followers.")
+        self._obtain_metadata()
+        return NodeIterator(
+            self._context,
+            'e6306cc3dbe69d6a82ef8b5f8654c50b',
+            lambda d: d["data"]["user"]["edge_following_hashtag"],
+            lambda n: Hashtag(self._context, n),
+            {'id': str(self.userid)},
+            'https://www.instagram.com/{0}/'.format(self.username),
+        )
+
     def get_followers(self) -> NodeIterator['Profile']:
         """
         Retrieve list of followers of given profile.