'NoneType' object is not subscriptable

[g0tmi1k]

Error

From 2025-10-24, I've noticed I was getting the following error:

[i] Logging into Substack...
DEBUG:urllib3.connectionpool:Starting new HTTPS connection (1): substack.com:443
DEBUG:urllib3.connectionpool:https://substack.com:443 "GET /api/v1/user/profile/self HTTP/1.1" 200 None
[-] An unexpected error occurred: 'NoneType' object is not subscriptable

Code

This is my code which is in use:

print("[i] Logging into Substack...")
try:
    api = Api(
        debug=logging.getLogger().isEnabledFor(logging.DEBUG),
        publication_url=os.getenv("PUBLICATION_URL"),
        cookies_path=cookie,
        email=email,
        password=password
    )
    print("[+] Logged into Substack!")
except Exception as e:
    print(f"[-] An unexpected error occurred: {e}")
    sys.exit(1)

API (Today)

Digging into the API response as of today (2025-12-18):

$ curl \
    --silent \
    --header 'Cookie: substack.sid=[REMOVED]' \
   'https://substack.com/api/v1/user/profile/self' \
  | jq '.. | objects | keys'
[
  "bio",
  "can_dm",
  "facebookAccount",
  "followerCount",
  "followsViewer",
  "handle",
  "hasActivity",
  "hasGuestPost",
  "hasHiddenPublicationUsers",
  "hasHiddenSubscriptions",
  "hasLikes",
  "id",
  "isFollowing",
  "isSubscribed",
  "lists",
  "max_pub_tier",
  "name",
  "photo_url",
  "previousSlug",
  "previous_name",
  "primaryPublicationSubscriptionState",
  "profile_disabled",
  "profile_set_up_at",
  "publicationUsers",
  "reader_installed_at",
  "slug",
  "status",
  "subscriberCount",
  "subscriberCountNumber",
  "subscriberCountString",
  "subscribesToViewerSubdomain",
  "subscriptions",
  "subscriptionsTruncated",
  "theme",
  "tos_accepted_at",
  "twitterAccount",
  "userLinks",
  "visibleSubscriptionsCount"
]
[
  "id",
  "label",
  "type",
  "url",
  "value"
]
[
  "id",
  "label",
  "type",
  "url",
  "value"
]
[
  "id",
  "label",
  "type",
  "url",
  "value"
]
[]
[
  "badge",
  "bestsellerTier",
  "leaderboard",
  "paidPublicationIds",
  "subscriber",
  "subscriberTier",
  "vip"
]
$

API (Previously)

Looking at NHagar/substack_ap (WayBackMachine), was able to see a able to see the historical version of the API response:

$ cat /tmp/usage_walkthrough.json | jq 'keys'
[
  "bestseller_badge_disabled",
  "bestseller_tier",
  "bio",
  "followsViewer",
  "handle",
  "hasActivity",
  "hasGuestPost",
  "hasHiddenPublicationUsers",
  "hasLikes",
  "id",
  "isFollowing",
  "isPersonalEligible",
  "isSubscribed",
  "lists",
  "max_pub_tier",
  "name",
  "photo_url",
  "previousSlug",
  "previous_name",
  "primaryPublication",
  "profile_disabled",
  "profile_set_up_at",
  "publicationUsers",
  "rough_num_free_subscribers",
  "rough_num_free_subscribers_int",
  "slug",
  "subdomainUrl",
  "subscriberCount",
  "subscriberCountNumber",
  "subscriberCountString",
  "subscriptions",
  "subscriptionsTruncated",
  "tos_accepted_at",
  "userLinks",
  "visibleSubscriptionsCount"
]
$

Converted JSON: usage_walkthrough.json

Compare APIs

$ cat /tmp/usage_walkthrough.json| jq 'keys' > usage_walkthrough-keys
$ 
$ curl \
    --silent \
    --header 'Cookie: substack.sid=[REMOVED]' \
   'https://substack.com/api/v1/user/profile/self' | jq 'key > self-keys
$ 
$ diff self-keys usage_walkthrough-keys
1a2,3
>   "bestseller_badge_disabled",
>   "bestseller_tier",
3,5d4
<   "can_dm",
<   "facebookAccount",
<   "followerCount",
11d9
<   "hasHiddenSubscriptions",
14a13
>   "isPersonalEligible",
22c21
<   "primaryPublicationSubscriptionState",
---
>   "primaryPublication",
26c25,26
<   "reader_installed_at",
---
>   "rough_num_free_subscribers",
>   "rough_num_free_subscribers_int",
28c28
<   "status",
---
>   "subdomainUrl",
32d31
<   "subscribesToViewerSubdomain",
35d33
<   "theme",
37d34
<   "twitterAccount",
$

Following the error

Looking for /api/v1/user/profile/self:

$ git clone https://github.com/ma2za/python-substack
[...]
$
$ cd ./python-substack/
$
$ grep -Rn '/api/v1/user/profile/self'
$
$ grep -Rn '/self'
./substack/api.py:234:        response = self._session.get(f"{self.base_url}/user/profile/self")
$
$ grep -Rn -C 5 '/self'
./substack/api.py-229-
./substack/api.py-230-    def get_user_profile(self):
./substack/api.py-231-        """
./substack/api.py-232-        Gets the users profile
./substack/api.py-233-        """
./substack/api.py:234:        response = self._session.get(f"{self.base_url}/user/profile/self")
./substack/api.py-235-
./substack/api.py-236-        return Api._handle_response(response=response)
./substack/api.py-237-
./substack/api.py-238-    def get_user_settings(self):
./substack/api.py-239-        """

Can see that ./substack/api.py has get_user_profile().

---

$ grep -n get_user_profile ./substack/api.py | wc -l
       4
$
$ grep -n get_user_profile ./substack/api.py
179:        profile = self.get_user_profile()
205:        profile = self.get_user_profile()
218:        profile = self.get_user_profile()
230:    def get_user_profile(self):
$

And its only used called in 3 places.

$ grep -nB 6 self.get_user_profile ./substack/api.py
173-    def get_user_id(self):
174-        """
175-
176-        Returns:
177-
178-        """
179:        profile = self.get_user_profile()
--
199-
200-    def get_user_primary_publication(self):
201-        """
202-        Gets the users primary publication
203-        """
204-
205:        profile = self.get_user_profile()
--
212-
213-    def get_user_publications(self):
214-        """
215-        Gets the users publications
216-        """
217-
218:        profile = self.get_user_profile()
$ 

• get_user_id()
• get_user_primary_publication()
• get_user_publications()

get_user_id()

$ grep get_user_id ./substack/api.py
    def get_user_id(self):
$

...As our code doesn't use get_user_id, its not going to be this

get_user_primary_publication() / get_user_publications()

$ grep self.get_user_primary_publication ./substack/api.py
            user_publication = self.get_user_primary_publication()
$ 
$ grep self.get_user_publications ./substack/api.py
            user_publications = self.get_user_publications()
$ 

So let's see where these are being called:

class Api:
[...]
    def __init__(
[...]
        # if the user provided a publication url, then use that
        if publication_url:
[...]
            user_publications = self.get_user_publications()
[...]
        else:
            # get the users primary publication
            user_publication = self.get_user_primary_publication()

Bingo! This starts to match up whats in the code (right at the top)
So using Api(), if publication_url is set, get_user_publications() is called, Otherwise its get_user_primary_publication()
For what its worth, publication_url is set for me

get_user_publications()

Looking at the function:

    def get_user_publications(self):
        """
        Gets the users publications
        """

        profile = self.get_user_profile()

        # Loop through users "publicationUsers" list, and return a list
        # of dictionaries of "name", and "subdomain", and "id"
        user_publications = []
        for publication in profile["publicationUsers"]:
[...]

Its expecting publicationUsers in the API response back....
It USED to be there, but its not anymore!

get_user_primary_publication()

Looking at the function:

    def get_user_primary_publication(self):
        """
        Gets the users primary publication
        """

        profile = self.get_user_profile()
        primary_publication = profile["primaryPublication"]
[...]

Its expecting primaryPublication in the API response back....
Again It USED to be there, but its not anymore!

This may start to explain why its failing....

[g0tmi1k]

Example demo of what's expected

$ cat /tmp/usage_walkthrough.json | jq '. | .primaryPublication,.publicationUsers[0]'
{
  "id": 2335062,
  "subdomain": "stockinsider",
  "custom_domain_optional": false,
  "name": "The Stock Insider",
  "logo_url": "https://substack-post-media.s3.amazonaws.com/public/images/427805dc-33bf-4c72-8616-633b76cb9dbf_391x391.png",
  "author_id": 205727896,
  "user_id": 205727896,
  "handles_enabled": false,
  "explicit": false,
  "is_personal_mode": false
}
{
  "id": 2355890,
  "user_id": 205727896,
  "publication_id": 2335062,
  "role": "admin",
  "public": true,
  "is_primary": true,
  "publication": {
    "id": 2335062,
    "name": "The Stock Insider",
    "subdomain": "stockinsider",
    "custom_domain": null,
    "custom_domain_optional": false,
    "hero_text": "The best TradingView Indicator. 437% in 2024. (Independent audit in progress)",
    "logo_url": "https://substack-post-media.s3.amazonaws.com/public/images/427805dc-33bf-4c72-8616-633b76cb9dbf_391x391.png",
    "author_id": 205727896,
    "theme_var_background_pop": "#00C2FF",
    "created_at": "2024-02-08T20:30:47.517Z",
    "email_from_name": "Jack Roshi, MIT PhD",
    "copyright": "Daily Moat, LLC",
    "founding_plan_name": "Founding - Lifetime",
    "community_enabled": true,
    "invite_only": false,
    "payments_state": "enabled",
    "language": null,
    "explicit": false,
    "is_personal_mode": false,
    "author": {
      "id": 205727896,
      "name": "𝐉𝐚𝐜𝐤 𝐑𝐨𝐬𝐡𝐢, 𝐏𝐡𝐃",
      "handle": "jackroshi",
      "previous_name": "𝐉𝐚𝐜𝐤 𝐑𝐨𝐬𝐡𝐢, 𝐌𝐈𝐓 𝐏𝐡𝐃",
      "photo_url": "https://substackcdn.com/image/fetch/f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2F9b047f6a-be0f-4dfd-9f17-4a8e37e0f48b_493x493.png",
      "bio": "Follower of Christ. PhD in Applied Mathematics from Massachusetts Institute of Technology. Ex-Knight Capital Group. Professional dart thrower. Bullish on Substack.",
      "profile_set_up_at": "2024-02-08T20:29:57.020Z"
    }
  }
}
$

[ma2za]

Hi, thank you for reporting, I did not know that they changed a few fields. I made a new release 0.1.16, I don't have time for proper testing, can you let me know if it is alright?