
Hey,
So I am doing things in postman so its language agnostic...
I am posting a valid email and password to https://cloud.appwrite.io/v1/account/sessions/email
, providing the project Id, response format and key headers correctly.
The response I get back contains a secret, as I expect. However I am sure that previously (a few weeks ago?) this secret was returning the actual secret? Whereas now it is returning a malformed JWT.
When decoding from base64, it reads as a json object similar to (obviously redacted):
{
"id": "664.........",
"secret": "82d........."
}
For context, I am implementing SSR logging in. The docs for SSR login imply that we should expect just the secret to be returned, and not this base64 encoded object (as it shows setting a cookie as session
with this value, and then later using the value without any decoding etc in sessionClient.setSession
).
Anyway, I'm not using js so I can't just use the SDK. I am using C#, and have tried the SDK but to no avail (so am writing my own SDK, hence the postman).
Now, in postman, making a GET request to /account
and using the secret which I get from decoding the secret returned from /account/sessions/email
and then using just the secret
property, as the X-Appwrite-Session
header - it works and I get the data back that I expect.
I might be going mad, but I'm sure that I remember this working differently previously though? It feels like extra hoops to jump through to get the secret. Are the SSR docs out of date, or is something not going correct here?
Recommended threads
- Help with 409 Error on Relationship Setu...
I ran into a 409 document_already_exists issue. with AppWrite so I tried to debug. Here's what I've set up: Collection A has 3 attributes and a two-way 1-to-m...
- Claiming Guest-Created Documents After S...
I'm wondering if it's possible for a guest or anonymous user to create a document and attach an email address to it, so that the document can later be claimed b...
- Appwrite REST API Docs in LLM-Friendly F...
Is there a version of the Appwrite REST API documentation available in a format suitable for LLM vector embedding? Specifically, I’m referring to the following ...
