# היתכנות של מתן אפשרות ללקוח מפתח API משתמש להרשם ל-auth\_redirect חוקי

**URL:** https://meta.discourse.org/t/feasibility-of-allowing-a-user-api-key-client-to-register-a-valid-auth-redirect/312901
**Category:** Development
**Created:** [20 ביוני,‏ 2024,‏ 4:01pm UTC](https://meta.discourse.org/t/feasibility-of-allowing-a-user-api-key-client-to-register-a-valid-auth-redirect/312901 "2024-06-20T16:01:11Z")
**Posts on this page:** 1
**Showing post:** 8

<div class="post-metadata">

### Author: ![angus](https://sea3.discourse-cdn.com/meta/user_avatar/meta.discourse.org/angus/32/341715_2.png) [@angus](https://meta.discourse.org/u/angus)
#### Post date: [22 באוקטובר,‏ 2024,‏ 6:43pm UTC](https://meta.discourse.org/t/feasibility-of-allowing-a-user-api-key-client-to-register-a-valid-auth-redirect/312901/8 "2024-10-22T18:43:26Z")

</div>

I’ve rebased this PR as I’m back to focusing on ActivityPub, and this is a potential framework for one of it’s features, as discussed in the OP.

[https://github.com/discourse/discourse/pull/28119](https://github.com/discourse/discourse/pull/28119)

While rebasing I noticed that separating keys from clients as this PR does would also solve issues like the one addressed recently by @nat

[https://github.com/discourse/discourse/commit/ede06ffd439263c36ce7fad125efe556a1e8524f](https://github.com/discourse/discourse/commit/ede06ffd439263c36ce7fad125efe556a1e8524f)

Namely, the need to make this change, to destroy all old keys associated with a client, regardless of user, arises because keys and clients are in the same table. Separating them means you can just register a new key for the alternate user of the client.

---

_[View the full topic](https://meta.discourse.org/t/feasibility-of-allowing-a-user-api-key-client-to-register-a-valid-auth-redirect/312901)._