Closed hutchike closed 1 year ago
Would it be better to have a create_passwordless_session
method, and then have people do sign_in create_passwordless_session(user)
I think that is the better API.
I see how it would be nice to have a shortcut sometimes, but I think the explicitness of having to go through building the session is almost a feature as you're forced to recognise how it's actually the session that gets "signed in".
What you suggest @henrikbjorn is awfully close to the already existing build_passwordless_session
method. So close that I don't want to add a new api right now.
Thanks for considering the PR. I just wanted to remind that the original API allowed us to sign_in
an authenticatable object (e.g. a user), whereas the new version doesn't, so it's a breaking change that will trip many people up. I'm sure there's a more elegant way to restore the ability to sign in an authenticatable object, and I do think this is a feature that's worth retaining. Just my 2c. Thanks again.
Thank you for that note. I think you're right. I'll make sure to mention it in the changelog that it's going away.
The old version of Passwordless allowed me to sign in a user with a simple
sign_in user
call. The new version requires me to pass a session object instead, so I've added a simplesign_in_authenticatable user
method to complement it. Hope you like it. I use this when I'm auth'ing via a Google login instead of a sign-in link via email.