I would like to authenticate users against LDAP. I'm happy to write code. Is there a correct way of going about this (ie. via plugin), or do I need to resort to hacking the source?
I plan to use Net::LDAP Gem.
Excellent inquiry!
The short answer is that there will be a correct way to do this and many other forms of custom authentication by Wagn 1.9, but the implementation is still in progress. To hack or not to hack may depend on how your timeline gels with ours.
We're currently working on Wagn 1.8, which will include support for Rails 3 and some relevant simplifications of user/role representations.
If you have some flexibility in your timeline, I can go further into what the solution looks like, and it would probably be worth having a call about the authentication roadmap if you're interested.
ps. Apologies for the slow reply. I've been away/offline since last Wednesday for family reasons.
Checking in about LDAP/custom authentication, Wagn 1.9 having been a little while ago. :)
oy, I do owe you on this one, don't I? OK, will try for tonight. If I slack on this, you have full permission to hound me (though I shouldn't make that your responsibility!)
fwiw, you don't owe me. You owe whichever mystery user created this ticket way long ago. I'm a newb to wagn. :)
Heh, sorry. I thought you were identifying yourself as that mystery user :)
Obviously, it was a bit ambitious to think that we'd get all the new authentication stuff worked out by 1.9. But I'm quite confident that it will be in 1.13, because that's the main challenge left between here and 2.0.
As you can see on the Wagn 1.14 release card, there is not a ticket that addresses this directly, but there are two that are highly relevant: move account info into cards and Refactor Out Admin and Account Controllers.
Both of those are highly dependent upon the events api that has been central to all the Wagn 1.12 work. That API is largely in place and is used extensively in the current code, though there is still some cleanup work to do, particularly around validations and deletions.
If someone (this mystery user or other) were really interested in pushing a given authentication method forward, I'd be really interested in working together and using their case as a central user story in the development. It's been a long time coming, but we're finally very close to a fully-card-based authentication API.