Add Basic In Memory Auth about five15 HOT 5 CLOSED

The roles are: Admin, Manager, User ?

from five15.

I think those are fine until we find out that we need others.

from five15.

Does this story include anything for the Grails side?

I've added some users to the angular side. Feel free to login as user:user, manager:manager, or admin:admin. Nothing spectacular happens yet, but I do have the 'report' state limited to accounts with the 'user' role.

@craigburke I extended the data field in the state config object to show what roles a state expects users to have to allow a transition to the next state. What do you think of this approach? It feels clean to me. I'm also curious about what I need to do to make use of the annotate and closure-wrap modules in the asset pipeline.

Let me know what you think.

from five15.

@VoltiSubito I really like what you did with the roles. Alternatively we could add an $http interceptor to redirect to $state('login') when we get a 403 response: https://docs.angularjs.org/api/ng/service/$http

I think I like your approach better though. It has the advantage of not requiring each state to have an http request to trigger the 403 and not triggering any requests at all before you are authenticated. We'll have some redundancy on the backend but I think it's really slick 👍

As for the asset pipeline stuff. I actually put a couple pull requests in to the asset pipeline core project to help with those two modules. The closure wrap will automatically wrap any js file with the directive option:

//= wrapped

As for the annotation we'll actually have a decision to make on that front. Basically it'll boil down to using my annotate plugin (which will just work out of the box) or getting much better build performance using the soon to be available angularPass compile setting in new version of AP. It's faster but we'll have to put a special comment before each function like this:

// @ngInject 
function FooController($scope) { 
...
}

So we'll just have to see which approach makes most sense.

from five15.

We have a dummied up in memory thing now so I am going to go ahead and close this. We need to start building real auth but will track that as separate issues.

from five15.