iOS Tests broken when running on iOS 7.0 simulator

[rhgills]

Both the cocoapods and static library test targets appear broken when running against the iOS 7.0 simulator on a clean copy of the repository.

Steps to reproduce:

1. clone the repo, remembering to git submodule update --init
2. run the iOS cocoapods test target against iOS 7.0 (iPhone Retina 4-inch)
3. run the iOS static test target

It looks like it fixes itself after successfully running tests against iOS 6.1, which appear to be unaffected. This makes sense given the following warning:

"Pods-iOS Tests (Cocoapods) was rejected as an implicit dependency for 'libPods-iOS Tests (Cocoapods).a' because its architectures 'i386' didn't contain all required architectures 'i386 x86_64'"

And the subsequent error:

"Library not found for -lPods-iOS Tests (Cocoapods)"

[rhgills]

Also, I'm back again and plan to get back to contributing!

[jasperblues]

w00t! Welcome back!

I'm still chipping away where I can, but planning on putting in a solid week or two ASAP. . . I'm keen to get the ObjectGraph feature done (fairly easy). . .

How's your experimental factory injection going? I think you mentioned that you found a way to do it without libffi?

[rhgills]

Thanks!

Fantastic, by ObjectGraph I assume you mean the new scopes? Would be nice to have those!

The best I have for factory injection is the IMP with va_args (in #47 (comment)), but I've been thinking that definition inheritance might be a better route to solve the definition duplication problem (see #90). That was my original motivation for pursuing it. But, I recall you mentioning that you have a use case in which a parameterized method would be useful?

[jasperblues]

Re another use-case: I can't recall the details now!. . . . Here's the autofac docs: https://code.google.com/p/autofac/wiki/DelegateFactories

But unlike there's I was thinking of just declaring in a component:

• Here's the initializer. . . property x, y, z will be supplied at runtime (ie when you ask for the component from the container). . . Property a, b, c will be wired-up from the definitions.

So you could define assembly methods like:

• (id)knightWith:(Sword_)sword armor:(Armor_)armor

. . I've moved on from this feature for now though. Much more excited to do the ObjectGraph scope, which is relatively easier to implement and definitely quite useful.