Christian Edward Jackson-Gruber

Yeah - your issue is quite different than the issue described in this bug. On 5 September 2014 01:49, Eugene Beletskiy [email protected] wrote: > Meaning the classes generated by Android...

The runtime graph validation is only a hedge against adding a different set of modules at runtime vs. at compile time. If you are certain you are passing in the...

Acknowledged. On Mon Feb 02 2015 at 4:30:03 PM Dale King [email protected] wrote: > As I said we were only doing the validation in debug mode. If there were >...

Hah. I think you caught a corner case we were just discussing in the Dagger 2 initial work. We should definitely fix this in the 1.x branch though.

Since Module code generation is now required, I think generating this into the module handler is the wiser course. Shall we test for anyone's `@Nullable`? I think so, since it's...

Hmm. wait wait wait. You're pretty solidly changing the design and enabling optional injection by use of @Nullable. I know that's the effect of now, but do we really ever...

Ok. this is fair. This is less optional injection, and more explicitly binding a null, with constraints. Upon further consideration I'm more ok with it. I think we should support...

So for Dagger 2 we are steering away from this, and just forbidding null. Any serious objections?

Also, Optional injection in Guice has shifted to be, in effect, either injecting Optional or creating default bindings which can be overridden - I know Jesse and I talked about...

My only problem is that we can't forbid it at compile-time without nullability analysis beyond the scope of annotation processing. I'd love it if we could. But we can make...