Ron Rademaker
Ron Rademaker
What's the benefit of these changes that justify the BC breaks?
This PR breaks a lot of backwards compatibility by adding methods to an interface and dropping support for older PHP versions. Is there no way to accomplish these things without?
This has been discussed in a few PRs, and @yohang considers these breaks acceptable. I've also experienced this and for now I pinned my usage on 1.0.\* instead of ~1.0...
Yeah, they're little changes, I just haven't had the time to get to it. A few notes about what to do would be nice though.
:-1: because there will always be states you shouldn't be able to transition to or from. A final state may by definition not be used as from. An initial state...
I think the interrupted state doesn't have to be in the component, @sagikazarmark can simply add it to the statemachine.
I think it would have been better if an interface 'MultiGraphStatefuleInterface' was added that extends from StatefulInterface, this interface would require implementing something like setStateAccessor(StateAccessorInterface $stateAccessor, $graph). That way the...
Yes, I'll create a PR.
1.1 will break BC in it's current condition, for example because all the typehints from StateMachine are gone (so anyone who has extended the class will have a broken package...
Anyone who (like me) likes the 1.0 branch better than where master was heading is welcome to join me on https://github.com/RonRademaker/Finite, I'll actively maintain this 1.0 fork and use semver...