Fix up a case where we report errors poorly when state is wrong, usually...

[cgruber]

... becuase of inconsistency in what is passed in to ObjectGraph.create().

[cgruber]

Yeah 2.0 and all that, but this might help people debug, so it is probably worth rolling a 1.2.x at some point.

[JakeWharton]

Yeah there's a few tiny fixes to the 1.x stuff I'd like to make. Made a
pass over the issues last weekend.
On Aug 1, 2014 9:29 AM, "Christian Edward Gruber" [email protected]
wrote:

Yeah 2.0 and all that, but this might help people debug, so it is probably
worth rolling a 1.2.x at some point.

—
Reply to this email directly or view it on GitHub
#428 (comment).

[cgruber]

Hmm. The build is breaking. :(

[JakeWharton]

I added JDK 8 recently. See #425.

[cgruber]

Oh. Sorry - missed that. I had it in a branch locally but wasn't going to add it until i'd fixed the issues. :/ Ok. If you're satisfied - it does seem to be a failure in an unrelated invoker test.

[JakeWharton]

Yep. And the build passed on Oracle JDK 7 and OpenJDK 7

[cgruber]

Gotcha. Cool cool.