You can clone with
if I have a passing test, midje-mode correctly displays the out of the test run above the fact. If the test throws an exception, no output is displayed, even if the output occurred before the exception. This makes it difficult to fix certain test failures.
I'm using midje-mode with nrepl. Everything seems to work except for the output of failing cases just like @arohner reported (I think we mean the same).
;.;. Any intelligent fool can make things bigger, more complex, and more
;.;. violent. It takes a touch of genius -- and a lot of courage -- to move
;.;. in the opposite direction. -- Schumacher
(fact "successful" 1 => 1)
Stacktrace and expected vs actual are not properly inserted:
;.;. FAIL[0m "failing" at (NO_SOURCE_FILE:3)
(fact "failing" 1 => 2)
When evaluating the code through nrepl you do see the correct output:
[31mFAIL[0m "failing" at (NO_SOURCE_FILE:2)
Unfortunately, I don't have a clue where to dig in the midje-mode code to fix this
Unfortunately, in the ramp-up to Midje 1.5, I've been focusing on non-emacs uses, to make sure that the repl tools are generally useful. Also, my slime/swank setup has mysteriously stopped working and my emacs-of-choice (Aquamacs) doesn't work with nrepl.
Which is to say: I don't know when I can look at this. I've been having good luck with using the in-repl autotest, so I don't miss buffer annotation as much as I thought I would.