Closed qerub closed 10 years ago
Oh wow. Originally I actually did write explicit expansion tests... but replaced them with the "by result" tests there now, after Travis failed them on earlier Racket versions.
Apparently I managed to forget that you had explained the reason, and provided the work-around, all along, in conditionals.rkt
. O_o
I pushed https://github.com/greghendershott/rackjure/commit/46224a7bbbb5bda2e538ac55855daf65576c0e3d to a topic branch. I think it's better. What do you think?
Hmm, Racket HEAD does not think it's better, according to Travis:
$ /usr/racket/bin/raco test -x .
raco test: (submod "./conditionals.rkt" test)
SIGSEGV MAPERR si_code 1 fault on addr 0x10
...
O_o But all other Racket versions including 5.92 are all green. Hopefully that's a broken Racket HEAD, not me. But I'll investigate...
Apparently I managed to forget that you had explained the reason, and provided the work-around, all along, in conditionals.rkt. O_o
Actually… You did that. :)
I think it's better. What do you think?
I think it's wonderful! :+1:
I'd still like to replace some of the check-equal?
with check-expand-fully
, but I can do that in another commit.
Also, I added a little note to your commit.
Actually… You did that. :)
cough Guess I'm juggling too many projects; a couple which particularly seem to flush my L1 cache...and L2, and L3. And as we see here, run scrub
on my offline storage, too.
The test suite in
threading.rkt
is not in a very good shape. Suggestions:#%app
as they are.check-expansion
fromconditionals.rkt
for the rest.(module+ test …)
to split it up (don't know if it's possible to keepracket/base
though).