kadoban 2017-02-08 15:45:54
threestrikes: Perhaps you need to change something in the .cabal file to refer to the file you've added or whatever? I'm kind of guessing.
threestrikes 2017-02-08 15:50:52
I'll check there and see if anything sticks out. Thanks
kadoban 2017-02-08 15:52:19
Or it's screwing up at runtime I guess, so could be anything in the code too, I suppose? Not really sure what's going on there.
geekosaur 2017-02-08 15:55:01
likely missing from data-files: so the runtime data file magic didn't magic right
geekosaur 2017-02-08 15:55:19
(https://www.haskell.org/cabal/users-guide/developing-packages.html#accessing-data-files-from-package-code)
threestrikes 2017-02-08 15:55:53
yeah I was checking runtime first. I can see that the source file I changed throws this exception (\(e :: SomeException) -> getDataDir >>= \dataDir -> throwIDE (T.pack $ printf (__ "Can't load icons from %s %s") dataDir (s
threestrikes 2017-02-08 15:56:21
how e)))
threestrikes 2017-02-08 15:57:20
Legion is about to come on so this may have to wait lol
threestrikes 2017-02-08 15:58:36
Thanks geekosaur I'll check that as well
erisco 2017-02-08 16:05:18
is there anything that can be done about this? http://lpaste.net/352256
Welkin 2017-02-08 16:07:55
erisco: you and your type adventures
geekosaur 2017-02-08 16:08:58
good luck with that, then
Welkin 2017-02-08 16:10:24
what is the best resource for learning about dependent typing using idris?
Welkin 2017-02-08 16:11:25
I have the haskell paper "Dependently Typed Programming with Singletons"
geekosaur 2017-02-08 16:11:32
and I have no clue what a constraint would mean when promoted. that error suggests neither does ghc >.>
mniip 2017-02-08 16:11:41
geekosaur, it's new in TypeInType
lpaste_ 2017-02-08 16:11:43
xcmw revised "Why stack fail to build reactive banana and wxwidgets?": "Why does stack fail to build reactive banana and wxwidgets?" at http://lpaste.net/8538512301479690240
erisco 2017-02-08 16:12:00
I don't either, but maybe it makes sense to merely drop the constraints
kadoban 2017-02-08 16:12:27
xcmw: At a guess, lack of C libraries or something? Did you check the logs it points to?
geekosaur 2017-02-08 16:12:37
xcmw, I think you duyplicated the cabal file as the stack.yaml
geekosaur 2017-02-08 16:12:50
if that's really your stack.yaml then stack is likely *really* confused >.>
mniip 2017-02-08 16:13:12
erisco, I don't think you can promote constraints in any way
lpaste_ 2017-02-08 16:13:19
xcmw revised "Why does stack fail to build reactive banana and wxwidgets?": "Why does stack fail to build reactive banana and wxwidgets?" at http://lpaste.net/8538512301479690240
mniip 2017-02-08 16:13:22
you'd get impredicative kind polymorphism otherwise
lpaste_ 2017-02-08 16:13:47
xcmw revised "Why does stack fail to build reactive banana and wxwidgets?": "Why does stack fail to build reactive banana and wxwidgets?" at http://lpaste.net/8538512301479690240
xcmw 2017-02-08 16:14:16
There is nothing in the log not included in the error messsage
geekosaur 2017-02-08 16:14:17
also this looks like one of those cases where stack "helpfully" wrote the error output somewhere other than the terminal
geekosaur 2017-02-08 16:14:42
or maybe one of the corner cases where it forgets to log anything (make sure you;re running latest stack though, thought they fixed those)
geekosaur 2017-02-08 16:15:28
I am however guessing it's the Setup.hs that is failing to build, likely because you are missing the Wx devel libs for your platform
geekosaur 2017-02-08 16:15:47
(runtime libs are not enough, if you are on a fedora/rhel/centos or debian/ubuntu/mint)
xcmw 2017-02-08 16:16:14
I'm on osx
xcmw 2017-02-08 16:16:58
I installed wxmac from homebrew
geekosaur 2017-02-08 16:17:10
might need to tell it to look under /usr/local for wx then
erisco 2017-02-08 16:17:21
glguy, jle` equality or ordering on Fin seems odd. Any ideas?
geekosaur 2017-02-08 16:17:29
--extra-lib-dirs=/usr/local/lib --extra-include-dirs=/usr/local/include
erisco 2017-02-08 16:17:45
is it just inappropriate for either?
geekosaur 2017-02-08 16:17:46
(this usually can *not* be put in a cabal file. no idea if you can put it in stack.yaml)
erisco 2017-02-08 16:19:16
Data.Type.Fin has a * and * -> * kinded equality and ordering. Maybe it is best to just define both then
GreySunshine 2017-02-08 16:19:41
Hello!
erisco 2017-02-08 16:19:53
seems a bit hectic with adding these higher kinded classes but... okay
GreySunshine 2017-02-08 16:22:39
Guys I was wondering if xmonad is tested for properties, It is mathematically sound, then does it not need bug fixes? Is it in maintenance? It is tough to wrap around my this!
lpaste_ 2017-02-08 16:23:05
xcmw revised "Why does stack fail to build reactive banana and wxwidgets?": "Why does stack fail to build reactive banana and wxwidgets?" at http://lpaste.net/8538512301479690240
xcmw 2017-02-08 16:23:16
geekosaur: Added that same error
erisco 2017-02-08 16:23:37
though... what is reflexivity? normally I say eqRefl :: forall (a :: k). a := a
erisco 2017-02-08 16:23:48
but if there are two kinds what would that be
geekosaur 2017-02-08 16:23:48
only some things can be tested. most stuff can't be because you would have to mathematically model an X server, including the bugs inevitably present in nvidia's drivers...
geekosaur 2017-02-08 16:24:31
xcmw, at this point I think you have to convince stack to tell you the actual error.
erisco 2017-02-08 16:24:55
I am guessing it has to be something like eqRefl :: forall (a :: F j) (b :: F k). (a :== b) := (b :== a)
erisco 2017-02-08 16:25:19
where :== is the equality on * -> * sorted kinds
geekosaur 2017-02-08 16:25:19
you might have to open a stack bug about it not printing or logging any errors
erisco 2017-02-08 16:25:44
and := is equality on * i.e. it is (a :== b -> b :== a, b :== a -> a :== b)
xcmw 2017-02-08 16:25:54
geekosaur: It logged the error
xcmw 2017-02-08 16:26:09
[1 of 1] Compiling Main ( /private/var/folders/_m/1h8dd_fd3dx3qvtq87_c4q480000gn/T/stack74916/wxcore-0.11.1.2/Setup.lhs, /private/var/folders/_m/1h8dd_fd3dx3q
erisco 2017-02-08 16:26:11
certainly gets a bit weird
xcmw 2017-02-08 16:26:12
vtq87_c4q480000gn/T/stack74916/wxcore-0.11.1.2/.stack-work/dist/x86_64-osx/Cabal-1.24.0.0/setup/Main.o )
xcmw 2017-02-08 16:26:15
Linking /private/var/folders/_m/1h8dd_fd3dx3qvtq87_c4q480000gn/T/stack74916/wxcore-0.11.1.2/.stack-work/dist/x86_64-osx/Cabal-1.24.0.0/setup/setup ...
xcmw 2017-02-08 16:26:18
error: Unknown option "--sysconfdir=/Users/michaelmesser/Code/gui-test/.stack-work/install/x86_64-osx/lts-7.19/8.0.1/etc". Use "--help" to show valid options.
geekosaur 2017-02-08 16:26:27
o.O
geekosaur 2017-02-08 16:26:39
you have a cabal library mismatch, at minimum
geekosaur 2017-02-08 16:27:13
but you'll have to take this up with the stack devs probably
Welkin 2017-02-08 16:27:23
sorted kinds
Welkin 2017-02-08 16:27:27
I've never heard that before
Welkin 2017-02-08 16:27:35
there is only 1 sort in haskell I thought
Welkin 2017-02-08 16:27:37
BOX
xcmw 2017-02-08 16:27:53
geekosaur: Ok
geekosaur 2017-02-08 16:28:00
Welkin, not true since 7.10
Welkin 2017-02-08 16:28:04
oh?
Welkin 2017-02-08 16:28:22
where can I read about it?
geekosaur 2017-02-08 16:28:29
I think the levity stuff took its place, but it's stll a bit in flux
MarcelineVQ 2017-02-08 16:29:04
Welkin: there's a large section for things related to it in the manual https://downloads.haskell.org/~ghc/latest/docs/html/users_guide/glasgow_exts.html#kind-polymorphism-and-type-in-type
geekosaur 2017-02-08 16:29:27
a simple version in 7.10, a more complex one in 8.0.x, and they;re still working on it with more changes coming in 8.2 and likely in 8.4 because some stuff didn't make the cutoff
Welkin 2017-02-08 16:30:08
this makes me want to get back into learning about type systems and category theory again
MarcelineVQ 2017-02-08 16:30:41
Welkin: large as-in 9.10-9.15 are probably all relevant
Welkin 2017-02-08 16:33:20
thanks MarcelineVQ
erisco 2017-02-08 16:39:23
type family Sing' :: (a :: k) -> (b :: Sing a) * Expected a type, but `b' has kind `Sing a'
erisco 2017-02-08 16:39:25
hrm darn
erisco 2017-02-08 16:40:57
it doesn't even like (a :: k)
mniip 2017-02-08 16:41:36
erisco, that's the wrong ->
mniip 2017-02-08 16:41:48
you probably want
mniip 2017-02-08 16:41:57
type family Sing' (a :: k) :: Sing a
mniip 2017-02-08 16:41:59
or something
erisco 2017-02-08 16:42:19
oh I see, thanks
mniip 2017-02-08 16:42:34
what you're saying is that Sing' is of kind a -> b
mniip 2017-02-08 16:42:38
where a :: k, b :: Sing a
mniip 2017-02-08 16:43:04
or something
erisco 2017-02-08 16:44:06
type instance Sing' ('Z :: Nat) = ('SZ :: Sing 'Z) * Data constructor `SZ' cannot be used here (it comes from a data family instance)
erisco 2017-02-08 16:44:24
really? :s
