[jhc] Re: apparent jhc bug.

[David Roundy]

I see now what this bug comes from.  It's from foreign imports that
use the wrong IO.  So it's not correct code that's being flagged here,
but it's also a pretty unhelpful error message for the case where a
foreign import is declared with an inappropriate type.

David

On Mon, Sep 7, 2009 at 8:58 PM, David
Roundy<roundyd at physics.oregonstate.edu> wrote:
> Hi John,
>
> I'm getting the following failure, which looks to me like a compiler
> bug.  I've cleared my cache, and rebuilt jhc from scratch, so this
> time I think it's a real bug.
>
> Compiling...
> [ 1 of 13] System.C.Stdio   user error (lookupExtTypeInfo: (ELit
> ((#2#) (ELit (Jhc.Prim.World__::ESort #)) (ELit
> (Foreign.C.Types.CInt::ESort *))::ESort #),ELit ((#2#) (ELit
> (Jhc.Prim.World__::ESort #)) (ELit (Foreign.C.Types.CInt::ESort
> *))::ESort #)))
>
> The code is unfortunately somewhat large.  You can (once again) get it at
>
> darcs get http://abridgegame.org/concurrency
>
> You can either build it using a very recent franchise and ./Setup.hs
> configure --jhc, or you could use the attached base-alt.config, and
> compile with something like
>
> jhc -ijhc -ibase -ialt --noauto -isrc --dependencies base-alt.jhc-deps
> --build-hl base-alt.config
>
> As you may be able to gather, I've not yet been able to kick the habit
> of hacking on jhc's library.
>
> It's possible that this error is due to a bug I've introduced (I
> certainly introduced plenty along the way), but since it type checked,
> I think it ought to succeed in the compile stage.
> --
> David Roundy
>



-- 
David Roundy