Core package: attempt #1
bulat.ziganshin at gmail.com
Tue Oct 3 11:27:00 EDT 2006
Tuesday, October 3, 2006, 6:26:42 PM, you wrote:
>> i've uploaded my first attempt to define Core package as
> Cool. The one thing that worries my slightly is that HugsCore.Bool
> imports Prelude, and surely Prelude will end up importing
> HugsCore.Bool at some point? It seems that Core needs to not depend on
> Prelude, so that the Prelude itself can be written in a cross-compiler
> friendly way.
it's just a temporary hack - prim* operations returns result of type
Prelude.Bool or Prelude.Ordering and i can't make anything with it
unless i write the base library itself. for GHC i use the same
technique for Bool - you can find in GHcCore.Bool two definitions -
that will be used in final library and temporary, which imports
existing Bool definition
>> as you can see, compiler-specific definitions are very short here,
>> much of code is in compiler-independent modules. i hope that adding yhc
>> and jhc support should be rather simple
> As soon as you have a darcs repo and can make some headway towards
> compiling base, the Yhc team will submit an appropriate CoreYhc.
i will be happy to create repository, but it's only one variant of
design so the other versions may have nothing common with this one
but i have to say that this work will need your permanent intervention
- APIs should be developed in the way that will allow to implement them in
yhc. if substantial part of core library will be developed only with
GHC and Hugs in mind, you will got the same problems as you have with 'base' now :)
btw, can i download yhc for windows plus sources of yhc's base library?
i will go to check now..
Bulat mailto:Bulat.Ziganshin at gmail.com
More information about the Glasgow-haskell-users