June 2004 - Cvs-nhc98

cvs commit: nhc98 fixCVS
by Malcolm Wallace 13 Jun '04

13 Jun '04

malcolm 2004/06/13 06:30:46 PDT Modified files: . fixCVS Log: Fixing hsc2hs does not depend on the presence of cpphs. Revision Changes Path 1.3 +1 -1 nhc98/fixCVS

1 0

cvs commit: nhc98/src/runtime/Mk mkTuple13.c mkTuple14.c mkTuple15.c Makefile
by Sven Panne 13 Jun '04

13 Jun '04

panne 2004/06/13 06:07:37 PDT Modified files: src/runtime/Mk Makefile Added files: src/runtime/Mk mkTuple13.c mkTuple14.c mkTuple15.c Log: More 15-tuple stuff Revision Changes Path 1.6 +2 -1 nhc98/src/runtime/Mk/Makefile

1 0

cvs commit: nhc98/include mkByProg.h
by Sven Panne 13 Jun '04

13 Jun '04

panne 2004/06/13 06:07:36 PDT Modified files: include mkByProg.h Log: More 15-tuple stuff Revision Changes Path 1.7 +13 -1 nhc98/include/mkByProg.h

1 0

*.hi files
by Sven Panne 13 Jun '04

13 Jun '04

I have trouble understanding the relationship between the .hi file below nhc98/include/ and the ones of the same name generated during the build process (below nhc98/src/prelude, IIRC). Could somebody please enlighten me? My concrete problem is that I've changed CTypesExtra (types => newtypes and added quite a few instances), but the changes don't seem to have any effect for the nhc98 which is being built. :-( Cheers, S.

2 3

Mysterious message...
by Sven Panne 13 Jun '04

13 Jun '04

During my HOpenGL crusade against nhc98, I came across heavy resistance: :-] /usr/src/packages/BUILD/nhc98-1.17/script/nhc98 -c -cpp -I/usr/src/packages/BUILD/nhc98-1.17/include/packages/OpenGL -package base -Iinclude -DCALLCONV=ccall -o /usr/src/packages/BUILD/nhc98-1.17/targets/ix86-Linux/obj/libraries/OpenGL/Graphics/Rendering/OpenGL/GLU/Matrix.o Graphics/Rendering/OpenGL/GLU/Matrix.hs Fail: Can't find int for (NHC.Internal._mkIOok14,Identifier) make: *** [/usr/src/packages/BUILD/nhc98-1.17/targets/ix86-Linux/obj/libraries/OpenGL/Graphics/Rendering/OpenGL/GLU/Matrix.o] Fehler 1 What is nhc98 trying to tell me here? %-) Cheers, S.

2 2

cvs commit: nhc98/include/NHC Internal.hi
by Malcolm Wallace 13 Jun '04

13 Jun '04

malcolm 2004/06/13 05:35:15 PDT Modified files: include/NHC Internal.hi Log: Extend _mkIOok[0..12] up to arity 15. Revision Changes Path 1.3 +9 -0 nhc98/include/NHC/Internal.hi

1 0

cvs commit: nhc98/src/prelude/Internal Internal.hs
by Malcolm Wallace 13 Jun '04

13 Jun '04

malcolm 2004/06/13 05:35:14 PDT Modified files: src/prelude/Internal Internal.hs Log: Extend _mkIOok[0..12] up to arity 15. Revision Changes Path 1.3 +16 -0 nhc98/src/prelude/Internal/Internal.hs

1 0

cvs commit: nhc98/include/NHC FFI.hi
by Malcolm Wallace 13 Jun '04

13 Jun '04

malcolm 2004/06/13 03:46:41 PDT Modified files: include/NHC FFI.hi Log: Add missing exports: realloc, finalizerFree. Revision Changes Path 1.10 +6 -0 nhc98/include/NHC/FFI.hi

1 0

cvs commit: nhc98/src/prelude/FFI FFI.hs
by Malcolm Wallace 13 Jun '04

13 Jun '04

malcolm 2004/06/13 03:42:56 PDT Modified files: src/prelude/FFI FFI.hs Log: Missing exports: realloc, finalizerFree. Revision Changes Path 1.19 +2 -0 nhc98/src/prelude/FFI/FFI.hs

1 0

Type checking bug in nhc98
by Sven Panne 13 Jun '04

13 Jun '04

Consider the following module: ------------------------------------ module Foo where bar :: Int -> (Int, Int) bar z = (z, len) where (len, _) = baz 0 baz n = (n, []) ------------------------------------ The current nhc98 from HEAD tells me: ====== Errors after type inference/checking: Context for Prelude.Num needed in left hand pattern at 5:10. GHC and Hugs accept this without problems, nhc98 only when it gets some help, e.g. changing "0" to "(0::Int)". Cheers, S.

1 0