- Cvs-hugs - Haskell.org

cvs commit: hugs98/tests/libs except1.output hugs98/tests/rts ioerror1.output
by ross＠glass.cse.ogi.edu 22 Jul '03

22 Jul '03

ross 2003/07/22 04:53:32 PDT Modified files: tests/libs except1.output tests/rts ioerror1.output Log: update test output for the new Show for IOError Revision Changes Path 1.10 +3 -6 hugs98/tests/libs/except1.output 1.9 +2 -4 hugs98/tests/rts/ioerror1.output

1 0

cvs commit: hugs98/demos Ldfs.hs
by ross＠glass.cse.ogi.edu 22 Jul '03

22 Jul '03

ross 2003/07/22 03:33:43 PDT Modified files: demos Ldfs.hs Log: remove the graph stuff (now in Data.Graph and Data.Tree) from the demo Revision Changes Path 1.2 +17 -229 hugs98/demos/Ldfs.hs

1 0

proposed changes in module search behaviour
by Ross Paterson 22 Jul '03

22 Jul '03

I propose to make two changes to Hugs's module search behaviour: - remove the A.B.C.hs feature: no-one seems to be using it, GHC doesn't do it, and it's doubling the number of files tried. - don't search the current directory, unless it's explicitly given in the path (but put it at the front of the default path to preserve the current behaviour). At least people could then turn it off if required. Note that this relates only to resolving module names -- file names don't use the path. Also, when a file name is specified, the directory containing it is searched first for any modules imported by that file (and other modules found in that directory).

3 2

cvs commit: hugs98/libraries/Hugs Prelude.hs
by ross＠glass.cse.ogi.edu 21 Jul '03

21 Jul '03

ross 2003/07/21 11:03:01 PDT Modified files: libraries/Hugs Prelude.hs Log: copy new instance of Show IOException (aka IOError) from GHC. Revision Changes Path 1.29 +11 -12 hugs98/libraries/Hugs/Prelude.hs

1 0

cvs commit: hugs98/src prelude.h
by ross＠glass.cse.ogi.edu 21 Jul '03

21 Jul '03

ross 2003/07/21 03:46:48 PDT Modified files: src prelude.h Log: fix gcc version test Revision Changes Path 1.48 +3 -3 hugs98/src/prelude.h

1 0

foreign import "wrapper" bug on x86 platforms
by Sven Panne 21 Jul '03

21 Jul '03

FYI: I've found the reason why tests/ffi/Callback.hs fails on Intel platforms: The x86 part of mkThunk in src/builtin.c is broken. It simply squeezes an additional argument below the return address (the stable pointer for calling back later), but this word is never popped. We should probably include something like __obscure_ccall_ret_code from GHC's fptools/ghc/rts/Adjustor.c, but it is a bit late now for such severe stack juggling... Cheers, S.

1 0

cvs commit: hugs98/src/unix aclocal.m4 configure hugs-package.in
by Sven Panne 20 Jul '03

20 Jul '03

panne 2003/07/20 09:44:29 PDT Modified files: src/unix aclocal.m4 configure hugs-package.in Log: Synched with latest changes for GLUT libraries. Revision Changes Path 1.15 +18 -9 hugs98/src/unix/aclocal.m4 1.65 +18062 -7160hugs98/src/unix/configure 1.8 +2 -0 hugs98/src/unix/hugs-package.in

1 0

cvs commit: hugs98/src/unix Makefile
by Sven Panne 20 Jul '03

20 Jul '03

panne 2003/07/20 09:38:45 PDT Modified files: src/unix Makefile Log: Improved the command for nuking the configure cache a bit Revision Changes Path 1.7 +1 -1 hugs98/src/unix/Makefile

1 0

cvs commit: hugs98/tests/rts arith.output enum.output1 enum.output2 read.output
by ross＠glass.cse.ogi.edu 18 Jul '03

18 Jul '03

ross 2003/07/18 06:43:51 PDT Modified files: tests/rts arith.output enum.output1 enum.output2 read.output Log: update output, now that we're printing with more precision Revision Changes Path 1.7 +11 -11 hugs98/tests/rts/arith.output 1.8 +3 -3 hugs98/tests/rts/enum.output1 1.8 +3 -3 hugs98/tests/rts/enum.output2 1.5 +2 -2 hugs98/tests/rts/read.output

1 0

cvs commit: hugs98/src/unix hugs-package.in
by Sven Panne 18 Jul '03

18 Jul '03

panne 2003/07/18 01:41:37 PDT Modified files: src/unix hugs-package.in Log: Added a REAL hack for HOpenGL stuff. Probably somebody will hit me with a big stick for doing this... :-] Improvements highly welcome! Revision Changes Path 1.7 +16 -1 hugs98/src/unix/hugs-package.in

2 1