On Mon, Apr 14, 2003 at 11:18:39AM +0100, Alastair Reid wrote:
ross 2003/04/11 16:39:44 PDT
Modified files: libraries/base/Foreign/C Error.hs libraries/base/System/Posix Internals.hs libraries/network/Network Socket.hsc
Log: Hugs only: add a CBITS "pragma" to tell the Hugs conversion script which files from <package>/cbits should be linked with the module by ffihugs, e.g. {-# CBITS HsNet.c initWinSock.c ancilData.c #-} [...] Is this the best way to handle cbits though?
It's the best I could think of, but I'm not emotionally attached to it.
2) Can the problem of associating C files with Haskell ffi files be solved in a way that all compilers (and associated tools, makefiles, etc.) can support and exploit?
(e.g., other compilers support the CBITS pragma, or an ffi extension, or a package configuration file extension, or a text file which can be read both by GHC makefiles and by your package conversion script, etc.)
It seemed to me that Hugs is quite different from the compilers in this respect: they link all the cbits stuff together in a library, with which the whole package is linked, while Hugs links each module separately, so only Hugs needs an association between modules and C files.