I'd like to propose eliminating the option of using the old library setup, and then dismantling it. (The new setup includes compatibility stubs.)
A branch has been created for Nov2002, so if people could
commit their changes to both the main trunk and that
release branch (if both apply), then I don't see a problem
with making the changes you suggest on the main trunk.
--sigbjorn
----- Original Message -----
From: "Ross Paterson"
I'd like to propose eliminating the option of using the old library setup, and then dismantling it. (The new setup includes compatibility stubs.) _______________________________________________ Cvs-hugs mailing list Cvs-hugs@haskell.org http://www.haskell.org/mailman/listinfo/cvs-hugs
A branch has been created for Nov2002, so if people could commit their changes to both the main trunk and that release branch (if both apply), then I don't see a problem with making the changes you suggest on the main trunk.
Sounds good. There's a bunch of changes I'd like to make including: - get Float == float, Double == double - strip out greencard support since the ffi is sufficient - cleanup internal names used for all the ffi-related primitives to match the names used in the ffi spec. - add a few ffi-mandated C functions that Sven(?) pointed out (hs_free_funptr I think - maybe one or two others) Mind you, the next 2 months see me moving home, getting married and drumming up more consulting work so I may not get round to it right away... -- Alastair
On Mon, Dec 09, 2002 at 09:28:17PM +0000, Alastair Reid wrote:
Sounds good. There's a bunch of changes I'd like to make including:
[...] - strip out greencard support since the ffi is sufficient
I think I'd find this one the most useful -- I'm never sure which of those layers need to keep working.
participants (3)
-
Alastair Reid -
Ross Paterson -
Sigbjorn Finne