Hi,
On Sat, Jan 17, 2009 at 8:38 PM, Samy Al Bahra
I suggest that we stick to having individual maintainers for now and switch maintainers once bsd.haskell.mk is ready to a private mailing list consisting of us initial porters.
Okay.
I think this approach could even make us re-think whether a cabal2port converter is really needed.
No, it doesn't. Unless we have implemented a query-complete *.cabal parser in bsd.haskell.mk, there will remain to be massive productivity boosts with such a thing (ones much greater than having a bsd.haskell.mk when it comes to Cabal).
Okay. If I have some time, I will try to experiment with this .mk, and send a proposed version to this list.
We will make it easy for you to generate patches on a per-port basis and per-branch based on dependency. Since none of us are committers, how exactly you decide to merge is irrelevant to us as long as it doesn't involve breaking things. :-P
Okay. By the way, I am on the haskell@freebsd.org "list" by now ;P I assume it also means that I can maintain the ports assigned to this "identity".
Let us get some serious work under our belt before advertising our efforts.
No problem, I expected this answer.
- GHC 6.8.10: What is up with GHC? What are the exact obstacles to get the latest version ported? Has anybody tried it? (I am just curious.)
This is a non-trivial issue. This will be discussed in a separate thread as to keep requirements/directions crystal clear. I suggest an overhaul of how these ports are structured to begin with in order to save ports which do not build or function properly with GHC 6.10.
Okay. Cheers, :g