questions about ghc-clang-wrapper

I've just run the ghc-clang-wrapper script (from https://gist.github.com/tibbe/8448715) to try to get GHC 7.6.3 (part of Haskell Platform 2013-2.0.0) working with XCode 5.0.2 on a MacOS 10.8.5 machine. Things look good, but I have two questions: Question 1: How do I know this worked? I tested it by modifying a package that I'm writing so that it requires CPP to build correctly, and successfully built it with "cabal clean; cabal configure; cabal build". (More details available on request.) Does that mean I'm in good shape? (I'm asking because I was never entirely clear on what the incompatibility was; all I know is that it has something to do with cpp.) Question 2: Is there any harm in adding -Wno-return-type to extraClangArgs in ghc-clang-wrapper? This silences an annoying but apparently harmless warning I get from "cabal configure". Thanks, Richard

Just use GCC and follow the alternative directions for getting ghc to point
at a real GCC. They're linked to from the Mac Haskell platform page as the
alternative directions.
On Saturday, March 1, 2014, Richard Cobbe
I've just run the ghc-clang-wrapper script (from https://gist.github.com/tibbe/8448715) to try to get GHC 7.6.3 (part of Haskell Platform 2013-2.0.0) working with XCode 5.0.2 on a MacOS 10.8.5 machine. Things look good, but I have two questions:
Question 1: How do I know this worked? I tested it by modifying a package that I'm writing so that it requires CPP to build correctly, and successfully built it with "cabal clean; cabal configure; cabal build". (More details available on request.) Does that mean I'm in good shape?
(I'm asking because I was never entirely clear on what the incompatibility was; all I know is that it has something to do with cpp.)
Question 2: Is there any harm in adding -Wno-return-type to extraClangArgs in ghc-clang-wrapper? This silences an annoying but apparently harmless warning I get from "cabal configure".
Thanks,
Richard _______________________________________________ Haskell-Cafe mailing list Haskell-Cafe@haskell.org javascript:; http://www.haskell.org/mailman/listinfo/haskell-cafe

On Sat, Mar 1, 2014 at 7:35 PM, Richard Cobbe
Question 1: How do I know this worked? I tested it by modifying a package that I'm writing so that it requires CPP to build correctly, and successfully built it with "cabal clean; cabal configure; cabal build". (More details available on request.) Does that mean I'm in good shape?
If you didn't get a compile failure after a bunch of syntax errors reported by cpp, you're good. (The problem is that clang's cpp is more pedantic about C syntax than gcc's, and doesn't appreciate being run on Haskell source. The patch adds options to disable the pedanticism.) -- brandon s allbery kf8nh sine nomine associates allbery.b@gmail.com ballbery@sinenomine.net unix, openafs, kerberos, infrastructure, xmonad http://sinenomine.net

woops, yeah, brandon is correct.
you're fine
On Sat, Mar 1, 2014 at 7:44 PM, Brandon Allbery
On Sat, Mar 1, 2014 at 7:35 PM, Richard Cobbe
wrote: Question 1: How do I know this worked? I tested it by modifying a package that I'm writing so that it requires CPP to build correctly, and successfully built it with "cabal clean; cabal configure; cabal build". (More details available on request.) Does that mean I'm in good shape?
If you didn't get a compile failure after a bunch of syntax errors reported by cpp, you're good. (The problem is that clang's cpp is more pedantic about C syntax than gcc's, and doesn't appreciate being run on Haskell source. The patch adds options to disable the pedanticism.)
-- brandon s allbery kf8nh sine nomine associates allbery.b@gmail.com ballbery@sinenomine.net unix, openafs, kerberos, infrastructure, xmonad http://sinenomine.net
_______________________________________________ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
participants (3)
-
Brandon Allbery
-
Carter Schonwald
-
Richard Cobbe