On Mon, Oct 29, 2012 at 9:51 AM, Kirk, Benjamin (JSC-EG311) <benjamin.kirk-1@nasa.gov> wrote:

On Oct 22, 2012, at 11:05 AM, Cody Permann <codypermann@gmail.com> wrote:

> I like the "include both, selectable at configure-time" option best,
> if it's not too much work for Cody.  In exchange I can finally get
> around to reproducing/fixing that Nemesis regression he found...
> Sounds good to me!  Based on my result table that I sent a few minutes ago I think we're all ready to be done with this issue.  For us, using the release package will fix all of these issues.  Only the truly hardcore will want to mess with the development version anyway.

I hate to bring this up, but fparser is broken with gcc-4.7 using -std=c++0x.

On trunk, in m4/configure.m4 we add -std=c++0x only for gcc-4.3 (erroneously?).  if you update that switch statement to add the flag for gcc-4.7, fparser breaks.

Cody, could you verify on your system?

Ben,  I just manually added those flags on my system and it's working fine, both the devel and release versions.  I have GCC 4.6 but I'd be surprised if the issue was with 4.7 only.  What error are you seeing?


This gets to the larger question of when we will require C++11, which in my mind is probably a good year or so from now.  In the meantime, should we bother trying to track partially implemented subsets?