From: Dean M. B. <mik...@gm...> - 2010-11-28 05:54:09
|
On Sun, Nov 28, 2010 at 1:47 PM, Dean Michael Berris <mik...@gm...> wrote: > Hi Everyone, > > So today I revisited my implementation of the directives and it looks > like using Boost.Variant in the implementation seems to be one of the > biggest contributors to compile time -- well, second to Boost.Spirit > -- and removing it from the directive implementations seems to be > going well. > > I'll be pushing soon so expect a greatly reduced compile-time for the tests now. > Well, I jumped the gun on "greatly reduced". It's a little faster than usual. I keep forgetting that clang is just really way faster than GCC. But, it's niticeably quicker on GCC 4.5. Not sure with other compilers though. :D HTH! -- Dean Michael Berris deanberris.com |