|
Boost : |
From: Jeremy Siek (jsiek_at_[hidden])
Date: 2001-06-28 09:15:22
On Thu, 28 Jun 2001 jaakko.jarvi_at_[hidden] wrote:
> I did some testing and compiled a program consisiting solely of
> two argument tie expressions like:
> tie(i, j) = make_pair(1, 2)
>
> Compile time increased around 9% with gcc 3.0 when going from the tie
> function in utility.hpp to the tuple tie.
>
> As this was a program with around 5000 tie expressions, and nothing
> else, I'd suspect that in realistic programs the difference is
> unnoticeable.
Great, thanks for checking that!
> The other thing is the complexity and portability issue, less
> compilers support tuples than utility.hpp, so maybe the
> two argument tie could be preserved somehow (different name,
> different namespace, prevent two different definitions of
> tie(A, B) with #define guards).
Hmm, does the tuple tie() work with VC++ and Borland? The graph library
currently works (mostly) with these compilers and it would be a shame to
lose that portability just for tie().
Cheers,
Jeremy
----------------------------------------------------------------------
Jeremy Siek www: http://www.lsc.nd.edu/~jsiek/
Ph.D. Candidate, IU B'ton email: jsiek_at_[hidden]
Summer Manager, AT&T Research phone: (973) 360-8185
----------------------------------------------------------------------
Boost list run by bdawes at acm.org, gregod at cs.rpi.edu, cpdaniel at pacbell.net, john at johnmaddock.co.uk