From: Gary Powell (Gary.Powell_at_[hidden])
Date: 2001-02-12 12:38:21
>>I agree with you that a lambda function would probably be better.
> My question is: Is it worth dragging in an entire library [heavily
> templatized (It matters, at least for VC++ users)] for the sake of
> this one thing?
Yep! Once I use LL for one thing, I seem to have a
need for it again. :>
>>Plus, for orthogonality in the STL, we should have
> binders for member data as well as member functions.
LL includes generic binders as well. Member functions use the same syntax.
If you are stuck with VC++, I think Peter Dimov's expression library works
with it, and Peter Higley and my old ET library works as well (May 2000 C++
Report). ET requires that you declare the types of the arguments before you
use them, but it's not a horrible task.
On the other hand, I can see the value in some interesting and well
documented template tools like the one you are proposing. I just see them
going away when VC gets partial specialization and Koeing lookup.
Boost list run by bdawes at acm.org, gregod at cs.rpi.edu, cpdaniel at pacbell.net, john at johnmaddock.co.uk