Re: [boost] auto_vector - vector with auto_ptr semantics

From: David Abrahams (david.abrahams_at_[hidden])
Date: 2002-02-04 10:02:49

----- Original Message -----
From: "Brey, Edward D" <EdwardDBrey_at_[hidden]>

> This is interesting and new to me.  Would this be helpful for the removing
> algorithms, or just the sorting algorithms?

Well, instead of remove_... you'd want to use a form of partitioning. It's
similar, but not exactly the same. Is partition a sorting algorithm? ;^)

> Currently, unlike the sorting algorithms, the STLport removing algorithms
> are implemented in terms of operator=.  It might be that auto_vector would
> have problems with algorithms even beyond the "fixing swap" topic.

Yes, algorithms that use op=() would incur a penalty unless your indirect
adaptor was adapting a sequence of shared_ptrs. That's one reason I favor a
flexible framework of adaptors for this purpose.

-Dave