[PATCH] D18579: SetVector: Add front, pop_front and additional constructor.

[David Blaikie via llvm-commits]

Yeah, not sure - wouldn't mind some other voices here to see how people
feel.

Reserve isn't part of any STL concept so far as I can tell, so, yes, to
that degree it is an implementation detail.

On Mon, Apr 4, 2016 at 11:03 AM, Matthias Braun via llvm-commits <
llvm-commits at lists.llvm.org> wrote:

> MatzeB added a comment.
>
> In http://reviews.llvm.org/D18579#391254, @dblaikie wrote:
>
> > That seems like a somewhat strange API - calling reserve down into the
> >  vector (& possibly the set, for things like DenseSet) seems more
> >  appropriate, no?
>
>
> After the discussions here I felt that reserve()/resize() are details of
> the underlying set not necessarily a common agreed interface supported by
> all sets, what if the next set implementation has additional methods for
> tuning? An alternative to that strange constructor would be to expose the
> set object so people can call reserve()/resize() on it directly though it
> breaks isolation.
>
>
> Repository:
>   rL LLVM
>
> http://reviews.llvm.org/D18579
>
>
>
> _______________________________________________
> llvm-commits mailing list
> llvm-commits at lists.llvm.org
> http://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-commits
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20160404/5f95ce6e/attachment.html>