From: David Abrahams (dave_at_[hidden])
Date: 2006-10-20 20:48:29
Herve Bronnimann <hervebronnimann_at_[hidden]> writes:
> On Monday, October 16, 2006, at 02:39PM, David Abrahams <dave_at_[hidden]> wrote:
>>Herve Bronnimann <hervebronnimann_at_[hidden]> writes:
>>>> My question is: Is there something I am missing that would create
>>> problems in the usage of Boost.bind, if the second set of overloads
>>> were added (besides the inconvenience of more overloads and a longer
>>> source code to parse for the compiler)?
>>Seems like it could work for the cases where rvalues don't get passed
>>with non-const lvalues.
> I'm not sure if you are saying that it would work in cases where it
> should have, but didn't, work before; or whether it would compile in
> cases where it shouldn't.
I'm not saying what should or shouldn't work; I'm just saying it seems
as though your scheme could work for some cases (the ones you outlined).
> If you are you saying there are such cases where this would allow
> illegal bindings, then could you please provide a concrete example?
I'm not saying that.
-- Dave Abrahams Boost Consulting www.boost-consulting.com
Boost list run by bdawes at acm.org, david.abrahams at rcn.com, gregod at cs.rpi.edu, cpdaniel at pacbell.net, john at johnmaddock.co.uk