|
Boost : |
Subject: Re: [boost] [review] The review of Boost.DoubleEnded starts today: September 21 - September 30
From: Thorsten Ottosen (tottosen_at_[hidden])
Date: 2017-10-09 08:30:25
Den 06-10-2017 kl. 11:53 skrev Thorsten Ottosen via Boost:
> Den 05-10-2017 kl. 22:42 skrev Ion Gaztañaga via Boost:
>> On 05/10/2017 19:02, Thorsten Ottosen via Boost wrote:
>>> I was thinking along generic code along the line
>>>
>>> template< class BacKInsertionContiguousContainer >
>>> void foo( BacKInsertionContiguousContainer& cont )
>>> {
>>> Â Â Â cont.reserve( source.size() );
>>> Â Â Â for( ... )
>>> Â Â Â Â Â Â cont.push_back( ... );
>>>
>>> }
>>>
>>> why should that not work with both devector and vector?
>>
>> It would work, but it would not be as efficient as vector.
>
> It may be. But I guess my concern was to make it compile. Sure, if
> a call to reserve is conditioned on d.capacity() - d.size(), it can
> behave differently. Reserve maybe easier to do something about.
What if we did this:
capacity() -> alias for back_capacity
back_capacity() -> return size() + back_free_capacity()
front_capacity() -> return size() + front_free_capacity()
?
kind regards
Thorsten
Boost list run by bdawes at acm.org, gregod at cs.rpi.edu, cpdaniel at pacbell.net, john at johnmaddock.co.uk