Re: [boost] [Review] Formal Review of Proposed Boost.Ratio Library

Subject: Re: [boost] [Review] Formal Review of Proposed Boost.Ratio Library
From: Howard Hinnant (howard.hinnant_at_[hidden])
Date: 2010-10-07 09:27:54

On Oct 7, 2010, at 8:40 AM, Stewart, Robert wrote:

> vicente.botet wrote:
>> From: "Stewart, Robert" <Robert.Stewart_at_[hidden]>
>> 
>>> The use of enable_if to control the contexts in which the
>>> copy constructor and copy assignment operator apply goes
>>> beyond the standard's specification.  That means boost::ratio
>>> behaves differently than will std::ratio.  I think this will
>>> lead to surprising results when one transitions from one to
>>> the other.
>> 
>> I 've made a request to add these constructors and this
>> request has not ben decided yet. I thinh the best for Bost
>> could be a flagf that intruduce this feature or not. So i
>> prupose that Boost.Ration includes thes constructors and
>> assignemes conditionally.
> 
> As I read the LGW issue, there was no support for adding that
> behavior.

This is correct.  However this is a good place to add that "no consensus" in the LWG issues list notes does not mean "no one in favor".  I argued in favor of this change as did Walter Brown.  As I recall people were mainly nervous about the timing of this change.  It was discussed post FCD.

I would be tempted to consider this extension to the standard semantics based on its technical merits alone, motivated by Beman's more general policy statement:

On Oct 2, 2010, at 8:11 AM, Beman Dawes wrote:

> One of the advantages of Boost is that we can add extensions and get
> user experience before something gets standardized. That's very
> helpful to the C++ committee.

-Howard