[cfe-commits] [libcxx] r119611 - /libcxx/trunk/include/algorithm

[Douglas Gregor]

On Nov 18, 2010, at 8:18 AM, Howard Hinnant wrote:

> On Nov 18, 2010, at 11:13 AM, Douglas Gregor wrote:
> 
>> 
>> On Nov 17, 2010, at 6:48 PM, Howard Hinnant wrote:
>> 
>>> On Nov 17, 2010, at 9:45 PM, Sean Hunt wrote:
>>> 
>>>> On 10-11-17 08:47 PM, Howard Hinnant wrote:
>>>>> Author: hhinnant
>>>>> Date: Wed Nov 17 19:47:02 2010
>>>>> New Revision: 119611
>>>>> 
>>>>> URL: http://llvm.org/viewvc/llvm-project?rev=119611&view=rev
>>>>> Log:
>>>>> LWG 1432
>>>>> 
>>>>> Modified:
>>>>>  libcxx/trunk/include/algorithm
>>>>> 
>>>>> Modified: libcxx/trunk/include/algorithm
>>>>> URL: http://llvm.org/viewvc/llvm-project/libcxx/trunk/include/algorithm?rev=119611&r1=119610&r2=119611&view=diff
>>>>> ==============================================================================
>>>>> --- libcxx/trunk/include/algorithm (original)
>>>>> +++ libcxx/trunk/include/algorithm Wed Nov 17 19:47:02 2010
>>>>> @@ -256,7 +256,7 @@
>>>>> 
>>>>> template<class RandomAccessIterator, class UniformRandomNumberGenerator>
>>>>>   void shuffle(RandomAccessIterator first, RandomAccessIterator last,
>>>>> -                 UniformRandomNumberGenerator&  g);
>>>>> +                 UniformRandomNumberGenerator&&  g);
>>>> 
>>>> Should this constructor also be protected by the rvalue references macro?
>>> 
>>> No, but only because it is a comment. :-)
>>> 
>>> Each public header begins with a synopsis in comments at the top, which is what we're looking at here.
>> 
>> 
>> I find that synopsis super-confusing, by the way ;)
>> 
>> 	- Doug
> 
> Ok, if that is the consensus of the community I can easily remove them.  On a previous product this was a customer-requested feature and subsequent feedback on it was positive.

Interesting! I've found it confusing when I've looked into the headers ("how is this thing really declared?"), but I don't have a typical user's view of libc++. In other words, don't change anything on my account :)

	- Doug