Re: [Boost-testing] best way to mark up test that only pass with C++0x

Subject: Re: [Boost-testing] best way to mark up test that only pass with C++0x
From: Dave Abrahams (dave_at_[hidden])
Date: 2011-09-30 10:59:51

on Thu Sep 29 2011, Eric Niebler <eric-AT-boostpro.com> wrote:

> On 9/26/2011 4:44 AM, Dave Abrahams wrote:
>> 
>
>> on Sun Sep 25 2011, Eric Niebler <eric-AT-boostpro.com> wrote:
>> 
>> You can always use those PP symbols to make the test pass
>> unconditionally when the necessary features are known to be missing.  I
>> don't think that drops any useful information on the floor.
>
> If folks are using to test results to determine which library features
> are supported on which platforms, then yes, this drops some very
> important information on the floor.

Except that we don't have a uniform discipline about how this is
treated, so you can't use the test results that way, at least reliably
and uniformly, today.  IMO that information belongs somewhere else, in a
first-class summary.

-- 
Dave Abrahams
BoostPro Computing
http://www.boostpro.com