Re: [boost] Intentions of Boost.TR1

Subject: Re: [boost] Intentions of Boost.TR1
From: Ryan Gallagher (ryan.gallagher_at_[hidden])
Date: 2009-05-01 13:07:45

David Abrahams <dave <at> boostpro.com> writes:
> 
> I just saw this post on comp.lang.c++.moderated.  I wonder if the
> Boost.TR1 documentation shouldn't clarify its intended use or degree of
> conformity.  I don't think TR1 conformity is something we have been
> paying much attention to, is it?
>
> From: Piotr Dobrogost <pd <at> 1.google.dobrogost.pl>
[...]
> But taking TR1 as
> a test case Dinkumware estimates conformance of free implementations
> as 15% for Boost, and 12% for Gcc in a comparison to their 100%
> (http://www.dinkumware.com/tr1_compare.aspx).
[...] 

That "tr1_compare" page is an "interesting" marketing document.  It's quite out
of date as it used the Boost 1.33.1 release, thus giving total failures for some
tests such as unordered_set.  It also doesn't seem to mention the math functions
supported.  (12% was only for the non-math/c99 portion of TR1.)  A request to
update should probably be made to them.

I'm not sure what it takes to get a "source license" from them, but if anyone
has it I wonder if they could investigate this "Quick Proofer" tool they
developed.  It would seem more believable if they would make this an open source
tool for use as a true acid test.

-Ryan