Re: [boost] Coming up on 1.44 code freeze

Subject: Re: [boost] Coming up on 1.44 code freeze
From: Niels Dekker - address until 2010-10-10 (niels_address_until_2010-10-10_at_[hidden])
Date: 2010-07-05 04:44:15

Marshall Clow wrote:
> I've been looking at the trunk and release branch, I've noticed that
> there are a lot of files that exist only in one or the other.
...
> Here's a list of what I found:

> Only in 'trunk/'
 ...
> libs/utility/initialized_test.cpp
> libs/utility/initialized_test_fail1.cpp
> libs/utility/initialized_test_fail2.cpp
> libs/utility/value_init_workaround_test.cpp

Thanks Marshall, I just added them, yesterday: 
https://svn.boost.org/trac/boost/changeset/63638

FYI, These tests verify the fixes of two utility/value_init tickets:

https://svn.boost.org/trac/boost/ticket/3472 "Setting value_initialized<T> 
to a value when T is a top-level const", resolved by adding a new template 
to value_init.hpp, boost::initialized<T>, which provides 
direct-initialization as well as value-initialization.

https://svn.boost.org/trac/boost/ticket/3869, "Unconditional call to memset 
in value_initialized()", resolved by making the existing memset workaround 
conditional, based on a new compiler defect macro, 
BOOST_NO_COMPLETE_VALUE_INITIALIZATION.

Kind regards,

  Niels

--
Niels Dekker
http://www.xs4all.nl/~nd/dekkerware
Scientific programmer at LKEB, Leiden University Medical Center