Re: [Boost-bugs] [Boost C++ Libraries] #4669: Lots of warnings from is_unsigned.hpp

Subject: Re: [Boost-bugs] [Boost C++ Libraries] #4669: Lots of warnings from is_unsigned.hpp
From: Boost C++ Libraries (noreply_at_[hidden])
Date: 2010-09-22 08:45:43


#4669: Lots of warnings from is_unsigned.hpp
------------------------------------------------+---------------------------
  Reporter: Per Ola Ingvarsson <pi@…> | Owner: johnmaddock
      Type: Bugs | Status: closed
 Milestone: To Be Determined | Component: type_traits
   Version: Boost 1.43.0 | Severity: Problem
Resolution: worksforme | Keywords:
------------------------------------------------+---------------------------
Changes (by johnmaddock):

  * status: new => closed
  * resolution: => worksforme

Comment:

 I'm unable to reproduce this here: Boost.Config ''never'' sets
 BOOST_NO_INCLASS_MEMBER_INITIALIZATION for any gcc compiler version (which
 is what would cause BOOST_STATIC_CONSTANT to use enum's rather than a true
 static constant).

 In addition all the type_traits are currently run with -Wall -pedantic
 -Werror and all the tests are passing even on the BSD platforms that are
 tested:
 http://beta.boost.org/development/tests/trunk/developer/type_traits.html

 However, some testing reveals that using an enum in this context causes
 is_unsigned to malfunction anyway, so I'll change the code to always use a
 static constant.

 I'd be interested in knowing why BOOST_NO_INCLASS_MEMBER_INITIALIZATION is
 being set on this platform/compiler combination - are you using a modified
 Boost version at all?

-- 
Ticket URL: <https://svn.boost.org/trac/boost/ticket/4669#comment:2>
Boost C++ Libraries <http://www.boost.org/>
Boost provides free peer-reviewed portable C++ source libraries.

This archive was generated by hypermail 2.1.7 : 2017-02-16 18:50:04 UTC