[prev in list] [next in list] [prev in thread] [next in thread] 

List:       boost
Subject:    Re: [boost] [1.34.0beta] many, many warnings... :(
From:       "Sohail Somani" <s.somani () fincad ! com>
Date:       2007-05-02 19:04:53
Message-ID: 1C1EBEF8DBACDC439D038EA051674EC78A46E8 () xbox ! financialcad ! com
[Download RAW message or body]

> -----Original Message-----
> From: boost-bounces@lists.boost.org 
> [mailto:boost-bounces@lists.boost.org] On Behalf Of Jody Hagins
> Sent: Wednesday, May 02, 2007 11:59 AM
> To: boost@lists.boost.org
> Subject: Re: [boost] [1.34.0beta] many, many warnings... :(
> 
> On Wed, 2 May 2007 10:59:56 -0700
> "Sohail Somani" <s.somani@fincad.com> wrote:
> 
> 
> > But most (not all, for sure) warnings can be disabled by
> > -fno-some-feature. The only places where you might have to 
> change the
> > code would be in the cases where the warning cannot be disabled. 
> 
> 
> That's the problem.  You don't want to disable the warning entirely. 
> That's worse than ignoring them.  You want to disable a warning at the
> point it is a warning.
> 
> This is a MAJOR drawback of gcc...

So then there are two options given the two major toolsets:
 * Change the code to quiet it (along the lines of BOOST_WARNING_XXX
macros) or fix it if it is a real warning
 * Add a warning flag when compiling the specific file

If the first option isn't feasible for a given warning, then the second
one may apply as a last resort.

And there is still the option of ignoring the warnings, but Boost is a
harder sell if it cannot compile without warnings. That's been my
experience.

Sohail

[prev in list] [next in list] [prev in thread] [next in thread]