Re: [boost] Re: Boost Library Guidelines

From: Beman Dawes (bdawes_at_[hidden])
Date: 2003-04-30 20:45:32

At 10:12 AM 4/30/2003, Douglas Gregor wrote:

 ><enters grumpy old developer mode>
 >I agree that it would be great from the user's point of view if all of
 >Boost compiled without warnings, but I'm afraid a no-warnings policy will
 >make the whack-a-mole game all that much worse. As it stands, some Boost
 >code is brittle on certain compilers: tweak in one place, and it breaks
 >a compiler; fix for that compiler and you've broken another compiler.

Pete Becker has pointed out that a no-warnings policy effectively redefines 
the language by eliminating uses that are actually valid C++. The 
redefinition is being done not by a standards body, but by individual
compiler vendors, who often change what constitutes a warning from release 
to release.

For Boost, eliminating all warnings would mean eliminating all warnings 
from recent versions of a number of compilers. That doesn't seem 
productive.

OTOH, there are a number of specific warnings we probably can agree are 
worrisome enough that we never want to see them. Maybe we should generally 
not report warnings, but do report the agreed upon set of warnings and 
pester Boost developers to clear them.

--Beman