[boost] Re: Boost Policy on compiler warnings?

From: David B. Held (dheld_at_[hidden])
Date: 2002-08-08 13:17:02

"Victor A. Wagner, Jr." <vawjr_at_[hidden]> wrote in message
news:4.3.1.2.20020808011945.036700a8_at_mail.rudbek.com...
> [...]
> I wonder if we could come up with some program to make an #ifdef'd
> source  file that had all the requisite "ignore this waring" directives in
> it...for all of the compilers, given a set for one.  Hmmmmm, PITA.

Well, even though many compilers will give spurious warnings, I don't
think those warnings are *always spurious*.  That is, there are surely
times when an otherwise spurious warning is really legitimate.  Such
an unconditional disabling of annoying warnings doesn't seem the
most prudent course of action.  Of course, a few that probably are never
important are ones about inlining.  But determining which warnings are
never useful, and which ones are sometimes useful seems to be a major
challenge.

Dave