Subject: Re: gcc-ansi -pedantic and long long on NetBSD
To: T. William Wells <bill@twwells.com>
From: Ted Lemon <mellon@ipd.wellsfargo.com>
List: current-users
Date: 11/01/1994 11:25:55
> Maintenance is something programmers don't like to think about,
> though the reality is that much of the cost of programs is in
> their maintenance. In a project of any size, the cost of one bug,
> or even one re-coding for portability, that never happens because
> -pedantic was used will vastly outweigh the cost of adhering to
> its requirements.

I haven't heard anybody disagreeing with this statement, and I
certainly don't.

I think that deciding what system header files look like is actually a
fairly important concern, although you may be right that this is the
wrong forum on which to discuss it.  If this belongs elsewhere, maybe
Chris or Theo can direct us there.

I jumped into this discussion (perhaps foolishly) because I've been
involved with gcc porting and bug fixing for a long time and have some
idea of how gcc developers solve problems like this.  I don't get to
choose what goes into the header files - I'm just trying to find out
what's needed so that I can suggest ways of accomplishing it.

My point from the previous message is that I don't see any evidence
that -pedantic is a good way to make programs more portable.  I
believe that -Wall does just as good a job of accomplishing what you
describe above.  If there's something useful that -pedantic does that
-Wall doesn't do, it would be helpful if you could tell us.

			       _MelloN_


--
Ted Lemon		      Wells Fargo Bank, Information Protection Division
mellon@ipd.wellsfargo.com					+1 415 477 5045