Message-ID: <8D53104ECD0CD211AF4000A0C9D60AE33CA38F@probe-2.Acclaim-Euro.net> From: Shawn Hargreaves To: djgpp AT delorie DOT com Cc: Eli Zaretskii , Andris Pavenis Subject: Re: Announce: Allegro 3.1 Date: Wed, 6 Jan 1999 15:33:16 -0000 MIME-Version: 1.0 X-Mailer: Internet Mail Service (5.0.1460.8) Content-Type: text/plain Reply-To: djgpp AT delorie DOT com Eli Zaretskii writes: >> Do you happen to know which warning option would turn off these >> complaints? > > I don't think there is such an option. > What options does Allegro use to compile itself? At the moment, -Wall, -W, and -Wno-unused. The gcc docs seem to imply that the same effect as -Wall can be obtained by manually listing all the individual options: do you happen to know if this is accurate, or if -Wall has any other effects that will not be duplicated this way? If possible, one solution might be to use many specific switches instead of the one generic option, although I don't think the same thing would apply to -W complaints (I can't see any way to manually enable those diagnostics). > So I guess that different teams and different volunteers who work on > different versions of the compiler do whatever they feel is right, > while other teams think that something else is right... And you get > to sort all that out. Absolutely, but it seems unfortunate that this should be at the discretion of the compiler authors rather than the end users. IMHO it would be much nicer if they could define a set of standard warning levels, and then leave the user to decide which level they prefer to enable. The C language, and the switches used to invoke a compiler, stay very much the same between each release (it would be ridiculous if egcs used different options to gcc). I think it is a shame that the warning generation is not subjected to the same rigorous compatibility standards as the rest of the system... Shawn Hargreaves.