On 24.01.2015, at 21:09, wm4 <nfx...@googlemail.com> wrote: > On Sat, 24 Jan 2015 18:37:01 +0000 > Derek Buitenhuis <derek.buitenh...@gmail.com> wrote: > >> On 1/24/2015 4:33 PM, wm4 wrote: >>> Which ones? We even expect C99 support from the compiler. >> >> Doesn't matter. It's the project's policy to have decls at >> block beginnings. Yes some of us think it's better. >> >> We know you don't. Don't start an ideological troll war. > > Having dumb policies is fine, but then don't use broken compilers as > excuse. Just say it's your policy to do it this way, even if there's no > technical necessity.
Maybe not really relevant, but since I wrote it... Skip it unless you have time to waste ;) I guess it's an old habit from when we still used to support gcc 2.95 :) Which btw last I tried some months ago still worked except for very few cases. Thing is, we require not that much of C99 and most is header or preprocessor-related or used very, very rarely. And while I don't know where they come from, every now and then I hear about someone asking for C89 support for some product (possibly custom in-house compilers for strange architectures/OS? Symbian maybe still hiding in some hole?). But with even tinycc supporting it I guess it's reasonable to say that any "technical" reasons that might exist are more related to someone's laziness than a real technical reason. That said, I still prefer it even though I write C++ every day. Maybe it's just an old habit and secretly longing back to writing PASCAL as in my childhood days ;) _______________________________________________ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel