[Ffmpeg-devel] [PATCH] Remove PRI macros definition
Sun Feb 25 22:17:41 CET 2007
On 25/02/07, M?ns Rullg?rd <mans at mansr.com> wrote:
> "Fran?ois Revol" <revol at free.fr> writes:
> >> >> You won't get warnings, but you will get random stuff printed and
> >> >> possibly crashes. These defines need to go for the same reason we
> >> >> removed the intXX_t typedefs.
> >> >
> >> > Then don't use them !
> >> >
> >> > This breaks the BeOS build.
> >> > Either put them back, move them to os_support.h with ifdef I don't
> >> > care
> >> > but fix this.
> >> I don't care how much you shout.
> > That's not shouting.
> > THAT IS.
> This is too!
> >> This can only be fixed reliably in
> >> the system headers.
> > error: portability redefined.
> > Previous definition in http://dict.die.net/portability/
> No, you're redefining "standards compliance".
> >> To me, breaking the Beos build is a very small price to pay for
> >> correct results on other systems. There are exactly 2 (two) known
> > 2?
> You and Cian Duffy.
> >> users of FFmpeg on Beos. I'm sorry, but if you choose to use an
> >> obscure system, you'll have to deal with the consequences yourself.
> > different != obscure.
> Where I'm from, 2 users counts as obscure.
BeOS builds of VLC usually get around 4,000 downloads and are distributed
with magnussoft/yellowTAB ZETA, which has totalled over 1M sales or
downloads. They use ffmpeg. There are two known active COMPILERS of ffmpeg
on BeOS, not users.
>> The Windows users finally accepted that we can't be expected to fix
> >> deficiencies in their system headers, especially when doing so
> >> reliably is impossible. Now it's your turn.
> > I'm not in the mood of even replying to this politely so I won't.
> > Fran?ois.
> M?ns Rullg?rd
> mans at mansr.com
> ffmpeg-devel mailing list
> ffmpeg-devel at mplayerhq.hu
"We're busy running out of time"
More information about the ffmpeg-devel