[Ffmpeg-devel] [PATCH] Wildcard and catalog image sequences

Michael Niedermayer michaelni
Wed Aug 30 16:31:57 CEST 2006


On Wed, Aug 30, 2006 at 04:04:47PM +0200, Michel Bardiaux wrote:
> >>>>>2. in C 0 is false not 0 is true, using 0 as true is broken as logical 
> >>>>>and/or
> >>>>> will no longer have their intuitive meaning, do you disagree here?
> >>>>Not at all. I knew you want patches to be as little intrusive as 
> >>>>possible, so I tried not to change anything to numbered sequences, 
> >>>>which are known to work. Hence I had to conform to the API defined by 
> >>>>get_frame_filename and filename_number_test, which is on the lines of 
> >>>>UNIX-style syscalls. Maybe this has to be changed, but shouldnt that be 
> >>>>a different patch?
> >>>of course it should be a seperate patch :)
> >>I wonder what your reaction would have been to a patch just to change 
> >>*that* (filename_number_test returning 0/1 instead of -1/0) coming out 
> >>of the blue... 
> >
> >i dont know ...
> But I can make an educated guess...
> >
> >
> >>Do you really want one?
> >
> >yes
> OK, phase 1 will be changing the API filename_number_test as above. No 
> change to find_image_range and get_frame_filename, right? 

exactly, 1/0 for "boolean", the other 2 functions arent boolean they rather
have an error/no error return value for which we generally use 
negative == error in lav*

> And you also 
> mentionned they should actually be av_ prefixed, can that go in the same 
> patch?


> After that, change to support only of numbered sequences in output and 
> cataloged sequences (file and stdin) for input?

if we keep supporting %d in output then maybe we should also keep support
for %d input, iam not sure though, comments from others welcome of course

Michael     GnuPG fingerprint: 9FF2128B147EF6730BADF133611EC787040B0FAB

In the past you could go to a library and read, borrow or copy any book
Today you'd get arrested for mere telling someone where the library is

More information about the ffmpeg-devel mailing list