On 21 July 2017 at 14:36, Nicolas George <geo...@nsup.org> wrote: > Le tridi 3 thermidor, an CCXXV, Derek Buitenhuis a écrit : > > +1 on exposing it as side data; it's consistent with previous APIs and > doesn't > > I will not oppose further; I do not buy the "Yet Another Field" argument > but I will debunk it another time. > > > FWIW, literally every other library exposes ICC data the same way (as a > dumb > > buffer), and it's IMO the most reasonable. It's not reasonable to put an > actual > > description of ICC data in the doxy (see: 300 page spec... it is > complictated). > > Just a reference to the spec is fine. > > I fully agree. I suggest something along the lines: > > The data contains an ICC profile as an opaque octet buffer > following the format described at $url with an optional name > defined in the metadata entry "name". > > Regards, >
Thanks for the suggestion, changed the description to what you suggested. I'll push both patches tomorrow unless someone finds something else wrong. _______________________________________________ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
