Thank you all for taking the time to reply! While I may have seemed inactive, I have read each and every reply multiple times to make sure I understood the points made correctly. It has really been food for thought and exactly what I hoped for, so thanks again!
I think it would be best to be independent of the http implementation that is used in the end, like Michael also suggested. I've started to define an API that libraries will need to use to make use of the ffserver code, I think this will still take a while. For the time being it will still be the lavf API (because it is already used as of now), but it should be trivial to add more libraries. This kinda bears in mind users that want a "mature" http implementation and those that do not particularly care and want "any" http implementation. I also proposed this on IRC, but was quickly discouraged, because of the complexity of #ifdefs, "hacks" and hidden incompatibilities. If this turns out to be too hard or too complex I will decide then (before the coding period starts) how to proceed. Once again, thank you all! _______________________________________________ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
