Hi all

ive promised to write a mail about SDR at the confernece about a seperate lib,
here is it (i in fact had this half written already before)

The SDR code in FFmpeg was and is intended to be a plain simple libavdevice input device
same as x11 and others
SDR is complex and not a simple memcpy() from a /dev file. To implement SDR many
design decissions have to be made, there are often more than one way things can be
done.
In the inside of a SDR input module these decissions are invissible,
they can be changed each commit as bugfixes and optimizatzions require.

OTOH, if SDR or parts of SDR are in a seperate library, some decissions now would
affect the public API and ABI, thats a added difficulty.

With audio resampling, FFmpeg supported resampling internally long before
providing libswresample. Audio resampling is also simpler than SDR.

Is a seperate library really the right choice for SDR in FFmpeg at this stage ?

Also code can always be moved into an seperate library but its much harder
to move or change it once such library is public.

Thx

-- 
Michael     GnuPG fingerprint: 9FF2128B147EF6730BADF133611EC787040B0FAB

It is dangerous to be right in matters on which the established authorities
are wrong. -- Voltaire