On Sun, Jul 29, 2018 at 12:58 PM, David Kastrup <dak@xxxxxxx> wrote:
Paul Davis <paul@xxxxxxxxxxxxxxxxxxxxx> writes:
> we have found bugs in libsmf that we have not pushed upstream for
> various reasonably defensible reasons.
Without wanting to dwell on it and just out of personal curiosity (sort
of having been on either side): what class of reasons would that be?
libsmf has been written with a clear tendency toward "if the SMF is not perfect, we cannot load it". Ardour wants a somewhat more tolerant behaviour, so quite a few of our changes have involved making that the case.
on top of that, we wanted to treat libsmf as a genuinely internal library that so if we wanted to amend it in more fundamental ways, we wouldn't need to stop and think about that for even a second. the first thing that we ever did with it, back in 2009, was to strip the code down to suit that purpose.
we probably could have/should have tried to push a few of our fixes in the direction of libsmf's current maintainer. that's my fault, but also a function of time overload.
_______________________________________________ Linux-audio-user mailing list Linux-audio-user@xxxxxxxxxxxxxxxxxxxx https://lists.linuxaudio.org/listinfo/linux-audio-user
