As already pointed, scale/resample being disableable is a deliberate decision for people who know they won't need pixel/sample format conversion and want to avoid depending on sws/lavr.
On Fri, 2 May 2014 10:57:58 -0700, Josh Allmann <[email protected]> wrote: > On 2 May 2014 10:16, Hendrik Leppkes <[email protected]> wrote: > > On Fri, May 2, 2014 at 5:54 PM, Josh Allmann <[email protected]> > > wrote: > >> Since the scaler is often auto-inserted in video filtergraphs, > >> this avoids a runtime error when filters are selectively enabled. > > > > I dunno, just because its commonly used at runtime shouldn't mandate > > including it all the time. > > If a user selectively disables it, its their own fault, isn't it. Its > > default on afterall. > > > > In that case, then why include fifos, buffers, ... ? 1) because without fifos, some parts of lavfi won't work at all. And without buffersrc/sink, you have no way of getting the data in or out, making lavfi pretty much completely useless. They really are more parts of lavfi internals than filters, and I've been considering making them into something that's not a filter. 2) they are also very small compared to the dependency on sws/lavr. > > If the scale filter is not unconditionally included, then it should > not be auto-inserted. The user should not have to be aware of lavfi's > plumbing -- that's the job of the build system. > > This is a compile-time bug manifesting as a runtime error. Basic > usability goes a long way; lavfi's usabillity is bad enough as-is, and > blaming the user is a really bad attitude to take. > --disable-all is IMO really not "basic usability". If you use that option, you're expected to know what you're doing. -- Anton Khirnov _______________________________________________ libav-devel mailing list [email protected] https://lists.libav.org/mailman/listinfo/libav-devel
