On Tue, Nov 05, 2019 at 10:13:52AM +0530, Gyan wrote: > > > On 05-11-2019 04:35 am, Michael Niedermayer wrote: > >On Sun, Nov 03, 2019 at 11:14:25AM +0530, Gyan wrote: > >>Helps better identification of expr eval failures. > >> > >>Gyan > >> eval.c | 21 +++++++++++++++++++++ > >> eval.h | 10 ++++++++++ > >> version.h | 4 ++-- > >> 3 files changed, 33 insertions(+), 2 deletions(-) > >>3dd142baa0144fd324eb9da8a9932cfd7ab2cd98 > >>0001-avutil-eval-add-function-to-track-variable-use.patch > >> From 19bce329464676f071707b99575f80e5abe1cd4c Mon Sep 17 00:00:00 2001 > >>From: Gyan Doshi <ffm...@gyani.pro> > >>Date: Sat, 2 Nov 2019 20:16:42 +0530 > >>Subject: [PATCH] avutil/eval: add function to track variable use > >> > >>Helps avoid multiple evals of cross-referenced expressions > >>and catch the use of non-applicable variables with respect > >>to eval or special mode in filters > >Maybe you should provide more details of the use case of this, maybe > >with an example. Because it seems not completely obvious > > An example of a cross-referenced expression would be 'y=x+rand(10)'. > Normally in filters, X would be evaluated first, then Y, then X again. The > 2nd eval of X will overwrite the first and can generate a different value. > With this func, we can avoid the 2nd eval. > > I'm in the process of modifying the scale filter to allow 'n', 't', 'pos' > variables in frame eval mode (it already supports dynamic output based on > incoming frame changes) and I would like to catch their presence in init > mode, since av_expr_eval will fail. Now, it can also fail due to circularly > referenced expressions e.g. 'x=y+10' and 'y=x+10' and there's no way to > detect whether it's this case or if inapplicable variables have been used.
please add this (or similar information) to the commit message thx [...] -- Michael GnuPG fingerprint: 9FF2128B147EF6730BADF133611EC787040B0FAB Opposition brings concord. Out of discord comes the fairest harmony. -- Heraclitus
signature.asc
Description: PGP signature
_______________________________________________ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org https://ffmpeg.org/mailman/listinfo/ffmpeg-devel To unsubscribe, visit link above, or email ffmpeg-devel-requ...@ffmpeg.org with subject "unsubscribe".
