On 29/04/15 17:36, Luca Barbato wrote: > On 28/04/15 15:21, Andreas Cadhalpun wrote: >> I think it's better to change the function not to need 64 blocks, >> as in the second patch I sent. > > Your patch assumes that the required block number information is fine > and returning that partial decode is better than returning no data, I'm > not sure about it and we already reject impossible block values. > > I do not know the format, so my assumption that the nblocks < 64 is > impossible is as good as yours. >
And looks like you are right, the reference encoder can produce such files. lu _______________________________________________ libav-devel mailing list [email protected] https://lists.libav.org/mailman/listinfo/libav-devel
