Am 05.11.21 um 04:52 schrieb Gyan Doshi: > > > On 2021-11-04 04:33 pm, Thilo Borgmann wrote: >> Am 03.11.21 um 17:41 schrieb Gyan Doshi: >>> >>> On 2021-11-03 03:16 pm, Thilo Borgmann wrote: >>>> Hi, >>>> >>>> this effectively reverts 071930de724166bfb90fc6d368c748771188fd94 and >>>> fixes the underlying issue by always preferring TFDT. >>>> >>>> Furthermore, the current solution fails if the -use_tfdt flag is given but >>>> no TFDT box is found in the stream. >>> Thanks for the heads-up. >>> >>> The original impetus for my patch was a client's sample file which had >>> rubbish sidx values but sane tfdt ones, likely due to a buggy muxer. >>> The motivation for an option was to give the user control. >>> >>> I think hardcoding a reversed preference can lead to the same impasse. How >>> about change the default value of use_tfdt and check for tfdt presence when >>> enabled? >> Changed the default to use TFDT and add warnings about fallback to SIDX or >> TFDT where useful. >> Looks better? > > LGTM.
Pushed with updated reference for tests/ref/seek/extra-mp4. Thanks! -Thilo _______________________________________________ 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".
