On 6/18/19 1:48 PM, Alfred E. Heggestad wrote: > On 18/06/2019 04:02, Steven Liu wrote: >> Alfred E. Heggestad <[email protected]> 于2019年6月17日周一 下午4:02写道： >>> >>> From 923da82598bddd1ed05750427dbc71e607d296a2 Mon Sep 17 00:00:00 2001 >>> From: "Alfred E. Heggestad" <[email protected]> >>> Date: Mon, 17 Jun 2019 09:59:04 +0200 >>> Subject: [PATCH] dash: change default MP4 extension to .m4s >>> >>> this was changed in commit 281a21ed50849e3c8c0d03005230e9fd07c24370 >>> --- >>> libavformat/dashenc.c | 2 +- >>> 1 file changed, 1 insertion(+), 1 deletion(-) >>> >>> diff --git a/libavformat/dashenc.c b/libavformat/dashenc.c >>> index 3fd7e78166..a51a1da0ca 100644 >>> --- a/libavformat/dashenc.c >>> +++ b/libavformat/dashenc.c >>> @@ -166,7 +166,7 @@ static struct format_string { >>> const char *str; >>> } formats[] = { >>> { SEGMENT_TYPE_AUTO, "auto" }, >>> - { SEGMENT_TYPE_MP4, "mp4" }, >>> + { SEGMENT_TYPE_MP4, "m4s" }, >>> { SEGMENT_TYPE_WEBM, "webm" }, >>> { 0, NULL } >>> }; >>> -- >>> 2.20.1 (Apple Git-117) >>> >>> _______________________________________________ >>> ffmpeg-devel mailing list >>> [email protected] >>> https://ffmpeg.org/mailman/listinfo/ffmpeg-devel >>> >>> To unsubscribe, visit link above, or email >>> [email protected] with subject "unsubscribe". >> >> >> >> LGTM >> > > the background for this is the extension for DASH media files > used to be *.m4s and it is now *.mp4 > > > the patch is a suggestion and should be checked by the DASH experts > > what is correct according to the standard ? > > the media-file is not really an .mp4 file, it cannot be > played with e.g. ffplay: > > $ ffplay chunk-stream1-00001.m4s Thanks for submitting the patch. I agree that m4s should be extension for media segments. mp4 should be used only for complete files. With respect to the patch, dashenc generates either multiple segments or a single file(with byte range as segments) based on "single_file" option. The default of mp4 is correct when "single_file" is enabled. But it is wrong when "single_file" is disabled. The proposed patch just reverses this situation. I would suggest the patch should handle both cases correctly. > > > > > /alfred > _______________________________________________ > ffmpeg-devel mailing list > [email protected] > https://ffmpeg.org/mailman/listinfo/ffmpeg-devel > > To unsubscribe, visit link above, or email > [email protected] with subject "unsubscribe".
_______________________________________________ ffmpeg-devel mailing list [email protected] https://ffmpeg.org/mailman/listinfo/ffmpeg-devel To unsubscribe, visit link above, or email [email protected] with subject "unsubscribe".
