January 2012 - ffmpeg-devel - lists.ffmpeg.org

[PATCH] mxfdec: support AVFMT_FLAG_IGNIDX.
by Reimar Döffinger 12 Mar '12

12 Mar '12

This allow users to work around regressions introduced by index support and helps with debugging. Signed-off-by: Reimar Döffinger <Reimar.Doeffinger(a)gmx.de> --- libavformat/mxfdec.c | 3 +++ 1 files changed, 3 insertions(+), 0 deletions(-) diff --git a/libavformat/mxfdec.c b/libavformat/mxfdec.c index 66af2e4..1498ebb 100644 --- a/libavformat/mxfdec.c +++ b/libavformat/mxfdec.c @@ -1529,6 +1529,9 @@ static int mxf_parse_handle_essence(MXFContext *mxf) /* remember where we were so we don't end up seeking further back than this */ mxf->last_forward_tell = avio_tell(pb); + if (mxf->fc->flags & AVFMT_FLAG_IGNIDX) + return 0; + if (!pb->seekable) { av_log(mxf->fc, AV_LOG_INFO, "file is not seekable - not parsing footer\n"); return -1; -- 1.7.8.2

3 4

MXF files from Avid containg DNXHD 120 and bigger
by Steve Dierker 12 Mar '12

12 Mar '12

Hello list, today I was trying to convert some DNXHD movies contained in MXF files created by Avid. For DnxHD 60, 90 and 90x it worked pretty fine but for everything with higher rates it didn't work. Later on I stumbled up on this http://roundup.libav.org/issue2160 So obviously it is a known bug and I wanted to what the progress is concerning fixing it and how I can participate? best regards, Steve

2 1

ffserver: fix missing buffer set after rewrite av_open_input_file to avformat_open_input
by Miroslav Slugeň 12 Mar '12

12 Mar '12

Hi, while rewriting functions "av_open_input_stream" and "av_open_input_file" to "avformat_open_input" Anton Khirnov forgot to add buf_size in ffserver needed for streaming. Simple attached patch is the easiest solution i found, there are also another solutions but it needs more dirty code. Please apply this patch to ffmpeg, ffserver is broken more than 5 months.

2 1

[PATCH] [matroska] Check return value of avio_seek and avoid modifying state if it fails
by Joakim Plate 12 Mar '12

12 Mar '12

--- libavformat/matroskadec.c | 24 ++++++++++++++---------- 1 files changed, 14 insertions(+), 10 deletions(-) diff --git a/libavformat/matroskadec.c b/libavformat/matroskadec.c index 3a5940b..7564c2d 100644 --- a/libavformat/matroskadec.c +++ b/libavformat/matroskadec.c @@ -2018,25 +2018,20 @@ static int matroska_read_seek(AVFormatContext *s, int stream_index, timestamp = FFMAX(timestamp, st->index_entries[0].timestamp); if ((index = av_index_search_timestamp(st, timestamp, flags)) < 0) { - avio_seek(s->pb, st->index_entries[st->nb_index_entries-1].pos, SEEK_SET); + if (avio_seek(s->pb, st->index_entries[st->nb_index_entries-1].pos, SEEK_SET) < 0) + return -1; matroska->current_id = 0; while ((index = av_index_search_timestamp(st, timestamp, flags)) < 0) { matroska_clear_queue(matroska); if (matroska_parse_cluster(matroska) < 0) break; } - } - - matroska_clear_queue(matroska); - if (index < 0) + matroska_clear_queue(matroska); return 0; + } index_min = index; for (i=0; i < matroska->tracks.nb_elem; i++) { - tracks[i].audio.pkt_cnt = 0; - tracks[i].audio.sub_packet_cnt = 0; - tracks[i].audio.buf_timecode = AV_NOPTS_VALUE; - tracks[i].end_timecode = 0; if (tracks[i].type == MATROSKA_TRACK_TYPE_SUBTITLE && !tracks[i].stream->discard != AVDISCARD_ALL) { index_sub = av_index_search_timestamp(tracks[i].stream, st->index_entries[index].timestamp, AVSEEK_FLAG_BACKWARD); @@ -2047,7 +2042,16 @@ static int matroska_read_seek(AVFormatContext *s, int stream_index, } } - avio_seek(s->pb, st->index_entries[index_min].pos, SEEK_SET); + if (avio_seek(s->pb, st->index_entries[index_min].pos, SEEK_SET) < 0) + return -1; + + matroska_clear_queue(matroska); + for (i=0; i < matroska->tracks.nb_elem; i++) { + tracks[i].audio.pkt_cnt = 0; + tracks[i].audio.sub_packet_cnt = 0; + tracks[i].audio.buf_timecode = AV_NOPTS_VALUE; + tracks[i].end_timecode = 0; + } matroska->current_id = 0; matroska->skip_to_keyframe = !(flags & AVSEEK_FLAG_ANY); matroska->skip_to_timecode = st->index_entries[index].timestamp; -- 1.7.6

2 6

[PATCH] Fix decoding of DNxHD video in MXF container
by Joseph Artsimovich 11 Mar '12

11 Mar '12

I found two problems with decoding of DNxHD encoded videos in MXF container. 1. The DNxHD codec signature in libavformat/mxf.c is wrong. 2. Frame height is calculated incorrectly for interlaced videos. What's stored in PictureEssenceDescriptor in MXF is not a frame height but field height. This is a general problem that should affect other codecs. I checked how D-10 (IMX) in MXF container ends up being handled correctly, and it turned out it's because of stream parsing. The patches fixing the decoding problems are attached. Note that I didn't look at MXF encoding, which might or might not be affected by problem 2. -- Joseph Artsimovich Senior C++ Applications Developer MirriAd Ltd

4 8

[PATCH] mxfdec: add timecode to metadata
by Matthieu Bouron 11 Mar '12

11 Mar '12

Hi there, Here is a patch to extract the starting timecode from mxf files. Feel free to comment, Matthieu

5 17

FFmpeg at Chemnitzer Linux-Tage
by Alexander Strasser 24 Feb '12

24 Feb '12

Hello all, we have registered FFmpeg for `Chemnitzer Linux-Tage' in Germany. Fortunately we were accepted as a project for the Linux-Live part of the exhibition. The event will take place on 17th & 18th of March in Chemnitz. More information can be found here: http://chemnitzer.linux-tage.de/2012/info/index?cookielang=en We invite every FFmpeg developer to drop by or even help out with the booth. Currently Carl Eugen Hoyos and Alexander Strasser are registered as booth staff and therefore will attend for sure. If you want to drop by or help out, it would be nice if you let us know about it here. There is also a lot to do: - We need a nice poster introducing FFmpeg(-usecases) - Designing/Ordering FFmpeg T-Shirts for the booth staff(/visitors?) - Preparing flyers - Probably more... If you have comments, ideas or help to offer, please tell us! Best Regards, Carl Eugen & Alexander

5 7

[PATCH] vf_overlay: add logic for avoiding overlaying frames with PTS > main frame PTS
by Stefano Sabatini 16 Feb '12

16 Feb '12

Also add debug logging messages for helping tracking down similar issues. Fix trac ticket #467. --- libavfilter/vf_overlay.c | 41 +++++++++++++++++++++++++++++------------ 1 files changed, 29 insertions(+), 12 deletions(-) diff --git a/libavfilter/vf_overlay.c b/libavfilter/vf_overlay.c index 062783e..9600061 100644 --- a/libavfilter/vf_overlay.c +++ b/libavfilter/vf_overlay.c @@ -25,6 +25,8 @@ * overlay one video on top of another */ +/* #define DEBUG */ + #include "avfilter.h" #include "libavutil/eval.h" #include "libavutil/avstring.h" @@ -75,7 +77,7 @@ typedef struct { uint8_t overlay_rgba_map[4]; uint8_t overlay_has_alpha; - AVFilterBufferRef *overpicref; + AVFilterBufferRef *overpicref, *overpicref_next; int main_pix_step[4]; ///< steps per pixel for each plane of the main output int overlay_pix_step[4]; ///< steps per pixel for each plane of the overlay @@ -144,6 +146,8 @@ static av_cold void uninit(AVFilterContext *ctx) if (over->overpicref) avfilter_unref_buffer(over->overpicref); + if (over->overpicref_next) + avfilter_unref_buffer(over->overpicref_next); } static int query_formats(AVFilterContext *ctx) @@ -302,22 +306,33 @@ static void start_frame(AVFilterLink *inlink, AVFilterBufferRef *inpicref) AVFilterBufferRef *outpicref = avfilter_ref_buffer(inpicref, ~0); AVFilterContext *ctx = inlink->dst; OverlayContext *over = ctx->priv; + av_unused AVFilterLink *outlink = ctx->outputs[0]; inlink->dst->outputs[0]->out_buf = outpicref; outpicref->pts = av_rescale_q(outpicref->pts, ctx->inputs[MAIN]->time_base, ctx->outputs[0]->time_base); if (!over->overpicref || over->overpicref->pts < outpicref->pts) { - AVFilterBufferRef *old = over->overpicref; - over->overpicref = NULL; - avfilter_request_frame(ctx->inputs[OVERLAY]); - if (over->overpicref) { - if (old) - avfilter_unref_buffer(old); - } else - over->overpicref = old; + if (!over->overpicref_next) + avfilter_request_frame(ctx->inputs[OVERLAY]); + + if (over->overpicref && over->overpicref_next && + over->overpicref_next->pts <= outpicref->pts) { + avfilter_unref_buffer(over->overpicref); + over->overpicref = over->overpicref_next; + over->overpicref_next = NULL; + } } +#ifdef DEBUG + av_log(ctx, AV_LOG_DEBUG, "main_pts:%"PRId64" main_pts_time:%f", + outpicref->pts, outpicref ->pts * av_q2d(outlink->time_base)); + if (over->overpicref) + av_log(ctx, AV_LOG_DEBUG, " over_pts:%"PRId64" over_pts_time:%f", + over->overpicref->pts, over->overpicref->pts * av_q2d(outlink->time_base)); + av_dlog(ctx, "\n"); +#endif + avfilter_start_frame(inlink->dst->outputs[0], outpicref); } @@ -326,9 +341,11 @@ static void start_frame_overlay(AVFilterLink *inlink, AVFilterBufferRef *inpicre AVFilterContext *ctx = inlink->dst; OverlayContext *over = ctx->priv; - over->overpicref = inpicref; - over->overpicref->pts = av_rescale_q(inpicref->pts, ctx->inputs[OVERLAY]->time_base, - ctx->outputs[0]->time_base); + inpicref->pts = av_rescale_q(inpicref->pts, ctx->inputs[OVERLAY]->time_base, + ctx->outputs[0]->time_base); + + if (!over->overpicref) over->overpicref = inpicref; + else over->overpicref_next = inpicref; } // divide by 255 and round to nearest -- 1.7.5.4

2 6

[PATCH] h264: Return the correct number of bytes for mid strea, extradata.
by Michael Niedermayer 16 Feb '12

16 Feb '12

Fixes the hang with Ticket952 Signed-off-by: Michael Niedermayer <michaelni(a)gmx.at> --- libavcodec/h264.c | 4 ++-- 1 files changed, 2 insertions(+), 2 deletions(-) diff --git a/libavcodec/h264.c b/libavcodec/h264.c index 962b5dc..a4892ac 100644 --- a/libavcodec/h264.c +++ b/libavcodec/h264.c @@ -1125,7 +1125,7 @@ int ff_h264_decode_extradata(H264Context *h, const uint8_t *buf, int size) if(decode_nal_units(h, buf, size) < 0) return -1; } - return 0; + return size; } av_cold int ff_h264_decode_init(AVCodecContext *avctx){ @@ -1170,7 +1170,7 @@ av_cold int ff_h264_decode_init(AVCodecContext *avctx){ } if(avctx->extradata_size > 0 && avctx->extradata && - ff_h264_decode_extradata(h, avctx->extradata, avctx->extradata_size)) + ff_h264_decode_extradata(h, avctx->extradata, avctx->extradata_size)<0) return -1; if(h->sps.bitstream_restriction_flag && s->avctx->has_b_frames < h->sps.num_reorder_frames){ -- 1.7.5.4

3 13

[RFC] lavu PTS printing utils
by Stefano Sabatini 09 Feb '12

09 Feb '12

Hi, I'm thinking to implement some convenience utils: static inline size_t av_get_ts_string(char *buf, size_t buf_size, int64_t ts) { return ts == (AV_NOPTS_VALUE) ? snprintf(buf, buf_size, "NOPTS") : snprintf(buf, buf_size, "%"PRId64"", ts); } static inline size_t av_get_ts_time_string(char *buf, size_t buf_size, int64_t ts, AVRational *tb) { return ts == (AV_NOPTS_VALUE) ? snprintf(buf, buf_size, "NOPTS") : snprintf(buf, buf_size, "%f", avq2d(tb) * ts; } since it seems I need to use them again and again (debugging timestamp issues in various components). Do you think this would be a good idea? If yes, where should I put them (libavutil/printutils.h may be an option)? -- FFmpeg = Forgiving Free Mastering Philosofic Everlasting Guide

6 21