lavf: Provide a monotonic timestamp to the outside world

This detects and removes single timestamp wraps, which fixes seeking for affected video files.

3 files changed, 151 insertions, 10 deletions

diff --git a/lib/ffmpeg/libavformat/avformat.h b/lib/ffmpeg/libavformat/avformat.h
index c76ac71aec..40f71cb3fc 100644
--- a/lib/ffmpeg/libavformat/avformat.h
+++ b/lib/ffmpeg/libavformat/avformat.h
@@ -627,6 +627,13 @@ typedef struct AVIndexEntry {
 #define AV_DISPOSITION_VISUAL_IMPAIRED   0x0100  /**< stream for visual impaired audiences */
 #define AV_DISPOSITION_CLEAN_EFFECTS     0x0200  /**< stream without voice */
 
+/*
+ * Options for behavior on timestamp wrap detection.
+ */
+#define AV_PTS_WRAP_IGNORE      0   ///< ignore the wrap
+#define AV_PTS_WRAP_ADD_OFFSET  1   ///< add the format specific offset on wrap detection
+#define AV_PTS_WRAP_SUB_OFFSET  -1  ///< subtract the format specific offset on wrap detection
+
 /**
  * Stream structure.
  * New fields can be added to the end with minor version bumps.
@@ -857,6 +864,24 @@ typedef struct AVStream {
 
     int pts_wrap_bits; /**< number of bits in pts (used for wrapping control) */
 #endif
+
+    /**
+     * Internal data to check for wrapping of the time stamp
+     */
+    int64_t pts_wrap_reference;
+
+    /**
+     * Options for behavior, when a wrap is detected.
+     *
+     * Defined by AV_PTS_WRAP_ values.
+     *
+     * If correction is enabled, there are two possibilities:
+     * If the first time stamp is near the wrap point, the wrap offset
+     * will be subtracted, which will create negative time stamps.
+     * Otherwise the offset will be added.
+     */
+    int pts_wrap_behavior;
+
 } AVStream;
 
 #define AV_PROGRAM_RUNNING 1
@@ -878,6 +903,10 @@ typedef struct AVProgram {
     int program_num;
     int pmt_pid;
     int pcr_pid;
+
+    int64_t pts_wrap_reference;    ///< reference dts for wrap detection
+    int pts_wrap_behavior;         ///< behavior on wrap detection
+
 } AVProgram;
 
 #define AVFMTCTX_NOHEADER      0x0001 /**< signal that no header is present
@@ -1234,6 +1263,14 @@ typedef struct AVFormatContext {
     /* av_seek_frame() support */
     int64_t data_offset; /**< offset of the first packet */
 #endif
+
+    /**
+     * Correct single timestamp overflows
+     * - encoding: unused
+     * - decoding: Set by user via AVOPtions (NO direct access)
+     */
+    unsigned int correct_ts_overflow;
+
 } AVFormatContext;
 
 typedef struct AVPacketList {
diff --git a/lib/ffmpeg/libavformat/options.c b/lib/ffmpeg/libavformat/options.c
index 31acbd37c1..666d05e055 100644
--- a/lib/ffmpeg/libavformat/options.c
+++ b/lib/ffmpeg/libavformat/options.c
@@ -127,6 +127,7 @@ static const AVOption options[]={
 {"careful",    "consider things that violate the spec and have not been seen in the wild as errors", 0, AV_OPT_TYPE_CONST, {.dbl = AV_EF_CAREFUL }, INT_MIN, INT_MAX, D, "err_detect"},
 {"compliant",  "consider all spec non compliancies as errors", 0, AV_OPT_TYPE_CONST, {.dbl = AV_EF_COMPLIANT }, INT_MIN, INT_MAX, D, "err_detect"},
 {"aggressive", "consider things that a sane encoder shouldnt do as an error", 0, AV_OPT_TYPE_CONST, {.dbl = AV_EF_AGGRESSIVE }, INT_MIN, INT_MAX, D, "err_detect"},
+{"correct_ts_overflow", "correct single timestamp overflows", OFFSET(correct_ts_overflow), AV_OPT_TYPE_INT, {.dbl = 1}, 0, 1, D},
 {NULL},
 };
 
diff --git a/lib/ffmpeg/libavformat/utils.c b/lib/ffmpeg/libavformat/utils.c
index 63f89dd2e3..89e5ed4f4d 100644
--- a/lib/ffmpeg/libavformat/utils.c
+++ b/lib/ffmpeg/libavformat/utils.c
@@ -122,6 +122,27 @@ static void frac_add(AVFrac *f, int64_t incr)
     f->num = num;
 }
 
+/**
+ * Wrap a given time stamp, if there is an indication for an overflow
+ *
+ * @param st stream
+ * @param timestamp the time stamp to wrap
+ * @return resulting time stamp
+ */
+static int64_t wrap_timestamp(AVStream *st, int64_t timestamp)
+{
+    if (st->pts_wrap_behavior != AV_PTS_WRAP_IGNORE &&
+        st->pts_wrap_reference != AV_NOPTS_VALUE && timestamp != AV_NOPTS_VALUE) {
+        if (st->pts_wrap_behavior == AV_PTS_WRAP_ADD_OFFSET &&
+            timestamp < st->pts_wrap_reference)
+            return timestamp + (1LL<<st->pts_wrap_bits);
+        else if (st->pts_wrap_behavior == AV_PTS_WRAP_SUB_OFFSET &&
+            timestamp >= st->pts_wrap_reference)
+            return timestamp - (1LL<<st->pts_wrap_bits);
+    }
+    return timestamp;
+}
+
 /** head of registered input format linked list */
 static AVInputFormat *first_iformat = NULL;
 /** head of registered output format linked list */
@@ -798,6 +819,8 @@ int av_read_packet(AVFormatContext *s, AVPacket *pkt)
         }
 
         st= s->streams[pkt->stream_index];
+        pkt->dts = wrap_timestamp(st, pkt->dts);
+        pkt->pts = wrap_timestamp(st, pkt->pts);
 
         switch(st->codec->codec_type){
         case AVMEDIA_TYPE_VIDEO:
@@ -948,8 +971,67 @@ static int is_intra_only(AVCodecContext *enc){
     return 0;
 }
 
+static int update_wrap_reference(AVFormatContext *s, AVStream *st, int stream_index)
+{
+    if (s->correct_ts_overflow && st->pts_wrap_bits != 64 &&
+        st->pts_wrap_reference == AV_NOPTS_VALUE && st->first_dts != AV_NOPTS_VALUE) {
+        int i;
+
+        // reference time stamp should be 60 s before first time stamp
+        int64_t pts_wrap_reference = st->first_dts - av_rescale(60, st->time_base.den, st->time_base.num);
+        // if first time stamp is not more than 1/8 and 60s before the wrap point, subtract rather than add wrap offset
+        int pts_wrap_behavior = (st->first_dts < (1LL<<st->pts_wrap_bits) - (1LL<<st->pts_wrap_bits-3)) ||
+            (st->first_dts < (1LL<<st->pts_wrap_bits) - av_rescale(60, st->time_base.den, st->time_base.num)) ?
+            AV_PTS_WRAP_ADD_OFFSET : AV_PTS_WRAP_SUB_OFFSET;
+
+        AVProgram *first_program = av_find_program_from_stream(s, NULL, stream_index);
+
+        if (!first_program) {
+            int default_stream_index = av_find_default_stream_index(s);
+            if (s->streams[default_stream_index]->pts_wrap_reference == AV_NOPTS_VALUE) {
+                for (i=0; i<s->nb_streams; i++) {
+                    s->streams[i]->pts_wrap_reference = pts_wrap_reference;
+                    s->streams[i]->pts_wrap_behavior = pts_wrap_behavior;
+                }
+            }
+            else {
+                st->pts_wrap_reference = s->streams[default_stream_index]->pts_wrap_reference;
+                st->pts_wrap_behavior = s->streams[default_stream_index]->pts_wrap_behavior;
+            }
+        }
+        else {
+            AVProgram *program = first_program;
+            while (program) {
+                if (program->pts_wrap_reference != AV_NOPTS_VALUE) {
+                    pts_wrap_reference = program->pts_wrap_reference;
+                    pts_wrap_behavior = program->pts_wrap_behavior;
+                    break;
+                }
+                program = av_find_program_from_stream(s, program, stream_index);
+            }
+
+            // update every program with differing pts_wrap_reference
+            program = first_program;
+            while(program) {
+                if (program->pts_wrap_reference != pts_wrap_reference) {
+                    for (i=0; i<program->nb_stream_indexes; i++) {
+                        s->streams[program->stream_index[i]]->pts_wrap_reference = pts_wrap_reference;
+                        s->streams[program->stream_index[i]]->pts_wrap_behavior = pts_wrap_behavior;
+                    }
+
+                    program->pts_wrap_reference = pts_wrap_reference;
+                    program->pts_wrap_behavior = pts_wrap_behavior;
+                }
+                program = av_find_program_from_stream(s, program, stream_index);
+            }
+        }
+        return 1;
+    }
+    return 0;
+}
+
 static void update_initial_timestamps(AVFormatContext *s, int stream_index,
-                                      int64_t dts, int64_t pts)
+                                      int64_t dts, int64_t pts, AVPacket *pkt)
 {
     AVStream *st= s->streams[stream_index];
     AVPacketList *pktl= s->packet_buffer;
@@ -973,6 +1055,16 @@ static void update_initial_timestamps(AVFormatContext *s, int stream_index,
         if(st->start_time == AV_NOPTS_VALUE && pktl->pkt.pts != AV_NOPTS_VALUE)
             st->start_time= pktl->pkt.pts;
     }
+
+    if (update_wrap_reference(s, st, stream_index) && st->pts_wrap_behavior == AV_PTS_WRAP_SUB_OFFSET) {
+        // correct first time stamps to negative values
+        st->first_dts = wrap_timestamp(st, st->first_dts);
+        st->cur_dts = wrap_timestamp(st, st->cur_dts);
+        pkt->dts = wrap_timestamp(st, pkt->dts);
+        pkt->pts = wrap_timestamp(st, pkt->pts);
+        pts = wrap_timestamp(st, pts);
+    }
+
     if (st->start_time == AV_NOPTS_VALUE)
         st->start_time = pts;
 }
@@ -1104,7 +1196,7 @@ static void compute_pkt_fields(AVFormatContext *s, AVStream *st,
             /* PTS = presentation timestamp */
             if (pkt->dts == AV_NOPTS_VALUE)
                 pkt->dts = st->last_IP_pts;
-            update_initial_timestamps(s, pkt->stream_index, pkt->dts, pkt->pts);
+            update_initial_timestamps(s, pkt->stream_index, pkt->dts, pkt->pts, pkt);
             if (pkt->dts == AV_NOPTS_VALUE)
                 pkt->dts = st->cur_dts;
 
@@ -1131,7 +1223,7 @@ static void compute_pkt_fields(AVFormatContext *s, AVStream *st,
             /* presentation is not delayed : PTS and DTS are the same */
             if(pkt->pts == AV_NOPTS_VALUE)
                 pkt->pts = pkt->dts;
-            update_initial_timestamps(s, pkt->stream_index, pkt->pts, pkt->pts);
+            update_initial_timestamps(s, pkt->stream_index, pkt->pts, pkt->pts, pkt);
             if(pkt->pts == AV_NOPTS_VALUE)
                 pkt->pts = st->cur_dts;
             pkt->dts = pkt->pts;
@@ -1147,7 +1239,7 @@ static void compute_pkt_fields(AVFormatContext *s, AVStream *st,
         if(pkt->dts == AV_NOPTS_VALUE)
             pkt->dts= st->pts_buffer[0];
         if(st->codec->codec_id == CODEC_ID_H264){ // we skipped it above so we try here
-            update_initial_timestamps(s, pkt->stream_index, pkt->dts, pkt->pts); // this should happen on the first packet
+            update_initial_timestamps(s, pkt->stream_index, pkt->dts, pkt->pts, pkt); // this should happen on the first packet
         }
         if(pkt->dts > st->cur_dts)
             st->cur_dts = pkt->dts;
@@ -1550,6 +1642,7 @@ int ff_add_index_entry(AVIndexEntry **index_entries,
 int av_add_index_entry(AVStream *st,
                        int64_t pos, int64_t timestamp, int size, int distance, int flags)
 {
+    timestamp = wrap_timestamp(st, timestamp);
     return ff_add_index_entry(&st->index_entries, &st->nb_index_entries,
                               &st->index_entries_allocated_size, pos,
                               timestamp, size, distance, flags);
@@ -1602,6 +1695,12 @@ int av_seek_frame_binary(AVFormatContext *s, int stream_index, int64_t target_ts
 }
 #endif
 
+static int64_t ff_read_timestamp(AVFormatContext *s, int stream_index, int64_t *ppos, int64_t pos_limit,
+                                 int64_t (*read_timestamp)(struct AVFormatContext *, int , int64_t *, int64_t ))
+{
+    return wrap_timestamp(s->streams[stream_index], read_timestamp(s, stream_index, ppos, pos_limit));
+}
+
 int ff_seek_frame_binary(AVFormatContext *s, int stream_index, int64_t target_ts, int flags)
 {
     AVInputFormat *avif= s->iformat;
@@ -1689,7 +1788,7 @@ int64_t ff_gen_search(AVFormatContext *s, int stream_index, int64_t target_ts,
 
     if(ts_min == AV_NOPTS_VALUE){
         pos_min = s->data_offset;
-        ts_min = read_timestamp(s, stream_index, &pos_min, INT64_MAX);
+        ts_min = ff_read_timestamp(s, stream_index, &pos_min, INT64_MAX, read_timestamp);
         if (ts_min == AV_NOPTS_VALUE)
             return -1;
     }
@@ -1705,7 +1804,7 @@ int64_t ff_gen_search(AVFormatContext *s, int stream_index, int64_t target_ts,
         pos_max = filesize - 1;
         do{
             pos_max -= step;
-            ts_max = read_timestamp(s, stream_index, &pos_max, pos_max + step);
+            ts_max = ff_read_timestamp(s, stream_index, &pos_max, pos_max + step, read_timestamp);
             step += step;
         }while(ts_max == AV_NOPTS_VALUE && pos_max >= step);
         if (ts_max == AV_NOPTS_VALUE)
@@ -1713,7 +1812,7 @@ int64_t ff_gen_search(AVFormatContext *s, int stream_index, int64_t target_ts,
 
         for(;;){
             int64_t tmp_pos= pos_max + 1;
-            int64_t tmp_ts= read_timestamp(s, stream_index, &tmp_pos, INT64_MAX);
+            int64_t tmp_ts= ff_read_timestamp(s, stream_index, &tmp_pos, INT64_MAX, read_timestamp);
             if(tmp_ts == AV_NOPTS_VALUE)
                 break;
             ts_max= tmp_ts;
@@ -1760,7 +1859,7 @@ int64_t ff_gen_search(AVFormatContext *s, int stream_index, int64_t target_ts,
             pos= pos_limit;
         start_pos= pos;
 
-        ts = read_timestamp(s, stream_index, &pos, INT64_MAX); //may pass pos_limit instead of -1
+        ts = ff_read_timestamp(s, stream_index, &pos, INT64_MAX, read_timestamp); //may pass pos_limit instead of -1
         if(pos == pos_max)
             no_change++;
         else
@@ -1788,9 +1887,9 @@ int64_t ff_gen_search(AVFormatContext *s, int stream_index, int64_t target_ts,
     ts  = (flags & AVSEEK_FLAG_BACKWARD) ?  ts_min :  ts_max;
 #if 0
     pos_min = pos;
-    ts_min = read_timestamp(s, stream_index, &pos_min, INT64_MAX);
+    ts_min = ff_read_timestamp(s, stream_index, &pos_min, INT64_MAX, read_timestamp);
     pos_min++;
-    ts_max = read_timestamp(s, stream_index, &pos_min, INT64_MAX);
+    ts_max = ff_read_timestamp(s, stream_index, &pos_min, INT64_MAX, read_timestamp);
     av_dlog(s, "pos=0x%"PRIx64" %"PRId64"<=%"PRId64"<=%"PRId64"\n",
             pos, ts_min, target_ts, ts_max);
 #endif
@@ -2992,6 +3091,8 @@ AVStream *avformat_new_stream(AVFormatContext *s, AVCodec *c)
     st->cur_dts = 0;
     st->first_dts = AV_NOPTS_VALUE;
     st->probe_packets = MAX_PROBE_PACKETS;
+    st->pts_wrap_reference = AV_NOPTS_VALUE;
+    st->pts_wrap_behavior = AV_PTS_WRAP_IGNORE;
 
     /* default pts setting is MPEG-like */
     avpriv_set_pts_info(st, 33, 1, 90000);
@@ -3025,6 +3126,8 @@ AVProgram *av_new_program(AVFormatContext *ac, int id)
         program->discard = AVDISCARD_NONE;
     }
     program->id = id;
+    program->pts_wrap_reference = AV_NOPTS_VALUE;
+    program->pts_wrap_behavior = AV_PTS_WRAP_IGNORE;
 
     return program;
 }