qtdemux: Always use `tfdt` if available in BYTE segments

This reverts the decision from
  https://bugzilla.gnome.org/show_bug.cgi?id=754230
where it was decided that we rather play safe and only use the `tfdt` if
it is "significantly different" to the sum of sample durations.

As the specification says

    If the time expressed in the track fragment decode time (‘tfdt’) box
    exceeds the sum of the durations of the samples in the preceding
    movie and movie fragments, then the duration of the last sample
    preceding this track fragment is extended such that the sum now
    equals the time given in this box.

we have to use the `tfdt` in general to allow for it to signal gaps in
the stream.

A muxer producing fragments might not yet know the full duration of the
last sample of a previous fragment if the next fragment starts with a
gap, and knowing the actual start of the next fragment would potentially
require to violate latency requirements.

Additionally, the existence of `tfdt` allows to avoid accumulating
rounding errors from summing up the durations.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/3586>

diff --git a/subprojects/gst-plugins-good/gst/isomp4/qtdemux.c b/subprojects/gst-plugins-good/gst/isomp4/qtdemux.c
index 845f1a0..7b1209b 100644 (file)
--- a/subprojects/gst-plugins-good/gst/isomp4/qtdemux.c
+++ b/subprojects/gst-plugins-good/gst/isomp4/qtdemux.c
@@ -3477,33 +3477,22 @@ qtdemux_parse_trun (GstQTDemux * qtdemux, GstByteReader * trun,
       GST_INFO_OBJECT (stream->pad, "first sample ts %" GST_TIME_FORMAT,
           GST_TIME_ARGS (gst_ts));
     } else {
-      /* subsequent fragments extend stream */
-      timestamp =
-          stream->samples[stream->n_samples - 1].timestamp +
-          stream->samples[stream->n_samples - 1].duration;
-
-      /* If this is a GST_FORMAT_BYTES stream and there's a significant
-       * difference (1 sec.) between decode_ts and timestamp, prefer the
-       * former */
-      if (has_tfdt && !qtdemux->upstream_format_is_time
-          && ABSDIFF (decode_ts, timestamp) >
-          MAX (stream->duration_last_moof / 2,
-              GSTTIME_TO_QTSTREAMTIME (stream, GST_SECOND))) {
-        GST_INFO_OBJECT (qtdemux,
-            "decode_ts (%" GST_TIME_FORMAT ") and timestamp (%" GST_TIME_FORMAT
-            ") are significantly different (more than %" GST_TIME_FORMAT
-            "), using decode_ts",
-            GST_TIME_ARGS (QTSTREAMTIME_TO_GSTTIME (stream, decode_ts)),
-            GST_TIME_ARGS (QTSTREAMTIME_TO_GSTTIME (stream, timestamp)),
-            GST_TIME_ARGS (QTSTREAMTIME_TO_GSTTIME (stream,
-                    MAX (stream->duration_last_moof / 2,
-                        GSTTIME_TO_QTSTREAMTIME (stream, GST_SECOND)))));
+      /* If this is a GST_FORMAT_BYTES stream and we have a tfdt then use it
+       * instead of the sum of sample durations */
+      if (has_tfdt && !qtdemux->upstream_format_is_time) {
         timestamp = decode_ts;
+        gst_ts = QTSTREAMTIME_TO_GSTTIME (stream, timestamp);
+        GST_INFO_OBJECT (qtdemux, "first sample ts %" GST_TIME_FORMAT
+            " (using tfdt)", GST_TIME_ARGS (gst_ts));
+      } else {
+        /* subsequent fragments extend stream */
+        timestamp =
+            stream->samples[stream->n_samples - 1].timestamp +
+            stream->samples[stream->n_samples - 1].duration;
+        gst_ts = QTSTREAMTIME_TO_GSTTIME (stream, timestamp);
+        GST_INFO_OBJECT (qtdemux, "first sample ts %" GST_TIME_FORMAT
+            " (extends previous samples)", GST_TIME_ARGS (gst_ts));
       }
-
-      gst_ts = QTSTREAMTIME_TO_GSTTIME (stream, timestamp);
-      GST_INFO_OBJECT (qtdemux, "first sample ts %" GST_TIME_FORMAT
-          " (extends previous samples)", GST_TIME_ARGS (gst_ts));
     }
   }