ffmpeg: add enc_time_base option

add a per-stream option for setting the encoder timebase.
the following values are allowed:
0 - for video, use 1/frame_rate, for audio use 1/sample_rate (this is
  the default)
-1 - match the input timebase (when possible)
>0 - set the timebase to provided number

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

doc/ffmpeg.texi

@@ -1176,6 +1176,30 @@ Try to make the choice automatically, in order to generate a sane output.
 
 Default value is -1.
 
+@item -enc_time_base[:@var{stream_specifier}] @var{timebase} (@emph{output,per-stream})
+Set the encoder timebase. @var{timebase} is a floating point number,
+and can assume one of the following values:
+
+@table @option
+@item 0
+Assign a default value according to the media type.
+
+For video - use 1/framerate, for audio - use 1/samplerate.
+
+@item -1
+Use the input stream timebase when possible.
+
+If an input stream is not available, the default timebase will be used.
+
+@item >0
+Use the provided number as the timebase.
+
+This field can be provided as a ratio of two integers (e.g. 1:24, 1:48000)
+or as a floating point number (e.g. 0.04166, 2.0833e-5)
+@end table
+
+Default value is 0.
+
 @item -shortest (@emph{output})
 Finish encoding when the shortest input stream ends.
 @item -dts_delta_threshold

ffmpeg.c

@@ -3234,6 +3234,30 @@ static void parse_forced_key_frames(char *kf, OutputStream *ost,
     ost->forced_kf_pts   = pts;
 }
 
+static void init_encoder_time_base(OutputStream *ost, AVRational default_time_base)
+{
+    InputStream *ist = get_input_stream(ost);
+    AVCodecContext *enc_ctx = ost->enc_ctx;
+    AVFormatContext *oc;
+
+    if (ost->enc_timebase.num > 0) {
+        enc_ctx->time_base = ost->enc_timebase;
+        return;
+    }
+
+    if (ost->enc_timebase.num < 0) {
+        if (ist) {
+            enc_ctx->time_base = ist->st->time_base;
+            return;
+        }
+
+        oc = output_files[ost->file_index]->ctx;
+        av_log(oc, AV_LOG_WARNING, "Input stream data not available, using default time base\n");
+    }
+
+    enc_ctx->time_base = default_time_base;
+}
+
 static int init_output_stream_encode(OutputStream *ost)
 {
     InputStream *ist = get_input_stream(ost);
@@ -3304,10 +3328,13 @@ static int init_output_stream_encode(OutputStream *ost)
         enc_ctx->sample_rate    = av_buffersink_get_sample_rate(ost->filter->filter);
         enc_ctx->channel_layout = av_buffersink_get_channel_layout(ost->filter->filter);
         enc_ctx->channels       = av_buffersink_get_channels(ost->filter->filter);
-        enc_ctx->time_base      = (AVRational){ 1, enc_ctx->sample_rate };
+
+        init_encoder_time_base(ost, av_make_q(1, enc_ctx->sample_rate));
         break;
+
     case AVMEDIA_TYPE_VIDEO:
-        enc_ctx->time_base = av_inv_q(ost->frame_rate);
+        init_encoder_time_base(ost, av_inv_q(ost->frame_rate));
+
         if (!(enc_ctx->time_base.num && enc_ctx->time_base.den))
             enc_ctx->time_base = av_buffersink_get_time_base(ost->filter->filter);
         if (   av_q2d(enc_ctx->time_base) < 0.001 && video_sync_method != VSYNC_PASSTHROUGH

ffmpeg.h

@@ -226,6 +226,8 @@ typedef struct OptionsContext {
     int        nb_program;
     SpecifierOpt *time_bases;
     int        nb_time_bases;
+    SpecifierOpt *enc_time_bases;
+    int        nb_enc_time_bases;
 } OptionsContext;
 
 typedef struct InputFilter {
@@ -453,6 +455,7 @@ typedef struct OutputStream {
     int64_t last_mux_dts;
     // the timebase of the packets sent to the muxer
     AVRational mux_timebase;
+    AVRational enc_timebase;
 
     int                    nb_bitstream_filters;
     uint8_t                  *bsf_extradata_updated;

ffmpeg_opt.c

@@ -1311,6 +1311,17 @@ static OutputStream *new_output_stream(OptionsContext *o, AVFormatContext *oc, e
         st->time_base = q;
     }
 
+    MATCH_PER_STREAM_OPT(enc_time_bases, str, time_base, oc, st);
+    if (time_base) {
+        AVRational q;
+        if (av_parse_ratio(&q, time_base, INT_MAX, 0, NULL) < 0 ||
+            q.den <= 0) {
+            av_log(NULL, AV_LOG_FATAL, "Invalid time base: %s\n", time_base);
+            exit_program(1);
+        }
+        ost->enc_timebase = q;
+    }
+
     ost->max_frames = INT64_MAX;
     MATCH_PER_STREAM_OPT(max_frames, i64, ost->max_frames, oc, st);
     for (i = 0; i<o->nb_max_frames; i++) {
@@ -3629,6 +3640,11 @@ const OptionDef options[] = {
 
     { "time_base", HAS_ARG | OPT_STRING | OPT_EXPERT | OPT_SPEC | OPT_OUTPUT, { .off = OFFSET(time_bases) },
         "set the desired time base hint for output stream (1:24, 1:48000 or 0.04166, 2.0833e-5)", "ratio" },
+    { "enc_time_base", HAS_ARG | OPT_STRING | OPT_EXPERT | OPT_SPEC | OPT_OUTPUT, { .off = OFFSET(enc_time_bases) },
+        "set the desired time base for the encoder (1:24, 1:48000 or 0.04166, 2.0833e-5). "
+        "two special values are defined - "
+        "0 = use frame rate (video) or sample rate (audio),"
+        "-1 = match source time base", "ratio" },
 
     { "bsf", HAS_ARG | OPT_STRING | OPT_SPEC | OPT_EXPERT | OPT_OUTPUT, { .off = OFFSET(bitstream_filters) },
         "A comma-separated list of bitstream filters", "bitstream_filters" },