lavfi: add asetpts audio filter

Based on an idea by Andrey Utkin <andrey.krieger.utkin@gmail.com>.

6 files changed, 92 insertions, 12 deletions

diff --git a/Changelog b/Changelog
index fd8e175e80..082aca7a56 100644
--- a/Changelog
+++ b/Changelog
@@ -43,6 +43,7 @@ version next:
 - MP2 encoding via TwoLAME
 - bmp parser
 - smptebars source
+- asetpts filter
 
 
 version 0.11:
diff --git a/doc/filters.texi b/doc/filters.texi
index e73fc09431..a5627525a9 100644
--- a/doc/filters.texi
+++ b/doc/filters.texi
@@ -3032,9 +3032,11 @@ Mark the frame as top-field-first.
 Mark the frame as progressive.
 @end table
 
-@section setpts
+@section asetpts, setpts
 
-Change the PTS (presentation timestamp) of the input video frames.
+Change the PTS (presentation timestamp) of the input frames.
+
+@code{asetpts} works on audio frames, @code{setpts} on video frames.
 
 Accept in input an expression evaluated through the eval API, which
 can contain the following constants:
@@ -3046,6 +3048,16 @@ the presentation timestamp in input
 @item N
 the count of the input frame, starting from 0.
 
+@item NB_CONSUMED_SAMPLES
+the number of consumed samples, not including the current frame (only
+audio)
+
+@item NB_SAMPLES
+the number of samples in the current frame (only audio)
+
+@item SAMPLE_RATE
+audio sample rate
+
 @item STARTPTS
 the PTS of the first video frame
 
@@ -3084,6 +3096,9 @@ setpts=N/(25*TB)
 
 # fixed rate 25 fps with some jitter
 setpts='1/(25*TB) * (N + 0.05 * sin(N*2*PI/25))'
+
+# apply an offset of 10 seconds to the input PTS
+setpts=PTS+10/TB
 @end example
 
 @section settb, asettb
diff --git a/libavfilter/Makefile b/libavfilter/Makefile
index b6bd37f148..d50c99d148 100644
--- a/libavfilter/Makefile
+++ b/libavfilter/Makefile
@@ -52,6 +52,7 @@ OBJS-$(CONFIG_AMIX_FILTER)                   += af_amix.o
 OBJS-$(CONFIG_ANULL_FILTER)                  += af_anull.o
 OBJS-$(CONFIG_ARESAMPLE_FILTER)              += af_aresample.o
 OBJS-$(CONFIG_ASETNSAMPLES_FILTER)           += af_asetnsamples.o
+OBJS-$(CONFIG_ASETPTS_FILTER)                += vf_setpts.o
 OBJS-$(CONFIG_ASETTB_FILTER)                 += f_settb.o
 OBJS-$(CONFIG_ASHOWINFO_FILTER)              += af_ashowinfo.o
 OBJS-$(CONFIG_ASPLIT_FILTER)                 += split.o
diff --git a/libavfilter/allfilters.c b/libavfilter/allfilters.c
index da1c8e62ae..5dc75a6b4d 100644
--- a/libavfilter/allfilters.c
+++ b/libavfilter/allfilters.c
@@ -42,6 +42,7 @@ void avfilter_register_all(void)
     REGISTER_FILTER (ANULL,       anull,       af);
     REGISTER_FILTER (ARESAMPLE,   aresample,   af);
     REGISTER_FILTER (ASETNSAMPLES, asetnsamples, af);
+    REGISTER_FILTER (ASETPTS,     asetpts,     af);
     REGISTER_FILTER (ASETTB,      asettb,      af);
     REGISTER_FILTER (ASHOWINFO,   ashowinfo,   af);
     REGISTER_FILTER (ASPLIT,      asplit,      af);
diff --git a/libavfilter/version.h b/libavfilter/version.h
index 60660d8090..a47262fb65 100644
--- a/libavfilter/version.h
+++ b/libavfilter/version.h
@@ -29,7 +29,7 @@
 #include "libavutil/avutil.h"
 
 #define LIBAVFILTER_VERSION_MAJOR  3
-#define LIBAVFILTER_VERSION_MINOR  6
+#define LIBAVFILTER_VERSION_MINOR  7
 #define LIBAVFILTER_VERSION_MICRO 100
 
 #define LIBAVFILTER_VERSION_INT AV_VERSION_INT(LIBAVFILTER_VERSION_MAJOR, \
diff --git a/libavfilter/vf_setpts.c b/libavfilter/vf_setpts.c
index 4e2c8f83b9..0f2382ae76 100644
--- a/libavfilter/vf_setpts.c
+++ b/libavfilter/vf_setpts.c
@@ -30,16 +30,20 @@
 #include "libavutil/mathematics.h"
 #include "avfilter.h"
 #include "internal.h"
+#include "audio.h"
 #include "video.h"
 
 static const char *const var_names[] = {
     "INTERLACED",  ///< tell if the current frame is interlaced
     "N",           ///< frame number (starting at zero)
+    "NB_CONSUMED_SAMPLES", ///< number of samples consumed by the filter (only audio)
+    "NB_SAMPLES",  ///< number of samples in the current frame (only audio)
     "POS",         ///< original position in the file of the frame
     "PREV_INPTS",  ///< previous  input PTS
     "PREV_OUTPTS", ///< previous output PTS
     "PTS",         ///< original pts in the file of the frame
-    "STARTPTS",   ///< PTS at start of movie
+    "SAMPLE_RATE", ///< sample rate (only audio)
+    "STARTPTS",    ///< PTS at start of movie
     "TB",          ///< timebase
     NULL
 };
@@ -47,10 +51,13 @@ static const char *const var_names[] = {
 enum var_name {
     VAR_INTERLACED,
     VAR_N,
+    VAR_NB_CONSUMED_SAMPLES,
+    VAR_NB_SAMPLES,
     VAR_POS,
     VAR_PREV_INPTS,
     VAR_PREV_OUTPTS,
     VAR_PTS,
+    VAR_SAMPLE_RATE,
     VAR_STARTPTS,
     VAR_TB,
     VAR_VARS_NB
@@ -59,6 +66,7 @@ enum var_name {
 typedef struct {
     AVExpr *expr;
     double var_values[VAR_VARS_NB];
+    enum AVMediaType type;
 } SetPTSContext;
 
 static av_cold int init(AVFilterContext *ctx, const char *args)
@@ -81,18 +89,24 @@ static av_cold int init(AVFilterContext *ctx, const char *args)
 
 static int config_input(AVFilterLink *inlink)
 {
-    SetPTSContext *setpts = inlink->dst->priv;
+    AVFilterContext *ctx = inlink->dst;
+    SetPTSContext *setpts = ctx->priv;
 
+    setpts->type = inlink->type;
     setpts->var_values[VAR_TB] = av_q2d(inlink->time_base);
 
-    av_log(inlink->src, AV_LOG_VERBOSE, "TB:%f\n", setpts->var_values[VAR_TB]);
+    if (setpts->type == AVMEDIA_TYPE_AUDIO)
+        setpts->var_values[VAR_SAMPLE_RATE] = inlink->sample_rate;
+
+    av_log(inlink->src, AV_LOG_VERBOSE, "TB:%f SAMPLE_RATE:%f\n",
+           setpts->var_values[VAR_TB], setpts->var_values[VAR_SAMPLE_RATE]);
     return 0;
 }
 
 #define D2TS(d)  (isnan(d) ? AV_NOPTS_VALUE : (int64_t)(d))
 #define TS2D(ts) ((ts) == AV_NOPTS_VALUE ? NAN : (double)(ts))
 
-static int start_frame(AVFilterLink *inlink, AVFilterBufferRef *inpicref)
+static int filter_frame(AVFilterLink *inlink, AVFilterBufferRef *inpicref)
 {
     SetPTSContext *setpts = inlink->dst->priv;
     double d;
@@ -103,28 +117,45 @@ static int start_frame(AVFilterLink *inlink, AVFilterBufferRef *inpicref)
 
     if (isnan(setpts->var_values[VAR_STARTPTS]))
         setpts->var_values[VAR_STARTPTS] = TS2D(inpicref->pts);
-
-    setpts->var_values[VAR_INTERLACED] = inpicref->video->interlaced;
     setpts->var_values[VAR_PTS       ] = TS2D(inpicref->pts);
     setpts->var_values[VAR_POS       ] = inpicref->pos == -1 ? NAN : inpicref->pos;
 
+    switch (inlink->type) {
+    case AVMEDIA_TYPE_VIDEO:
+        setpts->var_values[VAR_INTERLACED] = inpicref->video->interlaced;
+        break;
+
+    case AVMEDIA_TYPE_AUDIO:
+        setpts->var_values[VAR_NB_SAMPLES] = inpicref->audio->nb_samples;
+        break;
+    }
+
     d = av_expr_eval(setpts->expr, setpts->var_values, NULL);
     outpicref->pts = D2TS(d);
 
 #ifdef DEBUG
     av_log(inlink->dst, AV_LOG_DEBUG,
-           "n:%"PRId64" interlaced:%d pos:%"PRId64" pts:%"PRId64" t:%f -> pts:%"PRId64" t:%f\n",
+           "n:%"PRId64" interlaced:%d nb_samples:%d nb_consumed_samples:%d "
+           "pos:%"PRId64" pts:%"PRId64" t:%f -> pts:%"PRId64" t:%f\n",
            (int64_t)setpts->var_values[VAR_N],
            (int)setpts->var_values[VAR_INTERLACED],
+           (int)setpts->var_values[VAR_NB_SAMPLES],
+           (int)setpts->var_values[VAR_NB_CONSUMED_SAMPLES],
            inpicref ->pos,
            inpicref ->pts, inpicref ->pts * av_q2d(inlink->time_base),
            outpicref->pts, outpicref->pts * av_q2d(inlink->time_base));
+
 #endif
 
     setpts->var_values[VAR_N] += 1.0;
     setpts->var_values[VAR_PREV_INPTS ] = TS2D(inpicref ->pts);
     setpts->var_values[VAR_PREV_OUTPTS] = TS2D(outpicref->pts);
-    return ff_start_frame(inlink->dst->outputs[0], outpicref);
+
+    if (setpts->type == AVMEDIA_TYPE_AUDIO) {
+        setpts->var_values[VAR_NB_CONSUMED_SAMPLES] += inpicref->audio->nb_samples;
+        return ff_filter_samples(inlink->dst->outputs[0], outpicref);
+    } else
+        return ff_start_frame   (inlink->dst->outputs[0], outpicref);
 }
 
 static av_cold void uninit(AVFilterContext *ctx)
@@ -134,6 +165,36 @@ static av_cold void uninit(AVFilterContext *ctx)
     setpts->expr = NULL;
 }
 
+#if CONFIG_ASETPTS_FILTER
+AVFilter avfilter_af_asetpts = {
+    .name      = "asetpts",
+    .description = NULL_IF_CONFIG_SMALL("Set PTS for the output audio frame."),
+    .init      = init,
+    .uninit    = uninit,
+
+    .priv_size = sizeof(SetPTSContext),
+
+    .inputs = (const AVFilterPad[]) {
+        {
+            .name             = "default",
+            .type             = AVMEDIA_TYPE_AUDIO,
+            .get_audio_buffer = ff_null_get_audio_buffer,
+            .config_props     = config_input,
+            .filter_samples   = filter_frame,
+        },
+        { .name = NULL }
+    },
+    .outputs = (const AVFilterPad[]) {
+        {
+            .name             = "default",
+            .type             = AVMEDIA_TYPE_AUDIO,
+        },
+        { .name = NULL }
+    },
+};
+#endif /* CONFIG_ASETPTS_FILTER */
+
+#if CONFIG_SETPTS_FILTER
 AVFilter avfilter_vf_setpts = {
     .name      = "setpts",
     .description = NULL_IF_CONFIG_SMALL("Set PTS for the output video frame."),
@@ -146,9 +207,10 @@ AVFilter avfilter_vf_setpts = {
                                           .type             = AVMEDIA_TYPE_VIDEO,
                                           .get_video_buffer = ff_null_get_video_buffer,
                                           .config_props     = config_input,
-                                          .start_frame      = start_frame, },
+                                          .start_frame      = filter_frame, },
                                         { .name = NULL }},
     .outputs   = (const AVFilterPad[]) {{ .name             = "default",
                                           .type             = AVMEDIA_TYPE_VIDEO, },
                                         { .name = NULL}},
 };
+#endif /* CONFIG_SETPTS_FILTER */