doc/examples: Add example transcoding.c
authorAndrey Utkin <andrey.krieger.utkin@gmail.com>
Sun, 9 Mar 2014 00:20:41 +0000 (02:20 +0200)
committerMichael Niedermayer <michaelni@gmx.at>
Mon, 17 Mar 2014 19:56:33 +0000 (20:56 +0100)
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
configure
doc/Makefile
doc/examples/Makefile
doc/examples/transcoding.c [new file with mode: 0644]

index 7e51d06..8b67be7 100755 (executable)
--- a/configure
+++ b/configure
@@ -1272,6 +1272,7 @@ EXAMPLE_LIST="
     resampling_audio_example
     scaling_video_example
     transcode_aac_example
+    transcoding_example
 "
 
 EXTERNAL_LIBRARY_LIST="
@@ -2482,6 +2483,7 @@ remuxing_example_deps="avcodec avformat avutil"
 resampling_audio_example_deps="avutil swresample"
 scaling_video_example_deps="avutil swscale"
 transcode_aac_example_deps="avcodec avformat swresample"
+transcoding_example_deps="avfilter avcodec avformat avutil"
 
 # libraries
 avcodec_deps="avutil"
index 44049c4..10f03cc 100644 (file)
@@ -48,6 +48,7 @@ DOC_EXAMPLES-$(CONFIG_REMUXING_EXAMPLE)          += remuxing
 DOC_EXAMPLES-$(CONFIG_RESAMPLING_AUDIO_EXAMPLE)  += resampling_audio
 DOC_EXAMPLES-$(CONFIG_SCALING_VIDEO_EXAMPLE)     += scaling_video
 DOC_EXAMPLES-$(CONFIG_TRANSCODE_AAC_EXAMPLE)     += transcode_aac
+DOC_EXAMPLES-$(CONFIG_TRANSCODING_EXAMPLE)       += transcoding
 ALL_DOC_EXAMPLES_LIST = $(DOC_EXAMPLES-) $(DOC_EXAMPLES-yes)
 
 DOC_EXAMPLES       := $(DOC_EXAMPLES-yes:%=doc/examples/%$(PROGSSUF)$(EXESUF))
index 1553bab..646a63d 100644 (file)
@@ -22,6 +22,7 @@ EXAMPLES=       avio_reading                       \
                 resampling_audio                   \
                 scaling_video                      \
                 transcode_aac                      \
+                transcoding                        \
 
 OBJS=$(addsuffix .o,$(EXAMPLES))
 
diff --git a/doc/examples/transcoding.c b/doc/examples/transcoding.c
new file mode 100644 (file)
index 0000000..93187fa
--- /dev/null
@@ -0,0 +1,597 @@
+/*
+ * Copyright (c) 2010 Nicolas George
+ * Copyright (c) 2011 Stefano Sabatini
+ * Copyright (c) 2014 Andrey Utkin
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a copy
+ * of this software and associated documentation files (the "Software"), to deal
+ * in the Software without restriction, including without limitation the rights
+ * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+ * copies of the Software, and to permit persons to whom the Software is
+ * furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
+ * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+ * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
+ * THE SOFTWARE.
+ */
+
+/**
+ * @file
+ * API example for demuxing, decoding, filtering, encoding and muxing
+ * @example doc/examples/transcoding.c
+ */
+
+#include <libavcodec/avcodec.h>
+#include <libavformat/avformat.h>
+#include <libavfilter/avfiltergraph.h>
+#include <libavfilter/avcodec.h>
+#include <libavfilter/buffersink.h>
+#include <libavfilter/buffersrc.h>
+#include <libavutil/opt.h>
+#include <libavutil/pixdesc.h>
+
+static AVFormatContext *ifmt_ctx;
+static AVFormatContext *ofmt_ctx;
+typedef struct FilteringContext {
+    AVFilterContext *buffersink_ctx;
+    AVFilterContext *buffersrc_ctx;
+    AVFilterGraph *filter_graph;
+} FilteringContext;
+static FilteringContext *filter_ctx;
+
+static int open_input_file(const char *filename)
+{
+    int ret;
+    unsigned int i;
+
+    ifmt_ctx = NULL;
+    if ((ret = avformat_open_input(&ifmt_ctx, filename, NULL, NULL)) < 0) {
+        av_log(NULL, AV_LOG_ERROR, "Cannot open input file\n");
+        return ret;
+    }
+
+    if ((ret = avformat_find_stream_info(ifmt_ctx, NULL)) < 0) {
+        av_log(NULL, AV_LOG_ERROR, "Cannot find stream information\n");
+        return ret;
+    }
+
+    for (i = 0; i < ifmt_ctx->nb_streams; i++) {
+        AVStream *stream;
+        AVCodecContext *codec_ctx;
+        stream = ifmt_ctx->streams[i];
+        codec_ctx = stream->codec;
+        /* Reencode video & audio and remux subtitles etc. */
+        if (codec_ctx->codec_type == AVMEDIA_TYPE_VIDEO
+                || codec_ctx->codec_type == AVMEDIA_TYPE_AUDIO) {
+            /* Open decoder */
+            ret = avcodec_open2(codec_ctx,
+                    avcodec_find_decoder(codec_ctx->codec_id), NULL);
+            if (ret < 0) {
+                av_log(NULL, AV_LOG_ERROR, "Failed to open decoder for stream #%u\n", i);
+                return ret;
+            }
+        }
+    }
+
+    av_dump_format(ifmt_ctx, 0, filename, 0);
+    return 0;
+}
+
+static int open_output_file(const char *filename)
+{
+    AVStream *out_stream;
+    AVStream *in_stream;
+    AVCodecContext *dec_ctx, *enc_ctx;
+    AVCodec *encoder;
+    int ret;
+    unsigned int i;
+
+    ofmt_ctx = NULL;
+    avformat_alloc_output_context2(&ofmt_ctx, NULL, NULL, filename);
+    if (!ofmt_ctx) {
+        av_log(NULL, AV_LOG_ERROR, "Could not create output context\n");
+        return AVERROR_UNKNOWN;
+    }
+
+
+    for (i = 0; i < ifmt_ctx->nb_streams; i++) {
+        out_stream = avformat_new_stream(ofmt_ctx, NULL);
+        if (!out_stream) {
+            av_log(NULL, AV_LOG_ERROR, "Failed allocating output stream\n");
+            return AVERROR_UNKNOWN;
+        }
+
+        in_stream = ifmt_ctx->streams[i];
+        dec_ctx = in_stream->codec;
+        enc_ctx = out_stream->codec;
+
+        if (dec_ctx->codec_type == AVMEDIA_TYPE_VIDEO
+                || dec_ctx->codec_type == AVMEDIA_TYPE_AUDIO) {
+            /* in this example, we choose transcoding to same codec */
+            encoder = avcodec_find_encoder(dec_ctx->codec_id);
+
+            /* In this example, we transcode to same properties (picture size,
+             * sample rate etc.). These properties can be changed for output
+             * streams easily using filters */
+            if (dec_ctx->codec_type == AVMEDIA_TYPE_VIDEO) {
+                enc_ctx->height = dec_ctx->height;
+                enc_ctx->width = dec_ctx->width;
+                enc_ctx->sample_aspect_ratio = dec_ctx->sample_aspect_ratio;
+                /* take first format from list of supported formats */
+                enc_ctx->pix_fmt = encoder->pix_fmts[0];
+                /* video time_base can be set to whatever is handy and supported by encoder */
+                enc_ctx->time_base = dec_ctx->time_base;
+            } else {
+                enc_ctx->sample_rate = dec_ctx->sample_rate;
+                enc_ctx->channel_layout = dec_ctx->channel_layout;
+                enc_ctx->channels = av_get_channel_layout_nb_channels(enc_ctx->channel_layout);
+                /* take first format from list of supported formats */
+                enc_ctx->sample_fmt = encoder->sample_fmts[0];
+                enc_ctx->time_base = (AVRational){1, enc_ctx->sample_rate};
+            }
+
+            /* Third parameter can be used to pass settings to encoder */
+            ret = avcodec_open2(enc_ctx, encoder, NULL);
+            if (ret < 0) {
+                av_log(NULL, AV_LOG_ERROR, "Cannot open video encoder for stream #%u\n", i);
+                return ret;
+            }
+        } else if (dec_ctx->codec_type == AVMEDIA_TYPE_UNKNOWN) {
+            av_log(NULL, AV_LOG_FATAL, "Elementary stream #%d is of unknown type, cannot proceed\n", i);
+            return AVERROR_INVALIDDATA;
+        } else {
+            /* if this stream must be remuxed */
+            ret = avcodec_copy_context(ofmt_ctx->streams[i]->codec,
+                    ifmt_ctx->streams[i]->codec);
+            if (ret < 0) {
+                av_log(NULL, AV_LOG_ERROR, "Copying stream context failed\n");
+                return ret;
+            }
+        }
+
+        if (ofmt_ctx->oformat->flags & AVFMT_GLOBALHEADER)
+            enc_ctx->flags |= CODEC_FLAG_GLOBAL_HEADER;
+
+    }
+    av_dump_format(ofmt_ctx, 0, filename, 1);
+
+    if (!(ofmt_ctx->oformat->flags & AVFMT_NOFILE)) {
+        ret = avio_open(&ofmt_ctx->pb, filename, AVIO_FLAG_WRITE);
+        if (ret < 0) {
+            av_log(NULL, AV_LOG_ERROR, "Could not open output file '%s'", filename);
+            return ret;
+        }
+    }
+
+    /* init muxer, write output file header */
+    ret = avformat_write_header(ofmt_ctx, NULL);
+    if (ret < 0) {
+        av_log(NULL, AV_LOG_ERROR, "Error occurred when opening output file\n");
+        return ret;
+    }
+
+    return 0;
+}
+
+static int init_filter(FilteringContext* fctx, AVCodecContext *dec_ctx,
+        AVCodecContext *enc_ctx, const char *filter_spec)
+{
+    char args[512];
+    int ret = 0;
+    AVFilter *buffersrc = NULL;
+    AVFilter *buffersink = NULL;
+    AVFilterContext *buffersrc_ctx = NULL;
+    AVFilterContext *buffersink_ctx = NULL;
+    AVFilterInOut *outputs = avfilter_inout_alloc();
+    AVFilterInOut *inputs  = avfilter_inout_alloc();
+    AVFilterGraph *filter_graph = avfilter_graph_alloc();
+
+    if (!outputs || !inputs || !filter_graph) {
+        ret = AVERROR(ENOMEM);
+        goto end;
+    }
+
+    if (dec_ctx->codec_type == AVMEDIA_TYPE_VIDEO) {
+        buffersrc = avfilter_get_by_name("buffer");
+        buffersink = avfilter_get_by_name("buffersink");
+        if (!buffersrc || !buffersink) {
+            av_log(NULL, AV_LOG_ERROR, "filtering source or sink element not found\n");
+            ret = AVERROR_UNKNOWN;
+            goto end;
+        }
+
+        snprintf(args, sizeof(args),
+                "video_size=%dx%d:pix_fmt=%d:time_base=%d/%d:pixel_aspect=%d/%d",
+                dec_ctx->width, dec_ctx->height, dec_ctx->pix_fmt,
+                dec_ctx->time_base.num, dec_ctx->time_base.den,
+                dec_ctx->sample_aspect_ratio.num,
+                dec_ctx->sample_aspect_ratio.den);
+
+        ret = avfilter_graph_create_filter(&buffersrc_ctx, buffersrc, "in",
+                args, NULL, filter_graph);
+        if (ret < 0) {
+            av_log(NULL, AV_LOG_ERROR, "Cannot create buffer source\n");
+            goto end;
+        }
+
+        ret = avfilter_graph_create_filter(&buffersink_ctx, buffersink, "out",
+                NULL, NULL, filter_graph);
+        if (ret < 0) {
+            av_log(NULL, AV_LOG_ERROR, "Cannot create buffer sink\n");
+            goto end;
+        }
+
+        ret = av_opt_set_bin(buffersink_ctx, "pix_fmts",
+                (uint8_t*)&enc_ctx->pix_fmt, sizeof(enc_ctx->pix_fmt),
+                AV_OPT_SEARCH_CHILDREN);
+        if (ret < 0) {
+            av_log(NULL, AV_LOG_ERROR, "Cannot set output pixel format\n");
+            goto end;
+        }
+    } else if (dec_ctx->codec_type == AVMEDIA_TYPE_AUDIO) {
+        buffersrc = avfilter_get_by_name("abuffer");
+        buffersink = avfilter_get_by_name("abuffersink");
+        if (!buffersrc || !buffersink) {
+            av_log(NULL, AV_LOG_ERROR, "filtering source or sink element not found\n");
+            ret = AVERROR_UNKNOWN;
+            goto end;
+        }
+
+        if (!dec_ctx->channel_layout)
+            dec_ctx->channel_layout =
+                av_get_default_channel_layout(dec_ctx->channels);
+        snprintf(args, sizeof(args),
+                "time_base=%d/%d:sample_rate=%d:sample_fmt=%s:channel_layout=0x%"PRIx64,
+                dec_ctx->time_base.num, dec_ctx->time_base.den, dec_ctx->sample_rate,
+                av_get_sample_fmt_name(dec_ctx->sample_fmt),
+                dec_ctx->channel_layout);
+        ret = avfilter_graph_create_filter(&buffersrc_ctx, buffersrc, "in",
+                args, NULL, filter_graph);
+        if (ret < 0) {
+            av_log(NULL, AV_LOG_ERROR, "Cannot create audio buffer source\n");
+            goto end;
+        }
+
+        ret = avfilter_graph_create_filter(&buffersink_ctx, buffersink, "out",
+                NULL, NULL, filter_graph);
+        if (ret < 0) {
+            av_log(NULL, AV_LOG_ERROR, "Cannot create audio buffer sink\n");
+            goto end;
+        }
+
+        ret = av_opt_set_bin(buffersink_ctx, "sample_fmts",
+                (uint8_t*)&enc_ctx->sample_fmt, sizeof(enc_ctx->sample_fmt),
+                AV_OPT_SEARCH_CHILDREN);
+        if (ret < 0) {
+            av_log(NULL, AV_LOG_ERROR, "Cannot set output sample format\n");
+            goto end;
+        }
+
+        ret = av_opt_set_bin(buffersink_ctx, "channel_layouts",
+                (uint8_t*)&enc_ctx->channel_layout,
+                sizeof(enc_ctx->channel_layout), AV_OPT_SEARCH_CHILDREN);
+        if (ret < 0) {
+            av_log(NULL, AV_LOG_ERROR, "Cannot set output channel layout\n");
+            goto end;
+        }
+
+        ret = av_opt_set_bin(buffersink_ctx, "sample_rates",
+                (uint8_t*)&enc_ctx->sample_rate, sizeof(enc_ctx->sample_rate),
+                AV_OPT_SEARCH_CHILDREN);
+        if (ret < 0) {
+            av_log(NULL, AV_LOG_ERROR, "Cannot set output sample rate\n");
+            goto end;
+        }
+    } else {
+        ret = AVERROR_UNKNOWN;
+        goto end;
+    }
+
+    /* Endpoints for the filter graph. */
+    outputs->name       = av_strdup("in");
+    outputs->filter_ctx = buffersrc_ctx;
+    outputs->pad_idx    = 0;
+    outputs->next       = NULL;
+
+    inputs->name       = av_strdup("out");
+    inputs->filter_ctx = buffersink_ctx;
+    inputs->pad_idx    = 0;
+    inputs->next       = NULL;
+
+    if (!outputs->name || !inputs->name) {
+        ret = AVERROR(ENOMEM);
+        goto end;
+    }
+
+    if ((ret = avfilter_graph_parse_ptr(filter_graph, filter_spec,
+                    &inputs, &outputs, NULL)) < 0)
+        goto end;
+
+    if ((ret = avfilter_graph_config(filter_graph, NULL)) < 0)
+        goto end;
+
+    /* Fill FilteringContext */
+    fctx->buffersrc_ctx = buffersrc_ctx;
+    fctx->buffersink_ctx = buffersink_ctx;
+    fctx->filter_graph = filter_graph;
+
+end:
+    avfilter_inout_free(&inputs);
+    avfilter_inout_free(&outputs);
+
+    return ret;
+}
+
+static int init_filters(void)
+{
+    const char *filter_spec;
+    unsigned int i;
+    int ret;
+    filter_ctx = av_malloc(sizeof(FilteringContext) * ifmt_ctx->nb_streams);
+    if (!filter_ctx)
+        return AVERROR(ENOMEM);
+
+    for (i = 0; i < ifmt_ctx->nb_streams; i++) {
+        filter_ctx[i].buffersrc_ctx  = NULL;
+        filter_ctx[i].buffersink_ctx = NULL;
+        filter_ctx[i].filter_graph   = NULL;
+        if (!(ifmt_ctx->streams[i]->codec->codec_type == AVMEDIA_TYPE_AUDIO
+                || ifmt_ctx->streams[i]->codec->codec_type == AVMEDIA_TYPE_VIDEO))
+            continue;
+
+
+        if (ifmt_ctx->streams[i]->codec->codec_type == AVMEDIA_TYPE_VIDEO)
+            filter_spec = "null"; /* passthrough (dummy) filter for video */
+        else
+            filter_spec = "anull"; /* passthrough (dummy) filter for audio */
+        ret = init_filter(&filter_ctx[i], ifmt_ctx->streams[i]->codec,
+                ofmt_ctx->streams[i]->codec, filter_spec);
+        if (ret)
+            return ret;
+    }
+    return 0;
+}
+
+static int encode_write_frame(AVFrame *filt_frame, unsigned int stream_index, int *got_frame) {
+    int ret;
+    int got_frame_local;
+    AVPacket enc_pkt;
+    int (*enc_func)(AVCodecContext *, AVPacket *, const AVFrame *, int *) =
+        (ifmt_ctx->streams[stream_index]->codec->codec_type ==
+         AVMEDIA_TYPE_VIDEO) ? avcodec_encode_video2 : avcodec_encode_audio2;
+
+    if (!got_frame)
+        got_frame = &got_frame_local;
+
+    av_log(NULL, AV_LOG_INFO, "Encoding frame\n");
+    /* encode filtered frame */
+    enc_pkt.data = NULL;
+    enc_pkt.size = 0;
+    av_init_packet(&enc_pkt);
+    ret = enc_func(ofmt_ctx->streams[stream_index]->codec, &enc_pkt,
+            filt_frame, got_frame);
+    av_frame_free(&filt_frame);
+    if (ret < 0)
+        return ret;
+    if (!(*got_frame))
+        return 0;
+
+    /* prepare packet for muxing */
+    enc_pkt.stream_index = stream_index;
+    enc_pkt.dts = av_rescale_q_rnd(enc_pkt.dts,
+            ofmt_ctx->streams[stream_index]->codec->time_base,
+            ofmt_ctx->streams[stream_index]->time_base,
+            AV_ROUND_NEAR_INF|AV_ROUND_PASS_MINMAX);
+    enc_pkt.pts = av_rescale_q_rnd(enc_pkt.pts,
+            ofmt_ctx->streams[stream_index]->codec->time_base,
+            ofmt_ctx->streams[stream_index]->time_base,
+            AV_ROUND_NEAR_INF|AV_ROUND_PASS_MINMAX);
+    enc_pkt.duration = av_rescale_q(enc_pkt.duration,
+            ofmt_ctx->streams[stream_index]->codec->time_base,
+            ofmt_ctx->streams[stream_index]->time_base);
+
+    av_log(NULL, AV_LOG_DEBUG, "Muxing frame\n");
+    /* mux encoded frame */
+    ret = av_interleaved_write_frame(ofmt_ctx, &enc_pkt);
+    return ret;
+}
+
+static int filter_encode_write_frame(AVFrame *frame, unsigned int stream_index)
+{
+    int ret;
+    AVFrame *filt_frame;
+
+    av_log(NULL, AV_LOG_INFO, "Pushing decoded frame to filters\n");
+    /* push the decoded frame into the filtergraph */
+    ret = av_buffersrc_add_frame_flags(filter_ctx[stream_index].buffersrc_ctx,
+            frame, 0);
+    if (ret < 0) {
+        av_log(NULL, AV_LOG_ERROR, "Error while feeding the filtergraph\n");
+        return ret;
+    }
+
+    /* pull filtered frames from the filtergraph */
+    while (1) {
+        filt_frame = av_frame_alloc();
+        if (!filt_frame) {
+            ret = AVERROR(ENOMEM);
+            break;
+        }
+        av_log(NULL, AV_LOG_INFO, "Pulling filtered frame from filters\n");
+        ret = av_buffersink_get_frame(filter_ctx[stream_index].buffersink_ctx,
+                filt_frame);
+        if (ret < 0) {
+            /* if no more frames for output - returns AVERROR(EAGAIN)
+             * if flushed and no more frames for output - returns AVERROR_EOF
+             * rewrite retcode to 0 to show it as normal procedure completion
+             */
+            if (ret == AVERROR(EAGAIN) || ret == AVERROR_EOF)
+                ret = 0;
+            av_frame_free(&filt_frame);
+            break;
+        }
+
+        filt_frame->pict_type = AV_PICTURE_TYPE_NONE;
+        ret = encode_write_frame(filt_frame, stream_index, NULL);
+        if (ret < 0)
+            break;
+    }
+
+    return ret;
+}
+
+static int flush_encoder(unsigned int stream_index)
+{
+    int ret;
+    int got_frame;
+
+    if (!(ofmt_ctx->streams[stream_index]->codec->codec->capabilities &
+                CODEC_CAP_DELAY))
+        return 0;
+
+    while (1) {
+        av_log(NULL, AV_LOG_INFO, "Flushing stream #%u encoder\n", stream_index);
+        ret = encode_write_frame(NULL, stream_index, &got_frame);
+        if (ret < 0)
+            break;
+        if (!got_frame)
+            return 0;
+    }
+    return ret;
+}
+
+int main(int argc, char **argv)
+{
+    int ret;
+    AVPacket packet = { .data = NULL, .size = 0 };
+    AVFrame *frame = NULL;
+    enum AVMediaType type;
+    unsigned int stream_index;
+    unsigned int i;
+    int got_frame;
+    int (*dec_func)(AVCodecContext *, AVFrame *, int *, const AVPacket *);
+
+    if (argc != 3) {
+        av_log(NULL, AV_LOG_ERROR, "Usage: %s <input file> <output file>\n", argv[0]);
+        return 1;
+    }
+
+    av_register_all();
+    avfilter_register_all();
+
+    if ((ret = open_input_file(argv[1])) < 0)
+        goto end;
+    if ((ret = open_output_file(argv[2])) < 0)
+        goto end;
+    if ((ret = init_filters()) < 0)
+        goto end;
+
+    /* read all packets */
+    while (1) {
+        if ((ret = av_read_frame(ifmt_ctx, &packet)) < 0)
+            break;
+        stream_index = packet.stream_index;
+        type = ifmt_ctx->streams[packet.stream_index]->codec->codec_type;
+        av_log(NULL, AV_LOG_DEBUG, "Demuxer gave frame of stream_index %u\n",
+                stream_index);
+
+        if (filter_ctx[stream_index].filter_graph) {
+            av_log(NULL, AV_LOG_DEBUG, "Going to reencode&filter the frame\n");
+            frame = av_frame_alloc();
+            if (!frame) {
+                ret = AVERROR(ENOMEM);
+                break;
+            }
+            packet.dts = av_rescale_q_rnd(packet.dts,
+                    ifmt_ctx->streams[stream_index]->time_base,
+                    ifmt_ctx->streams[stream_index]->codec->time_base,
+                    AV_ROUND_NEAR_INF|AV_ROUND_PASS_MINMAX);
+            packet.pts = av_rescale_q_rnd(packet.pts,
+                    ifmt_ctx->streams[stream_index]->time_base,
+                    ifmt_ctx->streams[stream_index]->codec->time_base,
+                    AV_ROUND_NEAR_INF|AV_ROUND_PASS_MINMAX);
+            dec_func = (type == AVMEDIA_TYPE_VIDEO) ? avcodec_decode_video2 :
+                avcodec_decode_audio4;
+            ret = dec_func(ifmt_ctx->streams[stream_index]->codec, frame,
+                    &got_frame, &packet);
+            if (ret < 0) {
+                av_frame_free(&frame);
+                av_log(NULL, AV_LOG_ERROR, "Decoding failed\n");
+                break;
+            }
+
+            if (got_frame) {
+                frame->pts = av_frame_get_best_effort_timestamp(frame);
+                ret = filter_encode_write_frame(frame, stream_index);
+                av_frame_free(&frame);
+                if (ret < 0)
+                    goto end;
+            } else {
+                av_frame_free(&frame);
+            }
+        } else {
+            /* remux this frame without reencoding */
+            packet.dts = av_rescale_q_rnd(packet.dts,
+                    ifmt_ctx->streams[stream_index]->time_base,
+                    ofmt_ctx->streams[stream_index]->time_base,
+                    AV_ROUND_NEAR_INF|AV_ROUND_PASS_MINMAX);
+            packet.pts = av_rescale_q_rnd(packet.pts,
+                    ifmt_ctx->streams[stream_index]->time_base,
+                    ofmt_ctx->streams[stream_index]->time_base,
+                    AV_ROUND_NEAR_INF|AV_ROUND_PASS_MINMAX);
+
+            ret = av_interleaved_write_frame(ofmt_ctx, &packet);
+            if (ret < 0)
+                goto end;
+        }
+        av_free_packet(&packet);
+    }
+
+    /* flush filters and encoders */
+    for (i = 0; i < ifmt_ctx->nb_streams; i++) {
+        /* flush filter */
+        if (!filter_ctx[i].filter_graph)
+            continue;
+        ret = filter_encode_write_frame(NULL, i);
+        if (ret < 0) {
+            av_log(NULL, AV_LOG_ERROR, "Flushing filter failed\n");
+            goto end;
+        }
+
+        /* flush encoder */
+        ret = flush_encoder(i);
+        if (ret < 0) {
+            av_log(NULL, AV_LOG_ERROR, "Flushing encoder failed\n");
+            goto end;
+        }
+    }
+
+    av_write_trailer(ofmt_ctx);
+end:
+    av_free_packet(&packet);
+    av_frame_free(&frame);
+    for (i = 0; i < ifmt_ctx->nb_streams; i++) {
+        avcodec_close(ifmt_ctx->streams[i]->codec);
+        if (ofmt_ctx && ofmt_ctx->nb_streams > i && ofmt_ctx->streams[i] && ofmt_ctx->streams[i]->codec)
+            avcodec_close(ofmt_ctx->streams[i]->codec);
+        if (filter_ctx && filter_ctx[i].filter_graph)
+            avfilter_graph_free(&filter_ctx[i].filter_graph);
+    }
+    av_free(filter_ctx);
+    avformat_close_input(&ifmt_ctx);
+    if (ofmt_ctx && !(ofmt_ctx->oformat->flags & AVFMT_NOFILE))
+        avio_close(ofmt_ctx->pb);
+    avformat_free_context(ofmt_ctx);
+
+    if (ret < 0)
+        av_log(NULL, AV_LOG_ERROR, "Error occurred: %s\n", av_err2str(ret));
+
+    return ret ? 1 : 0;
+}