libhb: fix or simplify several hacks involved with Libav support

For files that are demuxed by Libav, we must share the format context
with the decoder iso that it can obtain the codec context for each stream.
The code that did this was very convoluted and difficult to understand.
It is simplified by simply passing the context in hb_title_t.

Reader was closing stream files before the decoder was finished with the
context.  This created the need to delay the actual close and cache
the context.  Changed reader so it behaves more like the rest of handbrake's
work objects which lets us explicitly close after the decoders are finished.

Libav does some probing of the file when av_find_stream_info is called.
This probing leaves the format context in a bad state for some files and
causes subsequent reads or seeks to misbehave.  So open 2 contexts in
ffmpeg_open.  One is used only for probing, and the other only for reading.

decavcodec.c had 2 separate decoders for files demuxed by hb and files
demuxed by Libav.  They have been combined and simplified.

Previously, it was not possible to decode one source audio track multiple
times in order to fan it out to multiple output tracks if the file is
demuxed by Libav.  We were using the codec context from the format context.
Since there is only one of these for each stream, we could only do one
decode for each stream.  Use avcodec_copy_context to make copies of
the codec context and allow multiple decodes. This allows removal of
a lot of special case code for Libav streams that was necessary to
duplicate the output of the decoder.

Patch Libav's mkv demux to fix a seek problem.  This has been pushed 
upstreams, so the next time we update Libav, we must remove this patch.


git-svn-id: svn://svn.handbrake.fr/HandBrake/trunk@4141 b64f7644-9d1e-0410-96f1-a4d463321fa5

1 files changed, 102 insertions, 243 deletions

diff --git a/libhb/stream.c b/libhb/stream.c
index 1bf17d5f0..84d6c1a63 100644
--- a/libhb/stream.c
+++ b/libhb/stream.c
@@ -124,6 +124,7 @@ typedef struct {
 
 struct hb_stream_s
 {
+    int     scan;
     int     frames;             /* video frames so far */
     int     errors;             /* total errors so far */
     int     last_error_frame;   /* frame # at last error message */
@@ -164,7 +165,8 @@ struct hb_stream_s
     hb_stream_type_t hb_stream_type;
     hb_title_t *title;
 
-    AVFormatContext *ffmpeg_ic;
+    AVFormatContext *ffmpeg_info_ic;
+    AVFormatContext *ffmpeg_reader_ic;
     AVPacket *ffmpeg_pkt;
     uint8_t ffmpeg_video_id;
 
@@ -209,10 +211,10 @@ static void hb_ts_stream_set_audio_list(hb_list_t *list_audio, hb_stream_t *stre
 static void hb_ps_stream_find_audio_ids(hb_stream_t *stream, hb_title_t *title);
 static off_t align_to_next_packet(hb_stream_t *stream);
 
-static int ffmpeg_open( hb_stream_t *stream, hb_title_t *title );
+static int ffmpeg_open( hb_stream_t *stream, hb_title_t *title, int scan );
 static void ffmpeg_close( hb_stream_t *d );
-static hb_title_t *ffmpeg_title_scan( hb_stream_t *stream );
-static hb_buffer_t *ffmpeg_read( hb_stream_t *stream );
+static hb_title_t *ffmpeg_title_scan( hb_stream_t *stream, hb_title_t *title );
+hb_buffer_t *hb_ffmpeg_read( hb_stream_t *stream );
 static int ffmpeg_seek( hb_stream_t *stream, float frac );
 static int ffmpeg_seek_ts( hb_stream_t *stream, int64_t ts );
 
@@ -610,7 +612,7 @@ static int audio_inactive( hb_stream_t *stream, int idx )
  ***********************************************************************
  *
  **********************************************************************/
-hb_stream_t * hb_stream_open( char *path, hb_title_t *title )
+hb_stream_t * hb_stream_open( char *path, hb_title_t *title, int scan )
 {
     FILE *f = fopen( path, "rb" );
     if ( f == NULL )
@@ -634,7 +636,7 @@ hb_stream_t * hb_stream_open( char *path, hb_title_t *title )
      * (even if we have saved state, the stream may have changed).
      */
     hb_stream_t *ss = hb_stream_lookup( path );
-    if ( title && ss && ss->hb_stream_type != ffmpeg )
+    if ( !scan && ss && ss->hb_stream_type != ffmpeg )
     {
         /*
          * copy the saved state since we might be encoding the same stream
@@ -643,6 +645,7 @@ hb_stream_t * hb_stream_open( char *path, hb_title_t *title )
         memcpy( d, ss, sizeof(*d) );
         d->file_handle = f;
         d->title = title;
+        d->scan = scan;
         d->path = strdup( path );
 
         if ( d->hb_stream_type == transport )
@@ -680,6 +683,7 @@ hb_stream_t * hb_stream_open( char *path, hb_title_t *title )
     }
     d->file_handle = f;
     d->title = title;
+    d->scan = scan;
     d->path = strdup( path );
     if (d->path != NULL )
     {
@@ -689,7 +693,7 @@ hb_stream_t * hb_stream_open( char *path, hb_title_t *title )
         }
         fclose( d->file_handle );
         d->file_handle = NULL;
-        if ( ffmpeg_open( d, title ) )
+        if ( ffmpeg_open( d, title, scan ) )
         {
             return d;
         }
@@ -831,7 +835,7 @@ void hb_stream_close( hb_stream_t ** _d )
      * if the stream was opened for a scan, cache the result, otherwise delete
      * the state.
      */
-    if ( stream->title == NULL )
+    if ( stream->scan )
     {
         hb_stream_delete_dynamic( stream );
         if ( stream_state_list == NULL )
@@ -868,13 +872,13 @@ static void hb_stream_delete_entry(hb_stream_t *stream, int indx)
  ***********************************************************************
  *
  **********************************************************************/
-hb_title_t * hb_stream_title_scan(hb_stream_t *stream)
+hb_title_t * hb_stream_title_scan(hb_stream_t *stream, hb_title_t * title)
 {
     if ( stream->hb_stream_type == ffmpeg )
-        return ffmpeg_title_scan( stream );
+        return ffmpeg_title_scan( stream, title );
 
     // 'Barebones Title'
-    hb_title_t *aTitle = hb_title_init( stream->path, 0 );
+    hb_title_t *aTitle = title;
     aTitle->type = HB_STREAM_TYPE;
     aTitle->index = 1;
 
@@ -1388,7 +1392,7 @@ hb_buffer_t * hb_stream_read( hb_stream_t * src_stream )
 {
     if ( src_stream->hb_stream_type == ffmpeg )
     {
-        return ffmpeg_read( src_stream );
+        return hb_ffmpeg_read( src_stream );
     }
     if ( src_stream->hb_stream_type == dvd_program )
     {
@@ -1483,28 +1487,15 @@ hb_buffer_t * hb_stream_read( hb_stream_t * src_stream )
     return hb_ts_stream_decode( src_stream );
 }
 
-void ffmpeg_flush_stream_buffers( hb_stream_t *stream )
-{
-    int i;
-    AVFormatContext *ic = stream->ffmpeg_ic;
-
-    for ( i = 0; i < ic->nb_streams; i++ )
-    {
-        if ( ic->streams[i]->codec && ic->streams[i]->codec->codec )
-        {
-            avcodec_flush_buffers( ic->streams[i]->codec );
-        }
-    }
-}
-
 int64_t ffmpeg_initial_timestamp( hb_stream_t * stream )
 {
-    AVStream *s = stream->ffmpeg_ic->streams[stream->ffmpeg_video_id];
-    if ( s->nb_index_entries < 1 )
+    AVFormatContext *ic = stream->ffmpeg_info_ic;
+    if ( ic->start_time != AV_NOPTS_VALUE && ic->start_time > 0 )
+        return ic->start_time;
+    else
         return 0;
-
-    return s->index_entries[0].timestamp;
 }
+
 int hb_stream_seek_chapter( hb_stream_t * stream, int chapter_num )
 {
 
@@ -1537,8 +1528,11 @@ int hb_stream_seek_chapter( hb_stream_t * stream, int chapter_num )
 
     if ( chapter_num > 1 && pos > 0 )
     {
-        av_seek_frame( stream->ffmpeg_ic, -1, pos, 0);
-        ffmpeg_flush_stream_buffers( stream );
+        AVStream *st = stream->ffmpeg_info_ic->streams[stream->ffmpeg_video_id];
+        // timebase must be adjusted to match timebase of stream we are
+        // using for seeking.
+        pos = av_rescale(pos, st->time_base.den, AV_TIME_BASE * (int64_t)st->time_base.num);
+        avformat_seek_file( stream->ffmpeg_reader_ic, stream->ffmpeg_video_id, 0, pos, pos, AVSEEK_FLAG_BACKWARD);
     }
     return 1;
 }
@@ -1655,7 +1649,6 @@ static void set_ts_audio_description( hb_audio_t *audio, iso639_lang_t *lang )
      * formatting routine in scan.c do all the stuff below.
      */
     const char *codec_name;
-    AVCodecContext *cc;
 
     // Names for streams we know about.
     if ( audio->config.in.stream_type == 0x80 && 
@@ -1705,24 +1698,6 @@ static void set_ts_audio_description( hb_audio_t *audio, iso639_lang_t *lang )
         // To distinguish, Bluray streams have a reg_desc of HDMV
         codec_name = "E-AC3";
     }
-    // For streams demuxed and decoded by ffmpeg, we have a cached context.
-    // Use it to get the name and profile information.  Obtaining
-    // the profile requires that ffmpeg has already probed the stream.
-    else if ( ( audio->config.in.codec & HB_ACODEC_FF_MASK ) &&
-              ( audio->config.in.codec & HB_ACODEC_FF_I_FLAG ) &&
-         ( cc = hb_ffmpeg_context( audio->config.in.codec_param ) ) &&
-         avcodec_find_decoder( cc->codec_id ) )
-    {
-        AVCodec *codec = avcodec_find_decoder( cc->codec_id );
-        codec_name = codec->name;
-
-        const char *profile_name;
-        profile_name = av_get_profile_name( codec, cc->profile );
-        if ( profile_name )
-        {
-            codec_name = profile_name;
-        }
-    }
     else if ( st2codec[audio->config.in.stream_type].kind == A )
     {
         codec_name = stream_type_name(audio->config.in.reg_desc,
@@ -1731,7 +1706,6 @@ static void set_ts_audio_description( hb_audio_t *audio, iso639_lang_t *lang )
     // For streams demuxed by us and decoded by ffmpeg, we can lookup the
     // decoder name.
     else if ( ( audio->config.in.codec & HB_ACODEC_FF_MASK ) &&
-              !( audio->config.in.codec & HB_ACODEC_FF_I_FLAG ) &&
               avcodec_find_decoder( audio->config.in.codec_param ) )
     {
         codec_name = avcodec_find_decoder( audio->config.in.codec_param )->name;
@@ -1750,25 +1724,13 @@ static void set_ts_audio_description( hb_audio_t *audio, iso639_lang_t *lang )
               strlen(lang->native_name) ? lang->native_name : lang->eng_name,
               codec_name );
 
-    if ( ( audio->config.in.codec & HB_ACODEC_FF_MASK) &&
-         ( audio->config.in.codec & HB_ACODEC_FF_I_FLAG ) )
-    {
-        int layout = audio->config.in.channel_layout;
-        char *desc = audio->config.lang.description +
-                        strlen( audio->config.lang.description );
-        sprintf( desc, " (%d.%d ch)",
-                 HB_INPUT_CH_LAYOUT_GET_DISCRETE_FRONT_COUNT(layout) +
-                     HB_INPUT_CH_LAYOUT_GET_DISCRETE_REAR_COUNT(layout),
-                 HB_INPUT_CH_LAYOUT_GET_DISCRETE_LFE_COUNT(layout) );
-    }
-
     snprintf( audio->config.lang.simple, sizeof( audio->config.lang.simple ), "%s",
               strlen(lang->native_name) ? lang->native_name : lang->eng_name );
     snprintf( audio->config.lang.iso639_2, sizeof( audio->config.lang.iso639_2 ),
               "%s", lang->iso639_2);
 }
 
-static void set_audio_description( hb_audio_t *audio, iso639_lang_t *lang )
+static void set_audio_description( hb_stream_t * stream, hb_audio_t *audio, iso639_lang_t *lang )
 {
     /* XXX
      * This is a duplicate of code in dvd.c - it should get factored out
@@ -1777,14 +1739,17 @@ static void set_audio_description( hb_audio_t *audio, iso639_lang_t *lang )
      * formatting routine in scan.c do all the stuff below.
      */
     const char *codec_name;
-    AVCodecContext *cc;
+    AVCodecContext *cc = NULL;
+
+    if ( stream && stream->ffmpeg_info_ic )
+    {
+        cc = stream->ffmpeg_info_ic->streams[audio->id]->codec;
+    }
 
     // For streams demuxed and decoded by ffmpeg, we have a cached context.
     // Use it to get the name and profile information.  Obtaining
     // the profile requires that ffmpeg has already probed the stream.
-    if ( ( audio->config.in.codec & HB_ACODEC_FF_MASK ) &&
-         ( audio->config.in.codec & HB_ACODEC_FF_I_FLAG ) &&
-         ( cc = hb_ffmpeg_context( audio->config.in.codec_param ) ) &&
+    if ( ( audio->config.in.codec & HB_ACODEC_FF_MASK ) && cc &&
          avcodec_find_decoder( cc->codec_id ) )
     {
         AVCodec *codec = avcodec_find_decoder( cc->codec_id );
@@ -1800,7 +1765,6 @@ static void set_audio_description( hb_audio_t *audio, iso639_lang_t *lang )
     // For streams demuxed by us and decoded by ffmpeg, we can lookup the
     // decoder name.
     else if ( ( audio->config.in.codec & HB_ACODEC_FF_MASK ) &&
-              !( audio->config.in.codec & HB_ACODEC_FF_I_FLAG ) &&
               avcodec_find_decoder( audio->config.in.codec_param ) )
     {
         codec_name = avcodec_find_decoder( audio->config.in.codec_param )->name;
@@ -1818,8 +1782,7 @@ static void set_audio_description( hb_audio_t *audio, iso639_lang_t *lang )
               strlen(lang->native_name) ? lang->native_name : lang->eng_name,
               codec_name );
 
-    if ( ( audio->config.in.codec & HB_ACODEC_FF_MASK) &&
-         ( audio->config.in.codec & HB_ACODEC_FF_I_FLAG ) )
+    if ( audio->config.in.channel_layout )
     {
         int layout = audio->config.in.channel_layout;
         char *desc = audio->config.lang.description +
@@ -2042,7 +2005,7 @@ static void add_audio_to_title(hb_title_t *title, int id)
             return;
 
     }
-    set_audio_description( audio, lang_for_code( 0 ) );
+    set_audio_description( NULL, audio, lang_for_code( 0 ) );
 
     // Sort by id when adding to the list
     int i;
@@ -3124,182 +3087,78 @@ static void hb_ts_stream_reset(hb_stream_t *stream)
 // ------------------------------------------------------------------
 // Support for reading media files via the ffmpeg libraries.
 
-static void ffmpeg_add_codec( hb_stream_t *stream, int stream_index )
+static int ffmpeg_open( hb_stream_t *stream, hb_title_t *title, int scan )
 {
-    // add a codec to the context here so it will be there when we
-    // read the first packet.
-    AVCodecContext *context = stream->ffmpeg_ic->streams[stream_index]->codec;
-    context->workaround_bugs = FF_BUG_AUTODETECT;
-    context->error_recognition = 1;
-    context->error_concealment = FF_EC_GUESS_MVS|FF_EC_DEBLOCK;
-    AVCodec *codec = avcodec_find_decoder( context->codec_id );
-    hb_ff_set_sample_fmt( context, codec );
-}
-
-// The ffmpeg stream reader / parser shares a lot of state with the 
-// decoder via a codec context kept in the AVStream of the reader's
-// AVFormatContext. Since decoding is done in a different thread we
-// have to somehow pass this codec context to the decoder and we have
-// to do it before the first packet is read (so we can't put the info
-// in the buf we'll send downstream). Decoders don't have any way to
-// get to the stream directly (they're not passed the title or job
-// pointers during a scan) so this is a back door for the decoder to
-// get the codec context. We just stick the stream pointer in the next
-// slot an array of pointers maintained as a circular list then return
-// the index into the list combined with the ffmpeg stream index as the
-// codec_param that will be passed to the decoder init routine. We make
-// the list 'big' (enough for 1024 simultaneously open ffmpeg streams)
-// so that we don't have to do a complicated allocator or worry about
-// deleting entries on close. 
-//
-// Entries can only be added to this list during a scan and are never
-// deleted so the list access doesn't require locking.
-static hb_stream_t **ffmpeg_streams;    // circular list of stream pointers
-static int ffmpeg_stream_cur;           // where we put the last stream pointer
-#define ffmpeg_sl_bits (10)             // log2 stream list size (in entries)
-#define ffmpeg_sl_size (1 << ffmpeg_sl_bits)
-
-// add a stream to the list & return the appropriate codec_param to access it
-static int ffmpeg_codec_param( hb_stream_t *stream, int stream_index )
-{
-    if ( !ffmpeg_streams )
-    {
-        ffmpeg_streams = calloc( ffmpeg_sl_size, sizeof(stream) );
-    }
-
-    // the title scan adds all the ffmpeg media streams at once so we
-    // only add a new entry to our stream list if the stream is different
-    // than last time.
-    int slot = ffmpeg_stream_cur;
-    if ( ffmpeg_streams[slot] != stream )
-    {
-        // new stream - put it in the next slot of the stream list
-        slot = ++ffmpeg_stream_cur & (ffmpeg_sl_size - 1);
-        ffmpeg_streams[slot] = stream;
-    }
-
-    ffmpeg_add_codec( stream, stream_index );
-
-    return ( stream_index << ffmpeg_sl_bits ) | slot;
-}
-
-// we're about to open 'title' to convert it - remap the stream associated
-// with the video & audio codec params of the title to refer to 'stream'
-// (the original scan stream was closed and no longer exists).
-static void ffmpeg_remap_stream( hb_stream_t *stream, hb_title_t *title )
-{
-    // all the video & audio came from the same stream so remapping
-    // the video's stream slot takes care of everything.
-    int slot = title->video_codec_param & (ffmpeg_sl_size - 1);
-    ffmpeg_streams[slot] = stream;
-
-    // add codecs for all the streams used by the title
-    ffmpeg_add_codec( stream, title->video_codec_param >> ffmpeg_sl_bits );
-
-    int i;
-    hb_audio_t *audio;
-    for ( i = 0; ( audio = hb_list_item( title->list_audio, i ) ); ++i )
-    {
-        if ( audio->config.in.codec & HB_ACODEC_FF_MASK )
-        {
-            ffmpeg_add_codec( stream,
-                              audio->config.in.codec_param >> ffmpeg_sl_bits );
-        }
-    }
-}
-
-void *hb_ffmpeg_context( int codec_param )
-{
-    if ( ffmpeg_streams == NULL )
-        return NULL;
-
-    int slot = codec_param & (ffmpeg_sl_size - 1);
-    int stream_index = codec_param >> ffmpeg_sl_bits;
-    return ffmpeg_streams[slot]->ffmpeg_ic->streams[stream_index]->codec;
-}
-
-void *hb_ffmpeg_avstream( int codec_param )
-{
-    if ( ffmpeg_streams == NULL )
-        return NULL;
-
-    int slot = codec_param & (ffmpeg_sl_size - 1);
-    int stream_index = codec_param >> ffmpeg_sl_bits;
-    return ffmpeg_streams[slot]->ffmpeg_ic->streams[stream_index];
-}
-
-static AVFormatContext *ffmpeg_deferred_close;
-
-static int ffmpeg_open( hb_stream_t *stream, hb_title_t *title )
-{
-    if ( ffmpeg_deferred_close )
-    {
-        av_close_input_file( ffmpeg_deferred_close );
-        ffmpeg_deferred_close = NULL;
-    }
-    AVFormatContext *ic;
+    AVFormatContext *info_ic = NULL, *reader_ic = NULL;
 
     av_log_set_level( AV_LOG_ERROR );
-    if ( av_open_input_file( &ic, stream->path, NULL, 0, NULL ) < 0 )
+
+    // FFMpeg has issues with seeking.  After av_find_stream_info, the
+    // streams are left in an indeterminate position.  So a seek is
+    // necessary to force things back to the beginning of the stream.
+    // But then the seek fails for some stream types.  So the safest thing
+    // to do seems to be to open 2 AVFormatContext.  One for probing info
+    // and the other for reading.
+    if ( av_open_input_file( &info_ic, stream->path, NULL, 0, NULL ) < 0 )
     {
         return 0;
     }
-    if ( av_find_stream_info( ic ) < 0 )
+    if ( av_find_stream_info( info_ic ) < 0 )
         goto fail;
 
-    stream->ffmpeg_ic = ic;
+    title->opaque_priv = (void*)info_ic;
+    stream->ffmpeg_info_ic = info_ic;
+    if ( av_open_input_file( &reader_ic, stream->path, NULL, 0, NULL ) < 0 )
+    {
+        goto fail;
+    }
+    stream->ffmpeg_reader_ic = reader_ic;
     stream->hb_stream_type = ffmpeg;
     stream->ffmpeg_pkt = malloc(sizeof(*stream->ffmpeg_pkt));
     av_init_packet( stream->ffmpeg_pkt );
     stream->chapter_end = INT64_MAX;
 
-    if ( title )
+    if ( !scan )
     {
         // we're opening for read. scan passed out codec params that
         // indexed its stream so we need to remap them so they point
         // to this stream.
         stream->ffmpeg_video_id = title->video_id;
-        ffmpeg_remap_stream( stream, title );
         av_log_set_level( AV_LOG_ERROR );
     }
     else
     {
         // we're opening for scan. let ffmpeg put some info into the
         // log about what we've got.
+        stream->ffmpeg_video_id = title->video_id;
         av_log_set_level( AV_LOG_INFO );
-        av_dump_format( ic, 0, stream->path, 0 );
+        av_dump_format( info_ic, 0, stream->path, 0 );
         av_log_set_level( AV_LOG_ERROR );
 
         // accept this file if it has at least one video stream we can decode
         int i;
-        for (i = 0; i < ic->nb_streams; ++i )
+        for (i = 0; i < info_ic->nb_streams; ++i )
         {
-            if ( ic->streams[i]->codec->codec_type == AVMEDIA_TYPE_VIDEO )
+            if ( info_ic->streams[i]->codec->codec_type == AVMEDIA_TYPE_VIDEO )
             {
                 break;
             }
         }
-        if ( i >= ic->nb_streams )
+        if ( i >= info_ic->nb_streams )
             goto fail;
     }
     return 1;
 
   fail:
-    av_close_input_file( ic );
+    if ( info_ic ) av_close_input_file( info_ic );
+    if ( reader_ic ) av_close_input_file( reader_ic );
     return 0;
 }
 
 static void ffmpeg_close( hb_stream_t *d )
 {
-    // XXX since we're sharing the CodecContext with the downstream
-    // decoder proc we can't close the stream. We need to reference count
-    // this so we can close it when both are done with their instance but
-    // for now just defer the close until the next stream open or close.
-    if ( ffmpeg_deferred_close )
-    {
-        av_close_input_file( ffmpeg_deferred_close );
-    }
-    ffmpeg_deferred_close = d->ffmpeg_ic;
+    av_close_input_file( d->ffmpeg_info_ic );
+    av_close_input_file( d->ffmpeg_reader_ic );
     if ( d->ffmpeg_pkt != NULL )
     {
         free( d->ffmpeg_pkt );
@@ -3309,7 +3168,7 @@ static void ffmpeg_close( hb_stream_t *d )
 
 static void add_ffmpeg_audio( hb_title_t *title, hb_stream_t *stream, int id )
 {
-    AVStream *st = stream->ffmpeg_ic->streams[id];
+    AVStream *st = stream->ffmpeg_info_ic->streams[id];
     AVCodecContext *codec = st->codec;
     AVMetadataTag *tag;
     int layout;
@@ -3345,13 +3204,13 @@ static void add_ffmpeg_audio( hb_title_t *title, hb_stream_t *stream, int id )
                ( codec->profile == FF_PROFILE_DTS_HD_MA ||
                  codec->profile == FF_PROFILE_DTS_HD_HRA ) )
             {
-                audio->config.in.codec = HB_ACODEC_DCA_HD | HB_ACODEC_FF_I_FLAG;
+                audio->config.in.codec = HB_ACODEC_DCA_HD;
             }
             else
             {
-                audio->config.in.codec = HB_ACODEC_FFMPEG | HB_ACODEC_FF_I_FLAG;
+                audio->config.in.codec = HB_ACODEC_FFMPEG;
             }
-            audio->config.in.codec_param = ffmpeg_codec_param( stream, id );
+            audio->config.in.codec_param = codec->codec_id;
 
             audio->config.in.bitrate = codec->bit_rate? codec->bit_rate : 1;
             audio->config.in.samplerate = codec->sample_rate;
@@ -3360,7 +3219,7 @@ static void add_ffmpeg_audio( hb_title_t *title, hb_stream_t *stream, int id )
         }
 
         tag = av_metadata_get( st->metadata, "language", NULL, 0 );
-        set_audio_description( audio, 
+        set_audio_description( stream, audio, 
             lang_for_code2( tag ? tag->value : "und" ) );
 
         hb_list_add( title->list_audio, audio );
@@ -3491,7 +3350,7 @@ static int ffmpeg_parse_vobsub_extradata( AVCodecContext *codec, hb_subtitle_t *
 
 static void add_ffmpeg_subtitle( hb_title_t *title, hb_stream_t *stream, int id )
 {
-    AVStream *st = stream->ffmpeg_ic->streams[id];
+    AVStream *st = stream->ffmpeg_info_ic->streams[id];
     AVCodecContext *codec = st->codec;
     
     hb_subtitle_t *subtitle = calloc( 1, sizeof(*subtitle) );
@@ -3560,7 +3419,7 @@ static char *get_ffmpeg_metadata_value( AVMetadata *m, char *key )
 
 static void add_ffmpeg_attachment( hb_title_t *title, hb_stream_t *stream, int id )
 {
-    AVStream *st = stream->ffmpeg_ic->streams[id];
+    AVStream *st = stream->ffmpeg_info_ic->streams[id];
     AVCodecContext *codec = st->codec;
     
     enum attachtype type;
@@ -3586,12 +3445,11 @@ static void add_ffmpeg_attachment( hb_title_t *title, hb_stream_t *stream, int i
     hb_list_add(title->list_attachment, attachment);
 }
 
-static hb_title_t *ffmpeg_title_scan( hb_stream_t *stream )
+static hb_title_t *ffmpeg_title_scan( hb_stream_t *stream, hb_title_t *title )
 {
-    AVFormatContext *ic = stream->ffmpeg_ic;
+    AVFormatContext *ic = stream->ffmpeg_info_ic;
 
     // 'Barebones Title'
-    hb_title_t *title = hb_title_init( stream->path, 0 );
     title->type = HB_FF_STREAM_TYPE;
     title->index = 1;
 
@@ -3629,16 +3487,18 @@ static hb_title_t *ffmpeg_title_scan( hb_stream_t *stream )
             }
             title->video_id = i;
             stream->ffmpeg_video_id = i;
+            if ( ic->streams[i]->sample_aspect_ratio.num &&
+                 ic->streams[i]->sample_aspect_ratio.den )
+            {
+                title->pixel_aspect_width = ic->streams[i]->sample_aspect_ratio.num;
+                title->pixel_aspect_height = ic->streams[i]->sample_aspect_ratio.den;
+            }
 
             if ( context->codec_id == CODEC_ID_H264 )
                 title->flags |= HBTF_NO_IDR;
 
-            // We have to use the 'internal' avcodec decoder because
-            // it needs to share the codec context from this video
-            // stream. The parser internal to av_read_frame
-            // passes a bunch of state info to the decoder via the context.
-            title->video_codec = WORK_DECAVCODECVI;
-            title->video_codec_param = ffmpeg_codec_param( stream, i );
+            title->video_codec = WORK_DECAVCODECV;
+            title->video_codec_param = context->codec_id;
         }
         else if ( ic->streams[i]->codec->codec_type == AVMEDIA_TYPE_AUDIO &&
                   avcodec_find_decoder( ic->streams[i]->codec->codec_id ) )
@@ -3718,7 +3578,7 @@ static int ffmpeg_is_keyframe( hb_stream_t *stream )
 {
     uint8_t *pkt;
 
-    switch ( stream->ffmpeg_ic->streams[stream->ffmpeg_video_id]->codec->codec_id )
+    switch ( stream->ffmpeg_info_ic->streams[stream->ffmpeg_video_id]->codec->codec_id )
     {
         case CODEC_ID_VC1:
             // XXX the VC1 codec doesn't mark key frames so to get previews
@@ -3740,7 +3600,7 @@ static int ffmpeg_is_keyframe( hb_stream_t *stream )
             // there are bframes or not so we have to look at the sequence
             // header to get that.
             pkt = stream->ffmpeg_pkt->data;
-            uint8_t *seqhdr = stream->ffmpeg_ic->streams[stream->ffmpeg_video_id]->codec->extradata;
+            uint8_t *seqhdr = stream->ffmpeg_info_ic->streams[stream->ffmpeg_video_id]->codec->extradata;
             int pshift = 2;
             if ( ( seqhdr[3] & 0x02 ) == 0 )
                 // no FINTERPFLAG
@@ -3760,13 +3620,13 @@ static int ffmpeg_is_keyframe( hb_stream_t *stream )
     return ( stream->ffmpeg_pkt->flags & AV_PKT_FLAG_KEY );
 }
 
-static hb_buffer_t * ffmpeg_read( hb_stream_t *stream )
+hb_buffer_t * hb_ffmpeg_read( hb_stream_t *stream )
 {
     int err;
     hb_buffer_t * buf;
 
   again:
-    if ( ( err = av_read_frame( stream->ffmpeg_ic, stream->ffmpeg_pkt )) < 0 )
+    if ( ( err = av_read_frame( stream->ffmpeg_reader_ic, stream->ffmpeg_pkt )) < 0 )
     {
         // av_read_frame can return EAGAIN.  In this case, it expects
         // to be called again to get more data.
@@ -3818,7 +3678,7 @@ static hb_buffer_t * ffmpeg_read( hb_stream_t *stream )
         {
             hb_log( "ffmpeg_read: pkt too big: %d bytes", stream->ffmpeg_pkt->size );
             av_free_packet( stream->ffmpeg_pkt );
-            return ffmpeg_read( stream );
+            return hb_ffmpeg_read( stream );
         }
         buf = hb_buffer_init( stream->ffmpeg_pkt->size );
         memcpy( buf->data, stream->ffmpeg_pkt->data, stream->ffmpeg_pkt->size );
@@ -3827,7 +3687,7 @@ static hb_buffer_t * ffmpeg_read( hb_stream_t *stream )
 
     // compute a conversion factor to go from the ffmpeg
     // timebase for the stream to HB's 90kHz timebase.
-    AVStream *s = stream->ffmpeg_ic->streams[stream->ffmpeg_pkt->stream_index];
+    AVStream *s = stream->ffmpeg_info_ic->streams[stream->ffmpeg_pkt->stream_index];
     double tsconv = 90000. * (double)s->time_base.num / (double)s->time_base.den;
 
     buf->start = av_to_hb_pts( stream->ffmpeg_pkt->pts, tsconv );
@@ -3859,8 +3719,8 @@ static hb_buffer_t * ffmpeg_read( hb_stream_t *stream )
      */
     enum CodecID ffmpeg_pkt_codec;
     enum AVMediaType codec_type;
-    ffmpeg_pkt_codec = stream->ffmpeg_ic->streams[stream->ffmpeg_pkt->stream_index]->codec->codec_id;
-    codec_type = stream->ffmpeg_ic->streams[stream->ffmpeg_pkt->stream_index]->codec->codec_type;
+    ffmpeg_pkt_codec = stream->ffmpeg_info_ic->streams[stream->ffmpeg_pkt->stream_index]->codec->codec_id;
+    codec_type = stream->ffmpeg_info_ic->streams[stream->ffmpeg_pkt->stream_index]->codec->codec_type;
     switch ( codec_type )
     {
         case AVMEDIA_TYPE_VIDEO:
@@ -3923,38 +3783,37 @@ static hb_buffer_t * ffmpeg_read( hb_stream_t *stream )
 
 static int ffmpeg_seek( hb_stream_t *stream, float frac )
 {
-    AVFormatContext *ic = stream->ffmpeg_ic;
+    AVFormatContext *ic = stream->ffmpeg_reader_ic;
     if ( frac > 0. )
     {
-        int64_t pos = (double)ic->duration * (double)frac;
-        if ( ic->start_time != AV_NOPTS_VALUE && ic->start_time > 0 )
-        {
-            pos += ic->start_time;
-        }
-        av_seek_frame( ic, -1, pos, 0 );
-        stream->need_keyframe = 1;
-        ffmpeg_flush_stream_buffers( stream );
+        int64_t pos = (double)stream->ffmpeg_info_ic->duration * (double)frac +
+                ffmpeg_initial_timestamp( stream );
+        avformat_seek_file( ic, -1, 0, pos, pos, AVSEEK_FLAG_BACKWARD);
     }
     else
     {
-        av_seek_frame( ic, -1, 0LL, AVSEEK_FLAG_BACKWARD );
-        ffmpeg_flush_stream_buffers( stream );
+        int64_t pos = ffmpeg_initial_timestamp( stream );
+        avformat_seek_file( ic, -1, 0, pos, pos, AVSEEK_FLAG_BACKWARD);
     }
+    stream->need_keyframe = 1;
     return 1;
 }
 
 // Assumes that we are always seeking forward
 static int ffmpeg_seek_ts( hb_stream_t *stream, int64_t ts )
 {
-    AVFormatContext *ic = stream->ffmpeg_ic;
+    AVFormatContext *ic = stream->ffmpeg_reader_ic;
     int64_t pos;
     int ret;
 
     pos = ts * AV_TIME_BASE / 90000 + ffmpeg_initial_timestamp( stream );
+    AVStream *st = stream->ffmpeg_info_ic->streams[stream->ffmpeg_video_id];
+    // timebase must be adjusted to match timebase of stream we are
+    // using for seeking.
+    pos = av_rescale(pos, st->time_base.den, AV_TIME_BASE * (int64_t)st->time_base.num);
     stream->need_keyframe = 1;
     // Seek to the nearest timestamp before that requested where
     // there is an I-frame
-    ret = av_seek_frame( ic, -1, pos, AVSEEK_FLAG_BACKWARD );
-    ffmpeg_flush_stream_buffers( stream );
+    ret = avformat_seek_file( ic, stream->ffmpeg_video_id, 0, pos, pos, 0);
     return ret;
 }