enhance miracle LOAD command to accept a service: prefix.

[melted] / src / modules / dv / producer_libdv.c

diff --git a/src/modules/dv/producer_libdv.c b/src/modules/dv/producer_libdv.c
index 3d4f243..30111da 100644 (file)
--- a/src/modules/dv/producer_libdv.c
+++ b/src/modules/dv/producer_libdv.c
@@ -154,7 +154,7 @@ static int producer_collect_info( producer_libdv this )
                        // Parse the header for meta info
                        dv_parse_header( this->dv_decoder, dv_data );
                        mlt_properties_set_double( properties, "aspect_ratio", 
-                               dv_format_wide( this->dv_decoder ) ? ( this->is_pal ? 512/351 : 96/79 ) : ( this->is_pal ? 128/117 : 72/79 ) );
+                               dv_format_wide( this->dv_decoder ) ? ( this->is_pal ? 512.0/351.0 : 96.0/79.0 ) : ( this->is_pal ? 128.0/117.0 : 72.0/79.0 ) );
                }
 
                mlt_pool_release( dv_data );
@@ -239,33 +239,43 @@ static int producer_get_audio( mlt_frame this, int16_t **buffer, mlt_audio_forma
        // Parse the header for meta info
        dv_parse_header( decoder, dv_data );
 
-       // Obtain required values
-       *frequency = decoder->audio->frequency;
-       *samples = decoder->audio->samples_this_frame;
-       *channels = decoder->audio->num_channels;
-
-       // Create a temporary workspace
-       for ( i = 0; i < 4; i++ )
-               audio_channels[ i ] = mlt_pool_alloc( DV_AUDIO_MAX_SAMPLES * sizeof( int16_t ) );
-
-       // Create a workspace for the result
-       *buffer = mlt_pool_alloc( *channels * DV_AUDIO_MAX_SAMPLES * sizeof( int16_t ) );
-
-       // Pass the allocated audio buffer as a property
-       mlt_properties_set_data( properties, "audio", *buffer, *channels * DV_AUDIO_MAX_SAMPLES * sizeof( int16_t ), ( mlt_destructor )mlt_pool_release, NULL );
-
-       // Decode the audio
-       dv_decode_full_audio( decoder, dv_data, audio_channels );
+       // Check that we have audio
+       if ( decoder->audio->num_channels > 0 )
+       {
+               // Obtain required values
+               *frequency = decoder->audio->frequency;
+               *samples = decoder->audio->samples_this_frame;
+               *channels = decoder->audio->num_channels;
+
+               // Create a temporary workspace
+               for ( i = 0; i < 4; i++ )
+                       audio_channels[ i ] = mlt_pool_alloc( DV_AUDIO_MAX_SAMPLES * sizeof( int16_t ) );
+       
+               // Create a workspace for the result
+               *buffer = mlt_pool_alloc( *channels * DV_AUDIO_MAX_SAMPLES * sizeof( int16_t ) );
+       
+               // Pass the allocated audio buffer as a property
+               mlt_properties_set_data( properties, "audio", *buffer, *channels * DV_AUDIO_MAX_SAMPLES * sizeof( int16_t ), ( mlt_destructor )mlt_pool_release, NULL );
        
-       // Interleave the audio
-       p = *buffer;
-       for ( i = 0; i < *samples; i++ )
-               for ( j = 0; j < *channels; j++ )
-                       *p++ = audio_channels[ j ][ i ];
-
-       // Free the temporary work space
-       for ( i = 0; i < 4; i++ )
-               mlt_pool_release( audio_channels[ i ] );
+               // Decode the audio
+               dv_decode_full_audio( decoder, dv_data, audio_channels );
+               
+               // Interleave the audio
+               p = *buffer;
+               for ( i = 0; i < *samples; i++ )
+                       for ( j = 0; j < *channels; j++ )
+                               *p++ = audio_channels[ j ][ i ];
+       
+               // Free the temporary work space
+               for ( i = 0; i < 4; i++ )
+                       mlt_pool_release( audio_channels[ i ] );
+       }
+       else
+       {
+               // No audio available on the frame, so get test audio (silence)
+               this->get_audio = NULL;
+               mlt_frame_get_audio( this, buffer, format, frequency, channels, samples );
+       }
 
        return 0;
 }