* filters.c : audio output filters management
*****************************************************************************
* Copyright (C) 2002 VideoLAN
- * $Id: filters.c,v 1.3 2002/08/12 09:34:15 sam Exp $
+ * $Id: filters.c,v 1.9 2002/09/02 23:17:06 massiot Exp $
*
* Authors: Christophe Massiot <massiot@via.ecp.fr>
*
#include "audio_output.h"
#include "aout_internal.h"
+/*****************************************************************************
+ * FindFilter: find an audio filter for a specific transformation
+ *****************************************************************************/
+static aout_filter_t * FindFilter( aout_instance_t * p_aout,
+ const audio_sample_format_t * p_input_format,
+ const audio_sample_format_t * p_output_format )
+{
+ aout_filter_t * p_filter = vlc_object_create( p_aout,
+ sizeof(aout_filter_t) );
+
+ if ( p_filter == NULL ) return NULL;
+ vlc_object_attach( p_filter, p_aout );
+
+ memcpy( &p_filter->input, p_input_format, sizeof(audio_sample_format_t) );
+ memcpy( &p_filter->output, p_output_format,
+ sizeof(audio_sample_format_t) );
+ p_filter->p_module = module_Need( p_filter, "audio filter", NULL );
+ if ( p_filter->p_module == NULL )
+ {
+ vlc_object_detach( p_filter );
+ vlc_object_destroy( p_filter );
+ return NULL;
+ }
+
+ return p_filter;
+}
+
+/*****************************************************************************
+ * SplitConversion: split a conversion in two parts
+ *****************************************************************************
+ * Returns the number of conversions required by the first part - 0 if only
+ * one conversion was asked.
+ * Beware : p_output_format can be modified during this function if the
+ * developer passed SplitConversion( toto, titi, titi, ... ). That is legal.
+ * SplitConversion( toto, titi, toto, ... ) isn't.
+ *****************************************************************************/
+static int SplitConversion( aout_instance_t * p_aout,
+ const audio_sample_format_t * p_input_format,
+ const audio_sample_format_t * p_output_format,
+ audio_sample_format_t * p_middle_format )
+{
+ vlc_bool_t b_format =
+ (p_input_format->i_format != p_output_format->i_format);
+ vlc_bool_t b_rate = (p_input_format->i_rate != p_output_format->i_rate);
+ vlc_bool_t b_channels =
+ (p_input_format->i_channels != p_output_format->i_channels);
+ int i_nb_conversions = b_format + b_rate + b_channels;
+
+ if ( i_nb_conversions <= 1 ) return 0;
+
+ memcpy( p_middle_format, p_output_format, sizeof(audio_sample_format_t) );
+
+ if ( i_nb_conversions == 2 )
+ {
+ if ( !b_format )
+ {
+ p_middle_format->i_rate = p_input_format->i_rate;
+ return 1;
+ }
+
+ if ( !b_rate )
+ {
+ p_middle_format->i_channels = p_input_format->i_channels;
+ return 1;
+ }
+
+ /* !b_channels */
+ p_middle_format->i_rate = p_input_format->i_rate;
+ return 1;
+ }
+
+ /* i_nb_conversion == 3 */
+ p_middle_format->i_channels = p_input_format->i_channels;
+ return 2;
+}
+
/*****************************************************************************
* aout_FiltersCreatePipeline: create a filters pipeline to transform a sample
* format to another
int aout_FiltersCreatePipeline( aout_instance_t * p_aout,
aout_filter_t ** pp_filters,
int * pi_nb_filters,
- audio_sample_format_t * p_input_format,
- audio_sample_format_t * p_output_format )
+ const audio_sample_format_t * p_input_format,
+ const audio_sample_format_t * p_output_format )
{
+ audio_sample_format_t temp_format;
+ int i_nb_conversions;
+
if ( AOUT_FMTS_IDENTICAL( p_input_format, p_output_format ) )
{
msg_Dbg( p_aout, "no need for any filter" );
return 0;
}
- pp_filters[0] = vlc_object_create( p_aout, sizeof(aout_filter_t) );
- if ( pp_filters[0] == NULL )
- {
- return -1;
- }
- vlc_object_attach( pp_filters[0], p_aout );
+ msg_Dbg( p_aout, "filter(s) format=%d->%d rate=%d->%d channels=%d->%d",
+ p_input_format->i_format, p_output_format->i_format,
+ p_input_format->i_rate, p_output_format->i_rate,
+ p_input_format->i_channels, p_output_format->i_channels );
/* Try to find a filter to do the whole conversion. */
- memcpy( &pp_filters[0]->input, p_input_format,
- sizeof(audio_sample_format_t) );
- memcpy( &pp_filters[0]->output, p_output_format,
- sizeof(audio_sample_format_t) );
- pp_filters[0]->p_module = module_Need( pp_filters[0], "audio filter",
- NULL );
- if ( pp_filters[0]->p_module != NULL )
+ pp_filters[0] = FindFilter( p_aout, p_input_format, p_output_format );
+ if ( pp_filters[0] != NULL )
{
msg_Dbg( p_aout, "found a filter for the whole conversion" );
*pi_nb_filters = 1;
return 0;
}
- /* Split the conversion : format | rate, or rate | format. */
- pp_filters[0]->output.i_format = pp_filters[0]->input.i_format;
- pp_filters[0]->p_module = module_Need( pp_filters[0], "audio filter",
- NULL );
- if ( pp_filters[0]->p_module == NULL )
+ /* We'll have to split the conversion. We always do the downmixing
+ * before the resampling, because the audio decoder can probably do it
+ * for us. */
+ i_nb_conversions = SplitConversion( p_aout, p_input_format,
+ p_output_format, &temp_format );
+ if ( !i_nb_conversions )
{
- /* Then, start with the format conversion. */
- memcpy( &pp_filters[0]->output, p_output_format,
- sizeof(audio_sample_format_t) );
- pp_filters[0]->output.i_rate = pp_filters[0]->input.i_rate;
- pp_filters[0]->p_module = module_Need( pp_filters[0], "audio filter",
- NULL );
- if ( pp_filters[0]->p_module == NULL )
- {
- msg_Err( p_aout, "couldn't find a filter for any conversion" );
- vlc_object_detach( pp_filters[0] );
- vlc_object_destroy( pp_filters[0] );
- return -1;
- }
+ /* There was only one conversion to do, and we already failed. */
+ msg_Err( p_aout, "couldn't find a filter for the conversion" );
+ return -1;
}
- /* Find a filter for the rest. */
- pp_filters[1] = vlc_object_create( p_aout, sizeof(aout_filter_t) );
- if ( pp_filters[1] == NULL )
+ pp_filters[0] = FindFilter( p_aout, p_input_format, &temp_format );
+ if ( pp_filters[0] == NULL && i_nb_conversions == 2 )
{
- vlc_object_detach( pp_filters[0] );
- vlc_object_destroy( pp_filters[0] );
- return -1;
+ /* Try with only one conversion. */
+ SplitConversion( p_aout, p_input_format, &temp_format,
+ &temp_format );
+ pp_filters[0] = FindFilter( p_aout, p_input_format,
+ &temp_format );
}
- vlc_object_attach( pp_filters[1], p_aout );
-
- memcpy( &pp_filters[1]->input, &pp_filters[0]->output,
- sizeof(audio_sample_format_t) );
- memcpy( &pp_filters[1]->output, p_output_format,
- sizeof(audio_sample_format_t) );
- pp_filters[1]->p_module = module_Need( pp_filters[1], "audio filter",
- NULL );
- if ( pp_filters[1]->p_module == NULL )
+ if ( pp_filters[0] == NULL )
{
msg_Err( p_aout,
- "couldn't find a filter for the 2nd part of the conversion" );
- vlc_object_detach( pp_filters[0] );
- vlc_object_destroy( pp_filters[0] );
- vlc_object_detach( pp_filters[1] );
- vlc_object_destroy( pp_filters[1] );
+ "couldn't find a filter for the first part of the conversion" );
return -1;
}
- msg_Dbg( p_aout, "filter pipeline made of two filters" );
- *pi_nb_filters = 2;
+ /* We have the first stage of the conversion. Find a filter for
+ * the rest. */
+ pp_filters[1] = FindFilter( p_aout, &pp_filters[0]->output,
+ p_output_format );
+ if ( pp_filters[1] == NULL )
+ {
+ /* Try to split the conversion. */
+ i_nb_conversions = SplitConversion( p_aout,
+ &pp_filters[0]->output,
+ p_output_format, &temp_format );
+ if ( !i_nb_conversions )
+ {
+ vlc_object_detach( pp_filters[0] );
+ vlc_object_destroy( pp_filters[0] );
+ msg_Err( p_aout,
+ "couldn't find a filter for the second part of the conversion" );
+ }
+ pp_filters[1] = FindFilter( p_aout, &pp_filters[0]->output,
+ &temp_format );
+ pp_filters[2] = FindFilter( p_aout, &temp_format,
+ p_output_format );
+
+ if ( pp_filters[1] == NULL || pp_filters[2] == NULL )
+ {
+ vlc_object_detach( pp_filters[0] );
+ vlc_object_destroy( pp_filters[0] );
+ if ( pp_filters[1] != NULL )
+ {
+ vlc_object_detach( pp_filters[1] );
+ vlc_object_destroy( pp_filters[1] );
+ }
+ if ( pp_filters[2] != NULL )
+ {
+ vlc_object_detach( pp_filters[2] );
+ vlc_object_destroy( pp_filters[2] );
+ }
+ msg_Err( p_aout,
+ "couldn't find filters for the second part of the conversion" );
+ }
+ *pi_nb_filters = 3;
+ }
+ else
+ {
+ *pi_nb_filters = 2;
+ }
+ /* We have enough filters. */
+ msg_Dbg( p_aout, "found %d filters for the whole conversion",
+ *pi_nb_filters );
return 0;
}
{
aout_filter_t * p_filter = pp_filters[i];
- int i_output_size = aout_FormatToByterate( &p_filter->output,
- p_filter->output.i_rate );
- int i_input_size = aout_FormatToByterate( &p_filter->input,
- p_filter->input.i_rate );
+ int i_output_size = p_filter->output.i_bytes_per_frame
+ * p_filter->output.i_rate
+ / p_filter->output.i_frame_length;
+ int i_input_size = p_filter->input.i_bytes_per_frame
+ * p_filter->input.i_rate
+ / p_filter->input.i_frame_length;
p_first_alloc->i_bytes_per_sec = __MAX( p_first_alloc->i_bytes_per_sec,
i_output_size );
p_first_alloc->i_bytes_per_sec = __MAX(
p_first_alloc->i_bytes_per_sec,
i_input_size );
+ p_filter->output_alloc.i_alloc_type = AOUT_ALLOC_NONE;
}
else
{
aout_buffer_t * p_output_buffer;
aout_BufferAlloc( &p_filter->output_alloc,
- (u64)((*pp_input_buffer)->i_nb_samples * 1000000)
- / p_filter->output.i_rate, *pp_input_buffer,
+ (mtime_t)(*pp_input_buffer)->i_nb_samples * 1000000
+ / p_filter->input.i_rate, *pp_input_buffer,
p_output_buffer );
if ( p_output_buffer == NULL )
{
msg_Err( p_aout, "out of memory" );
return;
}
- /* Please note that p_output_buffer->i_nb_samples shall be set by
- * the filter plug-in. */
+ /* Please note that p_output_buffer->i_nb_samples & i_nb_bytes
+ * shall be set by the filter plug-in. */
p_filter->pf_do_work( p_aout, p_filter, *pp_input_buffer,
p_output_buffer );
projects / vlc / blobdiff