// Read samples as unsigned 8-bit
long read_samples( Blip_Buffer& buf, unsigned char* out, long out_size )
{
// Limit number of samples read to those available
long count = buf.samples_avail();
if ( count > out_size )
count = out_size;
// Begin reading samples from Blip_Buffer
Blip_Reader reader;
int bass = reader.begin( buf );
for ( long n = count; n--; )
{
// Read 16-bit sample and convert to output format
long s = reader.read();
reader.next( bass );
if ( (short) s != s ) // clamp to 16 bits
s = 0x7fff - (s >> 24);
*out++ = (s >> 8) + 0x80;
}
// End reading and remove the samples
reader.end( buf );
buf.remove_samples( count );
return count;
}
// Read samples as floating-point, with values staying
// between 'low' and 'high' (defaults to -1.0 to 1.0).
long read_samples( Blip_Buffer& buf, float* out, long out_size,
float low = -1.0f, float high = 1.0f )
{
// Limit number of samples read to those available
long count = buf.samples_avail();
if ( count > out_size )
count = out_size;
// Begin reading samples from Blip_Buffer
Blip_Reader reader;
int bass = reader.begin( buf );
float factor = (high - low) * 0.5f;
float offset = low + 1.0f * factor;
unsigned long const sample_range = 1UL << blip_sample_bits;
factor *= 1.0f / (sample_range / 2);
for ( long n = count; n--; )
{
// Read sample at full resolution and convert to output format
long s = reader.read_raw();
reader.next( bass );
*out++ = s * factor + offset;
if ( (unsigned long) (s + sample_range / 2) >= sample_range )
{
// clamp
out [-1] = high;
if ( s < 0 )
out [-1] = low;
}
}
// End reading and remove the samples
reader.end( buf );
buf.remove_samples( count );
return count;
}