Merge branch 'copy-bufs'

Author: dudk

.travis.yml

@@ -1,14 +1,13 @@
 language: go
-go: 
- - "1.11.x"
+go:
+ - "1.x"
 
 os: osx
 
 install: true
 
 env:
   global:
-    - GO111MODULE=on
     - CODECOV_TOKEN="4831691a-c57e-465a-9590-5298c70d066b"
 
 script: go test -race -coverprofile=coverage.txt -covermode=atomic ./...

go.sum

@@ -2,6 +2,7 @@ github.com/davecgh/go-spew v1.1.0 h1:ZDRjVQ15GmhC3fiQ8ni8+OwkZQO4DARzQgrnXU1Liz8
 github.com/davecgh/go-spew v1.1.0/go.mod h1:J7Y8YcW2NihsgmVo/mv3lAwl/skON4iLHjSsI+c5H38=
 github.com/pmezard/go-difflib v1.0.0 h1:4DBwDE0NGyQoBHbLQYPwSUPoCMWR5BEzIk/f1lZbAQM=
 github.com/pmezard/go-difflib v1.0.0/go.mod h1:iKH77koFhYxTK1pcRnkKkqfTogsbg7gZNVY4sRDYZ/4=
+github.com/stretchr/objx v0.1.0 h1:4G4v2dO3VZwixGIRoQ5Lfboy6nUhCyYzaqnIAPPhYs4=
 github.com/stretchr/objx v0.1.0/go.mod h1:HFkY916IF+rwdDfMAkV7OtwuqBVzrE8GR6GFx+wExME=
 github.com/stretchr/testify v1.3.0 h1:TivCn/peBQ7UY8ooIcPgZFpTNSz0Q2U6UrFlUfqbe0Q=
 github.com/stretchr/testify v1.3.0/go.mod h1:M5WIy9Dh21IEIfnGCwXGc5bZfKNJtfHm1UVUgZn+9EI=

signal.go

@@ -61,61 +61,92 @@ func resolution(bitDepth BitDepth) int {
 	return 1<<(bitDepth-1) - 1
 }
 
-// AsFloat64 converts interleaved int signal to float64.
+// Size of non-interleaved data.
+func (ints InterInt) Size() int {
+	return int(math.Ceil(float64(len(ints.Data)) / float64(ints.NumChannels)))
+}
+
+// AsFloat64 allocates new Float64 buffer of the same
+// size and copies signal values there.
 func (ints InterInt) AsFloat64() Float64 {
 	if ints.Data == nil || ints.NumChannels == 0 {
 		return nil
 	}
 	floats := make([][]float64, ints.NumChannels)
-	bufSize := int(math.Ceil(float64(len(ints.Data)) / float64(ints.NumChannels)))
 
-	// get resolution of bit depth
+	for i := range floats {
+		floats[i] = make([]float64, ints.Size())
+	}
+	ints.CopyToFloat64(floats)
+	return floats
+}
+
+// CopyToFloat64 buffer the values of InterInt buffer.
+// If number of channels is not equal, function will panic.
+func (ints InterInt) CopyToFloat64(floats Float64) {
+	if ints.NumChannels != floats.NumChannels() {
+		panic(fmt.Errorf("unexpected number of channels in destination buffer: expected %v got %v", ints.NumChannels, floats.NumChannels()))
+	}
+	// get resolution of bit depth.
 	res := resolution(ints.BitDepth)
-	// determine the divider for bit depth conversion
+	// determine the divider for bit depth conversion.
 	divider := float64(res)
-	// determine the shift for signed-unsigned conversion
+	// determine the shift for signed-unsigned conversion.
 	shift := 0
 	if ints.Unsigned {
 		shift = res
 	}
 
 	for i := range floats {
-		floats[i] = make([]float64, bufSize)
-		pos := 0
-		for j := i; j < len(ints.Data); j = j + ints.NumChannels {
-			floats[i][pos] = float64(ints.Data[j]-shift) / divider
-			pos++
+		for pos, j := i, 0; pos < len(ints.Data) && j < len(floats[i]); pos, j = pos+ints.NumChannels, j+1 {
+			floats[i][j] = float64(ints.Data[pos]-shift) / divider
 		}
 	}
-	return floats
 }
 
-// AsInterInt converts float64 signal to interleaved int.
-// If unsigned is true, then all values are shifted and result will be in unsigned range.
-func (floats Float64) AsInterInt(bitDepth BitDepth, unsigned bool) []int {
-	var numChannels int
-	if numChannels = len(floats); numChannels == 0 {
-		return nil
+// AsInterInt allocates new interleaved int buffer of
+// the same size and copies signal values there.
+// If unsigned is true, then all values are shifted
+// and result will be in unsigned range.
+func (floats Float64) AsInterInt(bitDepth BitDepth, unsigned bool) InterInt {
+	numChannels := floats.NumChannels()
+	if numChannels == 0 {
+		return InterInt{}
 	}
 
+	ints := InterInt{
+		Data:        make([]int, len(floats[0])*numChannels),
+		NumChannels: numChannels,
+		BitDepth:    bitDepth,
+		Unsigned:    unsigned,
+	}
+
+	floats.CopyToInterInt(ints)
+	return ints
+}
+
+// CopyToInterInt buffer the values of Float64 buffer.
+// If number of channels is not equal, function will panic.
+func (floats Float64) CopyToInterInt(ints InterInt) {
+	if floats.NumChannels() != ints.NumChannels {
+		panic(fmt.Errorf("unexpected number of channels in destination buffer: expected %v got %v", floats.NumChannels(), ints.NumChannels))
+	}
 	// get resolution of bit depth
-	res := resolution(bitDepth)
+	res := resolution(ints.BitDepth)
 	// determine the multiplier for bit depth conversion
 	multiplier := float64(res)
 	// determine the shift for signed-unsigned conversion
 	shift := 0
-	if unsigned {
+	if ints.Unsigned {
 		shift = res
 	}
 
-	ints := make([]int, len(floats[0])*numChannels)
-
+	size := ints.Size()
 	for j := range floats {
-		for i := range floats[j] {
-			ints[i*numChannels+j] = int(floats[j][i]*multiplier) + shift
+		for i := 0; i < len(floats[j]) && i < size; i++ {
+			ints.Data[i*ints.NumChannels+j] = int(floats[j][i]*multiplier) + shift
 		}
 	}
-	return ints
 }
 
 // Float64Buffer returns an Float64 buffer of specified dimentions.
@@ -127,21 +158,21 @@ func Float64Buffer(numChannels, bufferSize int) Float64 {
 	return result
 }
 
-// NumChannels returns number of channels in this sample slice
+// NumChannels returns number of channels in this sample slice.
 func (floats Float64) NumChannels() int {
 	return len(floats)
 }
 
-// Size returns number of samples in single block in this sample slice
+// Size returns number of samples in single block in this sample slice.
 func (floats Float64) Size() int {
 	if floats.NumChannels() == 0 {
 		return 0
 	}
 	return len(floats[0])
 }
 
-// Append buffers set to existing one one
-// new buffer is returned if b is nil
+// Append buffers to existing one.
+// New buffer is returned if b is nil.
 func (floats Float64) Append(source Float64) Float64 {
 	if floats == nil {
 		floats = make([][]float64, source.NumChannels())
@@ -155,13 +186,12 @@ func (floats Float64) Append(source Float64) Float64 {
 	return floats
 }
 
-// Slice creates a new copy of buffer from start position with defined legth
-// if buffer doesn't have enough samples - shorten block is returned
-//
-// if start >= buffer size, nil is returned
-// if start + len >= buffer size, len is decreased till the end of slice
-// if start < 0, nil is returned
-func (floats Float64) Slice(start int, len int) Float64 {
+// Slice creates a new buffer that refers to floats data from start
+// position with defined legth. Shorten block is returned if buffer
+// doesn't have enough samples. If start is less than 0 or more than
+// buffer size, nil is returned. If len goes beyond the buffer size,
+// it's truncated up to length of the buffer.
+func (floats Float64) Slice(start, len int) Float64 {
 	if floats == nil || start >= floats.Size() || start < 0 {
 		return nil
 	}
@@ -171,7 +201,22 @@ func (floats Float64) Slice(start int, len int) Float64 {
 		if end > floats.Size() {
 			end = floats.Size()
 		}
-		result[i] = append(result[i], floats[i][start:end]...)
+		result[i] = floats[i][start:end]
 	}
 	return result
 }
+
+// Sum adds values from one buffer to another.
+// The lesser dimensions are used.
+func (floats Float64) Sum(b Float64) Float64 {
+	if floats == nil {
+		return nil
+	}
+
+	for i := 0; i < len(floats) && i < len(b); i++ {
+		for j := 0; j < len(floats[i]) && j < len(b[i]); j++ {
+			floats[i][j] += b[i][j]
+		}
+	}
+	return floats
+}