Use goreleaser to build and release (#244)

Instead of using gox on one side and an action to release, we can merge
them together with goreleaser which will build for extra targets (arm,
mips if needed in the future) and it also takes care of creating
checksums, a source archive, and a changelog and creating a release with
all the artifacts.

All binaries should respect the old naming convention, so any scripts
out there should still work.

Signed-off-by: Itxaka <igarcia@suse.com>

vendor/github.com/klauspost/compress/zstd/encoder.go

@@ -29,26 +29,28 @@ type Encoder struct {
 
 type encoder interface {
 	Encode(blk *blockEnc, src []byte)
+	EncodeNoHist(blk *blockEnc, src []byte)
 	Block() *blockEnc
 	CRC() *xxhash.Digest
 	AppendCRC([]byte) []byte
 	WindowSize(size int) int32
 	UseBlock(*blockEnc)
-	Reset()
+	Reset(d *dict, singleBlock bool)
 }
 
 type encoderState struct {
-	w             io.Writer
-	filling       []byte
-	current       []byte
-	previous      []byte
-	encoder       encoder
-	writing       *blockEnc
-	err           error
-	writeErr      error
-	nWritten      int64
-	headerWritten bool
-	eofWritten    bool
+	w                io.Writer
+	filling          []byte
+	current          []byte
+	previous         []byte
+	encoder          encoder
+	writing          *blockEnc
+	err              error
+	writeErr         error
+	nWritten         int64
+	headerWritten    bool
+	eofWritten       bool
+	fullFrameWritten bool
 
 	// This waitgroup indicates an encode is running.
 	wg sync.WaitGroup
@@ -70,27 +72,24 @@ func NewWriter(w io.Writer, opts ...EOption) (*Encoder, error) {
 	}
 	if w != nil {
 		e.Reset(w)
-	} else {
-		e.init.Do(func() {
-			e.initialize()
-		})
 	}
 	return &e, nil
 }
 
 func (e *Encoder) initialize() {
+	if e.o.concurrent == 0 {
+		e.o.setDefault()
+	}
 	e.encoders = make(chan encoder, e.o.concurrent)
 	for i := 0; i < e.o.concurrent; i++ {
-		e.encoders <- e.o.encoder()
+		enc := e.o.encoder()
+		e.encoders <- enc
 	}
 }
 
 // Reset will re-initialize the writer and new writes will encode to the supplied writer
 // as a new, independent stream.
 func (e *Encoder) Reset(w io.Writer) {
-	e.init.Do(func() {
-		e.initialize()
-	})
 	s := &e.state
 	s.wg.Wait()
 	s.wWg.Wait()
@@ -107,16 +106,17 @@ func (e *Encoder) Reset(w io.Writer) {
 		s.encoder = e.o.encoder()
 	}
 	if s.writing == nil {
-		s.writing = &blockEnc{}
+		s.writing = &blockEnc{lowMem: e.o.lowMem}
 		s.writing.init()
 	}
 	s.writing.initNewEncode()
 	s.filling = s.filling[:0]
 	s.current = s.current[:0]
 	s.previous = s.previous[:0]
-	s.encoder.Reset()
+	s.encoder.Reset(e.o.dict, false)
 	s.headerWritten = false
 	s.eofWritten = false
+	s.fullFrameWritten = false
 	s.w = w
 	s.err = nil
 	s.nWritten = 0
@@ -155,7 +155,7 @@ func (e *Encoder) Write(p []byte) (n int, err error) {
 		if err != nil {
 			return n, err
 		}
-		if debug && len(s.filling) > 0 {
+		if debugAsserts && len(s.filling) > 0 {
 			panic(len(s.filling))
 		}
 	}
@@ -175,14 +175,38 @@ func (e *Encoder) nextBlock(final bool) error {
 		return fmt.Errorf("block > maxStoreBlockSize")
 	}
 	if !s.headerWritten {
+		// If we have a single block encode, do a sync compression.
+		if final && len(s.filling) == 0 && !e.o.fullZero {
+			s.headerWritten = true
+			s.fullFrameWritten = true
+			s.eofWritten = true
+			return nil
+		}
+		if final && len(s.filling) > 0 {
+			s.current = e.EncodeAll(s.filling, s.current[:0])
+			var n2 int
+			n2, s.err = s.w.Write(s.current)
+			if s.err != nil {
+				return s.err
+			}
+			s.nWritten += int64(n2)
+			s.current = s.current[:0]
+			s.filling = s.filling[:0]
+			s.headerWritten = true
+			s.fullFrameWritten = true
+			s.eofWritten = true
+			return nil
+		}
+
 		var tmp [maxHeaderSize]byte
 		fh := frameHeader{
 			ContentSize:   0,
 			WindowSize:    uint32(s.encoder.WindowSize(0)),
 			SingleSegment: false,
 			Checksum:      e.o.crc,
-			DictID:        0,
+			DictID:        e.o.dict.ID(),
 		}
+
 		dst, err := fh.appendTo(tmp[:0])
 		if err != nil {
 			return err
@@ -262,7 +286,7 @@ func (e *Encoder) nextBlock(final bool) error {
 			// If we got the exact same number of literals as input,
 			// assume the literals cannot be compressed.
 			if len(src) != len(blk.literals) || len(src) != e.o.blockSize {
-				err = blk.encode()
+				err = blk.encode(src, e.o.noEntropy, !e.o.allLitEntropy)
 			}
 			switch err {
 			case errIncompressible:
@@ -292,12 +316,20 @@ func (e *Encoder) ReadFrom(r io.Reader) (n int64, err error) {
 	if debug {
 		println("Using ReadFrom")
 	}
-	// Maybe handle stuff queued?
+
+	// Flush any current writes.
+	if len(e.state.filling) > 0 {
+		if err := e.nextBlock(false); err != nil {
+			return 0, err
+		}
+	}
 	e.state.filling = e.state.filling[:e.o.blockSize]
 	src := e.state.filling
 	for {
 		n2, err := r.Read(src)
-		_, _ = e.state.encoder.CRC().Write(src[:n2])
+		if e.o.crc {
+			_, _ = e.state.encoder.CRC().Write(src[:n2])
+		}
 		// src is now the unfilled part...
 		src = src[n2:]
 		n += int64(n2)
@@ -307,14 +339,14 @@ func (e *Encoder) ReadFrom(r io.Reader) (n int64, err error) {
 			if debug {
 				println("ReadFrom: got EOF final block:", len(e.state.filling))
 			}
-			return n, e.nextBlock(true)
+			return n, nil
+		case nil:
 		default:
 			if debug {
 				println("ReadFrom: got error:", err)
 			}
 			e.state.err = err
 			return n, err
-		case nil:
 		}
 		if len(src) > 0 {
 			if debug {
@@ -362,6 +394,9 @@ func (e *Encoder) Close() error {
 	if err != nil {
 		return err
 	}
+	if e.state.fullFrameWritten {
+		return s.err
+	}
 	s.wg.Wait()
 	s.wWg.Wait()
 
@@ -421,19 +456,15 @@ func (e *Encoder) EncodeAll(src, dst []byte) []byte {
 		}
 		return dst
 	}
-	e.init.Do(func() {
-		e.o.setDefault()
-		e.initialize()
-	})
+	e.init.Do(e.initialize)
 	enc := <-e.encoders
 	defer func() {
 		// Release encoder reference to last block.
-		enc.Reset()
+		// If a non-single block is needed the encoder will reset again.
 		e.encoders <- enc
 	}()
-	enc.Reset()
-	blk := enc.Block()
-	single := len(src) > 1<<20
+	// Use single segments when above minimum window and below 1MB.
+	single := len(src) < 1<<20 && len(src) > MinWindowSize
 	if e.o.single != nil {
 		single = *e.o.single
 	}
@@ -442,11 +473,11 @@ func (e *Encoder) EncodeAll(src, dst []byte) []byte {
 		WindowSize:    uint32(enc.WindowSize(len(src))),
 		SingleSegment: single,
 		Checksum:      e.o.crc,
-		DictID:        0,
+		DictID:        e.o.dict.ID(),
 	}
 
 	// If less than 1MB, allocate a buffer up front.
-	if len(dst) == 0 && cap(dst) == 0 && len(src) < 1<<20 {
+	if len(dst) == 0 && cap(dst) == 0 && len(src) < 1<<20 && !e.o.lowMem {
 		dst = make([]byte, 0, len(src))
 	}
 	dst, err := fh.appendTo(dst)
@@ -454,26 +485,29 @@ func (e *Encoder) EncodeAll(src, dst []byte) []byte {
 		panic(err)
 	}
 
-	for len(src) > 0 {
-		todo := src
-		if len(todo) > e.o.blockSize {
-			todo = todo[:e.o.blockSize]
-		}
-		src = src[len(todo):]
+	// If we can do everything in one block, prefer that.
+	if len(src) <= maxCompressedBlockSize {
+		enc.Reset(e.o.dict, true)
+		// Slightly faster with no history and everything in one block.
 		if e.o.crc {
-			_, _ = enc.CRC().Write(todo)
+			_, _ = enc.CRC().Write(src)
 		}
-		blk.reset(nil)
-		blk.pushOffsets()
-		enc.Encode(blk, todo)
-		if len(src) == 0 {
-			blk.last = true
+		blk := enc.Block()
+		blk.last = true
+		if e.o.dict == nil {
+			enc.EncodeNoHist(blk, src)
+		} else {
+			enc.Encode(blk, src)
 		}
-		err := errIncompressible
+
 		// If we got the exact same number of literals as input,
 		// assume the literals cannot be compressed.
-		if len(blk.literals) != len(todo) || len(todo) != e.o.blockSize {
-			err = blk.encode()
+		err := errIncompressible
+		oldout := blk.output
+		if len(blk.literals) != len(src) || len(src) != e.o.blockSize {
+			// Output directly to dst
+			blk.output = dst
+			err = blk.encode(src, e.o.noEntropy, !e.o.allLitEntropy)
 		}
 
 		switch err {
@@ -481,13 +515,51 @@ func (e *Encoder) EncodeAll(src, dst []byte) []byte {
 			if debug {
 				println("Storing incompressible block as raw")
 			}
-			blk.encodeRaw(todo)
-			blk.popOffsets()
+			dst = blk.encodeRawTo(dst, src)
 		case nil:
+			dst = blk.output
 		default:
 			panic(err)
 		}
-		dst = append(dst, blk.output...)
+		blk.output = oldout
+	} else {
+		enc.Reset(e.o.dict, false)
+		blk := enc.Block()
+		for len(src) > 0 {
+			todo := src
+			if len(todo) > e.o.blockSize {
+				todo = todo[:e.o.blockSize]
+			}
+			src = src[len(todo):]
+			if e.o.crc {
+				_, _ = enc.CRC().Write(todo)
+			}
+			blk.pushOffsets()
+			enc.Encode(blk, todo)
+			if len(src) == 0 {
+				blk.last = true
+			}
+			err := errIncompressible
+			// If we got the exact same number of literals as input,
+			// assume the literals cannot be compressed.
+			if len(blk.literals) != len(todo) || len(todo) != e.o.blockSize {
+				err = blk.encode(todo, e.o.noEntropy, !e.o.allLitEntropy)
+			}
+
+			switch err {
+			case errIncompressible:
+				if debug {
+					println("Storing incompressible block as raw")
+				}
+				dst = blk.encodeRawTo(dst, todo)
+				blk.popOffsets()
+			case nil:
+				dst = append(dst, blk.output...)
+			default:
+				panic(err)
+			}
+			blk.reset(nil)
+		}
 	}
 	if e.o.crc {
 		dst = enc.AppendCRC(dst)