struct column and chunk, and optimization

save/chunk.go

@@ -0,0 +1,71 @@
+package save
+
+import (
+	"bytes"
+	"compress/gzip"
+	"compress/zlib"
+	"errors"
+	"github.com/Tnze/go-mc/nbt"
+	"io"
+)
+
+// Column is 16* chunk
+type Column struct {
+	DataVersion int
+	Level       struct {
+		Heightmaps map[string][]int64
+		Structures struct {
+			References map[string][]int64
+			Starts     map[string]struct {
+				ID string `nbt:"id"`
+			}
+		}
+		// Entities
+		// LiquidTicks
+		// PostProcessing
+		Sections []Chunk
+		// TileEntities
+		// TileTicks
+		InhabitedTime int64
+		IsLightOn     byte `nbt:"isLightOn"`
+		LastUpdate    int64
+		Status        string
+		PosX          int32 `nbt:"xPos"`
+		PosY          int32 `nbt:"yPos"`
+		Biomes        []int32
+	}
+}
+
+type Chunk struct {
+	Palette     []Block
+	Y           byte
+	BlockLight  []byte
+	BlockStates []int64
+	SkyLight    []byte
+}
+
+type Block struct {
+	Name       string
+	Properties map[string]interface{}
+}
+
+// Load read column data from []byte
+func (c *Column) Load(data []byte) (err error) {
+	var r io.Reader = bytes.NewReader(data[1:])
+
+	switch data[0] {
+	default:
+		err = errors.New("unknown compression")
+	case 1:
+		r, err = gzip.NewReader(r)
+	case 2:
+		r, err = zlib.NewReader(r)
+	}
+
+	if err != nil {
+		return err
+	}
+
+	err = nbt.NewDecoder(r).Decode(c)
+	return
+}