Merge remote-tracking branch 'origin/snbt_convertor' into snbt_convertor

# Conflicts:
#	nbt/snbt_scanner.go
#	nbt/snbt_scanner_test.go

nbt/encode.go

@@ -41,7 +41,7 @@ func (e *Encoder) marshal(val reflect.Value, tagType byte, tagName string) error
 func (e *Encoder) writeHeader(val reflect.Value, tagType byte, tagName string) (err error) {
 	if tagType == TagList {
 		eleType := getTagType(val.Type().Elem())
-		err = e.writeListHeader(eleType, tagName, val.Len())
+		err = e.writeListHeader(eleType, tagName, val.Len(), true)
 	} else {
 		err = e.writeTag(tagType, tagName)
 	}
@@ -224,9 +224,11 @@ func (e *Encoder) writeTag(tagType byte, tagName string) error {
 	return err
 }
 
-func (e *Encoder) writeListHeader(elementType byte, tagName string, n int) (err error) {
+func (e *Encoder) writeListHeader(elementType byte, tagName string, n int, writeTag bool) (err error) {
-	if err = e.writeTag(TagList, tagName); err != nil {
+	if writeTag {
-		return
+		if err = e.writeTag(TagList, tagName); err != nil {
+			return
+		}
 	}
 	if _, err = e.w.Write([]byte{elementType}); err != nil {
 		return

nbt/snbt_decode.go

@@ -39,7 +39,8 @@ func writeValue(e *Encoder, d *decodeState, tagName string) error {
 		e.writeTag(TagCompound, tagName)
 		return writeCompoundPayload(e, d)
 	case scanBeginList:
-		return writeListOrArray(e, d, tagName)
+		_, err := writeListOrArray(e, d, true, tagName)
+		return err
 	}
 }
 
@@ -66,6 +67,7 @@ func writeLiteralPayload(e *Encoder, v interface{}) error {
 }
 
 func writeCompoundPayload(e *Encoder, d *decodeState) error {
+	defer d.scanNext()
 	for {
 		d.scanWhile(scanSkipSpace)
 		if d.opcode == scanEndValue {
@@ -77,7 +79,12 @@ func writeCompoundPayload(e *Encoder, d *decodeState) error {
 		// read tag name
 		start := d.readIndex()
 		d.scanWhile(scanContinue)
-		tagName := string(d.data[start:d.readIndex()])
+		var tagName string
+		if tt, v := parseLiteral(d.data[start:d.readIndex()]); tt == TagString {
+			tagName = v.(string)
+		} else {
+			tagName = string(d.data[start:d.readIndex()])
+		}
 		// read value
 		if d.opcode == scanSkipSpace {
 			d.scanWhile(scanSkipSpace)
@@ -102,11 +109,12 @@ func writeCompoundPayload(e *Encoder, d *decodeState) error {
 	return nil
 }
 
-func writeListOrArray(e *Encoder, d *decodeState, tagName string) error {
+func writeListOrArray(e *Encoder, d *decodeState, writeTag bool, tagName string) (tagType byte, err error) {
 	d.scanWhile(scanSkipSpace)
 	if d.opcode == scanEndValue { // ']', empty TAG_List
-		e.writeListHeader(TagEnd, tagName, 0)
+		e.writeListHeader(TagEnd, tagName, 0, writeTag)
-		return nil
+		d.scanNext()
+		return TagList, nil
 	}
 
 	// We don't know the length of the List,
@@ -120,29 +128,42 @@ func writeListOrArray(e *Encoder, d *decodeState, tagName string) error {
 	case scanBeginLiteral:
 		d.scanWhile(scanContinue)
 		literal := d.data[start:d.readIndex()]
+		if d.opcode == scanSkipSpace {
+			d.scanWhile(scanSkipSpace)
+		}
 		if d.opcode == scanListType { // TAG_X_Array
 			var elemType byte
 			switch literal[0] {
 			case 'B':
-				e.writeTag(TagByteArray, tagName)
+				tagType = TagByteArray
 				elemType = TagByte
 			case 'I':
-				e.writeTag(TagIntArray, tagName)
+				tagType = TagIntArray
 				elemType = TagInt
 			case 'L':
-				e.writeTag(TagLongArray, tagName)
+				tagType = TagLongArray
 				elemType = TagLong
+			default:
+				return TagList, errors.New("unknown Array type")
+			}
+			if writeTag {
+				e.writeTag(tagType, tagName)
+			}
+			if d.opcode == scanSkipSpace {
+				d.scanWhile(scanSkipSpace)
+			}
+			d.scanWhile(scanSkipSpace)    // ;
+			if d.opcode == scanEndValue { // ]
+				// empty array
+				e.writeInt32(0)
+				break
 			}
 			for {
-				d.scanNext()
 				if d.opcode == scanSkipSpace {
 					d.scanWhile(scanSkipSpace)
 				}
-				if d.opcode == scanEndValue { // ]
-					break
-				}
 				if d.opcode != scanBeginLiteral {
-					return errors.New("not literal in Array")
+					return tagType, errors.New("not literal in Array")
 				}
 				start := d.readIndex()
 
@@ -150,7 +171,7 @@ func writeListOrArray(e *Encoder, d *decodeState, tagName string) error {
 				literal := d.data[start:d.readIndex()]
 				tagType, litVal := parseLiteral(literal)
 				if tagType != elemType {
-					return errors.New("unexpected element type in TAG_Array")
+					return tagType, errors.New("unexpected element type in TAG_Array")
 				}
 				switch elemType {
 				case TagByte:
@@ -161,6 +182,17 @@ func writeListOrArray(e *Encoder, d *decodeState, tagName string) error {
 					e2.writeInt64(litVal.(int64))
 				}
 				count++
+
+				if d.opcode == scanSkipSpace {
+					d.scanWhile(scanSkipSpace)
+				}
+				if d.opcode == scanEndValue { // ]
+					break
+				}
+				if d.opcode != scanListValue {
+					panic(phasePanicMsg)
+				}
+				d.scanWhile(scanSkipSpace) // ,
 			}
 			e.writeInt32(int32(count))
 			e.w.Write(buf.Bytes())
@@ -176,7 +208,7 @@ func writeListOrArray(e *Encoder, d *decodeState, tagName string) error {
 				tagType = t
 			}
 			if t != tagType {
-				return errors.New("different TagType in List")
+				return TagList, errors.New("different TagType in List")
 			}
 			writeLiteralPayload(e2, v)
 			count++
@@ -191,15 +223,41 @@ func writeListOrArray(e *Encoder, d *decodeState, tagName string) error {
 			if d.opcode != scanListValue {
 				panic(phasePanicMsg)
 			}
-			d.scanNext()
+			d.scanWhile(scanSkipSpace)
 			start = d.readIndex()
 			d.scanWhile(scanContinue)
 			literal = d.data[start:d.readIndex()]
 		}
-		e.writeListHeader(tagType, tagName, count)
+		e.writeListHeader(tagType, tagName, count, writeTag)
 		e.w.Write(buf.Bytes())
 	case scanBeginList: // TAG_List<TAG_List>
-		e.writeListHeader(TagList, tagName, count)
+		var elemType byte
+		for {
+			if d.opcode == scanSkipSpace {
+				d.scanWhile(scanSkipSpace)
+			}
+			if d.opcode != scanBeginList {
+				return TagList, errors.New("different TagType in List")
+			}
+			elemType, err = writeListOrArray(e2, d, false, "")
+			if err != nil {
+				return tagType, err
+			}
+			count++
+			if d.opcode == scanSkipSpace {
+				d.scanWhile(scanSkipSpace)
+			}
+			// ',' or ']'
+			if d.opcode == scanEndValue {
+				break
+			}
+			if d.opcode != scanListValue {
+				panic(phasePanicMsg)
+			}
+			// read '['
+			d.scanNext()
+		}
+		e.writeListHeader(elemType, tagName, count, writeTag)
 		e.w.Write(buf.Bytes())
 	case scanBeginCompound: // TAG_List<TAG_Compound>
 		for {
@@ -207,7 +265,7 @@ func writeListOrArray(e *Encoder, d *decodeState, tagName string) error {
 				d.scanWhile(scanSkipSpace)
 			}
 			if d.opcode != scanBeginCompound {
-				return errors.New("different TagType in List")
+				return TagList, errors.New("different TagType in List")
 			}
 			writeCompoundPayload(e2, d)
 			count++
@@ -215,7 +273,6 @@ func writeListOrArray(e *Encoder, d *decodeState, tagName string) error {
 				d.scanWhile(scanSkipSpace)
 			}
 			// read ',' or ']'
-			d.scanNext()
 			if d.opcode == scanSkipSpace {
 				d.scanWhile(scanSkipSpace)
 			}
@@ -228,10 +285,11 @@ func writeListOrArray(e *Encoder, d *decodeState, tagName string) error {
 			// read '{'
 			d.scanNext()
 		}
-		e.writeListHeader(TagCompound, tagName, count)
+		e.writeListHeader(TagCompound, tagName, count, writeTag)
 		e.w.Write(buf.Bytes())
 	}
-	return nil
+	d.scanNext()
+	return
 }
 
 // readIndex returns the position of the last byte read.
@@ -298,7 +356,7 @@ func parseLiteral(literal []byte) (byte, interface{}) {
 			if isNumber(c) {
 				continue
 			} else if integer {
-				if i == strlen-1 && isIntegerType(c) {
+				if i == strlen-1 && i != 0 && isIntegerType(c) {
 					numberType = c
 					strlen--
 				} else if i > 0 || i == 0 && c != '-' {

nbt/snbt_decode_test.go

@@ -25,22 +25,28 @@ func TestEncoder_WriteSNBT(t *testing.T) {
 		{`{}`, []byte{10, 0, 0, 0}},
 		{`{a:1b}`, []byte{10, 0, 0, 1, 0, 1, 'a', 1, 0}},
 		{`{ a : 1b }`, []byte{10, 0, 0, 1, 0, 1, 'a', 1, 0}},
-		{`{a:1,2:c}`, []byte{10, 0, 0, 3, 0, 1, 'a', 0, 0, 0, 1, 8, 0, 1, '2', 0, 1, 'c', 0}},
+		{`{b:1,2:c}`, []byte{10, 0, 0, 3, 0, 1, 'b', 0, 0, 0, 1, 8, 0, 1, '2', 0, 1, 'c', 0}},
-		{`{a:{b:{}}}`, []byte{10, 0, 0, 10, 0, 1, 'a', 10, 0, 1, 'b', 0, 0, 0}},
+		{`{c:{d:{}}}`, []byte{10, 0, 0, 10, 0, 1, 'c', 10, 0, 1, 'd', 0, 0, 0}},
+		{`{h:{},"i":{}}`, []byte{10, 0, 0, 10, 0, 1, 'h', 0, 10, 0, 1, 'i', 0, 0}},
 
 		{`[]`, []byte{9, 0, 0, 0, 0, 0, 0, 0}},
 		{`[1b,2b,3b]`, []byte{9, 0, 0, 1, 0, 0, 0, 3, 1, 2, 3}},
+		{`[ 1b , 2b , 3b ]`, []byte{9, 0, 0, 1, 0, 0, 0, 3, 1, 2, 3}},
 		{`[a,"b",'c']`, []byte{9, 0, 0, 8, 0, 0, 0, 3, 0, 1, 'a', 0, 1, 'b', 0, 1, 'c'}},
 		{`[{},{a:1b},{}]`, []byte{9, 0, 0, 10, 0, 0, 0, 3, 0, 1, 0, 1, 'a', 1, 0, 0}},
 		{`[ {  } , { a  : 1b  } , { } ] `, []byte{9, 0, 0, 10, 0, 0, 0, 3, 0, 1, 0, 1, 'a', 1, 0, 0}},
-		{`[[],[]]`, []byte{9, 0, 0, 9, 0, 0, 0, 2, 0, 0, 0, 0, 0, 0, 0, 0}},
+		{`[[],[]]`, []byte{9, 0, 0, 9, 0, 0, 0, 2, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0}},
 
-		{`[B;]`, []byte{7, 0, 0, 0, 0, 0, 0}},
+		{`[B; ]`, []byte{7, 0, 0, 0, 0, 0, 0}},
-		{`[B;1,2,3]`, []byte{7, 0, 0, 0, 0, 0, 3, 1, 2, 3}},
+		{`[B; 1b ,2B,3B]`, []byte{7, 0, 0, 0, 0, 0, 3, 1, 2, 3}},
 		{`[I;]`, []byte{11, 0, 0, 0, 0, 0, 0}},
-		{`[I;1,2,3]`, []byte{11, 0, 0, 0, 0, 0, 3, 0, 0, 0, 1, 0, 0, 0, 2, 0, 0, 0, 3}},
+		{`[I; 1, 2 ,3]`, []byte{11, 0, 0, 0, 0, 0, 3, 0, 0, 0, 1, 0, 0, 0, 2, 0, 0, 0, 3}},
 		{`[L;]`, []byte{12, 0, 0, 0, 0, 0, 0}},
-		{`[L;1,2,3]`, []byte{12, 0, 0, 0, 0, 0, 3, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 2, 0, 0, 0, 0, 0, 0, 0, 3}},
+		{`[ L; 1L,2L,3L]`, []byte{12, 0, 0, 0, 0, 0, 3, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 2, 0, 0, 0, 0, 0, 0, 0, 3}},
+
+		{`{d:[]}`, []byte{10, 0, 0, 9, 0, 1, 'd', 0, 0, 0, 0, 0, 0}},
+		{`{e:[]}`, []byte{10, 0, 0, 9, 0, 1, 'e', 0, 0, 0, 0, 0, 0}},
+		{`{f:[], g:[]}`, []byte{10, 0, 0, 9, 0, 1, 'f', 0, 0, 0, 0, 0, 9, 0, 1, 'g', 0, 0, 0, 0, 0, 0}},
 	}
 	for i := range testCases {
 		buf.Reset()
@@ -55,3 +61,13 @@ func TestEncoder_WriteSNBT(t *testing.T) {
 		}
 	}
 }
+
+func TestEncoder_WriteSNBT_bigTest(t *testing.T) {
+	var buf bytes.Buffer
+	e := NewEncoder(&buf)
+
+	err := e.WriteSNBT(bigTestSNBT)
+	if err != nil {
+		t.Error(err)
+	}
+}

nbt/snbt_scanner.go

@@ -9,10 +9,10 @@ const (
 	scanBeginLiteral           // end implied by next result != scanContinue
 	scanBeginCompound          // begin TAG_Compound (after left-brace )
 	scanBeginList              // begin TAG_List (after left-brack)
-	scanListValue              // just finished read list value
+	scanListValue              // just finished read list value (after comma)
 	scanListType               // just finished read list type (after "B;" or "L;")
 	scanCompoundTagName        // just finished read tag name (before colon)
-	scanCompoundValue          // just finished read value (before comma or right-brace )
+	scanCompoundValue          // just finished read value (after comma)
 	scanSkipSpace              // space byte; can skip; known to be last "continue" result
 	scanEndValue
 
@@ -176,7 +176,7 @@ func stateInSingleQuotedString(s *scanner, c byte) int {
 
 func stateInSingleQuotedStringEsc(s *scanner, c byte) int {
 	switch c {
-	case 'b', 'f', 'n', 'r', 't', '\\', '/', '\'':
+	case '\\', '\'':
 		s.step = stateInSingleQuotedString
 		return scanContinue
 	}
@@ -235,6 +235,9 @@ func stateListOrArrayT(s *scanner, c byte) int {
 }
 
 func stateArrayT(s *scanner, c byte) int {
+	if isSpace(c) {
+		return scanSkipSpace
+	}
 	if c == ']' { // empty array
 		return scanEndValue
 	}
@@ -258,10 +261,6 @@ func stateNum0(s *scanner, c byte) int {
 		s.step = stateNum1
 		return scanContinue
 	}
-	if isAllowedInUnquotedString(c) {
-		s.step = stateInUnquotedString
-		return scanContinue
-	}
 	return stateEndNumValue(s, c)
 }
 
@@ -274,10 +273,6 @@ func stateNum1(s *scanner, c byte) int {
 		s.step = stateNumDot
 		return scanContinue
 	}
-	if isAllowedInUnquotedString(c) {
-		s.step = stateInUnquotedString
-		return scanContinue
-	}
 	return stateEndNumValue(s, c)
 }
 
@@ -302,10 +297,6 @@ func stateNumDot0(s *scanner, c byte) int {
 		s.step = stateNumDot0
 		return scanContinue
 	}
-	if isAllowedInUnquotedString(c) {
-		s.step = stateInUnquotedString
-		return scanContinue
-	}
 	return stateEndNumDotValue(s, c)
 }
 
@@ -323,6 +314,10 @@ func stateEndNumValue(s *scanner, c byte) int {
 	case 'f', 'F', 'd', 'D':
 		return stateEndNumDotValue(s, c)
 	}
+	if isAllowedInUnquotedString(c) {
+		s.step = stateInUnquotedString
+		return scanContinue
+	}
 	return stateEndValue(s, c)
 }
 

nbt/snbt_scanner_test.go

@@ -41,13 +41,14 @@ func TestSNBT_number(t *testing.T) {
 }
 
 //go:embed bigTest_test.snbt
-var bigTest string
+var bigTestSNBT string
 
 func TestSNBT_compound(t *testing.T) {
 	goods := []string{
 		`{}`, `{name:3.14f}`, `{ "name" : 12345 }`,
 		`{ abc: { }}`, `{ "a b\"c": {}, def: 12345}`,
-		bigTest,
+		`{ ghi: [], klm: 1}`,
+		bigTestSNBT,
 	}
 	var s scanner
 	for _, str := range goods {
@@ -67,7 +68,7 @@ func TestSNBT_list(t *testing.T) {
 		`[{}, {}, {"a\"b":520}]`, // List of Compound
 		`[B,C,D]`, `[L, "abc"]`,  // List of string (like array)
 		`[B; 01B, 02B, 3B, 10B, 127B]`, // Array
-		`[I;]`,                         // Empty array
+		`[I;]`, `[B;   ]`,              // Empty array
 	}
 	var s scanner
 	scan := func(str string) bool {
@@ -91,10 +92,10 @@ func BenchmarkSNBT_bigTest(b *testing.B) {
 	var s scanner
 	for i := 0; i < b.N; i++ {
 		s.reset()
-		for _, c := range []byte(bigTest) {
+		for _, c := range []byte(bigTestSNBT) {
 			res := s.step(&s, c)
 			if res == scanError {
-				b.Errorf("scan valid data %q error: %v at [%d]", bigTest[:i], s.err, i)
+				b.Errorf("scan valid data %q error: %v at [%d]", bigTestSNBT[:i], s.err, i)
 				break
 			}
 		}