Modify comment

Author: qishenonly

config/options.go

@@ -3,26 +3,27 @@ package config
 import "os"
 
 type Options struct {
-	DirPath      string //数据库数据目录
-	DataFileSize int64  //数据文件的大小
-	SyncWrite    bool   // 每次写数据是否持久化
+	DirPath      string // Database data directory
+	DataFileSize int64  // Size of data files
+	SyncWrite    bool   // Whether to persist data on every write
 	IndexType    IndexerType
 	FIOType      FIOType
 }
 
-// IteratorOptions 索引迭代器配置项
+// IteratorOptions is the configuration for index iteration.
 type IteratorOptions struct {
-	// 遍历前缀为指定值的 Key，默认为空
+	// Prefix specifies the prefix value for keys to iterate over. Default is empty.
 	Prefix []byte
-	// 是否反向遍历，默认 false 是正向
+	// Reverse indicates whether to iterate in reverse order.
+	// Default is false for forward iteration.
 	Reverse bool
 }
 
-// WriteBatchOptions 批量写入配置项
+// WriteBatchOptions is the configuration for batch writing.
 type WriteBatchOptions struct {
-	// 一个批次当中最大的数据量
+	// MaxBatchNum is the maximum number of data entries in a batch.
 	MaxBatchNum uint
-	// 提交时是否 sync 持久化
+	// SyncWrites indicates whether to sync (persist) the data on batch commit.
 	SyncWrites bool
 }
 
@@ -37,10 +38,10 @@ const (
 type IndexerType = int8
 
 const (
-	// Btree 索引
+	// Btree
 	Btree IndexerType = iota + 1
 
-	// ART (Adpative Radix Tree) 自适应基数树
+	// ART (Adpative Radix Tree)
 	ART
 )
 

engine/batch.go

@@ -35,6 +35,7 @@ func (db *DB) NewWriteBatch(opt config.WriteBatchOptions) *WriteBatch {
 
 // Put Data batch write
 func (wb *WriteBatch) Put(key []byte, value []byte) error {
+	// Check if the key is empty
 	if len(key) == 0 {
 		return _const.ErrKeyIsEmpty
 	}
@@ -46,19 +47,25 @@ func (wb *WriteBatch) Put(key []byte, value []byte) error {
 		Key:   key,
 		Value: value,
 	}
+
+	// Add the LogRecord to the temporaryDataWrites map using the key as a string
 	wb.temporaryDataWrites[string(key)] = logRecord
 	return nil
 }
 
 // Delete Batch deletion of data
 func (wb *WriteBatch) Delete(key []byte) error {
+	// Check if the key is empty
 	if len(key) == 0 {
 		return _const.ErrKeyIsEmpty
 	}
+
+	// Acquire a lock to ensure thread safety
 	wb.lock.Lock()
 	defer wb.lock.Unlock()
 
-	// If the data does not exist, return it directly
+	// If the data does not exist, delete it from
+	// temporaryDataWrites if present and return directly
 	logRecordPst := wb.db.index.Get(key)
 	if logRecordPst == nil {
 		if wb.temporaryDataWrites[string(key)] != nil {
@@ -73,6 +80,7 @@ func (wb *WriteBatch) Delete(key []byte) error {
 		Type: data.LogRecordDeleted,
 	}
 	wb.temporaryDataWrites[string(key)] = logRecord
+
 	return nil
 }
 
@@ -127,9 +135,11 @@ func (wb *WriteBatch) Commit() error {
 	for _, record := range wb.temporaryDataWrites {
 		pst := positions[string(record.Key)]
 		if record.Type == data.LogRecordNormal {
+			// Put the record in the index if it is of type LogRecordNormal
 			wb.db.index.Put(record.Key, pst)
 		}
 		if record.Type == data.LogRecordDeleted {
+			// Delete the record from the index if it is of type LogRecordDeleted
 			wb.db.index.Delete(record.Key)
 		}
 	}
@@ -146,8 +156,13 @@ func encodeLogRecordKeyWithSeq(key []byte, seqNo uint64) []byte {
 	seq := make([]byte, binary.MaxVarintLen64)
 	n := binary.PutUvarint(seq[:], seqNo)
 
+	// Create a byte slice to hold the encoded key
 	encodeKey := make([]byte, n+len(key))
+
+	// Copy the sequence number bytes to the encodeKey slice
 	copy(encodeKey[:n], seq[:n])
+
+	// Copy the original key bytes to the encodeKey slice starting from offset n
 	copy(encodeKey[n:], key)
 
 	return encodeKey
@@ -156,6 +171,10 @@ func encodeLogRecordKeyWithSeq(key []byte, seqNo uint64) []byte {
 // Parse the LogRecord key to get the actual key and transaction sequence number seq
 func parseLogRecordKeyAndSeq(key []byte) ([]byte, uint64) {
 	seqNo, n := binary.Uvarint(key)
+
+	// Extract the real key from the remaining bytes
 	realKey := key[n:]
+
 	return realKey, seqNo
 }
+

engine/batch_test.go

@@ -18,7 +18,7 @@ func TestDB_WriteBatch(t *testing.T) {
 	assert.Nil(t, err)
 	assert.NotNil(t, db)
 
-	// 写数据之后不提交
+	// Do not submit data after writing
 	wb := db.NewWriteBatch(config.DefaultWriteBatchOptions)
 	err = wb.Put(randkv.GetTestKey(1), randkv.RandomValue(10))
 	assert.Nil(t, err)
@@ -28,7 +28,7 @@ func TestDB_WriteBatch(t *testing.T) {
 	_, err = db.Get(randkv.GetTestKey(1))
 	assert.Equal(t, _const.ErrKeyNotFound, err)
 
-	// 正常提交数据
+	// Normal submission data
 	err = wb.Commit()
 	assert.Nil(t, err)
 
@@ -72,7 +72,7 @@ func TestDB_WriteBatchRestart(t *testing.T) {
 	err = wb.Commit()
 	assert.Nil(t, err)
 
-	// 重启
+	// Restart
 	err = db.Close()
 	assert.Nil(t, err)
 
@@ -82,7 +82,7 @@ func TestDB_WriteBatchRestart(t *testing.T) {
 	_, err = db2.Get(randkv.GetTestKey(1))
 	assert.Equal(t, _const.ErrKeyNotFound, err)
 
-	// 判断事务序列号
+	// Judgment transaction sequence number
 	assert.Equal(t, uint64(2), db.transSeqNo)
 }
 
@@ -94,7 +94,7 @@ func TestDB_WriteBatch1(t *testing.T) {
 	assert.Nil(t, err)
 	assert.NotNil(t, db)
 
-	// 批量提交中间手动停止
+	// Manual stop during batch submission
 	wbopt := config.DefaultWriteBatchOptions
 	wbopt.MaxBatchNum = 1000000
 	wb := db.NewWriteBatch(wbopt)

engine/data/data_file.go

@@ -14,49 +14,50 @@ const (
 	MergeFinaFileSuffix = "mergeFina"
 )
 
-// DataFile 数据文件
+// DataFile represents a data file.
 type DataFile struct {
-	FileID    uint32           //文件id
-	WriteOff  int64            //文件写到了哪个位置
-	IoManager fileio.IOManager //io 读写操作
+	FileID    uint32           // File ID
+	WriteOff  int64            // Position where the file is currently being written
+	IoManager fileio.IOManager // IO read/write operations
 }
 
-// OpenDataFile 打开新的数据文件
-func OpenDataFile(dirPath string, fildID uint32, fileSize int64, fioType int8) (*DataFile, error) {
-	fileName := GetDataFileName(dirPath, fildID)
-	return newDataFile(fileName, fildID, fileSize, fioType)
+// OpenDataFile opens a new data file.
+func OpenDataFile(dirPath string, fileID uint32, fileSize int64, fioType int8) (*DataFile, error) {
+	fileName := GetDataFileName(dirPath, fileID)
+	return newDataFile(fileName, fileID, fileSize, fioType)
 }
 
-func GetDataFileName(dirPath string, fildID uint32) string {
-	return filepath.Join(dirPath, fmt.Sprintf("%09d", fildID)+DataFileSuffix)
+// GetDataFileName returns the file name for a data file.
+func GetDataFileName(dirPath string, fileID uint32) string {
+	return filepath.Join(dirPath, fmt.Sprintf("%09d", fileID)+DataFileSuffix)
 }
 
-// OpenHintFile 打开 Hint 索引文件
+// OpenHintFile opens the hint index file.
 func OpenHintFile(dirPath string, fileSize int64, fioType int8) (*DataFile, error) {
 	fileName := filepath.Join(dirPath, HintFileSuffix)
 	return newDataFile(fileName, 0, fileSize, fioType)
 }
 
-// OpenMergeFinaFile 打开标识 merge 完成的文件
+// OpenMergeFinaFile opens the file that indicates merge completion.
 func OpenMergeFinaFile(dirPath string, fileSize int64, fioType int8) (*DataFile, error) {
 	fileName := filepath.Join(dirPath, MergeFinaFileSuffix)
 	return newDataFile(fileName, 0, fileSize, fioType)
 }
 
-func newDataFile(dirPath string, fildID uint32, fileSize int64, fioType int8) (*DataFile, error) {
-	//初始化 IOManager 管理器接口
+func newDataFile(dirPath string, fileID uint32, fileSize int64, fioType int8) (*DataFile, error) {
+	// Initialize the IOManager interface
 	ioManager, err := fileio.NewIOManager(dirPath, fileSize, fioType)
 	if err != nil {
 		return nil, err
 	}
 	return &DataFile{
-		FileID:    fildID,
+		FileID:    fileID,
 		WriteOff:  0,
 		IoManager: ioManager,
 	}, nil
 }
 
-// ReadLogRecord 根据 offset 从数据文件中读取 logRecord
+// ReadLogRecord reads a log record from the data file based on the offset.
 func (df *DataFile) ReadLogRecord(offset int64) (*LogRecord, int64, error) {
 	fileSize, err := df.IoManager.Size()
 	if err != nil {
@@ -68,39 +69,39 @@ func (df *DataFile) ReadLogRecord(offset int64) (*LogRecord, int64, error) {
 		headerBytes = fileSize - offset
 	}
 
-	// 读取 header 信息
+	// Read header information
 	headerBuf, err := df.readNBytes(headerBytes, offset)
 	if err != nil {
 		return nil, 0, err
 	}
 
 	header, headerSize := decodeLogRecordHeader(headerBuf)
-	// 下面俩个条件表示读到了文件末尾，直接返回 EOF
+	// The following conditions indicate reaching the end of the file, directly return EOF
 	if header == nil {
 		return nil, 0, io.EOF
 	}
 	if header.crc == 0 && header.keySize == 0 && header.valueSize == 0 {
 		return nil, 0, io.EOF
 	}
 
-	// 取出对应的 key 和 value 的长度
+	// Retrieve the lengths of the key and value
 	keySize, valueSize := int64(header.keySize), int64(header.valueSize)
 	var recordSize = headerSize + keySize + valueSize
 
 	logRecord := &LogRecord{Type: header.recordType}
 
-	// 读取用户实际存储的 key/value 数据
+	// Read the actual user-stored key/value data
 	if keySize > 0 || valueSize > 0 {
 		kvBuf, err := df.readNBytes(keySize+valueSize, headerSize+offset)
 		if err != nil {
 			return nil, 0, err
 		}
-		// 解码
+		// Decode
 		logRecord.Key = kvBuf[:keySize]
 		logRecord.Value = kvBuf[keySize:]
 	}
 
-	// 校验 crc （检查数据的有效性）
+	// Verify CRC (check data integrity)
 	crc := getLogRecordCRC(logRecord, headerBuf[crc32.Size:headerSize])
 	if crc != header.crc {
 		return nil, 0, ErrInvalidCRC
@@ -117,7 +118,7 @@ func (df *DataFile) Write(buf []byte) error {
 	return nil
 }
 
-// WriteHintRecord 写入索引信息到 hint 文件中
+// WriteHintRecord writes index information to the hint file.
 func (df *DataFile) WriteHintRecord(key []byte, pst *LogRecordPst) error {
 	record := &LogRecord{
 		Key:   key,

engine/data/data_file_test.go

@@ -61,7 +61,7 @@ func TestDataFile_ReadLogRecord(t *testing.T) {
 	assert.Nil(t, err)
 	assert.NotNil(t, dataFile)
 
-	// 只有一条 LogRecord
+	// There is only one LogRecord
 	record1 := &LogRecord{
 		Key:   []byte("name"),
 		Value: []byte("flydb"),
@@ -76,7 +76,7 @@ func TestDataFile_ReadLogRecord(t *testing.T) {
 	assert.Equal(t, size, readSize1)
 	assert.Equal(t, record1, readRec1)
 
-	// 多条 LogRecord 从不同位置读取
+	// Multiple logrecords are read from different locations
 	record2 := &LogRecord{
 		Key:   []byte("name"),
 		Value: []byte("bitcask-kv"),
@@ -89,7 +89,7 @@ func TestDataFile_ReadLogRecord(t *testing.T) {
 	assert.Equal(t, size2, readSize2)
 	assert.Equal(t, record2, readRec2)
 
-	// 被删除的数据在数据文件的末尾
+	// The deleted data is at the end of the data file
 	record3 := &LogRecord{
 		Key:   []byte("name"),
 		Value: []byte("delete-data"),