README.md

bytex

go内置的bytes.Buffer功能很丰富，但它更多面向的是bytes这样的数据类型，bytex封装了bytes.Buffer，丰富了更多的基本类型操作。

在参考了MT协议情况下，实现了序列化io接口IBufferX。

从io的角度来说，bytes缓冲区分为read io与write io。

//创建一个可读的缓冲区，data是可读的数据，大部分情况来自网络数据，RPC数据，数据库或文件数据
func NewReadableBufferX(data []byte) *BufferX {
    var buffer = bytes.NewBuffer(data)
    var bufferX = &BufferX{buffer: buffer}
    return bufferX
}

//创建一个可写的缓冲区大小为1K，如果写入内容超过1K，此缓冲区会自动增长
func NewBufferX() *BufferX {
    var data = make([]byte, defaultByteBuff)
    var buffer = bytes.NewBuffer(data)
    buffer.Reset()
    var bufferX = &BufferX{buffer: buffer}
    return bufferX
}

//创建一个可写的缓冲区，大小为传入的size，如果写入内容超过此大小，此缓冲区会自动增长
//从效率上来说，传入的size应该与要写入的数据大小匹配，这样不会浪费内存，也不会因为
//在缓冲区不够时新开内存影响效率。
func NewSizedBufferX(size int) *BufferX {
    var data = make([]byte, size)
    var buffer = bytes.NewBuffer(data)
    buffer.Reset()
    var bufferX = &BufferX{buffer: buffer}
    return bufferX
}

//IBufferX buffer interface
type IBufferX interface {
    //Len : length
    Len() int
    //Read specific p, if length is not enough, return error
    Read(p []byte) error
    //ReadN n bytes, if length is not enough, return error
    ReadN(n int) ([]byte, error)
    //Write to buffer
    Write(p []byte)
    //Bytes left bytes
    Bytes() []byte
    //Reset reset cursor
    Reset()

    //ReWrite a buffer
    //重写缓冲区指定位置的数据，最常用的case是有些数据头部可能需要记录整段数据大小。
    //但最开始写入时，大小并不确定，可以先用0填充头部，在写完整段内容有了大小后在重写头部位置的数据。
    ReWrite(pos int, p []byte)
    //ReWriteU32 rewrite with a specific pos
    ReWriteU32(pos int, v uint32)

    //ReadU8 read byte
    ReadU8() (byte, error)
    //WriteU8 write byte
    WriteU8(byte)

    //ReadBool read bool
    ReadBool() (bool, error)
    //WriteBool write bool
    WriteBool(bool)

    //ReadLimitString read string
    ReadLimitString(limit uint32) (string, error)
    //WriteLimitString write string
    WriteLimitString(limit uint32, val string) error

    //ReadString read string
    ReadString() (string, error)
    //WriteString write string
    WriteString(val string)

    //ReadU16 read uint16
    ReadU16() (uint16, error)
    //WriteU16 write uint16
    WriteU16(uint16)

    //ReadI16 read int16
    ReadI16() (int16, error)
    //WriteI16 write int16
    WriteI16(int16)

    //ReadU32 read uint32
    ReadU32() (uint32, error)
    //WriteU32 write uint32
    WriteU32(uint32)

    //ReadI32 read int32
    ReadI32() (int32, error)
    //WriteI32 write int32
    WriteI32(int32)

    //ReadU64 read uint64
    ReadU64() (uint64, error)
    //WriteU64 write uint64
    WriteU64(uint64)

    //ReadI64 read int64
    ReadI64() (int64, error)
    //WriteI64 write int64
    WriteI64(int64)

    //ReadF64 read float64
    ReadF64() (float64, error)
    //WriteF64 write float64
    WriteF64(float64)
}