c#，关于Big Endian 和 Little Endian，以及转换类

Big Endian：最高字节在地址最低位，最低字节在地址最高位，依次排列。 
Little Endian：最低字节在最低位，最高字节在最高位，反序排列。

当在本地主机上，无需注意机器用的是Big Endian还是Little Endian。但是网络上都是用的是Big Endian，需要进行一个转换，但是c#提供的BitConverter默认使用的Little Endian，在需与网络通信时，反的字节序大有不便，特此提供一个可选Big Endian和Little Endian类方便转换。

为什么存在Little Endian？因为对于机器来说，Little Endian更有利于机器的运算。内存地址由低位到高位，在两个数相加运算，直接在高位添加进位数，不必移动内存地址

第一部分

using System;
using System.Runtime.InteropServices;

namespace MiscUtil.Conversion
{
    /// <summary>
    /// Equivalent of System.BitConverter, but with either endianness.
    /// </summary>
    public abstract class EndianBitConverter
    {
        #region Endianness of this converter
        /// <summary>
        /// Indicates the byte order ("endianess") in which data is converted using this class.
        /// </summary>
        /// <remarks>
        /// Different computer architectures store data using different byte orders. "Big-endian"
        /// means the most significant byte is on the left end of a word. "Little-endian" means the
        /// most significant byte is on the right end of a word.
        /// </remarks>
        /// <returns>true if this converter is little-endian, false otherwise.</returns>
        public abstract bool IsLittleEndian();

        /// <summary>
        /// Indicates the byte order ("endianess") in which data is converted using this class.
        /// </summary>
        public abstract Endianness Endianness { get; }
        #endregion

        #region Factory properties
        static LittleEndianBitConverter little = new LittleEndianBitConverter();
        /// <summary>
        /// Returns a little-endian bit converter instance. The same instance is
        /// always returned.
        /// </summary>
        public static LittleEndianBitConverter Little
        {
            get { return little; }
        }

        static BigEndianBitConverter big = new BigEndianBitConverter();
        /// <summary>
        /// Returns a big-endian bit converter instance. The same instance is
        /// always returned.
        /// </summary>
        public static BigEndianBitConverter Big
        {
            get { return big; }
        }
        #endregion

        #region Double/primitive conversions
        /// <summary>
        /// Converts the specified double-precision floating point number to a
        /// 64-bit signed integer. Note: the endianness of this converter does not
        /// affect the returned value.
        /// </summary>
        /// <param name="value">The number to convert. </param>
        /// <returns>A 64-bit signed integer whose value is equivalent to value.</returns>
        public long DoubleToInt64Bits(double value)
        {
            return BitConverter.DoubleToInt64Bits(value);
        }

        /// <summary>
        /// Converts the specified 64-bit signed integer to a double-precision
        /// floating point number. Note: the endianness of this converter does not
        /// affect the returned value.
        /// </summary>
        /// <param name="value">The number to convert. </param>
        /// <returns>A double-precision floating point number whose value is equivalent to value.</returns>
        public double Int64BitsToDouble (long value)
        {
            return BitConverter.Int64BitsToDouble(value);
        }

        /// <summary>
        /// Converts the specified single-precision floating point number to a
        /// 32-bit signed integer. Note: the endianness of this converter does not
        /// affect the returned value.
        /// </summary>
        /// <param name="value">The number to convert. </param>
        /// <returns>A 32-bit signed integer whose value is equivalent to value.</returns>
        public int SingleToInt32Bits(float value)
        {
            return new Int32SingleUnion(value).AsInt32;
        }

        /// <summary>
        /// Converts the specified 32-bit signed integer to a single-precision floating point
        /// number. Note: the endianness of this converter does not
        /// affect the returned value.
        /// </summary>
        /// <param name="value">The number to convert. </param>
        /// <returns>A single-precision floating point number whose value is equivalent to value.</returns>
        public float Int32BitsToSingle (int value)
        {
            return new Int32SingleUnion(value).AsSingle;
        }
        #endregion

        #region To(PrimitiveType) conversions
        /// <summary>
        /// Returns a Boolean value converted from one byte at a specified position in a byte array.
        /// </summary>
        /// <param name="value">An array of bytes.</param>
        /// <param name="startIndex">The starting position within value.</param>
        /// <returns>true if the byte at startIndex in value is nonzero; otherwise, false.</returns>
        public bool ToBoolean (byte[] value, int startIndex)
        {
            CheckByteArgument(value, startIndex, );
            return BitConverter.ToBoolean(value, startIndex);
        }

        /// <summary>
        /// Returns a Unicode character converted from two bytes at a specified position in a byte array.
        /// </summary>
        /// <param name="value">An array of bytes.</param>
        /// <param name="startIndex">The starting position within value.</param>
        /// <returns>A character formed by two bytes beginning at startIndex.</returns>
        public char ToChar (byte[] value, int startIndex)
        {
            return unchecked((char) (CheckedFromBytes(value, startIndex, )));
        }

        /// <summary>
        /// Returns a double-precision floating point number converted from eight bytes
        /// at a specified position in a byte array.
        /// </summary>
        /// <param name="value">An array of bytes.</param>
        /// <param name="startIndex">The starting position within value.</param>
        /// <returns>A double precision floating point number formed by eight bytes beginning at startIndex.</returns>
        public double ToDouble (byte[] value, int startIndex)
        {
            return Int64BitsToDouble(ToInt64(value, startIndex));
        }

        /// <summary>
        /// Returns a single-precision floating point number converted from four bytes
        /// at a specified position in a byte array.
        /// </summary>
        /// <param name="value">An array of bytes.</param>
        /// <param name="startIndex">The starting position within value.</param>
        /// <returns>A single precision floating point number formed by four bytes beginning at startIndex.</returns>
        public float ToSingle (byte[] value, int startIndex)
        {
            return Int32BitsToSingle(ToInt32(value, startIndex));
        }

        /// <summary>
        /// Returns a 16-bit signed integer converted from two bytes at a specified position in a byte array.
        /// </summary>
        /// <param name="value">An array of bytes.</param>
        /// <param name="startIndex">The starting position within value.</param>
        /// <returns>A 16-bit signed integer formed by two bytes beginning at startIndex.</returns>
        public short ToInt16 (byte[] value, int startIndex)
        {
            return unchecked((short) (CheckedFromBytes(value, startIndex, )));
        }

        /// <summary>
        /// Returns a 32-bit signed integer converted from four bytes at a specified position in a byte array.
        /// </summary>
        /// <param name="value">An array of bytes.</param>
        /// <param name="startIndex">The starting position within value.</param>
        /// <returns>A 32-bit signed integer formed by four bytes beginning at startIndex.</returns>
        public int ToInt32 (byte[] value, int startIndex)
        {
            return unchecked((int) (CheckedFromBytes(value, startIndex, )));
        }

        /// <summary>
        /// Returns a 64-bit signed integer converted from eight bytes at a specified position in a byte array.
        /// </summary>
        /// <param name="value">An array of bytes.</param>
        /// <param name="startIndex">The starting position within value.</param>
        /// <returns>A 64-bit signed integer formed by eight bytes beginning at startIndex.</returns>
        public long ToInt64 (byte[] value, int startIndex)
        {
            return CheckedFromBytes(value, startIndex, );
        }

        /// <summary>
        /// Returns a 16-bit unsigned integer converted from two bytes at a specified position in a byte array.
        /// </summary>
        /// <param name="value">An array of bytes.</param>
        /// <param name="startIndex">The starting position within value.</param>
        /// <returns>A 16-bit unsigned integer formed by two bytes beginning at startIndex.</returns>
        public ushort ToUInt16 (byte[] value, int startIndex)
        {
            return unchecked((ushort) (CheckedFromBytes(value, startIndex, )));
        }

        /// <summary>
        /// Returns a 32-bit unsigned integer converted from four bytes at a specified position in a byte array.
        /// </summary>
        /// <param name="value">An array of bytes.</param>
        /// <param name="startIndex">The starting position within value.</param>
        /// <returns>A 32-bit unsigned integer formed by four bytes beginning at startIndex.</returns>
        public uint ToUInt32 (byte[] value, int startIndex)
        {
            return unchecked((uint) (CheckedFromBytes(value, startIndex, )));
        }

        /// <summary>
        /// Returns a 64-bit unsigned integer converted from eight bytes at a specified position in a byte array.
        /// </summary>
        /// <param name="value">An array of bytes.</param>
        /// <param name="startIndex">The starting position within value.</param>
        /// <returns>A 64-bit unsigned integer formed by eight bytes beginning at startIndex.</returns>
        public ulong ToUInt64 (byte[] value, int startIndex)
        {
            return unchecked((ulong) (CheckedFromBytes(value, startIndex, )));
        }

        /// <summary>
        /// Checks the given argument for validity.
        /// </summary>
        /// <param name="value">The byte array passed in</param>
        /// <param name="startIndex">The start index passed in</param>
        /// <param name="bytesRequired">The number of bytes required</param>
        /// <exception cref="ArgumentNullException">value is a null reference</exception>
        /// <exception cref="ArgumentOutOfRangeException">
        /// startIndex is less than zero or greater than the length of value minus bytesRequired.
        /// </exception>
        static void CheckByteArgument(byte[] value, int startIndex, int bytesRequired)
        {
            if (value==null)
            {
                throw new ArgumentNullException("value");
            }
            if (startIndex <  || startIndex > value.Length-bytesRequired)
            {
                throw new ArgumentOutOfRangeException("startIndex");
            }
        }

        /// <summary>
        /// Checks the arguments for validity before calling FromBytes
        /// (which can therefore assume the arguments are valid).
        /// </summary>
        /// <param name="value">The bytes to convert after checking</param>
        /// <param name="startIndex">The index of the first byte to convert</param>
        /// <param name="bytesToConvert">The number of bytes to convert</param>
        /// <returns></returns>
        long CheckedFromBytes(byte[] value, int startIndex, int bytesToConvert)
        {
            CheckByteArgument(value, startIndex, bytesToConvert);
            return FromBytes(value, startIndex, bytesToConvert);
        }

        /// <summary>
        /// Convert the given number of bytes from the given array, from the given start
        /// position, into a long, using the bytes as the least significant part of the long.
        /// By the time this is called, the arguments have been checked for validity.
        /// </summary>
        /// <param name="value">The bytes to convert</param>
        /// <param name="startIndex">The index of the first byte to convert</param>
        /// <param name="bytesToConvert">The number of bytes to use in the conversion</param>
        /// <returns>The converted number</returns>
        protected abstract long FromBytes(byte[] value, int startIndex, int bytesToConvert);
        #endregion

        #region ToString conversions
        /// <summary>
        /// Returns a String converted from the elements of a byte array.
        /// </summary>
        /// <param name="value">An array of bytes.</param>
        /// <remarks>All the elements of value are converted.</remarks>
        /// <returns>
        /// A String of hexadecimal pairs separated by hyphens, where each pair
        /// represents the corresponding element in value; for example, "7F-2C-4A".
        /// </returns>
        public static string ToString(byte[] value)
        {
            return BitConverter.ToString(value);
        }

        /// <summary>
        /// Returns a String converted from the elements of a byte array starting at a specified array position.
        /// </summary>
        /// <param name="value">An array of bytes.</param>
        /// <param name="startIndex">The starting position within value.</param>
        /// <remarks>The elements from array position startIndex to the end of the array are converted.</remarks>
        /// <returns>
        /// A String of hexadecimal pairs separated by hyphens, where each pair
        /// represents the corresponding element in value; for example, "7F-2C-4A".
        /// </returns>
        public static string ToString(byte[] value, int startIndex)
        {
            return BitConverter.ToString(value, startIndex);
        }

        /// <summary>
        /// Returns a String converted from a specified number of bytes at a specified position in a byte array.
        /// </summary>
        /// <param name="value">An array of bytes.</param>
        /// <param name="startIndex">The starting position within value.</param>
        /// <param name="length">The number of bytes to convert.</param>
        /// <remarks>The length elements from array position startIndex are converted.</remarks>
        /// <returns>
        /// A String of hexadecimal pairs separated by hyphens, where each pair
        /// represents the corresponding element in value; for example, "7F-2C-4A".
        /// </returns>
        public static string ToString(byte[] value, int startIndex, int length)
        {
            return BitConverter.ToString(value, startIndex, length);
        }
        #endregion

        #region    Decimal conversions
        /// <summary>
        /// Returns a decimal value converted from sixteen bytes
        /// at a specified position in a byte array.
        /// </summary>
        /// <param name="value">An array of bytes.</param>
        /// <param name="startIndex">The starting position within value.</param>
        /// <returns>A decimal  formed by sixteen bytes beginning at startIndex.</returns>
        public decimal ToDecimal (byte[] value, int startIndex)
        {
            // HACK: This always assumes four parts, each in their own endianness,
            // starting with the first part at the start of the byte array.
            // On the other hand, there's no real format specified...
            int[] parts = new int[];
            for (int i=; i < ; i++)
            {
                parts[i] = ToInt32(value, startIndex+i*);
            }
            return new Decimal(parts);
        }

        /// <summary>
        /// Returns the specified decimal value as an array of bytes.
        /// </summary>
        /// <param name="value">The number to convert.</param>
        /// <returns>An array of bytes with length 16.</returns>
        public byte[] GetBytes(decimal value)
        {
            byte[] bytes = new byte[];
            int[] parts = decimal.GetBits(value);
            for (int i=; i < ; i++)
            {
                CopyBytesImpl(parts[i], , bytes, i*);
            }
            return bytes;
        }

        /// <summary>
        /// Copies the specified decimal value into the specified byte array,
        /// beginning at the specified index.
        /// </summary>
        /// <param name="value">A character to convert.</param>
        /// <param name="buffer">The byte array to copy the bytes into</param>
        /// <param name="index">The first index into the array to copy the bytes into</param>
        public void CopyBytes(decimal value, byte[] buffer, int index)
        {
            int[] parts = decimal.GetBits(value);
            for (int i=; i < ; i++)
            {
                CopyBytesImpl(parts[i], , buffer, i*+index);
            }
        }
        #endregion

        #region GetBytes conversions
        /// <summary>
        /// Returns an array with the given number of bytes formed
        /// from the least significant bytes of the specified value.
        /// This is used to implement the other GetBytes methods.
        /// </summary>
        /// <param name="value">The value to get bytes for</param>
        /// <param name="bytes">The number of significant bytes to return</param>
        byte[] GetBytes(long value, int bytes)
        {
            byte[] buffer = new byte[bytes];
            CopyBytes(value, bytes, buffer, );
            return buffer;
        }

        /// <summary>
        /// Returns the specified Boolean value as an array of bytes.
        /// </summary>
        /// <param name="value">A Boolean value.</param>
        /// <returns>An array of bytes with length 1.</returns>
        public byte[] GetBytes(bool value)
        {
            return BitConverter.GetBytes(value);
        }

        /// <summary>
        /// Returns the specified Unicode character value as an array of bytes.
        /// </summary>
        /// <param name="value">A character to convert.</param>
        /// <returns>An array of bytes with length 2.</returns>
        public byte[] GetBytes(char value)
        {
            return GetBytes(value, );
        }

        /// <summary>
        /// Returns the specified double-precision floating point value as an array of bytes.
        /// </summary>
        /// <param name="value">The number to convert.</param>
        /// <returns>An array of bytes with length 8.</returns>
        public byte[] GetBytes(double value)
        {
            return GetBytes(DoubleToInt64Bits(value), );
        }

        /// <summary>
        /// Returns the specified 16-bit signed integer value as an array of bytes.
        /// </summary>
        /// <param name="value">The number to convert.</param>
        /// <returns>An array of bytes with length 2.</returns>
        public byte[] GetBytes(short value)
        {
            return GetBytes(value, );
        }

        /// <summary>
        /// Returns the specified 32-bit signed integer value as an array of bytes.
        /// </summary>
        /// <param name="value">The number to convert.</param>
        /// <returns>An array of bytes with length 4.</returns>
        public byte[] GetBytes(int value)
        {
            return GetBytes(value, );
        }

        /// <summary>
        /// Returns the specified 64-bit signed integer value as an array of bytes.
        /// </summary>
        /// <param name="value">The number to convert.</param>
        /// <returns>An array of bytes with length 8.</returns>
        public byte[] GetBytes(long value)
        {
            return GetBytes(value, );
        }

        /// <summary>
        /// Returns the specified single-precision floating point value as an array of bytes.
        /// </summary>
        /// <param name="value">The number to convert.</param>
        /// <returns>An array of bytes with length 4.</returns>
        public byte[] GetBytes(float value)
        {
            return GetBytes(SingleToInt32Bits(value), );
        }

        /// <summary>
        /// Returns the specified 16-bit unsigned integer value as an array of bytes.
        /// </summary>
        /// <param name="value">The number to convert.</param>
        /// <returns>An array of bytes with length 2.</returns>
        public byte[] GetBytes(ushort value)
        {
            return GetBytes(value, );
        }

        /// <summary>
        /// Returns the specified 32-bit unsigned integer value as an array of bytes.
        /// </summary>
        /// <param name="value">The number to convert.</param>
        /// <returns>An array of bytes with length 4.</returns>
        public byte[] GetBytes(uint value)
        {
            return GetBytes(value, );
        }

        /// <summary>
        /// Returns the specified 64-bit unsigned integer value as an array of bytes.
        /// </summary>
        /// <param name="value">The number to convert.</param>
        /// <returns>An array of bytes with length 8.</returns>
        public byte[] GetBytes(ulong value)
        {
            return GetBytes(unchecked((long)value), );
        }

        #endregion

        #region CopyBytes conversions
        /// <summary>
        /// Copies the given number of bytes from the least-specific
        /// end of the specified value into the specified byte array, beginning
        /// at the specified index.
        /// This is used to implement the other CopyBytes methods.
        /// </summary>
        /// <param name="value">The value to copy bytes for</param>
        /// <param name="bytes">The number of significant bytes to copy</param>
        /// <param name="buffer">The byte array to copy the bytes into</param>
        /// <param name="index">The first index into the array to copy the bytes into</param>
        void CopyBytes(long value, int bytes, byte[] buffer, int index)
        {
            if (buffer==null)
            {
                throw new ArgumentNullException("buffer", "Byte array must not be null");
            }
            if (buffer.Length < index+bytes)
            {
                throw new ArgumentOutOfRangeException("Buffer not big enough for value");
            }
            CopyBytesImpl(value, bytes, buffer, index);
        }

        /// <summary>
        /// Copies the given number of bytes from the least-specific
        /// end of the specified value into the specified byte array, beginning
        /// at the specified index.
        /// This must be implemented in concrete derived classes, but the implementation
        /// may assume that the value will fit into the buffer.
        /// </summary>
        /// <param name="value">The value to copy bytes for</param>
        /// <param name="bytes">The number of significant bytes to copy</param>
        /// <param name="buffer">The byte array to copy the bytes into</param>
        /// <param name="index">The first index into the array to copy the bytes into</param>
        protected abstract void CopyBytesImpl(long value, int bytes, byte[] buffer, int index);

        /// <summary>
        /// Copies the specified Boolean value into the specified byte array,
        /// beginning at the specified index.
        /// </summary>
        /// <param name="value">A Boolean value.</param>
        /// <param name="buffer">The byte array to copy the bytes into</param>
        /// <param name="index">The first index into the array to copy the bytes into</param>
        public void CopyBytes(bool value, byte[] buffer, int index)
        {
            CopyBytes(value ?  : , , buffer, index);
        }

        /// <summary>
        /// Copies the specified Unicode character value into the specified byte array,
        /// beginning at the specified index.
        /// </summary>
        /// <param name="value">A character to convert.</param>
        /// <param name="buffer">The byte array to copy the bytes into</param>
        /// <param name="index">The first index into the array to copy the bytes into</param>
        public void CopyBytes(char value, byte[] buffer, int index)
        {
            CopyBytes(value, , buffer, index);
        }

        /// <summary>
        /// Copies the specified double-precision floating point value into the specified byte array,
        /// beginning at the specified index.
        /// </summary>
        /// <param name="value">The number to convert.</param>
        /// <param name="buffer">The byte array to copy the bytes into</param>
        /// <param name="index">The first index into the array to copy the bytes into</param>
        public void CopyBytes(double value, byte[] buffer, int index)
        {
            CopyBytes(DoubleToInt64Bits(value), , buffer, index);
        }

        /// <summary>
        /// Copies the specified 16-bit signed integer value into the specified byte array,
        /// beginning at the specified index.
        /// </summary>
        /// <param name="value">The number to convert.</param>
        /// <param name="buffer">The byte array to copy the bytes into</param>
        /// <param name="index">The first index into the array to copy the bytes into</param>
        public void CopyBytes(short value, byte[] buffer, int index)
        {
            CopyBytes(value, , buffer, index);
        }

        /// <summary>
        /// Copies the specified 32-bit signed integer value into the specified byte array,
        /// beginning at the specified index.
        /// </summary>
        /// <param name="value">The number to convert.</param>
        /// <param name="buffer">The byte array to copy the bytes into</param>
        /// <param name="index">The first index into the array to copy the bytes into</param>
        public void CopyBytes(int value, byte[] buffer, int index)
        {
            CopyBytes(value, , buffer, index);
        }

        /// <summary>
        /// Copies the specified 64-bit signed integer value into the specified byte array,
        /// beginning at the specified index.
        /// </summary>
        /// <param name="value">The number to convert.</param>
        /// <param name="buffer">The byte array to copy the bytes into</param>
        /// <param name="index">The first index into the array to copy the bytes into</param>
        public void CopyBytes(long value, byte[] buffer, int index)
        {
            CopyBytes(value, , buffer, index);
        }

        /// <summary>
        /// Copies the specified single-precision floating point value into the specified byte array,
        /// beginning at the specified index.
        /// </summary>
        /// <param name="value">The number to convert.</param>
        /// <param name="buffer">The byte array to copy the bytes into</param>
        /// <param name="index">The first index into the array to copy the bytes into</param>
        public void CopyBytes(float value, byte[] buffer, int index)
        {
            CopyBytes(SingleToInt32Bits(value), , buffer, index);
        }

        /// <summary>
        /// Copies the specified 16-bit unsigned integer value into the specified byte array,
        /// beginning at the specified index.
        /// </summary>
        /// <param name="value">The number to convert.</param>
        /// <param name="buffer">The byte array to copy the bytes into</param>
        /// <param name="index">The first index into the array to copy the bytes into</param>
        public void CopyBytes(ushort value, byte[] buffer, int index)
        {
            CopyBytes(value, , buffer, index);
        }

        /// <summary>
        /// Copies the specified 32-bit unsigned integer value into the specified byte array,
        /// beginning at the specified index.
        /// </summary>
        /// <param name="value">The number to convert.</param>
        /// <param name="buffer">The byte array to copy the bytes into</param>
        /// <param name="index">The first index into the array to copy the bytes into</param>
        public void CopyBytes(uint value, byte[] buffer, int index)
        {
            CopyBytes(value, , buffer, index);
        }

        /// <summary>
        /// Copies the specified 64-bit unsigned integer value into the specified byte array,
        /// beginning at the specified index.
        /// </summary>
        /// <param name="value">The number to convert.</param>
        /// <param name="buffer">The byte array to copy the bytes into</param>
        /// <param name="index">The first index into the array to copy the bytes into</param>
        public void CopyBytes(ulong value, byte[] buffer, int index)
        {
            CopyBytes(unchecked((long)value), , buffer, index);
        }

        #endregion

        #region Private struct used for Single/Int32 conversions
        /// <summary>
        /// Union used solely for the equivalent of DoubleToInt64Bits and vice versa.
        /// </summary>
        [StructLayout(LayoutKind.Explicit)]
            struct Int32SingleUnion
        {
            /// <summary>
            /// Int32 version of the value.
            /// </summary>
            [FieldOffset()]
            int i;
            /// <summary>
            /// Single version of the value.
            /// </summary>
            [FieldOffset()]
            float f;

            /// <summary>
            /// Creates an instance representing the given integer.
            /// </summary>
            /// <param name="i">The integer value of the new instance.</param>
            internal Int32SingleUnion(int i)
            {
                this.f = ; // Just to keep the compiler happy
                this.i = i;
            }

            /// <summary>
            /// Creates an instance representing the given floating point number.
            /// </summary>
            /// <param name="f">The floating point value of the new instance.</param>
            internal Int32SingleUnion(float f)
            {
                this.i = ; // Just to keep the compiler happy
                this.f = f;
            }

            /// <summary>
            /// Returns the value of the instance as an integer.
            /// </summary>
            internal int AsInt32
            {
                get { return i; }
            }

            /// <summary>
            /// Returns the value of the instance as a floating point number.
            /// </summary>
            internal float AsSingle
            {
                get { return f; }
            }
        }
        #endregion
    }
}

EndianBitConverter

第二部分

namespace MiscUtil.Conversion
{
    /// <summary>
    /// Implementation of EndianBitConverter which converts to/from big-endian
    /// byte arrays.
    /// </summary>
    public sealed class BigEndianBitConverter : EndianBitConverter
    {
        /// <summary>
        /// Indicates the byte order ("endianess") in which data is converted using this class.
        /// </summary>
        /// <remarks>
        /// Different computer architectures store data using different byte orders. "Big-endian"
        /// means the most significant byte is on the left end of a word. "Little-endian" means the
        /// most significant byte is on the right end of a word.
        /// </remarks>
        /// <returns>true if this converter is little-endian, false otherwise.</returns>
        public sealed override bool IsLittleEndian()
        {
            return false;
        }

        /// <summary>
        /// Indicates the byte order ("endianess") in which data is converted using this class.
        /// </summary>
        public sealed override Endianness Endianness
        {
            get { return Endianness.BigEndian; }
        }

        /// <summary>
        /// Copies the specified number of bytes from value to buffer, starting at index.
        /// </summary>
        /// <param name="value">The value to copy</param>
        /// <param name="bytes">The number of bytes to copy</param>
        /// <param name="buffer">The buffer to copy the bytes into</param>
        /// <param name="index">The index to start at</param>
        protected override void CopyBytesImpl(long value, int bytes, byte[] buffer, int index)
        {
            int endOffset = index+bytes-;
            for (int i=; i < bytes; i++)
            {
                buffer[endOffset-i] = unchecked((byte)(value&0xff));
                value = value >> ;
            }
        }

        /// <summary>
        /// Returns a value built from the specified number of bytes from the given buffer,
        /// starting at index.
        /// </summary>
        /// <param name="buffer">The data in byte array format</param>
        /// <param name="startIndex">The first index to use</param>
        /// <param name="bytesToConvert">The number of bytes to use</param>
        /// <returns>The value built from the given bytes</returns>
        protected override long FromBytes(byte[] buffer, int startIndex, int bytesToConvert)
        {
            long ret = ;
            for (int i=; i < bytesToConvert; i++)
            {
                ret = unchecked((ret << ) | buffer[startIndex+i]);
            }
            return ret;
        }
    }
}

BigEndianBitConverter

第三部分

namespace MiscUtil.Conversion
{
    /// <summary>
    /// Endianness of a converter
    /// </summary>
    public enum Endianness
    {
        /// <summary>
        /// Little endian - least significant byte first
        /// </summary>
        LittleEndian,
        /// <summary>
        /// Big endian - most significant byte first
        /// </summary>
        BigEndian
    }
}

Endianness

第四部分

namespace MiscUtil.Conversion
{
    /// <summary>
    /// Implementation of EndianBitConverter which converts to/from little-endian
    /// byte arrays.
    /// </summary>
    public sealed class LittleEndianBitConverter : EndianBitConverter
    {
        /// <summary>
        /// Indicates the byte order ("endianess") in which data is converted using this class.
        /// </summary>
        /// <remarks>
        /// Different computer architectures store data using different byte orders. "Big-endian"
        /// means the most significant byte is on the left end of a word. "Little-endian" means the
        /// most significant byte is on the right end of a word.
        /// </remarks>
        /// <returns>true if this converter is little-endian, false otherwise.</returns>
        public sealed override bool IsLittleEndian()
        {
            return true;
        }

        /// <summary>
        /// Indicates the byte order ("endianess") in which data is converted using this class.
        /// </summary>
        public sealed override Endianness Endianness
        {
            get { return Endianness.LittleEndian; }
        }

        /// <summary>
        /// Copies the specified number of bytes from value to buffer, starting at index.
        /// </summary>
        /// <param name="value">The value to copy</param>
        /// <param name="bytes">The number of bytes to copy</param>
        /// <param name="buffer">The buffer to copy the bytes into</param>
        /// <param name="index">The index to start at</param>
        protected override void CopyBytesImpl(long value, int bytes, byte[] buffer, int index)
        {
            for (int i=; i < bytes; i++)
            {
                buffer[i+index] = unchecked((byte)(value&0xff));
                value = value >> ;
            }
        }

        /// <summary>
        /// Returns a value built from the specified number of bytes from the given buffer,
        /// starting at index.
        /// </summary>
        /// <param name="buffer">The data in byte array format</param>
        /// <param name="startIndex">The first index to use</param>
        /// <param name="bytesToConvert">The number of bytes to use</param>
        /// <returns>The value built from the given bytes</returns>
        protected override long FromBytes(byte[] buffer, int startIndex, int bytesToConvert)
        {
            long ret = ;
            for (int i=; i < bytesToConvert; i++)
            {
                ret = unchecked((ret << ) | buffer[startIndex+bytesToConvert--i]);
            }
            return ret;
        }
    }
}

LittleEndianBitConverter