Convert Double Value to Byte value for storing in SQLite Database

Question

i want to save a byte[] in a table of SQLite Database (Field datatype is BLOB).

In that array i want to save a negative Double value say -36524.063415093093 in byte format (Not in byte[] format)

Double b = -36524.063415093093;

//  Signature is 
//  byte[] arr = new Byte[] {Double, byte, byte, byte};

When i write Convert.ToByte(b) it gives error & if i write (byte)b it returns 84.

but how to i convert 84 back to -36524.063415093093?

Please correct if my approach is wrong?

Answer 1

Please correct if my approach is wrong?

Your approach is wrong ;-)

You cannot store a double (which requires 64 bits) into a single byte (which is only 8 bits). It's impossible

You can get a byte array, which is just all the 64 bits spread across 8 bytes. This would let you store your double in a byte array. But you indicate that you don't want that.

Answer 2

From MSDN:

The double keyword denotes a simple type that stores 64-bit floating-point values

So, you will get the array of bytes when you will convert double to byte[]. Conversion itself will looks like:

byte[] array = BitConverter.GetBytes(double)

Answer 3

A double uses 8 bytes of memory so you will probably need to convert it into a byte array using the BitConverter class.

Answer 4

but how to i convert 84 back to -36524.063415093093?

You can't. The reason you're getting 84 is that the conversion from double to byte discards information:

1. Conversion to an integer discards the fractional part of the number, leaving -36524 (0xffff7154).
2. Truncation to byte discards all but the last 8 bits, leaving 0x54, or 84.

There's no way to reconstruct the original value. How would your decoding routine know that you meant -36524.063415093093 instead of 84.0 or 2132.9?

If you want to store a double in a database, you need to store all 64 bits of it, not just 8 bits. The easiest way to do this would be to use a REAL column instead of encoding your data as a BLOB.

If you have a reason to use a BLOB, then please explain what it is.

Answer 5

It's possible to convert a double to bytes and back with a helper struct with explicit layout.

[StructLayout(LayoutKind.Explicit)]
public struct DoubleStruct
{
    [FieldOffset(0)]
    public double value;

    [FieldOffset(0)]
    public byte byte0;

    [FieldOffset(1)]
    public byte byte1;

    [FieldOffset(2)]
    public byte byte2;

    [FieldOffset(3)]
    public byte byte3;

    [FieldOffset(4)]
    public byte byte4;

    [FieldOffset(5)]
    public byte byte5;

    [FieldOffset(6)]
    public byte byte6;

    [FieldOffset(7)]
    public byte byte7;
}

The example below produces 2 identical structs:

DoubleStruct ds = new DoubleStruct { value = -36524.063415093093 };

DoubleStruct ds1 = new DoubleStruct
{
    byte0 = 0xDD,
    byte1 = 0x16,
    byte2 = 0x7F,
    byte3 = 0x07,
    byte4 = 0x82,
    byte5 = 0xD5,
    byte6 = 0xE1,
    byte7 = 0xC0
};

To convert the helper struct to byte array, use:

byte[] bytes = new[] { ds1.byte0, ds1.byte1, ds1.byte2, ds1.byte3,
                      ds1.byte4, ds1.byte5, ds1.byte6, ds1.byte7 };