mysql: decimal转换为二进制

转自mysql代码注释：decimal.c

------------------------------------------------------------------------------------------------------------

*
  Convert decimal to its binary fixed-length representation
  two representations of the same length can be compared with memcmp
  with the correct -1/0/+1 result


  SYNOPSIS
    decimal2bin()
      from    - value to convert
      to      - points to buffer where string representation should be stored
      precision/scale - see decimal_bin_size() below


  NOTE
    the buffer is assumed to be of the size decimal_bin_size(precision, scale)


  RETURN VALUE
    E_DEC_OK/E_DEC_TRUNCATED/E_DEC_OVERFLOW


  DESCRIPTION
    for storage decimal numbers are converted to the "binary" format.


    This format has the following properties:
      1. length of the binary representation depends on the {precision, scale}
      as provided by the caller and NOT on the intg/frac of the decimal to
      convert.
      2. binary representations of the same {precision, scale} can be compared
      with memcmp - with the same result as decimal_cmp() of the original
      decimals (not taking into account possible precision loss during
      conversion).


    This binary format is as follows:
      1. First the number is converted to have a requested precision and scale.
      2. Every full DIG_PER_DEC1 digits of intg part are stored in 4 bytes
         as is
      3. The first intg % DIG_PER_DEC1 digits are stored in the reduced
         number of bytes (enough bytes to store this number of digits -
         see dig2bytes)
      4. same for frac - full decimal_digit_t's are stored as is,
         the last frac % DIG_PER_DEC1 digits - in the reduced number of bytes.
      5. If the number is negative - every byte is inversed.
      5. The very first bit of the resulting byte array is inverted (because
         memcmp compares unsigned bytes, see property 2 above)


    Example:


      1234567890.1234


    internally is represented as 3 decimal_digit_t's


      1 234567890 123400000


    (assuming we want a binary representation with precision=14, scale=4)
    in hex it's


      00-00-00-01  0D-FB-38-D2  07-5A-EF-40


    now, middle decimal_digit_t is full - it stores 9 decimal digits. It goes
    into binary representation as is:




      ...........  0D-FB-38-D2 ............


    First decimal_digit_t has only one decimal digit. We can store one digit in
    one byte, no need to waste four:


                01 0D-FB-38-D2 ............


    now, last digit. It's 123400000. We can store 1234 in two bytes:


                01 0D-FB-38-D2 04-D2


    So, we've packed 12 bytes number in 7 bytes.
    And now we invert the highest bit to get the final result:


                81 0D FB 38 D2 04 D2


    And for -1234567890.1234 it would be


                7E F2 04 37 2D FB 2D