File      : lib_crc.txt
   Date      : 2008-04-20
   Author    : Lammert Bies
   E-mail    : info@lammertbies.nl
   Version   : 1.16

   Contents  : Documentation for the CRC functions in lib_crc.c

   Support   : http://www.lammertbies.nl/forum/viewforum.php?f=11



   The file lib_crc.c contains source code for functions to calculate
   five commonly used CRC values: CRC-16, CRC-32, CRC-DNP, CRC-SICK,
   CRC-Kermit and CRC-CCITT. The functions can be freely used.

   To calculate a CRC, the following three steps must be followed:

   1. Initialize the CRC value. For CRC-16, CRC-SICK CRC-Kermit and CRC-DNP
      the initial value of the CRC is 0. For CRC-CCITT and CRC-MODBUS,
      the value 0xffff is used. CRC-32 starts with an initial value
      of 0xffffffffL.

   2. For each byte of the data starting with the first byte, call the
      function update_crc_16(), update_crc_32(), update_crc_dnp(),
      update_crc_sick(), update_crc_kermit() or update_crc_ccitt()
      to recalculate the value of the CRC.

   3. Only for CRC-32: When all bytes have been processed, take the
      one's complement of the obtained CRC value.

   4. Only for CRC-DNP: After all input processing, the one's complement
      of the CRC is calcluated and the two bytes of the CRC are swapped.

   5. Only for CRC-Kermit and CRC-SICK: After all input processing, the
      one's complement of the CRC is calcluated and the two bytes of the CRC
      are swapped.



   An example of this calculation process can be found in the tst_crc.c
   sample program. The program and other CRC implementations can be
   tested with the test string "123456789" without the quotes. The
   results should be:

   CRC16        : BB3D
   CRC16 Modbus : 4B37
   CRC16 SICK   : 56A6
   CRC-CCITT    : 0x31C3 (starting value 0x0000)
   CRC-CCITT    : 0x29B1 (starting value 0xFFFF)
   CRC-CCITT    : 0xE5CC (starting value 0x1D0F)
   CRC-Kermit   : 0x8921
   CRC-DNP      : 82EA
   CRC32        : CBF43926



   The example program tst_crc.exe can be invoked in three ways:

   tst_crc -a

      The program will prompt for an input string. All characters in the
      input string are used for the CRC calculation, based on their ASCII
      value.

      Example input string: ABC
         CRC16              = 0x4521
         CRC16 (Modbus)     = 0x8550
         CRC16 (Sick)       = 0xC3C1
         CRC-CCITT (0x0000) = 0x3994
         CRC-CCITT (0xffff) = 0xF508
         CRC-CCITT (0x1d0f) = 0x2898
         CRC-CCITT (Kermit) = 0xE359
         CRC-DNP            = 0x5AD3
         CRC32              = 0xA3830348

   tst_crc -x

      The program will prompt for an input string. All characters will
      be filtered out, except for 0..9, a..f and A..F. The remaining characters
      will be paired, and every pair of two characters represent the hexadecimal
      value to be used for one byte in the CRC calculation. The result if an
      od number of valud characters is provided is undefined.

      Example input string: 41 42 43
         CRC16              = 0x4521
         CRC16 (Modbus)     = 0x8550
         CRC16 (Sick)       = 0xC3C1
         CRC-CCITT (0x0000) = 0x3994
         CRC-CCITT (0xffff) = 0xF508
         CRC-CCITT (0x1d0f) = 0x2898
         CRC-CCITT (Kermit) = 0xE359
         CRC-DNP            = 0x5AD3
         CRC32              = 0xA3830348

      You see, that the result is the same as for the ASCII input "ABC". This
      is, because A, B and C are represented in ASCII by the hexadecimal
      values 41, 42 and 43. So it is obvious that the result should be
      the same in both cases.

   tst_crc file1 file2 ...

      If neither the -a, nor the -x parameter is used, the test program
      assumes that the parameters are file names. Each file is opened and
      the CRC values are calculated.



   The newest version of these files can be found at:

        http://www.lammertbies.nl/download/lib_crc.zip

   On-line CRC calculations of strings can be performed at:

        http://www.lammertbies.nl/comm/info/crc-calculation.html

   Support for the CRC routines in this library in the Error Detection and
   Correction forum at

        http://www.lammertbies.nl/forum/viewforum.php?f=11