kcode(1)                         3/11/92                           kcode(1)

NAME
     kcode - kanji code detector

SYNOPSIS
     kcode [-q] [-v] [-l limit] [file...]

DESCRIPTION
     kcode examines the named files (stdin if none) and reports their kanji
     encoding.  The  command  recognizes  EUC,  Shift-JIS, JIS, Old JIS and
     NEC-JIS encoded text.

     If no input files or a single input file are specified,  kcode  prints
     one of the following format identifiers on stdout:

          jis       - (New) JIS
          ojis      - Old JIS
          njis      - NEC-JIS
          sjis      - Shift-JIS
          euc       - Extended UNIX Code
          ascii     - only 7-bit ASCII characters present
          unknown   - unknown encoding

     In addition,  the  message  "ambiguous input format: EUC or Shift-JIS"
     indicates that the input file contents may be either EUC or Shift-JIS,
     and could not be distinguished further. This most commonly happens  if
     an input file contains only half-size katakana.

     If two or more input files are specified, kcode prints the encoding of
     each  input  file  on a separate line, preceded by the file name and a
     colon.

OPTIONS
     -l limit
          Limit the amount of text read from each input  file  to  at  most
          limit  bytes.  This  is useful if a file is very large, and it is
          not  known  in  advance  whether  the  file  contains  any  kanji
          characters.  This  option may not be used in conjunction with the
          -v option.

     -q   Report the input encoding via the exit status, instead of writing
          a message to stdout. Exit status and encoding relate as follows:

               0    - (New) JIS
               1    - Old JIS
               2    - NEC-JIS
               3    - Shift-JIS
               4    - Extended UNIX Code
               90   - ambiguous input format: EUC or Shift-JIS
               91   - input only contains 7-bit ASCII characters
               92   - unknown encoding

          This option may be used only if no input file or a  single  input



Page 1                       Reliant UNIX 5.44                      3, 1911

kcode(1)                         3/11/92                           kcode(1)

          file is specified.

     -v   Normally, kcode only examines the input until it finds the  first
          byte  sequence  which uniquely identifies an encoding. If a valid
          encoding was found, and the -v option is set, kcode verifies that
          the  remaining input conforms to the found encoding. Invalid byte
          sequences in the input cause  the  encoding  to  be  reported  as
          unknown.  7-bit encoded input files must terminate in shifted-out
          mode to pass verification.

          -v cannot be used in conjunction with the -l limit option.

DIAGNOSTICS
     Normally, kcode returns an exit status of 0 (unless the  -q option  is
     set).  If  an  error occurs, a diagnostic is printed on stderr and the
     exit status is set to 99.

SEE ALSO
     kconv(1), kverify(1), fkcode(3K).



































Page 2                       Reliant UNIX 5.44                      3, 1911