Command line usage

  [positional]

    The input file to be processed.  Sysin is assumed if this parameter is not
    provided.
    isCaseSensitive

  [-o --outputfilename]

    The output file to be written.  Sysout is assumed if this parameter is not
    provided.  The same name as the input file is allowed.
    isCaseSensitive

  [-f --fixed]

    The record length in a fixed record length file.
    inclusive range from 1 to 4096

  [-r --replace]

    Overwrite the input file with the results.  --outputFile is not required
    with this parameter.  --outputFile is assumed.

  [--eol]

    The byte(s) that end each line in a variable length record file.
    none or at least 1
    Default: lf(10)

  [--eolOut]

    The byte(s) that end each line in a variable length record file.  This will
    be used to write the output file.  If this is not specified then the --eol
    value will be used.
    none or at least 1

  [-d --duplicate]

    Special handling of duplicate keyed rows.
    valid values are firstonly lastonly reverse original
    Default: original

  [-c --copy]

    No sort, just copy.
    valid values are reverse original bykey
    Default: bykey

  [--maxrows]

    Used for variable length input, estimate the number of rows.  Too low could
    cause problems.
    inclusive range from 2
    Default: 9223372036854775807

  [-k --keys]

    The sort keys.
    none or at least 1
      ( usage for keys

      <positional>

        The data type of the key in the file.
        valid values are string integer float binteger bfloat date

      [-f --field]

        If this is a CSV file then use this instead of offset and length.
        inclusive range from 0

      [-o --offset]

        The zero relative offset from the beginning of a row.
        inclusive range from 0
        Default: 0

      [-l --length]

        The length of the key in bytes.
        inclusive range from 1 to 255
        Default: 255

      [positional]

        The direction of the sort for this key. AASC and ADESC are absolute
        values of the key - the case of letters would not matter and the sign
        of numbers would not matter.
        valid values are asc desc aasc adesc
        Default: asc

      [-f --format]

        The parsing format for converting the contents of the key in the file
        to an internal representation. Use Java SimpleDateFormat rules for
        making the format.
        isCaseSensitive
      )

  [--csv]

    The definition of the CSV file being read as input.  Using this indicates
    that the input is in fact a CSV file and the keys paramater should use the
    --field arguments.
      ( usage for csv

      [-h --header]

        Skip over the first line for sorting and just write it to the beginning
        of the output file.

      [-q --quoteByte]

        This is the single character that you would like to use for a quote.
        Default: "(34)

      [-s --separatorByte]

        This is the single character that you would like to use for the field
        separator.
        Default: ,(44)
      )

  [--variableoutput]

    Use this to cause a fixed input to be written as variable.

  [--workDirectory]

    The directory where temp files will be handled.
    isCaseSensitive
    Default: /tmp

  [--cacheInput]

    Read the input file into memory.  This saves reading it again on multipass
    sorts.  The amount of memory required to hold the input file in core is
    equal to the size of the file.

  [--cacheWork]

    Work files are saved in memory.  Otherwise they are stored on disk.  The
    amount of memory required to hold work areas in memory is about (2 *
    (keySize + 24)).

  [--power]

    The depth of the funnel.  The bigger this number is, the more memory will
    be used.  This is computed when --max or -f is specified.
    inclusive range from 2 to 16
    Default: 16

  [--version]

    Display the version of Funnel

  [-?]

    Show this help message.