.TH "RECUT" "1" "31 Dec 2013" "0.1" "RECUT"
.SH NAME
recut \- reorder, duplicate and remove sections from each line
.SH SYNOPSIS
.B recut
.B \-b
.I field-list
.RB "[\| " \-o
.IR delim " \|]"
.RB "[\| " -- " \|]"
.RI "[\| " file "... \|]"

.B recut
.B \-c
.I field-list
.RB "[\| " \-o
.IR delim " \|]"
.RB "[\| " -- " \|]"
.RI "[\| " file "... \|]"

.B recut
.B \-f
.I field-list
.RB "[\| " \-d
.IR delim " \|]"
.RB "[\| " \-o
.IR delim " \|]"
.RB "[\| " \-s " \|]"
.RB "[\| " -- " \|]"
.RI "[\| " file "... \|]"

.B recut
.I long-options
.RB "[\| " -- " \|]"
.RI "[\| " file "... \|]"

.B recut
.B \-\-version
.SH DESCRIPTION
.B recut
reads all files (or standard input) and processes them line by line.
It splits each line into fields at a delimiter (by default a tab
character), and then outputs those fields according to a specified
field list. Fields can thus be suppressed, reordered and duplicated.
The output fields are separated by another delimiter (by default also
tab). The result is printed on standard output.

Instead of fields separated by a delimiter,
.B recut
can also reorder the bytes or characters of each line.

The special file name
.B \-
means the standard input.

A pair of dashes
.RB ( \-\- )
may be used to separate the options from the file arguments. This may be useful if a file name starts with a dash.

All output lines end with a newline. If the last line of the input is
missing a newline, one is added automatically.
.SH OPTIONS
.TP 10
.BI \-\-bytes= field-list
Outputs exactly the specified bytes from each input line, with the
.I output-delimiter
between them (see
.BR \-\-output-delimiter ).
The first byte of each line is numbered 1. It is not an error if the
.I field-list
specifies bytes beyond the length of the line: those bytes will not be
printed. But the
.I output-delimiter
will still be printed.
.TP
.BI \-b " field-list"
Same as
.BI \-\-bytes= field-list
.TP
.BI \-\-character= field-list
Outputs the specified characters from each line of the input, with the
.I output-delimiter
between them (see
.BR \-\-output-delimiter ).
The first character of each line is numbered 1. It is not an error if
the
.I field-list
specifies bytes beyond the length of the line: those bytes will not be
printed. But the
.I output-delimiter
will still be printed.
The input is interpreted according to the encoding part of the current locale.
.TP
.BI \-c " field-list"
Same as
.BI \-\-characters= field-list
.TP
.BI \-\-fields= field-list
Outputs the specified fields from each line of the input, with the
.I output-delimiter
between them (see
.BR \-\-output-delimiter ).
The first field is numbered 1. It is not an error if the
.I field-list
argument specifies fields beyond the length of the line: those fields
are considered to be empty. The
.I output-delimiter
will still be printed between them.
.TP
.BI \-f " field-list"
Same as
.BI \-\-fields= field-list
.TP
.BI \-\-delimiter= delim
Specifies the delimiter that separates the fields in each line. The
default is a tab character. Unlike in
.BR cut (1),
the
.I delim
may consist of more than one character. This option only has an effect
in combination with
.BR \-\-fields .
.TP
.BI \-d " delim"
Same as
.BI \-\-delimiter= delim
.TP
.BI \-\-output-delimiter= delim
Specifies the delimiter that is inserted between fields, bytes or characters in the output. The default is the empty string in the case of
.B \-\-bytes
and
.BR \-\-characters .
When
.B \-\-fields
is used, the default is to use the same delimiter as for splitting the input, i.e., the same as
.BR \-\-delimiter .
.TP
.BI \-o " delim"

.TP
.B \-\-only-delimited
Suppress lines that do not contain at least one occurrence of
.I delim
(see
.BR \-\-delimiter ).
Only has effect in combination with
.BR \-\-fields .
.TP
.B \-s
Same as
.B \-\-only-delimited
.TP
.B \-\-version
Print the version number of
.B recut
and exit.
.LP
The equals sign (=) after long options may be replaced by spaces.
.LP
The
.I field-list
argument is a comma- or space-separated list of ranges. Each range is of the form
.IR a\-b ,
where a and b are field numbers >= 1. This represents the range of fields from
.I a
to
.IR b ,
inclusive. If
.I a
is 1, it may be omitted. Omitting
.I b
means "all remaining fields of the line".
.I b
may be less than
.IR a ,
in which case the range is descending. Some examples:
.TP
.B 3\-5
The fields 3, 4 and 5.
.TP
.B 5\-3
The fields 5, 4 and 3, in that order.
.TP
.B \-5
The fields 1, 2, 3, 4 and 5.
.TP
.B 3\-
The fields 3, 4, etc., until and including the last field in the line.
(No fields are printed if the line has fewer than 3 fields.)
.TP
.B \-
All fields of the line.
.TP
.B 3,2,1
The fields 3, 2 and 1. (The same as 3\-1.)
.TP
.B 2,1,3\-
Fields 2 and 1 and then all other fields, if any. (Swaps the first two
fields.)
.LP
An empty field list is allowed, e.g.,
.B \-\-fields=""
(but all output lines will be empty).
.SH EXAMPLES
.TP 10
.B "recut \-d : \-f 1,3 /etc/passwd"
Output the 1st and 3rd column of the passwd file.
.TP
.B "recut \-f 1,1,2,2,3\-"
Duplicate the first two columns. Input is read from standard input.
.TP
.B "recut \-f 2,1,3\-"
Swap the first two columns. Input is read from standard input.
.TP
.B "recut \-b \- myfile.txt"
The output is the exact content of myfile.txt (except that a newline
is added at the end if myfile.txt doesn't end with a newline).
.TP
.B "recut \-d :: \-f 2"
Split lines from standard input into fields at occurrences of the
2-character string '::' and output only the second field of each line.
.TP
.B "recut -d ',' \-o ';' -f -"
Copy input to output, replacing all commas by semicolons. (This is
more efficiently done with
.BR "tr ',' ';'" )
.SH "SEE ALSO"
.BR setlocale (3)