.\"	$OpenBSD: sort.1,v 1.31 2007/08/21 21:22:37 millert Exp $
.\"	$FreeBSD$
.\"
.\" Copyright (c) 1991, 1993
.\"	The Regents of the University of California.  All rights reserved.
.\"
.\" This code is derived from software contributed to Berkeley by
.\" the Institute of Electrical and Electronics Engineers, Inc.
.\"
.\" Redistribution and use in source and binary forms, with or without
.\" modification, are permitted provided that the following conditions
.\" are met:
.\" 1. Redistributions of source code must retain the above copyright
.\"    notice, this list of conditions and the following disclaimer.
.\" 2. Redistributions in binary form must reproduce the above copyright
.\"    notice, this list of conditions and the following disclaimer in the
.\"    documentation and/or other materials provided with the distribution.
.\" 3. Neither the name of the University nor the names of its contributors
.\"    may be used to endorse or promote products derived from this software
.\"    without specific prior written permission.
.\"
.\" THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
.\" ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
.\" IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
.\" ARE DISCLAIMED.  IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
.\" FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
.\" DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
.\" OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
.\" HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
.\" LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
.\" OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
.\" SUCH DAMAGE.
.\"
.\"     @(#)sort.1	8.1 (Berkeley) 6/6/93
.\"
.Dd July 3, 2012
.Dt SORT 1
.Os
.Sh NAME
.Nm sort
.Nd sort or merge records (lines) of text and binary files
.Sh SYNOPSIS
.Nm sort
.Bk -words
.Op Fl bcCdfghiRMmnrsuVz
.Sm off
.Op Fl k\ \& Ar field1 Op , Ar field2
.Sm on
.Op Fl S Ar memsize
.Ek
.Op Fl T Ar dir
.Op Fl t Ar char
.Op Fl o Ar output
.Op Ar file ...
.Nm sort
.Fl Fl help
.Nm sort
.Fl Fl version
.Sh DESCRIPTION
The
.Nm
utility sorts text and binary files by lines.
A line is a record separated from the subsequent record by a
newline (default) or NUL \'\\0\' character (-z option).
A record can contain any printable or unprintable characters.
Comparisons are based on one or more sort keys extracted from
each line of input, and are performed lexicographically,
according to the current locale's collating rules and the
specified command-line options that can tune the actual
sorting behavior.
By default, if keys are not given,
.Nm
uses entire lines for comparison.
.Pp
The command line options are as follows:
.Bl -tag -width Ds
.It Fl c, Fl Fl check, Fl C, Fl Fl check=silent|quiet
Check that the single input file is sorted.
If the file is not sorted,
.Nm
produces the appropriate error messages and exits with code 1,
otherwise returns 0.
If
.Fl C
or
.Fl Fl check=silent
is specified,
.Nm
produces no output.
This is a "silent" version of
.Fl c.
.It Fl m , Fl Fl merge
Merge only.
The input files are assumed to be pre-sorted.
If they are not sorted the output order is undefined.
.It Fl o Ar output , Fl Fl output Ns = Ns Ar output
Print the output to the
.Ar output
file instead of the standard output.
.It Fl S Ar size, Fl Fl buffer-size Ns = Ns Ar size
Use
.Ar size
for the maximum size of the memory buffer.
Size modifiers %,b,K,M,G,T,P,E,Z,Y can be used.
If a memory limit is not explicitly specified,
.Nm
takes up to about 90% of available memory.
If the file size is too big to fit into the memory buffer,
the temporary disk files are used to perform the sorting.
.It Fl T Ar dir , Fl Fl temporary-directory Ns = Ns Ar dir
Store temporary files in the directory
.Ar dir .
The default path is the value of the environment variable
.Ev TMPDIR
or
.Pa /var/tmp
if
.Ev TMPDIR
is not defined.
.It Fl u , Fl Fl unique
Unique keys.
Suppress all lines that have a key that is equal to an already
processed one.
This option, similarly to
.Fl s ,
implies a stable sort.
If used with
.Fl c
or
.Fl C ,
.Nm
also checks that there are no lines with duplicate keys.
.It Fl s
Stable sort.
This option maintains the original record order of records that have
and equal key.
This is a non-standard feature, but it is widely accepted and used.
.It Fl Fl version
Print the version and silently exits.
.It Fl Fl help
Print the help text and silently exits.
.El
.Pp
The following options override the default ordering rules.
When ordering options appear independently of key field
specifications, they apply globally to all sort keys.
When attached to a specific key (see
.Fl k ) ,
the ordering options override all global ordering options for
the key they are attahced to.
.Bl -tag -width indent
.It Fl b, Fl Fl ignore-leading-blanks
Ignore leading blank characters when comparing lines.
.It Fl d , Fl Fl dictionary-order
Consider only blank spaces and alphanumeric characters in comparisons.
.It Fl f , Fl Fl ignore-case
Convert all lowercase characters to their uppercase equivalent
before comparison, that is, perform case-independent sorting.
.It Fl g, Fl Fl general-numeric-sort, Fl Fl sort=general-numeric
Sort by general numerical value.
As opposed to
.Fl n ,
this option handles general floating points, which have a much
permissive format than those allowed by
. Fl n ,
but it has a significant performance drawback.
.It Fl h, Fl Fl human-numeric-sort, Fl Fl sort=human-numeric
Sort by numerical value, but take into account the SI suffix,
if present.
Sort first by numeric sign (negative, zero, or
positive); then by SI suffix (either empty, or `k' or `K', or one
of `MGTPEZY', in that order); and finally by numeric value.
The SI suffix must immediately follow the number.
For example, '12345K' sorts before '1M', because M is "larger" than K.
This sort option is useful for sorting the output of a single invocation
of 'df' command with
.Fl h
or
.Fl H
options (human-readable).
.It Fl i , Fl Fl ignore-nonprinting
Ignore all non-printable characters.
.It Fl M, Fl Fl month-sort, Fl Fl sort=month
Sort by month abbreviations.
Unknown strings are considered smaller than the month names.
.It Fl n , Fl Fl numeric-sort, Fl Fl sort=numeric
Sort fields numerically by arithmetic value.
Fields are supposed to have optional blanks in the beginning, an
optional minus sign, zero or more digits (including decimal point and
possible thousand separators).
.It Fl R, Fl Fl random-sort, Fl Fl sort=random
Sort by a random order.
This is a random permutation of the inputs except that
the equal keys sort together.
It is implemented by hashing the input keys and sorting
the hash values.
The hash function is choosen randomly.
The hash function is randomized by
.Cm /dev/random
content, or by file content if it is specified by
.Fl Fl random-source .
Even if multiple sort fields are specified,
the same random hash function is used for all of them.
.It Fl r , Fl Fl reverse
Sort in reverse order.
.It Fl V, Fl Fl version-sort
Sort version numbers.
The input lines are treated as file names in form
PREFIX VERSION SUFFIX, where SUFFIX matches the regular expression
"(\.([A-Za-z~][A-Za-z0-9~]*)?)*".
The files are compared by their prefixes and versions (leading
zeros are ignored in version numbers, see example below).
If an input string does not match the pattern, then it is compared
using the byte compare function.
All string comparisions are performed in C locale, the locale
environment setting is ignored.
.Bl -tag -width indent
.It Example:
.It $ ls sort* | sort -V
.It sort-1.022.tgz
.It sort-1.23.tgz
.It sort-1.23.1.tgz
.It sort-1.024.tgz
.It sort-1.024.003.
.It sort-1.024.003.tgz
.It sort-1.024.07.tgz
.It sort-1.024.009.tgz
.El
.El
.Pp
The treatment of field separators can be altered using these options:
.Bl -tag -width indent
.It Fl b , Fl Fl ignore-leading-blanks
Ignore leading blank space when determining the start
and end of a restricted sort key (see
.Fl k
).
If
.Fl b
is specified before the first
.Fl k
option, it applies globally to all key specifications.
Otherwise,
.Fl b
can be attached independently to each
.Ar field
argument of the key specifications.
.Fl b .
.It Xo
.Sm off
.Fl k\ \& Ar field1 Op , Ar field2 , Fl Fl key Ns = Ns Ar field1 Op , Ar field2
.Sm on
.Xc
Define a restricted sort key that has the starting position
.Ar field1 ,
and optional ending position
.Ar field2
of a key field.
The
.Fl k
option may be specified multiple times,
in which case subsequent keys are compared when earlier keys compare equal.
The
.Fl k
option replaces the obsolete options
.Cm \(pl Ns Ar pos1
and
.Fl Ns Ar pos2 ,
but the old notation is also supported.
.It Fl t Ar char , Fl Fl field-separator Ns = Ns Ar char
Use
.Ar char
as a field separator character.
The initial
.Ar char
is not considered to be part of a field when determining key offsets.
Each occurrence of
.Ar char
is significant (for example,
.Dq Ar charchar
delimits an empty field).
If
.Fl t
is not specified, the default field separator is a sequence of
blank space characters, and consecutive blank spaces do
.Em not
delimit an empty field, however, the initial blank space
.Em is
considered part of a field when determining key offsets.
To use NUL as field separator, use
.Fl t
\'\\0\'.
.It Fl z , Fl Fl zero-terminated
Use NUL as record separator.
By default, records in the files are supposed to be separated by
the newline characters.
With this option, NUL (\'\\0\') is used as a record separator character.
.El
.Pp
Other options:
.Bl -tag -width indent
.It Fl Fl batch-size Ns = Ns Ar num
Specify maximum number of files that can be opened by
.Nm
at once.
This option affects behavior when having many input files or using
temporary files.
The default value is 16.
.It Fl Fl compress-program Ns = Ns Ar PROGRAM
Use PROGRAM to compress temporary files.
PROGRAM must compress standard input to standard output, when called
without arguments.
When called with argument
.Fl d
it must decompress standard input to standard output.
If PROGRAM fails,
.Nm
must exit with error.
An example of PROGRAM that can be used here is bzip2.
.It Fl Fl random-source Ns = Ns Ar filename
In random sort, the file content is used as the source of the 'seed' data
for the hash function choice.
Two invocations of random sort with the same seed data will use
the same hash function and will produce the same result if the input is
also identical.
By default, file
.Cm /dev/random
is used.
.It Fl Fl debug
Print some extra information about the sorting process to the
standard output.
%%THREADS%%.It Fl Fl parallel
%%THREADS%%Set the maximum number of execution threads.
%%THREADS%%Default number equals to the number of CPUs.
.It Fl Fl files0-from Ns = Ns Ar filename
Take the input file list from the file
.Ar filename.
The file names must be separated by NUL
(like the output produced by the command "find ... -print0").
.It Fl Fl radixsort
Try to use radix sort, if the sort specifications allow.
The radix sort can only be used for trivial locales (C and POSIX),
and it cannot be used for numeric or month sort.
Radix sort is very fast and stable.
.It Fl Fl mergesort
Use mergesort.
This is a universal algorithm that can always be used,
but it is not always the fastest.
.It Fl Fl qsort
Try to use quick sort, if the sort specifications allow.
This sort algorithm cannot be used with
.Fl u
and
.Fl s .
.It Fl Fl heapsort
Try to use heap sort, if the sort specifications allow.
This sort algorithm cannot be used with
.Fl u
and
.Fl s .
.It Fl Fl mmap
Try to use file memory mapping system call.
It may increase speed in some cases.
.El
.Pp
The following operands are available:
.Bl -tag -width indent
.It Ar file
The pathname of a file to be sorted, merged, or checked.
If no
.Ar file
operands are specified, or if a
.Ar file
operand is
.Fl ,
the standard input is used.
.El
.Pp
A field is defined as a maximal sequence of characters other than the
field separator and record separator (newline by default).
Initial blank spaces are included in the field unless
.Fl b
has been specified;
the first blank space of a sequence of blank spaces acts as the field
separator and is included in the field (unless
.Fl t
is specified).
For example, all blank spaces at the beginning of a line are
considered to be part of the first field.
.Pp
Fields are specified by the
.Sm off
.Fl k\ \& Ar field1 Op , Ar field2
.Sm on
command-line option.
If
.Ar field2
is missing, the end of the key defaults to the end of the line.
.Pp
The arguments
.Ar field1
and
.Ar field2
have the form
.Em m.n
.Em (m,n > 0)
and can be followed by one or more of the modifiers
.Cm b , d , f , i ,
.Cm n , g , M
and
.Cm r ,
which correspond to the options discussed above.
When
.Cm b
is specified it applies only to
.Ar field1
or
.Ar field2
where it is specified while the rest of the modifiers
apply to the whole key field regardless if they are
specified only with
.Ar field1
or
.Ar field2
or both.
A
.Ar field1
position specified by
.Em m.n
is interpreted as the
.Em n Ns th
character from the beginning of the
.Em m Ns th
field.
A missing
.Em \&.n
in
.Ar field1
means
.Ql \&.1 ,
indicating the first character of the
.Em m Ns th
field; if the
.Fl b
option is in effect,
.Em n
is counted from the first non-blank character in the
.Em m Ns th
field;
.Em m Ns \&.1b
refers to the first non-blank character in the
.Em m Ns th
field.
.No 1\&. Ns Em n
refers to the
.Em n Ns th
character from the beginning of the line;
if
.Em n
is greater than the length of the line, the field is taken to be empty.
.Pp
.Em n Ns th
positions are always counted from the field beginning, even if the field
is shorter than the number of specified positions.
Thus, the key can really start from a position in a subsequent field.
.Pp
A
.Ar field2
position specified by
.Em m.n
is interpreted as the
.Em n Ns th
character (including separators) from the beginning of the
.Em m Ns th
field.
A missing
.Em \&.n
indicates the last character of the
.Em m Ns th
field;
.Em m
= \&0
designates the end of a line.
Thus the option
.Fl k Ar v.x,w.y
is synonymous with the obsolete option
.Cm \(pl Ns Ar v-\&1.x-\&1
.Fl Ns Ar w-\&1.y ;
when
.Em y
is omitted,
.Fl k Ar v.x,w
is synonymous with
.Cm \(pl Ns Ar v-\&1.x-\&1
.Fl Ns Ar w\&.0 .
The obsolete
.Cm \(pl Ns Ar pos1
.Fl Ns Ar pos2
option is still supported, except for
.Fl Ns Ar w\&.0b ,
which has no
.Fl k
equivalent.
.Sh ENVIRONMENT
.Bl -tag -width Fl
.It Ev LC_COLLATE
Locale settings to be used to determine the collation for
sorting records.
.It Ev LC_CTYPE
Locale settings to be used to case conversion and classification
of characters, that is, which characters are considered
whitespaces, etc.
.It Ev LC_MESSAGES
Locale settings that determine the language of output messages
that
.Nm
prints out.
.It Ev LC_NUMERIC
Locale settings that determine the number format used in numeric sort.
.It Ev LC_TIME
Locale settings that determine the month format used in month sort.
.It Ev LC_ALL
Locale settings that override all of the above locale settings.
This environment variable can be used to set all these settings
to the same value at once.
.It Ev LANG
Used as a last resort to determine different kinds of locale-specific
behavior if neither the respective environment variable, nor
.Ev LC_ALL
are set.
%%NLS%%.It Ev NLSPATH
%%NLS%%Path to NLS catalogs.
.It Ev TMPDIR
Path to the directory in which temporary files will be stored.
Note that
.Ev TMPDIR
may be overridden by the
.Fl T
option.
.It Ev GNUSORT_NUMERIC_COMPATIBILITY
If defined
.Fl t
will not override the locale numeric symbols, that is, thousand
separators and decimal separators.
By default, if we specify
.Fl t
with the same symbol as the thousand separator or decimal point,
the symbol will be treated as the field separator.
Older behavior was less definite; the symbol was treated as both field
separator and numeric separator, simultaneously.
This environment variable enables the old behavior.
.El
.Sh FILES
.Bl -tag -width Pa -compact
.It Pa /var/tmp/.bsdsort.PID.*
Temporary files.
.It Pa /dev/random
Default seed file for the random sort.
.El
.Sh EXIT STATUS
The
.Nm
utility shall exit with one of the following values:
.Pp
.Bl -tag -width flag -compact
.It 0
Successfully sorted the input files or if used with
.Fl c
or
.Fl C ,
the input file already met the sorting criteria.
.It 1
On disorder (or non-uniqueness) with the
.Fl c
or
.Fl C
options.
.It 2
An error occurred.
.El
.Sh SEE ALSO
.Xr comm 1 ,
.Xr join 1 ,
.Xr uniq 1 .
.Sh STANDARDS
The
.Nm
utility is compliant with the
.St -p1003.1-2008
specification.
.Pp
The flags
.Op Fl ghRMSsTVz
are extensions to the POSIX specification.
.Pp
All long options are extensions to the specification, some of them are
provided for compatibility with GNU versions and some of them are
own extensions.
.Pp
The old key notations
.Cm \(pl Ns Ar pos1
and
.Fl Ns Ar pos2
come from older versions of
.Nm
and are still supported but their use is highly discouraged.
.Sh HISTORY
A
.Nm
command first appeared in
.At v3 .
.Sh AUTHORS
Gabor Kovesdan <gabor@FreeBSD.org>,
.Pp
Oleg Moskalenko <oleg.moskalenko@citrix.com>
.Sh NOTES
This implementation of
.Nm
has no limits on input line length (other than imposed by available
memory) or any restrictions on bytes allowed within lines.
.Pp
The performance depends highly on locale settings,
efficient choice of sort keys and key complexity.
The fastest sort is with locale C, on whole lines,
with option
.Fl s.
In general, locale C is the fastest, then single-byte
locales follow and multi-byte locales as the slowest but
the correct collation order is always respected.
As for the key specification, the simpler to process the
lines the faster the search will be.
.Pp
When sorting by arithmetic value, using
.Fl n
results in much better performance than
.Fl g
so its use is encouraged
whenever possible.