The sort command is used in Linux and Unix system to arrange the contents of the file in a specific order. Using sort command you order lines, alphabet and numeric in a file.
Sort command has many useful options to reverse order, sort a file numerically, sorting a table by column number, checks if a file is already sorted and removes duplicates.
In this tutorial, we learn about sort command in Linux with some practical examples.
Sort Command and its syntax
The following line shows the sort command syntax:
sort [options] [files]
It takes input from STDIN by default but can sort files when file names are provided as an argument and it can also sort numerically as well.
Sort command can be invoked by typing just sort in the terminal. It will then prompt for the input from STDIN. After entering STDIN, ctrl+d is entered for marking end of input.
The following output shows the input data from STDIN is rearranged in alphabetical order.
$ sort cappa beta delta theta alpha alpha beta cappa delta theta
Sort a file
By default, sort treats all the characters as string characters. The numeric characters are also treated as characters.
First, we create a file with numeric data to be sorted. The following command creates a file 'numeric.txt' with numeric data:
$ cat > numeric.txt 01 10 25 83 502 111
Now, let us sort the file 'numeric.txt' as follows:
$ sort numeric.txt 01 10 111 25 502 83
The output shown above is not what we generally expect. Of course, 111 is larger than 25 and 502 is greater than 83 as well. But if you see closely, the numbers are sorted in dictionary order.
For example, for 111 and 25, compare the first character, 1 precedes 2. Hence 111 is placed before 25.
Let's now check how it sorts when content is alphabets or words. The following file 'wordlist.txt' contain words starting with uppercase case and lower case and display sorted list:
$ sort wordlist.txt Aeroplane ant apple bark born Box Cat tom welcome
Sort options
Sort command comes with pretty good options to order file contents. Let's discuss some of them as follows:
sort -n: To sort a file numerically.
sort -u: Suppress lines that repeat an earlier key.
sort -k: Sorting a table by column number.
sort -t SEP: Use the provided separator to identify the fields.
sort -M: Sort as per the calendar month specified.
sort -b: Ignore blanks at the start of the line.
sort -r: Sorting in reverse order.
sort -o: Output to a new file.
Sort file numerically (-n Option)
If we want to sort the contents inside a file numerically, we use -n option.
In the following example, the file 'numeric.txt' contains numbers and data is sorted in numerical order:
$ sort -n numeric.txt 01 10 25 83 111 502
To reverse the number in the file, you use a combination of -nr option.
For example:
$ sort -nr numeric.txt 502 111 83 25 10 01
Sort and remove duplicate lines (-u Option)
To sort and remove duplicate lines from a file use -u option. The first command will list the content of the file duplicate.txt using cat command and second we will use -u option to remove duplicate lines.
Example:
$ cat duplicate.txt hello linux lInux Linux raghu world zzz zzz
Sample output where you can see the duplicate lines are removed 'duplicate.txt' file:
$ sort -u duplicate.txt hello linux lInux Linux raghu world zzz
Sort and Ignore Case (-f Option)
Like many other Linux and Unix tools, sort command is case sensitive by default. But if we need to ignore the case, we can use -f or --ignore-case option.
For example, in the 'duplicate.txt', there are three instances of the word 'linux' in different cases. By default, they all are printed with -u option. It means they are not unique for sort command.
But if we use -f option it would be ignored as shown below:
$ sort -f -u duplicate.txt hello linux raghu world zzz
Sort by column (-k Option)
The sort command can sort any table using column number using -k option. For example use '-k3' to sort third column in the table.
Let consider we have a table of contents in 'population.txt' file as follows:
$ cat population.txt Kids 500 India Youth 400 England Senior 600 USA Junior 9000 Australia Pensioners 650 China
In the following we sort the table using column number 2 in the population.txt file:
$ sort -k2 population.txt Youth 400 England Kids 500 India Senior 600 USA Pensioners 650 China Junior 9000 Australia
Sort multiple columns
In some situations, you may need to sort a table using multiple columns. Let's take an example of a file containing multiple columns of data.
Following the file 'columns.txt'contain multiple columns as listed:
cat columns.txt version1.2 10 25 version1.2 30 50 version1.1 10 30 version1.1 40 50 version1.2 40 50 version1.1 10 20 version1.1 5 8
Now check the command to sort columns 1, 2 and 3 in numerical order.
$ sort -k1,1 -k2,2n -k3,3n columns.txt version1.1 5 8 version1.1 10 20 version1.1 10 30 version1.1 40 50 version1.2 10 25 version1.2 30 50 version1.2 40 50
Sort by column field separator (delimiter -t Option)
By default, delimiter for columns is whitespace or tab. But we can use custom delimiter using -t or --field-separator option in sort.
Let's take an example, use colon (:) as the delimiter and '/etc/passwd' file to sort.
If we want to sort /etc/passwd file on the basis of UIDs, we use the following command:
$ sort -n -t ':' -k3 /etc/passwd root:x:0:0:root:/root:/bin/bash bin:x:1:1:bin:/bin:/sbin/nologin daemon:x:2:2:daemon:/sbin:/sbin/nologin adm:x:3:4:adm:/var/adm:/sbin/nologin lp:x:4:7:lp:/var/spool/lpd:/sbin/nologin sync:x:5:0:sync:/sbin:/bin/sync shutdown:x:6:0:shutdown:/sbin:/sbin/shutdown halt:x:7:0:halt:/sbin:/sbin/halt mail:x:8:12:mail:/var/spool/mail:/sbin/nologin news:x:9:13:news:/etc/news: uucp:x:10:14:uucp:/var/spool/uucp:/sbin/nologin operator:x:11:0:operator:/root:/sbin/nologin games:x:12:100:games:/usr/games:/sbin/nologin <---output truncated--->
Sort by month (-M Option)
Using -M option you can sort by month. If you have a file with months, sort command order by months.
For example, 'months.txt' file contain a list of month names.
$ cat months.txt sep august july dec feb may jan
We will now use -M option to sort 'months.txt' file by month.
$ sort -M months.txt jan feb may july august sep dec
Sort human readable numbers (-h Option)
Sort can order human-readable numeric like 2K, 5G, 3M using -h or --human-numeric-sort option.
For example, below file list human readable number
$ cat human_numeric.txt 4G 2K 3M 1G 34K 52M 200M
Following command use -h option to compare and sort human-readable numbers.
$ sort -h human_numeric.txt 2K 34K 3M 52M 200M 1G 4G
Check for sorted input (-c Option)
The sort command can check if the input is already sorted or not using -c option. If it is not sorted, it returns the first unsorted line.
$ sort -c /etc/passwd sort: /etc/passwd:2: disorder: daemon:x:1:1:daemon:/usr/sbin:/bin/sh
Nothing is printed for sorted input:
$ sort /etc/passwd | sort -c $ echo $? 0
Sort with other commands
The sort is a text processing tool, so it can be used by piping in some input from another command.
In the following command we use tail command output as input for sort:
$ tail /etc/passwd | sort altair:x:1001:1001:Altair Ibn La Ahad,,,,:/home/altair:/bin/bash bind:x:120:132::/var/cache/bind:/bin/false dnsmasq:x:121:65534:dnsmasq,,,:/var/lib/misc:/bin/false mysql:x:115:129:MySQL Server,,,:/nonexistent:/bin/false postfix:x:105:126::/var/spool/postfix:/bin/false raghu:x:1000:1000:raghu,,,:/home/raghu:/bin/bash smmsp:x:119:131:Mail Submission Program,,,:/var/lib/sendmail:/bin/false smmta:x:118:130:Mail Transfer Agent,,,:/var/lib/sendmail:/bin/false sshd:x:117:65534::/var/run/sshd:/usr/sbin/nologin statd:x:116:65534::/var/lib/nfs:/bin/false
The sorting can be reversed with -r option.
Lets take another example to sort du command output in human-readable format by size, use the following command:
$ du -h | sort -h
Conclusion
In this tutorial, we learn about Linux sort and its option to order file contents. I hope you enjoyed reading and please comment below if you find any other examples.
