AWK - Quick Guide

AWK - Overview

AWK is an interpreted programming language. It is very powerful and specially designed for text processing. Its name is derived from the family names of its authors − Alfred Aho, Peter Weinberger, and Brian Kernighan.

The version of AWK that GNU/Linux distributes is written and maintained by the Free Software Foundation (FSF); it is often referred to as GNU AWK.

Types of AWK

Following are the variants of AWK −

AWK − Original AWK from AT & T Laboratory.
NAWK − Newer and improved version of AWK from AT & T Laboratory.
GAWK − It is GNU AWK. All GNU/Linux distributions ship GAWK. It is fully compatible with AWK and NAWK.

Typical Uses of AWK

Myriad of tasks can be done with AWK. Listed below are just a few of them −

Text processing,
Producing formatted text reports,
Performing arithmetic operations,
Performing string operations, and many more.

AWK - Environment

This chapter describes how to set up the AWK environment on your GNU/Linux system.

Installation Using Package Manager

Generally, AWK is available by default on most GNU/Linux distributions. You can use which command to check whether it is present on your system or not. In case you don’t have AWK, then install it on Debian based GNU/Linux using Advance Package Tool (APT) package manager as follows −

[jeryy]$ sudo apt-get update
[jeryy]$ sudo apt-get install gawk

Similarly, to install AWK on RPM based GNU/Linux, use Yellowdog Updator Modifier yum package manager as follows −

[root]# yum install gawk

After installation, ensure that AWK is accessible via command line.

[jerry]$ which awk

On executing the above code, you get the following result −

/usr/bin/awk

Installation from Source Code

As GNU AWK is a part of the GNU project, its source code is available for free download. We have already seen how to install AWK using package manager. Let us now understand how to install AWK from its source code.

The following installation is applicable to any GNU/Linux software, and for most other freely-available programs as well. Here are the installation steps −

Step 1 − Download the source code from an authentic place. The command-line utility wget serves this purpose.

[jerry]$ wget http://ftp.gnu.org/gnu/gawk/gawk-4.1.1.tar.xz

Step 2 − Decompress and extract the downloaded source code.

[jerry]$ tar xvf gawk-4.1.1.tar.xz

Step 3 − Change into the directory and run configure.

[jerry]$ ./configure

Step 4 − Upon successful completion, the configure generates Makefile. To compile the source code, issue a make command.

[jerry]$ make

Step 5 − You can run the test suite to ensure the build is clean. This is an optional step.

[jerry]$ make check

Step 6 − Finally, install AWK. Make sure you have super-user privileges.

[jerry]$ sudo make install

That is it! You have successfully compiled and installed AWK. Verify it by executing the awk command as follows −

[jerry]$ which awk

On executing this code, you get the following result −

/usr/bin/awk

AWK - Workflow

To become an expert AWK programmer, you need to know its internals. AWK follows a simple workflow − Read, Execute, and Repeat. The following diagram depicts the workflow of AWK −

Read

AWK reads a line from the input stream (file, pipe, or stdin) and stores it in memory.

Execute

All AWK commands are applied sequentially on the input. By default AWK execute commands on every line. We can restrict this by providing patterns.

Repeat

This process repeats until the file reaches its end.

Program Structure

Let us now understand the program structure of AWK.

BEGIN block

The syntax of the BEGIN block is as follows −

Syntax

BEGIN {awk-commands}

The BEGIN block gets executed at program start-up. It executes only once. This is good place to initialize variables. BEGIN is an AWK keyword and hence it must be in upper-case. Please note that this block is optional.

Body Block

The syntax of the body block is as follows −

Syntax

/pattern/ {awk-commands}

The body block applies AWK commands on every input line. By default, AWK executes commands on every line. We can restrict this by providing patterns. Note that there are no keywords for the Body block.

END Block

The syntax of the END block is as follows −

Syntax

END {awk-commands}

The END block executes at the end of the program. END is an AWK keyword and hence it must be in upper-case. Please note that this block is optional.

Let us create a file marks.txt which contains the serial number, name of the student, subject name, and number of marks obtained.

1)  Amit    Physics  80
2)  Rahul   Maths    90
3)  Shyam   Biology  87
4)  Kedar   English  85
5)  Hari    History  89

Let us now display the file contents with header by using AWK script.

Example

[jerry]$ awk 'BEGIN{printf "Sr No\tName\tSub\tMarks\n"} {print}' marks.txt

When this code is executed, it produces the following result −

Output

Sr No Name Sub Marks
1) Amit Physics 80
2) Rahul Maths 90
3) Shyam Biology 87
4) Kedar English 85
5) Hari History 89

At the start, AWK prints the header from the BEGIN block. Then in the body block, it reads a line from a file and executes AWK's print command which just prints the contents on the standard output stream. This process repeats until file reaches the end.

AWK - Basic Syntax

AWK is simple to use. We can provide AWK commands either directly from the command line or in the form of a text file containing AWK commands.

AWK Command Line

We can specify an AWK command within single quotes at command line as shown −

awk [options] file ...

Example

Consider a text file marks.txt with the following content −

1) Amit     Physics    80
2) Rahul    Maths      90
3) Shyam    Biology    87
4) Kedar    English    85
5) Hari     History    89

Let us display the complete content of the file using AWK as follows −

Example

[jerry]$ awk '{print}' marks.txt

On executing this code, you get the following result −

Output

1) Amit     Physics    80
2) Rahul    Maths      90
3) Shyam    Biology    87
4) Kedar    English    85
5) Hari     History    89

AWK Program File

We can provide AWK commands in a script file as shown −

awk [options] -f file ....

First, create a text file command.awk containing the AWK command as shown below −

{print}

Now we can instruct the AWK to read commands from the text file and perform the action. Here, we achieve the same result as shown in the above example.

Example

[jerry]$ awk -f command.awk marks.txt

On executing this code, you get the following result −

Output

1) Amit  Physics 80
2) Rahul Maths   90
3) Shyam Biology 87
4) Kedar English 85
5) Hari  History 89

AWK Standard Options

AWK supports the following standard options which can be provided from the command line.

The -v option

This option assigns a value to a variable. It allows assignment before the program execution. The following example describes the usage of the -v option.

Example

[jerry]$ awk -v name=Jerry 'BEGIN{printf "Name = %s\n", name}'

On executing this code, you get the following result −

Output

Name = Jerry

The --dump-variables[=file] option

It prints a sorted list of global variables and their final values to file. The default file is awkvars.out.

Example

[jerry]$ awk --dump-variables ''
[jerry]$ cat awkvars.out

On executing the above code, you get the following result −

Output

ARGC: 1
ARGIND: 0
ARGV: array, 1 elements
BINMODE: 0
CONVFMT: "%.6g"
ERRNO: ""
FIELDWIDTHS: ""
FILENAME: ""
FNR: 0
FPAT: "[^[:space:]]+"
FS: " "
IGNORECASE: 0
LINT: 0
NF: 0
NR: 0
OFMT: "%.6g"
OFS: " "
ORS: "\n"
RLENGTH: 0
RS: "\n"
RSTART: 0
RT: ""
SUBSEP: "\034"
TEXTDOMAIN: "messages"

The --help option

This option prints the help message on standard output.

Example

[jerry]$ awk --help

On executing this code, you get the following result −

Output

Usage: awk [POSIX or GNU style options] -f progfile [--] file ...
Usage: awk [POSIX or GNU style options] [--] 'program' file ...
POSIX options : GNU long options: (standard)
   -f progfile                --file=progfile
   -F fs                      --field-separator=fs
   -v var=val                 --assign=var=val
Short options : GNU long options: (extensions)
   -b                         --characters-as-bytes
   -c                         --traditional
   -C                         --copyright
   -d[file]                   --dump-variables[=file]
   -e 'program-text'          --source='program-text'
   -E file                    --exec=file
   -g                         --gen-pot
   -h                         --help
   -L [fatal]                 --lint[=fatal]
   -n                         --non-decimal-data
   -N                         --use-lc-numeric
   -O                         --optimize
   -p[file]                   --profile[=file]
   -P                         --posix
   -r                         --re-interval
   -S                         --sandbox
   -t                         --lint-old
   -V                         --version

The --lint[=fatal] option

This option enables checking of non-portable or dubious constructs. When an argument fatal is provided, it treats warning messages as errors. The following example demonstrates this −

Example

[jerry]$ awk --lint '' /bin/ls

On executing this code, you get the following result −

Output

awk: cmd. line:1: warning: empty program text on command line
awk: cmd. line:1: warning: source file does not end in newline
awk: warning: no program text at all!

The --posix option

This option turns on strict POSIX compatibility, in which all common and gawk-specific extensions are disabled.

The --profile[=file] option

This option generates a pretty-printed version of the program in file. Default file is awkprof.out. Below simple example illustrates this −

Example

[jerry]$ awk --profile 'BEGIN{printf"---|Header|--\n"} {print} 
END{printf"---|Footer|---\n"}' marks.txt > /dev/null 
[jerry]$ cat awkprof.out

On executing this code, you get the following result −

Output

# gawk profile, created Sun Oct 26 19:50:48 2014

   # BEGIN block(s)

   BEGIN {
      printf "---|Header|--\n"
   }

   # Rule(s) {
      print $0
   }

   # END block(s)

   END {
      printf "---|Footer|---\n"
   }

The --traditional option

This option disables all gawk-specific extensions.

The --version option

This option displays the version information of the AWK program.

Example

[jerry]$ awk --version

When this code is executed, it produces the following result −

Output

GNU Awk 4.0.1
Copyright (C) 1989, 1991-2012 Free Software Foundation.

AWK - Basic Examples

This chapter describes several useful AWK commands and their appropriate examples. Consider a text file marks.txt to be processed with the following content −

1) Amit     Physics   80
2) Rahul    Maths     90
3) Shyam    Biology   87
4) Kedar    English   85
5) Hari     History   89

Printing Column or Field

You can instruct AWK to print only certain columns from the input field. The following example demonstrates this −

Example

[jerry]$ awk '{print $3 "\t" $4}' marks.txt

On executing this code, you get the following result −

Output

Physics   80
Maths     90
Biology   87
English   85
History   89

In the file marks.txt, the third column contains the subject name and the fourth column contains the marks obtained in a particular subject. Let us print these two columns using AWK print command. In the above example, $3 and $4 represent the third and the fourth fields respectively from the input record.

Printing All Lines

By default, AWK prints all the lines that match pattern.

Example

[jerry]$ awk '/a/ {print $0}' marks.txt

On executing this code, you get the following result −

Output

2) Rahul    Maths     90
3) Shyam    Biology   87
4) Kedar    English   85
5) Hari     History   89

In the above example, we are searching form pattern a. When a pattern match succeeds, it executes a command from the body block. In the absence of a body block − default action is taken which is print the record. Hence, the following command produces the same result −

Example

[jerry]$ awk '/a/' marks.txt

Printing Columns by Pattern

When a pattern match succeeds, AWK prints the entire record by default. But you can instruct AWK to print only certain fields. For instance, the following example prints the third and fourth field when a pattern match succeeds.

Example

[jerry]$ awk '/a/ {print $3 "\t" $4}' marks.txt

On executing this code, you get the following result −

Output

Maths    90
Biology  87
English  85
History  89

Printing Column in Any Order

You can print columns in any order. For instance, the following example prints the fourth column followed by the third column.

Example

[jerry]$ awk '/a/ {print $4 "\t" $3}' marks.txt

On executing the above code, you get the following result −

Output

90   Maths
87   Biology
85   English
89   History

Counting and Printing Matched Pattern

Let us see an example where you can count and print the number of lines for which a pattern match succeeded.

Example

[jerry]$ awk '/a/{++cnt} END {print "Count = ", cnt}' marks.txt

On executing this code, you get the following result −

Output

Count = 4

In this example, we increment the value of counter when a pattern match succeeds and we print this value in the END block. Note that unlike other programming languages, there is no need to declare a variable before using it.

Printing Lines with More than 18 Characters

Let us print only those lines that contain more than 18 characters.

Example

[jerry]$ awk 'length($0) > 18' marks.txt

On executing this code, you get the following result −

Output

3) Shyam   Biology   87
4) Kedar   English   85

AWK provides a built-in length function that returns the length of the string. $0 variable stores the entire line and in the absence of a body block, default action is taken, i.e., the print action. Hence, if a line has more than 18 characters, then the comparison results true and the line gets printed.

AWK - Built-in Variables

AWK provides several built-in variables. They play an important role while writing AWK scripts. This chapter demonstrates the usage of built-in variables.

Standard AWK variables

The standard AWK variables are discussed below.

ARGC

It implies the number of arguments provided at the command line.

Example

[jerry]$ awk 'BEGIN {print "Arguments =", ARGC}' One Two Three Four

On executing this code, you get the following result −

Output

Arguments = 5

But why AWK shows 5 when you passed only 4 arguments? Just check the following example to clear your doubt.

ARGV

It is an array that stores the command-line arguments. The array's valid index ranges from 0 to ARGC-1.

Example

[jerry]$ awk 'BEGIN { 
   for (i = 0; i < ARGC - 1; ++i) { 
      printf "ARGV[%d] = %s\n", i, ARGV[i] 
   } 
}' one two three four

On executing this code, you get the following result −

Output

ARGV[0] = awk
ARGV[1] = one
ARGV[2] = two
ARGV[3] = three

CONVFMT

It represents the conversion format for numbers. Its default value is %.6g.

Example

[jerry]$ awk 'BEGIN { print "Conversion Format =", CONVFMT }'

On executing this code, you get the following result −

Output

Conversion Format = %.6g

ENVIRON

It is an associative array of environment variables.

Example

[jerry]$ awk 'BEGIN { print ENVIRON["USER"] }'

On executing this code, you get the following result −

Output

jerry

To find names of other environment variables, use env command.

FILENAME

It represents the current file name.

Example

[jerry]$ awk 'END {print FILENAME}' marks.txt

On executing this code, you get the following result −

Output

marks.txt

Please note that FILENAME is undefined in the BEGIN block.

FS

It represents the (input) field separator and its default value is space. You can also change this by using -F command line option.

Example

[jerry]$ awk 'BEGIN {print "FS = " FS}' | cat -vte

On executing this code, you get the following result −

Output

FS =  $

NF

It represents the number of fields in the current record. For instance, the following example prints only those lines that contain more than two fields.

Example

[jerry]$ echo -e "One Two\nOne Two Three\nOne Two Three Four" | awk 'NF > 2'

On executing this code, you get the following result −

Output

One Two Three
One Two Three Four

NR

It represents the number of the current record. For instance, the following example prints the record if the current record number is less than three.

Example

[jerry]$ echo -e "One Two\nOne Two Three\nOne Two Three Four" | awk 'NR < 3'

On executing this code, you get the following result −

Output

One Two
One Two Three

FNR

It is similar to NR, but relative to the current file. It is useful when AWK is operating on multiple files. Value of FNR resets with new file.

OFMT

It represents the output format number and its default value is %.6g.

Example

[jerry]$ awk 'BEGIN {print "OFMT = " OFMT}'

On executing this code, you get the following result −

Output

OFMT = %.6g

OFS

It represents the output field separator and its default value is space.

Example

[jerry]$ awk 'BEGIN {print "OFS = " OFS}' | cat -vte

On executing this code, you get the following result −

Output

OFS =  $

ORS

It represents the output record separator and its default value is newline.

Example

[jerry]$ awk 'BEGIN {print "ORS = " ORS}' | cat -vte

On executing the above code, you get the following result −

Output

ORS = $
$

RLENGTH

It represents the length of the string matched by match function. AWK's match function searches for a given string in the input-string.

Example

[jerry]$ awk 'BEGIN { if (match("One Two Three", "re")) { print RLENGTH } }'

On executing this code, you get the following result −

Output

RS

It represents (input) record separator and its default value is newline.

Example

[jerry]$ awk 'BEGIN {print "RS = " RS}' | cat -vte

On executing this code, you get the following result −

Output

RS = $
$

RSTART

It represents the first position in the string matched by match function.

Example

[jerry]$ awk 'BEGIN { if (match("One Two Three", "Thre")) { print RSTART } }'

On executing this code, you get the following result −

Output

SUBSEP

It represents the separator character for array subscripts and its default value is \034.

Example

[jerry]$ awk 'BEGIN { print "SUBSEP = " SUBSEP }' | cat -vte

On executing this code, you get the following result −

Output

SUBSEP = ^\$

$0

It represents the entire input record.

Example

[jerry]$ awk '{print $0}' marks.txt

On executing this code, you get the following result −

Output

1) Amit     Physics   80
2) Rahul    Maths     90
3) Shyam    Biology   87
4) Kedar    English   85
5) Hari     History   89

$n

It represents the n^th field in the current record where the fields are separated by FS.

Example

[jerry]$ awk '{print $3 "\t" $4}' marks.txt

On executing this code, you get the following result −

Output

Physics   80
Maths     90
Biology   87
English   85
History   89

GNU AWK Specific Variables

GNU AWK specific variables are as follows −

ARGIND

It represents the index in ARGV of the current file being processed.

Example

[jerry]$ awk '{ 
   print "ARGIND   = ", ARGIND; print "Filename = ", ARGV[ARGIND] 
}' junk1 junk2 junk3

On executing this code, you get the following result −

Output

ARGIND   =  1
Filename =  junk1
ARGIND   =  2
Filename =  junk2
ARGIND   =  3
Filename =  junk3

BINMODE

It is used to specify binary mode for all file I/O on non-POSIX systems. Numeric values of 1, 2, or 3 specify that input files, output files, or all files, respectively, should use binary I/O. String values of r or w specify that input files or output files, respectively, should use binary I/O. String values of rw or wr specify that all files should use binary I/O.

ERRNO

A string indicates an error when a redirection fails for getline or if close call fails.

Example

[jerry]$ awk 'BEGIN { ret = getline < "junk.txt"; if (ret == -1) print "Error:", ERRNO }'

On executing this code, you get the following result −

Output

Error: No such file or directory

FIELDWIDTHS

A space separated list of field widths variable is set, GAWK parses the input into fields of fixed width, instead of using the value of the FS variable as the field separator.

IGNORECASE

When this variable is set, GAWK becomes case-insensitive. The following example demonstrates this −

Example

[jerry]$ awk 'BEGIN{IGNORECASE = 1} /amit/' marks.txt

On executing this code, you get the following result −

Output

1) Amit  Physics   80

LINT

It provides dynamic control of the --lint option from the GAWK program. When this variable is set, GAWK prints lint warnings. When assigned the string value fatal, lint warnings become fatal errors, exactly like --lint=fatal.

Example

[jerry]$ awk 'BEGIN {LINT = 1; a}'

On executing this code, you get the following result −

Output

awk: cmd. line:1: warning: reference to uninitialized variable `a'
awk: cmd. line:1: warning: statement has no effect

PROCINFO

This is an associative array containing information about the process, such as real and effective UID numbers, process ID number, and so on.

Example

[jerry]$ awk 'BEGIN { print PROCINFO["pid"] }'

On executing this code, you get the following result −

Output

TEXTDOMAIN

It represents the text domain of the AWK program. It is used to find the localized translations for the program's strings.

Example

[jerry]$ awk 'BEGIN { print TEXTDOMAIN }'

On executing this code, you get the following result −

Output

messages

The above output shows English text due to en_IN locale

AWK - Operators

Like other programming languages, AWK also provides a large set of operators. This chapter explains AWK operators with suitable examples.

S.No.	Operators & Description
1	Arithmetic Operators AWK supports the following arithmetic operators.
2	Increment and Decrement Operators AWK supports the following increment and decrement operators.
3	Assignment Operators AWK supports the following assignment operators.
4	Relational Operators AWK supports the following relational operators.
5	Logical Operators AWK supports the following logical operators.
6	Ternary Operator We can easily implement a condition expression using ternary operator.
7	Unary Operators AWK supports the following unary operators.
8	Exponential Operators There are two formats of exponential operators.
9	String Concatenation Operator Space is a string concatenation operator that merges two strings.
10	Array Membership Operator It is represented by in. It is used while accessing array elements.
11	Regular Expression Operators This example explains the two forms of regular expressions operators.

AWK - Regular Expressions

AWK is very powerful and efficient in handling regular expressions. A number of complex tasks can be solved with simple regular expressions. Any command-line expert knows the power of regular expressions.

This chapter covers standard regular expressions with suitable examples.

Dot

It matches any single character except the end of line character. For instance, the following example matches fin, fun, fan etc.

Example

[jerry]$ echo -e "cat\nbat\nfun\nfin\nfan" | awk '/f.n/'

On executing the above code, you get the following result −

Output

fun
fin
fan

Start of line

It matches the start of line. For instance, the following example prints all the lines that start with pattern The.

Example

[jerry]$ echo -e "This\nThat\nThere\nTheir\nthese" | awk '/^The/'

On executing this code, you get the following result −

Output

There
Their

End of line

It matches the end of line. For instance, the following example prints the lines that end with the letter n.

Example

[jerry]$ echo -e "knife\nknow\nfun\nfin\nfan\nnine" | awk '/n$/'

Output

On executing this code, you get the following result −

fun
fin
fan

Match character set

It is used to match only one out of several characters. For instance, the following example matches pattern Call and Tall but not Ball.

Example

[jerry]$ echo -e "Call\nTall\nBall" | awk '/[CT]all/'

Output

On executing this code, you get the following result −

Call
Tall

Exclusive set

In exclusive set, the carat negates the set of characters in the square brackets. For instance, the following example prints only Ball.

Example

[jerry]$ echo -e "Call\nTall\nBall" | awk '/[^CT]all/'

On executing this code, you get the following result −

Output

Ball

Alteration

A vertical bar allows regular expressions to be logically ORed. For instance, the following example prints Ball and Call.

Example

[jerry]$ echo -e "Call\nTall\nBall\nSmall\nShall" | awk '/Call|Ball/'

On executing this code, you get the following result −

Output

Call
Ball

Zero or One Occurrence

It matches zero or one occurrence of the preceding character. For instance, the following example matches Colour as well as Color. We have made u as an optional character by using ?.

Example

[jerry]$ echo -e "Colour\nColor" | awk '/Colou?r/'

On executing this code, you get the following result −

Output

Colour
Color

Zero or More Occurrence

It matches zero or more occurrences of the preceding character. For instance, the following example matches ca, cat, catt, and so on.

Example

[jerry]$ echo -e "ca\ncat\ncatt" | awk '/cat*/'

On executing this code, you get the following result −

Output

ca
cat
catt

One or More Occurrence

It matches one or more occurrence of the preceding character. For instance below example matches one or more occurrences of the 2.

Example

[jerry]$ echo -e "111\n22\n123\n234\n456\n222"  | awk '/2+/'

On executing the above code, you get the following result −

Output

Grouping

Parentheses () are used for grouping and the character | is used for alternatives. For instance, the following regular expression matches the lines containing either Apple Juice or Apple Cake.

Example

[jerry]$ echo -e "Apple Juice\nApple Pie\nApple Tart\nApple Cake" | awk 
   '/Apple (Juice|Cake)/'

On executing this code, you get the following result −

Output

Apple Juice
Apple Cake

AWK - Arrays

AWK has associative arrays and one of the best thing about it is – the indexes need not to be continuous set of number; you can use either string or number as an array index. Also, there is no need to declare the size of an array in advance – arrays can expand/shrink at runtime.

Its syntax is as follows −

Syntax

array_name[index] = value

Where array_name is the name of array, index is the array index, and value is any value assigning to the element of the array.

Creating Array

To gain more insight on array, let us create and access the elements of an array.

Example

[jerry]$ awk 'BEGIN {
   fruits["mango"] = "yellow";
   fruits["orange"] = "orange"
   print fruits["orange"] "\n" fruits["mango"]
}'

On executing this code, you get the following result −

Output

orange
yellow

In the above example, we declare the array as fruits whose index is fruit name and the value is the color of the fruit. To access array elements, we use array_name[index] format.

Deleting Array Elements

For insertion, we used assignment operator. Similarly, we can use delete statement to remove an element from the array. The syntax of delete statement is as follows −

Syntax

delete array_name[index]

The following example deletes the element orange. Hence the command does not show any output.

Example

[jerry]$ awk 'BEGIN {
   fruits["mango"] = "yellow";
   fruits["orange"] = "orange";
   delete fruits["orange"];
   print fruits["orange"]
}'

Multi-Dimensional arrays

AWK only supports one-dimensional arrays. But you can easily simulate a multi-dimensional array using the one-dimensional array itself.

For instance, given below is a 3x3 two-dimensional array −

100   200   300
400   500   600
700   800   900

In the above example, array[0][0] stores 100, array[0][1] stores 200, and so on. To store 100 at array location [0][0], we can use the following syntax −

Syntax

array["0,0"] = 100

Though we gave 0,0 as index, these are not two indexes. In reality, it is just one index with the string 0,0.

The following example simulates a 2-D array −

Example

[jerry]$ awk 'BEGIN {
   array["0,0"] = 100;
   array["0,1"] = 200;
   array["0,2"] = 300;
   array["1,0"] = 400;
   array["1,1"] = 500;
   array["1,2"] = 600;

   # print array elements
   print "array[0,0] = " array["0,0"];
   print "array[0,1] = " array["0,1"];
   print "array[0,2] = " array["0,2"];
   print "array[1,0] = " array["1,0"];
   print "array[1,1] = " array["1,1"];
   print "array[1,2] = " array["1,2"];
}'

On executing this code, you get the following result −

Output

array[0,0] = 100
array[0,1] = 200
array[0,2] = 300
array[1,0] = 400
array[1,1] = 500
array[1,2] = 600

You can also perform a variety of operations on an array such as sorting its elements/indexes. For that purpose, you can use assort and asorti functions

AWK - Control Flow

Like other programming languages, AWK provides conditional statements to control the flow of a program. This chapter explains AWK's control statements with suitable examples.

If statement

It simply tests the condition and performs certain actions depending upon the condition. Given below is the syntax of if statement −

Syntax

if (condition)
   action

We can also use a pair of curly braces as given below to execute multiple actions −

Syntax

if (condition) {
   action-1
   action-1
   .
   .
   action-n
}

For instance, the following example checks whether a number is even or not −

Example

[jerry]$ awk 'BEGIN {num = 10; if (num % 2 == 0) printf "%d is even number.\n", num }'

On executing the above code, you get the following result −

Output

10 is even number.

If Else Statement

In if-else syntax, we can provide a list of actions to be performed when a condition becomes false.

The syntax of if-else statement is as follows −

Syntax

if (condition)
   action-1
else
   action-2

In the above syntax, action-1 is performed when the condition evaluates to true and action-2 is performed when the condition evaluates to false. For instance, the following example checks whether a number is even or not −

Example

[jerry]$ awk 'BEGIN {
   num = 11; if (num % 2 == 0) printf "%d is even number.\n", num; 
      else printf "%d is odd number.\n", num 
}'

On executing this code, you get the following result −

Output

11 is odd number.

If-Else-If Ladder

We can easily create an if-else-if ladder by using multiple if-else statements. The following example demonstrates this −

Example

[jerry]$ awk 'BEGIN {
   a = 30;
   
   if (a==10)
   print "a = 10";
   else if (a == 20)
   print "a = 20";
   else if (a == 30)
   print "a = 30";
}'

On executing this code, you get the following result −

Output

a = 30

AWK - Loops

This chapter explains AWK's loops with suitable example. Loops are used to execute a set of actions in a repeated manner. The loop execution continues as long as the loop condition is true.

For Loop

The syntax of for loop is −

Syntax

for (initialization; condition; increment/decrement)
   action

Initially, the for statement performs initialization action, then it checks the condition. If the condition is true, it executes actions, thereafter it performs increment or decrement operation. The loop execution continues as long as the condition is true. For instance, the following example prints 1 to 5 using for loop −

Example

[jerry]$ awk 'BEGIN { for (i = 1; i <= 5; ++i) print i }'

On executing this code, you get the following result −

Output

While Loop

The while loop keeps executing the action until a particular logical condition evaluates to true. Here is the syntax of while loop −

Syntax

while (condition)
   action

AWK first checks the condition; if the condition is true, it executes the action. This process repeats as long as the loop condition evaluates to true. For instance, the following example prints 1 to 5 using while loop −

Example

[jerry]$ awk 'BEGIN {i = 1; while (i < 6) { print i; ++i } }'

On executing this code, you get the following result −

Output

Do-While Loop

The do-while loop is similar to the while loop, except that the test condition is evaluated at the end of the loop. Here is the syntax of do-whileloop −

Syntax

do
   action
while (condition)

In a do-while loop, the action statement gets executed at least once even when the condition statement evaluates to false. For instance, the following example prints 1 to 5 numbers using do-while loop −

Example

[jerry]$ awk 'BEGIN {i = 1; do { print i; ++i } while (i < 6) }'

On executing this code, you get the following result −

Output

Break Statement

As its name suggests, it is used to end the loop execution. Here is an example which ends the loop when the sum becomes greater than 50.

Example

[jerry]$ awk 'BEGIN {
   sum = 0; for (i = 0; i < 20; ++i) { 
      sum += i; if (sum > 50) break; else print "Sum =", sum 
   } 
}'

On executing this code, you get the following result −

Output

Sum = 0
Sum = 1
Sum = 3
Sum = 6
Sum = 10
Sum = 15
Sum = 21
Sum = 28
Sum = 36
Sum = 45

Continue Statement

The continue statement is used inside a loop to skip to the next iteration of the loop. It is useful when you wish to skip the processing of some data inside the loop. For instance, the following example uses continue statement to print the even numbers between 1 to 20.

Example

[jerry]$ awk 'BEGIN {
   for (i = 1; i <= 20; ++i) {
      if (i % 2 == 0) print i ; else continue
   } 
}'

On executing this code, you get the following result −

Output

Exit Statement

It is used to stop the execution of the script. It accepts an integer as an argument which is the exit status code for AWK process. If no argument is supplied, exit returns status zero. Here is an example that stops the execution when the sum becomes greater than 50.

Example

[jerry]$ awk 'BEGIN {
   sum = 0; for (i = 0; i < 20; ++i) {
      sum += i; if (sum > 50) exit(10); else print "Sum =", sum 
   } 
}'

Output

On executing this code, you get the following result −

Sum = 0
Sum = 1
Sum = 3
Sum = 6
Sum = 10
Sum = 15
Sum = 21
Sum = 28
Sum = 36
Sum = 45

Let us check the return status of the script.

Example

[jerry]$ echo $?

On executing this code, you get the following result −

Output

AWK - Built-in Functions

AWK has a number of functions built into it that are always available to the programmer. This chapter describes Arithmetic, String, Time, Bit manipulation, and other miscellaneous functions with suitable examples.

S.No.	Built in functions & Description
1	Arithmetic Functions AWK has the following built-in arithmetic functions.
2	String Functions AWK has the following built-in String functions.
3	Time Functions AWK has the following built-in time functions.
4	Bit Manipulation Functions AWK has the following built-in bit manipulation functions.
5	Miscellaneous Functions AWK has the following miscellaneous functions.

AWK - User Defined Functions

Functions are basic building blocks of a program. AWK allows us to define our own functions. A large program can be divided into functions and each function can be written/tested independently. It provides re-usability of code.

Given below is the general format of a user-defined function −

Syntax

function function_name(argument1, argument2, ...) { 
   function body
}

In this syntax, the function_name is the name of the user-defined function. Function name should begin with a letter and the rest of the characters can be any combination of numbers, alphabetic characters, or underscore. AWK's reserve words cannot be used as function names.

Functions can accept multiple arguments separated by comma. Arguments are not mandatory. You can also create a user-defined function without any argument.

function body consists of one or more AWK statements.

Let us write two functions that calculate the minimum and the maximum number and call these functions from another function called main. The functions.awk file contains −

Example

# Returns minimum number
function find_min(num1, num2){
   if (num1 < num2)
   return num1
   return num2
}
# Returns maximum number
function find_max(num1, num2){
   if (num1 > num2)
   return num1
   return num2
}
# Main function
function main(num1, num2){
   # Find minimum number
   result = find_min(10, 20)
   print "Minimum =", result
  
   # Find maximum number
   result = find_max(10, 20)
   print "Maximum =", result
}
# Script execution starts here
BEGIN {
   main(10, 20)
}

On executing this code, you get the following result −

Output

Minimum = 10
Maximum = 20

AWK - Output Redirection

So far, we displayed data on standard output stream. We can also redirect data to a file. A redirection appears after the print or printf statement. Redirections in AWK are written just like redirection in shell commands, except that they are written inside the AWK program. This chapter explains redirection with suitable examples.

Redirection Operator

The syntax of the redirection operator is −

Syntax

print DATA > output-file

It writes the data into the output-file. If the output-file does not exist, then it creates one. When this type of redirection is used, the output-file is erased before the first output is written to it. Subsequent write operations to the same output-file do not erase the output-file, but append to it. For instance, the following example writes Hello, World !!! to the file.

Let us create a file with some text data.

Example

[jerry]$ echo "Old data" > /tmp/message.txt
[jerry]$ cat /tmp/message.txt

On executing this code, you get the following result −

Output

Old data

Now let us redirect some contents into it using AWK's redirection operator.

Example

[jerry]$ awk 'BEGIN { print "Hello, World !!!" > "/tmp/message.txt" }'
[jerry]$ cat /tmp/message.txt

On executing this code, you get the following result −

Output

Hello, World !!!

Append Operator

The syntax of append operator is as follows −

Syntax

print DATA >> output-file

It appends the data into the output-file. If the output-file does not exist, then it creates one. When this type of redirection is used, new contents are appended at the end of file. For instance, the following example appends Hello, World !!! to the file.

Let us create a file with some text data.

Example

[jerry]$ echo "Old data" > /tmp/message.txt 
[jerry]$ cat /tmp/message.txt

On executing this code, you get the following result −

Output

Old data

Now let us append some contents to it using AWK's append operator.

Example

[jerry]$ awk 'BEGIN { print "Hello, World !!!" >> "/tmp/message.txt" }'
[jerry]$ cat /tmp/message.txt

On executing this code, you get the following result −

Output

Old data
Hello, World !!!

Pipe

It is possible to send output to another program through a pipe instead of using a file. This redirection opens a pipe to command, and writes the values of items through this pipe to another process to execute the command. The redirection argument command is actually an AWK expression. Here is the syntax of pipe −

Syntax

print items | command

Let us use tr command to convert lowercase letters to uppercase.

Example

[jerry]$ awk 'BEGIN { print "hello, world !!!" | "tr [a-z] [A-Z]" }'

On executing this code, you get the following result −

Output

HELLO, WORLD !!!

Two way communication

AWK can communicate to an external process using |&, which is two-way communication. For instance, the following example uses tr command to convert lowercase letters to uppercase. Our command.awk file contains −

Example

BEGIN {
   cmd = "tr [a-z] [A-Z]"
   print "hello, world !!!" |& cmd
   close(cmd, "to")
   
   cmd |& getline out
   print out;
   close(cmd);
}

On executing this code, you get the following result −

Output

HELLO, WORLD !!!

Does the script look cryptic? Let us demystify it.

The first statement, cmd = "tr [a-z] [A-Z]", is the command to which we establish the two-way communication from AWK.
The next statement, i.e., the print command provides input to the tr command. Here &| indicates two-way communication.
The third statement, i.e., close(cmd, "to"), closes the to process after competing its execution.
The next statement cmd |& getline out stores the output into out variable with the aid of getline function.
The next print statement prints the output and finally the close function closes the command.

AWK - Pretty Printing

So far we have used AWK's print and printf functions to display data on standard output. But printf is much more powerful than what we have seen before. This function is borrowed from the C language and is very helpful while producing formatted output. Below is the syntax of the printf statement −

Syntax

printf fmt, expr-list

In the above syntax fmt is a string of format specifications and constants. expr-list is a list of arguments corresponding to format specifiers.

Escape Sequences

Similar to any string, format can contain embedded escape sequences. Discussed below are the escape sequences supported by AWK −

New Line

The following example prints Hello and World in separate lines using newline character −

Example

[jerry]$ awk 'BEGIN { printf "Hello\nWorld\n" }'

On executing this code, you get the following result −

Output

Hello
World

Horizontal Tab

The following example uses horizontal tab to display different field −

Example

[jerry]$ awk 'BEGIN { printf "Sr No\tName\tSub\tMarks\n" }'

On executing the above code, you get the following result −

Output

Sr No   Name    Sub Marks

Vertical Tab

The following example uses vertical tab after each filed −

Example

[jerry]$ awk 'BEGIN { printf "Sr No\vName\vSub\vMarks\n" }'

On executing this code, you get the following result −

Output

Sr No
   Name
      Sub
         Marks

Backspace

The following example prints a backspace after every field except the last one. It erases the last number from the first three fields. For instance, Field 1 is displayed as Field, because the last character is erased with backspace. However, the last field Field 4 is displayed as it is, as we did not have a \b after Field 4.

Example

[jerry]$ awk 'BEGIN { printf "Field 1\bField 2\bField 3\bField 4\n" }'

On executing this code, you get the following result −

Output

Field Field Field Field 4

Carriage Return

In the following example, after printing every field, we do a Carriage Return and print the next value on top of the current printed value. It means, in the final output, you can see only Field 4, as it was the last thing to be printed on top of all the previous fields.

Example

[jerry]$ awk 'BEGIN { printf "Field 1\rField 2\rField 3\rField 4\n" }'

On executing this code, you get the following result −

Output

Field 4

Form Feed

The following example uses form feed after printing each field.

Example

[jerry]$ awk 'BEGIN { printf "Sr No\fName\fSub\fMarks\n" }'

On executing this code, you get the following result −

Output

Sr No
   Name
      Sub
         Marks

Format Specifier

As in C-language, AWK also has format specifiers. The AWK version of the printf statement accepts the following conversion specification formats −

%c

It prints a single character. If the argument used for %c is numeric, it is treated as a character and printed. Otherwise, the argument is assumed to be a string, and the only first character of that string is printed.

Example

[jerry]$ awk 'BEGIN { printf "ASCII value 65 = character %c\n", 65 }'

Output

On executing this code, you get the following result −

ASCII value 65 = character A

%d and %i

It prints only the integer part of a decimal number.

Example

[jerry]$ awk 'BEGIN { printf "Percentags = %d\n", 80.66 }'

On executing this code, you get the following result −

Output

Percentags = 80

%e and %E

It prints a floating point number of the form [-]d.dddddde[+-]dd.

Example

[jerry]$ awk 'BEGIN { printf "Percentags = %E\n", 80.66 }'

On executing this code, you get the following result −

Output

Percentags = 8.066000e+01

The %E format uses E instead of e.

Example

[jerry]$ awk 'BEGIN { printf "Percentags = %e\n", 80.66 }'

On executing this code, you get the following result −

Output

Percentags = 8.066000E+01

%f

It prints a floating point number of the form [-]ddd.dddddd.

Example

[jerry]$ awk 'BEGIN { printf "Percentags = %f\n", 80.66 }'

On executing this code, you get the following result −

Output

Percentags = 80.660000

%g and %G

Uses %e or %f conversion, whichever is shorter, with non-significant zeros suppressed.

Example

[jerry]$ awk 'BEGIN { printf "Percentags = %g\n", 80.66 }'

Output

On executing this code, you get the following result −

Percentags = 80.66

The %G format uses %E instead of %e.

Example

[jerry]$ awk 'BEGIN { printf "Percentags = %G\n", 80.66 }'

On executing this code, you get the following result −

Output

Percentags = 80.66

%o

It prints an unsigned octal number.

Example

[jerry]$ awk 'BEGIN { printf "Octal representation of decimal number 10 = %o\n", 10}'

On executing this code, you get the following result −

Output

Octal representation of decimal number 10 = 12

%u

It prints an unsigned decimal number.

Example

[jerry]$ awk 'BEGIN { printf "Unsigned 10 = %u\n", 10 }'

On executing this code, you get the following result −

Output

Unsigned 10 = 10

%s

It prints a character string.

Example

[jerry]$ awk 'BEGIN { printf "Name = %s\n", "Sherlock Holmes" }'

On executing this code, you get the following result −

Output

Name = Sherlock Holmes

%x and %X

It prints an unsigned hexadecimal number. The %X format uses uppercase letters instead of lowercase.

Example

[jerry]$ awk 'BEGIN { 
   printf "Hexadecimal representation of decimal number 15 = %x\n", 15
}'

On executing this code, you get the following result −

Output

Hexadecimal representation of decimal number 15 = f

Now let use %X and observe the result −

Example

[jerry]$ awk 'BEGIN { 
   printf "Hexadecimal representation of decimal number 15 = %X\n", 15
}'

On executing this code, you get the following result −

Output

Hexadecimal representation of decimal number 15 = F

%%

It prints a single % character and no argument is converted.

Example

[jerry]$ awk 'BEGIN { printf "Percentags = %d%%\n", 80.66 }'

On executing this code, you get the following result −

Output

Percentags = 80%

Optional Parameters with %

With % we can use following optional parameters −

Width

The field is padded to the width. By default, the field is padded with spaces but when 0 flag is used, it is padded with zeroes.

Example

[jerry]$ awk 'BEGIN { 
   num1 = 10; num2 = 20; printf "Num1 = %10d\nNum2 = %10d\n", num1, num2 
}'

On executing this code, you get the following result −

Output

Num1 =         10
Num2 =         20

Leading Zeros

A leading zero acts as a flag, which indicates that the output should be padded with zeroes instead of spaces. Please note that this flag only has an effect when the field is wider than the value to be printed. The following example describes this −

Example

[jerry]$ awk 'BEGIN { 
   num1 = -10; num2 = 20; printf "Num1 = %05d\nNum2 = %05d\n", num1, num2 
}'

On executing this code, you get the following result −

Output

Num1 = -0010
Num2 = 00020

Left Justification

The expression should be left-justified within its field. When the input-string is less than the number of characters specified, and you want it to be left justified, i.e., by adding spaces to the right, use a minus symbol (–) immediately after the % and before the number.

In the following example, output of the AWK command is piped to the cat command to display the END OF LINE($) character.

Example

[jerry]$ awk 'BEGIN { num = 10; printf "Num = %-5d\n", num }' | cat -vte

On executing this code, you get the following result −

Output

Num = 10   $

Prefix Sign

It always prefixes numeric values with a sign, even if the value is positive.

Example

[jerry]$ awk 'BEGIN { 
   num1 = -10; num2 = 20; printf "Num1 = %+d\nNum2 = %+d\n", num1, num2 
}'

On executing this code, you get the following result −

Output

Num1 = -10
Num2 = +20

Hash

For %o, it supplies a leading zero. For %x and %X, it supplies a leading 0x or 0X respectively, only if the result is non-zero. For %e, %E, %f, and %F, the result always contains a decimal point. For %g and %G, trailing zeros are not removed from the result. The following example describes this −

Example

[jerry]$ awk 'BEGIN { 
   printf "Octal representation = %#o\nHexadecimal representaion = %#X\n", 10, 10
}'

On executing this code, you get the following result −

Output

Octal representation = 012
Hexadecimal representation = 0XA