An AWK program reads in a file and divides this input logically into records (defaults to each line read). Furthermore, while processing one record, it is logically divided into fields. Fields are by default separated by white-space characters, not including the newline character which is the default RS (record separator). $0 represents the entire record. $x represents the x-th field.

An AWK program is defined by a set of what are known as “pattern action” pairs. A pattern is a condition to be satisfied for the record, and when it does, the corresponding action is executed for that record. See the various types of patterns in the section “Code Blocks” further down.

A special thank goes to my colleague Cesar A. Murakami, for this tutorial’s main content:

Some Special variables (Input/Output):

• FS: Input field separator (char or regex)
• RS: Input record separator
• NF: quantity of fields of the current record
• NR: Current input record number
• OFS: Output field separator
• ORS: Output record separator

Some more special variables (files and match):

• ARGC: number of parameters passed to awk
• ARGV: the parameters (files)
• FILENAME: the current file being processed
• FNR: record number of FILENAME
• RSTART: starting position of string matched by match() function
• RLENGTH: length of string matched by match() function

Conditionals:

if (var1 > 2) {
  print "greater than 2";
  var1 = var1 + 1;
}
if ( var1 in arr )
  printf "%d is an array index\n",var1;
else {
  arr[var1] = var2;
  printf "%d was not an array index\n",var1;
}

Loops:

for (i=1; i<=10; i++) {
  print i;
}
for (i in arr) {
  print arr[i];
}
while ( x != 0 ) {
  do_something;
  print x;
}

Functions:

function mysum(param1, param2) {
  return param1 + param2;
}

Code Blocks ('PATTERN {ACTION}'):

BEGIN {
  # Code to be executed before record processing.
}
/regex/ {
  # Code to be executed when $0 contains a substring that are matched by regex.
}
$1 == "XYZ" {
  # Code to be executed when the first field is equal to "XYZ".
 }
$2 ~ /regex/ {
  # Code to be executed when the second field contains a substring that are matched by regex.
}
! PATTERN {
  # Code to be executed when PATTERN is not matched / satisfied.
}
PAT1, PAT2 {
  # Code to be executed for record range (PAT1 is matched on start record, PAT2 on final, both lines included in the range).
}
{
  # Code to be executed for every record.
}
END {
  # Code to be executed after record processing.
}