Pattern scanning and text processing language
Print a single field from every line
awk '{print $3}' [path/to/file]
Print the last field, whatever the field count
awk '{print $NF}' [path/to/file]
Split on something other than whitespace
awk -F ':' '{print $1, $7}' /etc/passwd
Print a field only from lines matching a pattern
awk '/[error]/ {print $1}' [path/to/file]
Keep the lines where a field passes a numeric test
awk '$3 > [100]' [path/to/file]
Combine a field test with an action
awk -F ',' '$2 == "[EU]" {print $1}' [path/to/file]
Sum a column
awk '{sum += $1} END {print sum}' [path/to/file]
Average a column
awk '{sum += $2} END {print sum / NR}' [path/to/file]
Count how often each value appears
awk '{count[$1]++} END {for (key in count) print count[key], key}' [path/to/file]
Select a range of line numbers
awk 'NR >= [10] && NR <= [20]' [path/to/file]
Skip the header line
awk 'NR > 1' [path/to/file]
Read one separator and write another
awk -F ',' -v OFS='\t' '{print $1, $3}' [path/to/file]
Pass a shell value into the program
awk -v threshold=[100] '$1 > threshold' [path/to/file]
Drop duplicate lines while keeping the original order
awk '!seen[$0]++' [path/to/file]
Print a formatted table with a header
awk -F ':' 'BEGIN {printf "%-20s %6s\n", "USER", "UID"} $3 >= 1000 {printf "%-20s %6d\n", $1, $3}' /etc/passwd
awk [-F fs] [-v var=val] [-f progfile] [--posix] [--] 'program' [files]
-F fs
Field separator (fs); default whitespace or TAB
-f file
Read awk program from file instead of command line
-v var=val
Assign val to var before program runs (repeatable)
--
End options; treat following as filenames
-V
Print version and exit (gawk)
--help
Print help and exit (gawk)
--posix
Enforce POSIX compatibility (gawk)
-mf n
Limit function args to n (debugging; gawk)
-mr n
Limit record size to n bytes (debugging; gawk)
-W traditional
Use original awk behavior (gawk)
awk is a pattern-scanning and text-processing language designed for extracting and transforming structured data. It reads input line by line, splits each line into fields, and applies user-defined rules consisting of patterns and actions.
An awk program is a sequence of pattern { action } rules. For each input line, awk tests the patterns and executes the associated actions for any that match. If no pattern is given, the action applies to every line. If no action is given, matching lines are printed.
Fields are accessed as $1, $2, etc., with $0 representing the entire line. The default field separator is whitespace, changeable with -F. Built-in variables include NR (current line number), NF (number of fields in current line), FS (field separator), and OFS (output field separator).
Special patterns BEGIN and END execute actions before and after all input is processed, useful for initialization and summary output. Awk supports variables, arrays, arithmetic, string functions, printf formatting, and control flow statements, making it a complete programming language for text processing.
$0: Full line $1: First field NF: Fields count NR: Record (line) number FILENAME: Current input file FS / OFS: Input / output field separator RS / ORS: Input / output record separator Use $(n) for nth field
Different awk implementations (gawk, mawk, nawk) have varying feature sets. Regular expressions and string functions may behave differently across implementations. Associative arrays are unordered. Floating-point arithmetic may produce rounding errors. Very large files are processed efficiently but complex programs with many arrays can consume significant memory.
awk was created by Alfred Aho, Peter Weinberger, and Brian Kernighan at Bell Labs in 1977, with the name derived from their initials. It was redesigned and expanded in 1985 as "new awk" (nawk). GNU awk (gawk) added many extensions including networking and internationalization. Awk is a standard POSIX utility available on virtually all Unix-like systems.
dnf: sudo dnf install gawk
pacman: sudo pacman -S gawk
apk: sudo apk add gawk
zypper: sudo zypper install gawk
brew: brew install gawk
nix: nix profile install nixpkgs#gawk
sed(1), grep(1), cut(1), perl(1), gawk(1)
[Source code](https://cgit.git.savannah.gnu.org/cgit/gawk.git)
[Homepage](https://www.gnu.org/software/gawk/)
[Documentation](https://www.gnu.org/software/gawk/manual/)