awk

Quick, line-oriented text processing. Awk reads records (lines), splits into fields, and runs actions on matches.

Basics

• Default field separator: whitespace. Fields: $1, $2, …, $NF; whole line: $0
• Change field separator: -F',' or FS=","; output FS: OFS
• Blocks: BEGIN { init } pattern { action } END { finalize }

Common one-liners

• Print column 1 and 3, CSV input

awk -F, '{ print $1, $3 }' file.csv

• Filter rows matching regex and print 2nd field

awk '/ERROR/ { print $2 }' app.log

• Sum a numeric column and print total

awk '{ sum += $3 } END { print sum }' data.txt

• Set output delimiter

awk -F, 'BEGIN{OFS=","} { print $1, $2, $5 }' file.csv

• Replace value in a column (in-memory)

awk '$2=="staging" {$2="prod"} {print}' OFS='\t' input.tsv

• Header + computed column

awk 'BEGIN{OFS=","; print "name,size_kb"} {print $1, $2/1024}' sizes.txt

Patterns and conditions

• Range match (between markers)

awk '/BEGIN/,/END/' file

• Numeric condition

awk '$5 > 100' data.txt

• Multiple conditions

awk '$2=="ok" && $5>=200 {print $1,$5}' OFS="," file

Editing files in-place (GNU awk via shell)

• With temp file

awk '{gsub(/foo/,"bar"); print}' input > tmp && mv tmp input

Tips

• Use -v to pass variables

awk -v thr=10 '$3>thr' file

• Use -F'\t' for TSV; for multi-char FS: -F"[|]+"
• For complex scripts, put in a .awk file and run: awk -f script.awk input