awk¶

read awk command from file¶

awk -f func.awk

print selected fields¶

$0 is for the entire line

NF number of fields

awk '{print $1}'      #select 1st field
awk '{print $1, $NF}' #select 1st and last fields

reverse words in line¶

awk '{for (i=NF; i>=0; i--) {printf "%s ", $i;} printf "\n"}'

sum of a field¶

END will only execute once

/^total/{next} exclude first line starts with total

ls -l | awk '/^total/{next} {sum += $5}; END {print sum}'

group another field by one field¶

delimiter is space
awk associate array is a dictionary

NF > 8 only when the number of fields > 8

awk -F" " 'NF > 8 {v[$1] += $2}; END {for (k in v) {printf "%s:%d\n", k, v[k]}}'

select table rows based on columns¶

Regex patterns should be put in // and not quoted. Include the header as well.

#MaxDataDiskCount MemoryInMb Name NumberOfCores OsDiskSizeInMb ResourceDiskSizeInMb
az vm list-sizes --location "australiaeast" -o table \
  | awk '(($3 == "Name" || $3 ~ /^Standard_DS/) \
      && ($2 == "MemoryInMb" || $2 >= 32000) \
      && ($4 == "NumberOfCores" || ($4 >= 4 && $4 <= 20)))'

Convert string to number input: path_to_my_file.txt

NAME                    CPU(cores)   MEMORY(bytes)
dev-dashboard-6570x79   2m           533Mi
dev-task-734449-2h882   1096m        4442Mi
dev-task-734449-d5k9m   1m           4531Mi
dev-task-734449-d996d   37m          6232Mi

code: select names starting with dev-task and cpu > 1000m. $2 + 0 will convert string to number.

pods_to_delete=$(kubectl top po -n dev \
| awk '$1 ~ /^dev-task/ && $2 ~ /m/ { sub("m", "", $2); if ($2 + 0 > 1000) print $1 }')

# test
awk '$1 ~ /^dev-task/ && $2 ~ /m/ { sub("m", "", $2); if ($2 + 0 > 1000) print $1 }' "path_to_my_file.txt"