textprocessing.md

1. grep
2. sed
3. awk
4. cut
5. tr
6. column
7. Cases

• Line count

grep

Parallel log grep:

ls /var/log/HOSTS/*/*.service.{a,b,c,d}.ru/archive/service.log-20241003*.gz 2>/dev/null | \
xargs -P 20 -I {} grep search_string {} > results.log

sed

# remove lines containing 'DEBUG'
sed '/DEBUG/d' service.log

# the same and update service.log with sed output (GNU)
sed -i '/DEBUG/d' service.log

# the same for BSD sed
sed -i '' '/DEBUG/d' service.log

awk

cut

tr

column

(printf "PERM LINKS OWNER GROUP SIZE DAY MONTH YEAR NAME\n"; ls -l | sed 1d) | column -t

Cases

Line count

grep -c . file.txt
grep -c ^ file.txt
awk 'END{print NR}' file.txt

# wc counts only New line characters, thus result may differ from grep and awk
wc -l file.txt