Perl/Linux one-liners

Collection of tricks of Perl/Linux one-liners

Print hello world

perl -e 'print "Hello world!\n"'
perl -le 'print qq/Hello world!/'

Grep-like pattern searcher

cat input.txt | perl -ne 'print if /regex/'
seq 1 100 | perl -ne 'print if /^9/'
seq 1 100 | perl -ne 'print if /9$/'

Sed-like string replacement

cat input.txt | perl -pe 's/old/new/'
seq 1 10 | perl -pe 's/1/one/'

Exchange first and second column

cat input.txt | perl -alne 'print "$F[1]\t$F[0]"'
seq 1 10 | paste - - | perl -alne 'print "$F[1]\t$F[0]"'
seq 1 10 | paste - - | perl -alne 'print join("\t", @F[1,0])'
seq 1 10 | paste - - | perl -alne 'print join(chr(9), @F[1,0])'
seq 1 10 | paste - - | perl -alne 'print join(chr 9, @F[1,0])'
seq 1 10 | paste - - | perl -alne 'print join chr 9, @F[1,0]'

Print the length of each line

cat input.txt | perl -lne 'print length($_)'
seq 1 10 | perl -lne 'print length($_)'

Print the maximum line lengths (available for GNU version wc, not available for BSD version wc)

wc -L input.txt
wc --max-line-length input.txt

Split a text file into characters and count the number of them

cat input.txt | perl -pe 's//\n/g' | sort | uniq -c
seq 1 10 | perl -pe 's//\n/g' | sort | uniq -c

Print the sum of a list of numbers

seq 1 100 | perl -lne '$sum += $_; END{print $sum}'
seq 1 100 | perl -lne 'BEGIN {print qq/Sum: /} $sum += $_; END{print $sum}'
seq 1 100 | perl -ne 'BEGIN {print qq/Sum: /} $sum += $_; END{print "$sum\n"}'
seq 1 100 | perl -nE 'BEGIN {print qq/Sum: /} $sum += $_; END{say $sum}'

Min, max, sum, mean of a list of number using List::Util and datamash

seq 1 100 | perl -MList::Util=min,max,sum -lne 'push @s, $_;END{print join qq/\t/, min(@s),max(@s),sum(@s)}'
seq 1 100 | datamash --header-out min 1 max 1 sum 1 mean 1