Perl, as one programming language, is not well designed, but it could be handy for simple text transformation.

§Command Options

The most common form of using Perl, for me, is one-liner, and it usually goes like this:

1
perl -i -pe 's!orignal!new!' <file>

Let decipher the cryptic options:

  • -p: do the transformation for matched patterns or print if unmatched
1
2
3
4
5
while (<>) {
      # your code goes here
} continue {
      print or die "-p destination: $!\n";
}

  • -i: in-place modification instead of writing to STDOUT

  • -e: evaluate CODE; where we specify the one-liner script

§Match multiple lines

1
perl -i -pe 'BEGIN { undef $/ }; s!orignal!new!s' <file>

Firstly, unset the variable of $/ so that the whole file is slurped into memory, instead of being read line by line. Secondly, append s option to the substitute code so that . could match newline.

§Reference: