Shell Programming and Scripting

Hello,

Although I have found similar questions, I could not find
advice that could help with my problem. The issue:

I am trying to replace all occurrences of a regex, but
I cannot make the regex groups work together.

This is a simple input test file:

Code:

The Vedanta Philosophy
~~~~~~~~~~~~~~~

Body text

This is what the output should be:

Code:

## The Vedanta Philosophy

Body text

This is my code:

Code:

#!/usr/bin/perl -w

use strict;

while (<>) {
    s/(^.*)\n(^~+)/## $1/igm;
    print;
}


sub usage {
    (my $progname = $0) =~ s/.*[\\\/]//;
    die "Usage: $progname [<file>...]\n";
}

The two groups

Code:

(^.*)

and

Code:

(^~+)

match the test file independently, but when
I put them together in the s///igm line, they do not work at all.

I read the perl regex introduction and tutorial, and
searched for hints where I am making a mistake, with no avail.



Thank you for any help or indication on how to solve this problem.

Hi samask,

Some issues:

1.- while (<>) reads one line at a time and your regex tries to work over multiple lines.
2.- '^' marks the beginning of the line and it is zero-width, put it out of the parentheses.
3.- There is no need of the 'g' switch. Your regex only will match at most once.
4.- No need of 'i' switch. You are not using any alphabetic character.

One solution:

Code:

$ cat script.pl
use warnings;
use strict;

local $/;
my $file = <>;
$file =~ s/^(.*)\n^(~+)/## $1/m;
print $file;
$ cat infile
The Vedanta Philosophy
~~~~~~~~~~~~~~~

Body text
$ perl script.pl infile
## The Vedanta Philosophy

Body text

Regards,
Birei

you can do below if interested with one liner:-

Code:

perl -07 -wlpe 's/(.*)\s+~+/## $1/s ;' infile.txt

Wow, thank you so much birei!

Thank you for the explanation, and also for your code.
It is just elegant, and it works perfectly.

Thank you so much once again.

Sam

---------- Post updated at 11:31 ---------- Previous update was at 11:29 ----------

Thank you ahmad.diab!

I like one liners, but, as I will add more regex logic, it may be more practical as separate script.

Thank you!

Sam