CODEWITHSUNDEEP
regexperl
Proteen
Proteen

Reputation: 107

Matching a specific unknown multiple times using regex in perl

Example buffer:

Line 1 is my favourite line
Line 2 is bad
Line 3 is bad
Line 1 is still my favourite line

How do I use regex to match the 2 sentences that contain "Line 1" ?

Please note that the number '1' is not known to me, Only thing that's known is that the multiple occurences have the same number.

Upvotes: 1

Views: 115

Answers (2)

jing yu
jing yu

Reputation: 264

use 5.14.0; #including features 'say' and smart match (~~)
use warnings;


my @lines = (); #lines that occurred

while (<DATA>){
    my $line = /^(Line\s*+\d+\b)/ ? $1 : ''; #\b to avoid '2' matches '22'
    $line ~~  @lines and do {say "$line occurred before."; next;}; 
    push @lines, $line;
}


__DATA__
Line 1 is my favourite line
Line 2 is bad
Line 3 is bad
Line 1 is still my favourite line
Line 22 is bad

Output:

Smartmatch is experimental at ./tst.pl line 16.
Line 1 occurred before.

Whilst the above code is good to pick up line numbers that have occurred, it doesn't tell you what those lines are. If you do want that feature, then try the following code:

my @lines = ();

while (<DATA>){
    my $line = /^(Line\s*+\d+\b)/ ? $1 : ''; #\b to avoid '2' matches '22'
    next unless $line;
    push @lines, $_;
    my @occurred = grep {/$line/} @lines;
    @occurred > 1 and print for @occurred;
}

Output:

Line 1 is my favourite line
Line 1 is still my favourite line

Upvotes: 0

mpapec
mpapec

Reputation: 50677

my $s = q{Line 1 is my favourite line
Line 2 is bad
Line 3 is bad
Line 1 is still my favourite line
};

my ($l1, undef, $l2) = $s =~ /(Line \s* ([0-9]+) .*) [\w\W]*? (Line \s* \2 .*)/x;
print "$l1\n$l2\n";

output

Line 1 is my favourite line
Line 1 is still my favourite line

Upvotes: 1

Related Questions