Using the 'and' operator inside Perl regex

Question

I'm using Perl to read in a file line by line and die if a condition is met. The condition being that a line has to begin with any of the letters AGCT and the rest of the line can only contain the letters AGCT.

my $fasta = $ARGV[0];

open(FASTA, $fasta) || die("Couldn't read file $fasta\n");
local $/ = "\n>";
while (my $line = <>) {
  if ( $line =~ /^[AGCT]/ && /[AGCT]/ ) {
    die;
  }
}
close FASTA;

I know the syntax in the regexp is wrong, I have tried many variations but can't get it, any ideas?

You open FASTA, but then you do nothing with it. Your while loop reads from ARGV instead. — melpomene
– melpomene, Commented Mar 16, 2018 at 13:47

simbabque · Accepted Answer · 2018-03-16 15:16:44Z

10

Your regular expression syntax is correct. Your Perl expression in the if condition is wrong.

if ( 
    $line =~ /^[AGCT]/   # this tests $line
    && /[AGCT]/          # this defaults to $_
) {

You need to use $line =~ // explicitly both times.

In addition, your second pattern does not do what you want. You are missing a *$, and it would make sense to include the beginning character, too. It should read

/^[AGCT][AGCT]*$/

As you can see, essentially you can just have one pattern and be done with it.

if ( $line =~ /^[AGCT]+$/ ) { ... }

You can shorten it even further by doing the opposite pattern and saying it shouldn't match.

if ( $line !~ /[^AGCT]/ ) { ... }

This is a bit confusing because of the double negation ( !~ and [^] though.

In any case, you should chomp your input first. I would write your program like this:

use strict;
use warnings;

# the \n removes the line number from die
open my $fh, '<', $fasta or die "Couldn't read file $fasta";

local $/ = "\n>";
while (my $line = <>) {
  chomp $line;
  die if $line =~ /[^AGCT]/;
}

That program will die if any of the lines contains something that is not A, G, C or T. I do believe that's what you wanted to do.

edited Mar 16, 2018 at 15:16

answered Mar 16, 2018 at 13:39

simbabque

54.4k8 gold badges77 silver badges141 bronze badges

Sign up to request clarification or add additional context in comments.

8 Comments

choroba Over a year ago

or $line =~ /[^AGCT]/

melpomene Over a year ago

@choroba That's slightly different in that simbabque's regex allows a trailing newline whereas yours doesn't. Also, simbabque's regex doesn't accept empty strings.

ysth Over a year ago

@choroba and if you were going to do that, it might as well be $line =~ y/AGCT//c

ikegami Over a year ago

$line =~ /^[AGCT]+$/ is not the same as $line !~ /[^AGCT]/. The latter matches an empty string.

ikegami Over a year ago

There's also the issue that $line might be terminated by \n>. It probably can even include newlines.

|

Collectives™ on Stack Overflow

Using the 'and' operator inside Perl regex

1 Answer 1

8 Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

8 Comments

Your Answer

Sign up or log in

Post as a guest

Related