Perl: Filtering through an Array to Make a New Array

Question

I'm trying to filter an array of a delimited text file in my program. The array from this text file looks like this:

YCL049C                   1     511.2465  0 0 MFSK
YCL049C                   2    4422.3098  0 0 YLVTASSLFVALT
YCL049C                   3    1131.5600  0 0 DFYQVSFVK
YCL049C                   4    1911.0213  0 0 SIAPAIVNSSVIFHDVSR
YCL049C                   5     774.4059  0 0 GVAMGNVK
..
.

and the code I have for this section of the program is:

my @msfile_filtered;
my $msline;
foreach $msline (@msfile) {

    my ($name, $pnum, $m2c, $charge, $missed, $sequence) = split (" ", $msline);
    if (defined $amino) {

        if ($amino =~ /$sequence/i) {

            push (@msfile_filtered, $msline);

        }

    }
    else {

        push (@msfile_filtered, $msline);

    }

}

$amino will just be a letter that will be input by the user, and corresponds to the last field $sequence. It is not essential that the user actually inputs $amino, so I need to duplicate this array and keep it unchanged if this is the case (hence the else statement). At the minute the @msfile_filtered array is empty, but I am unsure why, any ideas?

EDIT: just to clarify, there is only one space between each field, I copy and pasted this from notpad++, so extra spaced were added. The file itself will only have one space between fields.

Thanks in advance!

Joe Z · Accepted Answer · 2013-11-03 14:56:49Z

3

The regex that tries to find matching rows is backwards. To find a needle in a haystack, you need to write $haystack =~ /needle/, not the other way around.

Also, to simplify your logic, if $amino is undef, skip the loop entirely. I would rewrite your code as follows:

if (defined $amino)
{
    foreach $msline (@msfile)
    {
        my ($name, $pnum, $m2c, $charge, $missed, $sequence) = split(" ", $msline);
        push @msfile_filtered, $msline if ($sequence =~ /$amino/i);
    }
} else
{
    @msfile_filtered = @msfile;
}

You could simplify this further down to a single grep statement, but that begins to get hard to read. An example of such a line might be:

@msfile_filtered =
    defined $amino
        ? grep { ( split(" ", $_ ) )[5] =~ /$amino/i } @msfile
        : @msfile;

edited Nov 3, 2013 at 14:56

answered Nov 3, 2013 at 14:28

Joe Z

18k3 gold badges30 silver badges39 bronze badges

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

5 Comments

Joe Z Over a year ago

I used the example that was in the original question, right at the top. (The five rows that begin with YCL049C.)

user2941526 Over a year ago

Thanks for your help! when no input for $amino is entered, the loop to skipped find and the array remains the same, but when I enter a value for $amino, the @msfile_filtered does not contain only sequences with the specified letter. In fact it contains nothing at all. I'll have a play around and report back

user2941526 Over a year ago

I am using getopts, and the statement "my $amino = (defined $opt_a);" to try to turn the command line enter to $amino. When I enter "-a A" into the command line for example, "print $amino" returns "1".

hwnd Over a year ago

One reason this is not working is because you are trying to match your pattern in reverse, you need to flip flop $sequence and $amino here. Also you could just use split directly in this case if you know for sure there is always one space between the given fields. e.g foreach (@msfile) { push @msfile_filtered, $_ if (split)[5] =~ /$amino/i; }

Joe Z Over a year ago

@user2941526: I don't think you want $amino = (defined $opt_a). I think just $amino = $opt_a will do.

foibs · Accepted Answer · 2013-11-03 14:31:31Z

1

The split is should take more than one whitespaces, and the regex vars are vice versa.

First debug to check that values are correct after the split.

Also, you must swap your regex variables like this:

 if ($sequence =~ /$amino/i) {

Now you're checking if $amino contains $sequence, which obviously it doesn't

answered Nov 3, 2013 at 14:31

foibs

3,4141 gold badge21 silver badges13 bronze badges

3 Comments

user2941526 Over a year ago

split is working corretly, the copy and paste from notepad++ was a bit misleading. The file itself only has one space between fields

Joe Z Over a year ago

@folbs: The regex is indeed the root cause.

RobEarl Over a year ago

Splitting on a single literal space is a special case. It trims leading whitespace then split on /\s+/. See perldoc.perl.org/functions/split.html

Collectives™ on Stack Overflow

Perl: Filtering through an Array to Make a New Array

2 Answers 2

5 Comments

3 Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

5 Comments

3 Comments

Your Answer

Sign up or log in

Post as a guest

Related