A bit of a stab in the dark without seeing a sample of your data files but, do you need to chomp off the line terminator from each $line? If the text you wish to match is in the middle of the $record then the line terminator at the end of the pattern means you will never succeed.

I hope my guess is close to the mark and that this is helpful.

Without seeing any sample input, the following is just guesswork.

You're currently matching the entire record against the fourth ID; I think you want to match the fourth element of each record against all IDs.

I suspect

foreach my $records(@merged_array){
    
    if($records =~ /$sample_ids[3]/){
        print $records;
    }
}
[download]

should probably be more like

foreach my $records (@merged_array) {
    my @elements = split /,/, $records;

    for my $sample_id (@sample_ids) {
        if ($elements[3] =~ /$sample_id/) {
            print $records;
        }
    }
}
[download]

You were right I need to match the fourth element against each of the records. I added the code you suggested and no value was returned I decided to do a check and I did this:

foreach my $records (@merged_array) {
    my @elements = split /,/, $records;

        print $elements[0];
[download]

The Output from this was:  "John Doe"John Doe"John Doe"John Doe"John Doe

I realized that some of the fields themselves had commas within and that I couldn't consistently get the field I wanted by "print18" in this case So back to the drawing board for the next few minutes

It looks like your iterating over lines from Errors-email.txt (and it would have helped to see some of these lines, since they contain the IDs):

...
open (ERRORFILE,"<", "Errors-email.txt") || die "File not found\n";  
        @error_array = <ERRORFILE>;
...
foreach my $line(@error_array){
#
# Some $line substitutions here...
...
push(@sample_ids,$line);
...
[download]

Then you have the following:

...
foreach my $records(@merged_array){
    
    if($records =~ /$sample_ids[3]/){
...
[download]

Where @merged_array contains lines from your cvs file.

Does @sample_ids contain only the IDs? Did you try printing $sample_ids[3] to see what you get? Since you successfully used UCD67832 in the regex, it's evident that $sample_ids[3] does not equal UCD67832. What does it equal? Is UCD67832 a different element of @sample_ids? Does @sample_ids comtain whole lines instead of IDs?

I can't see that you've actually isolated the ID (unless it's via the set of substitutions), so I'd tend to focus on grabbing the ID from $line around those substitutions, and then push(@sample_ids,$id);

Hope this helps!

Thanks a lot for this post. I went back and checked

Here is what the contents of the error file look like

**************************************
Errors for file: Merged_CSVfiles_1.txt
**************************************

Sample_identifier 11SS00342 / IRD Clinic  clinic_HIV000140180:
*  The following fields contain invalid values: Blah blah blah blah
*
Sample_identifier 11SS00336 / IRD Clinic  clinic_HIV000140174:
*  The following fields contain invalid values: Yada yada yada
*
Sample_identifier 11SS00303 / IRD Clinic  clinic_HIV000140141:
*  The following fields contain invalid values: yeah yeah yeah
[download]

Just to check that the array @sample_ids contains the actual ids I printed the contents of the sample id' and got the following

                    
    push(@sample_ids,$line);
        print(@sample_ids);
[download]

I get the following output which is the list of the ID's (this is just a truncated list, just 8 of the 438 errors) UCD11-02580-V UCD11-02581-P UCD11-02581-V UCD11-02583-P UCD11-02583-V UCD11-02584-P UCD11-02584-V UCD11-02585-P I also did the following as a check

                    
push(@sample_ids,$line);
                        
    }# End of the if loop
}#End of the for-each loop        

print $sample_ids[3];
[download]

and I got: 11SS00304

What I did realized though was that:

foreach my $records (@merged_array) {
    my @elements = split /,/, $records;

        print $elements[0];
[download]

The Output from this was: "John Doe"John Doe"John Doe"John Doe"John Doe

I realized that some of the fields themselves had commas within and that I couldn't consistently get the field I wanted by "print18" in this case

You're most welcome, MyJeweledPerls!

If getting the field you want from the cvs lines is still presenting a problem, consider using Text::CSV to open and parse your cvs file. Its documentation shows how to do this, so you can get that field.