rnaeye has asked for the wisdom of the Perl Monks concerning the following question:

Hi guys, I have following code that takes multiple files, split fields, do simple arithmetic operation, and write results of each file into a separate output file. Now, I need to modify the code, and print the result of each input file as a seperate column in single output file (instead of multiple output files). I was wondering if you guys can give me a pointer. Thank you. 

 
#!/usr/bin/perl
use warnings;
use 5.12.4;
# USAGE: <perlscript> <*.txt>



my (%hash, $key, $value);
my ($col1, $col5, $sum, $rowdata, $line, $outfile, $infile, @column5);



foreach my $file (@ARGV) {
        
    #name output file
    $outfile = $file . '.csv';
    
    #file check
    die "Oops! A file called '$outfile' already exists.\n" if -e $outf
+ile;
    open (IN,  "<", "$file" )       or die "Cannot open file $!";
    open (OUT, ">", "$outfile")     or die "Cannot open file: $outfile
+ $!";
    
    while (<IN>) {

        chomp ($line = $_);
        next if $line =~ /NoCoordinateCount.*/;
        next if $line =~ /^$/;
        ($col1, undef, undef, undef, $col5 ) = split(/\s/, $line);
        $sum += $col5;
        $key = $col1;
        $value = $col5;
        $hash{$key} = $value;
    }
    

    foreach (sort keys %hash) {
        print OUT $_, "," , $hash{$_}/$sum, "\n";
    }

    $sum = 0; #reset sum for each file


}

[download]

Replies are listed 'Best First'.
Re: Appending new column data to CSV file?
by moritz (Cardinal) on Aug 04, 2013 at 16:44 UTC

    Here is a pointer: Instead of printing the data, store it in a data structure that is declared outside the outer foreach loop. Then when you are done with all input files, write the output into a single file.

    If you haven't yet worked with nested data structures in Perl, I can recommend perlreftut, perldsc and perllol (the latter for arrays of arays).

    Perl 6 - the future is here, just unevenly distributed
[reply]
Re: Appending new column data to CSV file?
by Cristoforo (Curate) on Aug 04, 2013 at 20:47 UTC
    In the while loop, these 2 statements: 
            $sum += $col5;
        push @numbers, $col5;

    [download]
    $sum and @numbers need to be declared before the while loop that reads the file. Then after each file is read, statements: 
        while (my ($i, $num) = each @numbers) {
        push @{ $output[$i] }, $num/$sum;
    }

    [download]
    You would need to declare @output before the foreach loop begins.

    Then after all the files have been read and the info collected, you could print it out like: print OUT join(",", @$_), "\n" for @output;

[reply]
[d/l]
[select]
Re: Appending new column data to CSV file?
by Laurent_R (Canon) on Aug 04, 2013 at 19:06 UTC

    Instead of printing your hash values divided by $sum, change directly your hash and push a reference to it into a global array of hashes. Once you have readd all your files, use that data structure to print your data structure in the required format. 

    $_ = $_/$sum foreach (values %hash);
push @AoH, \%hash;

    [download]

    It would probably be better to declare your hash within the "foreach my file (@ARGV)" loop to make sure you get a brand new and empty hash each time, but, on the other side, @AoH needs to be declared at the top of your program.

    At the end, you just need to go through the @AoH structure to print your output. (Of course, this is just one possibility, there are quite a few others.)

[reply]
[d/l]
Re: Appending new column data to CSV file?
by locked_user sundialsvc4 (Abbot) on Aug 05, 2013 at 12:35 UTC

    Aside from suggesting Text::CSV, consider the very-useful join() function, which can do simple things like join("," 1, 2, 3); with ease.

Back to Seekers of Perl Wisdom