Re: New to Perl - locating max, subtracting initial, printing output, all across several files in differing directories

[blindluke]

Welcome to the Monastery.

"Learning Perl" is a great book to read cover to cover, doing the tasks, reading the chapters in order. Not the best reference to pick out sections from and taking out stuff specific to the task you have at hand. There are better sources of knowledge for this type of search, a great point to start would be the Perl FAQ. Read the FAQ, try to get the bits you need, play around, break stuff, and read "Learning Perl" at your own pace. Make sure that the edition you have is at least the fourth one, and if you have the most recent one (6th edition), take everything you read in Chapter 15 (the one about the smart match operator) with a grain of salt. Some things have changed.

As to your problem, here is a small example that could get you started.

#!/usr/bin/perl
use warnings;
use strict;

my $maxvalue;
my $value_at_3k;

while (my $line = <DATA>) {
    my ($time, $value) = split ' ', $line;
    if ($value > $maxvalue) {
        $maxvalue = $value;
    }
    unless (defined $value_at_3k) {
        if ($time > 3000) {
            $value_at_3k = $value;
        }
    }
}

print "Maxvalue was $maxvalue \n";
print "Value at 3k was $value_at_3k \n";

__DATA__
2999.98 280
2999.99 281
3000.01 285
3000.02 290
3000.03 320
3000.04 350
3000.05 325
3000.06 275
3000.07 270
3000.08 260
3000.09 235

Start by reading the FAQ, then try to change the code so it reads the data from a file (with the filename declared explicitly in the code). Then wrap the code in a subroutine that accepts the filename as a parameter. Then try iterating over a directory - first printing the filenames, then passing them as arguments to your sub. Then wrap it up. Good luck with the task ahead of you.

And it's good to learn. You are very right about this.

[jtrousde]

Thanks for the responses! I can sort of see what's going on with your sample code. I assume I'm doing to replace DATA with the directory. Maybe not. I'm trying to make sense of the FAQ at the moment.

I suppose I actually need to modify my original intentions.

The data I need to read exists in directories within a parent directory. For example, I have directories 1, 2, 3, ... , 10, all within a parent directory (call it /users/parent/).

Within those folders is a file (file will always have the same name) with six columns with headers. I'm assuming the $value and $time are the headers of columns within my data file.

Additionally, I no longer need the difference. What I'd like to do is just pull the maximum value of a specific column within a specific range of column 1 (3000 to 3640), as well as the value in column two corresponding to the time closest to 3000 in column 1. These will be exported as two different columns to another file: max value in column one, "initial" value in column 2.

Ugh. Perl's notation is so different from what I've seen elsewhere. I feel like I'm drinking from a firehouse for a fairly simple problem.

EDIT: Working on it, btw. But it will take some time. Thanks for getting me started :)