First if you want to find the week number :
See DateTime

#!/usr/bin/perl -w 
use strict;
use DateTime;

my $date = q{20060911};
my $year = substr( $date, 0, 4 );
my $month = substr( $date, 4, 2 );
my $day = substr( $date, 6, 2 );
print DateTime->new( year => $year, month => $month, day => $day )->we
+ek_number;
[download]

hth,
PooLpi

Hello poolpi,

Thanks a lot for your help

For the moment, I do not have the right to install this new module. But I will have the right in a few days. Could you tell me in which directory I should install it ? Because all the .pm files are in several directories and I don't know in which directory the file should be.

a guide to installing modules

• remove the @ALL_WEEK_DAYS definitions (they're not relevant to the question at hand)
• turn the code that determines $week into a subroutine
• remove commented out code

The first argument of find() is a reference to a subroutine. This is where you tell find what you want done with the filenames it finds. Find will pass the names it finds into $_ . Say you wanted to print a list of all the files below your current directory:

find sub{print $_,"\n"}, "./"

The second parameter to find() takes an array of directories. If you only have one directory, as above, that is fine. Remember that perl passes parameters in a flat list so a single scalar will pass the same as an array of one element.

You can try this from the command line to get familiar with the concept:

%perl -MFile::Find -e 'find sub{print $_,"\n"}, "./"'

Of course find() really shines when iterating over a list of directories: find sub ( do-something }, @directories Note: This can return a lot of files as find() works recursively down through all the sub directories.

Find does not return the found files so this:  @array = find(); Doesn't work. You have to store the files passed to you yourself if you want to keep them. Like this: find sub { push @filelist, $_ }, @directories;

Of course if you subroutine is more complicated then it can be written separately and a reference to it passed to find().

sub my_Sub{
    do-something-with-$_
    .
    .
}
find \&my_Sub, @directories;
[download]

Hi again, steph_bow - I guess since I got you started down this road, I might as well keep you going. :) By the way, I note that you're still using a 'BEGIN' block in your code without any specific reason; you should probably read up on these special blocks (the section is called 'BEGIN, UNITCHECK, CHECK, INIT and END') to find out what they're for - and you shouldn't use them gratuitously.

File::Find is a nicely-documented module; you should read up on it by typing 'perldoc File::Find' (the exact case is important) at your command line. The short version, however, is that the 'find' function takes a sub reference as its first argument and a list of directories as the second - as starbolin mentioned. From what I see in your code, you mostly need to learn how to permute lists of directories - something like the following:

(pseudocode)
(A B, C) x (D, E, F) = (A/D, A/E, A/F, B/D, B/E, B/F, C/D, C/E, C/F)

In the example that I gave you the last time, I cheated a bit and used brace expansion in the glob statement - i.e., I used the shell to do the work (*bad, bad me...* :) This time, however, I'll do it "the right way", by using 'map':

use warnings;
use strict;

### Make up some numbers :)
my ($week, $date) = qw/36 250/;
my @regs = qw(LAR30 TYE68 SOK96);
### Interpolate the directory lists
@dirs = map {
    "../ALL_WEEKS/semaine_$week/semaine_${week}_TACOT/day_$date/r_$dat
+e/$_"
} @regs;

### Now that we have our list of directories, process them
find(\&process, @dirs);

sub process {
    ### Do stuff to the files returned by File::Find (represented by '
+$_')
    print "File name: $_\n";
}
[download]

-- 
Human history becomes more and more a race between education and catastrophe. -- HG Wells