Cut and paste from 425695...

use Date::Tie;
use strict;

tie my %dt, 'Date::Tie', 
    weekday => 1,
    week => 1,
    weekyear => 2005,
    hour => 0, 
    minute => 0, 
    second => 0;

$dt{weekday} = 0; # sunday

while ( $dt{weekyear} == 2005 )
{
    print "$dt{month}-$dt{day}-$dt{year}\n";
    $dt{week}++;
}

01-02-2005
01-09-2005
01-16-2005
01-23-2005
...
[download]

There are so many ways this could be done using Date::Manip, all of them slightly complex... ;)

But there is one other complexity to consider not related to manipulating the particular date module you use. And that is the concept of falling back to the starting date for a given week as the header for your week's summary.

What happens if the first day of the year is Wednesday, and you summarize everything that occurred in the first week of the year, given that the weeks begins on Sunday? You fall back to a previous year. Is this desired behavior?

What happens if the first day of the month is Wednesday, and you summarize everything that occurred in the first week of the month? ...same sort of problem. Things that happened on Wednesday April 1st will fall under the heading of the last Sunday in March. Again, is this desired behavior?

Some companies will actually use what they call a 4-5-4 calendar, where years always start on Sunday, and months always start on Sunday. This simplifies such problems. But then your company months start and end a day or two or three differently from real-world months. It takes a little getting used to, but ensures that when you're looking at everything that happened in week 1 of the year, week 1 started on a Sunday, and the entire week falls into the same year. And you can look at weeks of months in the same way; weeks never span months under such a calendar.

Now that there is some room, a minor rework of that code:

foreach my $week (sort keys %IssuesByWeek) {
        my $issues=$IssuesByWeek{$week};
        print "$week:\n\t",
              join ("\n\t", 
                map{ "$_ => $issues->{$_}" } 
                ( sort { $issues->{$b} <=> $issues->{$a} } 
                  keys %$issues )[0..4]
              ),"\n";
}
[download]

Also using the temporary value makes things a little easier to read later on and also saves on the unecessary extra hash lookups.

This is what I ended up with after your original suggestion. A hell of a line of code but it does make a pretty table:

print OUTPUT "<tr bgcolor=#C0C0C0><td><b>$newdate</b></td><td align=ri
+ght>#</td><td align=right>%</td></tr>\n<tr><td>",join ("</td></tr><tr
+><td>", map
{"<font size=-1>$_</td><td align=right><font size=-1>$IssuesByWeek{$da
+te}{$_}</td><td align=right><font size=-1>".int(($IssuesByWeek{$date}
+{$_}/$Weeks{$date})
*100)."\%</td>" } (sort {$IssuesByWeek{$date}{$b} <=> $IssuesByWeek{$d
+ate}{$a} } keys %{$IssuesByWeek{$date}} )[0..4]),"</tr><tr><td><b><fo
+nt size=-1>TOTAL TI
CKETS</font></b></td><td align=right>$Weeks{$date}</td></td><tr><td><b
+r><b>MACs</b></td><td align=right><br>$MACs{$date}</td><td> </td></tr
+>\n";
[download]

You should post that mess under Obfuscation

-Kevin

HTML::Template

First, your hash keys make it difficult to sort on them chronologically which presumably you would want to do:

for my $week ( map { $_->[0] }
               sort { $a->[1] <=> $b->[1] || $a->[2] <=> $b->[2] }
               map { [$_, split /\./] } keys %IssuesByWeek
             ) {
    # ...
}
[download]

for my $key ( keys %IssuesByWeek ) {
    # New key in format of YYYY-MM-DD
    $IssuesByWeek{ get_sunday( $key ) } = delete $IssuesByWeek{ $key }
+;
}
[download]

The second thing is that it is generally a waste to sort a list to find the top or bottom of the list. There are routines known as water-mark algorithms that do this much more efficiently. Instead of sorting the entire list and then throwing away most of the work, I would suggest using something similar to:

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

my @list = map { int(rand 1000) + 1 } 1 .. 50;

print "Top 5:\n";
print "\t$_\n" for top_x(\@list, 5);

sub top_x {
    my ($list, $x) = @_;
    $x--;
    my @top;
    $#top = $x;
    for my $item ( @$list ) {
        next if defined $top[ -1 ] && $item <= $top[ -1 ];
        for my $id ( 0 .. $#top ) {
            $top[ $id ] = $item and last if ! defined $top[ $id ];
            if ( $item > $top[ $id ] ) {
                @top[ $id .. $#top ] = ($item, @top[ $id .. $#top - 1]
+);
                #splice(@top, $id, 0, $item), pop @top;
                last;
            }
        }
    }
    return @top;
}
[download]

Second, there's only 53 Sundays in 2005, so it should be easy to calculate which ones go where.

my @sundays = ([]);
my @date = (2005, 1, 1);
$date[2]++ while Day_Of_Week(@date) != 7;

DATE: {
    push @sundays, \@date;
    @date = Add_Delta_Days( @date, 7 );
    last if $date[0] == 2006;
    redo DATE;
}

# Now, $sundays[1] == ( 2005, 1, 2 )
# $sundays[2] == (2005, 1, 9 )
# etc.
[download]

Alternately, you could just do:

my @sunday_date = Monday_of_Week(Week_of_Year( @starting_date ));
$sunday_date[2]--;
[download]