Re: avoid repitive code while sorting hash arrays

[Chuck Fox]

Rob and all the other perl wonder workers who contribute to this list,

Awesome,  While I have been reading and writing perl for a few years 
now, I am always amazed at the code reduction that can occur when you 
properly apply the power of perl. This is the most instructive forum 
that I have ever joined.  Thanks to you and all the others that post to 
this forum.

Chuck

[EMAIL PROTECTED] wrote:

Ramprasad wrote:
 

suppose I have a hash like

     %users =(
          'cvs' => {
                       'uname' => 'cvs',
                       'uid' => 582,
                       'gid' => 500
                     },
            'radvd' => {
                         'uname' => 'radvd',
                         'uid' => 75,
                         'gid' => 75
                       },
            'goofy' => {
                         'uname' => 'goofy',
                         'uid' => 515,
                         'gid' => 515
                       },
                 );
Now I want to sort on either of the fields 'uname' 'uid' 'gid'
and  ascending or descending Then my sort function is pretty repitive
code. I have to check if the field is numeric or not and then if the
sorting is ascending or descending.
assume $hash = \%users;
......
 if($sortkey eq 'uname'){
if($direction eq 'a'){
   @allunames = sort{ $$hash{$a}{$sortkey} cmp $$hash{$b}{$sortkey} }
keys %hash;
} else {
   @allunames = sort{ $$hash{$b}{$sortkey} cmp $$hash{$a}{$sortkey} }
keys %hash;
}
    } else {
if($direction eq 'a'){
   @allunames = sort{ $$hash{$a}{$sortkey} <=> $$hash{$b}{$sortkey} }
keys %hash;
} else {
   @allunames = sort{ $$hash{$b}{$sortkey} <=> $$hash{$a}{$sortkey} }
keys %hash;
}
    }
..............
Can someone suggest a better way.
   

Hi Ram.

Will this do?

sort {
 my ($va, $vb) = map $$hash{$_}{$sortkey}, $direction eq 'a' ? ($a, $b) : ($b, $a);
 $sortkey eq 'uname' ? $va cmp $vb : $va <=> $vb;
} keys %$hash;
Cheers,

Rob