Extracting non-consecutive but related items from an array.

[blackadder]

Hi Guys;

I have data like this;

server_A, Perl, UK
server_A, Word, UK
server_A, Outlook, UK
server_A, Excel, UK
server_B, Reuters, NL
server_B, TradeXL, NL
server_B, Thompsons, NL
server_B, Bloomberg, NL
server_B, Tibco, NL
server_c, BasketLink, USA
server_c, Evolution, USA
server_c, Lotus, USA
server_c, TIB, USA
server_A, Python, UK

And I need to produce this output;

server_A
         Perl  UK
         Word  UK
         Outlook  UK
         Excel  UK 
         Python  UK

server_B
         Reuters  NL
         TradeXL  NL
         Thompsons  NL
         Bloomberg  NL
         Tibco  NL

server_c
         BasketLink  USA
         Evolution  USA
         Lotus  USA
         TIB  USA

So, I wrote this code;

#! c:/perl/bin/perl.exe
use strict;
open (LST,"c:/work/test_data.lst") || die "\n$!\n";
chomp (my @data_array = <LST>);
my $snap_shot;
my %seen;
my @cleaned_data;
foreach my $data ( @data_array)
{
    my $rec;
    my ($server, @info) = split (/,/,$data);
    if (! $seen{$server})
    {
        $rec->{Server_Name} = $server;
        print "$server\n";
        $seen{$server}++;
        print "\t@info\n";
        @{$rec->{Apps_Info}} = @info;        
    }
    else
    {
        print "\t@info\n";
        @{$rec->{Apps_Info}} = @info;
    }    
    push (@cleaned_data, $rec);
}

Basically I need help with the logic of the script. It works fine if all entries relating to one item are repeated consecutively. However if entries were not listed consecutively, then this script will not produce the required output. I suppose the easy way around it is to sort the list alphabetically first and then run the script. The way I thought of doing it is by storing and indexing server names in an array then looping through the array and the list to pluck out the related item, this approach was taking ages and involved a lot of coding. So I just wondered how would you or could this be achieved easier and faster but from within the code?

Thanks

[robartes]

This is just the thing hashes exist for:

#Untested
use strict;
my %data;
#assume file is opened as <INPUT>
while (<INPUT>) {
  chomp;
  my ($server,@apps_info)=split /,/;
  push @{$data{$server}}, \@apps_info;
}
foreach my $server ( keys %data ) {
  print "$server\n\t";
  for ( @{$data{$server}} ) {
    print join " ", split /,/;
    print "\n\t";
  } 
}

Basically, pull all of the data into a hash with the server name as keys.

Update: Updated as per Tomtom's remark

[Tomtom]

Isn't it supposed to be $server instead of $_ ?

[robartes]

Isn't it supposed to be $server instead of $_ ?

Yup - that's why it says #Untested :). Thanks!

[blackadder]

Yes, I think its $server too....Thanks guys.

But, Can I have simple breakup of this code?

print
    $_,
    $/,
    map { "\t@$_$/"} @{$clean_data{$_}}
    for
        sort
            keys %clean_data;

And, why did we have to reference \ @apps_info? couldn't we use it without referencing?

[PodMaster]

Unless I'm overlooking something, you're overcomplicating it

use strict;
use warnings;

my %clean_data;
chomp (my @data_array = <DATA>);


for my $data ( @data_array)
{
    my ($server, @info) = split (/,/,$data);
    $data = \@info;
    push @{$clean_data{$server}}, $data;
}

#use Data::Dumper;die Dumper\%clean_data;
print
    $_,
    $/,
    map { "\t@$_$/"} @{$clean_data{$_}}
    for
        sort
            keys %clean_data;

__DATA__
server_A, Perl, UK
server_A, Word, UK
server_A, Outlook, UK
server_A, Excel, UK
server_B, Reuters, NL
server_B, TradeXL, NL
server_B, Thompsons, NL
server_B, Bloomberg, NL
server_B, Tibco, NL
server_c, BasketLink, USA
server_c, Evolution, USA
server_c, Lotus, USA
server_c, TIB, USA
server_A, Python, UK

[Anonymous Monk]

my %servers;
while (<>) {
    chomp;
    next unless 3 == my ($server, $loc, $country) = split /,\s*/;
    push @{$servers{$server}}, [$loc, $country];
}
local $" = " ";
while (my ($server, $info) = each %servers) {
    print "$server\n";
    print "\t@$_\n" foreach @$info;
}

[sapnac]

#use strict;
use FileHandle;

# Global Declaration 
my($spool,%serverdet);

open (spooler ,"data.dat") or  die "Error opening the  file\n";

while($spool=<spooler>) {
    chomp($spool);
    ($server,$dat1,$dat2)=split(/,/,$spool);
    if($serverdet{$server}) {
        $serverdet{$server} .= ";".$dat1.",".$dat2;
    }else{
        $serverdet{$server} = $dat1.",".$dat2;
    }
}

foreach $key (sort keys %serverdet) {
    print "Server $key\n";
    @val = split(/;/,$serverdet{$key});
    foreach $val  (sort @val)    {
     print $val."\n";
    }
}

Hope this helps!



code tags added by holli

[TedPride]

use strict;
use warnings;

my %h;
while (<DATA>) {
    chomp; @_ = split /, /, $_, 2;
    push @{$h{lc($_[0])}}, $_[1];
}
for (sort keys %h) {
    print "$_\n";
    print "         $_\n" for sort @{$h{$_}};
    print "\n";
}

__DATA__
server_A, Perl, UK
server_A, Word, UK
server_A, Outlook, UK
server_A, Excel, UK
server_B, Reuters, NL
server_B, TradeXL, NL
server_B, Thompsons, NL
server_B, Bloomberg, NL
server_B, Tibco, NL
server_c, BasketLink, USA
server_c, Evolution, USA
server_c, Lotus, USA
server_c, TIB, USA
server_A, Python, UK