DavidBer has asked for the wisdom of the Perl Monks concerning the following question: ⭐ (hashes)

I am trying to read in a file. The structure of the file is city:country. What I need to do is to show cities within a country. My thought was to read the file into a hash, with it like this:

US:Miami
US:New York
US:Chicago

I understand that an @ array basically uses a number as the identifier of the value. How does a hash do it?

Originally posted as a Categorized Question.

  • Comment on Can a hash name have the same value twice.

Replies are listed 'Best First'.
Re: Can a hash name have the same value twice.⭐
by reptile (Monk) on May 03, 2000 at 00:58 UTC

    There is a much better answer in this case than storing multiple bits of information in the hash keys. I would personally never do that unless I absolutely had to. The example given in this question is actually a really simple data structure: One name to Multiple values, or in perl, a hash whose values are array references. Enough lecturing now. 

    my %countries = ();
my $foo = "US:New York";
my ($country, $city) = (split /:/, $foo);
push @{ $countries{$country} }, $city;

    [download]

    This code assumes you understand references in perl.

    Then to print all the cities within the country: 

    foreach my $city (@{$countries{$country}}) {
    print "$city\n";
}

    [download]

    It's trivial to include even more information than that as well; turn that array reference into a hash reference by city, containing maybe another hash reference with even more information or an array with a list of people who live there.

[reply]
[d/l]
[select]
Re: Can a hash name have the same value twice.⭐
by chromatic (Archbishop) on May 03, 2000 at 07:19 UTC
    If you need to store more information, consider a more complex data structure. reptile suggests a hash of lists. I would choose a hash of hashes. 
    my %data = (
    'US' => {
        Florida => 'Miami',
        Illinois => 'Chicago',
        New York => 'New York City' },
    'Canada' => {
        British Columbia => 'Alberta',
        Saskatchewan => 'Saskatoon',
        Ontario => 'Toronto' }
);

    [download]
    The appropriate perldoc pages are perldsc and perlref.

    (Oh, a hash is an associative array. That means that it takes the first element -- New York, for example -- and runs it through a hashing algorithm to generate a unique index, which helps it locate the key -- New York City -- in an internal data structure. If the hashing algorithm is good, "New York" will always come up with the same hash value, which is why duplicate keys are not very useful.)

[reply]
[d/l]
Re: Can a hash name have the same value twice.
by Roger (Parson) on Sep 10, 2003 at 03:59 UTC
    Create a two dimensional hash like this:
    @cities = qw/ US:Miami AU:Melbourne US:Chicago /;

map {my($co,$ci)=split/:/,$_; $n{$co}{$ci}=1 } @cities;

foreach $x (sort keys %n) {
   foreach (sort keys %{$n{$x}}) {
        printf "$x - $_\n";
   }
}

    [download]
    Which will print out:

    AU - Melbourne
    US - Chicago
    US - Miami
[reply]
[d/l]
Re: Can a hash name have the same value twice.
by turnstep (Parson) on May 02, 2000 at 23:35 UTC
    Here is a simple way to do it: 
    open(MYFILE, "$cityfile") || die "Could not open $cityfile: $!\n";
while(<MYFILE>) {
  chomp;
  ($city, $country) = split(/:/, $_);
  $place{$country} .= "$city#";
}
close(MYFILE); 
for $x (keys %place) {
  print "$x:\n";
  for (split(/#/, $place{$x})) {
    print "\t$_\n";
  }
}

    [download]
[reply]
[d/l]
Re: Can a hash name have the same value twice.
by Roger (Parson) on Sep 10, 2003 at 04:05 UTC
    This is to amend my previous post, the foreach loop can certainly be rewritten with a one liner:
    map {my $key=$_;map {print "$key - $_\n"}
sort keys %{$n{$_}} } sort keys %n;

    [download]
[reply]
[d/l]
Re: Can a hash name have the same value twice.
by Bloodelf (Beadle) on Jun 19, 2001 at 02:53 UTC
    If you want to put all of the the values for each key together, then give this a go.(I'm Assuming the file is in the above format.) 
    open (FILE,"c:/file.txt");
%hash=();
while (<FILE>)
      {chomp;
      split /:/;
      if ($hash{$_[0]}){$hash{$_[0]}.=",$_[1]";}
      else {$hash{$_[0]}=$_[1];}
      }

    [download]
[reply]
[d/l]
      HI, I'm not able to execute the script u have written. I have a requirement like this: original file: 38881,408 38884,596 38884,597 38884,607 38884,608 38884,609 38884,615 output expected: 38881:408 38884:596,597,607,608,609,615 Could you plz. help me to get this output done
[reply]
Re: Can a hash name have the same value twice.
by Anonymous Monk on Mar 24, 2004 at 02:09 UTC

    Re: Can a hash name have the same value twice.

    Originally posted as a Categorized Answer.

[reply]
Re: Can a hash name have the same value twice.
by Doraemon (Beadle) on May 11, 2004 at 13:35 UTC
    Personally i think Roger's idea on the hash (with empty value) not quite efficient (but his code looks good, though :) ). I prefer the previous approach (reptile)

    Originally posted as a Categorized Answer.

[reply]

Back to Seekers of Perl Wisdom