easy way to extract a hash from an array?

rmexico has asked for the wisdom of the Perl Monks concerning the following question:

Replies are listed 'Best First'.
Re: easy way to extract a hash from an array? by davido (Cardinal) on Feb 08, 2006 at 16:53 UTC
Your first snippet may contain a bug. You're pushing the same %person into @PEOPLE every time through the loop, unless %person is declared lexically on each iteration of the loop (ie, there has to be a `my %person` in the loop). There isn't a way to avoid looping when you're looking for something inside of an array. You can grep, but that's still a loop: `my( @matches ) = grep { $_->{'fname'} eq 'John' } @PEOPLE;` [download] Dave	[reply] [d/l] [select]
Re: easy way to extract a hash from an array? by nobull (Friar) on Feb 08, 2006 at 18:20 UTC
If you are doing a lot of such lookups then you may find it worthwhile to index @PEOPLE. `my %PEOPLE_index; for my $ppl (@PEOPLE) { while ( my ( $k,$v) = each %$ppl ) { push @{$PEOPLE_index{$k}{$v}} => $ppl; } } for my $ppl ( @{$PEOPLE_index{fname}{John}} ) { #.... }` [download]	[reply] [d/l]
Re: easy way to extract a hash from an array? by MCS (Monk) on Feb 08, 2006 at 17:15 UTC
How else would you find something other than look for it? Personally I think two lines of code isn't exactly long winded. You could write a subroutine if you have to do this a few times that accepts your AoH and the string you are searching for. But yes, you still have to loop through it.	[reply]
Re: easy way to extract a hash from an array? by beachbum (Beadle) on Feb 08, 2006 at 18:47 UTC
If your goal is to search on first name, then you might not want to use an array anyhow.... `my %people = (); while (blah) { ## do something $people{$fname}{$lname}++; } my $search_term = 'John'; if (exists $people{$search_term}){ # do stuff. }` [download]	[reply] [d/l]
Re: easy way to extract a hash from an array? by johngg (Canon) on Feb 09, 2006 at 10:28 UTC
Given that you have reasons to hold your people in a list then a quick way to search for particular first names is to index them as other replies have suggested. You can build the index as you build the @PEOPLE list and if you make the index a HoL with key being name and value being record no. in @PEOPLE you can extract all of your Johns using a slice. `#!/usr/local/bin/perl -w # use strict; our @PEOPLE = (); our $rhFirstNameIdx = {}; our $recCt = 0; while(<DATA>) { chomp; my ($fname, $lname) = split; push @PEOPLE, {fname => $fname, lname => $lname}; push @{$rhFirstNameIdx->{$fname}}, $recCt ++; } our $wanted = 'John'; our @found = @PEOPLE[@{$rhFirstNameIdx->{$wanted}}]; print "First - $_->{fname}, Last - $_->{lname}\n" for @found; __END__ Fred Bloggs Charley Farley John Doe Peter Piper Mary Poppins John Smith Bill Bailey` [download] The little script above illustrates this and prints out the entries for John Doe and John Smith. Cheers JohnGG	[reply] [d/l]