Beefy Boxes and Bandwidth Generously Provided by pair Networks
Pathologically Eclectic Rubbish Lister

Re: Re: Removing Duplicate Array Elements!!!

by tachyon (Chancellor)
on Sep 27, 2001 at 17:33 UTC ( #115087=note: print w/replies, xml ) Need Help??

in reply to Re: Removing Duplicate Array Elements!!!
in thread Removing Duplicate Array Elements!!!

While using map certainly works, using map in a void context like you do is inefficient. Map returns an array for you regardless of whether you use it or not. As a result using map as a surrogate for loop is bad practice. In golf you use it because it is 2 chars shorter than a for loop. That is about the only (valid) reason to use it in a void context :-) All you actually want is:

$help{$_}++ for @a

This is far more efficient, it is also shorter for what it's worth. Thanks to chipmunk for straigtening me out on this some months ago.

my @a= (1,2,3,4,5,6,6,5,4,3,2,1); my %help; my @wasted_array = map { $help{$_}=1 } @a; @a = keys %help; print @wasted_array, "\n", @a;




Replies are listed 'Best First'.
Re: Re: Re: Removing Duplicate Array Elements!!!
by gbarr (Monk) on Sep 27, 2001 at 19:45 UTC
    If you are looking an efficient method to do this then try.

    my %hash;
    @hash{@a} = ();
    @a = keys %hash;

    But that only works, just as your example, for non-references as the keys of a hash can only be strings. If @a may contain references that you want to preserve then use

    my %hash;
    @hash{@a} = @a;
    @a = values %hash;

    This is slightly less efficient as it causes an extra copy of each element in @a

    I generally write this as

    @a = do { my %h; @h{@a} = @a; values %h }; # unique

      As you say, the slice method is faster again. I will be using it from now on. Thanks.

      use Benchmark; @a= (0..100,0..100); $mine = <<'CODE'; my %hash; @hash{$_}++ for @a; @a = keys %hash; CODE $yours1 = <<'CODE'; my %hash; @hash{@a} = (); @a = keys %hash; CODE $yours2 = <<'CODE'; my %hash; @hash{@a} = @a; @a = values %hash; CODE $his = <<'CODE'; my %hash; map { $hash{$_}=1 } @a; @a = keys %hash; CODE timethese ( 10000, { 'mine' => $mine, 'yours1' => $yours1, 'yours2' => $yours2, 'his' => $his } ); __END__ C:\>perl Benchmark: timing 10000 iterations of his, mine, yours1, yours2... his: 17 wallclock secs (17.03 usr + 0.00 sys = 17.03 CPU) @ 58 +7.20/s (n=10000) mine: 15 wallclock secs (15.05 usr + 0.00 sys = 15.05 CPU) @ 66 +4.45/s (n=10000) yours1: 12 wallclock secs (11.76 usr + 0.00 sys = 11.76 CPU) @ 85 +0.34/s (n=10000) yours2: 15 wallclock secs (14.78 usr + 0.00 sys = 14.78 CPU) @ 67 +6.59/s (n=10000) C:\>




      Hi thank u so much for your code is working it has solved my problem ..but as a beginner i want know wats happening there ..could plz explain me the code..thanks in advance.

Log In?

What's my password?
Create A New User
Node Status?
node history
Node Type: note [id://115087]
and all is quiet...

How do I use this? | Other CB clients
Other Users?
Others about the Monastery: (6)
As of 2017-01-22 09:40 GMT
Find Nodes?
    Voting Booth?
    Do you watch meteor showers?

    Results (187 votes). Check out past polls.