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

In a utility-module I have the folling snippet (it's a slightly extended utils.pm of the nagios-distribution): 

%ERRORS=('OK'=>0,'WARNING'=>1,'CRITICAL'=>2,'UNKNOWN'=>3,'DEPENDENT'=>
+4);
%ERROR_TEXT= map { $ERRORS{$_} => $_ } keys %ERRORS;

[download]

Is there a more concise way to 'reverse' a hash?

regards,
tomte

Hlade's Law:

If you have a difficult task, give it to a lazy person --
they will find an easier way to do it.

Replies are listed 'Best First'.
Re: 'reversing' a hash (reverse)
by particle (Vicar) on Jun 18, 2003 at 17:02 UTC
    %ERROR_TEXT= reverse %ERRORS;

    see reverse for more info. by the way, it's usually called 'inverting' a hash.

    ~Particle *accelerates*

[reply]
[d/l]
[select]
Re: 'reversing' a hash
by dws (Chancellor) on Jun 18, 2003 at 17:05 UTC
    Is there a more concise way to 'reverse' a hash?

      %ERROR_TEXT = reverse %ERRORS; is the idiomatic way. It's more concise, but it liable to catch a few less seasoned readers by surprise.

[reply]
[d/l]
Re: 'reversing' a hash
by shotgunefx (Parson) on Jun 18, 2003 at 21:46 UTC
    Just a thought, using reverse as noted above is the easiest way, though if you're worried about duplicate keys, you can always use a 
     foreach(keys %hash ){ 
 } 

 or 
 while (my ($k,$v) = each %hash){
 }

    [download]
    so you can handle duplicates in a reasonable way.

    -Lee

    "To be civilized is to deny one's nature."
[reply]
[d/l]
Re: 'reversing' a hash
by BrowserUk (Patriarch) on Jun 18, 2003 at 22:10 UTC

    I can't thing of any particular advantage for this but...

    @inverted{ values %hash } = keys %hash;

    Examine what is said, not who speaks.
    "Efficiency is intelligent laziness." -David Dunham
    "When I'm working on a problem, I never think about beauty. I think only how to solve the problem. But when I have finished, if the solution is not beautiful, I know it is wrong." -Richard Buckminster Fuller


[reply]
[d/l]

      I couldn't resist ;-) 

      use Data::Dumper;
use Benchmark qw(:all);

%ERRORS=('OK'=>0,'WARNING'=>1,'CRITICAL'=>2,'UNKNOWN'=>3,'DEPENDENT'=>
+4);

cmpthese(1000000, {
        'map' => sub {
            %ERRORS_TEXT = map { $ERRORS{$_} => $_ } keys %ERRORS;
        },
        'reverse' => sub {
            %ERRORS_TEXT = reverse %ERRORS;
        },
        'values' => sub {
            @ERRORS_TEXT{ values %ERRORS } = keys %ERRORS; 
        },
    }
);
__END__
First run:

Benchmark: timing 1000000 iterations of map, reverse, values...
       map: 27 wallclock secs (24.66 usr +  0.03 sys = 24.69 CPU) @ 40
+502.23/s (n=1000000)
   reverse: 10 wallclock secs ( 9.03 usr +  0.00 sys =  9.03 CPU) @ 11
+0741.97/s (n=1000000)
    values:  8 wallclock secs ( 7.70 usr +  0.04 sys =  7.74 CPU) @ 12
+9198.97/s (n=1000000)
            Rate     map reverse  values
map      40502/s      --    -63%    -69%
reverse 110742/s    173%      --    -14%
values  129199/s    219%     17%      --

Second run:
Benchmark: timing 1000000 iterations of map, reverse, values...
       map: 27 wallclock secs (24.49 usr +  0.03 sys = 24.52 CPU) @ 40
+783.03/s (n=1000000)
   reverse:  8 wallclock secs ( 8.54 usr +  0.02 sys =  8.56 CPU) @ 11
+6822.43/s (n=1000000)
    values:  7 wallclock secs ( 7.55 usr +  0.01 sys =  7.56 CPU) @ 13
+2275.13/s (n=1000000)
            Rate     map reverse  values
map      40783/s      --    -65%    -69%
reverse 116822/s    186%      --    -12%
values  132275/s    224%     13%      --

Third run:
Benchmark: timing 1000000 iterations of map, reverse, values...
       map: 26 wallclock secs (24.37 usr +  0.06 sys = 24.43 CPU) @ 40
+933.28/s (n=1000000)
   reverse: 11 wallclock secs ( 8.63 usr +  0.03 sys =  8.66 CPU) @ 11
+5473.44/s (n=1000000)
    values:  9 wallclock secs ( 7.57 usr +  0.01 sys =  7.58 CPU) @ 13
+1926.12/s (n=1000000)
            Rate     map reverse  values
map      40933/s      --    -65%    -69%
reverse 115473/s    182%      --    -12%
values  131926/s    222%     14%      --

      [download]

      Looks like I could gain a wee bit of speedup using your code, if it would matter, which it doesn't in my particular case; for the loss of readability, though IMHO :)

      regards,
      tomte

      Hlade's Law:

      If you have a difficult task, give it to a lazy person --
      they will find an easier way to do it.

[reply]
[d/l]

        Wow! A whole 1.3 microseconds saved. And to think I didn't see any merit in it:)

        On a slightly serious note, I think a novice perler might actually find the "values become keys and keys become values" slightly more obvious than the reverse method, which relies on a pretty good understanding of how hashes work in list contexts, on their first viewing, but once you understand that, the presence of the big clue, R-E-V-E-R-S-E does make it pretty clear:)

        Examine what is said, not who speaks.
        "Efficiency is intelligent laziness." -David Dunham
        "When I'm working on a problem, I never think about beauty. I think only how to solve the problem. But when I have finished, if the solution is not beautiful, I know it is wrong." -Richard Buckminster Fuller


[reply]
Re: 'reversing' a hash
by Tomte (Priest) on Jun 18, 2003 at 17:06 UTC

    ++to you both, I'd known it ;-)
    Man, do I just love perl.

    regards,
    tomte

    Hlade's Law:

    If you have a difficult task, give it to a lazy person --
    they will find an easier way to do it.

[reply]
A reply falls below the community's threshold of quality. You may see it by logging in.

Back to Seekers of Perl Wisdom