Beefy Boxes and Bandwidth Generously Provided by pair Networks
There's more than one way to do things
 
PerlMonks  

RE: Shift, Pop, Unshift and Push with Impunity!

by Anonymous Monk
on Jun 15, 2000 at 22:47 UTC ( [id://18367]=note: print w/replies, xml ) Need Help??


in reply to Shift, Pop, Unshift and Push with Impunity!

About hash table lookup. In theory, if you have infinite ammount of memory, you can approach O(1) time. As I understand all really implemented hash algorithms with finite memory will take, strictly speaking, O(n) time. There is an example of unsuccessful order of data that will make Perl's implementation of hashes to take O(n) for searching. In practice you'll hardly will be much worse than O(1).

Replies are listed 'Best First'.
Re: Shift, Pop, Unshift and Push with Impunity!
by Dominus (Parson) on Jan 05, 2001 at 07:46 UTC
    Says nobody in particular:
    There is an example of unsuccessful order of data that will make Perl's implementation of hashes to take O(n) for searching.
    When Hashes Go Wrong

      The following is a slight modification of your program. It (ab)uses the fact that Perl only checks whether to split buckets when you use a new bucket. It does not make any assumptions about Perl's hashing algorithm. It is therefore able to generate the special keys much more quickly than yours does.
      my $HOW_MANY = shift || 100_000; my ($s, $e1, $e2); $s = time; for ($i=0; $i<$HOW_MANY; $i++) { $q{$i}=1; } $e1 = time - $s; print qq{ Normally, it takes me about $e1 second(s) to put $HOW_MANY items into +a hash. Now I'm going to construct $HOW_MANY special keys and see how long it takes to put those into a hash instead. }; undef %q; print "Generating $HOW_MANY keys\n"; my @keys; my $i = 0; while (@keys < $HOW_MANY) { $i++; my %hash = (0 => 0, $i => 0); if (%hash =~ /1/) { # Goes in the same bucket as 0 push @keys, $i; } } print "Putting the $HOW_MANY special keys into the hash.\n"; $i = 0; $lasttime = $s = time; foreach $key (@keys) { $h{$key} = 1; $i++; if (time() - $lasttime > 5) { my $h = %h + 0; print "I have put $i keys into the hash, using $h bucket(s)\n" ; $lasttime = time; } } $e2 = time - $s; print qq{ The $HOW_MANY special keys took $e2 seconds to put into the hash, instead of $e1 seconds. };

      Update: (Much later.) Perl changed when it splits buckets so this program no longer runs slowly. Furthermore Perl's hashing algorithm is now random, so Dominus' program no longer runs slowly. Shucks.

Log In?
Username:
Password:

What's my password?
Create A New User
Domain Nodelet?
Node Status?
node history
Node Type: note [id://18367]
help
Chatterbox?
and the web crawler heard nothing...

How do I use this?Last hourOther CB clients
Other Users?
Others making s'mores by the fire in the courtyard of the Monastery: (5)
As of 2024-03-28 08:45 GMT
Sections?
Information?
Find Nodes?
Leftovers?
    Voting Booth?

    No recent polls found