From: Travis Whitton <whitton@...>
Date: 2003-02-28T03:20:34+09:00
Subject: Re: Is Ruby slower?

> Of course, to be fair, in its more traditional strong areas, (regexp matching, hash
> access, etc.) Perl shows an equal or better advantage.

The only reason Perl's hashes are faster than Ruby's is because Perl uses Bob
Jenkins' hashing algorithm, which is _supposedely_ the best known hashing
algorithm. Take note, that Perl was using this very simple hashing function
until after version 5.005:

// Snipped out of the Ruby source code
#elif HASH_PERL
    register int val = 0;

    while ((c = *string++) != '\0') {
        val = val*33 + c;
    }

    return val + (val>>5);
#else

It wouldn't take much to use Bob Jenkins' algorithm in Ruby. The current Perl
implementation looks like this:

#define PERL_HASH(hash,str,len) \
     STMT_START { \
        register const char *s_PeRlHaSh_tmp = str; \
        register const unsigned char *s_PeRlHaSh = (const unsigned char *)s_PeRlHaSh_tmp; \
        register I32 i_PeRlHaSh = len; \
        register U32 hash_PeRlHaSh = 0; \
        while (i_PeRlHaSh--) { \
            hash_PeRlHaSh += *s_PeRlHaSh++; \
            hash_PeRlHaSh += (hash_PeRlHaSh << 10); \
            hash_PeRlHaSh ^= (hash_PeRlHaSh >> 6); \
        } \
        hash_PeRlHaSh += (hash_PeRlHaSh << 3); \
        hash_PeRlHaSh ^= (hash_PeRlHaSh >> 11); \
        (hash) = (hash_PeRlHaSh + (hash_PeRlHaSh << 15)); \
    } STMT_END

All it would take would be converting it over to fit the variables in st.c and
adding another preprocessor conditional such as #elif HASH_JENKINS so as not to
disturb the 3 other strhash implementations. Anyway, just an idea. You can read
about Bob Jenkins hashing function at
http://burtleburtle.net/bob/hash/doobs.html

Cheers,
Travis