Re: OT: Google PR, PHP -> PERL ?

This has nothing to do with Google's page rank algorithm, if just allows you to access the Google page rank info as described here This is actually just a PHP implementation of Bob Jenkins' 32 bit integer hashing algorithm. I wrote a Perl XS module implementing this algorithm a couple of years ago and it has been on CPAN since 2003 - Digest::JHash. You will need to change the line c=0 in JHash.xs to c=0xe6359a60 before you compile it to get the Google init constant right. Then you can just:

use Digest::JHash 'jhash';
print 'Checksum: 6' . jhash('info:http://www.google.com/');
__DATA__
Checksum: 64222138902
[download]

to get the correct checksums. I might point out that this is against the Google TOS (well if you get the checksum wrong that is where they refer you).

Google's pagerank algorithm is a complex beast. You can read about the original implementation in the original paper from Brin and Page at The Anatomy of a Large-Scale Hypertextural Search Engine

Here is the XS code as a standalone Inline C widget:

use Inline C;

print '6' . jhash('info:http://www.google.com/');

__END__
__C__
typedef unsigned long UB4;
const UB4 INIT  = 0xe6359a60;  /* Google Magic Value */ 

#define MIX(a,b,c) \
{ \
  a -= b; a -= c; a ^= (c>>13); \
  b -= c; b -= a; b ^= (a<<8);  \
  c -= a; c -= b; c ^= (b>>13); \
  a -= b; a -= c; a ^= (c>>12); \
  b -= c; b -= a; b ^= (a<<16); \
  c -= a; c -= b; c ^= (b>>5);  \
  a -= b; a -= c; a ^= (c>>3);  \
  b -= c; b -= a; b ^= (a<<10); \
  c -= a; c -= b; c ^= (b>>15); \
}

unsigned long jhash( SV* str )
{
    STRLEN rawlen;
    char* p;
    UB4 a, b, c, len, length;

    /* extract the string data and string length from the perl scalar 
+*/
    p = (char*)SvPV(str, rawlen);
    length = len = (UB4)rawlen;

    if ( length == 0 ) {
        printf( "Recieved a null or undef string!\n" );
      return 0;
    }

    a = b = 0x9e3779b9;        /* golden ratio suggested by Jenkins 0x
+9E3779B9 */
    c = INIT;
    while (len >= 12) {
        a += ((UB4)p[0]+((UB4)p[1]<<8)+((UB4)p[2]<<16)+((UB4)p[3]<<24)
+);
        b += ((UB4)p[4]+((UB4)p[5]<<8)+((UB4)p[6]<<16)+((UB4)p[7]<<24)
+);
        c += ((UB4)p[8]+((UB4)p[9]<<8)+((UB4)p[10]<<16)+((UB4)p[11]<<2
+4));
        MIX(a, b, c);
        p += 12;
        len -= 12;
    }
    c += length;

    switch(len) {
        case 11: c+=((UB4)p[10]<<24);
        case 10: c+=((UB4)p[9]<<16);
        case 9:  c+=((UB4)p[8]<<8);
        case 8:  b+=((UB4)p[7]<<24);
        case 7:  b+=((UB4)p[6]<<16);
        case 6:  b+=((UB4)p[5]<<8);
        case 5:  b+=((UB4)p[4]);
        case 4:  a+=((UB4)p[3]<<24);
        case 3:  a+=((UB4)p[2]<<16);
        case 2:  a+=((UB4)p[1]<<8);
        case 1:  a+=((UB4)p[0]);
    }

    MIX(a, b, c);

    return(c);
}
[download]

cheers

tachyon

Comment on Re: OT: Google PR, PHP -> PERL ? Select or Download Code

Replies are listed 'Best First'.
Re^2: OT: Google PR, PHP -> PERL ? by 2ge (Scribe) on Aug 28, 2004 at 23:56 UTC
Hello Tachyon! Thank you very much for helping, ofcourse everything works, when I install my new linux. Working under win system is really terrible, I can't even install Jhash properly. Good job, thanks again!	[reply]