On Fri, Apr 08, 2011 at 04:33:27PM +0200, Stevan Baji?? wrote: > > > On Fri, 8 Apr 2011 09:27:11 -0500, Kenneth Marshall wrote: > > > On > Fri, Apr 08, 2011 at 04:03:07PM +0200, Stevan Baji?? wrote: > >> On Fri, 8 > Apr 2011 08:57:19 -0500, Kenneth Marshall wrote: > >> > >>> On > >> Fri, Apr > 08, 2011 at 03:12:25PM +0200, Stevan Baji?? wrote: > >> > >>>> On Fri, > 8 > >> Apr 2011 14:35:32 +0300, Ibrahim Harrani wrote: > > >>> > >>>> > Hi, > >>> > >>>> Is > >>> there any way to decode tokens(tokenizer osb) in > the > >> r-left:#1010ff 2px solid; margin-left:5px; > width:100%">dspam_token_data table? > >> 00%">lockquote type= > "padding-left:5px; border-left:#1010ff 2px solid; margin-left:5px; > width:100%"> Xperia(TM) PLAY #1010ff 2px solid; margin-left:5px; > width:100%"> https://lists.sourceforge.net/lists/li [1] > >> e" > style="padding-left:5px; border-left:#1010ff 2px solid; margin-left:5px; > width:100%"> user [3] The token is nothing else than a CRC64 ISO hash. > Assuming you have insane amount of memory > >> > >>> t:#1010ff 2px solid; > margin-left:5px; width:100%"> > >> left:5px; width:100%"> Actually, you > would only be able to find A token that generated the same hash, not > neccessarily the token that WAS USED to generate the hash value. That is > so because > >> > >>> right. Especially with that weak CRC64-ISO that DSPAM > is using. Better would be to use CRC64 ECMA-182, but that is another > >> > ay... even that CRC64 ECMA-182 is not collision free. -- Kind Regards > from Switzerland, Stevan Baji?? > >> > >>> > > I checked into different > hashes and the CRC64 was good enough in the collision domain to make it > not worth pursuing and alternative since then you could not upgrade your > existing instances. Ken > > The CRC64-ISO is considered weak for hashing. > Read this document here if you need more info -> > http://www.cs.ucl.ac.uk/staff/d.jones/crcnote.pdf > > -- > Kind Regards > from Switzerland, > > Stevan Baji?? > > > Links: > ------ > [1] > https://lists.sourceforge.net/lists/li
The hash that I have worked with is Bob Jenkins' lookup3.c hash. Here is his evaluation of a number of hash functions: http://www.burtleburtle.net/bob/hash/doobs.html He does a good job of analyzing the various usual choices. The question is are the collisions with the hash function used in DSPAM impacting its effectiveness to a degree in which it would make sense to change the function? Cheers, Ken ------------------------------------------------------------------------------ Xperia(TM) PLAY It's a major breakthrough. An authentic gaming smartphone on the nation's most reliable network. And it wants your games. http://p.sf.net/sfu/verizon-sfdev _______________________________________________ Dspam-user mailing list Dspam-user@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/dspam-user
