Re: Memory storage errors
[email protected] Mon, 25 May 2009 12:11:54 -1000
| Newsgroups | gmane.mail.spam.crm114 |
|---|---|
| Message-ID | <OF855A8B43.89F8DF4D-ON0A2575C1.0078D0C5-0A2575C1.0079E475@thdfsg.com> |
> [email protected] (Bill Yerazunis) wrote on 05/25/2009 04:50:15 AM: >> From: [email protected] >> Looking through the mailing list archives and READMEs I found the >> -w flag, which solves the problem if I make the argument big enough. >> But this confuses me since I thought the decision=5Flength parameter in >> mailfilter.cf cut off the amount of input mailreaver would use from each >> email. >> >> Is this a bug? > > Yes, and no. > It's a case of "bad spec yields silly code". > > If this is OK, then this is an easy "bug" to fix - insert > (caution- UNTESTED CODE) just before line 78 of mailreaver.crm: > > ################################## > # Step 0 - read in the input file > ################################## > window > input [ stdin 0 :*:decision_length:] > > ... and that should do it. Adding the above into mailreaver tended to cause segmentation faults with the Hyperspace classifier. In practice I'm calling mailreaver from a perl script anyway, so I just wrote a short wrapper around the file slurping part to grab the first several hundred lines of the email to be classified. (Makes for a nice speed boost too.) Dean Takemori Systems Support Supervisor TD Food Group [email protected] ------------------------------------------------------------------------------ Register Now for Creativity and Technology (CaT), June 3rd, NYC. CaT is a gathering of tech-side developers & brand creativity professionals. Meet the minds behind Google Creative Lab, Visual Complexity, Processing, & iPhoneDevCamp asthey present alongside digital heavyweights like Barbarian Group, R/GA, & Big Spaceship. http://www.creativitycat.com _______________________________________________ Crm114-general mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/crm114-general