Sorry for the confusion on this... I have some custom code that splits MailScanner's archive of all incoming mail into known spam and ham, then I manually move what's left in the archive into the spam/ham mbox files and then run sa-learn on the spam/ham mailboxes and then append them to archives of known spam and known ham.
Yesterday though, I accidentally copied a message to the 'spam' mailbox and ran sa-learn on it, and that message was really 'ham'. If I accidentally run sa-learn on a message as spam that really was *ham*, can I simply run "sa-learn --ham" on the message to cancel it out without using the "--forget" flag first? Or do I need to rebuild my entire Bayes database? Sorry for the newb questions, just confused about this. ------------------------------------------------------- This SF.Net email sponsored by: ApacheCon 2003, 16-19 November in Las Vegas. Learn firsthand the latest developments in Apache, PHP, Perl, XML, Java, MySQL, WebDAV, and more! http://www.apachecon.com/ _______________________________________________ Spamassassin-talk mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/spamassassin-talk
