I am unclear as to how to feed mail that is spam and gets by the filters back into sa so that it can learn them. can someone walk me how to do this, especially if the mail to be marked as spam is on a windows box now and not the linux box (I use doze for my mail client, and linux for serving...
Well, you didn't specify what mailclient you are using.. however, I can see you're using eudora.
It's impossible to do in the general case with email that has been downloaded to Eudora. When eudora downloads mail it *PERMANENTLY* discards what it considers to be "redundant" mime sections. This includes any text/plain sections in multipart/alternative messages. This information is gone and lost forever and cannot be reconstructed in any way.
Since the nature of the mime boundary layouts, and the contents of *all* mime sections are part of what sa-learn tokenizes, this damage makes the email significantly different than what it once was.
Single-part messages can be done.. use eudora's "view source" to get the full body with HTML tags, and use the "blah blah blah" button to get the full headers. You can assemble these two together to create something representative of the original that you can feed to sa-learn... Not easy.. but possible.
However multi-part messages are impossible to repair and you should NOT use this method on them.
------------------------------------------------------- This SF.net email is sponsored by: SF.net Giveback Program. Does SourceForge.net help you be more productive? Does it help you create better code? SHARE THE LOVE, and help us help YOU! Click Here: http://sourceforge.net/donate/ _______________________________________________ Spamassassin-talk mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/spamassassin-talk
