Some people use various IMAP clients to do it, but I use the Outlook object model using Win32::OLE (probably because of my Windows background) and a COM object created by Dmitry Streblechenko (www.dimastr.com) called Outlook Redemption..
<slightlyOT> There's a charge for it's use with any commercial application (that you might try to sell it with), but I use the developer version. No restrictions are present. SaveAs creates a nice RFC822 mail message as a TXT file that sa-learn can gobble up. And it bypasses Outlook restrictions on programmatic access so that none of those annoying Outlook pop-up messages come up asking you to allow access for X minutes. He's got some really cool new security features allowing you to change the name of the COM object and the GUID, or even keep the DLL unregistered (and load access to it only when needed) so that would be hackers/viri shouldn't be able to use it against you in a fight for control over your PC's email. </slightlyOT> The problem you may be encountering is that once the user forwards a SPAM to someone to feed into bayes, Exchange strips the header info. That's why the public folder works so well, because the user drags it over to the public folder (headers in tact) and you can get to it without problem from there as well. Steven -----Original Message----- From: Jon Dossey [mailto:[EMAIL PROTECTED] Sent: Tuesday, June 21, 2005 12:10 PM To: users@spamassassin.apache.org Subject: Exchange/Outlook - how do you learn spam? I'm sure a lot of us have a similar setup, linux/bsd mx gateways (running SA) relaying mail to Exchange, and Outlook clients. I'm just curious how everyone handles learning? It seems like a lot of people recommend a public folder for users to dump spam in, but how do you get it back out into a useable format that sa-learn will understand? Saving messages out of Outlook (for me anyway) into a txt file removes all the internet headers. So how else do you handle getting your messages back out of exchange/outlook, and sa-learn'ed? .jon
