i would presume the idea is that it would add any email addresses learned as spam to the auto whitelist?
i dont know if it exists or not, it was just written about in michael bell's how to but doesnt seem to be a feature of the actual software
Ahh.. well, I don't think there's any support for that.. sa-learn is specifically be a bayes tool, which is why it struck me as odd to expect it to manipulate the AWL database.
Also your question asked about the -a parameter. All that does is cause the AWL to be used in score calculations, which is even more strange for sa-learn to want to do, since it doesn't calculate scores at all.
The parameters that would modify the AWL database would be --add-to-whitelist or --add-to-blacklist. But I don't think sa-learn supports anything AWL related..
If this really bothers you, you can make a shell script that calls sa-learn and spamassassin --add-to-blacklist at the same time.
Bear in mind however that you need to have a realistic understanding of what to expect from the AWL in this. The AWL is NOT a static whitelist, and the effect of adding an email to the AWL with --add-to-whitelist will decay over time. (after 100 emails from that sender, they will only get a -1.0 score adjust from your one call to ). The AWL is not, and will likely never be an adequate replacement for whitelist_from commands.
------------------------------------------------------- This SF.Net email sponsored by: Free pre-built ASP.NET sites including Data Reports, E-commerce, Portals, and Forums are available now. Download today and enter to win an XBOX or Visual Studio .NET. http://aspnet.click-url.com/go/psa00100003ave/direct;at.aspnet_072303_01/01 _______________________________________________ Spamassassin-talk mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/spamassassin-talk
