IssueTrackerProduct

SpamBayes is probably the best non-enterprise anti-spam tool there is today. It assumes nothing and you have to teach it from scratch. That's a bit annoying but once you've trained it it rarely gets it wrong like Gmail does

UPDATE: As of the 0.6.11 release, this feature has been improved to cases where all parsed email headers are converted into lowercase.

As of the 0.6.9 release emails that come in via the inbound email feature and has the X-Spambayes-Classification header set to spam will be deleted and not uploaded into the issuetracker. If you for some reason want to switch this off you can go into Constants.py and set:

 SPAMBAYES_CHECK = None

Setting up the SpamBayes is outside the scope of the issuetracker. It's actually not that hard if you've got some sys admin skills. All you have to do is to set up the proxy for your POP3 server and instead of having the POP3 server address in your POP3 setup of the issuetracker, you have the address of the machine running the SpamBayes proxy. In ths case of the real.issuetrackerproduct.com it's set up to localhost.

The neat thing about this is that it works seemlessly. The anti-spamming shouldn't be done by the IssueTrackerProduct when there already are much better tools for it. The IssueTrackerProduct should only be seen as the end client that needs to decide what to do with this classification. Now, all that is needed is to keep training the SpamBayes proxy with a few more spam emails.

Request for feedback If you think spam emails should not be deleted and only just ignored by the issuetracker, then give your feedback on the mailing list