The pop3filter Manual

Most of the configuration sits in the file spamtokens.py. You can edit the following variables, in valid python syntax:

# list of spam blacklists. You can add your own if you wish
BLACKLIST_DOMAINS = [
    'spews.relays.osirusoft.com',
    'dns.rfc-ignorant.org',
    ...
    ]

This is the list of DNSBL you want to use. Read here for a comparison of various blacklists you might want to use.

# list of IP addresses that will not be checked. 
BLACKLIST_IGNORE = [
    '127.0.*',
    '10.*',
    ('212.14.77.97','212.14.77.126'),
    ]

A list of IP addresses that you don't want to block. IP-Address rules can be:

# All email addresses in this list will be ignored and not filtered. 
# These email addresses are also allowed to send HTML mail
IGNORE_THESE_SENDERS = [
    ]

When pop3filter processes mail, it first checks if the mail is on any of the IGNORE rules. This way, you can make sure your personal contacts to not get filtered. You can use the following expressions here:

Note: All rules in pop3filter that are not IP addresses work like this.

IGNORE_THESE_RECEIVERS = [
    ]

You can add a list of receivers here, that if specified will cause the mail to be ignored by pop3filter.

IGNORE_THESE_SUBJECTS = [ "fax empfangen" ]

You can add a list of subjects here, that if found will cause the mail to be ignored by pop3filter. In the example, the Fax mails from our server are ignored, because they have a fixed subject. Note: It is probably better to not use subjects as ignore filter rules.

# this is a list of header keywords, with associated rule tables
IGNORE_RULES = {
    'subject' : IGNORE_THESE_SUBJECTS,
    'from' : IGNORE_THESE_SENDERS,
    'to' : IGNORE_THESE_RECEIVERS,
}

As you can see, the actual keywords used for the IGNORE processing is configurable. If, for example, you want to exclude messages from a specific host with a proper message ID, you could add 

    'message-id' : IGNORE_THESE_MESSAGE_IDs

and supply a list of message ID patterns to be ignored.

FILTER_THESE_SENDERS = [
    "*.tw",
    "undisclosed recipients",
    ]

This is my personal remove list for contacts. Any mail which fits this sender description will be marked as spam. Others may use more relaxed rules (especially, if you live in taiwan or russia).

FILTER_THESE_RECEIVERS = [
    "undisclosed recipients",
    ]

If any of these receivers match, the message will be marked as spam.

FILTER_THESE_MESSAGE_IDS = [
    "*.tw",
    ]

If the message-ID matches taiwan or russia, the message will be marked as spam. Others may use more relaxed rules (especially, if you live in taiwan or russia).

FILTER_THESE_SUBJECTS = [
    ]

You can add specific subject patterns here to mark messages as spam.

FILTER_THESE_ADDRESSES = [
    ]

You can add specific IP addresses (or IP address ranges, see above) here; if a message uses these, it will be marked as spam.

# this is a list of header keywords, with associated rule tables
FILTER_RULES = {
    'subject' : FILTER_THESE_SUBJECTS,
    'from' : FILTER_THESE_SENDERS,
    'to' : FILTER_THESE_RECEIVERS,
    'message-id': FILTER_THESE_MESSAGE_IDS,
}

Again, as with the IGNORE_RULES, you can add your own filter rules here.

# these tokens, if found in the Subject: line, will indicate spam mail
FILTER_SUBJECT_TOKENS = ['lowest', 'rate', 'mortgages', 'registered',
'analyst', 'invest', 'information', 
...

A list of subject tokens that if found, will result in the message being tagged as spam.

Using HAMSTER rules

You can use a HAMSTER rules file, for example the one provided at http://spamabwehr.sakrak.net/spam-filter/ if you edit the registry entry

HKEY_LOCAL_MACHINE\Software\p-nand-q.com\HAMSTER_RULES