Following up on my earlier post on this subject, it appears that the Rules
processing operation may get confused when a Rule refers to people and there
is more than one name in both the Rule and the address list.



Here is what I think I am seeing (in the examples below, A, B, C, D, E, X, Y
and Z represent properly structured email addresses):

-------------------------------

Scenario 1:

If I create a Rule which says:



Apply this rule after the message arrives

Where the To line contains 'A' or 'B' or 'C' or 'D' or 'E'

Delete it

and Stop processing more rules



and I get an EMAIL which has ONLY ONE of the addresses A, B, C, D or E in
the TO: line,
the rule works and the email is moved to the Deleted Items folder.

-------------------------------

Scenario 2:

If I use the same Rule as in scenario 1, but the incoming EMAIL contains
MORE THAN ONE of the addresses A-E,
the rule fails.
(NOTE: I have not yet done a systematic and exhaustive test to see if the
failure is due solely to having more than one of A-E in the address list, or
if perhaps there is only a failure if there is also an address X, Y, or Z in
the TO: field which is NOT also contained in the rule.)

-------------------------------

Scenario 3:

If I modify the same RULE so that it contains ONLY ONE of the addresses A-E,
and I get an email with any one or more of the addresses A-E,
the rule works.

-------------------------------



In other words, Rules processing seems to fail if there are multiple
addresses in both the rule and the email. In my view of the world, there is
no reason why Scenario 2 should not delete an email if ANY of the TO:
addresses in the email matches ANY of the addresses in the rule.



While Rules processing does allow both AND and OR logic to be used in a
Rule, one cannot mix AND and OR logic in a single Rule. In the case where
an email has more than one address (such as A-E and X-Z) being tested
against a rule and the test is to match the email address on an OR basis
against the address list in the rule, a failure of any one address in the
email address list to match any of the addresses in the rules should not
cause the rule to fail, as this is essentially saying that the rules
processing is translated (per the above example) as:



IF email-address-A
matches any of rule-addresses
AND
email-address-B
matches any of rule-addresses
AND
email-address-C
matches any of rule-addresses
AND
email-address-D
matches any of rule-addresses
AND
email-address-E
matches any of rule-addresses
AND
email-address-X
matches any of rule-addresses
AND
email-address-Y
matches any of rule-addresses
AND
email-address-Z
matches any of rule-addresses
THEN execute rule



Since most spam that I see has lots of entries in the TO: address list, it
is far easier from a rules management standpoint to have a single rule which
refers to a large single list of email addresses than it is to have hundreds
of rules each of which refers to only one address.



I am using Excel to manage my rules address list, copying all of the TO:
entries in each new large batch of unfiltered spam into Excel and using the
Filter function to cut the list down to one copy of each entry, sorting the
list, then pasting the list back into a single OE6 rule (after inserting the
proper ' or ' linkages between list elements).