Questions about Rules in Mulberry Mail

This message contains references to words that are common in spam so I have 
taken special care to try to avoid their use while still conveying the subject 
matter as it pertains to applying rules to filter electronic mail.

How can I change the order in which rules are applied? The only ways I know are 
to clear the trigger for each rule that I want to move a rule above and then to 
set that trigger again (very tedious process when trying to move the 140th rule 
to the top of the list!) or to edit the preferences file when Mulberry Mail is 
not running (very dangerous process).

How can I filter using a specific literal string of characters without the 
space suppression that Mulberry Mail normally uses when processing a rule? An 
example of my problem here was trying to filter messages containing the word 
s-e-x (dashes added to get this past any such filter). This rule trapped the 
phrase "Jes explains" in a Linux Journal announcement because the space between 
words is ignored both in the message and in the search criteria. I see no way 
to tell Mulberry Mail to honor space characters. Another technique I tried 
before learning about the space suppression issue was to look for something 
like " PhD " which includes both a leading and trailing space. Now I see that 
this is hopeless. Some digital signatures are trapped by this rule.

How can I specify a filter in which I want the search criteria to be considered 
as a pattern to find obfuscated words? For example, instead of the letter "a" 
some messages contain "@" or "4". instead of the letter "i" some messages 
contain "!" or "|" or lower case "L" or the number "1". Some messages will 
repeat a letter to try to get past rules. A sample of these techniques used in 
combination would be vvVv||@aA@9g9rRaaA. In this example I have used the letter 
"v" both upper and lower case four times in succession, the pipe character 
twice in succession to stand for the letter "i", the @ symbol along with both 
upper and lower case "a" for the letter "a", a string of nines and lower case 
"g" characters to represent the letter "g" etc. I see no method of applying 
character substitutions to catch obfuscation like this. I recognize that this 
is not a trivial task for someone to knock out in a few hours (days? weeks? 
months?).

I think the answer to each of these questions is that there is no current means 
to accomplish what I want in a reasonable manner but I can always hope.

-- 
Jim