Paul,
>\bred\b)).*?(?!\g{-1})){3})
notice the '3' - which meens, that three bad words must occur in the mail
to be not spamlover
There is a (non) magic logic behind.
assume we have three bad words defind: black yellow green
counted (only the count is relevant) will be a bad word, if this word
follows not the same word
so:
black black black -> 1
black black yellow ->2
yellow black black ->2
yellow yellow black black -> 2
but
black yellow black ->3
yellow black black green ->3
yellow black black yellow ->3
yellow black green -> 3
black green black green black green ->6
any text between the words plays no rule
the search will be stopped, if the count number (here 3) is reached
Thomas
Von: "Paul K. Dickson" <[email protected]>
An: ASSP development mailing list <[email protected]>
Datum: 25.05.2010 20:42
Betreff: Re: [Assp-test] Way to ignore spamloverre via another
regex?
Thomas:
No luck. Something is amiss with the bad words part. Web interface and
sending emails still matches even if the mail has a bad word from the bad
word section in it.
I sent an email with "great!" and "puke" in it and it was recognized as
spamlover. Just to make sure I was assessing correctly I sent the same
email again but with lots of nasty words that are in bombre. I still got
the "spam but passing because spamlover blah blah blah".
It see's "puke" and reports a spamlover match.
Here is what I have of your suggestion:
#assp-do-not-optimize-regex
# if you use V2 -
# write the line above anywhere
# to prevent assp from trying
# to optimize this file (regex)
#
# any regex of your choice
#....
#....
### begin of the selection part
# the first good word must be
# at the end of the first line
(?=.*?(\bblue\b
\bgreat\!\b
#another word
#....
# any regex is possible in this part
#....
#additional word(s)
#....
# the last good and the first bad
# word must be in this line
# at the begin and end
\bwhite\b))(?!.*?(?:(\b(?:\bblack\b
\bpuke\b
#....
# You may also include some other file with some
# realy bad words (e.g.)
# include files/blackre.txt
#....
# the last bad word must be
# at the begin of this line
\bred\b)).*?(?!\g{-1})){3})
### end of the selection part
# now can follow the next/different part
# of the regex -
# if you want, in the same way like above
Here is the log with the bad bad word starred out:
May-25-10 14:29:09 12148-10442 [OKAddress] 74.125.82.48
<[email protected]> to: [email protected] [scoring:-5]
sender in OK Address Cache, counter = 10;
May-25-10 14:29:09 12148-10442 [BombBlack] 74.125.82.48
<[email protected]> to: [email protected] [scoring:20]
--
blackRe: '****** (20)' -- [Fwd test bad];
May-25-10 14:29:09 12148-10442 74.125.82.48 <[email protected]> to:
[email protected] Regex:SpamLover 'puke';
May-25-10 14:29:09 12148-10442 [BombRaw] 74.125.82.48
<[email protected]> to: [email protected] [spam passed]
--
passing because spamlover, otherwise blocked (bombRe: '***** (30)') --
[Fwd
test bad] -> /usr/local/assp/spam/10442.eml;
> From: Thomas Eckardt <[email protected]>
> Reply-To: ASSP development mailing list
<[email protected]>
> Date: Tue, 25 May 2010 16:45:42 +0200
> To: ASSP development mailing list <[email protected]>
> Subject: Re: [Assp-test] Way to ignore spamloverre via another regex?
>
> Paul,
>
> make your spamloversRe file this way.
>
> <START>
> assp-do-not-optimize-regex
> # if you use V2 -
> # write the line above anywhere
> # to prevent assp from trying
> # to optimize this file (regex)
> ....
> # any regex of your choice
> ....
> ....
> ### begin of the selection part
> # the first good word must be
> # at the end of the first line
> (?=.*?(\bblue\b
> another word
> ....
> # any regex is possible in this part
> ....
> additional word(s)
> ....
> # the last good and the first bad
> # word must be in this line
> # at the begin and end
> \bwhite\b))(?!.*?(?:(\b(?:\bblack\b
> \byellow\b
> ....
> \bgreen\b
> ....
> # You may also include some other file with some
> # realy bad words (e.g.)
> # include files/blackre.txt
> ....
> # the last bad word must be
> # at the begin of this line
> \bred\b)).*?(?!\g{-1})){3})
> ### end of the selection part
> # now can follow the next/different part
> # of the regex -
> # if you want, in the same way like above
> ....
> ....
> <EOF>
>
> Not absolute sure, but I think it works. This way you have it as
flexible
> as possible.
> You can add and delete words like before.
>
> Thomas
>
>
>
> Von: "Paul K. Dickson" <[email protected]>
> An: ASSP development mailing list <[email protected]>
> Datum: 25.05.2010 15:53
> Betreff: Re: [Assp-test] Way to ignore spamloverre via another
> regex?
>
>
>
>> I think this methode is better than having a 'notSpamLoversRe', because
> a
>> matching entry in this 'notSpamLoversRe' will revert every matched
>> spamloverRe entry - not only one.
>
> Thanks for the suggestion. However I'd rather have something like a
> notspamloverre because otherwise I'd have to enter that regex, often
times
> almost duplicated, all over the place. Easier just to put it once in
one
> file IMO. To each their own :)
>
>
> Paul K. Dickson
> Systems Administrator
> Interagency Information Technologies
> Frederick County Government
> Frederick, MD 21701
> [email protected]
> 301-600-2399/x12399
>
>
>
>> From: Thomas Eckardt <[email protected]>
>> Reply-To: ASSP development mailing list
> <[email protected]>
>> Date: Tue, 25 May 2010 15:45:51 +0200
>> To: ASSP development mailing list <[email protected]>
>> Subject: Re: [Assp-test] Way to ignore spamloverre via another regex?
>>
>> Paul, you can try the following regular expression in spamLoverRe:
>>
>>
>
(?=.*?\b(good1|good2|good3)\b)(?!.*?(?:(\b(?:bad1|bad2|bad3|bad4)\b).*?(?!\g{-
>> 1})){digit
>> of bad count})
>>
>> where good1, good2, good3 are the good words, which consider
'spamlover'
>>
>> but only
>>
>> if the bad words bad1, bad2, bad3, bad4 will not be found a 'digits of
>> bad count' times
>>
>> for example:
>>
>> If you want to consider a mail 'spamlover', if it contains any of the
>> words 'blue' or 'white' -
>> but not if it contains additional 'black' , 'yellow' , 'green' , 'red'
> (in
>> any combination) at least three times - than
>>
>>
>
(?=.*?\b(blue|white)\b)(?!.*?(?:(\b(?:black|yellow|green|red)\b).*?(?!\g{-1}))
>> {3})
>>
>> should do it.
>>
>> I think this methode is better than having a 'notSpamLoversRe', because
> a
>> matching entry in this 'notSpamLoversRe' will revert every matched
>> spamloverRe entry - not only one.
>>
>>
>>
>> Thomas
>>
>>
>>
>> Von: "Paul K. Dickson" <[email protected]>
>> An: ASSP development mailing list <[email protected]>
>> Datum: 24.05.2010 21:43
>> Betreff: [Assp-test] Way to ignore spamloverre via another
regex?
>>
>>
>>
>> I there a way to do that? For instance, I want 90% of mail to be spam
>> lover
>> that contains word N. However, there are a fair amount of mails that
> come
>> in that have word N in it that are also obviously spam by other vulgar
>> words
>> in them.
>>
>> So I¹m in a bit of a catch 22 situation. Love all mail w/ word N in it
> no
>> matter how vulgar or spammy, resulting in 10% of it being nasty spam,
or
>> remove word N from spamloverre and risk getting mail I want tagged but
>> passed completely blocked.
>>
>>
>>
>> Is there a way to say ³use spamloverre EXCEPT if it contains X bad
>> word(s)²?
>>
>>
>>
>>
>> Paul K. Dickson
>> Systems Administrator
>> Interagency Information Technologies
>> Frederick County Government
>> Frederick, MD 21701
>> [email protected]
>> 301-600-2399/x12399
>>
>>
>>
>
------------------------------------------------------------------------------
>>
>> _______________________________________________
>> Assp-test mailing list
>> [email protected]
>> https://lists.sourceforge.net/lists/listinfo/assp-test
>>
>>
>>
>> DISCLAIMER:
>> *******************************************************
>> This email and any files transmitted with it may be confidential,
> legally
>> privileged and protected in law and are intended solely for the use of
> the
>>
>> individual to whom it is addressed.
>> This email was multiple times scanned for viruses. There should be no
>> known virus in this email!
>> *******************************************************
>>
>>
>>
>
------------------------------------------------------------------------------
>>
>> _______________________________________________
>> Assp-test mailing list
>> [email protected]
>> https://lists.sourceforge.net/lists/listinfo/assp-test
>
>
>
------------------------------------------------------------------------------
>
> _______________________________________________
> Assp-test mailing list
> [email protected]
> https://lists.sourceforge.net/lists/listinfo/assp-test
>
>
>
>
> DISCLAIMER:
> *******************************************************
> This email and any files transmitted with it may be confidential,
legally
> privileged and protected in law and are intended solely for the use of
the
>
> individual to whom it is addressed.
> This email was multiple times scanned for viruses. There should be no
> known virus in this email!
> *******************************************************
>
>
>
------------------------------------------------------------------------------
>
> _______________________________________________
> Assp-test mailing list
> [email protected]
> https://lists.sourceforge.net/lists/listinfo/assp-test
------------------------------------------------------------------------------
_______________________________________________
Assp-test mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/assp-test
DISCLAIMER:
*******************************************************
This email and any files transmitted with it may be confidential, legally
privileged and protected in law and are intended solely for the use of the
individual to whom it is addressed.
This email was multiple times scanned for viruses. There should be no
known virus in this email!
*******************************************************
------------------------------------------------------------------------------
_______________________________________________
Assp-test mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/assp-test
