From usr7-28hl at xemaps.com  Mon Nov  1 03:23:24 2004
From: usr7-28hl at xemaps.com (usr7-28hl@xemaps.com)
Date: Mon Nov  1 02:23:32 2004
Subject: [Spambayes] IMAP Error: Must specify server details first
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E8FBC8C0@its-xchg4.massey.ac.nz>
References: <ECBA357DDED63B4995F5C1F5CBE5B1E8FBC8C0@its-xchg4.massey.ac.nz>
Message-ID: <41859E1C.1000609@mailsnare.net>

I thought I had posted this followup message on October 7th and 
wondered why I hadn't received a reply. In the meantime, I've 
been unable to use SpamBayes. I just discovered the message in my 
Drafts folder! Sigh....

Tony Meyer wrote:

  >>Wow, Tony, such service! Do you find time for anything else?
> 
> Well, a great many other people also work on the project, of course :)

Of course. I was referring to your numerous posts to this list,
and to the fact that you replied to my first message even
immediately.

> And
> sometimes (like now) replies can take some time (sorry) - and even then,
> there are still 212 spambayes@python.org messages that I would like time to
> address after this one...

Sorry to be impatient....

[snip]

>>Now, tonight, when I start the script:
>>
>>sb_imapfilter.py -b -v
>>
>>I'm not getting a browser window. The verbose output gives me 
>>three messages, which look okay, with no errors, but no 
>>browser. And opening a browser window and entering the URL 
>>manually gives me an error.

[snip[

> It's very strange that you don't get an error, and can't access the web
> interface.  Has anything else changed in this time?  If you telnet to port
> 8880 when imapfilter is running can you connect?  (Or you could use netstat
> to see if it's listed).

Just tried it again. Here's what I entered at the command prompt,
and the results in each case:

1. >sb_imapfilter.py -c -t -l 5
SpamBayes IMAP Filter Version 0.4 (May 2004)
and engine SpamBayes Engine Version 0.3 (January 2004).

2. >sb_imapfilter.py -b -v
SpamBayes IMAP Filter Version 0.4 (May 2004)
and engine SpamBayes Engine Version 0.3 (January 2004).

Loading database C:\...\SpamBayes\Proxy\hammie.db...
Loading state from c:\...\SpamBayes\Proxy\hammie.db database
c:\...\SpamBayes\Proxy\hammie.db is an existing database, with 0
spam and 0 ham
Done.

3. >telnet localhost:8880
Connecting to localhost:8880...Could not open connection to the
host, on post 23: Connect failed

From jpaonessa at sympatico.ca  Mon Nov  1 03:46:53 2004
From: jpaonessa at sympatico.ca (Joseph Paonessa)
Date: Mon Nov  1 03:46:58 2004
Subject: [Spambayes] spambay
Message-ID: <20041101024656.TNUJ15612.tomts16-srv.bellnexxia.net@financesecretar>

I can't uninstall spambays.  I have XP home and using microsoft access
2003..I tried going to add or remove...It was removed but its still in my
outlook...then I right clicked start and tried to 1 by 1 get rid of
programs..but its still there. I mean here...
 
please advise
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.python.org/pipermail/spambayes/attachments/20041031/03df1dd6/attachment.html
From tameyer at ihug.co.nz  Mon Nov  1 05:21:05 2004
From: tameyer at ihug.co.nz (Tony Meyer)
Date: Mon Nov  1 05:21:12 2004
Subject: [Spambayes] RE: [spambayes-dev] A query on Frequently Asked
	Question 4.6
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E86ED5DE@its-xchg4.massey.ac.nz>
Message-ID: <ECBA357DDED63B4995F5C1F5CBE5B1E86C37AF@its-xchg4.massey.ac.nz>

> The FAQ item 4.6 above addressed something that I've 
> wondered about since I got SB going around July this year,
> that is, "Where does all the 'saved' spam go?" <smile>. 

The messages that end up in your mailer stay there, unless you tell it to do
something with it (with the Outlook plug-in, this means they stay in the
designated spam folder).  The copies of the messages that sb_server makes
are moved to a folder where they are stored for a set amount of time (by
default 7 days) and then are permanently deleted (like if you choose
"discard" from the review page).

The information that SpamBayes collects from messages that it's trained on
is stored in a database (called hammie.db or default_bayes_database.db by
default).  This is just a list of all the tokens ('words') that have been in
all of those messages and a count of how many good & how many bad messages
each token has been seen.  It's not possible (with an unpatched SpamBayes)
to go backwards from the database to a particular message - to 'untrain',
you have to have a copy of the original message around.

> I do have a lot of spam now, 
> 16,054 spams and 1,725 hams. 
> The thing is where do I go to cut down on the amount of spam? 
> I see the comment, "Warning: you have much more spam 
> than ham - SpamBayes 
> works best with approximately even numbers of ham and spam." 
> in the 'Status and 
> Configuration' section.

Firstly, the golden rule is "if it's working, don't change anything" (where
working would be defined as classifying well enough that you are happy with
the results).  All sets of mail are different, but it does appear that
results are better when the balance is reasonably even - and certainly that
large imbalances (eg 50::1) can cause weird results.  You're at 9.3::1,
which isn't extremely bad, although it wouldn't be surprising if results
weren't optimum.

That's quite a large database, too, really.  Many of the developers have
databases with less than a thousand messages in total, with good results.
However, at the moment, it's only possible to add to the db, not remove from
it.  The general practice is to retrain from scratch when that's desired -
since SpamBayes learns quickly, that's not usually a problem.

I'd suggest reading (if you haven't already) the information about training
on the wiki: <http://entrian.com/sbwiki>.  In particular, doing 'mistake
based training' or 'nonedge training' may help keep the db small, *may* help
the imbalance, and may help the results.

We're trying to figure out ways to help with the imbalance issue for future
releases, although there's no clear solution as yet.

(You could put aside the current databases, try out, with fresh ones, a new
training regime and see if that works.  If it doesn't, you can just put the
old database back!).

> 4.3   How do I train SpamBayes (forward/bounce method)?
[...]
> ...I saw from that item 4.3 that there are locations at...
> spambayes_spam@localhost
> ...and at...
> spambayes_ham@localhost
> ...are these locations on this my computer or are they 
> located at my ISP's server? 

These aren't locations, they are special email addresses.  The idea is: 

  All your outgoing mail goes via a SMTP server (at your ISP, probably).
  Rather than have to use the review page, you can bounce/forward messages
  to SpamBayes to indicate that they need to be trained.  The way this
  works is to have all outgoing mail go through a proxy on its way to the
  SMTP server (like incoming mail goes through a proxy on its way from the
  POP server).  This SpamBayes SMTP proxy examines the recipient list of
  each outgoing message and if it is going to one of those two special
  addresses (which don't exist), then they are intercepted and the messages
  are used for training instead.  Some people use this method for training,
  but it's not particularly popular, and does have some flaws.  It's better
  to use the review page in most cases.  (There's another, unfinished,
script
  that offers all-in-the-mail-client training, which I have more hope
  for).

=Tony.Meyer

-- 
Please always include the list (spambayes@python.org) in your replies
(reply-all), and please don't send me personal mail about SpamBayes.
http://www.massey.ac.nz/~tameyer/writing/reply_all.html explains this.

From ececconi at visteon.com  Mon Nov  1 09:19:44 2004
From: ececconi at visteon.com (Cecconi, Enrico (E.))
Date: Mon Nov  1 09:45:00 2004
Subject: [Spambayes] spam with 100% score not moved to junk ++
Message-ID: <1468DAC3F7E2704686B5AEE348137CD40315C495@vistex33.dunton.visteon.com>

Hi,

the following 2 emails have been filtered and score 100% as spam but were
not moved to my designated junk folder.


 <<cheap oem software shipping worldwide.  Andrew butane>>  <<Heighten
sensation>> 

any ides why?

rgds,


Enrico Cecconi M.Sc.
Systems Engineer (Audio)
Visteon (UK) Ltd.
GB-28/H Electronics Tech. Centre
Dunton Engineering Centre
Laindon SS15 6EE
* +44-1268-497813
email: ececconi@visteon.com
> CONFIDENTIALITY NOTICE: This e-mail message including attachments, if any,
> is intended only for the person or entity to which it is addressed and may
> contain confidential and /or privileged material. Any unauthorized review,
> use, disclosure or distribution is prohibited. If you are not the intended
> recipient, please contact the sender by reply e-mail and destroy all
> copies of the original message. If you are the intended recipient but do
> not wish to receive communications through this medium, please do advise
> the sender immediately.
forgot to say am using

Windows 98SE, Outlook 2000 (9.0.0.2711), Spambayes outlook addin Bin Ver
1.0rc2 (june2004)


 <<spambayes1.log>> 
-------------- next part --------------
An embedded message was scrubbed...
From: Terry Lennox <reannon.rylan2002@hispeed.ch>
Subject: cheap oem software shipping worldwide.  Andrew butane
Date: Sat, 30 Oct 2004 07:36:01 -0500
Size: 3366
Url: http://mail.python.org/pipermail/spambayes/attachments/20041101/614e17d0/attachment.mht
-------------- next part --------------
An embedded message was scrubbed...
From: sharonda bailey <worrisas@mail.bulgaria.com>
Subject: Heighten sensation
Date: Sat, 30 Oct 2004 05:15:06 -0500
Size: 1640
Url: http://mail.python.org/pipermail/spambayes/attachments/20041101/614e17d0/attachment-0001.mht
-------------- next part --------------
A non-text attachment was scrubbed...
Name: spambayes1.log
Type: application/octet-stream
Size: 1249 bytes
Desc: not available
Url : http://mail.python.org/pipermail/spambayes/attachments/20041101/614e17d0/spambayes1.obj
From ececconi at visteon.com  Mon Nov  1 09:05:44 2004
From: ececconi at visteon.com (Cecconi, Enrico (E.))
Date: Mon Nov  1 10:32:55 2004
Subject: [Spambayes] spam with 100% score not moved to junk
Message-ID: <1468DAC3F7E2704686B5AEE348137CD40315C493@vistex33.dunton.visteon.com>

Hi,

the following 2 emails have been filtered and score 100% as spam but were
not moved to my designated junk folder.


 <<cheap oem software shipping worldwide.  Andrew butane>>  <<Heighten
sensation>> 

any ides why?

rgds,


Enrico Cecconi M.Sc.
Systems Engineer (Audio)
Visteon (UK) Ltd.
GB-28/H Electronics Tech. Centre
Dunton Engineering Centre
Laindon SS15 6EE
* +44-1268-497813
email: ececconi@visteon.com
> CONFIDENTIALITY NOTICE: This e-mail message including attachments, if any,
> is intended only for the person or entity to which it is addressed and may
> contain confidential and /or privileged material. Any unauthorized review,
> use, disclosure or distribution is prohibited. If you are not the intended
> recipient, please contact the sender by reply e-mail and destroy all
> copies of the original message. If you are the intended recipient but do
> not wish to receive communications through this medium, please do advise
> the sender immediately.
> 
> 
 <<spambayes1.log>> 
-------------- next part --------------
An embedded message was scrubbed...
From: Terry Lennox <reannon.rylan2002@hispeed.ch>
Subject: cheap oem software shipping worldwide.  Andrew butane
Date: Sat, 30 Oct 2004 07:36:01 -0500
Size: 3366
Url: http://mail.python.org/pipermail/spambayes/attachments/20041101/0c468260/attachment.mht
-------------- next part --------------
An embedded message was scrubbed...
From: sharonda bailey <worrisas@mail.bulgaria.com>
Subject: Heighten sensation
Date: Sat, 30 Oct 2004 05:15:06 -0500
Size: 1640
Url: http://mail.python.org/pipermail/spambayes/attachments/20041101/0c468260/attachment-0001.mht
-------------- next part --------------
A non-text attachment was scrubbed...
Name: spambayes1.log
Type: application/octet-stream
Size: 1249 bytes
Desc: not available
Url : http://mail.python.org/pipermail/spambayes/attachments/20041101/0c468260/spambayes1.obj
From scott at lentigo.net  Mon Nov  1 14:07:29 2004
From: scott at lentigo.net (Scott Burns)
Date: Mon Nov  1 14:07:35 2004
Subject: [Spambayes] IMAP Error: Must specify server details first
In-Reply-To: <41859E1C.1000609@mailsnare.net>
References: <ECBA357DDED63B4995F5C1F5CBE5B1E8FBC8C0@its-xchg4.massey.ac.nz>
	<41859E1C.1000609@mailsnare.net>
Message-ID: <41863511.4050500@lentigo.net>

I think your telnet syntax is wrong, try:

   > telnet localhost 8880

instead.  It doesn't look like your telnet recognizes ":" as a delimiter 
between host and port number so it's trying to connect to a host named 
"localhost:8880" on the telnet port of 23.

...s.

usr7-28hl@xemaps.com wrote:
[snip]
> Just tried it again. Here's what I entered at the command prompt,
> and the results in each case:
> 
> 1. >sb_imapfilter.py -c -t -l 5
> SpamBayes IMAP Filter Version 0.4 (May 2004)
> and engine SpamBayes Engine Version 0.3 (January 2004).
> 
> 2. >sb_imapfilter.py -b -v
> SpamBayes IMAP Filter Version 0.4 (May 2004)
> and engine SpamBayes Engine Version 0.3 (January 2004).
> 
> Loading database C:\...\SpamBayes\Proxy\hammie.db...
> Loading state from c:\...\SpamBayes\Proxy\hammie.db database
> c:\...\SpamBayes\Proxy\hammie.db is an existing database, with 0
> spam and 0 ham
> Done.
> 
> 3. >telnet localhost:8880
> Connecting to localhost:8880...Could not open connection to the
> host, on post 23: Connect failed

-- 
Scott Burns <scott@lentigo.net> <http://www.lentigo.net/scott>
pub  1024D/9DA64618 2001-11-17 Scott Burns <scott@lentigo.net>
Fingerprint: 2F1B A22E 33C3 FD3D BBE4 D5E2 728B 4753 9DA6 4618

From papaDoc at videotron.ca  Mon Nov  1 15:35:40 2004
From: papaDoc at videotron.ca (Remi Ricard)
Date: Mon Nov  1 15:35:43 2004
Subject: [Spambayes] Re: [spambayes-dev] A query on Frequently Asked
	Question 4.6
In-Reply-To: <VA.00000997.00ebeed7@ntlworld.com>
References: <VA.00000997.00ebeed7@ntlworld.com>
Message-ID: <418649BC.3060902@videotron.ca>

Hi Harry,

>     The FAQ item 4.6 above addressed something that I've wondered about since 
>I got SB going around July this year, that is, "Where does all the 'saved' spam 
>go?" <smile>. 
>     I do have a lot of spam now, 
>     16,054 spams and 1,725 hams. 
>     The thing is where do I go to cut down on the amount of spam? 
>     I see the comment, "Warning: you have much more spam than ham - SpamBayes 
>works best with approximately even numbers of ham and spam." in the 'Status and 
>Configuration' section.
>     I'll have another look and see if I can figure out where all the spam is 
>located...
>  
>
It could help if you tell us which version of  Spambayes and how you use it.
If you use sb_server they should be in your inbox or if you have set a 
filter rule then you should not.

If you use something else they should be in one of your folder since 
spambayes don't delete anything.

>================
>4.3   How do I train SpamBayes (forward/bounce method)?
>
>Alternatively, when you receive an incorrectly classified message, you can 
>forward it to the SMTP proxy for training. If the message should have been 
>classified as spam, forward or bounce the message to spambayes_spam@localhost, 
>and if the message should have been classified as ham, forward it to 
>spambayes_ham@localhost. You can still review the training through the web 
>interface, if you wish to do so. You should ensure that the "lookup message in 
>cache" option is set to True/Yes before you use this.
>
>Note that some mail clients (particularly Outlook Express) do not forward all 
>headers when you bounce, forward or redirect mail. We do not recommend using 
>the SMTP proxy with these clients.
>===============
>     ...I saw from that item 4.3 that there are locations at...
>     spambayes_spam@localhost
>     ...and at...
>     spambayes_ham@localhost
>     ...are these locations on this my computer or are they located at my ISP's 
>server? 
>  
>
It is "on" your computer  what  really  append is  your email program 
will send the message with spambayes_spam@localhost
to the port x (in think it is 25) on your computer then spambayes 
(working as a proxy) will see the message and grab it to train
on it and then discard it.


>     This is a standalone machine with a cable-modem broadband connection to 
>NTL...
>     I've just used XP's Search facility to see if,  spambayes_spam@localhost , 
>exists as a file on this machine it didn't find such and I didn't expect that 
>it would. 
>  
>
I think you can find the info in the file options.py but the info might 
be split in 2 part
"spambayes_spam" and localhost since the email address for the spam is 
configurable (spambayes _spam)


>     Are these thousands of spam actually located on my ISP's server, that is 
>on NTL's server? 
>  
>
No
But this could be yes if your email program is set to keep email on server.

>     If so how do I go about accessing this that I might reduce the amount of 
>spam to something near the same 'ham' figure of 1,725? 
>  
>
You can do it simply by not training on all spam that comes in.
The problem now is you can't reduce the number of spam you can only add 
info to the database.
So you can train on more ham to have a balanced database.

But I think, it might be better to start from scratch

>     
>     Excuse the long explanation. 
>  
>
It is always better to have more info so don't excuse yourself  ;-)
See the start of the email (We need more info......)

Remi
From cinnamonbelle at wildmail.com  Mon Nov  1 16:21:33 2004
From: cinnamonbelle at wildmail.com (C B)
Date: Mon Nov  1 16:21:24 2004
Subject: [Spambayes] SpamBayes feature suggestion
Message-ID: <B29BF5F0AE449B54891312AE59B8A54F@cinnamonbelle.wildmail.com>

Am so far impressed with SpamBayes' intuitiveness, ease of use, and - 
though it's not quite near perfect yet - its accuracy... actually I 
have a suggestion and a half.

1) Would be great if SpamBayes could suggest and implement a settings 
change at install that lets a person right-click on their junk e-mail 
folder and see a menu selection for 'empty junk e-mail folder' - just 
like the trash folder. Maybe similar functionality for the junk 
suspects folder.

2) It might get prohibitively bulky to start processing individual 
blacklists - or for that matter whitelists - as part of SpamBayes 
(which after all is supposed to be Bayesian)... but if there's a way 
to do that fast, it would help. I find Microsoft Outlook's message 
rules processing way too slow to want to use it on incoming e-mail... 
no matter how few things I'm asking it to find.

Carry on. :)  Great work... so far better than a commercial packaged-
with-updates anti-spam consumer product I've tried.


Care2 make the world greener!
Support Care2 Email: 1,400 whales may be killed this year. Take action! http://www.care2.com/go/z/17954
From reed at Penningtonlawfirm.com  Mon Nov  1 18:42:13 2004
From: reed at Penningtonlawfirm.com (Reed Odeneal)
Date: Mon Nov  1 18:42:26 2004
Subject: [Spambayes] Problems
Message-ID: <EF8DFC98058A824BAA11BF59C93EE11194003D@comserver.PLF>

We tried:

Renamed the database, as instructed
Re-trained from junk suspects/ham
After trying this, the problem still exists, Recovering junk suspects to
the inbox takes 30 seconds or more for the email to show up into the
inbox. It does the same thing if I do a drag and drop or by using the
toolbar. The database size was right around 1.6 megs before it was
renamed/recreated.

Ran a detect and repair on outlook2k3, and the problem still exists.

Any more recommendations?



-----Original Message-----
From: Tony Meyer [mailto:tameyer@ihug.co.nz] 
Sent: Thursday, October 28, 2004 8:03 PM
To: Reed Odeneal; spambayes@python.org
Subject: RE: [Spambayes] Problems

> The problem that we are having is when you click to
> recover an email from spam, out of the Junk Suspects
> folder, it takes AT LEAST 15 seconds for the email to
> show up in the inbox. It did not use to do this.

How many messages are in your database?  (The SpamBayes Manager dialog
shows
this).

If you move a message manually (e.g. drag and drop), is it quick?  If
so,
then it's the training that's taking a long time (which is the mostly
likely
explanation).

You could try renaming the database files ("Show Data Directory" on the
Advanced tab of the Manager dialog will show you them - they are
default_bayes_database.db and default_message_database.db), and
retraining,
and see if that fixes the problem.  If it does, we would be interested
to
know, as there have been a couple of other 'slow training' reports
recently,
which indicates that there might be a problem to fix.

=Tony.Meyer

-- 
Please always include the list (spambayes@python.org) in your replies
(reply-all), and please don't send me personal mail about SpamBayes.
http://www.massey.ac.nz/~tameyer/writing/reply_all.html explains this.



From papaDoc at videotron.ca  Mon Nov  1 23:34:16 2004
From: papaDoc at videotron.ca (Remi Ricard)
Date: Mon Nov  1 23:34:33 2004
Subject: [Spambayes] Re: [spambayes-dev] A query on Frequently Asked
	Question 4.6
In-Reply-To: <VA.0000099e.00ad4a76@ntlworld.com>
References: <VA.00000997.00ebeed7@ntlworld.com> <418649BC.3060902@videotron.ca>
	<VA.0000099e.00ad4a76@ntlworld.com>
Message-ID: <4186B9E8.4070409@videotron.ca>

Hi Harry,

>      I'll try and answer your queries as well as I can. 
>  
>
Good

>"SpamBayes POP3 Proxy Version 1.0 (July 2004)
>Spambayes Web Interface, 19:32 on Monday November 01 2004. Spambayes.org
>  
>
>     I don't know what you mean when you say, "If your use sb_server they should be...". Until I set the filter in 'Virtual Access' all of the spam had landed into my Inbox. Then with the filter in place they all wound up in that 'Spam' folder which was fine. 
>     **** I've just noticed in the Advanced Configuration section...
>  
>
With this info I can say that you are using a recent version of 
sb_server (This is the command line name for the Spambayes POP3 proxy).

>  
>
>>>Train when filtering: Train when filtering? After filtering a message, hammie can then train itself on the judgement (ham or spam). This can speed things up with a procmail-based solution. If you do enable this, please make sure to retrain any mistakes. Otherwise, your word database will slowly become 
>>>      
>>>
>useless.<<
>      ...I left that set at its default that is with the radio button at 'No'; which still reads as being the correct position but perhaps it should be at 'Yes'. Perhaps I'll try running it on 'yes' a bit later not at the moment till I hear from you. 
>  
>
The problem with this is if you receive more spam than ham your database 
will be more and more unbalance.
I think you should enable this after you have train on some spam and ham 
then turn this options off after you find SpamBayes to be accurate enough.

>>>    If so how do I go about accessing this that I might reduce the amount of 
>>>spam to something near the same 'ham' figure of 1,725? 
>>> 
>>>
>>>      
>>>
>>You can do it simply by not training on all spam that comes in.
>>The problem now is you can't reduce the number of spam you can only add 
>>info to the database.
>>So you can train on more ham to have a balanced database.
>>    
>>
>
>      Oh, my goodness, by "...simply not training on all..."; so does the mean I should have been using the 'discard' radio buttons on lots, if not all of the spam that came in around the time that, "Warning: you have much more spam than ham - SpamBayes works best with approximately even numbers of ham and 
>spam." message appeared? As you said, 'simply' - dumb. 
>  
>
Usually if there is a button in the web interface the button is there 
for something ;-)


>===============
>  
>
>>But I think, it might be better to start from scratch
>>    
>>
>
>     Will that be accomplished simply by an uninstall using Add/Remove programs and then a re-install?
>
As Tony said in another mail you can save/rename the old databases and 
restart from scratch. If something breaks you will have the old
ones.
Look for the files hammie.db and spambayes.messageinfo.db in 
c:\Documents and Settings\your_user_id\Application Data\Spambays\Proxy

> Why I'm asking that is, won't there be some data files that have to be cleared out too? Or will the Add/Remove action get rid of any such files? 
>  
>
I don't know since I'm using an old version installed from cvs (If it 
works don't fix it)

Remi
From tameyer at ihug.co.nz  Tue Nov  2 01:22:01 2004
From: tameyer at ihug.co.nz (Tony Meyer)
Date: Tue Nov  2 01:24:45 2004
Subject: [Spambayes] spambay
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E80140DB4A@its-xchg4.massey.ac.nz>
Message-ID: <ECBA357DDED63B4995F5C1F5CBE5B1E8FBC993@its-xchg4.massey.ac.nz>

> I can't uninstall spambays.

Please see FAQ 3.16:

<http://spambayes.org/faq.html#how-do-i-uninstall-the-plug-in>

As described in there, the toolbar isn't removed on uninstall (this is an
unresolved bug), so you have to manually remove it.

=Tony.Meyer

-- 
Please always include the list (spambayes@python.org) in your replies
(reply-all), and please don't send me personal mail about SpamBayes.
http://www.massey.ac.nz/~tameyer/writing/reply_all.html explains this.

From tameyer at ihug.co.nz  Tue Nov  2 01:35:43 2004
From: tameyer at ihug.co.nz (Tony Meyer)
Date: Tue Nov  2 01:36:23 2004
Subject: [Spambayes] SpamBayes feature suggestion
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E80140DBFC@its-xchg4.massey.ac.nz>
Message-ID: <ECBA357DDED63B4995F5C1F5CBE5B1E86C37B1@its-xchg4.massey.ac.nz>

> 1) Would be great if SpamBayes could suggest and implement a settings 
> change at install that lets a person right-click on their junk e-mail 
> folder and see a menu selection for 'empty junk e-mail folder' - just 
> like the trash folder. Maybe similar functionality for the junk 
> suspects folder.

There's a tracker for this request:

[ 788755 ] Easy-empty Spam, Potential Spam folders
<http://sf.net/tracker/index.php?func=detail&aid=788755&group_id=61702&atid=
498106>

The status at the moment is that the SpamBayes menu will include a "Empty
Spam Folder" item in the first 1.1 release (or if you run from CVS Head,
already does).  As I understand it, adding an item to the context
(right-click) menu is extremely difficult, or not possible.  However,
Outlook 2003 includes a "Junk Mail" folder that has this functionality, and
you can use that folder as the SpamBayes spam folder, so as people gradually
move to newer Outlook versions, they'll get this functionality, too.

> 2) It might get prohibitively bulky to start processing individual 
> blacklists - or for that matter whitelists - as part of SpamBayes 
> (which after all is supposed to be Bayesian)... but if there's a way 
> to do that fast, it would help. I find Microsoft Outlook's message 
> rules processing way too slow to want to use it on incoming e-mail... 
> no matter how few things I'm asking it to find.

Have you read FAQ 6.6?  It covers (in depth) why there isn't any
white/black-listing with SpamBayes:

<http://spambayes.org/faq.html#why-don-t-you-add-whitelisting-blacklisting-t
o-spambayes>

=Tony.Meyer

-- 
Please always include the list (spambayes@python.org) in your replies
(reply-all), and please don't send me personal mail about SpamBayes.
http://www.massey.ac.nz/~tameyer/writing/reply_all.html explains this.

From tameyer at ihug.co.nz  Tue Nov  2 01:40:22 2004
From: tameyer at ihug.co.nz (Tony Meyer)
Date: Tue Nov  2 01:40:34 2004
Subject: [Spambayes] spam with 100% score not moved to junk
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E80140DBF8@its-xchg4.massey.ac.nz>
Message-ID: <ECBA357DDED63B4995F5C1F5CBE5B1E86C37B2@its-xchg4.massey.ac.nz>

> the following 2 emails have been filtered and score 100% as 
> spam but were not moved to my designated junk folder.
[...]
> any ides why?

Unfortunately, the log file you included was not the one that was active
during the time that those messages were filtered (there's no record of any
filtering in the log at all).

When you say that they were filtered and score 100% - how do you know that?
Do you have the spam field visible and saw the score appear there?  If you
used the "Show spam clues" option that gives you a current score, but
doesn't tell you whether the message was filtered, or what it originally
scored.

Is this something that happens often, or has only happened once?

=Tony.Meyer

-- 
Please always include the list (spambayes@python.org) in your replies
(reply-all), and please don't send me personal mail about SpamBayes.
http://www.massey.ac.nz/~tameyer/writing/reply_all.html explains this.

From tameyer at ihug.co.nz  Tue Nov  2 01:45:48 2004
From: tameyer at ihug.co.nz (Tony Meyer)
Date: Tue Nov  2 01:45:58 2004
Subject: [Spambayes] Interesting Request
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E80140D9E0@its-xchg4.massey.ac.nz>
Message-ID: <ECBA357DDED63B4995F5C1F5CBE5B1E86C37B3@its-xchg4.massey.ac.nz>

> Is there any way to create a Windows based DLL of the main 
> SpamBayes engine? I have multiple ideas for how to use an API 
> like this and am 110% convinced this is the best spam 
> detection routine on the market.

What you are after, presumably, is the functionality contained in the
tokenizer.py and classifier.py scripts.  These take a message and convert it
to a set of tokens and take a set of tokens and a database and give a score
(respectively).  They're very simple to use - something like:

>>> import email
>>> msg = email.message_from_file(open("d:\\example.txt"))
>>> from spambayes.tokenizer import tokenize
>>> tokens = tokenize(msg)
>>> from spambayes.classifier import Classifier
>>> Classifier().spamprob(tokens)
0.5

(The score is 0.5, because the classifier hasn't learnt anything).  The
source is all well documented, and there are several simple scripts
(sb_filter, for example) that give good examples of how to do simple tasks
like scoring or training a message.

To create a DLL, you'd just write a Python script that did what you wanted,
then use something like py2exe to turn it into a DLL.

=Tony.Meyer

-- 
Please always include the list (spambayes@python.org) in your replies
(reply-all), and please don't send me personal mail about SpamBayes.
http://www.massey.ac.nz/~tameyer/writing/reply_all.html explains this.

From papaDoc at videotron.ca  Tue Nov  2 03:01:49 2004
From: papaDoc at videotron.ca (Remi Ricard)
Date: Tue Nov  2 03:01:52 2004
Subject: [Spambayes] Re: [spambayes-dev] A query on Frequently Asked
	Question 4.6
In-Reply-To: <VA.000009a2.01546501@ntlworld.com>
References: <VA.00000997.00ebeed7@ntlworld.com> <418649BC.3060902@videotron.ca>
	<VA.0000099e.00ad4a76@ntlworld.com>
	<4186B9E8.4070409@videotron.ca> <VA.000009a2.01546501@ntlworld.com>
Message-ID: <4186EA8D.5010801@videotron.ca>

Hi Harry,

>>As Tony said in another mail you can save/rename the old databases and 
>>restart from scratch. If something breaks you will have the old
>>ones.
>>Look for the files hammie.db and spambayes.messageinfo.db in 
>>c:\Documents and Settings\your_user_id\Application Data\Spambays\Proxy
>>    
>>
>     I don't know where that email from Tony is located (if emailed me I may have lost it 
>while dumping spam) 
>
I forwarded the message to you. Usually Tony send a reply directly to 
the person asking for info and a copy to the mailing list.



>     (I had to find them using Search - just looking in the Documents and Settings bit I 
>couldn't see the 'Application Data' part of the path. There will be a way to show this as it 
>must be hidden). 
>  
>
This directory by default is hiddent
Launch the explorer ("Windows key"-E)
Go to Tools->folder options
Then View
In the windows Select "Show-hidden  files and folder"



>==========
>  
>
>>I don't know since I'm using an old version installed from cvs (If it 
>>works don't fix it)
>>    
>>
>
>     That lost me somewhat.
>
What I was trying to say is I'm running an old version since my Linux 
box (the one doing the filtering) as problem generating a db for my 
windows box. (They don't use the same version of the database) and I 
need a backup since I can't live whitout SpamBayes and sometime
my connection to my server at home dies and I do the filtering on the 
windows machine at work.


> 
>     At present when I try to run the *Review Messages* it no longer sends the spam to that 
>'Spam' folder as it did for weeks and weeks. What it does is break down and presents me with 
>a short page that looks like this...
>     
>========
>  
>
>>500 Server error
>>
>>Traceback (most recent call last):
>>
>>  File "spambayes\Dibbler.pyc", line 461, in found_terminator
>>
>>  File "spambayes\ProxyUI.pyc", line 391, in onReview
>>
>>  File "spambayes\Corpus.pyc", line 214, in takeMessage
>>
>>  File "spambayes\FileCorpus.pyc", line 140, in addMessage
>>
>>  File "spambayes\Corpus.pyc", line 134, in addMessage
>>
>>  File "spambayes\storage.pyc", line 588, in onAddMessage
>>
>>  File "spambayes\storage.pyc", line 598, in train
>>
>>  File "spambayes\message.pyc", line 260, in setId
>>
>>  File "spambayes\message.pyc", line 118, in _getState
>>
>>  File "shelve.pyc", line 118, in __getitem__
>>
>>  File "bsddb\__init__.pyc", line 116, in __getitem__
>>
>>DBRunRecoveryError: (-30982, 'DB_RUNRECOVERY: Fatal error, run database 
>>recovery -- fatal region error detected; run recovery')
>>    
>>
>===========
>     ...I'm not sure if I've already shown that error message you. I have done to someone 
>  
>
Oupssss. Did you stop the server before moving/renaming the files.


><s>. 
>     I have the feeling that the answer is in that last bit where it says, "run database 
>recovery"; mainly because it tells me to <s>. But I don't know how. 
>     Is it a case of going...    Start=>Run=>cmd  and then in the resulting black-background 
>box - which would then be showing  c:\Documents and Settings\Harry Sigerson>  - that I type 
>in DB_RUNRECOVERY   and enter it?
>
This is copied form a previous mail from Tony
-----

This means that your database is corrupted, and cannot be recovered (despite
the wording of the message).  You need to delete the database file and start
training again.  The configuration page tells you where the database file
is, but it's most likely to be called "hammie.db" and be in the same place
as your configuration file.  This is probably C:\Documents and
Settings\[user]\Application Data\SpamBayes\Proxy.  Simply delete the
"hammie.db" and "spambayes.messageinfo.db" files, and all should work again.

Database corruption is quite uncommon these days, but we're not certain how
it can be caused.  Do you have any idea what might have caused yours?  Did
SpamBayes or your computer crash while training, or something like that?
Note that it might not have been just before the error occurred - sometimes

> 
>     
>     This is the area that I am not too conversant with. I am reluctant to act blindly in 
>case I totally banjax this machine. 
>     That was why I thought to go down the reinstall route since I have that latest upgrade 
>stashed away in a partition full of self-extracting application files. 
>  
>
This can be a safe route....

Remi
P.S. If you send the message directly to the mailing list there is more 
chance to have good answer ;-). Since I have subscribed to the mailing 
list I will see the message (But you can still send it also to me).
From relson at osagesoftware.com  Tue Nov  2 03:03:00 2004
From: relson at osagesoftware.com (David Relson)
Date: Tue Nov  2 03:03:03 2004
Subject: [Spambayes] Important: new feature!
In-Reply-To: <1ED4ECF91CDED24C8D012BCF2B034F13064C02CD@its-xchg4.massey.ac.nz>
References: <1ED4ECF91CDED24C8D012BCF2B034F1306FA09A4@its-xchg4.massey.ac.nz>
	<1ED4ECF91CDED24C8D012BCF2B034F13064C02CD@its-xchg4.massey.ac.nz>
Message-ID: <20041101210300.148462dc@osage.osagesoftware.com>

On Fri, 9 Jul 2004 15:04:40 +1200
Tony Meyer wrote:

> > I understand what you're saying, and I've re-trained 
> > spambayes on equal numbers of HAM and SPAM, 300 each. Yet, it 
> > still has a tendency to classify some HAM as spam. A 
> > significant percentage is like maybe 25% of my regular 
> > announcement mail gets spamflagged...
> 
> Something is definitely not working here - a 25% false positive rate
> is extraordinarily high.  You shouldn't even get as high as 2.5%. 
> Perhaps you could send the spam clues report for one of these messages
> (before any training on it) to the list?  We could take a look at it
> and see if there's an obvious reason that it's failing to be
> classified correctly.

To ask the simple question, are you sure that all your training messages
are correctly classified?  An error or two could have a _big_ effect.
From tameyer at ihug.co.nz  Tue Nov  2 07:07:30 2004
From: tameyer at ihug.co.nz (Tony Meyer)
Date: Tue Nov  2 07:08:09 2004
Subject: [Spambayes] Installation Question
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E80140D94F@its-xchg4.massey.ac.nz>
Message-ID: <ECBA357DDED63B4995F5C1F5CBE5B1E86C37B6@its-xchg4.massey.ac.nz>

>>> I downloaded and ran the installation program, but
>>> nothing seems to happen.

[I ask about log files]

> **Yes, actually there are 4.  The first 2 say "registration 
> complete"; the 4th says:
> "Loaded bayes database from 'C:\Documents and 
[...]
> SpamBayes: Watching (for filtering) in 'Personal Folders/Inbox'
> SpamBayes: Watching (for filtering) in 'Hotmail/Inbox'
> SpamBayes: Watching (for incremental training) in 'Personal 
> Folders/Junk
> E-Mail'
> Processing missed spam in folder 'Inbox' by starting a timer
> Processing missed spam in folder 'Inbox' by starting a timer
[...]
> SpamBayes - Disconnecting from Outlook
> SpamBayes has processed zero messages
> Addin terminating: 2 COM client and 2 COM servers exist."

This is strange - the log indicates that SpamBayes is running happily.

Is the SpamBayes toolbar definitely not visible?  If you right-click on the
toolbars and choose "Customize", is the SpamBayes toolbar listed there?

Check to see if SpamBayes is disabled by Outlook: Help->About Microsoft
Outlook->Disabled Items.

The troubleshooting guide has some other tips to try and get the toolbar to
appear, so it would be worth trying those:

<http://cvs.sourceforge.net/viewcvs.py/*checkout*/spambayes/spambayes/Outloo
k2000/docs/troubleshooting.html?rev=HEAD>

(My mailer will break that URL - it should be all on one line)

=Tony.Meyer

-- 
Please always include the list (spambayes@python.org) in your replies
(reply-all), and please don't send me personal mail about SpamBayes.
http://www.massey.ac.nz/~tameyer/writing/reply_all.html explains this.

From ececconi at visteon.com  Tue Nov  2 14:09:53 2004
From: ececconi at visteon.com (Cecconi, Enrico (E.))
Date: Tue Nov  2 14:10:10 2004
Subject: Resend to all: [Spambayes] spam with 100% score not moved to junk
	 
Message-ID: <1468DAC3F7E2704686B5AEE348137CD40315C4A3@vistex33.dunton.visteon.com>

Tony,

Yes I did use the "Show spam clues for current message" which appears under
the "SpamBayes" button on my Outlook toolbar. I used it when the email was
still in my inbox (highlighted).

I've been using SpamBayes for about a year without problems. This has been
happening for the last few months only. I've uninstalled and reinstalled and
re-run the training but it made no difference. I've also unchecked and
rechecked most of the boxes in SpamBayes manager but still can't get any
filtering to happen. There was, however one exception, yesterday one email
was put into my "unsure" folder!?! That's the first time ever that that has
happened, as far as I can remember.

Following on from your email, I've added the "Junk E-mail Type" field to my
inbox. It is empty.

BTW The log file I sent was the one with the latest date and I had only just
started Outlook for the first time, at that stage, yesterday.

regards,

EC

-----Original Message-----
From: Tony Meyer [mailto:tameyer@ihug.co.nz]
Sent: 02 November 2004 00:40
To: 'Cecconi, Enrico (E.)'; spambayes@python.org
Subject: RE: [Spambayes] spam with 100% score not moved to junk


> the following 2 emails have been filtered and score 100% as 
> spam but were not moved to my designated junk folder.
[...]
> any ides why?

Unfortunately, the log file you included was not the one that was active
during the time that those messages were filtered (there's no record of any
filtering in the log at all).

When you say that they were filtered and score 100% - how do you know that?
Do you have the spam field visible and saw the score appear there?  If you
used the "Show spam clues" option that gives you a current score, but
doesn't tell you whether the message was filtered, or what it originally
scored.

Is this something that happens often, or has only happened once?

=Tony.Meyer

-- 
Please always include the list (spambayes@python.org) in your replies
(reply-all), and please don't send me personal mail about SpamBayes.
http://www.massey.ac.nz/~tameyer/writing/reply_all.html explains this.
From cgreenleaf at tocnh.com  Tue Nov  2 14:07:46 2004
From: cgreenleaf at tocnh.com (Cyndy Greenleaf)
Date: Tue Nov  2 14:19:55 2004
Subject: [Spambayes] messages
Message-ID: <D7863BD5EF46A947B8AA049E8557E35CAC9D46@orthoadc.ortho.tocnh.com>

Spambayes has been deleting messages from my sister and leaving porn on
my e mail. What's up? How do I fix this?
 
Cyndy Greenleaf
Nashua Office Manager
The Orthopaedic Center
603-689-1645
cgreenleaf@tocnh.com
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.python.org/pipermail/spambayes/attachments/20041102/e9a9393d/attachment.html
From papaDoc at videotron.ca  Tue Nov  2 14:34:16 2004
From: papaDoc at videotron.ca (Remi Ricard)
Date: Tue Nov  2 14:34:19 2004
Subject: [Fwd: [Spambayes] [spambayes-dev] A query on Frequently Asked
	Question 4.6]
In-Reply-To: <VA.000009a3.007f3ab3@ntlworld.com>
References: <4186E714.6010504@videotron.ca> <VA.000009a3.007f3ab3@ntlworld.com>
Message-ID: <41878CD8.6060904@videotron.ca>

Hi Harry,

This a cut and paste of the message from Tony so you can delete the one 
with the attached email
------------------------------

The FAQ item 4.6 above addressed something that I've 
> wondered about since I got SB going around July this year,
> that is, "Where does all the 'saved' spam go?" <smile>. 


The messages that end up in your mailer stay there, unless you tell it to do
something with it (with the Outlook plug-in, this means they stay in the
designated spam folder).  The copies of the messages that sb_server makes
are moved to a folder where they are stored for a set amount of time (by
default 7 days) and then are permanently deleted (like if you choose
"discard" from the review page).

The information that SpamBayes collects from messages that it's trained on
is stored in a database (called hammie.db or default_bayes_database.db by
default).  This is just a list of all the tokens ('words') that have been in
all of those messages and a count of how many good & how many bad messages
each token has been seen.  It's not possible (with an unpatched SpamBayes)
to go backwards from the database to a particular message - to 'untrain',
you have to have a copy of the original message around.


>> I do have a lot of spam now, 
>> 16,054 spams and 1,725 hams. 
>> The thing is where do I go to cut down on the amount of spam? 
>> I see the comment, "Warning: you have much more spam 
>> than ham - SpamBayes 
>> works best with approximately even numbers of ham and spam." 
>> in the 'Status and 
>> Configuration' section.
>  
>

Firstly, the golden rule is "if it's working, don't change anything" (where
working would be defined as classifying well enough that you are happy with
the results).  All sets of mail are different, but it does appear that
results are better when the balance is reasonably even - and certainly that
large imbalances (eg 50::1) can cause weird results.  You're at 9.3::1,
which isn't extremely bad, although it wouldn't be surprising if results
weren't optimum.

That's quite a large database, too, really.  Many of the developers have
databases with less than a thousand messages in total, with good results.
However, at the moment, it's only possible to add to the db, not remove from
it.  The general practice is to retrain from scratch when that's desired -
since SpamBayes learns quickly, that's not usually a problem.

I'd suggest reading (if you haven't already) the information about training
on the wiki: <http://entrian.com/sbwiki>.  In particular, doing 'mistake
based training' or 'nonedge training' may help keep the db small, *may* help
the imbalance, and may help the results.

We're trying to figure out ways to help with the imbalance issue for future
releases, although there's no clear solution as yet.

(You could put aside the current databases, try out, with fresh ones, a new
training regime and see if that works.  If it doesn't, you can just put the
old database back!).


>> 4.3   How do I train SpamBayes (forward/bounce method)?
>  
>
[...]

>> ...I saw from that item 4.3 that there are locations at...
>> spambayes_spam@localhost
>> ...and at...
>> spambayes_ham@localhost
>> ...are these locations on this my computer or are they 
>> located at my ISP's server? 
>  
>

These aren't locations, they are special email addresses.  The idea is: 

  All your outgoing mail goes via a SMTP server (at your ISP, probably).
  Rather than have to use the review page, you can bounce/forward messages
  to SpamBayes to indicate that they need to be trained.  The way this
  works is to have all outgoing mail go through a proxy on its way to the
  SMTP server (like incoming mail goes through a proxy on its way from the
  POP server).  This SpamBayes SMTP proxy examines the recipient list of
  each outgoing message and if it is going to one of those two special
  addresses (which don't exist), then they are intercepted and the messages
  are used for training instead.  Some people use this method for training,
  but it's not particularly popular, and does have some flaws.  It's better
  to use the review page in most cases.  (There's another, unfinished,
script
  that offers all-in-the-mail-client training, which I have more hope
  for).
---------------------------------------


>In article <4186E714.6010504@videotron.ca>, Remi Ricard wrote:
>  
>
>>[Attachment decoded to FILE://\PROGRA~1\VA\DOWNLOAD\_Spambayes_ RE_ _spambayes-dev_ A query on Frequently Asked Question 4.6]
>>[Unable to display message/rfc822 in \PROGRA~1\VA\DOWNLOAD\msw00811.tmp]
>>
>>    
>>
>Remi, 
>     I got this from you and it landed, as it should, in Virtual Access' (VA's) DOWNLOAD folder (curiously it landed there and 
>decoded itself in the process - I have VA set that any attachment that comes into DOWNLOAD, comes in as an 'Extract' and needs 
>me to okay things before it decodes - stranger and stranger). 
>     In any case it would not open up. I tried to open it up using Word but it told me that it was '...not a valid archive'. 
>     Perhaps there is another way of opening it? 
>  
>
You can try with notepad since an email is simply a text file.
What are you using to read your mail ?


>     ++++++++++
>     Whoa! Now I find that every time my cursor crosses that entry in the VA  DOWNLOAD folder I immediately get an XP  'Save 
>As' box and no amount of cancelling will get rid of it. 
>     I shall wait till I hear from you before trying to delete that attachment. 
>     
>     I'll send this off just now I have another email on hold just now about the SpamBayes saga <s>. I hope you have the 
>patience to follow this. 
>  
>
I'm using Win2K with Thunderbird so I can't hep you on this one sorry .


Remi
From sales at smsmachine.com  Tue Nov  2 16:04:41 2004
From: sales at smsmachine.com (Sales SMS Machine Tools)
Date: Tue Nov  2 16:06:03 2004
Subject: [Spambayes] (no subject)
Message-ID: <004101c4c0ed$484ffbc0$0501a8c0@smsmachine.com>

Junk folder has disappeared from Folder List. Using Outlook 2000.
Junk Suspect folder still there.
Please advise.

Thanks

Ruthanne
SMS Machine Tools
From papaDoc at videotron.ca  Tue Nov  2 16:47:24 2004
From: papaDoc at videotron.ca (Remi Ricard)
Date: Tue Nov  2 16:47:26 2004
Subject: [Spambayes] Re: [spambayes-dev] A query on Frequently Asked
	Question 4.6
In-Reply-To: <VA.000009a7.00bfb965@ntlworld.com>
References: <VA.00000997.00ebeed7@ntlworld.com> <418649BC.3060902@videotron.ca>
	<VA.0000099e.00ad4a76@ntlworld.com>
	<4186B9E8.4070409@videotron.ca> <VA.000009a2.01546501@ntlworld.com>
	<4186EA8D.5010801@videotron.ca> <VA.000009a7.00bfb965@ntlworld.com>
Message-ID: <4187AC0C.70201@videotron.ca>

Hi Harry,

>>What I was trying to say is I'm running an old version since my Linux 
>>box (the one doing the filtering) as problem generating a db for my 
>>windows box. (They don't use the same version of the database) and I 
>>need a backup since I can't live whitout SpamBayes and sometime
>>my connection to my server at home dies and I do the filtering on the 
>>windows machine at work.
>>    
>>
>     Right, I see; I'm all interest. I am trying to get round to installing Linux and going down 
>that road, not only less well travelled but upon whose yellow bricks these tootsies have yet to 
>tread. 
>     I note that you have your own server at home - nice one. 
>  
>
Please don't DOS it since it is only a poor x486  running Freesco..... ;-)

>  
>
>>Oupssss. Did you stop the server before moving/renaming the files.
>>    
>>
>     Remi, I'm not sure about the significance of that but I didn't 'move' those two database 
>files. All I did was copy them elsewhere and then change their names. Is there something wrong 
>in doing that? The 'originals' are still located at the end of the correct path. 
>
>    With regard to Tony's information and query...
>  
>
>>Database corruption is quite uncommon these days, but we're not certain how
>>it can be caused.  Do you have any idea what might have caused yours?  Did
>>SpamBayes or your computer crash while training, or something like that?
>>Note that it might not have been just before the error occurred - sometimes
>>    
>>
This might have been caused by the hard reboot you did to make your PC 
reach its homepage...


>     Well, I'll put this on Hold and go and delete, hammie.db and its pal, 
>spambayes.messageinfo.db  and enjoying a revived SpamBayes, will come back and let you know...
>     {pause for a minute to count to ten - I lost a chunk of this post - anyway forging on}
>     ...with the hidden files enabled in the Documents and Settings there are five sections; 
>Administrator; All Users; Default User; H Sigerson; Harry Sigerson. 
>     It's a stand alone machine and I am the Administrator. Only the last two went all the way 
>to Proxy and the two database files. 
>  
>
This is because you were login as those users when you installed Spambayes.


>     
>     When I went down the, C:\Documents and Settings\Harry Sigerson\Application Data\Spam 
>Bayes\Proxy\  path there where the twosome   hammie.db   and   spambayes.messageinfo.db. 
>     When I tried to delete them they wouldn't go; they were being used elsewhere. 
>  
>
Check if Spambayes is still running. You should stop it and then delete 
the two files.

>     
>     When I went down the other, C:\Documents and Settings\H Sigerson\Application Data\Spam 
>Bayes\Proxy\   there they were again but this time deleting them was straight forward and they 
>went off to the Recycle Bin and got dumped from there. So far so good. 
>  
>
This is because Spambayes don't use the two files. This can be an old 
version or the files for another user. Check under which name you login.

>     
>     I went back to the first, the  \Harry Sigerson\  one and again tried to dump them from 
>there  once again they refused to go.  I've tried rebooting the machine to see if that helped 
>but no; they were still gone for the one and still in this one. 
>     
>     I've tried and will now do so again, to send and receive some email but I again got that 
>set of code error lines which tells that these two database files are still in place. Anyway 
>back in mo...
>     ...no go. I got the error lines again so the database files are still in place. 
>     
>     I cant' recall why there is both  \H Sigerson\   and a    \Harry Sigerson\   user. I know 
>that it dates back to when I was being advised that it was a bad idea to run a stand alone 
>machine as Administrator. So some degree of advice and 'learning' went into things. 
>     At the moment, in Administrative Tools from the Control Panel, then into Computer 
>Management and select the 'Groups's folder of 'Local Users and Groups' and then clicking 
>Administrators, there are two entries  Administrator  and   Harry Sigerson. 
>     In the same location but clicking on Power Users, the only entry there is  H Sigerson. 
>     Leaving that 'Groups' folder and opening its twin, the 'Users' folder, the 'Harry Sigerson 
>Properties' box's 'Members of' tab calls him up as one of the Administrators and one of the 
>Users. 
>     H Sigerson, in his 'Properties' box is called up as a Member of Power Users and Users. 
>     
>     I don't know if that gives any clues as to why I can't delete those two database files from 
>what is supposed to be in the control of an Administrator (Harry Sigerson) I just don't know. 
>     Deleting   hammie.db   and   spambayes.messageinfo.db    is proving difficult. 
>     I suppose uninstalling and then re-installing is one way to go but it would be so much more 
>satisfying to just find out why those database files cannot be dumped. 
>  
>
I'm more a programmer and a Unix administrator than a Windows 
administrator so I don't want to comment on that and give the chance to 
people to laugh at me. ;-)

Remi

From tameyer at ihug.co.nz  Tue Nov  2 22:05:08 2004
From: tameyer at ihug.co.nz (Tony Meyer)
Date: Tue Nov  2 22:05:20 2004
Subject: [Spambayes] messages
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E80140DE25@its-xchg4.massey.ac.nz>
Message-ID: <ECBA357DDED63B4995F5C1F5CBE5B1E86C37B8@its-xchg4.massey.ac.nz>

> Spambayes has been deleting messages from my sister

Please see FAQ 3.12:

<http://spambayes.org/faq.html#some-of-my-mail-is-going-missing>

> and leaving porn on my e mail.

I'm not sure what you mean by this.  Do you mean that some spam isn't being
classified correctly?  If so, please see FAQ 4.7:

<http://spambayes.org/faq.html#why-did-spambayes-mark-this-obvious-spam-unsu
re>

If you mean that you are getting mail from spambayes@python.org that is
spam, please see FAQ 5.8:

<http://spambayes.org/faq.html#why-does-the-spambayes-python-org-mailing-lis
t-get-spam>

If you mean that SpamBayes is somehow *causing* you to get spam, you're
wrong - this is not the case.  Please explain why you think this, and we'll
happily explain why you are mistaken.

=Tony.Meyer

-- 
Please always include the list (spambayes@python.org) in your replies
(reply-all), and please don't send me personal mail about SpamBayes.
http://www.massey.ac.nz/~tameyer/writing/reply_all.html explains this.

From tameyer at ihug.co.nz  Tue Nov  2 22:02:01 2004
From: tameyer at ihug.co.nz (Tony Meyer)
Date: Tue Nov  2 22:13:09 2004
Subject: [Spambayes] (no subject)
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E80140DE62@its-xchg4.massey.ac.nz>
Message-ID: <ECBA357DDED63B4995F5C1F5CBE5B1E8FBC9A0@its-xchg4.massey.ac.nz>

> Junk folder has disappeared from Folder List. Using Outlook 
> 2000. Junk Suspect folder still there. Please advise.

Please see FAQ 3.13:

<http://spambayes.org/faq.html#help-i-deleted-the-unsure-spam-folder>

=Tony.Meyer

-- 
Please always include the list (spambayes@python.org) in your replies
(reply-all), and please don't send me personal mail about SpamBayes.
http://www.massey.ac.nz/~tameyer/writing/reply_all.html explains this.

From postmaster at cp.finances.gouv.fr  Tue Nov  2 23:17:22 2004
From: postmaster at cp.finances.gouv.fr (postmaster@cp.finances.gouv.fr)
Date: Tue Nov  2 23:45:59 2004
Subject: [Spambayes] Notification e-manager Av1-bercy (Attachment Removal)
Message-ID: <20041102221728.D4DE013AC4C@av1-acct.cp.finances.gouv.fr>

**************** eManager Notification *****************

The following mail was blocked since it contains sensitive content.

Source mailbox: <spambayes@python.org>
Destination mailbox(es): josiane.faure@cp.finances.gouv.fr
Policy: Attachment Removal
Attachment file name: message.scr - audio/x-wav
Action: Replaced with text

La pi?ce jointe a ?t? suprim?e. Elle est interdite en raison de sa non-conformit? avec les r?gles d'utilisation de la messagerie en usage ? la DGCP

******************* End of message *********************
-------------- next part --------------
Received: from relaismsg 
	with id iA2MHOe17461
	for <josiane.faure@cp.finances.gouv.fr>; Tue, 2 Nov 2004 23:17:24 +0100 (MET)
Received: from relaismsg 
	by relaismsg.minefi.gouv.fr with id iA2MGgb28451
	for <josiane.faure@cp.finances.gouv.fr>; Tue, 2 Nov 2004 23:16:46 +0100 (MET)
Message-Id: <200411022216.iA2MGgb28451@relaismsg.minefi.gouv.fr>
From: spambayes@python.org
To: josiane.faure@cp.finances.gouv.fr
Subject: Mail Delivery (failure josiane.faure@cp.finances.gouv.fr)
Date: Tue, 2 Nov 2004 23:16:47 +0100
MIME-Version: 1.0
Content-Type: multipart/related;
	type="multipart/alternative";
	boundary="----=_NextPart_000_001B_01C0CA80.6B015D10"
X-Priority: 3
X-MSMail-Priority: Normal
From usr7-28hl at xemaps.com  Wed Nov  3 07:29:09 2004
From: usr7-28hl at xemaps.com (Keith Russell)
Date: Wed Nov  3 07:29:14 2004
Subject: [Spambayes] Unable to Access SpamBayes Browser Interface Using
	sb_imapfilter.py
Message-ID: <41887AB5.2030008@xemaps.com>

I'm reposting this problem because it was buried in another 
thread and apparently got lost:

When I start the IMAP script I'm not getting a browser window. 
The verbose output gives me three messages, which look okay, with 
no errors, but no browser. And opening a browser window and 
entering the URL manually gives me an error. I also can't telnet 
to localhost:8880. Here are the specifics:

1. >sb_imapfilter.py -c -t -l 5
SpamBayes IMAP Filter Version 0.4 (May 2004)
and engine SpamBayes Engine Version 0.3 (January 2004).

2. >sb_imapfilter.py -b -v
SpamBayes IMAP Filter Version 0.4 (May 2004)
and engine SpamBayes Engine Version 0.3 (January 2004).

Loading database C:\...\SpamBayes\Proxy\hammie.db...
Loading state from c:\...\SpamBayes\Proxy\hammie.db database
c:\...\SpamBayes\Proxy\hammie.db is an existing database, with 0
spam and 0 ham
Done.

4. http://localhost:8880
The page cannot be displayed
The page you are looking for is currently unavailable. The Web 
site might be experiencing technical difficulties, or you may 
need to adjust your browser settings.

3. >telnet localhost:8880
Connecting to localhost:8880...Could not open connection to the
host, on post 23: Connect failed

Can anyone help me out with this?

Thanks in advance....

-- 
Keith
From gillipops at blueyonder.co.uk  Wed Nov  3 15:01:12 2004
From: gillipops at blueyonder.co.uk (gillian coxson)
Date: Wed Nov  3 15:01:16 2004
Subject: [Spambayes] Incredamail
Message-ID: <000601c4c1ad$942a0020$727a2952@yoursq5tujzjzg>


-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.python.org/pipermail/spambayes/attachments/20041103/b82b9954/attachment.htm
From marki at athena.dev.unc.edu  Wed Nov  3 17:41:44 2004
From: marki at athena.dev.unc.edu (Ingram, Mark)
Date: Wed Nov  3 17:58:22 2004
Subject: [Spambayes] outlook plugin /Norton Antivirus problem
Message-ID: <2E6AF87F7F0FD711BDBC00B0D0F95A4D04AEAFF6@athena.dev.unc.edu>

I'm wondering if anyone has seen this problem and has a fix?
If I receive an email with a virus attached and I tell SpamBayes that I want
to delete as spam or if it automatically files it in my Junk Emails folder,
Norton Antivirus comes up saying that it has detected a virus. I have
reading (preview) panes turned off for all folders. The only way this can
happen is if SpamBayes opens the letter in the background to search for spam
patterns. Is this the case?   How do I keep it from opening the virus
infected email?
tx in advance!

Mark
marki@athena.dev.unc.edu



From srhurley at gapllc.com  Wed Nov  3 20:50:48 2004
From: srhurley at gapllc.com (Sherry Hurley)
Date: Wed Nov  3 21:07:46 2004
Subject: [Spambayes] Toolbars
Message-ID: <40F84740A5209C4BB753263F98A18CEFA73956@gapllc-server.gapllc.local>

I am not seeing, anywhere, the Delete Spam or Recover Spam toolbars.
There are not present in any drop down and there is plenty of space on
the toolbar line.  I have uninstalled and installed the current version,
no help.
 
If you could point me in the right direction I would appreciate it.
 
Outlook 2002
Win 2k SP 1
Spambayes 1.0
 
Thanks,
 
Beryl Tesler
btesler@tcg-systems.com
 
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.python.org/pipermail/spambayes/attachments/20041103/9a63c07c/attachment.html
From CBaumgar at sentinel.com  Wed Nov  3 22:20:57 2004
From: CBaumgar at sentinel.com (Baumgartner, Chris)
Date: Wed Nov  3 22:20:59 2004
Subject: [Spambayes] Junk Suspect File Disappeared
Message-ID: <9A2AF53E63133F458926DD2173FDDD34B3E88D@rodan.sentinel.com>

Good day.  I have been using SpamBayes for about six months, and it is
working fine except that my Junk Suspects folder has disappeared.
Daily, some important emails are placed into the Junk Suspects folder
and I then move them to the inbox.  Please help me recover my Junk
Suspects folder.  Thanks!

 

Christopher Baumgartner

Sentinel Technologies, Inc.

2550 Warrenville Road

Downers Grove, IL  60515

630/769-4337

630/769-1399 fax

cbaumgar@sentinel.com

 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.python.org/pipermail/spambayes/attachments/20041103/62decc0f/attachment.htm
From tameyer at ihug.co.nz  Wed Nov  3 23:05:42 2004
From: tameyer at ihug.co.nz (Tony Meyer)
Date: Wed Nov  3 23:07:20 2004
Subject: [Spambayes] Junk Suspect File Disappeared
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E80140E0DF@its-xchg4.massey.ac.nz>
Message-ID: <ECBA357DDED63B4995F5C1F5CBE5B1E8FBC9AA@its-xchg4.massey.ac.nz>

> my Junk Suspects folder has disappeared.

Please see FAQ 3.13:

<http://spambayes.org/faq.html#help-i-deleted-the-unsure-spam-folder>

=Tony.Meyer

-- 
Please always include the list (spambayes@python.org) in your replies
(reply-all), and please don't send me personal mail about SpamBayes.
http://www.massey.ac.nz/~tameyer/writing/reply_all.html explains this.

From Greg at attractor.com  Wed Nov  3 23:37:52 2004
From: Greg at attractor.com (Greg@attractor.com)
Date: Wed Nov  3 23:38:30 2004
Subject: [Spambayes] Question about Spambay
Message-ID: <61B9B9E6B74ED411B1F800508BD905D8865492@ATTRACTOR-1>

Hello,

We are running Spambay version 1.0 and someome here cannot seem to keep
certain emails on his 'good' list.  He recovers them from the junk folder
and then he gets another one in there.  The are from his brother and are
good emails.  How does he view the list that he has made? To make sure that
the right people are on it?  Feel free to call me at 650 685-8540   he is
really frustrated by this.

Thanks

greg

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.python.org/pipermail/spambayes/attachments/20041103/8de56e1d/attachment.html
From tameyer at ihug.co.nz  Thu Nov  4 00:05:07 2004
From: tameyer at ihug.co.nz (Tony Meyer)
Date: Thu Nov  4 00:06:22 2004
Subject: [Spambayes] Question about Spambay
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E80140E1A1@its-xchg4.massey.ac.nz>
Message-ID: <ECBA357DDED63B4995F5C1F5CBE5B1E86C37BD@its-xchg4.massey.ac.nz>

> someome here cannot seem to keep certain emails on his
> 'good' list.  He recovers them from the junk folder and
> then he gets another one in there.

Please see FAQs 4.7, 4.8 and 4.9:

<http://spambayes.org/faq.html#why-did-spambayes-mark-this-obvious-spam-unsu
re>
<http://spambayes.org/faq.html#ok-i-trained-on-that-message-but-it-still-thi
nks-it-s-unsure>
<http://spambayes.org/faq.html#spambayes-doesn-t-seem-to-catch-much-spam-wha
t-gives>

> How does he view the list that he has made? 
> To make sure that the right people are on it?

That isn't how SpamBayes works.  What you are referring to is called a
"whitelist" - FAQ 6.6 discusses why SpamBayes does not include whitelisting:

<http://spambayes.org/faq.html#why-don-t-you-add-whitelisting-blacklisting-t
o-spambayes>

What would be of use to you is viewing the 'clues' for the misclassified
messages (*before* any training on them is done).  You can do this via the
"Show spam clues for this message" function from the SpamBayes button on the
toolbar.  This will show you the 'clues' that SpamBayes used to make the
classification decision that it made.  If you don't understand them, feel
free to send an example message on to the list and someone will try and help
make it more clear.

=Tony.Meyer

-- 
Please always include the list (spambayes@python.org) in your replies
(reply-all), and please don't send me personal mail about SpamBayes.
http://www.massey.ac.nz/~tameyer/writing/reply_all.html explains this.

From tameyer at ihug.co.nz  Thu Nov  4 00:27:22 2004
From: tameyer at ihug.co.nz (Tony Meyer)
Date: Thu Nov  4 00:29:25 2004
Subject: [Spambayes] Toolbars
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E80140E0DE@its-xchg4.massey.ac.nz>
Message-ID: <ECBA357DDED63B4995F5C1F5CBE5B1E86C37BE@its-xchg4.massey.ac.nz>

> I am not seeing, anywhere, the Delete Spam or Recover Spam
> toolbars.  There are not present in any drop down and there
> is plenty of space on the toolbar line.  I have uninstalled
> and installed the current version, no help.

Right-click on empty space in the toolbars and choose "Customize".  Does the
SpamBayes toolbar appear there?  (If so, select it).

Check to see if Outlook has disabled SpamBayes.  Help->About Microsoft
Outlook->Disabled Items.

Check to see if SpamBayes is listed in the add-ins list.
Tools->Options->Other->Advanced Options->COM Add-ins.

If none of this helps, please send us a copy of any SpamBayes log files.
The troubleshooting guide explains where to find these.

<http://cvs.sf.net/viewcvs.py/*checkout*/spambayes/spambayes/Outlook2000/doc
s/troubleshooting.html?rev=HEAD>

=Tony.Meyer

-- 
Please always include the list (spambayes@python.org) in your replies
(reply-all), and please don't send me personal mail about SpamBayes.
http://www.massey.ac.nz/~tameyer/writing/reply_all.html explains this.

From tameyer at ihug.co.nz  Thu Nov  4 04:12:33 2004
From: tameyer at ihug.co.nz (Tony Meyer)
Date: Thu Nov  4 04:13:14 2004
Subject: [Spambayes] Unable to Access SpamBayes Browser Interface
	Usingsb_imapfilter.py
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E80140E0A0@its-xchg4.massey.ac.nz>
Message-ID: <ECBA357DDED63B4995F5C1F5CBE5B1E8FBC9B0@its-xchg4.massey.ac.nz>

> I'm reposting this problem because it was buried in another 
> thread and apparently got lost:

There was a response from Scott Burns:

<http://mail.python.org/pipermail/spambayes/2004-November/015109.html>

He's correct that the problem here:

> 3. >telnet localhost:8880
> Connecting to localhost:8880...Could not open connection to 
> the host, on post 23: Connect failed

Is that you need to do "telnet localhost 8880" (no colon) in order to test
the connection.

When you know if you can telnet to it, you'll know whether the web interface
isn't being served, or if it's a problem connecting to it in your browser.
The latter is likely to be something to do with proxies or firewalls that
you are using.  The former could be a SpamBayes bug.

=Tony.Meyer

-- 
Please always include the list (spambayes@python.org) in your replies
(reply-all), and please don't send me personal mail about SpamBayes.
http://www.massey.ac.nz/~tameyer/writing/reply_all.html explains this.

From tameyer at ihug.co.nz  Thu Nov  4 04:34:32 2004
From: tameyer at ihug.co.nz (Tony Meyer)
Date: Thu Nov  4 04:36:12 2004
Subject: [Spambayes] outlook plugin /Norton Antivirus problem
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E80140E0D9@its-xchg4.massey.ac.nz>
Message-ID: <ECBA357DDED63B4995F5C1F5CBE5B1E8FBC9B2@its-xchg4.massey.ac.nz>

> I'm wondering if anyone has seen this problem and has a fix?
> If I receive an email with a virus attached and I tell 
> SpamBayes that I want to delete as spam or if it 
> automatically files it in my Junk Emails folder, Norton 
> Antivirus comes up saying that it has detected a virus. I 
> have reading (preview) panes turned off for all folders. The 
> only way this can happen is if SpamBayes opens the letter in 
> the background to search for spam
> patterns. Is this the case?   How do I keep it from opening the virus
> infected email?

Is this perhaps mail arriving on an Exchange/IMAP/Hotmail store and being
moved to a "Junk Emails" folder on a local (pst) store?  That would explain
this, as Norton is unaware of the files on the remote machine, but becomes
aware of it once it arrives on the local one.

FWIW, SpamBayes doesn't "open" the mail in any way.  There's no way that
SpamBayes would trigger an email virus.

=Tony.Meyer

-- 
Please always include the list (spambayes@python.org) in your replies
(reply-all), and please don't send me personal mail about SpamBayes.
http://www.massey.ac.nz/~tameyer/writing/reply_all.html explains this.

From tameyer at ihug.co.nz  Thu Nov  4 05:41:28 2004
From: tameyer at ihug.co.nz (Tony Meyer)
Date: Thu Nov  4 05:42:12 2004
Subject: Resend to all: [Spambayes] spam with 100% score not moved to junk
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E80140DE1F@its-xchg4.massey.ac.nz>
Message-ID: <ECBA357DDED63B4995F5C1F5CBE5B1E86C37C2@its-xchg4.massey.ac.nz>

> Yes I did use the "Show spam clues for current message" which 
> appears under the "SpamBayes" button on my Outlook toolbar. I 
> used it when the email was still in my inbox (highlighted).
> 
> I've been using SpamBayes for about a year without problems. 
> This has been happening for the last few months only. I've 
> uninstalled and reinstalled and re-run the training but it 
> made no difference. I've also unchecked and rechecked most of 
> the boxes in SpamBayes manager but still can't get any 
> filtering to happen. There was, however one exception, 
> yesterday one email was put into my "unsure" folder!?! That's 
> the first time ever that that has happened, as far as I can remember.

I wonder if this is a case of:

[ 972359 ] some 100% SPAM sent to unsure
<http://sf.net/tracker/index.php?func=detail&aid=972359&group_id=61702&atid=
498103>

There, Mark hypothesises that the problem is with the content-type magic.
One of the example messages you sent looked like that might be the case, but
not the other.  Hopefully they are related, since we're already looking into
that one, so will find a solution at some point.

> Following on from your email, I've added the "Junk E-mail 
> Type" field to my inbox. It is empty.

The field should be named "Spam" and be in the "User-defined" category.

=Tony.Meyer

-- 
Please always include the list (spambayes@python.org) in your replies
(reply-all), and please don't send me personal mail about SpamBayes.
http://www.massey.ac.nz/~tameyer/writing/reply_all.html explains this.

From ececconi at visteon.com  Thu Nov  4 09:11:21 2004
From: ececconi at visteon.com (Cecconi, Enrico (E.))
Date: Thu Nov  4 09:11:46 2004
Subject: [Spambayes] spam with 100% score not moved to junk
Message-ID: <1468DAC3F7E2704686B5AEE348137CD40315C4B1@vistex33.dunton.visteon.com>

OK, so now i've added "spam" field from the "user defined" cat - it's still
empty.

The "unsure" thing is a side issue. It doesn't really worry me, I just
included to mention it in case it was significant. The main thing is SB used
to work on 90% of my spam (after trining) now it never works. It behaves as
if it's installed but not enabled, however reinstalling made no difference. 

What other checks can I do to see what is inhibiting the filtering?

What is the significance of "NOTE: Skipping folder for this session -
temporarily unavailable" in the attached log files (from today)? There was
one spam (attatched) in the inbox today. It was not filtered or moved by SB
but "show spam clues" gave it a 100% score.

regards,

EC

-----Original Message-----
From: Tony Meyer [mailto:tameyer@ihug.co.nz]
Sent: 04 November 2004 04:41
To: 'Cecconi, Enrico (E.)'; spambayes@python.org
Subject: RE: Resend to all: [Spambayes] spam with 100% score not moved
to junk


> Yes I did use the "Show spam clues for current message" which 
> appears under the "SpamBayes" button on my Outlook toolbar. I 
> used it when the email was still in my inbox (highlighted).
> 
> I've been using SpamBayes for about a year without problems. 
> This has been happening for the last few months only. I've 
> uninstalled and reinstalled and re-run the training but it 
> made no difference. I've also unchecked and rechecked most of 
> the boxes in SpamBayes manager but still can't get any 
> filtering to happen. There was, however one exception, 
> yesterday one email was put into my "unsure" folder!?! That's 
> the first time ever that that has happened, as far as I can remember.

I wonder if this is a case of:

[ 972359 ] some 100% SPAM sent to unsure
<http://sf.net/tracker/index.php?func=detail&aid=972359&group_id=61702&atid=
498103>

There, Mark hypothesises that the problem is with the content-type magic.
One of the example messages you sent looked like that might be the case, but
not the other.  Hopefully they are related, since we're already looking into
that one, so will find a solution at some point.

> Following on from your email, I've added the "Junk E-mail 
> Type" field to my inbox. It is empty.

The field should be named "Spam" and be in the "User-defined" category.

=Tony.Meyer

-- 
Please always include the list (spambayes@python.org) in your replies
(reply-all), and please don't send me personal mail about SpamBayes.
http://www.massey.ac.nz/~tameyer/writing/reply_all.html explains this.

-------------- next part --------------
A non-text attachment was scrubbed...
Name: spambayes1.log
Type: application/octet-stream
Size: 816 bytes
Desc: not available
Url : http://mail.python.org/pipermail/spambayes/attachments/20041104/8a33e77d/spambayes1-0001.obj
-------------- next part --------------
A non-text attachment was scrubbed...
Name: spambayes2.log
Type: application/octet-stream
Size: 950 bytes
Desc: not available
Url : http://mail.python.org/pipermail/spambayes/attachments/20041104/8a33e77d/spambayes2-0001.obj
-------------- next part --------------
An embedded message was scrubbed...
From: Montagu Xabaomlc <formulatedsal@benno-shop.ch>
Subject: Do you want to experience great next da =?utf-8?q?y ??=
Date: Wed, 3 Nov 2004 23:32:12 -0500 
Size: 785
Url: http://mail.python.org/pipermail/spambayes/attachments/20041104/8a33e77d/attachment-0001.mht
From johnt at amileasing.com  Thu Nov  4 14:11:23 2004
From: johnt at amileasing.com (Tsombakos, John)
Date: Thu Nov  4 15:37:00 2004
Subject: [Spambayes] Outlook [still!] freezing every day
Message-ID: <98385AF6669BD511B69800508BB34FEE069B62E1@mail.amileasing.com>

Sorry to be a pain, but this problem is still happening. As I've said, the
only work around is to turn spambayes off when I leave at night, then turn
it back on and then manually have it filter my inbox. If I forget to turn it
off, I come back in to a hung up Outlook. It runs all day just fine.

I've included yet another log. It looks like that last thing it does is
start a "new message timer", which never fires and subsequently takes
Outlook down with it.



-----Original Message-----
From: Tsombakos, John 
Sent: Monday, October 18, 2004 8:09 AM
To: spambayes@python.org
Subject: [Spambayes] Outlook [still!] freezing every day





Help!

I've installed the 1.0 version of Spambayes (Outlook plugin - Outlook 2000,
exchange server 5), but it is still freezing up Outlook. It doesn't happen
during the day and the junk mail flows in, just overnight.


-------------- next part --------------
A non-text attachment was scrubbed...
Name: spambayes2.log
Type: application/octet-stream
Size: 78322 bytes
Desc: not available
Url : http://mail.python.org/pipermail/spambayes/attachments/20041104/1b50886a/spambayes2-0001.obj
From daniele at glotrans.com  Thu Nov  4 17:30:47 2004
From: daniele at glotrans.com (Daniele Rossdeutscher)
Date: Thu Nov  4 17:31:25 2004
Subject: [Spambayes] spambayes 1.0
Message-ID: <002701c4c28b$a445b150$6601a8c0@dnai.com>

can you send me the link to download the 1.0 release? I've installed 0.86
and it's completely messed up Outlook. I can't find 1.0 anywhere on the
website...
thanks,
Daniele

PS please don't email me back in html, I can't read those messages anymore.

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.python.org/pipermail/spambayes/attachments/20041104/b412a0a1/attachment.html
From harrysigerson at ntlworld.com  Thu Nov  4 19:22:25 2004
From: harrysigerson at ntlworld.com (Harry Sigerson)
Date: Thu Nov  4 19:22:20 2004
Subject: [Spambayes] spambayes 1.0
In-Reply-To: <002701c4c28b$a445b150$6601a8c0@dnai.com>
References: <002701c4c28b$a445b150$6601a8c0@dnai.com>
Message-ID: <VA.000009b8.023fb2cb@ntlworld.com>

In article <002701c4c28b$a445b150$6601a8c0@dnai.com>, Daniele Rossdeutscher 
wrote:
> can you send me the link to download the 1.0 release? I've installed 0.86
> and it's completely messed up Outlook. I can't find 1.0 anywhere on the
> website... 
> thanks,
>
    Try...
http://sourceforge.net/projects/spambayes/index.html
    ...that should lead you there. 
    
    Harry. 




From Jared.Likes at BellSouth.com  Thu Nov  4 19:36:42 2004
From: Jared.Likes at BellSouth.com (Likes, Jared)
Date: Thu Nov  4 19:52:37 2004
Subject: [Spambayes] Windows Installer - please zip
Message-ID: <CDC61CFAAF6D494CA910D682CC6E94401522A216@bremoc2p-55>

Skipped content of type multipart/alternative-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: image/jpeg
Size: 2743 bytes
Desc: image001.jpg
Url : http://mail.python.org/pipermail/spambayes/attachments/20041104/8eb2d279/attachment.jpe
From rmalayter at bai.org  Thu Nov  4 20:06:59 2004
From: rmalayter at bai.org (Ryan Malayter)
Date: Thu Nov  4 20:07:02 2004
Subject: [Spambayes] Windows Installer - please zip
Message-ID: <792DE28E91F6EA42B4663AE761C41C2A032163C2@cliff.bai.org>

Obviously, your network administrators have implemented EXE blocking as
a security precaution. They do not want you installing unapproved
software on your network.

Rather than trying to subvert their policy - which is very sensible - I
would contact them and ask them to download and install SpamBayes for
you.

Regards,
	Ryan


________________________________

	From: spambayes-bounces@python.org
[mailto:spambayes-bounces@python.org] On Behalf Of Likes, Jared
	Sent: Thursday, November 04, 2004 12:37 PM
	To: spambayes@python.org
	Subject: [Spambayes] Windows Installer - please zip
	
	
	 

	I cannot download the Window Outlook addin installer because it
is an exe file and our firewall blocks exe files. (so does our email
gateway).

	Thanks,

	Jared
From Jared.Likes at bellsouth.com  Thu Nov  4 20:24:14 2004
From: Jared.Likes at bellsouth.com (Likes, Jared)
Date: Thu Nov  4 20:25:12 2004
Subject: [Spambayes] Windows Installer - please zip
Message-ID: <CDC61CFAAF6D494CA910D682CC6E94401522A26C@bremoc2p-55>

That would make sense and they probably inspect zip files as well... And
I have asked our proxy filter admins to do that in the past for things
on sourceforge.net. 

But still, I don't understand why you would put out a naked exe file on
the web.

-----Original Message-----
From: Ryan Malayter [mailto:rmalayter@bai.org] 
Sent: Thursday, November 04, 2004 2:07 PM
To: Likes, Jared; spambayes@python.org
Subject: RE: [Spambayes] Windows Installer - please zip

Obviously, your network administrators have implemented EXE blocking as
a security precaution. They do not want you installing unapproved
software on your network.

Rather than trying to subvert their policy - which is very sensible - I
would contact them and ask them to download and install SpamBayes for
you.

Regards,
	Ryan


________________________________

	From: spambayes-bounces@python.org
[mailto:spambayes-bounces@python.org] On Behalf Of Likes, Jared
	Sent: Thursday, November 04, 2004 12:37 PM
	To: spambayes@python.org
	Subject: [Spambayes] Windows Installer - please zip
	
	
	 

	I cannot download the Window Outlook addin installer because it
is an exe file and our firewall blocks exe files. (so does our email
gateway).

	Thanks,

	Jared

The information transmitted is intended only for the person or entity to which it is addressed and may contain confidential, proprietary, and/or privileged material. Any review, retransmission, dissemination or other use of, or taking of any action in reliance upon this information by persons or entities other than the intended recipient is prohibited. If you received this in error, please contact the sender and delete the material from all computers. 113


From jsp at PKC.com  Thu Nov  4 20:32:46 2004
From: jsp at PKC.com (Jesse Pelton)
Date: Thu Nov  4 20:32:51 2004
Subject: [Spambayes] Windows Installer - please zip
Message-ID: <9147d325abbb8a61410c233895dd50bb418a83d1@vtmail.pkc.com>

For convenience.  It's no more or less secure than a zip file, and it
avoids the need to unzip.  (Not all Windows installations have unzipping
in the shell.)  However, it seems like it ought to be accompanied by an
MD5 sum (at a minimum). 

> -----Original Message-----
> From: spambayes-bounces@python.org 
> [mailto:spambayes-bounces@python.org] On Behalf Of Likes, Jared
> Sent: Thursday, November 04, 2004 2:24 PM
> To: Ryan Malayter
> Cc: spambayes@python.org
> Subject: RE: [Spambayes] Windows Installer - please zip
> 
> That would make sense and they probably inspect zip files as 
> well... And
> I have asked our proxy filter admins to do that in the past for things
> on sourceforge.net. 
> 
> But still, I don't understand why you would put out a naked 
> exe file on
> the web.
> 
> -----Original Message-----
> From: Ryan Malayter [mailto:rmalayter@bai.org] 
> Sent: Thursday, November 04, 2004 2:07 PM
> To: Likes, Jared; spambayes@python.org
> Subject: RE: [Spambayes] Windows Installer - please zip
> 
> Obviously, your network administrators have implemented EXE 
> blocking as
> a security precaution. They do not want you installing unapproved
> software on your network.
> 
> Rather than trying to subvert their policy - which is very 
> sensible - I
> would contact them and ask them to download and install SpamBayes for
> you.
> 
> Regards,
> 	Ryan
> 
> 
> ________________________________
> 
> 	From: spambayes-bounces@python.org
> [mailto:spambayes-bounces@python.org] On Behalf Of Likes, Jared
> 	Sent: Thursday, November 04, 2004 12:37 PM
> 	To: spambayes@python.org
> 	Subject: [Spambayes] Windows Installer - please zip
> 	
> 	
> 	 
> 
> 	I cannot download the Window Outlook addin installer because it
> is an exe file and our firewall blocks exe files. (so does our email
> gateway).
> 
> 	Thanks,
> 
> 	Jared
> 
> The information transmitted is intended only for the person 
> or entity to which it is addressed and may contain 
> confidential, proprietary, and/or privileged material. Any 
> review, retransmission, dissemination or other use of, or 
> taking of any action in reliance upon this information by 
> persons or entities other than the intended recipient is 
> prohibited. If you received this in error, please contact the 
> sender and delete the material from all computers. 113
> 
> 
> _______________________________________________
> Spambayes@python.org
> http://mail.python.org/mailman/listinfo/spambayes
> Check the FAQ before asking: http://spambayes.sf.net/faq.html
> 
From martinzs at earthlink.net  Thu Nov  4 20:24:39 2004
From: martinzs at earthlink.net (Martin Sipkoff)
Date: Thu Nov  4 20:34:50 2004
Subject: [Spambayes] question
Message-ID: <005801c4c2a3$ee3a1c20$6402a8c0@Martin>

If I classify an email as spam, but which to change this classification later, how do I do so? What happens to emails that are classified as spam? Where do they go?

Thanks.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.python.org/pipermail/spambayes/attachments/20041104/2a6d9795/attachment.htm
From tameyer at ihug.co.nz  Thu Nov  4 23:06:23 2004
From: tameyer at ihug.co.nz (Tony Meyer)
Date: Thu Nov  4 23:07:23 2004
Subject: [Spambayes] Windows Installer - please zip
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E80140E2EA@its-xchg4.massey.ac.nz>
Message-ID: <ECBA357DDED63B4995F5C1F5CBE5B1E8FBC9B9@its-xchg4.massey.ac.nz>

> For convenience.  It's no more or less secure than a zip 
> file, and it avoids the need to unzip.  (Not all Windows 
> installations have unzipping in the shell.)

It's not at all uncommon, either.  Hardly any Windows installers I download
are zipped.  I would much rather just get the .exe.

> However, it 
> seems like it ought to be accompanied by an MD5 sum (at a minimum). 

Feel free to open a feature request for this, and we can add it to the
release process.

=Tony.Meyer

-- 
Please always include the list (spambayes@python.org) in your replies
(reply-all), and please don't send me personal mail about SpamBayes.
http://www.massey.ac.nz/~tameyer/writing/reply_all.html explains this.

From tameyer at ihug.co.nz  Thu Nov  4 23:11:31 2004
From: tameyer at ihug.co.nz (Tony Meyer)
Date: Thu Nov  4 23:12:08 2004
Subject: [Spambayes] question
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E801520AF4@its-xchg4.massey.ac.nz>
Message-ID: <ECBA357DDED63B4995F5C1F5CBE5B1E86C37C3@its-xchg4.massey.ac.nz>

I assume (from the headers of your message) that you're using Outlook
Express and sb_server.  If that's not the case, then this is all wrong -
please re-ask the questions, but giving us more information about what
you're using.

> If I classify an email as spam, but which to change
> this classification later, how do I do so?

While the SpamBayes copy of the message is still in the cache (by default, 7
days), you can use the "find message" query on the main page of the web
interface to bring up a review page with that message on it.  If you train
it as ham from there it will automatically be untrained as spam first.

Once the message is gone from the cache, it's very difficult to do.  You can
paste the message (including headers) into the "train message" box on the
main page of the web interface, which will at least counteract the spam
training, but this won't do the 'untraining' that really should be done.

> What happens to emails that are classified as spam?
> Where do they go?

With sb_server, all mail is just tagged with the classification.  It's up to
you what happens to the mail.  Typically users get the mailer (e.g. OE) to
move it to another folder where it can be periodically reviewed and deleted.

=Tony.Meyer

-- 
Please always include the list (spambayes@python.org) in your replies
(reply-all), and please don't send me personal mail about SpamBayes.
http://www.massey.ac.nz/~tameyer/writing/reply_all.html explains this.

From tameyer at ihug.co.nz  Thu Nov  4 23:40:13 2004
From: tameyer at ihug.co.nz (Tony Meyer)
Date: Thu Nov  4 23:40:55 2004
Subject: [Spambayes] RE: [spambayes-dev] wish: keep spam-mail from
	downloading to OE6
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E80140E27B@its-xchg4.massey.ac.nz>
Message-ID: <ECBA357DDED63B4995F5C1F5CBE5B1E86C37C4@its-xchg4.massey.ac.nz>

> Is it possible to not send mail marked "spam" to OE but
> instead just keep it in the SB's cache. While
> training I still can view these messages if I want to. 

This can't really be done with just a POP3 proxy.  Basically, OE asks the
server how many messages there are to download and then asks to download
each of them.  To not retrieve spam, sb_server would either have to modify
the message count or fail to deliver the message when asked.

The former can't be done, because SpamBayes doesn't know the classification
of messages until it retrieves them (it needs the whole message to tell).
To correctly modify the list, it would have to classify *all* messages just
to get the list, which isn't practical.

The latter doesn't work, because then the mailer (e.g. OE) doesn't
understand why the message failed to download and will report this as an
error.

One way to achieve the same result would be to have a filter periodically
run that connects to your POP3 server, classifies all the messages, and
removes the spam.  An example of such a filter is on the wiki
<http://entrian.com/sbwiki> and in the SpamBayes source (CVS and,
eventually, 1.1).  This is more complex to run, however, and doesn't really
suit most situations.

> PS I somehow can't get the auto-train option to work

What option?  Do you mean [Hammie] train_on_filter ?  If so, that's not an
option that's available to sb_server - pretty much all options relevant to
sb_server are available via the web interface.

=Tony.Meyer

-- 
Please always include the list (spambayes@python.org) in your replies
(reply-all), and please don't send me personal mail about SpamBayes.
http://www.massey.ac.nz/~tameyer/writing/reply_all.html explains this.

From tameyer at ihug.co.nz  Fri Nov  5 00:00:53 2004
From: tameyer at ihug.co.nz (Tony Meyer)
Date: Fri Nov  5 00:01:30 2004
Subject: [Spambayes] spam with 100% score not moved to junk
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E80140E27C@its-xchg4.massey.ac.nz>
Message-ID: <ECBA357DDED63B4995F5C1F5CBE5B1E86C37C5@its-xchg4.massey.ac.nz>

> The main thing is SB used to work on 90% of my spam (after 
> trining) now it never works. It behaves as if it's installed 
> but not enabled, however reinstalling made no difference. 
> 
> What is the significance of "NOTE: Skipping folder for this 
> session - temporarily unavailable" in the attached log files 
> (from today)?

Ack.  I can't believe I missed that in the original log (sorry).

*That* is the cause of the problem here - basically SpamBayes can't get hold
of the folder it's trying to watch, so it never installs the 'event hook'
that would tell it about new mail arriving, so it never knows about any of
the mail.

The question then, of course, is why the folder isn't available.  Outlook is
telling SpamBayes that the folder is "temporarily available" - can you think
of any reason why this might be the case?

You could try going into the configuration (SpamBayes Manager->Filtering)
and (a) seeing if the folder is offered in the list of folders to filter and
(b) reselecting it if it is.

You could try selecting another folder on the same store and seeing if it
gives the same error.

Microsoft indicates that possible causes for this error include:

  * Insufficient permissions for the mailbox.

  * If you try to open more than 255 mailboxes in one process (the process
here would be Outlook).  It also says that the actual limit may be lower as
some index values may be reserved.

  * If an incorrect password is provided and a prompt for the password
cannot be displayed.

Anything sound relevant?

=Tony.Meyer

-- 
Please always include the list (spambayes@python.org) in your replies
(reply-all), and please don't send me personal mail about SpamBayes.
http://www.massey.ac.nz/~tameyer/writing/reply_all.html explains this.

From mumms at sprouting.com  Fri Nov  5 02:20:38 2004
From: mumms at sprouting.com (Mumm's)
Date: Fri Nov  5 02:20:38 2004
Subject: [Spambayes] spammers have found work around?
Message-ID: <JHEJIIIFANHBHCLJIDGJOEHFGIAA.mumms@sprouting.com>

I've been using spambayes quite a while now with remarkably good
results, almost 0 spam for many months.
Recently, though, a few messages have been sneaking through in the 0%
to 9% range.
Do you think that spammers are reacting to spambays and other good
intelligent filter spam blockers by crafting spam especially to get
through. Anyone else noticing this change?
Will spambayes always be able to learn these new techniques?

Regards ..... Jim Mumm

-- Certified Organic seeds for sprouting --
 MUMM'S SPROUTING SEED LTD., Box 80, Parkside, SK, Canada S0J 2A0
mumms@sprouting.com     ph 306 747 2935   fx 306 747 3618
www.sprouting.com



From tim.peters at gmail.com  Fri Nov  5 03:20:20 2004
From: tim.peters at gmail.com (Tim Peters)
Date: Fri Nov  5 03:20:32 2004
Subject: [Spambayes] spammers have found work around?
In-Reply-To: <JHEJIIIFANHBHCLJIDGJOEHFGIAA.mumms@sprouting.com>
References: <JHEJIIIFANHBHCLJIDGJOEHFGIAA.mumms@sprouting.com>
Message-ID: <1f7befae04110418201632ffbd@mail.gmail.com>

[Mumm's]
> I've been using spambayes quite a while now with remarkably good
> results, almost 0 spam for many months.
> Recently, though, a few messages have been sneaking through in the 0%
> to 9% range.
> Do you think that spammers are reacting to spambays and other good
> intelligent filter spam blockers by crafting spam especially to get
> through.

Spammers have been working as hard as they can to evade filters of all
kinds for years, but I doubt anyone is targeting SB specifically. 
We're too small a target, and this kind of classifier is difficult to
sidestep reliably because it's so personalized.  Spam is a business,
and spammers get much higher return on investment by learning to fool
"one filter to rule them all" systems used by large organizations
(corporations, universities, ISPs).

Oversimplified, say you're a well-heeled spammer (as some are).  You
*buy* one of these systems, and then send spam to yourself until you
find a way to fool it.  To a first approximation, then, that spam has
a decent chance of fooling many installations of the same system.  You
send out millions of spam then as fast as you can, before the filter
vendor has a chance to change the system to plug the hole you found.

Now you can do the same thing with your own trained SpamBayes, but it
won't do you much good:  the training you do leaves you with a
different database than the training I do, and unless you spend real
money and effort to investigate me, you'll have no idea how to make
your spam look hammy to *my* classifier.  But if you could afford to
do intelligent targeted marketing, you'd get a higher return by moving
to a more traditional form of targeted advertising, trying to sell me
high-ticket legitimate items instead of assorted bottom-feeding scams.
 You'd be out of the spam business then.

That's why SB is hard to beat on a large scale.  It's not trying to
identify spam, it's trying to separate ham from spam according to an
individual's tastes.

> Anyone else noticing this change?

I notice that spam changes all the time.  For example, "Rolex" spam
has become very heavy over the past two weeks in my mix.  A few of
those were Unsure for me at first.  I trained on 2, and haven't seen
another one rate unsure since then.

BTW, I throw away my database a few times each year and start over
from scratch.  I know that this is fun, and slashes the database size.
 I suspect it helps recognition accuracy too, but don't know that for
a fact.  If you feel like you're in a rut, try it!  One common cause
for deteriorating accuracy is training a message into the wrong
category (ham as spam, or vice versa), and that's very hard to detect
after the fact.  As the months wear on, that's simply *going* to
happen sooner or later.  Starting over is often the easiest way to
recover from that.

> Will spambayes always be able to learn these new techniques?

Answering that requires perfect foreknowledge, so, yes, of course
<wink>.  We haven't made significant changes to the classifier or the
tokenizer in many months, but I haven't noticed any decrease in
effectiveness, despite that both the content and form of spam keeps
changing.

A good development is the increasing number of email clients that
refuse to download images in HTML email automatically.  I wish that
were universal.  So long as the spammer has to put stuff *in the msg*
itself that's visible to you, they have to make their sales pitch and
their URL visible to classifiers too, and then it can be analyzed. 
When all they give is a URL that automatically downloads a .gif or
.jpg containing an image of the sales message, that's very hard to
analyze.  But if email clients stop downloading that stuff, the
response rate on spam using that trick will fall to 0, and spammers
will stop doing that.  It's easy to forget that their goal isn't to
irritate you, it's to extract money from you.  To do that, they have
to make a visible sales pitch.
From tameyer at ihug.co.nz  Fri Nov  5 05:03:19 2004
From: tameyer at ihug.co.nz (Tony Meyer)
Date: Fri Nov  5 05:04:13 2004
Subject: [Spambayes] Outlook [still!] freezing every day
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E80140E2E2@its-xchg4.massey.ac.nz>
Message-ID: <ECBA357DDED63B4995F5C1F5CBE5B1E86C37C6@its-xchg4.massey.ac.nz>

> Sorry to be a pain, but this problem is still happening.

Every day I look at these logs and try and think of something to try to
figure out what it could be, but can't.

> If I forget to turn it off, I come back 
> in to a hung up Outlook. It runs all day just fine.

Does anything else happen at night?  Like some sort of backup system or
something?

How are you turning SpamBayes off?  With the "enable" button on the Manager
dialog?  What happens if you leave that enabled, but set it to Filter no
folders - does it still happen?

> I've included yet another log. It looks like that last thing 
> it does is start a "new message timer", which never fires and 
> subsequently takes Outlook down with it.

This is the case with a couple of the other logs, too, although not all of
them.  What happens if you turn background filtering off (Advanced tab)?
Does it still happen?

=Tony.Meyer

-- 
Please always include the list (spambayes@python.org) in your replies
(reply-all), and please don't send me personal mail about SpamBayes.
http://www.massey.ac.nz/~tameyer/writing/reply_all.html explains this.

From tameyer at ihug.co.nz  Fri Nov  5 05:04:12 2004
From: tameyer at ihug.co.nz (Tony Meyer)
Date: Fri Nov  5 05:06:18 2004
Subject: [Spambayes] RE: [spambayes-dev] thanks for spambayes
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E80140E2A6@its-xchg4.massey.ac.nz>
Message-ID: <ECBA357DDED63B4995F5C1F5CBE5B1E8FBC9C4@its-xchg4.massey.ac.nz>

> I just wanted to tank you all for the wonderful spambayes.
> It is saving me time day after day. I love it.
> 
> Thanks, and keep up the good work.

On behalf of all the developers, thanks for the kind words - we really
appreciate them :)

=Tony.Meyer

From ccanderson at comcast.net  Fri Nov  5 06:46:15 2004
From: ccanderson at comcast.net (Charles Anderson)
Date: Fri Nov  5 06:46:40 2004
Subject: [Spambayes] Missing Recover button
Message-ID: <000101c4c2fa$c4913640$7000a8c0@Anderson2>

I am using Spambayes with Outlook 2002 (10.6515.6626) SP3.  I have lost the
"Recover from Spam" Button and I can not figure out how to get it back.  I
have deleted and reinstalled Spambayes and deleted the Spambayes toolbar.
When I click on down arrow and then click on the "add or remove buttons"
Spambayes is grayed out and not accessible.
 
Any suggestions?  I do like your product and it seems to catch most of the
spam that I receive.
 
Thank you for your assistance.
 
Chuck Anderson
ccanderson@comcast.net.
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.python.org/pipermail/spambayes/attachments/20041105/017e6a72/attachment.html
From ccanderson at comcast.net  Fri Nov  5 06:53:33 2004
From: ccanderson at comcast.net (Charles Anderson)
Date: Fri Nov  5 06:53:51 2004
Subject: [Spambayes] Missing Recover button
Message-ID: <000601c4c2fb$cc3bd6b0$7000a8c0@Anderson2>

Resending this message with additional information.
 
I am using Spambayes with Outlook 2002 (10.6515.6626) SP3 and Windows XP
Service Pack 2.  I have lost the "Recover from Spam" Button and I can not
figure out how to get it back.  I have deleted and reinstalled Spambayes and
deleted the Spambayes toolbar.  When I click on down arrow and then click on
the "add or remove buttons"  Spambayes is grayed out and not accessible.  I
could not find any .log files,  I did look through the FAQ section and I did
not see anything that seemed to be related to this problem.
 
Any suggestions?  I do like your product and it seems to catch most of the
spam that I receive.
 
Thank you for your assistance.
 
Chuck Anderson
ccanderson@comcast.net.
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.python.org/pipermail/spambayes/attachments/20041105/9fa5173f/attachment.htm
From ccanderson at comcast.net  Fri Nov  5 06:58:07 2004
From: ccanderson at comcast.net (Charles Anderson)
Date: Fri Nov  5 06:58:14 2004
Subject: [Spambayes] Missing Recover button-found
Message-ID: <000b01c4c2fc$6cd015a0$7000a8c0@Anderson2>

I am sorry,  I found the FAQ section for Outlook and found the answer after
I have sent 2 other e-mails.  Thanks for the assistance.  Now I know why I
didn't see the Recover button.
 
 
Resending this message with additional information.
 
I am using Spambayes with Outlook 2002 (10.6515.6626) SP3 and Windows XP
Service Pack 2.  I have lost the "Recover from Spam" Button and I can not
figure out how to get it back.  I have deleted and reinstalled Spambayes and
deleted the Spambayes toolbar.  When I click on down arrow and then click on
the "add or remove buttons"  Spambayes is grayed out and not accessible.  I
could not find any .log files,  I did look through the FAQ section and I did
not see anything that seemed to be related to this problem.
 
Any suggestions?  I do like your product and it seems to catch most of the
spam that I receive.
 
Thank you for your assistance.
 
Chuck Anderson
ccanderson@comcast.net.
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.python.org/pipermail/spambayes/attachments/20041105/3481b5f9/attachment.html
From falcon_050 at hotmail.com  Fri Nov  5 08:29:17 2004
From: falcon_050 at hotmail.com (falcon_050)
Date: Fri Nov  5 08:30:05 2004
Subject: [Spambayes] Re: [spambayes-dev] wish: autotraining
References: <ECBA357DDED63B4995F5C1F5CBE5B1E86C37C4@its-xchg4.massey.ac.nz>
Message-ID: <BAY13-DAV16T1ePyu910000da0c@hotmail.com>

Hi Tony,

Tx for your response. The setting I am referring to is in the advanced settings page. I thought it would mean that the autotrained spam-messages would not appear in the (manual) traininglist marked as spam. Do I read that it wont work because this is not a "procmail-based solution" ?

TIA,

Valk

Training Options 
      Train when filtering:    No  
       Yes         Train when filtering: Train when filtering? After filtering a message, hammie can then train itself on the judgement (ham or spam). This can speed things up with a procmail-based solution. If you do enable this, please make sure to retrain any mistakes. Otherwise, your word database will slowly become useless. 
      Current Value:  Yes 
        


  ----- Original Message ----- 
  From: Tony Meyer 
  To: 'falcon_050' 
  Cc: 
  Sent: Thursday, November 04, 2004 11:40 PM
  Subject: RE: [spambayes-dev] wish: keep spam-mail from downloading to OE6


  > Is it possible to not send mail marked "spam" to OE but
  > instead just keep it in the SB's cache. While
  > training I still can view these messages if I want to. 

  This can't really be done with just a POP3 proxy.  Basically, OE asks the
  server how many messages there are to download and then asks to download
  each of them.  To not retrieve spam, sb_server would either have to modify
  the message count or fail to deliver the message when asked.

  The former can't be done, because SpamBayes doesn't know the classification
  of messages until it retrieves them (it needs the whole message to tell).
  To correctly modify the list, it would have to classify *all* messages just
  to get the list, which isn't practical.

  The latter doesn't work, because then the mailer (e.g. OE) doesn't
  understand why the message failed to download and will report this as an
  error.

  One way to achieve the same result would be to have a filter periodically
  run that connects to your POP3 server, classifies all the messages, and
  removes the spam.  An example of such a filter is on the wiki
  <http://entrian.com/sbwiki> and in the SpamBayes source (CVS and,
  eventually, 1.1).  This is more complex to run, however, and doesn't really
  suit most situations.

  > PS I somehow can't get the auto-train option to work

  What option?  Do you mean [Hammie] train_on_filter ?  If so, that's not an
  option that's available to sb_server - pretty much all options relevant to
  sb_server are available via the web interface.

  =Tony.Meyer

  -- 
  Please always include the list (spambayes@python.org) in your replies
  (reply-all), and please don't send me personal mail about SpamBayes.
  http://www.massey.ac.nz/~tameyer/writing/reply_all.html explains this.

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.python.org/pipermail/spambayes/attachments/20041105/fbd89bd1/attachment.htm
From ececconi at visteon.com  Fri Nov  5 09:30:23 2004
From: ececconi at visteon.com (Cecconi, Enrico (E.))
Date: Fri Nov  5 09:30:45 2004
Subject: [Spambayes] spam with 100% score not moved to junk
Message-ID: <1468DAC3F7E2704686B5AEE348137CD40315C4BB@vistex33.dunton.visteon.com>

I checked that "Inbox" was in the list. It was. I unchecked it and rechecked
it. I then started "filter messages" and SB started and filtered as
expected, some to junk, some to unsure. SB obcviously still works butn not
at start up or without manual intervention.

I then wondered if there was a timing problem with my network logon (I have
Outlook in "startup")and SpamBayes (although no secondary request for
password has ever appeared). So I closed Outlook, deleted all the SB logs
and then restarted Outlook. This resulted in the attatched log which has
"Processing missed spam in folder 'Inbox' by starting a timer" entry in the
last line. What is that all about? What timer?

rgds,

EC

-----Original Message-----
From: Tony Meyer [mailto:tameyer@ihug.co.nz]
Sent: 04 November 2004 23:01
To: 'Cecconi, Enrico (E.)'; spambayes@python.org
Subject: RE: [Spambayes] spam with 100% score not moved to junk


> The main thing is SB used to work on 90% of my spam (after 
> trining) now it never works. It behaves as if it's installed 
> but not enabled, however reinstalling made no difference. 
> 
> What is the significance of "NOTE: Skipping folder for this 
> session - temporarily unavailable" in the attached log files 
> (from today)?

Ack.  I can't believe I missed that in the original log (sorry).

*That* is the cause of the problem here - basically SpamBayes can't get hold
of the folder it's trying to watch, so it never installs the 'event hook'
that would tell it about new mail arriving, so it never knows about any of
the mail.

The question then, of course, is why the folder isn't available.  Outlook is
telling SpamBayes that the folder is "temporarily available" - can you think
of any reason why this might be the case?

You could try going into the configuration (SpamBayes Manager->Filtering)
and (a) seeing if the folder is offered in the list of folders to filter and
(b) reselecting it if it is.

You could try selecting another folder on the same store and seeing if it
gives the same error.

Microsoft indicates that possible causes for this error include:

  * Insufficient permissions for the mailbox.

  * If you try to open more than 255 mailboxes in one process (the process
here would be Outlook).  It also says that the actual limit may be lower as
some index values may be reserved.

  * If an incorrect password is provided and a prompt for the password
cannot be displayed.

Anything sound relevant?

=Tony.Meyer

-- 
Please always include the list (spambayes@python.org) in your replies
(reply-all), and please don't send me personal mail about SpamBayes.
http://www.massey.ac.nz/~tameyer/writing/reply_all.html explains this.

-------------- next part --------------
A non-text attachment was scrubbed...
Name: spambayes1.log
Type: application/octet-stream
Size: 809 bytes
Desc: not available
Url : http://mail.python.org/pipermail/spambayes/attachments/20041105/d853239a/spambayes1.obj
From noreply at spambayes  Fri Nov  5 13:25:04 2004
From: noreply at spambayes (noreply@spambayes)
Date: Fri Nov  5 13:21:59 2004
Subject: [Spambayes] Mail Account
Message-ID: <200411051221.iA5CLoM15828@uusnwa0n.utc.com>

------------------  Virus Warning Message (on uusnwa0n)

Found virus WORM_NETSKY.P in file data.rtf                                                                           .scr (in details_spambayes.zip)
The file is deleted.

---------------------------------------------------------
-------------- next part --------------
Your mail account is expired.
See the details to reactivate it.

-------------- next part --------------

------------------  Virus Warning Message (on uusnwa0n)

details_spambayes.zip is removed from here because it contains a virus.

---------------------------------------------------------
From CTIvey at southernco.com  Fri Nov  5 13:55:59 2004
From: CTIvey at southernco.com (Ivey, Charles T.)
Date: Fri Nov  5 14:13:05 2004
Subject: [Spambayes] Icon
Message-ID: <420C032031E05E4D957C12DF078B61CBBAAFB0@alxapex21.southernco.com>

I am missing my Recover form Spam icon. Can you tell me how to recover
it?

Thanks,
Charles
From bnorman at bak.rr.com  Fri Nov  5 17:16:43 2004
From: bnorman at bak.rr.com (Bill Norman)
Date: Fri Nov  5 17:16:54 2004
Subject: [Spambayes] mailing list
Message-ID: <200411051616.iA5GGlkM011182@ms-smtp-01-eri0.socal.rr.com>

Skipped content of type multipart/alternative-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: image/jpeg
Size: 3981 bytes
Desc: not available
Url : http://mail.python.org/pipermail/spambayes/attachments/20041105/d6cbdbb5/attachment.jpe
From robert at imagine.co.uk  Fri Nov  5 17:58:21 2004
From: robert at imagine.co.uk (Robert Neuschul)
Date: Fri Nov  5 17:58:24 2004
Subject: [Spambayes] A most peculair problem with SB 1
Message-ID: <VA.0000574a.01ad9bdb@imagine.co.uk>

Hi

I've just encountered a strange problem: SB up to and including RC1 ran 
perfectly well on our old Windows server. Having had to build a new 
2003 server I wanted to setup Spambayes, and decided to take advantage 
of the new full release.

I installed version 2.3.3 of Python.org's Python, added the relevant 
PyWin for 2.3.3 from Sourceforge, and downloaded the full SB 1 
Sourcecode version.

SB was then expanded to a temp directory alongside Python's install 
directory, and I ran "python setup.py install" - which duly compiles 
and copies everything it says it wants into the relevant Python 
subdirectories. 
Having got that far I then wanted to install the POP3 proxy service: 
opening a DOS session at the Python/Scripts directory and running 
"python pop3proxy_service.py install" returns several errors [see 
below] which I can't interpret.

C:\Python23\Scripts>python pop3proxy_service.py install
Traceback (most recent call last):
  File "C:\Python23\Scripts\pop3proxy_service.py", line 82, in ?
    import sb_server
  File "C:\Python23\Scripts\sb_server.py", line 104, in ?
    import spambayes.message
  File "c:\Python23\Lib\site-packages\spambayes\message.py", line 215, 
in ?
    msginfoDB = MessageInfoPickle(message_info_db_name)
  File "c:\Python23\Lib\site-packages\spambayes\message.py", line 147, 
in __init
__
    self.load()
  File "c:\Python23\Lib\site-packages\spambayes\message.py", line 159, 
in load
    self.db = pickle.load(fp)
  File "c:\Python23\lib\pickle.py", line 1390, in load
    return Unpickler(file).load()
  File "c:\Python23\lib\pickle.py", line 872, in load
    dispatch[key](self)
KeyError: '\x00'

Note that every file referenced in those error messages does actually 
exist in the Python tree - either in the Lib subtree or in the Scripts 
folder.

Thinking I'd made a serious error, or misconfigured something somewhere 
I decided to uninstall Python and SB entirely, reboot and try again 
with the pre-compiled version of SB1. This time any attempt to run the 
server returns similar but not identical errors.

C:\SpamBayes\bin>sb_server
Traceback (most recent call last):
  File "sb_server.py", line 104, in ?
  File "spambayes\message.pyc", line 215, in ?
  File "spambayes\message.pyc", line 147, in __init__
  File "spambayes\message.pyc", line 159, in load
  File "pickle.pyc", line 1390, in load
  File "pickle.pyc", line 872, in load
KeyError: '\x00'

I'm now completely baffled: can anyone offer any clues as to what's 
going on? None of the previous versions have ever given me any trouble 
with installation so I really can't understand this.

Many thanks,

Robert Neuschul


From tcrispino at sbiplaw.com  Fri Nov  5 18:26:44 2004
From: tcrispino at sbiplaw.com (Tracy Crispino)
Date: Fri Nov  5 18:25:37 2004
Subject: [Spambayes] question
Message-ID: <356D264FFE32D71198A200C09F1A59332B0E97@SWANSONBRATS01>

I installed spambayes but it seems to be filtering someone else's "IN" box
that I have rights too and not mine.  What can I do to change this?

 

Tracy E. Crispino

Legal Assistant

Swanson & Bratschun, L.L.C.

Telephone:  (303) 268-0066

Facsimile:   (303) 268-0065

***********************************************************

This e-mail message and any attachments are confidential and may be
attorney-client privileged.  If you are not the intended recipient please
notify Swanson & Bratschun, L.L.C. immediately by telephone at 303-268-0066
or by e-mail to tcrispino@sbiplaw.com and destroy all copies of this
message.

 

 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.python.org/pipermail/spambayes/attachments/20041105/b598eb03/attachment.htm
From bruce at argusology.com  Sat Nov  6 01:16:34 2004
From: bruce at argusology.com (bruce)
Date: Sat Nov  6 01:16:53 2004
Subject: [Spambayes] TypeError:object does not support item assignment
Message-ID: <16780.6114.220466.218858@argusology.com>

I'm having difficulty getting spambayes to work with my pop mail server and need some help troubleshooting.

- Linux version 2.4.27 
- spambayes-1.0rc2
I'm using stunnel to connect to my remote mail server, but it makes no
difference whether I connect directly to mailServer:110 - 
I get the same error messages. If I bypass spambayes, all is fine (except for the loads of spam, of course).

   [bruce@argusology mail]$ /usr/bin/sb_server.py
   SpamBayes POP3 Proxy Version 1.0 (July 2004)
   and engine SpamBayes Engine Version 0.3 (January 2004).

   Loading database... Listener on port 8110 is proxying localhost:110
   User interface url is http://localhost:8880/

when I try to connect to the pop server:
   [bruce@argusology bruce]$ telnet localhost 8110
   Trying 127.0.0.1...
   Connected to localhost.
   Escape character is '^]'.

I get ([bruce@argusology mail]):
   error: uncaptured python exception, closing channel <__main__.BayesProxyListener listening :8110 at 0x40acfe4c> /
       (exceptions.TypeError:object does not support item assignment
   [/usr/lib/python2.4/asyncore.py|read|69]
   [/usr/lib/python2.4/asyncore.py|handle_read_event|385]
   [/usr/lib/python2.4/site-packages/spambayes/Dibbler.py|handle_accept|284]
   [/usr/bin/sb_server.py|__init__|374]
   [/usr/bin/sb_server.py|__init__|199]
   [/usr/bin/sb_server.py|__init__|145]
   [/usr/lib/python2.4/asyncore.py|create_socket|261]
   [/usr/lib/python2.4/asyncore.py|add_channel|245])


then I hit <enter>, and ([bruce@argusology bruce]):
   Connection closed by foreign host.

and I get ([bruce@argusology mail]):
   error: uncaptured python exception, closing channel <__main__.BayesProxy connected 127.0.0.1:34689 at 0x40c1d86c> /
       (exceptions.AttributeError:'_socketobject' object has no attribute 'serverSocket'
   [/usr/lib/python2.4/asyncore.py|read|69]
   [/usr/lib/python2.4/asyncore.py|handle_read_event|391]
   [/usr/lib/python2.4/asynchat.py|handle_read|137]
   [/usr/bin/sb_server.py|found_terminator|290]
   [/usr/lib/python2.4/asyncore.py|__getattr__|366])
   Traceback (most recent call last):
     File "/usr/bin/sb_server.py", line 961, in ?
       run()
     File "/usr/bin/sb_server.py", line 955, in run
       start()
     File "/usr/bin/sb_server.py", line 892, in start
       main(state.servers, state.proxyPorts, state.uiPort, state.launchUI)
     File "/usr/bin/sb_server.py", line 872, in main
       Dibbler.run(launchBrowser=launchUI)
     File "/usr/lib/python2.4/site-packages/spambayes/Dibbler.py", line 701, in run    asyncore.loop(map=context._map)
     File "/usr/lib/python2.4/asyncore.py", line 192, in loop
       poll_fun(timeout, map)
     File "/usr/lib/python2.4/asyncore.py", line 133, in poll
       read(obj)
     File "/usr/lib/python2.4/asyncore.py", line 73, in read
       obj.handle_error()
     File "/usr/lib/python2.4/site-packages/spambayes/Dibbler.py", line 204, in handle_error
       asynchat.async_chat.handle_error(self)
     File "/usr/lib/python2.4/asyncore.py", line 421, in handle_error
       self.close()
     File "/usr/bin/sb_server.py", line 405, in close
       if not self.isClosed:
     File "/usr/lib/python2.4/asyncore.py", line 366, in __getattr__
       return getattr(self.socket, attr)
   AttributeError: '_socketobject' object has no attribute 'isClosed'

I would appreciate it greatly if someone could point me in the right direction.

Thanks,
Bruce
From usr7-28hl at xemaps.com  Sat Nov  6 07:07:03 2004
From: usr7-28hl at xemaps.com (usr7-28hl@xemaps.com)
Date: Sat Nov  6 07:07:08 2004
Subject: [Spambayes] Unable to Access SpamBayes Browser Interface
	Usingsb_imapfilter.py
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E8FBC9B0@its-xchg4.massey.ac.nz>
References: <ECBA357DDED63B4995F5C1F5CBE5B1E8FBC9B0@its-xchg4.massey.ac.nz>
Message-ID: <418C6A07.8050105@mailsnare.net>

Hi, Tony.

I seem to be having trouble either posting to the list or 
receiving my list posts. I posted a reply yesterday and haven't 
seen it; I apologize is this is a repeat.

Tony Meyer wrote:

> There was a response from Scott Burns:
> 
> <http://mail.python.org/pipermail/spambayes/2004-November/015109.html>
> 
> He's correct that the problem here:
> 
> 
>>3. >telnet localhost:8880
>>Connecting to localhost:8880...Could not open connection to 
>>the host, on post 23: Connect failed

For some reason, I didn't receive Scott's reply. (Thanks anyway, 
Scott!) I even searched my entire account for a message from him 
and didn't see one.

> Is that you need to do "telnet localhost 8880" (no colon) in order to test
> the connection.

I knew that! ;-) It is a little embarrassing; the "on post 23" is 
somewhat of a giveaway, isn't it? Anyway, I'm pretty sure I did 
test this using both formats at one time....

> When you know if you can telnet to it, you'll know whether the web interface
> isn't being served, or if it's a problem connecting to it in your browser.
> The latter is likely to be something to do with proxies or firewalls that
> you are using.  The former could be a SpamBayes bug.

A "telnet localhost 8880" also fails. So where do we go from 
here, since it seems as though I'm the only one with this bug, or 
whatever it may be?

Thanks.

-- 
Keith
From VB124880 at ncr.com  Sat Nov  6 15:50:56 2004
From: VB124880 at ncr.com (Bartoloma, Victor)
Date: Sat Nov  6 16:49:38 2004
Subject: [Spambayes] Fails to filter unless Filter now is selected.
Message-ID: <E5DA3625B5F00546BFF42C3FA9D476470A64E1D4@susday231.corp.ncr.com>

Skipped content of type multipart/alternative-------------- next part --------------
A non-text attachment was scrubbed...
Name: Outlook.ini
Type: application/octet-stream
Size: 5962 bytes
Desc: Outlook.ini
Url : http://mail.python.org/pipermail/spambayes/attachments/20041106/ee872511/Outlook-0001.obj
-------------- next part --------------
A non-text attachment was scrubbed...
Name: default_bayes_database.db
Type: application/octet-stream
Size: 90112 bytes
Desc: default_bayes_database.db
Url : http://mail.python.org/pipermail/spambayes/attachments/20041106/ee872511/default_bayes_database-0001.obj
-------------- next part --------------
A non-text attachment was scrubbed...
Name: default_message_database.db
Type: application/octet-stream
Size: 24576 bytes
Desc: default_message_database.db
Url : http://mail.python.org/pipermail/spambayes/attachments/20041106/ee872511/default_message_database-0001.obj
From tbaker818 at yahoo.com  Sat Nov  6 23:04:37 2004
From: tbaker818 at yahoo.com (Laurence Baker)
Date: Sat Nov  6 23:04:41 2004
Subject: [Spambayes] Fatal error, run database recovery
Message-ID: <20041106220439.BE4771E4002@bag.python.org>

Hello,

 

I shut down Outlook while Spambayes was deleting confirmed spam and ever
since, even after an uninstall and reinstall, it won't respond to any of the
button commands.

 

Here is what the log says (sounds like I need to run recovery on the
database, but I didn't see anything about that in the troubleshooting
guide). Any help would be much appreciated.

 

Folders/Inbox - pythoncom error: Python error invoking COM method.

Traceback (most recent call last):

  File "win32com\server\policy.pyc", line 275, in _Invoke_

  File "win32com\server\policy.pyc", line 280, in _invoke_

  File "win32com\server\policy.pyc", line 542, in _invokeex_

  File "addin.pyc", line 637, in OnClick

  File "addin.pyc", line 161, in TrainAsSpam

  File "train.pyc", line 52, in train_message

  File "spambayes\classifier.pyc", line 273, in learn

  File "spambayes\classifier.pyc", line 384, in _add_msg

  File "spambayes\storage.pyc", line 281, in _wordinfoset

  File "shelve.pyc", line 130, in __setitem__

  File "bsddb\__init__.pyc", line 120, in __setitem__

bsddb._db.DBRunRecoveryError: (-30982, 'DB_RUNRECOVERY: Fatal error, run
database recovery -- fatal region error detected; run recovery')

 

Thank you,

 

Laurence Baker

http://www.netmarkhome.com

 

 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.python.org/pipermail/spambayes/attachments/20041106/9835e4cc/attachment.html
From robert at imagine.co.uk  Sun Nov  7 04:53:46 2004
From: robert at imagine.co.uk (Robert Neuschul)
Date: Sun Nov  7 04:53:49 2004
Subject: [Spambayes] A most peculair problem with SB 1
Message-ID: <VA.0000574c.03896fbf@imagine.co.uk>

Sorry if you're seeing this message for the 2nd time. 

Hi

I've just encountered a strange problem: SB up to and including RC1 ran 
perfectly well on our old Windows server. Having had to build a new 
2003 server I wanted to setup Spambayes, and decided to take advantage 
of the new full release.

I installed version 2.3.3 of Python.org's Python, added the relevant 
PyWin for 2.3.3 from Sourceforge, and downloaded the full SB 1 
Sourcecode version.

SB was then expanded to a temp directory alongside Python's install 
directory, and from there I ran "python setup.py install" - which duly 
compiles and copies everything it says it wants into the relevant Python 
subdirectories. 
Having got that far I then wanted to install the POP3 proxy service: 
opening a DOS session at the Python/Scripts directory and running 
"python pop3proxy_service.py install" returns several errors [see 
below] which I can't interpret.

C:\Python23\Scripts>python pop3proxy_service.py install
Traceback (most recent call last):
  File "C:\Python23\Scripts\pop3proxy_service.py", line 82, in ?
    import sb_server
  File "C:\Python23\Scripts\sb_server.py", line 104, in ?
    import spambayes.message
  File "c:\Python23\Lib\site-packages\spambayes\message.py", line 215, 
in ?
    msginfoDB = MessageInfoPickle(message_info_db_name)
  File "c:\Python23\Lib\site-packages\spambayes\message.py", line 147, 
in __init
__
    self.load()
  File "c:\Python23\Lib\site-packages\spambayes\message.py", line 159, 
in load
    self.db = pickle.load(fp)
  File "c:\Python23\lib\pickle.py", line 1390, in load
    return Unpickler(file).load()
  File "c:\Python23\lib\pickle.py", line 872, in load
    dispatch[key](self)
KeyError: '\x00'

Note that every file referenced in those error messages does actually 
exist in the Python directory tree.

Thinking I'd made a serious error, or misconfigured something somewhere 
I decided to uninstall Python and SB entirely, reboot and try again 
with the pre-compiled version of SB1. This time any attempt to run the 
server returns similar but not identical errors.

C:\SpamBayes\bin>sb_server
Traceback (most recent call last):
  File "sb_server.py", line 104, in ?
  File "spambayes\message.pyc", line 215, in ?
  File "spambayes\message.pyc", line 147, in __init__
  File "spambayes\message.pyc", line 159, in load
  File "pickle.pyc", line 1390, in load
  File "pickle.pyc", line 872, in load
KeyError: '\x00'

I'm now completely baffled: can anyone offer any clues as to what's 
going on? None of the previous versions have ever given me any trouble 
with installation - both on Win 2000 server and on Win 2003 server, so I 
really don't understand this.

Many thanks,

Robert Neuschul


From anthony at interlink.com.au  Sun Nov  7 16:59:11 2004
From: anthony at interlink.com.au (Anthony Baxter)
Date: Sun Nov  7 16:59:28 2004
Subject: [Spambayes] Windows Installer - please zip
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E8FBC9B9@its-xchg4.massey.ac.nz>
References: <ECBA357DDED63B4995F5C1F5CBE5B1E8FBC9B9@its-xchg4.massey.ac.nz>
Message-ID: <418E464F.7060405@interlink.com.au>

Tony Meyer wrote:

>>However, it 
>>seems like it ought to be accompanied by an MD5 sum (at a minimum). 
> 
> 
> Feel free to open a feature request for this, and we can add it to the
> release process.

You might like to PGP/GPG sign the releases. I do this for Python, and
have nothing but positive feedback. The signature is a separate .asc
file alongside the release file, and has zero effect on the
PGP-ignorant. It also adds very very little work to the release
process.

Anthony
From jsp at PKC.com  Sun Nov  7 23:27:36 2004
From: jsp at PKC.com (Jesse Pelton)
Date: Sun Nov  7 23:31:18 2004
Subject: [Spambayes] Windows Installer - please zip
Message-ID: <4a5603b0b25a46a522310e8682fd6f4b418ea232@vtmail.pkc.com>

Anthony, would you be willing to add your comment to bug 1061119 (https://sourceforge.net/tracker/?func=detail&aid=1061119&group_id=61702&atid=498106)?


-----Original Message-----
From: spambayes-bounces@python.org on behalf of Anthony Baxter
Sent: Sun 11/7/2004 10:59 AM
To: Tony Meyer
Cc: spambayes@python.org
Subject: Re: [Spambayes] Windows Installer - please zip
 
Tony Meyer wrote:

>>However, it 
>>seems like it ought to be accompanied by an MD5 sum (at a minimum). 
> 
> 
> Feel free to open a feature request for this, and we can add it to the
> release process.

You might like to PGP/GPG sign the releases. I do this for Python, and
have nothing but positive feedback. The signature is a separate .asc
file alongside the release file, and has zero effect on the
PGP-ignorant. It also adds very very little work to the release
process.

Anthony
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.python.org/pipermail/spambayes/attachments/20041107/5a211903/attachment.htm
From tameyer at ihug.co.nz  Mon Nov  8 00:29:40 2004
From: tameyer at ihug.co.nz (Tony Meyer)
Date: Mon Nov  8 00:30:23 2004
Subject: [Spambayes] mailing list
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E801520C99@its-xchg4.massey.ac.nz>
Message-ID: <ECBA357DDED63B4995F5C1F5CBE5B1E8FBC9C5@its-xchg4.massey.ac.nz>

> HOW DO I GET OFF THE MAILING LIST?

This information is included with every message that you receive:

You can do this via the web:
<http://mail.python.org/mailman/listinfo/spambayes>

Or via email:
mailto:spambayes-request@python.org?subject=unsubscribe

=Tony.Meyer

-- 
Please always include the list (spambayes@python.org) in your replies
(reply-all), and please don't send me personal mail about SpamBayes.
http://www.massey.ac.nz/~tameyer/writing/reply_all.html explains this.

From tameyer at ihug.co.nz  Mon Nov  8 00:32:47 2004
From: tameyer at ihug.co.nz (Tony Meyer)
Date: Mon Nov  8 00:33:36 2004
Subject: [Spambayes] TypeError:object does not support item assignment
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E801520C9E@its-xchg4.massey.ac.nz>
Message-ID: <ECBA357DDED63B4995F5C1F5CBE5B1E86C37C7@its-xchg4.massey.ac.nz>

> I'm having difficulty getting spambayes to work with my pop 
> mail server and need some help troubleshooting.
> 
> - Linux version 2.4.27 
> - spambayes-1.0rc2
[...]
> I get ([bruce@argusology mail]):
>    error: uncaptured python exception, closing channel 
> <__main__.BayesProxyListener listening :8110 at 0x40acfe4c> /
>        (exceptions.TypeError:object does not support item assignment
>    [/usr/lib/python2.4/asyncore.py|read|69]
>    [/usr/lib/python2.4/asyncore.py|handle_read_event|385]

I suspect this is because you are using Python 2.4.  There were a couple of
issues with using 2.4 that have been since fixed in CVS, but were not
included in 1.0rc2 or 1.0.  If you have Python 2.3/2.2 available, try
running sb_server with that, and see if it works.  If it does, then you can
either using that Python version or use spambayes from CVS until the next
release comes out.

=Tony.Meyer

-- 
Please always include the list (spambayes@python.org) in your replies
(reply-all), and please don't send me personal mail about SpamBayes.
http://www.massey.ac.nz/~tameyer/writing/reply_all.html explains this.

From tameyer at ihug.co.nz  Mon Nov  8 01:35:58 2004
From: tameyer at ihug.co.nz (Tony Meyer)
Date: Mon Nov  8 01:37:41 2004
Subject: [Spambayes] Windows Installer - please zip
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E801520CA8@its-xchg4.massey.ac.nz>
Message-ID: <ECBA357DDED63B4995F5C1F5CBE5B1E86C37C8@its-xchg4.massey.ac.nz>

[Anthony]
> You might like to PGP/GPG sign the releases. I do this for 
> Python, and have nothing but positive feedback. The signature 
> is a separate .asc file alongside the release file, and has 
> zero effect on the PGP-ignorant. It also adds very very 
> little work to the release process.

Any chance you want to add a wee how-to to the end of README-DEVEL.txt?  I'm
happy to do this, but would rather not stuff it up :)

Is there any point in retrospectively signing the 1.0 release, or should I
just start this from 1.0.1 and 1.1a1?  (Assuming that I do the next release,
which is probably a reasonable assumption).

=Tony.Meyer

From tameyer at ihug.co.nz  Mon Nov  8 03:05:03 2004
From: tameyer at ihug.co.nz (Tony Meyer)
Date: Mon Nov  8 03:06:47 2004
Subject: [Spambayes] Re: [spambayes-dev] wish: autotraining
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E801520C8F@its-xchg4.massey.ac.nz>
Message-ID: <ECBA357DDED63B4995F5C1F5CBE5B1E86C37C9@its-xchg4.massey.ac.nz>

> The setting I am referring to is in the advanced settings
> page. 

Sorry - I didn't realise that we exposed that option.  I've checked in a
change to the text to make it clear that it doesn't apply to the POP3 proxy.

> I thought it would mean that the autotrained spam-messages
> would not appear in the (manual) traininglist marked as spam.

It's not a particularly suitable solution for the POP3 proxy for a couple of
reasons:

 1. Mis-training a message has very negative consequences for future
accuracy.  This means that if the proxy trained based on the classification,
and you didn't correct the mistake, classification accuracy would get
progressively worse.

 2. This would (without other modifications) mean that you were using "train
on everything".  This is not the recommended method of training, and doesn't
lead to optimal results.  <http://entrian.com/sbwiki/TrainingIdeas> has more
information about training techniques.

> Do I read that it wont work because this is not a
> "procmail-based solution" ?

Yes, only the sb_filter.py script makes use of this option.

=Tony.Meyer

-- 
Please always include the list (spambayes@python.org) in your replies
(reply-all), and please don't send me personal mail about SpamBayes.
http://www.massey.ac.nz/~tameyer/writing/reply_all.html explains this.

From tameyer at ihug.co.nz  Mon Nov  8 03:14:44 2004
From: tameyer at ihug.co.nz (Tony Meyer)
Date: Mon Nov  8 03:15:48 2004
Subject: [Spambayes] A most peculair problem with SB 1
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E801520C9A@its-xchg4.massey.ac.nz>
Message-ID: <ECBA357DDED63B4995F5C1F5CBE5B1E86C37CA@its-xchg4.massey.ac.nz>

> Having got that far I then wanted to install the POP3 proxy service: 
> opening a DOS session at the Python/Scripts directory and running 
> "python pop3proxy_service.py install" returns several errors [see 
> below] which I can't interpret.
[...]
>   File "c:\Python23\lib\pickle.py", line 872, in load
>     dispatch[key](self)
> KeyError: '\x00'

This error occurs when there's a problem loading the (pickled) database -
from the traceback, the 'messageinfo' one.  There are a couple of likely
causes:

 1. The file that the config files points to as the 'messageinfo' database
isn't a pickle, it's a bsddb db.  In this case the [Storage]
persistent_use_database ("Use database for storage") option needs to be set
to "dbm", or the file needs to be removed to start with a fresh (empty)
pickle.

 2. Something has gone wrong with the messageinfo pickle (i.e. when it was
being saved).  CVS now saves to a temp file first to avoid this, but 1.0
doesn't, so it could be the case.

Otherwise, this might be a SpamBayes bug that's causing it to try and load
the database as a pickle when it has been told not to (I hope not, but it
could be...).

=Tony.Meyer

-- 
Please always include the list (spambayes@python.org) in your replies
(reply-all), and please don't send me personal mail about SpamBayes.
http://www.massey.ac.nz/~tameyer/writing/reply_all.html explains this.

From tameyer at ihug.co.nz  Mon Nov  8 03:25:58 2004
From: tameyer at ihug.co.nz (Tony Meyer)
Date: Mon Nov  8 03:26:29 2004
Subject: [Spambayes] question
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E801520C9B@its-xchg4.massey.ac.nz>
Message-ID: <ECBA357DDED63B4995F5C1F5CBE5B1E86C37CB@its-xchg4.massey.ac.nz>

> I installed spambayes but it seems to be filtering
> someone else's "IN" box that I have rights too and
> not mine.  What can I do to change this?

1. Click on the "SpamBayes" button on the SpamBayes toolbar.

2. Choose "SpamBayes Manager..." from the dropdown menu that appears.

3. Click on the "Filtering" tab.

4. Click the top right "Browse" button.

5. Untick the box next to the folder that should *not* be filtered.

6. Tick the boxes next to any folders that *should* be filtered.  You may
need to expand the list out by clicking on the little "+" boxes.

7. Click "OK".

8. Click "Close".

=Tony.Meyer

-- 
Please always include the list (spambayes@python.org) in your replies
(reply-all), and please don't send me personal mail about SpamBayes.
http://www.massey.ac.nz/~tameyer/writing/reply_all.html explains this.

From tameyer at ihug.co.nz  Mon Nov  8 03:34:45 2004
From: tameyer at ihug.co.nz (Tony Meyer)
Date: Mon Nov  8 03:36:13 2004
Subject: [Spambayes] Fails to filter unless Filter now is selected.
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E801520CA1@its-xchg4.massey.ac.nz>
Message-ID: <ECBA357DDED63B4995F5C1F5CBE5B1E86C37CC@its-xchg4.massey.ac.nz>

> <<default_bayes_database.db>> <<default_message_database.db>> 

Nothing appears to be wrong with these database files.

Do any errors appear?  Check that the correct folders are selected (ones
where mail will be *after* all Outlook rules have run).

The real clues to what might be going wrong are in your log file.  The
troubleshooting guide (SpamBayes->Help->Troubleshooting Guide) explains
where to find those.

=Tony.Meyer

-- 
Please always include the list (spambayes@python.org) in your replies
(reply-all), and please don't send me personal mail about SpamBayes.
http://www.massey.ac.nz/~tameyer/writing/reply_all.html explains this.

From tameyer at ihug.co.nz  Mon Nov  8 03:08:06 2004
From: tameyer at ihug.co.nz (Tony Meyer)
Date: Mon Nov  8 03:43:29 2004
Subject: [Spambayes] Icon
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E801520C95@its-xchg4.massey.ac.nz>
Message-ID: <ECBA357DDED63B4995F5C1F5CBE5B1E8FBC9CF@its-xchg4.massey.ac.nz>

> I am missing my Recover form Spam icon. Can you tell me how 
> to recover it?

Please see FAQ 3.15:

<http://spambayes.org/faq.html#the-recover-from-spam-button-no-longer-appear
s>

=Tony.Meyer

-- 
Please always include the list (spambayes@python.org) in your replies
(reply-all), and please don't send me personal mail about SpamBayes.
http://www.massey.ac.nz/~tameyer/writing/reply_all.html explains this.

From tameyer at ihug.co.nz  Mon Nov  8 03:07:07 2004
From: tameyer at ihug.co.nz (Tony Meyer)
Date: Mon Nov  8 03:43:32 2004
Subject: [Spambayes] Fatal error, run database recovery
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E801520CA2@its-xchg4.massey.ac.nz>
Message-ID: <ECBA357DDED63B4995F5C1F5CBE5B1E8FBC9CE@its-xchg4.massey.ac.nz>

> bsddb._db.DBRunRecoveryError: (-30982, 'DB_RUNRECOVERY:
> Fatal error, run database recovery -- fatal region error
> detected; run recovery')
 
Please see FAQ 5.2:

<http://spambayes.org/faq.html#i-get-a-dbrunrecoveryerror-message>

=Tony.Meyer

-- 
Please always include the list (spambayes@python.org) in your replies
(reply-all), and please don't send me personal mail about SpamBayes.
http://www.massey.ac.nz/~tameyer/writing/reply_all.html explains this.

From aogilvie at computerworks.bc.ca  Mon Nov  8 07:33:32 2004
From: aogilvie at computerworks.bc.ca (Allan Ogilvie)
Date: Mon Nov  8 07:59:51 2004
Subject: [Spambayes] Bounce Backs
Message-ID: <003201c4c55c$de8d45c0$6400a8c0@cworks01>

I am happy with Spam Bayes as a spam filter. So far it has given me much
better filtering controls on incoming spam emails.

Is there a way to "bounce-back" email addresses that I have identified as
spam?  This would send back an invalid email address to the sender.
Hopefully this will help prevent further spamming from that person.
Currently, all I can do is just filter continuous spamming from the same
person into a spam folder.
Allan Ogilvie

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.python.org/pipermail/spambayes/attachments/20041107/172eac94/attachment.html
From robert at imagine.co.uk  Mon Nov  8 12:14:13 2004
From: robert at imagine.co.uk (Robert Neuschul)
Date: Mon Nov  8 13:28:43 2004
Subject: [Spambayes] A most peculair problem with SB 1
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E86C37CA@its-xchg4.massey.ac.nz>
References: <ECBA357DDED63B4995F5C1F5CBE5B1E86C37CA@its-xchg4.massey.ac.nz>
Message-ID: <VA.0000574d.006dfd78@imagine.co.uk>

Tony

Thanks for the reply.

In article <ECBA357DDED63B4995F5C1F5CBE5B1E86C37CA@its-xchg4.massey.ac.nz>, 
Tony Meyer wrote:
>  1. The file that the config files points to as the 'messageinfo' database
> isn't a pickle, it's a bsddb db.  In this case the [Storage]
> persistent_use_database ("Use database for storage") option needs to be set
> to "dbm", or the file needs to be removed to start with a fresh (empty)
> pickle.

Well I hadn't even got as far as editing the ini file so if this was pointing 
at a bsddb.db then it was not of my doing. I shall try deleting the .db and 
try it again.

>  2. Something has gone wrong with the messageinfo pickle (i.e. when it was
> being saved).  CVS now saves to a temp file first to avoid this, but 1.0
> doesn't, so it could be the case.

Now that sounds somewhat more likely. I'll try downloading from the CVS and 
test that version.

> Otherwise, this might be a SpamBayes bug that's causing it to try and load
> the database as a pickle when it has been told not to (I hope not, but it
> could be...).

One hopes it's not that :-)

Robert.


From hendrik.mertens at pandora.be  Mon Nov  8 14:34:13 2004
From: hendrik.mertens at pandora.be (Hendrik Mertens)
Date: Mon Nov  8 14:34:26 2004
Subject: [Spambayes] Puzzling
Message-ID: <000301c4c597$a2ec3860$1502a8c0@DellPC>

 

Dear,

 

SpamBayes does a fine job. This is just to inform you about
something that puzzles me.

 

At a rate of four times a day I receive an e-mail message with
"Enjoy Unlimited PayPerView in Complete Freedom" in the subject
line (internet header below). These people are extremely
irritating - I don't even live in the US, which is clear by my
e-mail address, but they don't bother. In all, I must have
received their message at least two hundred times.

 

I keep deleting it as spam, but still it manages to avoid being
filtered by SpamBayes. Can I do something about this?

 

Kind regards,

 

Hendrik Mertens

 

 

 

Return-Path: <Tyler@Powerful.Acernautic.com>

Delivered-To: hendrik.mertens@telenet.be

Received: (qmail 26553 invoked from network); 8 Nov 2004 07:16:32
-0000

Received: from unknown (HELO elektra.telenet-ops.be)
([195.130.136.19])

          (envelope-sender <Tyler@Powerful.Acernautic.com>)

          by ate.telenet-ops.be (qmail-ldap-1.03) with SMTP

          for <hendrik.mertens@telenet.be>; 8 Nov 2004 07:16:32
-0000

Received: from kelmis.telenet-ops.be (kelmis.telenet-ops.be
[195.130.132.52])

         by elektra.telenet-ops.be (8.12.8/8.12.8) with ESMTP id
iA87GVTK024708

         for <hendrik.mertens@telenet.be>; Mon, 8 Nov 2004
08:16:32 +0100

Received: from host013.acernautic.com (host013.acernautic.com
[216.108.233.13])

         by kelmis.telenet-ops.be (Postfix) with ESMTP id
3391D4BDC6

         for <hendrik.mertens@pandora.be>; Mon,  8 Nov 2004
08:16:30 +0100 (MET)

To: hendrik.mertens@pandora.be

From: "Tyler" <Tyler@Powerful.Acernautic.com>

Reply-To: "Powerful"
<Reply.537499054.817961@Powerful.Acernautic.com>

X-Mailer: IIS 6.0 .Net CLR 2004 Build 7.0.2416

X-MimeOLE: Produced By Microsoft Mime-OLE V.537499054

Date: Sun, 7 Nov 2004 11:07:34 -0800

Subject: [SPAM] Enjoy Unlimited PayPerView in Complete Freedom

MIME-Version: 1.0

Content-Type: multipart/alternative;
boundary=----=_NxtPrt_boundary_000_aPpABvk398f_v.2.0.ENC7bin

Message-Id: <20041108071630.3391D4BDC6@kelmis.telenet-ops.be>

X-Brightmail-Tracker: AAAAAQEqHrs=

 

 

 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.python.org/pipermail/spambayes/attachments/20041108/457565e7/attachment.htm
From dtesler at tcg-systems.com  Mon Nov  8 17:56:18 2004
From: dtesler at tcg-systems.com (David Tesler)
Date: Mon Nov  8 17:55:20 2004
Subject: [Spambayes] Spambayes toolbar is missing
Message-ID: <20041108165518.F0C041E4004@bag.python.org>

Hi, 

 

I have a client that uses Spambayes, (I also use it myself) and she said her
Spambayes toolbar at Outlook is missing. I have attempted to reinstall a new
copy of Spambayes, and it still didn't show up. I was wondering if you guys
know what it could possibly be.

 

Thank you!

 

David Tesler

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.python.org/pipermail/spambayes/attachments/20041108/0a4449d7/attachment.htm
From richie at entrian.com  Mon Nov  8 18:49:16 2004
From: richie at entrian.com (Richie Hindle)
Date: Mon Nov  8 18:49:20 2004
Subject: [Spambayes] Bounce Backs
In-Reply-To: <003201c4c55c$de8d45c0$6400a8c0@cworks01>
References: <003201c4c55c$de8d45c0$6400a8c0@cworks01>
Message-ID: <ibcvo0d52tpu7f61fu4s1livougnrh2gck@4ax.com>


[Allan]
> Is there a way to "bounce-back" email addresses that I have identified as
> spam?  This would send back an invalid email address to the sender.
> Hopefully this will help prevent further spamming from that person.

Please see FAQ 6.5,
http://spambayes.sourceforge.net/faq.html#why-can-t-i-bounce-spam-back-to-the-sender

-- 
Richie Hindle
richie@entrian.com

From gerard at signet.nl  Mon Nov  8 22:06:34 2004
From: gerard at signet.nl (Gerard Haagh)
Date: Mon Nov  8 22:07:58 2004
Subject: [Spambayes] message.py contains bug (fix attached)
Message-ID: <007001c4c5d6$d4251610$feffa8c0@AMD64310001>

Skipped content of type multipart/alternative-------------- next part --------------
#! /usr/bin/env python

"""message.py - Core Spambayes classes.

Classes:
    Message - an email.Message.Message, extended with spambayes methods
    SBHeaderMessage - A Message with spambayes header manipulations
    MessageInfoDB - persistent state storage for Message

Abstract:

    MessageInfoDB is a simple shelve persistency class for the persistent
    state of a Message obect.  For the moment, the db name is hard-coded,
    but we'll have to do this a different way.  Mark Hammond's idea is to
    have a master database, that simply keeps track of the names and instances
    of other databases, such as the wordinfo and msginfo databases.  The
    MessageInfoDB currently does not provide iterators, but should at some
    point.  This would allow us to, for example, see how many messages
    have been trained differently than their classification, for fp/fn
    assessment purposes.

    Message is an extension of the email package Message class, to
    include persistent message information. The persistent state
    -currently- consists of the message id, its current
    classification, and its current training.  The payload is not
    persisted. Payload persistence is left to whatever mail client
    software is being used.

    SBHeaderMessage extends Message to include spambayes header specific
    manipulations.

Usage:
    A typical classification usage pattern would be something like:

    >>> msg = spambayes.message.SBHeaderMessage()
    >>> msg.setPayload(substance) # substance comes from somewhere else
    >>> id = msg.setIdFromPayload()

    >>> if id is None:
    >>>     msg.setId(time())   # or some unique identifier

    >>> msg.delSBHeaders()      # never include sb headers in a classification

    >>> # bayes object is your responsibility
    >>> (prob, clues) = bayes.spamprob(msg.asTokens(), evidence=True)

    >>> msg.addSBHeaders(prob, clues)


    A typical usage pattern to train as spam would be something like:

    >>> msg = spambayes.message.SBHeaderMessage()
    >>> msg.setPayload(substance) # substance comes from somewhere else
    >>> id = msg.setId(msgid)     # id is a fname, outlook msg id, something...

    >>> msg.delSBHeaders()        # never include sb headers in a train

    >>> if msg.getTraining() == False:   # could be None, can't do boolean test
    >>>     bayes.unlearn(msg.asTokens(), False)  # untrain the ham

    >>> bayes.learn(msg.asTokens(), True) # train as spam
    >>> msg.rememberTraining(True)


To Do:
    o Master DB module, or at least make the msginfodb name an options parm
    o Figure out how to safely add message id to body (or if it can be done
      at all...)
    o Suggestions?

    """

# This module is part of the spambayes project, which is Copyright 2002-3
# The Python Software Foundation and is covered by the Python Software
# Foundation license.

__author__ = "Tim Stone <tim@fourstonesExpressions.com>"
__credits__ = "Mark Hammond, Tony Meyer, all the spambayes contributors."

from __future__ import generators

try:
    True, False
except NameError:
    # Maintain compatibility with Python 2.2
    True, False = 1, 0
    def bool(val):
        return not not val

import os
import types
import math
import re
import errno
import shelve
import pickle

import email
import email.Message
import email.Parser
import email.Header

from spambayes import dbmstorage
from spambayes.Options import options, get_pathname_option
from spambayes.tokenizer import tokenize

from cStringIO import StringIO

CRLF_RE = re.compile(r'\r\n|\r|\n')

class MessageInfoBase(object):
    def __init__(self, db_name):
        self.db_name = db_name

    def _getState(self, msg):
        if self.db is not None:
            try:
                attributes = self.db[msg.getId()]
            except KeyError:
                pass
            else:
                if not isinstance(attributes, types.ListType):
                    # Old-style message info db, which only
                    # handles storing 'c' and 't'.
                    (msg.c, msg.t) = attributes
                    return
                for att, val in attributes:
                    setattr(msg, att, val)

    def _setState(self, msg):
        if self.db is not None:
            attributes = []
            for att in msg.stored_attributes:
                attributes.append((att, getattr(msg, att)))
            self.db[msg.getId()] = attributes
            self.store()

    def _delState(self, msg):
        if self.db is not None:
            del self.db[msg.getId()]
            self.store()

class MessageInfoPickle(MessageInfoBase):
    def __init__(self, db_name, pickle_type=1):
        MessageInfoBase.__init__(self, db_name)
        self.mode = pickle_type
        self.load()

    def load(self):
        try:
            fp = open(self.db_name, 'rb')
        except IOError, e:
            if e.errno == errno.ENOENT:
                # New pickle
                self.db = {}
            else:
                raise
        else:
            self.db = pickle.load(fp)
            fp.close()

    def close(self):
        # we keep no resources open - nothing to do
        pass

    def store(self):
        fp = open(self.db_name, 'wb')
        pickle.dump(self.db, fp, self.mode)
        fp.close()

class MessageInfoDB(MessageInfoBase):
    def __init__(self, db_name, mode='c'):
        MessageInfoBase.__init__(self, db_name)
        self.mode = mode
        self.load()

    def load(self):
        try:
            self.dbm = dbmstorage.open(self.db_name, self.mode)
            self.db = shelve.Shelf(self.dbm)
        except dbmstorage.error:
            # This probably means that we don't have a dbm module
            # available.  Print out a warning, and continue on
            # (not persisting any of this data).
            if options["globals", "verbose"]:
                print "Warning: no dbm modules available for MessageInfoDB"
            self.dbm = self.db = None

    def __del__(self):
        self.close()

    def close(self):
        # Close our underlying database.  Better not assume all databases
        # have close functions!
        def noop(): pass
        getattr(self.db, "close", noop)()
        getattr(self.dbm, "close", noop)()

    def store(self):
        if self.db is not None:
            self.db.sync()

# This should come from a Mark Hammond idea of a master db
# For the moment, we get the name of another file from the options,
# so that these files don't litter lots of working directories.
# Once there is a master db, this option can be removed.
message_info_db_name = get_pathname_option("Storage", "messageinfo_storage_file")
if options["Storage", "persistent_use_database"] is True or \
   options["Storage", "persistent_use_database"] == "True" or \
   options["Storage", "persistent_use_database"] == "dbm":
    msginfoDB = MessageInfoDB(message_info_db_name)
elif options["Storage", "persistent_use_database"] is False or \
     options["Storage", "persistent_use_database"] == "False" or \
     options["Storage", "persistent_use_database"] == "pickle":
    msginfoDB = MessageInfoPickle(message_info_db_name)

class Message(email.Message.Message):
    '''An email.Message.Message extended for Spambayes'''

    def __init__(self):
        email.Message.Message.__init__(self)

        # persistent state
        self.stored_attributes = ['c', 't',]
        self.id = None
        self.c = None
        self.t = None

        # non-persistent state includes all of email.Message.Message state

    # This function (and it's hackishness) can be avoided by using the
    # message_from_string and sbheadermessage_from_string functions
    # at the end of the module.  i.e. instead of doing this:
    #   >>> msg = spambayes.message.SBHeaderMessage()
    #   >>> msg.setPayload(substance)
    # you do this:
    #   >>> msg = sbheadermessage_from_string(substance)
    # imapfilter has an example of this in action
    def setPayload(self, payload):
        prs = email.Parser.Parser()
        fp = StringIO(payload)
        # this is kindof a hack, due to the fact that the parser creates a
        # new message object, and we already have the message object
        prs._parseheaders(self, fp)
        # we may want to do some header parsing error handling here
        # to try to extract important headers regardless of malformations
        prs._parsebody(self, fp)

    def setId(self, id):
        if self.id and self.id != id:
            raise ValueError, "MsgId has already been set, cannot be changed"

        if id is None:
            raise ValueError, "MsgId must not be None"

        if not type(id) in types.StringTypes:
            raise TypeError, "Id must be a string"

        self.id = id
        msginfoDB._getState(self)

    def getId(self):
        return self.id

    def asTokens(self):
        return tokenize(self.as_string())

    def tokenize(self):
        return self.asTokens()

    def _force_CRLF(self, data):
        """Make sure data uses CRLF for line termination."""
        return CRLF_RE.sub('\r\n', data)

    def as_string(self, unixfrom=False):
        # The email package stores line endings in the "internal" Python
        # format ('\n').  It is up to whoever transmits that information to
        # convert to appropriate line endings (according to RFC822, that is
        # \r\n *only*).  imaplib *should* take care of this for us (in the
        # append function), but does not, so we do it here
        try:
            return self._force_CRLF(\
                email.Message.Message.as_string(self, unixfrom))
        except TypeError:
            parts = []
            for part in self.get_payload():
                parts.append(email.Message.Message.as_string(part, unixfrom))
            return self._force_CRLF("\n".join(parts))

    def modified(self):
        if self.id:    # only persist if key is present
            msginfoDB._setState(self)

    def GetClassification(self):
        if self.c == 's':
            return options['Headers','header_spam_string']
        elif self.c == 'h':
            return options['Headers','header_ham_string']
        elif self.c == 'u':
            return options['Headers','header_unsure_string']
        return None

    def RememberClassification(self, cls):
        # this must store state independent of options settings, as they
        # may change, which would really screw this database up

        if cls == options['Headers','header_spam_string']:
            self.c = 's'
        elif cls == options['Headers','header_ham_string']:
            self.c = 'h'
        elif cls == options['Headers','header_unsure_string']:
            self.c = 'u'
        else:
            raise ValueError, \
                  "Classification must match header strings in options"
        self.modified()

    def GetTrained(self):
        return self.t

    def RememberTrained(self, isSpam):
        # isSpam == None means no training has been done
        self.t = isSpam
        self.modified()

    def __repr__(self):
        return "spambayes.message.Message%r" % repr(self.__getstate__())

    def __getstate__(self):
        return (self.id, self.c, self.t)

    def __setstate__(self, t):
        (self.id, self.c, self.t) = t


class SBHeaderMessage(Message):
    '''Message class that is cognizant of Spambayes headers.
    Adds routines to add/remove headers for Spambayes'''

    def __init__(self):
        Message.__init__(self)

    def setIdFromPayload(self):
        try:
            self.setId(self[options['Headers','mailid_header_name']])
        except ValueError:
            return None

        return self.id

    def addSBHeaders(self, prob, clues):
        """Add hammie header, and remember message's classification.  Also,
        add optional headers if needed."""

        if prob < options['Categorization','ham_cutoff']:
            disposition = options['Headers','header_ham_string']
            disptype = 'ham'
        elif prob > options['Categorization','spam_cutoff']:
            disposition = options['Headers','header_spam_string']
            disptype = 'spam'
        else:
            disposition = options['Headers','header_unsure_string']
            disptype = 'unsure'
        self.RememberClassification(disposition)
        self[options['Headers','classification_header_name']] = disposition

        if options['Headers','include_score']:
            disp = "%.*f" % (options["Headers", "header_score_digits"], prob)
            if options["Headers", "header_score_logarithm"]:
                if prob<=0.005 and prob>0.0:
                    x=-math.log10(prob)
                    disp += " (%d)"%x
                if prob>=0.995 and prob<1.0:
                    x=-math.log10(1.0-prob)
                    disp += " (%d)"%x
            self[options['Headers','score_header_name']] = disp

        if options['Headers','include_thermostat']:
            thermostat = '**********'
            self[options['Headers','thermostat_header_name']] = \
                               thermostat[:int(prob*10)]

        if options['Headers','include_evidence']:
            hco = options['Headers','clue_mailheader_cutoff']
            sco = 1 - hco
            evd = []
            for word, score in clues:
                if (word[0] == '*' or score <= hco or score >= sco):
                    if isinstance(word, types.UnicodeType):
                        word = email.Header.Header(word,
                                                   charset='utf-8').encode()
                    evd.append("%r: %.2f" % (word, score))

            # Line-wrap this header, because it can get very long.  We don't
            # use email.Header.Header because that can explode with unencoded
            # non-ASCII characters.  We can't use textwrap because that's 2.3.
            wrappedEvd = []
            headerName = options['Headers','evidence_header_name']
            lineLength = len(headerName) + len(': ')
            for component, index in zip(evd, range(len(evd))):
                wrappedEvd.append(component)
                lineLength += len(component)
                if index < len(evd)-1:
                    if lineLength + len('; ') + len(evd[index+1]) < 78:
                        wrappedEvd.append('; ')
                    else:
                        wrappedEvd.append(';\n\t')
                        lineLength = 8
            self[headerName] = "".join(wrappedEvd)

        # These are pretty ugly, but no-one has a better idea about how to
        # allow filtering in 'stripped down' mailers like Outlook Express,
        # so for the moment, they stay in.
        # options["Headers", "notate_to"] (and notate_subject) can be
        # either a single string (like "spam") or a tuple (like
        # ("unsure", "spam")).  In Python 2.3 checking for a string in
        # something that could be a string or a tuple works fine, but
        # it dies in Python 2.2, because you can't do 'string in string',
        # only 'character in string', so we allow for that.
        if isinstance(options["Headers", "notate_to"], types.StringTypes):
            notate_to = (options["Headers", "notate_to"],)
        else:
            notate_to = options["Headers", "notate_to"]
        if disptype in notate_to:
            try:
                self.replace_header("To", "%s,%s" % (disposition,
                                                     self["To"]))
            except KeyError:
                self["To"] = disposition

        if isinstance(options["Headers", "notate_subject"], types.StringTypes):
            notate_subject = (options["Headers", "notate_subject"],)
        else:
            notate_subject = options["Headers", "notate_subject"]
        if disptype in notate_subject:
            try:
                self.replace_header("Subject", "%s,%s" % (disposition,
                                                          self["Subject"]))
            except KeyError:
                self["Subject"] = disposition

        if options['Headers','add_unique_id']:
            self[options['Headers','mailid_header_name']] = self.id

    def currentSBHeaders(self):
        """Return a dictionary containing the current values of the
        SpamBayes headers.  This can be used to restore the values
        after using the delSBHeaders() function."""
        headers = {}
        for header_name in [options['Headers','classification_header_name'],
                            options['Headers','mailid_header_name'],
                            options['Headers','classification_header_name'] + "-ID",
                            options['Headers','thermostat_header_name'],
                            options['Headers','evidence_header_name'],
                            options['Headers','score_header_name'],
                            options['Headers','trained_header_name'],
                            ]:
            value = self[header_name]
            if value is not None:
                headers[header_name] = value
        return headers

    def delSBHeaders(self):
        del self[options['Headers','classification_header_name']]
        del self[options['Headers','mailid_header_name']]
        del self[options['Headers','classification_header_name'] + "-ID"]  # test mode header
        del self[options['Headers','thermostat_header_name']]
        del self[options['Headers','evidence_header_name']]
        del self[options['Headers','score_header_name']]
        del self[options['Headers','trained_header_name']]

# These perform similar functions to email.message_from_string()
def message_from_string(s, _class=Message, strict=False):
    return email.message_from_string(s, _class, strict)

def sbheadermessage_from_string(s, _class=SBHeaderMessage, strict=False):
    return email.message_from_string(s, _class, strict)
From gravenhof at telenet.be  Mon Nov  8 21:47:43 2004
From: gravenhof at telenet.be (Hostellerie  Gravenhof)
Date: Mon Nov  8 22:15:26 2004
Subject: [Spambayes] (no subject)
Message-ID: <MHEBKHLEHJCJAANLGHHCGENKEMAA.gravenhof@telenet.be>

Skipped content of type multipart/alternative-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: image/jpeg
Size: 2950 bytes
Desc: not available
Url : http://mail.python.org/pipermail/spambayes/attachments/20041108/20edc150/attachment.jpe
From tim.peters at gmail.com  Mon Nov  8 22:25:46 2004
From: tim.peters at gmail.com (Tim Peters)
Date: Mon Nov  8 22:25:56 2004
Subject: [Spambayes] (no subject)
In-Reply-To: <MHEBKHLEHJCJAANLGHHCGENKEMAA.gravenhof@telenet.be>
References: <MHEBKHLEHJCJAANLGHHCGENKEMAA.gravenhof@telenet.be>
Message-ID: <1f7befae041108132571a1a22d@mail.gmail.com>

[Hostellerie  Gravenhof <gravenhof@telenet.be>]
> How can we find files back, which spanbayes deleted? This is an urgent
> matter.

SpamBayes never deletes email, so the question as asked has no answer.
 It would help now if you explained which version of SpamBayes you're
using (for example, the Outlook plugin, or a POP3 proxy?), and
precisely what makes you believe email has been deleted.

If it's the Outlook plugin, then at worst SpamBayes moved email into
the Unsure or Spam folders you told it to use.  Users sometimes delete
those folders themselves by mistake, in which case they can usually
still be found as subfolders in Deleted Items.  SpamBayes never
deletes email, or folders, or anything else.
From kennypitt at hotmail.com  Mon Nov  8 22:56:14 2004
From: kennypitt at hotmail.com (Kenny Pitt)
Date: Mon Nov  8 22:57:06 2004
Subject: [Spambayes] (no subject)
In-Reply-To: <MHEBKHLEHJCJAANLGHHCGENKEMAA.gravenhof@telenet.be>
Message-ID: <BAY16-DAV9vzSoBJHsx00022681@hotmail.com>

Hostellerie Gravenhof wrote:
> How can we find files back, which spanbayes deleted? This is an urgent
> matter.

Check FAQ #3.12:
http://spambayes.sourceforge.net/faq.html#some-of-my-mail-is-going-missing
 
-- 
Kenny Pitt
 

________________________________

From: spambayes-bounces+kennypitt=hotmail.com@python.org
[mailto:spambayes-bounces+kennypitt=hotmail.com@python.org] On Behalf Of
Hostellerie Gravenhof
Sent: Monday, November 08, 2004 3:48 PM
To: spambayes@python.org
Subject: [Spambayes] (no subject)


 


 

Met vriendelijke groet,

Pieter Van Den Heuvel

Zaakvoerder n.v. Pica
Hostellerie 't Gravenhof, 
Oostendestraat 343,
8820 Torhout
Tel:  050/21.23.14
Fax: 050/21.69.36

 


From aw-confirm at ebay.com  Mon Nov  8 21:34:27 2004
From: aw-confirm at ebay.com (eBay Customer Care)
Date: Mon Nov  8 23:00:41 2004
Subject: [Spambayes] scheduled account maintenance
Message-ID: <200411082034.iA8KYRhW010341@host8.globalsecureserver.com>

An HTML attachment was scrubbed...
URL: http://mail.python.org/pipermail/spambayes/attachments/20041108/2ebd55fb/attachment.htm
From tameyer at ihug.co.nz  Mon Nov  8 23:01:01 2004
From: tameyer at ihug.co.nz (Tony Meyer)
Date: Mon Nov  8 23:01:38 2004
Subject: [Spambayes] message.py contains bug (fix attached)
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E801521484@its-xchg4.massey.ac.nz>
Message-ID: <ECBA357DDED63B4995F5C1F5CBE5B1E8FBC9D7@its-xchg4.massey.ac.nz>

> When classifying a message in "class SBHeaderMessage"
> and notate_to and/or notate_subject is
> in effect, and you have set the header_XXX_string to a
> non-default value, no notation is made
> at all in the to or subject fields.

Thanks for the report.  This bug has already been identified, and a fix has
been checked into CVS.  The fix will appear in the next release (it was
identified too late to make it into 1.0).

(The fix is a more correct fix, allowing notate_to/notate_subject to work
correctly with any value).

=Tony.Meyer

-- 
Please always include the list (spambayes@python.org) in your replies
(reply-all), and please don't send me personal mail about SpamBayes.
http://www.massey.ac.nz/~tameyer/writing/reply_all.html explains this.

From tameyer at ihug.co.nz  Tue Nov  9 00:19:25 2004
From: tameyer at ihug.co.nz (Tony Meyer)
Date: Tue Nov  9 00:20:01 2004
Subject: [Spambayes] Puzzling
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E8015213AD@its-xchg4.massey.ac.nz>
Message-ID: <ECBA357DDED63B4995F5C1F5CBE5B1E8FBC9D8@its-xchg4.massey.ac.nz>

> At a rate of four times a day I receive an e-mail message
> with "Enjoy Unlimited PayPerView in Complete Freedom" in
> the subject line (internet header below).
[...]
> I keep deleting it as spam, but still it manages to avoid
> being filtered by SpamBayes. Can I do something about this?

Could you do a "Show spam clues for this message" on one of these (before
training on it) and send the generated message to the list?

The only way to really tell why a message scores what it does is to see the
clues, since everyone has a different training set.

=Tony.Meyer

-- 
Please always include the list (spambayes@python.org) in your replies
(reply-all), and please don't send me personal mail about SpamBayes.
http://www.massey.ac.nz/~tameyer/writing/reply_all.html explains this.

From tameyer at ihug.co.nz  Tue Nov  9 00:38:46 2004
From: tameyer at ihug.co.nz (Tony Meyer)
Date: Tue Nov  9 00:39:23 2004
Subject: [Spambayes] Spambayes toolbar is missing
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E801521400@its-xchg4.massey.ac.nz>
Message-ID: <ECBA357DDED63B4995F5C1F5CBE5B1E8FBC9D9@its-xchg4.massey.ac.nz>

> her Spambayes toolbar at Outlook is missing. I have
> attempted to reinstall a new copy of Spambayes, and
> it still didn't show up. I was wondering if you guys
> know what it could possibly be.

Please try the steps in the troubleshooting guide (there are many possible
causes).

You can get to it (on your working copy) via
SpamBayes->Help->Troubleshooting Guide, or there is a copy online at:

<http://cvs.sf.net/viewcvs.py/*checkout*/spambayes/spambayes/Outlook2000/doc
s/troubleshooting.html?rev=HEAD>

(That should all be one line - my mailer will wrap it).

=Tony.Meyer

-- 
Please always include the list (spambayes@python.org) in your replies
(reply-all), and please don't send me personal mail about SpamBayes.
http://www.massey.ac.nz/~tameyer/writing/reply_all.html explains this.

From LEDWYNNE at aol.com  Tue Nov  9 03:30:34 2004
From: LEDWYNNE at aol.com (LEDWYNNE@aol.com)
Date: Tue Nov  9 03:30:51 2004
Subject: [Spambayes] Where is my "spam" and "suspect" file
Message-ID: <c8.516d9306.2ec185ca@aol.com>

Help. I have no idea where to find my spam folder. I suspect that the daily 
messages that I used to get are in this filter....?
ledwynne@aol.com
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.python.org/pipermail/spambayes/attachments/20041108/d2402bf3/attachment.htm
From tameyer at ihug.co.nz  Tue Nov  9 03:56:32 2004
From: tameyer at ihug.co.nz (Tony Meyer)
Date: Tue Nov  9 03:57:20 2004
Subject: [Spambayes] Where is my "spam" and "suspect" file
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E80152154F@its-xchg4.massey.ac.nz>
Message-ID: <ECBA357DDED63B4995F5C1F5CBE5B1E86C37D2@its-xchg4.massey.ac.nz>

> I have no idea where to find my spam folder.

1. Click the "SpamBayes" button on the SpamBayes toolbar.
2. Choose "SpamBayes Manager..." from the dropdown menu that appears.
3. Click the "Filtering" tab.
4. Click the "Browse" button in the centre right.

This will bring up a dialog that shows (in tree form) the location of the
currently selected spam folder.  From this you should be able to easily find
the folder in Outlook's folder list (View->Folder List).

You could also use Outlook's "Advanced Find" function to find a message that
you suspect/know is in there.

=Tony.Meyer

-- 
Please always include the list (spambayes@python.org) in your replies
(reply-all), and please don't send me personal mail about SpamBayes.
http://www.massey.ac.nz/~tameyer/writing/reply_all.html explains this.

From robert at imagine.co.uk  Tue Nov  9 04:43:56 2004
From: robert at imagine.co.uk (Robert Neuschul)
Date: Tue Nov  9 04:43:58 2004
Subject: [Spambayes] A most peculair problem with SB 1
In-Reply-To: <VA.0000574d.006dfd78@imagine.co.uk>
References: <ECBA357DDED63B4995F5C1F5CBE5B1E86C37CA@its-xchg4.massey.ac.nz>
	<VA.0000574d.006dfd78@imagine.co.uk>
Message-ID: <VA.0000574f.03f81bf6@imagine.co.uk>

Tony

> Well I hadn't even got as far as editing the ini file so if this was pointing 
> at a bsddb.db then it was not of my doing. I shall try deleting the .db and 
> try it again.

Further to this - I'm now thoroughly confused:

1] I can't find any trace of an ini file anywhere on the system after running 
setup.py
2] I can't find any database files on the sytem. I can see the pickle and hammie 
py scripts. I've been searching for .db files - is this the wrong thing to be 
searching for? And where should I be looking for them?

I have to say that I'm very confused by the new location of files: previously 
when everything was retained within the spambayes directory I knew where to look 
for everything and could find it; now with all the scripts apparently 
distributed in at least 2 places within the python tree I'm completely confused. 

> > doesn't, so it could be the case.
> 
> Now that sounds somewhat more likely. I'll try downloading from the CVS and 
> test that version.

I should say that I haven't yet tried the CVS version - I was trying to see if I 
could spot what was going on with the version 1.0 I still have installed.

Robert.




From tameyer at ihug.co.nz  Tue Nov  9 05:05:12 2004
From: tameyer at ihug.co.nz (Tony Meyer)
Date: Tue Nov  9 05:06:31 2004
Subject: [Spambayes] A most peculair problem with SB 1
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E80152156C@its-xchg4.massey.ac.nz>
Message-ID: <ECBA357DDED63B4995F5C1F5CBE5B1E86C37D3@its-xchg4.massey.ac.nz>

> Further to this - I'm now thoroughly confused:
> 
> 1] I can't find any trace of an ini file anywhere on the 
> system after running setup.py

If you've not done any configuration, then there won't be one.  If you have,
it ought to be there somewhere.

> 2] I can't find any database files on the sytem. I can see 
> the pickle and hammie py scripts.

If you haven't done any classifying/filtering, there won't be any.  If you
have, they ought to be there somewhere.

> I've been searching for .db files
> - is this the wrong thing to be searching for?

By default the databases will be called "hammie.db" and
"spambayes.messageinfo.db", yes.

> And where should I be looking for them?

The SpamBayes files will be located (unless you specify full paths) in the
same directory as the last configuration file loaded.  If a configuration
file isn't found, then the default place is in the current working directory
unless you're on Windows (with pywin32 installed or using the binary), in
which case it's a "Spambayes\Proxy" directory in the Windows Application
Data directory.

> I have to say that I'm very confused by the new location of 
> files: previously when everything was retained within the spambayes
> directory I knew where to look for everything and could find
> it; now with all the scripts apparently distributed in at
> least 2 places within the python tree I'm completely confused. 

I presume that in the past you didn't run "setup.py install".  That copies
the scripts to Python's scripts directory and the spambayes package
directory to Lib/site-packages.  That's the standard way that Python
modules/scripts are installed.  If you don't want to do that, you don't have
to run "setup.py install" - as long as the path to the main spambayes
directory is on your PYTHONPATH, you'll be fine (this is what I do, for
example, because I have multiple copies of the spambayes source).

If the path isn't on the PYTHONPATH (various Python directories are, and
IIRC directories on the environment variable PATH), then you can just
add/modify the PYTHONPATH environment variable to include tha path.

Let me know if there definitely are no database files and no configuration
file.  That would mean that something's going wrong when running with
defaults and fresh (non-existing) databases, and that means a SpamBayes bug.

=Tony.Meyer

-- 
Please always include the list (spambayes@python.org) in your replies
(reply-all), and please don't send me personal mail about SpamBayes.
http://www.massey.ac.nz/~tameyer/writing/reply_all.html explains this.

From ececconi at visteon.com  Tue Nov  9 12:02:25 2004
From: ececconi at visteon.com (Cecconi, Enrico (E.))
Date: Tue Nov  9 12:04:38 2004
Subject: [Spambayes] spam with 100% score not moved to junk
Message-ID: <1468DAC3F7E2704686B5AEE348137CD40315C4CA@vistex33.dunton.visteon.com>

Hi Tony,

Spam filtering is back on now. Going to "filter now" and running through
"start filtering" seems to have reset everything. Thanks for your help.

Regards,

EC

Enrico Cecconi M.Sc.
Systems Engineer (Audio)
Visteon (UK) Ltd.
GB-28/H Electronics Tech. Centre
Dunton Engineering Centre
Laindon SS15 6EE
* +44-1268-497813
email: ececconi@visteon.com
> CONFIDENTIALITY NOTICE: This e-mail message including attachments, if any,
is intended only for the person or entity to which it is addressed and may
contain confidential and /or privileged material. Any unauthorized review,
use, disclosure or distribution is prohibited. If you are not the intended
recipient, please contact the sender by reply e-mail and destroy all copies
of the original message. If you are the intended recipient but do not wish
to receive communications through this medium, please do advise the sender
immediately.
> 
> 


-----Original Message-----
From: Tony Meyer [mailto:tameyer@ihug.co.nz]
Sent: 04 November 2004 23:01
To: 'Cecconi, Enrico (E.)'; spambayes@python.org
Subject: RE: [Spambayes] spam with 100% score not moved to junk


> The main thing is SB used to work on 90% of my spam (after 
> trining) now it never works. It behaves as if it's installed 
> but not enabled, however reinstalling made no difference. 
> 
> What is the significance of "NOTE: Skipping folder for this 
> session - temporarily unavailable" in the attached log files 
> (from today)?

Ack.  I can't believe I missed that in the original log (sorry).

*That* is the cause of the problem here - basically SpamBayes can't get hold
of the folder it's trying to watch, so it never installs the 'event hook'
that would tell it about new mail arriving, so it never knows about any of
the mail.

The question then, of course, is why the folder isn't available.  Outlook is
telling SpamBayes that the folder is "temporarily available" - can you think
of any reason why this might be the case?

You could try going into the configuration (SpamBayes Manager->Filtering)
and (a) seeing if the folder is offered in the list of folders to filter and
(b) reselecting it if it is.

You could try selecting another folder on the same store and seeing if it
gives the same error.

Microsoft indicates that possible causes for this error include:

  * Insufficient permissions for the mailbox.

  * If you try to open more than 255 mailboxes in one process (the process
here would be Outlook).  It also says that the actual limit may be lower as
some index values may be reserved.

  * If an incorrect password is provided and a prompt for the password
cannot be displayed.

Anything sound relevant?

=Tony.Meyer

-- 
Please always include the list (spambayes@python.org) in your replies
(reply-all), and please don't send me personal mail about SpamBayes.
http://www.massey.ac.nz/~tameyer/writing/reply_all.html explains this.
From lchmiel at ippt.gov.pl  Tue Nov  9 12:18:21 2004
From: lchmiel at ippt.gov.pl (Leszek Chmielewski)
Date: Tue Nov  9 12:18:37 2004
Subject: [Spambayes] Default settings in "Filtering"?
Message-ID: <NEBBLHGMAKJBCABHNDFDCEIOENAA.lchmiel@ippt.gov.pl>

Dear All,

What are the default settings in the SB Manager "Filtering": 
- Certain Spam threshold
- Possible Spam threshold
?

I have widened the range for "Uncertain" too much.

Leszek
_________________________
dr Leszek Chmielewski
mailto:lchmiel@ippt.gov.pl
http://www.ippt.gov.pl/~lchmiel/
Division of Vision and Measurement Systems
IFTR PAS
Swietokrzyska 21, PL 00-049 Warszawa
From Amir_Katz at bmc.com  Tue Nov  9 12:21:37 2004
From: Amir_Katz at bmc.com (Katz, Amir)
Date: Tue Nov  9 12:22:01 2004
Subject: [Spambayes] spam with 100% score not moved to junk
Message-ID: <F8B9B12641FB6B4FAC761BDEF9D63B040174E818@tlv-ex-01.adprod.bmc.com>

I wonder if this is the same problem that I was complaining a few weeks ago
(mail titled "SB plug-in and Outlook 2003", Oct. 16, 2004), as the error
message is the same. A reminder - my complaint is that when I start Outlook
2003 in "work offline" mode and download the messages, they are not getting
filtered by the plug-in.

Amir 

-----Original Message-----
From: Cecconi, Enrico (E.) [mailto:ececconi@visteon.com] 
Sent: Friday, November 05, 2004 10:30
To: 'Tony Meyer'; spambayes@python.org
Subject: RE: [Spambayes] spam with 100% score not moved to junk

I checked that "Inbox" was in the list. It was. I unchecked it and rechecked
it. I then started "filter messages" and SB started and filtered as
expected, some to junk, some to unsure. SB obcviously still works butn not
at start up or without manual intervention.

I then wondered if there was a timing problem with my network logon (I have
Outlook in "startup")and SpamBayes (although no secondary request for
password has ever appeared). So I closed Outlook, deleted all the SB logs
and then restarted Outlook. This resulted in the attatched log which has
"Processing missed spam in folder 'Inbox' by starting a timer" entry in the
last line. What is that all about? What timer?

rgds,

EC

-----Original Message-----
From: Tony Meyer [mailto:tameyer@ihug.co.nz]
Sent: 04 November 2004 23:01
To: 'Cecconi, Enrico (E.)'; spambayes@python.org
Subject: RE: [Spambayes] spam with 100% score not moved to junk


> The main thing is SB used to work on 90% of my spam (after
> trining) now it never works. It behaves as if it's installed but not 
> enabled, however reinstalling made no difference.
> 
> What is the significance of "NOTE: Skipping folder for this session - 
> temporarily unavailable" in the attached log files (from today)?

Ack.  I can't believe I missed that in the original log (sorry).

*That* is the cause of the problem here - basically SpamBayes can't get hold
of the folder it's trying to watch, so it never installs the 'event hook'
that would tell it about new mail arriving, so it never knows about any of
the mail.

The question then, of course, is why the folder isn't available.  Outlook is
telling SpamBayes that the folder is "temporarily available" - can you think
of any reason why this might be the case?

You could try going into the configuration (SpamBayes Manager->Filtering)
and (a) seeing if the folder is offered in the list of folders to filter and
(b) reselecting it if it is.

You could try selecting another folder on the same store and seeing if it
gives the same error.

Microsoft indicates that possible causes for this error include:

  * Insufficient permissions for the mailbox.

  * If you try to open more than 255 mailboxes in one process (the process
here would be Outlook).  It also says that the actual limit may be lower as
some index values may be reserved.

  * If an incorrect password is provided and a prompt for the password
cannot be displayed.

Anything sound relevant?

=Tony.Meyer

--
Please always include the list (spambayes@python.org) in your replies
(reply-all), and please don't send me personal mail about SpamBayes.
http://www.massey.ac.nz/~tameyer/writing/reply_all.html explains this.

From alfio at canada.com  Tue Nov  9 14:07:37 2004
From: alfio at canada.com (Alfio)
Date: Tue Nov  9 14:05:54 2004
Subject: [Spambayes] Outlook 2000 - read messages flag
Message-ID: <NCEDKHDAJPFGDKNEGCJAGELOCNAA.alfio@canada.com>

Hello,

Spambayes is just great, though I'm having a minor issue with the 'read
message' flag. Even though the spam message is read, the unread message icon
in the toolbar is active. Only way to have the icon disappear is to delete
the mail. This means I have to always check my mail (when I see the icon),
and when I see no new messages in my inbox, then I have to go into the spam
folder and delete the mail. Any ideas?

Thanks
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.python.org/pipermail/spambayes/attachments/20041109/f442e123/attachment.html
From Amir_Katz at bmc.com  Tue Nov  9 15:45:17 2004
From: Amir_Katz at bmc.com (Katz, Amir)
Date: Tue Nov  9 15:45:37 2004
Subject: [Spambayes] Outlook 2000 - read messages flag
Message-ID: <F8B9B12641FB6B4FAC761BDEF9D63B0402FD0739@tlv-ex-01.adprod.bmc.com>

As far as I know, there is no resolution for this as there is no API to set
or clear this flag.
 
See FAQ 3.8: http://spambayes.sourceforge.net/faq.html#id25
<http://spambayes.sourceforge.net/faq.html#id25> 
 
Amir 
 

  _____  

From: Alfio [mailto:alfio@canada.com] 
Sent: Tuesday, November 09, 2004 15:08
To: spambayes@python.org
Subject: [Spambayes] Outlook 2000 - read messages flag


Hello,
 
Spambayes is just great, though I'm having a minor issue with the 'read
message' flag. Even though the spam message is read, the unread message icon
in the toolbar is active. Only way to have the icon disappear is to delete
the mail. This means I have to always check my mail (when I see the icon),
and when I see no new messages in my inbox, then I have to go into the spam
folder and delete the mail. Any ideas?
 
Thanks
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.python.org/pipermail/spambayes/attachments/20041109/81de4bec/attachment.html
From BounceReplies at fmglobal.com  Tue Nov  9 16:54:35 2004
From: BounceReplies at fmglobal.com (BounceReplies@fmglobal.com)
Date: Tue Nov  9 16:55:29 2004
Subject: [Spambayes] RE: Re: Protected Mail Delivery (FMGMSWRET-02EX)
Message-ID: <20041109155528.D03D41E4002@bag.python.org>

Your message has not been delivered to its intended recipient at FM Global. The e-mail's content and/or its attached files do not comply with FM Global's Electronic Communications Policy. 

If this is a legitimate business e-mail, you MUST forward this e-mail to ReturnedMail@fmglobal.com.  You will receive an acknowledgement once it has been received. We will then review the message content and determine if it should be released to the intended recipient. You may also contact the FM Global Corporate Help Desk in Johnston, R.I., USA at +1 (1)401 275 3000 x 1299.  


From talyos at 012.net.il  Tue Nov  9 20:11:40 2004
From: talyos at 012.net.il (Tal Yosef)
Date: Tue Nov  9 21:12:50 2004
Subject: [Spambayes] (no subject)
Message-ID: <000801c4c68f$f1e387e0$27b8b250@home>


-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.python.org/pipermail/spambayes/attachments/20041109/67a42eba/attachment.html
From tameyer at ihug.co.nz  Tue Nov  9 21:59:58 2004
From: tameyer at ihug.co.nz (Tony Meyer)
Date: Tue Nov  9 22:00:35 2004
Subject: [Spambayes] Default settings in "Filtering"?
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E801521606@its-xchg4.massey.ac.nz>
Message-ID: <ECBA357DDED63B4995F5C1F5CBE5B1E86C37D4@its-xchg4.massey.ac.nz>

> What are the default settings in the SB Manager "Filtering": 
> - Certain Spam threshold

90.0

> - Possible Spam threshold

15.0

=Tony.Meyer

-- 
Please always include the list (spambayes@python.org) in your replies
(reply-all), and please don't send me personal mail about SpamBayes.
http://www.massey.ac.nz/~tameyer/writing/reply_all.html explains this.

From gary at championapparel.com  Tue Nov  9 23:31:30 2004
From: gary at championapparel.com (Gary Poeppelman)
Date: Tue Nov  9 23:21:31 2004
Subject: [Spambayes] help
Message-ID: <001501c4c6ab$df288b20$2f64a8c0@S0030021071>

I installed this and now all e-mails pop up for three seconds and
disappear. I can't find them, I do not have a junk folder. 
 
Gary Poeppelman
Champion Apparel and Promotions, Inc.
937-886-9012
Fax 937-886-9015
Toll Free 877-221-7400
Check out our web site:  www.championapparel.com
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.python.org/pipermail/spambayes/attachments/20041109/840f84bf/attachment.html
From tameyer at ihug.co.nz  Tue Nov  9 23:30:07 2004
From: tameyer at ihug.co.nz (Tony Meyer)
Date: Tue Nov  9 23:30:43 2004
Subject: [Spambayes] help
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E80152175C@its-xchg4.massey.ac.nz>
Message-ID: <ECBA357DDED63B4995F5C1F5CBE5B1E8FBC9E9@its-xchg4.massey.ac.nz>

> I installed this and now all e-mails pop up for three
> seconds and disappear. I can't find them, I do not have a
> junk folder. 

Please see FAQ 3.12:

<http://spambayes.org/faq.html#some-of-my-mail-is-going-missing>

=Tony.Meyer

-- 
Please always include the list (spambayes@python.org) in your replies
(reply-all), and please don't send me personal mail about SpamBayes.
http://www.massey.ac.nz/~tameyer/writing/reply_all.html explains this.

From robert at imagine.co.uk  Wed Nov 10 01:14:30 2004
From: robert at imagine.co.uk (Robert Neuschul)
Date: Wed Nov 10 01:14:33 2004
Subject: [Spambayes] A most peculair problem with SB 1
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E86C37D3@its-xchg4.massey.ac.nz>
References: <ECBA357DDED63B4995F5C1F5CBE5B1E86C37D3@its-xchg4.massey.ac.nz>
Message-ID: <VA.00005752.0249d91c@imagine.co.uk>

Tony

One of us is now very confused, and I suspect it has to be me.

> If you've not done any configuration, then there won't be one.  If you have,
> it ought to be there somewhere.

OK.

> If you haven't done any classifying/filtering, there won't be any.  If you
> have, they ought to be there somewhere.

As I said, I couldn't start Spambayes in order to do any classifying, or 
anything else. Which means the error messages I posted are somewhat puzzling.

> The SpamBayes files will be located (unless you specify full paths) 

I wasn't aware of any option to specify paths: none of the documentation I've 
read on installing SB makes a reference to such an option for the full source 
version: and certainly none of the scripts seem to give any indication of a 
path-setting capability. 
The only time I've ever seen such a path selection option is when deploying 
the compiled Windows executable version on other machines. Can you please 
clarify where and how such a path option is available and how it should be 
used, or point to documentation that explains how this option is to be used.

> same directory as the last configuration file loaded.  If a configuration
> file isn't found, then the default place is in the current working directory
> unless you're on Windows (with pywin32 installed or using the binary), in
> which case it's a "Spambayes\Proxy" directory in the Windows Application
> Data directory.

This is very ambiguous phrasing: according to instructions on Windows one

a] installs Python and simply accepts the default install location - which for 
Python.org distributions is usually %SYSTEMBOOTDISK%\PythonNN where NN is the 
major version number.

b] expands the full source version of SB to a temporary directory on the boot 
disk, and then from a DOS/CMD sesssion in that temporary directory one runs 
"python setup.py install" which then compiles and locates files into the 
python tree. Once this is completed one may then delete the entire temporary 
SB directory. 
One should then run either the relevant server startup script such as 
sb_server or the server service install such as script pop3proxy_service.py, 
which are now located within the Python script tree.

In this context 

1] is "working directory" intended to refer to the temporary directory from 
which one has just installed, or to the relevant Python\LIB\ sub directory or 
the Python\Script directory. And if not these directories, then where might SB 
- as you put it - *expect* to find such a configuration file /apart/ from the 
user data area?

2] Having just done a parallel install onto a known-good and known-clean XP 
system, exercising some care I can now detect what should be happening: the 
process appears to be installing the proxy and DBs to 

%SYSTEMDRIVE%\Documents and Settings\CurrentUser\Application Data\Spambayes

Which is, in my view, a terribly bad place to be putting anything on a Windows 
2000/2003 server, and is even worse when SB is or will be running as a 
service.

Any data subsequently entered into the DBs will not get backed up correctly 
and may also not be handled correctly by Shadow Volume services or by any user 
profile management which affects the user's data tree as implemented under 
2000/2003 GP in Active Directory. 

No Windows service should *ever* be installing or relying upon configuration 
or other data located in the *user* Data area and most especially not in the 
Administrator's data area or the data area of any user with administrator 
privileges. It's a fundamental breach of permissions and security and it's bad 
Windows programming practice.

And finally: the reason that the "service.py install" failed on my subsequent 
attempts is quite simple: there appears to have already been a Spambayes 
folder in that location from my first attempt at an installation, and 
subsequent attempts to install failed to write a new copy. I didn't even 
consider looking in that area since nothing in the documentation even 
suggested that this was where such data was located. 
Previous versions of SB appear correctly and much more logically to have held 
such data within the SB directory tree.

> I presume that in the past you didn't run "setup.py install".  

On the contrary: I did exactly that and had previous version of Spambayes 
running perfectly well - but on other hardware and with Windows 2000 Server.

However this is the first time that I've noticed Spambayes compile and copy 
itself into the Python tree. 
In the past with older versions, once Python was installed in c:\python, I 
merely ran SB from an entirely different directory - e.g. c:\spambayes. One 
then ran the pop3 service from there. This no longer works once one has run 
the setup script.

Having now realised what was going on with the Spambayes directory in the 
Application Data Area and sorted out the issue of being unable to write the 
proxy and DB correctly I have managed to get the pop3proxy_service installed 
and W2003 claims the service is running: however that service bluntly refuses 
to proxy workstation mail clients [and there are no server logs to indicate 
what is going on] - the mail client however returns an error 

"ERR Connect failed.WS_ERR(11004), Name lookup - Valid name, no data record 
(mailclnt #356)" 

Which is a complete fabrication: the data record does exist on my own Linux 
server - I can see the record as I type this, and there's also no problem with 
the mail client - it too is a Sourceforge project that's been working fine for 
nearly 15 years and it worked fine with the previous local install of 
Spambayes under W2000S.

Things get worse: the only way to get at the configuration or training pages 
is to run sb_server.py in a local cmd session on the server: this was not 
previously the case, and to run such a cmd session requires that one is logged 
in as a administrator on the server - a less than optimum solution, which once 
again weakens the security of the server.

> Let me know if there definitely are no database files and no configuration
> file.  That would mean that something's going wrong when running with
> defaults and fresh (non-existing) databases, and that means a SpamBayes bug.

As I already said: there were no db and ini files previously - just the proxy 
directory. Those files are now present since I managed to remove the previous 
copy of the application data area. 

Given that I can correctly install this same combination of components to both 
a Windows 2000 Server and an XP workstation, and have it up and working in a 
matter of minutes, and that I'm not altogether inexpert with Windows systems, 
I can only conclude that the problem lies in the interaction of Windows 2003 
and SpamBayes.

I'm begining to wish I'd never even tried this new version. Do please let me 
know what more I can do to help debug this setup.

Sorry if this is a bit lengthy.

Robert.


From MehdiAlimadadi at yahoo.com  Wed Nov 10 02:06:13 2004
From: MehdiAlimadadi at yahoo.com (Mehdi Alimadadi)
Date: Wed Nov 10 02:08:14 2004
Subject: [Spambayes] Please do a plug-in version for Outlook Express as well!
Message-ID: <000a01c4c6c1$bfcdf470$ce736395@MYLAPTOP>

Hi there,

Please do a plug-in version for Outlook Express as well!

Thanks,
MA

From tameyer at ihug.co.nz  Wed Nov 10 02:57:23 2004
From: tameyer at ihug.co.nz (Tony Meyer)
Date: Wed Nov 10 02:58:01 2004
Subject: [Spambayes] A most peculair problem with SB 1
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E8015217A1@its-xchg4.massey.ac.nz>
Message-ID: <ECBA357DDED63B4995F5C1F5CBE5B1E8FBC9ED@its-xchg4.massey.ac.nz>

> One of us is now very confused, and I suspect it has to be me.

I wouldn't bet against it being both of us <wink>.

> As I said, I couldn't start Spambayes in order to do any
> classifying, or anything else. Which means the error messages
> I posted are somewhat puzzling.

Troubling, to me, as I suspect that this means there's a bug somewhere.

> I wasn't aware of any option to specify paths: none of the
> documentation I've read on installing SB makes a reference
> to such an option for the full source version: and certainly
> none of the scripts seem to give any indication of a path-setting
> capability. 
[...]
> Can you please clarify where and how such a path option is available 
> and how it should be used, or point to documentation that explains how 
> this option is to be used.

There is confusion here, I think.  I was referring to various options, like
the database names, where you can specify a full path (e.g.
'C:\spambayes\hammie.db' or a relative one 'spambayes\hammie.db').  These
are all exposed via the web configuration - the help strings for these
options explain that if they are relative then they will be relative to the
configuration file.

[...]
> In this context
> 
> 1] is "working directory" intended to refer to the temporary
> directory from which one has just installed, or to the relevant
> Python\LIB\ sub directory or the Python\Script directory.

When I used "current working directory" (CWD) I meant according to the OS
(in Python terms, the result of 'os.getcwd()').  If the scripts are run from
a shell (DOS/CMD session), then it'll be whatever directory is currently
displayed at the prompt.  So doing:

  C:\Python23\Scripts> sb_server.py

The CWD would be C:\Python23\Scripts.  Doing:

  C:\Python23> Scripts\sb_server.py

The CWD would be C:\Python23.  The fact that it can change so easily (and is
somewhat unpredictable if not run from a shell) is the reason that
everything is (by default) located relative to the config file, which should
exist somewhere after an initial run, and a main reason that on Windows we
try to put the config file (and hence everything else) in the appropriate
Application Data directory.

When running as a service, I have absolutely no idea what would be the CWD.
However, you can't use the service unless the pywin32 extensions are
available, so the CWD would only come into play if there *was* a
bayescustomize.ini file there, which obviously won't be the case with a
fresh install.

> And if not these directories, then where might SB - as you put it
> - *expect* to find such a configuration file /apart/ from the 
> user data area?

SpamBayes will look for (in this order): any files specified in the
environment variable BAYESCUSTOMIZE, a file called "bayescustomize.ini" in
the CWD, a file called ".spambayesrc" in the home directory (this is
intended for non-Windows, though it will work, and I can't recall exactly
where each Windows flavour considers the home directory to be), and finally,
if on Windows and the pywin32 extensions are available, a file called
"Spambayes\Proxy\bayescustomize.ini" in the Windows Application Data
directory.

> 2] Having just done a parallel install onto a known-good and
> known-clean XP system, exercising some care I can now detect what
> should be happening: the process appears to be installing the proxy
> and DBs to 
> 
> %SYSTEMDRIVE%\Documents and Settings\CurrentUser\Application
> Data\Spambayes
> 
> Which is, in my view, a terribly bad place to be putting
> anything on a Windows 2000/2003 server, and is even worse when
> SB is or will be running as a service.
> 
> Any data subsequently entered into the DBs will not get
> backed up correctly and may also not be handled correctly by
> Shadow Volume services or by any user profile management which
> affects the user's data tree as implemented under 2000/2003 GP
> in Active Directory. 

Is this because it is located under the 'wrong' user?  This is only when
running as a service, correct?  I don't see how it otherwise applies (in
that case, being located there ensures that it can be backed up and handled
by user profile management).

> No Windows service should *ever* be installing or relying
> upon configuration or other data located in the *user* Data area
> and most especially not in the Administrator's data area or the
> data area of any user with administrator privileges. It's a
> fundamental breach of permissions and security and it's bad 
> Windows programming practice.

The rules were designed before the service version of sb_server existed.
The location is the appropriate place (I suppose you could debate whether
'Application Data' or 'Local Settings\Application Data' was more
appropriate) for a regular application to put its data.

It would be reasonably straightforward to add something to the
pop3proxy_service.py script so that when the service was installed an
initial bayescustomize.ini file (it could be empty, which would still work)
was created in a location in which it would be found (we can add one more
place to look if necessary, or just use the BAYESCUSTOMIZE environment
variable).  If this sounds like a suitable solution, where (in your opinion)
should data for a service be located by default?

> And finally: the reason that the "service.py install" failed
> on my subsequent attempts is quite simple: there appears to have
> already been a Spambayes folder in that location from my first
> attempt at an installation, and subsequent attempts to install
> failed to write a new copy.

That should not be the case.  SpamBayes tries to find the file, and if it
doesn't, then it checks if the directory exists, and if it doesn't, creates
it.  At least, this is what it is meant to do, and does here.  If it's doing
something else, then that's a bug.

> I didn't even consider looking in that area since nothing in the 
> documentation even suggested that this was where such data was 
> located.

The top of the configuration page (if it were possible to get that far, of
course) indicates where data is being stored (well, it says where the
configuration file is, which is a hint that everything else is there).  The
troubleshooting guide for sb_server
('spambayes\windows\docs\troubleshooting.html' in the source - it gets
installed if you use the binary).  When running as a service, the service
log should include a "using config file X" entry each time the service is
started.

I'm happy to put mention of this elsewhere (everyone is happy to admit that
the documentation needs work) if you can suggest a logical location.

> Previous versions of SB appear correctly and much more
> logically to have held such data within the SB directory tree.

This hasn't changed since September 16 2003, so "previous versions" would
have to be very old ones (1.0a5 and older).  If a "bayescustomize.ini" file
was found in one of the viable locations (as above), then the Application
Data directory would never have been used, though.

> > I presume that in the past you didn't run "setup.py install".
> 
> On the contrary: I did exactly that and had previous version
> of Spambayes running perfectly well - but on other hardware and
> with Windows 2000 Server.
> 
> However this is the first time that I've noticed Spambayes
> compile and copy itself into the Python tree. 

That's all that "setup.py install" does, so I presume it did it in the past
and you just didn't notice?

> In the past with older versions, once Python was installed in
> c:\python, I merely ran SB from an entirely different directory
> - e.g. c:\spambayes. One then ran the pop3 service from there.
> This no longer works once one has run the setup script.

This should still work without problems, and is a perfectly valid way to do
it.  What happens if you try this now?

[...]
> I have managed to get the pop3proxy_service installed
> and W2003 claims the service is running: however that service 
> bluntly refuses to proxy workstation mail clients [and there are
> no server logs to indicate what is going on]

SpamBayes ought to be creating logs if it is running as a binary.  They will
be in the temp directory called 'SpamBayesServiceX.log' (where X is from 1
to 5, with 1 being the most recent).  If it's running from source, then
output gets redirected to the win32trace collector (using PythonWin is the
easiest way to view this, although I think there's a standalone tool as
well).

> - the mail client however returns an error
[...]
> Things get worse: the only way to get at the configuration or
> training pages is to run sb_server.py in a local cmd session on
> the server
[...]

Hmm.  Trying the service myself, I can't get it regularly started when
running from source (CVS or 1.0).  It *does* work if I run in debug mode
(i.e. 'spambayes\windows\pop3proxy_service.py debug'), or if I use the 1.0
binary.  I'll look into why this is.

You're trying to connect to the web interface from the same machine that
it's running on, right?  If not, then you'll need to explicitly allow remote
access via the [pop3proxy] allow_remote_connections and [html_ui]
allow_remote_connections options.  (You can either do this manually, if you
identify the correct configuration file, by adding entries that look like:

[pop3proxy]
allow_remote_connections:130.123.238.51,130.123.238.52,localhost
[html_ui]
allow_remote_connections:130.123.238.51,localhost

or you can connect to the interface locally and do it via Advanced
Configuration page.  However, a flaw with 1.0 (fixed for new versions) means
that the three options (pop3proxy, smtpproxy, html_ui) are indistinguishable
- they are in that order, though).

If access doesn't work from the local machine, then either it's not running
at all, or something odd is happening (can't use that port, maybe?).  The
logs, if they can be found, should have some sort of explanation.

> Given that I can correctly install this same combination of 
> components to both a Windows 2000 Server and an XP workstation,
> and have it up and working in a matter of minutes, and that
> I'm not altogether inexpert with Windows systems, I can only
> conclude that the problem lies in the interaction of Windows 2003 
> and SpamBayes.

So you *only* experience problems when it's Windows 2003?  Not just any time
the new source is used?  If so, do you have any knowledge about things
Windows 2003 does differently?  (I know nothing at all about that).

If you run the pop3proxy_service.py script in debug mode, does it work?  If
so, I presume the problem is the same one that I have, and so once I figure
out what it is, all will be good.  If not, what errors does it give?

=Tony.Meyer

-- 
Please always include the list (spambayes@python.org) in your replies
(reply-all), and please don't send me personal mail about SpamBayes.
http://www.massey.ac.nz/~tameyer/writing/reply_all.html explains this.

From tameyer at ihug.co.nz  Wed Nov 10 03:00:23 2004
From: tameyer at ihug.co.nz (Tony Meyer)
Date: Wed Nov 10 03:00:57 2004
Subject: [Spambayes] Please do a plug-in version for Outlook Express as
	well!
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E8015217B7@its-xchg4.massey.ac.nz>
Message-ID: <ECBA357DDED63B4995F5C1F5CBE5B1E86C37D6@its-xchg4.massey.ac.nz>

> Please do a plug-in version for Outlook Express as well!

Have you used the POP3 proxy, which works with Outlook Express?  i.e. see
FAQ 2.1:

<http://spambayes.sourceforge.net/faq.html#does-spambayes-work-with-outlook-
express>

If you have:

The problem is that Outlook Express is a very stripped down mailer and
doesn't really offer much in the way for developers creating plug-ins.  It
would be much better for use to make improvements to the existing POP3 proxy
(sb_server) - if you have any improvements you would like to see, please let
us know!

=Tony.Meyer

-- 
Please always include the list (spambayes@python.org) in your replies
(reply-all), and please don't send me personal mail about SpamBayes.
http://www.massey.ac.nz/~tameyer/writing/reply_all.html explains this.

From tameyer at ihug.co.nz  Wed Nov 10 03:35:48 2004
From: tameyer at ihug.co.nz (Tony Meyer)
Date: Wed Nov 10 03:36:23 2004
Subject: [Spambayes] A most peculair problem with SB 1
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E8015217D0@its-xchg4.massey.ac.nz>
Message-ID: <ECBA357DDED63B4995F5C1F5CBE5B1E8FBC9EF@its-xchg4.massey.ac.nz>

> Hmm.  Trying the service myself, I can't get it regularly started when
> running from source (CVS or 1.0).  It *does* work if I run in 
> debug mode (i.e. 'spambayes\windows\pop3proxy_service.py debug'), or if 
> I use the 1.0 binary.  I'll look into why this is.

Further to this - the error that appears in the Event Log is:

"""
Python could not import the service's module 
  File "C:\spambayes\windows\pop3proxy_service.py", line 82, in ?
    import sb_server
  File "C:\spambayes\scripts\sb_server.py", line 115, in ?
    from spambayes.FileCorpus import FileCorpus, ExpiryFileCorpus
  File "C:\spambayes\spambayes\FileCorpus.py", line 92, in ?
    import sys, os, gzip, fnmatch, getopt, time, stat
  File "C:\Python24\Lib\gzip.py", line 9, in ?
    import zlib 
exceptions.ImportError: dynamic module does not define init function
(initzlib)
"""

Which is very odd.  It occurs with Python 2.3 or 2.4, so that doesn't appear
to matter.  zlib.pyd is where it should be, and does have initzlib in its
export table.

If I try adding "import zlib" to an otherwise functioning (non-SpamBayes)
Python service, I also get this error, so it's not SpamBayes related.  It is
very odd, however.  I'll keep trying to track this down.

If you get the same error, it would at least explain some things...

=Tony.Meyer

-- 
Please always include the list (spambayes@python.org) in your replies
(reply-all), and please don't send me personal mail about SpamBayes.
http://www.massey.ac.nz/~tameyer/writing/reply_all.html explains this.

From mfakhrin.kamaruddin at emerge.com.my  Wed Nov 10 07:57:47 2004
From: mfakhrin.kamaruddin at emerge.com.my (Fakhrin)
Date: Wed Nov 10 08:43:41 2004
Subject: [Spambayes] Hi all....
Message-ID: <000001c4c6f7$b5cf4ce0$0401a8c0@pc4>

Multi layered filtering is the best solutions right? But, I'm wondering, what do they use to integrate all the filtering techniques into one solution? Like, do they use any framework, middleware, applications or anything to integrate all the bayesian filtering, black and whitelist, etc altogether into one solution?



Best Regards,

Mohd Fakhrin Kamaruddin
Emerge Systems (M) Sdn Bhd
27-3, Block D1, Jalan PJU 1/41, Dataran Prima,
47301 Petaling Jaya, Selangor,
Malaysia
Mobile: (6012) 320 7469
Tel: (603) 7880 5580
Fax: (603) 7880 5578
E-mail: mfakhrin.kamaruddin@emerge.com.my
www.webhosting.com.my
www.webautomate.net
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.python.org/pipermail/spambayes/attachments/20041110/2e55b842/attachment.htm
From jhockin at cox.net  Wed Nov 10 09:19:10 2004
From: jhockin at cox.net (Jim Hockin)
Date: Wed Nov 10 09:23:40 2004
Subject: [Spambayes] How does SpamBayes training count "spam" and "good" mail
Message-ID: <000801c4c6fd$f510a1f0$6f01a8c0@degas>

Anyone know how SpamBayes training counts "spam" and "good" mail? In the SpamBayes 1.0 for Outlook, I designated initial training to look at about 300 "spam" messages in a SPAM folder and about 300 "good" messages in a GOOD folder, checkmarked it to rebuild entire database, and it comes back with "Completed training with 91 spam and 9 good messages" and the Training Database Status tells me the "Database only has 9 good and 91 spam - you should consider performing additional training." Can anyone tell me why it doesn't see the other 500 messages I gave it? What am I missing here?

Jim Hockin
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.python.org/pipermail/spambayes/attachments/20041110/321194a6/attachment.htm
From Amir_Katz at bmc.com  Wed Nov 10 10:15:28 2004
From: Amir_Katz at bmc.com (Katz, Amir)
Date: Wed Nov 10 10:26:36 2004
Subject: [Spambayes] Outlook 2003 is eating our lunch ?
Message-ID: <F8B9B12641FB6B4FAC761BDEF9D63B0403004963@tlv-ex-01.adprod.bmc.com>

I'm no big fan of Micro$oft, but I must admit that the junk mail filter that
comes with Outlook 2003 client is not half bad. It intercepts almost all
e-mails which are in Korean and other Asian languages, as well as many of
the normal spam. From a quick count that I did, in the last 48 hours or so I
got 304 spams, and Outlook intercepted 204 of them.
 
Amir 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.python.org/pipermail/spambayes/attachments/20041110/b4e13a18/attachment.html
From nimrod229 at bezeqint.net  Wed Nov 10 11:47:03 2004
From: nimrod229 at bezeqint.net (shanegbi)
Date: Wed Nov 10 11:46:58 2004
Subject: [Spambayes] uninstalling
Message-ID: <000501c4c712$9e0939c0$385efea9@homev>

would you pls advise how to uinstall the software
thks
S.Hanegbi
From black at www.blackxtreme.nl  Wed Nov 10 12:05:22 2004
From: black at www.blackxtreme.nl (black@www.blackxtreme.nl)
Date: Wed Nov 10 12:05:24 2004
Subject: [Spambayes] Automated reply from black@www.blackxtreme.nl
Message-ID: <200411101105.iAAB5MW10827@ws5.vip.nl>

(dit is een automatische reply op uw mail)

Dank je voor je mail, je krijgt zsm reactie van me.
Thank you for your mail, i will contact you asap.


NOTE FOR YOU
________________________________________________________
Voor de leukste Striptease shows en Special acts bekend van tv
kijk snel op onze site en zoek de leukste shows uit

website:	www.BlackXtreme.nl
email:	Esther@BlackXtreme.nl (ook MSN)
mobiel:	0031(0)6 - 53 296 488 bgg 06 18 66 50 37
imail:	estherbx@imail.nl (to send a quicknote to my mobile)
MSN Patric: blackxcrow@msn.com

Please visite our website if you are looking for Special Acts for your party
BlackXtreme winner EroticAward2000 "the best performance of Europe"	
------------------------------------------------------------------------------------------------------------------
Speakers aan en kies je favoriete dancemusic:
www.blackxtreme.nl/patricmusicroom.html
xxx have fun
----------------------------------------------------------------------------------------------------------------
From lchmiel at ippt.gov.pl  Wed Nov 10 13:00:23 2004
From: lchmiel at ippt.gov.pl (Leszek Chmielewski)
Date: Wed Nov 10 13:00:53 2004
Subject: [Spambayes] Outlook 2000 - read messages flag
In-Reply-To: <NCEDKHDAJPFGDKNEGCJAGELOCNAA.alfio@canada.com>
Message-ID: <NEBBLHGMAKJBCABHNDFDGEJFENAA.lchmiel@ippt.gov.pl>

> then I have to go into the spam folder and delete the mail. Any ideas?

You can click the "Mark as read" and immediately also "Mark as unread" on
any unread message. This in not automatic, but you don't have to delete or
view anything.

Leszek

From jsp at PKC.com  Wed Nov 10 14:15:31 2004
From: jsp at PKC.com (Jesse Pelton)
Date: Wed Nov 10 14:15:36 2004
Subject: [Spambayes] Outlook 2003 is eating our lunch ?
Message-ID: <7610c1e74d2856e40851cd0a5f149f1041921476@vtmail.pkc.com>

I imagine there will be a place for the Outlook plugin for some time to
come. I'm not quite sure why Microsoft hasn't implemented Bayesian
filtering; it's easy to understand, easy to use, and highly effective
(at least as implemented in SpamBayes). If Outlook caught 204 out of 304
spams, it may not be half bad, but it's about a third bad. SpamBayes
does much better for me, and if it doesn't, it's not hard to retrain it.
With Outlook's filtering, you take what you get.


________________________________

	From: spambayes-bounces@python.org
[mailto:spambayes-bounces@python.org] On Behalf Of Katz, Amir
	Sent: Wednesday, November 10, 2004 4:15 AM
	To: spambayes@python.org
	Subject: [Spambayes] Outlook 2003 is eating our lunch ?
	
	
	I'm no big fan of Micro$oft, but I must admit that the junk mail
filter that comes with Outlook 2003 client is not half bad. It
intercepts almost all e-mails which are in Korean and other Asian
languages, as well as many of the normal spam. From a quick count that I
did, in the last 48 hours or so I got 304 spams, and Outlook intercepted
204 of them.
	 
	Amir 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.python.org/pipermail/spambayes/attachments/20041110/ea05837d/attachment.html
From jsp at PKC.com  Wed Nov 10 14:23:39 2004
From: jsp at PKC.com (Jesse Pelton)
Date: Wed Nov 10 14:23:43 2004
Subject: [Spambayes] uninstalling
Message-ID: <1ee02bafead1094e14a3f9e12f116d714192165c@vtmail.pkc.com>

If you're running the Outlook plugin, please see
http://spambayes.sourceforge.net/faq.html#how-do-i-uninstall-the-plug-in
.

> -----Original Message-----
> From: spambayes-bounces@python.org 
> [mailto:spambayes-bounces@python.org] On Behalf Of shanegbi
> Sent: Wednesday, November 10, 2004 5:47 AM
> To: spambayes@python.org
> Subject: [Spambayes] uninstalling
> 
> would you pls advise how to uinstall the software
> thks
> S.Hanegbi
From hendrik.mertens at pandora.be  Tue Nov  9 14:15:47 2004
From: hendrik.mertens at pandora.be (Hendrik Mertens)
Date: Wed Nov 10 14:34:23 2004
Subject: [Spambayes] As requested by Tony Meyer / Spam Clues: [SPAM] Enjoy
	Unlimited PayPerView in Complete Freedom
Message-ID: <001601c4c65e$3a3d2e10$1502a8c0@DellPC>

Skipped content of type multipart/alternative-------------- next part --------------
An embedded message was scrubbed...
From: "Jessica" <Jessica@Powerful.Acernautic.com>
Subject: [SPAM] Enjoy Unlimited PayPerView in Complete Freedom
Date: Mon, 8 Nov 2004 19:39:16 +0100
Size: 2438
Url: http://mail.python.org/pipermail/spambayes/attachments/20041109/f6fe4d75/attachment-0001.mht
From lchmiel at ippt.gov.pl  Wed Nov 10 15:10:19 2004
From: lchmiel at ippt.gov.pl (Leszek Chmielewski)
Date: Wed Nov 10 15:10:34 2004
Subject: [Spambayes] Outlook 2000 - read messages flag
In-Reply-To: <NCEDKHDAJPFGDKNEGCJAGELOCNAA.alfio@canada.com>
Message-ID: <NEBBLHGMAKJBCABHNDFDIEKGENAA.lchmiel@ippt.gov.pl>

> then I have to go into the spam folder and delete the mail. Any ideas?

Outlook hides this icon if you click on the "Mark as read" button (Ctrl+Q),
while the bar is on _any_ message, also a one which was already read. It
seems this takes place starting from Service Pack 3.

Leszek

From robert at imagine.co.uk  Wed Nov 10 15:48:51 2004
From: robert at imagine.co.uk (Robert Neuschul)
Date: Wed Nov 10 15:48:52 2004
Subject: [Spambayes] A most peculair problem with SB 1
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E8FBC9ED@its-xchg4.massey.ac.nz>
References: <ECBA357DDED63B4995F5C1F5CBE5B1E8FBC9ED@its-xchg4.massey.ac.nz>
Message-ID: <VA.00005753.00d4e121@imagine.co.uk>

Tony

This is going to take some time to respond to since there are now 
several issues cropping up. To keep things a bit simpler I shall deal 
with each of them as distinct reply messages.

You will note that my root message concerned itself explicitly with the 
installation and use of SB on Windows 2003. In your message you ask if 
things are different on 2003 - and the answer is yes: very different in 
all sorts of ways. 

For a systems administrator on 2003 things such as Active Directory, 
User Controls, Securities and Permissions, ACLs and Group Policies have 
all changed in a number of ways that are now far more consistent and 
logical [for example Exchange and Active Directory are now properly 
integrated - if Exchange is installed and one creates a new user logon 
for the lan one has automatically created an Exchange account - 
Exchange 'objects' are system objects, and vice versa - see below for 
more on this], and in many many regards these elements of 2003 now have 
much more in common with standard LDAP/X400/X500 and Unix-style 
processes and procedures than they do with previous Windows 2000 
servers. 
In addition, as implied above, 2003 is now almost entirely object 
oriented in its "thinking" and in its management. The top level objects 
are the forests, zones and domains, making Administration across 
multiple failover servers much simpler [but also much more powerful] 
and more logical.

The issue of installing a service, and where that service should look 
for its configuration data, and where and how it should store any 
output data or logging is therefore a fraught one. The one thing that 
one can say with great certainty is that none of these things should 
ever be stored in the user data or user profile areas - the handling of 
these areas by Windows 2003 is now - like so much else - completely 
different, and that handling will vary depending upon the precise 
combination of network and server settings in use. That 
variability/changeability is utterly crucial to the statement that 
services data should not be stored there.

All sorts of things *will* go pear shaped if an advanced or non-default 
systems/network configuration is in use - for example when the user who 
installs SB has Administrator rights and is using terminal services to 
the server to install a product such as Spambayes, and also has the 
equivalent of roaming profiles [now deprecated as a method - there are 
alternative and better ways of doing the same thing] with cached local 
proxying from a workstation [aka Offline working mode] - then all their 
/subsequent/ session data from the roaming [non terminal] login ends up 
in the system-defined directory called \\domain\servername\Users and 
will fail to synchronise with the matching directory called 
\\server\Documents and Settings\User\Application Data\Spambayes, which 
Spambayes will have created during its install. 

That issue relating to the user data config area will get significantly 
more complex and nasty when there are 2 or more servers in use with 
automated failover between them: synchronisation of data and services - 
replication - in the event of a failure in what we used to call the PDC 
under Win 2000 is now capable of being fully automated - but will only 
work reliably if data and configurations are in places which the 
participating replication partner systems understand - and C:\Documents 
and Settings\User_Name\Application Data is not generally considered to 
be one of those safe places.

Service data [whether config or output] is - at least in my opinion - 
normally best stored either within a subdirectory of the directory 
structure of the application itself - in this case that would be 
somewhere such as c:\Python23\Spambayesconfig  - or in a location 
within the Windows tree such as %SYSTEMROOT%\System32\Spambayesconfig 
or similar. 
I'd also suggest that log files should not be sent to temp directories 
which get scavenged by 2003, and that the WMI interface be used to send 
warnings etc to the Windows Application Event log.

That way, synchronisation and failover and backup and versioning can 
all proceed correctly without impact on the performance of SB.

As a part of the changes in the way 2003 is designed and works, 
securities and permissions are somewhat changed as well: and in turn 
the way in which services are allowed to interact with both the wider 
system [which now logically includes any additional 2003 servers within 
the forest/zone/domain] and the desktop and any network resources has 
also changed. I can't tell you precisely how they've changed because 
this isn't my area of exertise and I therefore haven't looked into the 
programming of these services.

Hope this begins to clarify some of the issues. Sorry if there's any 
ambiguity or obscurity - if you have any questions please ask and 'll 
endeavour to answer them.
Other parts of my answer to follow later :-)

Robert


From GSBauer at dow.com  Wed Nov 10 15:34:49 2004
From: GSBauer at dow.com (Bauer, Gregory (GS))
Date: Wed Nov 10 17:08:46 2004
Subject: [Spambayes] Question
Message-ID: <BBF7AF102090E549BF3D1845B4FE479A0D1B39@USMDLMDOWM031.dow.com>

I've accidentally deleted my Spam Bayes Junk Email Suspects folder and files in Microsoft outlook.  Can you please advise instructions on how to recover the file.

Thanks,

Greg Bauer
Alliance Development Director
The Dow Chemical Company
Advanced Electronic Materials
4435 First Street #239
Livermore, CA 94551-2215
Phone: 925-454-2989
FAX:    925-454-3008
Cell:    925-899-7781
email address: gsbauer@dow.com
Visit our website at www.silknetalliance.com <www.silknetalliance.com> 

From tameyer at ihug.co.nz  Wed Nov 10 22:34:43 2004
From: tameyer at ihug.co.nz (Tony Meyer)
Date: Wed Nov 10 22:35:48 2004
Subject: [Spambayes] How does SpamBayes training count "spam" and "good"
	mail
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E80152185C@its-xchg4.massey.ac.nz>
Message-ID: <ECBA357DDED63B4995F5C1F5CBE5B1E86C37D7@its-xchg4.massey.ac.nz>

> Anyone know how SpamBayes training counts "spam" and
> "good" mail? In the SpamBayes 1.0 for Outlook, I designated
> initial training to look at about 300 "spam" messages
> in a SPAM folder and about 300 "good" messages in a GOOD
> folder, checkmarked it to rebuild entire database, and it
> comes back with "Completed training with 91 spam and 9 good
> messages" and the Training Database Status tells me the
> "Database only has 9 good and 91 spam - you should consider
> performing additional training." Can anyone tell me why it
> doesn't see the other 500 messages I gave it? What am I missing
> here?

The log file <0.5 wink>.  Seriously, that'll explain why messages were not
trained.  Wild guesses:

  1.  Some of the mail has never been received (drafts, sent items, ...)
  2.  Some of the mail are not what Outlook calls IPM.Note (regular
messages), for example some Exchange autogenerated messages.
  3.  There was a periodic problem accessing the folder (i.e. run scanpst
over it if it's a PST, or check the connectivity if it's
IMAP/Hotmail/Exchange).

If these don't apply, please feel free to send in a copy of the log file for
the period in which you tried this, and we'll try and explain whatever
errors are there.

=Tony.Meyer

-- 
Please always include the list (spambayes@python.org) in your replies
(reply-all), and please don't send me personal mail about SpamBayes.
http://www.massey.ac.nz/~tameyer/writing/reply_all.html explains this.

From tameyer at ihug.co.nz  Wed Nov 10 22:43:50 2004
From: tameyer at ihug.co.nz (Tony Meyer)
Date: Wed Nov 10 22:44:26 2004
Subject: [Spambayes] As requested by Tony Meyer / Spam Clues: [SPAM]
	EnjoyUnlimited PayPerView in Complete Freedom
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E801521901@its-xchg4.massey.ac.nz>
Message-ID: <ECBA357DDED63B4995F5C1F5CBE5B1E86C37D8@its-xchg4.massey.ac.nz>

"""
Combined Score: 100% (1)
Internal ham score (*H*): 2.22045e-016
Internal spam score (*S*): 1
"""

These scores are prior to training (i.e. moving to the spam folder or
clicking "Delete as Spam"), right?

Was this message not moved to the spam folder?  It should have been - it
scored a resounding 100% - it can't get any higher.

If it wasn't - was it moved to the unsure folder, or did it stay in the
inbox?

Next time this happens, could you send us a copy of your most recent log
file?  The troubleshooting guide (SpamBayes->Help->Troubleshooting Guide)
explains where to find that.

I wonder if perhaps this is a case of this bug:

[ 972359 ] some 100% SPAM sent to unsure
http://sourceforge.net/tracker/index.php?func=detail&aid=972359&group_id=617
02&atid=498103

If so, then the best I can offer is that we are working on getting it
solved, so it should be at some point, hopefully soon.

=Tony.Meyer

-- 
Please always include the list (spambayes@python.org) in your replies
(reply-all), and please don't send me personal mail about SpamBayes.
http://www.massey.ac.nz/~tameyer/writing/reply_all.html explains this.

From tameyer at ihug.co.nz  Wed Nov 10 22:45:51 2004
From: tameyer at ihug.co.nz (Tony Meyer)
Date: Wed Nov 10 22:46:27 2004
Subject: [Spambayes] Question
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E801521952@its-xchg4.massey.ac.nz>
Message-ID: <ECBA357DDED63B4995F5C1F5CBE5B1E8FBC9F2@its-xchg4.massey.ac.nz>

> I've accidentally deleted my Spam Bayes Junk Email Suspects 
> folder and files in Microsoft outlook.  Can you please advise 
> instructions on how to recover the file.

(First, check that the folder isn't just in your Deleted Items folder - if
so, just drag it out).

Open the SpamBayes Manager from the SpamBayes toolbar.  Click the
"Filtering" tab.  Use the lower two Browse buttons to bring up the selector
for the unsure/spam folder.  Select the new folder you wish to use (if you
haven't recreated one, you can do it from that dialog).

=Tony.Meyer

-- 
Please always include the list (spambayes@python.org) in your replies
(reply-all), and please don't send me personal mail about SpamBayes.
http://www.massey.ac.nz/~tameyer/writing/reply_all.html explains this.

From tameyer at ihug.co.nz  Wed Nov 10 22:47:51 2004
From: tameyer at ihug.co.nz (Tony Meyer)
Date: Wed Nov 10 22:48:23 2004
Subject: [Spambayes] Hi all....
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E801521848@its-xchg4.massey.ac.nz>
Message-ID: <ECBA357DDED63B4995F5C1F5CBE5B1E8FBC9F3@its-xchg4.massey.ac.nz>

> Multi layered filtering is the best solutions right?
> But, I'm wondering, what do they use to integrate all
> the filtering techniques into one solution? Like, do they
> use any framework, middleware, applications or anything
> to integrate all the bayesian filtering, black and
> whitelist, etc altogether into one solution?

AFAIK, this is a per-application decision, and there are a variety of
techniques that can be and are used.  If you really want to find out, I'd
say the best plan would be to find a few open-source examples of multi-layer
filtering, and ask them what they do, and draw conclusions from that.

=Tony.Meyer

-- 
Please always include the list (spambayes@python.org) in your replies
(reply-all), and please don't send me personal mail about SpamBayes.
http://www.massey.ac.nz/~tameyer/writing/reply_all.html explains this.

From tameyer at ihug.co.nz  Thu Nov 11 00:39:07 2004
From: tameyer at ihug.co.nz (Tony Meyer)
Date: Thu Nov 11 00:39:43 2004
Subject: [Spambayes] A most peculair problem with SB 1
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E801521921@its-xchg4.massey.ac.nz>
Message-ID: <ECBA357DDED63B4995F5C1F5CBE5B1E86C37DB@its-xchg4.massey.ac.nz>

> This is going to take some time to respond to since there are now 
> several issues cropping up.

Indeed.  It's good to get them sorted though - the service hasn't really
been all that developed, IIRC, so the help is appreciated :)  (as is the
explanation of the 2003 Server stuff, which I am not familiar with at all)

> To keep things a bit simpler I shall deal 
> with each of them as distinct reply messages.

Sounds good.

> The issue of installing a service, and where that service should look 
> for its configuration data, and where and how it should store any 
> output data or logging is therefore a fraught one. The one thing that 
> one can say with great certainty is that none of these things should 
> ever be stored in the user data or user profile areas
[...]
> Service data [whether config or output] is - at least in my opinion - 
> normally best stored either within a subdirectory of the directory 
> structure of the application itself - in this case that would be 
> somewhere such as c:\Python23\Spambayesconfig  - or in a location 
> within the Windows tree such as %SYSTEMROOT%\System32\Spambayesconfig 
> or similar. 

OK then, so what about I change pop3proxy_service.py to (when run with
"install") look for an existing configuration file and if there isn't one
create one in the parent directory of the script.  This would mean:

  * Binary: in the (eg) \Program Files\SpamBayes directory, so the cache
directories would be at the same level as the 'bin', 'lib', and 'doc'
directories.

  * Source (outside Python): in the root spambayes directory, so the cache
directories would be at the same level as the 'contrib', 'spambayes',
'windows' (etc) directories.

  * Source (inside Python): in the root Python directory, so the cache
directories would be at the same level as the 'Lib', 'DLLS', 'Scripts' (etc)
directories.

The directory that is chosen gets:
  
  * the bayescustomize.ini file
  * the hammie.db database
  * the spambayes.messageinfo.db database
  * the pop3proxy-ham-cache, pop3proxy-spam-cache, pop3proxy-unknown-cache
directories
(IIRC, that's all at the moment).

Alternatively, I could get it to create a (eg) "SpamBayesData" directory in
the parent directory of the script.  It would then be in the places
described above, but have the three files and three directories inside it.
That's tidier, although slightly more complicated.

> I'd also suggest that log files should not be sent to temp 
> directories which get scavenged by 2003, and that the WMI interface
> be used to send warnings etc to the Windows Application Event log.

This is do-able as well.  I'll try and create something that does this later
today.  (As an aside: it would be nice in many ways if all the spambayes
code was converted to use the Python logging module rather than print
statements (the Outlook plug-in actually has it's own logging system), and
things like this would be much simpler and easier to change as necessary.
That's a fair bit of work, though - maybe it'll make it to 1.1, but maybe it
won't).

> Hope this begins to clarify some of the issues.

It does indeed, thanks :)

=Tony.Meyer

-- 
Please always include the list (spambayes@python.org) in your replies
(reply-all), and please don't send me personal mail about SpamBayes.
http://www.massey.ac.nz/~tameyer/writing/reply_all.html explains this.

From robert at imagine.co.uk  Thu Nov 11 01:54:51 2004
From: robert at imagine.co.uk (Robert Neuschul)
Date: Thu Nov 11 01:54:53 2004
Subject: [Spambayes] A most peculair problem with SB 1
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E8FBC9EF@its-xchg4.massey.ac.nz>
References: <ECBA357DDED63B4995F5C1F5CBE5B1E8FBC9EF@its-xchg4.massey.ac.nz>
Message-ID: <VA.00005754.02ffa5e2@imagine.co.uk>

Tony

OK to other stuff:

> Further to this - the error that appears in the Event Log is:
> 
> """
> Python could not import the service's module 
>   File "C:\spambayes\windows\pop3proxy_service.py", line 82, in ?
>     import sb_server
>   File "C:\spambayes\scripts\sb_server.py", line 115, in ?
>     from spambayes.FileCorpus import FileCorpus, ExpiryFileCorpus
>   File "C:\spambayes\spambayes\FileCorpus.py", line 92, in ?
>     import sys, os, gzip, fnmatch, getopt, time, stat
>   File "C:\Python24\Lib\gzip.py", line 9, in ?
>     import zlib 
> exceptions.ImportError: dynamic module does not define init function
> (initzlib)
> """

Which - apart from the exceptions line at the end seems to me to be similar to 
my original error posting.

> If you get the same error, it would at least explain some things...

Running in debug mode there are two major differences:

1] SB works correctly, most of the time: I can run the config page locally or 
remotely without having to run sb_server.py from a cmd session at the same 
time. When it is running correctly it doesn't reurn any errors.

2] I can send/receive email.

What is possibly wrong [i.e. sometimes doesn't work] is that 

a] SB sometimes throws an exception in Python, and 

b] SB doesn't appear to be recording any debug or other logs anywhere.

I trapped the exception and it's listed below.

C:\Python23\Scripts>pop3proxy_service.py debug
Debugging service pop3proxy - press Ctrl+C to stop.
Loading database...Info 0x40001002 - The pop3proxy service has started as user 
'Administrator', using config file 'C:\Documents and 
Settings\Administrator\Application Data\SpamBayes\Proxy\bayescustomize.ini'.
 Listener on port 1100 is proxying my.domain.mx.server:110
User interface url is http://localhost:8880/
error: uncaptured python exception, closing channel <sb_server.BayesProxy 
connected 192.168.0.50:3142 at 0xdb5ad0> (socket.error:(10053, 'Software 
caused connection abort') [C:\Python23\lib\asynchat.py|handle_read|88] 
[C:\Python23\Scripts\sb_server.py|recv|397] 
[C:\Python23\lib\asyncore.py|recv|356])

As soon as I stop the debug mode and restart the proxy in ordinary 'services' 
mode we're back to square one: no comms and no access via browser to 
home/config/message review etc. It makes no difference whether I start the 
service from the MMC console or from a command line instruction.

Now whether or not SB *is* actually running as a service or whether the 2003 
services console is lying to me and telling me that the service is running 
when it isn't I haven't yet determined: the next logical step was to examine 
the task manager processes console - and that clearly shows that 
pythonservices.exe is running - but I cannot deduce from that if SB itself is 
functional as a service.

Hope this is helping

RObert.




From robert at imagine.co.uk  Thu Nov 11 02:19:33 2004
From: robert at imagine.co.uk (Robert Neuschul)
Date: Thu Nov 11 02:19:35 2004
Subject: [Spambayes] A most peculair problem with SB 1
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E86C37DB@its-xchg4.massey.ac.nz>
References: <ECBA357DDED63B4995F5C1F5CBE5B1E86C37DB@its-xchg4.massey.ac.nz>
Message-ID: <VA.00005755.031654b5@imagine.co.uk>

Tony

> OK then, so what about I change pop3proxy_service.py to (when run with
> "install") look for an existing configuration file and if there isn't one
> create one in the parent directory of the script.  This would mean:
> 
>   * Binary: in the (eg) \Program Files\SpamBayes directory, so the cache
> directories would be at the same level as the 'bin', 'lib', and 'doc'
> directories.
> 
>   * Source (outside Python): in the root spambayes directory, so the cache
> directories would be at the same level as the 'contrib', 'spambayes',
> 'windows' (etc) directories.
> 
>   * Source (inside Python): in the root Python directory, so the cache
> directories would be at the same level as the 'Lib', 'DLLS', 'Scripts' (etc)
> directories.

Any of these /should/ work: we'll only know by testing :-)
If necessary I'll setup a dedicated sandpit box for this and we can find out 
how it handles - I'm nervous of playing too much more with the existing box 
since it's actually the domain primary :-)

> The directory that is chosen gets:
>   
>   * the bayescustomize.ini file
>   * the hammie.db database
>   * the spambayes.messageinfo.db database
>   * the pop3proxy-ham-cache, pop3proxy-spam-cache, pop3proxy-unknown-cache
> directories
> (IIRC, that's all at the moment).
> 
> Alternatively, I could get it to create a (eg) "SpamBayesData" directory in
> the parent directory of the script.  It would then be in the places
> described above, but have the three files and three directories inside it.
> That's tidier, although slightly more complicated.

Again; these sound like eminently sane proposals: my concern would be that 
whatever directory name is chosen and used makes it blatantly clear to 
inexperienced first-time installers from reading the directory name that 
"this" is where one looks for the config and databases etc. It took me, as a 
relatively experienced user of Windows systems, some time to figure out where 
everything was - the subdirectory names aren't always instantly intuitive.

> This is do-able as well.  I'll try and create something that does this later
> today.  (As an aside: it would be nice in many ways if all the spambayes
> code was converted to use the Python logging module rather than print
> statements (the Outlook plug-in actually has it's own logging system), and
> things like this would be much simpler and easier to change as necessary.
> That's a fair bit of work, though - maybe it'll make it to 1.1, but maybe it
> won't).

I can sympathise with all of that: Windows logging is a nightmare at the best 
of times - each of the commercial application developers seems to think 
they've found a new and better way: it can be quite stunning to run a global 
search on *.log and find out just how much unread stuff has accumulated on the 
system. At least with WMI one has the option to place logs into the core 
repository - which most experienced users /do/ know about - and my memory of 
the Python logging is that it can use Windows Instrumentation fairly 
effectively. 

ATB,

Robert.


From tameyer at ihug.co.nz  Thu Nov 11 02:36:05 2004
From: tameyer at ihug.co.nz (Tony Meyer)
Date: Thu Nov 11 02:36:41 2004
Subject: [Spambayes] A most peculair problem with SB 1
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E801521A9D@its-xchg4.massey.ac.nz>
Message-ID: <ECBA357DDED63B4995F5C1F5CBE5B1E8FBC9FC@its-xchg4.massey.ac.nz>

>> Further to this - the error that appears in the Event Log is:
[...]
>> exceptions.ImportError: dynamic module does not define init function
>> (initzlib)
> 
> Which - apart from the exceptions line at the end seems to me 
> to be similar to my original error posting.

This one I have managed to (just) fix, thanks to some help from Mark.  The
problem is that I have a zlib.dll in my system32 directory, which is ahead
of the Python DLLs directory in sys.path, and so is found first (and is not
the same DLL).  Mark hinted that he might fix up pywin32 to handle this
better, but for the moment having pop3proxy_service.py do a reorganisation
of sys.path fixes the problem.

If you want to try it, you could put this after the "import sys, os" line of
pop3proxy_service.py:

"""
for dir in sys.path[:-1]:
    if dir.lower().endswith("windows\\system32"):
        sys.path.remove(dir)
        assert dir not in sys.path, \
               "Please remove multiple copies of windows\system32 in path"
        sys.path.append(dir)
"""

> What is possibly wrong [i.e. sometimes doesn't work] is that 
[...]
> b] SB doesn't appear to be recording any debug or other logs anywhere.

I've written up a Handler to redirect messages into the event log, to avoid
this.  A patch (also include the above) is attached, if you want to try it
out.  (The patch includes a hack to avoid win32trace even if running from
source).

[The patch requires Python 2.3 or higher at the moment, but your traceback
indicated that's what you're using anyway.]

The other stuff should be going to win32trace.  If you open PythonWin and do
Tools->Trace Debugger Collecting Tool, it should appear there.  I'm not sure
how else one gets hold of it.

> I trapped the exception and it's listed below.
[...]
> error: uncaptured python exception, closing channel 
> <sb_server.BayesProxy 
> connected 192.168.0.50:3142 at 0xdb5ad0> 
> (socket.error:(10053, 'Software 
> caused connection abort') 
> [C:\Python23\lib\asynchat.py|handle_read|88] 
> [C:\Python23\Scripts\sb_server.py|recv|397] 
> [C:\Python23\lib\asyncore.py|recv|356])

I hate this error - it gets thrown up when all sorts of things happen.  The
most likely one is that the socket that was connected died unexpectedly, but
I don't see how that would be the case here.

If you can, try the patch to get it working outside of debug mode, and see
if the same error occurs (it ought to correctly go to the event log, even
without the other patch since it's an exception).

> Hope this is helping

Yup, thanks :)

=Tony.Meyer

-- 
Please always include the list (spambayes@python.org) in your replies
(reply-all), and please don't send me personal mail about SpamBayes.
http://www.massey.ac.nz/~tameyer/writing/reply_all.html explains this.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: pop3proxy_service.py.diff
Type: application/octet-stream
Size: 5800 bytes
Desc: not available
Url : http://mail.python.org/pipermail/spambayes/attachments/20041111/f681f384/pop3proxy_service.py-0001.obj
From jhockin at cox.net  Thu Nov 11 12:24:52 2004
From: jhockin at cox.net (Jim Hockin)
Date: Thu Nov 11 12:29:20 2004
Subject: [Spambayes] How does SpamBayes training count "spam" and "good"
	mail
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E86C37D7@its-xchg4.massey.ac.nz>
Message-ID: <MOEKKNJAECPFAMBOJJPMOEHHCBAA.jhockin@cox.net>

Thanks for the good advice, Tony.

The "Scanpst" utility said the pst file was just fine.
The SpamBayes log said only nine, and 91 "new" messages were found...

GetNextPage with current 0 IDD_WIZARD_WELCOME
GetNextPage with current 3 IDD_WIZARD_FOLDERS_TRAIN
Checked 9 in folder Inbox - 9 new entries found.
Checked 93 in folder SPAM - 91 new entries found.

...When in fact there were about 300 messages in each folder.

Low and behold, it appears the Internet headers had been wiped on almost all
of my messages.
One of the other anti-spam programs I was testing could have done this.
It could have been SpamProxy Pro.

Thanks, again for the help.

jph



-----Original Message-----
From: Tony Meyer [mailto:tameyer@ihug.co.nz]
Sent: Wednesday, November 10, 2004 1:35 PM
To: 'Jim Hockin'; spambayes@python.org
Subject: RE: [Spambayes] How does SpamBayes training count "spam" and
"good" mail


> Anyone know how SpamBayes training counts "spam" and
> "good" mail? In the SpamBayes 1.0 for Outlook, I designated
> initial training to look at about 300 "spam" messages
> in a SPAM folder and about 300 "good" messages in a GOOD
> folder, checkmarked it to rebuild entire database, and it
> comes back with "Completed training with 91 spam and 9 good
> messages" and the Training Database Status tells me the
> "Database only has 9 good and 91 spam - you should consider
> performing additional training." Can anyone tell me why it
> doesn't see the other 500 messages I gave it? What am I missing
> here?

The log file <0.5 wink>.  Seriously, that'll explain why messages were not
trained.  Wild guesses:

  1.  Some of the mail has never been received (drafts, sent items, ...)
  2.  Some of the mail are not what Outlook calls IPM.Note (regular
messages), for example some Exchange autogenerated messages.
  3.  There was a periodic problem accessing the folder (i.e. run scanpst
over it if it's a PST, or check the connectivity if it's
IMAP/Hotmail/Exchange).

If these don't apply, please feel free to send in a copy of the log file for
the period in which you tried this, and we'll try and explain whatever
errors are there.

=Tony.Meyer

--
Please always include the list (spambayes@python.org) in your replies
(reply-all), and please don't send me personal mail about SpamBayes.
http://www.massey.ac.nz/~tameyer/writing/reply_all.html explains this.



From chanmo2 at bellsouth.net  Thu Nov 11 13:50:19 2004
From: chanmo2 at bellsouth.net (ch'an mo)
Date: Thu Nov 11 13:50:36 2004
Subject: [Spambayes] How does SpamBayes training count "spam" and
	"good"mail
In-Reply-To: <MOEKKNJAECPFAMBOJJPMOEHHCBAA.jhockin@cox.net>
Message-ID: <000601c4c7ed$047cc980$0e624ed8@system>



ch'an mo

-----Original Message-----
From: spambayes-bounces@python.org [mailto:spambayes-bounces@python.org] On
Behalf Of Jim Hockin
Sent: Thursday, November 11, 2004 6:25 AM
To: Tony Meyer; spambayes@python.org
Subject: RE: [Spambayes] How does SpamBayes training count "spam" and
"good"mail

Thanks for the good advice, Tony.

The "Scanpst" utility said the pst file was just fine.
The SpamBayes log said only nine, and 91 "new" messages were found...

GetNextPage with current 0 IDD_WIZARD_WELCOME
GetNextPage with current 3 IDD_WIZARD_FOLDERS_TRAIN
Checked 9 in folder Inbox - 9 new entries found.
Checked 93 in folder SPAM - 91 new entries found.

...When in fact there were about 300 messages in each folder.

Low and behold, it appears the Internet headers had been wiped on almost all
of my messages.
One of the other anti-spam programs I was testing could have done this.
It could have been SpamProxy Pro.

Thanks, again for the help.

jph



-----Original Message-----
From: Tony Meyer [mailto:tameyer@ihug.co.nz]
Sent: Wednesday, November 10, 2004 1:35 PM
To: 'Jim Hockin'; spambayes@python.org
Subject: RE: [Spambayes] How does SpamBayes training count "spam" and
"good" mail


> Anyone know how SpamBayes training counts "spam" and
> "good" mail? In the SpamBayes 1.0 for Outlook, I designated
> initial training to look at about 300 "spam" messages
> in a SPAM folder and about 300 "good" messages in a GOOD
> folder, checkmarked it to rebuild entire database, and it
> comes back with "Completed training with 91 spam and 9 good
> messages" and the Training Database Status tells me the
> "Database only has 9 good and 91 spam - you should consider
> performing additional training." Can anyone tell me why it
> doesn't see the other 500 messages I gave it? What am I missing
> here?

The log file <0.5 wink>.  Seriously, that'll explain why messages were not
trained.  Wild guesses:

  1.  Some of the mail has never been received (drafts, sent items, ...)
  2.  Some of the mail are not what Outlook calls IPM.Note (regular
messages), for example some Exchange autogenerated messages.
  3.  There was a periodic problem accessing the folder (i.e. run scanpst
over it if it's a PST, or check the connectivity if it's
IMAP/Hotmail/Exchange).

If these don't apply, please feel free to send in a copy of the log file for
the period in which you tried this, and we'll try and explain whatever
errors are there.

=Tony.Meyer

--
Please always include the list (spambayes@python.org) in your replies
(reply-all), and please don't send me personal mail about SpamBayes.
http://www.massey.ac.nz/~tameyer/writing/reply_all.html explains this.



_______________________________________________
Spambayes@python.org
http://mail.python.org/mailman/listinfo/spambayes
Check the FAQ before asking: http://spambayes.sf.net/faq.html


From chanmo2 at bellsouth.net  Thu Nov 11 14:02:42 2004
From: chanmo2 at bellsouth.net (ch'an mo)
Date: Thu Nov 11 14:02:59 2004
Subject: [Spambayes] How does SpamBayes training count "spam" and
	"good"mail
Message-ID: <000701c4c7ee$be8caa10$0e624ed8@system>

Please help me. I started with Spam Bayes v0.8 and it got me up graded up to
2.3.4 Then that wouldn't come onto Outlook, so I deleted v0.8 and still
can't get 2.3.4, what do I do now? I have the bar still there (of course it
doesn't work as I deleted v0.8). Now what? I would like to continue with
Spam Bayes.  

ch'an mo

-----Original Message-----
From: spambayes-bounces@python.org [mailto:spambayes-bounces@python.org] On
Behalf Of Jim Hockin
Sent: Thursday, November 11, 2004 6:25 AM
To: Tony Meyer; spambayes@python.org
Subject: RE: [Spambayes] How does SpamBayes training count "spam" and
"good"mail

Thanks for the good advice, Tony.

The "Scanpst" utility said the pst file was just fine.
The SpamBayes log said only nine, and 91 "new" messages were found...

GetNextPage with current 0 IDD_WIZARD_WELCOME
GetNextPage with current 3 IDD_WIZARD_FOLDERS_TRAIN
Checked 9 in folder Inbox - 9 new entries found.
Checked 93 in folder SPAM - 91 new entries found.

...When in fact there were about 300 messages in each folder.

Low and behold, it appears the Internet headers had been wiped on almost all
of my messages.
One of the other anti-spam programs I was testing could have done this.
It could have been SpamProxy Pro.

Thanks, again for the help.

jph



-----Original Message-----
From: Tony Meyer [mailto:tameyer@ihug.co.nz]
Sent: Wednesday, November 10, 2004 1:35 PM
To: 'Jim Hockin'; spambayes@python.org
Subject: RE: [Spambayes] How does SpamBayes training count "spam" and
"good" mail


> Anyone know how SpamBayes training counts "spam" and
> "good" mail? In the SpamBayes 1.0 for Outlook, I designated
> initial training to look at about 300 "spam" messages
> in a SPAM folder and about 300 "good" messages in a GOOD
> folder, checkmarked it to rebuild entire database, and it
> comes back with "Completed training with 91 spam and 9 good
> messages" and the Training Database Status tells me the
> "Database only has 9 good and 91 spam - you should consider
> performing additional training." Can anyone tell me why it
> doesn't see the other 500 messages I gave it? What am I missing
> here?

The log file <0.5 wink>.  Seriously, that'll explain why messages were not
trained.  Wild guesses:

  1.  Some of the mail has never been received (drafts, sent items, ...)
  2.  Some of the mail are not what Outlook calls IPM.Note (regular
messages), for example some Exchange autogenerated messages.
  3.  There was a periodic problem accessing the folder (i.e. run scanpst
over it if it's a PST, or check the connectivity if it's
IMAP/Hotmail/Exchange).

If these don't apply, please feel free to send in a copy of the log file for
the period in which you tried this, and we'll try and explain whatever
errors are there.

=Tony.Meyer

--
Please always include the list (spambayes@python.org) in your replies
(reply-all), and please don't send me personal mail about SpamBayes.
http://www.massey.ac.nz/~tameyer/writing/reply_all.html explains this.



_______________________________________________
Spambayes@python.org
http://mail.python.org/mailman/listinfo/spambayes
Check the FAQ before asking: http://spambayes.sf.net/faq.html


From RRasnake at aintl.com  Thu Nov 11 14:41:39 2004
From: RRasnake at aintl.com (Roger Rasnake)
Date: Thu Nov 11 14:41:45 2004
Subject: [Spambayes] No "Delete as Spam" button in Junk Suspects
Message-ID: <F0A2E20726E77445BB749EA67AA8D4FB01F80AFC@aib-ex1.aiweb.com>

Since I updated recently to the 1.0 version of Spambayes, I no longer
have a "delete as spam" button in the Junk Suspects folder, only
"Recover from Spam."  Before, of course, both buttons appeared.  I can,
of course, just delete them; but since I've assumed that "deleting as
spam" added those to the training database, that possibility is now
lost.  I tried reinstalling Spambayes, in case it was perhaps a problem
on my computer, but there was no change.  Any thoughts?  Or is this a
change in the program?
 
Thanks,
 
Roger
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.python.org/pipermail/spambayes/attachments/20041111/51b22717/attachment.htm
From Robert.Mezzone at PJSolomon.Com  Thu Nov 11 14:51:55 2004
From: Robert.Mezzone at PJSolomon.Com (Robert Mezzone)
Date: Thu Nov 11 14:52:04 2004
Subject: [Spambayes] No "Delete as Spam" button in Junk Suspects
Message-ID: <1254A68F4903D411B24800508B1220E90629F62D@solomon1.pjsc.internal>

I don't know how to restore the button, but in the meantime you can drag the
messages to you Junk Email folder and the program will train on the
messages. At least that's my experience moving messages thru OWA and my
Blacberry. I'm sure Tony can resolve your missing button issue.

hth.

Robert


-----Original Message-----
From: spambayes-bounces@python.org <spambayes-bounces@python.org>
To: spambayes@python.org <spambayes@python.org>
Sent: Thu Nov 11 08:41:39 2004
Subject: [Spambayes] No "Delete as Spam" button in Junk Suspects

Since I updated recently to the 1.0 version of Spambayes, I no longer have a
"delete as spam" button in the Junk Suspects folder, only "Recover from
Spam."  Before, of course, both buttons appeared.  I can, of course, just
delete them; but since I've assumed that "deleting as spam" added those to
the training database, that possibility is now lost.  I tried reinstalling
Spambayes, in case it was perhaps a problem on my computer, but there was no
change.  Any thoughts?  Or is this a change in the program?
 
Thanks,
 
Roger
From chanmo2 at bellsouth.net  Thu Nov 11 14:58:19 2004
From: chanmo2 at bellsouth.net (ch'an mo)
Date: Thu Nov 11 14:58:35 2004
Subject: [Spambayes] How does SpamBayes training count "spam" and
	"good"mail
Message-ID: <000501c4c7f6$83c22ec0$0e624ed8@system>

Please help me. I lost Span Bayes

ch'an mo

-----Original Message-----
From: spambayes-bounces@python.org [mailto:spambayes-bounces@python.org] On
Behalf Of Jim Hockin
Sent: Thursday, November 11, 2004 6:25 AM
To: Tony Meyer; spambayes@python.org
Subject: RE: [Spambayes] How does SpamBayes training count "spam" and
"good"mail

Thanks for the good advice, Tony.

The "Scanpst" utility said the pst file was just fine.
The SpamBayes log said only nine, and 91 "new" messages were found...

GetNextPage with current 0 IDD_WIZARD_WELCOME
GetNextPage with current 3 IDD_WIZARD_FOLDERS_TRAIN
Checked 9 in folder Inbox - 9 new entries found.
Checked 93 in folder SPAM - 91 new entries found.

...When in fact there were about 300 messages in each folder.

Low and behold, it appears the Internet headers had been wiped on almost all
of my messages.
One of the other anti-spam programs I was testing could have done this.
It could have been SpamProxy Pro.

Thanks, again for the help.

jph



-----Original Message-----
From: Tony Meyer [mailto:tameyer@ihug.co.nz]
Sent: Wednesday, November 10, 2004 1:35 PM
To: 'Jim Hockin'; spambayes@python.org
Subject: RE: [Spambayes] How does SpamBayes training count "spam" and
"good" mail


> Anyone know how SpamBayes training counts "spam" and
> "good" mail? In the SpamBayes 1.0 for Outlook, I designated
> initial training to look at about 300 "spam" messages
> in a SPAM folder and about 300 "good" messages in a GOOD
> folder, checkmarked it to rebuild entire database, and it
> comes back with "Completed training with 91 spam and 9 good
> messages" and the Training Database Status tells me the
> "Database only has 9 good and 91 spam - you should consider
> performing additional training." Can anyone tell me why it
> doesn't see the other 500 messages I gave it? What am I missing
> here?

The log file <0.5 wink>.  Seriously, that'll explain why messages were not
trained.  Wild guesses:

  1.  Some of the mail has never been received (drafts, sent items, ...)
  2.  Some of the mail are not what Outlook calls IPM.Note (regular
messages), for example some Exchange autogenerated messages.
  3.  There was a periodic problem accessing the folder (i.e. run scanpst
over it if it's a PST, or check the connectivity if it's
IMAP/Hotmail/Exchange).

If these don't apply, please feel free to send in a copy of the log file for
the period in which you tried this, and we'll try and explain whatever
errors are there.

=Tony.Meyer

--
Please always include the list (spambayes@python.org) in your replies
(reply-all), and please don't send me personal mail about SpamBayes.
http://www.massey.ac.nz/~tameyer/writing/reply_all.html explains this.



_______________________________________________
Spambayes@python.org
http://mail.python.org/mailman/listinfo/spambayes
Check the FAQ before asking: http://spambayes.sf.net/faq.html


From RRasnake at aintl.com  Thu Nov 11 14:59:14 2004
From: RRasnake at aintl.com (Roger Rasnake)
Date: Thu Nov 11 14:59:18 2004
Subject: [Spambayes] No "Delete as Spam" button in Junk Suspects
Message-ID: <F0A2E20726E77445BB749EA67AA8D4FB01F80B02@aib-ex1.aiweb.com>

Dear Robert,

I also just read the troubleshooting guide as politely instructed in the
automated response from spambayes (I forgot that that was available) and
followed the instructions on deleting the toolbar.  That solved the
problem!

R. 

-----Original Message-----
From: Robert Mezzone [mailto:Robert.Mezzone@PJSolomon.Com] 
Sent: Thursday, November 11, 2004 8:52 AM
To: Roger Rasnake; 'spambayes@python.org'
Subject: Re: [Spambayes] No "Delete as Spam" button in Junk Suspects

I don't know how to restore the button, but in the meantime you can drag
the messages to you Junk Email folder and the program will train on the
messages. At least that's my experience moving messages thru OWA and my
Blacberry. I'm sure Tony can resolve your missing button issue.

hth.

Robert


-----Original Message-----
From: spambayes-bounces@python.org <spambayes-bounces@python.org>
To: spambayes@python.org <spambayes@python.org>
Sent: Thu Nov 11 08:41:39 2004
Subject: [Spambayes] No "Delete as Spam" button in Junk Suspects

Since I updated recently to the 1.0 version of Spambayes, I no longer
have a "delete as spam" button in the Junk Suspects folder, only
"Recover from Spam."  Before, of course, both buttons appeared.  I can,
of course, just delete them; but since I've assumed that "deleting as
spam" added those to the training database, that possibility is now
lost.  I tried reinstalling Spambayes, in case it was perhaps a problem
on my computer, but there was no change.  Any thoughts?  Or is this a
change in the program?
 
Thanks,
 
Roger


From Robert.Mezzone at PJSolomon.Com  Thu Nov 11 15:10:24 2004
From: Robert.Mezzone at PJSolomon.Com (Robert Mezzone)
Date: Thu Nov 11 15:10:27 2004
Subject: [Spambayes] How does SpamBayes training count "spam" and "goo
	d"mail
Message-ID: <1254A68F4903D411B24800508B1220E90629F62F@solomon1.pjsc.internal>

Not sure what you mean by lost. Try this, assuming you are using Outlook.

Help, About Microsoft Outlook. Click the disabled items folder. Is SpamBayes
listed here, if so enable it.

Robert


-----Original Message-----
From: spambayes-bounces@python.org <spambayes-bounces@python.org>
To: Spambayes@python.org <Spambayes@python.org>
Sent: Thu Nov 11 08:58:19 2004
Subject: RE: [Spambayes] How does SpamBayes training count "spam" and
"good"mail

Please help me. I lost Span Bayes

ch'an mo

-----Original Message-----
From: spambayes-bounces@python.org [mailto:spambayes-bounces@python.org] On
Behalf Of Jim Hockin
Sent: Thursday, November 11, 2004 6:25 AM
To: Tony Meyer; spambayes@python.org
Subject: RE: [Spambayes] How does SpamBayes training count "spam" and
"good"mail

Thanks for the good advice, Tony.

The "Scanpst" utility said the pst file was just fine.
The SpamBayes log said only nine, and 91 "new" messages were found...

GetNextPage with current 0 IDD_WIZARD_WELCOME
GetNextPage with current 3 IDD_WIZARD_FOLDERS_TRAIN
Checked 9 in folder Inbox - 9 new entries found.
Checked 93 in folder SPAM - 91 new entries found.

...When in fact there were about 300 messages in each folder.

Low and behold, it appears the Internet headers had been wiped on almost all
of my messages.
One of the other anti-spam programs I was testing could have done this.
It could have been SpamProxy Pro.

Thanks, again for the help.

jph



-----Original Message-----
From: Tony Meyer [mailto:tameyer@ihug.co.nz]
Sent: Wednesday, November 10, 2004 1:35 PM
To: 'Jim Hockin'; spambayes@python.org
Subject: RE: [Spambayes] How does SpamBayes training count "spam" and
"good" mail


> Anyone know how SpamBayes training counts "spam" and
> "good" mail? In the SpamBayes 1.0 for Outlook, I designated
> initial training to look at about 300 "spam" messages
> in a SPAM folder and about 300 "good" messages in a GOOD
> folder, checkmarked it to rebuild entire database, and it
> comes back with "Completed training with 91 spam and 9 good
> messages" and the Training Database Status tells me the
> "Database only has 9 good and 91 spam - you should consider
> performing additional training." Can anyone tell me why it
> doesn't see the other 500 messages I gave it? What am I missing
> here?

The log file <0.5 wink>.  Seriously, that'll explain why messages were not
trained.  Wild guesses:

  1.  Some of the mail has never been received (drafts, sent items, ...)
  2.  Some of the mail are not what Outlook calls IPM.Note (regular
messages), for example some Exchange autogenerated messages.
  3.  There was a periodic problem accessing the folder (i.e. run scanpst
over it if it's a PST, or check the connectivity if it's
IMAP/Hotmail/Exchange).

If these don't apply, please feel free to send in a copy of the log file for
the period in which you tried this, and we'll try and explain whatever
errors are there.

=Tony.Meyer

--
Please always include the list (spambayes@python.org) in your replies
(reply-all), and please don't send me personal mail about SpamBayes.
http://www.massey.ac.nz/~tameyer/writing/reply_all.html explains this.



_______________________________________________
Spambayes@python.org
http://mail.python.org/mailman/listinfo/spambayes
Check the FAQ before asking: http://spambayes.sf.net/faq.html


_______________________________________________
Spambayes@python.org
http://mail.python.org/mailman/listinfo/spambayes
Check the FAQ before asking: http://spambayes.sf.net/faq.html
From Robert.Mezzone at PJSolomon.Com  Thu Nov 11 15:11:50 2004
From: Robert.Mezzone at PJSolomon.Com (Robert Mezzone)
Date: Thu Nov 11 15:11:59 2004
Subject: [Spambayes] How does SpamBayes training count "spam" and "goo
	d"mail
Message-ID: <1254A68F4903D411B24800508B1220E90629F630@solomon1.pjsc.internal>

I meant to say "Disabled Items" button. Sorry about that.

Robert


-----Original Message-----
From: Robert Mezzone <Robert.Mezzone@PJSolomon.Com>
To: 'chanmo2@bellsouth.net' <chanmo2@bellsouth.net>; 'spambayes@python.org'
<spambayes@python.org>
Sent: Thu Nov 11 09:10:24 2004
Subject: Re: [Spambayes] How does SpamBayes training count "spam" and
"good"mail

Not sure what you mean by lost. Try this, assuming you are using Outlook.

Help, About Microsoft Outlook. Click the disabled items folder. Is SpamBayes
listed here, if so enable it.

Robert


-----Original Message-----
From: spambayes-bounces@python.org <spambayes-bounces@python.org>
To: Spambayes@python.org <Spambayes@python.org>
Sent: Thu Nov 11 08:58:19 2004
Subject: RE: [Spambayes] How does SpamBayes training count "spam" and
"good"mail

Please help me. I lost Span Bayes

ch'an mo

-----Original Message-----
From: spambayes-bounces@python.org [mailto:spambayes-bounces@python.org] On
Behalf Of Jim Hockin
Sent: Thursday, November 11, 2004 6:25 AM
To: Tony Meyer; spambayes@python.org
Subject: RE: [Spambayes] How does SpamBayes training count "spam" and
"good"mail

Thanks for the good advice, Tony.

The "Scanpst" utility said the pst file was just fine.
The SpamBayes log said only nine, and 91 "new" messages were found...

GetNextPage with current 0 IDD_WIZARD_WELCOME
GetNextPage with current 3 IDD_WIZARD_FOLDERS_TRAIN
Checked 9 in folder Inbox - 9 new entries found.
Checked 93 in folder SPAM - 91 new entries found.

...When in fact there were about 300 messages in each folder.

Low and behold, it appears the Internet headers had been wiped on almost all
of my messages.
One of the other anti-spam programs I was testing could have done this.
It could have been SpamProxy Pro.

Thanks, again for the help.

jph



-----Original Message-----
From: Tony Meyer [mailto:tameyer@ihug.co.nz]
Sent: Wednesday, November 10, 2004 1:35 PM
To: 'Jim Hockin'; spambayes@python.org
Subject: RE: [Spambayes] How does SpamBayes training count "spam" and
"good" mail


> Anyone know how SpamBayes training counts "spam" and
> "good" mail? In the SpamBayes 1.0 for Outlook, I designated
> initial training to look at about 300 "spam" messages
> in a SPAM folder and about 300 "good" messages in a GOOD
> folder, checkmarked it to rebuild entire database, and it
> comes back with "Completed training with 91 spam and 9 good
> messages" and the Training Database Status tells me the
> "Database only has 9 good and 91 spam - you should consider
> performing additional training." Can anyone tell me why it
> doesn't see the other 500 messages I gave it? What am I missing
> here?

The log file <0.5 wink>.  Seriously, that'll explain why messages were not
trained.  Wild guesses:

  1.  Some of the mail has never been received (drafts, sent items, ...)
  2.  Some of the mail are not what Outlook calls IPM.Note (regular
messages), for example some Exchange autogenerated messages.
  3.  There was a periodic problem accessing the folder (i.e. run scanpst
over it if it's a PST, or check the connectivity if it's
IMAP/Hotmail/Exchange).

If these don't apply, please feel free to send in a copy of the log file for
the period in which you tried this, and we'll try and explain whatever
errors are there.

=Tony.Meyer

--
Please always include the list (spambayes@python.org) in your replies
(reply-all), and please don't send me personal mail about SpamBayes.
http://www.massey.ac.nz/~tameyer/writing/reply_all.html explains this.



_______________________________________________
Spambayes@python.org
http://mail.python.org/mailman/listinfo/spambayes
Check the FAQ before asking: http://spambayes.sf.net/faq.html


_______________________________________________
Spambayes@python.org
http://mail.python.org/mailman/listinfo/spambayes
Check the FAQ before asking: http://spambayes.sf.net/faq.html
From robert at imagine.co.uk  Thu Nov 11 19:11:39 2004
From: robert at imagine.co.uk (Robert Neuschul)
Date: Thu Nov 11 19:11:41 2004
Subject: [Spambayes] A most peculair problem with SB 1
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E8FBC9FC@its-xchg4.massey.ac.nz>
References: <ECBA357DDED63B4995F5C1F5CBE5B1E8FBC9FC@its-xchg4.massey.ac.nz>
Message-ID: <VA.00005757.019db964@imagine.co.uk>

Tony

Many thanks for that: I'm afraid I need a bit of hand-holding here.

Namely some nursery school instructions on how to apply the two 
changes: do I simply open the scripts in an editor and cut/paste the 
changes into place, or am I supposed to do something else? If so then 
what exactly. Is there anything else I need to do afterwards?

Robert.


From AOL at AOL.com  Thu Nov 11 23:39:03 2004
From: AOL at AOL.com (AOL Security)
Date: Fri Nov 12 00:48:16 2004
Subject: [Spambayes] Suspension Notice Please Re-enter Details
Message-ID: <20041111234815.42A351E4003@bag.python.org>

An HTML attachment was scrubbed...
URL: http://mail.python.org/pipermail/spambayes/attachments/20041111/0cc3ab18/attachment.htm
From tameyer at ihug.co.nz  Fri Nov 12 01:04:14 2004
From: tameyer at ihug.co.nz (Tony Meyer)
Date: Fri Nov 12 01:06:36 2004
Subject: [Spambayes] A most peculair problem with SB 1
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E801622CED@its-xchg4.massey.ac.nz>
Message-ID: <ECBA357DDED63B4995F5C1F5CBE5B1E8FBCA05@its-xchg4.massey.ac.nz>

> Namely some nursery school instructions on how to apply the two 
> changes: do I simply open the scripts in an editor and cut/paste the 
> changes into place

Either that, or if there is a 'patch' tool there you can give the diff and
the original to it and it'll do it for you.  Anyway, to save the bother,
attached is a changed version (slightly further along) of the script - just
replace the 1.0 one with this one.

> Is there anything else I need to do afterwards?

Shouldn't be, no.

=Tony.Meyer

-- 
Please always include the list (spambayes@python.org) in your replies
(reply-all), and please don't send me personal mail about SpamBayes.
http://www.massey.ac.nz/~tameyer/writing/reply_all.html explains this.
-------------- next part --------------
# Run the sb_server as a WinNT service.  Should work on Windows 2000
# and Windows XP.
#
# * Install as a service using "pop3proxy_service.py install"
# * Start the service (Use Control Panel etc, or
#   "pop3proxy_service.py start".  Check the event
#   log should anything go wrong.
# * To debug the service: "pop3proxy_service.py debug"
#   Service then runs in the command prompt, showing all
#   print statements.
# * To remove the service: "pop3proxy_service.py remove"

# This module is part of the spambayes project, which is Copyright 2002-4
# The Python Software Foundation and is covered by the Python Software
# Foundation license.

# Originally written by Mark Hammond.

import os
import sys
import logging

import servicemanager

class ServiceEventLogHandler(logging.Handler):
    """Dispatches logging events to the win32 services event log.

    Requires pywin32.    
    """
    import servicemanager
    def emit(self, record):
        """Emit a record.

        If a formatter is specified, it is used to format the record.
        This record is then written to the win32 services event log,
        with the type set to the appropriate type based on the level.
        """
        try:
            servicemgr = self.servicemanager
            level = record.levelno
            msg = self.format(record)
            if level >= logging.ERROR:
                servicemgr.LogErrorMsg(msg)
            elif level >= logging.WARNING:
                servicemgr.LogWarningMsg(msg)
            elif level >= logging.INFO:
                servicemgr.LogInfoMsg(msg)
            elif level >= logging.DEBUG:
                # What should we do with this?  It's very low-level
                # to be going into the log, but then it only gets
                # added if the logger's level is set low enough.
                # For now, nothing (absorb), and reconsider this
                # when we are actually using the logging module properly.
                pass
            else:
                # Really low; just absorb these for now.
                pass
        except:
            self.handleError(record)

    def handleError(self, record):
        """
        Handle errors which occur during an emit() call.

        sys.stderr does nowwhere, so redirect this into the event log, too.
        """
        if raiseExceptions:
            try:
                import cStringIO as StringIO
            except ImportError:
                import StringIO
            import traceback
            ei = sys.exc_info()
            msg = StringIO.StringIO()
            traceback.print_exception(ei[0], ei[1], ei[2], None, msg)
            msg.seek(0)
            self.servicemanager.LogErrorMsg(msg)
            del ei


class ServiceEventLogHandlerWrapper(object):
    """Pretend that the ServiceEventLogHandler is a file-like object,
    so we can use it while we don't use the proper logging module."""
    def __init__(self, service_name, level=logging.INFO):
        self.log = ServiceEventLogHandler()
        self.name = service_name
        self.level = level
        self.data = ""
    def write(self, data):
        # This could use the higher-up stuff, but don't for now.
        # Buffer until newline.
        self.data += data
        if '\n' not in data:
            return
        # Skip blank lines
        if not self.data.strip():
            return
        record = logging.LogRecord(self.name, self.level, "", "",
                                   self.data, None, None)
        self.log.emit(record)
        self.data = ""


# Messages from pop3proxy will go nowhere when executed as a service
# Try and detect that print will go nowhere and redirect.
# redirect output somewhere useful when running as a service.
import win32api
try:
    win32api.GetConsoleTitle()
except win32api.error:
    # No console - if we are running from Python sources,
    # redirect to win32traceutil, but if running from a binary
    # install, redirect to the services event log.
    # We used to redirect to log files (in the temp folder, in
    # the form SpamBayesService%d.log), but that is apparently
    # not necessarily a good place, so we moved to the official
    # location.
    # Want to move to logging module later, so for now, we
    # hack together a simple logging strategy.
    if hasattr(sys, "frozen"):
        sys.stdout = ServiceEventLogHandlerWrapper("pop3proxy")
        sys.stderr = ServiceEventLogHandlerWrapper("pop3proxy",
                                                   logging.ERROR)
    else:
        import win32traceutil

# If running from sources, patch up sys.path
if not hasattr(sys, "frozen"):
    # We are in the 'spambayes\win32' directory.  We
    # need the parent on sys.path, so 'spambayes.spambayes' is a package,
    # and 'pop3proxy' is a module
    try:
        # module imported by service manager, or 2.3 (in which __main__
        # exists, *and* sys.argv[0] is always already absolute)
        this_filename=__file__
    except NameError:
        this_filename = sys.argv[0]
    if not os.path.isabs(sys.argv[0]):
        # Python 2.3 __main__
        # patch up sys.argv, as our cwd will confuse service registration code
        sys.argv[0] = os.path.abspath(sys.argv[0])
        this_filename = sys.argv[0]

    sb_dir = os.path.dirname(os.path.dirname(this_filename))
    sb_scripts_dir = os.path.join(sb_dir,"scripts")

    sys.path.insert(0, sb_dir)
    sys.path.insert(-1, sb_scripts_dir)
    # and change directory here, so pop3proxy uses the default
    # config file etc
    os.chdir(sb_dir)

    # Fix to handle problem if there is a zlib.dll in the SYSTEM32 directory.
    # (The Python DLL directory must come before that in sys.path)
    # This is a bit hackish, but shouldn't do any real harm.
    from win32com.shell import shell, shellcon
    sys32path = shell.SHGetFolderPath(0, shellcon.CSIDL_SYSTEM, 0, 0)
    for path in sys.path[:-1]:
        if path == sys32path:
            sys.path.remove(path)
            assert path not in sys.path, \
                   "Please remove multiple copies of windows\system32 in path"
            sys.path.append(path) # put it at the *end*
    del sys32path
    del shell
    del shellcon
    del path

# Rest of the standard Python modules we use.
import traceback
import threading
import cStringIO

# The spambayes imports we need.
import sb_server

# The win32 specific modules.
import win32serviceutil, win32service
import pywintypes, win32con, winerror
from ntsecuritycon import *

class Service(win32serviceutil.ServiceFramework):
    # The script name was changed to "sb_server" but I'll leave this as pop3proxy
    # overwise people might accidently run two proxies.
    _svc_name_ = "pop3proxy"
    _svc_display_name_ = "SpamBayes Service"
    _svc_deps_ =  ['tcpip'] # We depend on the tcpip service.
    def __init__(self, args):
        win32serviceutil.ServiceFramework.__init__(self, args)
        self.event_stopped = threading.Event()
        self.event_stopping = threading.Event()
        self.thread = None

    def SvcStop(self):
        self.ReportServiceStatus(win32service.SERVICE_STOP_PENDING)
        self.event_stopping.set()
        sb_server.stop()

    def SvcDoRun(self):
        import servicemanager
        # Setup our state etc
        try:
            sb_server.prepare(can_stop=False)
        except sb_server.AlreadyRunningException:
            msg = "The SpamBayes proxy service could not be started, as "\
                  "another SpamBayes server is already running on this machine"
            servicemanager.LogErrorMsg(msg)
            errCode = winerror.ERROR_SERVICE_SPECIFIC_ERROR
            self.ReportServiceStatus(win32service.SERVICE_STOPPED,
                                     win32ExitCode=errCode, svcExitCode=1)
            return
        assert not sb_server.state.launchUI, "Service can't launch a UI"

        # Start the thread running the server.
        thread = threading.Thread(target=self.ServerThread)
        thread.start()

        # Write an event log record - in debug mode we will also
        # see this message printed.
        from spambayes.Options import optionsPathname
        extra = " as user '%s', using config file '%s'" \
                % (win32api.GetUserName(),
                   optionsPathname)
        servicemanager.LogMsg(
            servicemanager.EVENTLOG_INFORMATION_TYPE,
            servicemanager.PYS_SERVICE_STARTED,
            (self._svc_name_, extra)
            )

        try:
            # Thread running - wait for the stopping event.
            self.event_stopping.wait()
            # Either user requested stop, or thread done - wait for it
            # to actually stop, but reporting we are still alive.
            # Wait up to 60 seconds for shutdown before giving up and
            # exiting uncleanly - we wait for current proxy connections
            # to close, but you have to draw the line somewhere.
            for i in range(60):
                self.ReportServiceStatus(win32service.SERVICE_STOP_PENDING)
                self.event_stopped.wait(1)
                if self.event_stopped.isSet():
                    break
                print "The service is still shutting down..."
            else:
                # eeek - we timed out - give up in disgust.
                print "The worker failed to stop - aborting it anyway"
        except KeyboardInterrupt:
            pass

        # Write another event log record.
        s = sb_server.state
        status = " after %d sessions (%d ham, %d spam, %d unsure)" % \
                (s.totalSessions, s.numHams, s.numSpams, s.numUnsure)

        servicemanager.LogMsg(
            servicemanager.EVENTLOG_INFORMATION_TYPE,
            servicemanager.PYS_SERVICE_STOPPED,
            (self._svc_name_, status)
            )

    def ServerThread(self):
        try:
            try:
                sb_server.start()
            except SystemExit:
                # user requested shutdown
                print "pop3proxy service shutting down due to user request"
            except:
                # Otherwise an error we should log.
                ob = cStringIO.StringIO()
                traceback.print_exc(file=ob)

                message = "The pop3proxy service failed with an " \
                          "unexpected error\r\n\r\n" + ob.getvalue()

                # print it too, so any other log we have gets it.
                print message
                # Log an error event to the event log.
                import servicemanager
                servicemanager.LogErrorMsg(message)
        finally:
            self.event_stopping.set()
            self.event_stopped.set()

if __name__=='__main__':
    win32serviceutil.HandleCommandLine(Service)
From tim.peters at gmail.com  Fri Nov 12 04:01:57 2004
From: tim.peters at gmail.com (Tim Peters)
Date: Fri Nov 12 04:02:00 2004
Subject: [Spambayes] Professional Jealousy
Message-ID: <1f7befae041111190158392928@mail.gmail.com>

This is a hoot:

    http://www.send-safe.com/manual/#Test

That's the online manual for a bulk-email *sending* program. 
According to it, they now integrate SpamAssassin into their mailer
software, so that spammers can test their email against SA more
conveniently.  Of course "more technology" is the answer to all social
problems <wink>.
From richie at entrian.com  Fri Nov 12 09:22:20 2004
From: richie at entrian.com (Richie Hindle)
Date: Fri Nov 12 09:22:28 2004
Subject: [Spambayes] Professional Jealousy
In-Reply-To: <1f7befae041111190158392928@mail.gmail.com>
References: <1f7befae041111190158392928@mail.gmail.com>
Message-ID: <cis8p0t46lm2st472s3lkvn7l8l6rhe32a@4ax.com>


[Tim]
> they now integrate SpamAssassin into their mailer
> software, so that spammers can test their email against SA

This is *exactly* the sort of thing that the originators of Free
Software / Open Source had in mind.

-- 
Richie Hindle
richie@entrian.com

From Ralf.Meuser at tk-logistik.com  Fri Nov 12 11:55:12 2004
From: Ralf.Meuser at tk-logistik.com (Ralf Meuser)
Date: Fri Nov 12 11:55:22 2004
Subject: [Spambayes] Spambayes problem
Message-ID: <8E94307D6F01CE438E6F3B02CF3BB64634E914@SERVEUR1.tk-transports.com>

Hi there

 

I'm using Spambayes 1.0 on a Windows server 2003 with termial server.

All user have Outlook 2003 and until now it worked without problem.

This morning on user had a problem. Outlook starts and the Spambayes
toolbar show up but doesn't work.

Incomming mails are not verifiyed.

 

The log tell:

raceback (most recent call last):

  File "<string>", line 1, in ?

  File "boot_com_servers.py", line 62, in DllRegisterServer

  File "win32com\server\register.pyc", line 405, in RegisterClasses

  File "win32com\server\register.pyc", line 177, in RegisterServer

  File "win32com\server\register.pyc", line 32, in _set_string

pywintypes.error: (5, 'RegSetValue', 'Acc\xe8s refus\xe9.')

 

Can somebody help me out?

Thanks in advance

 

Ralf

 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.python.org/pipermail/spambayes/attachments/20041112/98f4399d/attachment.htm
From glenn at muellercorp.com  Fri Nov 12 15:22:11 2004
From: glenn at muellercorp.com (glenn)
Date: Fri Nov 12 14:04:57 2004
Subject: [Spambayes] (no subject)
Message-ID: <MJEKKCMNHGIMNDOFFFALAEODFDAA.glenn@muellercorp.com>

I have downloaded the zipfile but can't install. There seems to be no setup
to be found, please help

Best Regards,

Glenn Mueller
Vice President
Mueller Corporation
530 spring st
East Bridgewater, MA 02333
Phone: 508-583-2800
Fax: 508-378-4744
Email: glenn@muellercorp.com
Website: www.muellercorp.com

From jsp at PKC.com  Fri Nov 12 14:55:35 2004
From: jsp at PKC.com (Jesse Pelton)
Date: Fri Nov 12 14:55:38 2004
Subject: [Spambayes] (no subject)
Message-ID: <9f3cfb15cd4aa369a69f72c60c21940f4194c0d8@vtmail.pkc.com>

What platform are you trying to install on?  SpamBayes is available in
several flavors.  If you want the Outlook plugin (as appears likely from
your message headers), you've downloaded the wrong file.  Go to
http://spambayes.sourceforge.net/windows.html and follow the "download
the installation program" link.  (Note that the Outlook plugin does not
and cannot work with Outlook Express, but other flavors do.)

> -----Original Message-----
> From: spambayes-bounces@python.org 
> [mailto:spambayes-bounces@python.org] On Behalf Of glenn
> Sent: Friday, November 12, 2004 9:22 AM
> To: spambayes@python.org
> Subject: [Spambayes] (no subject)
> 
> I have downloaded the zipfile but can't install. There seems 
> to be no setup
> to be found, please help
> 
> Best Regards,
> 
> Glenn Mueller
> Vice President
> Mueller Corporation
> 530 spring st
> East Bridgewater, MA 02333
> Phone: 508-583-2800
> Fax: 508-378-4744
> Email: glenn@muellercorp.com
> Website: www.muellercorp.com
From betamagalhaes34kjd4 at hotmail.com  Fri Nov 12 17:56:21 2004
From: betamagalhaes34kjd4 at hotmail.com (Roberta Magalhães)
Date: Fri Nov 12 17:56:11 2004
Subject: [Spambayes] Cadastros de e-mails
Message-ID: <20041112165608.7F55F1E4009@bag.python.org>

Cadastros de e-mails para mala direta e e-mail marketing. Listas
atualizadas e personalizadas. Visite agora:

http://www.gueb.de/dvgamail


MALA DIRETA,  email,  E-MAILS PARA MALAS DIRETAS,  E-mails por
segmenta??o,  listas,  emails por segmento,  Programas gr?tis,  Listas, 
divulga??o por e-mail,  spam,  softwares gratuitos,  mala direta,  listas
de e-mails,  divulga??o de sites e  promo??o de home pages,

http://www.gueb.de/dvgamail
From ddeutsch at cityofbowie.org  Fri Nov 12 21:10:00 2004
From: ddeutsch at cityofbowie.org (David Deutsch)
Date: Fri Nov 12 21:18:06 2004
Subject: [Spambayes] startup error
Message-ID: <0651F89741FB624D84D37576727CF71701536CAB@BOWIE1.cityofbowie.org>

A non-text attachment was scrubbed...
Name: spambayes1.log
Type: application/octet-stream
Size: 2628 bytes
Desc: not available
Url : http://mail.python.org/pipermail/spambayes/attachments/20041112/864fc166/spambayes1.obj
From kennypitt at hotmail.com  Sat Nov 13 00:14:20 2004
From: kennypitt at hotmail.com (Kenny Pitt)
Date: Sat Nov 13 00:15:12 2004
Subject: [Spambayes] RE: [spambayes-dev] Update
In-Reply-To: <B6430707D34115468F35BD7AA430DFC004EB5295@eczoc01exc02.Entel.Entelcorp.com>
Message-ID: <BAY16-DAV12ch9wPiyA00029dc9@hotmail.com>

You should be able to install the 1.0 version directly over your existing
1.0rc2.  The installation does not affect your configuration or training
data.
 
For future reference, the  <mailto:spambayes@python.org>
spambayes@python.org mailing list is the most appropriate forum for this
type of question.  The  <mailto:spambayes-dev@python.org>
spambayes-dev@python.org list is primarily for discussion about the
development of SpamBayes.
 
-- 
Kenny Pitt
 


  _____  

From: spambayes-dev-bounces@python.org
[mailto:spambayes-dev-bounces@python.org] On Behalf Of Tallard Cornejo
Eduardo Antonio
Sent: Friday, November 12, 2004 4:03 PM
To: spambayes-dev@python.org
Subject: [spambayes-dev] Update



I have the version 1rc2. I want to install the version 1.0, but I don't miss
my actual configuration, what can I do?

 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.python.org/pipermail/spambayes/attachments/20041112/9fd72bb0/attachment.htm
From robert at imagine.co.uk  Sat Nov 13 00:30:34 2004
From: robert at imagine.co.uk (Robert Neuschul)
Date: Sat Nov 13 00:30:35 2004
Subject: [Spambayes] A most peculair problem with SB 1
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E8FBCA05@its-xchg4.massey.ac.nz>
References: <ECBA357DDED63B4995F5C1F5CBE5B1E8FBCA05@its-xchg4.massey.ac.nz>
Message-ID: <VA.00005758.00a957af@imagine.co.uk>

Tony

You're a star :-)

Thanks, I'll install this evening and report back later.

RObert.


From addmd at prodigy.net  Sat Nov 13 11:39:10 2004
From: addmd at prodigy.net (addmd)
Date: Sat Nov 13 11:39:16 2004
Subject: [Spambayes] Configure box won't show
Message-ID: <4195E44E.959CBC18@prodigy.net>

I just installed SpamBayes version 1.0 and I can't get the
"Configure" box to appear on my computer so that I can enter my
POP3 e-mail server address.  When I "right-click" the tray icon
and select "Configure" nothing at all happens, although in the
"read me" page it shows a window that should appear.

I have checked as many messages on the sourceforge page to see if

any of them appear to be the same problem.  I didn't want to
bother you if the question had already been answered, but I
didn't come across anything that seemed similar.

I would greatly appeciate any help with this.

Thanks,

Victor




From robert at imagine.co.uk  Sat Nov 13 12:51:22 2004
From: robert at imagine.co.uk (Robert Neuschul)
Date: Sat Nov 13 12:51:22 2004
Subject: [Spambayes] A most peculair problem with SB 1
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E8FBCA05@its-xchg4.massey.ac.nz>
References: <ECBA357DDED63B4995F5C1F5CBE5B1E8FBCA05@its-xchg4.massey.ac.nz>
Message-ID: <VA.00005759.00253218@imagine.co.uk>

Tony

Tried that: sadly it's not working. Perhaps I'm doing something wrong. 
Stopped the running non-functional service, replaced the file in 
python/scripts with your version, restarted the service and got the 
following

C:\Python23\Scripts>pop3proxy_service.py start
Traceback (most recent call last):
  File "C:\Python23\Scripts\pop3proxy_service.py", line 157, in ?
    sys32path = shell.SHGetFolderPath(0, shellcon.CSIDL_SYSTEM, 0, 0)
AttributeError: 'module' object has no attribute 'CSIDL_SYSTEM'

C:\Python23\Scripts>

A restart in debug mode gets the same output.

Regardless of how it is started [start or debug] W2003 reports the 
service as running, but the web UI is not available.

Not sure what's going on.

Robert.


From richie at entrian.com  Sat Nov 13 15:44:14 2004
From: richie at entrian.com (Richie Hindle)
Date: Sat Nov 13 15:44:16 2004
Subject: [Spambayes] Configure box won't show
In-Reply-To: <4195E44E.959CBC18@prodigy.net>
References: <4195E44E.959CBC18@prodigy.net>
Message-ID: <rd7cp0trcgcrnqgmuglaqb8o4kbqccot2p@4ax.com>


[Victor]
> I just installed SpamBayes version 1.0 and I can't get the
> "Configure" box to appear on my computer so that I can enter my
> POP3 e-mail server address.  When I "right-click" the tray icon
> and select "Configure" nothing at all happens

All that command tries to do is launch your web browser pointing to
http://localhost:8880/config - what happens if you type that address
into your browser's address bar by hand?

-- 
Richie Hindle
richie@entrian.com

From Poppe at organic.nl  Sat Nov 13 22:24:19 2004
From: Poppe at organic.nl (Poppe Braam)
Date: Sat Nov 13 22:24:20 2004
Subject: [Spambayes] (no subject)
Message-ID: <284F197CBE5F1441B02DAEDAC2843C2326593B@fs2.domein.local>

Hello, I would like to know if yr programme. which I am using in the Netherlands
is also available in China.
My business relations there would like to use it.

Friendly greetings
Poppe Braam
CEO Do-It b.v. Holland
poppe@organic.nl

From richie at entrian.com  Sat Nov 13 22:30:06 2004
From: richie at entrian.com (Richie Hindle)
Date: Sat Nov 13 22:30:09 2004
Subject: [Spambayes] (no subject)
In-Reply-To: <284F197CBE5F1441B02DAEDAC2843C2326593B@fs2.domein.local>
References: <284F197CBE5F1441B02DAEDAC2843C2326593B@fs2.domein.local>
Message-ID: <j2vcp05t2el998tfec73bl3980399aqupo@4ax.com>


[Poppe]
> Hello, I would like to know if yr programme. which I am using in the Netherlands
> is also available in China.
> My business relations there would like to use it.

Yes, SpamBayes is freely available worldwide, via http://spambayes.org
or http://spambayes.sourceforge.net/

If for some reason your associate can't access those sites, you are free
to email the software to him.

-- 
Richie Hindle
richie@entrian.com

From ricardobssilverio2kd at mail.com  Sun Nov 14 00:05:58 2004
From: ricardobssilverio2kd at mail.com (Ricardo Silverio)
Date: Sun Nov 14 00:05:43 2004
Subject: [Spambayes] e-mails para mala direta
Message-ID: <20041113230541.277811E4007@bag.python.org>

MALA DIRETA,  email,  E-MAILS PARA MALAS DIRETAS,  E-mails por
segmenta??o,  listas,  emails por segmento Cadastros de e-mails para mala
direta e e-mail marketing. Listas atualizadas e personalizadas. 

http://www.estacion.de/maladireta


Listas de e-mails,  divulga??o de sites e  promo??o de home
pages,Programas gr?tis,  Listas,  divulga??o por e-mail,  spam,  softwares
gratuitos,  mala direta,  

http://www.estacion.de/maladireta
From jg at greenmoss.net  Sun Nov 14 05:44:53 2004
From: jg at greenmoss.net (Goldsmith)
Date: Sun Nov 14 05:44:18 2004
Subject: [Spambayes] New Spambayes feature?
Message-ID: <000401c4ca04$b1632040$606cc43f@RANDOM>

I was thinking that spambayes could include something in its detection
engine which would convert ASCII letters with accents above them commonly
used to bypass filtering (e.g. "?f y?u n??d h?gh qu?l?ty m?d?c?t??n ?nd
w?uld l?v? t? s?v? ?n ?utr?g??us r?t??l pr?c?ng") to their equivalent, e.g.
the filter would check "?" as "o" against its token index. It would be easy
to do and would prevent the requirement of many additional tokens for every
variation of "pricing" or "medication" which might contain any number of
combinations of accented and unaccented characters.

Thanks. Maybe just respond with a "Good Idea" "Bad Idea" or "Go away" so I
can know someone received this.

-B

From brian at moonstarlodge.com  Sun Nov 14 14:42:32 2004
From: brian at moonstarlodge.com (Brian Lovelady)
Date: Sun Nov 14 14:54:34 2004
Subject: [Spambayes] Disappearing messages
Message-ID: <001501c4ca4f$cdf2b900$6401a8c0@BKL>

Skipped content of type multipart/alternative-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: image/jpeg
Size: 5675 bytes
Desc: not available
Url : http://mail.python.org/pipermail/spambayes/attachments/20041114/acd626bd/attachment.jpe
From skip at pobox.com  Sun Nov 14 14:58:39 2004
From: skip at pobox.com (Skip Montanaro)
Date: Sun Nov 14 14:58:28 2004
Subject: [Spambayes] New Spambayes feature?
In-Reply-To: <000401c4ca04$b1632040$606cc43f@RANDOM>
References: <000401c4ca04$b1632040$606cc43f@RANDOM>
Message-ID: <16791.25743.136522.529901@montanaro.dyndns.org>

>>>>> "B" == Goldsmith  <jg@greenmoss.net> writes:

    B> I was thinking that spambayes could include something in its
    B> detection engine which would convert ASCII letters with accents above
    B> them commonly used to bypass filtering

There is a replace_nonascii_chars option which will replace such characters
with question marks.  I implemented what you asked for awhile ago, but we
didn't add it as it didn't seem to have any effect on accuracy.  Database
size isn't generally a prime consideration, but if you think this particular
technique is bloating databases perhaps it bears looking at again.

Skip
From LarsK at kappenberg.de  Sun Nov 14 15:59:32 2004
From: LarsK at kappenberg.de (Lars Kappenberg)
Date: Sun Nov 14 15:59:43 2004
Subject: [Spambayes] SpamBayes Data directory
Message-ID: <3434740D6BF805489859DAAC4BA8B88505786D@KS01.kappenberg.local>

Hi *,

first of all, many thanks for this great piece of software.
I am using SpamBayes for quite a long time and am very happy about it.

After setting up my new notebook a question arised:
Is it possible to modify the path to the SpamBayes database dir, which
is

"%USERPROFILE%\Application Data\SpamBayes" by default?

I would love to have the db on a network volume (using it with my PC and
my Notebook [not at the same time]...)

I already tried a Shortcut and a symbolic link (since a junction does
not work with Network volumes).

Regards,
Lars
From rolph at 110.net  Sun Nov 14 19:15:39 2004
From: rolph at 110.net (Don Rolph)
Date: Sun Nov 14 19:15:45 2004
Subject: [Spambayes] why doesn't the pop3 proxy filter all messages?
Message-ID: <5.1.0.14.2.20041114130956.02636b70@localhost>

I have installed spambayes 1.0 as a pop3/smtp proxy with Eudora.  I find 
that it filters some of the incoming pop3 mail but not all.  I run the 
identical test mail through the mail system and sometimes the pop3 proxy 
will classify the mail and provide a spam score and other times it passes 
the mail through without any checking or providing any spam score.

Thoughts?  I know that the pop3 proxy is passing the mail because:

o  the active sessions counter gets incremented
o  the mail is delivered and the client does not know how to contact the 
external pop3 server itself

This beast is in general working great with excellent selectivity on spam, 
BUT it only seems to be working on some of the messages.

Thanks in advance!


Regards.

Don Rolph rolph@110.net


From weber201 at npgcable.com  Sun Nov 14 20:36:26 2004
From: weber201 at npgcable.com (The Webers)
Date: Sun Nov 14 20:36:41 2004
Subject: [Spambayes] Filter Based on Recognized Email ISPs - Bounce Back
	Spam to Sending IP Address
Message-ID: <000001c4ca81$3edfe8f0$a6147918@P08287>

I've been out of the technical area for sometime but it seems to me that all
this spam should be able to be bounced back to the sending server.   

 

Couldn't a front end app could be developed that: 

 

1.	- Contains a list of all recognized Email providers (Yahoo, MSN,
NPGCABLE ..) and allow that Email to come through 
2.	- provide option to filters on userid, subject, text.
3.	- Email NOT passing that filter be rejected back to the sender but
send a copy to a separate folder for user to review and add Email address to
filter to allow future Email from that specific user to pass through. 
4.	Email not passing the filter should be analyzed to find the sending
IP and bounce it back to the route it came from.  This might help getting a
portion of the spam stopped or at least place less work on the recipient.

 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.python.org/pipermail/spambayes/attachments/20041114/f4e468f1/attachment.html
From marionrogers at lineone.net  Mon Nov 15 01:10:29 2004
From: marionrogers at lineone.net (marionrogers@lineone.net)
Date: Mon Nov 15 01:10:30 2004
Subject: [Spambayes] Problem with POP3 Proxy: [unsure how to make SpamBayes
	check new mail]
Message-ID: <20041115001029.581BB1E4002@bag.python.org>

A non-text attachment was scrubbed...
Name: SpamBayesServer1.log
Type: application/octet-stream
Size: 673 bytes
Desc: not available
Url : http://mail.python.org/pipermail/spambayes/attachments/20041115/20212d37/SpamBayesServer1.obj
-------------- next part --------------
I am using SpamBayes POP3 Proxy Version 1.0rc2 (June 2004) (binary),
with version 2.3.3 (#51, Feb 13 2004, 14:39:56) [MSC v.1200 32 bit
(Intel)] of Python; my operating system is Windows 5.1.2600.2 (Service
Pack 1).  I have trained 2 ham and 40 spam.
 
The problem I am having is I have only very basic knowledge of
computing and have no idea what I need to do in order for SpamBayes to
start checking incoming mail and what happens to the mail one sorted
between ham and spam.  At present I check incoming mail in Ticali,
delete the 99% which is smam without opening it, then download ham via
Outlook Express.  Any help would be much appreciated.
From TJLWBECGSGWU at spammotel.com  Mon Nov 15 01:19:23 2004
From: TJLWBECGSGWU at spammotel.com (Mathew Hendry)
Date: Mon Nov 15 01:19:41 2004
Subject: [Spambayes] Re: Filter Based on Recognized Email ISPs - Bounce Back
	Spam to Sending IP Address
References: <000001c4ca81$3edfe8f0$a6147918@P08287>
Message-ID: <k3qfp052tv7neq5et2b1deurvrnpikafjj@4ax.com>

On Sun, 14 Nov 2004 12:36:26 -0700, "The Webers" <weber201@npgcable.com>
wrote:

>Couldn't a front end app could be developed that: 
>
>1.	- Contains a list of all recognized Email providers (Yahoo, MSN,
>NPGCABLE ..)

You mean something like http://www.senderbase.org ?

>and allow that Email to come through
>2.	- provide option to filters on userid, subject, text.

Okay, there are plenty of filters that can do that. Most e-mail programs
have such features built in.

>3.	- Email NOT passing that filter be rejected back to the sender but
>send a copy to a separate folder for user to review and add Email address to
>filter to allow future Email from that specific user to pass through.
>4.	Email not passing the filter should be analyzed to find the sending
>IP and bounce it back to the route it came from.  This might help getting a
>portion of the spam stopped or at least place less work on the recipient.

It doesn't make much sense to talk about bouncing to an IP address: bounces
are e-mails, sent to an e-mail address. The problem is, which e-mail
address? Bouncing to the claimed sender is of course a bad idea, because
spammers invariably forge this information. For more on this see
http://spamlinks.openrbl.org/filter-bounce-fake.htm .

OTOH, http://www.spamcop.net sounds close to what you're after. It doesn't
send bounces; it sends complaints to the ISPs responsible for the offending
IP addresses. Repeat offenders are added to an IP blacklist available to
anyone through the bl.spamcop.net DNS zone.

Combine that with a filtering program that can be configured to refer to
bl.spamcop.net, spamassassin for instance, and you're set.

-- Mat.


From tameyer at ihug.co.nz  Mon Nov 15 01:37:00 2004
From: tameyer at ihug.co.nz (Tony Meyer)
Date: Mon Nov 15 01:37:36 2004
Subject: [Spambayes] SpamBayes Data directory
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E801623095@its-xchg4.massey.ac.nz>
Message-ID: <ECBA357DDED63B4995F5C1F5CBE5B1E86C37E5@its-xchg4.massey.ac.nz>

> After setting up my new notebook a question arised:
> Is it possible to modify the path to the SpamBayes database 
> dir, which is
> 
> "%USERPROFILE%\Application Data\SpamBayes" by default?

Yes.  Please see FAQ 3.14:

<http://spambayes.org/faq.html#how-can-i-change-the-directory-that-spambayes
-stores-my-data-in>

=Tony.Meyer

-- 
Please always include the list (spambayes@python.org) in your replies
(reply-all), and please don't send me personal mail about SpamBayes.
http://www.massey.ac.nz/~tameyer/writing/reply_all.html explains this.

From tameyer at ihug.co.nz  Mon Nov 15 01:41:02 2004
From: tameyer at ihug.co.nz (Tony Meyer)
Date: Mon Nov 15 01:41:31 2004
Subject: [Spambayes] Disappearing messages
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E801623091@its-xchg4.massey.ac.nz>
Message-ID: <ECBA357DDED63B4995F5C1F5CBE5B1E86C37E6@its-xchg4.massey.ac.nz>

> lately I have noticed that some messages are 
> getting deleted before they are placed in a folder.
> Is this something that SB does or should I be
> looking at Outlook or some other possible problem?

SpamBayes doesn't have the ability to delete any mail, so that definitely
isn't the case.  It is possible that SpamBayes is *moving* the mail - if you
haven't checked your unsure (Possible Junk) or spam (Junk) folders for these
messages, then it would be worth doing that.

Some of the information in FAQ 3.12 might help:

<http://spambayes.org/faq.html#some-of-my-mail-is-going-missing>

You could also check the SpamBayes log (SpamBayes
Manager->Advanced->Diagnostics->View Log) - it has an entry for each message
that it 'sees', so you could at least tell if the messages got that far
(although SpamBayes fires after Outlook, so if Outlook has done something
with it, then SpamBayes will have no idea the message existed).

=Tony.Meyer

-- 
Please always include the list (spambayes@python.org) in your replies
(reply-all), and please don't send me personal mail about SpamBayes.
http://www.massey.ac.nz/~tameyer/writing/reply_all.html explains this.

From tameyer at ihug.co.nz  Mon Nov 15 01:55:34 2004
From: tameyer at ihug.co.nz (Tony Meyer)
Date: Mon Nov 15 01:56:12 2004
Subject: [Spambayes] Filter Based on Recognized Email ISPs - Bounce
	BackSpam to Sending IP Address
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E80162309E@its-xchg4.massey.ac.nz>
Message-ID: <ECBA357DDED63B4995F5C1F5CBE5B1E8FBCA09@its-xchg4.massey.ac.nz>

> I've been out of the technical area for sometime but
> it seems to me that all this spam should be able to be
> bounced back to the sending server.   

FAQ 6.5 covers this (in a very brief way):

<http://spambayes.org/faq.html#why-can-t-i-bounce-spam-back-to-the-sender>

> Couldn't a front end app could be developed that: 
> 
> - Contains a list of all recognized Email providers
> (Yahoo, MSN, NPGCABLE ..) and allow that Email to come through 

You could never collect a list that included every ISP in every country in
the world, let alone all company domains and everyone that has their own
domain.  (Well, maybe you could, but it would include every domain there is,
so would be pointless).

This would also be worthless against almost all spam, which tends to forge
sender addresses.

> - provide option to filters on userid, subject, text. 

Rule-based filtering, like that, is generally less effective than solutions
like SpamBayes (and all the others).

=Tony.Meyer

-- 
Please always include the list (spambayes@python.org) in your replies
(reply-all), and please don't send me personal mail about SpamBayes.
http://www.massey.ac.nz/~tameyer/writing/reply_all.html explains this.

From tameyer at ihug.co.nz  Mon Nov 15 05:00:20 2004
From: tameyer at ihug.co.nz (Tony Meyer)
Date: Mon Nov 15 05:00:58 2004
Subject: [Spambayes] startup error
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E80162305E@its-xchg4.massey.ac.nz>
Message-ID: <ECBA357DDED63B4995F5C1F5CBE5B1E8FBCA0B@its-xchg4.massey.ac.nz>

It appears that something is wrong with your configuration (accessing a
folder, from the log).  Have you tried reconfiguring, as the error message
suggests?  If so, what happened then?

=Tony.Meyer

-- 
Please always include the list (spambayes@python.org) in your replies
(reply-all), and please don't send me personal mail about SpamBayes.
http://www.massey.ac.nz/~tameyer/writing/reply_all.html explains this.

From tameyer at ihug.co.nz  Mon Nov 15 05:02:09 2004
From: tameyer at ihug.co.nz (Tony Meyer)
Date: Mon Nov 15 05:02:46 2004
Subject: [Spambayes] why doesn't the pop3 proxy filter all messages?
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E80162309B@its-xchg4.massey.ac.nz>
Message-ID: <ECBA357DDED63B4995F5C1F5CBE5B1E8FBCA0C@its-xchg4.massey.ac.nz>

> I have installed spambayes 1.0 as a pop3/smtp proxy with 
> Eudora.  I find that it filters some of the incoming pop3
> mail but not all.  I run the identical test mail through the
> mail system and sometimes the pop3 proxy will classify the
> mail and provide a spam score and other times it passes 
> the mail through without any checking or providing any spam score.

If you examine the headers of the messages that are not filtered, do they
have an "X-Spambayes-Exception" header?  If so, what does that header
contain?

=Tony.Meyer

-- 
Please always include the list (spambayes@python.org) in your replies
(reply-all), and please don't send me personal mail about SpamBayes.
http://www.massey.ac.nz/~tameyer/writing/reply_all.html explains this.

From tameyer at ihug.co.nz  Mon Nov 15 05:06:30 2004
From: tameyer at ihug.co.nz (Tony Meyer)
Date: Mon Nov 15 05:07:05 2004
Subject: [Spambayes] Problem with POP3 Proxy: [unsure how to make
	SpamBayescheck new mail]
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E8016235FE@its-xchg4.massey.ac.nz>
Message-ID: <ECBA357DDED63B4995F5C1F5CBE5B1E86C37E9@its-xchg4.massey.ac.nz>

> The problem I am having is I have only very basic knowledge 
> of computing and have no idea what I need to do in order for 
> SpamBayes to start checking incoming mail and what happens to 
> the mail one sorted between ham and spam.

You need to go to the SpamBayes configuration page and enter your incoming
mail server details (I suspect you have done this).  Since you are using
Outlook Express, you will also need to tick the "unsure" and "spam" boxes
for the "Notate Subject" and/or "Notate To" options.

You then need to change Outlook Express to collect mail via SpamBayes.  You
do this by doing Tools->Accounts->Mail, selecting your mail account, then
click Properties, then Servers.  Change the incoming server to "localhost".

This should then have all incoming mail classified as ham, spam, or unsure.
This classification will appear in the subject and/or to headers, depending
on which option you used above.  You can then set up an Outlook Express rule
(like any other rule) to (eg) move these messages somewhere appropriate.

The classification you do on the "review" page on the web interface is
*training* SpamBayes, so that it understands what mail you want, and what
you don't.  You should update this periodically, especially if there are any
errors.

=Tony.Meyer

-- 
Please always include the list (spambayes@python.org) in your replies
(reply-all), and please don't send me personal mail about SpamBayes.
http://www.massey.ac.nz/~tameyer/writing/reply_all.html explains this.

From tameyer at ihug.co.nz  Mon Nov 15 05:09:25 2004
From: tameyer at ihug.co.nz (Tony Meyer)
Date: Mon Nov 15 05:10:17 2004
Subject: [Spambayes] A most peculair problem with SB 1
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E80162306A@its-xchg4.massey.ac.nz>
Message-ID: <ECBA357DDED63B4995F5C1F5CBE5B1E86C37EA@its-xchg4.massey.ac.nz>

> Tried that: sadly it's not working. Perhaps I'm doing 
> something wrong. 
> Stopped the running non-functional service, replaced the file in 
> python/scripts with your version, restarted the service and got the 
> following
> 
> C:\Python23\Scripts>pop3proxy_service.py start
> Traceback (most recent call last):
>   File "C:\Python23\Scripts\pop3proxy_service.py", line 157, in ?
>     sys32path = shell.SHGetFolderPath(0, shellcon.CSIDL_SYSTEM, 0, 0)
> AttributeError: 'module' object has no attribute 'CSIDL_SYSTEM'

What version of pywin32 are you using?  Mine (203) has
shellcon.CSIDL_SYSTEM, but it seems yours doesn't.

If you go to line 157 of "C:\Python23\Scripts\pop3proxy_service.py" and
change "shellcon.CSIDL_SYSTEM" to "37" (no quotes), that should fix it (so
that you can try the new script, at least).

=Tony.Meyer

-- 
Please always include the list (spambayes@python.org) in your replies
(reply-all), and please don't send me personal mail about SpamBayes.
http://www.massey.ac.nz/~tameyer/writing/reply_all.html explains this.

From tameyer at ihug.co.nz  Mon Nov 15 05:29:39 2004
From: tameyer at ihug.co.nz (Tony Meyer)
Date: Mon Nov 15 05:30:18 2004
Subject: [Spambayes] Spambayes problem
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E801622E19@its-xchg4.massey.ac.nz>
Message-ID: <ECBA357DDED63B4995F5C1F5CBE5B1E86C37EB@its-xchg4.massey.ac.nz>

> I'm using Spambayes 1.0 on a Windows server 2003 with
> termial server.
> All user have Outlook 2003 and until now it worked without problem.
> This morning on user had a problem. Outlook starts and the
> Spambayes toolbar show up but doesn't work.
> Incomming mails are not verifiyed.
> 
> The log tell:
> raceback (most recent call last):
>   File "<string>", line 1, in ?
>   File "boot_com_servers.py", line 62, in DllRegisterServer
>   File "win32com\server\register.pyc", line 405, in RegisterClasses
>   File "win32com\server\register.pyc", line 177, in RegisterServer
>   File "win32com\server\register.pyc", line 32, in _set_string
> pywintypes.error: (5, 'RegSetValue', 'Acc\xe8s refus\xe9.')

This is trying to set a value in the registry and is (it appears) being
denied access.  Have the permissions of the user's account changed recently?

However, AFAICT, this code should only be executed when you are trying to
register (install) the plug-in.  Did the user try to do that?  If so, then
the actual problem might be in an older log.

=Tony.Meyer

-- 
Please always include the list (spambayes@python.org) in your replies
(reply-all), and please don't send me personal mail about SpamBayes.
http://www.massey.ac.nz/~tameyer/writing/reply_all.html explains this.

From ede.winkell at hccnet.nl  Mon Nov 15 10:17:26 2004
From: ede.winkell at hccnet.nl (E. de Winkel)
Date: Mon Nov 15 10:25:35 2004
Subject: [Spambayes] downloading SpamBayes
Message-ID: <001801c4caf3$ecbba280$0101a8c0@desktop>

dear sir, madam,
after reading a recent article in a dutch computermagazine I became interested in Spambayes and tried to download this programm. Since I am fluent in English, German and Dutch I tried several (Belgian, Swiss, German and American) download sites but on neither of these I could find the right place. Possibly because I am not a "member" of neither of these institutions.
please advice how to download.
regards
Ermin de Winkel
ede.winkel@hccnet.nl
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.python.org/pipermail/spambayes/attachments/20041115/faab54fb/attachment.html
From jsp at PKC.com  Mon Nov 15 14:11:43 2004
From: jsp at PKC.com (Jesse Pelton)
Date: Mon Nov 15 14:11:45 2004
Subject: [Spambayes] downloading SpamBayes
Message-ID: <779b6becec47b2e7c113b44cacd215ce4198ab0f@vtmail.pkc.com>

Have you tried http://spambayes.sourceforge.net/download.html?
<https://sourceforge.net/project/showfiles.php?group_id=61702> 


________________________________

	From: spambayes-bounces@python.org
[mailto:spambayes-bounces@python.org] On Behalf Of E. de Winkel
	Sent: Monday, November 15, 2004 4:17 AM
	To: spambayes@python.org
	Subject: [Spambayes] downloading SpamBayes
	
	
	dear sir, madam,
	after reading a recent article in a dutch computermagazine I
became interested in Spambayes and tried to download this programm.
Since I am fluent in English, German and Dutch I tried several (Belgian,
Swiss, German and American) download sites but on neither of these I
could find the right place. Possibly because I am not a "member" of
neither of these institutions.
	please advice how to download.
	regards
	Ermin de Winkel
	ede.winkel@hccnet.nl

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.python.org/pipermail/spambayes/attachments/20041115/4ad119f1/attachment.html
From james at cleaningsystemsinc.com  Mon Nov 15 16:41:48 2004
From: james at cleaningsystemsinc.com (James Beno)
Date: Mon Nov 15 16:40:15 2004
Subject: [Spambayes] help needed
Message-ID: <A907CCAB918F1A4F9B142C2CBB51F2C1398D7F@ntcsi2.netcsi.csi.lan>

I have used Spambayes for the last several months and love it.  Recently
my company gave me a new PC.  It appears that I have SpamBayes in my
programs but the tool bar does not show up in Outlook, nor does it work.
I have not been successful at downloading it either.  Please provide
suggestions on what I should do.
Thanks,
Jim Beno
jbeno@cleaningsystemsinc.com
920-337-2175 x 113
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.python.org/pipermail/spambayes/attachments/20041115/2cd2471e/attachment.html
From shlif at optonline.net  Mon Nov 15 18:00:20 2004
From: shlif at optonline.net (Abe Shliferstein)
Date: Mon Nov 15 18:01:09 2004
Subject: [Spambayes] Linking up to existing training
Message-ID: <0I78001LTCKGGF@mta7.srv.hcvlny.cv.net>

I use Outlook on two networked machines (one at a time, of course).  I
recently had some file problems on one machine.  After fixing them up, I
found that SpamBayes was totally untrained on that machine ("Database has no
training information"), but still fine on the other.  I know there's a way
to make both instances share the same training database but I can't remember
how to link to the good one.  Could you please tell me what steps I need to
take?

 

Thanks.  By the way, I think SpamBayes is the best anti-spam tool and I keep
spreading the word.

 

Abe

---

Abe Shliferstein

shlif@optonline.net

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.python.org/pipermail/spambayes/attachments/20041115/544a5447/attachment.htm
From cmuser at shaggy.smf.ebay.com  Mon Nov 15 17:46:36 2004
From: cmuser at shaggy.smf.ebay.com (Unexpected reply handler)
Date: Mon Nov 15 18:21:22 2004
Subject: [Spambayes] Re: approved
In-Reply-To: <200411151627.iAFGRhNk013983@mailhost2.sjc.ebay.com>
References: <200411151627.iAFGRhNk013983@mailhost2.sjc.ebay.com>
Message-ID: <200411151646.iAFGka000783@shaggy.smf.ebay.com>

Thank you for your response. Please don't reply to this message - it is an automated response and your reply will not be received.
 
If you have a question for eBay Customer Support, please visit the following eBay Help page. This page will help you locate the answer to your question, or assist you in contacting us:

 http://pages.ebay.com/help/index.html

If you would like to change your notification preferences, which determine what type of email you receive from eBay, please follow the steps below:

1. Click "My eBay" located at the top of all eBay pages. You may be asked to sign in.

2. Click the "eBay Preferences" link located under the "My Account" heading.

3. Click the "view/change" link to the right of "Notification Preferences." You may be asked to sign in once more.

4. On the "Change Your Notification Preferences" page, check the boxes to indicate the types of messages you'd like to receive from eBay. Then, uncheck the boxes to indicate the types of messages you don't want to receive from us.

5. Once you're done, be sure to click the "Save Changes" button at the top or bottom of the page. 

Again, thanks for writing eBay.


-- 
From GCE at gcevaluators.com  Mon Nov 15 20:18:53 2004
From: GCE at gcevaluators.com (GCE)
Date: Mon Nov 15 20:18:57 2004
Subject: [Spambayes] Spambayes and Netscape 4.X
Message-ID: <01b401c4cb47$f2359190$0500a8c0@superbubba>

First off, IT'S NOT MY FAULT!  I have begged my wife on several occasions to
upgrade her mail to something other than netscape communicator but she
won't.  

 

I have set up spambayes as a pop3 filter and it marks everything as spam.
The problem I have is that netscape isn't filtering the spam to the spam
folder.

 

I have set up a filter within Netscape to do this but netscape seems to
ignore it.  The filter reads as such.

 

Header of the message contains X-Spambayes Classification: spam move to
folder spam

 

What in the hell am I doing wrong?

 

Michael

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.python.org/pipermail/spambayes/attachments/20041115/7d7a586c/attachment.html
From fabienbaudon at free.fr  Mon Nov 15 20:47:09 2004
From: fabienbaudon at free.fr (Fabien Baudon)
Date: Mon Nov 15 20:46:41 2004
Subject: [Spambayes] Install problem
Message-ID: <000701c4cb4b$e51bb300$39ecfe52@gadsweb>

I use Windows 2000 NT and Outlook express 6.
I can't install spambayes outlook addin.
The message from the log file is:
    Registered: SpamBayes.OutlookAddin
    Registration complete.
I don't find the Tab labelled Other in the Menu Tools/Option of Outlook. So,
I can't check COM Add-Ins.
What can I do? Is the Addin compatible with Outlook express 6
Thank you for your answer.


From richie at entrian.com  Mon Nov 15 22:09:34 2004
From: richie at entrian.com (Richie Hindle)
Date: Mon Nov 15 22:09:36 2004
Subject: [Spambayes] Spambayes and Netscape 4.X
In-Reply-To: <01b401c4cb47$f2359190$0500a8c0@superbubba>
References: <01b401c4cb47$f2359190$0500a8c0@superbubba>
Message-ID: <4m6ip0hp0r0eet92n1f6b90i3025ro09fb@4ax.com>


[Michael]
> First off, IT'S NOT MY FAULT!  I have begged my wife on several occasions to
> upgrade her mail to something other than netscape communicator but she
> won't. 

Commiserations!  8-)

> Header of the message contains X-Spambayes Classification: spam move to
> folder spam

That should be X-Spambayes-Classification, with two hyphens and no
spaces.

-- 
Richie Hindle
richie@entrian.com

From kennypitt at hotmail.com  Mon Nov 15 22:25:12 2004
From: kennypitt at hotmail.com (Kenny Pitt)
Date: Mon Nov 15 22:26:04 2004
Subject: [Spambayes] Install problem
In-Reply-To: <000701c4cb4b$e51bb300$39ecfe52@gadsweb>
Message-ID: <BAY16-DAV14fAj81EYt000350e2@hotmail.com>

Fabien Baudon wrote:
> I use Windows 2000 NT and Outlook express 6.
> I can't install spambayes outlook addin.
> The message from the log file is:
>     Registered: SpamBayes.OutlookAddin
>     Registration complete.
> I don't find the Tab labelled Other in the Menu Tools/Option of
> Outlook. So, I can't check COM Add-Ins. What can I do? Is the Addin
> compatible with Outlook express 6 Thank you for your answer. 

Outlook Express is not the same thing as Outlook, and is not supported by
the Outlook Addin version of SpamBayes.  You'll need to use the POP3 Proxy
version known as sb_server.  Note that the binary installer for this is the
same as for the Outlook Addin, but you need to choose the "Server/Proxy
application" option instead of the "Microsoft Outlook Addin" when
installing.  Also see FAQ #2.1:

http://spambayes.sourceforge.net/faq.html#does-spambayes-work-with-outlook-e
xpress

-- 
Kenny Pitt

From MEMBER_SERVICES at healthnet.com  Tue Nov 16 09:26:26 2004
From: MEMBER_SERVICES at healthnet.com (MEMBER_SERVICES@healthnet.com)
Date: Tue Nov 16 10:26:39 2004
Subject: [Spambayes] (no subject)
Message-ID: <OF88256F4E.002F2C3B-ON88256F4E.002F2C3B-88256F4E.002E5D94@fhs.com>

Thank you for your inquiry. Health Net of California strives to respond to
all Internet inquiries within two business days of receipt. Please be
advised that complex issues may require additional time for resolution.



This  message,together  with  any  attachments, is
intended only for the use of the individual or entity
to which it is addressed. It may contain information
that is confidential and prohibited from disclosure.
If you are not the intended recipient, you are hereby
notified that  any dissemination  or copying of this
message or any attachment is strictly prohibited. If
you have received this message in error, please notify
the  original  sender immediately by telephone or by
return e-mail and delete this message, along with any
attachments, from your computer.  Thank you.


From EMei at atla.com  Tue Nov 16 19:14:58 2004
From: EMei at atla.com (Mei, Elaine)
Date: Tue Nov 16 19:15:44 2004
Subject: [Spambayes] Slow filtering junk mails
Message-ID: <2E1BDA7F69603A4885F621D302BA1B7DFDBA4D@email.atla.com>

Hello,

 

I recently reformatted my harddrive and re-installed Windows XP, and put
MS office and Spambayes back into the computer.   What I noticed is that
the filtering processing has been very slow since installing Spambayes.
I tried software upgrade and deleting the data files for Spambayes and
it didn't solve the problem.  Any suggestions to proceed further with
this?

 

 

 

Elaine Mei 
250 S. Wacker, Suite 1600 
Chicago, IL 60606-5889 USA 
Toll-free North America: 888-665-ATLA 
Chicago Area: 312-454-5100 
Website: http://www.atla.com <http://www.atla.com/>  

 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.python.org/pipermail/spambayes/attachments/20041116/c6a644d4/attachment.html
From mailgroups at murdochfl.com  Tue Nov 16 22:21:18 2004
From: mailgroups at murdochfl.com (Bob Murdoch)
Date: Tue Nov 16 22:18:46 2004
Subject: [Spambayes] Yahoo addresses
In-Reply-To: <2E1BDA7F69603A4885F621D302BA1B7DFDBA4D@email.atla.com>
Message-ID: <20041116211846.206A91E4006@bag.python.org>

I belong to a number of mailings.  Members on these lists often use
'disposable' email addresses, such as those available from Yahoo.

I have Spambayes installed with Outlook XP.  Mail from these mailing
lists whose original poster had a Yahoo address continually winds up
in my Junk E-mail folder, no matter how many times I select the
Recover from Spam option.

Is there a white list or some other such feature that allows me to
mark all messages from a mailing list as valid rather than spam?

tia,

Bob M..

From enewger at gmx.de  Tue Nov 16 22:33:26 2004
From: enewger at gmx.de (Eckhardt Newger)
Date: Tue Nov 16 22:36:29 2004
Subject: [Spambayes] How to recover spam to a specific ham folder
Message-ID: <cndv7n.1fg.1@ID-40943.user.dfncis.de>

Hello,

I've split up my ham folder into several subfolders each representing
one of my active mail accounts. Rules in Outlook XP direct all emails to
the appropriate subfolder while being received. Is there a way to
specify for a false positive which subfolder an email will redirected to
when using the 'Recover from spam' button? So far I noticed that all
recovered emails end up in just one of those subfolders ( the first one
I set up) not respecting the email address they have been send to.

Many thanks for any help.

Best regards

From tameyer at ihug.co.nz  Wed Nov 17 01:07:41 2004
From: tameyer at ihug.co.nz (Tony Meyer)
Date: Wed Nov 17 01:07:46 2004
Subject: [Spambayes] How to recover spam to a specific ham folder
In-Reply-To: <cndv7n.1fg.1@ID-40943.user.dfncis.de>
Message-ID: <JIEMKPKMIMNMNCGKKKGOOEAJCAAA.tameyer@ihug.co.nz>

> Is there a way to specify for a false positive which
> subfolder an email will redirected to
> when using the 'Recover from spam' button? So far I noticed
> that all recovered emails end up in just one of those
> subfolders ( the first one I set up) not respecting the
> email address they have been send to.

The 'Recover from spam' button is meant to send the message back to the
folder it was in before SpamBayes moved it (here, presumably the correct
subfolder).  However, if it can't manage to do that, then it just sends it
back to the 'main' Inbox, whatever Outlook thinks that is.

Usually problems with figuring out the original folder occur with Hotmail or
IMAP stores - are you using one of those?

Your log files (SpamBayes->Help->Troubleshooting Guide explains where those
are) might have additional clues about what is going wrong here.

=Tony.Meyer

From tameyer at ihug.co.nz  Wed Nov 17 01:07:40 2004
From: tameyer at ihug.co.nz (Tony Meyer)
Date: Wed Nov 17 01:07:47 2004
Subject: [Spambayes] Yahoo addresses
In-Reply-To: <20041116211846.206A91E4006@bag.python.org>
Message-ID: <JIEMKPKMIMNMNCGKKKGOMEAJCAAA.tameyer@ihug.co.nz>

> I have Spambayes installed with Outlook XP.  Mail from these mailing
> lists whose original poster had a Yahoo address continually winds up
> in my Junk E-mail folder, no matter how many times I select the
> Recover from Spam option.

If you use the "show clues for this message" function on one of those
(before training), you might be able to understand why they are scoring what
they do.  If not, feel free to forward one of those 'show clues' messages to
the list, and we'll try and explain what's going on.

> Is there a white list or some other such feature that allows me to
> mark all messages from a mailing list as valid rather than spam?

Discussion of whitelisting is available here:

<http://spambayes.org/faq.html#why-don-t-you-add-whitelisting-blacklisting-t
o-spambayes>

=Tony.Meyer

From tameyer at ihug.co.nz  Wed Nov 17 01:10:12 2004
From: tameyer at ihug.co.nz (Tony Meyer)
Date: Wed Nov 17 01:10:18 2004
Subject: [Spambayes] help needed
In-Reply-To: <A907CCAB918F1A4F9B142C2CBB51F2C1398D7F@ntcsi2.netcsi.csi.lan>
Message-ID: <JIEMKPKMIMNMNCGKKKGOOEAKCAAA.tameyer@ihug.co.nz>

> Recently my company gave me a new PC.  It appears
> that I have SpamBayes in my programs but the tool
> bar does not show up in Outlook, nor does it work.
> I have not been successful at downloading it either.
> Please provide suggestions on what I should do.

What happens if you do try and download it?  You can get it from:

<http://spambayes.org/download.html>

You probably just need to reinstall the plug-in, but you'll need to get hold
of the installer first.

If it already is installed, but just not registered, you could try running
the registration application.  This is called 'register_outlook_addin.exe'
and is in the 'bin' directory in the directory SpamBayes was installed into
(by default "C:\Program Files\Spambayes").

=Tony.Meyer

From tameyer at ihug.co.nz  Wed Nov 17 01:19:31 2004
From: tameyer at ihug.co.nz (Tony Meyer)
Date: Wed Nov 17 01:19:36 2004
Subject: [Spambayes] startup error
In-Reply-To: <0651F89741FB624D84D37576727CF7170108CB15@BOWIE1.cityofbowie.org>
Message-ID: <JIEMKPKMIMNMNCGKKKGOGEALCAAA.tameyer@ihug.co.nz>

[Tony Meyer]
> It appears that something is wrong with your configuration
> (accessing a folder, from the log).  Have you tried
> reconfiguring, as the error message suggests?  If so, what happened
> then?

[Bob Boller]
> The configuration icon (in fact none of the icons) is not there.
> I have uninstalled and reinstalled several with the same message
> each time. Could there be a registry item stuck?

If the toolbar isn't there at all, then getting that back is the first step
(and that may fix whatever the other problem is).  There are various methods
to try and do this outlined in the troubleshooting guide.  A copy is
installed on the machine (normally you could get to it via
SpamBayes->Help->Troubleshooting Guide), or you can use the online copy:

<http://cvs.sf.net/viewcvs.py/*checkout*/spambayes/spambayes/Outlook2000/doc
s/troubleshooting.html?rev=HEAD>

=Tony.Meyer

From tameyer at ihug.co.nz  Wed Nov 17 01:20:37 2004
From: tameyer at ihug.co.nz (Tony Meyer)
Date: Wed Nov 17 01:20:42 2004
Subject: [Spambayes] why doesn't the pop3 proxy filter all messages?
In-Reply-To: <5.1.0.14.2.20041115061018.02444ac0@pop3.110.net>
Message-ID: <JIEMKPKMIMNMNCGKKKGOMEALCAAA.tameyer@ihug.co.nz>

[Don Rolph]
> I have installed spambayes 1.0 as a pop3/smtp proxy with
> Eudora.  I find that it filters some of the incoming pop3
> mail but not all.  I run the identical test mail through the
> mail system and sometimes the pop3 proxy will classify the
> mail and provide a spam score and other times it passes
> the mail through without any checking or providing any spam score.

[Tony Meyer]
>If you examine the headers of the messages that are not filtered, do they
>have an "X-Spambayes-Exception" header?  If so, what does that header
>contain?

[Don Rolph]
> There are no X-Spambayes headers of any type.  There are no
> errors in the log files.

Are you positive that Eudora is connecting via localhost and not directly to
the POP3 server?  It should not be possible for a message to get through the
SpamBayes POP3 proxy without getting either a classification header or an
exception header.

What happens if you telnet to the proxy?  For example (do this from a
Command/DOS prompt):

telnet localhost 110
[wait for the connection]
USER [username]
PASS [password]
RETR 1
QUIT

That should print out to the console the first message waiting to be
downloaded (it will not be deleted, so it'll still be delivered in a regular
mail check).  Does that include any SpamBayes headers?

> I have not found out how to turn on the global verbose option
> to get the pop3 log.

If you run sb_server from a command line, you can add "-o
globals:verbose:True" to the end.  Otherwise, you need to open up the
configuration file (probably called "bayescustomize.ini", and probably in
"C:\Documents and Settings\{username}\Application Data\Spambayes\Proxy" -
the FAQ explains alternative locations) and add these lines:

[globals]
verbose:True

However, I suspect that you won't get anything in the logs.  I could be
wrong, though.

=Tony.Meyer

From tameyer at ihug.co.nz  Wed Nov 17 01:36:26 2004
From: tameyer at ihug.co.nz (Tony Meyer)
Date: Wed Nov 17 01:36:30 2004
Subject: [Spambayes] Linking up to existing training
In-Reply-To: <0I78001LTCKGGF@mta7.srv.hcvlny.cv.net>
Message-ID: <JIEMKPKMIMNMNCGKKKGOEEANCAAA.tameyer@ihug.co.nz>

> I use Outlook on two networked machines (one at a
> time, of course).
[...]
> I know there's a way to make both
> instances share the same training database but I can't
> remember how to link to the good one.  Could you
> please tell me what steps I need to take?

FAQ 3.14 covers setting the data directory to somewhere else (this is where
the database is stored, so most likely what you're after, assuming the
machine can access the data on the other one):

<http://spambayes.org/faq.html#how-can-i-change-the-directory-that-spambayes
-stores-my-data-in>

You do need to ensure that you're never in the situation where both machines
are accessing the database at once, though, or there's a good chance that
it'll get corrupted.

=Tony.Meyer

From tameyer at ihug.co.nz  Wed Nov 17 01:41:43 2004
From: tameyer at ihug.co.nz (Tony Meyer)
Date: Wed Nov 17 01:41:47 2004
Subject: [Spambayes] Slow filtering junk mails
In-Reply-To: <2E1BDA7F69603A4885F621D302BA1B7DFDBA4D@email.atla.com>
Message-ID: <JIEMKPKMIMNMNCGKKKGOAEAOCAAA.tameyer@ihug.co.nz>

> I recently reformatted my harddrive and re-installed
> Windows XP, and put MS office and Spambayes back into
> the computer.   What I noticed is that the filtering
> processing has been very slow since installing Spambayes.
> I tried software upgrade and deleting the data files
> for Spambayes and it didn't solve the problem.  Any
> suggestions to proceed further with this?

My first suggestion would have been to remove the training data and retrain,
unfortunately, but since you've already tried that...

What's slow?  Just filtering the messages as they arrive in the inbox?
Clicking on the "Delete as spam" and "Recover from spam" buttons as well?
Starting up Outlook?

If it's just the former, then maybe it's just the background filtering
values?  If you go into the SpamBayes Manager dialog, then the Advanced tab,
you can adjust these.  If you're not using any Outlook rules, then you can
even disable background filtering altogether.

Otherwise - roughly how much slower would you say everything is?  Twice as
slow?  10 times as slow?  There have been a few (though not many) other
reports of slowness, which is concerning, but there's no obvious cause (at
least yet).  We're still looking into those, unfortunately...

=Tony.Meyer

From mailgroups at murdochfl.com  Wed Nov 17 04:08:05 2004
From: mailgroups at murdochfl.com (Bob Murdoch)
Date: Wed Nov 17 04:05:29 2004
Subject: [Spambayes] Yahoo addresses
In-Reply-To: <JIEMKPKMIMNMNCGKKKGOMEAJCAAA.tameyer@ihug.co.nz>
Message-ID: <20041117030527.C95FD1E4006@bag.python.org>

Tony, 

> -----Original Message-----
> From: Tony Meyer [mailto:tameyer@ihug.co.nz] 
> Sent: Tuesday, November 16, 2004 7:08 PM
> 
> If you use the "show clues for this message" function on one 
> of those (before training), you might be able to understand 
> why they are scoring what they do.  If not, feel free to 
> forward one of those 'show clues' messages to the list, and 
> we'll try and explain what's going on.
> 

Thanks for the offer.  Here's one that is sitting in the Junk E-mail
folder right now:

Combined Score: 0% (1.11022e-016)
Internal ham score (*H*): 1
Internal spam score (*S*): 2.22045e-016

# ham trained on: 11067
# spam trained on: 2816

87 Significant Tokens
token                               spamprob         #ham  #spam
'suggestion'                        0.00215414        104      0
'version?'                          0.00738916         30      0
'header:Received:9'                 0.0163827        2609     11
'versions'                          0.0238391         331      2
'subject:support'                   0.0260702        8377     57
'firebird'                          0.0282926        3922     29
'header:In-Reply-To:1'              0.0290818        6043     46
'skip:f 40'                         0.0300911        7481     59
'url:67folb'                        0.0302692        6932     55
'yahoo!'                            0.0304382        9396     75
'reply-to:addr:yahoogroups.com'     0.0304602        9389     75
'url:docs'                          0.0304602        9389     75
'to:addr:yahoogroups.com'           0.0304728        9385     75
'url:terms'                         0.0308535        9389     76
'aage'                              0.0314234         128      1
'subject:firebird'                  0.0317295        6603     55
'reply-to:addr:firebird-support'    0.0325554        6430     55
'to:addr:firebird-support'          0.0325799        6425     55
'group,'                            0.0333268        9240     81
'url:yahoo'                         0.0335356        9407     83
'groups'                            0.0355302        9394     88
'subject:] '                        0.0357374       10610    100
'url:firebird-support'              0.0369658        5637     55
'exists'                            0.0370612         210      2
'url:tm'                            0.0387088        6935     71
'sponsor'                           0.0412744        6943     76
'ib/fb.'                            0.0412844           5      0
'subject:Limits'                    0.0412844           5      0
'url:groups'                        0.0442376        6373     75
'url:group'                         0.0442907        6365     75
'web,'                              0.0443107        6362     75
'skip:- 70'                         0.0467778        7052     88
'links'                             0.0479997        6396     82
'1.5.'                              0.0486592         158      2
'subject'                           0.0533724        9414    135
'url:us'                            0.0555939        6947    104
'group'                             0.0636147        6598    114
'subject:operator'                  0.0652174           3      0
'subject:-'                         0.065235        10365    184
'skip:- 20'                         0.0711668        7030    137
'message-id:@eGroups.com'           0.0741458        2162     44
'x-mailer:yahoo groups message poster' 0.0741458        2162     44
'url:yqlsaa'                        0.0864342        2910     70
'subject:['                         0.0918335        4083    105
'aage.'                             0.0918367           2      0
'sweep'                             0.0936782        1523     40
'use'                               0.0986524       10056    280
'blocker.'                          0.0991036        1431     40
'pop-up'                            0.0991036        1431     40
'toolbar.'                          0.0991036        1431     40
'url:esiiaa'                        0.0991036        1431     40
'url:l5yrja'                        0.0991036        1431     40
'url:click'                         0.103353         7196    211
'clean'                             0.106282         1522     46
'companion'                         0.107932         1431     44
'url:info'                          0.109057         9409    293
'ads.'                              0.114228         1434     47
'reply-to:no real name:2**0'        0.115949        10699    357
'(for'                              0.121962          143      5
'send'                              0.126389         9319    343
'to:'                               0.12909          9467    357
'free!'                             0.136233         1621     65
'unsubscribe'                       0.142761         9252    392
'header:User-Agent:1'               0.144991         3802    164
'using'                             0.145481         2956    128
'visit'                             0.165135         6677    336
'email'                             0.170982        10386    545
'subject:: '                        0.175712         3172    172
'thanks'                            0.182812         1582     90
'still'                             0.186087         1101     64
'size'                              0.194004          458     28
'limitation'                        0.196574           49      3
'header:Reply-To:1'                 0.197409        10818    677
'subject:Delphi'                    0.210823           89      6
'make'                              0.229174         2803    212
'know'                              0.24713          2000    167
'that'                              0.265611         6618    609
'older'                             0.269515          107     10
'from'                              0.279117        10039    989
'header:Return-path:1'              0.279199          386     38
'this'                              0.282022        10726   1072
'your'                              0.327367        10199   1263
'with'                              0.34073          6806    895
'the'                               0.356808        10641   1502
'subject:with'                      0.365366          717    105
'get'                               0.365705         5542    813
'for'                               0.395639         8603   1433

From bob at 1776.com  Wed Nov 17 00:51:19 2004
From: bob at 1776.com (Robert K. Coe)
Date: Wed Nov 17 12:38:48 2004
Subject: [Spambayes] Slow filtering junk mails
In-Reply-To: <2E1BDA7F69603A4885F621D302BA1B7DFDBA4D@email.atla.com>
Message-ID: <000001c4cc4b$3500b9b0$6601a8c0@CambridgeMA.GOV>

Maybe the reinstallation left some of the system configuration parameters (paging file size, for example)?at sub-optimal values. If you're filtering a lot of messages simultaneously, you might be chewing up a lot of memory.

-----Original Message-----
From: spambayes-bounces@python.org [mailto:spambayes-bounces@python.org]On Behalf Of Mei, Elaine
Sent: Tuesday, November 16, 2004 1:15 PM
To: spambayes@python.org
Subject: [Spambayes] Slow filtering junk mails



Hello,

?

I recently reformatted my harddrive and re-installed Windows XP, and put MS office and Spambayes back into the computer.?? What I noticed is that the filtering processing has been very slow since installing Spambayes.?? I tried software upgrade and deleting the data files for Spambayes and it didn?t solve the problem.? Any suggestions to proceed further with this?

?

?

?

Elaine Mei 
250 S. Wacker, Suite 1600 
Chicago, IL 60606-5889 USA 
Toll-free North America: 888-665-ATLA 
Chicago Area: 312-454-5100 
Website: http://www.atla.com 

?

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.python.org/pipermail/spambayes/attachments/20041116/8a044a77/attachment.htm
From cvitale at qualityoncology.com  Wed Nov 17 13:57:03 2004
From: cvitale at qualityoncology.com (Connie Vitale)
Date: Wed Nov 17 14:10:51 2004
Subject: [Spambayes] (no subject)
Message-ID: <BC32CA758269D41192C6009027B0D5430238814A@MAILSERVER>

Skipped content of type multipart/alternative-------------- next part --------------
A non-text attachment was scrubbed...
Name: Blank Bkgrd.gif
Type: image/gif
Size: 145 bytes
Desc: not available
Url : http://mail.python.org/pipermail/spambayes/attachments/20041117/8036cfba/BlankBkgrd.gif
From black at www.blackxtreme.nl  Wed Nov 17 15:25:32 2004
From: black at www.blackxtreme.nl (black@www.blackxtreme.nl)
Date: Wed Nov 17 15:25:35 2004
Subject: [Spambayes] Automated reply from black@www.blackxtreme.nl
Message-ID: <200411171425.iAHEPWx15853@ws5.vip.nl>

(dit is een automatische reply op uw mail)

Dank je voor je mail, je krijgt zsm reactie van me.
Thank you for your mail, i will contact you asap.


NOTE FOR YOU
________________________________________________________
Voor de leukste Striptease shows en Special acts bekend van tv
kijk snel op onze site en zoek de leukste shows uit

website:	www.BlackXtreme.nl
email:	Esther@BlackXtreme.nl (ook MSN)
mobiel:	0031(0)6 - 53 296 488 bgg 06 18 66 50 37
imail:	estherbx@imail.nl (to send a quicknote to my mobile)
MSN Patric: blackxcrow@msn.com

Please visite our website if you are looking for Special Acts for your party
BlackXtreme winner EroticAward2000 "the best performance of Europe"	
------------------------------------------------------------------------------------------------------------------
Speakers aan en kies je favoriete dancemusic:
www.blackxtreme.nl/patricmusicroom.html
xxx have fun
----------------------------------------------------------------------------------------------------------------
From rv at bakker-sl.nl  Wed Nov 17 15:46:17 2004
From: rv at bakker-sl.nl (Ronald Verheij)
Date: Wed Nov 17 15:46:12 2004
Subject: [Spambayes] Training spambayes,
Message-ID: <006101c4ccb4$31bc2550$aa640a0a@pcsbh2>

Hi,

I try to train my installed spambayes but if I brwose and choose an .dbx file it looks like it takes hours and it is not doing anything I think.
*************************************************
Ronald Verheij
Bakker Sliedrecht Electro Industrie
Telefoon: 0184 436671
Email: rv@bakker-sl.nl
Website: www.bakker-sl.nl
*************************************************

*************************************************************************************
** eSafe scanned this email for viruses, vandals and malicious content **
*************************************************************************************
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.python.org/pipermail/spambayes/attachments/20041117/ca872515/attachment.html
From EMei at atla.com  Wed Nov 17 16:17:05 2004
From: EMei at atla.com (Mei, Elaine)
Date: Wed Nov 17 16:17:50 2004
Subject: [Spambayes] Slow filtering junk mails
Message-ID: <2E1BDA7F69603A4885F621D302BA1B7DFDBA4F@email.atla.com>

Outlook is the client I use for email.   Before reformatting it, it can
filter incoming mails (Inbox of 100) in a few seconds.  Now the
filtration process is like 1 per second or longer.   When I have a few
hundred emails in inbox, it takes a quite some time to go.  

Elaine

-----Original Message-----
From: Tony Meyer [mailto:tameyer@ihug.co.nz] 
Sent: Tuesday, November 16, 2004 6:42 PM
To: Mei, Elaine; spambayes@python.org
Subject: RE: [Spambayes] Slow filtering junk mails

> I recently reformatted my harddrive and re-installed
> Windows XP, and put MS office and Spambayes back into
> the computer.   What I noticed is that the filtering
> processing has been very slow since installing Spambayes.
> I tried software upgrade and deleting the data files
> for Spambayes and it didn't solve the problem.  Any
> suggestions to proceed further with this?

My first suggestion would have been to remove the training data and
retrain,
unfortunately, but since you've already tried that...

What's slow?  Just filtering the messages as they arrive in the inbox?
Clicking on the "Delete as spam" and "Recover from spam" buttons as
well?
Starting up Outlook?

If it's just the former, then maybe it's just the background filtering
values?  If you go into the SpamBayes Manager dialog, then the Advanced
tab,
you can adjust these.  If you're not using any Outlook rules, then you
can
even disable background filtering altogether.

Otherwise - roughly how much slower would you say everything is?  Twice
as
slow?  10 times as slow?  There have been a few (though not many) other
reports of slowness, which is concerning, but there's no obvious cause
(at
least yet).  We're still looking into those, unfortunately...

=Tony.Meyer


From enewger at gmx.de  Wed Nov 17 16:52:49 2004
From: enewger at gmx.de (Eckhardt Newger)
Date: Wed Nov 17 16:53:19 2004
Subject: [Spambayes] Re: How to recover spam to a specific ham folder
Message-ID: <cnfvl3.164.1@ID-40943.user.dfncis.de>

Hi Tony,

Thanks for your reply.

No, I neither use Hotmail nor IMAP.

What I can see though is that all emails first show up in the main
entrance folder, then they are being further processed: Outlook rules
move them to the appropriate subfolders, Spambayes checks for ham/spam
status and moves them to the corresponding folders (spam or unsure
messages) if necessary. An explanation for the problem could be, if
Spambayes comes before Outlook, it moves the emails from the main folder
to the spam folder or to the unsure messages folder, an undo would just
put them back into the main folder.

I changed my Spymbayes configuration and took out the main folder as ham
folder, so Spambayes can only operate on the subfolders. Now it should
restore correctly, because now it knows the subfolder an email came
from. But Spambayes doesn't do as exspected: An email with target
subfolder B and moved by Spambayes to the spam folder is recovered to
subfolder A (the first one in Spambayes' ham folder list).

I have no clue what to do now.

Eckhardt Newger



-----Urspr?ngliche Nachricht-----
Von: Tony Meyer [mailto:tameyer@ihug.co.nz]
Gesendet: Mittwoch, 17. November 2004 01:08
An: Eckhardt Newger; spambayes@python.org
Betreff: RE: [Spambayes] How to recover spam to a specific ham folder

> Is there a way to specify for a false positive which subfolder an
> email will redirected to when using the 'Recover from spam' button? So

> far I noticed that all recovered emails end up in just one of those
> subfolders ( the first one I set up) not respecting the email address
> they have been send to.

The 'Recover from spam' button is meant to send the message back to the
folder it was in before SpamBayes moved it (here, presumably the correct
subfolder).  However, if it can't manage to do that, then it just sends
it back to the 'main' Inbox, whatever Outlook thinks that is.

Usually problems with figuring out the original folder occur with
Hotmail or IMAP stores - are you using one of those?

Your log files (SpamBayes->Help->Troubleshooting Guide explains where
those
are) might have additional clues about what is going wrong here.

=Tony.Meyer
From SBruchesky at indoff1.com  Wed Nov 17 21:19:26 2004
From: SBruchesky at indoff1.com (Bruchesky, Syndi)
Date: Wed Nov 17 21:27:30 2004
Subject: [Spambayes] no filterable mail items selected.
Message-ID: <6AB05F9D9EBCD311A41300508B72DDA703248846@www.indoff1.com>

I've been using SB for a while now and have had no problems.  I encourage
friends and family to use it as well.

One particular friend tried to install it and filter his pre-sorted emails
(SPAM and Good) folders.  It will not train on them.  It keeps saying 2
good, 2 spam, when there are actually about 50 in the SPAM folder and 25 in
the Inbox (good) folder.

 

The following is the latest log file for SB. I've tried installing and
uninstalling several times.  When I try to "delete it as spam"  one of the
emails in the Inbox, or "recover from SPAM" on a good message, I get the
error message "no filterable mail items selected".  However, on new emails,
all emails are going into the Junk Suspects and I CAN click and "delete as
spam" or "recover from spam".  Being that emails come in 20's and 30's or
more at a time, its going to take forever to train everything that comes in.

This pc is running XPPro, SP1, P4, 512MB Ram, Office 2003, SB ver1.0rc1.

Any ideas?  Here is the latest log file for your perusal. - Syndi

Loaded bayes database from 'C:\Documents and 
Settings\Administrator\Application 
Data\SpamBayes\default_bayes_database.db'
Loaded message database from 'C:\Documents and 
Settings\Administrator\Application 
Data\SpamBayes\default_message_database.db'
Bayes database initialized with 0 spam and 0 good 
messages
SpamBayes Outlook Addin Binary Version 1.0rc1 (May 
2004) starting (with engine SpamBayes Engine Version 
0.3 (January 2004))
on Windows 5.1.2600 (Service Pack 1)
using Python 2.3.3 (#51, Feb 13 2004, 14:39:56) [MSC 
v.1200 32 bit (Intel)]
SpamBayes: Watching (for filtering) in 'Personal Folders/Inbox'
SpamBayes: Watching (for incremental training) 
in 'Personal Folders/SPAM'Processing missed spam in folder 'Inbox' by
starting a 
timer
FAILED to add the toolbar 
item 'SpamBayesCommand.Manager' - (-
2147352567, 'Exception occurred.', (0, None, None, 
None, 0, -2147467259), None)
The above toolbar message is common - recreating the 
toolbar...
ERROR: 'No filterable mail items are selected'
SpamBayes - Dis

 

 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.python.org/pipermail/spambayes/attachments/20041117/d0d42a1e/attachment.htm
From postmaster at apotek.no  Wed Nov 17 21:45:40 2004
From: postmaster at apotek.no (postmaster@apotek.no)
Date: Wed Nov 17 21:55:50 2004
Subject: [Spambayes] Meldingen: Re: approved fra deg innehold en exe fil
Message-ID: <20041117205549.95D661E4006@bag.python.org>

Dette er en automatisk generert melding.

Meldingen var sent fra  spambayes@python.org  til f?lgende:
notteroy@apotek.no


P? grunn av virusfaren med exe-filer har meldingen blitt stoppet.
I disse dager kan du motta denne meldingen uten ? ha sendt en mail til mottakeren. Da er adressen din misbrukt av et virus.

Ta kontakt med intranettadministrasjon@apotek.no for sp?rsm?l

--

A mail message with subject "Re: approved" has been found containing an executable file.

The message was sent from  spambayes@python.org  to the following:
notteroy@apotek.no


Due to the danger of viruses in executable files the message has been stopped.
For more information contact intranettadministrasjon@apotek.no

From chanmo2 at bellsouth.net  Wed Nov 17 16:37:25 2004
From: chanmo2 at bellsouth.net (ch'an mo)
Date: Thu Nov 18 07:07:09 2004
Subject: [Spambayes] Off the List Please
Message-ID: <002401c4ccbb$5bc67560$22354ed8@system>

Please take me off the lists. No time to read them; Plus they don't help me.

 

Lucille Pelletier (Known also as ch'an mo)

 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.python.org/pipermail/spambayes/attachments/20041117/d043630d/attachment.htm
From kjelloff at broadpark.no  Thu Nov 18 10:55:09 2004
From: kjelloff at broadpark.no (Kjell G Offerdal)
Date: Thu Nov 18 10:54:27 2004
Subject: [Spambayes] Ten points
Message-ID: <IHEDIGADJPJJFCAFJCGFOEIPCGAA.kjelloff@broadpark.no>

Skipped content of type multipart/alternative-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: image/jpeg
Size: 2743 bytes
Desc: not available
Url : http://mail.python.org/pipermail/spambayes/attachments/20041118/578a8ee5/attachment-0001.jpe
From jsp at PKC.com  Thu Nov 18 13:36:46 2004
From: jsp at PKC.com (Jesse Pelton)
Date: Thu Nov 18 13:36:49 2004
Subject: [Spambayes] Off the List Please
Message-ID: <7f7d9ac074851b5508f88d4019703db4419c975f@vtmail.pkc.com>

Please see http://mail.python.org/mailman/listinfo/spambayes
<BLOCKED::http://mail.python.org/mailman/listinfo/spambayes> .
 
Does anyone know if the list manager could append the standard message
footer (with information about how to get to the list manager and the
FAQ) to HTML messages as well as plain text messages? That might
minimize this sort of request in the future.


________________________________

	From: spambayes-bounces@python.org
[mailto:spambayes-bounces@python.org] On Behalf Of ch'an mo
	Sent: Wednesday, November 17, 2004 10:37 AM
	To: Spambayes@python.org
	Subject: [Spambayes] Off the List Please
	
	

	Please take me off the lists. No time to read them; Plus they
don't help me.

	 

	Lucille Pelletier (Known also as ch'an mo)

	 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.python.org/pipermail/spambayes/attachments/20041118/b00b1573/attachment.htm
From rcoe at CambridgeMA.GOV  Thu Nov 18 16:02:33 2004
From: rcoe at CambridgeMA.GOV (Coe, Bob)
Date: Thu Nov 18 16:02:36 2004
Subject: [Spambayes] Off the List Please
Message-ID: <42393C9DA7930245AB540667607F4F5022C340@SPIKE.city>

What makes you think the people you're aiming at would read that footer if you did append it? After all, every person now trying to get off the list subscribed to the list in the first place, whereupon he was told how to get off. Plus, every subscriber receives a monthly reminder containing the same information. The users who don't read and/or file the reminders won't read the footers either.

-----Original Message-----
From: spambayes-bounces+rcoe=cambridgema.gov@python.org [mailto:spambayes-bounces+rcoe=cambridgema.gov@python.org]On Behalf Of Jesse Pelton
Sent: Thursday, November 18, 2004 7:37 AM
To: ch'an mo; Spambayes@python.org
Subject: RE: [Spambayes] Off the List Please


Please see http://mail.python.org/mailman/listinfo/spambayes <BLOCKED::http://mail.python.org/mailman/listinfo/spambayes> .
 
Does anyone know if the list manager could append the standard message footer (with information about how to get to the list manager and the FAQ) to HTML messages as well as plain text messages? That might minimize this sort of request in the future.


  _____  

From: spambayes-bounces@python.org [mailto:spambayes-bounces@python.org] On Behalf Of ch'an mo
Sent: Wednesday, November 17, 2004 10:37 AM
To: Spambayes@python.org
Subject: [Spambayes] Off the List Please



Please take me off the lists. No time to read them; Plus they don't help me.

 

Lucille Pelletier (Known also as ch'an mo)

 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.python.org/pipermail/spambayes/attachments/20041118/40d4bdf8/attachment.html
From jsp at PKC.com  Thu Nov 18 16:27:39 2004
From: jsp at PKC.com (Jesse Pelton)
Date: Thu Nov 18 16:27:43 2004
Subject: [Spambayes] Off the List Please
Message-ID: <ce2e0045ca0eae16aca4c41144ff7904419cbf6c@vtmail.pkc.com>

A vain hope, perhaps. But if it's not there, there's no hope at all.
Besides, it just seems appropriate. At present, every plain text message
forwarded by the list manager includes that footer, but many SpamBayes
users are composing HTML messages with Outlook, probably not even
realizing that they can compose plain text messages, or that there are
reasons for doing so. This means that on HTML-capable mail readers, a
substantial fraction of the messages do not display that footer. That's
a missed opportunity to educate.


________________________________

	From: Coe, Bob [mailto:rcoe@CambridgeMA.GOV] 
	Sent: Thursday, November 18, 2004 10:03 AM
	To: Jesse Pelton
	Cc: spambayes@python.org
	Subject: RE: [Spambayes] Off the List Please
	
	
	What makes you think the people you're aiming at would read that
footer if you did append it? After all, every person now trying to get
off the list subscribed to the list in the first place, whereupon he was
told how to get off. Plus, every subscriber receives a monthly reminder
containing the same information. The users who don't read and/or file
the reminders won't read the footers either.

		-----Original Message-----
		From: spambayes-bounces+rcoe=cambridgema.gov@python.org
[mailto:spambayes-bounces+rcoe=cambridgema.gov@python.org]On Behalf Of
Jesse Pelton
		Sent: Thursday, November 18, 2004 7:37 AM
		To: ch'an mo; Spambayes@python.org
		Subject: RE: [Spambayes] Off the List Please
		
		
		Please see
http://mail.python.org/mailman/listinfo/spambayes
<BLOCKED::http://mail.python.org/mailman/listinfo/spambayes> .
		 
		Does anyone know if the list manager could append the
standard message footer (with information about how to get to the list
manager and the FAQ) to HTML messages as well as plain text messages?
That might minimize this sort of request in the future.


________________________________

			From: spambayes-bounces@python.org
[mailto:spambayes-bounces@python.org] On Behalf Of ch'an mo
			Sent: Wednesday, November 17, 2004 10:37 AM
			To: Spambayes@python.org
			Subject: [Spambayes] Off the List Please
			
			

			Please take me off the lists. No time to read
them; Plus they don't help me.

			 

			Lucille Pelletier (Known also as ch'an mo)

			 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.python.org/pipermail/spambayes/attachments/20041118/2e9f66d8/attachment.htm
From tim.peters at gmail.com  Thu Nov 18 17:20:06 2004
From: tim.peters at gmail.com (Tim Peters)
Date: Thu Nov 18 17:20:12 2004
Subject: [Spambayes] Off the List Please
In-Reply-To: <ce2e0045ca0eae16aca4c41144ff7904419cbf6c@vtmail.pkc.com>
References: <ce2e0045ca0eae16aca4c41144ff7904419cbf6c@vtmail.pkc.com>
Message-ID: <1f7befae0411180820467b5536@mail.gmail.com>

Mailman already attaches the list info to HTML msgs, in a

    Content-Type: text/plain; charset="us-ascii"
    MIME-Version: 1.0
    Content-Transfer-Encoding: 7bit
    Content-Disposition: inline

MIME section, part of a top-level MIME multipart/mixed structure. If
an email reader doesn't display the list info inline, as it was asked
to do, then I'd call it a flaw in that email client.  Mailman also
includes a List-Unsubscribe header in every message it sends, spelling
out two ways to unsubscribe.

I agree with Bob that nothing would make a real difference.  A small
percentage of subscribers have no idea how to unsubscribe, and decades
of experience on many lists says that will never change, no matter how
clear or prominent the instructions.

For the same reasons, a small percentage of people still open virus
attachments, and there's even emprical evidence backing the fantastic
claim that some still send money in response to spam <wink>.

the-problem-here-ain't-technical-it's-human-ly y'rs  - tim
From Ow.Mun.Heng at wdc.com  Thu Nov 18 17:22:55 2004
From: Ow.Mun.Heng at wdc.com (Ow Mun Heng)
Date: Thu Nov 18 17:24:24 2004
Subject: [Spambayes] Re: confirm 7392dc2ec2fcb9b1c845e211e5d22d15699ecd3e
In-Reply-To: <mailman.0.1100794514.10199.spambayes@python.org>
References: <mailman.0.1100794514.10199.spambayes@python.org>
Message-ID: <1100794975.4518.79.camel@localhost>

On Fri, 2004-11-19 at 00:15, spambayes-request@python.org wrote:
> Mailing list subscription confirmation notice for mailing list
> Spambayes
> 
> We have received a request from 60.48.86.188 for subscription of your
> email address, "ow.mun.heng@wdc.com", to the spambayes@python.org
> mailing list.  To confirm that you want to be added to this mailing
> list, simply reply to this message, keeping the Subject: header
> intact.  Or visit this web page:
> 
>     http://mail.python.org/mailman/confirm/spambayes/7392dc2ec2fcb9b1c845e211e5d22d15699ecd3e
> 
> 
> Or include the following line -- and only the following line -- in a
> message to spambayes-request@python.org:
> 
>     confirm 7392dc2ec2fcb9b1c845e211e5d22d15699ecd3e
> 
> Note that simply sending a `reply' to this message should work from
> most mail readers, since that usually leaves the Subject: line in the
> right form (additional "Re:" text in the Subject: is okay).
> 
> If you do not wish to be subscribed to this list, please simply
> disregard this message.  If you think you are being maliciously
> subscribed to the list, or have any other questions, send them to
spambayes-owner@python.org.

-- 
Ow Mun Heng
Gentoo/Linux on D600 1.4Ghz 
CPU kernel 2.6.9-gentoo-r1 
Neuromancer 00:22:46 up 3:41, 2 users, load average: 0.19, 0.35, 0.39 

From jsp at PKC.com  Thu Nov 18 18:30:49 2004
From: jsp at PKC.com (Jesse Pelton)
Date: Thu Nov 18 18:30:52 2004
Subject: [Spambayes] Off the List Please
Message-ID: <8991024abacc1dd02c7652c37be247b9419cdc4a@vtmail.pkc.com>

Huh.  I'm stuck with Outlook here at work, and it doesn't display the
footer (maybe because it's text/plain?).  Perhaps part of the solution
is indeed technical, but in Microsoft's court.  I imagine a lot of the
people who ask how to get off the list are also using Outlook and cannot
see the footer.  You can lead a horse to water, but you can't make him
drink, especially if (as in this case) he doesn't know it's there.  I
wonder what would happen if there were a text/html inline attachment.

Anyway, it was just an idea, intended to reduce the load on the good
people who respond to postings to this list.  If it's a bad or
unworkable idea, I'm happy to release it back into the nothingness from
which it came. 

> -----Original Message-----
> From: Tim Peters [mailto:tim.peters@gmail.com] 
> Sent: Thursday, November 18, 2004 11:20 AM
> To: Jesse Pelton
> Cc: spambayes@python.org
> Subject: Re: [Spambayes] Off the List Please
> 
> Mailman already attaches the list info to HTML msgs, in a
> 
>     Content-Type: text/plain; charset="us-ascii"
>     MIME-Version: 1.0
>     Content-Transfer-Encoding: 7bit
>     Content-Disposition: inline
> 
> MIME section, part of a top-level MIME multipart/mixed structure. If
> an email reader doesn't display the list info inline, as it was asked
> to do, then I'd call it a flaw in that email client.  Mailman also
> includes a List-Unsubscribe header in every message it sends, spelling
> out two ways to unsubscribe.
> 
> I agree with Bob that nothing would make a real difference.  A small
> percentage of subscribers have no idea how to unsubscribe, and decades
> of experience on many lists says that will never change, no matter how
> clear or prominent the instructions.
> 
> For the same reasons, a small percentage of people still open virus
> attachments, and there's even emprical evidence backing the fantastic
> claim that some still send money in response to spam <wink>.
> 
> the-problem-here-ain't-technical-it's-human-ly y'rs  - tim
> 
From lynn at herecomestheguide.com  Thu Nov 18 20:39:31 2004
From: lynn at herecomestheguide.com (Lynn Broadwell)
Date: Thu Nov 18 20:57:22 2004
Subject: [Spambayes] Will Spambayse work on a Mac?
Message-ID: <BDC23A73.6BF8%lynn@herecomestheguide.com>

Thanks in advance for your fast reply.
Lynn Broadwell, Publisher

Hopscotch Press, Inc.
Publisher of Here Comes The Guide
930 Carleton Street, Berkeley CA 94710
510/548-0400, fax 510/548-0144
http://www.herecomestheguide.com
lynn@herecomestheguide.com



From skip at pobox.com  Thu Nov 18 21:06:33 2004
From: skip at pobox.com (Skip Montanaro)
Date: Thu Nov 18 21:06:17 2004
Subject: [Spambayes] Off the List Please
In-Reply-To: <8991024abacc1dd02c7652c37be247b9419cdc4a@vtmail.pkc.com>
References: <8991024abacc1dd02c7652c37be247b9419cdc4a@vtmail.pkc.com>
Message-ID: <16797.201.159420.638579@montanaro.dyndns.org>


    Jesse> Huh.  I'm stuck with Outlook here at work, and it doesn't display
    Jesse> the footer (maybe because it's text/plain?).  Perhaps part of the
    Jesse> solution is indeed technical, but in Microsoft's court.  

(Should I try channelling Tim?  Sounds dangerous...)

Tim uses Outlook.  Here are some telltale headers from an old message of his
I happened to have laying about:

    X-MSMail-Priority: Normal
    X-Mailer: Microsoft Outlook IMO, Build 9.0.6604 (9.0.2911.0)
    X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2800.1165

While I'm sure he knows Mailman's behavior well enough to not need to see
header fields or the footer to know what they contain, I suppose he may also
know how to drive Outlook well enough to expose those bits.  Personally, I
stay far enough away from Windows that I haven't any idea how to tell you to
do that.  Tim, Tony or one of the other Outlook users will have to provide
those tips...

    Jesse> You can lead a horse to water, but you can't make him drink,
    Jesse> especially if (as in this case) he doesn't know it's there.

Isn't that why horses have long noses?  So they can smell water a good ways
off? <wink>

Skip
From Amir_Katz at bmc.com  Thu Nov 18 21:11:26 2004
From: Amir_Katz at bmc.com (Katz, Amir)
Date: Thu Nov 18 21:11:38 2004
Subject: [Spambayes] Spambayes proxy server error
Message-ID: <F8B9B12641FB6B4FAC761BDEF9D63B04031EC0A5@tlv-ex-01.adprod.bmc.com>

Hi, I can't get the server to train. I tried to mark all messages as
'discard', but clicking 'train' fails with the following output.

This is SB 1.0 on Win XP with Outlook express.

Anyone?

Thanks.
 
==================================================================== 

500 Server error
Traceback (most recent call last):

  File "spambayes\Dibbler.pyc", line 461, in found_terminator

  File "spambayes\ProxyUI.pyc", line 391, in onReview

  File "spambayes\Corpus.pyc", line 214, in takeMessage

  File "spambayes\FileCorpus.pyc", line 140, in addMessage

  File "spambayes\Corpus.pyc", line 134, in addMessage

  File "spambayes\storage.pyc", line 588, in onAddMessage

  File "spambayes\storage.pyc", line 598, in train

  File "spambayes\message.pyc", line 260, in setId

  File "spambayes\message.pyc", line 118, in _getState

  File "shelve.pyc", line 118, in __getitem__

  File "bsddb\__init__.pyc", line 116, in __getitem__

DBRunRecoveryError: (-30982, 'DB_RUNRECOVERY: Fatal error, run database
recovery -- fatal region error detected; run recovery')

From skip at pobox.com  Thu Nov 18 21:13:19 2004
From: skip at pobox.com (Skip Montanaro)
Date: Thu Nov 18 21:12:19 2004
Subject: [Spambayes] Will Spambayse work on a Mac?
In-Reply-To: <BDC23A73.6BF8%lynn@herecomestheguide.com>
References: <BDC23A73.6BF8%lynn@herecomestheguide.com>
Message-ID: <16797.607.200371.92436@montanaro.dyndns.org>


    Lynn> Thanks in advance for your fast reply.

I use it on my Mac.  I happen to use the Unix side of things, but the
SpamBayes POP3 proxy also works.  There is no tight integration available
for Mac Mail like there is for Outlook on Windows however.

-- 
Skip Montanaro
skip@mojam.com
http://www.mojam.com/
From tim.peters at gmail.com  Thu Nov 18 22:32:52 2004
From: tim.peters at gmail.com (Tim Peters)
Date: Thu Nov 18 22:32:56 2004
Subject: [Spambayes] Off the List Please
In-Reply-To: <16797.201.159420.638579@montanaro.dyndns.org>
References: <8991024abacc1dd02c7652c37be247b9419cdc4a@vtmail.pkc.com>
	<16797.201.159420.638579@montanaro.dyndns.org>
Message-ID: <1f7befae041118133239a378d3@mail.gmail.com>

[Skip Montanaro]
> (Should I try channelling Tim?  Sounds dangerous...)

I predict you'll survive <wink>.

> Tim uses Outlook.
...
> While I'm sure he knows Mailman's behavior well enough to not need
> to see header fields or the footer to know what they contain, I
> suppose he may also know how to drive Outlook well enough to
> expose those bits.

I actually don't know.  I read my spambayes subscription via a gmail
account these days, and the list info is definitely visible at the
bottom of HTML msgs from the spambayes list viewed via gmail.  gmail
does with it what it's supposed to do with it.

All versions of Outlook (but not of Outlook Express) destroy incoming
MIME structure in a non-reversible way, and it could very well be that
they simply throw away the list info Mailman is trying to display when
they receive an HTML msg.  The info is still in the headers then (most
easily revealed by doing a "show clues" via SpamBayes), but Outlook
does make those clumsy to get at.

So, for all I know, it's true that no Outlook user ever sees the list
info on HTML msgs.  There would be little Mailman-community knowledge
about this, because by default Mailman throws away all text/html MIME
sections, reflecting (I guess) geeks' widespread hatred of HTML email;
the spambayes lists are rare exceptions.

But Outlook certainly *does* show the list info on plain-text msgs
sent to the list; I have thousands of those archived in my Outlook,
and you don't have to do anything special to see it.  Despite that we
allow it here, relatively little traffic to this list is in HTML
format, so "even Outlook users" <wink> don't have much excuse.

Anyway, it's not something I want to spend more time on.  The admin
pages make clear that an overwhelmingly majority of folks who want to
unsubscribe manage to do so on their own.
From rcoe at CambridgeMA.GOV  Fri Nov 19 13:56:08 2004
From: rcoe at CambridgeMA.GOV (Coe, Bob)
Date: Fri Nov 19 13:56:11 2004
Subject: [Spambayes] RE: Off the List Please
Message-ID: <42393C9DA7930245AB540667607F4F5022C341@SPIKE.city>

In my environment (Outlook 2000 and Exchange Server) HTML messages from the list have the footer as a plaintext attachment. But in any case, the reference in the footer is only an unexplained link. I doubt that an unsophisticated user will realize that that's where the unsubscribe information is.

Bob


> -----Original Message-----
> From: spambayes-bounces+rcoe=cambridgema.gov@python.org
> [mailto:spambayes-bounces+rcoe=cambridgema.gov@python.org]On Behalf Of
> Tim Peters
> Sent: Thursday, November 18, 2004 4:33 PM
> To: skip@pobox.com
> Cc: spambayes@python.org
> Subject: Re: [Spambayes] Off the List Please
> 
> 
> [Skip Montanaro]
> > (Should I try channelling Tim?  Sounds dangerous...)
> 
> I predict you'll survive <wink>.
> 
> > Tim uses Outlook.
> ...
> > While I'm sure he knows Mailman's behavior well enough to not need
> > to see header fields or the footer to know what they contain, I
> > suppose he may also know how to drive Outlook well enough to
> > expose those bits.
> 
> I actually don't know.  I read my spambayes subscription via a gmail
> account these days, and the list info is definitely visible at the
> bottom of HTML msgs from the spambayes list viewed via gmail.  gmail
> does with it what it's supposed to do with it.
> 
> All versions of Outlook (but not of Outlook Express) destroy incoming
> MIME structure in a non-reversible way, and it could very well be that
> they simply throw away the list info Mailman is trying to display when
> they receive an HTML msg.  The info is still in the headers then (most
> easily revealed by doing a "show clues" via SpamBayes), but Outlook
> does make those clumsy to get at.
> 
> So, for all I know, it's true that no Outlook user ever sees the list
> info on HTML msgs.  There would be little Mailman-community knowledge
> about this, because by default Mailman throws away all text/html MIME
> sections, reflecting (I guess) geeks' widespread hatred of HTML email;
> the spambayes lists are rare exceptions.
> 
> But Outlook certainly *does* show the list info on plain-text msgs
> sent to the list; I have thousands of those archived in my Outlook,
> and you don't have to do anything special to see it.  Despite that we
> allow it here, relatively little traffic to this list is in HTML
> format, so "even Outlook users" <wink> don't have much excuse.
> 
> Anyway, it's not something I want to spend more time on.  The admin
> pages make clear that an overwhelmingly majority of folks who want to
> unsubscribe manage to do so on their own.
> _______________________________________________
> Spambayes@python.org
> http://mail.python.org/mailman/listinfo/spambayes
> Check the FAQ before asking: http://spambayes.sf.net/faq.html
> 
From jsp at PKC.com  Fri Nov 19 14:40:48 2004
From: jsp at PKC.com (Jesse Pelton)
Date: Fri Nov 19 14:40:51 2004
Subject: [Spambayes] RE: Off the List Please
Message-ID: <5f63155931ac000b1f540b2ce8a49181419df7db@vtmail.pkc.com>

Ouch!  I don't think of myself as an unsophisticated user, and I had no
idea what was in those attachments.  I never open attachments unless I
have good reason to AND I'm certain of their source, and I think many
e-mail users are equally cautious.  Thanks for clarifying that.

It's a good news/bad news situation for Outlook users.  Outlook makes
the footer available for HTML messages, but in a form that virtually no
one is ever going to see, since it ignores the inline disposition.

I think Tim's observation that almost everyone who wants to unsubscribe
figures it out on their own supports Bob's contention that some people
won't figure it out whatever you do, so even if the footers of HTML
messages were easily visible to Outlook users, it would probably make
little difference.  (Short form: I'll shut up now.  Thanks for bearing
with me.)
From ljw674t at smsu.edu  Fri Nov 19 17:19:00 2004
From: ljw674t at smsu.edu (Wright, Lyle J)
Date: Fri Nov 19 17:19:13 2004
Subject: [Spambayes] It all started with SP2
Message-ID: <4170477726671041983398A3D169E68FC4C360@mercury.WESTPLAINS.SMSU.EDU>

Could you please help me get SpamBayes to work again?
I've been using it successfully for over a year and once Microsoft SP2
was installed it quit working. Now, SpamBayes loads correctly, assigns
Spam values correctly, puts messages in the "Spam" and "Possible Spam"
folders correctly, but if I try to access any functions, it performs the
function but then it also shuts down Outlook. Thus, I can not click on
any Spambayes toolbar buttons (or dropdown list to configure) without it
shutting down Outlook as it performs the function.
I have read through your troubleshooting guide, reset the Outlook
toolbars by deleting outcmd.dat. Uninstalled SpamBayes and reinstalled
it. Nothing works to eliminate this problem. I also registered on
SourceForge to study the bugs there. It is similar to 956834, but not
the same. I am stuck and would appreciate your insights. Since I am not
a programmer, I don't know where to start to try to fix this. In the log
file below, I recovered a message from 'Possible Spam' folder, and it
performed the function, but then shut down Outlook.
*	the version of Windows you are using-Window XP Version 2002-SP2
*	the version of SpamBayes - Version 1.0
*	any log files. - see below
Warning: option experimental_ham_spam_imbalance_adjustment in section
Classifier is deprecated
Loaded bayes database from 'C:\Documents and
Settings\ljw674t\Application Data\SpamBayes\default_bayes_database.db'
Loaded message database from 'C:\Documents and
Settings\ljw674t\Application Data\SpamBayes\default_message_database.db'
Bayes database initialized with 574 spam and 210 good messages
SpamBayes Outlook Addin Binary Version 1.0 (July 2004) starting (with
engine SpamBayes Engine Version 0.3 (January 2004))
on Windows 5.1.2600 (Service Pack 2)
using Python 2.3.3 (#51, Dec 18 2003, 20:22:39) [MSC v.1200 32 bit
(Intel)]
SpamBayes: Watching (for filtering) in 'Mailbox - Wright, Lyle J/Inbox'
SpamBayes: Watching (for filtering) in 'Lyle's PC/Inbox'
SpamBayes: Watching (for incremental training) in 'Mailbox - Wright,
Lyle J/Spam'
Processing missed spam in folder 'Inbox' by starting a timer
Processing missed spam in folder 'Inbox' by starting a timer
Creating new SpamBayes toolbar to host our buttons
Recovering to folder 'Inbox' and ham training message 'AWADmail Issue
143' -  Training on message 'AWADmail Issue 143' in 'Mailbox - Wright,
Lyle J/Possible Spam -  trained as good


 
Thanks - Lyle
***********************
Lyle J. Wright - Coordinator & Adjunct Faculty
Small Business Development Center (SBDC)
Southwest Missouri State University - West Plains
128 Garfield
West Plains, MO 65775
FON: (417) 255-7966
FAX: (417) 255-7963
Missouri Business Development Network  http://www.missouribusiness.net/ 
***********************


From crescent at lungsfordanelle.com  Sat Nov 20 17:22:41 2004
From: crescent at lungsfordanelle.com (Crescent DeCiantis)
Date: Sat Nov 20 17:22:39 2004
Subject: [Spambayes] Outlook '97?
Message-ID: <007101c4cf1d$292a0f40$72c70944@cx1260346a>

Is there anyway this could work with Outlook '97?  I'd like to help my wife 
with her spam issues. 


From rolph at 110.net  Sat Nov 20 18:07:00 2004
From: rolph at 110.net (The Rolphs)
Date: Sat Nov 20 18:07:13 2004
Subject: [Spambayes] Outlook '97?
In-Reply-To: <007101c4cf1d$292a0f40$72c70944@cx1260346a>
Message-ID: <5.1.0.14.2.20041120120627.020b2138@localhost>

Have you considered using the pop3 server model?  I am using this with 
eudora quite successfully.

At 11:22 AM 11/20/2004 -0500, Crescent DeCiantis wrote:
>Is there anyway this could work with Outlook '97?  I'd like to help my 
>wife with her spam issues.
>
>_______________________________________________
>Spambayes@python.org
>http://mail.python.org/mailman/listinfo/spambayes
>Check the FAQ before asking: http://spambayes.sf.net/faq.html


Regards.

The Rolphs rolph@110.net



From cowgirlup72701 at yahoo.com  Sun Nov 21 00:39:21 2004
From: cowgirlup72701 at yahoo.com (cindy preston)
Date: Sun Nov 21 00:39:25 2004
Subject: [Spambayes] Note on SPAM 
Message-ID: <20041120233921.83335.qmail@web50510.mail.yahoo.com>

hey,sugar im missing you
wish we had met be-4 we got like we r  now.i could love you i think u r a sexy man so good looking
i want to lay in that bed with you and hold you
 how long sence you had a girl?
can u have sex?
 would u want to?
want to be your baby
cindy
 

__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.python.org/pipermail/spambayes/attachments/20041120/078a45d1/attachment.html
From cowgirlup72701 at yahoo.com  Sun Nov 21 00:45:52 2004
From: cowgirlup72701 at yahoo.com (cindy preston)
Date: Sun Nov 21 00:45:56 2004
Subject: [Spambayes] Note on SPAM 
Message-ID: <20041120234552.80783.qmail@web50502.mail.yahoo.com>

i lost tou,how do i keep u where i can find you easy/
 

__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.python.org/pipermail/spambayes/attachments/20041120/1f524c90/attachment.htm
From f.orourke at optusnet.com.au  Sun Nov 21 21:51:16 2004
From: f.orourke at optusnet.com.au (Frank O'Rourke)
Date: Sun Nov 21 21:46:42 2004
Subject: [Spambayes] Problem with POP3 Proxy: (Incredimail instillation)
Message-ID: <000601c4d00b$d9f8dec0$a1aa1dd3@oemcomputer>


-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.python.org/pipermail/spambayes/attachments/20041122/b0133bf0/attachment.htm
From rmalayter at bai.org  Mon Nov 22 01:08:22 2004
From: rmalayter at bai.org (Ryan Malayter)
Date: Mon Nov 22 01:08:24 2004
Subject: [Spambayes] Outlook 2003 is eating our lunch ?
Message-ID: <792DE28E91F6EA42B4663AE761C41C2A0336C97D@cliff.bai.org>

I saw a presentation at Microsoft TechEd a couple of years back about
the spam filter technology going in to Outlook 2003 / Exchange 2003. It
did not impress me that much. From what I recall, they use a positive
and negative scoring system based on tokens. The tokens and their
"weights" are determined collectively by the huge number of Hotmail
users marking messages as spam and ham. I recall the positive/negative
token wieghts were combined with simple addition in the examples
presented; I am not sure if that's how it works in the actual code
version.

Basically, it sounded a lot like SpamAssasin's point-based rule system.
The reason it works reasonably well for a lot of people is because of
the huge Hotmail corpus that they use to statistically determine the
point values for each token. They have also released a few updates to
the rule sets since Outlook 2003 RTM, these perform marginally better
against current spam.

That said, I get something near 98% capture rate with zero false
positives using SpamBayes, so I won't be ditching it anytime soon in
favor of Outlook 2003's filter or the Exchange 2003 Intelligent Message
Filter.

Regards,
	Ryan

________________________________

	From: spambayes-bounces@python.org
[mailto:spambayes-bounces@python.org] On Behalf Of Jesse Pelton
	Sent: Wednesday, November 10, 2004 7:16 AM
	To: Katz, Amir; spambayes@python.org
	Subject: RE: [Spambayes] Outlook 2003 is eating our lunch ?
	
	
	I imagine there will be a place for the Outlook plugin for some
time to come. I'm not quite sure why Microsoft hasn't implemented
Bayesian filtering; it's easy to understand, easy to use, and highly
effective (at least as implemented in SpamBayes). If Outlook caught 204
out of 304 spams, it may not be half bad, but it's about a third bad.
SpamBayes does much better for me, and if it doesn't, it's not hard to
retrain it. With Outlook's filtering, you take what you get.


________________________________

		From: spambayes-bounces@python.org
[mailto:spambayes-bounces@python.org] On Behalf Of Katz, Amir
		Sent: Wednesday, November 10, 2004 4:15 AM
		To: spambayes@python.org
		Subject: [Spambayes] Outlook 2003 is eating our lunch ?
		
		
		I'm no big fan of Micro$oft, but I must admit that the
junk mail filter that comes with Outlook 2003 client is not half bad. It
intercepts almost all e-mails which are in Korean and other Asian
languages, as well as many of the normal spam. From a quick count that I
did, in the last 48 hours or so I got 304 spams, and Outlook intercepted
204 of them.
		 
		Amir 

From tim.peters at gmail.com  Mon Nov 22 01:48:26 2004
From: tim.peters at gmail.com (Tim Peters)
Date: Mon Nov 22 01:48:29 2004
Subject: [Spambayes] Outlook 2003 is eating our lunch ?
In-Reply-To: <792DE28E91F6EA42B4663AE761C41C2A0336C97D@cliff.bai.org>
References: <792DE28E91F6EA42B4663AE761C41C2A0336C97D@cliff.bai.org>
Message-ID: <1f7befae04112116484745a726@mail.gmail.com>

[Ryan Malayter]
> I saw a presentation at Microsoft TechEd a couple of years back about
> the spam filter technology going in to Outlook 2003 / Exchange 2003. It
> did not impress me that much. ...

There used to be great info here:

    http://www.mapilab.com/articles/outlook_spam_filter.html

They reverse-engineered the code and data files, and learned a lot
about 2003's filtering internals.  Alas, it's all been removed now
("due to legal issue").

> ...
> The reason it works reasonably well for a lot of people is because of
> the huge Hotmail corpus that they use to statistically determine the
> point values for each token. They have also released a few updates to
> the rule sets since Outlook 2003 RTM, these perform marginally better
> against current spam.

The biggest problem is that it doesn't learn, and everyone's 2003
comes with the same scoring.  That includes copies of 2003 that
spammers buy, so they can easily fine-tune their campaign to evade
2003's filtering just by sending msgs to themselves to see how they
score.  Of course they score the same for everyone else too.

> ...
> That said, I get something near 98% capture rate with zero false
> positives using SpamBayes, so I won't be ditching it anytime soon in
> favor of Outlook 2003's filter or the Exchange 2003 Intelligent Message
> Filter.

Me neither.  2003's filter is fast, fixed-size, and does a very good
job on the spam it was trained against.  That's about it.  If the
spammers would just agree to send nothing except messages 2003 was
trained it, it would be even better than SB <wink>.
From tameyer at ihug.co.nz  Mon Nov 22 05:07:02 2004
From: tameyer at ihug.co.nz (Tony Meyer)
Date: Mon Nov 22 05:07:38 2004
Subject: [Spambayes] Spambayes proxy server error
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E801763A28@its-xchg4.massey.ac.nz>
Message-ID: <ECBA357DDED63B4995F5C1F5CBE5B1E86C37F5@its-xchg4.massey.ac.nz>

> Hi, I can't get the server to train. I tried to mark all 
> messages as 'discard', but clicking 'train' fails with the 
> following output.
> 
> This is SB 1.0 on Win XP with Outlook express.
[...]
> DBRunRecoveryError: (-30982, 'DB_RUNRECOVERY: Fatal error, 
> run database recovery -- fatal region error detected; run recovery')

This (unfortunately) is a case of FAQ 5.2:

<http://spambayes.org/faq.html#i-get-a-dbrunrecoveryerror-message>

=Tony.Meyer

-- 
Please always include the list (spambayes@python.org) in your replies
(reply-all), and please don't send me personal mail about SpamBayes.
http://www.massey.ac.nz/~tameyer/writing/reply_all.html explains this.

From tameyer at ihug.co.nz  Mon Nov 22 05:10:44 2004
From: tameyer at ihug.co.nz (Tony Meyer)
Date: Mon Nov 22 05:11:19 2004
Subject: [Spambayes] Yahoo addresses
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E801763A03@its-xchg4.massey.ac.nz>
Message-ID: <ECBA357DDED63B4995F5C1F5CBE5B1E86C37F6@its-xchg4.massey.ac.nz>

[Tony Meyer]
>> If you use the "show clues for this message" function on one
>> of those (before training), you might be able to understand 
>> why they are scoring what they do.  If not, feel free to 
>> forward one of those 'show clues' messages to the list, and 
>> we'll try and explain what's going on.

[Bob Murdoch]
> Thanks for the offer.  Here's one that is sitting in the Junk E-mail
> folder right now:
> 
> Combined Score: 0% (1.11022e-016)
> Internal ham score (*H*): 1
> Internal spam score (*S*): 2.22045e-016
[...]

This has a score of 0%, which is as non-junk as it can get.  Have you done
any training since this message was filtered?  (The "show clues" function
shows what the message would currently score, not what it originally did).
If not, did the message definitely end up in the junk folder?  If so, then
if could you send us a copy of your most recent log file
(SpamBayes->Help->Troubleshooting Guide explains where they are) when it
happens next that would be great, since that shouldn't happen.

=Tony.Meyer

-- 
Please always include the list (spambayes@python.org) in your replies
(reply-all), and please don't send me personal mail about SpamBayes.
http://www.massey.ac.nz/~tameyer/writing/reply_all.html explains this.

From tameyer at ihug.co.nz  Mon Nov 22 05:32:26 2004
From: tameyer at ihug.co.nz (Tony Meyer)
Date: Mon Nov 22 05:33:03 2004
Subject: [Spambayes] Ten points
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E801763A1A@its-xchg4.massey.ac.nz>
Message-ID: <ECBA357DDED63B4995F5C1F5CBE5B1E8FBCA25@its-xchg4.massey.ac.nz>

> I had Spam problems - problem solved with Your program. Thanks!!!!!!

You're welcome.  On behalf of the team, thanks for the kind words :)

=Tony.Meyer

From ligon123 at caramail.com  Mon Nov 22 12:26:04 2004
From: ligon123 at caramail.com (=?windows-1252?Q?Chico de Par=EDs?=)
Date: Mon Nov 22 12:26:16 2004
Subject: [Spambayes] How to save Spambayes filter ?
Message-ID: <1101122764028690@lycos-europe.com>

Hi,
I am currently assesing the possibility to use Spambayes. It looks light, performant... but how can I save the filter if I am to change my PC ?
Where is it exported... on the pst file ?
Does the filter grow ... grow ... grow or is it possible to "compact" it ?
What if I migrate from Outlook XP to Outlook 2003 concerning the filter, can I keep it or will I have to start from 0 ?

Regards,
Leenull

C est le moment de dynamiser votre bo?te mail en d?couvrant les offres CaraMail Max et Pro - http://www.caramail.com

From bunkie at stevensleinweber.com  Mon Nov 22 16:52:36 2004
From: bunkie at stevensleinweber.com (Bunkie McBride)
Date: Mon Nov 22 16:53:22 2004
Subject: [Spambayes] Problems --- 
Message-ID: <PJEBKIPJPGHIKMEMAJAOKEOFCGAA.bunkie@stevensleinweber.com>

I cannot get the Spambayes program to load completely.  I get an error
message at the very end saying it could not load the addin to Outlook.

Bunkie


-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.python.org/pipermail/spambayes/attachments/20041122/874b1bf4/attachment.htm
From usr7-28hl at xemaps.com  Mon Nov 22 21:05:32 2004
From: usr7-28hl at xemaps.com (usr7-28hl@xemaps.com)
Date: Mon Nov 22 21:05:38 2004
Subject: [Spambayes] Unable to Access SpamBayes Browser Interface Using
	sb_imapfilter.py
In-Reply-To: <418C6A07.8050105@mailsnare.net>
References: <ECBA357DDED63B4995F5C1F5CBE5B1E8FBC9B0@its-xchg4.massey.ac.nz>
	<418C6A07.8050105@mailsnare.net>
Message-ID: <41A2468C.60400@mailsnare.net>

Hello, everyone.

Did my message (quoted below) get lost, or did I just miss the 
replies again?

Any help would be GREATLY appreciated.

-- 
Keith

I wrote:

> Hi, Tony.
> 
> I seem to be having trouble either posting to the list or receiving my 
> list posts. I posted a reply yesterday and haven't seen it; I apologize 
> is this is a repeat.
> 
> Tony Meyer wrote:
> 
>> There was a response from Scott Burns:
>>
>> <http://mail.python.org/pipermail/spambayes/2004-November/015109.html>
>>
>> He's correct that the problem here:
>>
>>
>>> 3. >telnet localhost:8880
>>> Connecting to localhost:8880...Could not open connection to the host, 
>>> on post 23: Connect failed
> 
> 
> For some reason, I didn't receive Scott's reply. (Thanks anyway, Scott!) 
> I even searched my entire account for a message from him and didn't see 
> one.
> 
>> Is that you need to do "telnet localhost 8880" (no colon) in order to 
>> test
>> the connection.
> 
> 
> I knew that! ;-) It is a little embarrassing; the "on post 23" is 
> somewhat of a giveaway, isn't it? Anyway, I'm pretty sure I did test 
> this using both formats at one time....
> 
>> When you know if you can telnet to it, you'll know whether the web 
>> interface
>> isn't being served, or if it's a problem connecting to it in your 
>> browser.
>> The latter is likely to be something to do with proxies or firewalls that
>> you are using.  The former could be a SpamBayes bug.
> 
> 
> A "telnet localhost 8880" also fails. So where do we go from here, since 
> it seems as though I'm the only one with this bug, or whatever it may be?
> 
> Thanks.
> 

From richie at entrian.com  Mon Nov 22 22:20:11 2004
From: richie at entrian.com (Richie Hindle)
Date: Mon Nov 22 22:20:14 2004
Subject: [Spambayes] Unable to Access SpamBayes Browser Interface Using
	sb_imapfilter.py
In-Reply-To: <41887AB5.2030008@xemaps.com>
References: <41887AB5.2030008@xemaps.com>
Message-ID: <rsl4q0pqvl3avisechaep2rqclb0clgknj@4ax.com>


[Keith]
> 3. >telnet localhost:8880
> Connecting to localhost:8880...Could not open connection to the
> host, on post 23: Connect failed

Your command line there is wrong - telnet expects a space between the
server name and the port number rather than a colon, like this:

  telnet localhost 8880

Could you try that and let us know the result?

-- 
Richie Hindle
richie@entrian.com

From tameyer at ihug.co.nz  Tue Nov 23 00:06:40 2004
From: tameyer at ihug.co.nz (Tony Meyer)
Date: Tue Nov 23 00:07:20 2004
Subject: [Spambayes] How to save Spambayes filter ?
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E801764559@its-xchg4.massey.ac.nz>
Message-ID: <ECBA357DDED63B4995F5C1F5CBE5B1E86C37F9@its-xchg4.massey.ac.nz>

> but how can I save the filter if I am to change my PC ?

Please see FAQ 3.6:

<http://spambayes.org/faq.html#can-i-share-move-my-training-data-from-one-co
mputer-to-another>

> Does the filter grow ... grow ... grow or is it possible to 
> "compact" it ?

In typical operation the database only grows.  However, it will only get
larger when new tokens are found - after a certain amount of training new
tokens are reasonably rare, and so it doesn't grow all that quickly,
particularly if you use an effectively training strategy (e.g. see
<http://entrian.com/sbwiki/TrainingIdeas>).

Since SpamBayes learns _very_ quickly, it's simple to just retrain from
scratch if you find that the database has grown too large.  There has been
some discussion about pruning the database on the developers' list, but not
recently, and with no real resolution (mostly because of the lack of need, I
suspect). 

> What if I migrate from Outlook XP to Outlook 2003 concerning the
> filter, can I keep it or will I have to start from 0 ?

If it's on the same machine, with the same (Windows) username, then you
shouldn't need to do anything at all, and your training data should still be
there.  If it's a different machine/username, then you'll need to copy the
database over (see FAQ 3.6), but that's easy to do.

=Tony.Meyer

-- 
Please always include the list (spambayes@python.org) in your replies
(reply-all), and please don't send me personal mail about SpamBayes.
http://www.massey.ac.nz/~tameyer/writing/reply_all.html explains this.

From tameyer at ihug.co.nz  Tue Nov 23 00:22:27 2004
From: tameyer at ihug.co.nz (Tony Meyer)
Date: Tue Nov 23 00:23:04 2004
Subject: [Spambayes] Unable to Access SpamBayes Browser Interface
	Usingsb_imapfilter.py
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E80176455E@its-xchg4.massey.ac.nz>
Message-ID: <ECBA357DDED63B4995F5C1F5CBE5B1E86C37FB@its-xchg4.massey.ac.nz>

> Did my message (quoted below) get lost, or did I just miss the 
> replies again?

It made it to the list, but imapfilter questions tend to get handled by me
and I was busy and then away all of last week, so without access to mail for
the most part.  Sorry about that.

> A "telnet localhost 8880" also fails. So where do we go from here, 
> since it seems as though I'm the only one with this bug, or 
> whatever it may be?

What if you try another port?  (4321 or something).  (To check that it's not
that port 8880 is unavailable, although I would expect an error in that
case).  To do that do:

> sb_imapfilter.py -b -v -o html_ui:port:4321

What happens if you run:

> sb_server.py -b

Can you get to the web interface then?  (If you can, then it must be an
imapfilter problem.  If you can't, then it's either something very odd in
the lower level web interface code, or something outside the SpamBayes
code).

=Tony.Meyer

-- 
Please always include the list (spambayes@python.org) in your replies
(reply-all), and please don't send me personal mail about SpamBayes.
http://www.massey.ac.nz/~tameyer/writing/reply_all.html explains this.

From tameyer at ihug.co.nz  Tue Nov 23 00:25:26 2004
From: tameyer at ihug.co.nz (Tony Meyer)
Date: Tue Nov 23 00:26:01 2004
Subject: [Spambayes] Problems --- 
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E80176455C@its-xchg4.massey.ac.nz>
Message-ID: <ECBA357DDED63B4995F5C1F5CBE5B1E8FBCA2F@its-xchg4.massey.ac.nz>

> I cannot get the Spambayes program to load completely. 
> I get an error message at the very end saying it could
> not load the addin to Outlook. 
 
Can you send us a copy of your most recent log file?  The troubleshooting
guide explains where to find those:

<http://cvs.sf.net/viewcvs.py/*checkout*/spambayes/spambayes/Outlook2000/doc
s/troubleshooting.html?rev=HEAD>

=Tony.Meyer

-- 
Please always include the list (spambayes@python.org) in your replies
(reply-all), and please don't send me personal mail about SpamBayes.
http://www.massey.ac.nz/~tameyer/writing/reply_all.html explains this.

From tameyer at ihug.co.nz  Tue Nov 23 01:17:46 2004
From: tameyer at ihug.co.nz (Tony Meyer)
Date: Tue Nov 23 01:19:16 2004
Subject: [Spambayes] Training spambayes,
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E801763A0C@its-xchg4.massey.ac.nz>
Message-ID: <ECBA357DDED63B4995F5C1F5CBE5B1E8FBCA32@its-xchg4.massey.ac.nz>

> I try to train my installed spambayes but if I brwose
> and choose an .dbx file it looks like it takes hours
> and it is not doing anything I think.

How large is the dbx file?  (i.e. how many messages are in it?)

Do any error messages appear in your log file?  The troubleshooting guide
explains where those are.

=Tony.Meyer

-- 
Please always include the list (spambayes@python.org) in your replies
(reply-all), and please don't send me personal mail about SpamBayes.
http://www.massey.ac.nz/~tameyer/writing/reply_all.html explains this.

From tameyer at ihug.co.nz  Tue Nov 23 01:21:22 2004
From: tameyer at ihug.co.nz (Tony Meyer)
Date: Tue Nov 23 01:21:57 2004
Subject: [Spambayes] no filterable mail items selected.
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E801763A13@its-xchg4.massey.ac.nz>
Message-ID: <ECBA357DDED63B4995F5C1F5CBE5B1E8FBCA33@its-xchg4.massey.ac.nz>

> One particular friend tried to install it and filter
> his pre-sorted emails (SPAM and Good) folders.  It will
> not train on them.  It keeps saying 2 good, 2 spam, when
> there are actually about 50 in the SPAM folder and 25
> in the Inbox (good) folder.

I suspect (the brand new) FAQ 5.6 applies here:

<http://spambayes.org/faq.html#i-get-an-error-message-no-filterable-messages
-are-selected>

> Being that emails come in 20's and 30's or more at a time,
> its going to take forever to train everything that comes in.

The training that we recommend with the Outlook plug-in is 'mistake based
training'.  This means you should only train on messages that end up in the
unsure/possible folder and any good mail in the junk folder or junk in the
good folders.  See <http://entrian.com/sbwiki/TrainingIdeas> for more
details.

With this method, you should not need to train on very many of the incoming
messages at all.  After even just 5 or so messages of each type, most mail
should be correctly classified.  After a couple of days or so SpamBayes
should be classifying with really high accuracy.

=Tony.Meyer

-- 
Please always include the list (spambayes@python.org) in your replies
(reply-all), and please don't send me personal mail about SpamBayes.
http://www.massey.ac.nz/~tameyer/writing/reply_all.html explains this.

From tameyer at ihug.co.nz  Tue Nov 23 04:21:39 2004
From: tameyer at ihug.co.nz (Tony Meyer)
Date: Tue Nov 23 04:22:15 2004
Subject: [Spambayes] It all started with SP2
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E801763A32@its-xchg4.massey.ac.nz>
Message-ID: <ECBA357DDED63B4995F5C1F5CBE5B1E86C3800@its-xchg4.massey.ac.nz>

> I've been using it successfully for over a year and once Microsoft SP2
> was installed it quit working.

Was it definitely immediately after SP2 was installed, and not just at a
similar time?

> Now, SpamBayes loads correctly, assigns
> Spam values correctly, puts messages in the "Spam" and "Possible Spam"
> folders correctly, but if I try to access any functions, it 
> performs the function but then it also shuts down Outlook.

When does the shutdown occur?  After the action, or before?  If you click
the drop-down list, does it appear, and the shutdown occur when you select
an item, or not appear?  When Outlook shuts down, does it give any sort of
error message?  Is there any message from Outlook when you open it again?

> In the log file below, I recovered a message from 'Possible Spam' folder,
> and it performed the function, but then shut down Outlook.
> *	the version of Windows you are using-Window XP Version 2002-SP2
> *	the version of SpamBayes - Version 1.0
> *	any log files. - see below
[...]

The problem with this (and with 956834, which is probably caused by the same
thing) is that there's no indication of an error occurring at all, which
makes it very difficult to try and figure out what's causing it.

Is Outlook completely up-to-date?  Have you tried running scanpst over any
pst files?  (Unlikely to be related, but can't hurt to be sure).

Do you have any other Outlook plug-ins?  Any other Outlook troubles before
this?

I might try building a test version of the plug-in that uses a temporary
toolbar (we did this once, a long time back) rather than a permanent one,
and see if that has any effect. Would you be willing to download an
installer for that test version and give that a go?  (It wouldn't be any
different in any other way and you could just reinstall the regular one if
it didn't have any effect).

=Tony.Meyer

-- 
Please always include the list (spambayes@python.org) in your replies
(reply-all), and please don't send me personal mail about SpamBayes.
http://www.massey.ac.nz/~tameyer/writing/reply_all.html explains this.

From usr7-28hl at xemaps.com  Tue Nov 23 06:18:30 2004
From: usr7-28hl at xemaps.com (usr7-28hl@xemaps.com)
Date: Tue Nov 23 06:18:30 2004
Subject: [Spambayes] Unable to Access SpamBayes Browser Interface Using
	sb_imapfilter.py
In-Reply-To: <rsl4q0pqvl3avisechaep2rqclb0clgknj@4ax.com>
References: <41887AB5.2030008@xemaps.com>
	<rsl4q0pqvl3avisechaep2rqclb0clgknj@4ax.com>
Message-ID: <41A2C826.9080702@mailsnare.net>

Richie Hindle wrote:

> [Keith]
> 
>>3. >telnet localhost:8880
>>Connecting to localhost:8880...Could not open connection to the
>>host, on post 23: Connect failed
> 
> 
> Your command line there is wrong - telnet expects a space between the
> server name and the port number rather than a colon, like this:
> 
>   telnet localhost 8880
> 
> Could you try that and let us know the result?

Thanks for replying, Richie.

Actually, though, I think you missed part of my message. The part 
you quoted is actually from an earlier post. Scott and Tony had 
pointed out my syntax error, and in my followup message I 
reported that I had gotten the same result with the correct 
syntax. :-(

-- 
Keith
From michael at kimballpottery.com  Tue Nov 23 15:15:35 2004
From: michael at kimballpottery.com (Michael Kimball)
Date: Tue Nov 23 15:16:36 2004
Subject: [Spambayes] email evades Spambayes
Message-ID: <41A34607.1050107@kimballpottery.com>

When I d/l'ed my email this am this one came through with no indication 
that SpamBayes had checked it.  Everything else had 
X-Spambayes-Classification headers with the ham/spam/unsure value.  Is 
there anything in this that can let it evade SpamBayes?  Or did 
SpamBayes just hiccup?  I've set the proxies so AVG collects my mail, 
passes it to SpamBayes which passes it to Thunderbird.


 From - Tue Nov 23 08:48:53 2004
X-Account-Key: account1
X-UIDL: UID206-1100919104
X-Mozilla-Status: 0001
X-Mozilla-Status2: 00000000
Return-Path: <FOXVTAYXDVHST@inbox.ru>
Delivered-To: 337-michael@kimballpottery.com
Received: (qmail 1201 invoked from network); 23 Nov 2004 12:54:33 -0000
Received: from adsl-69-225-160-10.dsl.pltn13.pacbell.net (69.225.160.10)
by mail.webinmotion.ca with SMTP; 23 Nov 2004 12:54:31 -0000
X-Message-Info: 49XA59CYGxskb521T19fNdoqKS06daUA1rKLntS8L99
Received: (from silty@69.225.160.10)
by citron8[1
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii



-- 
No virus found in this incoming message.
Checked by AVG Anti-Virus.
Version: 7.0.289 / Virus Database: 265.4.1 - Release Date: 11/19/2004


-- 
No virus found in this outgoing message when I SENT it.  Who knows what it has picked up on it's way through the interweb.
Checked by AVG Anti-Virus.
Version: 7.0.289 / Virus Database: 265.4.1 - Release Date: 11/19/2004

From Amir_Katz at bmc.com  Tue Nov 23 16:09:12 2004
From: Amir_Katz at bmc.com (Katz, Amir)
Date: Tue Nov 23 16:26:00 2004
Subject: [Spambayes] Spambayes proxy server error
Message-ID: <F8B9B12641FB6B4FAC761BDEF9D63B040330C3A5@tlv-ex-01.adprod.bmc.com>

No one has answered and rightly so - I found out that this is covered in the
FAQ. Case closed.


Amir 

-----Original Message-----
From: Katz, Amir [mailto:Amir_Katz@bmc.com] 
Sent: Thursday, November 18, 2004 22:11
To: spambayes@python.org
Subject: [Spambayes] Spambayes proxy server error

Hi, I can't get the server to train. I tried to mark all messages as
'discard', but clicking 'train' fails with the following output.

This is SB 1.0 on Win XP with Outlook express.

Anyone?

Thanks.
 
==================================================================== 

500 Server error
Traceback (most recent call last):

  File "spambayes\Dibbler.pyc", line 461, in found_terminator

  File "spambayes\ProxyUI.pyc", line 391, in onReview

  File "spambayes\Corpus.pyc", line 214, in takeMessage

  File "spambayes\FileCorpus.pyc", line 140, in addMessage

  File "spambayes\Corpus.pyc", line 134, in addMessage

  File "spambayes\storage.pyc", line 588, in onAddMessage

  File "spambayes\storage.pyc", line 598, in train

  File "spambayes\message.pyc", line 260, in setId

  File "spambayes\message.pyc", line 118, in _getState

  File "shelve.pyc", line 118, in __getitem__

  File "bsddb\__init__.pyc", line 116, in __getitem__

DBRunRecoveryError: (-30982, 'DB_RUNRECOVERY: Fatal error, run database
recovery -- fatal region error detected; run recovery')

_______________________________________________
Spambayes@python.org
http://mail.python.org/mailman/listinfo/spambayes
Check the FAQ before asking: http://spambayes.sf.net/faq.html
From csc at csc.noaa.gov  Tue Nov 23 17:19:32 2004
From: csc at csc.noaa.gov (Automatic Reply)
Date: Tue Nov 23 17:19:36 2004
Subject: [Spambayes] Mail System (csc@csc.noaa.gov)
Message-ID: <200411231619.iANGJWDX025881@csc.noaa.gov>

Hello,
Thank you for your email to the NOAA Coastal Services Center.
Your inquiry/request has been routed to the appropriate person
who will respond as soon as possible.  Inquiries are normally
answered within seven business days.

Coastal Information Services/LIB

--
(This is an automated response to your recent e-mail
message to the NOAA Coastal Services Center)
From d.horalek at volny.cz  Tue Nov 23 19:42:25 2004
From: d.horalek at volny.cz (=?iso-8859-2?Q?Marika_Hor=E1lkov=E1?=)
Date: Tue Nov 23 19:42:47 2004
Subject: [Spambayes] (no subject)
Message-ID: <000a01c4d18c$2d03b1e0$c051183e@DOMAMEP75LWOMV>


S pozdravem 

Marika Hor?lkov?
GSM: 602 837 960
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.python.org/pipermail/spambayes/attachments/20041123/ffe6164d/attachment.htm
From d.horalek at volny.cz  Tue Nov 23 19:44:43 2004
From: d.horalek at volny.cz (=?iso-8859-2?Q?Marika_Hor=E1lkov=E1?=)
Date: Tue Nov 23 19:45:05 2004
Subject: [Spambayes] (no subject)
Message-ID: <002601c4d18c$7f31c010$c051183e@DOMAMEP75LWOMV>


S pozdravem 

Marika Hor?lkov?
GSM: 602 837 960
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.python.org/pipermail/spambayes/attachments/20041123/775420b9/attachment.html
From jfodor at rfcinc.com  Tue Nov 23 23:25:36 2004
From: jfodor at rfcinc.com (jfodor)
Date: Tue Nov 23 23:25:41 2004
Subject: [Spambayes] Out of Office Assistant
Message-ID: <70721244157CD211AD5D00600845E8C8E3E22C@BARNEY2>

I am wondering why, when using Microsoft Outlook 2000, SpamBayes does not
forward ALL mail when the out of office assistant is turned on.  I would
like to configure it so that all mail is forwarded to a remote location to
ensure that an accidental SPAM message gets forwarded.  So far, I have only
been able to forward messages that are for sure HAM.  Could someone help me
out with this?

James Fodor
RF Communications
(630) 420-8882 x 133
1701 Quincy Ave, Suite 29
Naperville, IL  60540


-- 
No virus found in this outgoing message.
Checked by AVG Anti-Virus.
Version: 7.0.289 / Virus Database: 265.4.1 - Release Date: 11/19/2004
 
From RMezzone at PJSolomon.com  Tue Nov 23 23:46:53 2004
From: RMezzone at PJSolomon.com (Robert Mezzone)
Date: Tue Nov 23 23:46:58 2004
Subject: [Spambayes] Out of Office Assistant
Message-ID: <1254A68F4903D411B24800508B1220E90629F77E@solomon1.pjsc.internal>

I can't address the OOF question but I have mine setup so all Ham, Spam and
Possible Spam is forwarded to my Blackberry. I moved the Spambayes related
folders so they are subfolders of my Inbox, this way everything is forwarded
and I can move things between folders on my BB and the changes synch with my
desktop.

Not sure if this helps...

Robert


-----Original Message-----
From: spambayes-bounces+rmezzone=pjsolomon.com@python.org
<spambayes-bounces+rmezzone=pjsolomon.com@python.org>
To: 'spambayes@python.org' <spambayes@python.org>
Sent: Tue Nov 23 17:25:36 2004
Subject: [Spambayes] Out of Office Assistant

I am wondering why, when using Microsoft Outlook 2000, SpamBayes does not
forward ALL mail when the out of office assistant is turned on.  I would
like to configure it so that all mail is forwarded to a remote location to
ensure that an accidental SPAM message gets forwarded.  So far, I have only
been able to forward messages that are for sure HAM.  Could someone help me
out with this?

James Fodor
RF Communications
(630) 420-8882 x 133
1701 Quincy Ave, Suite 29
Naperville, IL  60540


-- 
No virus found in this outgoing message.
Checked by AVG Anti-Virus.
Version: 7.0.289 / Virus Database: 265.4.1 - Release Date: 11/19/2004
 
_______________________________________________
Spambayes@python.org
http://mail.python.org/mailman/listinfo/spambayes
Check the FAQ before asking: http://spambayes.sf.net/faq.html
From jfodor at rfcinc.com  Tue Nov 23 23:51:45 2004
From: jfodor at rfcinc.com (jfodor)
Date: Tue Nov 23 23:51:48 2004
Subject: [Spambayes] Out of Office Assistant
Message-ID: <70721244157CD211AD5D00600845E8C8E3E22F@BARNEY2>

Did you have any special rules set up to forward that email?  How are you
forwarding it if you are not using the out of office assistant?

James

-----Original Message-----
From: Robert Mezzone [mailto:RMezzone@PJSolomon.com]
Sent: Tuesday, November 23, 2004 4:47 PM
To: 'jfodor@rfcinc.com'; 'spambayes@python.org'
Subject: Re: [Spambayes] Out of Office Assistant


I can't address the OOF question but I have mine setup so all Ham, Spam and
Possible Spam is forwarded to my Blackberry. I moved the Spambayes related
folders so they are subfolders of my Inbox, this way everything is forwarded
and I can move things between folders on my BB and the changes synch with my
desktop.

Not sure if this helps...

Robert


-----Original Message-----
From: spambayes-bounces+rmezzone=pjsolomon.com@python.org
<spambayes-bounces+rmezzone=pjsolomon.com@python.org>
To: 'spambayes@python.org' <spambayes@python.org>
Sent: Tue Nov 23 17:25:36 2004
Subject: [Spambayes] Out of Office Assistant

I am wondering why, when using Microsoft Outlook 2000, SpamBayes does not
forward ALL mail when the out of office assistant is turned on.  I would
like to configure it so that all mail is forwarded to a remote location to
ensure that an accidental SPAM message gets forwarded.  So far, I have only
been able to forward messages that are for sure HAM.  Could someone help me
out with this?

James Fodor
RF Communications
(630) 420-8882 x 133
1701 Quincy Ave, Suite 29
Naperville, IL  60540


-- 
No virus found in this outgoing message.
Checked by AVG Anti-Virus.
Version: 7.0.289 / Virus Database: 265.4.1 - Release Date: 11/19/2004
 
_______________________________________________
Spambayes@python.org
http://mail.python.org/mailman/listinfo/spambayes
Check the FAQ before asking: http://spambayes.sf.net/faq.html

-- 
No virus found in this incoming message.
Checked by AVG Anti-Virus.
Version: 7.0.289 / Virus Database: 265.4.1 - Release Date: 11/19/2004
 

-- 
No virus found in this outgoing message.
Checked by AVG Anti-Virus.
Version: 7.0.289 / Virus Database: 265.4.1 - Release Date: 11/19/2004
 
From RMezzone at PJSolomon.com  Tue Nov 23 23:54:06 2004
From: RMezzone at PJSolomon.com (Robert Mezzone)
Date: Tue Nov 23 23:54:11 2004
Subject: [Spambayes] Out of Office Assistant
Message-ID: <1254A68F4903D411B24800508B1220E90629F77F@solomon1.pjsc.internal>

I have my BB desktop software configured to forward all messages in my Inbox
and it's sub folders. I haven't played with Outlook rules in a while, but I
would imagine they have something similiar. Then again I could be wrong...
Good luck.

Robert


-----Original Message-----
From: jfodor <jfodor@rfcinc.com>
To: Robert Mezzone <RMezzone@PJSolomon.com>; jfodor <jfodor@rfcinc.com>;
'spambayes@python.org' <spambayes@python.org>
Sent: Tue Nov 23 17:51:45 2004
Subject: RE: [Spambayes] Out of Office Assistant

Did you have any special rules set up to forward that email?  How are you
forwarding it if you are not using the out of office assistant?

James

-----Original Message-----
From: Robert Mezzone [mailto:RMezzone@PJSolomon.com]
Sent: Tuesday, November 23, 2004 4:47 PM
To: 'jfodor@rfcinc.com'; 'spambayes@python.org'
Subject: Re: [Spambayes] Out of Office Assistant


I can't address the OOF question but I have mine setup so all Ham, Spam and
Possible Spam is forwarded to my Blackberry. I moved the Spambayes related
folders so they are subfolders of my Inbox, this way everything is forwarded
and I can move things between folders on my BB and the changes synch with my
desktop.

Not sure if this helps...

Robert


-----Original Message-----
From: spambayes-bounces+rmezzone=pjsolomon.com@python.org
<spambayes-bounces+rmezzone=pjsolomon.com@python.org>
To: 'spambayes@python.org' <spambayes@python.org>
Sent: Tue Nov 23 17:25:36 2004
Subject: [Spambayes] Out of Office Assistant

I am wondering why, when using Microsoft Outlook 2000, SpamBayes does not
forward ALL mail when the out of office assistant is turned on.  I would
like to configure it so that all mail is forwarded to a remote location to
ensure that an accidental SPAM message gets forwarded.  So far, I have only
been able to forward messages that are for sure HAM.  Could someone help me
out with this?

James Fodor
RF Communications
(630) 420-8882 x 133
1701 Quincy Ave, Suite 29
Naperville, IL  60540


-- 
No virus found in this outgoing message.
Checked by AVG Anti-Virus.
Version: 7.0.289 / Virus Database: 265.4.1 - Release Date: 11/19/2004
 
_______________________________________________
Spambayes@python.org
http://mail.python.org/mailman/listinfo/spambayes
Check the FAQ before asking: http://spambayes.sf.net/faq.html

-- 
No virus found in this incoming message.
Checked by AVG Anti-Virus.
Version: 7.0.289 / Virus Database: 265.4.1 - Release Date: 11/19/2004
 

-- 
No virus found in this outgoing message.
Checked by AVG Anti-Virus.
Version: 7.0.289 / Virus Database: 265.4.1 - Release Date: 11/19/2004
 
From tameyer at ihug.co.nz  Wed Nov 24 00:21:58 2004
From: tameyer at ihug.co.nz (Tony Meyer)
Date: Wed Nov 24 00:22:35 2004
Subject: [Spambayes] Spambayes proxy server error
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E8017ACF2D@its-xchg4.massey.ac.nz>
Message-ID: <ECBA357DDED63B4995F5C1F5CBE5B1E8FBCA42@its-xchg4.massey.ac.nz>

> No one has answered and rightly so - I found out that this is 
> covered in the FAQ. Case closed.

I answered:

<http://mail.python.org/pipermail/spambayes/2004-November/015331.html>

But all I did was point to the FAQ, anyway...

=Tony.Meyer

-- 
Please always include the list (spambayes@python.org) in your replies
(reply-all), and please don't send me personal mail about SpamBayes.
http://www.massey.ac.nz/~tameyer/writing/reply_all.html explains this.

From tameyer at ihug.co.nz  Wed Nov 24 00:26:32 2004
From: tameyer at ihug.co.nz (Tony Meyer)
Date: Wed Nov 24 00:26:40 2004
Subject: [Spambayes] Out of Office Assistant
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E8017ACF31@its-xchg4.massey.ac.nz>
Message-ID: <ECBA357DDED63B4995F5C1F5CBE5B1E8FBCA43@its-xchg4.massey.ac.nz>

> I am wondering why, when using Microsoft Outlook 2000, 
> SpamBayes does not forward ALL mail when the out of office 
> assistant is turned on.  I would like to configure it so that 
> all mail is forwarded to a remote location to ensure that an 
> accidental SPAM message gets forwarded.

Maybe I'm missing something, but shouldn't you just then untick the "Enable
SpamBayes" box when you turn Out of Office on, then?

If I want to auto-forward mail, I use the Outlook rules, not the Out of
Office stuff, so I'm not sure when that kicks in.  If it's at the same time
as the rules, though, then SpamBayes ought to run after it (assuming you
have background filtering enabled).  If it runs in some sort of batch mode
or something, then maybe you can get the Out of Office tool to forward mail
arriving in more than one folder?  Or increase the background filtering
times?

=Tony.Meyer

-- 
Please always include the list (spambayes@python.org) in your replies
(reply-all), and please don't send me personal mail about SpamBayes.
http://www.massey.ac.nz/~tameyer/writing/reply_all.html explains this.

From michael at kimballpottery.com  Wed Nov 24 05:33:49 2004
From: michael at kimballpottery.com (Michael Kimball)
Date: Wed Nov 24 05:37:20 2004
Subject: [Spambayes] email evades Spambayes
In-Reply-To: <5.1.0.14.2.20041123211026.02217e88@pop3.110.net>
References: <5.1.0.14.2.20041123211026.02217e88@pop3.110.net>
Message-ID: <41A40F2D.9010101@kimballpottery.com>

Don Rolph wrote:
> I am finding that if your email client uses any non-basic pop3 commands, 
> that the pop3 proxy sometimes hangs.  If it hangs, then it passes the 
> email without filtering it.
> 
> What email client are you using?

I'm using Thunderbird 0.9 (essentially the same as Netscape 7.2)

> 
> At 09:15 AM 11/23/2004 -0500, you wrote:
> 
>> When I d/l'ed my email this am this one came through with no 
>> indication that SpamBayes had checked it.  Everything else had 
>> X-Spambayes-Classification headers with the ham/spam/unsure value.  Is 
>> there anything in this that can let it evade SpamBayes?  Or did 
>> SpamBayes just hiccup?  I've set the proxies so AVG collects my mail, 
>> passes it to SpamBayes which passes it to Thunderbird.
>>


-- 
No virus found in this outgoing message when I SENT it.  Who knows what it has picked up on it's way through the interweb.
Checked by AVG Anti-Virus.
Version: 7.0.289 / Virus Database: 265.4.1 - Release Date: 11/19/2004

From mikec at mikecauser.com  Wed Nov 24 12:28:05 2004
From: mikec at mikecauser.com (Mike Causer)
Date: Wed Nov 24 12:28:11 2004
Subject: [Spambayes] Exit codes 11 and -6 from sb_filter
Message-ID: <20041124112805.4849bd1e.mikec@mikecauser.com>

Today a message sneaked through with no classification.  Looking in the
procmail log showed that sb_filter.py had returned an exit status 11.
The offending message is plain-text and has nothing wrong with it that I
can spot, and when sb_filter.py is run via the command line it gets
an X-Spambayes-Classification: spam; 0.92 quite correctly.   A grep
around in the Spambayes code and in the main Python 2.3 code doesn't
turn up any explicit sys.exit(11) and attempts to get a full trace out
of sb_filter while it's running from procmail have failed.

There's another message which also fails classification via procmail
with exit status of -6,  but correctly gets a ham 0.6 on the commandline.


Need help from someone with a bigger bag of clues I think...


Mike
-- 
Mike Causer                          Email - mailto:mikec@mikecauser.com
GPG KeyID 1C2DDA07                       WWW - http://www.mikecauser.com
Flood the fen again! - Wicken Fen enlargement - http://www.wicken.org.uk
From baraa-f at arabsites.com  Wed Nov 24 13:20:27 2004
From: baraa-f at arabsites.com (baraa-f@arabsites.com)
Date: Wed Nov 24 13:18:37 2004
Subject: [Spambayes] Postcard
Message-ID: <200411241218.iAOCITb04934@uusnwa08.utc.com>

------------------  Virus Warning Message (on uusnwa08)

Found virus WORM_NETSKY.P in file details.txt                                                                     .pif (in postcard.zip)
The file is deleted.

---------------------------------------------------------
-------------- next part --------------
Congratulations!,

 your best friend.

-------------- next part --------------

------------------  Virus Warning Message (on uusnwa08)

postcard.zip is removed from here because it contains a virus.

---------------------------------------------------------
From anthony at interlink.com.au  Wed Nov 24 14:10:38 2004
From: anthony at interlink.com.au (Anthony Baxter)
Date: Wed Nov 24 14:10:53 2004
Subject: [Spambayes] alternative kmail integration setup
Message-ID: <200411250010.38808.anthony@interlink.com.au>

I've documented my setup of kmail and spambayes here:

http://www.interlink.com.au/anthony/tech/kmail/

The major difference from Toby's setup is that I've hooked the "train as spam"
and "train as ham" buttons into spambayes directly, so that there's no need 
for an extra training step. A future enhancement will be to file a copy of
non-edge ham/spam to a "to be trained on" folder. This is nice - it gives me
very tight integration with the mailer itself. (I first tried using 
spamassassin, as it's supported by kmail - eeesh. you don't notice
spambayes until it's not there...)

From robert at imagine.co.uk  Wed Nov 24 14:55:30 2004
From: robert at imagine.co.uk (Robert Neuschul)
Date: Wed Nov 24 14:55:50 2004
Subject: [Spambayes] Msg for Tony about earlier problems.
Message-ID: <VA.00005768.0101e695@imagine.co.uk>

Tony

After messing around with your latest patches for the 
pop3proxy_server.py I wasn't getting anywhere; some of the time things 
worked, some of the time they didn't - with no clear indication of why 
things were so variable. 

In some desperation I uninstalled everything, cleaned the Windows 2003 
registry to remove all references to Python and Spambayes, and all 
instances of the service. I then deleted all the old files [cache DB 
etc] and rebooted the server. I then did a clean install of Python 
2.3.3 with the latest matching Pywin [these are the same versions 
previously installed], and rebooted the server again. I then did some 
careful analysis of the local environment to be sure that Python and 
PyWin were working OK - which they were.

At that point I installed the current Spambayes release 1 source 
version [not the latest CVS version], and then copied across your 
patched version of pop3proxy_server.py and installed it. 

Hey presto: all is now working fine. 

So far it's been running for just over a week and the only real issue I 
have isn't with SB itself it's with configuring the AV not to touch the 
SB cache, where there are [as one might expect] a lot of infected items 
which never make it into our mail system thanks to pre-processing of SB 
headers.

I /think/ that's cracked the problem. Many thanks for your patience and 
the time spent on debugging.

All the best,

Robert


From anthony at interlink.com.au  Wed Nov 24 15:21:55 2004
From: anthony at interlink.com.au (Anthony Baxter)
Date: Wed Nov 24 15:22:10 2004
Subject: [Spambayes] alternative kmail integration setup
In-Reply-To: <200411250010.38808.anthony@interlink.com.au>
References: <200411250010.38808.anthony@interlink.com.au>
Message-ID: <200411250121.56265.anthony@interlink.com.au>

On Thursday 25 November 2004 00:10, Anthony Baxter wrote:
> I've documented my setup of kmail and spambayes here:
>
> http://www.interlink.com.au/anthony/tech/kmail/
>
> The major difference from Toby's setup is that I've hooked the "train as
> spam" and "train as ham" buttons into spambayes directly, so that there's
> no need for an extra training step. A future enhancement will be to file a
> copy of non-edge ham/spam to a "to be trained on" folder. This is nice - it
> gives me very tight integration with the mailer itself. (I first tried
> using spamassassin, as it's supported by kmail - eeesh. you don't notice
> spambayes until it's not there...)

I forgot to add - the only really pressing piece of integration that's missing
is to show the spambayes scoring header in the column view for the unsure
and spam folders. I've not yet been able to find any way to customise this -
if anyone knows, I'd love to hear it - or else I can continue attempting to 
find it via googling... from the look of this bug: 
http://bugs.kde.org/show_bug.cgi?id=39782 
I'd say that I'm completely out of luck here.

Anthony
From Jennifer at Mobilechamber.com  Wed Nov 24 15:25:16 2004
From: Jennifer at Mobilechamber.com (Jennifer Stewart)
Date: Wed Nov 24 15:37:51 2004
Subject: [Spambayes] deleted my junk suspects folder, now messed up
Message-ID: <D5719B6000144E4794AF4BC229EE9B4663598B@MACCPDC1.Mobileareachamber.com>

OK, I accidentally deleted my junk suspects box, now it says its not
configured correctly and our computer person is out until Monday.  Any
help?  I tried things on my own but nothing seems to work.  Can you make
the junk and junk suspects folders where you can't delete the  without
being prompted?

 

Jennifer Stewart

 

Administrative Assistant, Economic Development 

451 Government Street

Mobile, Al 36602

251-431-8636

251-431-8608 Fax

jennifer@mobilechamber.com

visit our website:   <http://www.mobilechamber.com>
www.mobilechamber.com


The Mobile Area Chamber of Commerce serves as a progressive advocate for
business needs to promote the Mobile area's economic well-being.
Focusing on our community, the Mobile Chamber builds jobs, a competitive
workforce, a positive business climate, higher expectations for our
community's development and brings value to its members. Aggressive
economic development marketing programs by the Mobile Chamber have
helped bring in 232 new businesses and led to 809 expansions of existing
industries since 1985, representing a capital investment of $7.9 billion
and 28,579 direct new jobs. 

 

 



This E-Mail transmission (and/or the documents accompanying it) may contain information belonging to the sender which is confidential, privileged and/or exempt from disclosure under applicable law.  The information is intended only for the use of the individual(s) or entity named above.   If you are not the intended recipient, you are hereby  notified that any disclosure, copying, distribution or the taking of any action in reliance on the contents of this information is strictly prohibited.  If you have received this E-Mail transmission in error, please immediately notify us by return E-Mail or telephone to arrange for return of its contents including any documents.
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.python.org/pipermail/spambayes/attachments/20041124/a65ae5db/attachment.htm
From tameyer at ihug.co.nz  Thu Nov 25 00:58:34 2004
From: tameyer at ihug.co.nz (Tony Meyer)
Date: Thu Nov 25 00:59:10 2004
Subject: [Spambayes] deleted my junk suspects folder, now messed up
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E8017AD183@its-xchg4.massey.ac.nz>
Message-ID: <ECBA357DDED63B4995F5C1F5CBE5B1E86C3805@its-xchg4.massey.ac.nz>

> OK, I accidentally deleted my junk suspects box

Please see FAQ 3.13:

<http://spambayes.org/faq.html#help-i-deleted-the-unsure-spam-folder>

> Can you make the junk and junk suspects folders where
> you can't delete the  without being prompted?

Possibly, although it would be difficult, since it's interfering with normal
Outlook actions.  I think I vaguely recall this being mentioned at some
point on the development list.  We're definitely trying to come up with
something to make this (fairly common) problem less common.

=Tony.Meyer

-- 
Please always include the list (spambayes@python.org) in your replies
(reply-all), and please don't send me personal mail about SpamBayes.
http://www.massey.ac.nz/~tameyer/writing/reply_all.html explains this.

From tameyer at ihug.co.nz  Thu Nov 25 01:01:56 2004
From: tameyer at ihug.co.nz (Tony Meyer)
Date: Thu Nov 25 01:02:32 2004
Subject: [Spambayes] spambay
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E8017AD0AC@its-xchg4.massey.ac.nz>
Message-ID: <ECBA357DDED63B4995F5C1F5CBE5B1E8FBCA54@its-xchg4.massey.ac.nz>

>>> I can't uninstall spambays.
>> 
>> Please see FAQ 3.16:
>> 
>> <http://spambayes.org/faq.html#how-do-i-uninstall-the-plug-in>
>>
>> As described in there, the toolbar isn't removed on uninstall
>> (this is an unresolved bug), so you have to manually remove it.
>
> Okay Tony you sent me this but it is not helping...i have 
> deleted from add remove and it is still here!! 

By still there, do you mean that you have a non-functioning (i.e. clicking
on it does nothing) toolbar?  If so, then as I said above, and as the FAQ
explains, that is a bug, which means you need to manually remove the toolbar
yourself.  The FAQ explains how to do that.

If you mean that you uninstalled via Add/Remove Programs and you still have
a *functioning* SpamBayes, then:

 * Did you install the source version as well?  If so, you'll need to run
addin.py --unregister to remove that.

 * Are you positive?  It seems very unlikely.

 * Are the files in C:\Program Files\SpamBayes still there?  (Or wherever
you chose to install it).

=Tony.Meyer

-- 
Please always include the list (spambayes@python.org) in your replies
(reply-all), and please don't send me personal mail about SpamBayes.
http://www.massey.ac.nz/~tameyer/writing/reply_all.html explains this.

From anthony at interlink.com.au  Thu Nov 25 02:44:32 2004
From: anthony at interlink.com.au (Anthony Baxter)
Date: Thu Nov 25 02:44:48 2004
Subject: [Spambayes] alternative kmail integration setup
In-Reply-To: <200411250121.56265.anthony@interlink.com.au>
References: <200411250010.38808.anthony@interlink.com.au>
	<200411250121.56265.anthony@interlink.com.au>
Message-ID: <200411251244.33135.anthony@interlink.com.au>


There's an open wishlist item for getting this setup automatically
added to KDE's antispam setup wizard now, and it sounds like it'll
be done for the next release of kde.

http://bugs.kde.org/show_bug.cgi?id=93851   
From ta-meyer at ihug.co.nz  Thu Nov 25 07:59:17 2004
From: ta-meyer at ihug.co.nz (Tony Meyer)
Date: Thu Nov 25 07:59:57 2004
Subject: [Spambayes] ANNOUNCE: SpamBayes release 1.0.1
Message-ID: <ECBA357DDED63B4995F5C1F5CBE5B1E86C3823@its-xchg4.massey.ac.nz>

The SpamBayes team is pleased to announce release 1.0.1 of SpamBayes.

As is now usual, this is both a release of the source code and of an
installation program for all Microsoft Windows users.

This is a bug-fix release that fixes a number of minor issues with the 1.0
release, but includes no new functionality, and is entirely compatible with
the 1.0 release.  A 1.1a1 release, including many new features, will
probably be released early in the new year.

Details about the bugs that have been fixed in this release can be found at

    https://sourceforge.net/project/shownotes.php?release_id=285346

You can get the release via the 'Download' page at

    http://spambayes.org/download.html

Enjoy the new release and your spam-free mailbox :-)

As always, thanks to everyone involved in this release.

Tony.
(on behalf of the SpamBayes team)

--- What is SpamBayes? ---

The SpamBayes project is working on developing a Bayesian (of sorts)
anti-spam filter (in Python), initially based on the work of Paul Graham,
but since modified with ideas from Robinson, Peters, et al.

The project includes a number of different applications, all using the same
core code, ranging from a plug-in for Microsoft Outlook, to a POP3 proxy, to
various command-line tools.

The Windows installation program will install either the Outlook add-in (for
Microsoft Outlook users), or the SpamBayes server program (for all other
POP3 mail client users, including Microsoft Outlook Express). All Windows
users (including existing users of the Outlook add-in) are encouraged to use
the installation program.

If you wish to use the source-code version, you will also need to install
Python - see README.txt in the source tree for more information.

From gerrit at nl.linux.org  Thu Nov 25 08:11:40 2004
From: gerrit at nl.linux.org (Gerrit)
Date: Thu Nov 25 08:12:33 2004
Subject: [Spambayes] Re: [Spambayes-announce] ANNOUNCE: SpamBayes release
	1.0.1
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E86C3823@its-xchg4.massey.ac.nz>
References: <ECBA357DDED63B4995F5C1F5CBE5B1E86C3823@its-xchg4.massey.ac.nz>
Message-ID: <20041125071140.GA13681@nl.linux.org>

Tony Meyer wrote:
> The SpamBayes team is pleased to announce release 1.0.1 of SpamBayes.

~/cvs/spambayes/spambayes$ cvs update -PCdfR
cvs update: Updating .
cvs update: Updating resources
cvs update: Updating test
~/cvs/spambayes/spambayes$ cat __init__.py
# package marker.

__version__ = '1.0rc1'


Is this a mistake?

regards,
Gerrit Holl.

-- 
Weather in Lulea / Kallax, Sweden 25/11 07:20:
	-5.0?C   wind 7.2 m/s WNW (34 m above NAP)
-- 
In the councils of government, we must guard against the acquisition of
unwarranted influence, whether sought or unsought, by the
military-industrial complex. The potential for the disastrous rise of
misplaced power exists and will persist.
    -Dwight David Eisenhower, January 17, 1961
From ta-meyer at ihug.co.nz  Thu Nov 25 08:21:30 2004
From: ta-meyer at ihug.co.nz (Tony Meyer)
Date: Thu Nov 25 08:22:07 2004
Subject: [Spambayes] RE: [Spambayes-announce] ANNOUNCE: SpamBayes release
	1.0.1
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E8017AD324@its-xchg4.massey.ac.nz>
Message-ID: <ECBA357DDED63B4995F5C1F5CBE5B1E8FBCA64@its-xchg4.massey.ac.nz>

[Tony Meyer]
>> The SpamBayes team is pleased to announce release 1.0.1 of 
>> SpamBayes.

[Gerrit]
> ~/cvs/spambayes/spambayes$ cvs update -PCdfR
> cvs update: Updating .
> cvs update: Updating resources
> cvs update: Updating test
> ~/cvs/spambayes/spambayes$ cat __init__.py
> # package marker.
> 
> __version__ = '1.0rc1'
> 
> Is this a mistake?

You had me worried for a minute :)  No, it's not.

You're doing a CVS update from CVS Head above (presumably).  The code in
there will eventually be 1.1a1 - the 1.0 code was forked off into a separate
branch (release_1_0_branch, IIRC) at the time of 1.0rc1.  The __init__.py
file isn't usually updated until a new release is made, so the CVS Head one
is still sitting there as it was when it was forked.  If you get the 1.0.1
source archive the __init__.py in there will have __version__ == '1.0.1'.

I suppose we could have the CVS version be something like "CVS" and just
change it for a release, but there doesn't really seem to be much point.

Make sense?

=Tony.Meyer

-- 
Please always include the list (spambayes@python.org) in your replies
(reply-all), and please don't send me personal mail about SpamBayes.
http://www.massey.ac.nz/~tameyer/writing/reply_all.html explains this.

From maureen at nw.com.au  Thu Nov 25 12:55:28 2004
From: maureen at nw.com.au (Maureen Cox)
Date: Thu Nov 25 12:55:43 2004
Subject: [Spambayes] Using spambayes
Message-ID: <005101c4d2e5$abe08ba0$2bf012cb@mauputer>

 
Hello
 
I have recently downloaded spambayes on to my computer.  
 
It seems to be working ok in that my junk mail appears to be ending up
in the junkmail folder.   My question is this:
Do I then have to delete all the messages which appear in the junkmail
folder and if I keep doing this, will the junk mail eventually be
filtered and stop appearing ??
 
Sounds like a pretty basic question, but I am not a computer whiz and am
wondering if I am doing all the steps correctly.
 
I look forward to your response
 
Maureen
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.python.org/pipermail/spambayes/attachments/20041125/9790e705/attachment.html
From m.cervo at spliethoff.com  Thu Nov 25 15:24:12 2004
From: m.cervo at spliethoff.com (Cervo, Michiel)
Date: Thu Nov 25 15:24:17 2004
Subject: [Spambayes] Spambayes problem!
Message-ID: <0DDCD682C35AA74CA374BC7DBF6C9BAA07586724@amsmail1.SPLGROUP>


Hello

I have 2 problems with spambayes, I did all the options that where shown in the Troubleshooting Guide.
One problem is that I can't get a second user on a computer with Outlook. If I try so I get this message:

Traceback (most recent call last):
  File "addin.py", line 1546, in ?
  File "win32com\server\register.pyc", line 483, in UseCommandLine
  File "win32com\server\register.pyc", line 413, in RegisterClasses
  File "win32com\server\register.pyc", line 181, in RegisterServer
  File "win32com\server\register.pyc", line 32, in _set_string
pywintypes.error: (5, 'RegSetValue', 'Access is denied.')

Instead of 
Registered: SpamBayes.OutlookAddin 
Registration complete.
Registration (in HKEY_LOCAL_MACHINE) complete.

What is the problem?!?!?!


My second problem is that the button's won't work. If I select "Filter messages"  from the spambayes menu the spambayes program filter the messages for me, if I start spambayes it filers my messages olso. But the button's won't work. I did my best but it didn't helpt.


Thanks in advantage,

Michiel Cervo
Spliethoff - IT Department
Mail: m.cervo@spliethoff.com
Telefoon 020 - 4488 497

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.python.org/pipermail/spambayes/attachments/20041125/98acffcf/attachment.html
From skip at pobox.com  Thu Nov 25 16:12:59 2004
From: skip at pobox.com (Skip Montanaro)
Date: Thu Nov 25 16:13:20 2004
Subject: [Spambayes] Re: [Spambayes-announce] ANNOUNCE: SpamBayes release
	1.0.1
In-Reply-To: <20041125071140.GA13681@nl.linux.org>
References: <ECBA357DDED63B4995F5C1F5CBE5B1E86C3823@its-xchg4.massey.ac.nz>
	<20041125071140.GA13681@nl.linux.org>
Message-ID: <16805.63099.513576.842563@montanaro.dyndns.org>


    >> The SpamBayes team is pleased to announce release 1.0.1 of SpamBayes.

    gerrit> ~/cvs/spambayes/spambayes$ cat __init__.py
    gerrit> # package marker.

    gerrit> __version__ = '1.0rc1'

    gerrit> Is this a mistake?

Looks that way.  I checked in the obvious fix...

Skip
From skip at pobox.com  Thu Nov 25 16:14:16 2004
From: skip at pobox.com (Skip Montanaro)
Date: Thu Nov 25 16:14:28 2004
Subject: [Spambayes] RE: [Spambayes-announce] ANNOUNCE: SpamBayes release
	1.0.1
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E8FBCA64@its-xchg4.massey.ac.nz>
References: <ECBA357DDED63B4995F5C1F5CBE5B1E8017AD324@its-xchg4.massey.ac.nz>
	<ECBA357DDED63B4995F5C1F5CBE5B1E8FBCA64@its-xchg4.massey.ac.nz>
Message-ID: <16805.63176.429685.971690@montanaro.dyndns.org>


    >> __version__ = '1.0rc1'
    >> 
    >> Is this a mistake?

    Tony> You had me worried for a minute :)  No, it's not.

    Tony> You're doing a CVS update from CVS Head above (presumably).  The
    Tony> code in there will eventually be 1.1a1 - the 1.0 code was forked
    Tony> off into a separate branch (release_1_0_branch, IIRC) at the time
    Tony> of 1.0rc1.

Ah, crap.  Oh well.

Skip
From b.easton at exemail.com.au  Thu Nov 25 21:00:04 2004
From: b.easton at exemail.com.au (Brent Easton)
Date: Thu Nov 25 21:00:10 2004
Subject: [Spambayes] Spambates Exceptions
Message-ID: <200411260700040033.026BB5A1@jay.exetel.com.au>

Hi, 

Have recently upgrades to v1.0 spambayes. Using Calypso mail client with spambayes proxy server. I now get the following header added to every email:

X-Spambayes-Exception: Traceback (most recent call last): . File "sb_server.pyc", line 475, in onRetr . File "spambayes\classifier.pyc", line 190, in chi2_spamprob . File "spambayes\classifier.pyc", line 493, in _getclues . File "spambayes\classifier.pyc", line 508, in _worddistanceget . File "spambayes\classifier.pyc", line 308, in probability .AssertionError


Thanks,
Brent.
____________________________________________________________
Brent Easton                       
Analyst/Programmer                               
University of Western Sydney                                   
Email: b.easton@uws.edu.au

From tameyer at ihug.co.nz  Fri Nov 26 01:20:52 2004
From: tameyer at ihug.co.nz (Tony Meyer)
Date: Fri Nov 26 01:21:28 2004
Subject: [Spambayes] Spambates Exceptions
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E8017AD46A@its-xchg4.massey.ac.nz>
Message-ID: <ECBA357DDED63B4995F5C1F5CBE5B1E8FBCA68@its-xchg4.massey.ac.nz>

> Have recently upgrades to v1.0 spambayes. Using Calypso mail 
> client with spambayes proxy server. I now get the following 
> header added to every email:
> 
> X-Spambayes-Exception: Traceback (most recent call last): . 
> File "sb_server.pyc", line 475, in onRetr . File 
> "spambayes\classifier.pyc", line 190, in chi2_spamprob . File 
> "spambayes\classifier.pyc", line 493, in _getclues . File 
> "spambayes\classifier.pyc", line 508, in _worddistanceget . 
> File "spambayes\classifier.pyc", line 308, in probability 
> .AssertionError

It's very rare to see this error these days.  What it means it that your
database has an entry that has been seen in more ham than the total number
of ham messages you have trained (obviously impossible).  The only cause I
can think of for this is that training was interrupted at just the right (or
wrong, depending on your point of view) moment.

There are two options to fix this:

  1.  The best option, by far, is to retrain from scratch.  Delete the two
database files and start training again - training is very quick, so it'll
hardly take any time to get back to high accuracy.  The FAQ explains how to
go about doing this.

  2.  You can manually repair this error in the database.  You'd have to
install Python (if you haven't already) and use the source version.  You use
the sb_dbexpimp.py script to convert the database to a CSV file, which you
can then open (eg in Excel) and change the numbers at the top to be at least
as large as the largest values in each of the ham & spam columns.  You then
use the sb_dbexpimp.py script to convert the fixed database back to the
original format.  We don't recommend this, as the problem might indicate
larger issues with the database, so it's much more reliable to just retrain.

Sorry the news isn't better!

=Tony.Meyer

-- 
Please always include the list (spambayes@python.org) in your replies
(reply-all), and please don't send me personal mail about SpamBayes.
http://www.massey.ac.nz/~tameyer/writing/reply_all.html explains this.

From tameyer at ihug.co.nz  Fri Nov 26 01:23:50 2004
From: tameyer at ihug.co.nz (Tony Meyer)
Date: Fri Nov 26 01:24:27 2004
Subject: [Spambayes] Spambayes problem!
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E8017AD3D6@its-xchg4.massey.ac.nz>
Message-ID: <ECBA357DDED63B4995F5C1F5CBE5B1E86C3824@its-xchg4.massey.ac.nz>

> I have 2 problems with spambayes, I did all the options
> that where shown in the Troubleshooting Guide. 
> One problem is that I can't get a second user on a computer
> with Outlook. If I try so I get this message:
> 
> Traceback (most recent call last): 
>   File "addin.py", line 1546, in ? 
>   File "win32com\server\register.pyc", line 483, in UseCommandLine 
>   File "win32com\server\register.pyc", line 413, in RegisterClasses 
>   File "win32com\server\register.pyc", line 181, in RegisterServer 
>   File "win32com\server\register.pyc", line 32, in _set_string 
> pywintypes.error: (5, 'RegSetValue', 'Access is denied.') 
[...]
> What is the problem?!?!?!

It appears that you don't have access to set the required registry key.
You'll need to run the registration utility as a user with suitably high
permission to be able to enable SpamBayes for all users (e.g. as
administrator).

> My second problem is that the button's won't work. If I select
> "Filter messages"  from the spambayes menu the spambayes program
> filter the messages for me, if I start spambayes it filers my
> messages olso. But the button's won't work.

This is with the working, first, user, correct?  We'd need to see your log
file (instructions about finding them are in the troubleshooting guide) to
know for sure, but the most common cause of these symptoms is a corrupted
database, and the solution is to retrain from scratch.

=Tony.Meyer

-- 
Please always include the list (spambayes@python.org) in your replies
(reply-all), and please don't send me personal mail about SpamBayes.
http://www.massey.ac.nz/~tameyer/writing/reply_all.html explains this.

From tameyer at ihug.co.nz  Fri Nov 26 01:30:27 2004
From: tameyer at ihug.co.nz (Tony Meyer)
Date: Fri Nov 26 01:31:25 2004
Subject: [Spambayes] Using spambayes
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E8017AD39B@its-xchg4.massey.ac.nz>
Message-ID: <ECBA357DDED63B4995F5C1F5CBE5B1E86C3825@its-xchg4.massey.ac.nz>

> Do I then have to delete all the messages which appear in
> the junkmail folder

It's up to you.  Once they are there, they're of no use to SpamBayes.  See
FAQ 3.17:

<http://spambayes.org/faq.html#is-it-ok-to-delete-the-messages-in-the-spam-f
older>

> and if I keep doing this, will the junk
> mail eventually be filtered and stop appearing ??

I'm not sure what you mean by this, so if my answer doesn't make sense,
please clarify and I'll try again :)

The way SpamBayes works is that it initially knows nothing about what you
think is good mail (ham) and what you think is bad mail (spam).  You teach
it by giving it examples of ham and spam, and it then classifies incoming
mail based on that knowledge.

The way that we recommend Outlook users train is to use the "Delete as spam"
and "Recover from spam" buttons (or just drag the messages to the ham/spam
folders) on all messages that end up in the unsure (Possible Junk) folder
and any messages that are incorrect (ham in the spam folder, spam not
moved).  It shouldn't take very much training before classification is very
good.

=Tony.Meyer

-- 
Please always include the list (spambayes@python.org) in your replies
(reply-all), and please don't send me personal mail about SpamBayes.
http://www.massey.ac.nz/~tameyer/writing/reply_all.html explains this.

From ta-meyer at ihug.co.nz  Fri Nov 26 01:33:40 2004
From: ta-meyer at ihug.co.nz (Tony Meyer)
Date: Fri Nov 26 01:34:16 2004
Subject: [Spambayes] RE: [Spambayes-announce] ANNOUNCE: SpamBayes release
	1.0.1
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E8017AD3F0@its-xchg4.massey.ac.nz>
Message-ID: <ECBA357DDED63B4995F5C1F5CBE5B1E86C3826@its-xchg4.massey.ac.nz>

>     >> __version__ = '1.0rc1'
>     >> 
>     >> Is this a mistake?
[...]
> Ah, crap.  Oh well.

:)  Seems to me it doesn't really matter what's there for CVS users.

What does Python do, out of curiosity?  (i.e. when 2.3 was first branched
and before 2.4a1, what did sys.version have?)

=Tony.Meyer

From anthony at interlink.com.au  Fri Nov 26 01:42:09 2004
From: anthony at interlink.com.au (Anthony Baxter)
Date: Fri Nov 26 01:42:24 2004
Subject: [Spambayes] RE: [Spambayes-announce] ANNOUNCE: SpamBayes release
	1.0.1
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E86C3826@its-xchg4.massey.ac.nz>
References: <ECBA357DDED63B4995F5C1F5CBE5B1E86C3826@its-xchg4.massey.ac.nz>
Message-ID: <200411261142.10399.anthony@interlink.com.au>

On Friday 26 November 2004 11:33, Tony Meyer wrote:
> :)  Seems to me it doesn't really matter what's there for CVS users.
>
> What does Python do, out of curiosity?  (i.e. when 2.3 was first branched
> and before 2.4a1, what did sys.version have?)

CVS gets updated to 2.3 final.
2.3 is cut from the CVS trunk.
release23-maint branch is created.
Trunk and branch get the version string updated to 2.3+ (this doesn't always
happen)
CVS branch gets 2.3.1rc1 
  "       "         "    2.3.1
  "       "       version updated to 2.3.1+

&c &c
From b.easton at exemail.com.au  Fri Nov 26 01:44:09 2004
From: b.easton at exemail.com.au (Brent Easton)
Date: Fri Nov 26 01:44:16 2004
Subject: [Spambayes] Spambates Exceptions
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E8FBCA68@its-xchg4.massey.ac.nz>
References: <ECBA357DDED63B4995F5C1F5CBE5B1E8FBCA68@its-xchg4.massey.ac.nz>
Message-ID: <200411261144090908.036FCF14@jay.exetel.com.au>

Hi Tony,

Thanks a lot for that. I ended up peeking at the source and finding the assert statement. I don't know python, but I guessed it looked like a corrupted db. I have decided to retrain.

Thanks again,
Brent.


*********** REPLY SEPARATOR  ***********

On 26/11/2004 at 1:20 PM Tony Meyer  wrote:

>> Have recently upgrades to v1.0 spambayes. Using Calypso mail 
>> client with spambayes proxy server. I now get the following 
>> header added to every email:
>> 
>> X-Spambayes-Exception: Traceback (most recent call last): . 
>> File "sb_server.pyc", line 475, in onRetr . File 
>> "spambayes\classifier.pyc", line 190, in chi2_spamprob . File 
>> "spambayes\classifier.pyc", line 493, in _getclues . File 
>> "spambayes\classifier.pyc", line 508, in _worddistanceget . 
>> File "spambayes\classifier.pyc", line 308, in probability 
>> .AssertionError
>
>It's very rare to see this error these days.  What it means it that your
>database has an entry that has been seen in more ham than the total number
>of ham messages you have trained (obviously impossible).  The only cause I
>can think of for this is that training was interrupted at just the right
>(or
>wrong, depending on your point of view) moment.
>
>There are two options to fix this:
>
>  1.  The best option, by far, is to retrain from scratch.  Delete the two
>database files and start training again - training is very quick, so it'll
>hardly take any time to get back to high accuracy.  The FAQ explains how to
>go about doing this.
>
>  2.  You can manually repair this error in the database.  You'd have to
>install Python (if you haven't already) and use the source version.  You
>use
>the sb_dbexpimp.py script to convert the database to a CSV file, which you
>can then open (eg in Excel) and change the numbers at the top to be at
>least
>as large as the largest values in each of the ham & spam columns.  You then
>use the sb_dbexpimp.py script to convert the fixed database back to the
>original format.  We don't recommend this, as the problem might indicate
>larger issues with the database, so it's much more reliable to just
>retrain.
>
>Sorry the news isn't better!
>
>=Tony.Meyer
>
>-- 
>Please always include the list (spambayes@python.org) in your replies
>(reply-all), and please don't send me personal mail about SpamBayes.
>http://www.massey.ac.nz/~tameyer/writing/reply_all.html explains this.


____________________________________________________________
Brent Easton                       
Analyst/Programmer                               
University of Western Sydney                                   
Email: b.easton@uws.edu.au

From tameyer at ihug.co.nz  Fri Nov 26 01:51:28 2004
From: tameyer at ihug.co.nz (Tony Meyer)
Date: Fri Nov 26 01:52:03 2004
Subject: [Spambayes] RE: [Spambayes-announce] ANNOUNCE: SpamBayes
	release1.0.1
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E8017AD4D4@its-xchg4.massey.ac.nz>
Message-ID: <ECBA357DDED63B4995F5C1F5CBE5B1E86C3828@its-xchg4.massey.ac.nz>

[Tony]
> > What does Python do, out of curiosity?  (i.e. when 2.3 was first 
> > branched and before 2.4a1, what did sys.version have?)

[Anthony]
> CVS gets updated to 2.3 final.
> 2.3 is cut from the CVS trunk.
> release23-maint branch is created.
> Trunk and branch get the version string updated to 2.3+ (this 
> doesn't always happen)
> CVS branch gets 2.3.1rc1 
>   "       "         "    2.3.1
>   "       "       version updated to 2.3.1+

So we could now have CVS Head's version 1.0+ and the maintenance branch
1.0.1+, then?  Sounds like a workable plan to me - any objections?  (I'll
update the "how to build a release" doc to include the step, too).

=Tony.Meyer

From mac_sutton at iprimus.com.au  Fri Nov 26 02:46:20 2004
From: mac_sutton at iprimus.com.au (Mac Sutton)
Date: Fri Nov 26 02:46:05 2004
Subject: [Spambayes] Hi guys
Message-ID: <000601c4d359$bbc338a0$55b132d2@mycomputer>

I would just like to now when will there be a version for outlook express coming out base on the same one as outlook uses.
Thank for your time and help have a nice day.


---
avast! Antivirus: Outbound message clean.
Virus Database (VPS): 0448-0, 11/23/2004
Tested on: 11/26/2004 12:46:23 PM
avast! - copyright (c) 1988-2004 ALWIL Software.
http://www.avast.com


-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.python.org/pipermail/spambayes/attachments/20041126/e1ec7f7d/attachment.htm
From tameyer at ihug.co.nz  Fri Nov 26 02:54:25 2004
From: tameyer at ihug.co.nz (Tony Meyer)
Date: Fri Nov 26 02:55:02 2004
Subject: [Spambayes] Hi guys
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E8017AD4ED@its-xchg4.massey.ac.nz>
Message-ID: <ECBA357DDED63B4995F5C1F5CBE5B1E8FBCA6F@its-xchg4.massey.ac.nz>

> I would just like to now when will there be a version
> for outlook express coming out base on the same one as 
> outlook uses.

You can already use SpamBayes with Outlook Express.  Please see FAQ 2.1:

<http://spambayes.org/faq.html#does-spambayes-work-with-outlook-express>

=Tony.Meyer

-- 
Please always include the list (spambayes@python.org) in your replies
(reply-all), and please don't send me personal mail about SpamBayes.
http://www.massey.ac.nz/~tameyer/writing/reply_all.html explains this.

From tameyer at ihug.co.nz  Fri Nov 26 03:00:15 2004
From: tameyer at ihug.co.nz (Tony Meyer)
Date: Fri Nov 26 03:00:52 2004
Subject: [Spambayes] RE: you have a patched pop3 proxy?
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E8017AD177@its-xchg4.massey.ac.nz>
Message-ID: <ECBA357DDED63B4995F5C1F5CBE5B1E8FBCA70@its-xchg4.massey.ac.nz>

> I heard rumors that you have a patched pop3 proxy.
> 
> What fixes have you incorporated?
> 
> How would one get access to it?

I don't, really.  I have the CVS version of SpamBayes, which will eventually
become 1.1 and includes a great many changes, but that's available to anyone
that wants to run from source.  There are no plans to build a binary version
of this until 1.1a1 is ready - I imagine that will be early in the new year.

All the changes in the 1.0 branch (i.e. all the bug fixes) that I have are
in the (released yesterday) 1.0.1 version.

I suspect you're basing this on the message I posted where I offered to make
a test version of the installer for someone (who hasn't since responded).
That would just have been the same as 1.0.1 but with a tentative change (and
maybe extra logging) specifically designed to try and fix one problem.  If I
was certain that it would if the problem, I'd just make the change, but I'm
not.  (That particular problem was to do with the Outlook plug-in, IIRC).

There's really just two ways to get access to the latest SpamBayes code:

  * The safe way - wait until a new version is released, and use that.

  * The bleeding edge way - run from CVS source and risk having things stop
working temporarily if things go wrong, or have to rebuild databases (etc)
if incompatible changes are made.


=Tony.Meyer

-- 
Please always include the list (spambayes@python.org) in your replies
(reply-all), and please don't send me personal mail about SpamBayes.
http://www.massey.ac.nz/~tameyer/writing/reply_all.html explains this.

From tameyer at ihug.co.nz  Fri Nov 26 03:06:50 2004
From: tameyer at ihug.co.nz (Tony Meyer)
Date: Fri Nov 26 03:07:52 2004
Subject: [Spambayes] email evades Spambayes
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E8017ACF2C@its-xchg4.massey.ac.nz>
Message-ID: <ECBA357DDED63B4995F5C1F5CBE5B1E86C3829@its-xchg4.massey.ac.nz>

> When I d/l'ed my email this am this one came through with no 
> indication that SpamBayes had checked it.  Everything else had 
> X-Spambayes-Classification headers with the ham/spam/unsure 
> value.  Is there anything in this that can let it evade SpamBayes?
> Or did SpamBayes just hiccup?

This is really weird - and a bit concerning since there was a similar report
recently.  I can't think of any way that mail could be downloaded through
SpamBayes and not get either a X-Spambayes-Classification header or a
X-Spambayes-Exception header, or crash.

If this happens again, could you switch to running sb_server in verbose mode
(put

[globals]
verbose:True

in your bayescustomize.ini file) and see if there's anything unusual in the
_pop3proxy.log file that will get created?

If it doesn't happen again, it was probably just cosmic alignment or
something <wink>.

=Tony.Meyer

-- 
Please always include the list (spambayes@python.org) in your replies
(reply-all), and please don't send me personal mail about SpamBayes.
http://www.massey.ac.nz/~tameyer/writing/reply_all.html explains this.

From tameyer at ihug.co.nz  Fri Nov 26 03:26:18 2004
From: tameyer at ihug.co.nz (Tony Meyer)
Date: Fri Nov 26 03:26:42 2004
Subject: [Spambayes] why doesn't the pop3 proxy filter all messages?
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E801763A01@its-xchg4.massey.ac.nz>
Message-ID: <ECBA357DDED63B4995F5C1F5CBE5B1E86C382A@its-xchg4.massey.ac.nz>

> Ok, it looks like the problem is the CAPA command from 
> eudora.  It looks like this is confusing the proxy server.

This should not be the case.  CAPA should work, although it should strip out
PIPELINING and STLS if they are present in the server's response.  You
should be able to telnet to the proxy and give it a "CAPA" command and get a
response.  Does that work?

> My hypothesis is that if I can get Eudora to not send the 
> CAPA command, that we should work better.  Does this sound
> reasonable?

If there is a problem with the CAPA command (it's possible, although
unlikely, since I can't see how this wouldn't have hit many people already),
then it's a SpamBayes problem, so we need to fix something.

[Next message]
> Ok, when I set never for secure sockets when receiving 
> options, then the CAPA is not sent and it looks like 
> things work.

I presume you weren't actually using SSL (TLS) in any case, since that
wouldn't have worked through the proxy.

I still don't understand:

  (a) Why CAPA is causing problems
  (b) How mail can be delivered if the proxy dies after a CAPA (is there a
second, direct, connection to the mail server?)

=Tony.Meyer

-- 
Please always include the list (spambayes@python.org) in your replies
(reply-all), and please don't send me personal mail about SpamBayes.
http://www.massey.ac.nz/~tameyer/writing/reply_all.html explains this.

From tameyer at ihug.co.nz  Fri Nov 26 03:28:23 2004
From: tameyer at ihug.co.nz (Tony Meyer)
Date: Fri Nov 26 03:29:05 2004
Subject: [Spambayes] email evades Spambayes
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E8017AD150@its-xchg4.massey.ac.nz>
Message-ID: <ECBA357DDED63B4995F5C1F5CBE5B1E8FBCA73@its-xchg4.massey.ac.nz>

> There is a bug of some sort in the pop3 proxy which causes it 
> to pass all mail without filtering if it hangs.

Are you positive?  I cannot reproduce this (Eudora, Thunderbird, Outlook
Express) - if the proxy hangs, then the mail clients generates an error, as
one would expect.  Are you certain there isn't a secondary connection
directly to the server that's working and gets the mail instead?  A test
would be whether mail that's downloaded when the proxy hangs has the
SpamBayes headers.

=Tony.Meyer

-- 
Please always include the list (spambayes@python.org) in your replies
(reply-all), and please don't send me personal mail about SpamBayes.
http://www.massey.ac.nz/~tameyer/writing/reply_all.html explains this.

From tameyer at ihug.co.nz  Fri Nov 26 03:31:37 2004
From: tameyer at ihug.co.nz (Tony Meyer)
Date: Fri Nov 26 03:33:28 2004
Subject: [Spambayes] Msg for Tony about earlier problems.
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E8017AD172@its-xchg4.massey.ac.nz>
Message-ID: <ECBA357DDED63B4995F5C1F5CBE5B1E8FBCA74@its-xchg4.massey.ac.nz>

[Description of doing a clean install]
> Hey presto: all is now working fine. 

:)

> So far it's been running for just over a week and the only 
> real issue I have isn't with SB itself it's with configuring
> the AV not to touch the SB cache, where there are [as one might
> expect] a lot of infected items which never make it into our
> mail system thanks to pre-processing of SB headers.

This is similar to the problem that some people have where their anti-virus
software finds viruses in the cache directories.  I can't think of any way
to work around this, though, other than manually telling the AV software
that it's not a problem (as you describe).

> I /think/ that's cracked the problem. Many thanks for your 
> patience and the time spent on debugging.

No worries - let me know if there are any more problems.  Thanks for your
help in figuring out how the service should operate - I'll commit the
updated script at some point so it'll behave much better in 1.1.

=Tony.Meyer

-- 
Please always include the list (spambayes@python.org) in your replies
(reply-all), and please don't send me personal mail about SpamBayes.
http://www.massey.ac.nz/~tameyer/writing/reply_all.html explains this.

From rolph at 110.net  Fri Nov 26 03:38:26 2004
From: rolph at 110.net (The Rolphs)
Date: Fri Nov 26 03:38:40 2004
Subject: [Spambayes] email evades Spambayes
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E86C3829@its-xchg4.massey.ac .nz>
References: <ECBA357DDED63B4995F5C1F5CBE5B1E8017ACF2C@its-xchg4.massey.ac.nz>
Message-ID: <5.1.0.14.2.20041125213604.0224de60@localhost>

I did t his.

I can confirm the following:

If the CAPA command is given SOME of the time the server connection fails 
to close after the QUIT command.  Under these conditions, the pop3 proxy 
passes the commands for the next session(s) without filtering the 
email.  It is moderately reproducible.

I also have tow cases under 1.0 where this occurred even with very simple 
pop3 commands sequences, but I have been unable to crisply reproduce this.

At 03:06 PM 11/26/2004 +1300, Tony Meyer wrote:
> > When I d/l'ed my email this am this one came through with no
> > indication that SpamBayes had checked it.  Everything else had
> > X-Spambayes-Classification headers with the ham/spam/unsure
> > value.  Is there anything in this that can let it evade SpamBayes?
> > Or did SpamBayes just hiccup?
>
>This is really weird - and a bit concerning since there was a similar report
>recently.  I can't think of any way that mail could be downloaded through
>SpamBayes and not get either a X-Spambayes-Classification header or a
>X-Spambayes-Exception header, or crash.
>
>If this happens again, could you switch to running sb_server in verbose mode
>(put
>
>[globals]
>verbose:True
>
>in your bayescustomize.ini file) and see if there's anything unusual in the
>_pop3proxy.log file that will get created?
>
>If it doesn't happen again, it was probably just cosmic alignment or
>something <wink>.
>
>=Tony.Meyer
>
>--
>Please always include the list (spambayes@python.org) in your replies
>(reply-all), and please don't send me personal mail about SpamBayes.
>http://www.massey.ac.nz/~tameyer/writing/reply_all.html explains this.
>
>_______________________________________________
>Spambayes@python.org
>http://mail.python.org/mailman/listinfo/spambayes
>Check the FAQ before asking: http://spambayes.sf.net/faq.html


Regards.

The Rolphs rolph@110.net



From jwilliam at xmission.com  Fri Nov 26 04:02:22 2004
From: jwilliam at xmission.com (Jerry Williams)
Date: Fri Nov 26 04:02:26 2004
Subject: [Spambayes] "Check for new version" proxy settings?
Message-ID: <DMEHJCBFHDFHAHCIBLJPGEOBFHAA.jwilliam@xmission.com>

The release notes for 1.0.1 talk about putting proxy detail in the
configuration file.

But I can't find any information on what file and what information needs to
be put there in what format.

Thanks in advance for any pointers on how to do this.

From skip at pobox.com  Fri Nov 26 04:28:14 2004
From: skip at pobox.com (Skip Montanaro)
Date: Fri Nov 26 04:28:24 2004
Subject: [Spambayes] RE: [Spambayes-announce] ANNOUNCE: SpamBayes release
	1.0.1
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E86C3826@its-xchg4.massey.ac.nz>
References: <ECBA357DDED63B4995F5C1F5CBE5B1E8017AD3F0@its-xchg4.massey.ac.nz>
	<ECBA357DDED63B4995F5C1F5CBE5B1E86C3826@its-xchg4.massey.ac.nz>
Message-ID: <16806.41678.425431.520324@montanaro.dyndns.org>


    Tony> What does Python do, out of curiosity?  (i.e. when 2.3 was first
    Tony> branched and before 2.4a1, what did sys.version have?)

Current cvs head seems like it says "2.4c1".  Since that's the current
release, I infer that cvs head matches the current version more-or-less.  I
think that fairly soon after the 2.4 release the cvs head version will read
2.5a0.

Skip


From rolph at 110.net  Fri Nov 26 04:31:57 2004
From: rolph at 110.net (The Rolphs)
Date: Fri Nov 26 04:32:12 2004
Subject: [Spambayes] why doesn't the pop3 proxy filter all messages?
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E86C382A@its-xchg4.massey.ac .nz>
References: <ECBA357DDED63B4995F5C1F5CBE5B1E801763A01@its-xchg4.massey.ac.nz>
Message-ID: <5.1.0.14.2.20041125222445.022616e0@localhost>

I can confirm that sending the CAPA causes the session to fail to close 
reliably from a telnet session through the proxy.  Not using the proxy 
results in correct receipt of mail by the client.  Using the proxy results 
in all mail being received, but many of the messages do not have the 
X-Spambayes headers.

what I think would resolve this is for the proxy to send an extra <CR><LF> 
after the QUIT command to ensure that the session has closed.

I have had two of these cases after changing the  check forr SSL setting, 
but the problem is greatly reduced.

At 03:26 PM 11/26/2004 +1300, Tony Meyer wrote:
> > Ok, it looks like the problem is the CAPA command from
> > eudora.  It looks like this is confusing the proxy server.
>
>This should not be the case.  CAPA should work, although it should strip out
>PIPELINING and STLS if they are present in the server's response.  You
>should be able to telnet to the proxy and give it a "CAPA" command and get a
>response.  Does that work?
>
> > My hypothesis is that if I can get Eudora to not send the
> > CAPA command, that we should work better.  Does this sound
> > reasonable?
>
>If there is a problem with the CAPA command (it's possible, although
>unlikely, since I can't see how this wouldn't have hit many people already),
>then it's a SpamBayes problem, so we need to fix something.
>
>[Next message]
> > Ok, when I set never for secure sockets when receiving
> > options, then the CAPA is not sent and it looks like
> > things work.
>
>I presume you weren't actually using SSL (TLS) in any case, since that
>wouldn't have worked through the proxy.
>
>I still don't understand:
>
>   (a) Why CAPA is causing problems
>   (b) How mail can be delivered if the proxy dies after a CAPA (is there a
>second, direct, connection to the mail server?)
>
>=Tony.Meyer
>
>--
>Please always include the list (spambayes@python.org) in your replies
>(reply-all), and please don't send me personal mail about SpamBayes.
>http://www.massey.ac.nz/~tameyer/writing/reply_all.html explains this.


Regards.

The Rolphs rolph@110.net



From ta-meyer at ihug.co.nz  Fri Nov 26 04:34:57 2004
From: ta-meyer at ihug.co.nz (Tony Meyer)
Date: Fri Nov 26 04:35:33 2004
Subject: [Spambayes] RE: [Spambayes-announce] ANNOUNCE: SpamBayes release
	1.0.1
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E8017AD51E@its-xchg4.massey.ac.nz>
Message-ID: <ECBA357DDED63B4995F5C1F5CBE5B1E8FBCA75@its-xchg4.massey.ac.nz>

> Current cvs head seems like it says "2.4c1".  Since that's 
> the current release, I infer that cvs head matches the 
> current version more-or-less.  I think that fairly soon after 
> the 2.4 release the cvs head version will read 2.5a0.

Hmmm...this doesn't match the all-powerful-release-manager's answer at all
:)

So should spambayes have "1.0+" or "1.1a0" for CVS head?  And "1.0.1",
"1.0.1+", or "1.0.2a0" for the maintenance branch?

=Tony.Meyer

From anthony at interlink.com.au  Fri Nov 26 04:51:42 2004
From: anthony at interlink.com.au (Anthony Baxter)
Date: Fri Nov 26 04:51:57 2004
Subject: [Spambayes] Hi guys
In-Reply-To: <000601c4d359$bbc338a0$55b132d2@mycomputer>
References: <000601c4d359$bbc338a0$55b132d2@mycomputer>
Message-ID: <200411261451.42990.anthony@interlink.com.au>

On Friday 26 November 2004 12:46, Mac Sutton wrote:
> I would just like to now when will there be a version for outlook express
> coming out base on the same one as outlook uses. Thank for your time and

It's unlikely that there will ever be a version for outlook express. It's a
completely different beast to outlook, and none of the SB developers are
inclined to work on it.
From anthony at interlink.com.au  Fri Nov 26 04:55:48 2004
From: anthony at interlink.com.au (Anthony Baxter)
Date: Fri Nov 26 04:56:05 2004
Subject: [Spambayes] RE: [Spambayes-announce] ANNOUNCE: SpamBayes release
	1.0.1
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E8FBCA75@its-xchg4.massey.ac.nz>
References: <ECBA357DDED63B4995F5C1F5CBE5B1E8FBCA75@its-xchg4.massey.ac.nz>
Message-ID: <200411261455.48599.anthony@interlink.com.au>

On Friday 26 November 2004 14:34, Tony Meyer wrote:
> > Current cvs head seems like it says "2.4c1".  Since that's
> > the current release, I infer that cvs head matches the
> > current version more-or-less.  I think that fairly soon after
> > the 2.4 release the cvs head version will read 2.5a0.
>
> Hmmm...this doesn't match the all-powerful-release-manager's answer at all

It's not too far off. I said in the original that I don't bother updating to
2.x+ all the time. I forgot about 2.(x+1)a0 - usually that happens a 
bit down the track, once there's actually been changes to the trunk...

From rolph at 110.net  Fri Nov 26 05:02:58 2004
From: rolph at 110.net (The Rolphs)
Date: Fri Nov 26 05:03:21 2004
Subject: [Spambayes] email evades Spambayes
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E8FBCA73@its-xchg4.massey.ac .nz>
References: <ECBA357DDED63B4995F5C1F5CBE5B1E8017AD150@its-xchg4.massey.ac.nz>
Message-ID: <5.1.0.14.2.20041125230215.022762f0@localhost>

I can confirm there is no secondary connection.  A check of netstat -a 
doesn't show it and the mail client only knows about  the localhost servers.

At 03:28 PM 11/26/2004 +1300, Tony Meyer wrote:
> > There is a bug of some sort in the pop3 proxy which causes it
> > to pass all mail without filtering if it hangs.
>
>Are you positive?  I cannot reproduce this (Eudora, Thunderbird, Outlook
>Express) - if the proxy hangs, then the mail clients generates an error, as
>one would expect.  Are you certain there isn't a secondary connection
>directly to the server that's working and gets the mail instead?  A test
>would be whether mail that's downloaded when the proxy hangs has the
>SpamBayes headers.
>
>=Tony.Meyer
>
>--
>Please always include the list (spambayes@python.org) in your replies
>(reply-all), and please don't send me personal mail about SpamBayes.
>http://www.massey.ac.nz/~tameyer/writing/reply_all.html explains this.


Regards.

The Rolphs rolph@110.net



From tameyer at ihug.co.nz  Fri Nov 26 05:09:40 2004
From: tameyer at ihug.co.nz (Tony Meyer)
Date: Fri Nov 26 05:10:57 2004
Subject: [Spambayes] RE: [Spambayes-announce] ANNOUNCE: SpamBayes
	release1.0.1
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E8017AD530@its-xchg4.massey.ac.nz>
Message-ID: <ECBA357DDED63B4995F5C1F5CBE5B1E86C382C@its-xchg4.massey.ac.nz>

> It's not too far off. I said in the original that I don't 
> bother updating to 2.x+ all the time. I forgot about 2.(x+1)a0 -
> usually that happens a bit down the track, once there's actually
> been changes to the trunk...

Ah, ok.  So as a draft to add to README-DEVEL (at the end of the process):

 o Change spambayes/__init__.py to contain the new version number with a
   '+' at the end, to indicate to CVS users that they are using code after
   the release, and check this change in.  After a number of changes have
   been checked in, this can be incremented and have "a0" added to the end.
   For example, with a 1.1 release:
       [before the release process] '1.1rc1'
       [during the release process] '1.1'
       [after the release process]  '1.1+'
       [later]                      '1.2a0'

Seem ok?  That'd make CVS Head '1.1a0' and the maintenance branch '1.0.1+'
now.

=Tony.Meyer

From tameyer at ihug.co.nz  Fri Nov 26 05:25:05 2004
From: tameyer at ihug.co.nz (Tony Meyer)
Date: Fri Nov 26 05:25:42 2004
Subject: [Spambayes] "Check for new version" proxy settings?
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E8017AD510@its-xchg4.massey.ac.nz>
Message-ID: <ECBA357DDED63B4995F5C1F5CBE5B1E8FBCA78@its-xchg4.massey.ac.nz>

> The release notes for 1.0.1 talk about putting proxy detail
> in the configuration file.
> 
> But I can't find any information on what file and what
> information needs to be put there in what format.
> 
> Thanks in advance for any pointers on how to do this.

Firstly, if the "Check for new version" function already works for you, then
you don't need to do this.  The information can usually be retrieved
automatically and so you don't need to do anything.

Secondly, if all you have is the location of an 'autoconfigure' script (as
IE uses, but also supported by Firefox and others), that won't work.
Supporting those is a lot of work, and not really worth it just for a
version check.

However, if it doesn't currently work, and you have the proxy's address (and
optionally a username & password), then you need to put these in the
'default_bayes_customize.ini' file (for Outlook) or
'Proxy\bayescustomze.ini' file (for non-Outlook) that's in your data
directory (the FAQ explains where to find that).

Add these lines to the start or end of the file:

[globals]
proxy_username:foo
proxy_password:bar
proxy_server:proxy.example.com:8891

If you don't need a username/password, then those can be left out.  If the
port is 8880, then you can leave the colon and port off the end of the
server name.

=Tony.Meyer

-- 
Please always include the list (spambayes@python.org) in your replies
(reply-all), and please don't send me personal mail about SpamBayes.
http://www.massey.ac.nz/~tameyer/writing/reply_all.html explains this.

From tameyer at ihug.co.nz  Fri Nov 26 05:45:26 2004
From: tameyer at ihug.co.nz (Tony Meyer)
Date: Fri Nov 26 05:46:03 2004
Subject: [Spambayes] email evades Spambayes
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E8017AD508@its-xchg4.massey.ac.nz>
Message-ID: <ECBA357DDED63B4995F5C1F5CBE5B1E86C382E@its-xchg4.massey.ac.nz>

> If the CAPA command is given SOME of the time the server 
> connection fails to close after the QUIT command.

What do you get back from the CAPA command (a) with the proxy and (b) with a
direct connection to the server?  We modify the CAPA response, so maybe I
can't duplicate this since my response doesn't get duplicated (I don't have
access to a POP3 server that has a CAPA response that would be modified).
However, it does pass the test script...

> Under these conditions, the pop3 proxy passes the commands for
> the next session(s) without filtering the email.  It is
> moderately reproducible.

This is also very strange (a second connection should be through the proxy,
no matter what stage the first connection is at).  What happens then, if you
cause this problem with a telnet connection, and before ending the session
(i.e. while it is hung) open another telnet connection to the proxy?  Does
this then result in a connection where you have direct access to the server?

Does anything get printed to the (spambayes) log?  (The troubleshooting
guide explains where this is, or it's the console window when running from
source).

> what I think would resolve this is for the proxy to send an 
> extra <CR><LF> after the QUIT command to ensure that the session
> has closed.

I don't believe that this would solve the problem.  The proxy sends all
outgoing data exactly as is, no matter what.  There's no reason that any
sequence of actions should effect the server's response to a QUIT command,
unless there's something odd about that specific POP server.  (Do you only
use the one POP server?  If there's more than one, does the problem happen
with both?)

> I have had two of these cases after changing the check forr 
> SSL setting, but the problem is greatly reduced.

Without checking (and assuming that the problem is with CAPA), I assume this
is because Eudora doesn't use CAPA in most circumstances if it doesn't need
to check for STARTTLS ability.

=Tony.Meyer

-- 
Please always include the list (spambayes@python.org) in your replies
(reply-all), and please don't send me personal mail about SpamBayes.
http://www.massey.ac.nz/~tameyer/writing/reply_all.html explains this.

From tameyer at ihug.co.nz  Fri Nov 26 05:53:43 2004
From: tameyer at ihug.co.nz (Tony Meyer)
Date: Fri Nov 26 05:54:18 2004
Subject: [Spambayes] Re: How to recover spam to a specific ham folder
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E801763A0F@its-xchg4.massey.ac.nz>
Message-ID: <ECBA357DDED63B4995F5C1F5CBE5B1E86C382F@its-xchg4.massey.ac.nz>

> What I can see though is that all emails first show up in the 
> main entrance folder, then they are being further processed: 
> Outlook rules move them to the appropriate subfolders, 
> Spambayes checks for ham/spam status and moves them to the 
> corresponding folders (spam or unsure
> messages) if necessary. An explanation for the problem could 
> be, if Spambayes comes before Outlook, it moves the emails 
> from the main folder to the spam folder or to the unsure 
> messages folder, an undo would just put them back into the 
> main folder.

If background filtering is enabled (it is by default), then SpamBayes tries
to run after all Outlook rules (you can adjust the values - Advanced tab of
the Manager dialog - to ensure that this is the case).

> I changed my Spymbayes configuration and took out the main 
> folder as ham folder, so Spambayes can only operate on the 
> subfolders. Now it should restore correctly, because now it 
> knows the subfolder an email came from. But Spambayes doesn't 
> do as exspected: An email with target subfolder B and moved 
> by Spambayes to the spam folder is recovered to subfolder A 
> (the first one in Spambayes' ham folder list).

This is very odd.  If SpamBayes can't figure out which folder to restore to
(i.e. if the error is printed to the log), then it should try to put it in
what Outlook thinks is the default inbox, which I presume would be the main
folder, and not subfolder A.

Now that you've made the change and the main folder isn't in the watched
folders list, do you get an error printed in your log (the troubleshooting
guide explains where that is) saying "Error locating origin of message"?

If you do get the error, then the problem is with (a) saving/retrieving the
original folder info and (b) subfolder A being returned as the default
inbox.

If you do not get the error, then the problem must be with saving the
original folder info - i.e. SpamBayes is saving subfolder A as the original
folder, rather than subfolder B.  If that is the case, are there any
additional Outlook rules running on either subfolder?

=Tony.Meyer

-- 
Please always include the list (spambayes@python.org) in your replies
(reply-all), and please don't send me personal mail about SpamBayes.
http://www.massey.ac.nz/~tameyer/writing/reply_all.html explains this.

From tameyer at ihug.co.nz  Fri Nov 26 06:21:57 2004
From: tameyer at ihug.co.nz (Tony Meyer)
Date: Fri Nov 26 06:22:33 2004
Subject: [Spambayes] It all started with SP2
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E8017AD16C@its-xchg4.massey.ac.nz>
Message-ID: <ECBA357DDED63B4995F5C1F5CBE5B1E8FBCA7D@its-xchg4.massey.ac.nz>

> 6. The other outlook plugin is Plaxxo, but it hasn't really
> given me any troubles prior to, or since.

I downloaded that in case it was some sort of conflict, but I don't have any
troubles running them both, so I presume it's not.

> 7. I think I could try a download if that might work. Since
> our server blocks executables, I'd have to figure out a way 
> to do that with your help, but I can download and install 
> software in my PC from the web.

I can put it online easily enough.  What I'm testing is whether setting the
toolbar to temporary instead of permanent will change anything (other than
that, it is exactly identical to 1.0.1).  If this version doesn't help, then
you can just uninstall and delete it and put the regular one back.

To try it out:

  1.  Download it from
<http://www.massey.ac.nz/~tameyer/spambayes-tester-261104.exe>

  2.  Close Outlook.

  3.  Uninstall the existing SpamBayes (with the regular Windows Add/Remove
Programs control panel)

  4.  Completely reset the Outlook toolbars by removing the file \Documents
and Settings\{username}\Application Data\Microsoft\Outlook\outcmd.dat.

  5.  Run the installer.

  6.  Open Outlook and try the toolbar out.

=Tony.Meyer

-- 
Please always include the list (spambayes@python.org) in your replies
(reply-all), and please don't send me personal mail about SpamBayes.
http://www.massey.ac.nz/~tameyer/writing/reply_all.html explains this.

From tameyer at ihug.co.nz  Fri Nov 26 06:42:34 2004
From: tameyer at ihug.co.nz (Tony Meyer)
Date: Fri Nov 26 06:43:10 2004
Subject: [Spambayes] Please do a plug-in version for Outlook Express as
	well!
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E801521A9E@its-xchg4.massey.ac.nz>
Message-ID: <ECBA357DDED63B4995F5C1F5CBE5B1E8FBCA7F@its-xchg4.massey.ac.nz>

> Thanks for the through explanation. I have lots of programs 
> running on my computer: ZoneAlarm, AVG, SpywareGuard, Spybot,
> MrPostman, etc. to name a few.  I just was trying to reduce
> the number of running programs by one!

A plug-in will take up more or less the same amount of memory and processing
time, so all you'd really gain is reducing the list of running programs in
Task Manager (which, given how many processes are running with an average XP
system, is nearly pointless).

> About the proxy, it seems to me that it doesn't come with a 
> default list of filter words.  It would be better to have a list
> of words in addition to the learning feature, so it will be
> ready to do something as soon as it is installed :)

This has been discussed a number of times.  The arguments against are:

  (a) You really need to supply both sample ham and spam messages to start
off with, but how do you choose these?  Are there examples of ham and spam
that will really be universally considered such?

  (b) SpamBayes learns really fast - you only need to train on a dozen
messages and results are fairly satisfactory*.  So you're not really gaining
that much.

  (c) The success of filters like this lies in the fact that it learns what
you personally consider good & bad mail, and this works against that.

All that said, it would be simple enough to do if someone was going to do
the work in (a) of picking the messages (and could be optional).  IIRC, one
of the developers (Skip?) even started work along these lines, but there
really isn't much demand for it, and so it's never got very far.

If you strongly feel that it would be a good addition, the best move would
be to add a tracker to the feature requests (RFE) on sourceforge:
<http://sf.net/projects/spambayes>.  That way it won't get lost in list
traffic, and it's easy for other people to add their pleas.

=Tony.Meyer

* I use the 'bigrams' option, still experimental with 1.0.1, which probably
makes learning quicker.  But it should 
still be quick even without that.

-- 
Please always include the list (spambayes@python.org) in your replies
(reply-all), and please don't send me personal mail about SpamBayes.
http://www.massey.ac.nz/~tameyer/writing/reply_all.html explains this.

From robert at imagine.co.uk  Fri Nov 26 08:59:59 2004
From: robert at imagine.co.uk (Robert Neuschul)
Date: Fri Nov 26 08:59:59 2004
Subject: [Spambayes] Msg for Tony about earlier problems.
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E8FBCA74@its-xchg4.massey.ac.nz>
References: <ECBA357DDED63B4995F5C1F5CBE5B1E8FBCA74@its-xchg4.massey.ac.nz>
Message-ID: <VA.0000576b.00194be4@imagine.co.uk>

Tony

In article <ECBA357DDED63B4995F5C1F5CBE5B1E8FBCA74@its-xchg4.massey.ac.nz>, 
Tony Meyer wrote:
> This is similar to the problem that some people have where their anti-virus
> software finds viruses in the cache directories.  I can't think of any way
> to work around this, though, other than manually telling the AV software
> that it's not a problem (as you describe).

What I'd really like is for the AV to be able to proxy the proxy :-)

IOW, I'd like SB to be able to pass the mail directly to the AV in memory in 
real-time before it even hits the disk: if the item is infected it gets 
dumped by SB [pipe to dev/null or w.h.y.] without further inspection and with 
nothing even appearing in the SB training/cache etc. If the item is clean it 
goes on to normal SB processing: that way the cache and DB would always be 
clean and wouldn't need exclusions in any AV config.
It would place a heck of a load on the AV being robust and reliable, but it 
would cut the load on SB processing.
I'm sure I could spend a day or so messing with some real-time AV processing 
and get something like that to work, but there are other things in my life 
right now :-)

Robert.


From patric at www.blackxtreme.nl  Sat Nov 27 12:08:01 2004
From: patric at www.blackxtreme.nl (patric@www.blackxtreme.nl)
Date: Sat Nov 27 12:26:05 2004
Subject: [Spambayes] Automated reply from patric@www.blackxtreme.nl
Message-ID: <200411271108.iARB81N03645@ws5.vip.nl>

(dit is een automatische reply op uw mail)

Dank je voor je email, je krijgt zsm reactie van me. 
ps voor aanvraag/info boekingen stuur je mail door naar info@blackxtreme.nl dan krijg je sneller reactie!!!

Thank you for your email, i will contact you asap. 
ps for booking/info forward your email to info@blackxtreme.nl for quick response!!!


@ NOTE FOR YOU
________________________________________________________
Voor de leukste Striptease shows en Special acts bekend van tv
kijk snel op onze site www.blackxtreme.nl en zoek de leukste shows uit.

Email/: patric@blackxtreme.nl 
Info tel: 06 - 1866 50 37
MSN Patric: blackxcrow@msn.com (geen email)

Please visite our website if you are looking for Special Acts for your party
BlackXtreme winner EroticAward2000 "the best performance of Europe" 
www.blackxtreme.nl
------------------------------------------------------------------------------------------------------------------
Bezoek ook eens ons webwinkeltje:
http://www.blackxtreme.nl/Mijnwinkeltje.html
Lekker relax luisteren naar je favo muziekzender ga dan naar:
http://www.blackxtreme.nl/patricmusicroom.html

xxx have fun
----------------------------------------------------------------------------------------------------------------
From usr7-28hl at xemaps.com  Sun Nov 28 02:44:13 2004
From: usr7-28hl at xemaps.com (Keith Russell)
Date: Sun Nov 28 02:44:18 2004
Subject: [Spambayes] Unable to Access SpamBayes Browser Interface Using
	sb_imapfilter.py
In-Reply-To: <41A2C826.9080702@mailsnare.net>
References: <41887AB5.2030008@xemaps.com>	<rsl4q0pqvl3avisechaep2rqclb0clgknj@4ax.com>
	<41A2C826.9080702@mailsnare.net>
Message-ID: <41A92D6D.60609@mailsnare.net>

Sorry for my frequent reposts on this topic in the past. I don't 
know why I'm missing messages from the list, but I just checked 
the archives and found that Tony had replied to this thread--and 
it never arrived in my inbox.

So here's my response to Tony....

Tony Meyer wrote:

>> A "telnet localhost 8880" also fails. So where do we go from here, 
>> since it seems as though I'm the only one with this bug, or 
>> whatever it may be?
> 
> What if you try another port?  (4321 or something).  (To check that it's not
> that port 8880 is unavailable, although I would expect an error in that
> case).  To do that do:
> 
>> sb_imapfilter.py -b -v -o html_ui:port:4321

Same results as before. The process starts with the expected 
output, ending in "Done." But the configuration page doesn't 
open, and http://localhost:4321 fails with "Cannot find 
server/The page cannot be displayed". Also "telnet localhost 
4321" gives an error.

> What happens if you run:
> 
>> sb_server.py -b
> 
> Can you get to the web interface then?  (If you can, then it must be an
> imapfilter problem.  If you can't, then it's either something very odd in
> the lower level web interface code, or something outside the SpamBayes
> code).

It works perfectly. The configuration page opens automatically. I 
also noticed that a message displays saying, "User interface url 
is http://localhost:8880". I don't get this with 
sb_imapfilter.py. Should I?

I should say that all these tests today were done after 
downloading and installing the new official release.

Thanks, everyone, for your help. I hate to give up on SpamBayes, 
but this problem first appeared on September 30, and I need to 
find something that works.

-- 
Keith
From Amir_Katz at bmc.com  Sun Nov 28 13:59:48 2004
From: Amir_Katz at bmc.com (Katz, Amir)
Date: Sun Nov 28 13:59:54 2004
Subject: [Spambayes] Using spambayes
Message-ID: <F8B9B12641FB6B4FAC761BDEF9D63B04033ED161@tlv-ex-01.adprod.bmc.com>

One good practice I can offer to MS Outlook users - let SB move the spams to
a folder of your choice ("Junk E-mail" would be a good choice in Outlook
2003), but then move them to a special folder in a separate PST file. This
way, you keep all the spam messages but your normal PST does not grow. Then,
if at some point you run into an unrecoverable SB problem, you can build the
knowledge base from scratch by using the corpus that you kept on the side.

Amir 

-----Original Message-----
From: Tony Meyer [mailto:tameyer@ihug.co.nz] 
Sent: Friday, November 26, 2004 02:30
To: 'Maureen Cox'; spambayes@python.org
Subject: RE: [Spambayes] Using spambayes

> Do I then have to delete all the messages which appear in
> the junkmail folder

It's up to you.  Once they are there, they're of no use to SpamBayes.  See
FAQ 3.17:

<http://spambayes.org/faq.html#is-it-ok-to-delete-the-messages-in-the-spam-f
older>

> and if I keep doing this, will the junk mail eventually be filtered and
stop appearing ??

I'm not sure what you mean by this, so if my answer doesn't make sense,
please clarify and I'll try again :)

The way SpamBayes works is that it initially knows nothing about what you
think is good mail (ham) and what you think is bad mail (spam).  You teach
it by giving it examples of ham and spam, and it then classifies incoming
mail based on that knowledge.

The way that we recommend Outlook users train is to use the "Delete as spam"
and "Recover from spam" buttons (or just drag the messages to the ham/spam
folders) on all messages that end up in the unsure (Possible Junk) folder
and any messages that are incorrect (ham in the spam folder, spam not
moved).  It shouldn't take very much training before classification is very
good.

=Tony.Meyer

-- 
Please always include the list (spambayes@python.org) in your replies
(reply-all), and please don't send me personal mail about SpamBayes.
http://www.massey.ac.nz/~tameyer/writing/reply_all.html explains this.

_______________________________________________
Spambayes@python.org
http://mail.python.org/mailman/listinfo/spambayes
Check the FAQ before asking: http://spambayes.sf.net/faq.html
From tameyer at ihug.co.nz  Mon Nov 29 04:52:29 2004
From: tameyer at ihug.co.nz (Tony Meyer)
Date: Mon Nov 29 04:53:03 2004
Subject: [Spambayes] Using spambayes
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E8017AD9C0@its-xchg4.massey.ac.nz>
Message-ID: <ECBA357DDED63B4995F5C1F5CBE5B1E8FBCA8F@its-xchg4.massey.ac.nz>

> One good practice I can offer to MS Outlook users - let SB 
> move the spams to a folder of your choice ("Junk E-mail" 
> would be a good choice in Outlook 2003), but then move them 
> to a special folder in a separate PST file. This way, you 
> keep all the spam messages but your normal PST does not grow. 

FWIW, I do this, too.  If you're going to keep your spam messages, then it
would definitely be worth moving them other than your main pst file, since
it certainly seems that Outlook performance suffers as a pst file gets
large.

=Tony.Meyer

From tameyer at ihug.co.nz  Mon Nov 29 05:10:49 2004
From: tameyer at ihug.co.nz (Tony Meyer)
Date: Mon Nov 29 05:44:13 2004
Subject: [Spambayes] Unable to Access SpamBayes Browser Interface
	Usingsb_imapfilter.py
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E8017AD8F8@its-xchg4.massey.ac.nz>
Message-ID: <ECBA357DDED63B4995F5C1F5CBE5B1E8FBCA91@its-xchg4.massey.ac.nz>

> Thanks, everyone, for your help. I hate to give up on SpamBayes, 
> but this problem first appeared on September 30, and I need to 
> find something that works.

If you mean that you want to stop trying to figure out what's wrong and move
onto something else, then I understand.  The 'related' page on our website
links to various other filters that all should work very well -
unfortunately I'm not familiar enough with them to know which work with IMAP
- I gather POPfile has experimental IMAP support, so maybe that would be a
good choice.

<http://spambayes.org/related.html>

If this is the case, then you can ignore the rest of this.

> Sorry for my frequent reposts on this topic in the past. I don't 
> know why I'm missing messages from the list, but I just checked 
> the archives and found that Tony had replied to this thread--and 
> it never arrived in my inbox.

Maybe you have mailman set to avoid sending you copies of messages that
include you in the recipient list (this is the default, I believe), but for
some reason (bad reply address?) you aren't getting the remaining one?  You
can go into your mailman settings (the link to that is at the end of each
message) and alter that; maybe that will help.

[trying sb_server instead]
> It works perfectly.

So imapfilter is at fault, then.

> I also noticed that a message displays saying, "User interface url 
> is http://localhost:8880". I don't get this with sb_imapfilter.py. 
> Should I?

Yes, you should.

This is quite strange.  The problem must be with imapfilter since sb_server
works, but imapfilter does almost nothing when -c or -t aren't used.  The
process doesn't stop (right?) so it's either hanging or running after
failing to register the web interface.

Can you try the attached sb_imapfilter.py script and tell me what the output
is?  I've added various debugging prints.

=Tony.Meyer

-- 
Please always include the list (spambayes@python.org) in your replies
(reply-all), and please don't send me personal mail about SpamBayes.
http://www.massey.ac.nz/~tameyer/writing/reply_all.html explains this.
-------------- next part --------------
#!/usr/bin/env python

"""An IMAP filter.  An IMAP message box is scanned and all non-scored
messages are scored and (where necessary) filtered.

The original filter design owed much to isbg by Roger Binns
(http://www.rogerbinns.com/isbg).

Usage:
    sb_imapfilter [options]

        note: option values with spaces in them must be enclosed
              in double quotes

        options:
            -p  dbname  : pickled training database filename
            -d  dbname  : dbm training database filename
            -t          : train contents of spam folder and ham folder
            -c          : classify inbox
            -h          : display this message
            -v          : verbose mode
            -P          : security option to prompt for imap password,
                          rather than look in options["imap", "password"]
            -e y/n      : expunge/purge messages on exit (y) or not (n)
            -i debuglvl : a somewhat mysterious imaplib debugging level
                          (4 is a good level, and suitable for bug reports)
            -l minutes  : period of time between filtering operations
            -b          : Launch a web browser showing the user interface.
                          (If not specified, and neither the -c or -t
                          options are used, then this will default to the
                          value in your configuration file).
            -o section:option:value :
                          set [section, option] in the options database
                          to value


Examples:

    Classify inbox, with dbm database
        sb_imapfilter -c -d bayes.db

    Train Spam and Ham, then classify inbox, with dbm database
        sb_imapfilter -t -c -d bayes.db

    Train Spam and Ham only, with pickled database
        sb_imapfilter -t -p bayes.db

Warnings:
    o This is alpha software!  The filter is currently being developed and
      tested.  We do *not* recommend using it on a production system unless
      you are confident that you can get your mail back if you lose it.  On
      the other hand, we do recommend that you test it for us and let us
      know if anything does go wrong.
    o By default, the filter does *not* delete, modify or move any of your
      mail.  Due to quirks in how imap works, new versions of your mail are
      modified and placed in new folders, but the originals are still
      available.  These are flagged with the /Deleted flag so that you know
      that they can be removed.  Your mailer may not show these messages
      by default, but there should be an option to do so.  *However*, if
      your mailer automatically purges/expunges (i.e. permanently deletes)
      mail flagged as such, *or* if you set the imap_expunge option to
      True, then this mail will be irretrievably lost.
"""

todo = """
To Do:
    o IMAPMessage and IMAPFolder currently carry out very simple checks
      of responses received from IMAP commands, but if the response is not
      "OK", then the filter terminates.  Handling of these errors could be
      much nicer.
    o IMAP over SSL is relatively untested.
    o Develop a test script, like spambayes/test/test_pop3proxy.py that
      runs through some tests (perhaps with a *real* imap server, rather
      than a dummy one).  This would make it easier to carry out the tests
      against each server whenever a change is made.
    o IMAP supports authentication via other methods than the plain-text
      password method that we are using at the moment.  Neither of the
      servers I have access to offer any alternative method, however.  If
      someone's does, then it would be nice to offer this.
    o Usernames should be able to be literals as well as quoted strings.
      This might help if the username/password has special characters like
      accented characters.
    o Suggestions?
"""

# This module is part of the spambayes project, which is Copyright 2002-3
# The Python Software Foundation and is covered by the Python Software
# Foundation license.

__author__ = "Tony Meyer <ta-meyer@ihug.co.nz>, Tim Stone"
__credits__ = "All the Spambayes folk."

from __future__ import generators

try:
    True, False
except NameError:
    # Maintain compatibility with Python 2.2
    True, False = 1, 0

import socket
import os
import re
import time
import sys
import getopt
import types
import traceback
import email
import email.Parser
from getpass import getpass
from email.Header import Header
from email.Utils import parsedate
try:
    import cStringIO as StringIO
except ImportError:
    import StringIO

from spambayes.Options import options, get_pathname_option
from spambayes import tokenizer, storage, message, Dibbler
from spambayes.UserInterface import UserInterfaceServer
from spambayes.ImapUI import IMAPUserInterface
from spambayes.Version import get_version_string

from imaplib import IMAP4
from imaplib import Time2Internaldate
try:
    if options["imap", "use_ssl"]:
        from imaplib import IMAP4_SSL as BaseIMAP
    else:
        from imaplib import IMAP4 as BaseIMAP
except ImportError:
    from imaplib import IMAP4 as BaseIMAP

# global IMAPlib object
global imap
imap = None

# A flag can have any character in the ascii range 32-126
# except for (){ %*"\
FLAG_CHARS = ""
for i in range(32, 127):
    if not chr(i) in ['(', ')', '{', ' ', '%', '*', '"', '\\']:
        FLAG_CHARS += chr(i)
FLAG = r"\\?[" + re.escape(FLAG_CHARS) + r"]+"
# The empty flag set "()" doesn't match, so that extract returns
# data["FLAGS"] == None
FLAGS_RE = re.compile(r"(FLAGS) (\((" + FLAG + r" )*(" + FLAG + r")\))")
INTERNALDATE_RE = re.compile(r"(INTERNALDATE) (\"\d{1,2}\-[A-Za-z]{3,3}\-" +
                             r"\d{2,4} \d{2,2}\:\d{2,2}\:\d{2,2} " +
                             r"[\+\-]\d{4,4}\")")
RFC822_RE = re.compile(r"(RFC822) (\{[\d]+\})")
BODY_PEEK_RE = re.compile(r"(BODY\[\]) (\{[\d]+\})")
RFC822_HEADER_RE = re.compile(r"(RFC822.HEADER) (\{[\d]+\})")
UID_RE = re.compile(r"(UID) ([\d]+)")
FETCH_RESPONSE_RE = re.compile(r"([0-9]+) \(([" + \
                               re.escape(FLAG_CHARS) + r"\"\{\}\(\)\\ ]*)\)?")
LITERAL_RE = re.compile(r"^\{[\d]+\}$")

def _extract_fetch_data(response):
    '''Extract data from the response given to an IMAP FETCH command.'''
    # Response might be a tuple containing literal data
    # At the moment, we only handle one literal per response.  This
    # may need to be improved if our code ever asks for something
    # more complicated (like RFC822.Header and RFC822.Body)
    if type(response) == types.TupleType:
        literal = response[1]
        response = response[0]
    else:
        literal = None
    # the first item will always be the message number
    mo = FETCH_RESPONSE_RE.match(response)
    data = {}
    if mo is None:
        print """IMAP server gave strange fetch response.  Please
        report this as a bug."""
        print response
    else:
        data["message_number"] = mo.group(1)
        response = mo.group(2)
    # We support the following FETCH items:
    #  FLAGS
    #  INTERNALDATE
    #  RFC822
    #  UID
    #  RFC822.HEADER
    #  BODY.PEEK
    # All others are ignored.
    for r in [FLAGS_RE, INTERNALDATE_RE, RFC822_RE, UID_RE,
              RFC822_HEADER_RE, BODY_PEEK_RE]:
        mo = r.search(response)
        if mo is not None:
            if LITERAL_RE.match(mo.group(2)):
                data[mo.group(1)] = literal
            else:
                data[mo.group(1)] = mo.group(2)
    return data

class IMAPSession(BaseIMAP):
    '''A class extending the IMAP4 class, with a few optimizations'''

    def __init__(self, server, port, debug=0, do_expunge=False):
        try:
            BaseIMAP.__init__(self, server, port)
        except (BaseIMAP.error, socket.gaierror, socket.error):
            print "Cannot connect to server %s on port %s" % (server, port)
            sys.exit(-1)
        self.debug = debug
        # For efficiency, we remember which folder we are currently
        # in, and only send a select command to the IMAP server if
        # we want to *change* folders.  This function is used by
        # both IMAPMessage and IMAPFolder.
        self.current_folder = None
        self.do_expunge = do_expunge
        self.logged_in = False

    def login(self, username, pwd):
        try:
            BaseIMAP.login(self, username, pwd)  # superclass login
        except BaseIMAP.error, e:
            if str(e) == "permission denied":
                print "There was an error logging in to the IMAP server."
                print "The userid and/or password may be incorrect."
                sys.exit()
            else:
                raise
        self.logged_in = True

    def logout(self):
        # sign off
        if self.do_expunge:
            # we may never have logged in, in which case we do nothing
            if self.logged_in:
                # expunge messages from the spam and unsure folders
                for fol in ["spam_folder",
                            "unsure_folder",]:
                    self.select(options["imap", fol])
                    self.expunge()
                # expunge messages from the ham and spam training folders
                for fol_list in ["ham_train_folders",
                                 "spam_train_folders",]:
                    for fol in options["imap", fol_list]:
                        self.select(fol)
                        self.expunge()
        BaseIMAP.logout(self)  # superclass logout

    def SelectFolder(self, folder):
        '''A method to point ensuing imap operations at a target folder'''
        if self.current_folder != folder:
            if self.current_folder != None:
                if self.do_expunge:
                    # It is faster to do close() than a single
                    # expunge when we log out (because expunge returns
                    # a list of all the deleted messages which we don't do
                    # anything with)
                    imap.close()
            # We *always* use SELECT and not EXAMINE, because this
            # speeds things up considerably.
            if folder == "":
                # This is Python bug #845560 - if the empty string is
                # passed, we get a traceback, not just an 'invalid folder'
                # error, so print out a warning and exit.
                print "Tried to select an invalid folder"
                sys.exit(-1)
            response = self.select(folder, None)
            if response[0] != "OK":
                print "Invalid response to select %s:\n%s" % (folder,
                                                              response)
                sys.exit(-1)
            self.current_folder = folder
            return response

    def folder_list(self):
        '''Return a alphabetical list of all folders available on the
        server'''
        response = self.list()
        if response[0] != "OK":
            return []
        all_folders = response[1]
        folders = []
        for fol in all_folders:
            # Sigh.  Some servers may give us back the folder name as a
            # literal, so we need to crunch this out.
            if isinstance(fol, ()):
                r = re.compile(r"{\d+}")
                m = r.search(fol[0])
                if not m:
                    # Something is wrong here!  Skip this folder
                    continue
                fol = '%s"%s"' % (fol[0][:m.start()], fol[1])
            r = re.compile(r"\(([\w\\ ]*)\) ")
            m = r.search(fol)
            if not m:
                # Something is not good with this folder, so skip it.
                continue
            name_attributes = fol[:m.end()-1]
            # IMAP is a truly odd protocol.  The delimiter is
            # only the delimiter for this particular folder - each
            # folder *may* have a different delimiter
            self.folder_delimiter = fol[m.end()+1:m.end()+2]
            # a bit of a hack, but we really need to know if this is
            # the case
            if self.folder_delimiter == ',':
                print "WARNING: Your imap server uses a comma as the " \
                      "folder delimiter.  This may cause unpredictable " \
                      "errors."
            folders.append(fol[m.end()+4:].strip('"'))
        folders.sort()
        return folders

    def FindMessage(self, id):
        '''A (potentially very expensive) method to find a message with
        a given spambayes id (header), and return a message object (no
        substance).'''
        # If efficiency becomes a concern, what we could do is store a
        # dict of key-to-folder, and look in that folder first.  (It might
        # have moved independantly of us, so we would still have to search
        # if we didn't find it).  For the moment, we do an search through
        # all folders, alphabetically.
        for folder_name in self.folder_list():
            fol = IMAPFolder(folder_name)
            for msg in fol:
                if msg.id == id:
                    return msg
        return None

class IMAPMessage(message.SBHeaderMessage):
    def __init__(self):
        message.Message.__init__(self)
        self.folder = None
        self.previous_folder = None
        self.rfc822_command = "(BODY.PEEK[])"
        self.rfc822_key = "BODY[]"
        self.got_substance = False
        self.invalid = False

    def setFolder(self, folder):
        self.folder = folder

    def _check(self, response, command):
        if response[0] != "OK":
            print "Invalid response to %s:\n%s" % (command, response)
            sys.exit(-1)

    def extractTime(self):
        # When we create a new copy of a message, we need to specify
        # a timestamp for the message.  If the message has a valid date
        # header we use that.  Otherwise, we use the current time.
        message_date = self["Date"]
        if message_date is not None:
            parsed_date = parsedate(message_date)
            if parsed_date is not None:
                try:
                    return Time2Internaldate(time.mktime(parsed_date))
                except ValueError:
                    # Invalid dates can cause mktime() to raise a
                    # ValueError, for example:
                    #   >>> time.mktime(parsedate("Mon, 06 May 0102 10:51:16 -0100"))
                    #   Traceback (most recent call last):
                    #     File "<interactive input>", line 1, in ?
                    #   ValueError: year out of range
                    # (Why this person is getting mail from almost two
                    # thousand years ago is another question <wink>).
                    # In any case, we just pass and use the current date.
                    pass
                except OverflowError:
                    pass
        return Time2Internaldate(time.time())

    def get_substance(self):
        '''Retrieve the RFC822 message from the IMAP server and set as the
        substance of this message.'''
        if self.got_substance:
            return
        if not self.uid or not self.id:
            print "Cannot get substance of message without an id and an UID"
            return
        imap.SelectFolder(self.folder.name)
        try:
            response = imap.uid("FETCH", self.uid, self.rfc822_command)
        except IMAP4.error:
            self.rfc822_command = "RFC822"
            self.rfc822_key = "RFC822"
            response = imap.uid("FETCH", self.uid, self.rfc822_command)
        if response[0] != "OK":
            self.rfc822_command = "RFC822"
            self.rfc822_key = "RFC822"
            response = imap.uid("FETCH", self.uid, self.rfc822_command)
        self._check(response, "uid fetch")
        data = _extract_fetch_data(response[1][0])
        # Annoyingly, we can't just pass over the RFC822 message to an
        # existing message object (like self) and have it parse it. So
        # we go through the hoops of creating a new message, and then
        # copying over all its internals.
        try:
            new_msg = email.Parser.Parser().parsestr(data[self.rfc822_key])
        # We use a general 'except' because the email package doesn't
        # always return email.Errors (it can return a TypeError, for
        # example) if the email is invalid.  In any case, we want
        # to keep going, and not crash, because we might leave the
        # user's mailbox in a bad state if we do.  Better to soldier
        # on.
        except:
            # Yikes!  Barry set this to return at this point, which
            # would work ok for training (IIRC, that's all he's
            # using it for), but for filtering, what happens is that
            # the message ends up blank, but ok, so the original is
            # flagged to be deleted, and a new (almost certainly
            # unsure) message, *with only the spambayes headers* is
            # created.  The nice solution is still to do what sb_server
            # does and have a X-Spambayes-Exception header with the
            # exception data and then the original message.
            self.invalid = True

            # This is nicked from sb_server - thanks Richie!
            stream = StringIO.StringIO()
            traceback.print_exc(None, stream)
            details = stream.getvalue()

            # Build the header.  This will strip leading whitespace from
            # the lines, so we add a leading dot to maintain indentation.
            detailLines = details.strip().split('\n')
            dottedDetails = '\n.'.join(detailLines)
            headerName = 'X-Spambayes-Exception'
            header = Header(dottedDetails, header_name=headerName)

            # Insert the header, converting email.Header's '\n' line
            # breaks to IMAP4's '\r\n'.
            # (Also insert the id header, otherwise we'll keep doing
            # this message over and over again).
            headers, body = re.split(r'\n\r?\n', data["RFC822"], 1)
            header = re.sub(r'\r?\n', '\r\n', str(header))
            headers += "\n%s: %s\r\n%s: %s\r\n\r\n" % \
                       (headerName, header,
                        options["Headers", "mailid_header_name"], self.id)
            self.invalid_content = headers + body

            # Print the exception and a traceback.
            print >>sys.stderr, details
        else:
            self._headers = new_msg._headers
            self._unixfrom = new_msg._unixfrom
            self._payload = new_msg._payload
            self._charset = new_msg._charset
            self.preamble = new_msg.preamble
            self.epilogue = new_msg.epilogue
            self._default_type = new_msg._default_type
            if not self.has_key(options["Headers", "mailid_header_name"]):
                self[options["Headers", "mailid_header_name"]] = self.id
        self.got_substance = True
        if options["globals", "verbose"]:
            sys.stdout.write(chr(8) + "*")

    def MoveTo(self, dest):
        '''Note that message should move to another folder.  No move is
        carried out until Save() is called, for efficiency.'''
        if self.previous_folder is None:
            self.previous_folder = self.folder
        self.folder = dest

    def as_string(self, unixfrom=False):
        # Basically the same as the parent class's except that we handle
        # the case where the data was unparsable, so we haven't done any
        # filtering, and we are not actually a proper email.Message object.
        if self.invalid:
            return self._force_CRLF(self.invalid_content)
        else:
            return message.SBHeaderMessage.as_string(self, unixfrom)

    def Save(self):
        '''Save message to imap server.'''
        # we can't actually update the message with IMAP
        # so what we do is create a new message and delete the old one
        if self.folder is None:
            raise RuntimeError, """Can't save a message that doesn't
            have a folder."""
        if not self.id:
            raise RuntimeError, """Can't save a message that doesn't have
            an id."""
        response = imap.uid("FETCH", self.uid, "(FLAGS INTERNALDATE)")
        self._check(response, 'fetch (flags internaldate)')
        data = _extract_fetch_data(response[1][0])
        if data.has_key("INTERNALDATE"):
            msg_time = data["INTERNALDATE"]
        else:
            msg_time = self.extractTime()
        if data.has_key("FLAGS"):
            flags = data["FLAGS"]
            # The \Recent flag can be fetched, but cannot be stored
            # We must remove it from the list if it is there.
            flags = re.sub(r"\\Recent ?| ?\\Recent", "", flags)
        else:
            flags = None

        for flgs, tme in [(flags, msg_time),
                          (None, msg_time),
                          (flags, Time2Internaldate(time.time())),
                          (None, Time2Internaldate(time.time()))]:
            response = imap.append(self.folder.name, flgs, tme,
                                   self.as_string())
            if response[0] == "OK":
                break
        self._check(response, 'append')

        if self.previous_folder is None:
            imap.SelectFolder(self.folder.name)
        else:
            imap.SelectFolder(self.previous_folder.name)
            self.previous_folder = None
        response = imap.uid("STORE", self.uid, "+FLAGS.SILENT", "(\\Deleted \\Seen)")
        self._check(response, 'store')
        # Not all IMAP servers immediately offer the new message
        # (stupidly), but we need to find it.  Generally a 'no-op' will
        # allow the server time to handle it, so do that.
        # See [ 941596 ] sb_imapfilter.py not adding headers / moving messages
        imap.noop()

        # We need to update the uid, as it will have changed.
        # Although we don't use the UID to keep track of messages, we do
        # have to use it for IMAP operations.
        imap.SelectFolder(self.folder.name)
        response = imap.uid("SEARCH", "(UNDELETED HEADER %s \"%s\")" % \
                            (options["Headers", "mailid_header_name"],
                             self.id.replace('\\',r'\\').replace('"',r'\"')))
        self._check(response, 'search')
        new_id = response[1][0]

        # See [ 870799 ] imap trying to fetch invalid message UID
        # It seems that although the save gave a "NO" response to the
        # first save, the message was still saved (without the flags,
        # probably).  This isn't really good behaviour on the server's
        # part, but, as usual, we try and deal with it.  So, if we get
        # more than one undeleted message with the same SpamBayes id,
        # delete all of them apart from the last one, and use that.
        multiple_ids = new_id.split()
        for id_to_remove in multiple_ids[:-1]:
            response = imap.uid("STORE", id_to_remove, "+FLAGS.SILENT",
                                "(\\Deleted \\Seen)")
            self._check(response, 'store')
        if multiple_ids:
            new_id = multiple_ids[-1]
        else:
            # Let's hope it doesn't, but, just in case, if the search
            # turns up empty, we make the assumption that the new
            # message is the last one with a recent flag
            response = imap.uid("SEARCH", "RECENT")
            new_id = response[1][0]
            if new_id.find(' ') > -1:
                ids = new_id.split(' ')
                new_id = ids[-1]
            # Ok, now we're in trouble if we still haven't found it.
            # We make a huge assumption that the new message is the one
            # with the highest UID (they are sequential, so this will be
            # ok as long as another message hasn't also arrived)
            if new_id == "":
                response = imap.uid("SEARCH", "ALL")
                new_id = response[1][0]
                if new_id.find(' ') > -1:
                    ids = new_id.split(' ')
                    new_id = ids[-1]
        self.uid = new_id

# This performs a similar function to email.message_from_string()
def imapmessage_from_string(s, _class=IMAPMessage, strict=False):
    return email.message_from_string(s, _class, strict)


class IMAPFolder(object):
    def __init__(self, folder_name):
        self.name = folder_name
        # Unique names for cached messages - see _generate_id below.
        self.lastBaseMessageName = ''
        self.uniquifier = 2

    def __cmp__(self, obj):
        '''Two folders are equal if their names are equal'''
        if obj is None:
            return False
        return cmp(self.name, obj.name)

    def _check(self, response, command):
        if response[0] != "OK":
            print "Invalid response to %s:\n%s" % (command, response)
            sys.exit(-1)

    def __iter__(self):
        '''IMAPFolder is iterable'''
        for key in self.keys():
            try:
                yield self[key]
            except KeyError:
                pass

    def recent_uids(self):
        '''Returns uids for all the messages in the folder that
        are flagged as recent, but not flagged as deleted.'''
        imap.SelectFolder(self.name, True)
        response = imap.uid("SEARCH", "RECENT UNDELETED")
        self._check(response, "SEARCH RECENT UNDELETED")
        return response[1][0].split(' ')

    def keys(self):
        '''Returns *uids* for all the messages in the folder not
        marked as deleted.'''
        imap.SelectFolder(self.name)
        response = imap.uid("SEARCH", "UNDELETED")
        self._check(response, "SEARCH UNDELETED")
        if response[1][0]:
            return response[1][0].split(' ')
        else:
            return []

    def __getitem__(self, key):
        '''Return message (no substance) matching the given *uid*.'''
        # We don't retrieve the substances of the message here - you need
        # to call msg.get_substance() to do that.
        imap.SelectFolder(self.name)
        # Using RFC822.HEADER.LINES would be better here, but it seems
        # that not all servers accept it, even though it is in the RFC
        response = imap.uid("FETCH", key, "RFC822.HEADER")
        self._check(response, "uid fetch header")
        data = _extract_fetch_data(response[1][0])

        msg = IMAPMessage()
        msg.setFolder(self)
        msg.uid = key
        r = re.compile(re.escape(options["Headers",
                                         "mailid_header_name"]) + \
                       "\:\s*(\d+(\-\d)?)")
        mo = r.search(data["RFC822.HEADER"])
        if mo is None:
            msg.setId(self._generate_id())
            # Unfortunately, we now have to re-save this message, so that
            # our id is stored on the IMAP server.  Before anyone suggests
            # it, we can't store it as a flag, because user-defined flags
            # aren't supported by all IMAP servers.
            # This will need to be done once per message.
            msg.get_substance()
            msg.Save()
        else:
            msg.setId(mo.group(1))

        if options["globals", "verbose"]:
            sys.stdout.write(".")
        return msg

    # Lifted straight from pop3proxy.py (under the name getNewMessageName)
    def _generate_id(self):
        # The message id is the time it arrived, with a uniquifier
        # appended if two arrive within one clock tick of each other.
        messageName = "%10.10d" % long(time.time())
        if messageName == self.lastBaseMessageName:
            messageName = "%s-%d" % (messageName, self.uniquifier)
            self.uniquifier += 1
        else:
            self.lastBaseMessageName = messageName
            self.uniquifier = 2
        return messageName

    def Train(self, classifier, isSpam):
        '''Train folder as spam/ham'''
        num_trained = 0
        for msg in self:
            if msg.GetTrained() == (not isSpam):
                msg.get_substance()
                msg.delSBHeaders()
                classifier.unlearn(msg.asTokens(), not isSpam)
                # Once the message has been untrained, it's training memory
                # should reflect that on the off chance that for some reason
                # the training breaks, which happens on occasion (the
                # tokenizer is not yet perfect)
                msg.RememberTrained(None)

            if msg.GetTrained() is None:
                msg.get_substance()
                saved_headers = msg.currentSBHeaders()
                msg.delSBHeaders()
                classifier.learn(msg.asTokens(), isSpam)
                num_trained += 1
                msg.RememberTrained(isSpam)
                if isSpam:
                    move_opt_name = "move_trained_spam_to_folder"
                else:
                    move_opt_name = "move_trained_ham_to_folder"
                if options["imap", move_opt_name] != "":
                    # We need to restore the SpamBayes headers.
                    for header, value in saved_headers.items():
                        msg[header] = value
                    msg.MoveTo(IMAPFolder(options["imap",
                                                  move_opt_name]))
                    msg.Save()
        return num_trained

    def Filter(self, classifier, spamfolder, unsurefolder):
        count = {}
        count["ham"] = 0
        count["spam"] = 0
        count["unsure"] = 0
        for msg in self:
            if msg.GetClassification() is None:
                msg.get_substance()
                (prob, clues) = classifier.spamprob(msg.asTokens(),
                                                    evidence=True)
                # add headers and remember classification
                msg.delSBHeaders()
                msg.addSBHeaders(prob, clues)

                cls = msg.GetClassification()
                if cls == options["Headers", "header_ham_string"]:
                    # we leave ham alone
                    count["ham"] += 1
                elif cls == options["Headers", "header_spam_string"]:
                    msg.MoveTo(spamfolder)
                    count["spam"] += 1
                else:
                    msg.MoveTo(unsurefolder)
                    count["unsure"] += 1
                msg.Save()
        return count


class IMAPFilter(object):
    def __init__(self, classifier):
        self.spam_folder = IMAPFolder(options["imap", "spam_folder"])
        self.unsure_folder = IMAPFolder(options["imap", "unsure_folder"])
        self.classifier = classifier

    def Train(self):
        if options["globals", "verbose"]:
            t = time.time()

        total_ham_trained = 0
        total_spam_trained = 0

        if options["imap", "ham_train_folders"] != "":
            ham_training_folders = options["imap", "ham_train_folders"]
            for fol in ham_training_folders:
                # Select the folder to make sure it exists
                imap.SelectFolder(fol)
                if options['globals', 'verbose']:
                    print "   Training ham folder %s" % (fol)
                folder = IMAPFolder(fol)
                num_ham_trained = folder.Train(self.classifier, False)
                total_ham_trained += num_ham_trained
                if options['globals', 'verbose']:
                    print "       %s trained." % (num_ham_trained)

        if options["imap", "spam_train_folders"] != "":
            spam_training_folders = options["imap", "spam_train_folders"]
            for fol in spam_training_folders:
                # Select the folder to make sure it exists
                imap.SelectFolder(fol)
                if options['globals', 'verbose']:
                    print "   Training spam folder %s" % (fol)
                folder = IMAPFolder(fol)
                num_spam_trained = folder.Train(self.classifier, True)
                total_spam_trained += num_spam_trained
                if options['globals', 'verbose']:
                    print "       %s trained." % (num_spam_trained)

        if total_ham_trained or total_spam_trained:
            self.classifier.store()

        if options["globals", "verbose"]:
            print "Training took %.4f seconds, %s messages were trained" \
                  % (time.time() - t, total_ham_trained + total_spam_trained)

    def Filter(self):
        if options["globals", "verbose"]:
            t = time.time()
        count = {}
        count["ham"] = 0
        count["spam"] = 0
        count["unsure"] = 0

        # Select the spam folder and unsure folder to make sure they exist
        imap.SelectFolder(self.spam_folder.name)
        imap.SelectFolder(self.unsure_folder.name)

        for filter_folder in options["imap", "filter_folders"]:
            # Select the folder to make sure it exists
            imap.SelectFolder(filter_folder)
            folder = IMAPFolder(filter_folder)
            subcount = folder.Filter(self.classifier, self.spam_folder,
                          self.unsure_folder)
            for key in count.keys():
                count[key] += subcount.get(key, 0)

        if options["globals", "verbose"]:
            if count is not None:
                print "\nClassified %s ham, %s spam, and %s unsure." % \
                      (count["ham"], count["spam"], count["unsure"])
            print "Classifying took %.4f seconds." % (time.time() - t,)


def run():
    global imap
    try:
        opts, args = getopt.getopt(sys.argv[1:], 'hbPtcvl:e:i:d:p:o:')
    except getopt.error, msg:
        print >>sys.stderr, str(msg) + '\n\n' + __doc__
        sys.exit()

    doTrain = False
    doClassify = False
    doExpunge = options["imap", "expunge"]
    imapDebug = 0
    sleepTime = 0
    promptForPass = False
    launchUI = False
    server = ""
    username = ""

    for opt, arg in opts:
        if opt == '-h':
            print >>sys.stderr, __doc__
            sys.exit()
        elif opt == "-b":
            launchUI = True
        elif opt == '-t':
            doTrain = True
        elif opt == '-P':
            promptForPass = True
        elif opt == '-c':
            doClassify = True
        elif opt == '-v':
            options["globals", "verbose"] = True
        elif opt == '-e':
            if arg == 'y':
                doExpunge = True
            else:
                doExpunge = False
        elif opt == '-i':
            imapDebug = int(arg)
        elif opt == '-l':
            sleepTime = int(arg) * 60
        elif opt == '-o':
            options.set_from_cmdline(arg, sys.stderr)
    bdbname, useDBM = storage.database_type(opts)

    # Let the user know what they are using...
    print get_version_string("IMAP Filter")
    print "and engine %s.\n" % (get_version_string(),)

    if (launchUI and (doClassify or doTrain)):
        print """-b option is exclusive with -c and -t options.
The user interface will be launched, but no classification
or training will be performed.
"""

    if options["globals", "verbose"]:
        print "Loading database %s..." % (bdbname),

    classifier = storage.open_storage(bdbname, useDBM)

    if options["globals", "verbose"]:
        print "Done."

    if options["imap", "server"]:
        # The options class is ahead of us here:
        #   it knows that imap:server will eventually be able to have
        #   multiple values, but for the moment, we just use the first one
        server = options["imap", "server"]
        if len(server) > 0:
            server = server[0]
        username = options["imap", "username"]
        if len(username) > 0:
            username = username[0]
        if not promptForPass:
            pwd = options["imap", "password"]
            if len(pwd) > 0:
                pwd = pwd[0]
    else:
        pwd = None
        if not launchUI:
            print "You need to specify both a server and a username."
            sys.exit()

    if promptForPass:
        pwd = getpass()

    if server.find(':') > -1:
        server, port = server.split(':', 1)
        port = int(port)
    else:
        if options["imap", "use_ssl"]:
            port = 993
        else:
            port = 143

    imap_filter = IMAPFilter(classifier)

    # Web interface
    if not (doClassify or doTrain):
        print "XXX Serving web interface"
        if server != "":
            imap = IMAPSession(server, port, imapDebug, doExpunge)
        print "XXX With imap =", imap
        httpServer = UserInterfaceServer(options["html_ui", "port"])
        print "XXX And httpServer =", httpServer
        httpServer.register(IMAPUserInterface(classifier, imap, pwd,
                                              IMAPSession))
        print "XXX Registered, will now run"
        Dibbler.run(launchBrowser=launchUI or options["html_ui",
                                                      "launch_browser"])
    else:
        while True:
            imap = IMAPSession(server, port, imapDebug, doExpunge)
            imap.login(username, pwd)

            if doTrain:
                if options["globals", "verbose"]:
                    print "Training"
                imap_filter.Train()
            if doClassify:
                if options["globals", "verbose"]:
                    print "Classifying"
                imap_filter.Filter()

            imap.logout()

            if sleepTime:
                time.sleep(sleepTime)
            else:
                break

if __name__ == '__main__':
    run()
From usr7-28hl at xemaps.com  Mon Nov 29 07:45:31 2004
From: usr7-28hl at xemaps.com (Keith Russell)
Date: Mon Nov 29 07:45:38 2004
Subject: [Spambayes] Unable to Access SpamBayes Browser Interface
	Usingsb_imapfilter.py
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E8FBCA91@its-xchg4.massey.ac.nz>
References: <ECBA357DDED63B4995F5C1F5CBE5B1E8FBCA91@its-xchg4.massey.ac.nz>
Message-ID: <41AAC58B.9030409@mailsnare.net>

Thanks very much for the reply, Tony.

Tony Meyer wrote:

> If you mean that you want to stop trying to figure out what's wrong and move
> onto something else, then I understand.  The 'related' page on our website
> links to various other filters that all should work very well -
> unfortunately I'm not familiar enough with them to know which work with IMAP
> - I gather POPfile has experimental IMAP support, so maybe that would be a
> good choice.

I guess I intended a double meaning. I've considered POPfile, but 
IMAP is still very experimental, and much more difficult to 
install than SpamBayes' IMAP support. And while I like POPFile, 
there are a couple of things about SpamBayes I prefer.

So as long as I can continue to get help, I'm willing to stick 
with it, and possibly make a contribution to the development 
effort, as well.

> <http://spambayes.org/related.html>
> 
> If this is the case, then you can ignore the rest of this.

No way! :-)

> Maybe you have mailman set to avoid sending you copies of messages that
> include you in the recipient list (this is the default, I believe), but for
> some reason (bad reply address?) you aren't getting the remaining one?  You
> can go into your mailman settings (the link to that is at the end of each
> message) and alter that; maybe that will help.

Thanks. I'll check my settings. But I did get this latest message 
of yours.

>>I also noticed that a message displays saying, "User interface url 
>>is http://localhost:8880". I don't get this with sb_imapfilter.py. 
>>Should I?
> 
> 
> Yes, you should.
> 
> This is quite strange.  The problem must be with imapfilter since sb_server
> works, but imapfilter does almost nothing when -c or -t aren't used.  The
> process doesn't stop (right?) so it's either hanging or running after
> failing to register the web interface.

No, it doesn't stop. The last thing I see (in verbose mode) is 
"Done."

> Can you try the attached sb_imapfilter.py script and tell me what the output
> is?  I've added various debugging prints.

Thanks for this. Here's what I get:

 >sb_imapfilter.py -b
SpamBayes IMAP Filter Version 0.5 (November 2004)
and engine SpamBayes Engine Version 0.3 (January 2004).

XXX Serving web interface


 >sb_imapfilter.py -b -v
SpamBayes IMAP Filter Version 0.5 (November 2004)
and engine SpamBayes Engine Version 0.3 (January 2004).

Loading database C:\...\SpamBayes\Proxy\hammie.db...
Loading state from c:\...\SpamBayes\Proxy\hammie.db database
c:\...\SpamBayes\Proxy\hammie.db is an existing database, with 0
spam and 0 ham
Done.
XXX Serving web interface


Thanks again for your help.

-- 
Keith
From anthony at interlink.com.au  Mon Nov 29 08:20:02 2004
From: anthony at interlink.com.au (Anthony Baxter)
Date: Mon Nov 29 08:21:15 2004
Subject: [Spambayes] spambayes and kmail
Message-ID: <200411291820.03150.anthony@interlink.com.au>

So here's the situation with spambayes and kmail. The next pointfix
release of kde (which includes kmail) will include limited support for
automatically setting up spambayes - basically just filtering for 'spam'.
The next major release (3.4) will include the unsure folder and the like.

This is all pretty damn cool, in my opinion... it gives us integration 
with a full-featured GUI mailer on Unix, in a way that's similar to 
the Outlook interface (although, obviously without the nice admin
GUI - perhaps it would be worth writing code to generate a prefs GUI
using different toolkits? There's code in shtoom that does this that
could be lifted...

Anthony

From virusalert at tiscali.cz  Mon Nov 29 08:41:37 2004
From: virusalert at tiscali.cz (Content-filter at avas7.tiscali.cz)
Date: Mon Nov 29 08:41:40 2004
Subject: [Spambayes] VIRUS (Worm.Ganda-A) IN MAIL TO YOU (from
	<?@ppp5113.praha.tiscali.cz>)
Message-ID: <VR03119-01-96@avas7.tiscali.cz>

VIRUS ALERT

Our content checker found
    virus: Worm.Ganda-A
in an email to you from unknown sender:

  ?@ppp5113.praha.tiscali.cz

According to the 'Received:' trace, the message originated at:
  [213.235.68.159]
  -en--ek (ppp5113.praha.tiscali.cz [213.235.68.159])

Our internal reference code for this message is 03119-01-96.
Not quarantined.

Please contact your system administrator for details.

---@@@---

From: antivir@cz.tiscali.com
Subject: VIRUS: Upozorn?n?
Body: 
V p??choz?m emailu byl nalezen virus: Worm.Ganda-A v souboru .
P?vodn? zpr?va: G.W Bush animation. 
Komu: spambayes@python.org 
Od: <-voigtova@podborany.net> (nemus? b?t spr?vn? adresa odesilatele).
------------------------
Pokud tyto zpr?vy nechcete doru?ovat nastavte si nov? pravidlo pro zpracov?n?
p??choz? po?ty
http://servicedesk.tiscali.cz/servicedesk/mail/avas/vir_smazat.php
Antivirus m??ete deaktivovat na t?to adrese :
http://servicedesk.tiscali.cz/servicedesk/mail/avas/

??????????
tak toto je pokus.

=======================
f: "Content-filter at avas7.tiscali.cz" <virusalert@tiscali.cz>
T: spambayes@python.org
C: 
B: 
p: 
h: avas7.tiscali.cz
n: 03119-01-96
b: 5a84818422b64886d2d0ba8f1bf969a7
d: Mon, 29 Nov 2004 08:41:37 +0100 (CET)
u: 1101714097
U: 20041129T074137Z
s: <-voigtova@podborany.net>
t: -en--ek (ppp5113.praha.tiscali.cz [213.235.68.159])
a: 
g: 
e: 213.235.68.159
l: 1
o: ?@ppp5113.praha.tiscali.cz
S: 
R: spambayes@python.org
D: 
O: <spambayes@python.org>
N: <spambayes@python.org>:
   550 5.7.1 Message content rejected, id=03119-01-96 - VIRUS: Worm.Ganda-A
j: G.W Bush animation.
m: <20041129074122.60E2A190C6E@avas7.tiscali.cz>
r: 
Q: 
H: Received: from -en--ek (ppp5113.praha.tiscali.cz [213.235.68.159]), 	by avas7.tiscali.cz (Postfix) with SMTP id 60E2A190C6E, 	for <spambayes@python.org>; Mon, 29 Nov 2004 08:41:22 +0100 (CET), To: <spambayes@python.org>, From: -voigtova@podborany.net, Subject: G.W Bush animation., MIME-Version: 1.0, Content-type: multipart/mixed; boundary="part1", Message-Id: <20041129074122.60E2A190C6E@avas7.tiscali.cz>, Date: Mon, 29 Nov 2004 08:41:22 +0100 (CET)
z: 62692
i: virus-20041129-084137-03119-01-96
q: 
v: p003: Worm.Ganda-A FOUND
V: Worm.Ganda-A
F: 
W: ClamAV-clamd
X: 
A: 
c: -
1: 0
2: 0
k: 0
T: spambayes@python.org
.: 
R: spambayes@python.org
D: 
O: <spambayes@python.org>
N: <spambayes@python.org>:
   550 5.7.1 Message content rejected, id=03119-01-96 - VIRUS: Worm.Ganda-A
c: -
0: 
1: 0
2: 0
k: 0



From virusalert at tiscali.cz  Mon Nov 29 09:03:56 2004
From: virusalert at tiscali.cz (Content-filter at avas7.tiscali.cz)
Date: Mon Nov 29 09:03:59 2004
Subject: [Spambayes] VIRUS (Worm.Ganda-A) IN MAIL TO YOU (from
	<?@ppp5046.praha.tiscali.cz>)
Message-ID: <VR04364-01-22@avas7.tiscali.cz>

VIRUS ALERT

Our content checker found
    virus: Worm.Ganda-A
in an email to you from unknown sender:

  ?@ppp5046.praha.tiscali.cz

According to the 'Received:' trace, the message originated at:
  [213.235.68.92]
  -en--ek (ppp5046.praha.tiscali.cz [213.235.68.92])

Our internal reference code for this message is 04364-01-22.
Not quarantined.

Please contact your system administrator for details.

---@@@---

From: antivir@cz.tiscali.com
Subject: VIRUS: Upozorn?n?
Body: 
V p??choz?m emailu byl nalezen virus: Worm.Ganda-A v souboru .
P?vodn? zpr?va:  
Komu: spambayes@python.org 
Od: <-voigtova@podborany.net> (nemus? b?t spr?vn? adresa odesilatele).
------------------------
Pokud tyto zpr?vy nechcete doru?ovat nastavte si nov? pravidlo pro zpracov?n?
p??choz? po?ty
http://servicedesk.tiscali.cz/servicedesk/mail/avas/vir_smazat.php
Antivirus m??ete deaktivovat na t?to adrese :
http://servicedesk.tiscali.cz/servicedesk/mail/avas/

??????????
tak toto je pokus.

=======================
f: "Content-filter at avas7.tiscali.cz" <virusalert@tiscali.cz>
T: spambayes@python.org
C: 
B: 
p: 
h: avas7.tiscali.cz
n: 04364-01-22
b: 87de5b292de930668cea7b52fa6a2fc4
d: Mon, 29 Nov 2004 09:03:56 +0100 (CET)
u: 1101715436
U: 20041129T080356Z
s: <-voigtova@podborany.net>
t: -en--ek (ppp5046.praha.tiscali.cz [213.235.68.92])
a: 
g: 
e: 213.235.68.92
l: 1
o: ?@ppp5046.praha.tiscali.cz
S: 
R: spambayes@python.org
D: 
O: <spambayes@python.org>
N: <spambayes@python.org>:
   550 5.7.1 Message content rejected, id=04364-01-22 - VIRUS: Worm.Ganda-A
j: 
m: <20041129080340.2B89A190C81@avas7.tiscali.cz>
r: 
Q: 
H: Received: from -en--ek (ppp5046.praha.tiscali.cz [213.235.68.92]), 	by avas7.tiscali.cz (Postfix) with SMTP id 2B89A190C81, 	for <spambayes@python.org>; Mon, 29 Nov 2004 09:03:40 +0100 (CET), MIME-Version: 1.0, Content-type: multipart/mixed; boundary="part1", Message-Id: <20041129080340.2B89A190C81@avas7.tiscali.cz>, Date: Mon, 29 Nov 2004 09:03:40 +0100 (CET), From: -voigtova@podborany.net, To: undisclosed-recipients:;
z: 62537
i: virus-20041129-090356-04364-01-22
q: 
v: p003: Worm.Ganda-A FOUND
V: Worm.Ganda-A
F: 
W: ClamAV-clamd
X: 
A: 
c: -
1: 0
2: 0
k: 0
T: spambayes@python.org
.: 
R: spambayes@python.org
D: 
O: <spambayes@python.org>
N: <spambayes@python.org>:
   550 5.7.1 Message content rejected, id=04364-01-22 - VIRUS: Worm.Ganda-A
c: -
0: 
1: 0
2: 0
k: 0



From wvanandel at exploration.org  Mon Nov 29 16:33:47 2004
From: wvanandel at exploration.org (Wayne Van Andel)
Date: Mon Nov 29 16:45:08 2004
Subject: [Spambayes] question regarding Outlook2000?
Message-ID: <915A62E64D05D311853D0008C79F8774DA2DFB@EP_ADMINSVR_1>

Spambayes,  I just read about Spambayes in my Lockergnome eletter, and it
sounds great.  My configurations consists of an Outlook2000 client running
on a Win2kPro platform that communicates via our LAN to an Exchange5.5
Server.  I do not have "personal folders" (*.pst file) setup on my client,
so that my mail remains on the server, even though my Inbox opens up on the
client.  Can I still use the Spambayes plugin on my Outlook2000 in this
environment?  
   Thanks for listening, Wayne Van Andel wvanandel@exploration.org
From tameyer at ihug.co.nz  Mon Nov 29 22:43:12 2004
From: tameyer at ihug.co.nz (Tony Meyer)
Date: Mon Nov 29 22:43:54 2004
Subject: [Spambayes] question regarding Outlook2000?
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E8017ADC8D@its-xchg4.massey.ac.nz>
Message-ID: <ECBA357DDED63B4995F5C1F5CBE5B1E8FBCA93@its-xchg4.massey.ac.nz>

> an Outlook2000 client running on a Win2kPro platform that
> communicates via our LAN to an Exchange5.5
> Server.  I do not have "personal folders" (*.pst file) setup 
> on my client, so that my mail remains on the server, even though
> my Inbox opens up on the client.  Can I still use the Spambayes
> plugin on my Outlook2000 in this environment?  

Yes.  Please see FAQ 3.3:

<http://spambayes.org/faq.html#will-spambayes-work-with-outlook-connecting-t
o-an-exchange-server>

=Tony.Meyer

-- 
Please always include the list (spambayes@python.org) in your replies
(reply-all), and please don't send me personal mail about SpamBayes.
http://www.massey.ac.nz/~tameyer/writing/reply_all.html explains this.

From ta-meyer at ihug.co.nz  Mon Nov 29 23:01:49 2004
From: ta-meyer at ihug.co.nz (Tony Meyer)
Date: Mon Nov 29 23:02:26 2004
Subject: [Spambayes] Download verification checking request
Message-ID: <ECBA357DDED63B4995F5C1F5CBE5B1E8FBCA94@its-xchg4.massey.ac.nz>

As per

[ 1061119 ] Provide verification and/or security measures for downloads
<http://sf.net/tracker/?func=detail&atid=498106&aid=1061119&group_id=61702>

the 1.0.1 release (and future releases) includes some measures to try and
let people confirm that their download is valid.  Specifically, the file
sizes and MD5 checksums are listed on the download page:

<http://spambayes.org/download.html>

and there are PGP signatures for each of the three files, signed by the
release manager (me, for this one).

I'm pretty new to this stuff - I'm pretty sure that the MD5s and sizes are
correct, but not 100% about the signatures.  If someone familiar with this
stuff could use the signatures to verify the files, and let me know (either
here or via the tracker above) so that I know I've done the process
correctly, that would be fantastic.

Thanks!

=Tony.Meyer

-- 
Please always include the list (spambayes@python.org) in your replies
(reply-all), and please don't send me personal mail about SpamBayes.
http://www.massey.ac.nz/~tameyer/writing/reply_all.html explains this.

From tameyer at ihug.co.nz  Tue Nov 30 00:04:37 2004
From: tameyer at ihug.co.nz (Tony Meyer)
Date: Tue Nov 30 00:06:49 2004
Subject: [Spambayes] Unable to Access SpamBayes Browser
	InterfaceUsingsb_imapfilter.py
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E8017ADBC1@its-xchg4.massey.ac.nz>
Message-ID: <ECBA357DDED63B4995F5C1F5CBE5B1E8FBCA9B@its-xchg4.massey.ac.nz>

> I guess I intended a double meaning. I've considered POPfile, but 
> IMAP is still very experimental, and much more difficult to 
> install than SpamBayes' IMAP support. And while I like POPFile, 
> there are a couple of things about SpamBayes I prefer.

Sounds good to me :)

> So as long as I can continue to get help, I'm willing to stick 
> with it, and possibly make a contribution to the development 
> effort, as well.

Helping identify problems/bugs is enough of a contribution really.

> > Can you try the attached sb_imapfilter.py script and tell 
> > me what the output is?  I've added various debugging prints.
> 
> Thanks for this. Here's what I get:
> 
>  >sb_imapfilter.py -b
> SpamBayes IMAP Filter Version 0.5 (November 2004)
> and engine SpamBayes Engine Version 0.3 (January 2004).
> 
> XXX Serving web interface

That's not very far at all!  It appears to be dying when creating the IMAP
session, then, which is strange.  Attached is a new version with more
debugging prints in new places (this ought to be enough to narrow it down to
a specific place).  If you could do the same thing with this one (just
sb_imapfilter.py -b is enough) that would be great.

=Tony.Meyer

-- 
Please always include the list (spambayes@python.org) in your replies
(reply-all), and please don't send me personal mail about SpamBayes.
http://www.massey.ac.nz/~tameyer/writing/reply_all.html explains this.
-------------- next part --------------
#!/usr/bin/env python

"""An IMAP filter.  An IMAP message box is scanned and all non-scored
messages are scored and (where necessary) filtered.

The original filter design owed much to isbg by Roger Binns
(http://www.rogerbinns.com/isbg).

Usage:
    sb_imapfilter [options]

        note: option values with spaces in them must be enclosed
              in double quotes

        options:
            -p  dbname  : pickled training database filename
            -d  dbname  : dbm training database filename
            -t          : train contents of spam folder and ham folder
            -c          : classify inbox
            -h          : display this message
            -v          : verbose mode
            -P          : security option to prompt for imap password,
                          rather than look in options["imap", "password"]
            -e y/n      : expunge/purge messages on exit (y) or not (n)
            -i debuglvl : a somewhat mysterious imaplib debugging level
                          (4 is a good level, and suitable for bug reports)
            -l minutes  : period of time between filtering operations
            -b          : Launch a web browser showing the user interface.
                          (If not specified, and neither the -c or -t
                          options are used, then this will default to the
                          value in your configuration file).
            -o section:option:value :
                          set [section, option] in the options database
                          to value


Examples:

    Classify inbox, with dbm database
        sb_imapfilter -c -d bayes.db

    Train Spam and Ham, then classify inbox, with dbm database
        sb_imapfilter -t -c -d bayes.db

    Train Spam and Ham only, with pickled database
        sb_imapfilter -t -p bayes.db

Warnings:
    o This is alpha software!  The filter is currently being developed and
      tested.  We do *not* recommend using it on a production system unless
      you are confident that you can get your mail back if you lose it.  On
      the other hand, we do recommend that you test it for us and let us
      know if anything does go wrong.
    o By default, the filter does *not* delete, modify or move any of your
      mail.  Due to quirks in how imap works, new versions of your mail are
      modified and placed in new folders, but the originals are still
      available.  These are flagged with the /Deleted flag so that you know
      that they can be removed.  Your mailer may not show these messages
      by default, but there should be an option to do so.  *However*, if
      your mailer automatically purges/expunges (i.e. permanently deletes)
      mail flagged as such, *or* if you set the imap_expunge option to
      True, then this mail will be irretrievably lost.
"""

todo = """
To Do:
    o IMAPMessage and IMAPFolder currently carry out very simple checks
      of responses received from IMAP commands, but if the response is not
      "OK", then the filter terminates.  Handling of these errors could be
      much nicer.
    o IMAP over SSL is relatively untested.
    o Develop a test script, like spambayes/test/test_pop3proxy.py that
      runs through some tests (perhaps with a *real* imap server, rather
      than a dummy one).  This would make it easier to carry out the tests
      against each server whenever a change is made.
    o IMAP supports authentication via other methods than the plain-text
      password method that we are using at the moment.  Neither of the
      servers I have access to offer any alternative method, however.  If
      someone's does, then it would be nice to offer this.
    o Usernames should be able to be literals as well as quoted strings.
      This might help if the username/password has special characters like
      accented characters.
    o Suggestions?
"""

# This module is part of the spambayes project, which is Copyright 2002-3
# The Python Software Foundation and is covered by the Python Software
# Foundation license.

__author__ = "Tony Meyer <ta-meyer@ihug.co.nz>, Tim Stone"
__credits__ = "All the Spambayes folk."

from __future__ import generators

try:
    True, False
except NameError:
    # Maintain compatibility with Python 2.2
    True, False = 1, 0

import socket
import os
import re
import time
import sys
import getopt
import types
import traceback
import email
import email.Parser
from getpass import getpass
from email.Header import Header
from email.Utils import parsedate
try:
    import cStringIO as StringIO
except ImportError:
    import StringIO

from spambayes.Options import options, get_pathname_option
from spambayes import tokenizer, storage, message, Dibbler
from spambayes.UserInterface import UserInterfaceServer
from spambayes.ImapUI import IMAPUserInterface
from spambayes.Version import get_version_string

from imaplib import IMAP4
from imaplib import Time2Internaldate
try:
    if options["imap", "use_ssl"]:
        from imaplib import IMAP4_SSL as BaseIMAP
    else:
        from imaplib import IMAP4 as BaseIMAP
except ImportError:
    from imaplib import IMAP4 as BaseIMAP

# global IMAPlib object
global imap
imap = None

# A flag can have any character in the ascii range 32-126
# except for (){ %*"\
FLAG_CHARS = ""
for i in range(32, 127):
    if not chr(i) in ['(', ')', '{', ' ', '%', '*', '"', '\\']:
        FLAG_CHARS += chr(i)
FLAG = r"\\?[" + re.escape(FLAG_CHARS) + r"]+"
# The empty flag set "()" doesn't match, so that extract returns
# data["FLAGS"] == None
FLAGS_RE = re.compile(r"(FLAGS) (\((" + FLAG + r" )*(" + FLAG + r")\))")
INTERNALDATE_RE = re.compile(r"(INTERNALDATE) (\"\d{1,2}\-[A-Za-z]{3,3}\-" +
                             r"\d{2,4} \d{2,2}\:\d{2,2}\:\d{2,2} " +
                             r"[\+\-]\d{4,4}\")")
RFC822_RE = re.compile(r"(RFC822) (\{[\d]+\})")
BODY_PEEK_RE = re.compile(r"(BODY\[\]) (\{[\d]+\})")
RFC822_HEADER_RE = re.compile(r"(RFC822.HEADER) (\{[\d]+\})")
UID_RE = re.compile(r"(UID) ([\d]+)")
FETCH_RESPONSE_RE = re.compile(r"([0-9]+) \(([" + \
                               re.escape(FLAG_CHARS) + r"\"\{\}\(\)\\ ]*)\)?")
LITERAL_RE = re.compile(r"^\{[\d]+\}$")

def _extract_fetch_data(response):
    '''Extract data from the response given to an IMAP FETCH command.'''
    # Response might be a tuple containing literal data
    # At the moment, we only handle one literal per response.  This
    # may need to be improved if our code ever asks for something
    # more complicated (like RFC822.Header and RFC822.Body)
    if type(response) == types.TupleType:
        literal = response[1]
        response = response[0]
    else:
        literal = None
    # the first item will always be the message number
    mo = FETCH_RESPONSE_RE.match(response)
    data = {}
    if mo is None:
        print """IMAP server gave strange fetch response.  Please
        report this as a bug."""
        print response
    else:
        data["message_number"] = mo.group(1)
        response = mo.group(2)
    # We support the following FETCH items:
    #  FLAGS
    #  INTERNALDATE
    #  RFC822
    #  UID
    #  RFC822.HEADER
    #  BODY.PEEK
    # All others are ignored.
    for r in [FLAGS_RE, INTERNALDATE_RE, RFC822_RE, UID_RE,
              RFC822_HEADER_RE, BODY_PEEK_RE]:
        mo = r.search(response)
        if mo is not None:
            if LITERAL_RE.match(mo.group(2)):
                data[mo.group(1)] = literal
            else:
                data[mo.group(1)] = mo.group(2)
    return data

class IMAPSession(BaseIMAP):
    '''A class extending the IMAP4 class, with a few optimizations'''

    def __init__(self, server, port, debug=0, do_expunge=False):
        print "XXX Initialising base IMAP", port
        try:
            BaseIMAP.__init__(self, server, port)
        except (BaseIMAP.error, socket.gaierror, socket.error):
            print "Cannot connect to server %s on port %s" % (server, port)
            sys.exit(-1)
        print "XXX 1"
        self.debug = debug
        # For efficiency, we remember which folder we are currently
        # in, and only send a select command to the IMAP server if
        # we want to *change* folders.  This function is used by
        # both IMAPMessage and IMAPFolder.
        print "XXX 2"
        self.current_folder = None
        print "XXX 3"
        self.do_expunge = do_expunge
        print "XXX 4"
        self.logged_in = False
        print "XXX 5"

    def login(self, username, pwd):
        try:
            BaseIMAP.login(self, username, pwd)  # superclass login
        except BaseIMAP.error, e:
            if str(e) == "permission denied":
                print "There was an error logging in to the IMAP server."
                print "The userid and/or password may be incorrect."
                sys.exit()
            else:
                raise
        self.logged_in = True

    def logout(self):
        # sign off
        if self.do_expunge:
            # we may never have logged in, in which case we do nothing
            if self.logged_in:
                # expunge messages from the spam and unsure folders
                for fol in ["spam_folder",
                            "unsure_folder",]:
                    self.select(options["imap", fol])
                    self.expunge()
                # expunge messages from the ham and spam training folders
                for fol_list in ["ham_train_folders",
                                 "spam_train_folders",]:
                    for fol in options["imap", fol_list]:
                        self.select(fol)
                        self.expunge()
        BaseIMAP.logout(self)  # superclass logout

    def SelectFolder(self, folder):
        '''A method to point ensuing imap operations at a target folder'''
        if self.current_folder != folder:
            if self.current_folder != None:
                if self.do_expunge:
                    # It is faster to do close() than a single
                    # expunge when we log out (because expunge returns
                    # a list of all the deleted messages which we don't do
                    # anything with)
                    imap.close()
            # We *always* use SELECT and not EXAMINE, because this
            # speeds things up considerably.
            if folder == "":
                # This is Python bug #845560 - if the empty string is
                # passed, we get a traceback, not just an 'invalid folder'
                # error, so print out a warning and exit.
                print "Tried to select an invalid folder"
                sys.exit(-1)
            response = self.select(folder, None)
            if response[0] != "OK":
                print "Invalid response to select %s:\n%s" % (folder,
                                                              response)
                sys.exit(-1)
            self.current_folder = folder
            return response

    def folder_list(self):
        '''Return a alphabetical list of all folders available on the
        server'''
        response = self.list()
        if response[0] != "OK":
            return []
        all_folders = response[1]
        folders = []
        for fol in all_folders:
            # Sigh.  Some servers may give us back the folder name as a
            # literal, so we need to crunch this out.
            if isinstance(fol, ()):
                r = re.compile(r"{\d+}")
                m = r.search(fol[0])
                if not m:
                    # Something is wrong here!  Skip this folder
                    continue
                fol = '%s"%s"' % (fol[0][:m.start()], fol[1])
            r = re.compile(r"\(([\w\\ ]*)\) ")
            m = r.search(fol)
            if not m:
                # Something is not good with this folder, so skip it.
                continue
            name_attributes = fol[:m.end()-1]
            # IMAP is a truly odd protocol.  The delimiter is
            # only the delimiter for this particular folder - each
            # folder *may* have a different delimiter
            self.folder_delimiter = fol[m.end()+1:m.end()+2]
            # a bit of a hack, but we really need to know if this is
            # the case
            if self.folder_delimiter == ',':
                print "WARNING: Your imap server uses a comma as the " \
                      "folder delimiter.  This may cause unpredictable " \
                      "errors."
            folders.append(fol[m.end()+4:].strip('"'))
        folders.sort()
        return folders

    def FindMessage(self, id):
        '''A (potentially very expensive) method to find a message with
        a given spambayes id (header), and return a message object (no
        substance).'''
        # If efficiency becomes a concern, what we could do is store a
        # dict of key-to-folder, and look in that folder first.  (It might
        # have moved independantly of us, so we would still have to search
        # if we didn't find it).  For the moment, we do an search through
        # all folders, alphabetically.
        for folder_name in self.folder_list():
            fol = IMAPFolder(folder_name)
            for msg in fol:
                if msg.id == id:
                    return msg
        return None

class IMAPMessage(message.SBHeaderMessage):
    def __init__(self):
        message.Message.__init__(self)
        self.folder = None
        self.previous_folder = None
        self.rfc822_command = "(BODY.PEEK[])"
        self.rfc822_key = "BODY[]"
        self.got_substance = False
        self.invalid = False

    def setFolder(self, folder):
        self.folder = folder

    def _check(self, response, command):
        if response[0] != "OK":
            print "Invalid response to %s:\n%s" % (command, response)
            sys.exit(-1)

    def extractTime(self):
        # When we create a new copy of a message, we need to specify
        # a timestamp for the message.  If the message has a valid date
        # header we use that.  Otherwise, we use the current time.
        message_date = self["Date"]
        if message_date is not None:
            parsed_date = parsedate(message_date)
            if parsed_date is not None:
                try:
                    return Time2Internaldate(time.mktime(parsed_date))
                except ValueError:
                    # Invalid dates can cause mktime() to raise a
                    # ValueError, for example:
                    #   >>> time.mktime(parsedate("Mon, 06 May 0102 10:51:16 -0100"))
                    #   Traceback (most recent call last):
                    #     File "<interactive input>", line 1, in ?
                    #   ValueError: year out of range
                    # (Why this person is getting mail from almost two
                    # thousand years ago is another question <wink>).
                    # In any case, we just pass and use the current date.
                    pass
                except OverflowError:
                    pass
        return Time2Internaldate(time.time())

    def get_substance(self):
        '''Retrieve the RFC822 message from the IMAP server and set as the
        substance of this message.'''
        if self.got_substance:
            return
        if not self.uid or not self.id:
            print "Cannot get substance of message without an id and an UID"
            return
        imap.SelectFolder(self.folder.name)
        try:
            response = imap.uid("FETCH", self.uid, self.rfc822_command)
        except IMAP4.error:
            self.rfc822_command = "RFC822"
            self.rfc822_key = "RFC822"
            response = imap.uid("FETCH", self.uid, self.rfc822_command)
        if response[0] != "OK":
            self.rfc822_command = "RFC822"
            self.rfc822_key = "RFC822"
            response = imap.uid("FETCH", self.uid, self.rfc822_command)
        self._check(response, "uid fetch")
        data = _extract_fetch_data(response[1][0])
        # Annoyingly, we can't just pass over the RFC822 message to an
        # existing message object (like self) and have it parse it. So
        # we go through the hoops of creating a new message, and then
        # copying over all its internals.
        try:
            new_msg = email.Parser.Parser().parsestr(data[self.rfc822_key])
        # We use a general 'except' because the email package doesn't
        # always return email.Errors (it can return a TypeError, for
        # example) if the email is invalid.  In any case, we want
        # to keep going, and not crash, because we might leave the
        # user's mailbox in a bad state if we do.  Better to soldier
        # on.
        except:
            # Yikes!  Barry set this to return at this point, which
            # would work ok for training (IIRC, that's all he's
            # using it for), but for filtering, what happens is that
            # the message ends up blank, but ok, so the original is
            # flagged to be deleted, and a new (almost certainly
            # unsure) message, *with only the spambayes headers* is
            # created.  The nice solution is still to do what sb_server
            # does and have a X-Spambayes-Exception header with the
            # exception data and then the original message.
            self.invalid = True

            # This is nicked from sb_server - thanks Richie!
            stream = StringIO.StringIO()
            traceback.print_exc(None, stream)
            details = stream.getvalue()

            # Build the header.  This will strip leading whitespace from
            # the lines, so we add a leading dot to maintain indentation.
            detailLines = details.strip().split('\n')
            dottedDetails = '\n.'.join(detailLines)
            headerName = 'X-Spambayes-Exception'
            header = Header(dottedDetails, header_name=headerName)

            # Insert the header, converting email.Header's '\n' line
            # breaks to IMAP4's '\r\n'.
            # (Also insert the id header, otherwise we'll keep doing
            # this message over and over again).
            headers, body = re.split(r'\n\r?\n', data["RFC822"], 1)
            header = re.sub(r'\r?\n', '\r\n', str(header))
            headers += "\n%s: %s\r\n%s: %s\r\n\r\n" % \
                       (headerName, header,
                        options["Headers", "mailid_header_name"], self.id)
            self.invalid_content = headers + body

            # Print the exception and a traceback.
            print >>sys.stderr, details
        else:
            self._headers = new_msg._headers
            self._unixfrom = new_msg._unixfrom
            self._payload = new_msg._payload
            self._charset = new_msg._charset
            self.preamble = new_msg.preamble
            self.epilogue = new_msg.epilogue
            self._default_type = new_msg._default_type
            if not self.has_key(options["Headers", "mailid_header_name"]):
                self[options["Headers", "mailid_header_name"]] = self.id
        self.got_substance = True
        if options["globals", "verbose"]:
            sys.stdout.write(chr(8) + "*")

    def MoveTo(self, dest):
        '''Note that message should move to another folder.  No move is
        carried out until Save() is called, for efficiency.'''
        if self.previous_folder is None:
            self.previous_folder = self.folder
        self.folder = dest

    def as_string(self, unixfrom=False):
        # Basically the same as the parent class's except that we handle
        # the case where the data was unparsable, so we haven't done any
        # filtering, and we are not actually a proper email.Message object.
        if self.invalid:
            return self._force_CRLF(self.invalid_content)
        else:
            return message.SBHeaderMessage.as_string(self, unixfrom)

    def Save(self):
        '''Save message to imap server.'''
        # we can't actually update the message with IMAP
        # so what we do is create a new message and delete the old one
        if self.folder is None:
            raise RuntimeError, """Can't save a message that doesn't
            have a folder."""
        if not self.id:
            raise RuntimeError, """Can't save a message that doesn't have
            an id."""
        response = imap.uid("FETCH", self.uid, "(FLAGS INTERNALDATE)")
        self._check(response, 'fetch (flags internaldate)')
        data = _extract_fetch_data(response[1][0])
        if data.has_key("INTERNALDATE"):
            msg_time = data["INTERNALDATE"]
        else:
            msg_time = self.extractTime()
        if data.has_key("FLAGS"):
            flags = data["FLAGS"]
            # The \Recent flag can be fetched, but cannot be stored
            # We must remove it from the list if it is there.
            flags = re.sub(r"\\Recent ?| ?\\Recent", "", flags)
        else:
            flags = None

        for flgs, tme in [(flags, msg_time),
                          (None, msg_time),
                          (flags, Time2Internaldate(time.time())),
                          (None, Time2Internaldate(time.time()))]:
            response = imap.append(self.folder.name, flgs, tme,
                                   self.as_string())
            if response[0] == "OK":
                break
        self._check(response, 'append')

        if self.previous_folder is None:
            imap.SelectFolder(self.folder.name)
        else:
            imap.SelectFolder(self.previous_folder.name)
            self.previous_folder = None
        response = imap.uid("STORE", self.uid, "+FLAGS.SILENT", "(\\Deleted \\Seen)")
        self._check(response, 'store')
        # Not all IMAP servers immediately offer the new message
        # (stupidly), but we need to find it.  Generally a 'no-op' will
        # allow the server time to handle it, so do that.
        # See [ 941596 ] sb_imapfilter.py not adding headers / moving messages
        imap.noop()

        # We need to update the uid, as it will have changed.
        # Although we don't use the UID to keep track of messages, we do
        # have to use it for IMAP operations.
        imap.SelectFolder(self.folder.name)
        response = imap.uid("SEARCH", "(UNDELETED HEADER %s \"%s\")" % \
                            (options["Headers", "mailid_header_name"],
                             self.id.replace('\\',r'\\').replace('"',r'\"')))
        self._check(response, 'search')
        new_id = response[1][0]

        # See [ 870799 ] imap trying to fetch invalid message UID
        # It seems that although the save gave a "NO" response to the
        # first save, the message was still saved (without the flags,
        # probably).  This isn't really good behaviour on the server's
        # part, but, as usual, we try and deal with it.  So, if we get
        # more than one undeleted message with the same SpamBayes id,
        # delete all of them apart from the last one, and use that.
        multiple_ids = new_id.split()
        for id_to_remove in multiple_ids[:-1]:
            response = imap.uid("STORE", id_to_remove, "+FLAGS.SILENT",
                                "(\\Deleted \\Seen)")
            self._check(response, 'store')
        if multiple_ids:
            new_id = multiple_ids[-1]
        else:
            # Let's hope it doesn't, but, just in case, if the search
            # turns up empty, we make the assumption that the new
            # message is the last one with a recent flag
            response = imap.uid("SEARCH", "RECENT")
            new_id = response[1][0]
            if new_id.find(' ') > -1:
                ids = new_id.split(' ')
                new_id = ids[-1]
            # Ok, now we're in trouble if we still haven't found it.
            # We make a huge assumption that the new message is the one
            # with the highest UID (they are sequential, so this will be
            # ok as long as another message hasn't also arrived)
            if new_id == "":
                response = imap.uid("SEARCH", "ALL")
                new_id = response[1][0]
                if new_id.find(' ') > -1:
                    ids = new_id.split(' ')
                    new_id = ids[-1]
        self.uid = new_id

# This performs a similar function to email.message_from_string()
def imapmessage_from_string(s, _class=IMAPMessage, strict=False):
    return email.message_from_string(s, _class, strict)


class IMAPFolder(object):
    def __init__(self, folder_name):
        self.name = folder_name
        # Unique names for cached messages - see _generate_id below.
        self.lastBaseMessageName = ''
        self.uniquifier = 2

    def __cmp__(self, obj):
        '''Two folders are equal if their names are equal'''
        if obj is None:
            return False
        return cmp(self.name, obj.name)

    def _check(self, response, command):
        if response[0] != "OK":
            print "Invalid response to %s:\n%s" % (command, response)
            sys.exit(-1)

    def __iter__(self):
        '''IMAPFolder is iterable'''
        for key in self.keys():
            try:
                yield self[key]
            except KeyError:
                pass

    def recent_uids(self):
        '''Returns uids for all the messages in the folder that
        are flagged as recent, but not flagged as deleted.'''
        imap.SelectFolder(self.name, True)
        response = imap.uid("SEARCH", "RECENT UNDELETED")
        self._check(response, "SEARCH RECENT UNDELETED")
        return response[1][0].split(' ')

    def keys(self):
        '''Returns *uids* for all the messages in the folder not
        marked as deleted.'''
        imap.SelectFolder(self.name)
        response = imap.uid("SEARCH", "UNDELETED")
        self._check(response, "SEARCH UNDELETED")
        if response[1][0]:
            return response[1][0].split(' ')
        else:
            return []

    def __getitem__(self, key):
        '''Return message (no substance) matching the given *uid*.'''
        # We don't retrieve the substances of the message here - you need
        # to call msg.get_substance() to do that.
        imap.SelectFolder(self.name)
        # Using RFC822.HEADER.LINES would be better here, but it seems
        # that not all servers accept it, even though it is in the RFC
        response = imap.uid("FETCH", key, "RFC822.HEADER")
        self._check(response, "uid fetch header")
        data = _extract_fetch_data(response[1][0])

        msg = IMAPMessage()
        msg.setFolder(self)
        msg.uid = key
        r = re.compile(re.escape(options["Headers",
                                         "mailid_header_name"]) + \
                       "\:\s*(\d+(\-\d)?)")
        mo = r.search(data["RFC822.HEADER"])
        if mo is None:
            msg.setId(self._generate_id())
            # Unfortunately, we now have to re-save this message, so that
            # our id is stored on the IMAP server.  Before anyone suggests
            # it, we can't store it as a flag, because user-defined flags
            # aren't supported by all IMAP servers.
            # This will need to be done once per message.
            msg.get_substance()
            msg.Save()
        else:
            msg.setId(mo.group(1))

        if options["globals", "verbose"]:
            sys.stdout.write(".")
        return msg

    # Lifted straight from pop3proxy.py (under the name getNewMessageName)
    def _generate_id(self):
        # The message id is the time it arrived, with a uniquifier
        # appended if two arrive within one clock tick of each other.
        messageName = "%10.10d" % long(time.time())
        if messageName == self.lastBaseMessageName:
            messageName = "%s-%d" % (messageName, self.uniquifier)
            self.uniquifier += 1
        else:
            self.lastBaseMessageName = messageName
            self.uniquifier = 2
        return messageName

    def Train(self, classifier, isSpam):
        '''Train folder as spam/ham'''
        num_trained = 0
        for msg in self:
            if msg.GetTrained() == (not isSpam):
                msg.get_substance()
                msg.delSBHeaders()
                classifier.unlearn(msg.asTokens(), not isSpam)
                # Once the message has been untrained, it's training memory
                # should reflect that on the off chance that for some reason
                # the training breaks, which happens on occasion (the
                # tokenizer is not yet perfect)
                msg.RememberTrained(None)

            if msg.GetTrained() is None:
                msg.get_substance()
                saved_headers = msg.currentSBHeaders()
                msg.delSBHeaders()
                classifier.learn(msg.asTokens(), isSpam)
                num_trained += 1
                msg.RememberTrained(isSpam)
                if isSpam:
                    move_opt_name = "move_trained_spam_to_folder"
                else:
                    move_opt_name = "move_trained_ham_to_folder"
                if options["imap", move_opt_name] != "":
                    # We need to restore the SpamBayes headers.
                    for header, value in saved_headers.items():
                        msg[header] = value
                    msg.MoveTo(IMAPFolder(options["imap",
                                                  move_opt_name]))
                    msg.Save()
        return num_trained

    def Filter(self, classifier, spamfolder, unsurefolder):
        count = {}
        count["ham"] = 0
        count["spam"] = 0
        count["unsure"] = 0
        for msg in self:
            if msg.GetClassification() is None:
                msg.get_substance()
                (prob, clues) = classifier.spamprob(msg.asTokens(),
                                                    evidence=True)
                # add headers and remember classification
                msg.delSBHeaders()
                msg.addSBHeaders(prob, clues)

                cls = msg.GetClassification()
                if cls == options["Headers", "header_ham_string"]:
                    # we leave ham alone
                    count["ham"] += 1
                elif cls == options["Headers", "header_spam_string"]:
                    msg.MoveTo(spamfolder)
                    count["spam"] += 1
                else:
                    msg.MoveTo(unsurefolder)
                    count["unsure"] += 1
                msg.Save()
        return count


class IMAPFilter(object):
    def __init__(self, classifier):
        self.spam_folder = IMAPFolder(options["imap", "spam_folder"])
        self.unsure_folder = IMAPFolder(options["imap", "unsure_folder"])
        self.classifier = classifier

    def Train(self):
        if options["globals", "verbose"]:
            t = time.time()

        total_ham_trained = 0
        total_spam_trained = 0

        if options["imap", "ham_train_folders"] != "":
            ham_training_folders = options["imap", "ham_train_folders"]
            for fol in ham_training_folders:
                # Select the folder to make sure it exists
                imap.SelectFolder(fol)
                if options['globals', 'verbose']:
                    print "   Training ham folder %s" % (fol)
                folder = IMAPFolder(fol)
                num_ham_trained = folder.Train(self.classifier, False)
                total_ham_trained += num_ham_trained
                if options['globals', 'verbose']:
                    print "       %s trained." % (num_ham_trained)

        if options["imap", "spam_train_folders"] != "":
            spam_training_folders = options["imap", "spam_train_folders"]
            for fol in spam_training_folders:
                # Select the folder to make sure it exists
                imap.SelectFolder(fol)
                if options['globals', 'verbose']:
                    print "   Training spam folder %s" % (fol)
                folder = IMAPFolder(fol)
                num_spam_trained = folder.Train(self.classifier, True)
                total_spam_trained += num_spam_trained
                if options['globals', 'verbose']:
                    print "       %s trained." % (num_spam_trained)

        if total_ham_trained or total_spam_trained:
            self.classifier.store()

        if options["globals", "verbose"]:
            print "Training took %.4f seconds, %s messages were trained" \
                  % (time.time() - t, total_ham_trained + total_spam_trained)

    def Filter(self):
        if options["globals", "verbose"]:
            t = time.time()
        count = {}
        count["ham"] = 0
        count["spam"] = 0
        count["unsure"] = 0

        # Select the spam folder and unsure folder to make sure they exist
        imap.SelectFolder(self.spam_folder.name)
        imap.SelectFolder(self.unsure_folder.name)

        for filter_folder in options["imap", "filter_folders"]:
            # Select the folder to make sure it exists
            imap.SelectFolder(filter_folder)
            folder = IMAPFolder(filter_folder)
            subcount = folder.Filter(self.classifier, self.spam_folder,
                          self.unsure_folder)
            for key in count.keys():
                count[key] += subcount.get(key, 0)

        if options["globals", "verbose"]:
            if count is not None:
                print "\nClassified %s ham, %s spam, and %s unsure." % \
                      (count["ham"], count["spam"], count["unsure"])
            print "Classifying took %.4f seconds." % (time.time() - t,)


def run():
    global imap
    try:
        opts, args = getopt.getopt(sys.argv[1:], 'hbPtcvl:e:i:d:p:o:')
    except getopt.error, msg:
        print >>sys.stderr, str(msg) + '\n\n' + __doc__
        sys.exit()

    doTrain = False
    doClassify = False
    doExpunge = options["imap", "expunge"]
    imapDebug = 0
    sleepTime = 0
    promptForPass = False
    launchUI = False
    server = ""
    username = ""

    for opt, arg in opts:
        if opt == '-h':
            print >>sys.stderr, __doc__
            sys.exit()
        elif opt == "-b":
            launchUI = True
        elif opt == '-t':
            doTrain = True
        elif opt == '-P':
            promptForPass = True
        elif opt == '-c':
            doClassify = True
        elif opt == '-v':
            options["globals", "verbose"] = True
        elif opt == '-e':
            if arg == 'y':
                doExpunge = True
            else:
                doExpunge = False
        elif opt == '-i':
            imapDebug = int(arg)
        elif opt == '-l':
            sleepTime = int(arg) * 60
        elif opt == '-o':
            options.set_from_cmdline(arg, sys.stderr)
    bdbname, useDBM = storage.database_type(opts)

    # Let the user know what they are using...
    print get_version_string("IMAP Filter")
    print "and engine %s.\n" % (get_version_string(),)

    if (launchUI and (doClassify or doTrain)):
        print """-b option is exclusive with -c and -t options.
The user interface will be launched, but no classification
or training will be performed.
"""

    if options["globals", "verbose"]:
        print "Loading database %s..." % (bdbname),

    classifier = storage.open_storage(bdbname, useDBM)

    if options["globals", "verbose"]:
        print "Done."

    if options["imap", "server"]:
        # The options class is ahead of us here:
        #   it knows that imap:server will eventually be able to have
        #   multiple values, but for the moment, we just use the first one
        server = options["imap", "server"]
        if len(server) > 0:
            server = server[0]
        username = options["imap", "username"]
        if len(username) > 0:
            username = username[0]
        if not promptForPass:
            pwd = options["imap", "password"]
            if len(pwd) > 0:
                pwd = pwd[0]
    else:
        pwd = None
        if not launchUI:
            print "You need to specify both a server and a username."
            sys.exit()

    if promptForPass:
        pwd = getpass()

    if server.find(':') > -1:
        server, port = server.split(':', 1)
        port = int(port)
    else:
        if options["imap", "use_ssl"]:
            port = 993
        else:
            port = 143

    imap_filter = IMAPFilter(classifier)

    # Web interface
    if not (doClassify or doTrain):
        print "XXX Serving web interface", imap
        if server != "":
            imap = IMAPSession(server, port, imapDebug, doExpunge)
        print "XXX With imap =", imap
        httpServer = UserInterfaceServer(options["html_ui", "port"])
        print "XXX And httpServer =", httpServer
        httpServer.register(IMAPUserInterface(classifier, imap, pwd,
                                              IMAPSession))
        print "XXX Registered, will now run"
        Dibbler.run(launchBrowser=launchUI or options["html_ui",
                                                      "launch_browser"])
    else:
        while True:
            imap = IMAPSession(server, port, imapDebug, doExpunge)
            imap.login(username, pwd)

            if doTrain:
                if options["globals", "verbose"]:
                    print "Training"
                imap_filter.Train()
            if doClassify:
                if options["globals", "verbose"]:
                    print "Classifying"
                imap_filter.Filter()

            imap.logout()

            if sleepTime:
                time.sleep(sleepTime)
            else:
                break

if __name__ == '__main__':
    run()
From anthony at interlink.com.au  Tue Nov 30 04:16:15 2004
From: anthony at interlink.com.au (Anthony Baxter)
Date: Tue Nov 30 04:17:42 2004
Subject: [Spambayes] Download verification checking request
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E8FBCA94@its-xchg4.massey.ac.nz>
References: <ECBA357DDED63B4995F5C1F5CBE5B1E8FBCA94@its-xchg4.massey.ac.nz>
Message-ID: <200411301416.17348.anthony@interlink.com.au>

Suggestions:

Use an ascii sig (gpg -ba)

Put the sig files on the website directly, they're too small to justify 
putting in the Files section.

Put your key id on the download page - that way people can more
easily fetch it from the keyservers.

From usr7-28hl at xemaps.com  Tue Nov 30 04:52:25 2004
From: usr7-28hl at xemaps.com (Keith Russell)
Date: Tue Nov 30 04:52:32 2004
Subject: [Spambayes] Unable to Access SpamBayes Browser
	InterfaceUsingsb_imapfilter.py
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E8FBCA9B@its-xchg4.massey.ac.nz>
References: <ECBA357DDED63B4995F5C1F5CBE5B1E8FBCA9B@its-xchg4.massey.ac.nz>
Message-ID: <41ABEE79.5000504@mailsnare.net>

Thanks, Tony.

Tony Meyer wrote:

>>Thanks for this. Here's what I get:
>>
>> >sb_imapfilter.py -b
>>SpamBayes IMAP Filter Version 0.5 (November 2004)
>>and engine SpamBayes Engine Version 0.3 (January 2004).
>>
>>XXX Serving web interface
> 
> 
> That's not very far at all!  It appears to be dying when creating the IMAP
> session, then, which is strange.  Attached is a new version with more
> debugging prints in new places (this ought to be enough to narrow it down to
> a specific place).  If you could do the same thing with this one (just
> sb_imapfilter.py -b is enough) that would be great

Here you go:

SpamBayes IMAP Filter Version 0.5 (November 2004)
and engine SpamBayes Engine Version 0.3 (January 2004).

XXX Serving web interface None
XXX Initialising base IMAP 993

-- 
Keith
From tameyer at ihug.co.nz  Tue Nov 30 05:12:46 2004
From: tameyer at ihug.co.nz (Tony Meyer)
Date: Tue Nov 30 05:13:48 2004
Subject: [Spambayes] Unable to Access SpamBayes
	BrowserInterfaceUsingsb_imapfilter.py
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E80189369F@its-xchg4.massey.ac.nz>
Message-ID: <ECBA357DDED63B4995F5C1F5CBE5B1E86C3842@its-xchg4.massey.ac.nz>

> Here you go:
> 
> SpamBayes IMAP Filter Version 0.5 (November 2004)
> and engine SpamBayes Engine Version 0.3 (January 2004).
> 
> XXX Serving web interface None
> XXX Initialising base IMAP 993

Ah ha!  It appears that it will hang when it manages to connect to a remote
server, but that server doesn't give out an initial welcome message (it will
wait for it indefinitely).  Regular IMAP does this, but IMAP over SSL
doesn't.  I'll have to figure out exactly how I'll modify the code to avoid
this happening to anyone, but I suspect that the fix in your case is to open
your configuration file (probably C:\Documents and
Settings\{username}\Application Data\Proxy\bayescustomize.ini) and add in
these lines:

[imap]
use_ssl:True

(If there's already an [imap] section, you can just add the use_ssl line
under that).

=Tony.Meyer

-- 
Please always include the list (spambayes@python.org) in your replies
(reply-all), and please don't send me personal mail about SpamBayes.
http://www.massey.ac.nz/~tameyer/writing/reply_all.html explains this.

From usr7-28hl at xemaps.com  Tue Nov 30 05:45:12 2004
From: usr7-28hl at xemaps.com (Keith Russell)
Date: Tue Nov 30 05:45:39 2004
Subject: [Spambayes] Unable to Access SpamBayes
	BrowserInterfaceUsingsb_imapfilter.py
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E86C3842@its-xchg4.massey.ac.nz>
References: <ECBA357DDED63B4995F5C1F5CBE5B1E86C3842@its-xchg4.massey.ac.nz>
Message-ID: <41ABFAD8.8070408@mailsnare.net>

Tony Meyer wrote:
> I suspect that the fix in your case is to open
> your configuration file (probably C:\Documents and
> Settings\{username}\Application Data\Proxy\bayescustomize.ini)

This threw me for a second, because I couldn't find this path. 
It's actually C:\Documents and
Settings\{username}\Application 
Data\SpamBayes\Proxy\bayescustomize.ini :-)

> and add in
> these lines:
> 
> [imap]
> use_ssl:True
> 
> (If there's already an [imap] section, you can just add the use_ssl line
> under that).

There was an IMAP section, with four entry lines, including 
"use_ssl:False". I changed it to "True", and FINALLY, SUCCESS!

Thank you very much. Now I can get back to actually configuring 
and testing it, and see if I can make it work!

-- 
Keith

From tameyer at ihug.co.nz  Tue Nov 30 05:55:02 2004
From: tameyer at ihug.co.nz (Tony Meyer)
Date: Tue Nov 30 05:55:40 2004
Subject: [Spambayes] Unable to Access
	SpamBayesBrowserInterfaceUsingsb_imapfilter.py
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E8018936B3@its-xchg4.massey.ac.nz>
Message-ID: <ECBA357DDED63B4995F5C1F5CBE5B1E8FBCAA5@its-xchg4.massey.ac.nz>

> This threw me for a second, because I couldn't find this path. 
> It's actually C:\Documents and
> Settings\{username}\Application 
> Data\SpamBayes\Proxy\bayescustomize.ini :-)

Sorry - typing too fast.

> There was an IMAP section, with four entry lines, including 
> "use_ssl:False". I changed it to "True", and FINALLY, SUCCESS!

Fantastic :)

> Thank you very much. Now I can get back to actually configuring 
> and testing it, and see if I can make it work!

Let me know if you have any more troubles (although it really ought to be
relatively smooth from here on).  I'll figure out a way to catch this
problem and warn users somehow.

BTW, when 1.1 comes out (1.1a1 should be early next year) there are lots of
imapfilter improvements, so if you think of any others that you'd like, let
us know :)

=Tony.Meyer

-- 
Please always include the list (spambayes@python.org) in your replies
(reply-all), and please don't send me personal mail about SpamBayes.
http://www.massey.ac.nz/~tameyer/writing/reply_all.html explains this.

From tameyer at ihug.co.nz  Tue Nov 30 06:59:54 2004
From: tameyer at ihug.co.nz (Tony Meyer)
Date: Tue Nov 30 07:01:10 2004
Subject: [spambayes-dev] Re: [Spambayes] Download verification checking
	request
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E80189368D@its-xchg4.massey.ac.nz>
Message-ID: <ECBA357DDED63B4995F5C1F5CBE5B1E86C3843@its-xchg4.massey.ac.nz>

> Suggestions:

Thanks :)

> Use an ascii sig (gpg -ba)

Done.

> Put the sig files on the website directly, they're too small 
> to justify putting in the Files section.

Done.

> Put your key id on the download page - that way people can 
> more easily fetch it from the keyservers.

Done.

I swapped to GnuPG, too, because using the command line was just so much
easier than using the GUI PGP.  I've added instructions about doing this to
README-DEVEL.txt, so hopefully it should all go smoothly next time :)

=Tony.Meyer

From kennypitt at hotmail.com  Tue Nov 30 14:08:37 2004
From: kennypitt at hotmail.com (Kenny Pitt)
Date: Tue Nov 30 14:09:12 2004
Subject: [Spambayes] deleted my junk suspects folder, now messed up
In-Reply-To: <D5719B6000144E4794AF4BC229EE9B4663598B@MACCPDC1.Mobileareachamber.com>
Message-ID: <BAY16-DAV11E94CE2DF801581CA0D00A1BE0@phx.gbl>

See FAQ 3.13:
 
http://spambayes.sourceforge.net/faq.html#help-i-deleted-the-unsure-spam-fol
der
 
This happens quite regularly because Outlook makes it all too easy to
accidentally delete the entire folder instead of just the messages in it.
Unfortunately, there is no way to prevent the folders from being deleted.
If you use Outlook 2003, you can set Outlook's "Junk E-mail" folder as your
SpamBayes spam folder and Outlook will not let this folder be deleted, but
there is still nothing you can do except be careful with your suspects
folder.
 
-- 
Kenny Pitt
 

  _____  

From: spambayes-bounces@python.org [mailto:spambayes-bounces@python.org] On
Behalf Of Jennifer Stewart
Sent: Wednesday, November 24, 2004 9:25 AM
To: spambayes@python.org
Subject: [Spambayes] deleted my junk suspects folder, now messed up
Importance: High



OK, I accidentally deleted my junk suspects box, now it says its not
configured correctly and our computer person is out until Monday.  Any help?
I tried things on my own but nothing seems to work.  Can you make the junk
and junk suspects folders where you can't delete the  without being
prompted?

 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.python.org/pipermail/spambayes/attachments/20041130/7bee5508/attachment.htm
From scott at lentigo.net  Tue Nov 30 14:38:00 2004
From: scott at lentigo.net (Scott Burns)
Date: Tue Nov 30 14:38:11 2004
Subject: [Spambayes] Download verification checking request
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E8FBCA94@its-xchg4.massey.ac.nz>
References: <ECBA357DDED63B4995F5C1F5CBE5B1E8FBCA94@its-xchg4.massey.ac.nz>
Message-ID: <41AC77B8.2080904@lentigo.net>

I grabbed your public key, the spambayes-1.0.1.tar.gz tarball, and 
spambayes-1.0.1.tar.gz.asc sig file.  I followed the instructions on the 
website (testing those, too) and everything worked just fine. ...s.

Tony Meyer wrote:
> As per
> 
> [ 1061119 ] Provide verification and/or security measures for downloads
> <http://sf.net/tracker/?func=detail&atid=498106&aid=1061119&group_id=61702>
> 
> the 1.0.1 release (and future releases) includes some measures to try and
> let people confirm that their download is valid.  Specifically, the file
> sizes and MD5 checksums are listed on the download page:
> 
> <http://spambayes.org/download.html>
> 
> and there are PGP signatures for each of the three files, signed by the
> release manager (me, for this one).
> 
> I'm pretty new to this stuff - I'm pretty sure that the MD5s and sizes are
> correct, but not 100% about the signatures.  If someone familiar with this
> stuff could use the signatures to verify the files, and let me know (either
> here or via the tracker above) so that I know I've done the process
> correctly, that would be fantastic.
> 
> Thanks!
> 
> =Tony.Meyer

-- 
Scott Burns <scott@lentigo.net> <http://www.lentigo.net/scott>
pub  1024D/9DA64618 2001-11-17 Scott Burns <scott@lentigo.net>
Fingerprint: 2F1B A22E 33C3 FD3D BBE4 D5E2 728B 4753 9DA6 4618

From ta-meyer at ihug.co.nz  Tue Nov 30 22:37:47 2004
From: ta-meyer at ihug.co.nz (Tony Meyer)
Date: Tue Nov 30 22:38:27 2004
Subject: [Spambayes] Download verification checking request
In-Reply-To: <ECBA357DDED63B4995F5C1F5CBE5B1E80189379D@its-xchg4.massey.ac.nz>
Message-ID: <ECBA357DDED63B4995F5C1F5CBE5B1E8FBCAA8@its-xchg4.massey.ac.nz>

> I grabbed your public key, the spambayes-1.0.1.tar.gz tarball, and 
> spambayes-1.0.1.tar.gz.asc sig file.  I followed the 
> instructions on the website (testing those, too) and everything
> worked just fine. ...s.

Thanks for letting me know; much appreciated.

=Tony.Meyer

From SMTP at epitech.net  Tue Nov 30 22:24:33 2004
From: SMTP at epitech.net (SMTP@epitech.net)
Date: Wed Dec  1 01:58:31 2004
Subject: [Spambayes] Virus found in a message you sent
Message-ID: <20041201005830.5E9DB1E4005@bag.python.org>

A virus was found in a message sent by this
account.

--- Scan information follows ---

Result: Virus Detected
Virus Name: W32.Sober.I@mm!zip
File Attachment: oh_nono.txt.zip
Attachment Status: deleted

--- Original message information follows ---

From: spambayes@python.org
To: demaille@inf.enst.fr
Date: Tue, 30 Nov 2004 21:10:44 GMT
Subject: Oh God it's
Message-Id: <e833.1051b4b5dfccb8@python.org>
Received: from infres.enst.fr ([137.194.192.1])
 by itoth2.epitech.net (SAVSMTP 3.1.2.35) with SMTP id M2004113022242816358
 for <akim@epita.fr>; Tue, 30 Nov 2004 22:24:29 +0100