From michaelmurdock at gmail.com  Sun Oct  2 08:35:16 2005
From: michaelmurdock at gmail.com (Michael Murdock)
Date: Sun, 2 Oct 2005 00:35:16 -0600
Subject: [spambayes-dev] 2 types of spam
Message-ID: <bc98f3670510012335q51e9117crcd934acdd4ed248@mail.gmail.com>

 One feature that seems to be missing from Spambayes - and other spam
> checkers is the ability to distinguish & then sort unwanted spam from
> wanted
> spam. Unwanted being the stuff I know I never want to look at eg Viagra
> adverts, Cheap USA mortgage offers, Nigerian business opportunities etc
> etc. Wanted spam is the stuff I do check from time to time eg
> Lastminute.com <http://Lastminute.com>
> offers, Misco catalogues etc.
>
> Any chance of offering such a facility. It would seem that you have all
> the
> basic tools to make it happen.

 The first type of email that you describe is spam. The second type
of email is unsolicited. But this doesn't make it spam, since you
don't mind receiving it.
 In principle, you should be able to train the models on the first type
of email, by calling it spam. And on the second type of email you
train the models to accept it by calling it ham. Presumably, and with
enough examples, SpamBayes should learn, at least somewhat, to
mark as spam the nigerian biz opportunities and mark as ham the
LastMinute.com offers. So I guess my uneducated answer is that
you are right: SpamBayes does have all the basic tools to make it
happen - it's all about the training.
 ~Michael.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.python.org/pipermail/spambayes-dev/attachments/20051002/a6794977/attachment.htm

From scott at hypexr.org  Mon Oct  3 00:44:56 2005
From: scott at hypexr.org (Scott Rippee)
Date: Sun, 2 Oct 2005 15:44:56 -0700
Subject: [spambayes-dev] 2 Types of Spam
In-Reply-To: <NKEMIEKKAOOABPKDMBMJKEEACIAA.mike@horrocky.com>
References: <NKEMIEKKAOOABPKDMBMJKEEACIAA.mike@horrocky.com>
Message-ID: <20051002224456.GB4023@mrcel.hypexr.org>

The kind of mail that you are talking about it usually easy enough to filter with normal filtering rules (ie. keyword matches).  It would however be easy enough to do what you are talking about (at least in Linux using my technique: procmail, sb_filter.py, sp_mboxtrain.py).  

You could have two spam databases.  The first one would be for your normal spam and the second for the other emails.  When you check spam with your procmail rules you would run it through the first filter and then run it through the second filter for the sb_filter classification.  Then you would move your special spam into its own location and add a additional scheduled training for the second database with the spam you want to check on occasions.

On Wed, Sep 28, 2005 at 10:46:39PM +0100, Mike and Lynn Horrocks wrote:
> One feature that seems to be missing from Spambayes  - and other spam
> checkers is the ability to distinguish & then sort unwanted spam from wanted
> spam. Unwanted being the stuff I know I never want to look at eg Viagra
> adverts, Cheap USA mortgage offers,  Nigerian business opportunities etc
> etc. Wanted spam is the stuff I do check from time to time eg Lastminute.com
> offers, Misco catalogues etc.
> 
> Any chance of offering such a facility. It would seem that you have all the
> basic tools to make it happen.
> 
> Mike


-- 
 Scott Rippee
   scott at hypexr.org
   http://www.hypexr.org

From kenny.pitt at gmail.com  Fri Oct  7 19:49:17 2005
From: kenny.pitt at gmail.com (Kenny Pitt)
Date: Fri, 7 Oct 2005 13:49:17 -0400
Subject: [spambayes-dev] [Spambayes] Bit of help on adding another
	statistic = )
In-Reply-To: <20051005101115.IAGU9927.ibm69aec.bellsouth.net@ENLIGHTENED>
References: <E94364AE-39EF-4B59-889D-244BAF5B2934@ihug.co.nz>
	<20051005101115.IAGU9927.ibm69aec.bellsouth.net@ENLIGHTENED>
Message-ID: <2a052b990510071049s25b4e48epb9f7ea59d864f5e4@mail.gmail.com>

FYI, if you are interested in modifying the spambayes source code then
your questions are probably best answered on the spambayes-dev list.

On 10/5/05, Erik Brown <kirebrow at yahoo.com> wrote:
> I tried to hack the Stats.py to display in larger font below the spam
> savings the total number of messages classified - the total number of
> messages classified incorrectly.

First off, using a different font for any of the elements in a Windows
dialog box requires a fare bit of low-level Windows programming. If
you can forego having your new statistics in a different font then the
task will be much easier. If you really want to go to the trouble,
though, you can find the dialog box code under the Outlook2000\dialogs
source directory.

> Just for kicks, I am providing a screen shot of what I'm trying to
> accomplish.  My current numbers after 860 messages classified is only 2
> misclassified.  So the number would be 100 - 0.2326 = 99.7674.

That is certainly a valid statistic, but may be misleading because it
doesn't take into account good messages that were removed from the
Inbox as unsure. These messages are considered "misclassified", but
you have to go looking for them somewhere other than your normal Inbox
folder.

> The number on the screen shot indicates what I am trying to do.  Can you
> provide me the code so that I can hack my local copy perhaps?

If you are already hacking Stats.py then you should already have all
the source code. Stats.py generates a string containing all the lines
of statistics, and the string is then displayed in the SpamBayes
Manager dialog in Outlook.

There are several functions that you would need to modify.
_CalculateAdditional computes the actual numeric percentages as
floating point values. _AddPercentStrings then adds displayable string
versions of each percentage value. Finally, GetStats formats all the
statistics into the strings that you see displayed. Each call to
push(...) in GetStats creates a new line in the stats display.

The percentage you are interested in can be computed as 100.0 *
(num_seen - num_incorrect) / num_seen.

--
Kenny Pitt

From galvescarvalho at yahoo.com.br  Sat Oct  8 17:19:30 2005
From: galvescarvalho at yahoo.com.br (Geraldo Alves C.)
Date: Sat, 8 Oct 2005 12:19:30 -0300
Subject: [spambayes-dev] 2.000 Modelos de Cartas Comerciais,
	Modelos de Propostas, contratos e Documentos
Message-ID: <20051008151948.9DFB31E400F@bag.python.org>

Os melhores Modelos de Contratos e Cartas Comerciais. Como escrever uma
proposta, ducumento ou carta comercial:

Visite agora:

http://www.gueb.de/modelosdecartascomerciais

Modelos de cartas de agradecimento, modelos de cartas de demiss?o, modelos
de cartas com mensagem mensagens de p?sames, modelos de convites, modelos
de declara??es, modelos de cartas de solicita??o de empregos, modelos de
cartas de cobran?as, modelos de cartas de recomenda??o, modelos de cartas
de recomenda??es, redigir cartas comerciais, modelos de propostas
comerciais, modelos de respostas a propostas comerciais, dicas de como
redigir, dicas de reda??o de cartas comerciais, modelos de pedidos,
modelos de atestados m?dicos.
Modelos de cartas em ingl?s, modelos de cartas de reclama??o, modelos de
cartas de refer?ncias, dicas de como escrever cartas comerciais, dicas de
como redigir cartas comerciais, modelos de convites para festas, modelos
de convites para eventos, modelos de cartas de pedidos de demiss?o,
modelos de cartas de felicita??es, modelos de cartas formais, modelos de
recibos, proposta de presta??o de servi?os, Modelos de Contratos
Como escrever um curr?culo, como redigir documentos:

Visite agora:

http://www.gueb.de/modelosdecartascomerciais

Modelos de cartas Convites, Propostas, Atas, Contratos, Agradecimentos,
Empregos, Solicita??o, Solicita??es, Apresenta??o, Apresenta??es,
Demiss?o, Demiss?es, Cobran?as, Pedidos, Atestados, Declara??es,
Declara??o, P?sames, Condol?ncias, Batizados, Fomal, Formais, Redigir,
Escrever, Escritas, Textos, Mensagens, Mensagem, Ingl?s, Clientes,
Fornecedores, Empresas, Neg?cios, Marketing, Comunicados, Comunica??o,
Secret?rias, Amigos, Parentes, Vendedores, Vendedoras, Refer?ncias,
Social, Sociais, Respostas, Dicas, M?dicos, Advogados, Duplicatas,
Felicita??o, Felicita??es, Recomenda??o, Recomenda??es, Funcion?rios,
Reclama??o, Reclama??es, Comunicar, Comunicados, Documentos, Viagens,
Viagem, Solenidades, Confraterniza??o, Confraterniza??es, Resultados,
Pre?os, Aumentos, Entregas, F?rias, Estudantes, Lan?amentos, Novos, Aviso
Pr?vio, Est?gios, Escolares, Protestos, D?bitos, Exposi??o, Exposi??es,
Feiras, Acordos, Recibos, Procura??es, Edital, Editais" 
modelos de cartas de demiss?o, modelos de cartas com mensagem mensagens de
p?sames, modelos de convites, modelos de declara??es, modelos de cartas de
solicita??o de empregos, modelos de cartas de cobran?as, modelos de cartas
de recomenda??o, modelos de cartas de recomenda??es, redigir cartas
comerciais, modelos de propostas comerciais, modelos de respostas a
propostas comerciais, dicas de como redigir, dicas de reda??o de cartas
comerciais, modelos de pedidos.

Visite agora:

http://www.gueb.de/modelosdecartascomerciais

modelos de cartas de demiss?o, modelos de cartas com mensagem mensagens de
p?sames, modelos de convites, modelos de declara??es, modelos de cartas de
solicita??o de empregos, modelos de cartas de cobran?as, modelos de cartas
de recomenda??o, modelos de cartas de recomenda??es, redigir cartas
comerciais, modelos de propostas comerciais, modelos de respostas a
propostas comerciais, dicas de como redigir, dicas de reda??o de cartas
comerciais, modelos de pedidos, modelos de atestados m?dicos.
Modelos de cartas em ingl?s, modelos de cartas de reclama??o, modelos de
cartas de refer?ncias, dicas de como escrever cartas comerciais, dicas de
como redigir cartas comerciais, modelos de convites para festas, modelos
de convites para eventos, modelos de cartas de pedidos de demiss?o,
modelos de cartas de felicita??es, modelos de cartas formais, modelos de
recibos, proposta de presta??o de servi?os, Modelos de Contratos
Cartas Comerciais, cartas, contartos, modelos, empresa, contabilidade,
advocacia, advogado, direito, procura??o, memorando.


http://www.gueb.de/modelosdecartascomerciais

propostas comerciais, dicas de como redigir, dicas de reda??o de cartas
comerciais, modelos de pedidos, modelos de atestados m?dicos.
Modelos de cartas em ingl?s, modelos de cartas de reclama??o, modelos de
cartas de refer?ncias, dicas de como escrever cartas comerciais, dicas de
como redigir cartas comerciais, modelos de convites para 

From kobermr at students.udmercy.edu  Tue Oct 18 04:30:08 2005
From: kobermr at students.udmercy.edu (Matt Kober)
Date: Mon, 17 Oct 2005 22:30:08 -0400
Subject: [spambayes-dev]  Business opportunity
Message-ID: <001e01c5d38b$db8b46e0$6501a8c0@MattsLaptop>

I saw some thread online about coral calcium from tonga.  Are you still
doing that?

 

Matt

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.python.org/pipermail/spambayes-dev/attachments/20051017/559087f0/attachment.htm

From canonrt at optonline.net  Sat Oct 22 12:42:37 2005
From: canonrt at optonline.net (canonrt)
Date: Sat, 22 Oct 2005 06:42:37 -0400
Subject: [spambayes-dev] Help
Message-ID: <000001c5d6f5$51e01a20$6501a8c0@canonsworld>

My spam bayes just stopped working!
 
I tried re loading it, re installing it, re downloading it, but I am
still unable to wake it up!
 
I really miss this great program!
 
What do I do?..................John
 
 
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.python.org/pipermail/spambayes-dev/attachments/20051022/2eb1dbfc/attachment.htm

From robert.mezzone at pjsolomon.com  Mon Oct 24 21:46:46 2005
From: robert.mezzone at pjsolomon.com (Robert Mezzone)
Date: Mon, 24 Oct 2005 15:46:46 -0400
Subject: [spambayes-dev] Help
Message-ID: <D71A62BA7CA7AB40ACBA500EF30BB63D01A6881C@pjscms>

Check your disabled items in Outlook, assuming you are using Outlook.
"Help" "About Microsoft Outlook" "Disabled Items" button.
 
hth.

________________________________

From: spambayes-dev-bounces at python.org
[mailto:spambayes-dev-bounces at python.org] On Behalf Of canonrt
Sent: Saturday, October 22, 2005 6:43 AM
To: spambayes-dev at python.org
Subject: [spambayes-dev] Help


My spam bayes just stopped working!
 
I tried re loading it, re installing it, re downloading it, but I am
still unable to wake it up!
 
I really miss this great program!
 
What do I do?..................John
 
 
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.python.org/pipermail/spambayes-dev/attachments/20051024/7804afde/attachment.htm

From avw at charter.net  Thu Oct 27 18:23:30 2005
From: avw at charter.net (avw)
Date: Thu, 27 Oct 2005 09:23:30 -0700
Subject: [spambayes-dev] Recover deleted mail
Message-ID: <000001c5db12$c5b16a00$6502a8c0@ALAN>

I accidentally clicked on "Delete as spam" and deleted an email I want. Is
there some place I can recover this from? It is not in the "Deleted items"
file.

 

Alan

 

 

 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.python.org/pipermail/spambayes-dev/attachments/20051027/e8dad215/attachment.htm

From robert.mezzone at pjsolomon.com  Thu Oct 27 18:27:47 2005
From: robert.mezzone at pjsolomon.com (Robert Mezzone)
Date: Thu, 27 Oct 2005 12:27:47 -0400
Subject: [spambayes-dev] Recover deleted mail
Message-ID: <D71A62BA7CA7AB40ACBA500EF30BB63D01664036@pjscms>

In the spam folder created by spambayes. That's where it goes by default when you click delete as spam. Assuming the message was in your inbox when you clicked the delete as spam button.

-----Original Message-----
From: spambayes-dev-bounces at python.org <spambayes-dev-bounces at python.org>
To: spambayes-dev at python.org <spambayes-dev at python.org>
Sent: Thu Oct 27 12:23:30 2005
Subject: [spambayes-dev] Recover deleted mail

I accidentally clicked on ?Delete as spam? and deleted an email I want. Is there some place I can recover this from? It is not in the ?Deleted items? file.

 

Alan

 

 

 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.python.org/pipermail/spambayes-dev/attachments/20051027/8b8f1583/attachment.html

From tameyer at ihug.co.nz  Sun Oct 30 04:17:29 2005
From: tameyer at ihug.co.nz (Tony Meyer)
Date: Sun, 30 Oct 2005 16:17:29 +1300
Subject: [spambayes-dev] Default storage backend
In-Reply-To: <c997m1dknqib1b36n9a73bnpn30ga9eprp@4ax.com>
References: <6.2.3.4.0.20051026125848.0247e630@pop3.highstream.net>
	<hgevl1hb2101cs146b0fmnvo13ib91b5hd@4ax.com>
	<57E1ED29-CB4A-455D-9C81-6F06C16CD5CB@ihug.co.nz>
	<c997m1dknqib1b36n9a73bnpn30ga9eprp@4ax.com>
Message-ID: <3F676D29-B4C1-4071-A9F4-ADF6B5E0CF59@ihug.co.nz>

[Switching to spambayes-dev to try and catch the attention of the  
right people]

[Tony]
>> I wonder if we ought to switch to defaulting to ZODB for 1.1a2;
>> I've been using it for over a year without any database problems

[Richie]
> +1 provided there aren't multi-process issues - as I understand it,  
> ZODB
> (without ZEO) isn't multi-process safe.  Which means it should be  
> fine for
> the POP3 proxy and (I assume?) the Outlook add-in, but maybe not  
> for the
> Unix side of things?

Good question.  ZODB will fail to open the database if another  
process already has it open.

Do those people using SpamBayes in the *nix world have two SpamBayes  
processes running at the same time?  I presume two reading processes  
would work fine with bsddb.

=Tony.Meyer

From skip at pobox.com  Sun Oct 30 16:30:57 2005
From: skip at pobox.com (skip@pobox.com)
Date: Sun, 30 Oct 2005 09:30:57 -0600
Subject: [spambayes-dev] Default storage backend
In-Reply-To: <3F676D29-B4C1-4071-A9F4-ADF6B5E0CF59@ihug.co.nz>
References: <6.2.3.4.0.20051026125848.0247e630@pop3.highstream.net>
	<hgevl1hb2101cs146b0fmnvo13ib91b5hd@4ax.com>
	<57E1ED29-CB4A-455D-9C81-6F06C16CD5CB@ihug.co.nz>
	<c997m1dknqib1b36n9a73bnpn30ga9eprp@4ax.com>
	<3F676D29-B4C1-4071-A9F4-ADF6B5E0CF59@ihug.co.nz>
Message-ID: <17252.59185.65428.738295@montanaro.dyndns.org>


    Tony> Do those people using SpamBayes in the *nix world have two
    Tony> SpamBayes processes running at the same time?  I presume two
    Tony> reading processes would work fine with bsddb.

I don't, but we encounter people on the spambayes from time-to-time that
mistakenly try accessing their training databases from multiple processes.
I think most of the time they simply don't realize they are doing that.  I
don't know if ZODB's refusal to run in those cases would provide better
clues that they are doing something wrong than the current corrupt database
does.  At any rate, it would eliminate that particular cause of corruption.

Skip

From jepler at unpythonic.net  Sun Oct 30 16:32:25 2005
From: jepler at unpythonic.net (jepler@unpythonic.net)
Date: Sun, 30 Oct 2005 09:32:25 -0600
Subject: [spambayes-dev] Default storage backend
In-Reply-To: <3F676D29-B4C1-4071-A9F4-ADF6B5E0CF59@ihug.co.nz>
References: <6.2.3.4.0.20051026125848.0247e630@pop3.highstream.net>
	<hgevl1hb2101cs146b0fmnvo13ib91b5hd@4ax.com>
	<57E1ED29-CB4A-455D-9C81-6F06C16CD5CB@ihug.co.nz>
	<c997m1dknqib1b36n9a73bnpn30ga9eprp@4ax.com>
	<3F676D29-B4C1-4071-A9F4-ADF6B5E0CF59@ihug.co.nz>
Message-ID: <20051030153225.GA7932@unpythonic.net>

On unix, my two setups are:
1. A higher-volume setting (ca 1000 messages/day) 

   pickle-format database

   patched sb_xmlrpcserver running all the time

   sb_client connects and sends requests to server to score or train
   messages, or to flush database to disk

2. A lower-volume setting (ca 100 messages/day)

   pickle-format database

   sb_filter used to score messages

   sb_mboxtrain used manually to train messages

In each case there's only a single writer to the database, unless
something goes wrong.  In one case there's only ever a single reader of
the database (with sb_xmlrpcserver) but in the second case there may be
many readers (sb_filter)

Jeff
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
Url : http://mail.python.org/pipermail/spambayes-dev/attachments/20051030/53efd633/attachment.pgp

From skip at pobox.com  Sun Oct 30 16:44:11 2005
From: skip at pobox.com (skip@pobox.com)
Date: Sun, 30 Oct 2005 09:44:11 -0600
Subject: [spambayes-dev] Default storage backend
In-Reply-To: <20051030153225.GA7932@unpythonic.net>
References: <6.2.3.4.0.20051026125848.0247e630@pop3.highstream.net>
	<hgevl1hb2101cs146b0fmnvo13ib91b5hd@4ax.com>
	<57E1ED29-CB4A-455D-9C81-6F06C16CD5CB@ihug.co.nz>
	<c997m1dknqib1b36n9a73bnpn30ga9eprp@4ax.com>
	<3F676D29-B4C1-4071-A9F4-ADF6B5E0CF59@ihug.co.nz>
	<20051030153225.GA7932@unpythonic.net>
Message-ID: <17252.59979.367365.305437@montanaro.dyndns.org>


    Jeff> On unix, my two setups are:
    Jeff> 1. A higher-volume setting (ca 1000 messages/day) 
    ...
    Jeff> 2. A lower-volume setting (ca 100 messages/day)
    ...

I use only a single writer and handle the high volume (1000-2000
messages/day is probably about right for me) in a couple ways.  First, I use
sb_bnclient.  Second, I always retrain from scratch, training into a new
database that is then mv'd into place.  Third, I export that database and
copy it to my higher performance mail server, where a similar setup uses
that database to skim off the most egregious spam first, allowing the rest
to be transferred to my laptop.

Skip


From tim.peters at gmail.com  Mon Oct 31 21:17:30 2005
From: tim.peters at gmail.com (Tim Peters)
Date: Mon, 31 Oct 2005 15:17:30 -0500
Subject: [spambayes-dev] Default storage backend
In-Reply-To: <3F676D29-B4C1-4071-A9F4-ADF6B5E0CF59@ihug.co.nz>
References: <6.2.3.4.0.20051026125848.0247e630@pop3.highstream.net>
	<hgevl1hb2101cs146b0fmnvo13ib91b5hd@4ax.com>
	<57E1ED29-CB4A-455D-9C81-6F06C16CD5CB@ihug.co.nz>
	<c997m1dknqib1b36n9a73bnpn30ga9eprp@4ax.com>
	<3F676D29-B4C1-4071-A9F4-ADF6B5E0CF59@ihug.co.nz>
Message-ID: <1f7befae0510311217t42c2897bx6511b0e661252c0d@mail.gmail.com>

[Tony Meyer]
>>> I wonder if we ought to switch to defaulting to ZODB for 1.1a2;
>>> I've been using it for over a year without any database problems

[Richie Hindle]
>> +1 provided there aren't multi-process issues - as I understand it,
>> ZODB (without ZEO) isn't multi-process safe.  Which means it should be
>> fine for the POP3 proxy and (I assume?) the Outlook add-in, but maybe
>> not for the Unix side of things?

[Tony]
> Good question.  ZODB will fail to open the database if another
> process already has it open.

That depends on how you open it.  The actual restriction is that, in
the absence of ZEO, at most one process at a time can open a
FileStorage with write access.  Any number of processes can open a
FileStorage simultaneously with read access.  The only code difference
is

    storage = FileStorage('Data.fs')  # read/write access

vs

    storage = FileStorage('Data.fs', read_only=True) # read-only access

With read-only access, you can load anything from the database, and
even change stuff (in memory), but you'll get a ReadOnlyError
exception if you try to _commit_ any changes.

> Do those people using SpamBayes in the *nix world have two SpamBayes
> processes running at the same time?  I presume two reading processes
> would work fine with bsddb.

Works fine with ZODB too, provided the readers ask for read_only mode.

With ZEO, multiple processes can write too.