http://permalink.gmane.org/gmane.mail.spam.crm114 hourly 1 1901-01-01T00:00+00:00 http://gmane.org/img/gmane-25t.png http://gmane.org http://permalink.gmane.org/gmane.mail.spam.crm114/9591 <pre> More likely to be a change/update in TRE, but we're both running the same version. - Bill ------------------------------------------------------------------------------ For Developers, A Lot Can Happen In A Second. Boundary is the first to Know...and Tell You. Monitor Your Applications in Ultra-Fine Resolution. Try it FREE! http://p.sf.net/sfu/Boundary-d2dvs2 </pre> wsy-MXLR4ItNKm8< at >public.gmane.org 2012-04-13T13:58:06 http://permalink.gmane.org/gmane.mail.spam.crm114/9590 <pre>On Thu, Apr 12, 2012 at 08:36:51AM -0400, wsy-MXLR4ItNKm8&lt; at &gt;public.gmane.org spake thusly: You mean my fix works, but it is really intended that I be able to put in raw spaces, right? Weird. Oops...I forgot to mention my crm version in the initial posting: # crm -v This is CRM114, version 20090423-BlameSteveJobs (TRE 0.8.0 (BSD)) Copyright 2001-2006 William S. Yerazunis This software is licensed under the GPL with ABSOLUTELY NO WARRANTY Yours is somewhat newer than mine. Any chance this is a bug which has been fixed? </pre> Tracy Reed 2012-04-12T18:21:31 http://permalink.gmane.org/gmane.mail.spam.crm114/9589 <pre> Only marginally unintentional parsing. But yeah, your solution is correct. (the "." is matching an arbitrary next character). Lemme look at the source code. Mailreaver:502 is: match &lt;fromend nomultiline&gt; (:w: :pm: :pat:) [:priolist:] /(.)(.+)/ So it *should* have picked up everything after the first character as the priority pattern. Worse: it works for me correctly: bash-4.2$ crm -v This is CRM114, version 20100106-BlameMichelson (TRE 0.8.0 (BSD)) Copyright 2001-2009 William S. Yerazunis This software is licensed under the GPL with ABSOLUTELY NO WARRANTY bash-4.2$ bash-4.2$ crm '-{ match &lt;fromend nomultiline&gt; (:a: :b: :c:) /(.)(.+)/; output /1st char: :*:b:\n/; output /Rest of chars: :*:c:\n/; liaf}' foo bar baz wugga happy go lucky 1st char: f Rest of chars: oo bar 1st char: b Rest of chars: az wugga 1st char: h Rest of chars: appy go lucky bash-4.2$ No problem. But sadly, no clue either. - Bill ------------------------------------------------------------------------------ For</pre> wsy-MXLR4ItNKm8< at >public.gmane.org 2012-04-12T12:36:51 http://permalink.gmane.org/gmane.mail.spam.crm114/9588 <pre>I have been successfully using priolist.mfp to match email addresses that I want to black/whitelist for ages. Now I want to block on a particular phrase such as: yahoo messenger online now which is from a particularly egregious sort of spam/scam my organization is receiving. I have tried putting the following combinations, none of which have worked: -yahoo messenger online now -yahoo\ messenger\ online\ now -yahoo\smessenger\sonline\snow -"yahoo messenger online now" and probably various others which I cannot now reproduce. I really expected the first one to work. I have finally stumbled upon a combination which does work: -yahoo.messenger.online.now Why would . work (I'm plenty familiar with regex and know it will match anything) but \s to match the spaces or simply raw spaces not work? Is there some unintentional parsing being done on the whitespace I put in which is breaking things? Thanks! </pre> Tracy Reed 2012-04-12T00:14:20 http://permalink.gmane.org/gmane.mail.spam.crm114/9587 <pre>Hi Bill, It's been a long time since I last wrote to the list. I'm interested in playing around with libcrm and the SVM classifier, but am a bit confused. 1. Where might I find the "best" version of the library? The link on the wiki seems pretty old. For me, "best" means reliable enough for doing personal spam filtering. 2. The HowTo mentions CRM114_SVM, but I see that there is also CRM114_LIBSVM and that the two choices follow different code paths in crm114_base. Which should I use, or why might I choose one vs. the other? Best Regards, Steve ------------------------------------------------------------------------------ This SF email is sponsosred by: Try Windows Azure free for 90 days Click Here http://p.sf.net/sfu/sfd2d-msazure </pre> Steve Pellegrin 2012-03-29T01:55:14 http://permalink.gmane.org/gmane.mail.spam.crm114/9586 <pre> I *think* there's supposed to be a space there, but it's been so long since I touched the code that I can't remember! I *think* that a column 1 '#' can be used as a comment. I am pretty sure that will NOT work. I don't think so. It's reading a line at a time. But it's always possible that there's a bug. - Bill ------------------------------------------------------------------------------ This SF email is sponsosred by: Try Windows Azure free for 90 days Click Here http://p.sf.net/sfu/sfd2d-msazure </pre> wsy-MXLR4ItNKm8< at >public.gmane.org 2012-03-26T17:52:58 http://permalink.gmane.org/gmane.mail.spam.crm114/9585 <pre> Well, in newer versions, the whitelist/blacklist/priolist is deprecated into JUST one: priolist. A priolist file looks like this: a + or a -, then a space, then a pattern. A leading "+" means "whitelist", a leading "-" means "blacklist". As a regex, it looks like this: (+|-) pattern For example, to whitelist my lawer, my doctor, and blacklist my ex-girlfriend: + my_lawyer-gexy9bi/sWtBDgjK7y7TUQ&lt; at &gt;public.gmane.org + my_doctor-rimE70MFspc&lt; at &gt;public.gmane.org - my_ex_girlfriend-XJeLq6sSUTW+dUnouKm4lg&lt; at &gt;public.gmane.org + my_new_girlfriend-Tpms0nXwIR954TAoqtyWWQ&lt; at &gt;public.gmane.org So, to whitelist or blacklist anyone, just add them (or their domain) to the priolist. Note that the pattern just needs to *match*. It can match anywhere in the entire message, so if my ex-girlfriend mentiones my doctor's email address, it will still come through. Note also that the priolist is executed in strict order. Thus, if my ex-girlfriend mentions my doctor, it comes through (the doctor outranks my ex-girlfriend) but if sh</pre> wsy-MXLR4ItNKm8< at >public.gmane.org 2012-03-26T12:16:31 http://permalink.gmane.org/gmane.mail.spam.crm114/9584 <pre>I see whitelisting mentioned in this thread, so I thought to offer what I find a very useful and trouble-free whitelist system based on shell scripts. (My current understanding is that one crm-114 setup includes whitelisting, and one does not.) I have this system integrated with my maildrop .mailfilter, and call it when/as needed from the keyboard in mutt to add senders to my .whitelist. http://impressive.net/people/gerald/2000/12/spam-filtering.html </pre> Eric d'Halibut 2012-03-23T03:25:03 http://permalink.gmane.org/gmane.mail.spam.crm114/9583 <pre>martin-Gf0sdYSFR0SsTnJN9+BGXg&lt; at &gt;public.gmane.org said: Ok, /me is feeling rather embarrased now. I just checked my entire setup, and it looks like something hasn't been working since some time in May last year, at least based on the date on my spam.css and nonspam.css files. I think the problem is in the dovecot antispam plugin, am investigating. Martin </pre> Martin Lucina 2012-03-22T15:12:29 http://permalink.gmane.org/gmane.mail.spam.crm114/9582 <pre>wsy-MXLR4ItNKm8&lt; at &gt;public.gmane.org said: I'm currently using the default from mailfilter.cf: clf: /osb unique microgroom/ Do I have anything to gain from SVM? (Speaking as a user, and slightly distracted at the moment, so I haven't read up on it) It's definitely 0x20. Same here. I'll see what happens when I retrain and if the problem persists will just add the Koreans to the whitelist. Martin </pre> Martin Lucina 2012-03-22T14:50:21 http://permalink.gmane.org/gmane.mail.spam.crm114/9581 <pre> Only if you want the SVM classifier. The base classifer (Markovian) is unchanged. Go into EMACS or another byte-accurate editor (or a hex editor) and check! It can be quite important. (i.e. is there really a hex 0x20 between each word? Or does the Hangul representation you get have a "space" in the glyphset that is NOT 0x20, and it gets used because that way you don't have to change glyphsets twice for every word. Yes, there is. That's what the whitelist is for. Although one might regard it as "cheating", I am a firm believer in "whatever gets you through the night." - Bill Yerazunis ------------------------------------------------------------------------------ This SF email is sponsosred by: Try Windows Azure free for 90 days Click Here http://p.sf.net/sfu/sfd2d-msazure </pre> wsy-MXLR4ItNKm8< at >public.gmane.org 2012-03-22T14:32:52 http://permalink.gmane.org/gmane.mail.spam.crm114/9580 <pre>Hi Bill, wsy-MXLR4ItNKm8&lt; at &gt;public.gmane.org said: Added the Cc: to the list back in, have now received email from the system confirming that I'm subscribed. That never happened for me. I kept getting a trickle of both SPAM and non-SPAM mail going into the "UNSURE" folder. Should I also upgrade to the latest codebase? If so, which one? I'm currently using 20090807-BlameThorstenAndJenny (TRE 0.7.5 (LGPL)). Assuming the various different Korean character sets all have what renders as "space" on my screen in the same place as "ascii space", which is a pretty safe assumption, then yes. Hangul is alphabet-based rather than glyph-based, the only reason it looks superficially similar to Japanese is that they use the trick of combining multiple charaters to form a single glyph. Alternatively, given that the volume of mail I get in Korean is quite low, is there a way to tell the system to just pass through mail from certain senders, completely ignoring it for training purposes? Replaying email will be a bit h</pre> Martin Lucina 2012-03-22T14:24:53 http://permalink.gmane.org/gmane.mail.spam.crm114/9579 <pre> Yes, such a toolkit (actually, a complete package) exists. It's actually a commercial product; the core bits of it are libcrm114 (which is LGPLed) but you pay for the rest. Oh, and it's all in Japanese. I don't think there's an English-language port of the GUI nor of the documentation. Your choice: 1) build a toolkit yourself. or 2) learn Japanese - Bill Yerazunis ------------------------------------------------------------------------------ Keep Your Developer Skills Current with LearnDevNow! The most comprehensive online learning library for Microsoft developers is just $99.99! Visual Studio, SharePoint, SQL - plus HTML5, CSS3, MVC3, Metro Style Apps, more. Free future releases when you subscribe now! http://p.sf.net/sfu/learndevnow-d2d </pre> wsy-MXLR4ItNKm8< at >public.gmane.org 2012-02-28T13:24:04 http://permalink.gmane.org/gmane.mail.spam.crm114/9578 <pre>I was discussing a security issue with my new employer today, about scanning backups of servers that should not have confidential data on them for precisely such data. In the short term, scanning the email would do, especially if attachmants can be scanned. And I thought of CRM114 for the task, instead of the very slow and painful tools that are often used now. Is there a toolkit for such scanning? I'd much prefer to avoid take on the full integration project, but if anyone's already got such a toolkit assembled, even if it's a commercial toolkit, I'd love to review it for use at my new workplace. ------------------------------------------------------------------------------ Keep Your Developer Skills Current with LearnDevNow! The most comprehensive online learning library for Microsoft developers is just $99.99! Visual Studio, SharePoint, SQL - plus HTML5, CSS3, MVC3, Metro Style Apps, more. Free future releases when you subscribe now! http://p.sf.net/sfu/learndevnow-d2d____________________________________</pre> Nico Kadel-Garcia 2012-02-28T03:22:22 http://permalink.gmane.org/gmane.mail.spam.crm114/9577 <pre>Hi, All. I'm back to looking at CRM114, and I decided to update some of the documentation, working from "the (possibly) slightly unstable latest mainline version" -- thanks for the continued good work, and I hope the updates prove useful. This is the first. /Jskud --- crm114-20120205-ORIG/CLASSIFY_DETAILS.txt2009-09-11 11:25:57.000000000 -0700 +++ crm114-20120205/CLASSIFY_DETAILS.txt2012-02-06 07:49:00.000000000 -0800 &lt; at &gt;&lt; at &gt; -18,7 +18,8 &lt; at &gt;&lt; at &gt; The current distribution builds in this set of classifiers. The classifiers are: -1) SBPH Markovian (the default) This is an extension of Bayesian +1) SBPH Markovian (the default) - This classifier uses + Sparse Binary Polynomial Hashing (SBPH), an extension of Bayesian classification, mapping features in the input text into a Markov Random Field. This turns each token in the input into 2^(N-1) features, which gives high accuracy but at high computation &lt; at &gt;&lt; at &gt; -52,7 +53,7 &lt; at &gt;&lt; at &gt; other classifiers. It _will_ work against binary files, though, which n</pre> Jskud.CRM114-hFAK3oOPH3QAvxtiuMwx3w< at >public.gmane.org 2012-02-07T06:27:59 http://permalink.gmane.org/gmane.mail.spam.crm114/9576 <pre>Hi, All. I'm back to looking at CRM114, and I decided to update some of the documentation, working from "the (possibly) slightly unstable latest mainline version" -- thanks for the continued good work, and I hope the updates prove useful. This is the second of two. I mostly reworked the formatting to make it consistent, and made the step titles consistent as well, fixing a few obvious typos in the process. /Jskud --- crm114-20120205-ORIG/CRM114_Mailfilter_HOWTO.txt2009-09-11 11:25:57.000000000 -0700 +++ crm114-20120205/CRM114_Mailfilter_HOWTO.txt2012-02-06 19:52:36.000000000 -0800 &lt; at &gt;&lt; at &gt; -7,7 +7,7 &lt; at &gt;&lt; at &gt; The CRM114 &amp; Mailfilter HOWTO -Bill Yerazunis, 2003-09-18 -(last update 2009-03-02) +(last update 2012-02-06) This is the CRM114 Mailfilter HOWTO. It describes how to set up CRM114 &lt; at &gt;&lt; at &gt; -31,7 +31,7 &lt; at &gt;&lt; at &gt; ---------------------------------------------------------- -That said, I hope CRM114, Mailreaver, and Mailreaver is useful to you; +That said, I hope CRM114, Mailfilter, and Mailre</pre> Jskud.CRM114-hFAK3oOPH3QAvxtiuMwx3w< at >public.gmane.org 2012-02-07T06:31:46 http://permalink.gmane.org/gmane.mail.spam.crm114/9575 <pre> Hi Khalid, This would explain a lot. Certainly I can package a shared library, preferrably in addition to the static library. The Makefile would have to produce libcrm114.so which I would package as libcrm114.so.1.0.0 or similar (according to the actual version) and add a symbolic link libcrm114.so in the package. This means we would need proper versioning so that several versions could co-exist in the future. In an ideal world the upstream source should be named to show the version; libcrm114-1.0.0.tar.gz as an example. The major version number is to be counted up when a new version is no longer backward compatible to previous versions. The middle number counts up for every release introducing a backward compatible change in the API and the minor number is for bugfix releases. I really would like to see this done in the Makefile upstream rather than adding a patch to my spec file. Bill: any chance that you consider to change your Makefile to produce an additional dynamic library? Dynami</pre> Thomas Spahni 2012-01-11T16:33:50 http://permalink.gmane.org/gmane.mail.spam.crm114/9574 <pre> Hello Khalid Something is broken, either with my package for libcrm114.a or something else. When I set up pecl-crm114 I get this during the configure step: checking for grep that handles long lines and -e... /usr/bin/grep checking for egrep... /usr/bin/grep -E checking for a sed that does not truncate output... /usr/bin/sed checking for cc... cc checking whether the C compiler works... yes checking for C compiler default output file name... a.out checking for suffix of executables... checking whether we are cross compiling... no checking for suffix of object files... o checking whether we are using the GNU C compiler... yes checking whether cc accepts -g... yes checking for cc option to accept ISO C89... none needed checking how to run the C preprocessor... cc -E checking for icc... no checking for suncc... no checking whether cc understands -c and -o together... yes checking for system library directory... lib checking if compiler supports -R... no checking if compiler supports -Wl,-rpath,... yes check</pre> Thomas Spahni 2012-01-08T18:36:37 http://permalink.gmane.org/gmane.mail.spam.crm114/9573 <pre> Hi all I packaged libcrm114 as crm114-devel-static for the openSUSE distribution. Package details are here: https://build.opensuse.org/package/show?package=crm114-devel-static&amp;project=home%3Avodoo and the rpm's may be downloaded from the archive here: http://download.opensuse.org/repositories/home:/vodoo/openSUSE_11.3/ http://download.opensuse.org/repositories/home:/vodoo/openSUSE_11.4/ http://download.opensuse.org/repositories/home:/vodoo/openSUSE_12.1/ I added a man page with contents borrowed from HOWTO.txt. The file layout of the package is using these locations: /usr/lib/libcrm114.a /usr/include/crm114_*.h /usr/share/doc/packages/crm114-devel-static Replace /usr/lib by /usr/lib64 for the x86_64 packages. Anyone may feel free to use my spec file as a starting point for other distributions. Comments are welcome. Khalid Ahsein: Can you please adapt your config.m4 to make pecl-crm114-0.9.1.tgz link with the installed libcrm114 library if available? Then I could package it was well. Thank you</pre> Thomas Spahni 2012-01-06T09:43:10 http://permalink.gmane.org/gmane.mail.spam.crm114/9572 <pre>wsy-MXLR4ItNKm8&lt; at &gt;public.gmane.org replied to my note: &lt;&gt; &gt; So, setting up crm114 for someone else led me to a seriously overdue &lt;&gt; &gt; cleanup of my own rewrites.mfp file. I decided that it's probably &lt;&gt; &gt; just doing string replacements to normalize the data, so it should &lt;&gt; &gt; deal just fine with an IPv6 colon separated address. &lt;&gt; &gt; &lt;&gt; &gt; So far, it seems to do so just fine, so, Bill, I think you can call &lt;&gt; &gt; crm114 "IPv6 compliant" &lt;&gt; &lt;&gt; So - all those colons didn't get the knickers in a twist? :-) Apparently not &lt;grin&gt; &lt;&gt; I guess :*: and :+: don't appear much in IP6 addresses - or, if they &lt;&gt; do, then whatever follows it until the next colon doesn't map &lt;&gt; to a preexisting variable. Nope, those two strings will never appear in an address. A v6 address is comprised of eight units of four hex digits separated by colons: 2001:0470:1f07:61e:52e5:49ff:fe55:6731 Leading zeros can be dropped and units that are all zero can be dropped, ONCE, in an address: 2001:470:1f07:61e::3(global address in the </pre> R A Lichtensteiger 2012-01-05T20:25:18 http://permalink.gmane.org/gmane.mail.spam.crm114/9571 <pre> So - all those colons didn't get the knickers in a twist? :-) I guess :*: and :+: don't appear much in IP6 addresses - or, if they do, then whatever follows it until the next colon doesn't map to a preexisting variable. That's one bug in the syntax; I wrote CRM114's expansion syntax fifteen years ago and had no idea about IPV6. Then again, RFC 2460 only came out in 1998 which was about the same time. That's good to know. :) - Crash ------------------------------------------------------------------------------ Ridiculously easy VDI. With Citrix VDI-in-a-Box, you don't need a complex infrastructure or vast IT resources to deliver seamless, secure access to virtual desktops. With this all-in-one solution, easily deploy virtual desktops for less than the cost of PCs and save 60% on VDI infrastructure costs. Try it free! http://p.sf.net/sfu/Citrix-VDIinabox </pre> wsy-MXLR4ItNKm8< at >public.gmane.org 2012-01-05T19:57:43 Search the mailing list at Gmane query http://search.gmane.org/?group=$group=gmane.mail.spam.crm114