Date: Sat, 21 Aug 2004 22:23:15 +0200 From: Radim Kolar <hsn@netmag.cz> To: FreeBSD-gnats-submit@FreeBSD.org Subject: ports/70801: [NEW PORT] mail/crm114-devel: An Markov based SpamFilter Message-ID: <E1BycOR-000CFR-Nc@sanatana.dharma> Resent-Message-ID: <200408212040.i7LKeZL6084320@freefall.freebsd.org>
next in thread | raw e-mail | index | archive | help
>Number: 70801 >Category: ports >Synopsis: [NEW PORT] mail/crm114-devel: An Markov based SpamFilter >Confidential: no >Severity: non-critical >Priority: low >Responsible: freebsd-ports-bugs >State: open >Quarter: >Keywords: >Date-Required: >Class: change-request >Submitter-Id: current-users >Arrival-Date: Sat Aug 21 20:40:35 GMT 2004 >Closed-Date: >Last-Modified: >Originator: Radim Kolar >Release: FreeBSD 5.2-CURRENT i386 >Organization: >Environment: System: FreeBSD sanatana.dharma 5.2-CURRENT FreeBSD 5.2-CURRENT #0: Thu Aug 19 19:04:54 CEST 2004 >Description: This is an development version of crm114. It has manpages. This version is quite stable now. Major improvements: new classification methods added! CRM114 is a system to examine incoming e-mail, system log streams, data files or other data streams, and to sort, filter, or alter the incoming files or data streams according to the user's wildest desires. Criteria for categorization of data can be by satisfaction of regexes, by sparse binary polynomial matching with a Bayesian Chain Rule evaluator, or by other means. Accuracy of the SBPH/BCR classifier has been seen in excess of 99 per cent, for 1/4 megabyte of learning text. In other words, CRM114 learns, and it learns fast. WWW: http://crm114.sourceforge.net/ Generated with FreeBSD Port Tools 0.63 >How-To-Repeat: >Fix: --- crm114-devel-20040816_3.shar begins here --- # This is a shell archive. Save it in a file, remove anything before # this line, and then unpack it by entering "sh file". Note, it may # create directories; files and directories will be owned by you and # have default permissions. # # This archive contains: # # crm114-devel # crm114-devel/Makefile # crm114-devel/README.html # crm114-devel/distinfo # crm114-devel/pkg-descr # crm114-devel/pkg-plist # crm114-devel/files # crm114-devel/files/patch-msync # crm114-devel/files/patch-gcc34 # echo c - crm114-devel mkdir -p crm114-devel > /dev/null 2>&1 echo x - crm114-devel/Makefile sed 's/^X//' >crm114-devel/Makefile << 'END-of-crm114-devel/Makefile' X# New ports collection makefile for: crm114 X# Date created: 23 February 2004 X# Whom: Meno Abels <meno.abels@adviser.com> X# X# $FreeBSD$ X# X XPORTNAME= crm114 XPORTVERSION= 20040816 XPORTREVISION= 3 X# WARNING change versionname in CRM114RELEASENAME too. X# portlint didn't like this variable at this place. X# So I have to put it down in this file. XCATEGORIES= mail XMASTER_SITES= ${MASTER_SITE_SOURCEFORGE} XMASTER_SITE_SUBDIR= crm114 XPKGNAMESUFFIX= -devel X# Oliver Eikemeier recommended the following construction X# to avoid the reduncany of the CRM114RELEASENAME and .src and X# .css extention. XDISTNAME= ${PORTNAME}-${PORTVERSION}.${CRM114RELEASENAME}-auto.${PORTREVISION} X XMAINTAINER= ports@freebsd.org XCOMMENT= An Markov based SpamFilter X XLIB_DEPENDS= tre.5:${PORTSDIR}/textproc/libtre X# there is no other then formail on this planet XRUN_DEPENDS= procmail:${PORTSDIR}/mail/procmail XCONFLICTS= crm114-2* XCRM114RELEASENAME=BlameClockworkOrange X XUSE_GETTEXT= yes XUSE_ICONV= yes XGNU_CONFIGURE= yes XUSE_REINPLACE= yes XUSE_BZIP2= yes XPLIST= ${WRKDIR}/plist XMAN1= crm.1 crm114.1 cssdiff.1 cssmerge.1 cssutil.1 X XCONFIGURE_ENV+= CPPFLAGS=-I${LOCALBASE}/include XCONFIGURE_ENV+= LDFLAGS=-L${LOCALBASE}/lib X X.ifdef WITH_OPTIMIZED_CFLAGS XCFLAGS+= -Wuninitialized -ffast-math -finline-functions -fomit-frame-pointer -fexpensive-optimizations -fforce-mem -fforce-addr -O3 X.endif X X# Peter Jeremy recommended the use of b64decode and md5 20040302 Xpre-build: X ${REINPLACE_CMD} 's|:mime_decoder: /mewdecode/|:mime_decoder: /b64decode -pr/|' ${WRKSRC}/mailfilter/mailfilter.cf X ${REINPLACE_CMD} 's|/md5sum/|/md5 -r/|' ${WRKSRC}/mailfilter/mailfilter.crm X Xpre-everything:: X.ifndef WITH_OPTIMIZED_CFLAGS X @${ECHO_MSG} "===> Define WITH_OPTIMIZED_CFLAGS for compiling with optimalization" X.endif X Xpre-install: X ${CAT} ${PKGDIR}/pkg-plist > ${PLIST} X ${FIND} ${WRKSRC}/examples -name *.mfp* -execdir ${ECHO_CMD} "share/doc/crm114/examples/crmfilter/{}" \; >> ${PLIST} X ${ECHO_CMD} "@dirrm share/doc/crm114/examples/crmfilter" >> ${PLIST} X ${FIND} ${WRKSRC}/tests -name *.crm -execdir ${ECHO_CMD} "share/doc/crm114/examples/tests/{}" \; >> ${PLIST} X ${ECHO_CMD} "@dirrm share/doc/crm114/examples/tests" >> ${PLIST} X ${FIND} ${WRKSRC}/examples \( -name *.crm -or -name *.sh -or -name *.dat -or -name *.log \) -execdir ${ECHO_CMD} "share/doc/crm114/examples/{}" \; >> ${PLIST} X ${ECHO_CMD} "@dirrm share/doc/crm114/examples" >> ${PLIST} X ${FIND} ${WRKSRC}/man \( -name *.azm -or -name \*.html -or -name *.txt \) -execdir ${ECHO_CMD} "share/doc/crm114/{}" \; >> ${PLIST} X ${FIND} ${WRKSRC}/docs \( -name *.txt -or -name *.recipe \) -execdir ${ECHO_CMD} "share/doc/crm114/{}" \; >> ${PLIST} X.for f in AUTHORS COPYING ChangeLog HACKING NEWS README README.1st TODO X ${ECHO_CMD} "share/doc/crm114/"$f >> ${PLIST} X.endfor X ${ECHO_CMD} "@dirrm share/doc/crm114" >> ${PLIST} X X.include <bsd.port.mk> END-of-crm114-devel/Makefile echo x - crm114-devel/README.html sed 's/^X//' >crm114-devel/README.html << 'END-of-crm114-devel/README.html' X<!DOCTYPE html PUBLIC "-//IETF//DTD HTML 2.0//EN"> X X<html> X <head> X <title>The FreeBSD Ports Collection (myports/crm114-devel)</title> X </head> X X <body> X <h1>The FreeBSD Ports Collection (myports/crm114-devel)</h1> X <hr> X X <p>You are now in the directory for the port "myports/crm114-devel".</p> X X <p>The package name of this port is "crm114-devel-20040419_1".</p> X X <p>This is the one-line description for this port:</p> X <hr> X X <p> XAn Markov based SpamFilter X </p> X <hr> X X <p>Please read the "<a href="pkg-descr">description file</a>" for a X longer description and/or visit the <a href="http://crm114.sourceforge.net/">web site</a> for futher informations.</p> X X <p>If needed, you may contact the <a href= X "mailto:hsn@netmag.cz">maintainer</a> of this port or the port <a href= X "mailto:ports@FreeBSD.org">mailing-list</a>.</p> X X <p>This port requires package(s) "" to build.</p> X X <p>This port requires package(s) "" to run.</p> X X <p>Go to the <a href="../../README.html">top of the ports tree</a> for X a summary on how to use the ports collection.</p> X <hr> X X <p><a href="../README.html">Go up one level</a> | <a href= X "../../README.html">Go to top of ports tree</a></p> X </body> X</html> X END-of-crm114-devel/README.html echo x - crm114-devel/distinfo sed 's/^X//' >crm114-devel/distinfo << 'END-of-crm114-devel/distinfo' XMD5 (crm114-20040816.BlameClockworkOrange-auto.3.tar.bz2) = c77d90692731b64de5b8583b3289e1b7 XSIZE (crm114-20040816.BlameClockworkOrange-auto.3.tar.bz2) = 332797 END-of-crm114-devel/distinfo echo x - crm114-devel/pkg-descr sed 's/^X//' >crm114-devel/pkg-descr << 'END-of-crm114-devel/pkg-descr' XCRM114 is a system to examine incoming e-mail, system log streams, Xdata files or other data streams, and to sort, filter, or alter the Xincoming files or data streams according to the user's wildest Xdesires. Criteria for categorization of data can be by satisfaction Xof regexes, by sparse binary polynomial matching with a Bayesian XChain Rule evaluator, or by other means. Accuracy of the SBPH/BCR Xclassifier has been seen in excess of 99 per cent, for 1/4 megabyte Xof learning text. In other words, CRM114 learns, and it learns fast. X XWWW: http://crm114.sourceforge.net/ END-of-crm114-devel/pkg-descr echo x - crm114-devel/pkg-plist sed 's/^X//' >crm114-devel/pkg-plist << 'END-of-crm114-devel/pkg-plist' Xbin/crm Xbin/crm114 Xbin/cssdiff Xbin/cssmerge Xbin/cssutil Xlibexec/crm114/mailfilter.cf Xlibexec/crm114/mailfilter.crm Xlibexec/crm114/rewriteutil.crm Xlibexec/crm114/classifymail.crm X@dirrm libexec/crm114 END-of-crm114-devel/pkg-plist echo c - crm114-devel/files mkdir -p crm114-devel/files > /dev/null 2>&1 echo x - crm114-devel/files/patch-msync sed 's/^X//' >crm114-devel/files/patch-msync << 'END-of-crm114-devel/files/patch-msync' X--- src/crm_markovian.c.orig Wed Apr 21 05:38:44 2004 X+++ src/crm_markovian.c Thu May 6 11:28:17 2004 X@@ -636,7 +636,6 @@ X learn_end_regex_loop: X regcomp_failed: X // and remember to let go of the mmap and the pattern bufffer X- msync (hashes, hfsize * sizeof (FEATUREBUCKET_TYPE), MS_SYNC); X munmap (hashes, hfsize * sizeof (FEATUREBUCKET_TYPE)); X X // Because mmap/munmap doesn't set atime, nor set the "modified" END-of-crm114-devel/files/patch-msync echo x - crm114-devel/files/patch-gcc34 sed 's/^X//' >crm114-devel/files/patch-gcc34 << 'END-of-crm114-devel/files/patch-gcc34' X--- src/crm_var_hash_table.c.orig Sat Aug 21 21:31:16 2004 X+++ src/crm_var_hash_table.c Sat Aug 21 21:33:02 2004 X@@ -580,7 +580,7 @@ X if ( oldend < newstart) X { X // nothing to be done here - not overlapping X- goto end_of_tests; X+ continue; X }; X X // dead zone entirely after current var X@@ -591,7 +591,7 @@ X if ( newend < oldstart ) X { X // nothing to be done here - not overlapping X- goto end_of_tests; X+ continue; X }; X X // If we get this far, the dead zone in some way overlaps with X@@ -630,7 +630,7 @@ X // newstart is a "good" char, but since oldend is X // noninclusive, this is right. X oldend = newstart; X- goto end_of_tests; X+ continue; X }; X X // dead zone overlaps back of current var; trim the front off X@@ -647,7 +647,7 @@ X // Newend is the first char that ISN'T in the var, so this X // is correct. X oldstart = newend ; X- goto end_of_tests; X+ continue; X }; X // dead zone split by current var - the dead zone is actually X // split into two distinct pieces. In this case, we need to X@@ -677,7 +677,6 @@ X crm_compress_tdw_section (oldtext, newend, oldend) X + crm_compress_tdw_section(oldtext, oldstart, newstart)); X } X- end_of_tests: X }; X }; X // END-of-crm114-devel/files/patch-gcc34 exit --- crm114-devel-20040816_3.shar ends here --- >Release-Note: >Audit-Trail: >Unformatted:
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?E1BycOR-000CFR-Nc>