FuzzyOcr is a SpamAssassin plugin which uses an Optical Character

Recognition (OCR) program to scan image attachments for spammy content.

It works!

9 files changed, 191 insertions, 0 deletions

diff --git a/mail/Makefile b/mail/Makefile
index bb569dfa9e6..71a8d0d21a2 100644
--- a/mail/Makefile
+++ b/mail/Makefile
@@ -288,6 +288,7 @@
     SUBDIR += p5-Email-Simple-Creator
     SUBDIR += p5-Email-Valid
     SUBDIR += p5-Email-Valid-Loose
+    SUBDIR += p5-FuzzyOcr
     SUBDIR += p5-IMAP-Admin
     SUBDIR += p5-IMAP-Client
     SUBDIR += p5-IMAP-Sieve
diff --git a/mail/p5-FuzzyOcr-devel/Makefile b/mail/p5-FuzzyOcr-devel/Makefile
new file mode 100644
index 00000000000..73ceb3f9aca
--- /dev/null
+++ b/mail/p5-FuzzyOcr-devel/Makefile
@@ -0,0 +1,75 @@
+# New ports collection makefile for:	FuzzyOcr
+# Date Created:				2006-11-03
+# Whom:					Shaun Amott <shaun@FreeBSD.org>
+#
+# $FreeBSD$
+#
+
+PORTNAME=	FuzzyOcr
+DISTVERSION=	2.3rc1
+CATEGORIES=	mail perl5
+MASTER_SITES=	http://users.own-hero.net/~decoder/${PORTNAME:L}/	\
+		http://mirror.inerd.com/FreeBSD/distfiles/${PORTNAME}/
+PKGNAMEPREFIX=	p5-
+DISTNAME=	${PORTNAME:L}-${DISTVERSION}
+
+MAINTAINER=	shaun@FreeBSD.org
+COMMENT=	Plugin for SpamAssassin which scans image attachments for spam
+
+RUN_DEPENDS=	giftopnm:${PORTSDIR}/graphics/netpbm			\
+		import:${PORTSDIR}/graphics/ImageMagick			\
+		gocr:${PORTSDIR}/graphics/gocr				\
+		gif2ps:${PORTSDIR}/graphics/libungif			\
+		spamassassin:${PORTSDIR}/mail/p5-Mail-SpamAssassin	\
+		${SITE_PERL}/${PERL_ARCH}/Digest/MD5.pm:${PORTSDIR}/security/p5-Digest-MD5	\
+		${SITE_PERL}/${PERL_ARCH}/String/Approx.pm:${PORTSDIR}/devel/p5-String-Approx
+
+USE_PERL5=	yes
+NO_BUILD=	yes
+
+WRKSRC=		${WRKDIR}
+
+PORTDOCS=	FAQ samples
+PORTDOCS_F=	FAQ samples/README samples/animated-gif.eml		\
+		samples/corrupted-gif.eml samples/jpeg.eml samples/png.eml
+
+SUB_FILES=	pkg-message
+
+#-----------------------------------------------------------------------
+# The plugin is installed twice: once where it should go, and once where
+# it actually works from. This will hopefully be fixed at some point.
+#-----------------------------------------------------------------------
+
+SA_PLUGINDIR?=	Mail/SpamAssassin/Plugin
+SA_CONFIGDIR?=	etc/mail/spamassassin
+
+PLIST_FILES=	${SA_CONFIGDIR}/FuzzyOcr.pm				\
+		%%SITE_PERL%%/${SA_PLUGINDIR}/FuzzyOcr.pm		\
+		%%EXAMPLESDIR%%/FuzzyOcr.cf %%EXAMPLESDIR%%/FuzzyOcr.words
+PLIST_DIRS=	%%EXAMPLESDIR%%
+
+post-patch:
+	@${REINPLACE_CMD} -e 's,/usr/bin,${LOCALBASE}/bin,g'		\
+	                  -e 's,/bin/grep,/usr/bin/grep,g'		\
+	                  -e 's,/etc/mail,${LOCALBASE}/etc/mail,g'	\
+	                  ${WRKSRC}/FuzzyOcr.??
+	@${REINPLACE_CMD} -e 's,$$ENV{HOME},($$ENV{HOME} ? $$ENV{HOME} : "/nonexistent"),g' \
+	                 ${WRKSRC}/FuzzyOcr.pm
+
+do-install:
+	@${MKDIR} ${EXAMPLESDIR}
+	${INSTALL_DATA} ${WRKSRC}/FuzzyOcr.pm ${SITE_PERL}/${SA_PLUGINDIR}
+	${INSTALL_DATA} ${WRKSRC}/FuzzyOcr.pm ${PREFIX}/${SA_CONFIGDIR}
+	${INSTALL_DATA} ${WRKSRC}/FuzzyOcr.cf ${EXAMPLESDIR}
+	${INSTALL_DATA} ${WRKSRC}/FuzzyOcr.words.sample ${EXAMPLESDIR}/FuzzyOcr.words
+
+post-install:
+.if !defined(NOPORTDOCS)
+	@${MKDIR} ${DOCSDIR}/samples
+.  for f in ${PORTDOCS_F}
+	${INSTALL_DATA} ${WRKSRC}/${f} ${DOCSDIR}/${f}
+.  endfor
+.endif
+	@${CAT} ${PKGMESSAGE}
+
+.include <bsd.port.mk>
diff --git a/mail/p5-FuzzyOcr-devel/distinfo b/mail/p5-FuzzyOcr-devel/distinfo
new file mode 100644
index 00000000000..a0ae182da04
--- /dev/null
+++ b/mail/p5-FuzzyOcr-devel/distinfo
@@ -0,0 +1,3 @@
+MD5 (fuzzyocr-2.3rc1.tar.gz) = 43879c3f30045b0067d21f9e44b90697
+SHA256 (fuzzyocr-2.3rc1.tar.gz) = bdd8c3524e6445ea79968497c84a9d5e58491ccab9e05aa0784179bec8c24131
+SIZE (fuzzyocr-2.3rc1.tar.gz) = 75295
diff --git a/mail/p5-FuzzyOcr-devel/files/pkg-message.in b/mail/p5-FuzzyOcr-devel/files/pkg-message.in
new file mode 100644
index 00000000000..11e9ed3223d
--- /dev/null
+++ b/mail/p5-FuzzyOcr-devel/files/pkg-message.in
@@ -0,0 +1,5 @@
+------------------------------------------------------------------------
+ To complete the install of FuzzyOcr, copy FuzzyOcr.cf and
+ FuzzyOcr.words from %%EXAMPLESDIR%% into your
+ SpamAssassin configuration directory.
+------------------------------------------------------------------------
diff --git a/mail/p5-FuzzyOcr-devel/pkg-descr b/mail/p5-FuzzyOcr-devel/pkg-descr
new file mode 100644
index 00000000000..a7069d137cb
--- /dev/null
+++ b/mail/p5-FuzzyOcr-devel/pkg-descr
@@ -0,0 +1,12 @@
+FuzzyOcr is a plugin for SpamAssassin which scans image attachments for
+spam keywords, using an Optical Character Recognition program.
+
+This plugin can be used to detect spam that puts all the real spam
+content in an attached image. The mail itself only random text and
+random html, without any URLs or identifiable information.
+
+It is also able to do approximate matches on words, so errors in
+recognition or attempts to obfuscate the text inside the image will not
+cause the detection to fail.
+
+WWW: http://wiki.apache.org/spamassassin/FuzzyOcrPlugin
diff --git a/mail/p5-FuzzyOcr/Makefile b/mail/p5-FuzzyOcr/Makefile
new file mode 100644
index 00000000000..73ceb3f9aca
--- /dev/null
+++ b/mail/p5-FuzzyOcr/Makefile
@@ -0,0 +1,75 @@
+# New ports collection makefile for:	FuzzyOcr
+# Date Created:				2006-11-03
+# Whom:					Shaun Amott <shaun@FreeBSD.org>
+#
+# $FreeBSD$
+#
+
+PORTNAME=	FuzzyOcr
+DISTVERSION=	2.3rc1
+CATEGORIES=	mail perl5
+MASTER_SITES=	http://users.own-hero.net/~decoder/${PORTNAME:L}/	\
+		http://mirror.inerd.com/FreeBSD/distfiles/${PORTNAME}/
+PKGNAMEPREFIX=	p5-
+DISTNAME=	${PORTNAME:L}-${DISTVERSION}
+
+MAINTAINER=	shaun@FreeBSD.org
+COMMENT=	Plugin for SpamAssassin which scans image attachments for spam
+
+RUN_DEPENDS=	giftopnm:${PORTSDIR}/graphics/netpbm			\
+		import:${PORTSDIR}/graphics/ImageMagick			\
+		gocr:${PORTSDIR}/graphics/gocr				\
+		gif2ps:${PORTSDIR}/graphics/libungif			\
+		spamassassin:${PORTSDIR}/mail/p5-Mail-SpamAssassin	\
+		${SITE_PERL}/${PERL_ARCH}/Digest/MD5.pm:${PORTSDIR}/security/p5-Digest-MD5	\
+		${SITE_PERL}/${PERL_ARCH}/String/Approx.pm:${PORTSDIR}/devel/p5-String-Approx
+
+USE_PERL5=	yes
+NO_BUILD=	yes
+
+WRKSRC=		${WRKDIR}
+
+PORTDOCS=	FAQ samples
+PORTDOCS_F=	FAQ samples/README samples/animated-gif.eml		\
+		samples/corrupted-gif.eml samples/jpeg.eml samples/png.eml
+
+SUB_FILES=	pkg-message
+
+#-----------------------------------------------------------------------
+# The plugin is installed twice: once where it should go, and once where
+# it actually works from. This will hopefully be fixed at some point.
+#-----------------------------------------------------------------------
+
+SA_PLUGINDIR?=	Mail/SpamAssassin/Plugin
+SA_CONFIGDIR?=	etc/mail/spamassassin
+
+PLIST_FILES=	${SA_CONFIGDIR}/FuzzyOcr.pm				\
+		%%SITE_PERL%%/${SA_PLUGINDIR}/FuzzyOcr.pm		\
+		%%EXAMPLESDIR%%/FuzzyOcr.cf %%EXAMPLESDIR%%/FuzzyOcr.words
+PLIST_DIRS=	%%EXAMPLESDIR%%
+
+post-patch:
+	@${REINPLACE_CMD} -e 's,/usr/bin,${LOCALBASE}/bin,g'		\
+	                  -e 's,/bin/grep,/usr/bin/grep,g'		\
+	                  -e 's,/etc/mail,${LOCALBASE}/etc/mail,g'	\
+	                  ${WRKSRC}/FuzzyOcr.??
+	@${REINPLACE_CMD} -e 's,$$ENV{HOME},($$ENV{HOME} ? $$ENV{HOME} : "/nonexistent"),g' \
+	                 ${WRKSRC}/FuzzyOcr.pm
+
+do-install:
+	@${MKDIR} ${EXAMPLESDIR}
+	${INSTALL_DATA} ${WRKSRC}/FuzzyOcr.pm ${SITE_PERL}/${SA_PLUGINDIR}
+	${INSTALL_DATA} ${WRKSRC}/FuzzyOcr.pm ${PREFIX}/${SA_CONFIGDIR}
+	${INSTALL_DATA} ${WRKSRC}/FuzzyOcr.cf ${EXAMPLESDIR}
+	${INSTALL_DATA} ${WRKSRC}/FuzzyOcr.words.sample ${EXAMPLESDIR}/FuzzyOcr.words
+
+post-install:
+.if !defined(NOPORTDOCS)
+	@${MKDIR} ${DOCSDIR}/samples
+.  for f in ${PORTDOCS_F}
+	${INSTALL_DATA} ${WRKSRC}/${f} ${DOCSDIR}/${f}
+.  endfor
+.endif
+	@${CAT} ${PKGMESSAGE}
+
+.include <bsd.port.mk>
diff --git a/mail/p5-FuzzyOcr/distinfo b/mail/p5-FuzzyOcr/distinfo
new file mode 100644
index 00000000000..a0ae182da04
--- /dev/null
+++ b/mail/p5-FuzzyOcr/distinfo
@@ -0,0 +1,3 @@
+MD5 (fuzzyocr-2.3rc1.tar.gz) = 43879c3f30045b0067d21f9e44b90697
+SHA256 (fuzzyocr-2.3rc1.tar.gz) = bdd8c3524e6445ea79968497c84a9d5e58491ccab9e05aa0784179bec8c24131
+SIZE (fuzzyocr-2.3rc1.tar.gz) = 75295
diff --git a/mail/p5-FuzzyOcr/files/pkg-message.in b/mail/p5-FuzzyOcr/files/pkg-message.in
new file mode 100644
index 00000000000..11e9ed3223d
--- /dev/null
+++ b/mail/p5-FuzzyOcr/files/pkg-message.in
@@ -0,0 +1,5 @@
+------------------------------------------------------------------------
+ To complete the install of FuzzyOcr, copy FuzzyOcr.cf and
+ FuzzyOcr.words from %%EXAMPLESDIR%% into your
+ SpamAssassin configuration directory.
+------------------------------------------------------------------------
diff --git a/mail/p5-FuzzyOcr/pkg-descr b/mail/p5-FuzzyOcr/pkg-descr
new file mode 100644
index 00000000000..a7069d137cb
--- /dev/null
+++ b/mail/p5-FuzzyOcr/pkg-descr
@@ -0,0 +1,12 @@
+FuzzyOcr is a plugin for SpamAssassin which scans image attachments for
+spam keywords, using an Optical Character Recognition program.
+
+This plugin can be used to detect spam that puts all the real spam
+content in an attached image. The mail itself only random text and
+random html, without any URLs or identifiable information.
+
+It is also able to do approximate matches on words, so errors in
+recognition or attempts to obfuscate the text inside the image will not
+cause the detection to fail.
+
+WWW: http://wiki.apache.org/spamassassin/FuzzyOcrPlugin