Dropping maintainership of security/p5-Crypt-OpenSSL-RSA
Hi, could anybody please be so kind as to remove my maintainership from security/p5-Crypt-OpenSSL-RSA? -- Thanks, Andreas Index: Makefile === RCS file: /cvs/ports/security/p5-Crypt-OpenSSL-RSA/Makefile,v retrieving revision 1.15 diff -u -r1.15 Makefile --- Makefile14 Mar 2011 09:17:02 - 1.15 +++ Makefile7 Jan 2012 09:46:39 - @@ -7,11 +7,9 @@ MODULES = cpan DISTNAME = Crypt-OpenSSL-RSA-0.26 -REVISION = 0 +REVISION = 1 CATEGORIES = security HOMEPAGE = http://perl-openssl.sourceforge.net/ - -MAINTAINER = Andreas Voegele po...@andreasvoegele.com # Perl PERMIT_PACKAGE_CDROM = Yes
UPDATE: devel/p5-File-ChangeNotify 0.20
Hi, the attached patch updates devel/p5-File-ChangeNotify to version 0.20. The update fixes a bug in the kqueue-based file watcher. Upstream changed the license from same as perl to Artistic 2.0. Tested on i386 and amd64. Comments? Ok? Index: Makefile === RCS file: /cvs/ports/devel/p5-File-ChangeNotify/Makefile,v retrieving revision 1.5 diff -u -r1.5 Makefile --- Makefile3 Dec 2010 11:44:31 - 1.5 +++ Makefile13 Dec 2011 08:24:23 - @@ -3,23 +3,24 @@ COMMENT= watch for changes to files, cross-platform style MODULES= cpan -DISTNAME= File-ChangeNotify-0.13 -REVISION= 0 +DISTNAME= File-ChangeNotify-0.20 CATEGORIES=devel -# Perl +# Artistic 2.0 PERMIT_PACKAGE_CDROM = Yes PERMIT_PACKAGE_FTP = Yes PERMIT_DISTFILES_CDROM = Yes PERMIT_DISTFILES_FTP = Yes CONFIGURE_STYLE = modbuild -USE_GROFF =Yes RUN_DEPENDS = devel/p5-Class-MOP \ devel/p5-Moose \ devel/p5-MooseX-Params-Validate \ - devel/p5-MooseX-SemiAffordanceAccessor + devel/p5-MooseX-SemiAffordanceAccessor \ + devel/p5-namespace-autoclean + +REGRESS_DEPENDS = devel/p5-Test-Exception MAKE_ENV +=TEST_POD=1 IS_MAINTAINER=1 Index: distinfo === RCS file: /cvs/ports/devel/p5-File-ChangeNotify/distinfo,v retrieving revision 1.2 diff -u -r1.2 distinfo --- distinfo3 Jul 2010 03:41:42 - 1.2 +++ distinfo13 Dec 2011 08:24:23 - @@ -1,5 +1,5 @@ -MD5 (File-ChangeNotify-0.13.tar.gz) = eR70uZegW1BwTiwo+Ph4oA== -RMD160 (File-ChangeNotify-0.13.tar.gz) = Jltiid+UQa63uHcm5mBS2KnQ69s= -SHA1 (File-ChangeNotify-0.13.tar.gz) = WcEiFJwmsqrqCjYp2xNnbAlSbqM= -SHA256 (File-ChangeNotify-0.13.tar.gz) = yHAoVdthRnb9v2P9raOA6tj2FAE11Y1TkUQuDYmpaig= -SIZE (File-ChangeNotify-0.13.tar.gz) = 18155 +MD5 (File-ChangeNotify-0.20.tar.gz) = m7aPbhSUNqznyxPI+c22vQ== +RMD160 (File-ChangeNotify-0.20.tar.gz) = HZ0XOB8M165qhqHzMuU2XwEYGDg= +SHA1 (File-ChangeNotify-0.20.tar.gz) = lP4BxmtHbC42dj51YJBeSu1ckAY= +SHA256 (File-ChangeNotify-0.20.tar.gz) = qdVKmhK/J9HKxmW5WZULpvhETd6E4occK9IEJ2OMCgA= +SIZE (File-ChangeNotify-0.20.tar.gz) = 19642
Re: UPDATE: p5-Moose 1.25 ...
Hi Ingo, Hi Andreas, Andreas Vögele wrote on Tue, Nov 22, 2011 at 03:34:47PM +0100: I decided to remove USE_GROFF from the Moose port. Sounds reasonable for a Perl port, in particular since it seems you checked the output quality, given the issues you found. There are two minor issues though. Both groff and mandoc render =item B as used in Moose's Pod documents wrongly. Humm. Looking e.g. at Moose::Object.3p, i don't really see a problem, neither with groff nor with mandoc, and both agree. Can you explain what exactly is wrong? An example is attached. mandoc doesn't emphasise the method declaration's second argument. Groff, on the other hand, underlines the second argument. This happens with $obj, @obj and %obj, but not with obj, for example, so this is probably a bug in pod2man too. pod2man example.pod | mandoc | less pod2man example.pod | nroff -man | less -r Furthermore, mandoc doesn't wrap long URLs. You mean, as in Moose::Spec::Role.3p, below SEE ALSO, Roles? Groff doesn't wrap that one either, so both formatters agree. You're right, Groff also produces long lines. BTW, see Lhttp://blogs.perl.org/users/brian_d_foy/2011/11/help-update-podperldoc.html. -- Regards, Andreas =over =item Two arg method B $obj-($x, $y) bla. =item Long URL Lhttp://www.gigamonkeys.com/book/object-reorientation-generic-functions.html =back
UPDATE: p5-File-ChangeNotify 0.20
Hi, the attached patch updates devel/p5-File-ChangeNotify to version 0.20. The update fixes a bug in the kqueue-based file watcher and is also required by Moose 1.25+. The module's license changed from same as perl to Artistic 2.0. I checked the manual pages and removed USE_GROFF. See Lhttps://metacpan.org/source/DROLSKY/File-ChangeNotify-0.20/Changes for the list of changes since version 0.13. p5-File-ChangeNotify is used by Catalyst. Tested on i386 and amd64. -- Regards, Andreas diff --git a/devel/p5-File-ChangeNotify/Makefile b/devel/p5-File-ChangeNotify/Makefile index 1ad7747..2430456 100644 --- a/devel/p5-File-ChangeNotify/Makefile +++ b/devel/p5-File-ChangeNotify/Makefile @@ -3,23 +3,24 @@ COMMENT= watch for changes to files, cross-platform style MODULES= cpan -DISTNAME= File-ChangeNotify-0.13 -REVISION= 0 +DISTNAME= File-ChangeNotify-0.20 CATEGORIES=devel -# Perl +# Artistic 2.0 PERMIT_PACKAGE_CDROM = Yes PERMIT_PACKAGE_FTP = Yes PERMIT_DISTFILES_CDROM = Yes PERMIT_DISTFILES_FTP = Yes CONFIGURE_STYLE = modbuild -USE_GROFF =Yes RUN_DEPENDS = devel/p5-Class-MOP \ devel/p5-Moose \ devel/p5-MooseX-Params-Validate \ - devel/p5-MooseX-SemiAffordanceAccessor + devel/p5-MooseX-SemiAffordanceAccessor \ + devel/p5-namespace-autoclean + +REGRESS_DEPENDS = devel/p5-Test-Exception MAKE_ENV +=TEST_POD=1 IS_MAINTAINER=1 diff --git a/devel/p5-File-ChangeNotify/distinfo b/devel/p5-File-ChangeNotify/distinfo index ee2e30d..c82b17d 100644 --- a/devel/p5-File-ChangeNotify/distinfo +++ b/devel/p5-File-ChangeNotify/distinfo @@ -1,5 +1,5 @@ -MD5 (File-ChangeNotify-0.13.tar.gz) = eR70uZegW1BwTiwo+Ph4oA== -RMD160 (File-ChangeNotify-0.13.tar.gz) = Jltiid+UQa63uHcm5mBS2KnQ69s= -SHA1 (File-ChangeNotify-0.13.tar.gz) = WcEiFJwmsqrqCjYp2xNnbAlSbqM= -SHA256 (File-ChangeNotify-0.13.tar.gz) = yHAoVdthRnb9v2P9raOA6tj2FAE11Y1TkUQuDYmpaig= -SIZE (File-ChangeNotify-0.13.tar.gz) = 18155 +MD5 (File-ChangeNotify-0.20.tar.gz) = m7aPbhSUNqznyxPI+c22vQ== +RMD160 (File-ChangeNotify-0.20.tar.gz) = HZ0XOB8M165qhqHzMuU2XwEYGDg= +SHA1 (File-ChangeNotify-0.20.tar.gz) = lP4BxmtHbC42dj51YJBeSu1ckAY= +SHA256 (File-ChangeNotify-0.20.tar.gz) = qdVKmhK/J9HKxmW5WZULpvhETd6E4occK9IEJ2OMCgA= +SIZE (File-ChangeNotify-0.20.tar.gz) = 19642
UPDATE: p5-Any-Moose 0.18 and p5-Mouse 0.97
Hi,
the attached patches update devel/p5-Any-Moose to version 0.18 and
devel/p5-Mouse to 0.97. p5-Mouse now provides Mouse::XS, which is about
two times faster than Mouse::PurePerl. p5-Any-Moose now depends on
p5-Mouse = 0.40. I checked the manual pages and removed USE_GROFF.
p5-Mouse is now SHARED_ONLY, which keeps the port simple. p5-Any-Moose
and p5-Mouse are used by p5-Data-AMF and p5-GnuPG-Interface. p5-Data-AMF
depends on the SHARED_ONLY ports p5-DateTime and p5-XML-LibXML anyway.
p5-GnuPG-Interface and gnupg aren't built on the vax platform, so
nothing is lost.
See Lhttps://metacpan.org/source/SARTAK/Any-Moose-0.18/Changes and
Lhttps://metacpan.org/source/GFUJI/Mouse-0.97/Changes for the list of
changes.
No feedback from the maintainer for three weeks. Both ports are
available at
Lhttps://github.com/jasperla/openbsd-wip/tree/master/devel in case
anybody wants to change anything.
Tested on i386 and amd64.
--
Regards,
Andreas
diff --git a/devel/p5-Any-Moose/Makefile b/devel/p5-Any-Moose/Makefile
index 781b090..3328446 100644
--- a/devel/p5-Any-Moose/Makefile
+++ b/devel/p5-Any-Moose/Makefile
@@ -3,10 +3,8 @@
COMMENT = use Moose or Mouse modules
MODULES = cpan
-DISTNAME = Any-Moose-0.09
-REVISION = 0
+DISTNAME = Any-Moose-0.18
CATEGORIES = devel
-USE_GROFF =Yes
# Perl
PERMIT_PACKAGE_CDROM = Yes
@@ -14,8 +12,9 @@ PERMIT_PACKAGE_FTP = Yes
PERMIT_DISTFILES_CDROM = Yes
PERMIT_DISTFILES_FTP = Yes
-CPAN_AUTHOR = FLORA
+RUN_DEPENDS = devel/p5-Mouse=0.40
-RUN_DEPENDS = devel/p5-Mouse=0.21
+REGRESS_DEPENDS = devel/p5-Moose \
+ devel/p5-MooseX-Types
.include bsd.port.mk
diff --git a/devel/p5-Any-Moose/distinfo b/devel/p5-Any-Moose/distinfo
index 2b05361..884605b 100644
--- a/devel/p5-Any-Moose/distinfo
+++ b/devel/p5-Any-Moose/distinfo
@@ -1,5 +1,5 @@
-MD5 (Any-Moose-0.09.tar.gz) = +64PvYia0+Br835A4eb+LA==
-RMD160 (Any-Moose-0.09.tar.gz) = i4QjM61Iql8//fpBtO9u3HRmkj8=
-SHA1 (Any-Moose-0.09.tar.gz) = 2+wRL7jd/lU4DZYv4L44IHqllVI=
-SHA256 (Any-Moose-0.09.tar.gz) = 1Au8jdYkfBDykg6dBc6F8u2BSgP+lGTAmHW6Rz8cr3M=
-SIZE (Any-Moose-0.09.tar.gz) = 11474
+MD5 (Any-Moose-0.18.tar.gz) = OwPnw+JZMAUQWqg0f5Vx1Q==
+RMD160 (Any-Moose-0.18.tar.gz) = dXShi+rav9OdMeoLavy7kAgmk4U=
+SHA1 (Any-Moose-0.18.tar.gz) = sWr1270fwVsv7LATukYkSPbCjmM=
+SHA256 (Any-Moose-0.18.tar.gz) = Cxtw02Gck0Y0zrSm/xLeJl5j2rKXh1GrgWTJ90x14ac=
+SIZE (Any-Moose-0.18.tar.gz) = 15319
diff --git a/devel/p5-Mouse/Makefile b/devel/p5-Mouse/Makefile
index e77bcf7..4527edc 100644
--- a/devel/p5-Mouse/Makefile
+++ b/devel/p5-Mouse/Makefile
@@ -1,12 +1,12 @@
# $OpenBSD: Makefile,v 1.7 2010/12/03 11:44:34 ajacoutot Exp $
+SHARED_ONLY = Yes
+
COMMENT = Moose minus the antlers
MODULES = cpan
-DISTNAME = Mouse-0.24
-REVISION = 0
+DISTNAME = Mouse-0.97
CATEGORIES = devel
-USE_GROFF =Yes
MAINTAINER = Simon Bertrang si...@openbsd.org
@@ -16,10 +16,17 @@ PERMIT_PACKAGE_FTP =Yes
PERMIT_DISTFILES_CDROM = Yes
PERMIT_DISTFILES_FTP = Yes
-CPAN_AUTHOR = SARTAK
+CPAN_AUTHOR = GFUJI
-RUN_DEPENDS = devel/p5-Class-Method-Modifiers
-BUILD_DEPENDS =${RUN_DEPENDS}
-REGRESS_DEPENDS = devel/p5-Test-Exception
+REGRESS_DEPENDS = devel/p5-Declare-Constraints-Simple \
+ devel/p5-IO-String \
+ devel/p5-Locale-US \
+ devel/p5-Params-Coerce \
+ devel/p5-Path-Class \
+ devel/p5-Test-Deep \
+ devel/p5-Test-LeakTrace \
+ devel/p5-Test-Output \
+ textproc/p5-Regexp-Common \
+ www/p5-libwww
.include bsd.port.mk
diff --git a/devel/p5-Mouse/distinfo b/devel/p5-Mouse/distinfo
index 4694ee5..da497fa 100644
--- a/devel/p5-Mouse/distinfo
+++ b/devel/p5-Mouse/distinfo
@@ -1,5 +1,5 @@
-MD5 (Mouse-0.24.tar.gz) = 7H1XYCf0qPAaT7DIR6GVJA==
-RMD160 (Mouse-0.24.tar.gz) = QKf21i//YxqNpadhlB+2msu8pNE=
-SHA1 (Mouse-0.24.tar.gz) = N0aXtY3QKjqJoG2iQJb6AENdgJw=
-SHA256 (Mouse-0.24.tar.gz) = OP7xCY8HZdFV/q+AJ7bO3Z4tk4wih4RQc/jYYgrD48s=
-SIZE (Mouse-0.24.tar.gz) = 114627
+MD5 (Mouse-0.97.tar.gz) = kqizh/lhnTK85mhcHdNw5Q==
+RMD160 (Mouse-0.97.tar.gz) = S+nhKiyGAJgTGYnvDuUtoTGpQME=
+SHA1 (Mouse-0.97.tar.gz) = EteF4siCHFLuBxTJTMNhQRLP56I=
+SHA256 (Mouse-0.97.tar.gz) = 37Hzq7xG52RqaF76TlgExM+XcuuAesar+PHRUptXuXo=
+SIZE (Mouse-0.97.tar.gz) = 229149
diff --git a/devel/p5-Mouse/pkg/PLIST b/devel/p5-Mouse/pkg/PLIST
index 34d8452..0aa4839 100644
--- a/devel/p5-Mouse/pkg/PLIST
+++ b/devel/p5-Mouse/pkg/PLIST
@@ -1,33 +1,72 @@
@comment $OpenBSD: PLIST,v 1.3 2009/05/06 22:54:06 simon Exp $
-${P5SITE}/Mouse/
-${P5SITE}/Mouse.pm
-${P5SITE}/Mouse/Meta/
-${P5SITE}/Mouse/Meta/Attribute.pm
-${P5SITE}/Mouse/Meta/Class.pm
-${P5SITE}/Mouse/Meta/Method/
-${P5SITE}/Mouse/Meta/Method/Accessor.pm
-${P5SITE}/Mouse/Meta/Method/Constructor.pm
UPDATE: p5-Package-Stash 0.33, p5-Package-Stash-XS 0.25, p5-namespace-clean 0.21, p5-namespace-autoclean 0.13
Hi,
the attached patches update p5-Package-Stash, p5-Package-Stash-XS,
p5-namespace-clean, p5-namespace-autoclean and also p5-Class-MOP. A lot
of Perl modules depend on recent versions of these modules.
The ports have to be updated together. p5-namespace-autoclean depends on
p5-namespace-clean = 0.20. p5-namespace-clean depends on
p5-Package-Stash = 0.23. p5-Package-Stash 0.33 utilises
p5-Package-Stash-XS if installed.
One of p5-Class-MOP's tests needs to be updated since a message changed
in Package::Stash. Furthermore, p5-Task-Weaken is only a build
dependency and BUILD_DEPENDS doesn't have to be set to RUN_DEPENDS.
USE_GROFF was removed from all five ports.
No maintainer feedback for two weeks. The ports are available at
Lhttps://github.com/jasperla/openbsd-wip/tree/master/devel. See the
UPDATE files in the Git repository for the changelogs.
Tested on i386 and amd64.
--
Regards,
Andreas
diff --git a/devel/p5-Package-Stash/Makefile b/devel/p5-Package-Stash/Makefile
index 182a3e7..e8ce20b 100644
--- a/devel/p5-Package-Stash/Makefile
+++ b/devel/p5-Package-Stash/Makefile
@@ -3,10 +3,8 @@
COMMENT= routines for manipulating stashes
MODULES= cpan
-DISTNAME= Package-Stash-0.05
-REVISION= 0
+DISTNAME= Package-Stash-0.33
CATEGORIES=devel
-USE_GROFF =Yes
MAINTAINER=Jim Razmus II j...@openbsd.org
@@ -16,6 +14,10 @@ PERMIT_PACKAGE_FTP= Yes
PERMIT_DISTFILES_CDROM=Yes
PERMIT_DISTFILES_FTP= Yes
-REGRESS_DEPENDS= devel/p5-Test-Exception
+RUN_DEPENDS = devel/p5-Dist-CheckConflicts \
+ devel/p5-Package-DeprecationManager
+
+REGRESS_DEPENDS = devel/p5-Test-Fatal \
+ devel/p5-Test-Requires
.include bsd.port.mk
diff --git a/devel/p5-Package-Stash/distinfo b/devel/p5-Package-Stash/distinfo
index 25d2c8a..6138abf 100644
--- a/devel/p5-Package-Stash/distinfo
+++ b/devel/p5-Package-Stash/distinfo
@@ -1,5 +1,5 @@
-MD5 (Package-Stash-0.05.tar.gz) = 17OFesofIrLaBs/NMSPL7w==
-RMD160 (Package-Stash-0.05.tar.gz) = TUoduL7GOvE6idTRtJjPj/rGyqw=
-SHA1 (Package-Stash-0.05.tar.gz) = OEOPduPMFLJa+zuy6bPKxt6p92M=
-SHA256 (Package-Stash-0.05.tar.gz) =
5yRZg67ugURM/frSIki3HwdEIp+qFgUHhGiqXSWeGXY=
-SIZE (Package-Stash-0.05.tar.gz) = 17023
+MD5 (Package-Stash-0.33.tar.gz) = /h9u+g29RvhSCQO0NvGF8Q==
+RMD160 (Package-Stash-0.33.tar.gz) = /MyL9B9zpfkvDnHDfSzW19jp0/I=
+SHA1 (Package-Stash-0.33.tar.gz) = m+sQKiFMEwfZ2UgIYZfzN7TN/mQ=
+SHA256 (Package-Stash-0.33.tar.gz) =
9l/YFdUVBv/EgCU/3tZ8BNMow2B8Q0HHu3pb9nAWpVU=
+SIZE (Package-Stash-0.33.tar.gz) = 29841
diff --git a/devel/p5-Package-Stash/pkg/PLIST b/devel/p5-Package-Stash/pkg/PLIST
index f400b5b..ccc76ac 100644
--- a/devel/p5-Package-Stash/pkg/PLIST
+++ b/devel/p5-Package-Stash/pkg/PLIST
@@ -1,4 +1,8 @@
@comment $OpenBSD: PLIST,v 1.1.1.1 2010/06/29 13:56:06 jim Exp $
-${P5SITE}/Package/
+bin/package-stash-conflicts
+${P5SITE}/Package/Stash/
${P5SITE}/Package/Stash.pm
+${P5SITE}/Package/Stash/Conflicts.pm
+${P5SITE}/Package/Stash/PP.pm
@man man/man3p/Package::Stash.3p
+@man man/man3p/Package::Stash::PP.3p
diff --git a/devel/p5-Package-Stash-XS/Makefile
b/devel/p5-Package-Stash-XS/Makefile
index a2ba375..605d013 100644
--- a/devel/p5-Package-Stash-XS/Makefile
+++ b/devel/p5-Package-Stash-XS/Makefile
@@ -4,7 +4,7 @@ COMMENT = faster and more correct implementation of
Package::Stash
SHARED_ONLY = Yes
MODULES = cpan
-DISTNAME = Package-Stash-XS-0.22
+DISTNAME = Package-Stash-XS-0.25
CATEGORIES = devel
# Perl
diff --git a/devel/p5-Package-Stash-XS/distinfo
b/devel/p5-Package-Stash-XS/distinfo
index 6a45dd4..1f99280 100644
--- a/devel/p5-Package-Stash-XS/distinfo
+++ b/devel/p5-Package-Stash-XS/distinfo
@@ -1,5 +1,5 @@
-MD5 (Package-Stash-XS-0.22.tar.gz) = DHT6p5IBXl4+uFKxDJ4dcg==
-RMD160 (Package-Stash-XS-0.22.tar.gz) = Bh2fOb33mpNYJbqbyUmm74lbjFA=
-SHA1 (Package-Stash-XS-0.22.tar.gz) = JZIUsrUG+S7pNqW3zSv3hog1OXM=
-SHA256 (Package-Stash-XS-0.22.tar.gz) =
9pyXJ2s+APNxDbOD4L1PXcrOw3YCZNCSf0CcgM6pKOw=
-SIZE (Package-Stash-XS-0.22.tar.gz) = 65842
+MD5 (Package-Stash-XS-0.25.tar.gz) = UuzhiELEbaj7ekoU3qkQqw==
+RMD160 (Package-Stash-XS-0.25.tar.gz) = Mq+Zz635DybnkUedoy9lqIK4QJE=
+SHA1 (Package-Stash-XS-0.25.tar.gz) = 5LExwDeSEIqHKJKJvJq3XqVVwuE=
+SHA256 (Package-Stash-XS-0.25.tar.gz) =
c5xYENaaKDRrNYhuczWIlTjRRJ8TZWFkINjpExqshcQ=
+SIZE (Package-Stash-XS-0.25.tar.gz) = 67155
diff --git a/devel/p5-Class-MOP/Makefile b/devel/p5-Class-MOP/Makefile
index 390893f..1849639 100644
--- a/devel/p5-Class-MOP/Makefile
+++ b/devel/p5-Class-MOP/Makefile
@@ -6,8 +6,8 @@ COMMENT=meta object protocol for perl 5
MODULES= cpan
DISTNAME= Class-MOP-1.12
+REVISION= 0
CATEGORIES=devel
-USE_GROFF =Yes
# perl
PERMIT_PACKAGE_CDROM= Yes
@@ -15,14 +15,14 @@ PERMIT_PACKAGE_FTP= Yes
PERMIT_DISTFILES_CDROM=Yes
PERMIT_DISTFILES_FTP=
UPDATE: p5-Lingua-EN-Inflect to 1.893
Hi,
the current version of DBIx::Class::Schema::Loader depends on a more
recent version of Lingua::EN::Inflect. A patch that updates
textproc/p5-Lingua-EN-Inflect to version 1.893 is attached. I've just
got an OK from Sam Smith, the port's maintainer.
See
Lhttps://metacpan.org/source/DCONWAY/Lingua-EN-Inflect-1.893/Changes
for a list of changes since version 1.88.
--
Regards,
Andreas
diff -ur /usr/ports/textproc/p5-Lingua-EN-Inflect/Makefile
/usr/ports/mystuff/textproc/p5-Lingua-EN-Inflect/Makefile
--- /usr/ports/textproc/p5-Lingua-EN-Inflect/Makefile Fri Dec 3 12:46:48 2010
+++ /usr/ports/mystuff/textproc/p5-Lingua-EN-Inflect/Makefile Mon Nov 14
18:10:01 2011
@@ -3,8 +3,7 @@
COMMENT= convert singular to plural, select a or an
MODULES= cpan
-DISTNAME= Lingua-EN-Inflect-1.88
-REVISION= 2
+DISTNAME= Lingua-EN-Inflect-1.893
CATEGORIES=textproc
USE_GROFF =Yes
diff -ur /usr/ports/textproc/p5-Lingua-EN-Inflect/distinfo
/usr/ports/mystuff/textproc/p5-Lingua-EN-Inflect/distinfo
--- /usr/ports/textproc/p5-Lingua-EN-Inflect/distinfo Thu Apr 5 19:26:16 2007
+++ /usr/ports/mystuff/textproc/p5-Lingua-EN-Inflect/distinfo Mon Nov 14
18:10:56 2011
@@ -1,5 +1,5 @@
-MD5 (Lingua-EN-Inflect-1.88.tar.gz) = 5RAO5Uv7OJ98zuiK9wq2oA==
-RMD160 (Lingua-EN-Inflect-1.88.tar.gz) = EbF5wHjqzIUdy8DW66WDfc3KecE=
-SHA1 (Lingua-EN-Inflect-1.88.tar.gz) = +Lu5g2tCMrLMwBGrKv86b9v5NtE=
-SHA256 (Lingua-EN-Inflect-1.88.tar.gz) =
OflbrZBUs7/1qVUwnL1hE4B7eq4LkeRFKuzOtsks6cg=
-SIZE (Lingua-EN-Inflect-1.88.tar.gz) = 37566
+MD5 (Lingua-EN-Inflect-1.893.tar.gz) = XOUIjICGF4UWUetuJ4hJuw==
+RMD160 (Lingua-EN-Inflect-1.893.tar.gz) = KDTlWw0E8GQKUftOYL/Q0h7dZzE=
+SHA1 (Lingua-EN-Inflect-1.893.tar.gz) = +INdsFs4GYLVkYCGIjQ+LrEUeiQ=
+SHA256 (Lingua-EN-Inflect-1.893.tar.gz) =
vnZShPaufSzKn2M8mGf90IhPK3uTt1unglpbfIfrEsg=
+SIZE (Lingua-EN-Inflect-1.893.tar.gz) = 45630
diff -ur /usr/ports/textproc/p5-Lingua-EN-Inflect/pkg/PLIST
/usr/ports/mystuff/textproc/p5-Lingua-EN-Inflect/pkg/PLIST
--- /usr/ports/textproc/p5-Lingua-EN-Inflect/pkg/PLIST Tue Jan 4 20:55:58 2005
+++ /usr/ports/mystuff/textproc/p5-Lingua-EN-Inflect/pkg/PLIST Mon Nov 14
18:15:23 2011
@@ -2,9 +2,4 @@
${P5SITE}/Lingua/
${P5SITE}/Lingua/EN/
${P5SITE}/Lingua/EN/Inflect.pm
-${P5SITE}/Lingua/EN/demo_NO.pl
-${P5SITE}/Lingua/EN/demo_NUM.pl
-${P5SITE}/Lingua/EN/demo_PL.pl
-${P5SITE}/Lingua/EN/demo_eq.pl
-${P5SITE}/Lingua/EN/demo_inflect.pl
@man man/man3p/Lingua::EN::Inflect.3p
NEW: p5-DateTime-Format-SQLite
Hi, I'm currently updating DBIx::Class, Moose and Catalyst. See https://github.com/voegelas/openbsd-wip/wiki in case you're interested. While I'm waiting for feedback from the maintainers of several ports, it would be great if the attached port was imported into OpenBSD's ports tree. DateTime::Format::SQLite is used by the test suites of DBIx::Class and DBIx::Class::Validation. But IMHO it's also of general use. The module understands the formats used by SQLite for its date, datetime and time functions. It can be used to parse these formats in order to create DateTime objects, and it can take a DateTime object and produce a timestring accepted by SQLite. -- Regards, Andreas p5-DateTime-Format-SQLite-0.11.tar.gz Description: application/gzip
NEW: p5-MooseX-Types-JSON
Hi, the attached port is used by the test suite of recent DBIx::Class versions as well as the dbicadmin tool. It is in the REGRESS_DEPENDS list of my upcoming p5-DBIx-Class update. MooseX::Types::JSON provides string type constraints that match JSON. -- Regards, Andreas p5-MooseX-Types-JSON-0.02.tar.gz Description: application/gzip
NEW: p5-Data-Compare
Hi, the attached port is required by recent versions of DBIx::Class. It will be in the RUN_DEPENDS list of my upcoming p5-DBIx-Class update. Data::Compare compares two perl data structures recursively. -- Regards, Andreas p5-Data-Compare-1.22.tar.gz Description: application/gzip
NEW: p5-Data-Dumper-Concise
Hi, the attached port is required by recent versions of DBIx::Class. It will be in the RUN_DEPENDS list of my upcoming p5-DBIx-Class update. The module is also required by Devel::REPL, which Carp::REPL and CatalystX::REPL depend on. Data::Dumper::Concise is similar to Data::Dumper but dumps perl data structures with less indentation and newlines. Wen Heping sent this port to ports@ in May but it wasn't imported. Wen, I didn't change the MAINTAINER entry. Or do you object to maintain this port? -- Regards, Andreas p5-Data-Dumper-Concise-2.020.tar.gz Description: application/gzip
OBSOLETE: p5-Catalyst-Engine-HTTP-POE
Hi, the port www/p5-Catalyst-Engine-HTTP-POE is not usable and could be removed from OpenBSD's ports tree. The Restarter engine used by this module was removed in Catalyst 5.80008. Catalyst 5.80013 was imported two years ago, i.e. this port has been unusable for two years. No update is available. The module is unmaintained and deprecated in favour of Catalyst::Engine::HTTP::Prefork. See the section Refactoring / cleanups in Lhttps://metacpan.org/source/FLORA/Catalyst-Runtime-5.80008/Changes. See Lhttps://metacpan.org/source/AGRUNDMA/Catalyst-Engine-HTTP-POE-0.08/lib/Catalyst/Engine/HTTP/POE.pm for the source code. It uses the module Catalyst::Engine::HTTP::Restarter::Watcher, which doesn't exist anymore. -- Regards, Andreas
UPDATE: devel/p5-Test-WWW-Mechanize-1.30
Synchronize devel/p5-Test-WWW-Mechanize with www/p5-WWW-Mechanize, which
was recently updated.
More recent versions of Test::WWW::Mechanize depend on LWP 6.02, which
hasn't hit OpenBSD's ports tree yet.
Tested on i386 and amd64.
Used by www/p5-WWW-Mechanize-TreeBuilder, which still passes its tests
suite. Also used by devel/p5-SVN-Web as well as several Catalyst
modules. The p5-SVN-Web port currently doesn't pass its test suite no
matter whether p5-Test-WWW-Mechanize is updated or not.
I've got updates for p5-Test-WWW-Mechanize-Catalyst and other Catalyst
ports that I'll send to the respective maintainers. I've also forked
Jasper's openbsd-wip repository and started to push the updated ports to
the forked repository. See https://github.com/voegelas/openbsd-wip/wiki
for a list of ports.
diff --git a/devel/p5-Test-WWW-Mechanize/Makefile
b/devel/p5-Test-WWW-Mechanize/Makefile
index 2ff5ec8..6430c0e 100644
--- a/devel/p5-Test-WWW-Mechanize/Makefile
+++ b/devel/p5-Test-WWW-Mechanize/Makefile
@@ -1,13 +1,11 @@
# $OpenBSD: ports/devel/p5-Test-WWW-Mechanize/Makefile,v 1.11 2010/12/03
11:44:37 ajacoutot Exp $
-COMMENT= test suite using WWW::Mechanize
+COMMENT= test suite using WWW::Mechanize
MODULES= cpan
-DISTNAME= Test-WWW-Mechanize-1.12
-REVISION= 2
+DISTNAME= Test-WWW-Mechanize-1.30
CATEGORIES=www devel
-USE_GROFF =Yes
# Artitistic License
PERMIT_PACKAGE_CDROM= Yes
@@ -18,8 +16,11 @@ PERMIT_DISTFILES_FTP=Yes
BUILD_DEPENDS= ${RUN_DEPENDS}
RUN_DEPENDS= devel/p5-Carp-Assert-More \
devel/p5-Test-LongString \
+ www/p5-HTML-Lint \
+ www/p5-HTML-Tree \
www/p5-HTTP-Server-Simple \
- www/p5-WWW-Mechanize
+ www/p5-WWW-Mechanize \
+ www/p5-URI
MAKE_ENV= TEST_POD=Yes
diff --git a/devel/p5-Test-WWW-Mechanize/distinfo
b/devel/p5-Test-WWW-Mechanize/distinfo
index 969ada0..1066be5 100644
--- a/devel/p5-Test-WWW-Mechanize/distinfo
+++ b/devel/p5-Test-WWW-Mechanize/distinfo
@@ -1,5 +1,5 @@
-MD5 (Test-WWW-Mechanize-1.12.tar.gz) = IgntCXRafzTaMcNGGrc0/g==
-RMD160 (Test-WWW-Mechanize-1.12.tar.gz) = fnVWWvPlxZAKoW7pfPUXFkhuKi0=
-SHA1 (Test-WWW-Mechanize-1.12.tar.gz) = 3VzvhnJvlfUZf5NNK5nt/W5S/LQ=
-SHA256 (Test-WWW-Mechanize-1.12.tar.gz) =
vurdFLEX9dSDH9+eIrQddHs4P47UZWXrNKlr1C7GrIw=
-SIZE (Test-WWW-Mechanize-1.12.tar.gz) = 11710
+MD5 (Test-WWW-Mechanize-1.30.tar.gz) = oNcNUB4m45j0e/Czpb4Egw==
+RMD160 (Test-WWW-Mechanize-1.30.tar.gz) = 2h8Srn/qyUhrEMN8/TVzYQ6iJvg=
+SHA1 (Test-WWW-Mechanize-1.30.tar.gz) = AlA31S0JaPnlpC6bu/DFAOeMqCk=
+SHA256 (Test-WWW-Mechanize-1.30.tar.gz) =
p3yRE5KT/OyldL4dpWL4b0yerJzRYC/rTYJjV0cuC0A=
+SIZE (Test-WWW-Mechanize-1.30.tar.gz) = 19234
diff --git a/devel/p5-Test-WWW-Mechanize/patches/patch-t_content_lacks_t
b/devel/p5-Test-WWW-Mechanize/patches/patch-t_content_lacks_t
new file mode 100644
index 000..719c9fe
--- /dev/null
+++ b/devel/p5-Test-WWW-Mechanize/patches/patch-t_content_lacks_t
@@ -0,0 +1,12 @@
+$OpenBSD$
+--- t/content_lacks.t.orig Mon Apr 5 07:49:01 2010
t/content_lacks.t Tue Oct 25 14:23:31 2011
+@@ -35,7 +35,7 @@ test_out( q{not ok 1 - Shouldn't say it's a test page}
+ test_fail(+4);
+ test_diag(q(searched: html\x{0a}head\x{0a}titleTest
Page/title...) );
+ test_diag(q( and found: Test Page) );
+-test_diag(q( at position: 33) );
++test_diag(q( at position: 33 (line 3 column 16)) );
+ $mech-content_lacks( 'Test Page', q{Shouldn't say it's a test page} );
+ test_test( 'Handles not finding it' );
+
UPDATE: devel/p5-MooseX-POE-0.213
The regression tests of p5-MooseX-POE fail because of changes in Moose and other modules. The attached patch updates p5-MooseX-POE as well as the ports that p5-MooseX-POE depends on. I've moved the dependency on p5-MooseX-Declare from RUN_DEPENDS to REGRESS_DEPENDS since that module is optional. p5-MooseX-POE is required by devel/p5-POE-Component-Supervisor, which still passes its regression tests after the update. I noticed that the tests provided by p5-POE-Component-Supervisor are time critical, though. They may fail in QEMU or on slow machines. These failures are not related to the updated ports. devel/p5-MooseX-Async is no longer required and could be removed from the ports tree. There aren't any ports left that depend on p5-MooseX-Async. I've removed USE_GROFF and verified that all manual pages are displayed fine. Tested on i386 and amd64. diff --git a/devel/p5-B-Hooks-OP-Check/Makefile b/devel/p5-B-Hooks-OP-Check/Makefile index fc2d0f0..8898c6c 100644 --- a/devel/p5-B-Hooks-OP-Check/Makefile +++ b/devel/p5-B-Hooks-OP-Check/Makefile @@ -4,11 +4,9 @@ SHARED_ONLY= Yes COMMENT= wrap OP check callbacks -DISTNAME= B-Hooks-OP-Check-0.18 -REVISION= 1 +DISTNAME= B-Hooks-OP-Check-0.19 CATEGORIES=devel -USE_GROFF =Yes # perl PERMIT_PACKAGE_CDROM= Yes diff --git a/devel/p5-B-Hooks-OP-Check/distinfo b/devel/p5-B-Hooks-OP-Check/distinfo index 804ed26..7d0cdfa 100644 --- a/devel/p5-B-Hooks-OP-Check/distinfo +++ b/devel/p5-B-Hooks-OP-Check/distinfo @@ -1,5 +1,5 @@ -MD5 (B-Hooks-OP-Check-0.18.tar.gz) = DtfoUZZ8bt/fSiwYWJWFQw== -RMD160 (B-Hooks-OP-Check-0.18.tar.gz) = EtaOUG3ZKOZCcavkIFmCXKVxc5Y= -SHA1 (B-Hooks-OP-Check-0.18.tar.gz) = H3aGHCrnrBdY8QwWqyi41byAAyA= -SHA256 (B-Hooks-OP-Check-0.18.tar.gz) = EGaTBhvo4NzI4jo1gAtD3n5QY5hk6knURvOAEl9A4iw= -SIZE (B-Hooks-OP-Check-0.18.tar.gz) = 60537 +MD5 (B-Hooks-OP-Check-0.19.tar.gz) = pq+JDsyyZvatnLKV63xXCw== +RMD160 (B-Hooks-OP-Check-0.19.tar.gz) = Peh67e001bumai9lHGKpZpgC4dc= +SHA1 (B-Hooks-OP-Check-0.19.tar.gz) = IYInYFjiShnbO0qA7PCOXEIFbOs= +SHA256 (B-Hooks-OP-Check-0.19.tar.gz) = L/9KhgyuT6abkRZKHi05a/3dhGa90E1xjmlfcna84V4= +SIZE (B-Hooks-OP-Check-0.19.tar.gz) = 64368 diff --git a/devel/p5-Devel-Declare/Makefile b/devel/p5-Devel-Declare/Makefile index 9185a3e..6527c2d 100644 --- a/devel/p5-Devel-Declare/Makefile +++ b/devel/p5-Devel-Declare/Makefile @@ -4,8 +4,7 @@ SHARED_ONLY=Yes COMMENT= adding keywords to perl, in perl -DISTNAME= Devel-Declare-0.005011 -REVISION= 1 +DISTNAME= Devel-Declare-0.006007 CATEGORIES=devel @@ -16,11 +15,11 @@ PERMIT_DISTFILES_CDROM= Yes PERMIT_DISTFILES_FTP= Yes MODULES= cpan -USE_GROFF =Yes -BUILD_DEPENDS= devel/p5-B-Hooks-OP-Check -RUN_DEPENDS= devel/p5-B-Hooks-OP-Check \ +BUILD_DEPENDS= devel/p5-B-Hooks-OP-Check=0.19 +RUN_DEPENDS= devel/p5-B-Hooks-OP-Check=0.19 \ devel/p5-B-Hooks-EndOfScope \ devel/p5-Sub-Name +REGRESS_DEPENDS= devel/p5-Test-Warn .include bsd.port.mk diff --git a/devel/p5-Devel-Declare/distinfo b/devel/p5-Devel-Declare/distinfo index a5d626a..cf38fa5 100644 --- a/devel/p5-Devel-Declare/distinfo +++ b/devel/p5-Devel-Declare/distinfo @@ -1,5 +1,5 @@ -MD5 (Devel-Declare-0.005011.tar.gz) = CoH7PFQHsuj7YBdHHGdiKQ== -RMD160 (Devel-Declare-0.005011.tar.gz) = 3i0cYQV23aHj2J5mX6YY9JZftAo= -SHA1 (Devel-Declare-0.005011.tar.gz) = LKOhz1kGx9g02Wg4rDBbOOos64Y= -SHA256 (Devel-Declare-0.005011.tar.gz) = LF8szyIxs8MihcT2z1F6aIgUNMz31fpJMd+SMMYlHIo= -SIZE (Devel-Declare-0.005011.tar.gz) = 44831 +MD5 (Devel-Declare-0.006007.tar.gz) = rGSc92xcV4Za0rq7MhZtiA== +RMD160 (Devel-Declare-0.006007.tar.gz) = JsR4NvZLD+GqgF10wSOFq75/xN4= +SHA1 (Devel-Declare-0.006007.tar.gz) = 3MoEW4YBXnh86gyJK8PzwrjvvhY= +SHA256 (Devel-Declare-0.006007.tar.gz) = 4fGENoXvxKFBHJZ+cy0/UzB8bbx8lnKcIbmihozkVC8= +SIZE (Devel-Declare-0.006007.tar.gz) = 55678 diff --git a/devel/p5-MooseX-Declare/Makefile b/devel/p5-MooseX-Declare/Makefile index d247446..4d6c16a 100644 --- a/devel/p5-MooseX-Declare/Makefile +++ b/devel/p5-MooseX-Declare/Makefile @@ -2,8 +2,7 @@ COMMENT= declarative syntax for Moose -DISTNAME= MooseX-Declare-0.32 -REVISION= 0 +DISTNAME= MooseX-Declare-0.35 CATEGORIES=devel @@ -14,10 +13,9 @@ PERMIT_DISTFILES_CDROM= Yes PERMIT_DISTFILES_FTP= Yes MODULES= cpan -USE_GROFF =Yes RUN_DEPENDS= devel/p5-List-MoreUtils \ - devel/p5-MooseX-Method-Signatures \ + devel/p5-MooseX-Method-Signatures=0.37 \ devel/p5-MooseX-Role-Parameterized \ devel/p5-MooseX-Types \ devel/p5-namespace-autoclean diff --git
Re: UPDATE: exim 4.77
Exim 4.77 works fine here on i386 and amd64. I use the non-flavoured version with Perl, SpamAssassin, DKIM, IPv6 etc.
Re: spamassassin update, please test
Am 06/22/11 16:36, schrieb Stuart Henderson: please test and report back, thanks. Works fine here, including sa-update and sa-compile. Running on i386 with Exim.
Re: [Update] databases/p5-DBD-Mock
Wen Heping writes: Hi, Here is an update of databases/p5-DBD-Mock to 1.39. Tested on Loongson and regress test OK. Comments? OK ? I've been using version 1.39 of DBD::Mock for months because of a local port of Fey that requires an updated version. p5-DBD-Mock is used by the regression tests of the ports p5-CGI-Application-Plugin-DBH and p5-DBIx-DataModel. Both ports still pass their regression tests successfully. Tested on amd64. Wen, there's a message from Marc Espie's on the removal of USE_GROFF: Lhttp://article.gmane.org/gmane.os.openbsd.ports/45798. Also, there's still an issue where -- is rendered as O- in POD-based manual pages; see Lhttp://mdocml.bsd.lv/archives/tech/0255.html. Run pod2man DBD-Mock-1.39/lib/DBD/Mock.pm | mandoc and look for O- in the output. Regards, Andreas
Re: Update to exim 4.75
Felix Kronlage writes: Hi, attached diff updates exim to 4.75. One patch is removed, since this is now part of exim. Works fine here on i386.
Re: [Update]devel/p5-Package-Stash
wen heping writes:
Here's an update of devel/p5-Package-Stash to 0.27 ,
it is required by some others ports' update.
Upstream is already at version 0.28, which only fixes META.json though.
Tested on loongson OK. Regress test OK too.
Comments? Ok?
wen
REVISION= 0
I think this line can be removed as the package version got bumped.
+RUN_DEPENDS= devel/p5-Package-DeprecationManager \
+ devel/p5-Package-Stash-XS=0.22
By always depending on p5-Package-Stash-XS the port can only be used on
platforms that support shared objects. But I think that's okay as the
ports that depend on p5-Package-Stash depend on shared libraries anyway.
Attached is a port of Dist::CheckConflicts, which is used by
Package::Stash during configuration. I don't know if this port is of any
use but as I created it a while ago I've decided to attach the archive.
The port could be added to the build dependencies:
BUILD_DEPENDS = ${RUN_DEPENDS} devel/p5-Dist-CheckConflicts
IIRC the update of p5-Package-Stash also requires updates to
p5-namespace-clean and p5-namespace-autoclean. Patches are attached.
+REGRESS_DEPENDS= devel/p5-Test-Exception \
As far as I can see Test::Exception is no longer used. According to the
changelog Test::Fatal is used instead.
Regards,
Andreas
p5-Dist-CheckConflicts-0.02.tar.gz
Description: application/gzip
Index: Makefile
===
RCS file: /cvs/ports/devel/p5-namespace-autoclean/Makefile,v
retrieving revision 1.4
diff -u -r1.4 Makefile
--- Makefile3 Dec 2010 11:44:38 - 1.4
+++ Makefile4 Apr 2011 09:35:18 -
@@ -3,8 +3,8 @@
COMMENT = keep imports out of your namespace
MODULES = cpan
-DISTNAME = namespace-autoclean-0.09
-REVISION = 0
+DISTNAME = namespace-autoclean-0.12
+CPAN_AUTHOR = BOBTFISH
CATEGORIES = devel
USE_GROFF =Yes
@@ -18,7 +18,7 @@
RUN_DEPENDS = devel/p5-B-Hooks-EndOfScope \
devel/p5-Class-MOP \
- devel/p5-namespace-clean
+ devel/p5-namespace-clean=0.20
REGRESS_DEPENDS = devel/p5-Moose
Index: distinfo
===
RCS file: /cvs/ports/devel/p5-namespace-autoclean/distinfo,v
retrieving revision 1.1.1.1
diff -u -r1.1.1.1 distinfo
--- distinfo26 Oct 2009 16:30:35 - 1.1.1.1
+++ distinfo4 Apr 2011 09:35:18 -
@@ -1,5 +1,5 @@
-MD5 (namespace-autoclean-0.09.tar.gz) = eSQetJKc327s1uXMTB7y+A==
-RMD160 (namespace-autoclean-0.09.tar.gz) = NDAY12b4JCmkfuyaWNybT0OWDrs=
-SHA1 (namespace-autoclean-0.09.tar.gz) = /woR7/51wKjeZBjl+K0Y/Cgk6X4=
-SHA256 (namespace-autoclean-0.09.tar.gz) =
hzDhWhDlmB5BfQb6NpDNVNW8hFNl4q80hxNEhpVbMok=
-SIZE (namespace-autoclean-0.09.tar.gz) = 10993
+MD5 (namespace-autoclean-0.12.tar.gz) = bEM/3LBw7sA+ILHi00u/Eg==
+RMD160 (namespace-autoclean-0.12.tar.gz) = LxL/ZgVrRV6qfYIKkBYI8WLp4wI=
+SHA1 (namespace-autoclean-0.12.tar.gz) = c/APOOUfSAYUZlbTb6Efpi7LMcc=
+SHA256 (namespace-autoclean-0.12.tar.gz) =
76YqKYnHPJeeLORqm64pMfp/owwQq66DnMlhTbwtr4g=
+SIZE (namespace-autoclean-0.12.tar.gz) = 12768
Index: Makefile
===
RCS file: /cvs/ports/devel/p5-namespace-clean/Makefile,v
retrieving revision 1.8
diff -u -r1.8 Makefile
--- Makefile3 Dec 2010 11:44:38 - 1.8
+++ Makefile4 Apr 2011 09:34:40 -
@@ -3,8 +3,8 @@
COMMENT = keep imports and functions out of your namespace
MODULES = cpan
-DISTNAME = namespace-clean-0.18
-REVISION = 0
+DISTNAME = namespace-clean-0.20
+CPAN_AUTHOR = FLORA
CATEGORIES = devel
USE_GROFF =Yes
@@ -17,7 +17,7 @@
PERMIT_DISTFILES_FTP = Yes
RUN_DEPENDS = devel/p5-B-Hooks-EndOfScope=0.07 \
- devel/p5-Package-Stash \
+ devel/p5-Package-Stash=0.22 \
devel/p5-Sub-Identify \
devel/p5-Sub-Name
BUILD_DEPENDS =${RUN_DEPENDS}
Index: distinfo
===
RCS file: /cvs/ports/devel/p5-namespace-clean/distinfo,v
retrieving revision 1.4
diff -u -r1.4 distinfo
--- distinfo30 Jun 2010 18:16:27 - 1.4
+++ distinfo4 Apr 2011 09:34:40 -
@@ -1,5 +1,5 @@
-MD5 (namespace-clean-0.18.tar.gz) = qWSst3MkxoahENOLcwKoeQ==
-RMD160 (namespace-clean-0.18.tar.gz) = gPRdVgU41WN8CrLSB6Ani6+q5/U=
-SHA1 (namespace-clean-0.18.tar.gz) = lxGpLey1NDyTb0rdK4gJOTdRufY=
-SHA256 (namespace-clean-0.18.tar.gz) =
kmk88Cjfg+J54vpYatdV05J7M9X1V+p7G2bCBA9YTMc=
-SIZE (namespace-clean-0.18.tar.gz) = 17063
+MD5 (namespace-clean-0.20.tar.gz) = 5zdi2cu7zVcRJuxyS0bJYg==
+RMD160 (namespace-clean-0.20.tar.gz) = x5V2iF2kFfWYdh28eVj9vJqXZMA=
+SHA1 (namespace-clean-0.20.tar.gz) = LUerf3ZxlJNWSkPVXO4b+7l/YP8=
+SHA256 (namespace-clean-0.20.tar.gz) =
Nfx+TSPkEycXmRCQrttCeLPhoPK+JD44yZYEiBVDWtU=
+SIZE (namespace-clean-0.20.tar.gz) =
Re: exim Update to 4.73 (security)
Felix Kronlage writes: attached is a diff to update to exim 4.73. I saw no regression here, however exim has a few variations, so I'm happy about any feedback. Works fine here on i386 without flavours. I've attached a patch for files/Makefile that brings the file in line with src/EDITME. BTW, as the Exim port has always been built with ALT_CONFIG_ROOT_ONLY CVE-2010-4345 is not an issue on OpenBSD. CVE-2010-4344 is only an issue on OpenBSD 4.6 and before. Index: files/Makefile === RCS file: /cvs/ports/mail/exim/files/Makefile,v retrieving revision 1.12 diff -u -r1.12 Makefile --- files/Makefile 15 Dec 2009 11:32:18 - 1.12 +++ files/Makefile 7 Jan 2011 07:15:25 - @@ -1,4 +1,4 @@ -# $Cambridge: exim/exim-src/src/EDITME,v 1.23 2009/11/20 12:18:19 nm4 Exp $ +# $Cambridge: exim/src/src/EDITME,v 1.27 2010/06/12 15:21:25 jetmore Exp $ ## # The Exim mail transport agent # @@ -131,8 +131,7 @@ # group that is used for Exim processes when they no longer need to be root. In # particular, this applies when receiving messages and when doing remote # deliveries. (Local deliveries run as various non-root users, typically as the -# owner of a local mailbox.) Specifying these values as root is very strongly -# discouraged. +# owner of a local mailbox.) Specifying these values as root is not supported. EXIM_USER=ref:_exim @@ -352,6 +351,25 @@ WITH_OLD_DEMIME=yes +# If you're using ClamAV and are backporting fixes to an old version, instead +# of staying current (which is the more usual approach) then you may need to +# use an older API which uses a STREAM command, now deprecated, instead of +# zINSTREAM. If you need to set this, please let the Exim developers know, as +# if nobody reports a need for it, we'll remove this option and clean up the +# code. zINSTREAM was introduced with ClamAV 0.95. +# +# WITH_OLD_CLAMAV_STREAM=yes + +#-- +# By default Exim includes code to support DKIM (DomainKeys Identified +# Mail, RFC4871) signing and verification. Verification of signatures is +# turned on by default. See the spec for information on conditionally +# disabling it. To disable the inclusion of the entire feature, set +# DISABLE_DKIM to yes + +# DISABLE_DKIM=yes + + #-- # Compiling Exim with experimental features. These are documented in # experimental-spec.txt. Experimental means that the way these features are @@ -412,14 +430,13 @@ #-- -# By default, Exim insists that its configuration file be owned either by root -# or by the Exim user. You can specify one additional permitted owner here. +# By default, Exim insists that its configuration file be owned by root. You +# can specify one additional permitted owner here. # CONFIGURE_OWNER= # If the configuration file is group-writeable, Exim insists by default that it -# is owned by root or the Exim user. You can specify one additional permitted -# group owner here. +# is owned by root. You can specify one additional permitted group owner here. # CONFIGURE_GROUP= @@ -441,32 +458,31 @@ #-- # The -C option allows Exim to be run with an alternate runtime configuration -# file. When this is used by root or the Exim user, root privilege is retained -# by the binary (for any other caller, it is dropped). You can restrict the -# location of alternate configurations by defining a prefix below. Any file -# used with -C must then start with this prefix (except that /dev/null is also -# permitted if the caller is root, because that is used in the install script). -# If the prefix specifies a directory that is owned by root, a compromise of -# the Exim account does not permit arbitrary alternate configurations to be -# used. The prefix can be more restrictive than just a directory (the second -# example). +# file. When this is used by root, root privilege is retained by the binary +# (for any other caller including the Exim user, it is dropped). You can +# restrict the location of alternate configurations by defining a prefix below. +# Any file used with -C must then start with this prefix (except that /dev/null +# is also permitted if the caller is root, because that is used in the install +# script). If the prefix specifies a directory that is owned by root, a +# compromise of the Exim account does not permit arbitrary alternate +# configurations to be used. The prefix can be more restrictive than just a +# directory (the second example). # ALT_CONFIG_PREFIX=/some/directory/ # ALT_CONFIG_PREFIX=/some/directory/exim.conf-
UPDATE: devel/p5-MooseX-SemiAffordanceAccessor
Hi, attached is a patch that updates devel/p5-MooseX-SemiAffordanceAccessor from version 0.05 to version 0.08. Version 0.08 works with Moose 1.08, which is currently in OpenBSD's port tree, but also with newer versions of Moose. The module is used by p5-Audio-MPD and p5-File-ChangeNotify, which still pass their tests. Tested on i386. Upstream's most recent log entries: 0.08 2010-07-20 - Fix a bad example in the SYNOPSIS. 0.07 2010-07-14 - Fix bad uri for bugtracker in metadata 0.06 2010-07-14 - Use modern Moose APIs, to avoid warnings with the next Moose release. Index: Makefile === RCS file: /cvs/ports/devel/p5-MooseX-SemiAffordanceAccessor/Makefile,v retrieving revision 1.4 diff -u -r1.4 Makefile --- Makefile3 Dec 2010 11:44:33 - 1.4 +++ Makefile1 Jan 2011 10:36:15 - @@ -3,10 +3,8 @@ COMMENT = name your accessors foo() and set_foo() MODULES = cpan -DISTNAME = MooseX-SemiAffordanceAccessor-0.05 -REVISION = 0 +DISTNAME = MooseX-SemiAffordanceAccessor-0.08 CATEGORIES = devel -USE_GROFF =Yes # Perl PERMIT_PACKAGE_CDROM = Yes @@ -15,8 +13,6 @@ PERMIT_DISTFILES_FTP = Yes RUN_DEPENDS = devel/p5-Moose - -CONFIGURE_STYLE = modbuild MAKE_ENV +=TEST_POD=1 IS_MAINTAINER=1 Index: distinfo === RCS file: /cvs/ports/devel/p5-MooseX-SemiAffordanceAccessor/distinfo,v retrieving revision 1.1.1.1 diff -u -r1.1.1.1 distinfo --- distinfo27 Oct 2009 00:35:51 - 1.1.1.1 +++ distinfo1 Jan 2011 10:36:15 - @@ -1,5 +1,5 @@ -MD5 (MooseX-SemiAffordanceAccessor-0.05.tar.gz) = TpFxhdUKDeLcmUKmQ+t2vQ== -RMD160 (MooseX-SemiAffordanceAccessor-0.05.tar.gz) = po8TG1ykvMVIG9JrXOp7NYJco+A= -SHA1 (MooseX-SemiAffordanceAccessor-0.05.tar.gz) = jIVwvSXZ3LiOdlH50EgS1qtLT0U= -SHA256 (MooseX-SemiAffordanceAccessor-0.05.tar.gz) = MIRnEBhl0rNOQvMs8koATOURnS5So/jwSO2uLRxzAic= -SIZE (MooseX-SemiAffordanceAccessor-0.05.tar.gz) = 4687 +MD5 (MooseX-SemiAffordanceAccessor-0.08.tar.gz) = f7QHAxWyTPJoqbfN+zUHLw== +RMD160 (MooseX-SemiAffordanceAccessor-0.08.tar.gz) = rqIwvVpVIz2LcLM4x9hZKXoKBVg= +SHA1 (MooseX-SemiAffordanceAccessor-0.08.tar.gz) = mh0PJJ4raf/oPCTldL8qu0Z+UqQ= +SHA256 (MooseX-SemiAffordanceAccessor-0.08.tar.gz) = vQTt+eFweNTh8lnQCOIVspCDP1VwzNGOtY/fbSvb4nE= +SIZE (MooseX-SemiAffordanceAccessor-0.08.tar.gz) = 9157
UPDATE: devel/p5-File-ChangeNotify
Hi,
attached is a patch that updates devel/p5-File-ChangeNotify from version
0.13 to 0.19. The updated version works with OpenBSD's Moose 1.08 but
also with more recent Moose versions.
The patch also fixes a small bug in KQueue.pm, that I reported to upstream.
It seems that the port so far didn't take into account that the
p5-IO-KQueue package could be installed on the system. Since the
KQueue-based file watcher quickly exceeds the maximum number of open
files in a default OpenBSD installation, I changed ChangeNotify.pm to
prefer the Default watcher even if IO::KQueue is available. I think this
is more reasonable than to ask the users to raise their openfiles and
maxfiles limits from 128 and 7030 to much higher values.
Tested on i386.
The following ports depend on p5-File-ChangeNotify:
- audio/squeezecenter
- sysutils/p5-Filesys-Notify-Simple
- www/p5-Catalyst-Devel
Upstream's change log:
0.19 2010-10-17
- The all.t test didn't run any tests at all if neither the Inotify nor
KQueue watchers could be loaded. This was treated as a failure by test
harnesses. Now we always test the Default class.
0.18 2010-10-15
- Always make a Default watcher object if we cannot load an OS-specific
class.
- Add Test::Exception as a test prereq.
0.17 2010-10-04
- Loading File::ChangeNotify always loads the Default watcher class.
This will give a useful error message if this module's prereqs are
not loaded.
0.16 2010-07-12
- Changes to avoid a warning about a useless coercion from future
versions of Moose.
0.15 2010-07-09
- Add a missing prereq, namespace::autoclean.
0.14 2010-07-08
- Running the tests left behind a lot of temp directories that should
have been cleaned up, but weren't. Reported by Peter Edwards. RT
#59125.
- License is now Artistic 2.0
Index: Makefile
===
RCS file: /cvs/ports/devel/p5-File-ChangeNotify/Makefile,v
retrieving revision 1.5
diff -u -r1.5 Makefile
--- Makefile3 Dec 2010 11:44:31 - 1.5
+++ Makefile1 Jan 2011 12:18:29 -
@@ -3,23 +3,24 @@
COMMENT= watch for changes to files, cross-platform style
MODULES= cpan
-DISTNAME= File-ChangeNotify-0.13
-REVISION= 0
+DISTNAME= File-ChangeNotify-0.19
CATEGORIES=devel
-# Perl
+# Artistic 2.0
PERMIT_PACKAGE_CDROM = Yes
PERMIT_PACKAGE_FTP = Yes
PERMIT_DISTFILES_CDROM = Yes
PERMIT_DISTFILES_FTP = Yes
CONFIGURE_STYLE = modbuild
-USE_GROFF =Yes
+BUILD_DEPENDS = ${RUN_DEPENDS}
RUN_DEPENDS = devel/p5-Class-MOP \
devel/p5-Moose \
devel/p5-MooseX-Params-Validate \
- devel/p5-MooseX-SemiAffordanceAccessor
+ devel/p5-MooseX-SemiAffordanceAccessor \
+ devel/p5-namespace-autoclean
+REGRESS_DEPENDS = devel/p5-Test-Exception
MAKE_ENV +=TEST_POD=1 IS_MAINTAINER=1
Index: distinfo
===
RCS file: /cvs/ports/devel/p5-File-ChangeNotify/distinfo,v
retrieving revision 1.2
diff -u -r1.2 distinfo
--- distinfo3 Jul 2010 03:41:42 - 1.2
+++ distinfo1 Jan 2011 12:18:29 -
@@ -1,5 +1,5 @@
-MD5 (File-ChangeNotify-0.13.tar.gz) = eR70uZegW1BwTiwo+Ph4oA==
-RMD160 (File-ChangeNotify-0.13.tar.gz) = Jltiid+UQa63uHcm5mBS2KnQ69s=
-SHA1 (File-ChangeNotify-0.13.tar.gz) = WcEiFJwmsqrqCjYp2xNnbAlSbqM=
-SHA256 (File-ChangeNotify-0.13.tar.gz) =
yHAoVdthRnb9v2P9raOA6tj2FAE11Y1TkUQuDYmpaig=
-SIZE (File-ChangeNotify-0.13.tar.gz) = 18155
+MD5 (File-ChangeNotify-0.19.tar.gz) = VQwjU6nvRVIDu8AC7DGr+A==
+RMD160 (File-ChangeNotify-0.19.tar.gz) = AR00UgFZUSdOzuYz8k9v6hycoho=
+SHA1 (File-ChangeNotify-0.19.tar.gz) = kHUhoX7Qh7eHUQZHMhP2WNo+WaE=
+SHA256 (File-ChangeNotify-0.19.tar.gz) =
8UfmQPMm0uG3XrKZ+1LXcDXbEE+evsc71IuEhBOXAdA=
+SIZE (File-ChangeNotify-0.19.tar.gz) = 19908
Index: patches/patch-lib_File_ChangeNotify_Watcher_KQueue_pm
===
RCS file: patches/patch-lib_File_ChangeNotify_Watcher_KQueue_pm
diff -N patches/patch-lib_File_ChangeNotify_Watcher_KQueue_pm
--- /dev/null 1 Jan 1970 00:00:00 -
+++ patches/patch-lib_File_ChangeNotify_Watcher_KQueue_pm 1 Jan 2011
12:18:29 -
@@ -0,0 +1,23 @@
+$OpenBSD$
+--- lib/File/ChangeNotify/Watcher/KQueue.pm.orig Sun Oct 17 22:00:14 2010
lib/File/ChangeNotify/Watcher/KQueue.pmSat Jan 1 10:21:31 2011
+@@ -184,7 +184,7 @@ sub _watch_file {
+
+ # Don't panic if we can't open a file
+ open my $fh, '', $file or warn Can't open '$file': $!;
+-return unless $fh;
++return unless $fh defined fileno($fh);
+
+ # Store this filehandle (this will automatically nuke any existing events
+ # assigned to the file)
+@@ -229,6 +229,10 @@ On FreeBSD, you can check (and alter) your system's se
+ necessary. The important keys are: Ckern.maxfiles and
+ Ckern.maxfilesperproc. You can see how many
Re: UPDATE: devel/p5-File-ChangeNotify
Mikolaj Kucharski writes: On Sat, Jan 01, 2011 at 01:23:12PM +0100, Andreas Vögele wrote: It seems that the port so far didn't take into account that the p5-IO-KQueue package could be installed on the system. Since the KQueue-based file watcher quickly exceeds the maximum number of open files in a default OpenBSD installation, I changed ChangeNotify.pm to prefer the Default watcher even if IO::KQueue is available. I think this is more reasonable than to ask the users to raise their openfiles and maxfiles limits from 128 and 7030 to much higher values. Ok. What about users, who altered their system already, and have raised kern.maxfiles and openfiles-cur in login.conf(5) for example, for gam_server (see gamin package). Not that I am using ChangeNotify.pm, but what is default watcher in that module? The default is the Default watcher unless you have happened to install the p5-IO-KQueue package, which the current p5-File-ChangeNotify package does NOT depend on. If anybody uses p5-File-ChangeNotify _with_ p5-IO-KQueue _and_ has raised the resource limits that person should have complained a long time ago about the missing dependency on p5-IO-KQueue. Which watcher is more efficient? Does your change (to not use IO::KQueue as default watcher) help the user of ChangeNotify.pm in any way? In my opinion robustness is more important than efficiency. The ports and packages are supposed to make life simple for the users. If the KQueue-based watcher becomes the default, most users who install p5-File-ChangeNotify - very likely because of a dependency - will have to raise their resource limits. The default for the openfiles resource limit is 128, which means that the KQueue-based watcher will fail on any directory that contains more than 127 files. The Default watcher on the other hand is slower but can handle large directories without any modification to the system's configuration. As wrote above, I'm not using module in subject, but I have concerns for changing defaults.. I'm asking as I feel I'm missing something to understand reasoning. I doubt that the defaults are changed for anybody. On the contrary, so far the port doesn't depend on p5-IO-KQueue, which suggests that whoever created this port didn't consider the KQueue-based watcher. Actually, if the p5-IO-KQueue package is installed the regression tests of the current port fail because of a bug in KQueue.pm. My guess is that the port was merely created to satisfy dependencies. Also, if p5-IO-KQueue is pulled in by some other package, the current p5-File-ChangeNotify package might suddenly start to fail because the installation of p5-IO-KQueue changes the default from the Default watcher to the KQueue watcher. This is not what users expect. Furthermore, it's not possible to automatically decide at runtime whether to use the Default or the KQueue watcher. When the watcher is instantiated the size of the directory is not known. Also, directories might grow during runtime. Using the KQueue-based watcher is dangerous unless you know for sure that the size of the watched directories will never exceed your resource limits. A hybrid watcher that tries to use KQueue and falls back to the Default mechanism if a threshold of available file descriptors is reached could make sense though. But my goal is just to get this port updated so that eventually Moose can be updated.
Re: UPDATE: devel/p5-File-ChangeNotify
Andreas Vögele writes: Mikolaj Kucharski writes: On Sat, Jan 01, 2011 at 01:23:12PM +0100, Andreas Vögele wrote: It seems that the port so far didn't take into account that the p5-IO-KQueue package could be installed on the system. Since the KQueue-based file watcher quickly exceeds the maximum number of open files in a default OpenBSD installation, I changed ChangeNotify.pm to prefer the Default watcher even if IO::KQueue is available. I think this is more reasonable than to ask the users to raise their openfiles and maxfiles limits from 128 and 7030 to much higher values. Ok. What about users, who altered their system already, and have raised kern.maxfiles and openfiles-cur in login.conf(5) for example, for gam_server (see gamin package). Not that I am using ChangeNotify.pm, but what is default watcher in that module? The default is the Default watcher unless you have happened to install the p5-IO-KQueue package, which the current p5-File-ChangeNotify package does NOT depend on. [...] I will put the KQueue-based stuff into a sub package. That way the Default watcher stays the default until the optional sub package is installed, which will switch the default to the KQueue-based watcher.
Re: UPDATE: devel/p5-File-ChangeNotify
Mikolaj Kucharski writes:
So, File::ChangeNotify doesn't handle resource limits gracefuly,
yeah[ref#2]? If that is true, your changes make sense for me.
Yes, exactly. The KQueue-based watcher class doesn't fall back to
polling if it runs out of file descriptors.
I've attached an updated diff. This version of the port creates a
p5-File-ChangeNotify-Watcher-KQueue package in addition to the main
package. The main package provides the slow default watcher that doesn't
require any further configuration. As soon as the sub package is
installed the KQueue-based watcher becomes the default.
Index: Makefile
===
RCS file: /cvs/ports/devel/p5-File-ChangeNotify/Makefile,v
retrieving revision 1.5
diff -u -r1.5 Makefile
--- Makefile3 Dec 2010 11:44:31 - 1.5
+++ Makefile1 Jan 2011 17:25:18 -
@@ -1,25 +1,34 @@
# $OpenBSD: Makefile,v 1.5 2010/12/03 11:44:31 ajacoutot Exp $
-COMMENT= watch for changes to files, cross-platform style
+COMMENT-main= watch for changes to files, cross-platform style
+COMMENT-kqueue=watch for changes to files (KQueue-based)
MODULES= cpan
-DISTNAME= File-ChangeNotify-0.13
-REVISION= 0
+DISTNAME= File-ChangeNotify-0.19
+PKGNAME-main= p5-File-ChangeNotify-0.19
+PKGNAME-kqueue=p5-File-ChangeNotify-Watcher-KQueue-0.19
CATEGORIES=devel
-# Perl
+# Artistic 2.0
PERMIT_PACKAGE_CDROM = Yes
PERMIT_PACKAGE_FTP = Yes
PERMIT_DISTFILES_CDROM = Yes
PERMIT_DISTFILES_FTP = Yes
+MULTI_PACKAGES = -main -kqueue
+
CONFIGURE_STYLE = modbuild
-USE_GROFF =Yes
-RUN_DEPENDS = devel/p5-Class-MOP \
- devel/p5-Moose \
- devel/p5-MooseX-Params-Validate \
- devel/p5-MooseX-SemiAffordanceAccessor
+BUILD_DEPENDS =${RUN_DEPENDS-main} \
+ devel/p5-IO-KQueue
+RUN_DEPENDS-main = devel/p5-Class-MOP \
+ devel/p5-Moose \
+ devel/p5-MooseX-Params-Validate \
+ devel/p5-MooseX-SemiAffordanceAccessor \
+ devel/p5-namespace-autoclean
+RUN_DEPENDS-kqueue = devel/p5-File-ChangeNotify \
+ devel/p5-IO-KQueue
+REGRESS_DEPENDS = devel/p5-Test-Exception
MAKE_ENV +=TEST_POD=1 IS_MAINTAINER=1
Index: distinfo
===
RCS file: /cvs/ports/devel/p5-File-ChangeNotify/distinfo,v
retrieving revision 1.2
diff -u -r1.2 distinfo
--- distinfo3 Jul 2010 03:41:42 - 1.2
+++ distinfo1 Jan 2011 17:25:18 -
@@ -1,5 +1,5 @@
-MD5 (File-ChangeNotify-0.13.tar.gz) = eR70uZegW1BwTiwo+Ph4oA==
-RMD160 (File-ChangeNotify-0.13.tar.gz) = Jltiid+UQa63uHcm5mBS2KnQ69s=
-SHA1 (File-ChangeNotify-0.13.tar.gz) = WcEiFJwmsqrqCjYp2xNnbAlSbqM=
-SHA256 (File-ChangeNotify-0.13.tar.gz) =
yHAoVdthRnb9v2P9raOA6tj2FAE11Y1TkUQuDYmpaig=
-SIZE (File-ChangeNotify-0.13.tar.gz) = 18155
+MD5 (File-ChangeNotify-0.19.tar.gz) = VQwjU6nvRVIDu8AC7DGr+A==
+RMD160 (File-ChangeNotify-0.19.tar.gz) = AR00UgFZUSdOzuYz8k9v6hycoho=
+SHA1 (File-ChangeNotify-0.19.tar.gz) = kHUhoX7Qh7eHUQZHMhP2WNo+WaE=
+SHA256 (File-ChangeNotify-0.19.tar.gz) =
8UfmQPMm0uG3XrKZ+1LXcDXbEE+evsc71IuEhBOXAdA=
+SIZE (File-ChangeNotify-0.19.tar.gz) = 19908
Index: patches/patch-lib_File_ChangeNotify_Watcher_KQueue_pm
===
RCS file: patches/patch-lib_File_ChangeNotify_Watcher_KQueue_pm
diff -N patches/patch-lib_File_ChangeNotify_Watcher_KQueue_pm
--- /dev/null 1 Jan 1970 00:00:00 -
+++ patches/patch-lib_File_ChangeNotify_Watcher_KQueue_pm 1 Jan 2011
17:25:18 -
@@ -0,0 +1,23 @@
+$OpenBSD$
+--- lib/File/ChangeNotify/Watcher/KQueue.pm.orig Sun Oct 17 22:00:14 2010
lib/File/ChangeNotify/Watcher/KQueue.pmSat Jan 1 10:21:31 2011
+@@ -184,7 +184,7 @@ sub _watch_file {
+
+ # Don't panic if we can't open a file
+ open my $fh, '', $file or warn Can't open '$file': $!;
+-return unless $fh;
++return unless $fh defined fileno($fh);
+
+ # Store this filehandle (this will automatically nuke any existing events
+ # assigned to the file)
+@@ -229,6 +229,10 @@ On FreeBSD, you can check (and alter) your system's se
+ necessary. The important keys are: Ckern.maxfiles and
+ Ckern.maxfilesperproc. You can see how many files your system current has
+ open with Ckern.openfiles.
++
++On OpenBSD, the Csysctl keys are Ckern.maxfiles and Ckern.nfiles.
++Per-process limits are set in F/etc/login.conf. See Llogin.conf(5) and
++LBSD::Resource for details.
+
+ =head1 SUPPORT
+
Index: pkg/DESCR
===
RCS file: pkg/DESCR
diff -N pkg/DESCR
--- pkg/DESCR 27 Oct 2009 00:31:20 - 1.1.1.1
+++ /dev/null 1 Jan 1970 00:00:00 -
@@ -1,2 +0,0 @@
-Watch for changes to files, cross-platform style via
UPDATE: security/p5-Crypt-OpenSSL-RSA
Hi,
attached is a patch that updates p5-Crypt-OpenSSL-RSA to version 0.26.
The new version fixes a regression test that failed on Red Hat Linux,
i.e. nothing changes on OpenBSD. But I've updated my mail address,
inserted spaces, checked the manual page and removed USE_GROFF. Tested
on i386.
Index: Makefile
===
RCS file: /cvs/ports/security/p5-Crypt-OpenSSL-RSA/Makefile,v
retrieving revision 1.13
diff -u -r1.13 Makefile
--- Makefile20 Nov 2010 17:22:44 - 1.13
+++ Makefile26 Dec 2010 11:38:54 -
@@ -1,29 +1,27 @@
# $OpenBSD: Makefile,v 1.13 2010/11/20 17:22:44 espie Exp $
-SHARED_ONLY= Yes
+SHARED_ONLY = Yes
-COMMENT= RSA encoding and decoding using OpenSSL
+COMMENT = RSA encoding and decoding using OpenSSL
-MODULES= cpan
-USE_GROFF =Yes
+MODULES = cpan
-DISTNAME= Crypt-OpenSSL-RSA-0.25
-REVISION= 1
-CATEGORIES=security
-HOMEPAGE= http://perl-openssl.sourceforge.net/
+DISTNAME = Crypt-OpenSSL-RSA-0.26
+CATEGORIES = security
+HOMEPAGE = http://perl-openssl.sourceforge.net/
-MAINTAINER=Andreas Voegele andr...@kurzegasse.de
+MAINTAINER = Andreas Voegele po...@andreasvoegele.com
# Perl
-PERMIT_PACKAGE_CDROM= Yes
-PERMIT_PACKAGE_FTP=Yes
-PERMIT_DISTFILES_CDROM=Yes
-PERMIT_DISTFILES_FTP= Yes
+PERMIT_PACKAGE_CDROM = Yes
+PERMIT_PACKAGE_FTP = Yes
+PERMIT_DISTFILES_CDROM = Yes
+PERMIT_DISTFILES_FTP = Yes
-WANTLIB= crypto ssl
+WANTLIB = crypto ssl
-RUN_DEPENDS= security/p5-Crypt-OpenSSL-Random=0.03 \
+RUN_DEPENDS = security/p5-Crypt-OpenSSL-Random=0.03 \
security/p5-Crypt-OpenSSL-Bignum=0.03
-BUILD_DEPENDS= ${RUN_DEPENDS}
+BUILD_DEPENDS =${RUN_DEPENDS}
.include bsd.port.mk
Index: distinfo
===
RCS file: /cvs/ports/security/p5-Crypt-OpenSSL-RSA/distinfo,v
retrieving revision 1.6
diff -u -r1.6 distinfo
--- distinfo18 Jul 2007 09:41:07 - 1.6
+++ distinfo26 Dec 2010 11:38:54 -
@@ -1,5 +1,5 @@
-MD5 (Crypt-OpenSSL-RSA-0.25.tar.gz) = /fGckJP0fP+4Ua6TfQU8FA==
-RMD160 (Crypt-OpenSSL-RSA-0.25.tar.gz) = x9+EJSOPyNYtshe2jvlEGkMNvaE=
-SHA1 (Crypt-OpenSSL-RSA-0.25.tar.gz) = QVl7/Aaqj4FtnU5PDcKEBevpxAc=
-SHA256 (Crypt-OpenSSL-RSA-0.25.tar.gz) =
eXc0O11bswXEzXHT80OqEvc+eOMYXpkjV5sb1c1FtgE=
-SIZE (Crypt-OpenSSL-RSA-0.25.tar.gz) = 19133
+MD5 (Crypt-OpenSSL-RSA-0.26.tar.gz) = uvh18B7jm4gzW48JYv5LvA==
+RMD160 (Crypt-OpenSSL-RSA-0.26.tar.gz) = FIRvD7BP4NoOP84jdJVVzxJVrXI=
+SHA1 (Crypt-OpenSSL-RSA-0.26.tar.gz) = xWbKE+tYcSEFYJ/cAwc4dSbSYNA=
+SHA256 (Crypt-OpenSSL-RSA-0.26.tar.gz) =
9C8nb/5a4XdcS9dvMCZIPw05l1983Qe/SOzPuMRShrs=
+SIZE (Crypt-OpenSSL-RSA-0.26.tar.gz) = 19364
UPDATE: devel/p5-Devel-Caller
Hi, attached is a patch that updates p5-Devel-Caller to version 0.26 and removes USE_GROFF, which isn't required by this port. The last two entries from upstream's changelog: 2.05 Thursday 8th April, 2010 Don't call B::PADOP-gv. Work by Florian Ragwitz 2.04 Tuesday 16th February, 2010 Use CxTYPE macro rather than directly inspecting cx-cx_type Without the fix from version 2.04 a test fails when testing Devel::LexAlias. p5-MooseX-Params-Validate depends on p5-Devel-Caller and still passes its regression tests. Tested on i386. Index: Makefile === RCS file: /cvs/ports/devel/p5-Devel-Caller/Makefile,v retrieving revision 1.5 diff -u -r1.5 Makefile --- Makefile3 Dec 2010 11:44:30 - 1.5 +++ Makefile26 Dec 2010 19:12:00 - @@ -4,11 +4,9 @@ SHARED_ONLY = Yes MODULES = cpan -DISTNAME = Devel-Caller-2.03 -REVISION= 1 +DISTNAME = Devel-Caller-2.05 CATEGORIES = devel CPAN_AUTHOR = RCLAMP -USE_GROFF =Yes # Perl PERMIT_PACKAGE_CDROM = Yes Index: distinfo === RCS file: /cvs/ports/devel/p5-Devel-Caller/distinfo,v retrieving revision 1.1.1.1 diff -u -r1.1.1.1 distinfo --- distinfo27 Oct 2009 00:28:50 - 1.1.1.1 +++ distinfo26 Dec 2010 19:12:00 - @@ -1,5 +1,5 @@ -MD5 (Devel-Caller-2.03.tar.gz) = VPb87oWHfANdecDFxPZAPw== -RMD160 (Devel-Caller-2.03.tar.gz) = V1P3BoriYvykk/D+O1y+1kIUzG4= -SHA1 (Devel-Caller-2.03.tar.gz) = EqLBmR2wRcWBhBPmiUmHgnN8O5k= -SHA256 (Devel-Caller-2.03.tar.gz) = J0MiqgsqHY3iXbcBsO9lhMflcsqjW6VxbaAr2SxE+IQ= -SIZE (Devel-Caller-2.03.tar.gz) = 5497 +MD5 (Devel-Caller-2.05.tar.gz) = LRlzGM6OjXNvI7f3Ueiwkw== +RMD160 (Devel-Caller-2.05.tar.gz) = m4BtrRaQDnvF1TU7dSGYJPUUCQQ= +SHA1 (Devel-Caller-2.05.tar.gz) = V0EgIa4Hvtn/SVVSPOxYp1YORvs= +SHA256 (Devel-Caller-2.05.tar.gz) = 3PtZAEQnfhJeePeBoVAZipTIl2mvb6q49USpFt+7Q4g= +SIZE (Devel-Caller-2.05.tar.gz) = 5716 Index: patches/patch-lib_Devel_Caller_pm === RCS file: patches/patch-lib_Devel_Caller_pm diff -N patches/patch-lib_Devel_Caller_pm --- /dev/null 1 Jan 1970 00:00:00 - +++ patches/patch-lib_Devel_Caller_pm 26 Dec 2010 19:12:00 - @@ -0,0 +1,11 @@ +$OpenBSD$ +--- lib/Devel/Caller.pm.orig Thu Apr 8 14:39:39 2010 lib/Devel/Caller.pmSun Dec 26 20:01:30 2010 +@@ -210,6 +210,7 @@ constants are returned as Cundef in both cases + Ccalled_as_method returns true if the subroutine at $level was + called as a method. + ++=back + + =head1 BUGS +
NEW: devel/p5-Devel-REPL and required modules
Hi, Devel::REPL is an interactive shell for Perl. You can run the command re.pl and then evaluate Perl statements. For example: % re.pl $ use MIME::Base64 $ encode_base64(hello, world) aGVsbG8sIHdvcmxk See http://search.cpan.org/dist/Devel-REPL/ for details. Tested on i386. I've attached ports for Devel::REPL and several modules that are required by Devel::REPL. The Devel::LexAlias module depends on the updated p5-Devel-Caller port that I sent earlier. Also, Devel::LexAlias builds a shared object. The Clipboard module depends on xclip, i.e. xbase is required. mandoc formats all manual pages properly. The new ports are: devel/p5-Devel-REPL - an interactive shell for Perl devel/p5-MooseX-Object-Pluggable - make your Moose-based classes pluggable devel/p5-Data-Dumper-Concise - dump with less indentation and newlines plus sub deparsing devel/p5-Devel-ArgNames - figure out the names of variables passed into subroutines devel/p5-Lexical-Persistence - persistent lexical variable values for arbitrary calls devel/p5-Devel-LexAlias - alias lexical variables www/p5-App-Nopaste - easy access to any pastebin www/p5-WWW-Pastebin-PastebinCom-Create - paste text to http://pastebin.com x11/p5-Clipboard - copy and paste text onto and from the clipboard devel/p5-Config-GitLike - git-compatible config file parsing p5-App-Nopaste-0.24-port.tar.gz Description: application/gzip p5-Clipboard-0.13-port.tar.gz Description: application/gzip p5-Config-GitLike-1.04-port.tar.gz Description: application/gzip p5-Data-Dumper-Concise-2.012-port.tar.gz Description: application/gzip p5-Devel-ArgNames-0.03-port.tar.gz Description: application/gzip p5-Devel-LexAlias-0.04-port.tar.gz Description: application/gzip p5-Devel-REPL-1.003012-port.tar.gz Description: application/gzip p5-Lexical-Persistence-1.020-port.tar.gz Description: application/gzip p5-MooseX-Object-Pluggable-0.0011-port.tar.gz Description: application/gzip p5-WWW-Pastebin-PastebinCom-Create-0.003-port.tar.gz Description: application/gzip
Re: UPDATE: p5-Alien-wxWidgets-0.47.diff
Cesare Gargano writes: Oh, I agree... I forgot to apply your patch, sorry :( This could be ok? Cesare On Tue, Dec 08, 2009 at 06:49:42PM +0100, Landry Breuil wrote: On Tue, Dec 08, 2009 at 05:13:21PM +, Cesare Gargano wrote: Hi! This updates p5-Alien-wxWidgets to 0.47 I replaced a 1 with a 2 in the patch, I don't like their method to get our wxWidgets version ;( No, sorry, this won't work. Please keep the patch as is. There is a reason i did it that way : we want to be able to update wxwidgets without breaking/patching alien-wxwidgets. I don't care if you like it or not, but hardcoding so versions like this in patches is a sure way to break things at one moment. Landry BTW, upstream is interested in getting Alien::wxWidgets fixed on OpenBSD. Some time ago James Wright created a bug report, which is available at Lhttps://rt.cpan.org/Public/Bug/Display.html?id=41678.
Re: UPDATE: exim-4.71
Björn Ketelaars writes: OK, I think I've got them all: * Update to Exim 4.71 by Bernd; * Added a bit more information to MESSAGE concerning the use of mailwrapper; * Removed information from MESSAGE concerning converting from Exim 3; * Removed Exim 3 converting script from package; * Changed no_x11 flavor to -x11 subpackage (also added PSEUDO_FLAVOR no_x11 for those who like to compile in a X11-less environment); I'd call the subpackage eximon and not exim-x11, after all it contains eximon and its highly unlikely that anything else will ever go into this subpackage. * Removed WITH_OLD_DEMIM from no_exiscan flavor. This exim condition is deprecated; I wouldn't remove WITH_OLD_DEMIME as long as upstream supports the demime condition. Removing this feature might break existing setups. At least, there ought to be a hint in the MESSAGE file on how to replace the demime condition. There's a chapter called Content scanning at ACL time in the Exim manual that explains the demime and decode conditions. * Removed iconv flavor, whereby libiconv is changed from optional to mandatory; I'd also remove the no_exiscan flavor. Firstly, that flavor comes from a time when exiscan was distributed as a separate patch. Secondly, removing this flavor won't break existing setups. Thirdly, I doubt that anybody uses the no_exiscan flavor. Everbody I know uses Exim because of its content scanning features. Personally, I'd also put all the other flavors into a single flavor called heavy. Even if you are going to use LDAP only nothing will break if the PostgreSQL, MySQL and SASL client libraries get installed. Can OpenBSD's ports system handle upgrades from exim, exim-iconv and exim-no_exiscan to exim and from all other flavors to exim-heavy? * Use of syslog instead of seperate logfiles under /var/spool/exim/logs/. I don't like this change very much. I've always bothered that the log files aren't stored under /var/log/exim and, actually, I also use syslog, but this is a significant change and IMHO packages should try to keep changes as minimal as possible. For example, people might have set up scripts that monitor the log files stored in /var/spool/exim/logs/. At least, this change ought to be documented and announced properly. Kind regards, Andreas
Re: UPDATE: exim-4.71
Felix Kronlage schrieb: I agree. Does anything speak against reducing the number of flavour by: [...] - removing sqlite3 flavor (I don't see the sense in this flavour, the port had it when I took it over from the previous maintainer, but I fail to the see the reason for it.) When I worked on the Exim port I kept the existing flavors only because I didn't know whether OpenBSD's port system could handle upgrades from one flavor to another smoothly. I'd put the mysql, postgresql, sqlite3, ldap and sasl flavors into a single flavor called heavy. The database flavors only pull in the client libraries, not the server packages. As far as I can see these flavors don't need to be kept separate. BTW, SQLite3 support in Exim is useful. For example, I use SQLite3 databases to store greylisting data. Regards, Andreas
Re: UPDATE: exim-4.71
Felix Kronlage writes: On Sat, Dec 12, 2009 at 01:10:08PM +0100, Björn Ketelaars wrote: * Use of syslog instead of seperate logfiles under /var/spool/exim/logs/. /var/spool/exim/logs is the standard location for the exim logs, I'm not sure I see the reason to change this. What is the reasoning behind this change? Of course, I don't know what motivated Björn, but sometimes I prefer to keep all log files in a separate file system mounted at /var/log so that the rest of /var isn't filled when the log files explode.
Re: UPDATE: exim-4.71
Björn Ketelaars writes: On 12-12-2009 15:52, Andreas Vögele wrote: [...] When I worked on the Exim port I kept the existing flavors only because I didn't know whether OpenBSD's port system could handle upgrades from one flavor to another smoothly. I'd put the mysql, postgresql, sqlite3, ldap and sasl flavors into a single flavor called heavy. The database flavors only pull in the client libraries, not the server packages. As far as I can see these flavors don't need to be kept separate. [...] If one want to keep changes as minimal as possible I would prefer to keep the database/ldap/sasl flavors separated. Yes, but I meant that from the end users perspective. If the mysql, postgresl, ldap etc. flavors are replaced by a single flavor or less flavors that work for them all and don't require unnecessary changes on the end users side, changes to the port are IMHO fine.
FIX: mail/mozilla-thunderbird
The isp sub directory is missing from the mozilla-thunderbird package.
Without this directory RSS and movemail accounts cannot be created and
the junk filter cannot be told to handle message headers from SpamAssassin.
Regards,
Andreas
Index: Makefile
===
RCS file: /cvs/ports/mail/mozilla-thunderbird/Makefile,v
retrieving revision 1.82
diff -u -r1.82 Makefile
--- Makefile23 Jun 2009 17:05:00 - 1.82
+++ Makefile11 Jul 2009 18:48:36 -
@@ -7,7 +7,7 @@
VER= 2.0.0.22
DISTNAME= mozilla
-PKGNAME= mozilla-thunderbird-${VER}
+PKGNAME= mozilla-thunderbird-${VER}p0
SO_VERSION=13.0
# NOTE: Must bump minor version if any shlib's are removed from the
# components dir to avoid pkg_add -r issues. Whenever PKGNAME gets
@@ -105,7 +105,7 @@
MOZ= ${PREFIX}/mozilla-thunderbird
DATADIRS= chrome components defaults dictionaries extensions \
- greprefs icons init.d res
+ greprefs icons init.d res isp
post-extract:
@cp -f ${FILESDIR}/nsSound.cpp ${WRKSRC}/widget/src/gtk2/
Index: pkg/PLIST
===
RCS file: /cvs/ports/mail/mozilla-thunderbird/pkg/PLIST,v
retrieving revision 1.15
diff -u -r1.15 PLIST
--- pkg/PLIST 9 Jan 2009 15:54:31 - 1.15
+++ pkg/PLIST 11 Jul 2009 18:48:36 -
@@ -253,6 +253,13 @@
mozilla-thunderbird/icons/mozicon50.xpm
mozilla-thunderbird/init.d/
mozilla-thunderbird/init.d/README
+mozilla-thunderbird/isp/
+mozilla-thunderbird/isp/SpamAssassin.sfd
+mozilla-thunderbird/isp/SpamPal.sfd
+mozilla-thunderbird/isp/en-US/
+mozilla-thunderbird/isp/en-US/gmail.rdf
+mozilla-thunderbird/isp/movemail.rdf
+mozilla-thunderbird/isp/rss.rdf
@bin mozilla-thunderbird/mozilla-thunderbird-bin
@bin mozilla-thunderbird/mozilla-xremote-client
@bin mozilla-thunderbird/regxpcom
WIP: lyx-1.6.2
Here's a port for LyX 1.6.2, which was released this week. Tested on i386. Unfortunately, the spell checker window bug is still there. If a new and empty document is created and the spell checker is called through the tool bar or the menu bar, the spell checker's window isn't mappe/closed properly. At least if your window manager is cwm, dwm or twm. I can't reproduce this behaviour with fvwm and mwm. This bug might be related to Lhttp://bugzilla.lyx.org/show_bug.cgi?id=2218#c15. Apart from this problem LyX 1.6.2 seems to work fine. LyX 1.6 requires Qt 4. XForms is no longer supported. The port depends on the boost package from the ports tree instead of the included boost sources. patch-configure is gone. The port for 1.6.2 adds a new patch and the compiler flag -fno-inline to work around bugs in g++ 3.3. lyx-1.6.2.tar.gz Description: OpenBSD port of LyX 1.6.2
FIX: x11/wxWidgets
The socket code in wxWidgets doesn't initialize the servent_data structure that is passed to getservbyname_r(). Programs that use this code die with a segfault. The attached patch zero-fills the buffer with memset(). I've sent a bug report to the upstream maintainers (ticket #10332). ? wxWidgets-2.8.7-port.diff Index: Makefile === RCS file: /cvs/ports/x11/wxWidgets/Makefile,v retrieving revision 1.13 diff -u -r1.13 Makefile --- Makefile 11 Nov 2008 17:19:34 - 1.13 +++ Makefile 5 Jan 2009 17:26:24 - @@ -5,7 +5,7 @@ COMMENT= c++ cross-platform GUI toolkit DISTNAME= wxWidgets-2.8.7 -PKGNAME= wxWidgets-gtk2-2.8.7p0 +PKGNAME= wxWidgets-gtk2-2.8.7p1 SHARED_LIBS= wx_base 1.0 \ wx_base_net 1.0 \ wx_base_odbc 1.0 \ Index: patches/patch-src_unix_gsocket_cpp === RCS file: patches/patch-src_unix_gsocket_cpp diff -N patches/patch-src_unix_gsocket_cpp --- /dev/null 1 Jan 1970 00:00:00 - +++ patches/patch-src_unix_gsocket_cpp 5 Jan 2009 17:26:24 - @@ -0,0 +1,11 @@ +$OpenBSD$ +--- src/unix/gsocket.cpp.orig Wed Nov 21 13:43:17 2007 src/unix/gsocket.cpp Mon Jan 5 16:26:31 2009 +@@ -2116,6 +2116,7 @@ GSocketError GAddress_INET_SetPortName(GAddress *addre + char buffer[1024]; + #endif + struct servent serv; ++ memset(buffer, 0, sizeof(buffer)); + se = wxGetservbyname_r(port, protocol, serv, + (void*)buffer, sizeof(buffer)); + if (!se)
Re: OpenBSD LyX port update to 1.6 ?
Zvezdan Petkovic writes: On Dec 3, 2008, at 11:15 AM, Ljung, Peter wrote: Hi, I noticed that LyX has updated to version 1.6. recently. I'm just wondering if you are planning for an update on the OpenBSD port to version 1.6. [...] I'm sorry to disappoint you, but I dropped the maintenance of LyX, Xcircuit, and Unison ports this summer due to the lack of time. I do not know if anybody has taken over the maintenance or not. If nobody else maintains LyX now, you will have to rely on the kindness of OpenBSD port committers to do that when they have time. [...] I've attached a port for LyX 1.6.1 which needs testing. The port requires QT 4 and takes quite some time to build. The flavors are gone since XForms is no longer supported by LyX. As I am new to LyX I would like to hear success and bug reports from more experienced LyX users and also from users on platforms other than i386. So far I've noticed an issue with the window manager dwm and the spell checker window. Sometimes the window isn't mapped properly. Regards, Andreas lyx-1.6.1-port.tar.gz Description: Lyx 1.6.1 for OpenBSD
Re: NEW: mail/sqlgrey (resubmit)
Pierre-Emmanuel André writes: Le Tue, 24 Jun 2008 09:38:11 +0200, Giovanni Bechis [EMAIL PROTECTED] a écrit : SQLgrey is a postfix policy service implementing a grey-listing policy, it can use Postgresql, Mysql or SQLite as backend and it can alert the sysadmin when the database backend is not available. Port tested (in a production environment for about a month) [...] Tested on @i386 with a PostgreSQL 8.3.1 backend. Works well ! I didn't test the OpenBSD port but I got sporadic stability problems when I used SQLgrey on a heavily loaded Debian-based mail server. After looking at the SQLgrey code I decided to use Postgrey instead, which works very well. Unless you really need to share the greylisting database I'd use Postgrey or spamd instead of SQLgrey.
Re: Emacs 22 port?
Brad Walker writes: On Mon, 2008-04-14 at 21:42 +0200, Eric Faurot wrote: On Sun, 13 Apr 2008 14:35:57 -0700 Matthew Dempsky [EMAIL PROTECTED] wrote: Is anyone working on an Emacs 22 port? I thought I'd check before making a redundant effort. This is a work-in-progress port of emacs 22.2 based on a post by Fulvio Ciriaco. Contributions are most welcome. I've attached a patch for Emacs 22 diffed against Eric's tarball[...]. This patch incorporates Matthew Dempsky's gtk FLAVOR patch and plist fixes[...] as well as an athena flavor. I also attempted to resolve the problems Eric mentioned. [...] Here's yet another proposal. * The port attached to the end of this message is named emacs instead of emacs22. It conflicts with emacs21. I think that there's no need to support different Emacs versions in the ports tree. Naming the port emacs eases package updates to newer Emacs versions. * The port isn't split into -main and -el. I don't see the point. Users who are low on disk space can use mg oder jmacs, i.e. joe. * By default Emacs is built against GTK+ since GTK+ will be the default toolkit in Emacs 23. There are two flavors: * no_x11: Don't build against any widget set. Don't build against image libraries like jpeg and png which arent't of much use on a text console. BTW, does anybody actually use the no_x11 flavor? * athena: Build against the Athena widget set instead of GTK+. This flavor will provide minimal GUI support. For example, when Emacs 23 is released the default flavor will support librsvg, dbus and all the other bells and whistles that the upstream developers are currently adding to Emacs. The athena flavor, on the other hand, will be kept small. What do you think? Tested on i386 only. The GCCARCH variable at the end of the makefile needs to be checked on other platforms. emacs.tar.gz Description: Emacs 22 port
Re: Emacs 22 port?
Eric Faurot writes: I just imported emacs22. Why did you hurry with the import? The port still lacks the patches from emacs21 and has only been tested on a few platforms. I called it emacs22, because there is no reason to change it to emacs. Some people will want emacs21, other emacs22. Who will maintain the emacs21 port? You? The default package stays with the athena tk. I just added the gtk flavor, and the no_x11 flavor remains, of course. Shouldn't the no_x11 port be built with the following options? --without-jpeg \ --without-tiff \ --without-gif \ --without-png Otherwise, LIB_DEPENDS-main should be set accordingly. What emacs23 will be is not relevant. emacs-el stays a subpackage because it is really heavy and mostly useless. I updated and attached the port that I sent to list this morning accordingly. I won't send a diff against emacs22 since I maintain a local port of Emacs 23 anyway and will focus on that port now. The updated port contains most of the patches from emacs21 adapted to Emacs 22 in case anyone is interested in incorporating these patches into the emacs22 port. emacs22.tar.gz Description: Updated port of Emacs 22
Re: Emacs 22 port?
Brad Walker writes: On Sat, 2008-06-14 at 22:25 +0200, Andreas Vögele wrote: Eric Faurot writes: The default package stays with the athena tk. I just added the gtk flavor, and the no_x11 flavor remains, of course. Shouldn't the no_x11 port be built with the following options? --without-jpeg \ --without-tiff \ --without-gif \ --without-png Otherwise, LIB_DEPENDS-main should be set accordingly. configure handles disabling that --without-x: [...] I see. Thanks a lot for the clarification. I got confused by the output of make lib-depends-check while testing the different flavors with an incomplete makefile. Time to go to bed.
Re: Emacs 22 port?
Eric Faurot writes: On Tue, 10 Jun 2008 17:02:01 +0200 [EMAIL PROTECTED] (Andreas Vögele) wrote: Brad Walker writes: Adding --localstatedir=/var places the game score files under /var/games/emacs, but then update-plist strips them. I think that you ought to use @sample in the PLIST file to create the directory and the files in /var/games. For example: [snip] Right, but unfortunately this is apparently not working either. Emacs does not seem to save scores in these files anyway, for some reason that I am not interrested in investigating further. Argh, I forgot about other changes I made to my copy of the Emacs CVS repository. Emacs checks whether the s-uid bit of the update-game-score is set and saves the score in ~/.emacs.d/games/ if it isn't. Changing this locally is easy but fixing it in a way that upstream will accept isn't. The emacs source package is so completely and desperatingly ugly. We really don't need this anyway: [...] Yes, let's forget about the system-wide high score files. Another point: I don't see a compelling reason for adding an -athena flavor instead of keeping it the default, as it is now. I find it just confusing. [...] Here is a new version with a -gtk flavor. GTK will be the default toolkit in Emacs 23. I don't know if the Emacs developers plan to drop support for the Athena widget set in a future Emacs version, but what about making GTK the default now and adding a -athena flavor that might be dropped later? Or will the ports for the different Emacs versions, i.e. emacs21, emacs22 etc., be kept separate anyway? Personally, I'm happy with the Athena widget set since I use Emacs with menu, tool and scroll bars disabled. But I wouldn't mind using a port built against GTK. I also left the parts for having score files in /var/games/emacs in case someone wants to figure out why it is not working. Here are the patches that I use locally. But I wouldn't include them in the port. Let Emacs save the game scores to ~/.emacs.d/games/ instead. $OpenBSD$ --- lisp/play/gamegrid.el.orig Tue May 6 09:24:05 2008 +++ lisp/play/gamegrid.el Fri Jun 13 08:20:27 2008 @@ -489,7 +489,7 @@ FILE is created there. (not (zerop (logand (file-modes (expand-file-name update-game-score exec-directory)) - #o4000) + #o6000) (cond ((file-name-absolute-p file) (gamegrid-add-score-insecure file score)) ((and gamegrid-shared-game-dir $OpenBSD$ --- lib-src/update-game-score.c.orig Sat May 10 01:19:13 2008 +++ lib-src/update-game-score.c Fri Jun 13 10:03:52 2008 @@ -222,7 +222,7 @@ main (argc, argv) if (optind+3 != argc) usage (EXIT_FAILURE); - running_suid = (getuid () != geteuid ()); + running_suid = (getuid () != geteuid ()) || (getgid () != getegid ()); prefix = get_prefix (running_suid, user_prefix); @@ -471,7 +471,7 @@ write_scores (filename, scores, count) fclose (f); if (rename (tempfile, filename) 0) return -1; - if (chmod (filename, 0644) 0) + if (chmod (filename, 0664) 0) return -1; return 0; }
Re: Emacs 22 port?
Brad Walker writes:
Adding --localstatedir=/var places the game score files
under /var/games/emacs, but then update-plist strips them.
I think that you ought to use @sample in the PLIST file to create the
directory and the files in /var/games. For example:
@mode g+w
@group games
@sample /var/games/emacs/
@mode
@group
share/examples/emacs/snake-scores
@mode g+w
@group games
@sample /var/games/emacs/snake-scores
@mode
@group
share/examples/emacs/tetris-scores
@mode g+w
@group games
@sample /var/games/emacs/tetris-scores
@mode
@group
Put a post-install target into the Makefile to copy the high score files
to the example directory:
post-install:
${INSTALL_DATA_DIR} ${PREFIX}/share/examples/emacs
${INSTALL_DATA} ${WRKINST}/var/games/emacs/snake-scores
${PREFIX}/share/examples/emacs
${INSTALL_DATA} ${WRKINST}/var/games/emacs/tetris-scores
${PREFIX}/share/examples/emacs
Finally, change the group of the update-game-score binary to the games
group and set the group s-bit so that Emacs can update the system-wide
high score files:
@mode g+s
@group games
libexec/emacs/${VERSION}/${GCCARCH}/update-game-score
@mode
@group
- there is a problem with chown commands failing during fake install.
The Makefile uses find $${dest} -exec chown $${LOGNAME:-$$USERNAME} {}
';' ;\ in several places, causing the chown errors since ports doesn't
set LOGNAME or USERNAME. Adding 'LOGNAME=${SHAREOWN}' to MAKE_ENV
quieted the chown errors.
Emacs 23, i.e. the development version, uses $USER and `id -un 2
/dev/null` in addition to $LOGNAME and $USERNAME:
## http://lists.gnu.org/archive/html/emacs-devel/2007-10/msg01672.html
## Needs to be the user running install, so configure can't set it.
set_installuser=for installuser in $${LOGNAME} $${USERNAME} $${USER} \
`id -un 2 /dev/null`; do \
[ -n $${installuser} ] break ; \
done
Re: Samba with kerberos
Jeff Ross writes:
Jeff Ross wrote:
Hi all,
I need a version of samba that is compiled with both ldap and
kerberos so that my samba server can join a MS Windows 2003 Active
Directory domain. [...]
I've applied the included patch and successfully build Samba with
env FLAVOR=ldap make package. I haven't tested the package yet,
though.
Index: Makefile
===
RCS file: /cvs/ports/net/samba/Makefile,v
retrieving revision 1.89
diff -u -r1.89 Makefile
--- Makefile 4 Jan 2008 17:48:37 - 1.89
+++ Makefile 8 Mar 2008 09:49:49 -
@@ -57,7 +57,7 @@
--with-syslog \
--with-utmp
-CONFIGURE_ENV= CPPFLAGS=-I${LOCALBASE}/include \
+CONFIGURE_ENV= CPPFLAGS=-I${LOCALBASE}/include -I/usr/include/kerberosV \
LDFLAGS=-L${LOCALBASE}/lib -Wl,--export-dynamic
FLAVORS=cups ldap
@@ -74,7 +74,7 @@
.endif
.if ${FLAVOR:L:Mldap}
-CONFIGURE_ARGS+= --with-ldap --without-ads
+CONFIGURE_ARGS+= --with-ldap --with-ads
LIB_DEPENDS+= ldap,lber::databases/openldap
BUILD_DEPENDS+= ::misc/libutf8
.else
Index: patches/patch-configure
===
RCS file: patches/patch-configure
diff -N patches/patch-configure
--- /dev/null 1 Jan 1970 00:00:00 -
+++ patches/patch-configure 8 Mar 2008 09:49:49 -
@@ -0,0 +1,12 @@
+$OpenBSD$
+--- configure.orig Mon Dec 10 17:04:31 2007
configure Sat Mar 8 10:08:58 2008
+@@ -48682,7 +48682,7 @@ fi
+ echo ${ECHO_T}$with_ads_support 6; }
+
+ FOUND_KRB5=no
+-KRB5_LIBS=
++KRB5_LIBS=-lssl -lcrypto
+
+ if test x$with_ldap_support != xyes; then
+
Re: OpenVPN update
Felix Kronlage [EMAIL PROTECTED] writes: hi, here comes an update to OpenVPN to 2.1RC7. Please test this in the scenarios where you use OpenVPN and report any breakage. Works fine for me on -current/i386. I don't run any OpenVPN servers under OpenBSD but I use OpenVPN daily to connect my notebook to several Debian-based OpenVPN servers. Right now I am connected to an OpenVPN server through a wireless network. And there's another OpenVPN tunnel on top of that tunnel. Everything works as before.
Re: UPDATE: exim-4.68
Felix Kronlage writes:
hi.
trivial update from 4.67 to exim 4.68 attached.
Changelog for the exim changes is here:
url: http://www.exim.org/ftp/ChangeLogs/ChangeLog-4.68
please test and comment.
Works fine for me on i386 but the version number is still hardcoded to
4.66:
./pkg/PLIST:bin/exim-4.66
./patches/patch-scripts_exim_install:+version=exim-4.66
I don't know why patch-scripts_exim_install was added by one of the
former maintainers. Without the patch the version number is set to
4.68-1 instead of 4.68. The number that follows the dash is the local
compilation number.
IMHO you could remove patch-scripts_exim_install. I think it's okay
to include the compilation number even if it will always equal 1.
Alternatively, here's a patch that excludes the compilation number
from the version without the need to hardcode the version into
patch-scripts_exim_install.
$OpenBSD$
--- scripts/exim_install.orig Thu Aug 30 16:31:06 2007
+++ scripts/exim_install Fri Sep 7 16:22:08 2007
@@ -219,7 +219,7 @@ while [ $# -gt 0 ]; do
if [ $name = exim${EXE} ]; then
version=exim-`./exim -bV -C /dev/null | \
- awk '/Exim version/ { OFS=; print $3,-,substr($4,2,length($4)-1) }'`${EXE}
+ awk '/Exim version/ { OFS=; print $3 }'`${EXE}
if [ ${version} = exim-${EXE} ]; then
echo $com
But I'm wondering whether the versioned binary is needed at all.
Here's a patch that disables the special handling of the exim binary.
With this patch $LOCALBASE/bin/exim will be the binary itself instead
of a a symbolic link to the versioned binary. A nice side effect is
that you can run make update-plist without the need to fix the file
mode of the exim binary in PLIST.
$OpenBSD$
--- scripts/exim_install.orig Thu Aug 30 16:31:06 2007
+++ scripts/exim_install Fri Sep 7 16:07:28 2007
@@ -217,7 +217,7 @@ while [ $# -gt 0 ]; do
# The exim binary is handled specially
- if [ $name = exim${EXE} ]; then
+ if false; then
version=exim-`./exim -bV -C /dev/null | \
awk '/Exim version/ { OFS=; print $3,-,substr($4,2,length($4)-1) }'`${EXE}
The only difference that I can see is that the output of ps will no
longer show the version:
26814 ?? Is 0:00.00 /usr/local/bin/exim -bd -q30m
versus
2491 ?? Is 0:00.00 /usr/local/bin/exim -bd -q30m (exim-4.68)
UPDATE: security/p5-Crypt-OpenSSL-RSA
The attached diff updates security/p5-Crypt-OpenSSL-RSA from version
0.24 to version 0.25. Upstream added a license file to the
distribution and fixed the regression tests. Accordingly, I removed
patches/patch-t_rsa_t. I also updated my mail address in the
Makefile.
Tests fine on -current/i386.
Index: Makefile
===
RCS file: /cvs/ports/security/p5-Crypt-OpenSSL-RSA/Makefile,v
retrieving revision 1.6
diff -u -u -r1.6 Makefile
--- Makefile 29 Dec 2006 10:19:52 - 1.6
+++ Makefile 18 Jul 2007 06:30:37 -
@@ -6,11 +6,11 @@
MODULES= cpan
-DISTNAME= Crypt-OpenSSL-RSA-0.24
+DISTNAME= Crypt-OpenSSL-RSA-0.25
CATEGORIES= security
HOMEPAGE= http://perl-openssl.sourceforge.net/
-MAINTAINER= Andreas Voegele [EMAIL PROTECTED]
+MAINTAINER= Andreas Voegele [EMAIL PROTECTED]
# Perl
PERMIT_PACKAGE_CDROM= Yes
Index: distinfo
===
RCS file: /cvs/ports/security/p5-Crypt-OpenSSL-RSA/distinfo,v
retrieving revision 1.5
diff -u -u -r1.5 distinfo
--- distinfo 5 Apr 2007 17:26:08 - 1.5
+++ distinfo 18 Jul 2007 06:30:37 -
@@ -1,5 +1,5 @@
-MD5 (Crypt-OpenSSL-RSA-0.24.tar.gz) = V2DiZ4zZ4aoztA1dIWiZyA==
-RMD160 (Crypt-OpenSSL-RSA-0.24.tar.gz) = kkaOKul/znwSeLJbr9dFjhwy32U=
-SHA1 (Crypt-OpenSSL-RSA-0.24.tar.gz) = 2CWGIv1QKS0JgKzWgrZSKASoh6U=
-SHA256 (Crypt-OpenSSL-RSA-0.24.tar.gz) = XelGvxnNljF0GKQipEyw6+puF+Qu8UhUieZ98Kasslk=
-SIZE (Crypt-OpenSSL-RSA-0.24.tar.gz) = 11847
+MD5 (Crypt-OpenSSL-RSA-0.25.tar.gz) = /fGckJP0fP+4Ua6TfQU8FA==
+RMD160 (Crypt-OpenSSL-RSA-0.25.tar.gz) = x9+EJSOPyNYtshe2jvlEGkMNvaE=
+SHA1 (Crypt-OpenSSL-RSA-0.25.tar.gz) = QVl7/Aaqj4FtnU5PDcKEBevpxAc=
+SHA256 (Crypt-OpenSSL-RSA-0.25.tar.gz) = eXc0O11bswXEzXHT80OqEvc+eOMYXpkjV5sb1c1FtgE=
+SIZE (Crypt-OpenSSL-RSA-0.25.tar.gz) = 19133
Index: patches/patch-t_rsa_t
===
RCS file: patches/patch-t_rsa_t
diff -N patches/patch-t_rsa_t
--- patches/patch-t_rsa_t 29 Dec 2006 10:19:52 - 1.1
+++ /dev/null 1 Jan 1970 00:00:00 -
@@ -1,12 +0,0 @@
-$OpenBSD: patch-t_rsa_t,v 1.1 2006/12/29 10:19:52 jasper Exp $
t/rsa.t.orig Mon Nov 13 01:01:04 2006
-+++ t/rsa.t Thu Dec 28 13:11:23 2006
-@@ -4,7 +4,7 @@ use Test;
- use Crypt::OpenSSL::Random;
- use Crypt::OpenSSL::RSA;
-
--BEGIN { plan tests = 45 + (UNIVERSAL::can(Crypt::OpenSSL::RSA, use_sha512_hash) ? 3*6 : 0) }
-+BEGIN { plan tests = 43 + (UNIVERSAL::can(Crypt::OpenSSL::RSA, use_sha512_hash) ? 4*5 : 0) }
-
- sub _Test_Encrypt_And_Decrypt
- {
Re: libsigsegv
Jasper Lievisse Adriaanse writes: hi, here's an updated version of the [libsigsegv] port, with help from [EMAIL PROTECTED] it has been tested on alpha, amd64, sparc, sparc64, mips64, i386 and powerpc. (note: for sparc64 you need to have -r1.51 of trap.c commited yesterday) ok to import? I can only build the updated libsigsegv package if I add SHARED_LIBS=sigsegv 0.0 to the Makefile. I haven't made any progress on the CLISP port. Gambit-C and PLT Scheme look more promising. I'll probably use one of these Scheme implementations instead of CLISP. Cheers, Andreas
Re: UPDATE: lang/clisp
Scott Vokes writes:
On 6/5/07, Andreas Vögele [EMAIL PROTECTED] wrote:
AFAIR the code in the gmalloc.c file bundled with clisp-2.41 is
broken. I doubt that the upstream clisp maintainers ever tried to
build that code. Try to build clisp with the gmalloc.c file from
clisp-2.33.2 instead.
I fixed it and will send the patch upstream.
BTW, I've never been able to get clisp running with libsigsegv. Even
clisp-2.33.2 doesn't work when linked to libsigsegv.
I couldn't get clisp to install with libsigsegv as a shared
library, but it works static.
Here are ports for clisp-2.41 and its dependency libsigsegv-2.4.
Please test and commit.
Hm, why did you put ${BASE_PKGPATH} into the PLIST files? I couldn't
build the libsigsegv port in /usr/ports/mystuff/devel/libsigsegv/ with
this variable in PLIST.
Also, I don't think that USE_GMAKE and REGRESS_TARGET are needed. The
libsigsegv ports builds fine with OpenBSD's make. And check is the
default regress target. On the other hand, the libsigsegv package
seems to support concurrent builds, i.e. you could add
SEPARATE_BUILD=concurrent to the Makefile. I don't think that the
libsigsegv port will run on all architectures supported by OpenBSD, so
ONLY_FOR_ARCHS may be required too.
diff -ur libsigsegv.orig/Makefile libsigsegv/Makefile
--- libsigsegv.orig/Makefile Sun Jun 24 21:52:31 2007
+++ libsigsegv/Makefile Wed Jun 27 12:33:37 2007
@@ -17,9 +17,7 @@
MASTER_SITES= ${MASTER_SITE_GNU:=libsigsegv/}
-USE_GMAKE= Yes
+SEPARATE_BUILD= concurrent
CONFIGURE_STYLE= gnu
-
-REGRESS_TARGET= check
.include bsd.port.mk
Only in libsigsegv: Makefile~
diff -ur libsigsegv.orig/pkg/PLIST libsigsegv/pkg/PLIST
--- libsigsegv.orig/pkg/PLIST Sun Jun 24 19:45:58 2007
+++ libsigsegv/pkg/PLIST Mon Jun 25 11:01:13 2007
@@ -1,4 +1,4 @@
@comment $OpenBSD$
include/sigsegv.h
-lib/${BASE_PKGPATH}.a
-lib/${BASE_PKGPATH}.la
+lib/libsigsegv.a
+lib/libsigsegv.la
Only in libsigsegv: w-libsigsegv-2.4
Re: UPDATE: lang/clisp
Scott Vokes writes:
On 6/5/07, Andreas Vögele [EMAIL PROTECTED] wrote:
AFAIR the code in the gmalloc.c file bundled with clisp-2.41 is
broken. I doubt that the upstream clisp maintainers ever tried to
build that code. Try to build clisp with the gmalloc.c file from
clisp-2.33.2 instead.
I fixed it and will send the patch upstream.
BTW, I've never been able to get clisp running with libsigsegv. Even
clisp-2.33.2 doesn't work when linked to libsigsegv.
I couldn't get clisp to install with libsigsegv as a shared
library, but it works static.
Here are ports for clisp-2.41 and its dependency libsigsegv-2.4.
Please test and commit.
I can build the clisp port, but make regress fails on my machine:
./foo -x (setq zz 10) (saveinitmem \foo\)
./foo: operating system error during load of initialization file
`/usr/ports/mystuff/lang/clisp/w-clisp-2.41/build-i386/foo'
[spvw_memfile.d:1685] errno = EFAULT: Bad address.
I also get a lot of messages from systrace when building clisp:
systrace: deny user: andreas, prog:
/usr/ports/mystuff/lang/clisp/w-clisp-2.41/build-i386/lisp.run, pid:
31280(0)[29784], policy: /usr/bin/env, filters: 193, syscall:
native-mincore(78), args: 12
I haven't looked into these problems yet.
BTW, there's no need to set PKGNAME in the port's Makefile since the
package version gets bumped. Also, if I were you I'd move the
libsigsegv port from /usr/ports/mystuff/libsigsegv to
/usr/ports/mystuff/devel/libsigsegv and change BUILD_DEPENDS
accordingly. I've attached a diff that also replaces ${BASE_PKGPATH}
with clisp in PLIST.
diff -ur clisp.orig/Makefile clisp/Makefile
--- clisp.orig/Makefile Sun Jun 24 20:42:43 2007
+++ clisp/Makefile Wed Jun 27 12:39:07 2007
@@ -5,7 +5,6 @@
COMMENT= ANSI Common Lisp compiler
DISTNAME= clisp-2.41
-PKGNAME= ${DISTNAME}
CATEGORIES= lang
HOMEPAGE= http://clisp.cons.org/
@@ -15,7 +14,7 @@
PERMIT_DISTFILES_CDROM= Yes
PERMIT_DISTFILES_FTP= Yes
-BUILD_DEPENDS= libsigsegv-=1:mystuff/libsigsegv
+BUILD_DEPENDS= libsigsegv-=1:devel/libsigsegv
WANTLIB= c m ncurses readline iconv intl sigsegv
diff -ur clisp.orig/pkg/PLIST clisp/pkg/PLIST
--- clisp.orig/pkg/PLIST Sat Jun 23 03:48:52 2007
+++ clisp/pkg/PLIST Wed Jun 27 17:53:10 2007
@@ -1,88 +1,89 @@
@comment $OpenBSD$
-bin/${BASE_PKGPATH}
-lib/${BASE_PKGPATH}/
-lib/${BASE_PKGPATH}/base/
-lib/${BASE_PKGPATH}/base/calls.o
-lib/${BASE_PKGPATH}/base/gettext.o
-lib/${BASE_PKGPATH}/base/libavcall.a
-lib/${BASE_PKGPATH}/base/libcallback.a
-lib/${BASE_PKGPATH}/base/libcharset.a
-lib/${BASE_PKGPATH}/base/libnoreadline.a
-lib/${BASE_PKGPATH}/base/lisp.a
-lib/${BASE_PKGPATH}/base/lisp.run
-lib/${BASE_PKGPATH}/base/lispinit.mem
-lib/${BASE_PKGPATH}/base/makevars
-lib/${BASE_PKGPATH}/base/modules.h
-lib/${BASE_PKGPATH}/base/modules.o
-lib/${BASE_PKGPATH}/base/readline.o
-lib/${BASE_PKGPATH}/base/regex.o
-lib/${BASE_PKGPATH}/base/regexi.o
-lib/${BASE_PKGPATH}/base/regexp.dvi
-lib/${BASE_PKGPATH}/${BASE_PKGPATH}-link
-lib/${BASE_PKGPATH}/data/
-lib/${BASE_PKGPATH}/data/Symbol-Table.text
-lib/${BASE_PKGPATH}/data/UnicodeDataFull.txt
-lib/${BASE_PKGPATH}/full/
-lib/${BASE_PKGPATH}/full/calls.o
-lib/${BASE_PKGPATH}/full/gettext.o
-lib/${BASE_PKGPATH}/full/libavcall.a
-lib/${BASE_PKGPATH}/full/libcallback.a
-lib/${BASE_PKGPATH}/full/libcharset.a
-lib/${BASE_PKGPATH}/full/libnoreadline.a
-lib/${BASE_PKGPATH}/full/lisp.a
-lib/${BASE_PKGPATH}/full/lisp.run
-lib/${BASE_PKGPATH}/full/lispinit.mem
-lib/${BASE_PKGPATH}/full/makevars
-lib/${BASE_PKGPATH}/full/modules.h
-lib/${BASE_PKGPATH}/full/modules.o
-lib/${BASE_PKGPATH}/full/readline.o
-lib/${BASE_PKGPATH}/full/regex.o
-lib/${BASE_PKGPATH}/full/regexi.o
-lib/${BASE_PKGPATH}/full/regexp.dvi
-lib/${BASE_PKGPATH}/linkkit/
-lib/${BASE_PKGPATH}/linkkit/${BASE_PKGPATH}.h
-lib/${BASE_PKGPATH}/linkkit/modprep.lisp
-lib/${BASE_PKGPATH}/linkkit/modules.c
-lib/${BASE_PKGPATH}/linkkit/modules.d
-share/doc/${BASE_PKGPATH}/
-share/doc/${BASE_PKGPATH}/ANNOUNCE
-share/doc/${BASE_PKGPATH}/CLOS-guide.txt
-share/doc/${BASE_PKGPATH}/COPYRIGHT
-share/doc/${BASE_PKGPATH}/GNU-GPL
-share/doc/${BASE_PKGPATH}/LISP-tutorial.txt
-share/doc/${BASE_PKGPATH}/MAGIC.add
-share/doc/${BASE_PKGPATH}/NEWS
-share/doc/${BASE_PKGPATH}/README
-share/doc/${BASE_PKGPATH}/README.de
-share/doc/${BASE_PKGPATH}/README.es
-share/doc/${BASE_PKGPATH}/SUMMARY
-share/doc/${BASE_PKGPATH}/${BASE_PKGPATH}.1
-share/doc/${BASE_PKGPATH}/${BASE_PKGPATH}.dvi
-share/doc/${BASE_PKGPATH}/${BASE_PKGPATH}.html
-share/doc/${BASE_PKGPATH}/${BASE_PKGPATH}.png
-share/doc/${BASE_PKGPATH}/${BASE_PKGPATH}.ps
-share/doc/${BASE_PKGPATH}/editors.txt
-share/doc/${BASE_PKGPATH}/impnotes.css
-share/doc/${BASE_PKGPATH}/impnotes.html
+bin/clisp
+lib/clisp/
+lib/clisp/base/
+lib/clisp/base/calls.o
+lib/clisp/base/gettext.o
+lib/clisp/base/libavcall.a
+lib/clisp/base/libcallback.a
+lib/clisp/base/libcharset.a
+lib/clisp/base/libnoreadline.a
+lib/clisp/base/lisp.a
+lib/clisp/base/lisp.run
+lib/clisp/base/lispinit.mem
+lib/clisp
Re: problem with version numbering in ICU 3.6
Vincent GROSS writes: Hi folks, I'm trying to make a somewhat decent port of ICU (http://www.icu-project.org), and I have a problem with version numbering : the current version is 3.6, but the version number appended to the shareds objects is 36.0 ... should I keep it like that or follow the OpenBSD recommandation and turn it into 3.6 ? (it can be done via MAKE_FLAGS) I've sent a mail to the ICU support mailing list to ask them about the version numbering scheme. Here's a port of icu4c 3.6 that I created several weeks ago while evaluating a software that requires ICU. Maybe there's something that you can use in your port. For example, there's a fix for the message WARNING: symbol(icudtXX_dat) size mismatch, relink your program in patch-source_stubdata_Makefile_in. I didn't change the library version since I needed the port for a couple of days only. The port builds fine on i386 and passes the regression tests. I haven't tested the port on any other architecture though. icu4c.tar.gz Description: icu4c-3.6 port
Re: Working towards updating lang/clisp
Scott Vokes writes: I have been trying to update the clisp port, but have run into a snag. [...] As of 2.39, a configure option, --with-gmalloc, was added to use the included GNU malloc instead of OpenBSD's (due to conflicts with clisp's memory management), but attempts to build it eventually grind to a halt with an error in gmalloc.c itself: [...] AFAIR the code in the gmalloc.c file bundled with clisp-2.41 is broken. I doubt that the upstream clisp maintainers ever tried to build that code. Try to build clisp with the gmalloc.c file from clisp-2.33.2 instead. BTW, I've never been able to get clisp running with libsigsegv. Even clisp-2.33.2 doesn't work when linked to libsigsegv.
UPDATE: p5-Crypt-OpenSSL-RSA 0.24
Hi,
the attached patch updates p5-Crypt-OpenSSL-RSA from version 0.19 to
0.24. The new version fixes bugs, adds new functions and, last but no
least, may be distributed under the same license terms as Perl.
I contacted Wesley Griffin, the current maintainer. He says that he
hasn't got the time to maintain the port any longer.
If nobody else is interested in maintaining this port, I'll take it.
I use p5-Crypt-OpenSSL-RSA occasionally to convert public keys from
OpenSSL's file format to the format used by Racoon.
p5-Crypt-OpenSSL-RSA is used by p5-Net-DNS-SEC, which is maintained by
Jasper Lievisse Adriaanse. I don't use p5-Net-DNS-SEC myself but the
regression tests of that module still pass.
Andreas
Index: Makefile
===
RCS file: /cvs/ports/security/p5-Crypt-OpenSSL-RSA/Makefile,v
retrieving revision 1.5
diff -u -r1.5 Makefile
--- Makefile 27 Dec 2004 13:28:34 - 1.5
+++ Makefile 29 Dec 2006 08:37:30 -
@@ -1,27 +1,27 @@
# $OpenBSD: Makefile,v 1.5 2004/12/27 13:28:34 alek Exp $
-NOT_FOR_ARCHS= ${NO_SHARED_ARCHS}
+SHARED_ONLY= Yes
COMMENT= RSA encoding and decoding using OpenSSL
-VERSION= 0.19
-DISTNAME= Crypt-OpenSSL-RSA-${VERSION}
-PKGNAME= p5-${DISTNAME}
-CATEGORIES= security perl5
+MODULES= cpan
+
+DISTNAME= Crypt-OpenSSL-RSA-0.24
+CATEGORIES= security
HOMEPAGE= http://perl-openssl.sourceforge.net/
-MASTER_SITES= ${MASTER_SITE_PERL_CPAN:=Crypt/}
-MAINTAINER= Wesley Griffin [EMAIL PROTECTED]
+MAINTAINER= Andreas Voegele [EMAIL PROTECTED]
+
+# Perl
+PERMIT_PACKAGE_CDROM= Yes
+PERMIT_PACKAGE_FTP= Yes
+PERMIT_DISTFILES_CDROM= Yes
+PERMIT_DISTFILES_FTP= Yes
-PERMIT_PACKAGE_CDROM= No license
-PERMIT_PACKAGE_FTP= No license
-PERMIT_DISTFILES_CDROM= No license
-PERMIT_DISTFILES_FTP= No license
WANTLIB= crypto ssl
-RUN_DEPENDS= :p5-Crypt-OpenSSL-Random-=0.03:security/p5-Crypt-OpenSSL-Random
+RUN_DEPENDS= :p5-Crypt-OpenSSL-Random-=0.03:security/p5-Crypt-OpenSSL-Random \
+ :p5-Crypt-OpenSSL-Bignum-=0.03:security/p5-Crypt-OpenSSL-Bignum
BUILD_DEPENDS= ${RUN_DEPENDS}
-
-CONFIGURE_STYLE= perl
.include bsd.port.mk
Index: distinfo
===
RCS file: /cvs/ports/security/p5-Crypt-OpenSSL-RSA/distinfo,v
retrieving revision 1.3
diff -u -r1.3 distinfo
--- distinfo 5 Jan 2005 17:26:42 - 1.3
+++ distinfo 29 Dec 2006 08:37:30 -
@@ -1,4 +1,4 @@
-MD5 (Crypt-OpenSSL-RSA-0.19.tar.gz) = 87a308bf752ffd74b8034a906184e116
-RMD160 (Crypt-OpenSSL-RSA-0.19.tar.gz) = 7134b8cdd5f27021b8a70c44893372f0cf7921ce
-SHA1 (Crypt-OpenSSL-RSA-0.19.tar.gz) = 7633a523588ae95551468ce7aec01fa3317f1ba9
-SIZE (Crypt-OpenSSL-RSA-0.19.tar.gz) = 11227
+MD5 (Crypt-OpenSSL-RSA-0.24.tar.gz) = 5760e2678cd9e1aa33b40d5d216899c8
+RMD160 (Crypt-OpenSSL-RSA-0.24.tar.gz) = 92468e2ae97fce7c1278b25bafd7458e1c32df65
+SHA1 (Crypt-OpenSSL-RSA-0.24.tar.gz) = d8258622fd50292d0980acd682b6522804a887a5
+SIZE (Crypt-OpenSSL-RSA-0.24.tar.gz) = 11847
Index: patches/patch-t_rsa_t
===
RCS file: patches/patch-t_rsa_t
diff -N patches/patch-t_rsa_t
--- /dev/null 1 Jan 1970 00:00:00 -
+++ patches/patch-t_rsa_t 29 Dec 2006 08:37:30 -
@@ -0,0 +1,12 @@
+$OpenBSD$
+--- t/rsa.t.orig Mon Nov 13 01:01:04 2006
t/rsa.t Thu Dec 28 13:11:23 2006
+@@ -4,7 +4,7 @@ use Test;
+ use Crypt::OpenSSL::Random;
+ use Crypt::OpenSSL::RSA;
+
+-BEGIN { plan tests = 45 + (UNIVERSAL::can(Crypt::OpenSSL::RSA, use_sha512_hash) ? 3*6 : 0) }
++BEGIN { plan tests = 43 + (UNIVERSAL::can(Crypt::OpenSSL::RSA, use_sha512_hash) ? 4*5 : 0) }
+
+ sub _Test_Encrypt_And_Decrypt
+ {
Index: pkg/DESCR
===
RCS file: /cvs/ports/security/p5-Crypt-OpenSSL-RSA/pkg/DESCR,v
retrieving revision 1.1.1.1
diff -u -r1.1.1.1 DESCR
--- pkg/DESCR 15 Oct 2002 11:20:10 - 1.1.1.1
+++ pkg/DESCR 29 Dec 2006 08:37:30 -
@@ -1,14 +1,17 @@
-This module is an XS perl module designed to provide basic RSA
-functionality. It does this by providing a glue to the RSA functions
-in the OpenSSL library. In particular, it provides the following
-functions:
+Crypt::OpenSSL::RSA is an XS perl module designed to provide basic RSA
+functionality. It does this by providing a glue to the RSA functions
+in the OpenSSL library. In particular, it provides the following functions:
-loadPublicKey, loadPrivateKey - load a key from a string
-getPublicKeyString, getPrivateKeyString - save a key to a string
-generateKey - make a new key
-encrypt, decrypt, sign, verify,
-use_pkcs1_oaep_padding, use_pkcs1_padding,
-use_sslv23_padding, use_no_padding,
-use_md5_hash, use_sha1_hash, use_ripemd60_hash
-size - returns the size, in bytes, of the key
-checkKey - checks the validity of the key
+ new_public_key, new_private_key - create a key from a string
+ generate_key - make a new key
+ get_private_key_string - save key to a string
+
Re: backuppc updated port
Joerg Zinke writes:
I've got an updated port for BackupPC 3.0.0beta2. I'll send you
the files as soon as I've fixed a couple of minor bugs.
can you send the files here to ports@ since other people (at least
me) are interested in that port too.
I've attached the current version of the BackupPC port. The
README.OpenBSD file is incomplete but should contain enough
information to get going.
did you managed it to get the cgi running in chroot? or did you just
disable the chroot for the cgi?
I run another non-chrooted httpd instance that only listens on
127.0.0.1 and runs with the privileges of the _backuppc user.
Connections can be forwarded with ProxyPass from the chrooted httpd
instance. Alternatively, SSH port forwarding can be used to access
the BackupPC web frontend.
I'm struggling with one problem.
On startup, BackupPC version 3 checks whether all executables that are
specified in the configuration file exist. BackupPC quits if an
executable is missing. Thus the current port depends on the optional
packages gtar, samba, bzip2, rsync and par2cmdline.
BackupPC version 2 didn't check for the presence of optional
executables. Thus it was possible to specify these executables in the
configuration file without the need to depend on the optional
packages.
On the one hand, I'd like to keep the port simple. On the other hand,
I don't want the BackupPC port to depend on gtar and samba, since most
people will probably use rsync to backup their hosts.
I can think of several solutions to this problem:
1. By default, the BackupPC port could depend on all packages. But we
could add flavors like no_gtar and no_samba that disable the
dependencies so that people can easily build their own packages
that don't depend on samba etc.
2. We could set the path names of the optional executables to the
empty string so that BackupPC doesn't look for these executables on
startup. But then users would have to set the path names manually
in the configuration when an optional package is installed.
3. We could disable the checks for the executables. The changes in
BackupPC would be rather small, but I'd like to avoid changes to
the upstream sources if possible.
Unfortunately, there's a bug in BackupPC's configure.pl. If a path is
set to the empty string, configure.pl will output the default path,
e.g. '/usr/bin/smbclient', instead of the empty string. I haven't
decided yet how to fix this bug. This bug affects solutions 1 and 2.
Personally, I'd go for solution 1. What do you think?
There's another issue I'm not happy with.
I used to keep the files for the second httpd instance in
/var/backuppc/www/{conf,htdocs,logs}. I now decided to put httpd.conf
into /etc/backuppc alongside the BackupPC configuration files. The
httpd log files are kept in /var/log/backuppc alongside the BackupPC
log files. IMHO it makes sense to keep these files together -
especially the configuration files. Also, I don't expect the file
names that are used by BackupPC and httpd to ever clash.
But OpenBSD's httpd insists on creating the file logs/etag-state in
the ServerRoot directory that I have set to /etc/backuppc. Since the
path name logs/etag-state cannot be configured I have to create a
symbolic link called /etc/backuppc/logs that points to
/var/log/backuppc. This also means that I have to add an @exec
statement to PLIST. Not nice, but I can't think of a better solution.
What else is there to say?
By default, the BackupPC httpd instance now listens on port 8081 on
localhost. The port can be set with BACKUPPC_HTTPD_PORT in the
Makefile. Any suggestions for another default port instead of 8081?
The _backuppc user has got the user and group id 599. This will
change if the BackupPC port should ever make it into the ports tree.
If you have used a previous version of the BackupPC port you must move
several files and directories. If a hard disk partition is mounted on
/var/backuppc that partition should now be mounted on
/var/db/backuppc.
The contents of /var/backuppc/conf needs to be moved to /etc/backuppc.
Don't forget to compare the old and new config.pl. The config files
of the individual hosts ought to be moved one level down into the
directory /etc/backuppc/pc. The contents of /var/backuppc/www/conf
should also be moved to /etc/backuppc. Make sure to update your
httpd.conf.
You can move the contents of /var/backuppc/logs and
/var/backuppc/www/logs to /var/log/backuppc if you would like to keep
your log files.
The sub directories cpool, pc, pool, trash, and .ssh have moved from
/var/backuppc to /var/db/backuppc. Make sure to set the home
directory of the user _backuppc to /var/db/backuppc.
BackupPC's httpd must now be started with httpd -d /etc/backuppc -f
/etc/backuppc/httpd.conf -u instead of httpd -d /var/backuppc/www
-u.
The directory /var/backuppc is no longer needed once you have moved
all your files.
Regards,
Andreas
port-p5-File-RsyncP.tar.gz
Description:
Re: backuppc updated port
Joerg Zinke writes:
On Sat, 25 Nov 2006 15:30:26 +0100
[EMAIL PROTECTED] (Andreas Vögele) wrote:
On the other hand, I don't want the BackupPC port to depend on gtar
and samba, since most people will probably use rsync to backup
their hosts.
Yeah thats true. I only use ssh for example.
I can think of several solutions to this problem:
1. By default, the BackupPC port could depend on all packages. But we
could add flavors like no_gtar and no_samba that disable the
dependencies [...]
2. We could set the path names of the optional executables to the
empty string so that BackupPC doesn't look for these executables on
startup. [...]
3. We could disable the checks for the executables. [...]
I think a mix of 1. + 2. could be the way to go. If the user decides
for no_gtar or no_samba as flavor, the port should not install them
as dependency _and_ should leave the config string empty.
Yes, that's actually what I meant and that's why I wrote that the bug
with the empty string also applies to solution 1.
Just disabling the checks as suggested in 3. would work and would
keep the port simple...
I don't know whether it's better to provide a package that pulls in
all the packages that BackupPC may use (which includes Samba) or if
it's better to provide a minimal package that leaves it to the user to
install additional packages.
Unfortunately, there's a bug in BackupPC's configure.pl. If a path
is set to the empty string, configure.pl will output the default
path, e.g. '/usr/bin/smbclient', instead of the empty string. I
haven't decided yet how to fix this bug. This bug affects
solutions 1 and 2.
Since 3.0.0 is beta it should not be too hard to get the bug fixed
in the upstream before final is released?
I don't think that the final version of BackupPC 3 will be released
before the end of the year.
I just haven't decided where to fix the bug. The script configure.pl
handles empty strings given on the command line properly. It seems
that the problem arises when the configuration is written to disk.
Changing that code may have unwanted side effects. I have to look
closer at the code and come up with a simple solution.
There's another issue I'm not happy with.
I used to keep the files for the second httpd instance in
/var/backuppc/www/{conf,htdocs,logs}. I now decided to put httpd.conf
into /etc/backuppc alongside the BackupPC configuration files. The
httpd log files are kept in /var/log/backuppc alongside the BackupPC
log files. [...]
But OpenBSD's httpd insists on creating the file logs/etag-state in
the ServerRoot directory that I have set to /etc/backuppc. Since the
path name logs/etag-state cannot be configured I have to create a
symbolic link called /etc/backuppc/logs that points to
/var/log/backuppc. This also means that I have to add an @exec
statement to PLIST. Not nice, but I can't think of a better solution.
Mhmm... what about this: keep webserver root in /var/backuppc/www/ and
put /etc/backuppc in /var/backuppc/www/etc/ and set a symbolic link (for
whole directory or maybe better for every single file in it from
/etc/backuppc - /var/backuppc/www/etc/
The /etc/backuppc directory isn't strictly required by BackupPC. But
I'd like to keep the config and log files separate from the
directories cpool, pc, pool and trash, where the backups are stored.
I usually store cpool, pc, pool and trash on a separate disk that
powers down if no backups are in progress. The config and log files
must not be stored on the backup disk since the BackupPC daemon
regularly reads the config files and may write to the log files.
Also, it's safer to keep the configuration files and the backups on
different disks. If the backup disk breaks down the configuration
files aren't lost and it's easy to setup a new backup disk. On the
other hand, if the disk with the configuration files breaks down it's
possible to retrieve the configuration files from the backup disk.
In practice, I use a cron job to backup the configuration files, but
nevertheless I think it's a good idea to keep the configuration files
and the backups separate.
I think it is not a good idea to have the webserver root in etc.
Well, that's what Debian does :-)
The contents of /var/backuppc/conf needs to be moved to /etc/backuppc.
Why not a symbolic link from /etc/backuppc - /var/backuppc/conf/ as
suggested above?
Or just ignore the /etc/backuppc idea?
For example /etc/bind or /etc/ apache do not exists on openbsd
too... it is no problem to have config files under /var/...
There's another reason not to put the config files into /etc. In
version 3, BackupPC's web frontend can be configured to write to the
config directory. It's rather strange to have a web frontend write to
/etc, which after all may be mounted read-only.
What about the following directory structure?
/var/backuppc/ (home directory of the _backuppc user)
/var/backuppc/.ssh/
/var/backuppc/conf/
/var
Re: backuppc updated port
Siju George writes: hi, I did receive a backuppc port for 3.9 from a dear friend here. I thought it will be included in 4.0 ports. A few people are asking me for the port and I have directed them here. May I know if the port was bandoned or is there an updated one that can run on 4.0 I've got an updated port for BackupPC 3.0.0beta2. I'll send you the files as soon as I've fixed a couple of minor bugs. The file system layout has changed. In version 2 everything was stored in /var/backuppc. In version 3 the configuration has moved to /etc/backuppc. The log files are now stored in /var/log/backuppc. The other directories have moved to /var/db/backuppc. I could keep the old layout but one benefit of the new layout is that you may put /var/db/backuppc on a hard disk that powers down when no backups are in progress. On the other hand,
WANTED: new maintainer for Exim port
I don't have the time to maintain the Exim port and am looking for a
new maintainer who would like to maintain the port.
A patch that removes the MAINTAINER variable from the port's Makefile
is attached.
Index: Makefile
===
RCS file: /cvs/ports/mail/exim/Makefile,v
retrieving revision 1.55
diff -u -r1.55 Makefile
--- Makefile 13 Oct 2006 15:18:54 - 1.55
+++ Makefile 28 Oct 2006 05:51:02 -
@@ -2,14 +2,13 @@
COMMENT= Flexible mail transfer agent
DISTNAME= exim-4.63
+PKGNAME= ${DISTNAME}p1
CATEGORIES= mail
MASTER_SITES= ftp://ftp.exim.org/pub/exim/exim4/ \
http://mirror.switch.ch/ftp/mirror/exim/exim/exim4/ \
ftp://mirror.switch.ch/mirror/exim/exim/exim4/
HOMEPAGE= http://www.exim.org/
-
-MAINTAINER= Andreas Voegele [EMAIL PROTECTED]
# GPL
PERMIT_PACKAGE_CDROM= Yes
WANTED: new maintainer for p5-Mail-SpamAssassin port
I don't have the time to maintain the p5-Mail-SpamAssassin port and am
looking for a new maintainer who would like to maintain the port.
A patch that removes the MAINTAINER variable from the port's Makefile
is attached.
Index: Makefile
===
RCS file: /cvs/ports/mail/p5-Mail-SpamAssassin/Makefile,v
retrieving revision 1.43
diff -u -r1.43 Makefile
--- Makefile 27 Oct 2006 10:12:44 - 1.43
+++ Makefile 28 Oct 2006 05:53:12 -
@@ -3,10 +3,8 @@
COMMENT= mailfilter to identify and mark spam
DISTNAME= Mail-SpamAssassin-3.1.7
-PKGNAME= p5-${DISTNAME}
+PKGNAME= p5-${DISTNAME}p1
CATEGORIES= mail perl5
-
-MAINTAINER= Andreas Voegele [EMAIL PROTECTED]
HOMEPAGE= http://spamassassin.apache.org/
Re: WANTED: new maintainer for p5-Mail-SpamAssassin port
Marc Balmer writes: Andreas Vögele wrote: I don't have the time to maintain the p5-Mail-SpamAssassin port and am looking for a new maintainer who would like to maintain the port. A patch that removes the MAINTAINER variable from the port's Makefile is attached. I will take that. Marc, I don't want to offend you but you still haven't updated the Horde port, which contains a security whole that was announced in May. Don't you think that you should take care of your existing ports before you take other ports? Antoine has done a lot of the work on the latest version of the SpamAssassin port, which I couldn't do since I was at a fair this week. I don't know if Antoine is interested in taking the port. But if he is than he should take the port IMHO. Steven has also done a lot of work on the port. At least, he pushed me to keep the port up-to-date :-)
Re: WANTED: new maintainer for Exim port
Felix Kronlage writes: On Sat, Oct 28, 2006 at 07:59:58AM +0200, Andreas Vögele wrote: I don't have the time to maintain the Exim port and am looking for a new maintainer who would like to maintain the port. I'm tempted to step up for this and take it. I use exim a lot and maintain lots of installations of it on openbsd. So if noone else feels a bit urge to maintain this, I'd like to. That would be great. All the Exim servers that I maintain do now run under Debian and it just takes to much time to check the OpenBSD port that I don't use regularly.
Re: p5-Mail-SpamAssassin: proposed change
Antoine Jacoutot writes: Hi. Here's a diff that change the _spamdaemon user's home directory from /nonexistent to /var/spamassassin (owned by _spamdaemon). This way, one can start spamd as the _spamdaemon user and make use of auto-whitelist, bayes... without error. [,,,] Let me know! That makes sense. Actually, I already run spamd locally with the following options, which is similar to the changes that you propose. /usr/local/bin/spamd -d -u spamassassin -H /var/spamassassin -r /var/spamassassin/spamd.pid We might also use /var/spamassassin instead of /var/db/spamassassin for the updated rules that sa-update downloads.
Re: p5-Mail-SpamAssassin: proposed change
David Krause writes: * Antoine Jacoutot [EMAIL PROTECTED] [061024 09:57]: Selon Andreas V?gele [EMAIL PROTECTED]: We might also use /var/spamassassin instead of /var/db/spamassassin for the updated rules that sa-update downloads. Indeed. New diff attached. What do you all think? Should the _spamdaemon user be able to the write to the updated rules directory? The updated rules are put into a subdirectory by sa-update, e.g. /var/db/spamassassin/3.001007. The _spamdaemon user doesn't need write access to that directory, only read access. In the current setup sa-update must be run by root, which is a bad idea anyway. We could add another user, e.g. _saupdate, that may be used to call sa-update, and change the owner of /var/db/spamassassin to that user.
Re: p5-Mail-SpamAssassin: proposed change
Antoine Jacoutot writes: So, should we commit this? I'll update the patch for SpamAssassin 3.1.7 that I sent to the list yesterday. I don't think that we have to keep these changes separate.
UPDATE: p5-Mail-SpamAssassin-3.1.7
The attached patch updates p5-Mail-SpamAssassin to version 3.1.7,
which is a bug fix release. See http://spamassassin.apache.org/ for
the release announcements.
I've cleaned up the Makefile. The port now uses DESTDIR to install
the package into the staging area. The files patch-Makefile_PL,
patch-lib_Mail_SpamAssassin_pm and patch-spamc_spamc_pod are gone.
I also decided to no longer install SpamAssassin's public keys as
config files. This keeps the port simple. Instead I added the
following line to the PLIST file:
@extraunexec rm -rf ${CONFDIR}/sa-update-keys
The keys will be installed on sa-update's first invocation. sa-update
will also create the directory /var/db/spamassassin.
Tested on i386. More testing is very welcome.
Index: Makefile
===
RCS file: /cvs/ports/mail/p5-Mail-SpamAssassin/Makefile,v
retrieving revision 1.42
diff -u -r1.42 Makefile
--- Makefile 21 Oct 2006 19:26:14 - 1.42
+++ Makefile 23 Oct 2006 09:14:46 -
@@ -2,8 +2,8 @@
COMMENT= mailfilter to identify and mark spam
-DISTNAME= Mail-SpamAssassin-3.1.4
-PKGNAME= p5-${DISTNAME}p1
+DISTNAME= Mail-SpamAssassin-3.1.7
+PKGNAME= p5-${DISTNAME}
CATEGORIES= mail perl5
MAINTAINER= Andreas Voegele [EMAIL PROTECTED]
@@ -27,29 +27,25 @@
:p5-IO-Socket-SSL-*:security/p5-IO-Socket-SSL \
:p5-libwww-*:www/p5-libwww \
:p5-Archive-Tar-=1.23:archivers/p5-Archive-Tar \
- :p5-IO-Zlib-=1.04:archivers/p5-IO-Zlib \
- :gnupg-*:security/gnupg
+ :p5-IO-Zlib-=1.04:archivers/p5-IO-Zlib
-RUN_DEPENDS= ${BUILD_DEPENDS}
+RUN_DEPENDS= ${BUILD_DEPENDS} \
+ :gnupg-*:security/gnupg
MAKE_ENV+= CC=${CC}
CONFIGURE_STYLE= perl
-CONFIGURE_ARGS+= RUN_NET_TESTS=no \
+CONFIGURE_ARGS+= DESTDIR=${WRKINST} \
SYSCONFDIR=${SYSCONFDIR} \
- LOCALSTATEDIR=/var/db \
+ LOCALSTATEDIR=${LOCALSTATEDIR} \
CONTACT_ADDRESS=the administrator of this system \
ENABLE_SSL=yes
CONFDIR= ${SYSCONFDIR}/mail/spamassassin
-RULES_DIR= ${PREFIX}/share/spamassassin
+LOCALSTATEDIR= /var/db/spamassassin
SAMPLE_CONFDIR= ${PREFIX}/share/examples/SpamAssassin
-SUBST_VARS= CONFDIR SAMPLE_CONFDIR
-
-FAKE_FLAGS= ${DESTDIRNAME}=${WRKINST} \
- DEFRULESDIR=${WRKINST}${RULES_DIR} \
- LOCALRULESDIR=${WRKINST}${SAMPLE_CONFDIR}
+SUBST_VARS= CONFDIR LOCALSTATEDIR
# regress insists on creating files in $HOME
PORTHOME= ${WRKDIR}/temp-home
@@ -57,10 +53,6 @@
FAQ= OpenBSD-SpamAssassin-mini-howto.html
DOCS= README UPGRADE LICENSE procmailrc.example \
sample-nonspam.txt sample-spam.txt
-LOCALCF= local.cf init.pre v310.pre v312.pre
-
-pre-configure:
- @perl -pi -e [EMAIL PROTECTED]@[EMAIL PROTECTED]@#${CONFDIR}# ${WRKSRC}/rules/*
post-install:
${INSTALL_DATA_DIR} ${PREFIX}/share/doc/SpamAssassin
@@ -70,19 +62,9 @@
${FILESDIR}/${FAQ} ${WRKSRC}/${FAQ}
cd ${WRKSRC} ${INSTALL_DATA} ${FAQ} ${DOCS} \
${PREFIX}/share/doc/SpamAssassin
- cd ${WRKSRC}/rules ${INSTALL_DATA} ${LOCALCF} ${SAMPLE_CONFDIR}
- cd ${WRKSRC} pod2man --section=1 spamc/spamc.pod \
+ cd ${WRKINST}${CONFDIR} mv local.cf *.pre ${SAMPLE_CONFDIR}
+ @cd ${WRKSRC} pod2man --section=1 spamc/spamc.pod \
${PREFIX}/man/man1/spamc.1
- cd ${WRKSRC} pod2man --section=8 spamd/spamd \
- ${PREFIX}/man/man8/spamd.8
- # Extract the default GnuPG keys for sa-update. The keys will
- # be installed as config files in ${CONFDIR}/sa-update-keys.
- # Otherwise, sa-update would install the keys on its first
- # invocation. The sa-update utility also creates the directory
- # /var/db/spamassassin.
- ${INSTALL_DATA_DIR} ${SAMPLE_CONFDIR}/sa-update-keys
- ${LOCALBASE}/bin/gpg --homedir=${SAMPLE_CONFDIR}/sa-update-keys --batch --import ${WRKSRC}/rules/sa-update-pubkey.txt
- rm -f ${SAMPLE_CONFDIR}/sa-update-keys/pubring.gpg~
pre-regress:
mkdir -p ${PORTHOME}
Index: distinfo
===
RCS file: /cvs/ports/mail/p5-Mail-SpamAssassin/distinfo,v
retrieving revision 1.26
diff -u -r1.26 distinfo
--- distinfo 9 Aug 2006 07:21:07 - 1.26
+++ distinfo 23 Oct 2006 09:14:46 -
@@ -1,4 +1,4 @@
-MD5 (Mail-SpamAssassin-3.1.4.tar.gz) = 6259d1b1c5ce34c37596fc262c0b9663
-RMD160 (Mail-SpamAssassin-3.1.4.tar.gz) = 8aebcc9fa4ed0eafadab85eaafd2f7b488a7ff91
-SHA1 (Mail-SpamAssassin-3.1.4.tar.gz) = df88c2e27c1c8ede60a1b967443a3b16cc74cd9f
-SIZE (Mail-SpamAssassin-3.1.4.tar.gz) = 1142066
+MD5 (Mail-SpamAssassin-3.1.7.tar.gz) = 4b342c63949d47f3ce56b3fc1c8881c1
+RMD160 (Mail-SpamAssassin-3.1.7.tar.gz) = b5e5cffdbcfba6e58346bdd51c8adba1547fcca4
+SHA1 (Mail-SpamAssassin-3.1.7.tar.gz) = 3d31eff0eb9a158fab308958d65cdca81b8944bc
+SIZE (Mail-SpamAssassin-3.1.7.tar.gz) = 1168183
Index: patches/patch-Makefile_PL
===
RCS file: patches/patch-Makefile_PL
diff -N patches/patch-Makefile_PL
--- patches/patch-Makefile_PL 6 Jun 2006 17:33:41 - 1.5
+++ /dev/null 1 Jan 1970 00:00:00 -
@@ -1,26
UPDATE: exim-4.63
Attached is an update to Exim 4.63, which fixes some bugs and adds
some new features. See URL:
http://permalink.gmane.org/gmane.mail.exim.announce/104 for Philip
Hazel's release announcement. Tested on i386.
Index: Makefile
===
RCS file: /cvs/ports/mail/exim/Makefile,v
retrieving revision 1.53
diff -u -r1.53 Makefile
--- Makefile 5 May 2006 06:02:30 - 1.53
+++ Makefile 1 Aug 2006 09:18:39 -
@@ -1,7 +1,7 @@
# $OpenBSD: Makefile,v 1.53 2006/05/05 06:02:30 sturm Exp $
COMMENT= Flexible mail transfer agent
-DISTNAME= exim-4.62
+DISTNAME= exim-4.63
CATEGORIES= mail
MASTER_SITES= ftp://ftp.exim.org/pub/exim/exim4/ \
http://mirror.switch.ch/ftp/mirror/exim/exim/exim4/ \
Index: distinfo
===
RCS file: /cvs/ports/mail/exim/distinfo,v
retrieving revision 1.8
diff -u -r1.8 distinfo
--- distinfo 5 May 2006 06:02:30 - 1.8
+++ distinfo 1 Aug 2006 09:18:39 -
@@ -1,4 +1,4 @@
-MD5 (exim-4.62.tar.gz) = 4142642dade5c24154ef2ea2ba562b12
-RMD160 (exim-4.62.tar.gz) = 8cd77cef20f974a900f4b3f6f6ac51e4929eaff3
-SHA1 (exim-4.62.tar.gz) = f0699f671b7517183d8fa3039a1949716d8a6f43
-SIZE (exim-4.62.tar.gz) = 2054559
+MD5 (exim-4.63.tar.gz) = 187b6f4242f869288fb4562df44d9c29
+RMD160 (exim-4.63.tar.gz) = 8c51efb20dab3787e5b0b8b2a196d5d879ed4843
+SHA1 (exim-4.63.tar.gz) = eb729a491ac315de805be4d59adc7a8854456fa9
+SIZE (exim-4.63.tar.gz) = 2066186
Index: patches/patch-scripts_exim_install
===
RCS file: /cvs/ports/mail/exim/patches/patch-scripts_exim_install,v
retrieving revision 1.5
diff -u -r1.5 patch-scripts_exim_install
--- patches/patch-scripts_exim_install 5 May 2006 06:02:30 - 1.5
+++ patches/patch-scripts_exim_install 1 Aug 2006 09:18:39 -
@@ -6,7 +6,7 @@
if [ $name = exim${EXE} ]; then
-version=exim-`./exim -bV -C /dev/null | \
- awk '/Exim version/ { OFS=; print $3,-,substr($4,2,length($4)-1) }'`${EXE}
-+version=exim-4.62
++version=exim-4.63
if [ ${version} = exim-${EXE} ]; then
echo $com
Index: pkg/PLIST
===
RCS file: /cvs/ports/mail/exim/pkg/PLIST,v
retrieving revision 1.16
diff -u -r1.16 PLIST
--- pkg/PLIST 5 May 2006 06:02:30 - 1.16
+++ pkg/PLIST 1 Aug 2006 09:18:39 -
@@ -4,7 +4,7 @@
bin/exicyclog
bin/exigrep
bin/exim
-bin/exim-4.62
+bin/exim-4.63
bin/exim_checkaccess
bin/exim_dbmbuild
bin/exim_dumpdb
Re: p5-Mail-SpamAssassin missing dependency?
Antti Harri wrote:
is SA missing runtime dependency or was there something weird with my
configuration?
After upgrading some packages to 3.9-stable I started getting this
from sa-update:
sa-update: importing default keyring to
'/etc/mail/spamassassin/sa-update-keys'... fatal: couldn't find GPG
in $PATH
I've attached a patch that adds the gnupg runtime dependency. The patch
also changes LOCALSTATEDIR from /var/lib to /var/db.
Index: Makefile
===
RCS file: /cvs/ports/mail/p5-Mail-SpamAssassin/Makefile,v
retrieving revision 1.39
diff -u -r1.39 Makefile
--- Makefile6 Jun 2006 17:33:41 - 1.39
+++ Makefile30 Jun 2006 12:47:00 -
@@ -3,7 +3,7 @@
COMMENT= mailfilter to identify and mark spam
DISTNAME= Mail-SpamAssassin-3.1.3
-PKGNAME= p5-${DISTNAME}
+PKGNAME= p5-${DISTNAME}p0
CATEGORIES=mail perl5
MAINTAINER=Andreas Voegele [EMAIL PROTECTED]
@@ -29,13 +29,15 @@
:p5-Archive-Tar-*:archivers/p5-Archive-Tar \
:p5-IO-Zlib-*:archivers/p5-IO-Zlib
-RUN_DEPENDS= ${BUILD_DEPENDS}
+RUN_DEPENDS= ${BUILD_DEPENDS} \
+ :gnupg-*:security/gnupg
MAKE_ENV+= CC=${CC}
CONFIGURE_STYLE= perl
CONFIGURE_ARGS+= RUN_NET_TESTS=no \
SYSCONFDIR=${SYSCONFDIR} \
+ LOCALSTATEDIR=/var/db \
CONTACT_ADDRESS=the administrator of this system \
ENABLE_SSL=yes
Index: patches/patch-sa-update_raw
===
RCS file: patches/patch-sa-update_raw
diff -N patches/patch-sa-update_raw
--- /dev/null 1 Jan 1970 00:00:00 -
+++ patches/patch-sa-update_raw 30 Jun 2006 12:47:00 -
@@ -0,0 +1,12 @@
+$OpenBSD$
+--- sa-update.raw.orig Wed May 24 22:07:30 2006
sa-update.raw Mon Jun 26 13:12:35 2006
+@@ -1120,7 +1120,7 @@ Options:
+
+ --updatedir pathDirectory to place updates, defaults to the
+ SpamAssassin site rules directory (def:
+- /var/lib/spamassassin/version)
++ /var/db/spamassassin/version)
+
+ --channel channel Retrieve updates from this channel
+ Use multiple times for multiple channels
Index: pkg/PLIST
===
RCS file: /cvs/ports/mail/p5-Mail-SpamAssassin/pkg/PLIST,v
retrieving revision 1.19
diff -u -r1.19 PLIST
--- pkg/PLIST 6 Jun 2006 17:33:42 - 1.19
+++ pkg/PLIST 30 Jun 2006 12:47:00 -
@@ -209,3 +209,4 @@
share/spamassassin/sa-update-pubkey.txt
share/spamassassin/triplets.txt
share/spamassassin/user_prefs.template
[EMAIL PROTECTED] /var/db/spamassassin/
NEW: www/horde 3.1.1
I've attached a tarball that contains a port of Horde 3.1.1 and several Horde applications. The port is based on Marc Balmer's port. A major difference is that all ports are stored beneath www/horde so that a common Makefile.inc can be used for all Horde applications. IMHO there are several advantages: * It's easier to maintain the Horde ports if a common Makefile is used. * It's easier to add new Horde applications. * It's easier to build and install all Horde applications if they are stored in a common directory. * There's only one directory that needs to be removed from the ports tree if one day a better, more secure alternative should become available ;-) I've also changed the file permissions of most configuration files in PLIST since only the conf.php files need to be writable by Horde. The test.php files are no longer readable by default. Administrators who would like to test their installation have to change the file permissions temporarily. The reason for this change is that OpenBSD's package system resets the file permissions when updating a package. I think it's better to restrict access to the test.php files. Otherwise administrators would have to fix the file permissions after every update. I've moved the Horde applications that already exist in the ports tree to the following places: devel/horde - www/horde/horde mail/imp - www/horde/imp www/turba - www/horde/turba www/kronolith - www/horde/kronolith www/nag - www/horde/nag databases/mnemo - www/horde/mnemo devel/chora - www/horde/chora I've also added four new applications that the Horde project has released in recent months: www/horde/ingo - Mail filter rules www/horde/gollem - File manager www/horde/passwd - Password changing application www/horde/forwards - Mail forwards What do you think about these changes? horde.tar.gz Description: application/gzip
UPDATE: mail/exim 4.62
Philip Hazel has released Exim 4.62, which is a bug fix release. A patch
that updates the port is attached.
There are still no flavors for DomainKeys, SPF and SRS. I've got ports
for libdomainkeys and libspf2 but I'm not eager to maintain these ports.
Index: Makefile
===
RCS file: /cvs/ports/mail/exim/Makefile,v
retrieving revision 1.52
diff -u -r1.52 Makefile
--- Makefile5 Feb 2006 14:30:11 - 1.52
+++ Makefile2 May 2006 17:58:52 -
@@ -1,8 +1,8 @@
# $OpenBSD: Makefile,v 1.52 2006/02/05 14:30:11 steven Exp $
COMMENT= Flexible mail transfer agent
-DISTNAME= exim-4.60
-PKGNAME= ${DISTNAME}p0
+DISTNAME= exim-4.62
+PKGNAME= ${DISTNAME}
CATEGORIES=mail
MASTER_SITES= ftp://ftp.exim.org/pub/exim/exim4/ \
http://mirror.switch.ch/ftp/mirror/exim/exim/exim4/ \
@@ -10,7 +10,7 @@
HOMEPAGE= http://www.exim.org/
-MAINTAINER=Andreas Voegele [EMAIL PROTECTED]
+MAINTAINER=Andreas Voegele [EMAIL PROTECTED]
# GPL
PERMIT_PACKAGE_CDROM= Yes
@@ -95,6 +95,9 @@
@cp ${FILESDIR}/eximon.conf ${WRKSRC}/Local
.endif
+pre-fake:
+ ${INSTALL_DATA_DIR} ${WRKINST}${SYSCONFDIR}/mail
+
post-install:
${INSTALL_DATA_DIR} ${PREFIX}/share/examples/exim
${INSTALL_DATA} ${WRKINST}${SYSCONFDIR}/exim/configure
${PREFIX}/share/examples/exim
@@ -104,5 +107,9 @@
${INSTALL_DATA} ${WRKSRC}/doc/README.SIEVE ${PREFIX}/share/doc/exim
${INSTALL_DATA} ${WRKSRC}/doc/filter.txt ${PREFIX}/share/doc/exim
${INSTALL_MAN} ${WRKSRC}/doc/exim.8 ${PREFIX}/man/man8
+ rm ${WRKINST}${SYSCONFDIR}/exim/configure
+ rmdir ${WRKINST}${SYSCONFDIR}/exim
+ rm ${WRKINST}${SYSCONFDIR}/mail/aliases
+ rmdir ${WRKINST}${SYSCONFDIR}/mail
.include bsd.port.mk
Index: distinfo
===
RCS file: /cvs/ports/mail/exim/distinfo,v
retrieving revision 1.7
diff -u -r1.7 distinfo
--- distinfo5 Feb 2006 14:30:11 - 1.7
+++ distinfo2 May 2006 17:58:52 -
@@ -1,4 +1,4 @@
-MD5 (exim-4.60.tar.gz) = 583be34bf0ca2d6bda9f6dea4823c574
-RMD160 (exim-4.60.tar.gz) = 772c5c0a98fc3b5789b065980831914f33cef5f7
-SHA1 (exim-4.60.tar.gz) = 79bce0491ebae54170382c0909c2480b90c25257
-SIZE (exim-4.60.tar.gz) = 2024104
+MD5 (exim-4.62.tar.gz) = 4142642dade5c24154ef2ea2ba562b12
+RMD160 (exim-4.62.tar.gz) = 8cd77cef20f974a900f4b3f6f6ac51e4929eaff3
+SHA1 (exim-4.62.tar.gz) = f0699f671b7517183d8fa3039a1949716d8a6f43
+SIZE (exim-4.62.tar.gz) = 2054559
Index: patches/patch-scripts_exim_install
===
RCS file: /cvs/ports/mail/exim/patches/patch-scripts_exim_install,v
retrieving revision 1.4
diff -u -r1.4 patch-scripts_exim_install
--- patches/patch-scripts_exim_install 15 Jan 2006 13:06:43 - 1.4
+++ patches/patch-scripts_exim_install 2 May 2006 17:58:52 -
@@ -6,7 +6,7 @@
if [ $name = exim${EXE} ]; then
-version=exim-`./exim -bV -C /dev/null | \
- awk '/Exim version/ { OFS=; print $3,-,substr($4,2,length($4)-1)
}'`${EXE}
-+version=exim-4.60
++version=exim-4.62
if [ ${version} = exim-${EXE} ]; then
echo $com
Index: pkg/DESCR
===
RCS file: /cvs/ports/mail/exim/pkg/DESCR,v
retrieving revision 1.9
diff -u -r1.9 DESCR
--- pkg/DESCR 25 Mar 2005 15:13:59 - 1.9
+++ pkg/DESCR 2 May 2006 17:58:52 -
@@ -11,7 +11,9 @@
* no_x11 - do not build eximon which requires X11
* mysql- support mysql queries for lookups
* postgresql - support postgresql queries for lookups
+* sqlite3 - support for sqlite queries for lookups
* ldap - support ldap (OpenLDAP) queries for lookups
* iconv- support character code conversions
+* sasl - support for SASL
All this flavors are independent of each other.
Index: pkg/PLIST
===
RCS file: /cvs/ports/mail/exim/pkg/PLIST,v
retrieving revision 1.15
diff -u -r1.15 PLIST
--- pkg/PLIST 15 Jan 2006 13:06:43 - 1.15
+++ pkg/PLIST 2 May 2006 17:58:52 -
@@ -4,7 +4,7 @@
bin/exicyclog
bin/exigrep
bin/exim
-bin/exim-4.60
+bin/exim-4.62
bin/exim_checkaccess
bin/exim_dbmbuild
bin/exim_dumpdb
Re: spamassassin update
Ben Lovett wrote:
I received this update from steven mestdagh. Since I'm no longer running
SpamAssassin on any systems, I offered it to steven, and he mentioned
he didn't have the time for it. So if there is anyone out there that
would like to take over, please do so.
I've attached an updated patch since there is a bug in the Razor2 plugin
that was only noticed after the release. I've also added myself as the
new maintainer.
Index: Makefile
===
RCS file: /cvs/ports/mail/p5-Mail-SpamAssassin/Makefile,v
retrieving revision 1.37
diff -u -r1.37 Makefile
--- Makefile8 Feb 2006 04:54:49 - 1.37
+++ Makefile19 Mar 2006 11:15:15 -
@@ -2,11 +2,11 @@
COMMENT= mailfilter to identify and mark spam
-DISTNAME= Mail-SpamAssassin-3.1.0
-PKGNAME= p5-${DISTNAME}p0
+DISTNAME= Mail-SpamAssassin-3.1.1
+PKGNAME= p5-${DISTNAME}
CATEGORIES=mail perl5
-MAINTAINER=Ben Lovett [EMAIL PROTECTED]
+MAINTAINER=Andreas Voegele [EMAIL PROTECTED]
HOMEPAGE= http://spamassassin.apache.org/
Index: distinfo
===
RCS file: /cvs/ports/mail/p5-Mail-SpamAssassin/distinfo,v
retrieving revision 1.23
diff -u -r1.23 distinfo
--- distinfo9 Oct 2005 09:05:43 - 1.23
+++ distinfo19 Mar 2006 11:15:15 -
@@ -1,4 +1,4 @@
-MD5 (Mail-SpamAssassin-3.1.0.tar.gz) = f70c1fcab3d9563731bbc307eda7d69e
-RMD160 (Mail-SpamAssassin-3.1.0.tar.gz) =
20dc1b8befcc72545b6136e58a9fcce143480ccd
-SHA1 (Mail-SpamAssassin-3.1.0.tar.gz) =
d887cbae5962cb03e45aaf71cd93881a2799
-SIZE (Mail-SpamAssassin-3.1.0.tar.gz) = 1185499
+MD5 (Mail-SpamAssassin-3.1.1.tar.gz) = f7844cbc149de3d7b09a4310f4ab6739
+RMD160 (Mail-SpamAssassin-3.1.1.tar.gz) =
35e3ff6f8396a9057d9f12acf62a304410d60096
+SHA1 (Mail-SpamAssassin-3.1.1.tar.gz) =
cda06e3d38d831521c59e50ec024e468b76035cb
+SIZE (Mail-SpamAssassin-3.1.1.tar.gz) = 1129275
Index: patches/patch-Makefile_PL
===
RCS file: /cvs/ports/mail/p5-Mail-SpamAssassin/patches/patch-Makefile_PL,v
retrieving revision 1.3
diff -u -r1.3 patch-Makefile_PL
--- patches/patch-Makefile_PL 9 Oct 2005 09:05:43 - 1.3
+++ patches/patch-Makefile_PL 19 Mar 2006 11:15:15 -
@@ -1,7 +1,7 @@
$OpenBSD: patch-Makefile_PL,v 1.3 2005/10/09 09:05:43 mbalmer Exp $
Makefile.PL.orig Wed Oct 5 09:35:06 2005
-+++ Makefile.PLWed Oct 5 09:35:33 2005
-@@ -809,7 +809,7 @@ sub MY::install {
+--- Makefile.PL.orig Fri Mar 10 20:30:24 2006
Makefile.PLSun Mar 19 11:59:51 2006
+@@ -857,7 +857,7 @@ sub MY::install {
foreach (@code) {
# Add our install targets as a dependency to all top-level install targets
@@ -10,7 +10,7 @@
# Now do the DESTDIR hack, if necessary.
next if !$mm_needs_destdir;
-@@ -1085,12 +1085,6 @@ qmail/qmail-spamc$(EXE_EXT): spamc/qmail
+@@ -1143,12 +1143,6 @@ qmail/qmail-spamc$(EXE_EXT): spamc/qmail
#bin__install: $(INST_SCRIPT)/sa-filter
## $(RM_F) $(B_SCRIPTDIR)/spamassassin
## $(SYMLINK) $(INST_SCRIPT)/sa-filter $(B_SCRIPTDIR)/spamassassin
Index: patches/patch-lib_Mail_SpamAssassin_Plugin_Razor2_pm
===
RCS file: patches/patch-lib_Mail_SpamAssassin_Plugin_Razor2_pm
diff -N patches/patch-lib_Mail_SpamAssassin_Plugin_Razor2_pm
--- /dev/null 1 Jan 1970 00:00:00 -
+++ patches/patch-lib_Mail_SpamAssassin_Plugin_Razor2_pm19 Mar 2006
11:15:15 -
@@ -0,0 +1,11 @@
+$OpenBSD$
+--- lib/Mail/SpamAssassin/Plugin/Razor2.pm.origFri Mar 10 20:29:54 2006
lib/Mail/SpamAssassin/Plugin/Razor2.pm Sun Mar 19 12:00:45 2006
+@@ -46,6 +46,7 @@ package Mail::SpamAssassin::Plugin::Razo
+
+ use Mail::SpamAssassin::Plugin;
+ use Mail::SpamAssassin::Logger;
++use Mail::SpamAssassin::Timeout;
+ use strict;
+ use warnings;
+ use bytes;
Index: patches/patch-lib_Mail_SpamAssassin_pm
===
RCS file:
/cvs/ports/mail/p5-Mail-SpamAssassin/patches/patch-lib_Mail_SpamAssassin_pm,v
retrieving revision 1.7
diff -u -r1.7 patch-lib_Mail_SpamAssassin_pm
--- patches/patch-lib_Mail_SpamAssassin_pm 31 Oct 2004 08:44:44 -
1.7
+++ patches/patch-lib_Mail_SpamAssassin_pm 19 Mar 2006 11:15:15 -
@@ -1,6 +1,6 @@
lib/Mail/SpamAssassin.pm.orig Tue Sep 21 15:05:27 2004
-+++ lib/Mail/SpamAssassin.pm Sun Oct 3 10:46:22 2004
-@@ -150,6 +150,7 @@ $HOME_URL = http://spamassassin.apache.
+--- lib/Mail/SpamAssassin.pm.orig Fri Mar 10 20:29:55 2006
lib/Mail/SpamAssassin.pm Sun Mar 19 11:59:51 2006
+@@ -151,6 +151,7 @@ $HOME_URL = http://spamassassin.apache.
'/etc/mail/spamassassin/user_prefs.template',
'/usr/local/share/spamassassin/user_prefs.template',
Re: spamassassin update
Ben Lovett wrote: I received this update from steven mestdagh. Since I'm no longer running SpamAssassin on any systems, I offered it to steven, and he mentioned he didn't have the time for it. So if there is anyone out there that would like to take over, please do so. Hm, most of the mail servers I manage still run Debian. But I'll maintain the SpamAssassin port if nobody else is interested.
Re: i386 breakage
Hannah Schroeter wrote:
On Thu, Feb 16, 2006 at 11:50:41AM +0100, steven mestdagh wrote:
from a bulk build started Feb 13: 6 packages not building on i386:
clisp-2.33.2p0 Cannot map memory...
Issues with randomized mmap/malloc, [...] There were different
suggestions for workarounds, among them was the suggestion to use the
(IIRC (s)brk based) gmalloc which is included in the clisp tree
anyway.
I'm not sure though whether I'll have the time to actually try this
out and submit patches in time for the release.
I've attached a simple patch that enables the included malloc. CLISP
can be built successfully with this patch and regression tests pass.
Index: Makefile
===
RCS file: /cvs/ports/lang/clisp/Makefile,v
retrieving revision 1.25
diff -u -r1.25 Makefile
--- Makefile16 Feb 2006 20:56:58 - 1.25
+++ Makefile18 Feb 2006 07:55:31 -
@@ -1,12 +1,11 @@
# $OpenBSD: Makefile,v 1.25 2006/02/16 20:56:58 naddy Exp $
-BROKEN=does not handle randomized mmap()
ONLY_FOR_ARCHS=i386
COMMENT= ANSI Common Lisp compiler
DISTNAME= clisp-2.33.2
-PKGNAME= ${DISTNAME}p0
+PKGNAME= ${DISTNAME}p1
CATEGORIES=lang
HOMEPAGE= http://clisp.cons.org/
Index: patches/patch-src_configure
===
RCS file: /cvs/ports/lang/clisp/patches/patch-src_configure,v
retrieving revision 1.1
diff -u -r1.1 patch-src_configure
--- patches/patch-src_configure 11 Feb 2005 12:14:37 - 1.1
+++ patches/patch-src_configure 18 Feb 2006 07:55:31 -
@@ -1,6 +1,15 @@
$OpenBSD: patch-src_configure,v 1.1 2005/02/11 12:14:37 alek Exp $
src/configure.orig Thu Feb 10 00:43:19 2005
-+++ src/configure Thu Feb 10 00:43:34 2005
+--- src/configure.orig Wed Jun 2 23:56:55 2004
src/configure Sat Feb 18 08:34:03 2006
+@@ -11437,7 +11437,7 @@ if test $cross_compiling = no; then
+ # Both are broken. When used with CLISP, the one in the default libc.a
+ # leads to a SIGSEGV, the one in libmalloc.a leads to a SIGBUS.
+ case $host_os in
+- hpux*) cl_cv_func_malloc_broken=yes ;;
++ hpux*|openbsd*) cl_cv_func_malloc_broken=yes ;;
+ *) cl_cv_func_malloc_broken=no ;;
+ esac
+ else
@@ -24255,7 +24255,7 @@ s,@host@,$host,;t t
s,@host_cpu@,$host_cpu,;t t
s,@host_vendor@,$host_vendor,;t t
Cyrus IMAP Kerberos flavour
The Kerberos flavour of Cyrus IMAP can't be built since libresolv was
removed from -current.
$ cd /usr/ports/mail/cyrus-imapd
$ env FLAVOR=kerberos make
[...]
checking for prop_get in -lsasl2... no
configure: error: Cannot continue without libsasl2.
I've attached a patch that changes the remaining places in configure
where -lresolv is used. Tested on amd64.
Index: patches/patch-configure
===
RCS file: /cvs/ports/mail/cyrus-imapd/patches/patch-configure,v
retrieving revision 1.1.1.1
diff -u -r1.1.1.1 patch-configure
--- patches/patch-configure 27 Mar 2005 18:01:50 - 1.1.1.1
+++ patches/patch-configure 7 Feb 2006 20:27:46 -
@@ -1,6 +1,6 @@
configure.orig Tue Nov 23 19:55:47 2004
-+++ configure Sat Jan 22 13:12:28 2005
-@@ -3286,10 +3286,9 @@ CFLAGS=$save_CFLAGS
+--- configure.orig Mon Feb 14 18:59:46 2005
configure Tue Feb 7 21:20:31 2006
+@@ -3287,10 +3287,9 @@
fi
if test $ac_cv_fpic = yes; then
@@ -13,7 +13,7 @@
echo $as_me:$LINENO: result: $ac_cv_fpic 5
echo ${ECHO_T}$ac_cv_fpic 6
-@@ -4489,7 +4488,7 @@ if test ${ac_cv_lib_resolv_res_search+s
+@@ -4490,7 +4489,7 @@
echo $ECHO_N (cached) $ECHO_C 6
else
ac_check_lib_save_LIBS=$LIBS
@@ -22,7 +22,20 @@
cat conftest.$ac_ext _ACEOF
#line $LINENO configure
/* confdefs.h. */
-@@ -6728,7 +6727,7 @@ echo $as_me: error: Berkeley DB 3.x or
+@@ -4538,9 +4537,9 @@
+ fi
+ echo $as_me:$LINENO: result: $ac_cv_lib_resolv_res_search 5
+ echo ${ECHO_T}$ac_cv_lib_resolv_res_search 6
+-if test $ac_cv_lib_resolv_res_search = yes; then
+- LIB_SOCKET=-lresolv $LIB_SOCKET
+-fi
++#if test $ac_cv_lib_resolv_res_search = yes; then
++# LIB_SOCKET=-lresolv $LIB_SOCKET
++#fi
+
+
+ fi
+@@ -6729,7 +6728,7 @@
@@ -31,7 +44,7 @@
CPPFLAGS=${BDB_INCADD} ${CPPFLAGS}
CYRUSDB_OBJS=$CYRUSDB_OBJS cyrusdb_berkeley.o
-@@ -9639,7 +9638,7 @@ if test ${ac_cv_lib_resolv_res_search+s
+@@ -9649,7 +9648,7 @@
echo $ECHO_N (cached) $ECHO_C 6
else
ac_check_lib_save_LIBS=$LIBS
@@ -40,7 +53,32 @@
cat conftest.$ac_ext _ACEOF
#line $LINENO configure
/* confdefs.h. */
-@@ -11814,7 +11813,7 @@ if test ${ac_cv_lib_resolv_res_search+s
+@@ -9697,16 +9696,16 @@
+ fi
+ echo $as_me:$LINENO: result: $ac_cv_lib_resolv_res_search 5
+ echo ${ECHO_T}$ac_cv_lib_resolv_res_search 6
+-if test $ac_cv_lib_resolv_res_search = yes; then
+- cat confdefs.h _ACEOF
+-#define HAVE_LIBRESOLV 1
+-_ACEOF
++#if test $ac_cv_lib_resolv_res_search = yes; then
++# cat confdefs.h _ACEOF
++##define HAVE_LIBRESOLV 1
++#_ACEOF
++#
++# LIBS=-lresolv $LIBS
++#
++#fi
+
+- LIBS=-lresolv $LIBS
+
+-fi
+-
+-
+ if test $with_statickrb = yes -a ! -d $with_krb; then
+ { { echo $as_me:$LINENO: error: --enable-statickrb specified but
--with-krb did not specify a valid directory 5
+ echo $as_me: error: --enable-statickrb specified but --with-krb did not
specify a valid directory 2;}
+@@ -11831,7 +11830,7 @@
echo $ECHO_N (cached) $ECHO_C 6
else
ac_check_lib_save_LIBS=$LIBS
@@ -49,7 +87,31 @@
cat conftest.$ac_ext _ACEOF
#line $LINENO configure
/* confdefs.h. */
-@@ -11892,7 +11891,7 @@ if test ${ac_cv_lib_gssapi_gss_unwrap+s
+@@ -11879,15 +11878,15 @@
+ fi
+ echo $as_me:$LINENO: result: $ac_cv_lib_resolv_res_search 5
+ echo ${ECHO_T}$ac_cv_lib_resolv_res_search 6
+-if test $ac_cv_lib_resolv_res_search = yes; then
+- cat confdefs.h _ACEOF
+-#define HAVE_LIBRESOLV 1
+-_ACEOF
++#if test $ac_cv_lib_resolv_res_search = yes; then
++# cat confdefs.h _ACEOF
++##define HAVE_LIBRESOLV 1
++#_ACEOF
++#
++# LIBS=-lresolv $LIBS
++#
++#fi
+
+- LIBS=-lresolv $LIBS
+-
+-fi
+-
+ if test -d ${gssapi}; then
+ gssapi_dir=${gssapi}/lib
+ GSSAPIBASE_LIBS=-L$gssapi_dir
+@@ -11909,7 +11908,7 @@
echo $ECHO_N (cached) $ECHO_C 6
else
ac_check_lib_save_LIBS=$LIBS
@@ -58,7 +120,7 @@
cat conftest.$ac_ext _ACEOF
#line $LINENO configure
/* confdefs.h. */
-@@ -12242,8 +12241,8 @@ fi
+@@ -12259,8 +12258,8 @@
GSSAPIBASE_STATIC_LIBS=$GSSAPIBASE_LIBS $gssapi_dir/libgssapi_krb5.a
$gssapi_dir/libkrb5.a $gssapi_dir/libk5crypto.a $gssapi_dir/libcom_err.a
elif test $gss_impl = heimdal; then
CPPFLAGS=$CPPFLAGS -DKRB5_HEIMDAL
@@ -69,7 +131,7 @@
elif test $gss_impl = cybersafe03; then
# Version of CyberSafe with two libraries
CPPFLAGS=$CPPFLAGS -D$platform -I${gssapi}/appsec-sdk/include
-@@ -12435,7 +12434,7 @@ if test ${ac_cv_lib_resolv_res_search+s
+@@ -12452,7 +12451,7 @@
echo $ECHO_N (cached) $ECHO_C 6
else
ac_check_lib_save_LIBS=$LIBS
@@ -78,7 +140,7 @@
cat conftest.$ac_ext _ACEOF
#line $LINENO configure
/* confdefs.h. */
-@@ -12483,9 +12482,9 @@ LIBS=$ac_check_lib_save_LIBS
+@@ -12500,9 +12499,9 @@
fi
echo $as_me:$LINENO: result: $ac_cv_lib_resolv_res_search 5
echo ${ECHO_T}$ac_cv_lib_resolv_res_search 6
@@ -91,7 +153,7 @@
SASL_MECHS=$SASL_MECHS libgssapiv2.la
SASL_STATIC_OBJS=$SASL_STATIC_OBJS
Re: PHP4 packages
Lars Hansson writes: On Thu, 2 Feb 2006 00:00:22 +1300 Shane Lazarus [EMAIL PROTECTED] wrote: Heya Given all of the PHP packages seem to require PHP5, where do I go to find PHP4 packages/ports, or how do I modify the existing requirements? Uh, I see php4 packages both for current and 3.8 on the mirrors: That's not the question. What the OP probably means is that all third-party PHP ports, like for example Horde, depend on the php5 port. You can't built packages that depend on php4. If I were the OP I would install the relevant third-party PHP packages manually without using the ports system. Alternatively, he could copy those ports to /usr/ports/mystuff and alter the dependencies.
Re: FIX: cyrus-sasl2
Antoine Jacoutot writes:
Hi, this is just a quick fix to replace hard coded /usr/local/etc to
${SYSCONFDIR} from the saslauthd man page.
Also, saslauthd is currently built without support for the kerberos5
authentication mechanism. Here's a patch that fixes the Kerberos
include path in saslauthd's configure script.
Index: patches/patch-saslauthd_configure
===
RCS file: /cvs/ports/security/cyrus-sasl2/patches/patch-saslauthd_configure,v
retrieving revision 1.10
diff -u -r1.10 patch-saslauthd_configure
--- patches/patch-saslauthd_configure 19 Dec 2005 18:11:27 - 1.10
+++ patches/patch-saslauthd_configure 24 Jan 2006 18:39:47 -
@@ -1,6 +1,6 @@
$OpenBSD: patch-saslauthd_configure,v 1.10 2005/12/19 18:11:27 jakob Exp $
--- saslauthd/configure.orig Sun May 15 07:59:04 2005
-+++ saslauthd/configureMon Dec 19 15:56:57 2005
saslauthd/configureTue Jan 24 19:33:31 2006
@@ -753,10 +753,10 @@ ac_env_CC_set=${CC+set}
ac_env_CC_value=$CC
ac_cv_env_CC_set=${CC+set}
@@ -44,6 +44,15 @@
*)
{ echo $as_me:$LINENO: WARNING: The system type is not recognized. If
you believe that CyberSafe GSSAPI works on this platform, please update the
configure script 5
echo $as_me: WARNING: The system type is not recognized. If you believe that
CyberSafe GSSAPI works on this platform, please update the configure script
2;}
+@@ -6391,7 +6394,7 @@ echo $as_me: error: CyberSafe was force
+ cmu_saved_CPPFLAGS=$CPPFLAGS
+
+ if test -d ${gssapi}; then
+-CPPFLAGS=$CPPFLAGS -I$gssapi/include
++CPPFLAGS=$CPPFLAGS -I$gssapi/include/kerberosV
+ # We want to keep -I in our CPPFLAGS, but only if we succeed
+ cmu_saved_CPPFLAGS=$CPPFLAGS
+ LDFLAGS=$LDFLAGS -L$gssapi/lib
@@ -6702,7 +6705,7 @@ if test ${ac_cv_lib_resolv_res_search+s
echo $ECHO_N (cached) $ECHO_C 6
else
Re: FIX: cyrus-sasl2
Antoine Jacoutot [EMAIL PROTECTED] writes:
On Wed, 1 Feb 2006, Andreas Vögele wrote:
Antoine Jacoutot writes:
Hi, this is just a quick fix to replace hard coded /usr/local/etc to
${SYSCONFDIR} from the saslauthd man page.
Also, saslauthd is currently built without support for the kerberos5
authentication mechanism. Here's a patch that fixes the Kerberos
include path in saslauthd's configure script.
Huh ? I use it all the time...
You mean /usr/local/sbin/saslauthd -a kerberos5, right ? If so then
it works fine.
I'm wondering how this is possible since the Kerberos include path in
saslauthd's configure script *is* wrong. The port in CVS only fixes
the include path in the top-level configure script.
Do you follow -current?
I can reproduce this problem on i386 and amd64:
$ cd /usr/ports/security/cyrus-sasl2
$ make
[...]
configure: configuring in saslauthd
[...]
checking krb5.h usability... no
checking krb5.h presence... no
checking for krb5.h... no
[...]
$ ./w-cyrus-sasl-2.1.21p0/cyrus-sasl-2.1.21/saslauthd/saslauthd -v
saslauthd 2.1.21
authentication mechanisms: getpwent rimap
With the fix that I posted the output is:
authentication mechanisms: getpwent kerberos5 rimap
Re: UPDATE: mail/exim
It seems that Ilya Voronin, the maintainer of the Exim port has vanished. I tried to contact him twice in the last three months and didn't get a response. I would like to maintain the Exim port. I've attached a port of Exim 4.60, which includes the setting WITH_OLD_DEMIME=yes as suggested by Tamas Tevesz. The file pkg/PFRAG.no-no_exiscan is no longer needed. Tested on i386 and amd64 with all flavors except -mysql, which was tested by Tamas Tevesz. exim-4.60-port.tar.gz Description: Port of Exim 4.60
UPDATE: mail/exim
I've updated the Exim port to version 4.60. I've sent the updated port to the current maintainer Ilya Voronin, but maybe there are people on this list who would like to test this update too. Noteworthy changes: The exiscan patch was included upstream. Upstream added a couple of features and fixed bugs. There are two new flavors: sqlite3 and sasl Authentication via saslauthd is enabled. This feature is completely independent of the sasl flavor and neither requires Exim to be linked against cyrus-sasl2 nor saslauthd to be present at runtime. exim-4.60-port.tar.gz Description: Exim 4.60 port
Re: lang/clisp build fails on i386 -current
Christian Weisgerber writes: J. Lievisse Adriaanse [EMAIL PROTECTED] wrote: When trying to build lang/clisp on both of my laptops (running -current), the build fails in make build because it can't map the memory correctly, it says something like: [...] Does anybody else has this problems too? Known problem, everybody does. Do you know if the OpenBSD kernel will be fixed before 3.8 is released? Thierry Deval wrote on 4th May: While a proper solution has to be devised, you can use the following diff that diminishes the impact of randomization on malloc... And makes clisp work again. [...] I still use the patch that Thierry provided.
