CVS: cvs.openbsd.org: ports

2016-11-28 Thread Anthony J . Bentley 
CVSROOT:/cvs
Module name:ports
Changes by: bent...@cvs.openbsd.org 2016/11/28 20:06:31

ports/www/netsurf/nsgenbind/patches

Update of /cvs/ports/www/netsurf/nsgenbind/patches
In directory cvs.openbsd.org:/tmp/cvs-serv30189/patches

Log Message:
Directory /cvs/ports/www/netsurf/nsgenbind/patches added to the repository

CVS: cvs.openbsd.org: ports

2016-11-28 Thread Stuart Henderson 
CVSROOT:/cvs
Module name:ports
Changes by: st...@cvs.openbsd.org   2016/11/28 17:35:33

Modified files:
mail/neomutt   : Makefile distinfo 
mail/neomutt/pkg: PLIST 

Log message:
update to neomutt-20161126

CVS: cvs.openbsd.org: ports

2016-11-28 Thread Stuart Henderson 
CVSROOT:/cvs
Module name:ports
Changes by: st...@cvs.openbsd.org   2016/11/28 17:27:23

Modified files:
graphics/opencv: Makefile distinfo 
graphics/opencv/patches: patch-apps_haartraining_CMakeLists_txt 
 patch-cmake_OpenCVDetectPython_cmake 
 
patch-modules_core_include_opencv2_core_version_hpp 
graphics/opencv/pkg: PLIST-docs PLIST-java PLIST-main 

Log message:
update to opencv-2.4.13, from maintainer Rafael Sadowski with a few tweaks by me
and ok with him.

Re: [UPDATE] nagios/cnagios to 0.33

2016-11-28 Thread Kirill Bychkov 
On Mon, November 28, 2016 17:29, Tim Kuijsten wrote:
> Op 28-11-16 om 15:11 schreef Kirill Bychkov:
>> On Mon, November 28, 2016 16:51, Tim Kuijsten wrote:
>>> Op 28-11-16 om 14:37 schreef Kirill Bychkov:
 On Mon, November 28, 2016 02:22, Tim Kuijsten wrote:
> Hi ports@,
>
> I've updated cnagios to the latest release. Quite some segfaults have
> been fixed since 0.30.
>
> Tested on amd64.
>
> -Tim
>
> ps. I'm still looking for an easy way to use it with icinga. Maybe a
> cnagios package on it's own instead of being a subpackage of nagios
> would be better. I'm not sure..
>

 Hmm. It tries to detect data version of status.dat on configure stage.
 If there is no file, configure fails.
>>>
>>> I think this is related to "Remove dependencies on live nagios instance
>>> #6" [1].
>>
>> That's it.
>>
>>>
 For my icinga installation it detects version3. This could be overrided
 with CONFIGURE_ARGS = --with-nagios-data=3.
 But what should be set for nagios itself? Any nagios users around?
>>>
>>> I think status file format 3 is the latest in both nagios and icinga.
>>
>> Thanks to sane defaults in OpenBSD packages! It is enough to install nagios
>> and run rcctls start nagios to make it run and collect some data for local
>> host!
>> But cnagios tells me that nagios uses data version 4:
>> [...]
>> nagios var dir... /var/nagios
>> nagios status file... /var/nagios/status.dat
>> nagios data version... 4
>> [...]
>>
>>
>>>
>>> -Tim
>>>
>>> [1] https://github.com/dannywarren/cnagios/issues/6
>>>

 WIP diff attached. Currently only icinga FLAVOR could be installed.
 Default flavor will fail if there is no /var/nagios/status.dat.

 P.S. Briefly tested with icinga.

>>
>> Updated patch is attached.
>> OK to commit?
>>
>
> There is a typo in the diff: *+Upsatream* commit
> 3bd27fb40e68f61ffd01bea6234b919a667b6fe4
>
Thanks! Fixed in my tree.

attn: collectd kvm plugins broken

2016-11-28 Thread Landry Breuil 
Hi,

since some weeks in current, kvm access is limited, and the code trying
to access it should move to use sysctl to collect system information,
which doesnt require root access. Thus, collectd swap, processes
and tcpconns plugins are broken at runtime.

https://github.com/collectd/collectd/search?q=kvm

If anyone feels bored and wants to tackle this, be my guest.

Landry

UPDATE: OfflineIMAP v7.0.10

2016-11-28 Thread Remi Locherer 
Hi,

Today OfflineIMAP v7.0.10 was released.

It works fine for me synching mails between local maildir and dovecot on
OpenBSD 6.0.

>From the announcemnet mail:
--
This release is mainly about improving reliability. The biggest changes are
about comparing the local and remote structures of folders.

The Gmail repository type allows to tune some predefined options for advanced
use cases.

Offlineimap learns where to find the default OpenSUSE certificate.

Some code refactoring and documentation improvements
--

I did not send an update for the previous release 7.0.9 which was
mainly fixes for the blinkenlights ui.


Remi


Index: Makefile
===
RCS file: /cvs/ports/mail/offlineimap/Makefile,v
retrieving revision 1.38
diff -u -p -r1.38 Makefile
--- Makefile23 Oct 2016 11:10:21 -  1.38
+++ Makefile28 Nov 2016 21:16:41 -
@@ -2,7 +2,7 @@
 
 COMMENT=   powerful IMAP/Maildir synchronization and reader support
 
-MODPY_EGG_VERSION = 7.0.8
+MODPY_EGG_VERSION = 7.0.10
 GH_ACCOUNT =   OfflineIMAP
 GH_PROJECT =   offlineimap
 GH_TAGNAME =   v${MODPY_EGG_VERSION}
Index: distinfo
===
RCS file: /cvs/ports/mail/offlineimap/distinfo,v
retrieving revision 1.26
diff -u -p -r1.26 distinfo
--- distinfo23 Oct 2016 11:10:21 -  1.26
+++ distinfo28 Nov 2016 21:16:41 -
@@ -1,2 +1,2 @@
-SHA256 (offlineimap-7.0.8.tar.gz) = 
Gdp3PBSG09BuvgTkmfSOV9QiQUlYJnKnSHax8VSBvWo=
-SIZE (offlineimap-7.0.8.tar.gz) = 691624
+SHA256 (offlineimap-7.0.10.tar.gz) = 
clQxwgd10lJO2CPaNWsLj3z2J75RjKr86AjyPXl9FUE=
+SIZE (offlineimap-7.0.10.tar.gz) = 694812

Re: UPDATE productivity/baikal 0.4.6

2016-11-28 Thread Remi Locherer 
On Mon, Nov 14, 2016 at 11:11:11PM +0100, Remi Locherer wrote:
> Hi
> 
> This is an update of baikal from 0.2.7 to 0.4.6.
> 
> I havn't tested an update with an actual database. An existing database
> will not be overwritten by this update.o
> 
> Remi

ping

Version 0.2.7 is from February 2014. There have been many releases
inbetween. I recommend to either update the port or delete it.

> 
> Index: Makefile
> ===
> RCS file: /cvs/ports/productivity/baikal/Makefile,v
> retrieving revision 1.12
> diff -u -p -r1.12 Makefile
> --- Makefile  4 Feb 2016 13:14:15 -   1.12
> +++ Makefile  14 Nov 2016 21:56:41 -
> @@ -2,35 +2,30 @@
>  
>  COMMENT= lightweight CalDAV and CardDAV server
>  
> -DISTNAME=baikal-regular-0.2.7
> -PKGNAME= ${DISTNAME:S/-regular//}
> -EXTRACT_SUFX=.tgz
> +VERSION= 0.4.6
> +DISTNAME=baikal-${VERSION}
> +MASTER_SITES=
> https://github.com/fruux/Baikal/releases/download/${VERSION}/
> +DISTFILES=   ${DISTNAME}.zip
>  CATEGORIES=  productivity www
> -REVISION=2
>  
> -HOMEPAGE=http://baikal-server.com/
> +HOMEPAGE=http://sabre.io/baikal/
>  
>  # GPLv3
>  PERMIT_PACKAGE_CDROM=Yes
>  
> -MASTER_SITES=http://baikal-server.com/get/
> -
>  MODULES= lang/php
>  
> -WRKDIST= ${WRKDIR}/baikal-regular/
> -
>  NO_BUILD=Yes
>  NO_TEST= Yes
>  PKG_ARCH=*
>  
> -PREFIX=  ${VARBASE}/www
> +PREFIX=  ${VARBASE}/www
>  INSTDIR= ${PREFIX}/baikal
> +WRKSRC=  ${WRKDIR}/baikal
>  SUBST_VARS=  INSTDIR
>  
>  do-install:
>   ${INSTALL_DATA_DIR} ${INSTDIR}
>   cd ${WRKSRC} && tar cf - . | (cd ${INSTDIR}; tar xf -)
> - find ${INSTDIR} -name '*.orig' -exec rm {} \;
> - mv ${INSTDIR}/Specific/db/db.sqlite{,.dist}
>  
>  .include 
> Index: distinfo
> ===
> RCS file: /cvs/ports/productivity/baikal/distinfo,v
> retrieving revision 1.3
> diff -u -p -r1.3 distinfo
> --- distinfo  9 Jun 2014 13:12:53 -   1.3
> +++ distinfo  14 Nov 2016 21:56:41 -
> @@ -1,2 +1,2 @@
> -SHA256 (baikal-regular-0.2.7.tgz) = 
> Jo2u7Ja58bv+4MFjmK4hOr/oJ8l+xIcXmYWCi2JnNZ8=
> -SIZE (baikal-regular-0.2.7.tgz) = 2273699
> +SHA256 (baikal-0.4.6.zip) = lG6OQWH374S+QkMLbp07t91Lu+JBtAm+IIwURH16p6Y=
> +SIZE (baikal-0.4.6.zip) = 3288478
> Index: patches/patch-Specific_virtualhosts_baikal_nginx
> ===
> RCS file: patches/patch-Specific_virtualhosts_baikal_nginx
> diff -N patches/patch-Specific_virtualhosts_baikal_nginx
> --- patches/patch-Specific_virtualhosts_baikal_nginx  13 Sep 2013 15:27:58 
> -  1.1
> +++ /dev/null 1 Jan 1970 00:00:00 -
> @@ -1,35 +0,0 @@
> -$OpenBSD: patch-Specific_virtualhosts_baikal_nginx,v 1.1 2013/09/13 15:27:58 
> ajacoutot Exp $
>  Specific/virtualhosts/baikal.nginx.orig  Fri Sep 13 15:10:51 2013
> -+++ Specific/virtualhosts/baikal.nginx   Fri Sep 13 15:12:06 2013
> -@@ -2,11 +2,11 @@ server {
> - listen   80;
> - server_name  dav.mydomain.com;
> - 
> --root  /var/www/dav.mydomain.com;
> -+root  /var/www/baikal/html;
> - index index.php;
> - 
> --rewrite ^/.well-known/caldav /cal.php redirect;
> --rewrite ^/.well-known/carddav /card.php redirect;
> -+rewrite ^/.well-known/caldav /cal.php redirect;
> -+rewrite ^/.well-known/carddav /card.php redirect;
> - 
> - charset utf-8;
> - 
> -@@ -18,7 +18,7 @@ server {
> - location ~ ^(.+\.php)(.*)$ {
> - try_files $fastcgi_script_name =404;
> - fastcgi_split_path_info  ^(.+\.php)(.*)$;
> --fastcgi_pass   unix:/var/run/php-fpm/php-fpm.sock;
> -+fastcgi_pass   127.0.0.1:9000;
> - fastcgi_param  SCRIPT_FILENAME  $document_root$fastcgi_script_name;
> - fastcgi_param  PATH_INFO$fastcgi_path_info;
> - include/etc/nginx/fastcgi_params;
> -@@ -26,6 +26,6 @@ server {
> - 
> - error_page   500 502 503 504  /50x.html;
> - location = /50x.html {
> --root   /var/www/nginx-default;
> -+root   /var/www/htdocs;
> - }
> - }
> Index: pkg/PLIST
> ===
> RCS file: /cvs/ports/productivity/baikal/pkg/PLIST,v
> retrieving revision 1.5
> diff -u -p -r1.5 PLIST
> --- pkg/PLIST 9 Jun 2014 13:12:53 -   1.5
> +++ pkg/PLIST 14 Nov 2016 21:56:42 -
> @@ -1,13 +1,13 @@
> -@comment $OpenBSD: PLIST,v 1.5 2014/06/09 13:12:53 jasper Exp $
> -@ask-update baikal-<0.2.4p1 Due to a bug in older versions of this package, 
> backup baikal/Specific/db/db.sqlite before updating and restore it afterwards.
> +@comment $OpenBSD$
>  baikal/
> -baikal/ChangeLog.md
> +baikal/CHANGELOG.md
>  baikal/Core/
>  baikal/Core/Distrib.php
>  baikal/Core/Frameworks/
>  baikal/Core/Frameworks/Baikal/
>  baikal/Core/Frameworks/Baikal/Core/
>

Re: [UPDATE] sysutils/collectd + [NEW] sysutils/riemann-c-client

2016-11-28 Thread Landry Breuil 
On Mon, Nov 28, 2016 at 10:39:32PM +0300, Pavel Korovin wrote:
> On 11/27, Landry Breuil wrote:
> > Finally got around testing this, and saw the same things. But if you run
> > the check_libriemann manually, you get this:
> > 
> > $ktrace tests/check_libriemann
> > Running suite(s): Riemann C client library tests
> > 97%: Checks: 37, Failures: 0, Errors: 1
> > tests/check_client.c:29:E:Client:test_riemann_client_connect:0: (after
> > this point) Test timeout expired
> > 
> > So something about fd redirections or workdir ?
> > 
> > Is sysutils/ the right directory for such a client library ?
>  
> 
> Hi Landry,
> 
> I suppose you have libcheck installed, because I don't even have test target
> available without it.
> And it seems that check target requires riemann, do you have it installed?

Pretty sure i had no riemann installed, that was on my work desktop,
will have to check tmrw. Either way, we should set NO_TEST=Yes if we
dont really know the right requirements and nobody wants to spend time
into doing this properly, or... do it properly :)

Landry

Re: [UPDATE] sysutils/collectd + [NEW] sysutils/riemann-c-client

2016-11-28 Thread Pavel Korovin 
On 11/27, Landry Breuil wrote:
> Finally got around testing this, and saw the same things. But if you run
> the check_libriemann manually, you get this:
> 
> $ktrace tests/check_libriemann
> Running suite(s): Riemann C client library tests
> 97%: Checks: 37, Failures: 0, Errors: 1
> tests/check_client.c:29:E:Client:test_riemann_client_connect:0: (after
> this point) Test timeout expired
> 
> So something about fd redirections or workdir ?
> 
> Is sysutils/ the right directory for such a client library ?
 

Hi Landry,

I suppose you have libcheck installed, because I don't even have test target
available without it.
And it seems that check target requires riemann, do you have it installed?

-- 
With best regards,
Pavel Korovin

Re: [UPDATE] devel/p5-Sub-Identify to 0.12

2016-11-28 Thread Alexander Bluhm 
On Fri, Nov 25, 2016 at 06:49:35PM -0700, Andrew Fresh wrote:
> On Fri, Nov 25, 2016 at 11:13:41PM +0100, Alexander Bluhm wrote:
> > On Thu, Nov 24, 2016 at 01:07:03PM -0700, Andrew Fresh wrote:
> > > devel/p5-namespace-autoclean fails tests both with and without these
> > > changes.
> > 
> > I am testing the Perl 5.24.1-RC4 update.  With that it passes.
> 
> Aha, for some reason it's missing the p5-Test-Requires hard requirement.
> With this patch, more tests run and they all pass, even after
> pkg_deleteing everything before `make test`.

Usually I do a 'grep use' in the t directory.  Not all dependencies
are recorded properly in various perl modules.

I think you should add devel/p5-Sub-Install although it is currently
installed by a transitive dependency.

> 
> Index: Makefile
> ===
> RCS file: /cvs/ports/devel/p5-namespace-autoclean/Makefile,v
> retrieving revision 1.10
> diff -u -p -r1.10 Makefile
> --- Makefile  20 Mar 2016 19:56:42 -  1.10
> +++ Makefile  25 Nov 2016 23:50:42 -
> @@ -20,4 +20,9 @@ RUN_DEPENDS =   devel/p5-B-Hooks-EndOfScop
>  TEST_DEPENDS = devel/p5-Moose \
> devel/p5-Sub-Name

The indentation is wrong here.  Space vs. Tab

>  
> +TEST_DEPENDS +=  devel/p5-Test-Requires \
> + devel/p5-Moo \
> + devel/p5-MooseX-Role-WithOverloading \
> + devel/p5-Mouse

Why another TEST_DEPENDS += block?  Can you put it in one?

> +
>  .include 

anyway OK bluhm@

Re: NEW: security/p5-Digest-SHA

2016-11-28 Thread Giovanni Bechis 

Il giorno Lunedi, 28 Novembre, 2016 18:48 CET, Andrew Fresh 
 ha scritto:

> On Mon, Nov 28, 2016 at 05:26:42PM +, Nigel Taylor wrote:
> > On 11/28/16 15:38, Jeremie Courreges-Anglas wrote:
> > > Giovanni Bechis  writes:
> > >
> > >> pkg/DESCR:
> > >> Digest::SHA is a complete implementation of the NIST Secure Hash
> > >> Standard.
> 
> > No you don't do this, fix what in base or wait for perl 5.26 to go into
> > base, your only adding what's going to have to be removed if put in
> > ports, unless
> >
> > ===>  Checking files for p5-Module-CoreList-5.20161120
>
> I do like how you pick a module that's in core, but also in ports as
> your example. (This one is pretty useful to be able to update outside of
> base perl though).
>
> 
> > $ perl -Ilib ./corelist -a Digest::SHA | tail
> >   v5.24.05.95
> >   v5.25.05.95
> >   v5.25.15.95
> >   v5.25.25.95
> >   v5.25.35.95
> >   v5.25.45.96
> >   v5.25.55.96
> >   v5.25.65.96
> >   v5.25.75.96
> >
> > As can be seen 5.96 is in perl v5.25.4 onwards, which means it goes into
> > base so far it's not being removed from perl core. Having two version of
> > the same thing is not a good idea, as can end up using the wrong version.
> >
> > $ corelist Digest::SHA
> >
> > Data for 2015-09-12
> > Digest::SHA was first released with perl v5.9.3
> >
> > $ perl -e 'use Digest::SHA; print $Digest::SHA::VERSION."\n";'
> > 5.88_01
> >
> > Also as in previous base releases these should also be patched in base,
> > if this need fixing.
>
> I agree that there's no real use in doing this, since I hope to get
> 5.24.1 imported shortly after it's released, which will include 5.95,
> plus the patch to remove '.' from @INC.
>
> AFAIK there is still "discussion" about how best to fix this "remove '.'
> from @INC" in the most backwards compatible way for 5.24.1 and 5.22.3.
>
>
> Handily for 5.26 there will be a flag to build perl without it in @INC
> at all, which I did backport to 5.24, tho that will not go in with the
> update, so we can see what fallout comes from that specifically.
>
> > The 5.88_01 suggests that the version in base might have been modified
> > in some way, or patched to fix problems in the past, which your now losing.
>
> It was, to remove the '.' from @INC, which is the change between 5.95
> and 5.96.
>
> https://metacpan.org/changes/distribution/Digest-SHA
my laptop did not notice it was on base as well ,I think I have something wrong 
in @INC.
I think it is wise to wait for newer Perl.
 Thanks & Cheers
   Giovanni

Re: NEW: security/p5-Digest-SHA

2016-11-28 Thread Andrew Fresh 
On Mon, Nov 28, 2016 at 05:26:42PM +, Nigel Taylor wrote:
> On 11/28/16 15:38, Jeremie Courreges-Anglas wrote:
> > Giovanni Bechis  writes:
> > 
> >> pkg/DESCR:
> >> Digest::SHA is a complete implementation of the NIST Secure Hash
> >> Standard.

> No you don't do this, fix what in base or wait for perl 5.26 to go into
> base, your only adding what's going to have to be removed if put in
> ports, unless
> 
> ===>  Checking files for p5-Module-CoreList-5.20161120

I do like how you pick a module that's in core, but also in ports as
your example. (This one is pretty useful to be able to update outside of
base perl though).


> $ perl -Ilib ./corelist -a Digest::SHA | tail
>   v5.24.05.95
>   v5.25.05.95
>   v5.25.15.95
>   v5.25.25.95
>   v5.25.35.95
>   v5.25.45.96
>   v5.25.55.96
>   v5.25.65.96
>   v5.25.75.96
> 
> As can be seen 5.96 is in perl v5.25.4 onwards, which means it goes into
> base so far it's not being removed from perl core. Having two version of
> the same thing is not a good idea, as can end up using the wrong version.
>
> $ corelist Digest::SHA
> 
> Data for 2015-09-12
> Digest::SHA was first released with perl v5.9.3
> 
> $ perl -e 'use Digest::SHA; print $Digest::SHA::VERSION."\n";'
> 5.88_01
> 
> Also as in previous base releases these should also be patched in base,
> if this need fixing.

I agree that there's no real use in doing this, since I hope to get
5.24.1 imported shortly after it's released, which will include 5.95,
plus the patch to remove '.' from @INC.

AFAIK there is still "discussion" about how best to fix this "remove '.'
from @INC" in the most backwards compatible way for 5.24.1 and 5.22.3.


Handily for 5.26 there will be a flag to build perl without it in @INC
at all, which I did backport to 5.24, tho that will not go in with the
update, so we can see what fallout comes from that specifically.

> The 5.88_01 suggests that the version in base might have been modified
> in some way, or patched to fix problems in the past, which your now losing.

It was, to remove the '.' from @INC, which is the change between 5.95
and 5.96.

https://metacpan.org/changes/distribution/Digest-SHA

Re: NEW: security/p5-Digest-SHA

2016-11-28 Thread Nigel Taylor 
On 11/28/16 15:38, Jeremie Courreges-Anglas wrote:
> Giovanni Bechis  writes:
> 
>> pkg/DESCR:
>> Digest::SHA is a complete implementation of the NIST Secure Hash
>> Standard.
>> It gives Perl programmers a convenient way to calculate
>> SHA-1, SHA-224, SHA-256, SHA-384, SHA-512, SHA-512/224, and SHA-512/256
>> message digests.
>> The module can handle all types of input, including partial-byte data.
>>
>>  Comments ? Ok ?
> 
> I see this on amd64:
> 
> /usr/bin/perl /usr/libdata/perl5/ExtUtils/xsubpp  -typemap 
> /usr/libdata/perl5/ExtUtils/typemap -typemap typemap  SHA.xs > SHA.xsc && mv 
> SHA.xsc SHA.c
> cc -c  -I.  -DNO_LOCALE_NUMERIC -DNO_LOCALE_COLLATE -fno-strict-aliasing 
> -fno-delete-null-pointer-checks -pipe -fstack-protector -I/usr/local/include 
> -O2-DVERSION=\"5.96\"  -DXS_VERSION=\"5.96\" -DPIC -fPIC  
> "-I/usr/libdata/perl5/amd64-openbsd/5.20.3/CORE"   SHA.c
> In file included from SHA.xs:35:
> src/sha.c: In function 'sharewind':
> src/sha.c:297: warning: array size (32) smaller than bound length (64)
> src/sha.c:297: warning: array size (32) smaller than bound length (64)
> src/sha.c:298: warning: array size (32) smaller than bound length (64)
> src/sha.c:298: warning: array size (32) smaller than bound length (64)
> src/sha.c:299: warning: array size (32) smaller than bound length (64)
> src/sha.c:299: warning: array size (32) smaller than bound length (64)
> 
> It doesn't *look* harmful but better fix it and give upstream a heads-up
> (-Wbounded is OpenBSD-only, afaik).
> 
> Why the extra newlines in pkg/DESCR?  Looks fine otherwise.
> 
No you don't do this, fix what in base or wait for perl 5.26 to go into
base, your only adding what's going to have to be removed if put in
ports, unless

===>  Checking files for p5-Module-CoreList-5.20161120
`/usr/ports/distfiles/Module-CoreList-5.20161120.tar.gz' is up to date.
>> (SHA256) Module-CoreList-5.20161120.tar.gz: OK
===>  Extracting for p5-Module-CoreList-5.20161120
===>  Patching for p5-Module-CoreList-5.20161120
===>  Configuring for p5-Module-CoreList-5.20161120
Checking if your kit is complete...
Looks good
Generating a Unix-style Makefile
Writing Makefile for Module::CoreList
Writing MYMETA.yml and MYMETA.json
===>  Building for p5-Module-CoreList-5.20161120
cp lib/Module/CoreList/TieHashDelta.pm
blib/lib/Module/CoreList/TieHashDelta.pm
cp lib/Module/CoreList.pod blib/lib/Module/CoreList.pod
cp lib/Module/CoreList.pm blib/lib/Module/CoreList.pm
cp lib/Module/CoreList/Utils.pm blib/lib/Module/CoreList/Utils.pm
cp corelist blib/script/corelist
/usr/bin/perl -MExtUtils::MY -e 'MY->fixin(shift)' -- blib/script/corelist
Manifying blib/man1/corelist.1
Manifying blib/man3/Module::CoreList.3p
Manifying blib/man3/Module::CoreList::Utils.3p
$ cd `make show=WRKSRC`
$ perl -Ilib ./corelist -a Digest::SHA | tail
  v5.24.05.95
  v5.25.05.95
  v5.25.15.95
  v5.25.25.95
  v5.25.35.95
  v5.25.45.96
  v5.25.55.96
  v5.25.65.96
  v5.25.75.96

As can be seen 5.96 is in perl v5.25.4 onwards, which means it goes into
base so far it's not being removed from perl core. Having two version of
the same thing is not a good idea, as can end up using the wrong version.


$ corelist Digest::SHA

Data for 2015-09-12
Digest::SHA was first released with perl v5.9.3

$ perl -e 'use Digest::SHA; print $Digest::SHA::VERSION."\n";'
5.88_01

Also as in previous base releases these should also be patched in base,
if this need fixing.

The 5.88_01 suggests that the version in base might have been modified
in some way, or patched to fix problems in the past, which your now losing.

CVS: cvs.openbsd.org: ports

2016-11-28 Thread Alexander Bluhm 
CVSROOT:/cvs
Module name:ports
Changes by: bl...@cvs.openbsd.org   2016/11/28 09:55:09

Modified files:
converters/p5-Net-IDN-Encode: Makefile 
Added files:
converters/p5-Net-IDN-Encode/patches: 
  patch-lib_Net_IDN_Punycode_xs 

Log message:
With the canaries turned on in malloc.conf, running the regression
tests of p5-Net-IDN-Encode crashed the Perl process.  There was no
proper protection against heap overflow in the XS part.  Grow the
destination string in encode_punycode() and decode_punycode() before
writing to it.  This has to be checked every time.
Take maintainer.
OK jca@

Re: NEW: security/p5-Digest-SHA

2016-11-28 Thread Jeremie Courreges-Anglas 
Giovanni Bechis  writes:

> pkg/DESCR:
> Digest::SHA is a complete implementation of the NIST Secure Hash
> Standard.
> It gives Perl programmers a convenient way to calculate
> SHA-1, SHA-224, SHA-256, SHA-384, SHA-512, SHA-512/224, and SHA-512/256
> message digests.
> The module can handle all types of input, including partial-byte data.
>
>  Comments ? Ok ?

I see this on amd64:

/usr/bin/perl /usr/libdata/perl5/ExtUtils/xsubpp  -typemap 
/usr/libdata/perl5/ExtUtils/typemap -typemap typemap  SHA.xs > SHA.xsc && mv 
SHA.xsc SHA.c
cc -c  -I.  -DNO_LOCALE_NUMERIC -DNO_LOCALE_COLLATE -fno-strict-aliasing 
-fno-delete-null-pointer-checks -pipe -fstack-protector -I/usr/local/include 
-O2-DVERSION=\"5.96\"  -DXS_VERSION=\"5.96\" -DPIC -fPIC  
"-I/usr/libdata/perl5/amd64-openbsd/5.20.3/CORE"   SHA.c
In file included from SHA.xs:35:
src/sha.c: In function 'sharewind':
src/sha.c:297: warning: array size (32) smaller than bound length (64)
src/sha.c:297: warning: array size (32) smaller than bound length (64)
src/sha.c:298: warning: array size (32) smaller than bound length (64)
src/sha.c:298: warning: array size (32) smaller than bound length (64)
src/sha.c:299: warning: array size (32) smaller than bound length (64)
src/sha.c:299: warning: array size (32) smaller than bound length (64)

It doesn't *look* harmful but better fix it and give upstream a heads-up
(-Wbounded is OpenBSD-only, afaik).

Why the extra newlines in pkg/DESCR?  Looks fine otherwise.

-- 
jca | PGP : 0x1524E7EE / 5135 92C1 AD36 5293 2BDF  DDCC 0DFA 74AE 1524 E7EE

Re: graphics/imlib2: imlib2-config vs. pkg-config

2016-11-28 Thread Christian Weisgerber 
On 2016-11-27, Jeremie Courreges-Anglas  wrote:

>>> > imlib2-config and pkg-config imlib2 don't agree:
>>> 
>>> I think that "-L/usr/local/lib -L/usr/X11R6/lib -lImlib2" and
>>> "-I/usr/local/include -I/usr/X11R6/include" should be enough.
>>
>> Yes to "-I/usr/local/include -I/usr/X11R6/include".
>>
>> However, in "-L/usr/local/lib -L/usr/X11R6/lib -lImlib2" the X11R6 path
>> is useless.  It should be either just
>>
>> -L/usr/local/lib -lImlib2
>
> I'm not sure I understand, afaik it would be enough if /usr/X11R6/lib
> was in the default search path of ld(1).

Oh, you're right.  I forgot that the interlibrary dependencies don't
include the path.  (Unless rpath is used, which we usually don't
do.)

-- 
Christian "naddy" Weisgerber  na...@mips.inka.de

Re: [UPDATE] nagios/cnagios to 0.33

2016-11-28 Thread Tim Kuijsten 

Op 28-11-16 om 15:11 schreef Kirill Bychkov:

On Mon, November 28, 2016 16:51, Tim Kuijsten wrote:

Op 28-11-16 om 14:37 schreef Kirill Bychkov:

On Mon, November 28, 2016 02:22, Tim Kuijsten wrote:

Hi ports@,

I've updated cnagios to the latest release. Quite some segfaults have
been fixed since 0.30.

Tested on amd64.

-Tim

ps. I'm still looking for an easy way to use it with icinga. Maybe a
cnagios package on it's own instead of being a subpackage of nagios
would be better. I'm not sure..



Hmm. It tries to detect data version of status.dat on configure stage.
If there is no file, configure fails.


I think this is related to "Remove dependencies on live nagios instance
#6" [1].


That's it.




For my icinga installation it detects version3. This could be overrided
with CONFIGURE_ARGS = --with-nagios-data=3.
But what should be set for nagios itself? Any nagios users around?


I think status file format 3 is the latest in both nagios and icinga.


Thanks to sane defaults in OpenBSD packages! It is enough to install nagios
and run rcctls start nagios to make it run and collect some data for local
host!
But cnagios tells me that nagios uses data version 4:
[...]
nagios var dir... /var/nagios
nagios status file... /var/nagios/status.dat
nagios data version... 4
[...]




-Tim

[1] https://github.com/dannywarren/cnagios/issues/6



WIP diff attached. Currently only icinga FLAVOR could be installed.
Default flavor will fail if there is no /var/nagios/status.dat.

P.S. Briefly tested with icinga.



Updated patch is attached.
OK to commit?



There is a typo in the diff: *+Upsatream* commit 
3bd27fb40e68f61ffd01bea6234b919a667b6fe4

Re: [UPDATE] nagios/cnagios to 0.33

2016-11-28 Thread Kirill Bychkov 
On Mon, November 28, 2016 16:51, Tim Kuijsten wrote:
> Op 28-11-16 om 14:37 schreef Kirill Bychkov:
>> On Mon, November 28, 2016 02:22, Tim Kuijsten wrote:
>>> Hi ports@,
>>>
>>> I've updated cnagios to the latest release. Quite some segfaults have
>>> been fixed since 0.30.
>>>
>>> Tested on amd64.
>>>
>>> -Tim
>>>
>>> ps. I'm still looking for an easy way to use it with icinga. Maybe a
>>> cnagios package on it's own instead of being a subpackage of nagios
>>> would be better. I'm not sure..
>>>
>>
>> Hmm. It tries to detect data version of status.dat on configure stage.
>> If there is no file, configure fails.
>
> I think this is related to "Remove dependencies on live nagios instance
> #6" [1].

That's it.

>
>> For my icinga installation it detects version3. This could be overrided
>> with CONFIGURE_ARGS = --with-nagios-data=3.
>> But what should be set for nagios itself? Any nagios users around?
>
> I think status file format 3 is the latest in both nagios and icinga.

Thanks to sane defaults in OpenBSD packages! It is enough to install nagios
and run rcctls start nagios to make it run and collect some data for local
host!
But cnagios tells me that nagios uses data version 4:
[...]
nagios var dir... /var/nagios
nagios status file... /var/nagios/status.dat
nagios data version... 4
[...]


>
> -Tim
>
> [1] https://github.com/dannywarren/cnagios/issues/6
>
>>
>> WIP diff attached. Currently only icinga FLAVOR could be installed.
>> Default flavor will fail if there is no /var/nagios/status.dat.
>>
>> P.S. Briefly tested with icinga.
>>

Updated patch is attached.
OK to commit?Index: Makefile
===
RCS file: /cvs/ports/net/nagios/cnagios/Makefile,v
retrieving revision 1.22
diff -u -p -r1.22 Makefile
--- Makefile	8 Jul 2016 16:48:21 -	1.22
+++ Makefile	28 Nov 2016 14:06:51 -
@@ -1,9 +1,11 @@
 # $OpenBSD: Makefile,v 1.22 2016/07/08 16:48:21 sthen Exp $
 
-COMMENT =		console interface for nagios
+COMMENT =		console interface for nagios/icinga
+
+GH_ACCOUNT =		dannywarren
+GH_PROJECT =		cnagios
+GH_TAGNAME =		0.33
 
-DISTNAME =		cnagios-0.30
-REVISION =		1
 CATEGORIES =		net
 
 WANTLIB =		c perl pthread curses m
@@ -13,12 +15,25 @@ PERMIT_PACKAGE_CDROM =	Yes
 
 NO_TEST =		Yes
 
-RUN_DEPENDS =		net/nagios/nagios>=3.0
 CONFIGURE_STYLE =	gnu
-CONFIGURE_ARGS =	--with-nagios-data=3 \
-			--with-etc-dir="${SYSCONFDIR}/cnagios" \
-			--with-var-dir="/var/nagios" \
+CONFIGURE_ARGS +=	--with-nagios-data=3 \
+			--with-etc-dir="${SYSCONFDIR}/cnagios"
+
+FLAVORS =		icinga
+FLAVOR ?=
+
+.if ${FLAVOR:Micinga}
+RUN_DEPENDS =		net/icinga/core
+CONFIGURE_ARGS =	--with-var-dir="/var/icinga" \
+			--with-nagios-data=3 \
+			--with-status-file="/var/icinga/status.dat"
+.else
+RUN_DEPENDS =		net/nagios/nagios>=3.0
+CONFIGURE_ARGS =	--with-var-dir="/var/nagios" \
+			--with-nagios-data=4 \
 			--with-status-file="/var/nagios/status.dat"
+.endif
+
 MODGNU_CONFIG_GUESS_DIRS=${WRKSRC}/config
 
 ALL_TARGET =		cnagios
Index: distinfo
===
RCS file: /cvs/ports/net/nagios/cnagios/distinfo,v
retrieving revision 1.7
diff -u -p -r1.7 distinfo
--- distinfo	1 Apr 2013 18:26:00 -	1.7
+++ distinfo	28 Nov 2016 14:06:51 -
@@ -1,2 +1,2 @@
-SHA256 (cnagios-0.30.tar.gz) = oHd0oOs7RZGSplgQICTYdNH/dvnx9/6PIG7LADb/FDg=
-SIZE (cnagios-0.30.tar.gz) = 81813
+SHA256 (cnagios-0.33.tar.gz) = YLM89YK1cAWCAG3gVXDFsy4k2qnwQjXmG+TNSZ/CVhE=
+SIZE (cnagios-0.33.tar.gz) = 82080
Index: patches/patch-version_h
===
RCS file: patches/patch-version_h
diff -N patches/patch-version_h
--- /dev/null	1 Jan 1970 00:00:00 -
+++ patches/patch-version_h	28 Nov 2016 14:06:51 -
@@ -0,0 +1,10 @@
+$OpenBSD$
+
+Bump version.
+Upsatream commit 3bd27fb40e68f61ffd01bea6234b919a667b6fe4
+
+--- version.h.orig	Mon Nov 28 16:11:54 2016
 version.h	Mon Nov 28 16:12:08 2016
+@@ -1 +1 @@
+-#define VERSION "0.32"
++#define VERSION "0.33"
Index: pkg/DESCR
===
RCS file: /cvs/ports/net/nagios/cnagios/pkg/DESCR,v
retrieving revision 1.1.1.1
diff -u -p -r1.1.1.1 DESCR
--- pkg/DESCR	5 Feb 2008 23:00:22 -	1.1.1.1
+++ pkg/DESCR	28 Nov 2016 14:06:51 -
@@ -1,4 +1,4 @@
-Cnagios is a full-screen terminal interface for viewing Nagios
+Cnagios is a full-screen terminal interface for viewing Nagios/Icinga
 HOST and SERVICE objects, and the durations of their current states.
 
 It's lightning fast because it's written in C using the curses

Re: [UPDATE] nagios/cnagios to 0.33

2016-11-28 Thread Tim Kuijsten 

Op 28-11-16 om 14:37 schreef Kirill Bychkov:

On Mon, November 28, 2016 02:22, Tim Kuijsten wrote:

Hi ports@,

I've updated cnagios to the latest release. Quite some segfaults have
been fixed since 0.30.

Tested on amd64.

-Tim

ps. I'm still looking for an easy way to use it with icinga. Maybe a
cnagios package on it's own instead of being a subpackage of nagios
would be better. I'm not sure..



Hmm. It tries to detect data version of status.dat on configure stage.
If there is no file, configure fails.


I think this is related to "Remove dependencies on live nagios instance 
#6" [1].



For my icinga installation it detects version3. This could be overrided
with CONFIGURE_ARGS = --with-nagios-data=3.
But what should be set for nagios itself? Any nagios users around?


I think status file format 3 is the latest in both nagios and icinga.

-Tim

[1] https://github.com/dannywarren/cnagios/issues/6



WIP diff attached. Currently only icinga FLAVOR could be installed.
Default flavor will fail if there is no /var/nagios/status.dat.

P.S. Briefly tested with icinga.

Re: [UPDATE] nagios/cnagios to 0.33

2016-11-28 Thread Kirill Bychkov 
On Mon, November 28, 2016 02:22, Tim Kuijsten wrote:
> Hi ports@,
>
> I've updated cnagios to the latest release. Quite some segfaults have
> been fixed since 0.30.
>
> Tested on amd64.
>
> -Tim
>
> ps. I'm still looking for an easy way to use it with icinga. Maybe a
> cnagios package on it's own instead of being a subpackage of nagios
> would be better. I'm not sure..
>

Hmm. It tries to detect data version of status.dat on configure stage.
If there is no file, configure fails.
For my icinga installation it detects version3. This could be overrided
with CONFIGURE_ARGS = --with-nagios-data=3.
But what should be set for nagios itself? Any nagios users around?

WIP diff attached. Currently only icinga FLAVOR could be installed.
Default flavor will fail if there is no /var/nagios/status.dat.

P.S. Briefly tested with icinga.
Index: Makefile
===
RCS file: /cvs/ports/net/nagios/cnagios/Makefile,v
retrieving revision 1.22
diff -u -p -r1.22 Makefile
--- Makefile	8 Jul 2016 16:48:21 -	1.22
+++ Makefile	28 Nov 2016 13:34:33 -
@@ -1,9 +1,11 @@
 # $OpenBSD: Makefile,v 1.22 2016/07/08 16:48:21 sthen Exp $
 
-COMMENT =		console interface for nagios
+COMMENT =		console interface for nagios/icinga
+
+GH_ACCOUNT =		dannywarren
+GH_PROJECT =		cnagios
+GH_TAGNAME =		0.33
 
-DISTNAME =		cnagios-0.30
-REVISION =		1
 CATEGORIES =		net
 
 WANTLIB =		c perl pthread curses m
@@ -13,12 +15,24 @@ PERMIT_PACKAGE_CDROM =	Yes
 
 NO_TEST =		Yes
 
-RUN_DEPENDS =		net/nagios/nagios>=3.0
 CONFIGURE_STYLE =	gnu
-CONFIGURE_ARGS =	--with-nagios-data=3 \
-			--with-etc-dir="${SYSCONFDIR}/cnagios" \
-			--with-var-dir="/var/nagios" \
+CONFIGURE_ARGS +=	--with-nagios-data=3 \
+			--with-etc-dir="${SYSCONFDIR}/cnagios"
+
+FLAVORS =		icinga
+FLAVOR ?=
+
+.if ${FLAVOR:Micinga}
+RUN_DEPENDS =		net/icinga/core
+CONFIGURE_ARGS =	--with-var-dir="/var/icinga" \
+			--with-nagios-data=3 \
+			--with-status-file="/var/icinga/status.dat"
+.else
+RUN_DEPENDS =		net/nagios/nagios>=3.0
+CONFIGURE_ARGS =	--with-var-dir="/var/nagios" \
 			--with-status-file="/var/nagios/status.dat"
+.endif
+
 MODGNU_CONFIG_GUESS_DIRS=${WRKSRC}/config
 
 ALL_TARGET =		cnagios
Index: distinfo
===
RCS file: /cvs/ports/net/nagios/cnagios/distinfo,v
retrieving revision 1.7
diff -u -p -r1.7 distinfo
--- distinfo	1 Apr 2013 18:26:00 -	1.7
+++ distinfo	28 Nov 2016 13:34:33 -
@@ -1,2 +1,2 @@
-SHA256 (cnagios-0.30.tar.gz) = oHd0oOs7RZGSplgQICTYdNH/dvnx9/6PIG7LADb/FDg=
-SIZE (cnagios-0.30.tar.gz) = 81813
+SHA256 (cnagios-0.33.tar.gz) = YLM89YK1cAWCAG3gVXDFsy4k2qnwQjXmG+TNSZ/CVhE=
+SIZE (cnagios-0.33.tar.gz) = 82080
Index: patches/patch-version_h
===
RCS file: patches/patch-version_h
diff -N patches/patch-version_h
--- /dev/null	1 Jan 1970 00:00:00 -
+++ patches/patch-version_h	28 Nov 2016 13:34:33 -
@@ -0,0 +1,10 @@
+$OpenBSD$
+
+Bump version.
+Upsatream commit 3bd27fb40e68f61ffd01bea6234b919a667b6fe4
+
+--- version.h.orig	Mon Nov 28 16:11:54 2016
 version.h	Mon Nov 28 16:12:08 2016
+@@ -1 +1 @@
+-#define VERSION "0.32"
++#define VERSION "0.33"
Index: pkg/DESCR
===
RCS file: /cvs/ports/net/nagios/cnagios/pkg/DESCR,v
retrieving revision 1.1.1.1
diff -u -p -r1.1.1.1 DESCR
--- pkg/DESCR	5 Feb 2008 23:00:22 -	1.1.1.1
+++ pkg/DESCR	28 Nov 2016 13:34:33 -
@@ -1,4 +1,4 @@
-Cnagios is a full-screen terminal interface for viewing Nagios
+Cnagios is a full-screen terminal interface for viewing Nagios/Icinga
 HOST and SERVICE objects, and the durations of their current states.
 
 It's lightning fast because it's written in C using the curses

CVS: cvs.openbsd.org: ports

2016-11-28 Thread Benoit Lecocq 
CVSROOT:/cvs
Module name:ports
Changes by: ben...@cvs.openbsd.org  2016/11/28 03:41:40

Modified files:
textproc/highlight: Makefile distinfo 
textproc/highlight/pkg: PLIST 

Log message:
Update to highlight-3.33.

CVS: cvs.openbsd.org: ports

2016-11-28 Thread Antoine Jacoutot 
CVSROOT:/cvs
Module name:ports
Changes by: ajacou...@cvs.openbsd.org   2016/11/28 03:14:23

Modified files:
print/cups-filters: Makefile distinfo 
print/cups-filters/pkg: PLIST 

Log message:
Update to cups-filters-1.12.0.

CVS: cvs.openbsd.org: ports

2016-11-28 Thread David Coppa 
CVSROOT:/cvs
Module name:ports
Changes by: dco...@cvs.openbsd.org  2016/11/28 02:15:13

Modified files:
devel/cmake: Makefile distinfo 
devel/cmake/patches: patch-CMakeLists_txt 
 patch-Modules_CMakeCInformation_cmake 
 patch-Modules_CMakeCXXInformation_cmake 
 patch-Modules_CMakeDetermineJavaCompiler_cmake 
 patch-Modules_CMakeFortranInformation_cmake 
 patch-Modules_CMakeUnixFindMake_cmake 
 patch-Modules_Compiler_G95-Fortran_cmake 
 patch-Modules_Compiler_GNU-Fortran_cmake 
 patch-Modules_Compiler_GNU_cmake 
 patch-Modules_FindBoost_cmake 
 patch-Modules_FindJNI_cmake 
 patch-Modules_FindJava_cmake 
 patch-Modules_FindKDE3_cmake 
 patch-Modules_FindLua_cmake 
 patch-Modules_FindMPEG2_cmake 
 patch-Modules_FindMPEG_cmake 
 patch-Modules_FindPythonInterp_cmake 
 patch-Modules_FindPythonLibs_cmake 
 patch-Modules_FindQt3_cmake 
 patch-Modules_FindQt4_cmake 
 patch-Modules_FindQt_cmake 
 patch-Modules_FindRuby_cmake 
 patch-Modules_FindSDL_cmake 
 patch-Modules_FindTCL_cmake 
 patch-Modules_FindTclStub_cmake 
 patch-Modules_FindTclsh_cmake 
 patch-Modules_FindWish_cmake 
 patch-Modules_FindXMLRPC_cmake 
 patch-Modules_Qt4ConfigDependentSettings_cmake 
 patch-Source_CMakeLists_txt 
 patch-Source_cmComputeLinkInformation_cxx 
 patch-Source_cmGeneratorTarget_cxx 
 patch-Source_cmTarget_cxx 
 patch-Source_cmake_cxx 
 patch-Source_kwsys_SystemTools_cxx 
 patch-Tests_CMakeTests_ModuleNoticesTest_cmake_in 
 patch-Tests_RunCMake_CMakeLists_txt 
 patch-bootstrap 
devel/cmake/pkg: PLIST 
Added files:
devel/cmake/patches: patch-Source_cmMessenger_cxx 

Log message:
Update to CMake-3.7.0

Thanks to naddy@ who ran a bulk build with it

CVS: cvs.openbsd.org: ports

2016-11-28 Thread David Coppa 
CVSROOT:/cvs
Module name:ports
Changes by: dco...@cvs.openbsd.org  2016/11/28 01:43:50

Modified files:
x11/kde4/qyoto/patches: patch-cmake_CMakeCSharpInformation_cmake 

Log message:
Unbreak building with upcoming CMake-3.7.0 (policy CMP0054)