CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: rsadow...@cvs.openbsd.org 2021/09/06 23:58:12 Modified files: multimedia/mkvtoolnix: Makefile distinfo multimedia/mkvtoolnix/patches: patch-Rakefile patch-ac_qt5_m4 Log message: Update mkvtoolnix to 60.0.0 With mkvtoolnix 59.0.0 the Qt library is required for building all applications. This means we need Qt5Core for no_x11. https://www.bunkus.org/blog/2021/07/mkvtoolnix-v59-0-0-released/ Feedack and test by Stefan Hagen
Re: www/unit
Hi,
here's the latest revision of www/unit/Makefile, it's
buildable, creates four packages:
1) unit
2) unit-perl
3) unit-python
4) unit-ruby
The unit-php module is postpone, it needs to have php
shared library enabled in the parent packages.
Could you help me with review.
Thank you.
--
Sergey Osokin
# $OpenBSD$
COMMENT-main= Dynamic web application server
COMMENT-perl= NGINX Unit perl module
COMMENT-python= NGINX Unit python module
COMMENT-ruby= NGINX Unit ruby module
VERSION=1.25.0
DISTNAME= unit-${VERSION}
PKGNAME-main= ${DISTNAME}
PKGNAME-perl= unit-perl-${VERSION}
PKGNAME-python= unit-python-${VERSION}
PKGNAME-ruby= unit-ruby-${VERSION}
CATEGORIES= www
HOMEPAGE= https://unit.nginx.org
MAINTAINER= Sergey A. Osokin
PERMIT_PACKAGE= Yes
MULTI_PACKAGES= -main ${MODULE_PACKAGES}
MODULE_PACKAGES=-perl -python -ruby
.include
FLAVOR ?=
WANTLIB-main= c crypto pcre2-8 ssl z
WANTLIB-perl= c m perl
WANTLIB-python= ${MODPY_WANTLIB}
WANTLIB-ruby= ${MODRUBY_WANTLIB}
MASTER_SITES= https://unit.nginx.org/download/
DISTFILES= ${DISTNAME}${EXTRACT_SUFX}
COMPILER= base-clang ports-gcc
COMPILER_LANGS= c
LIB_DEPENDS-main= devel/pcre2
RUN_DEPENDS-perl= www/unit,-main
LIB_DEPENDS-python= ${MODPY_LIB_DEPENDS}
RUN_DEPENDS-python= www/unit,-main ${MODPY_LIB_DEPENDS}
LIB_DEPENDS-ruby= ${MODRUBY_LIB_DEPENDS}
RUN_DEPENDS-ruby = www/unit,-main ${MODRUBY_LIB_DEPENDS}
CONFIGURE_STYLE=simple
UNIT_DIR= /var/unit
SUBST_VARS= UNIT_DIR
CFLAGS+=-I ${LOCALBASE}/include
LDFLAGS+= -L ${LOCALBASE}/lib
CONFIGURE_ENV+= LDFLAGS="${LDFLAGS}"
CONFIGURE_ARGS= --prefix=${UNIT_DIR} \
--bindir=${PREFIX}/bin \
--sbindir=${PREFIX}/sbin \
--libdir=${PREFIX}/lib \
--incdir=${PREFIX}/include \
--mandir=${PREFIX}/man \
--modules=${UNIT_DIR}/modules \
--log=logs/unit.log \
--pid=/var/run/unit.pid \
--openssl \
--user=unit \
--group=unit
.if ${BUILD_PACKAGES:M-python}
MODULES+= lang/python
CONFIGURE_ENV+= PYTHON=python${MODPY_VERSION} \
PYTHON_CONFIG=python${MODPY_VERSION}-config
.endif
.if ${BUILD_PACKAGES:M-ruby}
MODULES+= lang/ruby
CONFIGURE_ENV+= RUBY=ruby${MODRUBY_REV}
.endif
NO_TEST=Yes
post-configure:
.if ${BUILD_PACKAGES:M-perl}
cd ${WRKSRC} && ./configure perl
.endif
.if ${BUILD_PACKAGES:M-python}
cd ${WRKSRC} && ./configure python \
--config=python${MODPY_VERSION}-config
.endif
.if ${BUILD_PACKAGES:M-ruby}
cd ${WRKSRC} && ./configure ruby --ruby=${MODRUBY_FLAVOR} \
--module=ruby${MODRUBY_REV}
.endif
post-install:
${INSTALL_DATA_DIR} ${WRKINST}${UNIT_DIR}
${INSTALL_DATA_DIR} ${WRKINST}${UNIT_DIR}/modules
${INSTALL_DATA_DIR} ${WRKINST}${UNIT_DIR}/state
.include
Re: [s...@spacehopper.org: NEW: mail/dovecot-fts-flatcurve]
On 9/6/2021 4:53 PM, Stuart Henderson wrote: quick ping, would quite like to get this in so i don't forget about it and am then sad when i upgrade dovecot and it breaks ;) Looks Ok to me. - Forwarded message from Stuart Henderson - From: Stuart Henderson Date: Thu, 2 Sep 2021 22:56:30 +0100 To: ports Subject: NEW: mail/dovecot-fts-flatcurve Mail-Followup-To: ports This is a fairly new fts plugin for Dovecot, it's likely to be added to the main Dovecot distribution upstream sometime, but it seems useful already. It uses some Dovecot-internal features that means it doesn't have to do as much as the existing dovecot-fts-xapian, takes less memory in my tests, and is far simpler to setup than solr. ok to import this? Requires the just-committed addition of icu to the Dovecot build otherwise it fails at runtime. info: https://github.com/slusarz/dovecot-fts-flatcurve/ https://github.com/slusarz/dovecot-fts-flatcurve/issues/4 https://doc.dovecot.org/configuration_manual/fts/ This is a Dovecot FTS plugin to enable message indexing using the Xapian search engine library. It is intended to act as a simple interface to the Xapian storage/search query functionality. Unlike the other Xapian-based FTS plugin "dovecot-fts-xapian", flatcurve relies on Dovecot to do the necessary stemming, and supports match scoring and substring matches - it is RFC 3501 (IMAP4rev1) compliant. It does not support fuzzy searches. - End forwarded message -
Re: NEW: libgnt
On 9/6/2021 5:22 AM, Stuart Henderson wrote:
On 2021/09/06 11:18, Landry Breuil wrote:
Le Sun, Sep 05, 2021 at 08:58:35PM -0400, Brad Smith a écrit :
Attached is a port of libgnt.
Comment:
ncurses toolkit for text-mode graphical user interfaces
Description:
GNT is an ncurses toolkit for creating text-mode graphical user interfaces in a
fast and easy way. It is based on GLib and ncurses.
It was born out of the console-based UI, Finch, for the libpurple project, but
has now been split into its own independent repository.
and it is a requirement to update pidgin so it's a welcome addition to
the portstree !
just one strange thing: homepage states that it needs py3 to build, but
you enforce py2. I'm not certain we want to add more ports that depend
on py2, so can it be made to use py3 ?
Landry
There's an optional dep on python 2:
# Check for Python headers
python_dep = dependency('python2', required : false)
if not python_dep.found()
python_dep = dependency('python-2.7', required : false)
endif
gnt_config.set('USE_PYTHON', python_dep.found())
If that's not required by other ports, it might be better to neuter it, indeed.
Looking at what it does it looks like there is a hook to run Python
scripts in some
manner. I'm not sure what exactly it does. Documentation seems to be
very minimal.
I see FreeBSD and NetBSD do not have a Python dependency on their respective
packages. I pulled that out and removed the check to ensure a
reproducible build.
Re: NEW: libgnt
On Sun, Sep 05, 2021 at 08:58:35PM -0400, Brad Smith wrote: > Attached is a port of libgnt. > > Comment: > ncurses toolkit for text-mode graphical user interfaces > > Description: > GNT is an ncurses toolkit for creating text-mode graphical user interfaces in > a > fast and easy way. It is based on GLib and ncurses. > > It was born out of the console-based UI, Finch, for the libpurple project, but > has now been split into its own independent repository. Updated to remove the Python dependency. libgnt.tar.gz Description: application/tar-gz
Re: NEW: libgnt
On 9/6/2021 5:18 AM, Landry Breuil wrote: Le Sun, Sep 05, 2021 at 08:58:35PM -0400, Brad Smith a écrit : Attached is a port of libgnt. Comment: ncurses toolkit for text-mode graphical user interfaces Description: GNT is an ncurses toolkit for creating text-mode graphical user interfaces in a fast and easy way. It is based on GLib and ncurses. It was born out of the console-based UI, Finch, for the libpurple project, but has now been split into its own independent repository. and it is a requirement to update pidgin so it's a welcome addition to the portstree ! just one strange thing: homepage states that it needs py3 to build, but you enforce py2. I'm not certain we want to add more ports that depend on py2, so can it be made to use py3 ? Yes, that is coming. I was a little confused by that as well. It looks like the 2.x version supports Python 2. The main branch that will be a 3.x release to eventually go with Pidgin 3.x supports Python 3. To me the website should mention release dependencies not main branch dependencies that hasn't even had a release yet.
Re: Awesome package - missing dep
On Mon, Sep 06, 2021 at 09:53:59PM +0200, Stefan Hagen wrote: > Stefan Hagen wrote: > > Nik Reist wrote: > > > While trying to set awesome as my wm, .xsession-errors shows a > > > required namespace is not found. It is missing pango as a dependency > > > and installing it allows awesome to work properly. > > > > Thanks for reporting. Indeed awesome needs pango, but not directoy. > > > > Awesome is using lua-lgi, which provides gobject bindings. It can call > > gobject based libraries like gtk, webkit, gstreamer, pango... > > Lua-lgi itself only depends on devel/gobject-introspection. > > > > I think it doesn't make sense to add dependencies to lua-lgi, just > > because it can call those libraries. Therefore I propose to add pango > > as run-dependency to awesome. > > > > OK? > > Again, with revision bump. Committed. Thanks Nik for the report and Stefan for the fix
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: t...@cvs.openbsd.org2021/09/06 15:32:19 Modified files: x11/awesome: Makefile Log message: Add missing RDEP on pango. Reported by Nik Reist, diff from Stefan Hagen ok aja
www/unit
Hi,
hope you're doing well these days.
I'm working on creation a port for the NGINX Unit,
https://unit.nginx.org.
I've created Makefile for unitd, main binary, the daemon.
It's attached.
Next step is to create additional packages of Unit's modules: perl,
php, python, and ruby. So, I need your help on that.
On FreeBSD and NetBSD we have separate ports/packages for the languages.
To configure a module for Unit the configure script needs to be run
twice:
1. to configure unit daemon
2. to configure a unit module
Could you guide me about best practices in this case.
Should I create a separate port for every language?
Or it's possible to define all packages in one big Makefile?
Thank you.
--
Sergey Osokin
# $OpenBSD$
COMMENT=Dynamic web application server
VERSION=1.25.0
DISTNAME= unit-${VERSION}
PKGNAME=${DISTNAME}
CATEGORIES= www
HOMEPAGE= https://unit.nginx.org
MAINTAINER= Sergey A. Osokin
PERMIT_PACKAGE= Yes
WANTLIB=c crypto pcre2-8 ssl z
MASTER_SITES= https://unit.nginx.org/download/
DISTFILES= ${DISTNAME}${EXTRACT_SUFX}
COMPILER= base-clang ports-gcc
COMPILER_LANGS= c
LIB_DEPENDS=devel/pcre2
CONFIGURE_STYLE = simple
UNIT_DIR= /var/unit
SUBST_VARS= UNIT_DIR
CFLAGS+=-I ${LOCALBASE}/include
LDFLAGS+= -L ${LOCALBASE}/lib
CONFIGURE_ENV+= LDFLAGS="${LDFLAGS}"
CONFIGURE_ARGS =--prefix=${UNIT_DIR} \
--bindir=${PREFIX}/bin \
--sbindir=${PREFIX}/sbin \
--libdir=${PREFIX}/lib \
--incdir=${PREFIX}/include \
--mandir=${PREFIX}/man \
--modules=${UNIT_DIR}/modules \
--log=logs/unit.log \
--pid=/var/run/unit.pid \
--openssl \
--user=unit \
--group=unit
NO_TEST=Yes
post-install:
${INSTALL_DATA_DIR} ${WRKINST}${UNIT_DIR}
${INSTALL_DATA_DIR} ${WRKINST}${UNIT_DIR}/modules
${INSTALL_DATA_DIR} ${WRKINST}${UNIT_DIR}/state
.include
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: ajacou...@cvs.openbsd.org 2021/09/06 15:18:21 Modified files: graphics/tesseract/tesseract: Makefile Log message: Add missing LDEP. Triggered by a report from sthen@ about configure picking up curl.
Re: Awesome package - missing dep
On Mon, Sep 06, 2021 at 09:53:59PM +0200, Stefan Hagen wrote:
> Stefan Hagen wrote:
> > Nik Reist wrote:
> > > While trying to set awesome as my wm, .xsession-errors shows a
> > > required namespace is not found. It is missing pango as a dependency
> > > and installing it allows awesome to work properly.
> >
> > Thanks for reporting. Indeed awesome needs pango, but not directoy.
> >
> > Awesome is using lua-lgi, which provides gobject bindings. It can call
> > gobject based libraries like gtk, webkit, gstreamer, pango...
> > Lua-lgi itself only depends on devel/gobject-introspection.
> >
> > I think it doesn't make sense to add dependencies to lua-lgi, just
> > because it can call those libraries. Therefore I propose to add pango
> > as run-dependency to awesome.
That's the way to do it, ok aja
> >
> > OK?
>
> Again, with revision bump.
>
> Index: x11/awesome/Makefile
> ===
> RCS file: /cvs/ports/x11/awesome/Makefile,v
> retrieving revision 1.114
> diff -u -p -u -p -r1.114 Makefile
> --- x11/awesome/Makefile 9 Jun 2021 19:50:10 - 1.114
> +++ x11/awesome/Makefile 6 Sep 2021 19:51:53 -
> @@ -5,7 +5,7 @@ COMMENT= highly configurable framework
> V= 4.3
> DISTNAME=awesome-${V}
> CATEGORIES= x11
> -REVISION=2
> +REVISION=3
>
> HOMEPAGE=https://awesomewm.org/
>
> @@ -42,7 +42,8 @@ BUILD_DEPENDS= devel/pango \
>
> MODLUA_RUN_DEPENDS= devel/lua-lgi
>
> -RUN_DEPENDS= misc/rlwrap \
> +RUN_DEPENDS= devel/pango \
> + misc/rlwrap \
> shells/bash
>
> CONFIGURE_ARGS= -DCOMPRESS_MANPAGES=OFF \
>
--
Antoine
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: st...@cvs.openbsd.org 2021/09/06 14:56:50 Modified files: security/py-cryptography: Makefile distinfo security/py-cryptography/pkg: PLIST Added files: security/py-cryptography: crates.inc Log message: build the rust parts of py-cryptography. many thanks to semarie@ for help with this port and cargo.port.mk changes to accommodate this and tb@ for sparc64 tests.
[s...@spacehopper.org: NEW: mail/dovecot-fts-flatcurve]
quick ping, would quite like to get this in so i don't forget about it and am then sad when i upgrade dovecot and it breaks ;) - Forwarded message from Stuart Henderson - From: Stuart Henderson Date: Thu, 2 Sep 2021 22:56:30 +0100 To: ports Subject: NEW: mail/dovecot-fts-flatcurve Mail-Followup-To: ports This is a fairly new fts plugin for Dovecot, it's likely to be added to the main Dovecot distribution upstream sometime, but it seems useful already. It uses some Dovecot-internal features that means it doesn't have to do as much as the existing dovecot-fts-xapian, takes less memory in my tests, and is far simpler to setup than solr. ok to import this? Requires the just-committed addition of icu to the Dovecot build otherwise it fails at runtime. info: https://github.com/slusarz/dovecot-fts-flatcurve/ https://github.com/slusarz/dovecot-fts-flatcurve/issues/4 https://doc.dovecot.org/configuration_manual/fts/ This is a Dovecot FTS plugin to enable message indexing using the Xapian search engine library. It is intended to act as a simple interface to the Xapian storage/search query functionality. Unlike the other Xapian-based FTS plugin "dovecot-fts-xapian", flatcurve relies on Dovecot to do the necessary stemming, and supports match scoring and substring matches - it is RFC 3501 (IMAP4rev1) compliant. It does not support fuzzy searches. - End forwarded message - dovecot-fts-flatcurve.tgz Description: application/tar-gz
Re: NEW: libgnt
On 2021/09/06 16:09, Brad Smith wrote: > On Sun, Sep 05, 2021 at 08:58:35PM -0400, Brad Smith wrote: > > Attached is a port of libgnt. > > > > Comment: > > ncurses toolkit for text-mode graphical user interfaces > > > > Description: > > GNT is an ncurses toolkit for creating text-mode graphical user interfaces > > in a > > fast and easy way. It is based on GLib and ncurses. > > > > It was born out of the console-based UI, Finch, for the libpurple project, > > but > > has now been split into its own independent repository. > > Updated to remove the Python dependency. Thanks, this is OK sthen@
Re: Awesome package - missing dep
Stefan Hagen wrote:
> Nik Reist wrote:
> > While trying to set awesome as my wm, .xsession-errors shows a
> > required namespace is not found. It is missing pango as a dependency
> > and installing it allows awesome to work properly.
>
> Thanks for reporting. Indeed awesome needs pango, but not directoy.
>
> Awesome is using lua-lgi, which provides gobject bindings. It can call
> gobject based libraries like gtk, webkit, gstreamer, pango...
> Lua-lgi itself only depends on devel/gobject-introspection.
>
> I think it doesn't make sense to add dependencies to lua-lgi, just
> because it can call those libraries. Therefore I propose to add pango
> as run-dependency to awesome.
>
> OK?
Again, with revision bump.
Index: x11/awesome/Makefile
===
RCS file: /cvs/ports/x11/awesome/Makefile,v
retrieving revision 1.114
diff -u -p -u -p -r1.114 Makefile
--- x11/awesome/Makefile9 Jun 2021 19:50:10 - 1.114
+++ x11/awesome/Makefile6 Sep 2021 19:51:53 -
@@ -5,7 +5,7 @@ COMMENT=highly configurable framework
V= 4.3
DISTNAME= awesome-${V}
CATEGORIES=x11
-REVISION= 2
+REVISION= 3
HOMEPAGE= https://awesomewm.org/
@@ -42,7 +42,8 @@ BUILD_DEPENDS=devel/pango \
MODLUA_RUN_DEPENDS=devel/lua-lgi
-RUN_DEPENDS= misc/rlwrap \
+RUN_DEPENDS= devel/pango \
+ misc/rlwrap \
shells/bash
CONFIGURE_ARGS=-DCOMPRESS_MANPAGES=OFF \
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: abie...@cvs.openbsd.org 2021/09/06 13:46:48 Modified files: net/dendrite : Makefile distinfo modules.inc net/dendrite/pkg: PLIST Log message: Update dendrite to 0.5.0 - Only ship the bins needed to run. OK bentley@
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: abie...@cvs.openbsd.org 2021/09/06 13:45:06 Modified files: security/age : Makefile distinfo modules.inc Log message: Update age to 1.0.0 \o/ Release notes here: https://github.com/FiloSottile/age/releases/tag/v1.0.0
Re: Awesome package - missing dep
Nik Reist wrote: > While trying to set awesome as my wm, .xsession-errors shows a > required namespace is not found. It is missing pango as a dependency > and installing it allows awesome to work properly. Thanks for reporting. Indeed awesome needs pango, but not directoy. Awesome is using lua-lgi, which provides gobject bindings. It can call gobject based libraries like gtk, webkit, gstreamer, pango... Lua-lgi itself only depends on devel/gobject-introspection. I think it doesn't make sense to add dependencies to lua-lgi, just because it can call those libraries. Therefore I propose to add pango as run-dependency to awesome. OK? Index: x11/awesome/Makefile === RCS file: /cvs/ports/x11/awesome/Makefile,v retrieving revision 1.114 diff -u -p -u -p -r1.114 Makefile --- x11/awesome/Makefile9 Jun 2021 19:50:10 - 1.114 +++ x11/awesome/Makefile6 Sep 2021 19:04:23 - @@ -42,7 +42,8 @@ BUILD_DEPENDS=devel/pango \ MODLUA_RUN_DEPENDS=devel/lua-lgi -RUN_DEPENDS= misc/rlwrap \ +RUN_DEPENDS= devel/pango \ + misc/rlwrap \ shells/bash CONFIGURE_ARGS=-DCOMPRESS_MANPAGES=OFF \
Awesome package - missing dep
Greetings, While trying to set awesome as my wm, .xsession-errors shows a required namespace is not found. It is missing pango as a dependency and installing it allows awesome to work properly. Thanks, Nik Reist
Re: NEW: www/sfeed and www/sfeed_curses (RSS/Atom tool and reader)
Hello -- On 09/06/2021 02:19 PM, Hiltjo Posthuma wrote: On Fri, Sep 03, 2021 at 02:16:31PM +0200, Frederic Cambus wrote: On Sun, Aug 29, 2021 at 03:36:40PM +0200, Hiltjo Posthuma wrote: > > Other than that, both look good and following your readmes and sample > > sfeedrc gets everything working on my machine. Same here, works as advertised on my machine. > I've attached an updated tarball of the ports with your recommended changes. Looks good to me, OK fcambus@ to import. Hi, Sorry if I'm impatient, is it possible to import it now? :) It would be nice if it is packaged in the OpenBSD 7.0 release. Thanks, If someone wants to import with my ok too, go ahead. I won't be able to get around to this myself for the next several days. ~Brian
Re: NEW: www/sfeed and www/sfeed_curses (RSS/Atom tool and reader)
On Fri, Sep 03, 2021 at 02:16:31PM +0200, Frederic Cambus wrote: > On Sun, Aug 29, 2021 at 03:36:40PM +0200, Hiltjo Posthuma wrote: > > > > Other than that, both look good and following your readmes and sample > > > sfeedrc gets everything working on my machine. > > Same here, works as advertised on my machine. > > > I've attached an updated tarball of the ports with your recommended changes. > > Looks good to me, OK fcambus@ to import. Hi, Sorry if I'm impatient, is it possible to import it now? :) It would be nice if it is packaged in the OpenBSD 7.0 release. Thanks, -- Kind regards, Hiltjo
NEW: devel/re2
RE2 is a fast, safe, thread-friendly C++ library alternative to backtracking regular expression engines like those used in PCRE, Perl, and Python. https://github.com/google/re2 re2-20210901.tgz Description: application/tar-gz
Re: using login_oath not working
podolica schrieb am Mittwoch, 4. August 2021 um 09:29: > I've trying to use login_oath for 2fa with totp follwing this > > (https://dataswamp.org/~solene/2021-02-06-openbsd-2fa.html) > > tutorial and the pkg_readme file in > > /usr/local/share/doc/pkg-readmes/login_oath. > > First, I've copied the new login classes into the /etc/login.conf. > > Then, I've created a new user "testuser" and choose totppw as > > login class. Then, I've entered a password "xyz" and login into > > testuser using the root account and "su testuser". > > There, I've created a file using > > openssl rand -hex 20 > ~/.totp-key > > After a reboot, I've used the root account to generate the totp > > codes: > > $ oathtool --totp > > 123456 > > > and login in another terminal with username "testuser" and > > password "123456/xyz". > > But the login is failing. > > Maybe someone has an idea how I could get this to work? I tried again and still it is still not working. Any idea how to solve this issue?
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: es...@cvs.openbsd.org 2021/09/06 08:54:05 Modified files: graphics/gmic : Makefile distinfo graphics/gmic/patches: patch-src_gmic_cpp graphics/gmic-qt: Makefile Makefile.inc distinfo graphics/gmic-qt/krita: Makefile graphics/gmic-qt/pkg: DESCR Added files: graphics/gmic-qt/none: Makefile graphics/gmic-qt/none/pkg: PLIST Log message: update gmic/gmic-qt to 2.9.8. enable gmic-qt-none, a stand-alone version of gmic-qt that may be now used in batch-mode by software like blender thanks to Josh Grosse for useful comments
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: es...@cvs.openbsd.org 2021/09/06 08:49:25 ports/graphics/gmic-qt/none/pkg Update of /cvs/ports/graphics/gmic-qt/none/pkg In directory cvs.openbsd.org:/tmp/cvs-serv20685/pkg Log Message: Directory /cvs/ports/graphics/gmic-qt/none/pkg added to the repository
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: es...@cvs.openbsd.org 2021/09/06 08:49:07 ports/graphics/gmic-qt/none Update of /cvs/ports/graphics/gmic-qt/none In directory cvs.openbsd.org:/tmp/cvs-serv50320/none Log Message: Directory /cvs/ports/graphics/gmic-qt/none added to the repository
Re: [update] lang/node to 12.22.6
On 2021/09/05 20:16, Stuart Henderson wrote: > Build fails on i386: > > cc -o > /pobj/node-12.22.6/node-v12.22.6/out/Release/obj.target/zlib/deps/zlib/contrib/optimizations/inffast_chunk.o > ../deps/zlib/contrib/optimizations/ > inffast_chunk.c '-DV8_DEPRECATION_WARNINGS' > '-DV8_IMMINENT_DEPRECATION_WARNINGS' '-D__STDC_FORMAT_MACROS' > '-DOPENSSL_NO_PINSHARED' '-DOPENSSL_THREADS' '-DHAVE_HIDDEN' > '-DADLER32_SIMD_SSSE3' '-DINFLATE_CHUNK_SIMD_SSE2' '-DCRC32_S > IMD_SSE42_PCLMUL' -I../deps/zlib -pthread -Wall -Wextra > -Wno-unused-parameter -m32 -I/usr/local/include -I/usr/local/include -O3 > -fno-omit-frame-pointer -MMD -MF > /pobj/node-12.22.6/node-v12.22.6/out/Release/.deps//pobj/node-12.22.6/node-v12.22.6/out/Release/obj.target/zlib/deps/zlib/contrib/optimizations/inffast_chunk.o.d.raw > -O2 -pipe -c In file included from > ../deps/zlib/contrib/optimizations/inffast_chunk.c:10: I have repaired this. I guess with the dependency chains in npm they need all the help with gzip they can get ;) The update that was committed did not have patches in sync with the versions of files in the tarball, please run "make update-patches" for future ones.
Re: Have makesum target display hashes in common format
On Mon, Sep 06, 2021 at 02:01:08PM +0100, Stuart Henderson wrote:
> On 2021/09/06 14:48, Marc Espie wrote:
> > On Fri, Sep 03, 2021 at 01:34:57PM -0600, Aaron Bieber wrote:
> > >
> > > Kurt Mosiejczuk writes:
> > >
> > > > On Fri, Sep 03, 2021 at 12:04:42PM -0600, Aaron Bieber wrote:
> > > >
> > > >> How many is multiple? :D - the Go stuff is gonna get ugly(er?)!
> > > >
> > > >> net/go-ipfs for example, the output is useless because it's so big :D
> > > >
> > > > Does go actually publish base64 SHA256 hashes for their modules?
> > > > If not, it's not the size of the output that makes that useless.
> > >
> > > They do, but checking by hand is redundant as the checks are all built
> > > into the tooling.
> > >
> > > For example: https://sum.golang.org/lookup/suah.dev/ogvt@v1.1.0
> > >
> > > (the h1:blablabl stuff is an encoded sha256 hash)
> > >
> > > >
> > > >> Maybe it can be conditional.. somehow?
> > > >
> > > > Why? How often do you do a "make makesum"? I do it like once per port
> > > > update and verify my distfiles.
> > > >
> > >
> > > It seems like I do it quite often. Though maybe it just stands out
> > > because the Go lists are always sofaking big!
> > >
> > > > --Kurt
> >
> > I think it would be more sense to append it to a file, having the name
> > be /dev/null by default, for those few individuals who actually need this
> >
>
> I really would like it to not calculate at all if it's not needed, some
> of the distfiles are huge and sha256 and md5 is definitely non-trivial,
> especially over NFS
>
>
So, yeah, empty instead of redirect to /dev/null
I think it's nicer to >>${SOMEFILE}... people can always do /dev/tty if they
want to see it on screen, and that way it doesn't clutter the display after
the diff.
.e.g,
.if !empty(SOMEVAR)
md6 >>${SOMEVAR}; sha256 >>${SOMEVAR}
.endif
(or the corresponding exec form)
only thing missing would be a variable name.
LEGACY_CHECKSUMS ?
Re: [new] net/lurch - OMEMO support for libpurple
Stefan Hagen wrote: > Stuart Henderson wrote: >> On 2021/08/15 22:51, Stefan Hagen wrote: >>> Portcheck is happy, but make ports-lib-depends-check has issues: >>> >>> $ make port-lib-depends-check >>> >>> lurch-0.7.0-src(net/lurch): >>> Missing lib: jabber.0 (/usr/local/lib/purple-2/lurch.so) (NOT REACHABLE) >>> Extra: purple.7 >>> *** Error 1 in target 'port-lib-depends-check' (ignored) >> >> That needs lib/purple-2/jabber in WANTLIB and net/pidgin,-libpurple in >> LIB_DEPENDS, it does not want purple in WANTLIB. check-lib-depends will >> still warn because it doesn't handle subdirectories properly. > > Now that I know what to look for, I found the documentation in > library-specs(7). > >>> $ ldd /usr/local/lib/purple-2/lurch.so | grep jabber >>> [...] /usr/local/lib/purple-2/libjabber.so.0.0 >> >> Use objdump -p and look for NEEDED lines, ldd will show inter-library deps >> too which shouldn't be listed. > > Good to know. Thanks. > > Attached is the corrected port: > - updated WANTLIB > - added comment about portcheck + lib-depends-check errors > - fixed PKGNAME > - fixed some variables and tabs *ping* The errors thrown by lib-depends-check and portcheck are caused by the path in WANTLIB and can be ignored. lurch.tgz Description: application/tar-gz
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: st...@cvs.openbsd.org 2021/09/06 07:49:44 Modified files: lang/node/patches: patch-deps_zlib_zlib_gyp Log message: ...and another piece
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: st...@cvs.openbsd.org 2021/09/06 07:45:01 Added files: lang/node/patches: patch-deps_zlib_zlib_gyp Log message: same as previous but for the .gyp config
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: st...@cvs.openbsd.org 2021/09/06 07:32:26 Added files: lang/node/patches: patch-deps_zlib_BUILD_gn Log message: node has a patch to add SIMD etc optimizations to zlib on i386/amd64; disable on i386, it breaks build. it uses cpuid to detect features so _should_ work on amd64.
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: st...@cvs.openbsd.org 2021/09/06 07:27:41 Modified files: lang/node/patches: patch-Makefile patch-common_gypi patch-deps_npm_lib_npm_js patch-deps_npm_node_modules_node-gyp_lib_install_js patch-deps_openssl_config_bn_conf_asm_h patch-deps_openssl_config_bn_conf_no-asm_h patch-deps_openssl_config_dso_conf_asm_h patch-deps_openssl_config_dso_conf_no-asm_h patch-deps_openssl_config_opensslconf_asm_h patch-deps_openssl_config_opensslconf_no-asm_h patch-deps_openssl_openssl_asm_gypi patch-deps_v8_src_base_cpu_cc patch-lib_module_js Log message: regenerate patches, no pkg change
Re: Have makesum target display hashes in common format
On 2021/09/06 14:48, Marc Espie wrote: > On Fri, Sep 03, 2021 at 01:34:57PM -0600, Aaron Bieber wrote: > > > > Kurt Mosiejczuk writes: > > > > > On Fri, Sep 03, 2021 at 12:04:42PM -0600, Aaron Bieber wrote: > > > > > >> How many is multiple? :D - the Go stuff is gonna get ugly(er?)! > > > > > >> net/go-ipfs for example, the output is useless because it's so big :D > > > > > > Does go actually publish base64 SHA256 hashes for their modules? > > > If not, it's not the size of the output that makes that useless. > > > > They do, but checking by hand is redundant as the checks are all built > > into the tooling. > > > > For example: https://sum.golang.org/lookup/suah.dev/ogvt@v1.1.0 > > > > (the h1:blablabl stuff is an encoded sha256 hash) > > > > > > > >> Maybe it can be conditional.. somehow? > > > > > > Why? How often do you do a "make makesum"? I do it like once per port > > > update and verify my distfiles. > > > > > > > It seems like I do it quite often. Though maybe it just stands out > > because the Go lists are always sofaking big! > > > > > --Kurt > > I think it would be more sense to append it to a file, having the name > be /dev/null by default, for those few individuals who actually need this > I really would like it to not calculate at all if it's not needed, some of the distfiles are huge and sha256 and md5 is definitely non-trivial, especially over NFS
Re: Have makesum target display hashes in common format
On Fri, Sep 03, 2021 at 01:34:57PM -0600, Aaron Bieber wrote: > > Kurt Mosiejczuk writes: > > > On Fri, Sep 03, 2021 at 12:04:42PM -0600, Aaron Bieber wrote: > > > >> How many is multiple? :D - the Go stuff is gonna get ugly(er?)! > > > >> net/go-ipfs for example, the output is useless because it's so big :D > > > > Does go actually publish base64 SHA256 hashes for their modules? > > If not, it's not the size of the output that makes that useless. > > They do, but checking by hand is redundant as the checks are all built > into the tooling. > > For example: https://sum.golang.org/lookup/suah.dev/ogvt@v1.1.0 > > (the h1:blablabl stuff is an encoded sha256 hash) > > > > >> Maybe it can be conditional.. somehow? > > > > Why? How often do you do a "make makesum"? I do it like once per port > > update and verify my distfiles. > > > > It seems like I do it quite often. Though maybe it just stands out > because the Go lists are always sofaking big! > > > --Kurt I think it would be more sense to append it to a file, having the name be /dev/null by default, for those few individuals who actually need this
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: ajacou...@cvs.openbsd.org 2021/09/06 05:04:03 Added files: devel/glib2/patches: patch-glib_tests_test-printf_c Log message: Unbreak build: don't use printf %n. from jca@
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: rob...@cvs.openbsd.org 2021/09/06 05:00:06 Modified files: security/openpam/patches: patch-openpam_lib_libpam_openpam_dynamic_c Log message: unbreak previous patch
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: rob...@cvs.openbsd.org 2021/09/06 04:59:24 Modified files: security/openpam: Makefile Added files: security/openpam/patches: patch-openpam_lib_libpam_openpam_dynamic_c Log message: do not use %n in snprintf()
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: mar...@cvs.openbsd.org 2021/09/06 04:44:21 Modified files: net/py-tinytuya: Makefile distinfo Log message: Update py-tinytuya to 1.2.8.
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: bent...@cvs.openbsd.org 2021/09/06 03:42:05 Modified files: multimedia : Makefile Log message: +gaupol
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: bent...@cvs.openbsd.org 2021/09/06 03:40:48 Log message: Import gaupol-1.9. This package provides Gaupol, an editor for text-based subtitle files. It supports multiple subtitle file formats and provides means of creating, editing and translating subtitles and timing subtitles to match video. ok sthen@ From Anindya Mukherjee; thanks! Status: Vendor Tag: bentley Release Tags: bentley_20210906 N ports/multimedia/gaupol/Makefile N ports/multimedia/gaupol/distinfo N ports/multimedia/gaupol/pkg/DESCR N ports/multimedia/gaupol/pkg/PLIST No conflicts created by this import
Re: [Update] devel/avr/gcc: 5.4.0 -> 8.4.0
On Thu, Aug 05, 2021 at 07:54:35AM +0200, Martin Reindl wrote: > On Thu, Dec 03, 2020 at 02:44:22PM +0100, Manuel Giraud wrote: > > Hi, > > > > Here is an update of avr/gcc. I was able to build current qmk_firmware > > (https://qmk.fm/) for my keyboard (with 5.4.0, it was not building since > > august 2019). > > Taking this diff, here is an update to 8.5.0. Tested with ethersex.de for a > 644PU which does not work with the current avr-gcc. > > Also compile tested: > devel/arduino > devel/avr/gdb > devel/avr/libc (and ethersex runs this) > devel/simulavr > > Comments? OKs? Friendly ping to get this commited before 7.0? -m
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: lan...@cvs.openbsd.org 2021/09/06 03:30:56 Modified files: x11/xfce4/libxfce4ui: Makefile distinfo Log message: x11/xfce4/libxfce4ui: update to 4.16.1
Re: NEW: libgnt
On 2021/09/06 11:18, Landry Breuil wrote:
> Le Sun, Sep 05, 2021 at 08:58:35PM -0400, Brad Smith a écrit :
> > Attached is a port of libgnt.
> >
> > Comment:
> > ncurses toolkit for text-mode graphical user interfaces
> >
> > Description:
> > GNT is an ncurses toolkit for creating text-mode graphical user interfaces
> > in a
> > fast and easy way. It is based on GLib and ncurses.
> >
> > It was born out of the console-based UI, Finch, for the libpurple project,
> > but
> > has now been split into its own independent repository.
>
> and it is a requirement to update pidgin so it's a welcome addition to
> the portstree !
>
> just one strange thing: homepage states that it needs py3 to build, but
> you enforce py2. I'm not certain we want to add more ports that depend
> on py2, so can it be made to use py3 ?
>
> Landry
>
There's an optional dep on python 2:
# Check for Python headers
python_dep = dependency('python2', required : false)
if not python_dep.found()
python_dep = dependency('python-2.7', required : false)
endif
gnt_config.set('USE_PYTHON', python_dep.found())
If that's not required by other ports, it might be better to neuter it, indeed.
Re: NEW: libgnt
Le Sun, Sep 05, 2021 at 08:58:35PM -0400, Brad Smith a écrit : > Attached is a port of libgnt. > > Comment: > ncurses toolkit for text-mode graphical user interfaces > > Description: > GNT is an ncurses toolkit for creating text-mode graphical user interfaces in > a > fast and easy way. It is based on GLib and ncurses. > > It was born out of the console-based UI, Finch, for the libpurple project, but > has now been split into its own independent repository. and it is a requirement to update pidgin so it's a welcome addition to the portstree ! just one strange thing: homepage states that it needs py3 to build, but you enforce py2. I'm not certain we want to add more ports that depend on py2, so can it be made to use py3 ? Landry
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: lan...@cvs.openbsd.org 2021/09/06 02:58:08 Modified files: security/nss : Makefile distinfo Log message: security/nss: update to 3.70, will be required for gecko 93. see https://hg.mozilla.org/projects/nss/file/tip/doc/rst/releases/nss_3_70.rst
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: st...@cvs.openbsd.org 2021/09/06 02:30:28 Modified files: net/haproxy: Tag: OPENBSD_6_9 Makefile distinfo Log message: update haproxy in -stable to 2.3.13, fixing HTTP/2 parser vulnerabilities https://www.haproxy.com/blog/august-2021-haproxy-2-0-http-2-vulnerabilities-fixed/ from brad
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: lan...@cvs.openbsd.org 2021/09/06 02:20:20 Modified files: www/seamonkey : Makefile Log message: www/seamonkey: fix build on i386 make sure we dont try to detect libatomic from gcc-libs, otherwise the build fails later on as found out by sthen@.
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: lan...@cvs.openbsd.org 2021/09/06 01:32:47 Modified files: geo/postgis: Makefile distinfo geo/postgis/pkg: PLIST Log message: geo/postgis: update to 3.1.4
[maintainer-update] mail/mu 1.6.5 -> 1.6.6
Hello, Yet another bugfix release for mail/mu. @Omar: The mu-init(1) fix has been committed upstream and is part of the update. Port changes: * none, except version bump Software changes: * Bugfixes: https://github.com/djcb/mu/compare/1.6.5...1.6.6 portcheck, lib-depends-check, make test: all happy on amd64. Best regards, Stefan Index: mail/mu/Makefile === RCS file: /cvs/ports/mail/mu/Makefile,v retrieving revision 1.28 diff -u -p -u -p -r1.28 Makefile --- mail/mu/Makefile28 Aug 2021 23:05:40 - 1.28 +++ mail/mu/Makefile6 Sep 2021 06:33:05 - @@ -2,7 +2,7 @@ COMMENT= maildir indexer and searcher with emacs frontend -V= 1.6.5 +V= 1.6.6 GUILE_V= 2.2 DISTNAME= mu-$V Index: mail/mu/distinfo === RCS file: /cvs/ports/mail/mu/distinfo,v retrieving revision 1.15 diff -u -p -u -p -r1.15 distinfo --- mail/mu/distinfo28 Aug 2021 23:05:40 - 1.15 +++ mail/mu/distinfo6 Sep 2021 06:33:05 - @@ -1,2 +1,2 @@ -SHA256 (mu-1.6.5.tar.xz) = UECqFayzcikBGUaT3Gf8VLlVE0e5Ibm2kLFKGn7IOEc= -SIZE (mu-1.6.5.tar.xz) = 867240 +SHA256 (mu-1.6.6.tar.xz) = Qs09IPKmcP5XvtzeBJtrKwMP5RxQf9/f62qoZfH0Q7U= +SIZE (mu-1.6.6.tar.xz) = 866724
