UPDATE: math/octave
This brings octave to the latest version. Tested ok on amd64.
Would be nice if someone could build/test on other archs.
Index: Makefile
===
RCS file: /cvs/ports/math/octave/Makefile,v
retrieving revision 1.108
diff -u -p -u -r1.108 Makefile
--- Makefile8 Mar 2019 20:00:47 - 1.108
+++ Makefile25 Apr 2019 05:43:13 -
@@ -2,14 +2,13 @@
COMMENT = high-level language for numerical computations
-VERSION = 4.4.1
+VERSION = 5.1.0
DISTNAME = octave-${VERSION}
-REVISION = 1
CATEGORIES = math
-SHARED_LIBS += octave 11.0
-SHARED_LIBS += octgui 5.0
-SHARED_LIBS += octinterp 10.0
+SHARED_LIBS += octave 12.0
+SHARED_LIBS += octgui 6.0
+SHARED_LIBS += octinterp 11.0
HOMEPAGE = https://www.gnu.org/software/octave/
@@ -21,7 +20,7 @@ PERMIT_PACKAGE_CDROM =Yes
WANTLIB += ${MODFORTRAN_WANTLIB}
WANTLIB += FLAC GL GLU GraphicsMagick GraphicsMagick++ ICE
WANTLIB += Qt5Gui Qt5Network Qt5OpenGL Qt5PrintSupport Qt5Widgets Qt5Core
-WANTLIB += Qt5Help Qt5Sql qscintilla2_qt5
+WANTLIB += Qt5Help Qt5Xml Qt5Sql qscintilla2_qt5
WANTLIB += SM X11 X11-xcb Xdamage
WANTLIB += Xext Xfixes Xi Xmu Xrandr Xrender Xt Xxf86vm amd
WANTLIB += arpack blas btf bz2 c camd ccolamd cholmod colamd crypto curl
@@ -29,7 +28,7 @@ WANTLIB += curses cxsparse drm ereadline
WANTLIB += execinfo glib-2.0 graphite2 gthread-2.0 harfbuzz icudata
WANTLIB += fftw3f_threads fontconfig freetype gl2ps glapi glpk glut
WANTLIB += gmp hdf5 iconv icui18n icuuc intl klu lapack lcms2 ltdl lzma
-WANTLIB += m nghttp2 ogg pcre pcre2-16 png png16 portaudio pthread qhull
+WANTLIB += m nghttp2 ogg pcre pcre2-16 png16 portaudio pthread qhull
WANTLIB += qrupdate sndfile sndio ssl ${COMPILER_LIBCXX} suitesparseconfig
WANTLIB += umfpack util vorbis vorbisenc xcb xcb-dri2 xcb-dri3 xcb-glx
WANTLIB += xcb-present xcb-sync xcb-xfixes xshmfence z
Index: distinfo
===
RCS file: /cvs/ports/math/octave/distinfo,v
retrieving revision 1.21
diff -u -p -u -r1.21 distinfo
--- distinfo13 Jan 2019 22:43:59 - 1.21
+++ distinfo25 Apr 2019 05:43:13 -
@@ -1,2 +1,2 @@
-SHA256 (octave-4.4.1.tar.gz) = CfvQ8hL07yHlPx2cQc8wzj1/lFD7RJEWAeIe1kxnrpc=
-SIZE (octave-4.4.1.tar.gz) = 28626462
+SHA256 (octave-5.1.0.tar.gz) = 42sRJMrCfHyqUcxX3kCMMWdtXwCWNJtNULV7/hvNdJU=
+SIZE (octave-5.1.0.tar.gz) = 32034788
Index: patches/patch-configure
===
RCS file: /cvs/ports/math/octave/patches/patch-configure,v
retrieving revision 1.13
diff -u -p -u -r1.13 patch-configure
--- patches/patch-configure 13 Jan 2019 22:43:59 - 1.13
+++ patches/patch-configure 25 Apr 2019 05:43:13 -
@@ -5,7 +5,7 @@ Use GNU readline as our base readline la
Index: configure
--- configure.orig
+++ configure
-@@ -66327,13 +66327,13 @@ fi
+@@ -67197,13 +67197,13 @@ fi
ac_octave_save_LIBS="$LIBS"
LIBS="$TERM_LIBS"
@@ -22,7 +22,7 @@ Index: configure
cat confdefs.h - <<_ACEOF >conftest.$ac_ext
/* end confdefs.h. */
-@@ -66372,7 +66372,7 @@ fi
+@@ -67242,7 +67242,7 @@ fi
{ $as_echo "$as_me:${as_lineno-$LINENO}: result:
$ac_cv_lib_readline_rl_set_keyboard_input_timeout" >&5
$as_echo "$ac_cv_lib_readline_rl_set_keyboard_input_timeout" >&6; }
if test "x$ac_cv_lib_readline_rl_set_keyboard_input_timeout" = xyes; then :
Index: pkg/PLIST
===
RCS file: /cvs/ports/math/octave/pkg/PLIST,v
retrieving revision 1.25
diff -u -p -u -r1.25 PLIST
--- pkg/PLIST 13 Jan 2019 22:43:59 - 1.25
+++ pkg/PLIST 25 Apr 2019 05:43:14 -
@@ -135,7 +135,6 @@ include/octave-${VERSION}/octave/graphic
include/octave-${VERSION}/octave/graphics-props.cc
include/octave-${VERSION}/octave/graphics-toolkit.h
include/octave-${VERSION}/octave/graphics.h
-include/octave-${VERSION}/octave/gripes.h
include/octave-${VERSION}/octave/gsvd.h
include/octave-${VERSION}/octave/gtk-manager.h
include/octave-${VERSION}/octave/help.h
@@ -159,11 +158,9 @@ include/octave-${VERSION}/octave/libocti
include/octave-${VERSION}/octave/lo-amos-proto.h
include/octave-${VERSION}/octave/lo-arpack-proto.h
include/octave-${VERSION}/octave/lo-array-errwarn.h
-include/octave-${VERSION}/octave/lo-array-gripes.h
include/octave-${VERSION}/octave/lo-blas-proto.h
include/octave-${VERSION}/octave/lo-cutils.h
include/octave-${VERSION}/octave/lo-error.h
-include/octave-${VERSION}/octave/lo-fftpack-proto.h
include/octave-${VERSION}/octave/lo-hash.h
include/octave-${VERSION}/octave/lo-ieee.h
include/octave-${VERSION}/octave/lo-lapack-proto.h
@@ -636,6 +633,7 @@ include/octave-${VERSION}/octave/pr-outp
include/octave-${VERSION}/octave/procstream.h
include/octave-${VERSION}/octave/profiler.h
include/octave-${VERSION}/octave/pt-all.h
Re: [NEW] graphics/xdot-1.0 [It was: [NEW] math/xdot-1.0]
Hello,
gentle reminder. This is already ok sthen@
Anyone willing to import?
On 20/04/2019 12:31, Stuart Henderson wrote:
On 2019/04/20 12:25, Alessandro DE LAURENZIS wrote:
Hello Stuart,
On 18/04/2019 18:07, Stuart Henderson wrote:
[...]
- the pypi tarball doesn't include the tests, which I instead enabled
using the one from github; is this acceptable?
I usually prefer pypi over github, because pypi uses uploaded files
rather than autogenerated ones which are subject to change. It depends
how useful the tests are really.
So I reverted the port's Makefile to use pypi (I think the test suite is
useful mainly for development).
: RUN_DEPENDS = math/graphviz \
: devel/py-gobject3 \
That should be devel/py-gobject3${MODPY_FLAVOR}
Fixed.
Updated tarball attached.
Thanks - this one is OK sthen@ if someone would like to import.
--
Alessandro DE LAURENZIS
[mailto:jus...@atlantide.t28.net]
Web: http://www.atlantide.t28.net
LinkedIn: https://www.linkedin.com/in/delaurenzis/
UPDATE: telephony/coturn
Hi,
this updates telephony/coturn to 4.5.1.1. The update includes security
fixes and configuration changes.
Change log: https://coturn.net/turnserver/v4.5.1.1/ChangeLog-1.txt
Works fine for me on amd64.
OK?
Index: Makefile
===
RCS file: /cvs/ports/telephony/coturn/Makefile,v
retrieving revision 1.2
diff -u -p -r1.2 Makefile
--- Makefile11 Jan 2018 19:27:11 - 1.2
+++ Makefile25 Apr 2019 04:49:52 -
@@ -2,9 +2,8 @@
COMMENT = coturn STUN/TURN server
-V =4.5.0.7
+V =4.5.1.1
DISTNAME = turnserver-${V}
-REVISION = 0
CATEGORIES = telephony
Index: distinfo
===
RCS file: /cvs/ports/telephony/coturn/distinfo,v
retrieving revision 1.1.1.1
diff -u -p -r1.1.1.1 distinfo
--- distinfo23 Dec 2017 17:55:38 - 1.1.1.1
+++ distinfo25 Apr 2019 04:49:52 -
@@ -1,2 +1,2 @@
-SHA256 (turnserver-4.5.0.7.tar.gz) =
hiSMVBoRhOs4jFTUF4z/vxbvU1BPu2AQbldRlPB4siE=
-SIZE (turnserver-4.5.0.7.tar.gz) = 400577
+SHA256 (turnserver-4.5.1.1.tar.gz) =
4CDOkOoDASE0UdNwmRhf8l2T+X+g8rSL8hspRvw2lqQ=
+SIZE (turnserver-4.5.1.1.tar.gz) = 423160
Index: pkg/PLIST
===
RCS file: /cvs/ports/telephony/coturn/pkg/PLIST,v
retrieving revision 1.1.1.1
diff -u -p -r1.1.1.1 PLIST
--- pkg/PLIST 23 Dec 2017 17:55:38 - 1.1.1.1
+++ pkg/PLIST 25 Apr 2019 04:49:52 -
@@ -92,6 +92,7 @@ share/examples/turnserver/scripts/mobile
share/examples/turnserver/scripts/mobile/mobile_tls_client_c2c_tcp_relay.sh
share/examples/turnserver/scripts/mobile/mobile_udp_client.sh
share/examples/turnserver/scripts/oauth.sh
+share/examples/turnserver/scripts/pack.sh
share/examples/turnserver/scripts/peer.sh
share/examples/turnserver/scripts/readme.txt
share/examples/turnserver/scripts/restapi/
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: bcal...@cvs.openbsd.org 2019/04/24 20:18:58 Modified files: graphics/feh : Makefile Log message: Take MAINTAINER of feh. ok with sthen@
patch lang/guile force frame pointer
Hello,
Recently we found that lang/guile does not work with frame pointer
omission. The patch below adds the compiler option to always include the
frame pointer.
I found when adding this option that the guile runtime would stack
overflow, which is remedied by increasing the default stack size.
With this, guile will build and run on platforms which have
-fomit-frame-pointer on by default.
Any objections? We don't have to apply this now, but if we want to
enable frame pointer elimination again then it would be needed.
Index: Makefile
===
RCS file: /cvs/ports/lang/guile/Makefile,v
retrieving revision 1.47
diff -u -p -u -r1.47 Makefile
--- Makefile2 Nov 2018 18:48:39 - 1.47
+++ Makefile25 Apr 2019 01:00:30 -
@@ -36,6 +36,7 @@ USE_GMAKE=Yes
# warning: format '%ji' expects type 'intmax_t', but argument 4 has type
'scm_t_intmax'
CONFIGURE_ARGS+= --disable-error-on-warning
CONFIGURE_ENV= CPPFLAGS="-I${LOCALBASE}/include" \
+ CFLAGS="-fno-omit-frame-pointer" \
LDFLAGS="-L${LOCALBASE}/lib" \
PTHREAD_CFLAGS=-pthread \
PTHREAD_LIBS=-pthread
Index: patches/patch-libguile_eval_c
===
RCS file: patches/patch-libguile_eval_c
diff -N patches/patch-libguile_eval_c
--- /dev/null 1 Jan 1970 00:00:00 -
+++ patches/patch-libguile_eval_c 25 Apr 2019 01:00:30 -
@@ -0,0 +1,20 @@
+--- libguile/eval.c.orig Mon Dec 13 12:24:39 2010
libguile/eval.cSun Apr 14 10:29:56 2019
+@@ -3078,7 +3078,7 @@
+ long scm_eval_stack;
+
+ scm_t_option scm_eval_opts[] = {
+- { SCM_OPTION_INTEGER, "stack", 22000, "Size of thread stacks (in machine
words)." }
++ { SCM_OPTION_INTEGER, "stack", 44000, "Size of thread stacks (in machine
words)." }
+ };
+
+ scm_t_option scm_debug_opts[] = {
+@@ -3099,7 +3099,7 @@
+ { SCM_OPTION_INTEGER, "depth", 20, "Maximal length of printed backtrace." },
+ { SCM_OPTION_BOOLEAN, "backtrace", 0, "Show backtrace on error." },
+ { SCM_OPTION_BOOLEAN, "debug", 0, "Use the debugging evaluator." },
+- { SCM_OPTION_INTEGER, "stack", 2, "Stack size limit (measured in words;
0 = no check)." },
++ { SCM_OPTION_INTEGER, "stack", 44000, "Stack size limit (measured in words;
0 = no check)." },
+ { SCM_OPTION_SCM, "show-file-name", (unsigned long)SCM_BOOL_T, "Show file
names and line numbers in backtraces when not `#f'. A value of `base' displays
only base names, while `#t' displays full names."},
+ { SCM_OPTION_BOOLEAN, "warn-deprecated", 0, "Warn when deprecated features
are used." }
+ };
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: jer...@cvs.openbsd.org 2019/04/24 17:33:49 Modified files: audio : Makefile databases : Makefile devel : Makefile graphics : Makefile net: Makefile security : Makefile sysutils : Makefile textproc : Makefile www: Makefile x11: Makefile Log message: Stop building ports for ruby 2.4 libraries by default Per our current ruby support policy, as ruby 2.4's normal maintenance phase has ended.
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: k...@cvs.openbsd.org2019/04/24 17:01:24 Modified files: devel : Makefile Log message: Adding py-objdir
Re: gcc8 available on which archs?
In general, everything that's in ONLY_FOR_ARCHS and not BROKEN is in a working state. On Tue, 23 Apr 2019 17:14:09 +0200, Christian Weisgerber wrote: > With everybody eager to switch ports-gcc to gcc8, I'm still a bit > hazy about which architectures gcc8 is actually known to work on. > To the best of my knowledge: > > should be fine > aarch64 already switched > amd64 successful bulk build > powerpc successful partial build > sparc64 successful partial build > > optimistic > hppaexistence of adastrap 8.3.0 suggests working compiler works. > mips64 existence of adastrap 8.2.0 suggests working compiler works. > mips64el== mips64? should very likely work, but I do not have a machine. > don't know > arm ? works. > alpha ? (don't care) Completely untested, may change soon(ish). > > known bad > i386marked BROKEN > > -- > Christian "naddy" Weisgerber na...@mips.inka.de >
Re: gcc8 available on which archs?
On Tue, 23 Apr 2019 16:52:57 +0100, Stuart Henderson wrote: > On 2019/04/23 17:14, Christian Weisgerber wrote: > > With everybody eager to switch ports-gcc to gcc8, I'm still a bit > > hazy about which architectures gcc8 is actually known to work on. > > To the best of my knowledge: > > > > should be fine > > aarch64 already switched > > amd64 successful bulk build > > powerpc successful partial build > > sparc64 successful partial build > > > > optimistic > > hppaexistence of adastrap 8.3.0 suggests working compiler > > mips64 existence of adastrap 8.2.0 suggests working compiler > > mips64el== mips64? > > > > don't know > > arm ? > > alpha ? (don't care) > > > > known bad > > i386marked BROKEN > > > > -- > > Christian "naddy" Weisgerber na...@mips.inka.de > > From the last snaps: > > -rw-rw-r-- 1 openbsd openbsd 23870613 Apr 21 09:53 aarch64/gcc-8.3.0.tgz > -rw-rw-r-- 1 openbsd openbsd 28096948 Apr 21 18:07 amd64/gcc-8.3.0.tgz > -rw-rw-r-- 1 openbsd openbsd 27137929 Feb 25 05:54 arm/gcc-8.2.0p0.tgz > -rw-rw-r-- 1 openbsd openbsd 25484157 Apr 6 14:04 mips64/gcc-8.2.0p2.tgz > -rw-rw-r-- 1 openbsd openbsd 27639770 Mar 31 22:07 powerpc/gcc-8.2.0p2.tgz > -rw-rw-r-- 1 openbsd openbsd 22927965 Mar 19 14:44 sparc64/gcc-8.2.0p2.tgz > > I'll try removing the BROKEN-i386 marker and building there in case anything > changed in the 8.3.0 update. > > Won't build, but I'm on it.
Re: NEW: devel/qtav
On Sun, Apr 14, 2019 at 07:25:40PM +0200, Rafael Sadowski wrote:
> here's a new port for QtAV which can be used by the upcoming digikam
> 6.0.0 update. It's more or less a Qt5 QML FFmpeg player. This port is
> based on work from Vadim. After our FFmpeg update it's work quit well
> and fast. Test with the installed player and with digikam6.
devel is overloaded already, this port should go to multimedia I think.
With PKGNAME=${DISTNAME:L} you can ditch having V next to GH_TAGNAME and
the reason for explicit PKGNAME becomes clearer, but that's just a
nit-picky suggestion.
The majority of patches seem trivial to upstream, do you have plans to
do so?
Installed docs are broken, warnings during `make package':
Warning: symlink(s) point to non-existent
/usr/ports/pobj/qtav-1.12.0/fake-amd64/usr/local/share/doc/Player/../../Changelog
/usr/ports/pobj/qtav-1.12.0/fake-amd64/usr/local/share/doc/Player/changelog
Warning: symlink(s) point to non-existent
/usr/ports/pobj/qtav-1.12.0/fake-amd64/usr/local/share/doc/Player/../../README.md
/usr/ports/pobj/qtav-1.12.0/fake-amd64/usr/local/share/doc/Player/README
Warning: symlink(s) point to non-existent
/usr/ports/pobj/qtav-1.12.0/fake-amd64/usr/local/share/doc/QtAV/../../Changelog
/usr/ports/pobj/qtav-1.12.0/fake-amd64/usr/local/share/doc/QtAV/changelog
Warning: symlink(s) point to non-existent
/usr/ports/pobj/qtav-1.12.0/fake-amd64/usr/local/share/doc/QtAV/../../README.md
/usr/ports/pobj/qtav-1.12.0/fake-amd64/usr/local/share/doc/QtAV/README
Warning: symlink(s) point to non-existent
/usr/ports/pobj/qtav-1.12.0/fake-amd64/usr/local/share/doc/QtAVWidgets/../../Changelog
/usr/ports/pobj/qtav-1.12.0/fake-amd64/usr/local/share/doc/QtAVWidgets/changelog
Warning: symlink(s) point to non-existent
/usr/ports/pobj/qtav-1.12.0/fake-amd64/usr/local/share/doc/QtAVWidgets/../../README.md
/usr/ports/pobj/qtav-1.12.0/fake-amd64/usr/local/share/doc/QtAVWidgets/README
> Comment:
> FFmpeg-based Qt multimedia framework
>
> Description:
> QtAV is a multimedia playback library based on Qt and FFmpeg.
That is a bit scarse and pretty much equals COMMENT; this package ships
a full blown player (although not in $PATH), so at least that should be
made clear.
> WWW: http://www.qtav.org/
Has TLS.
Libs are LGPLv2.1 *or later*.
Tested successfully with the player on amd64. Preview images when
hovering over the time line are shown, but in their own X window which
messes with tiling window managers (or at least dwm in my case) as the
image is shows up anywhere but next to the cursor. No biggy and
probably rather a application agnostic WM problem, but worth noting.
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: k...@cvs.openbsd.org2019/04/24 15:49:58 Log message: objgraph is a module that lets you visually explore Python object graphs. It can draw object reference graphs with graphviz, or can make use of xdot for interactive use. Its origin was in a set of functions used while hunting for memory leaks in a Python program. Needed for devel/py-gevent tests Proposal by sthen@, additonal work by me OK sthen@ Status: Vendor Tag: kmos Release Tags: kmos_20190424 N ports/devel/py-objgraph/Makefile N ports/devel/py-objgraph/distinfo N ports/devel/py-objgraph/pkg/DESCR N ports/devel/py-objgraph/pkg/PLIST No conflicts created by this import
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: bl...@cvs.openbsd.org 2019/04/24 15:27:26 Modified files: net/scapy : Makefile Added files: net/scapy/patches: patch-scapy_arch_common_py Log message: In Scapy 2.4.2 probing for OpenBSD tcpdump is broken, so filters for sniff() did not work. Apply patch from upstream. OK sthen@ danj@
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: es...@cvs.openbsd.org 2019/04/24 15:15:23 Modified files: sysutils : Makefile Log message: +random_run
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: es...@cvs.openbsd.org 2019/04/24 15:14:37 Log message: random_run (rr) is a small command that runs something while randomizing its arguments, .e.g, rr mpv * Code and port by me. okay jca@ Status: Vendor Tag: espie Release Tags: ports N ports/sysutils/random_run/distinfo N ports/sysutils/random_run/Makefile N ports/sysutils/random_run/pkg/PLIST N ports/sysutils/random_run/pkg/DESCR No conflicts created by this import
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: st...@cvs.openbsd.org 2019/04/24 15:04:47 Modified files: devel/ccache : Makefile distinfo devel/ccache/patches: patch-Makefile_in patch-configure Log message: update to ccache-3.7, now moved to https://ccache.dev/
Re: NEW: devel/py-objgraph
On 2019/04/24 16:43, Kurt Mosiejczuk wrote: > On Wed, Apr 24, 2019 at 08:58:12PM +0100, Stuart Henderson wrote: > > py-gevent wants this for tests, it's simple and no complex chain of > > new dependencies. ok to import? > > > > > objgraph is a module that lets you visually explore Python object graphs. > > It can draw object reference graphs with graphviz, or can make use of > > xdot for interactive use. > > > > Its origin was in a set of functions used while hunting for memory leaks > > in a Python program. > > > > TEST_DEPENDS needs graphics/py-dot. py-dot needs an update and a > python3 FLAVOR. I'm working on those, but py-objgraph doesn't need > the python3 flavor of py-dot since it is just using dot. (In fact > using the python3 flavor would make it more complicated) > > Attached is my counter-proposal. Mostly yours with the addition and > the Makefile formatting made more fussy-compliant. :) > > --Kurt > > Oh, it doesn't use py-dot, just dot(1) from graphviz, I missed it because I already had graphviz installed and assumed py-graphviz depended on it. Could you just add math/graphviz instead of py-dot to TEST_DEPENDS please? Then it is OK sthen@.
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: bl...@cvs.openbsd.org 2019/04/24 14:33:53 Modified files: net/scapy : Makefile net/scapy/patches: patch-scapy_layers_l2_py Added files: net/scapy/patches: patch-scapy_packet_py Log message: Scapy 2.4.2 sends ARP requests with wrong MAC address. Apply patch from upstream. OK sthen@
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: sebas...@cvs.openbsd.org2019/04/24 14:26:55 Removed files: www/ruby-xmlrpc: Makefile distinfo www/ruby-xmlrpc/pkg: DESCR PLIST Log message: no need for ruby-xmlrpc, it's part of ruby package already didn't install due to conflicting already existing files
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: sebas...@cvs.openbsd.org2019/04/24 14:24:38 Modified files: www: Makefile Log message: unhook ruby-xmlrpc, it's already part of ruby package
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: sebas...@cvs.openbsd.org2019/04/24 14:23:21 Modified files: security/ruby-cms_scanner: Makefile Log message: xmlrpc gem is already part of ruby, no need to depend on an extra gem that doesn't install due to conflicts
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: st...@cvs.openbsd.org 2019/04/24 14:14:08 Modified files: www/py-gunicorn: Makefile distinfo www/py-gunicorn/patches: patch-requirements_test_txt www/py-gunicorn/pkg: PLIST Log message: update to py-gunicorn-19.9.0, tweaks/ok danj@
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: feine...@cvs.openbsd.org2019/04/24 14:13:54 Modified files: databases/py-pg_activity: Makefile distinfo Log message: Update to pg_activity 1.5.0
Re: a couple of other misc python updates
On 2019/04/24 14:21, Daniel Jakots wrote:
> On Tue, 23 Apr 2019 13:55:33 +0100, Stuart Henderson
> wrote:
>
> > I haven't gone through any testing with these yet, but if anyone
> > wants to help prepare for python 3.7 then testing these with
> > dependencies (with the in-tree Python versions) would be helpful.
> >
> > databases/py-psycopg2/Makefile
>
> ok
Thanks. Will wait a couple of days for shadchin@.
> > www/py-aiohttp/Makefile
>
> You're missing some rdep and we don't package some tests so NO_TEST is
> needed. Diff attached to fix, with my diff applied, ok
Thanks, committed.
> > www/py-gunicorn/Makefile
>
> comments inlined + diff attached
> with my diff applied, ok
>
> > www/py-tornado/Makefile
>
> I haven't looked at it
>
>
> > Index: www/py-gunicorn/Makefile
> > ===
> > RCS file: /cvs/ports/www/py-gunicorn/Makefile,v
> > retrieving revision 1.20
> > diff -u -p -r1.20 Makefile
> > --- www/py-gunicorn/Makefile4 May 2018 11:37:24 -
> > 1.20 +++ www/py-gunicorn/Makefile 20 Apr 2019 10:10:12 -
> > @@ -2,7 +2,7 @@
> >
> > COMMENT = Python WSGI HTTP server
> >
> > -MODPY_EGG_VERSION =19.8.1
> > +MODPY_EGG_VERSION =19.9.0
> > DISTNAME = gunicorn-${MODPY_EGG_VERSION}
> > PKGNAME = py-${DISTNAME}
> > CATEGORIES = www
> > @@ -25,7 +25,8 @@ RUN_DEPENDS +=www/py-aiohttp \
> > www/py-multidict
> > .endif
> > TEST_DEPENDS = devel/py-test${MODPY_FLAVOR} \
> > - devel/py-test-cov${MODPY_FLAVOR}
> > + devel/py-test-cov${MODPY_FLAVOR} \
> > + ${BASE_PKGPATH}=${MODPY_EGG_VERSION}
>
> for psycopg2, you used ${FULLPKGNAME}:${BUILD_PKGPATH}, what's the
> difference?
Cargo culted from different ports, we have quite a few different ways
to handle this situation ..
>
> You're missing a tdep on py-coverage and while their check was ok
> during last gunicorn update, we updated py-coverage and now we're
> outside the bound so we need to remove the check in
> patches/patch-requirements_test_txt
Thanks, applied.
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: st...@cvs.openbsd.org 2019/04/24 14:09:30 Modified files: devel/py-coveralls: Makefile devel/py-coveralls/pkg: PLIST Log message: oops, missing bits for py3
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: st...@cvs.openbsd.org 2019/04/24 14:07:11 Modified files: devel : Makefile devel/py-coveralls: Makefile devel/py-coveralls/pkg: PLIST Log message: build py-coveralls,python3
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: st...@cvs.openbsd.org 2019/04/24 14:00:03 Modified files: devel/py-gevent: Makefile Log message: add bits for tests, currently inactive pending importing py-objgraph
NEW: devel/py-objgraph
py-gevent wants this for tests, it's simple and no complex chain of new dependencies. ok to import? objgraph is a module that lets you visually explore Python object graphs. It can draw object reference graphs with graphviz, or can make use of xdot for interactive use. Its origin was in a set of functions used while hunting for memory leaks in a Python program. py-objgraph.tgz Description: application/tar-gz
Re: py-gevent update
> On 2019/04/23 18:16, Kurt Mosiejczuk wrote: > On Tue, Apr 23, 2019 at 01:52:24PM +0100, Stuart Henderson wrote: >> Unbreaks with python 3.7. No problems seen with dep's. OK? > > Needs a NO_TEST. Tests aren't bundled in with the new tarball. > > Otherwise it builds fine on arm64, sparc64, and amd64 here. > > With NO_TEST, OK kmos@ Hi! Please forgive me for this drive-by post, but I'm the maintainer of gevent and happened to see this message go by and it made me curious. Can you please help me understand what you mean by "Tests aren't bundled in with the new tarball?" I thought I was including them. It is true that `greentest/testrunner.py` is no more. That was replaced with `python -m gevent.tests`: $ python3.7 -m venv ./gevent $ cd ./gevent $ wget https://files.pythonhosted.org/packages/ed/27/6c49b70808f569b66ec7fac2e78f076e9b204db9cf5768740cff3d5a07ae/gevent-1.4.0.tar.gz $ tar -xf ./gevent-1.4.0.tar.gz $ . bin/activate (gevent) $ cd gevent-1.4.0/ (gevent) $ python setup.py build_ext -i ... (gevent) $ PYTHONPATH=src python -m gevent.tests Running tests in parallel with concurrency 7 + //python -u -mgevent.tests.test__GreenletExit + //python -u -mgevent.tests.test___config ... + GEVENT_DEBUG=error /private/tmp/gevent/bin/python -u -W ignore -m gevent.testing.monkey_test test_ssl.py - GEVENT_DEBUG=error /private/tmp/gevent/bin/python -u -W ignore -m gevent.testing.monkey_test test_asyncore.py [Ran 101 tests in 5.6s (skipped=50)] + GEVENT_DEBUG=error /private/tmp/gevent/bin/python -u -W ignore -m gevent.testing.monkey_test test_subprocess.py - GEVENT_DEBUG=error /private/tmp/gevent/bin/python -u -W ignore -m gevent.testing.monkey_test test_ftplib.py [Ran 88 tests in 6.9s] ... Of course, the test dependencies listed in setup.py must be available to run all the tests (I think only objgraph is a hard dependency, if the others are missing there will be degraded performance or spurious failures). Compared to 1.2, if the CFFI backends aren't built, there will be some extra test failures; likewise if dnspython isn't installed. If gevent itself isn't installed some tests will also fail when they spawn subprocesses that don't inherit PYTHONPATH (probably 1.2 was the same? I don't remember). Many of those are things I'm working to clean up for gevent 1.5 (e.g., it no longer requires the CFFI backends [https://github.com/gevent/gevent/issues/1366]). Let me know if I can help in any way. Thanks, Jason (I'm not subscribed so I would appreciate being CCd on replies.)
Re: NEW: devel/llvm (Extra Clang Tools)
On Wed, Apr 24 2019, Stuart Henderson wrote:
> On 2019/04/24 20:00, Jeremie Courreges-Anglas wrote:
>> On Tue, Apr 16 2019, Rafael Sadowski wrote:
>> > On Tue Apr 02, 2019 at 11:08:13AM +0200, Jeremie Courreges-Anglas wrote:
>>
>> [...]
>>
>> >> Hmmk so using just ON/OFF just looks cleaner. Unless someone goes on
>> >> a rampage to unify all cmake CONFIGURE_ARGS in the tree, I'd prefer to
>> >> minimize churn and keep llvm consistent.
>> >>
>> >> >>
>> >> >> > - Set CLANG_ENABLE_STATIC_ANALYZER=True, CLANG_INCLUDE_TESTS=True to
>> >> >> > build clang-tidy and enable test. (It is ignored if the folder
>> >> >> > extra
>> >> >> > don't exists)
>>
>> I had to look up whether specifying -DCLANG_INCLUDE_TESTS=True
>> and -DCLANG_ENABLE_STATIC_ANALYZER=True would change the content of
>> other subpackages. Both appear ON by default, so I see no need to
>> specify them.
>>
>> >> >> > - Adjust WANTLIB
>>
>> Thanks.
>>
>> [...]
>>
>> > New diff with following changes:
>> >
>> > - Rename extras to clang-extras
>>
>> I was only suggesting renaming the PKGNAME, not renaming the subpackage
>> and all affiliated variables. Thinking about it some more, I prefer
>> naming the package like done upstream: clang-tools-extra.
>>
>> > - More generic aka. lazy DESCR
>> > - Sorted PLIST
>>
>> Looks good, here's an updated diff that:
>> - keeps -extras as the subpackage name
>> - renames the produced package to clang-tools-extra
>> - fixes PSEUDO_FLAVORS
>> - amends DESCR (missing word, "tool" is repeated)
>>
>> Like sthen I wonder whether all those .a archives are useful.
>> *shrug*
>>
>> Previous versions of this diff have been tested on amd64 and sparc64.
>> ok jca@
>>
>>
>> Index: Makefile
>> ===
>> RCS file: /cvs/ports/devel/llvm/Makefile,v
>> retrieving revision 1.216
>> diff -u -p -r1.216 Makefile
>> --- Makefile 9 Mar 2019 06:18:11 - 1.216
>> +++ Makefile 24 Apr 2019 16:22:45 -
>> @@ -8,9 +8,10 @@ ONLY_FOR_ARCHS = ${LLVM_ARCHS}
>>
>> DPB_PROPERTIES = parallel
>>
>> -MULTI_PACKAGES =-main -python -lldb
>> +MULTI_PACKAGES =-main -python -lldb -extras
>> COMMENT-main = modular, fast C/C++/ObjC compiler, static analyzer and
>> tools
>> COMMENT-lldb = LLDB debugger
>> +COMMENT-extras =clang extra tools
>
> usual capitalisation is Clang, I'd prefer to stick to that (especially
> with it done like that for -python).
Right.
>> COMMENT-python =Python bindings for Clang
>>
>> LLVM_V =7.0.1
>> @@ -20,13 +21,15 @@ PKGSPEC-main = llvm-=${LLVM_V}
>> PKGNAME-main = llvm-${LLVM_V}
>> PKGNAME-python =py-llvm-${LLVM_V}
>> PKGNAME-lldb = lldb-${LLVM_V}
>> +PKGNAME-extras =clang-tools-extra-${LLVM_V}
>> REVISION-main = 9
>> REVISION-lldb = 0
>
> -main and -lldb both need bumps.
lol I focused so much on the possible changes due to the additional
CONFIGURE_ARGS, I completely missed the WANTLIB changes...
>> CATEGORIES =devel
>> DISTFILES = llvm-${LLVM_V}.src${EXTRACT_SUFX} \
>> cfe-${LLVM_V}.src${EXTRACT_SUFX} \
>> lld-${LLVM_V}.src${EXTRACT_SUFX} \
>> -lldb-${LLVM_V}.src${EXTRACT_SUFX}
>> +lldb-${LLVM_V}.src${EXTRACT_SUFX} \
>> +clang-tools-extra-${LLVM_V}.src${EXTRACT_SUFX}
>> MASTER_SITES = https://www.llvm.org/releases/${LLVM_V}/
>
> this immediately redirects to http://releases.llvm.org/7.0.1/ now
> (non https); MASTER_SITES could be updated to
> https://releases.llvm.org/${LLVM_V}/
Right.
> Otherwise OK sthen.
Updated diff below. Thanks a lot!
Index: Makefile
===
RCS file: /cvs/ports/devel/llvm/Makefile,v
retrieving revision 1.216
diff -u -p -r1.216 Makefile
--- Makefile9 Mar 2019 06:18:11 - 1.216
+++ Makefile24 Apr 2019 19:06:53 -
@@ -8,9 +8,10 @@ ONLY_FOR_ARCHS = ${LLVM_ARCHS}
DPB_PROPERTIES = parallel
-MULTI_PACKAGES = -main -python -lldb
+MULTI_PACKAGES = -main -python -lldb -extras
COMMENT-main = modular, fast C/C++/ObjC compiler, static analyzer and tools
COMMENT-lldb = LLDB debugger
+COMMENT-extras = Clang extra tools
COMMENT-python = Python bindings for Clang
LLVM_V = 7.0.1
@@ -20,14 +21,17 @@ PKGSPEC-main = llvm-=${LLVM_V}
PKGNAME-main = llvm-${LLVM_V}
PKGNAME-python = py-llvm-${LLVM_V}
PKGNAME-lldb = lldb-${LLVM_V}
-REVISION-main =9
-REVISION-lldb =0
+PKGNAME-extras = clang-tools-extra-${LLVM_V}
+REVISION-main =10
+REVISION-lldb =1
CATEGORIES = devel
+
DISTFILES =llvm-${LLVM_V}.src${EXTRACT_SUFX} \
cfe-${LLVM_V}.src${EXTRACT_SUFX} \
lld-${LLVM_V}.src${EXTRACT_SUFX} \
- lldb-${LLVM_V}.src${EXTRACT_SUFX}
-MASTER_SITES = https://www.llvm.org/releases/${LLVM_V}/
+ lldb-${LLVM_V}.src${EXTRACT_SUFX} \
+
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: st...@cvs.openbsd.org 2019/04/24 13:00:34 Modified files: www/py-aiohttp : Makefile distinfo www/py-aiohttp/pkg: PLIST Log message: update to py-aiohttp-3.5.4, tweaks/ok danj@
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: c...@cvs.openbsd.org2019/04/24 12:39:06 Modified files: x11/rxvt-unicode: Makefile x11/rxvt-unicode/pkg: DESCR Added files: x11/rxvt-unicode/patches: patch-configure Log message: rxvt-unicode: don't use libsupc++ with ports-gcc, like base-clang already does. This fixes the build with ports-gcc-8. Thanks to jca@ who pointed out that we could use libesupc++, and performed a simple test that showed there is not much incentive to do so. While here, remove trailing whitespaces in DESCR. OK bcallah@ jca@
Re: NEW: devel/llvm (Extra Clang Tools)
On 2019/04/24 20:00, Jeremie Courreges-Anglas wrote:
> On Tue, Apr 16 2019, Rafael Sadowski wrote:
> > On Tue Apr 02, 2019 at 11:08:13AM +0200, Jeremie Courreges-Anglas wrote:
>
> [...]
>
> >> Hmmk so using just ON/OFF just looks cleaner. Unless someone goes on
> >> a rampage to unify all cmake CONFIGURE_ARGS in the tree, I'd prefer to
> >> minimize churn and keep llvm consistent.
> >>
> >> >>
> >> >> > - Set CLANG_ENABLE_STATIC_ANALYZER=True, CLANG_INCLUDE_TESTS=True to
> >> >> > build clang-tidy and enable test. (It is ignored if the folder extra
> >> >> > don't exists)
>
> I had to look up whether specifying -DCLANG_INCLUDE_TESTS=True
> and -DCLANG_ENABLE_STATIC_ANALYZER=True would change the content of
> other subpackages. Both appear ON by default, so I see no need to
> specify them.
>
> >> >> > - Adjust WANTLIB
>
> Thanks.
>
> [...]
>
> > New diff with following changes:
> >
> > - Rename extras to clang-extras
>
> I was only suggesting renaming the PKGNAME, not renaming the subpackage
> and all affiliated variables. Thinking about it some more, I prefer
> naming the package like done upstream: clang-tools-extra.
>
> > - More generic aka. lazy DESCR
> > - Sorted PLIST
>
> Looks good, here's an updated diff that:
> - keeps -extras as the subpackage name
> - renames the produced package to clang-tools-extra
> - fixes PSEUDO_FLAVORS
> - amends DESCR (missing word, "tool" is repeated)
>
> Like sthen I wonder whether all those .a archives are useful.
> *shrug*
>
> Previous versions of this diff have been tested on amd64 and sparc64.
> ok jca@
>
>
> Index: Makefile
> ===
> RCS file: /cvs/ports/devel/llvm/Makefile,v
> retrieving revision 1.216
> diff -u -p -r1.216 Makefile
> --- Makefile 9 Mar 2019 06:18:11 - 1.216
> +++ Makefile 24 Apr 2019 16:22:45 -
> @@ -8,9 +8,10 @@ ONLY_FOR_ARCHS = ${LLVM_ARCHS}
>
> DPB_PROPERTIES = parallel
>
> -MULTI_PACKAGES = -main -python -lldb
> +MULTI_PACKAGES = -main -python -lldb -extras
> COMMENT-main = modular, fast C/C++/ObjC compiler, static analyzer and
> tools
> COMMENT-lldb = LLDB debugger
> +COMMENT-extras = clang extra tools
usual capitalisation is Clang, I'd prefer to stick to that (especially
with it done like that for -python).
> COMMENT-python = Python bindings for Clang
>
> LLVM_V = 7.0.1
> @@ -20,13 +21,15 @@ PKGSPEC-main =llvm-=${LLVM_V}
> PKGNAME-main = llvm-${LLVM_V}
> PKGNAME-python = py-llvm-${LLVM_V}
> PKGNAME-lldb = lldb-${LLVM_V}
> +PKGNAME-extras = clang-tools-extra-${LLVM_V}
> REVISION-main = 9
> REVISION-lldb = 0
-main and -lldb both need bumps.
> CATEGORIES = devel
> DISTFILES = llvm-${LLVM_V}.src${EXTRACT_SUFX} \
> cfe-${LLVM_V}.src${EXTRACT_SUFX} \
> lld-${LLVM_V}.src${EXTRACT_SUFX} \
> - lldb-${LLVM_V}.src${EXTRACT_SUFX}
> + lldb-${LLVM_V}.src${EXTRACT_SUFX} \
> + clang-tools-extra-${LLVM_V}.src${EXTRACT_SUFX}
> MASTER_SITES = https://www.llvm.org/releases/${LLVM_V}/
this immediately redirects to http://releases.llvm.org/7.0.1/ now
(non https); MASTER_SITES could be updated to
https://releases.llvm.org/${LLVM_V}/
Otherwise OK sthen.
> EXTRACT_SUFX = .tar.xz
>
> @@ -42,10 +45,10 @@ MAINTAINER= Jeremie Courreges-Anglas # BSD
> PERMIT_PACKAGE_CDROM = Yes
>
> -WANTLIB =${COMPILER_LIBCXX} c curses edit form m panel pthread \
> +WANTLIB =${COMPILER_LIBCXX} c edit form m panel pthread \
> ${MODPY_WANTLIB} z
>
> -PSEUDO_FLAVORS = no_lldb
> +PSEUDO_FLAVORS = no_lldb no_extras
> FLAVOR ?=
> NOT_FOR_ARCHS-lldb = ${GCC4_ARCHS}
>
> @@ -72,12 +75,16 @@ WANTLIB-python = clang
> RUN_DEPENDS-python = ${MODPY_RUN_DEPENDS}
> LIB_DEPENDS-python = ${BUILD_PKGPATH},-main
>
> -WANTLIB-lldb = clang
> +WANTLIB-lldb = ${COMPILER_LIBCXX} clang c curses edit form m
> panel
> +WANTLIB-lldb += ${MODPY_WANTLIB}
> LIB_DEPENDS-lldb = ${BUILD_PKGPATH},-main
> RUN_DEPENDS-lldb = ${MODPY_RUN_DEPENDS} \
> devel/py-six
>
> -SEPARATE_BUILD = Yes
> +WANTLIB-extras = ${COMPILER_LIBCXX} c m clang
> +LIB_DEPENDS-extras = ${BUILD_PKGPATH},-main
> +RUN_DEPENDS-extras = ${MODPY_RUN_DEPENDS}
> +
> CONFIGURE_ARGS = -DLLVM_ENABLE_FFI:Bool=False \
> -DLLVM_ENABLE_TERMINFO:Bool=False \
> -DLLVM_ENABLE_RTTI:Bool=True \
> @@ -111,7 +118,7 @@ PKG_ARGS += -Dpowerpc=1
> PKG_ARGS += -Dpowerpc=0
> .endif
>
> -TEST_TARGET =check
> +TEST_TARGET =check check-clang check-clang-tools
>
> # XXX sync
> GCC_VER =4.9.4
> @@ -127,12 +134,16 @@ post-extract:
> mv ${WRKDIR}/cfe-${LLVM_V}.src ${WRKSRC}/tools/clang
> mv ${WRKDIR}/lld-${LLVM_V}.src ${WRKSRC}/tools/lld
> mv
Re: a couple of other misc python updates
On Tue, 23 Apr 2019 13:55:33 +0100, Stuart Henderson
wrote:
> I haven't gone through any testing with these yet, but if anyone
> wants to help prepare for python 3.7 then testing these with
> dependencies (with the in-tree Python versions) would be helpful.
>
> databases/py-psycopg2/Makefile
ok
> www/py-aiohttp/Makefile
You're missing some rdep and we don't package some tests so NO_TEST is
needed. Diff attached to fix, with my diff applied, ok
> www/py-gunicorn/Makefile
comments inlined + diff attached
with my diff applied, ok
> www/py-tornado/Makefile
I haven't looked at it
> Index: www/py-gunicorn/Makefile
> ===
> RCS file: /cvs/ports/www/py-gunicorn/Makefile,v
> retrieving revision 1.20
> diff -u -p -r1.20 Makefile
> --- www/py-gunicorn/Makefile 4 May 2018 11:37:24 -
> 1.20 +++ www/py-gunicorn/Makefile 20 Apr 2019 10:10:12 -
> @@ -2,7 +2,7 @@
>
> COMMENT =Python WSGI HTTP server
>
> -MODPY_EGG_VERSION = 19.8.1
> +MODPY_EGG_VERSION = 19.9.0
> DISTNAME = gunicorn-${MODPY_EGG_VERSION}
> PKGNAME =py-${DISTNAME}
> CATEGORIES = www
> @@ -25,7 +25,8 @@ RUN_DEPENDS += www/py-aiohttp \
> www/py-multidict
> .endif
> TEST_DEPENDS = devel/py-test${MODPY_FLAVOR} \
> - devel/py-test-cov${MODPY_FLAVOR}
> + devel/py-test-cov${MODPY_FLAVOR} \
> + ${BASE_PKGPATH}=${MODPY_EGG_VERSION}
for psycopg2, you used ${FULLPKGNAME}:${BUILD_PKGPATH}, what's the
difference?
You're missing a tdep on py-coverage and while their check was ok
during last gunicorn update, we updated py-coverage and now we're
outside the bound so we need to remove the check in
patches/patch-requirements_test_txt
--- Makefile.orig Wed Apr 24 13:26:40 2019
+++ Makefile Wed Apr 24 13:31:04 2019
@@ -19,11 +19,16 @@ MODPY_SETUPTOOLS = Yes
MODPY_VERSION = ${MODPY_DEFAULT_VERSION_3}
-# py-multidict is python3 only
-RUN_DEPENDS = textproc/py-chardet${MODPY_FLAVOR} \
- www/py-multidict
+# only py-chardet isn't python3 only
+RUN_DEPENDS = devel/py-async-timeout \
+ textproc/py-chardet${MODPY_FLAVOR} \
+ www/py-multidict \
+ www/py-yarl
TEST_DEPENDS = ${RUN_DEPENDS} \
devel/py-test${MODPY_FLAVOR} \
www/py-gunicorn${MODPY_FLAVOR}
+
+# we don't have any port for pytest-{xdist,timeout}
+NO_TEST = Yes
.include
Index: Makefile
===
RCS file: /cvs/ports/www/py-gunicorn/Makefile,v
retrieving revision 1.20
diff -u -p -r1.20 Makefile
--- Makefile 4 May 2018 11:37:24 - 1.20
+++ Makefile 24 Apr 2019 18:20:39 -
@@ -2,7 +2,7 @@
COMMENT = Python WSGI HTTP server
-MODPY_EGG_VERSION = 19.8.1
+MODPY_EGG_VERSION = 19.9.0
DISTNAME = gunicorn-${MODPY_EGG_VERSION}
PKGNAME = py-${DISTNAME}
CATEGORIES = www
@@ -24,8 +24,10 @@ FLAVOR ?=
RUN_DEPENDS += www/py-aiohttp \
www/py-multidict
.endif
-TEST_DEPENDS = devel/py-test${MODPY_FLAVOR} \
- devel/py-test-cov${MODPY_FLAVOR}
+TEST_DEPENDS = devel/py-coverage${MODPY_FLAVOR} \
+ devel/py-test${MODPY_FLAVOR} \
+ devel/py-test-cov${MODPY_FLAVOR} \
+ ${BASE_PKGPATH}=${MODPY_EGG_VERSION}
.if ! ${FLAVOR:Mpython3}
TEST_DEPENDS += devel/py-mock
.endif
Index: distinfo
===
RCS file: /cvs/ports/www/py-gunicorn/distinfo,v
retrieving revision 1.13
diff -u -p -r1.13 distinfo
--- distinfo 4 May 2018 11:37:24 - 1.13
+++ distinfo 24 Apr 2019 18:20:39 -
@@ -1,2 +1,2 @@
-SHA256 (gunicorn-19.8.1.tar.gz) = vFkAWXnvttLdfVunLZn4qEIoYq0X/zoW6QBoRjDdKhA=
-SIZE (gunicorn-19.8.1.tar.gz) = 415541
+SHA256 (gunicorn-19.9.0.tar.gz) = +iZiCXxm+SD1P3BiHGxYyko8TTQ0IF5gjhIbWztx9PM=
+SIZE (gunicorn-19.9.0.tar.gz) = 415774
Index: patches/patch-requirements_test_txt
===
RCS file: /cvs/ports/www/py-gunicorn/patches/patch-requirements_test_txt,v
retrieving revision 1.3
diff -u -p -r1.3 patch-requirements_test_txt
--- patches/patch-requirements_test_txt 4 May 2018 11:37:24 - 1.3
+++ patches/patch-requirements_test_txt 24 Apr 2019 18:20:39 -
@@ -6,8 +6,9 @@ Index: requirements_test.txt
--- requirements_test.txt.orig
+++ requirements_test.txt
@@ -1,3 +1,3 @@
- coverage>=4.0,<4.4 # TODO: https://github.com/benoitc/gunicorn/issues/1548
--pytest==3.0.5
--pytest-cov==2.4.0
+-coverage>=4.0,<4.4 # TODO: https://github.com/benoitc/gunicorn/issues/1548
+-pytest==3.2.5 # TODO: upgrade to latest version requires drop support to Python 2.6
+-pytest-cov==2.5.1
++coverage
+pytest
+pytest-cov
Index: pkg/PLIST
===
RCS file: /cvs/ports/www/py-gunicorn/pkg/PLIST,v
retrieving revision 1.8
diff -u -p -r1.8 PLIST
--- pkg/PLIST 4 May 2018 11:37:24 - 1.8
+++ pkg/PLIST 24 Apr 2019 18:20:39 -
@@ -80,8
Re: NEW: devel/llvm (Extra Clang Tools)
On Tue, Apr 16 2019, Rafael Sadowski wrote:
> On Tue Apr 02, 2019 at 11:08:13AM +0200, Jeremie Courreges-Anglas wrote:
[...]
>> Hmmk so using just ON/OFF just looks cleaner. Unless someone goes on
>> a rampage to unify all cmake CONFIGURE_ARGS in the tree, I'd prefer to
>> minimize churn and keep llvm consistent.
>>
>> >>
>> >> > - Set CLANG_ENABLE_STATIC_ANALYZER=True, CLANG_INCLUDE_TESTS=True to
>> >> > build clang-tidy and enable test. (It is ignored if the folder extra
>> >> > don't exists)
I had to look up whether specifying -DCLANG_INCLUDE_TESTS=True
and -DCLANG_ENABLE_STATIC_ANALYZER=True would change the content of
other subpackages. Both appear ON by default, so I see no need to
specify them.
>> >> > - Adjust WANTLIB
Thanks.
[...]
> New diff with following changes:
>
> - Rename extras to clang-extras
I was only suggesting renaming the PKGNAME, not renaming the subpackage
and all affiliated variables. Thinking about it some more, I prefer
naming the package like done upstream: clang-tools-extra.
> - More generic aka. lazy DESCR
> - Sorted PLIST
Looks good, here's an updated diff that:
- keeps -extras as the subpackage name
- renames the produced package to clang-tools-extra
- fixes PSEUDO_FLAVORS
- amends DESCR (missing word, "tool" is repeated)
Like sthen I wonder whether all those .a archives are useful.
*shrug*
Previous versions of this diff have been tested on amd64 and sparc64.
ok jca@
Index: Makefile
===
RCS file: /cvs/ports/devel/llvm/Makefile,v
retrieving revision 1.216
diff -u -p -r1.216 Makefile
--- Makefile9 Mar 2019 06:18:11 - 1.216
+++ Makefile24 Apr 2019 16:22:45 -
@@ -8,9 +8,10 @@ ONLY_FOR_ARCHS = ${LLVM_ARCHS}
DPB_PROPERTIES = parallel
-MULTI_PACKAGES = -main -python -lldb
+MULTI_PACKAGES = -main -python -lldb -extras
COMMENT-main = modular, fast C/C++/ObjC compiler, static analyzer and tools
COMMENT-lldb = LLDB debugger
+COMMENT-extras = clang extra tools
COMMENT-python = Python bindings for Clang
LLVM_V = 7.0.1
@@ -20,13 +21,15 @@ PKGSPEC-main = llvm-=${LLVM_V}
PKGNAME-main = llvm-${LLVM_V}
PKGNAME-python = py-llvm-${LLVM_V}
PKGNAME-lldb = lldb-${LLVM_V}
+PKGNAME-extras = clang-tools-extra-${LLVM_V}
REVISION-main =9
REVISION-lldb =0
CATEGORIES = devel
DISTFILES =llvm-${LLVM_V}.src${EXTRACT_SUFX} \
cfe-${LLVM_V}.src${EXTRACT_SUFX} \
lld-${LLVM_V}.src${EXTRACT_SUFX} \
- lldb-${LLVM_V}.src${EXTRACT_SUFX}
+ lldb-${LLVM_V}.src${EXTRACT_SUFX} \
+ clang-tools-extra-${LLVM_V}.src${EXTRACT_SUFX}
MASTER_SITES = https://www.llvm.org/releases/${LLVM_V}/
EXTRACT_SUFX = .tar.xz
@@ -42,10 +45,10 @@ MAINTAINER= Jeremie Courreges-Anglas
Re: [NEW PORT]: rr
On Wed, Apr 24 2019, Marc Espie wrote:
> I fixed usage (missed -n), retagged as 1.1
>
> New tarball.
>
> - marked as only for clang archs (until the gcc 8 switch)
> - obey LDFLAGS/CXXFLAGS trivially
>
> - rename the package itself to random_run.
Nicer.
> I think keeping the short name for the command (rr) makes sense
> for now... I don't see mozilla's rr being ported any time soon.
Indeed, given the amount of #include >
>> As a result, I've written a new (smallish) command called rr:
>> ---
>>
>> rr (random run) runs a command with randomized arguments.
>>
>> Basically,
>>
>> $ rr mpv *.mp4
>> will give you a true shuffle, that doesn't play the same file twice.
>>
>> there are more options:
>>
>> $ rr xv -nolimit -- *
>> allows you to specify options that won't be jumbled.
>>
>> $ rr -n 500 -x '.*.pdf' xv -nolimit -- *
>> will run xv several times on all files in the current directory
>> except those matching '.*.pdf', limiting each run to 500 parameters.
>>
>> --
>>
>> Port attached
>
>
> # $OpenBSD$
>
> # TODO awaiting the gcc8 switch
> ONLY_FOR_ARCHS = ${CLANG_ARCHS}
>
> COMMENT = run command with randomized parameters
> CATEGORIES = sysutils
> MAINTAINER = Marc Espie
>
> GH_PROJECT = random_run
> GH_TAGNAME = 1.1
> GH_ACCOUNT = marcespie
>
> COMPILER =base-clang ports-gcc
>
> # BSD licence
> PERMIT_PACKAGE_CDROM = Yes
> NO_TESTS =Yes
>
>
NO_TESTS -> NO_TEST, extra empty line.
> WANTLIB = ${COMPILER_LIBCXX} c m
>
> MAKE_FLAGS = OPTIMIZE="${CXXFLAGS}" LDFLAGS="${LDFLAGS}"
>
> do-install:
> ${INSTALL_PROGRAM} ${WRKBUILD}/rr ${PREFIX}/bin
> ${INSTALL_MAN} ${WRKSRC}/rr.1 ${PREFIX}/man/man1
>
>
> .include
> SHA256 (random_run-1.1.tar.gz) = SkIj19KNe31ZKymf39/TVwQLv9jY1GdpSyY7h7B+aHY=
> SIZE (random_run-1.1.tar.gz) = 3475
> rr (random run) runs a command with randomized arguments.
>
> Basically,
>
> $ rr mpv *.mp4
> will give you a true shuffle, that doesn't play the same file twice.
>
> there are more options:
*There
>
> $ rr xv -nolimit -- *
> allows you to specify options that won't be jumbled.
>
> $ rr -n 500 -x '.*.pdf' xv -nolimit -- *
> will run xv several times on all files in the current directory
> except those matching '.*.pdf', limiting each run to 500 parameters.
>
> See the manpage for details
Missing period.
> @comment $OpenBSD: PLIST,v$
> @bin bin/rr
> @man man/man1/rr.1
>
--
jca | PGP : 0x1524E7EE / 5135 92C1 AD36 5293 2BDF DDCC 0DFA 74AE 1524 E7EE
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: st...@cvs.openbsd.org 2019/04/24 10:34:42 Modified files: devel/py-gevent: Makefile distinfo devel/py-gevent/patches: patch-setup_py devel/py-gevent/pkg: PLIST Log message: update to py-gevent-1.4.0, tweak/ok kmos@
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: st...@cvs.openbsd.org 2019/04/24 10:33:09 Modified files: www/tomcat : Makefile Added files: www/tomcat/v9 : Makefile distinfo www/tomcat/v9/patches: patch-conf_server_xml www/tomcat/v9/pkg: DESCR-examples DESCR-main PLIST-examples PLIST-main README-main tomcat.rc Log message: add tomcat-9.0.17
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: st...@cvs.openbsd.org 2019/04/24 10:32:19 Modified files: www/tomcat/v8 : Makefile distinfo www/tomcat/v8/pkg: PLIST-examples PLIST-main Log message: update to tomcat-8.5.39 use PKGSPEC, regen plist
Update: JRuby 9.2.6.0 -> 9.2.7.0
Simple update to the latest release of JRuby. Release announcement at
https://www.jruby.org/2019/04/09/jruby-9-2-7-0.html. One major
improvement is almost all refinements now work correctly.
Tested on amd64, the only supported arch. Will be committing in a
couple days unless I hear objections.
Thanks,
Jeremy
Index: Makefile
===
RCS file: /cvs/ports/lang/jruby/Makefile,v
retrieving revision 1.76
diff -u -p -r1.76 Makefile
--- Makefile24 Mar 2019 22:24:13 - 1.76
+++ Makefile14 Apr 2019 03:06:52 -
@@ -5,8 +5,7 @@ ONLY_FOR_ARCHS = amd64
COMMENT = pure-Java implementation of the Ruby language
-V =9.2.6.0
-REVISION = 0
+V =9.2.7.0
DISTNAME = jruby-dist-${V}-bin
PKGNAME = jruby-${V}
CATEGORIES = lang lang/ruby
Index: distinfo
===
RCS file: /cvs/ports/lang/jruby/distinfo,v
retrieving revision 1.48
diff -u -p -r1.48 distinfo
--- distinfo19 Feb 2019 05:23:53 - 1.48
+++ distinfo14 Apr 2019 03:07:03 -
@@ -1,6 +1,6 @@
SHA256 (jnr-jffi-1.2.2-0-g4c196bb.tar.gz) =
xK/m48Z/YA+fg4yFJqcRxceFnT0F98y255Ju9eSE7b0=
-SHA256 (jruby-dist-9.2.6.0-bin.tar.gz) =
cKH/Dhepi6pj6pLJH9OP8eVaIFbl1XugQJxFQ9KeDj0=
+SHA256 (jruby-dist-9.2.7.0-bin.tar.gz) =
2nwaXOkAFcC6/UvKA1IpTgj+HJ7ASaxR6C/lftUOE0g=
SHA256 (jruby-launcher-1.1.5-java.gem) =
IjxjzP/iW+z5izqdVk33ZXj1k6bs7tBZS/Gos1raxZg=
SIZE (jnr-jffi-1.2.2-0-g4c196bb.tar.gz) = 1759433
-SIZE (jruby-dist-9.2.6.0-bin.tar.gz) = 24072079
+SIZE (jruby-dist-9.2.7.0-bin.tar.gz) = 25630203
SIZE (jruby-launcher-1.1.5-java.gem) = 61440
Index: pkg/PLIST
===
RCS file: /cvs/ports/lang/jruby/pkg/PLIST,v
retrieving revision 1.43
diff -u -p -r1.43 PLIST
--- pkg/PLIST 19 Feb 2019 05:23:53 - 1.43
+++ pkg/PLIST 14 Apr 2019 03:07:21 -
@@ -27,6 +27,7 @@ jruby/lib/ruby/2.0/site_ruby/
jruby/lib/ruby/gems/
jruby/lib/ruby/gems/1.8/
jruby/lib/ruby/gems/1.8/cache/
+jruby/lib/ruby/gems/1.8/cache/bundler-1.16.6.gem
jruby/lib/ruby/gems/1.8/cache/did_you_mean-1.2.0.gem
jruby/lib/ruby/gems/1.8/cache/minitest-5.10.3.gem
jruby/lib/ruby/gems/1.8/cache/net-telnet-0.1.1.gem
@@ -35,6 +36,356 @@ jruby/lib/ruby/gems/1.8/cache/rake-12.3.
jruby/lib/ruby/gems/1.8/cache/test-unit-3.2.8.gem
jruby/lib/ruby/gems/1.8/cache/xmlrpc-0.3.0.gem
jruby/lib/ruby/gems/1.8/gems/
+jruby/lib/ruby/gems/1.8/gems/bundler-1.16.6/
+jruby/lib/ruby/gems/1.8/gems/bundler-1.16.6/CHANGELOG.md
+jruby/lib/ruby/gems/1.8/gems/bundler-1.16.6/LICENSE.md
+jruby/lib/ruby/gems/1.8/gems/bundler-1.16.6/README.md
+jruby/lib/ruby/gems/1.8/gems/bundler-1.16.6/bundler.gemspec
+jruby/lib/ruby/gems/1.8/gems/bundler-1.16.6/exe/
+jruby/lib/ruby/gems/1.8/gems/bundler-1.16.6/exe/bundle
+jruby/lib/ruby/gems/1.8/gems/bundler-1.16.6/exe/bundle_ruby
+jruby/lib/ruby/gems/1.8/gems/bundler-1.16.6/exe/bundler
+jruby/lib/ruby/gems/1.8/gems/bundler-1.16.6/lib/
+jruby/lib/ruby/gems/1.8/gems/bundler-1.16.6/lib/bundler/
+jruby/lib/ruby/gems/1.8/gems/bundler-1.16.6/lib/bundler.rb
+jruby/lib/ruby/gems/1.8/gems/bundler-1.16.6/lib/bundler/build_metadata.rb
+jruby/lib/ruby/gems/1.8/gems/bundler-1.16.6/lib/bundler/capistrano.rb
+jruby/lib/ruby/gems/1.8/gems/bundler-1.16.6/lib/bundler/cli/
+jruby/lib/ruby/gems/1.8/gems/bundler-1.16.6/lib/bundler/cli.rb
+jruby/lib/ruby/gems/1.8/gems/bundler-1.16.6/lib/bundler/cli/add.rb
+jruby/lib/ruby/gems/1.8/gems/bundler-1.16.6/lib/bundler/cli/binstubs.rb
+jruby/lib/ruby/gems/1.8/gems/bundler-1.16.6/lib/bundler/cli/cache.rb
+jruby/lib/ruby/gems/1.8/gems/bundler-1.16.6/lib/bundler/cli/check.rb
+jruby/lib/ruby/gems/1.8/gems/bundler-1.16.6/lib/bundler/cli/clean.rb
+jruby/lib/ruby/gems/1.8/gems/bundler-1.16.6/lib/bundler/cli/common.rb
+jruby/lib/ruby/gems/1.8/gems/bundler-1.16.6/lib/bundler/cli/config.rb
+jruby/lib/ruby/gems/1.8/gems/bundler-1.16.6/lib/bundler/cli/console.rb
+jruby/lib/ruby/gems/1.8/gems/bundler-1.16.6/lib/bundler/cli/doctor.rb
+jruby/lib/ruby/gems/1.8/gems/bundler-1.16.6/lib/bundler/cli/exec.rb
+jruby/lib/ruby/gems/1.8/gems/bundler-1.16.6/lib/bundler/cli/gem.rb
+jruby/lib/ruby/gems/1.8/gems/bundler-1.16.6/lib/bundler/cli/info.rb
+jruby/lib/ruby/gems/1.8/gems/bundler-1.16.6/lib/bundler/cli/init.rb
+jruby/lib/ruby/gems/1.8/gems/bundler-1.16.6/lib/bundler/cli/inject.rb
+jruby/lib/ruby/gems/1.8/gems/bundler-1.16.6/lib/bundler/cli/install.rb
+jruby/lib/ruby/gems/1.8/gems/bundler-1.16.6/lib/bundler/cli/issue.rb
+jruby/lib/ruby/gems/1.8/gems/bundler-1.16.6/lib/bundler/cli/list.rb
+jruby/lib/ruby/gems/1.8/gems/bundler-1.16.6/lib/bundler/cli/lock.rb
+jruby/lib/ruby/gems/1.8/gems/bundler-1.16.6/lib/bundler/cli/open.rb
+jruby/lib/ruby/gems/1.8/gems/bundler-1.16.6/lib/bundler/cli/outdated.rb
+jruby/lib/ruby/gems/1.8/gems/bundler-1.16.6/lib/bundler/cli/package.rb
+jruby/lib/ruby/gems/1.8/gems/bundler-1.16.6/lib/bundler/cli/platform.rb
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: st...@cvs.openbsd.org 2019/04/24 10:31:51 Modified files: www/tomcat/v7 : Makefile distinfo www/tomcat/v7/pkg: PLIST-examples PLIST-main Log message: update to tomcat-7.0.93 use PKGSPEC, regen plist
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: st...@cvs.openbsd.org 2019/04/24 10:31:06 Modified files: www/tomcat/v6 : Makefile www/tomcat/v6/pkg: PLIST-main Log message: use PKGSPEC, regen plist
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: st...@cvs.openbsd.org 2019/04/24 10:28:59 Modified files: www/tomcat : Makefile.inc Log message: - use FIX_EXTRACT_PERMISSIONS - use find ... -delete rather than | xargs rm
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: st...@cvs.openbsd.org 2019/04/24 10:26:59 Modified files: audio/fdk-aac : Makefile audio/hs-libmpd: Makefile audio/libmpdclient: Makefile audio/mpc : Makefile audio/mpd : Makefile audio/ncmpc: Makefile audio/ncmpcpp : Makefile audio/pianobar : Makefile audio/zeya : Makefile databases/hs-HDBC: Makefile databases/hs-HDBC-sqlite3: Makefile databases/mdbtools: Makefile databases/redis: Makefile devel/cmake: Makefile devel/cryptopp : Makefile devel/go-check-v1: Makefile devel/go-tools : Makefile devel/hs-ConfigFile: Makefile devel/hs-MissingH: Makefile devel/hs-convertible: Makefile devel/hs-hslogger: Makefile devel/jsoncpp : Makefile devel/libev: Makefile devel/lua-penlight: Makefile devel/lualdoc : Makefile devel/py-certifi: Makefile devel/py-dulwich: Makefile devel/py-fastimport: Makefile devel/py-gevent: Makefile fonts/noto : Makefile.inc fonts/noto/cjk : Makefile fonts/noto/emoji: Makefile fonts/noto/fonts: Makefile games/gnubg: Makefile games/pysol: Makefile games/xlennart : Makefile graphics/feh : Makefile graphics/gimp/deskew: Makefile graphics/img2pdf: Makefile graphics/imv : Makefile graphics/quirc : Makefile graphics/scour : Makefile math/libtommath: Makefile net/amule : Makefile net/go-net : Makefile net/hpodder: Makefile net/libtorrent : Makefile net/miniupnp/libnatpmp: Makefile net/mldonkey : Makefile net/pidgin-window-merge: Makefile net/py-geventhttpclient: Makefile net/rtorrent : Makefile net/telegram-purple: Makefile net/tkabber: Makefile.inc net/tkabber/base: Makefile net/tkabber/plugins: Makefile security/ccid : Makefile security/go-crypto: Makefile security/opensc: Makefile security/pcsc-lite: Makefile security/rhash : Makefile security/tclgpg: Makefile security/wpa_supplicant: Makefile sysutils/gdmap : Makefile telephony/baresip: Makefile.inc telephony/baresip/baresip: Makefile telephony/baresip/re: Makefile telephony/baresip/rem: Makefile telephony/baresip/restund: Makefile telephony/libzrtp: Makefile textproc/clit : Makefile textproc/go-text: Makefile textproc/go-xlsx: Makefile textproc/hs-HaXml: Makefile textproc/hs-libxml-sax: Makefile textproc/hs-polyparse: Makefile textproc/hs-xml-types: Makefile textproc/p5-Sourcecode-Spellchecker: Makefile textproc/txt2man: Makefile textproc/xlsx2csv: Makefile www/cntlm : Makefile www/minitube : Makefile www/php-predis : Makefile www/tomcat : Makefile.inc www/tomcat/v6 : Makefile www/tomcat/v7 : Makefile www/tomcat/v8 : Makefile x11/awesome: Makefile x11/faenza-icon-theme: Makefile x11/gmtk : Makefile x11/gnome-mplayer: Makefile x11/gtk2-murrine-engine: Makefile x11/hs-dbus: Makefile x11/i3 : Makefile x11/libxdg-basedir: Makefile x11/mlterm : Makefile x11/p5-AnyEvent-I3: Makefile x11/rxvt-unicode: Makefile x11/xclip : Makefile Log message: reset unresponsive maintainers, various porters have spent a bunch of time waiting for timeouts over the last year or more for many of these ports. ok danj@ jca@ kmos@
Re: [ports-gcc-8] Unbreak x11/rxvt-unicode
On Wed, Apr 24 2019, Charlene Wendling wrote: > Hi ports, > > It already happened with gcc-6, but i've a better fix now. > >> http://build-failures.rhaalovely.net/powerpc/2018-11-20/x11/rxvt-unicode.log > > I wasn't able to get why it happened with ports-gcc-8 and not > clang: as they share the same __cplusplus value, they should > have the same operator issues. Until i diffed 'make configure' > on amd64 and macppc to see the difference. > > I've found out that when using clang, rxvt-unicode is not built > with libsupc++. It's not the case with gcc, so i tried building > without libsupc++, and it works. I wonder if there is a eg++ bug > here. It's not an eg++ bug. Upstream wants to avoid libstdc++ and link against libsupc++ instead to get a smaller executable, etc. This is documented by the gcc folks: https://gcc.gnu.org/onlinedocs/libstdc++/faq.html#faq.what_is_libsupcxx But in ports-gcc the library is renamed: libsupc++ -> libesupc++. The diff below also fixes the build on sparc64, runtime doesn't seem affected. The diff doesn't change anything with clang on amd64 where 1. libesupc++ isn't in the library search path of ./configure 2. even with -L/usr/local/lib in LDFLAGS, the configure test also fails because of missing symbols Using -lesupc++ is probably what upstream would prefer because speed but the difference isn't impressive. Stupid benchmarks on sparc64 with DISPLAY unset: - cc + libesupc++ billy /usr/ports/x11/rxvt-unicode$ time for s in $(jot 100); do urxvt 2>/dev/null; done 0m13.23s real 0m00.10s user 0m01.12s system - c++ billy /usr/ports/x11/rxvt-unicode$ time for s in $(jot 100); do urxvt 2>/dev/null; done 0m15.15s real 0m00.07s user 0m01.28s system I'm leaning toward your solution since it reduces the difference between clang and "ports-gcc" archs. Your call, ok jca@ either way. PS: no need to wait for the usual 1 week timeout IMO Index: patches/patch-configure === RCS file: patches/patch-configure diff -N patches/patch-configure --- /dev/null 1 Jan 1970 00:00:00 - +++ patches/patch-configure 24 Apr 2019 15:34:51 - @@ -0,0 +1,16 @@ +$OpenBSD$ + +Try to use libesupc++ with ports-gcc, instead of libsupc++ from base-gcc. + +Index: configure +--- configure.orig configure +@@ -4499,7 +4499,7 @@ $as_echo_n "checking for working libsupc++... " >&6; } +save_CXX="$CXX" +save_LIBS="$LIBS" +CXX="$CC" +- LIBS="$LIBS -lsupc++" ++ LIBS="$LIBS -lesupc++" + + cat confdefs.h - <<_ACEOF >conftest.$ac_ext + /* end confdefs.h. */ -- jca | PGP : 0x1524E7EE / 5135 92C1 AD36 5293 2BDF DDCC 0DFA 74AE 1524 E7EE
Update: Ruby 2.6.2 -> 2.6.3
Simple update to the latest release of Ruby 2.6. Release announcement
at https://www.ruby-lang.org/en/news/2019/04/17/ruby-2-6-3-released/
Tested on amd64. Will be committing in a couple days unless I hear
objections.
Thanks,
Jeremy
Index: Makefile
===
RCS file: /cvs/ports/lang/ruby/2.6/Makefile,v
retrieving revision 1.3
diff -u -p -r1.3 Makefile
--- Makefile15 Mar 2019 16:44:49 - 1.3
+++ Makefile22 Apr 2019 23:20:20 -
@@ -1,6 +1,6 @@
# $OpenBSD: Makefile,v 1.3 2019/03/15 16:44:49 jeremy Exp $
-VERSION = 2.6.2
+VERSION = 2.6.3
DISTNAME = ruby-${VERSION}
SHARED_LIBS = ruby26 0.0
NEXTVER = 2.7
Index: distinfo
===
RCS file: /cvs/ports/lang/ruby/2.6/distinfo,v
retrieving revision 1.3
diff -u -p -r1.3 distinfo
--- distinfo15 Mar 2019 16:44:49 - 1.3
+++ distinfo22 Apr 2019 23:20:31 -
@@ -1,2 +1,2 @@
-SHA256 (ruby-2.6.2.tar.gz) = oEBdK/LC0vMyAztw3/NU0iSoZKsO3UYrekE0IEU7Sas=
-SIZE (ruby-2.6.2.tar.gz) = 1665
+SHA256 (ruby-2.6.3.tar.gz) = V3/TeV8iuNkcHU5nM2N7A5TUCC22WfzPIkx3Siscgvs=
+SIZE (ruby-2.6.3.tar.gz) = 16784748
Index: pkg/PLIST-main
===
RCS file: /cvs/ports/lang/ruby/2.6/pkg/PLIST-main,v
retrieving revision 1.3
diff -u -p -r1.3 PLIST-main
--- pkg/PLIST-main 15 Mar 2019 16:44:50 - 1.3
+++ pkg/PLIST-main 22 Apr 2019 23:33:08 -
@@ -42,7 +42,7 @@ include/ruby-${REV}/ruby/util.h
include/ruby-${REV}/ruby/version.h
include/ruby-${REV}/ruby/vm.h
include/ruby-${REV}/${SUB}/
-include/ruby-${REV}/${SUB}/rb_mjit_min_header-2.6.2.h
+include/ruby-${REV}/${SUB}/rb_mjit_min_header-2.6.3.h
include/ruby-${REV}/${SUB}/ruby/
include/ruby-${REV}/${SUB}/ruby/config.h
lib/libruby26.so
@@ -343,6 +343,7 @@ lib/ruby/${REV}/csv.rb
lib/ruby/${REV}/csv/core_ext/
lib/ruby/${REV}/csv/core_ext/array.rb
lib/ruby/${REV}/csv/core_ext/string.rb
+lib/ruby/${REV}/csv/delete_suffix.rb
lib/ruby/${REV}/csv/fields_converter.rb
lib/ruby/${REV}/csv/match_p.rb
lib/ruby/${REV}/csv/parser.rb
@@ -1674,7 +1675,7 @@ lib/ruby/gems/${REV}/specifications/defa
lib/ruby/gems/${REV}/specifications/default/bigdecimal-1.4.1.gemspec
lib/ruby/gems/${REV}/specifications/default/bundler-1.17.2.gemspec
lib/ruby/gems/${REV}/specifications/default/cmath-1.0.0.gemspec
-lib/ruby/gems/${REV}/specifications/default/csv-3.0.4.gemspec
+lib/ruby/gems/${REV}/specifications/default/csv-3.0.9.gemspec
lib/ruby/gems/${REV}/specifications/default/date-2.0.0.gemspec
lib/ruby/gems/${REV}/specifications/default/dbm-1.0.0.gemspec
lib/ruby/gems/${REV}/specifications/default/e2mmap-0.1.0.gemspec
Index: pkg/PLIST-ri_docs
===
RCS file: /cvs/ports/lang/ruby/2.6/pkg/PLIST-ri_docs,v
retrieving revision 1.3
diff -u -p -r1.3 PLIST-ri_docs
--- pkg/PLIST-ri_docs 15 Mar 2019 16:44:50 - 1.3
+++ pkg/PLIST-ri_docs 22 Apr 2019 23:33:49 -
@@ -3313,6 +3313,9 @@ share/ri/${REV}/system/CMath/tanh-c.ri
share/ri/${REV}/system/CMath/tanh-i.ri
share/ri/${REV}/system/CSV/
share/ri/${REV}/system/CSV/%3c%3c-i.ri
+share/ri/${REV}/system/CSV/DeleteSuffix/
+share/ri/${REV}/system/CSV/DeleteSuffix/cdesc-DeleteSuffix.ri
+share/ri/${REV}/system/CSV/DeleteSuffix/delete_suffix-i.ri
share/ri/${REV}/system/CSV/FieldsConverter/
share/ri/${REV}/system/CSV/FieldsConverter/add_converter-i.ri
share/ri/${REV}/system/CSV/FieldsConverter/cdesc-FieldsConverter.ri
@@ -,6 +3336,7 @@ share/ri/${REV}/system/CSV/MatchP/match%
share/ri/${REV}/system/CSV/Parser/
share/ri/${REV}/system/CSV/Parser/InputsScanner/
share/ri/${REV}/system/CSV/Parser/InputsScanner/cdesc-InputsScanner.ri
+share/ri/${REV}/system/CSV/Parser/InputsScanner/each_line-i.ri
share/ri/${REV}/system/CSV/Parser/InputsScanner/eos%3f-i.ri
share/ri/${REV}/system/CSV/Parser/InputsScanner/keep_back-i.ri
share/ri/${REV}/system/CSV/Parser/InputsScanner/keep_drop-i.ri
@@ -3347,6 +3351,7 @@ share/ri/${REV}/system/CSV/Parser/Invali
share/ri/${REV}/system/CSV/Parser/InvalidEncoding/cdesc-InvalidEncoding.ri
share/ri/${REV}/system/CSV/Parser/Scanner/
share/ri/${REV}/system/CSV/Parser/Scanner/cdesc-Scanner.ri
+share/ri/${REV}/system/CSV/Parser/Scanner/each_line-i.ri
share/ri/${REV}/system/CSV/Parser/Scanner/keep_back-i.ri
share/ri/${REV}/system/CSV/Parser/Scanner/keep_drop-i.ri
share/ri/${REV}/system/CSV/Parser/Scanner/keep_end-i.ri
@@ -3354,6 +3359,7 @@ share/ri/${REV}/system/CSV/Parser/Scanne
share/ri/${REV}/system/CSV/Parser/Scanner/new-c.ri
share/ri/${REV}/system/CSV/Parser/UnoptimizedStringIO/
share/ri/${REV}/system/CSV/Parser/UnoptimizedStringIO/cdesc-UnoptimizedStringIO.ri
+share/ri/${REV}/system/CSV/Parser/UnoptimizedStringIO/each_line-i.ri
share/ri/${REV}/system/CSV/Parser/UnoptimizedStringIO/eof%3f-i.ri
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: ste...@cvs.openbsd.org 2019/04/24 09:30:31 Modified files: math/lapack: Makefile distinfo math/lapack/files: Makefile math/lapack/pkg: PLIST Log message: update to 3.8.0
nginx 1.16
hi,
somewhat trivial update, seems to run fine here.
https://nginx.org/en/CHANGES-1.16 for the changes, some of the modules
*could* probably be upgraded too.
Landry
Index: Makefile
===
RCS file: /cvs/ports/www/nginx/Makefile,v
retrieving revision 1.130
diff -u -r1.130 Makefile
--- Makefile23 Apr 2019 06:20:37 - 1.130
+++ Makefile24 Apr 2019 15:19:06 -
@@ -15,7 +15,7 @@
COMMENT-perl= nginx perl scripting module
COMMENT-passenger= nginx passenger (ruby/python/nodejs) integration module
-VERSION= 1.14.2
+VERSION= 1.16.0
DISTNAME= nginx-${VERSION}
CATEGORIES=www
@@ -32,8 +32,6 @@
PKGNAME-perl= nginx-perl-${VERSION}
PKGNAME-passenger= nginx-passenger-${VERSION}
-REVISION-main= 0
-
MASTER_SITES= https://nginx.org/download/
MASTER_SITES0= https://github.com/simpl/ngx_devel_kit/archive/
MASTER_SITES1= https://github.com/nbs-system/naxsi/archive/
@@ -73,7 +71,7 @@
WANTLIB-geoip= GeoIP
WANTLIB-xslt= exslt xml2 xslt
WANTLIB-naxsi=
-WANTLIB-ldap_auth= ldap
+WANTLIB-ldap_auth= ldap-2.4
WANTLIB-lua= ${MODLUA_WANTLIB} m
WANTLIB-headers_more=
WANTLIB-perl= c m perl
Index: distinfo
===
RCS file: /cvs/ports/www/nginx/distinfo,v
retrieving revision 1.68
diff -u -r1.68 distinfo
--- distinfo6 Dec 2018 08:50:54 - 1.68
+++ distinfo24 Apr 2019 15:19:06 -
@@ -2,13 +2,13 @@
SHA256 (lua-nginx-module-v0.10.11.tar.gz) =
wPuR/P0cbn3sNMpkgm74H/66/e9hdNJURnY284BWZiY=
SHA256 (naxsi-0.55.3.tar.gz) = CzyV0lB3Lcia2LSeR8HgJMWuLHbAz/pEXp/gXE3RNJU=
SHA256 (nginx-1.14.0-chroot.patch) =
6dERcspRpgEau5QbXHC+K0r5C9Ogy/df6j8BpYrStL0=
-SHA256 (nginx-1.14.2.tar.gz) = AC2fYVTjMYhqLdTmBlhjycHPgpGul6ElUwhXLAK+l5c=
+SHA256 (nginx-1.16.0.tar.gz) = T9N2uteHl+fxgJSgDw8QiCWTJkNrU361r2mwG+LKE0U=
SHA256 (nginx-auth-ldap-0.20170725.tar.gz) =
gNbM6amHfVHewvhaEc580l7b0tYFwovChofsxWlSKe4=
SHA256 (ngx_devel_kit-v0.3.0.tar.gz) =
iOBamainQZBm9a51lm+x78QJutRSLRSYbaB0VUrmFhk=
SIZE (headers-more-nginx-module-v0.33.tar.gz) = 28130
SIZE (lua-nginx-module-v0.10.11.tar.gz) = 616653
SIZE (naxsi-0.55.3.tar.gz) = 187416
SIZE (nginx-1.14.0-chroot.patch) = 8220
-SIZE (nginx-1.14.2.tar.gz) = 1015384
+SIZE (nginx-1.16.0.tar.gz) = 1032345
SIZE (nginx-auth-ldap-0.20170725.tar.gz) = 18457
SIZE (ngx_devel_kit-v0.3.0.tar.gz) = 66455
Index: patches/patch-conf_mime_types
===
RCS file: /cvs/ports/www/nginx/patches/patch-conf_mime_types,v
retrieving revision 1.2
diff -u -r1.2 patch-conf_mime_types
--- patches/patch-conf_mime_types 28 Apr 2018 15:58:59 - 1.2
+++ patches/patch-conf_mime_types 24 Apr 2019 15:19:06 -
@@ -2,7 +2,7 @@
Index: conf/mime.types
--- conf/mime.types.orig
+++ conf/mime.types
-@@ -38,22 +38,30 @@ types {
+@@ -40,22 +40,30 @@ types {
application/vnd.ms-excel xls;
application/vnd.ms-fontobjecteot;
application/vnd.ms-powerpointppt;
@@ -39,7 +39,7 @@
application/x-perl pl pm;
application/x-pilot prc pdb;
application/x-rar-compressed rar;
-@@ -65,7 +73,6 @@ types {
+@@ -67,7 +75,6 @@ types {
application/x-x509-ca-cert der pem crt;
application/x-xpinstall xpi;
application/xhtml+xmlxhtml;
@@ -47,7 +47,7 @@
application/zip zip;
application/octet-stream bin exe dll;
-@@ -74,6 +81,7 @@ types {
+@@ -76,6 +83,7 @@ types {
application/octet-stream iso img;
application/octet-stream msi msp msm;
Index: pkg/PLIST-main
===
RCS file: /cvs/ports/www/nginx/pkg/PLIST-main,v
retrieving revision 1.2
diff -u -r1.2 PLIST-main
--- pkg/PLIST-main 4 Sep 2018 12:46:25 - 1.2
+++ pkg/PLIST-main 24 Apr 2019 15:19:06 -
@@ -2,6 +2,7 @@
@pkgpath www/nginx/stable
@pkgpath www/nginx/devel
@pkgpath www/nginx
+@rcscript ${RCDIR}/nginx
@man man/man8/nginx.8
@bin sbin/nginx
@mode 700
@@ -36,4 +37,3 @@
@sample ${SYSCONFDIR}/nginx/uwsgi_params
share/nginx/win-utf
@sample ${SYSCONFDIR}/nginx/win-utf
-@rcscript ${RCDIR}/nginx
Index: pkg/PLIST-naxsi
===
RCS file: /cvs/ports/www/nginx/pkg/PLIST-naxsi,v
retrieving revision 1.1
diff -u -r1.1 PLIST-naxsi
--- pkg/PLIST-naxsi 4 Nov 2016 22:13:43 - 1.1
+++ pkg/PLIST-naxsi 24 Apr 2019 15:19:06 -
@@ -1,5 +1,5 @@
@comment $OpenBSD: PLIST-naxsi,v 1.1 2016/11/04 22:13:43 jeremy Exp $
[update] games/taisei 1.0a -> 1.2
Here is an update for taisei. Our version is 7 years old, and many things changed upstream: - All patches are upstreamed, it uses C11 now - Moved from cmake to meson - It uses SDL2 instead of SDL (openal has been dropped) - Gameplay changes can be found there [0] It's almost a new port: - Requires OpenGL>=3.3 or OpenGL ES>=3.0, so only for "modern" architectures that may have compatible video controllers afaik. i386 has been added in case an amd64 machine runs an i386 install. - GH_ACCOUNT changed to a dedicated account - move HOMEPAGE to https - requires python for build only - regenerated LIB_DEPENDS/WANTLIB Testing (amd64 only): I have some glitches/freezes with my AMD Radeon R5 230 video card that don't happen on my laptop, that has an Intel HD4000 one. Otherwise, audio plays, my gamepad (a logitech dual action) works properly and i've met no issues playing the game (i'm still not able to finish it, it's meant to be hard). Comments/feedback are welcome :) [0] https://taisei-project.org/news taisei.diff Description: Binary data
Re: [NEW PORT]: rr
I fixed usage (missed -n), retagged as 1.1 New tarball. - marked as only for clang archs (until the gcc 8 switch) - obey LDFLAGS/CXXFLAGS trivially - rename the package itself to random_run. I think keeping the short name for the command (rr) makes sense for now... I don't see mozilla's rr being ported any time soon. > > As a result, I've written a new (smallish) command called rr: > --- > > rr (random run) runs a command with randomized arguments. > > Basically, > > $ rr mpv *.mp4 > will give you a true shuffle, that doesn't play the same file twice. > > there are more options: > > $ rr xv -nolimit -- * > allows you to specify options that won't be jumbled. > > $ rr -n 500 -x '.*.pdf' xv -nolimit -- * > will run xv several times on all files in the current directory > except those matching '.*.pdf', limiting each run to 500 parameters. > > -- > > Port attached random_run-1.1.tgz Description: random_run-1.1.tgz
Re: [NEW PORT]: rr
On Wed, Apr 24, 2019 at 04:30:23PM +0200, Jeremie Courreges-Anglas wrote:
> On Wed, Apr 24 2019, Marc Espie wrote:
> > I've been playing a bit with new C++, and to host stuff on github.
> >
> > As a result, I've written a new (smallish) command called rr:
> > ---
> >
> > rr (random run) runs a command with randomized arguments.
> >
> > Basically,
> >
> > $ rr mpv *.mp4
> > will give you a true shuffle, that doesn't play the same file twice.
> >
> > there are more options:
> >
> > $ rr xv -nolimit -- *
> > allows you to specify options that won't be jumbled.
> >
> > $ rr -n 500 -x '.*.pdf' xv -nolimit -- *
> > will run xv several times on all files in the current directory
> > except those matching '.*.pdf', limiting each run to 500 parameters.
> >
> > --
> >
> > Port attached
> >
> > # $OpenBSD$
> >
> > COMMENT = run command with randomized parameters
> > PKGNAME = rr-${GH_TAGNAME}
> > CATEGORIES =sysutils
> > MAINTAINER =Marc Espie
> >
> > GH_PROJECT =random_run
> > GH_TAGNAME =1.0
> > GH_ACCOUNT =marcespie
> >
> > COMPILER = base-clang ports-gcc
>
> Doesn't build with current ports-gcc (4.9.4) since it doesn't know about
> C++17. Builds fine with gcc-8.3.0 though.
So it's currently BROKEN for those arches, but not for long.
> The port doesn't respect CXXFLAGS/DEBUG/LDFLAGS. Suggestion:
>
> MAKE_FLAGS = OPTIMIZE="${CXXFLAGS}" \
> LDFLAGS="${LDFLAGS}"
Should work.
Re: [NEW PORT]: rr
On Wed, Apr 24, 2019 at 10:48:02AM -0400, Ted Unangst wrote: > Marc Espie wrote: > > I've been playing a bit with new C++, and to host stuff on github. > > > > As a result, I've written a new (smallish) command called rr: > > isn't rr the name of the mozilla reverse debugger? If I saw a pkg called rr, > that's what I would be expecting to get. > > https://github.com/mozilla/rr Basically, we don't have this in the ports tree. I consider the point moot unless somebody ports it. It's mostly impossible to write anything without stepping on another command's toes.
Re: [NEW PORT]: rr
On Wed, Apr 24, 2019 at 10:48:02AM -0400, Ted Unangst wrote: > Marc Espie wrote: > > I've been playing a bit with new C++, and to host stuff on github. > > > > As a result, I've written a new (smallish) command called rr: > > isn't rr the name of the mozilla reverse debugger? If I saw a pkg called rr, > that's what I would be expecting to get. > > https://github.com/mozilla/rr same for me, i had a "wtf espie ported rr !?" moment :)
Re: [NEW PORT]: rr
Marc Espie wrote: > I've been playing a bit with new C++, and to host stuff on github. > > As a result, I've written a new (smallish) command called rr: isn't rr the name of the mozilla reverse debugger? If I saw a pkg called rr, that's what I would be expecting to get. https://github.com/mozilla/rr
Re: [NEW PORT]: rr
On Wed, Apr 24 2019, Marc Espie wrote:
> I've been playing a bit with new C++, and to host stuff on github.
>
> As a result, I've written a new (smallish) command called rr:
> ---
>
> rr (random run) runs a command with randomized arguments.
>
> Basically,
>
> $ rr mpv *.mp4
> will give you a true shuffle, that doesn't play the same file twice.
>
> there are more options:
>
> $ rr xv -nolimit -- *
> allows you to specify options that won't be jumbled.
>
> $ rr -n 500 -x '.*.pdf' xv -nolimit -- *
> will run xv several times on all files in the current directory
> except those matching '.*.pdf', limiting each run to 500 parameters.
>
> --
>
> Port attached
>
> # $OpenBSD$
>
> COMMENT = run command with randomized parameters
> PKGNAME = rr-${GH_TAGNAME}
> CATEGORIES = sysutils
> MAINTAINER = Marc Espie
>
> GH_PROJECT = random_run
> GH_TAGNAME = 1.0
> GH_ACCOUNT = marcespie
>
> COMPILER =base-clang ports-gcc
Doesn't build with current ports-gcc (4.9.4) since it doesn't know about
C++17. Builds fine with gcc-8.3.0 though.
The port doesn't respect CXXFLAGS/DEBUG/LDFLAGS. Suggestion:
MAKE_FLAGS =OPTIMIZE="${CXXFLAGS}" \
LDFLAGS="${LDFLAGS}"
> # BSD licence
> PERMIT_PACKAGE_CDROM = Yes
> NO_TESTS =Yes
>
> WANTLIB = ${COMPILER_LIBCXX} c m
>
> do-install:
> ${INSTALL_PROGRAM} ${WRKBUILD}/rr ${PREFIX}/bin
> ${INSTALL_MAN} ${WRKSRC}/rr.1 ${PREFIX}/man/man1
>
>
> .include
> SHA256 (random_run-1.0.tar.gz) = HHcgSwtFu0/87XVkVu297fX0Q8eGnF5+dPIoB3ZyxPM=
> SIZE (random_run-1.0.tar.gz) = 3463
> rr (random run) runs a command with randomized arguments.
>
> Basically,
>
> $ rr mpv *.mp4
> will give you a true shuffle, that doesn't play the same file twice.
>
> there are more options:
>
> $ rr xv -nolimit -- *
> allows you to specify options that won't be jumbled.
>
> $ rr -n 500 -x '.*.pdf' xv -nolimit -- *
> will run xv several times on all files in the current directory
> except those matching '.*.pdf', limiting each run to 500 parameters.
>
> See the manpage for details
> @comment $OpenBSD: PLIST,v$
> @bin bin/rr
> @man man/man1/rr.1
>
--
jca | PGP : 0x1524E7EE / 5135 92C1 AD36 5293 2BDF DDCC 0DFA 74AE 1524 E7EE
[NEW PORT]: rr
I've been playing a bit with new C++, and to host stuff on github. As a result, I've written a new (smallish) command called rr: --- rr (random run) runs a command with randomized arguments. Basically, $ rr mpv *.mp4 will give you a true shuffle, that doesn't play the same file twice. there are more options: $ rr xv -nolimit -- * allows you to specify options that won't be jumbled. $ rr -n 500 -x '.*.pdf' xv -nolimit -- * will run xv several times on all files in the current directory except those matching '.*.pdf', limiting each run to 500 parameters. -- Port attached rr.tgz Description: rr-1.0.tgz
Re: [UPDATE]: sysutils/opam -> 2.0.4
Ping!
On Sat, Apr 13, 2019 at 23:50, wrote:
> ok krw@, but I assume for after the ports tree is unlocked?
>
> Ken
>
>> On Apr 13, 2019, at 6:53 AM, Christopher Zimmermann
>> wrote:
>>
>> Hello,
>>
>> this is a very simple update. OK?
>>
>> Christopher
>>
>>
>> Index: Makefile
>> ===
>> RCS file: /cvs/ports/sysutils/opam/Makefile,v
>> retrieving revision 1.11
>> diff -u -p -r1.11 Makefile
>> --- Makefile 12 Mar 2019 05:36:06 - 1.11
>> +++ Makefile 13 Apr 2019 13:49:57 -
>> @@ -4,8 +4,7 @@ COMMENT = OCaml source-based package ma
>>
>> CATEGORIES = sysutils devel
>>
>> -V = 2.0.3
>> -REVISION = 0
>> +V = 2.0.4
>> PKGNAME = opam-${V}
>> DISTNAME = opam-full-${V}
>>
>> Index: distinfo
>> ===
>> RCS file: /cvs/ports/sysutils/opam/distinfo,v
>> retrieving revision 1.4
>> diff -u -p -r1.4 distinfo
>> --- distinfo 12 Mar 2019 05:36:06 - 1.4
>> +++ distinfo 13 Apr 2019 13:49:57 -
>> @@ -1,2 +1,2 @@
>> -SHA256 (opam-full-2.0.3.tar.gz) =
>> BYnaTaGEWEpURdWThQCVNlNPYLwOJ3ciRbL0nl+o8OI=
>> -SIZE (opam-full-2.0.3.tar.gz) = 7870020
>> +SHA256 (opam-full-2.0.4.tar.gz) =
>> 3r+4KLQA+1EcopDxv8ko25HK107BzL3c/b/v8m9wmeU=
>> +SIZE (opam-full-2.0.4.tar.gz) = 7868547
>>
>>
>> --
>> http://gmerlin.de
>> OpenPGP: http://gmerlin.de/christopher.pub
>> CB07 DA40 B0B6 571D 35E2 0DEF 87E2 92A7 13E5 DEE1
[ports-gcc-8] Unbreak x11/rxvt-unicode
Hi ports,
It already happened with gcc-6, but i've a better fix now.
> http://build-failures.rhaalovely.net/powerpc/2018-11-20/x11/rxvt-unicode.log
I wasn't able to get why it happened with ports-gcc-8 and not
clang: as they share the same __cplusplus value, they should
have the same operator issues. Until i diffed 'make configure'
on amd64 and macppc to see the difference.
I've found out that when using clang, rxvt-unicode is not built
with libsupc++. It's not the case with gcc, so i tried building
without libsupc++, and it works. I wonder if there is a eg++ bug
here.
It appears that FreeBSD did the same exact thing before me for
other reasons [0].
It has been tested, with Perl matcher module activated, on:
- macppc/ports-gcc-8
- macppc/ports-gcc-4.9
- amd64/base-clang (it's supposed to be harmless there anyway)
Comments/feedback are welcome!
Charlène.
[0]
https://svnweb.freebsd.org/ports/head/x11/rxvt-unicode/files/patch-configure?revision=357486=markup
Index: Makefile
===
RCS file: /cvs/ports/x11/rxvt-unicode/Makefile,v
retrieving revision 1.44
diff -u -p -u -p -r1.44 Makefile
--- Makefile24 Oct 2018 14:28:13 - 1.44
+++ Makefile23 Apr 2019 22:07:59 -
@@ -3,9 +3,9 @@
COMMENT = clone of rxvt with Unicode and Xft support
DISTNAME = rxvt-unicode-9.22
-REVISION = 7
+REVISION = 8
CATEGORIES = x11
-FIX_EXTRACT_PERMISSIONS=Yes
+FIX_EXTRACT_PERMISSIONS= Yes
EXTRACT_SUFX = .tar.bz2
MASTER_SITES = http://dist.schmorp.de/rxvt-unicode/ \
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 23 Apr 2019 22:08:00 -
@@ -0,0 +1,52 @@
+$OpenBSD$
+
+Don't use gcc specific flags, because it causes various undefined
+reference errors with ports-gcc>=6 when -lsupc++ is used.
+
+Index: configure
+--- configure.orig
configure
+@@ -4493,43 +4493,6 @@ ac_compiler_gnu=$ac_cv_cxx_compiler_gnu
+
+
+ LINKER="$CXX"
+-if test x$GCC = xyes && test x$GXX = xyes; then
+- { $as_echo "$as_me:${as_lineno-$LINENO}: checking for working
libsupc++" >&5
+-$as_echo_n "checking for working libsupc++... " >&6; }
+- save_CXX="$CXX"
+- save_LIBS="$LIBS"
+- CXX="$CC"
+- LIBS="$LIBS -lsupc++"
+-
+-cat confdefs.h - <<_ACEOF >conftest.$ac_ext
+-/* end confdefs.h. */
+-struct test { }; void f() try { throw new test; } catch (...) { throw; }
+-int
+-main ()
+-{
+-
+- ;
+- return 0;
+-}
+-_ACEOF
+-if ac_fn_cxx_try_link "$LINENO"; then :
+- { $as_echo "$as_me:${as_lineno-$LINENO}: result: ok" >&5
+-$as_echo "ok" >&6; }
+- LINKER="$CC"
+-else
+-
+- { $as_echo "$as_me:${as_lineno-$LINENO}: result: no, making
everything bigger and slower" >&5
+-$as_echo "no, making everything bigger and slower" >&6; }
+- LIBS="$save_LIBS"
+-
+-
+-fi
+-rm -f core conftest.err conftest.$ac_objext \
+-conftest$ac_exeext conftest.$ac_ext
+- CXX="$save_CXX"
+-fi
+-LINKER=$LINKER
+-
+
+ if test -z "$orig_CXXFLAGS"; then
+ if test x$GCC = xyes && test "x$GXX" = xyes; then
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: st...@cvs.openbsd.org 2019/04/24 04:12:55 Modified files: archivers/lz4 : Makefile distinfo archivers/lz4/patches: patch-lib_Makefile Log message: update to lz4-1.9.1
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: rob...@cvs.openbsd.org 2019/04/24 03:56:33 Modified files: x11/mate/screensaver: Makefile distinfo Log message: update to 1.22.1
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: rob...@cvs.openbsd.org 2019/04/24 03:31:51 Modified files: x11/mate/pluma : Makefile distinfo x11/mate/pluma/pkg: PLIST Log message: update to 1.22.1
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: st...@cvs.openbsd.org 2019/04/24 02:48:23 Modified files: multimedia/gstreamer1/plugins-base: Makefile Log message: Add missing ldep on graphics/jpeg to fix "Missing library for jpeg>=0.0", this was previously pulled in indirectly via gtk+3 BUILD_DEPENDS which was removed (at least ->gdk-pixbuf->tiff->jpeg)
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: ajacou...@cvs.openbsd.org 2019/04/24 02:42:12 Modified files: meta/gnome : Makefile Log message: Update to gnome-3.30.2.3 (sync with gnome-desktop).
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: ajacou...@cvs.openbsd.org 2019/04/24 02:41:43 Modified files: x11/gnome/desktop: Makefile distinfo Log message: Update to gnome-desktop-3.30.2.3.
Re: py-gevent update
On 2019/04/23 18:16, Kurt Mosiejczuk wrote:
> On Tue, Apr 23, 2019 at 01:52:24PM +0100, Stuart Henderson wrote:
> > Unbreaks with python 3.7. No problems seen with dep's. OK?
>
> Needs a NO_TEST. Tests aren't bundled in with the new tarball.
>
> Otherwise it builds fine on arm64, sparc64, and amd64 here.
>
> With NO_TEST, OK kmos@
>
> --Kurt
>
Thanks - this is what I'll commit if I don't hear objections.
Index: Makefile
===
RCS file: /cvs/ports/devel/py-gevent/Makefile,v
retrieving revision 1.15
diff -u -p -r1.15 Makefile
--- Makefile14 Sep 2017 13:00:54 - 1.15
+++ Makefile24 Apr 2019 08:31:39 -
@@ -2,11 +2,10 @@
COMMENT = network library for easy and scalable concurrency
-MODPY_EGG_VERSION =1.2.2
+MODPY_EGG_VERSION =1.4.0
DISTNAME = gevent-${MODPY_EGG_VERSION}
PKGNAME = ${MODPY_PY_PREFIX}${DISTNAME}
MAINTAINER = Dmitrij D. Czarkoff
-REVISION = 1
CATEGORIES = devel
HOMEPAGE = http://www.gevent.org/
@@ -32,15 +31,13 @@ BUILD_DEPENDS = devel/libev:configure \
LIB_DEPENDS = devel/libev \
net/libcares
RUN_DEPENDS = devel/py-greenlet${MODPY_FLAVOR}
-TEST_DEPENDS = ${RUN_DEPENDS} \
- lang/python/${MODPY_VERSION},-tests \
- sysutils/py-psutil${MODPY_FLAVOR}
MAKE_ENV = CARES_EMBED=0 LIBEV_EMBED=0
-TEST_ENV = PYTHONPATH="${WRKBUILD}"
CFLAGS += -I${LOCALBASE}/include
LDFLAGS += -L${LOCALBASE}/lib
+NO_TEST = Yes
+
WRKBUILD = ${WRKSRC}/src
pre-patch:
@@ -51,9 +48,5 @@ post-build:
post-install:
rm -Rf ${PREFIX}/lib/python${MODPY_VERSION}/site-packages/greentest
-
-do-test:
- cd ${WRKBUILD}/greentest && ${SETENV} ${ALL_TEST_ENV} ${MODPY_BIN} \
- testrunner.py --config known_failures.py
.include
Index: distinfo
===
RCS file: /cvs/ports/devel/py-gevent/distinfo,v
retrieving revision 1.7
diff -u -p -r1.7 distinfo
--- distinfo13 Jun 2017 07:46:44 - 1.7
+++ distinfo24 Apr 2019 08:31:39 -
@@ -1,2 +1,2 @@
-SHA256 (gevent-1.2.2.tar.gz) = R5HIrpxX1vFTNUc24cyrHiuvbI2a5ad6mskPQeKWay0=
-SIZE (gevent-1.2.2.tar.gz) = 3108671
+SHA256 (gevent-1.4.0.tar.gz) = Hrf6O5vZF03+nDtZt6Cbdo7NSW3r/El2qVMKPhXJkNE=
+SIZE (gevent-1.4.0.tar.gz) = 5169595
Index: patches/patch-setup_py
===
RCS file: /cvs/ports/devel/py-gevent/patches/patch-setup_py,v
retrieving revision 1.3
diff -u -p -r1.3 patch-setup_py
--- patches/patch-setup_py 18 Jan 2017 12:42:28 - 1.3
+++ patches/patch-setup_py 24 Apr 2019 08:31:39 -
@@ -1,17 +1,21 @@
-$OpenBSD: patch-setup_py,v 1.3 2017/01/18 12:42:28 czarkoff Exp $
setup.py.orig Mon Jan 16 15:05:24 2017
-+++ setup.py Mon Jan 16 15:06:37 2017
-@@ -117,13 +117,6 @@ if ((len(sys.argv) >= 2
+$OpenBSD$
+
+Index: setup.py
+--- setup.py.orig
setup.py
+@@ -316,15 +316,6 @@ def make_long_description():
+
def run_setup(ext_modules, run_make):
- if run_make:
+-if run_make:
-if (not LIBEV_EMBED and not WIN and cffi_modules) or PYPY:
-# We're not embedding libev but we do want
-# to build the CFFI module. We need to configure libev
-# because the CORE Extension won't.
-# TODO: Generalize this.
--system(libev_configure_command)
+-if LIBEV_CFFI_MODULE in cffi_modules and not WIN:
+-system(libev_configure_command)
-
- MakeSdist.make()
-
setup(
+ name='gevent',
+ version=__version__,
Index: pkg/PLIST
===
RCS file: /cvs/ports/devel/py-gevent/pkg/PLIST,v
retrieving revision 1.6
diff -u -p -r1.6 PLIST
--- pkg/PLIST 14 Sep 2017 13:00:54 - 1.6
+++ pkg/PLIST 24 Apr 2019 08:31:39 -
@@ -4,6 +4,7 @@ lib/python${MODPY_VERSION}/site-packages
lib/python${MODPY_VERSION}/site-packages/gevent-${MODPY_EGG_VERSION}-py${MODPY_VERSION}.egg-info/PKG-INFO
lib/python${MODPY_VERSION}/site-packages/gevent-${MODPY_EGG_VERSION}-py${MODPY_VERSION}.egg-info/SOURCES.txt
lib/python${MODPY_VERSION}/site-packages/gevent-${MODPY_EGG_VERSION}-py${MODPY_VERSION}.egg-info/dependency_links.txt
+lib/python${MODPY_VERSION}/site-packages/gevent-${MODPY_EGG_VERSION}-py${MODPY_VERSION}.egg-info/entry_points.txt
lib/python${MODPY_VERSION}/site-packages/gevent-${MODPY_EGG_VERSION}-py${MODPY_VERSION}.egg-info/not-zip-safe
lib/python${MODPY_VERSION}/site-packages/gevent-${MODPY_EGG_VERSION}-py${MODPY_VERSION}.egg-info/requires.txt
lib/python${MODPY_VERSION}/site-packages/gevent-${MODPY_EGG_VERSION}-py${MODPY_VERSION}.egg-info/top_level.txt
@@ -11,15
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: lan...@cvs.openbsd.org 2019/04/24 02:14:17 Modified files: geo/pdal : Makefile distinfo geo/pdal/patches: patch-cmake_macros_cmake geo/pdal/pkg : PLIST Added files: geo/pdal/patches: patch-vendor_arbiter_arbiter_hpp Removed files: geo/pdal/patches: patch-cmake_modules_FindGEOS_cmake patch-dimbuilder_CMakeLists_txt patch-pdal_PluginDirectory_cpp patch-pdal_util_CMakeLists_txt patch-pdal_util_Utils_cpp patch-plugins_sqlite_io_SQLiteCommon_hpp patch-test_unit_PluginManagerTest_cpp Log message: Update to pdal 1.9.0. Should build fine with python 3.7.
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: bent...@cvs.openbsd.org 2019/04/24 02:04:06 Modified files: www/youtube-dl : Makefile distinfo www/youtube-dl/pkg: PLIST Log message: Update to youtube-dl-2019.04.24.
