Re: powerpc bulk build report

[Charlene Wendling]

It's not done yet ;)

One of the machines lost some nfs mounts, i'm restarting the bulk atm ;)

On Fri, 15 Oct 2021 09:13:44 -0600 (MDT)
c...@openbsd.org wrote:

> Bulk build on macppc-0.ports.openbsd.org
> 
> Started : Tue Sep 28 01:49:06 MDT 2021
> Finished: Fri Oct 15 09:12:08 MDT 2021
> Duration: 17 Days 7 hours 23 minutes
> 
> Built using OpenBSD 7.0 (GENERIC) #939: Mon Sep 27 18:51:05 MDT 2021
> 

Re: NEW: games/vvvvvv - a retro platformer with gravity-reversal puzzle mechanics

[Charlene Wendling]

On Mon, 4 Oct 2021 23:55:06 -0600
Thomas Frohwein  wrote:

> On Mon, Oct 04, 2021 at 07:32:56PM +0000, Charlene Wendling wrote:
> > Hi,
> > 
> > 
> > > There is an existing port draft by bcallah@ and cwen@ in
> > > openbsd-wip [2]. I've graciously adopted the patches to the
> > > bundled physfs for big endian compatibility and to use
> > > ~/.local/share/VV/ for the assets file data.zip.
> > 
> > Actually, that patch may not be needed, because VV does not
> > support big endian archs [0], as i've found out later. Yet i've
> > built your port on macppc, and indeed colors are off, most sprites
> > are not displayed, so i think it needs a 'NOT_FOR_ARCHS=$
> > {BE_ARCHS}' at least.
> > 
> > > Some comments on where I diverge from the version in openbsd-wip:
> > > [...]
> > > - Compiles with std=c++98, so looks to me like base-gcc can be
> > > included in COMPILER...
> > 
> > Due to [0], it won't be built on sparc64, the last GCC arch
> > we build packages for, so i think the COMPILER line can safely be
> > removed.
> > 
> > > [...]
> > 
> > Charlène.
> > 
> > [0] https://github.com/TerryCavanagh/VV/pull/461
> > 
> 
> Thanks, that makes sense. I added the NOT_FOR_ARCHS and removed the
> patch.
> 
> I left the COMPILER line with the comment as it's C++, in case that
> will be relevant in the future. I think if there's a decision at some
> point that this isn't needed with sparc64 in NOT_FOR_ARCHS, it can be
> addressed then IMO. (Comments welcome on what direction this topic is
> going...)
> 
> I've added an updated tarball; I tested it again with the free assets
> and still works. ok?


OK cwen@ port-wise; it would be nice if someone is OK with the
licensing terms though. 

Re: NEW: games/vvvvvv - a retro platformer with gravity-reversal puzzle mechanics

[Charlene Wendling]

Hi,


> There is an existing port draft by bcallah@ and cwen@ in openbsd-wip
> [2]. I've graciously adopted the patches to the bundled physfs for
> big endian compatibility and to use ~/.local/share/VV/ for the
> assets file data.zip.

Actually, that patch may not be needed, because VV does not support
big endian archs [0], as i've found out later. Yet i've built your port
on macppc, and indeed colors are off, most sprites are not displayed, so
i think it needs a 'NOT_FOR_ARCHS=${BE_ARCHS}' at least.

> Some comments on where I diverge from the version in openbsd-wip:
> [...]
> - Compiles with std=c++98, so looks to me like base-gcc can be
> included in COMPILER...

Due to [0], it won't be built on sparc64, the last GCC arch
we build packages for, so i think the COMPILER line can safely be
removed.

> [...]

Charlène.

[0] https://github.com/TerryCavanagh/VV/pull/461

Re: powerpc bulk build report

[Charlene Wendling]

Hi,

On Thu, 9 Sep 2021 22:27:27 +0200
Alexander Bluhm  wrote:

> On Thu, Sep 09, 2021 at 08:56:52AM -0600, c...@openbsd.org wrote:
> > http://build-failures.rhaalovely.net/powerpc/2021-08-21/misc/open62541,ns0_full.log
> 
> /usr/obj/ports/open62541-1.0.6-ns0_full/bin/cc ...
> namespace0_generated.c Segmentation fault (core dumped) 
> ninja: build stopped: subcommand failed.
> 
> The compiler needs a lot of stack memory.  The ulimit -s should be
> raised to 8 GB.  Feel free to skip the ns0_full flavor on small
> memory architectures.

I was going to _very_ sloppily mark the flavor NOT_FOR_ARCHS, until
sthen noticed that 8 GB of stack memory is a bit too much.

I found out after some research that i was using an old login.conf,
and definitely have said no to sysmerge at one point. Sigh.

Using the latest login.conf, where the stacksize is raised to 8MB by
default for pbuild, fixes the build at home, it should be fine on the
bulk machines as well.

Thanks for your feedback :)

Charlène.

> bluhm
> 

Re: powerpc bulk build report

[Charlene Wendling]

Hi,

This is the first bulk done with LLD as the default linker.

Please test your favorite packages and report issues if you have some :)

On Thu, 9 Sep 2021 08:56:52 -0600 (MDT)
c...@openbsd.org wrote:

> Bulk build on macppc-0.ports.openbsd.org
> 
> Started : Sat Aug 21 14:27:57 MDT 2021
> Finished: Thu Sep  9 08:56:41 MDT 2021
> Duration: 18 Days 18 hours 29 minutes

It looks like LLD is making bulks a bit faster (actually macppc-1 was
down for 2 days, hence the extra delay)

> Built using OpenBSD 7.0-beta (GENERIC) #926: Fri Aug 20 18:11:14 MDT
> 2021
> 
> Built 9517 packages
> [...]
> 
> Critical path missing pkgs:
> http://build-failures.rhaalovely.net/powerpc/2021-08-21/summary.log
> 
> Build failures: 10

> http://build-failures.rhaalovely.net/powerpc/2021-08-21/devel/xtensa-lx106-elf/gdb.log

gkoehler@ is already providing a diff for libgmp:

https://marc.info/?l=openbsd-ports&m=163116370903235&w=2

> http://build-failures.rhaalovely.net/powerpc/2021-08-21/mail/rspamd,hyperscan.log

rspamd is not built on 32 bit archs for now.

> http://build-failures.rhaalovely.net/powerpc/2021-08-21/misc/open62541,ns0_full.log

This one was already present during the previous bulk, but no report has
been sent, it's not due to LLD.

Re: [macppc / lld] devel/gmp text relocation

[Charlene Wendling]

On Thu, 9 Sep 2021 16:15:53 +0200
Christian Weisgerber  wrote:

> George Koehler:
> 
> > WRKSRC/mpn/asm-defs.m4:1054,
> > 
> > ifelse(`PIC_ALWAYS',`yes',`define(`PIC')')
> > 
> > Notice that `PIC_ALWAYS' is in quotes, so the macro PIC_ALWAYS is
> > not expanded.  This faulty check compares the literal string
> > "PIC_ALWAYS" with the literal string "yes"; these strings never
> > match, so this faulty check never defines PIC.  I don't fix this
> > faulty check, because it might affect other archs.
> 
> It appears that check is only used on powerpc*.  The only place I
> can find that defines PIC_ALWAYS is in GMP_ASM_POWERPC_PIC_ALWAYS,
> which is indeed only called on powerpc*.
> 
> So the fix might be as simple as repairing that ifelse?

Well, this fixes devel/xtensa-lx106-elf/gdb :

https://bin.charlenew.xyz/xtensa-lx106-elf-gdb-8.2.1p0.log



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

[macppc / lld] Unbreak games/corewars

[Charlene Wendling]

Hi,

Corewars' build failed in the current macppc bulk:

> https://bin.charlenew.xyz/corewars.fail.log

Once again the `-lm' linker flag is needed on macppc. The below diff
allows me to build and run corewars on macppc [0], with no regression on
amd64.

While here, i've added a missing tab for CONFIGURE_STYLE.

OK?

Charlène.


[0] https://bin.charlenew.xyz/corewars.fixed.log


Index: Makefile
===
RCS file: /cvs/ports/games/corewars/Makefile,v
retrieving revision 1.29
diff -u -p -u -p -r1.29 Makefile
--- Makefile28 Feb 2021 06:13:34 -  1.29
+++ Makefile4 Sep 2021 14:48:39 -
@@ -3,7 +3,7 @@
 COMMENT=   computer simulation game
 
 DISTNAME=  corewars-0.9.13
-REVISION=  10
+REVISION=  11
 
 CATEGORIES=games x11
 
@@ -23,7 +23,10 @@ LIB_DEPENDS= x11/gtk+2
 MASTER_SITES=  ${MASTER_SITE_SOURCEFORGE:=corewars/}
 
 AUTOCONF_VERSION=  2.59
-CONFIGURE_STYLE=autoconf
+CONFIGURE_STYLE=   autoconf
+
+# Needed for architectures not inlining sqrt(3), like powerpc
+CONFIGURE_ENV= LDFLAGS="${LDFLAGS} -lm"
 
 post-install:
${INSTALL_DATA_DIR} ${PREFIX}/share/doc/corewars

Re: [macppc / lld] Unbreak sysutils/gdmap

[Charlene Wendling]

On Fri, 27 Aug 2021 15:18:14 +
Charlene Wendling  wrote:

> Hi,
> 
> > https://bin.charlenew.xyz/gdmap-0.8.1p5.log
> 
> Once again, macppc needs the `-lm' linker flag.
> 
> With the below diff i can build gdmap [0] on macppc. While here, i've
> refreshed WANTLIB.
> 
> The runtime has been tested successfully on macppc and amd64.
> 
> Comments/feedback are welcome,
> 
> Charlène.
> 
> 
> [0] https://bin.charlenew.xyz/gdmap_fixed.log
> 

It's better with a comment :)

Index: Makefile
===
RCS file: /cvs/ports/sysutils/gdmap/Makefile,v
retrieving revision 1.9
diff -u -p -u -p -r1.9 Makefile
--- Makefile12 Jul 2019 20:49:43 -  1.9
+++ Makefile27 Aug 2021 16:21:30 -
@@ -3,17 +3,18 @@
 COMMENT =  graphical disk map
 DISTNAME = gdmap-0.8.1
 CATEGORIES =   sysutils x11
-REVISION = 5
+REVISION = 6
 
 HOMEPAGE = http://gdmap.sourceforge.net/
 
 # GPLv2
 PERMIT_PACKAGE =   Yes
 
-WANTLIB =  X11 Xcomposite Xcursor Xdamage Xext Xfixes Xi Xinerama Xrandr
-WANTLIB += Xrender atk-1.0 c cairo fontconfig freetype gdk-x11-2.0
-WANTLIB += gdk_pixbuf-2.0 gio-2.0 glib-2.0 gobject-2.0 gtk-x11-2.0 iconv
-WANTLIB += intl m pango-1.0 pangocairo-1.0 pangoft2-1.0 pthread xml2 z
+WANTLIB += X11 Xcomposite Xcursor Xdamage Xext Xfixes Xi Xinerama
+WANTLIB += Xrandr Xrender atk-1.0 c cairo fontconfig freetype
+WANTLIB += gdk-x11-2.0 gdk_pixbuf-2.0 gio-2.0 glib-2.0 gobject-2.0
+WANTLIB += gtk-x11-2.0 harfbuzz intl m pango-1.0 pangocairo-1.0
+WANTLIB += pangoft2-1.0 pthread xml2 z
 
 MASTER_SITES = ${MASTER_SITE_SOURCEFORGE:=gdmap/}
 
@@ -29,6 +30,9 @@ CONFIGURE_STYLE = autoconf
 AUTOCONF_VERSION = 2.61
 
 CONFIGURE_ENV =CPPFLAGS="-I${LOCALBASE}/include" \
-   LDFLAGS="-L${LOCALBASE}/lib"
+   LDFLAGS="-L${LOCALBASE}/lib ${LDFLAGS}"
+
+# Needed for architectures not inlining sqrt(3), like powerpc
+LDFLAGS += -lm
 
 .include 

[macppc / lld] Unbreak sysutils/gdmap

[Charlene Wendling]

Hi,

> https://bin.charlenew.xyz/gdmap-0.8.1p5.log

Once again, macppc needs the `-lm' linker flag.

With the below diff i can build gdmap [0] on macppc. While here, i've
refreshed WANTLIB.

The runtime has been tested successfully on macppc and amd64.

Comments/feedback are welcome,

Charlène.


[0] https://bin.charlenew.xyz/gdmap_fixed.log


Index: Makefile
===
RCS file: /cvs/ports/sysutils/gdmap/Makefile,v
retrieving revision 1.9
diff -u -p -u -p -r1.9 Makefile
--- Makefile12 Jul 2019 20:49:43 -  1.9
+++ Makefile27 Aug 2021 15:14:19 -
@@ -3,17 +3,18 @@
 COMMENT =  graphical disk map
 DISTNAME = gdmap-0.8.1
 CATEGORIES =   sysutils x11
-REVISION = 5
+REVISION = 6
 
 HOMEPAGE = http://gdmap.sourceforge.net/
 
 # GPLv2
 PERMIT_PACKAGE =   Yes
 
-WANTLIB =  X11 Xcomposite Xcursor Xdamage Xext Xfixes Xi Xinerama Xrandr
-WANTLIB += Xrender atk-1.0 c cairo fontconfig freetype gdk-x11-2.0
-WANTLIB += gdk_pixbuf-2.0 gio-2.0 glib-2.0 gobject-2.0 gtk-x11-2.0 iconv
-WANTLIB += intl m pango-1.0 pangocairo-1.0 pangoft2-1.0 pthread xml2 z
+WANTLIB += X11 Xcomposite Xcursor Xdamage Xext Xfixes Xi Xinerama
+WANTLIB += Xrandr Xrender atk-1.0 c cairo fontconfig freetype
+WANTLIB += gdk-x11-2.0 gdk_pixbuf-2.0 gio-2.0 glib-2.0 gobject-2.0
+WANTLIB += gtk-x11-2.0 harfbuzz intl m pango-1.0 pangocairo-1.0
+WANTLIB += pangoft2-1.0 pthread xml2 z
 
 MASTER_SITES = ${MASTER_SITE_SOURCEFORGE:=gdmap/}
 
@@ -29,6 +30,6 @@ CONFIGURE_STYLE = autoconf
 AUTOCONF_VERSION = 2.61
 
 CONFIGURE_ENV =CPPFLAGS="-I${LOCALBASE}/include" \
-   LDFLAGS="-L${LOCALBASE}/lib"
+   LDFLAGS="-L${LOCALBASE}/lib -lm"
 
 .include 

[macppc / lld] Unbreak japanese/kanjipad

[Charlene Wendling]

Hi,

Kanjipad is broken in the current macppc bulk:

> https://bin.charlenew.xyz/kanjipad.log

macppc does not inline sqrt(3) so it actually uses libm, unlike
other archs [0]. I don't know why this has not happened before.

The below diff adds the necessary `-lm' linker flag to allow kanjipad
to be built with lld on macppc [1]. While here i've refreshed WANTLIB.

The runtime is fine on macppc and amd64.

OK? 

Charlène.


[0] https://marc.info/?l=openbsd-ports&m=161299356616167&w=2
[1] https://bin.charlenew.xyz/kanjipad_fixed.log


Index: Makefile
===
RCS file: /cvs/ports/japanese/kanjipad/Makefile,v
retrieving revision 1.20
diff -u -p -u -p -r1.20 Makefile
--- Makefile12 Jul 2019 21:02:22 -  1.20
+++ Makefile26 Aug 2021 15:20:48 -
@@ -3,7 +3,7 @@
 COMMENT=   Japanese handwriting recognition
 
 DISTNAME=  kanjipad-2.0.0
-REVISION=  7
+REVISION=  8
 CATEGORIES=japanese
 
 HOMEPAGE=  https://fishsoup.net/software/kanjipad/
@@ -16,13 +16,18 @@ MASTER_SITES=   https://fishsoup.net/softw
 WANTLIB += X11 Xcomposite Xcursor Xdamage Xext Xfixes Xi Xinerama
 WANTLIB += Xrandr Xrender atk-1.0 c cairo fontconfig freetype
 WANTLIB += gdk-x11-2.0 gdk_pixbuf-2.0 gio-2.0 glib-2.0 gobject-2.0
-WANTLIB += gtk-x11-2.0 intl m pango-1.0 pangocairo-1.0 pangoft2-1.0
-WANTLIB += z
+WANTLIB += gtk-x11-2.0 harfbuzz intl pango-1.0 pangocairo-1.0
+WANTLIB += pangoft2-1.0 z
+
+# Needed for architectures not inlining sqrt(3), like powerpc
+WANTLIB += m
 
 LIB_DEPENDS=   devel/gettext,-runtime \
x11/gtk+2
 
-MAKE_FLAGS=OPTIMIZE='${CFLAGS}' PREFIX=${PREFIX}
+MAKE_FLAGS=OPTIMIZE="${CFLAGS}" LDFLAGS="${LDFLAGS}" PREFIX="${PREFIX}"
+
+LDFLAGS+=  -lm
 
 post-install:
${INSTALL_DATA_DIR} ${PREFIX}/share/doc/kanjipad

Unbreak archivers/innoextract on all clang+ld.bfd archs (macppc AND mips64) (Was: Re: mips64 bulk build report)

[Charlene Wendling]

Hi,

> http://build-failures.rhaalovely.net/mips64/2021-07-26/archivers/innoextract.log

I'm feeling compelled to fix my own mistake here :)

Back in the days this error has popped out on macppc, i had some very
low hope for mips64 to have libexecinfo support one day and as such
limited the fix to macppc.

The problem here is due `-Wl,--as-needed' that tends to misbehave on
ld.bfd archs, the whole explanation is in the diff actually.

In order to make it universal for all clang+ld.bfd archs, i'm proposing
to disable optimisations for these archs through PROPERTIES, so if
sparc64 joins the gang one day, there is no further things to do.

It's doing more than the patch used to do, hence a REVISION bump.
If really wanted i can provide a NO_AS_NEEDED option that would involve
only the removal of the linker flag. I'm not sure what is the best
course of action.

I've tested on macppc where it builds OK; there is no changes for
clang+ld.lld archs. I've no mips64 machine, so i can't test.

Comments and feedback are welcome,

Charlène.


Index: Makefile
===
RCS file: /cvs/ports/archivers/innoextract/Makefile,v
retrieving revision 1.17
diff -u -p -u -p -r1.17 Makefile
--- Makefile5 Nov 2020 11:23:30 -   1.17
+++ Makefile6 Aug 2021 13:34:30 -
@@ -3,7 +3,7 @@
 COMMENT=   unpack Inno Setup installers
 
 DISTNAME=  innoextract-1.9
-REVISION=  0
+REVISION=  1
 CATEGORIES=archivers
 HOMEPAGE=  https://constexpr.org/innoextract/
 MAINTAINER=Thomas Frohwein 
@@ -37,6 +37,17 @@ CONFIGURE_ARGS+=-DCMAKE_INSTALL_PREFIX="
-DDOXYGEN_EXECUTABLE="false"
 
 NO_TEST=   Yes
+
+.include 
+
+# Drop -Wl,--as-needed causes the std::unique_ptr test to fail on ld.bfd archs
+# using base-clang due to undefined reference errors, and user supplied linker
+# flags are removed during the test. A fallback exists, using std::auto_ptr,
+# but it has been disabled by C++17. Use the same code as other archs instead
+# of reenabling std::auto_ptr.
+.if ${PROPERTIES:Mclang} && !${PROPERTIES:Mlld}
+CONFIGURE_ARGS+=   -DSET_OPTIMIZATION_FLAGS=OFF
+.endif
 
 do-install:
${INSTALL_PROGRAM} ${WRKBUILD}/innoextract ${PREFIX}/bin
Index: patches/patch-cmake_BuildType_cmake
===
RCS file: patches/patch-cmake_BuildType_cmake
diff -N patches/patch-cmake_BuildType_cmake
--- patches/patch-cmake_BuildType_cmake 15 Dec 2020 19:38:22 -  1.1
+++ /dev/null   1 Jan 1970 00:00:00 -
@@ -1,24 +0,0 @@
-$OpenBSD: patch-cmake_BuildType_cmake,v 1.1 2020/12/15 19:38:22 cwen Exp $
-
-Fix the build on powerpc.
-Should be reviewed once lld becomes the default linker on powerpc
-
-Index: cmake/BuildType.cmake
 cmake/BuildType.cmake.orig
-+++ cmake/BuildType.cmake
-@@ -301,6 +301,15 @@ else(MSVC)
-   if(MACOS)
-   # TODO For some reason this check succeeds on macOS, 
but then
-   # flag causes the actual build to fail :(
-+  elseif(CMAKE_SYSTEM_NAME STREQUAL "OpenBSD"
-+ AND CMAKE_SYSTEM_PROCESSOR STREQUAL "powerpc")
-+  # XXX Need a review once lld is the default linker on 
powerpc
-+  # -Wl,--as-needed causes the std::unique_ptr test to
-+  # fail due to undefined reference errors, and user
-+  # supplied linker flags are removed during the test. A
-+  # fallback exists, using std::auto_ptr, but it has been
-+  # disabled by C++17. Use the same code as other archs
-+  # instead of reenabling std::auto_ptr.
-   else()
-   # Link as few libraries as possible
-   # This is much easier than trying to decide which 
libraries are needed for each

Re: sparc64 bulk build report

[Charlene Wendling]

On Sat, 31 Jul 2021 12:08:24 +1000
Jonathan Gray  wrote:

> On Sat, Jul 31, 2021 at 01:03:08AM +0000, Charlene Wendling wrote:
> > On Fri, 30 Jul 2021 23:43:07 +0200
> > Matthias Kilian  wrote:
> > 
> > > On Fri, Jul 30, 2021 at 11:36:47PM +0200, Matthias Kilian wrote:
> > > > FWIW, I've poppler-21.07.0 in my tree, an update to 21.08.0 will
> > > > probably available in two days. I can disable poppler-qt{5,6}
> > > > for sparc64.
> > > > 
> > > > > Note that the NOT_FOR_ARCHS-arm for qt5 looks suspect.
> > > > 
> > > > Well, this entry is more than 5 years old. Should I remove it
> > > > (there seems to be a qt5 package available for arm)?
> > > > 
> > > > > This builds on qt6, really ?
> > > > 
> > > > I don't see a qt6 package for arm on the mirrors.
> > > 
> > > Maybe something like this?
> > 
> > Not yet, in my humble opinion.
> > 
> > In fact, mips64 has the same issue, and probably other ld.bfd archs.
> > powerpc is not impacted but since bulks are so slow i have not hit
> > that yet (the bulk snapshot is from July 16).
> > 
> > One major change occurred , Mesa has been updated by July 22:
> > 
> > http://marc.info/?l=openbsd-cvs&m=162694920304776
> > 
> > Qt5 and Qt6 have started to be broken after the July 23 bulk on
> > sparc64, July 27 for mips64.
> > 
> > Taken from the mips64 bulk:
> > 
> > > clang++ -Wl,-rpath,/usr/X11R6/lib -o egl main.o
> > > -L/usr/X11R6/lib -lEGL -L/usr/local/lib
> > > -L/pobj/qtbase-5.15.2/qtbase-everywhere-src-5.15.2/build-octeon/config.tests/egl
> > >  /usr/X11R6/lib/libX11.so.
> > > 17.1: warning: sprintf() is often misused, please use
> > > snprintf() /usr/X11R6/lib/libX11.so.17.1: warning: strcat() is
> > > almost always misused, please use
> > > strlcat() /usr/X11R6/lib/libX11.so.17.1: warning: strcpy() is
> > > almost always misused, please use
> > > strlcpy() /usr/X11R6/lib/libEGL.so.1.1: undefined reference to
> > > `_mesa_sha1_compute' /usr/X11R6/lib/libEGL.so.1.1: undefined
> > > reference to `_mesa_sha1_format' /usr/X11R6/lib/libEGL.so.1.1:
> > > undefined reference to
> > > `util_get_process_name' /usr/X11R6/lib/libEGL.so.1.1: undefined
> > > reference to `ralloc_asprintf_append' /usr/X11R6/lib/libEGL.so.
> > > 1.1: undefined reference to
> > > `ralloc_free' /usr/X11R6/lib/libEGL.so.1.1: undefined reference
> > > to `ralloc_strdup' /usr/X11R6/lib/libEGL.so.1.1: undefined
> > > reference to `util_get_process_exec_path' clang++: error: linker
> > > command failed with exit code 1 (use -v to see invocation)
> > 
> > Since that test fails, all the EGL stuff is not built.

I've isolated the test and attached it, so i don't need to pull Qt5,
which would have otherwise taken a bit less than 2 days to get
built at home (i build everything from scratch due to being fed
up with library mismatches).

I've installed the latest macppc snapshot (#916: Fri Jul 30 16:26:40 MDT
2021) on my PowerBook. I was able to reproduce the issue.

> Does a libEGL built with the following help?

Yes, it passes the test! Thanks a lot!

> Follows how the meson build is linked.
> 
> Index: lib/mesa/mk/libEGL/Makefile
> ===
> RCS file: /cvs/xenocara/lib/mesa/mk/libEGL/Makefile,v
> retrieving revision 1.3
> diff -u -p -r1.3 Makefile
> --- lib/mesa/mk/libEGL/Makefile   22 Jul 2021 11:10:08
> - 1.3 +++ lib/mesa/mk/libEGL/Makefile 31 Jul 2021
> 02:04:01 - @@ -38,7 +38,8 @@ CPPFLAGS+=   -I$
> {MESA_SRC}/src/egl/main \ -I${MESA_SRC}/src/gbm/backends/dri \
>   -D_EGL_NATIVE_PLATFORM=_EGL_PLATFORM_X11
>  
> -LDADD+= ${.CURDIR}/../libmesa_util/${__objdir}/libmesa_util.a \
> +LDADD+= -Wl,--as-needed -Wl,--allow-shlib-undefined
> -Wl,--start-group \
> + ${.CURDIR}/../libmesa_util/${__objdir}/libmesa_util.a \
>   ${.CURDIR}/../libmesa_format/${__objdir}/libmesa_format.a \
>   -lz -lm \
>   -L${X11BASE}/lib -lX11-xcb -lX11 -lxcb -lxcb-xfixes
> @@ -55,7 +56,8 @@ LDADD+= ${.CURDIR}/../libloader_dri3_hel
>  .endif
>  
>  LDADD+= -L${.CURDIR}/../libgbm/${__objdir} -lgbm \
> - -L${.CURDIR}/../libglapi/${__objdir} -lglapi
> + -L${.CURDIR}/../libglapi/${__objdir} -lglapi \
> + -Wl,--end-group
>  
>  obj: _xenocara_obj
>  
> 


egl_test.tgz
Description: Binary data

Re: sparc64 bulk build report

[Charlene Wendling]

On Fri, 30 Jul 2021 23:43:07 +0200
Matthias Kilian  wrote:

> On Fri, Jul 30, 2021 at 11:36:47PM +0200, Matthias Kilian wrote:
> > FWIW, I've poppler-21.07.0 in my tree, an update to 21.08.0 will
> > probably available in two days. I can disable poppler-qt{5,6} for
> > sparc64.
> > 
> > > Note that the NOT_FOR_ARCHS-arm for qt5 looks suspect.
> > 
> > Well, this entry is more than 5 years old. Should I remove it (there
> > seems to be a qt5 package available for arm)?
> > 
> > > This builds on qt6, really ?
> > 
> > I don't see a qt6 package for arm on the mirrors.
> 
> Maybe something like this?

Not yet, in my humble opinion.

In fact, mips64 has the same issue, and probably other ld.bfd archs.
powerpc is not impacted but since bulks are so slow i have not hit
that yet (the bulk snapshot is from July 16).

One major change occurred , Mesa has been updated by July 22:

http://marc.info/?l=openbsd-cvs&m=162694920304776

Qt5 and Qt6 have started to be broken after the July 23 bulk on
sparc64, July 27 for mips64.

Taken from the mips64 bulk:

> clang++ -Wl,-rpath,/usr/X11R6/lib -o egl main.o   -L/usr/X11R6/lib -lEGL 
> -L/usr/local/lib  
> -L/pobj/qtbase-5.15.2/qtbase-everywhere-src-5.15.2/build-octeon/config.tests/egl
>  
> /usr/X11R6/lib/libX11.so.17.1: warning: sprintf() is often misused, please 
> use snprintf()
> /usr/X11R6/lib/libX11.so.17.1: warning: strcat() is almost always misused, 
> please use strlcat()
> /usr/X11R6/lib/libX11.so.17.1: warning: strcpy() is almost always misused, 
> please use strlcpy()
> /usr/X11R6/lib/libEGL.so.1.1: undefined reference to `_mesa_sha1_compute'
> /usr/X11R6/lib/libEGL.so.1.1: undefined reference to `_mesa_sha1_format'
> /usr/X11R6/lib/libEGL.so.1.1: undefined reference to `util_get_process_name'
> /usr/X11R6/lib/libEGL.so.1.1: undefined reference to `ralloc_asprintf_append'
> /usr/X11R6/lib/libEGL.so.1.1: undefined reference to `ralloc_free'
> /usr/X11R6/lib/libEGL.so.1.1: undefined reference to `ralloc_strdup'
> /usr/X11R6/lib/libEGL.so.1.1: undefined reference to 
> `util_get_process_exec_path'
> clang++: error: linker command failed with exit code 1 (use -v to see 
> invocation)

Since that test fails, all the EGL stuff is not built.

[macppc] Unbreak lang/gforth

[Charlene Wendling]

Hi,

After the libffi fix i've just sent to ports@ and that is needed on top
of that diff, gforth needs some tweaks to build on macppc.

By default it runs out of memory. So i tried to 

- use i386 patches: it runs out of memory
- build everything with -O1: it runs out of memory
- build everything with -O0: preforth segfaults

Finally, i managed to make it with ports-gcc and building with -O0 due to
memory usage [0]. It's also _way_ faster to build than with base-clang.

No REVISION bump is needed, it's broken since base-clang's introduction,
_SYSTEM_VERSION-powerpc has been bumped since then, and this diff only
impacts macppc.

The test suite is very light, so i tried several example of forth code
found on the web without issues.

Comments/feedback are welcome,

Charlène.


[0] https://bin.charlenew.xyz/gforth-0.7.3p3.log


Index: Makefile
===
RCS file: /cvs/ports/lang/gforth/Makefile,v
retrieving revision 1.25
diff -u -p -u -p -r1.25 Makefile
--- Makefile12 Jul 2019 20:47:18 -  1.25
+++ Makefile29 Jul 2021 14:09:53 -
@@ -42,6 +42,17 @@ SUBST_VARS +=KERNEL V
 PATCH_LIST =   patch-* patch_i386-*
 .endif
 
+# With base-clang on powerpc it either run out of memory with i386 patches or
+# -O1. Using -O0 makes preforth segfault. Use ports-gcc instead.
+MODULES += gcc4
+MODGCC4_ARCHS =powerpc
+MODGCC4_LANGS =c
+
+# We still need -O0 due to high memory usage though.
+.if ${MACHINE_ARCH} == "powerpc"
+CFLAGS +=  -O0
+.endif
+
 .include 
 
 # The filename of the gForth kernel is based on the wordsize

[macppc] devel/libffi: improve consumers build and probably runtime

[Charlene Wendling]

Hi,

George emitted the hypothesis that libffi may be partly broken on
macppc. He was right :)

After searching around, i found out that upstream has already a fix for
that [0]. Tests results have improved a lot:

Before the below diff:

   # of expected passes440
   # of unexpected failures104

After the below diff:

   # of expected passes530
   # of unexpected failures14

I have been able to build gforth on macppc and make it run with that
patched libffi. Finally. I'll send a diff for gforth after that one,
some extra bits are required.

I think no further consumer testing is really needed, given the
spectacular change in tests results, we should assume the libffi we're
shipping on macppc currently is in a pretty bad shape anyway.

Comments/feedback are welcome,

Charlène.


[0] https://github.com/libffi/libffi/commit/4d6d2866


Index: Makefile
===
RCS file: /cvs/ports/devel/libffi/Makefile,v
retrieving revision 1.43
diff -u -p -u -p -r1.43 Makefile
--- Makefile11 Jun 2021 15:37:14 -  1.43
+++ Makefile29 Jul 2021 13:54:53 -
@@ -3,7 +3,7 @@
 COMMENT=   Foreign Function Interface
 
 DISTNAME=  libffi-3.3
-REVISION=  0
+REVISION=  1
 SHARED_LIBS +=  ffi  1.2  # .6.4
 CATEGORIES=devel
 
Index: patches/patch-src_powerpc_sysv_S
===
RCS file: patches/patch-src_powerpc_sysv_S
diff -N patches/patch-src_powerpc_sysv_S
--- /dev/null   1 Jan 1970 00:00:00 -
+++ patches/patch-src_powerpc_sysv_S29 Jul 2021 13:54:53 -
@@ -0,0 +1,48 @@
+$OpenBSD$
+
+From 4d6d2866ae43e55325e8ee96561221804602cd7a Mon Sep 17 00:00:00 2001
+From: Samuel Holland 
+Date: Fri, 21 Feb 2020 21:06:15 -0600
+Subject: [PATCH] Update powerpc sysv assembly for ffi_powerpc.h changes (#541)
+
+Some of the flag bits were moved when adding powerpc64 vector support.
+
+Fixes #536
+
+Index: src/powerpc/sysv.S
+--- src/powerpc/sysv.S.orig
 src/powerpc/sysv.S
+@@ -104,17 +104,16 @@ ENTRY(ffi_call_SYSV)
+   bctrl
+ 
+   /* Now, deal with the return value.  */
+-  mtcrf   0x01,%r31 /* cr7  */
++  mtcrf   0x03,%r31 /* cr6-cr7  */
+   bt- 31,L(small_struct_return_value)
+   bt- 30,L(done_return_value)
+ #ifndef __NO_FPRS__
+   bt- 29,L(fp_return_value)
+ #endif
+   stw %r3,0(%r30)
+-  bf+ 28,L(done_return_value)
++  bf+ 27,L(done_return_value)
+   stw %r4,4(%r30)
+-  mtcrf   0x02,%r31 /* cr6  */
+-  bf  27,L(done_return_value)
++  bf  26,L(done_return_value)
+   stw %r5,8(%r30)
+   stw %r6,12(%r30)
+   /* Fall through...  */
+@@ -145,10 +144,9 @@ L(done_return_value):
+ #ifndef __NO_FPRS__
+ L(fp_return_value):
+   .cfi_restore_state
+-  bf  28,L(float_return_value)
++  bf  27,L(float_return_value)
+   stfd%f1,0(%r30)
+-  mtcrf   0x02,%r31 /* cr6  */
+-  bf  27,L(done_return_value)
++  bf  26,L(done_return_value)
+   stfd%f2,8(%r30)
+   b   L(done_return_value)
+ L(float_return_value):

Re: [NEW] mail/p5-OpenSMTPd-Filter 0.0.2

[Charlene Wendling]

On Tue, 27 Jul 2021 18:38:04 -0700
Andrew Hewus Fresh  wrote:

> This module is a helper to make writing OpenSMTPd
>  filters in perl easier.
> 
> https://metacpan.org/release/OpenSMTPd-Filter
> 
> While this module is somewhat new, I've been using it on my home
> server for a while with seemingly good success.  I've been wanting to
> figure out how to make it work with an async library of some sort so
> it could process mail in parallel which would be super important for
> a busy server.  However, on my very low throughput server that isn't a
> limitation, so I figure this might help other folks even though I
> haven't had time to figure out how to be async.
> 
> Questions, comments, OK for import?

I was not able to fetch the distfile. I managed to do so by adding
CPAN_AUTHOR=ANDREW to the Makefile.

Portgen spots the need for CPAN_AUTHOR and adds it to the port Makefile,
so the distfile is downloaded without issues. Maybe a "make clean=dist"
invocation on all generated ports should be added to portgen? 

It passes all test without issues in a clean chroot.

OK cwen@ with CPAN_AUTHOR set :)

> l8rZ,
> -- 
> andrew - http://afresh1.com
> 
> The programmer's national anthem is 'GH!!'.

[macppc] devel/clang-tools-extra: remove BROKEN

[Charlene Wendling]

Hi,

Since i've set up BROKEN for this port, due to time wasted because it's
lacking consumers, rust has taken out many ports, the bootstrapped
guile2 now builds in a reasonable time, and even recently we're finally
free from Qt4.

I'm proposing to bring back clang-tools-extra. It builds fine [0] if
you wonder.

OK?

Charlène.

[0] https://bin.charlenew.xyz/clang-tools-extra-11.1.0p0.log


Index: Makefile
===
RCS file: /cvs/ports/devel/clang-tools-extra/Makefile,v
retrieving revision 1.16
diff -u -p -u -p -r1.16 Makefile
--- Makefile18 May 2021 06:19:48 -  1.16
+++ Makefile27 Jul 2021 18:42:43 -
@@ -9,7 +9,6 @@
 # patches: rm patches/patch-*lld*
 
 ONLY_FOR_ARCHS =   ${LLVM_ARCHS}
-BROKEN-powerpc =   no consumers on powerpc, save 25 build hours in bulks
 
 COMMENT=   Clang extra tools
 

[macppc] Unbreak games/cataclysm-dda

[Charlene Wendling]

Hi,

It was expected to happen one day :)

>From the current macppc bulk:

> obj/tiles/active_item_cache_test.o: In function 
> `::_ZL29C_A_T_C_HT_E_S_T__(void)':
> active_item_cache_test.cpp:(.text+0x1b4): relocation truncated to fit: 
> R_PPC_PLTREL24 [...]
> active_item_cache_test.cpp:(.text+0xcb4): additional relocation overflows 
> omitted from the output

The below diff allowed me to build CDDA [0].

Runtime test: it is slow, but OK.

LDFLAGS is empty and this version has never been built on macppc, so
REVISION bump is not needed.

Comments/feedback are welcome,

Charlène.


[0] https://bin.charlenew.xyz/cataclysm-dda-0.Fv0.log


Index: Makefile
===
RCS file: /cvs/ports/games/cataclysm-dda/Makefile,v
retrieving revision 1.18
diff -u -p -u -p -r1.18 Makefile
--- Makefile14 Jul 2021 03:50:15 -  1.18
+++ Makefile27 Jul 2021 18:05:20 -
@@ -30,7 +30,7 @@ RUN_DEPENDS=  devel/desktop-file-utils \
 
 LIB_DEPENDS=   devel/gettext,-runtime
 
-MAKE_ENV=  CXX="${CXX}" CXXFLAGS="${CXXFLAGS}"
+MAKE_ENV=  CXX="${CXX}" CXXFLAGS="${CXXFLAGS}" LDFLAGS="${LDFLAGS}"
 
 MAKE_FLAGS=ASTYLE=0 RELEASE=1 USE_XDG_DIR=1 \
LINTJSON=0 LANGUAGES=all
@@ -52,6 +52,12 @@ MAN_TILES=   -tiles
 USE_GMAKE= Yes
 
 NO_TEST=   Yes
+
+# relocation truncated to fit: R_PPC_PLTREL24 -- to be removed once lld is the
+# default linker on powerpc
+.if ${MACHINE_ARCH} == "powerpc"
+LDFLAGS+=  -Wl,-relax
+.endif
 
 post-install:
${INSTALL_MAN} ${WRKSRC}/doc/cataclysm${MAN_TILES}.6 ${PREFIX}/man/man6/

Re: flameshot 0.10 broken .desktop (Re: CVS: cvs.openbsd.org: ports)

[Charlene Wendling]

On Mon, 26 Jul 2021 09:04:47 +0200
Matthieu Herrb  wrote:

> On Sun, Jul 25, 2021 at 06:16:05PM +0200, Matthieu Herrb wrote:
> > On Fri, Jul 16, 2021 at 10:47:39AM -0600, Denis Fondras wrote:
> > > CVSROOT:  /cvs
> > > Module name:  ports
> > > Changes by:   de...@cvs.openbsd.org   2021/07/16
> > > 10:47:39
> > > 
> > > Modified files:
> > >   graphics/flameshot: Makefile distinfo 
> > >   graphics/flameshot/pkg: PLIST 
> > > Added files:
> > >   graphics/flameshot/patches: patch-src_CMakeLists_txt 
> > > 
> > > Log message:
> > > Update to 0.10.0
> > > 
> > > Patch by Stefan Hagen 
> > > Help and OK sthen@
> > 
> > Hi,
> > 
> > 
> > Looks like this broke the .desktop file
> > (/usr/local/share/applications/org.flameshot.Flameshot.desktop).
> > It now contains:
> > 
> >Exec=/usr/bin/flameshot
> > 
> > Which obviously doesn't work.
> > 
> 
> Possible fix:
> 
> Index: Makefile
> ===
> RCS file: /cvs/OpenBSD/ports/graphics/flameshot/Makefile,v
> retrieving revision 1.5
> diff -u -p -u -r1.5 Makefile
> --- Makefile  16 Jul 2021 16:47:38 -  1.5
> +++ Makefile  26 Jul 2021 05:41:37 -
> @@ -1,6 +1,7 @@
>  # $OpenBSD: Makefile,v 1.5 2021/07/16 16:47:38 denis Exp $
>  
>  COMMENT =powerful yet simple to use screenshot software
> +REVISION =   0
>  CATEGORIES = graphics x11
>  
>  GH_ACCOUNT = flameshot-org
> @@ -25,6 +26,10 @@ RUN_DEPENDS =  devel/desktop-file-utils \
>   x11/gtk+3,-guic
>  
>  CONFIGURE_ARGS +=-DENABLE_CACHE=OFF
> +
> +post-patch:
> + perl -pi -e 's,/usr/bin/,${LOCALBASE}/bin/,' \
> + $
> {WRKSRC}/data/desktopEntry/package/org.flameshot.Flameshot.desktop 
>  NO_TEST =Yes
>  
> 
> -- 
> Matthieu Herrb
> 

By the way, while running configure i hit this:

 -- Found Git: /usr/local/bin/git (found version "2.32.0") 
 git found: /usr/local/bin/git in version 2.32.0
 fatal: not a git repository (or any parent up to mount point /)
 Stopping at filesystem boundary (GIT_DISCOVERY_ACROSS_FILESYSTEM not
 set).
 FLAMESHOT_GIT_HASH:

I have disabled git for the build in the below diff.

I don't know if using ${SUBST_CMD} with a patched .desktop file is
preferable or not (i would have done it that way).

OK cwen@ either way.


Index: Makefile
===
RCS file: /cvs/ports/graphics/flameshot/Makefile,v
retrieving revision 1.5
diff -u -p -u -p -r1.5 Makefile
--- Makefile16 Jul 2021 16:47:38 -  1.5
+++ Makefile26 Jul 2021 08:54:25 -
@@ -1,6 +1,7 @@
 # $OpenBSD: Makefile,v 1.5 2021/07/16 16:47:38 denis Exp $
 
 COMMENT =  powerful yet simple to use screenshot software
+REVISION = 0
 CATEGORIES =   graphics x11
 
 GH_ACCOUNT =   flameshot-org
@@ -24,7 +25,12 @@ LIB_DEPENDS =x11/qt5/qtsvg
 RUN_DEPENDS =  devel/desktop-file-utils \
x11/gtk+3,-guic
 
-CONFIGURE_ARGS +=  -DENABLE_CACHE=OFF
+CONFIGURE_ARGS +=  -DENABLE_CACHE=OFF \
+   -DCMAKE_DISABLE_FIND_PACKAGE_Git=TRUE
+
+post-patch:
+   perl -pi -e 's,/usr/bin/,${LOCALBASE}/bin/,' \
+   ${WRKSRC}/data/desktopEntry/package/org.flameshot.Flameshot.desktop
 
 NO_TEST =  Yes
 

[update] multimedia/atomicparsley 0.9.6 -> 20210715 (+unsigned char archs fixes)

[Charlene Wendling]

Hi,

Here is an update for atomicparsley. Upstream moved from bitbucket to
github and the project now uses cmake, among many other changes [0].

Port-wise, it's just a minimal port using cmake, i don't have much to
say, excepted that i added some fixes for unsigned char archs as
reported on macppc (it's upstreamed already).

I also shortened the version, that is too lengthy for my taste.

It builds and runs fine on macppc and amd64.

Comments/feedback are welcome,

Charlène.

[0]
https://github.com/wez/atomicparsley/compare/0.9.6...20210715.151551.e7ad03a


Index: Makefile
===
RCS file: /cvs/ports/multimedia/atomicparsley/Makefile,v
retrieving revision 1.14
diff -u -p -u -p -r1.14 Makefile
--- Makefile3 Feb 2020 20:40:39 -   1.14
+++ Makefile25 Jul 2021 14:09:53 -
@@ -2,33 +2,27 @@
 
 COMMENT=   read and set MPEG-4/3GPP metadata tags
 
-V= 0.9.6
-DISTNAME=  atomicparsley-$V
-REVISION=  4
-DISTFILES= ${DISTNAME}{$V}.tar.bz2
-WRKDIST=   ${WRKDIR}/wez-atomicparsley-da2f6e4fc120
+GH_ACCOUNT=wez
+GH_PROJECT=atomicparsley
+GH_TAGNAME=20210715.151551.e7ad03a
 
-CATEGORIES=multimedia
+DISTNAME=  atomicparsley-20210715
 
-HOMEPAGE=  https://bitbucket.org/wez/atomicparsley/
+CATEGORIES=multimedia
 
 # GPLv2 "or its successor"
 PERMIT_PACKAGE=Yes
 
-WANTLIB=   c m ${COMPILER_LIBCXX} z
+WANTLIB += ${COMPILER_LIBCXX} c m z
 
-COMPILER = base-clang ports-gcc base-gcc
+COMPILER = base-clang ports-gcc base-gcc
 
-MASTER_SITES=  https://bitbucket.org/wez/atomicparsley/get/
+MODULES =  devel/cmake
 
-CONFIGURE_STYLE=  autoreconf
-CONFIGURE_ENV=ac_cv_func_lroundf=yes \
- OBJCXX="${CXX}"
-AUTOCONF_VERSION= 2.69
-AUTOMAKE_VERSION= 1.15
-AUTORECONF=  ./autogen.sh
+# Don't use git to retrieve the build version
+CONFIGURE_ARGS +=  -DGIT= \
+   -DPACKAGE_VERSION="${PKGNAME} (${GH_TAGNAME})"
 
-NO_TEST=   Yes
-SEPARATE_BUILD= Yes
+NO_TEST =  Yes
 
 .include 
Index: distinfo
===
RCS file: /cvs/ports/multimedia/atomicparsley/distinfo,v
retrieving revision 1.3
diff -u -p -u -p -r1.3 distinfo
--- distinfo11 Jun 2016 22:12:13 -  1.3
+++ distinfo25 Jul 2021 14:09:53 -
@@ -1,2 +1,2 @@
-SHA256 (atomicparsley-0.9.6.tar.bz2) = 
4o1GcovoYhnmzkhpXqY32DHKAXDKa9rJmBCZaoKR7lA=
-SIZE (atomicparsley-0.9.6.tar.bz2) = 178956
+SHA256 (atomicparsley-20210715.tar.gz) = 
VG3LXztiWv9Pa/ItJ6CmNtFYVP1ylAKmkz0x89BBfg0=
+SIZE (atomicparsley-20210715.tar.gz) = 230214
Index: patches/patch-src_AtomicParsley_h
===
RCS file: patches/patch-src_AtomicParsley_h
diff -N patches/patch-src_AtomicParsley_h
--- /dev/null   1 Jan 1970 00:00:00 -
+++ patches/patch-src_AtomicParsley_h   25 Jul 2021 14:09:53 -
@@ -0,0 +1,17 @@
+$OpenBSD$
+
+Fix potential bug on unsigned char archs
+See https://github.com/wez/atomicparsley/pull/41
+
+Index: src/AtomicParsley.h
+--- src/AtomicParsley.h.orig
 src/AtomicParsley.h
+@@ -258,7 +258,7 @@ AtomicInfo *APar_UserData_atom_Init(const char *userda
+ 
+ /* ID3v2 (2.4) style metadata, non-external form */
+ AtomicInfo *APar_ID32_atom_Init(const char *frameID_str,
+-char meta_area,
++signed char meta_area,
+ const char *lang_str,
+ uint16_t id32_lang);
+ 
Index: patches/patch-src_CDtoc_cpp
===
RCS file: /cvs/ports/multimedia/atomicparsley/patches/patch-src_CDtoc_cpp,v
retrieving revision 1.1
diff -u -p -u -p -r1.1 patch-src_CDtoc_cpp
--- patches/patch-src_CDtoc_cpp 11 Jun 2016 22:11:53 -  1.1
+++ patches/patch-src_CDtoc_cpp 25 Jul 2021 14:09:53 -
@@ -1,12 +1,13 @@
 $OpenBSD: patch-src_CDtoc_cpp,v 1.1 2016/06/11 22:11:53 sthen Exp $
 src/CDtoc.cpp.orig Sat Jun 11 23:06:27 2016
-+++ src/CDtoc.cpp  Sat Jun 11 23:06:38 2016
-@@ -71,7 +71,7 @@ CD_TOC_* cdTOC = NULL;
-   uint8_t LEADOUT_TRACK_NUMBER = MACOSX_LEADOUT_TRACK;
- #elif defined (HAVE_LINUX_CDROM_H)
-   uint8_t LEADOUT_TRACK_NUMBER = CDROM_LEADOUT;
--#elif defined (_WIN32)
+Index: src/CDtoc.cpp
+--- src/CDtoc.cpp.orig
 src/CDtoc.cpp
+@@ -73,7 +73,7 @@ CD_TOC_ *cdTOC = NULL;
+ uint8_t LEADOUT_TRACK_NUMBER = MACOSX_LEADOUT_TRACK;
+ #elif defined(__linux__)
+ uint8_t LEADOUT_TRACK_NUMBER = CDROM_LEADOUT;
+-#elif defined(_WIN32)
 +#else
-   uint8_t LEADOUT_TRACK_NUMBER = 0xAA; //NOTE: for WinXP 
IOCTL_CDROM_READ_TOC_EX code, its 0xA2
+ uint8_t LEADOUT_TRACK_NUMBER =
+ 0xAA; // NOTE: for WinXP IOCTL_CDROM_READ_TOC_EX code, its 0xA2
  #endif
- 
Index: patches/patch-src_parsley_cpp
===
RCS file: patches/patch-src_parsley

[macppc, arm*] mail/trojita: fix misbehaving spinner

[Charlene Wendling]

Hi,

The loading spinner in trojita is misbehaving on my macppc machine.

Upstream has already dealt with the situation [0], and after applying
their patch the spinner is fine on macppc. There is indeed no
regression on amd64.

OK?

Charlène.


[0]
https://github.com/KDE/trojita/commit/e4478df9389894b6498ffe252310cdecd1e58904


Index: Makefile
===
RCS file: /cvs/ports/mail/trojita/Makefile,v
retrieving revision 1.34
diff -u -p -u -p -r1.34 Makefile
--- Makefile19 Jan 2021 06:41:35 -  1.34
+++ Makefile24 Jul 2021 19:08:33 -
@@ -3,7 +3,7 @@
 COMMENT =  fast Qt IMAP e-mail client
 
 DISTNAME = trojita-0.7
-REVISION = 12
+REVISION = 13
 
 SHARED_LIBS =  trojita_plugins 2.0 # 2.0
 
Index: patches/patch-src_Gui_Spinner_h
===
RCS file: patches/patch-src_Gui_Spinner_h
diff -N patches/patch-src_Gui_Spinner_h
--- /dev/null   1 Jan 1970 00:00:00 -
+++ patches/patch-src_Gui_Spinner_h 24 Jul 2021 19:08:33 -
@@ -0,0 +1,26 @@
+$OpenBSD$
+
+From e4478df9389894b6498ffe252310cdecd1e58904 Mon Sep 17 00:00:00 2001
+From: Vicente Bergas
+Date: Sat, 3 Feb 2018 19:02:38 +0100
+Subject: [PATCH] Fix signed vs. unsigned math on ARM
+
+On ARM systems, `char` is not signed, but unsigned, and therefore code
+like qMin(-11, ...) doesn't really do a correct thing.
+
+(impacts powerpc as well)
+
+Index: src/Gui/Spinner.h
+--- src/Gui/Spinner.h.orig
 src/Gui/Spinner.h
+@@ -82,8 +82,8 @@ private slots:
+ void updateAncestors();
+ void updateGeometry();
+ private:
+-uchar m_step;
+-char m_fadeStep;
++int m_step;
++int m_fadeStep;
+ int m_timer;
+ QTimer *m_startTimer;
+ QList m_ancestors;

[macppc, arm*] Fix an infinite loop in games/qqwing

[Charlene Wendling]

Hi,

On my macppc machine, i can create an infinite loop just by doing:

  $ qqwing --generate | qqwing --solve

The looped getchar(3) call never reaches EOF on unsigned char archs.

The below diff fixes that issue.

OK?

Charlène.


Index: Makefile
===
RCS file: /cvs/ports/games/qqwing/Makefile,v
retrieving revision 1.3
diff -u -p -u -p -r1.3 Makefile
--- Makefile12 Jul 2019 20:46:23 -  1.3
+++ Makefile24 Jul 2021 14:38:21 -
@@ -3,7 +3,7 @@
 COMMENT =  Sudoku generator and solver
 DISTNAME = qqwing-1.3.4
 SHARED_LIBS +=  qqwing0.0 # 3.0
-REVISION = 0
+REVISION = 1
 
 CATEGORIES =   games
 
Index: patches/patch-main_cpp
===
RCS file: patches/patch-main_cpp
diff -N patches/patch-main_cpp
--- /dev/null   1 Jan 1970 00:00:00 -
+++ patches/patch-main_cpp  24 Jul 2021 14:38:21 -
@@ -0,0 +1,17 @@
+$OpenBSD$
+
+Fix infinite loop while using `qqwing --solve` on unsigned char archs (powerpc
+and arm)
+
+Index: main.cpp
+--- main.cpp.orig
 main.cpp
+@@ -488,7 +488,7 @@ void printHelp(){
+ bool readPuzzleFromStdIn(int* puzzle){
+   int read = 0;
+   while (read < BOARD_SIZE){
+-  char c = getchar();
++  int c = getchar();
+   if (c == EOF) return false;
+   if (c >= '1' && c <='9'){
+   puzzle[read] = c-'0';

[macppc, arm*] Fix an infinite loop in converters/ish

[Charlene Wendling]

Hi,

Decoding using the -lf option on powerpc made ish freeze. A looped
fgetc(3) call never reached EOF, since the return value was stored
as a char, that is unsigned on powerpc and arm*, instead of int.

--8<--
$ cd /tmp && cp /bin/ls .
$ ish -ss ls
$ rm ls
$ ish -lf ls.ish
[...]
< Restore from ls.ish >
<<< ls for UNIX ( use shift_jis ) 2021/07/24 11:55:40 357460 byte >>>
ls :  Error!! File delete.
<<< ls for UNIX ( use shift_jis ) 2021/07/24 11:55:40 357460 byte >>>
ls :
o
Finished.
[infinite loop]
-->8--

With the below diff, ish properly quits once the decoding has been
done, and the decoded ls works. Tested on amd64 and macppc.

OK?

Charlène.


Index: Makefile
===
RCS file: /cvs/ports/converters/ish/Makefile,v
retrieving revision 1.24
diff -u -p -u -p -r1.24 Makefile
--- Makefile14 Jul 2019 02:16:51 -  1.24
+++ Makefile24 Jul 2021 08:31:46 -
@@ -4,6 +4,7 @@ COMMENT=binary-to-text file converter
 
 DISTNAME=  ish201a5
 PKGNAME=   ish-2.01a5
+REVISION=  0
 CATEGORIES=converters
 
 PERMIT_PACKAGE=no license
Index: patches/patch-decode_c
===
RCS file: /cvs/ports/converters/ish/patches/patch-decode_c,v
retrieving revision 1.1
diff -u -p -u -p -r1.1 patch-decode_c
--- patches/patch-decode_c  30 Dec 2011 23:14:34 -  1.1
+++ patches/patch-decode_c  24 Jul 2021 08:31:46 -
@@ -1,9 +1,11 @@
 $OpenBSD: patch-decode_c,v 1.1 2011/12/30 23:14:34 sthen Exp $
 
-Fix: When decoding, the status display has Y2K problem.
+Hunk #1: When decoding, the status display has Y2K problem.
+Hunk #2: fix an infinite loop on unsigned char archs (powerpc and arm*)
 
 decode.c.orig  Sun Jan  7 10:46:37 1996
-+++ decode.c   Fri Dec 30 23:09:44 2011
+Index: decode.c
+--- decode.c.orig
 decode.c
 @@ -273,8 +273,8 @@ ish_head *head;
tm.tm_min = (p[1]&0x07)<<3;
tm.tm_min += p[0]>>5;
@@ -15,3 +17,13 @@ Fix: When decoding, the status display h
  #ifdef __TURBOC__
memcpy( &timep[0], head->time, 4);
timep[1] = timep[0];
+@@ -675,8 +675,7 @@ FILE*ipath;
+ int maxlen;
+ char*buff;
+ {
+-int i,j;
+-charc;
++int c,i,j;
+ 
+ if((decok & 0x80) == 0)
+ j=78;

[update] audio/grip 4.2.0 -> 4.2.2 (+unsigned char archs fix)

[Charlene Wendling]

Hi,

Here is an update for grip to 4.2.2. Upstream changes:

--8<--
v4.2.2 2021-07-07  
  * Patch from D. J. Slattery: fix crash on startup due to invalid
pointer
v4.2.1 2021-05-30
  * Translation updates: Italian
  * Update documentation, IRC moving from Freenode to Libera
-->8--

Port-wise, it's a simple update, i've just updated WANTLIB, and added
x11/gtk+3,-guic in RUN_DEPENDS as portcheck warned.

I also added a fix to ensure the runtime is fine on unsigned char archs
when starting ripping with no track selected and a M3U playlist
already exists.

The port has been tested on amd64 and macppc.

Comments/feedback are welcome,

Charlène.


Index: Makefile
===
RCS file: /cvs/ports/audio/grip/Makefile,v
retrieving revision 1.69
diff -u -p -u -p -r1.69 Makefile
--- Makefile24 Jan 2021 18:24:27 -  1.69
+++ Makefile23 Jul 2021 21:43:07 -
@@ -2,7 +2,7 @@
 
 COMMENT=   front-end to external cd audio rippers and mp3 encoders
 
-DISTNAME=  grip-4.2.0
+DISTNAME=  grip-4.2.2
 CATEGORIES=audio
 
 HOMEPAGE=  http://sourceforge.net/projects/grip/
@@ -15,8 +15,9 @@ WANTLIB += Xfixes Xi Xinerama Xrandr Xre
 WANTLIB += cdda_paranoia crypto curl expat ffi fontconfig freetype
 WANTLIB += fribidi gdk-x11-2.0 gdk_pixbuf-2.0 gio-2.0 glib-2.0
 WANTLIB += gmodule-2.0 gobject-2.0 graphite2 gtk-x11-2.0 harfbuzz
-WANTLIB += iconv id3 intl m nghttp2 pango-1.0 pangocairo-1.0 pangoft2-1.0
-WANTLIB += pcre pixman-1 png ssl xcb xcb-render xcb-shm z
+WANTLIB += iconv id3 intl jpeg m nghttp2 pango-1.0 pangocairo-1.0
+WANTLIB += pangoft2-1.0 pcre pixman-1 png ssl xcb xcb-render xcb-shm
+WANTLIB += z
 
 COMPILER=  base-clang ports-gcc base-gcc
 
@@ -31,7 +32,8 @@ LIB_DEPENDS=  audio/cdparanoia \
net/curl \
x11/gtk+2
 
-RUN_DEPENDS=   audio/vorbis-tools
+RUN_DEPENDS=   audio/vorbis-tools \
+   x11/gtk+3,-guic
 
 BUILD_DEPENDS= devel/gettext,-tools
 
Index: distinfo
===
RCS file: /cvs/ports/audio/grip/distinfo,v
retrieving revision 1.11
diff -u -p -u -p -r1.11 distinfo
--- distinfo24 Jan 2021 18:24:27 -  1.11
+++ distinfo23 Jul 2021 21:43:07 -
@@ -1,2 +1,2 @@
-SHA256 (grip-4.2.0.tar.gz) = moU7VED4NxVUm6S9BkWdYxPBf5bb/7zAkyKCx1ucJeo=
-SIZE (grip-4.2.0.tar.gz) = 1209972
+SHA256 (grip-4.2.2.tar.gz) = nXtGgJeNYM8lyllNi9UdmsnVcHOCXfryWmKGZ9QFTHE=
+SIZE (grip-4.2.2.tar.gz) = 1210481
Index: patches/patch-src_rip_c
===
RCS file: /cvs/ports/audio/grip/patches/patch-src_rip_c,v
retrieving revision 1.8
diff -u -p -u -p -r1.8 patch-src_rip_c
--- patches/patch-src_rip_c 24 Jan 2021 18:24:27 -  1.8
+++ patches/patch-src_rip_c 23 Jul 2021 21:43:07 -
@@ -1,7 +1,11 @@
 $OpenBSD: patch-src_rip_c,v 1.8 2021/01/24 18:24:27 ajacoutot Exp $
 
 src/rip.c.orig Thu Apr 15 20:21:15 2004
-+++ src/rip.c  Tue May 13 16:15:31 2008
+Hunk #2: ensure EOF is properly detected on archs where char is unsigned
+ (powerpc and arm*)
+
+Index: src/rip.c
+--- src/rip.c.orig
 src/rip.c
 @@ -27,7 +27,7 @@
  #elif defined (HAVE_SYS_VFS_H)
  #include 
@@ -11,3 +15,12 @@ $OpenBSD: patch-src_rip_c,v 1.8 2021/01/
  #include 
  #include 
  #endif
+@@ -663,7 +663,7 @@ void RemoveM3U(GripInfo *ginfo)
+ {
+   FILE *fp;
+   EncodeTrack enc_track;
+-  char testchar;
++  int testchar;
+   GString *str;
+ 
+   if (!ginfo->have_disc)

Re: [macppc, hppa] lang/guile2: provide a valid bootstrap to reduce build time

[Charlene Wendling]

On Thu, 22 Jul 2021 14:02:11 -0400
George Koehler  wrote:

> On Thu, 22 Jul 2021 10:40:46 +
> Charlene Wendling  wrote:
> 
> > On Thu, 22 Jul 2021 00:08:10 -0400
> > George Koehler  wrote:
> > 
> > > One issue: the port is extracting your g2strap-2.2.7-32be.tgz
> > > twice, in both WRKDIR and WRKSRC.  If you can fix that (maybe with
> > > EXTRACT_ONLY), the diff will be ok gkoehler@
> > 
> > Nicely spotted ; i've a diff that fixes that below.
> 
> ok gkoehler@

Thanks. I'll commit it this weekend if there are no objections (i've
slipped it in the current bulk).

> > Guile 3.x has apparently the same issue :(
> 
> Upstream Guile might not know that their big-endian stuff is
> broken.  I recall trying to mail their bug tracker about it, but
> I suspect that my mail never arrived.

It landed on their bugtracker:

https://debbugs.gnu.org/cgi/bugreport.cgi?bug=26854#11


> --George
> 
> > Index: Makefile
> > ===
> > RCS file: /cvs/ports/lang/guile2/Makefile,v
> > retrieving revision 1.27
> > diff -u -p -u -p -r1.27 Makefile
> > --- Makefile3 Aug 2020 07:27:46 -   1.27
> > +++ Makefile22 Jul 2021 10:21:49 -
> > @@ -1,7 +1,8 @@
> >  # $OpenBSD: Makefile,v 1.27 2020/08/03 07:27:46 jasper Exp $
> >  
> > -# When updating, check that x11/gnome/aisleriot MODGNOME_CPPFLAGS
> > references the -# proper guile2 includes directory
> > +# When updating, check that x11/gnome/aisleriot MODGNOME_CPPFLAGS
> > references +# the proper guile2 includes directory. Also, 32-bit BE
> > arches bootstraps may +# need to be rebuilt.
> >  
> >  COMMENT=   GNU's Ubiquitous Intelligent Language for
> > Extension
> >  # '
> > @@ -25,6 +26,12 @@ WANTLIB += c curses ffi gc gmp iconv int
> >  WANTLIB += unistring
> >  
> >  MASTER_SITES=  ${MASTER_SITE_GNU:=guile/}
> > +MASTER_SITES0= https://charlenew.xyz/distfiles/
> > +
> > +DISTFILES= ${DISTNAME}${EXTRACT_SUFX}
> > +SUPDISTFILES=  ${BOOTSTRAP_32BE}:0
> > +
> > +EXTRACT_ONLY=  ${DISTNAME}${EXTRACT_SUFX}
> >  
> >  # internal compiler error: in unify_nodes, at
> >  # tree-ssa-structalias.c:1231
> >  COMPILER = base-clang ports-gcc
> > @@ -49,9 +56,22 @@ CONFIGURE_ARGS=  --program-suffix=$
> > {V}
> >  # warning: format '%ji' expects type 'intmax_t', but argument 4
> >  # has type 'scm_t_intmax'
> >  CONFIGURE_ARGS +=  --disable-error-on-warning
> >  
> > -# powerpc: Prevent "Segmentation fault (core dumped)" during build.
> > -post-patch:
> > -   mv ${WRKSRC}/prebuilt/32-bit-big-endian{,-broken}
> > +.include 
> > +
> > +# The prebuilt bootstrap makes the build segfault on 32-bit BE
> > arches. +# Provide ours instead.
> > +BOOTSTRAP_32BE=g2strap-2.2.7-32be.tgz
> > +.if !${PROPERTIES:Mlp64} && ${PROPERTIES:Mbe}
> > +BOOTSTRAP_TARGET=  32-bit-big-endian
> > +BOOTSTRAP= ${BOOTSTRAP_32BE}
> > +DISTFILES+=${SUPDISTFILES}
> > +.endif
> > +
> > +post-extract:
> > +.if "${BOOTSTRAP_TARGET}" != ""
> > +   mv ${WRKSRC}/prebuilt/${BOOTSTRAP_TARGET}{,-broken}
> > +   cd ${WRKSRC} && tar xzf ${FULLDISTDIR}/${BOOTSTRAP}
> > +.endif
> >  
> >  post-install:
> > install -d ${PREFIX}/share/guile/site/${V}/
> > @@ -59,3 +79,20 @@ post-install:
> > mv ${PREFIX}/share/aclocal/guile{,${V}}.m4
> >  
> >  .include 
> > +
> > +# The below bits only matter on 32-bit BE archs for now.
> > +
> > +G2PKG= ${PACKAGE_REPOSITORY}/${MACHINE_ARCH}/all/$
> > {FULLPKGNAME}.tgz +
> > +.PHONY: g2strap
> > +
> > +g2strap: ${G2PKG}
> > +.if "${BOOTSTRAP_TARGET}" != ""
> > +   mkdir -p ${WRKBUILD}/g2strap/prebuilt/${BOOTSTRAP_TARGET}
> > +   cp -a $
> > {WRKBUILD}/bootstrap/{Makefile.*,ice-9,language,srfi,system} \
> > + ${WRKBUILD}/g2strap/prebuilt/${BOOTSTRAP_TARGET}
> > +   tar -C ${WRKBUILD}/g2strap -czf ${FULLDISTDIR}/$
> > {BOOTSTRAP} .
> > +   @echo "The bootstrap can be found at: " ${FULLDISTDIR}/$
> > {BOOTSTRAP} +.else
> > +   @echo "You don't need to build a bootstrap on $
> > {MACHINE_ARCH}." +.endif
> > Index: distinfo
> > ===
> > RCS file: /cvs/ports/lang/guile2/distinfo,v
> > retrieving revision 1.9
> > diff -u -p -u -p -r1.9 distinfo
> > --- distinfo3 Aug 2020 07:27:46 -   1.9
> > +++ distinfo22 Jul 2021 10:21:49 -
> > @@ -1,2 +1,4 @@
> > +SHA256 (g2strap-2.2.7-32be.tgz) =
> > MZmNCiGNR+GcVb73Bk7qnVGOn86juxflnEDEGzDjkZo= SHA256
> > (guile-2.2.7.tar.gz) = RLTF+74lfM3r6hhCAhLJs+kMPIalSSDYVUA5/GdpoAc=
> > +SIZE (g2strap-2.2.7-32be.tgz) = 4057585 SIZE (guile-2.2.7.tar.gz)
> > = 18129833
> > 
> 
> 
> -- 
> George Koehler 

Re: NEW PORT + PATCH: libvidstab and ffmpeg support

[Charlene Wendling]

Hi Marc,

On Wed, 21 Jul 2021 12:51:18 +0200
Marc Espie  wrote:

> vid.stab is a video stabilization library that interfaces with ffmpeg.
> 
> From the xml plugins in kdenlive, I suspect it will work out of the
> box there as well.
> 
> Here's the patch to activate it in ffmpeg. From simple tests, it
> seems to work as advertized.
> (I have bumped avfilter because it gains a new interface AND a new
> library dependency)
> 
> (The port itself is attached)

Trying to build on powerpc led to some issues:

- The SSE checks don't work with OpenBSD (see the terrible
  CMakeModules/FindSSE.cmake), and reported SSE2 support on powerpc,
  breaking the build searching for headers that indeed cannot be found.

  I've disabled all SSE checks on !amd64, so it does not try to build
  the port with instruction sets that are not available. As such,
  libvidstab still uses SSE2 on amd64, and nothing on all other arches.

  I know some i386 support SSE2, but not all. I'll let sthen@ decide.

- I've met several instance of char used to store (f)getc return values.
  It should be int instead, and as such the patch i bring fixes possible
  infinite loops on powerpc and arm*.

- I've removed optimization flags because the port was build with -O3.


I've attached a new port with these fixes.

On amd64, i've tested with 'ffplay -vf vidstabdetect=show=1 [...]',
where it seems to work.

On macppc, the library is built successfully, but given how video
playback is slow there, i would not dare to use that filter. I'm
building ffmpeg atm. I'll report back if issues arise, i've some time,
because it should be tested on more archs, ffmpeg is indeed an important
port.

Charlène.


Index: Makefile
===
RCS file: /cvs/ports/graphics/ffmpeg/Makefile,v
retrieving revision 1.210
diff -u -p -r1.210 Makefile
--- Makefile9 Jul 2021 10:43:26 -   1.210
+++ Makefile21 Jul 2021 10:47:56 -
@@ -4,7 +4,7 @@ COMMENT=audio/video converter and strea
 
 V= 4.4
 DISTNAME=  ffmpeg-${V}
-REVISION=  0
+REVISION=  1
 EPOCH= 1
 CATEGORIES=graphics multimedia
 MASTER_SITES=  https://ffmpeg.org/releases/
@@ -12,7 +12,7 @@ EXTRACT_SUFX= .tar.xz
 
 SHARED_LIBS=   avcodec 25.0 \
avdevice13.0 \
-   avfilter10.0 \
+   avfilter11.0 \
avformat22.0 \
avresample  2.0 \
avutil  15.0 \
@@ -31,7 +31,7 @@ WANTLIB+= SDL2 X11 Xext Xv aom ass bz2 c
 WANTLIB+=  freetype fribidi gsm iconv lzma m mp3lame opus
 WANTLIB+=  pthread sndio speex ssl theoradec theoraenc v4l2
 WANTLIB+=  vorbis vorbisenc vpx x264 x265 xcb xcb-shape xcb-shm
-WANTLIB+=  xcb-xfixes xml2 xvidcore z
+WANTLIB+=  xcb-xfixes xml2 xvidcore vidstab z
 
 COMPILER=  base-clang ports-gcc
 COMPILER_LANGS=c
@@ -56,6 +56,7 @@ LIB_DEPENDS=  archivers/bzip2 \
multimedia/libass \
multimedia/libtheora \
multimedia/libv4l \
+   multimedia/libvidstab \
multimedia/libvpx \
multimedia/x264 \
multimedia/x265 \
@@ -65,7 +66,7 @@ LIB_DEPENDS=  archivers/bzip2 \
 # inter-library dependencies for the current configuration
 LIBavcodec_EXTRALIBS=-L${LOCALBASE}/lib -laom -ldav1d -lgsm -liconv -llzma 
-lmp3lame -lopus -lspeex -ltheoradec -ltheoraenc -lvorbis -lvorbisenc -lvpx 
-lx264 -lx265 -lxvidcore -lm -lz -pthread
 LIBavdevice_EXTRALIBS=-L${LOCALBASE}/lib -L${X11BASE}/lib -lX11 -lXext -lXv 
-lxcb-shape -lxcb-shm -lxcb-xfixes -lxcb -lsndio -lv4l2 -lm
-LIBavfilter_EXTRALIBS=-L${LOCALBASE}/lib -L${X11BASE}/lib -lass -lfontconfig 
-lfreetype -lfribidi -lm -lz -pthread
+LIBavfilter_EXTRALIBS=-L${LOCALBASE}/lib -L${X11BASE}/lib -lass -lfontconfig 
-lfreetype -lfribidi -lvidstab -lm -lz -pthread
 LIBavformat_EXTRALIBS=-L${LOCALBASE}/lib -lbz2 -lxml2 -lssl -lcrypto -lm -lz
 LIBavresample_EXTRALIBS=-lm
 LIBavutil_EXTRALIBS=-lm -pthread
@@ -118,6 +119,7 @@ CONFIGURE_ARGS+=--enable-shared \
--enable-libxvid \
--enable-nonfree \
--enable-openssl \
+   --enable-libvidstab \
--extra-cflags="-I${LOCALBASE}/include -I${X11BASE}/include" \
--extra-libs="-L${LOCALBASE}/lib -L${X11BASE}/lib" \
--extra-ldsoflags="${LDLIBFLAGS}" \



libvidstab.tgz
Description: Binary data

Re: [macppc, hppa] lang/guile2: provide a valid bootstrap to reduce build time

[Charlene Wendling]

On Thu, 22 Jul 2021 00:08:10 -0400
George Koehler  wrote:

> On Tue, 20 Jul 2021 15:09:04 +
> Charlene Wendling  wrote:
> 
> > Hi,
> > 
> > Long time ago, gkoehler@ found out the guile2 32 bit big endian
> > bootstrap files were leading to segfaults during the build [0] on
> > powerpc.
> > 
> > Not using a bootstrap increases the building time by a factor of ~
> > 6. Currently, guile2 build time is 25 hours in the powerpc bulk. To
> > give you an idea of what this represents on this arch, it's one
> > hour more than x11/qt5/qtbase, or twice the time needed to build
> > devel/boost.
> > 
> > The people at Macports saw gkoehler@'s mail, and managed to build
> > bootstraps [1].
> > 
> > I took their recipe and adapted to our port system, so we can
> > provide a valid 32-bit big endian bootstrap that actually works,
> > based on what we do with lang/gcc/8, with the ability for anyone to
> > produce a bootstrap.
> > 
> > This reduces the build time from 19 to 3 hours on my Mac Mini G4!
> > [2]
> 
> One issue: the port is extracting your g2strap-2.2.7-32be.tgz twice,
> in both WRKDIR and WRKSRC.  If you can fix that (maybe with
> EXTRACT_ONLY), the diff will be ok gkoehler@

Nicely spotted ; i've a diff that fixes that below.

> You don't bump REVISION, and I don't see a reason to bump.
> We're building the same guile2, but building faster.
> 
> There is still no 64-bit-big-endian bootstrap, so powerpc64 (and
> probably mips64 and sparc64) still go the slow way.  I have not tried
> to make a 64-bit-big-endian bootstrap.  My build times for guile2 are
> about
> 
>5 h 45 m   on macppc Cube G4 (450 MHz) with your g2strap
>4 h 50 m   on powerpc64 (2700 to 3800 MHz)
> 
> I have avoided building guile2 for a long time, because the build was
> too slow.  Now that we have a bootstrap, I ran "make test".  2 tests
> fail on all archs because OpenBSD's getrlimit(2) doesn't have
> RLIMIT_AS, but macppc has a 3rd failure in the ffi (foreign function
> interface to call C from Guile):
> 
> bad return from expression `(f-s64-)': expected -20; got
> -873842568 bad return from expression `(f-u64-)': expected
> 40; got 3421124728 bad return from expression `(f-s64-u8
> 10)': expected -10; got -4294967294 ...
> FAIL: test-ffi
> 
> guile2 uses devel/libffi so this might be some libffi problem.  I have
> not yet studied issue; I know that lang/ruby uses libffi and wonder if
> I can reproduce the problem by writing ruby code.
>
> --George
> 
> > This has no impact on !{macppc,hppa}. I know it's quite a big change
> > given the limited scope of that diff. But actually guile moved to
> > 3.x, and 2.2.x is now considered legacy [3], so it's very unlikely
> > we see an update for this version anyway. I'm hoping they did not
> > reproduce the mistake for 3.x.
> > 

Guile 3.x has apparently the same issue :(

> > Comments and feedback are welcome,
> > 
> > Charlène.
> > 
> > 
> > [0] https://marc.info/?l=openbsd-ports&m=157611545827058&w=2
> > [1] https://trac.macports.org/ticket/54124#comment:78
> > [2] https://bin.charlenew.xyz/guile2-2.2.7.log
> > [3] https://www.gnu.org/software/guile/download/#releases
> > 

Index: Makefile
===
RCS file: /cvs/ports/lang/guile2/Makefile,v
retrieving revision 1.27
diff -u -p -u -p -r1.27 Makefile
--- Makefile3 Aug 2020 07:27:46 -   1.27
+++ Makefile22 Jul 2021 10:21:49 -
@@ -1,7 +1,8 @@
 # $OpenBSD: Makefile,v 1.27 2020/08/03 07:27:46 jasper Exp $
 
-# When updating, check that x11/gnome/aisleriot MODGNOME_CPPFLAGS references 
the
-# proper guile2 includes directory
+# When updating, check that x11/gnome/aisleriot MODGNOME_CPPFLAGS references
+# the proper guile2 includes directory. Also, 32-bit BE arches bootstraps may
+# need to be rebuilt.
 
 COMMENT=   GNU's Ubiquitous Intelligent Language for Extension
 # '
@@ -25,6 +26,12 @@ WANTLIB += c curses ffi gc gmp iconv int
 WANTLIB += unistring
 
 MASTER_SITES=  ${MASTER_SITE_GNU:=guile/}
+MASTER_SITES0= https://charlenew.xyz/distfiles/
+
+DISTFILES= ${DISTNAME}${EXTRACT_SUFX}
+SUPDISTFILES=  ${BOOTSTRAP_32BE}:0
+
+EXTRACT_ONLY=  ${DISTNAME}${EXTRACT_SUFX}
 
 # internal compiler error: in unify_nodes, at tree-ssa-structalias.c:1231
 COMPILER = base-clang ports-gcc
@@ -49,9 +56,22 @@ CONFIGURE_ARGS=  --program-suffix=${V}
 # warning: format '%ji' expects type 'intmax_t', but argument 4 has type 
'scm_t_intmax

[macppc, hppa] lang/guile2: provide a valid bootstrap to reduce build time

[Charlene Wendling]

Hi,

Long time ago, gkoehler@ found out the guile2 32 bit big endian
bootstrap files were leading to segfaults during the build [0] on
powerpc.

Not using a bootstrap increases the building time by a factor of ~ 6.
Currently, guile2 build time is 25 hours in the powerpc bulk. To give you
an idea of what this represents on this arch, it's one hour more than
x11/qt5/qtbase, or twice the time needed to build devel/boost.

The people at Macports saw gkoehler@'s mail, and managed to build
bootstraps [1].

I took their recipe and adapted to our port system, so we can provide
a valid 32-bit big endian bootstrap that actually works, based on
what we do with lang/gcc/8, with the ability for anyone to
produce a bootstrap.

This reduces the build time from 19 to 3 hours on my Mac Mini G4! [2]

This has no impact on !{macppc,hppa}. I know it's quite a big change
given the limited scope of that diff. But actually guile moved to 3.x,
and 2.2.x is now considered legacy [3], so it's very unlikely we
see an update for this version anyway. I'm hoping they did not
reproduce the mistake for 3.x.

Comments and feedback are welcome,

Charlène.


[0] https://marc.info/?l=openbsd-ports&m=157611545827058&w=2
[1] https://trac.macports.org/ticket/54124#comment:78
[2] https://bin.charlenew.xyz/guile2-2.2.7.log
[3] https://www.gnu.org/software/guile/download/#releases


Index: Makefile
===
RCS file: /cvs/ports/lang/guile2/Makefile,v
retrieving revision 1.27
diff -u -p -u -p -r1.27 Makefile
--- Makefile3 Aug 2020 07:27:46 -   1.27
+++ Makefile20 Jul 2021 14:53:41 -
@@ -1,7 +1,8 @@
 # $OpenBSD: Makefile,v 1.27 2020/08/03 07:27:46 jasper Exp $
 
-# When updating, check that x11/gnome/aisleriot MODGNOME_CPPFLAGS references 
the
-# proper guile2 includes directory
+# When updating, check that x11/gnome/aisleriot MODGNOME_CPPFLAGS references
+# the proper guile2 includes directory. Also, 32-bit BE arches bootstraps may
+# need to be rebuilt.
 
 COMMENT=   GNU's Ubiquitous Intelligent Language for Extension
 # '
@@ -25,6 +26,9 @@ WANTLIB += c curses ffi gc gmp iconv int
 WANTLIB += unistring
 
 MASTER_SITES=  ${MASTER_SITE_GNU:=guile/}
+MASTER_SITES0= https://charlenew.xyz/distfiles/
+DISTFILES= ${DISTNAME}${EXTRACT_SUFX}
+SUPDISTFILES=  ${BOOTSTRAP_32BE}:0
 
 # internal compiler error: in unify_nodes, at tree-ssa-structalias.c:1231
 COMPILER = base-clang ports-gcc
@@ -49,9 +53,22 @@ CONFIGURE_ARGS=  --program-suffix=${V}
 # warning: format '%ji' expects type 'intmax_t', but argument 4 has type 
'scm_t_intmax'
 CONFIGURE_ARGS +=  --disable-error-on-warning
 
-# powerpc: Prevent "Segmentation fault (core dumped)" during build.
-post-patch:
-   mv ${WRKSRC}/prebuilt/32-bit-big-endian{,-broken}
+.include 
+
+# The prebuilt bootstrap makes the build segfault on 32-bit BE arches.
+# Provide ours instead.
+BOOTSTRAP_32BE=g2strap-2.2.7-32be.tgz
+.if !${PROPERTIES:Mlp64} && ${PROPERTIES:Mbe}
+BOOTSTRAP_TARGET=  32-bit-big-endian
+BOOTSTRAP= ${BOOTSTRAP_32BE}
+DISTFILES+=${SUPDISTFILES}
+.endif
+
+post-extract:
+.if "${BOOTSTRAP_TARGET}" != ""
+   mv ${WRKSRC}/prebuilt/${BOOTSTRAP_TARGET}{,-broken}
+   cd ${WRKSRC} && tar xzf ${FULLDISTDIR}/${BOOTSTRAP}
+.endif
 
 post-install:
install -d ${PREFIX}/share/guile/site/${V}/
@@ -59,3 +76,20 @@ post-install:
mv ${PREFIX}/share/aclocal/guile{,${V}}.m4
 
 .include 
+
+# The below bits only matter on 32-bit BE archs for now.
+
+G2PKG= ${PACKAGE_REPOSITORY}/${MACHINE_ARCH}/all/${FULLPKGNAME}.tgz
+
+.PHONY: g2strap
+
+g2strap: ${G2PKG}
+.if "${BOOTSTRAP_TARGET}" != ""
+   mkdir -p ${WRKBUILD}/g2strap/prebuilt/${BOOTSTRAP_TARGET}
+   cp -a ${WRKBUILD}/bootstrap/{Makefile.*,ice-9,language,srfi,system} \
+ ${WRKBUILD}/g2strap/prebuilt/${BOOTSTRAP_TARGET}
+   tar -C ${WRKBUILD}/g2strap -czf ${FULLDISTDIR}/${BOOTSTRAP} .
+   @echo "The bootstrap can be found at: " ${FULLDISTDIR}/${BOOTSTRAP}
+.else
+   @echo "You don't need to build a bootstrap on ${MACHINE_ARCH}."
+.endif
Index: distinfo
===
RCS file: /cvs/ports/lang/guile2/distinfo,v
retrieving revision 1.9
diff -u -p -u -p -r1.9 distinfo
--- distinfo3 Aug 2020 07:27:46 -   1.9
+++ distinfo20 Jul 2021 14:53:41 -
@@ -1,2 +1,4 @@
+SHA256 (g2strap-2.2.7-32be.tgz) = MZmNCiGNR+GcVb73Bk7qnVGOn86juxflnEDEGzDjkZo=
 SHA256 (guile-2.2.7.tar.gz) = RLTF+74lfM3r6hhCAhLJs+kMPIalSSDYVUA5/GdpoAc=
+SIZE (g2strap-2.2.7-32be.tgz) = 4057585
 SIZE (guile-2.2.7.tar.gz) = 18129833

Re: [macppc, base libexecinfo] Unbreak editors/emacs

[Charlene Wendling]

On Tue, 13 Jul 2021 13:54:05 +0200
Jeremie Courreges-Anglas  wrote:

> On Tue, Jul 13 2021, Charlene Wendling  wrote:
> > On Tue, 13 Jul 2021 01:51:28 +0200
> > Jeremie Courreges-Anglas  wrote:
> >
> >> On Mon, Jul 12 2021, Charlene Wendling 
> >> wrote:
> >> > Hi,
> >> >
> >> > This is the first bulk with base libexecinfo.
> >> >
> >> >> https://bin.charlenew.xyz/emacs_bulk_failure.log
> >> >
> >> > The last port commit brought libexecinfo support to emacs, and
> >> > leads to segfault during the build on macppc. I attach the
> >> > backtrace of a simple 'temacs' invocation, showing that it may
> >> > not be emacs' fault.
> >> >
> >> > I can't fix the libexecinfo issue.
> >> >
> >> > But the build works fine with the below diff [0] [1] [2]. It
> >> > reverts the latest commit on macppc, while leaving the other
> >> > arches as it is currently. As such no REVISION bump is needed.
> >> >
> >> > Comments/feedback are welcome,
> >> 
> >> ok jca@ for this diff but please amend the comment unless you're
> >> sure the problem comes from Lisp code.  AFAIK the emacs C code
> >> calls backtrace() early.  I'd suggest
> >> 
> >>   # backtrace() crashes at build time on powerpc
> >
> > I'm not sure, so i'll change the comment.
> >
> >> I'm slightly worried regarding libunwind status on powerpc.  What
> >> does regress/gnu/lib/libexecinfo say?
> >
> > This cannot be run on macppc:
> >
> > -->8--
> >  run-regress-skiparch 
> > # No unwind info on this arch
> > SKIPPED
> > --8<--
> >
> > It's limited to amd64 or llvm>=12, as mentioned in the regress'
> > Makefile cvs log.
> 
> Well could you please test with the diff below applied?
> 
> 
> Index: Makefile
> ===
> RCS file: /d/cvs/src/regress/gnu/lib/libexecinfo/Makefile,v
> retrieving revision 1.2
> diff -u -p -r1.2 Makefile
> --- Makefile  18 Jun 2021 14:09:31 -  1.2
> +++ Makefile  13 Jul 2021 11:49:56 -
> @@ -6,7 +6,7 @@
>  
>  # Only amd64 has unwind info by default in llvm11. llvm12 adds more
>  # arches.
>  CLANG_VERSION!=clang --version | head -1 | grep -E -o 'version
> [[:digit:]]+' | cut -d' ' -f 2 -CLANG_GE_12!=if [ ${CLANG_VERSION}
> -ge 12 ] ; then echo 1 ; fi +CLANG_GE_12!=if [ ${CLANG_VERSION} -ge
> 11 ] ; then echo 1 ; fi .if ${MACHINE} == "amd64" || ${CLANG_GE_12}
> == "1" 
>  PROG=t_backtrace
> 
> -- 
> jca | PGP : 0x1524E7EE / 5135 92C1 AD36 5293 2BDF  DDCC 0DFA 74AE
> 1524 E7EE
> 

Here you are. I did the same thing this morning but i wasn't very
sure if it was valid.

If i pass DEBUG="-O0 -g" to the Makefile, it does not segfault by the way. 

(gdb) run   
 
Starting program: /usr/src/regress/gnu/lib/libexecinfo/t_backtrace -r 1 
 

 
Program received signal SIGSEGV, Segmentation fault.
 
libunwind::CFI_Parser::findFDE (addressSpace=..., 
pc=3122944028,   
ehSectionStart=3123060828, sectionLength=, 
fdeHint=, fdeInfo=0xfffcbce8,   
cieInfo=0xfffcbcc8) at /usr/src/gnu/llvm/libunwind/src/DwarfParser.hpp:223  
 
223 pint_t cfiLength = addressSpace.get32(p);   
 
(gdb) bt
 
#0  libunwind::CFI_Parser::findFDE 
(addressSpace=..., pc=3122944028,   
ehSectionStart=3123060828, sectionLength=, 
fdeHint=, fdeInfo=0xfffcbce8, 
cieInfo=0xfffcbcc8) at /usr/src/gnu/llvm/libunwind/src/DwarfParser.hpp:223
#1  0xba24cfd0 in libunwind::UnwindCursor::getInfoFromDwarfSection (this=0xfffcc920, 
pc=3122944028, sects=..., fdeSectionOffsetHint=0)
at /usr/src/gnu/llvm/libunwind/src/UnwindCursor.hpp:1518
#2  0xba249ce4 in libunwind::UnwindCursor::setInfoBasedOnIPRegister (this=0xfffcc920, 
isReturnAddress=)
at /usr/src/gnu/llvm/libunwind/src/UnwindCursor.hpp:1927
#3  0xba2480a8 in __unw_init_local (cursor=0xfffcc920, context=0xfffcc578)
at /usr/src/gnu/llvm/libunwind/src/libunwind.cpp:76
#4  0xba24582c in _Unwind_Backtrace (callback=0xba24ec54 , 
ref=0xfffccd28)
at /usr/src/gnu/llvm/libunwind/src/UnwindLevel1-gcc-ext.c:112
#5  0xba24ebd4 in backtrace (arr=, len=)
at /usr/src/gnu/lib/libexecinfo/unwind.c:69
#6  0x2d1e070c in myfunc3 () 

Re: [macppc, base libexecinfo] Unbreak editors/emacs

[Charlene Wendling]

On Tue, 13 Jul 2021 01:51:28 +0200
Jeremie Courreges-Anglas  wrote:

> On Mon, Jul 12 2021, Charlene Wendling  wrote:
> > Hi,
> >
> > This is the first bulk with base libexecinfo.
> >
> >> https://bin.charlenew.xyz/emacs_bulk_failure.log
> >
> > The last port commit brought libexecinfo support to emacs, and
> > leads to segfault during the build on macppc. I attach the
> > backtrace of a simple 'temacs' invocation, showing that it may not
> > be emacs' fault.
> >
> > I can't fix the libexecinfo issue.
> >
> > But the build works fine with the below diff [0] [1] [2]. It reverts
> > the latest commit on macppc, while leaving the other arches as it
> > is currently. As such no REVISION bump is needed.
> >
> > Comments/feedback are welcome,
> 
> ok jca@ for this diff but please amend the comment unless you're sure
> the problem comes from Lisp code.  AFAIK the emacs C code calls
> backtrace() early.  I'd suggest
> 
>   # backtrace() crashes at build time on powerpc

I'm not sure, so i'll change the comment.

> I'm slightly worried regarding libunwind status on powerpc.  What does
> regress/gnu/lib/libexecinfo say?

This cannot be run on macppc:

-->8--
 run-regress-skiparch 
# No unwind info on this arch
SKIPPED
--8<--

It's limited to amd64 or llvm>=12, as mentioned in the regress'
Makefile cvs log.

The bulk is almost done, and i had to test a number of ports on my G3
that used to have ports-libexecinfo as a depend and met no such issues,
so i hope for the best.

> > Charlène.
> >
> > [0] https://bin.charlenew.xyz/emacs.log
> > [1] https://bin.charlenew.xyz/emacs.gtk2.log
> > [2] https://bin.charlenew.xyz/emacs.no_x11.log
> >
> > Index: Makefile
> > ===
> > RCS file: /cvs/ports/editors/emacs/Makefile,v
> > retrieving revision 1.99
> > diff -u -p -u -p -r1.99 Makefile
> > --- Makefile12 Jun 2021 23:17:50 -  1.99
> > +++ Makefile12 Jul 2021 18:59:58 -
> > @@ -15,7 +15,7 @@ MAINTAINER=   Jeremie
> > Courreges-Anglas  >  # GPLv3+
> >  PERMIT_PACKAGE=Yes
> >  
> > -WANTLIB =  c curses execinfo jansson gmp gnutls m
> > pthread xml2 z +WANTLIB =   c curses jansson gmp
> > gnutls m pthread xml2 z 
> >  MASTER_SITES=  ${MASTER_SITE_GNU:=emacs/}
> >  EXTRACT_SUFX=  .tar.xz
> > @@ -98,6 +98,13 @@ WANTLIB += xcb
> >  . else
> >  ERRORS+=   "Fatal: Conflicting flavor: ${FLAVOR}"
> >  . endif
> > +.endif
> > +
> > +# lisp code breaks the build with libexecinfo support
> > +.if ${MACHINE_ARCH} == "powerpc"
> > +CONFIGURE_ENV +=   ac_cv_header_execinfo_h=no
> > +.else
> > +WANTLIB += execinfo
> >  .endif
> >  
> >  post-install:
> >
> >
> 
> -- 
> jca | PGP : 0x1524E7EE / 5135 92C1 AD36 5293 2BDF  DDCC 0DFA 74AE
> 1524 E7EE
> 

[macppc, base libexecinfo] Unbreak editors/emacs

[Charlene Wendling]

Hi,

This is the first bulk with base libexecinfo.

> https://bin.charlenew.xyz/emacs_bulk_failure.log

The last port commit brought libexecinfo support to emacs, and leads to
segfault during the build on macppc. I attach the backtrace of a simple
'temacs' invocation, showing that it may not be emacs' fault.

I can't fix the libexecinfo issue.

But the build works fine with the below diff [0] [1] [2]. It reverts
the latest commit on macppc, while leaving the other arches as it
is currently. As such no REVISION bump is needed.

Comments/feedback are welcome,

Charlène.

[0] https://bin.charlenew.xyz/emacs.log
[1] https://bin.charlenew.xyz/emacs.gtk2.log
[2] https://bin.charlenew.xyz/emacs.no_x11.log


Index: Makefile
===
RCS file: /cvs/ports/editors/emacs/Makefile,v
retrieving revision 1.99
diff -u -p -u -p -r1.99 Makefile
--- Makefile12 Jun 2021 23:17:50 -  1.99
+++ Makefile12 Jul 2021 18:59:58 -
@@ -15,7 +15,7 @@ MAINTAINER=   Jeremie Courreges-Anglas 

temacs_backtrace_backtrace.log
Description: Binary data

Re: [update] www/snownews 1.6.10 -> 1.8

[Charlene Wendling]

On Mon, 5 Jul 2021 17:15:09 -0400
"trondd"  wrote:

> On Mon, July 5, 2021 9:53 am, Charlene Wendling wrote:
> > Hi,
> >
> > The below diff updates snownews to 1.8 with a new repo owner and
> > finally a lot of modern features [0]! I tend to separate upstream vs
> > ports changes, but it's unpractical here:
> >
> >> Added Atom feed support
> >   Modify COMMENT and DESCR accordingly. While here, update to valid
> >   feeds and a mix of RSS and Atom. HTTPS is used where applicable.
> >> Use curl for networking
> >   Add proper LIB_DEPENDS, remove README now that HTTPS is supported.
> >   There is a year 2038 printf() warning, but it would require to
> >   sprinkle CURLINFO_FILETIME_T instead of CURLINFO_FILETIME in many
> >   places first. Also, refresh WANTLIB, and remove the netio.c patch
> >   accordingly.
> >> Use the OPML format to store the feed list, and XDG dirs
> >   Snownews-1.8 will migrate your urls file to the OPML format in
> >   ~/.config/snownews where you can find other settings (key
> > bindings, colors etc.). The OPML conversion tools (snow2opml,
> > opml2snow) have been removed, so we don't need p5-XML-LibXML as a
> > RUN_DEPENDS.
> >> Use UTF-8 everywhere (including manpages)
> >   The upstream configuration system couldn't find our ncurses
> > library with wide chars support, so i needed to tweak CFLAGS and
> > LDFLAGS.
> >> Deal with lack of verbosity for the fake stage
> >   The most painful part of that update has been the fake stage,
> >   because upstream hides issued commands, so i preferred to
> >   unsilence them.
> >
> > All the rest is mostly updating old patches.
> >
> > Testing:
> >
> > It builds and works fine on amd64 and macppc.
> >
> > Note that feed items' text enclosed in '' is
> > not displayed (yet?).
> >
> > Comments and feedback are welcome,
> >
> > Charlène.
> >
> >
> > [0] https://github.com/msharov/snownews/releases
> >
> 
> Thanks.  I use snownews.  It converted my configs and works as
> epxected. I had to go back in and clean up all the calls to external
> conversion tools that the previous version needed but that's to be
> expected.  Worked either way.
> 
> Tim.
> 

You're welcome! I tried beforehand with a few feeds, but was not 100%
sure how the migration would really fare for current users. The commit
message will mention this, but given it works in either way it should be
smooth for everyone.

Charlène.

[update] www/snownews 1.6.10 -> 1.8

[Charlene Wendling]

Hi,

The below diff updates snownews to 1.8 with a new repo owner and
finally a lot of modern features [0]! I tend to separate upstream vs
ports changes, but it's unpractical here:

> Added Atom feed support
  Modify COMMENT and DESCR accordingly. While here, update to valid
  feeds and a mix of RSS and Atom. HTTPS is used where applicable.
> Use curl for networking
  Add proper LIB_DEPENDS, remove README now that HTTPS is supported.
  There is a year 2038 printf() warning, but it would require to
  sprinkle CURLINFO_FILETIME_T instead of CURLINFO_FILETIME in many
  places first. Also, refresh WANTLIB, and remove the netio.c patch
  accordingly.
> Use the OPML format to store the feed list, and XDG dirs
  Snownews-1.8 will migrate your urls file to the OPML format in
  ~/.config/snownews where you can find other settings (key bindings,
  colors etc.). The OPML conversion tools (snow2opml, opml2snow) have
  been removed, so we don't need p5-XML-LibXML as a RUN_DEPENDS.
> Use UTF-8 everywhere (including manpages)
  The upstream configuration system couldn't find our ncurses library
  with wide chars support, so i needed to tweak CFLAGS and LDFLAGS. 
> Deal with lack of verbosity for the fake stage
  The most painful part of that update has been the fake stage,
  because upstream hides issued commands, so i preferred to 
  unsilence them.

All the rest is mostly updating old patches.

Testing:

It builds and works fine on amd64 and macppc.

Note that feed items' text enclosed in '' is
not displayed (yet?).

Comments and feedback are welcome,

Charlène.


[0] https://github.com/msharov/snownews/releases


Index: Makefile
===
RCS file: /cvs/ports/www/snownews/Makefile,v
retrieving revision 1.41
diff -u -p -u -p -r1.41 Makefile
--- Makefile21 Nov 2020 22:03:13 -  1.41
+++ Makefile5 Jul 2021 13:01:43 -
@@ -1,13 +1,11 @@
 # $OpenBSD: Makefile,v 1.41 2020/11/21 22:03:13 kmos Exp $
 
-COMMENT=   text mode rss newsreader
+COMMENT=   text mode rss and atom newsreader
 CATEGORIES=www
 
-GH_ACCOUNT=kouya
+GH_ACCOUNT=msharov
 GH_PROJECT=snownews
-GH_TAGNAME=1.6.10
-
-HOMEPAGE=  https://github.com/kouya/snownews
+GH_TAGNAME=v1.8
 
 # GPLv3 only
 PERMIT_PACKAGE=Yes
@@ -15,21 +13,25 @@ PERMIT_PACKAGE= Yes
 # C11
 COMPILER=  base-clang ports-gcc
 
-WANTLIB += c curses iconv intl xml2 z
+WANTLIB += c crypto curl curses iconv intl xml2
 
 NO_TEST=   Yes
 
 USE_GMAKE= Yes
 
 BUILD_DEPENDS= devel/gettext,-tools
-RUN_DEPENDS=   textproc/p5-XML-LibXML
 LIB_DEPENDS=   devel/gettext,-runtime \
+   net/curl \
textproc/libxml
 
 CONFIGURE_STYLE=   simple
-CONFIGURE_ARGS=--prefix="\$${PREFIX}" \
-   --mandir="\$${PREFIX}/man" \
-   --builddir=${WRKDIR}
-MAKE_ENV=  COPTFLAGS="${CFLAGS}"
+CONFIGURE_ARGS+=   --builddir=${WRKDIR} --mandir=${PREFIX}/man
+MAKE_ENV=  cflags="${CFLAGS}" ldflags="${LDFLAGS}"
+
+# Needed for cchar_t, setcchar() and getcchar()
+CFLAGS+=   -D_XOPEN_SOURCE_EXTENDED
+# Fix linking errors due to upstream build system
+# not being able to find proper LDFLAGS for ncurses
+LDFLAGS+=  -lncursesw
 
 .include 
Index: distinfo
===
RCS file: /cvs/ports/www/snownews/distinfo,v
retrieving revision 1.17
diff -u -p -u -p -r1.17 distinfo
--- distinfo20 Nov 2020 20:48:11 -  1.17
+++ distinfo5 Jul 2021 13:01:43 -
@@ -1,2 +1,2 @@
-SHA256 (snownews-1.6.10.tar.gz) = jHgGeu914oPfSzzKHJZlh7ZlTp6Eo+bl64vdWCl5kkI=
-SIZE (snownews-1.6.10.tar.gz) = 189715
+SHA256 (snownews-1.8.tar.gz) = kNJhGz46ALwUqIaTZdNmrR2rF+oWh4V0QBWfxxN8O+0=
+SIZE (snownews-1.8.tar.gz) = 154813
Index: patches/patch-Config_mk_in
===
RCS file: /cvs/ports/www/snownews/patches/patch-Config_mk_in,v
retrieving revision 1.1
diff -u -p -u -p -r1.1 patch-Config_mk_in
--- patches/patch-Config_mk_in  20 Nov 2020 20:48:11 -  1.1
+++ patches/patch-Config_mk_in  5 Jul 2021 13:01:43 -
@@ -3,25 +3,25 @@ $OpenBSD: patch-Config_mk_in,v 1.1 2020/
 Index: Config.mk.in
 --- Config.mk.in.orig
 +++ Config.mk.in
-@@ -21,7 +21,7 @@ PREFIX   := @prefix@
- BINDIR:= @bindir@
- LOCALEPATH:= @localedir@
- MANPATH   := @mandir@
--BUILDDIR  := @builddir@/${NAME}
-+BUILDDIR  := @builddir@/build
- PKGDIR:= @pkgdir@
+@@ -19,20 +19,13 @@ mandir := @mandir@
+ man1dir   := @man1dir@
+ localedir := @localedir@
+ TMPDIR:= @TMPDIR@
+-builddir  := @builddir@/${name}
++builddir  := @builddir@/build
  O := .o/
  
-@@ -29,11 +29,5 @@ O   := .o/
+  Compiler options 
  
- CFLAGS:= -Wall -Wextra -Wredundant-decls -Wsha

Re: Mouse cursor size in chromium

[Charlene Wendling]

Hi,

I had a similar problem since i wanted a bigger cursor, and use
iridium.

I dropped .Xdefaults/.Xresources Xcursor.* properties and put
this in my .xsession (see `pkg_info xcursor-themes`)

--8<--
export XCURSOR_THEME=Adwaita
export XCURSOR_PATH="/usr/local/share/icons/"
export XCURSOR_SIZE=32
-->8--

My ~/.config/gtk-3.0/settings.ini was already set to

--8<--
gtk-cursor-theme-size=32
--8<--

Since then, cursors are of the same size and theme everywhere.

Charlène.

On Sun, 04 Jul 2021 07:18:41 -0700
Greg Steuck  wrote:

> I'm running chromium Version 91.0.4472.77 (Official Build) (64-bit)
> on a 6.9-mostly-current amd64. The machine has a 4k panel, so I added
> these to .Xdefaults:
> 
> Xft.dpi: 96
> Xcursor.size: 96
> Xcursor.theme: Adwaita
> 
> I launch xmonad from xdm using .xsession, so no display environment.
> Almost everywhere (e.g. root window, emacs, xterm, firefox) the mouse
> cursor is a nice large arrow.
> 
> Sadly, in chromium even with --force-device-scale-factor=2 the mouse
> cursor is imperceptibly tiny. I didn't find anything useful in
> chrome://settings and web searches are frustratingly useless. Anybody
> have a recipe?
> 
> Thanks
> Greg
> 

Re: luajit update to 2.0 git head

[Charlene Wendling]

On Sat,  3 Jul 2021 16:48:21 +
Charlene Wendling  wrote:

> Hi,
> 
> On Sat, 3 Jul 2021 15:21:55 +0100
> Stuart Henderson  wrote:
> 
>  
> > I've built all dependent ports on amd64 with no problems, and done
> > runtime tests of rspamd and luakit so far. I'll do some tests with
> > i386 too. Does anyone want to do other tests? (The powerpc-related
> > patch removed in the diff was committed upstream; I have not tested
> > it on ppc though).
> 
> It builds fine on powerpc with your diff and ports-gcc, but actually
> base-clang can be used instead. 
> 
> I've updated your diff with the required bits. It appears that luajit
> "wants" gcc>=4.3 to build on powerpc, but our base-clang poses as
> 4.2.1. Other arches are not impacted by the issue because the gcc
> version requirement is lower. I've disabled gcc version checking for
> all base-clang arches.
> 
> It builds [0] and works fine with hexchat. By the way, luajit is not
> enabled in rspamd for powerpc as well.
> 
> Charlène.
> 

(sigh, forgot the log)


[0] https://bin.charlenew.xyz/luajit_macppc_clang.log


Index: Makefile
===
RCS file: /cvs/ports/lang/luajit/Makefile,v
retrieving revision 1.31
diff -u -p -u -p -r1.31 Makefile
--- Makefile10 Aug 2020 00:56:40 -  1.31
+++ Makefile3 Jul 2021 16:40:02 -
@@ -1,38 +1,38 @@
 # $OpenBSD: Makefile,v 1.31 2020/08/10 00:56:40 jca Exp $
 
-# ../src/luajit2/src/lj_arch.h:55:2: error: #error "No support for
-# this architecture (yet)"
-ONLY_FOR_ARCHS = powerpc i386 amd64
-# arm needs missing EABI symbols
+ONLY_FOR_ARCHS = arm amd64 i386 powerpc
+# if changing the above and it actually builds, switch to using
+#  luajit in mail/rspamd
+# - arm added speculatively, so not added in rspamd yet.
+#
+# no aarch64 support in luajit 2.0 branch
 # games/tome4 is using embedded copy of luajit
-# if changing, sync in mail/rspamd
 
 SHARED_LIBS += luajit-${MODLUA_VERSION} 1.0 # 2.0.5
 
+# "Releases are only made occasionally. You're strongly encouraged to follow
+# the git branches"
+GH_ACCOUNT =   LuaJIT
+GH_PROJECT =   LuaJIT
+GH_COMMIT =98f95f69180d48ce49289d6428b46a9ccdd67a46
+
 COMMENT =  just-in-time compiler for Lua
 V =2.0.5
-REVISION = 2
-DISTNAME = LuaJIT-${V}
+DISTNAME = LuaJIT-${V}pl20210608
 PKGNAME =  ${DISTNAME:L}
 
 CATEGORIES =   lang
 
-HOMEPAGE = http://luajit.org/
+HOMEPAGE = https://luajit.org/
 
 # MIT
 PERMIT_PACKAGE =   Yes
 
 WANTLIB =  c m
 
-MASTER_SITES = ${HOMEPAGE}download/
-
 MODULES =  lang/lua
 MODLUA_SA= Yes
 MODLUA_VERSION=5.1
-
-MODULES += gcc4
-MODGCC4_ARCHS =powerpc
-MODGCC4_LANGS =c
 
 MAKE_FLAGS += \
CC="${CC}" \
Index: distinfo
===
RCS file: /cvs/ports/lang/luajit/distinfo,v
retrieving revision 1.8
diff -u -p -u -p -r1.8 distinfo
--- distinfo5 May 2017 10:21:47 -   1.8
+++ distinfo3 Jul 2021 16:40:02 -
@@ -1,2 +1,2 @@
-SHA256 (LuaJIT-2.0.5.tar.gz) = h0sfgpfGl4IfVh+bc7V//UGe2PQnjILgW0iAbTDB6Xk=
-SIZE (LuaJIT-2.0.5.tar.gz) = 849845
+SHA256 (LuaJIT-2.0.5pl20210608-98f95f69.tar.gz) = 
wmjMMUjuUzRqfRu4plfFqD7otJefP7Wwj67uj4mDfdQ=
+SIZE (LuaJIT-2.0.5pl20210608-98f95f69.tar.gz) = 844346
Index: patches/patch-Makefile
===
RCS file: /cvs/ports/lang/luajit/patches/patch-Makefile,v
retrieving revision 1.6
diff -u -p -u -p -r1.6 patch-Makefile
--- patches/patch-Makefile  5 May 2017 10:21:47 -   1.6
+++ patches/patch-Makefile  3 Jul 2021 16:40:02 -
@@ -1,6 +1,8 @@
 $OpenBSD: patch-Makefile,v 1.6 2017/05/05 10:21:47 jsg Exp $
 Makefile.orig  Tue May  2 04:11:00 2017
-+++ Makefile   Tue May  2 11:40:27 2017
+
+Index: Makefile
+--- Makefile.orig
 Makefile
 @@ -17,7 +17,7 @@ MAJVER=  2
  MINVER=  0
  RELVER=  5
@@ -15,7 +17,7 @@ $OpenBSD: patch-Makefile,v 1.6 2017/05/0
$(RM) $(INSTALL_TSYM) $(INSTALL_DYN) $(INSTALL_SHORT1) $(INSTALL_SHORT2)
cd src && test -f $(FILE_SO) && \
 -$(INSTALL_X) $(FILE_SO) $(INSTALL_DYN) && \
--$(LDCONFIG) $(INSTALL_LIB) && \
+-( $(LDCONFIG) $(INSTALL_LIB) || : ) && \
 -$(SYMLINK) $(INSTALL_SONAME) $(INSTALL_SHORT1) && \
 -$(SYMLINK) $(INSTALL_SONAME) $(INSTALL_SHORT2) || :
 +$(INSTALL_X) $(FILE_SO) $(INSTALL_DYN) || :
Index: patches/patch-src_Makefile
===
RCS file: /cvs/ports/lang/luajit/patches/patch-src_Makefile,v
retrieving revision 1.6
diff -u -p -u -p -r1.6 patch-src_Makefile
--- patches/patch-src_Makefile  5 May 2017 10:21:47 -   1.6
+++ patches/patch-src_Makefile  3 Jul 2

Re: luajit update to 2.0 git head

[Charlene Wendling]

Hi,

On Sat, 3 Jul 2021 15:21:55 +0100
Stuart Henderson  wrote:

 
> I've built all dependent ports on amd64 with no problems, and done
> runtime tests of rspamd and luakit so far. I'll do some tests with
> i386 too. Does anyone want to do other tests? (The powerpc-related
> patch removed in the diff was committed upstream; I have not tested
> it on ppc though).

It builds fine on powerpc with your diff and ports-gcc, but actually
base-clang can be used instead. 

I've updated your diff with the required bits. It appears that luajit
"wants" gcc>=4.3 to build on powerpc, but our base-clang poses as
4.2.1. Other arches are not impacted by the issue because the gcc
version requirement is lower. I've disabled gcc version checking for all
base-clang arches.

It builds [0] and works fine with hexchat. By the way, luajit is not
enabled in rspamd for powerpc as well.

Charlène.


Index: Makefile
===
RCS file: /cvs/ports/lang/luajit/Makefile,v
retrieving revision 1.31
diff -u -p -u -p -r1.31 Makefile
--- Makefile10 Aug 2020 00:56:40 -  1.31
+++ Makefile3 Jul 2021 16:40:02 -
@@ -1,38 +1,38 @@
 # $OpenBSD: Makefile,v 1.31 2020/08/10 00:56:40 jca Exp $
 
-# ../src/luajit2/src/lj_arch.h:55:2: error: #error "No support for
-# this architecture (yet)"
-ONLY_FOR_ARCHS = powerpc i386 amd64
-# arm needs missing EABI symbols
+ONLY_FOR_ARCHS = arm amd64 i386 powerpc
+# if changing the above and it actually builds, switch to using
+#  luajit in mail/rspamd
+# - arm added speculatively, so not added in rspamd yet.
+#
+# no aarch64 support in luajit 2.0 branch
 # games/tome4 is using embedded copy of luajit
-# if changing, sync in mail/rspamd
 
 SHARED_LIBS += luajit-${MODLUA_VERSION} 1.0 # 2.0.5
 
+# "Releases are only made occasionally. You're strongly encouraged to follow
+# the git branches"
+GH_ACCOUNT =   LuaJIT
+GH_PROJECT =   LuaJIT
+GH_COMMIT =98f95f69180d48ce49289d6428b46a9ccdd67a46
+
 COMMENT =  just-in-time compiler for Lua
 V =2.0.5
-REVISION = 2
-DISTNAME = LuaJIT-${V}
+DISTNAME = LuaJIT-${V}pl20210608
 PKGNAME =  ${DISTNAME:L}
 
 CATEGORIES =   lang
 
-HOMEPAGE = http://luajit.org/
+HOMEPAGE = https://luajit.org/
 
 # MIT
 PERMIT_PACKAGE =   Yes
 
 WANTLIB =  c m
 
-MASTER_SITES = ${HOMEPAGE}download/
-
 MODULES =  lang/lua
 MODLUA_SA= Yes
 MODLUA_VERSION=5.1
-
-MODULES += gcc4
-MODGCC4_ARCHS =powerpc
-MODGCC4_LANGS =c
 
 MAKE_FLAGS += \
CC="${CC}" \
Index: distinfo
===
RCS file: /cvs/ports/lang/luajit/distinfo,v
retrieving revision 1.8
diff -u -p -u -p -r1.8 distinfo
--- distinfo5 May 2017 10:21:47 -   1.8
+++ distinfo3 Jul 2021 16:40:02 -
@@ -1,2 +1,2 @@
-SHA256 (LuaJIT-2.0.5.tar.gz) = h0sfgpfGl4IfVh+bc7V//UGe2PQnjILgW0iAbTDB6Xk=
-SIZE (LuaJIT-2.0.5.tar.gz) = 849845
+SHA256 (LuaJIT-2.0.5pl20210608-98f95f69.tar.gz) = 
wmjMMUjuUzRqfRu4plfFqD7otJefP7Wwj67uj4mDfdQ=
+SIZE (LuaJIT-2.0.5pl20210608-98f95f69.tar.gz) = 844346
Index: patches/patch-Makefile
===
RCS file: /cvs/ports/lang/luajit/patches/patch-Makefile,v
retrieving revision 1.6
diff -u -p -u -p -r1.6 patch-Makefile
--- patches/patch-Makefile  5 May 2017 10:21:47 -   1.6
+++ patches/patch-Makefile  3 Jul 2021 16:40:02 -
@@ -1,6 +1,8 @@
 $OpenBSD: patch-Makefile,v 1.6 2017/05/05 10:21:47 jsg Exp $
 Makefile.orig  Tue May  2 04:11:00 2017
-+++ Makefile   Tue May  2 11:40:27 2017
+
+Index: Makefile
+--- Makefile.orig
 Makefile
 @@ -17,7 +17,7 @@ MAJVER=  2
  MINVER=  0
  RELVER=  5
@@ -15,7 +17,7 @@ $OpenBSD: patch-Makefile,v 1.6 2017/05/0
$(RM) $(INSTALL_TSYM) $(INSTALL_DYN) $(INSTALL_SHORT1) $(INSTALL_SHORT2)
cd src && test -f $(FILE_SO) && \
 -$(INSTALL_X) $(FILE_SO) $(INSTALL_DYN) && \
--$(LDCONFIG) $(INSTALL_LIB) && \
+-( $(LDCONFIG) $(INSTALL_LIB) || : ) && \
 -$(SYMLINK) $(INSTALL_SONAME) $(INSTALL_SHORT1) && \
 -$(SYMLINK) $(INSTALL_SONAME) $(INSTALL_SHORT2) || :
 +$(INSTALL_X) $(FILE_SO) $(INSTALL_DYN) || :
Index: patches/patch-src_Makefile
===
RCS file: /cvs/ports/lang/luajit/patches/patch-src_Makefile,v
retrieving revision 1.6
diff -u -p -u -p -r1.6 patch-src_Makefile
--- patches/patch-src_Makefile  5 May 2017 10:21:47 -   1.6
+++ patches/patch-src_Makefile  3 Jul 2021 16:40:02 -
@@ -1,7 +1,9 @@
 $OpenBSD: patch-src_Makefile,v 1.6 2017/05/05 10:21:47 jsg Exp $
 src/Makefile.orig  Tue May  2 04:11:00 2017
-+++ src/Makefile   Tue May  2 11:40:27 2017
-@@ -326,6 +326,10 @@ else
+
+Index: src/Makefile
+--- src/Makefile.orig
 src/Makefile
+@@ -331,6 +331,10 @@ else
ifeq (GNU/kFreeBSD,$(TARGET_SYS))
  TARGET_XLIBS+= -ldl
endif
@@ -12,7 +14,7 @@ 

Re: [update] games/hyperrogue 11.3o -> 12.0b

[Charlene Wendling]

On Tue, 29 Jun 2021 17:28:19 +
Brian Callahan  wrote:

> Hi Charlene --
> 
> On 06/29/2021 12:42 PM, Charlene Wendling wrote:
> > Hi,
> > 
> > I heard that a new Hyperrogue version came out, so i wanted to see
> > if segfaults at build time on macppc were fixed. They're not, sadly.
> > 
> 
> Thank you for doing the heavy lifting on this.
> One small nit, NO_TEST=Yes seems to be needed.

Oh, you're right, i'm always confused with that one.

Here is a fixed diff below if more people want to try.
 
> OK for me with that.
> 
> ~Brian
> 


Index: Makefile
===
RCS file: /cvs/ports/games/hyperrogue/Makefile,v
retrieving revision 1.16
diff -u -p -u -p -r1.16 Makefile
--- Makefile28 Jun 2020 02:37:25 -  1.16
+++ Makefile29 Jun 2021 19:24:58 -
@@ -1,6 +1,6 @@
 # $OpenBSD: Makefile,v 1.16 2020/06/28 02:37:25 bcallah Exp $
 
-V =11.3o
+V =12.0b
 COMMENT =  roguelike game in a non-Euclidean world
 CATEGORIES =   games x11
 
@@ -20,26 +20,41 @@ WANTLIB += c m png z
 # C++11
 COMPILER = base-clang ports-gcc
 
-BUILD_DEPENDS =${MODGNU_AUTOCONF_DEPENDS} \
-   ${MODGNU_AUTOMAKE_DEPENDS}
-
 LIB_DEPENDS =  devel/sdl-gfx \
devel/sdl-mixer \
devel/sdl-ttf \
graphics/glew \
graphics/png
 
-CONFIGURE_STYLE =  autoreconf
-CONFIGURE_ENV =CPPFLAGS="${CPPFLAGS} -I${LOCALBASE}/include 
-I${X11BASE}/include" \
-   LDFLAGS="${LDFLAGS} -L${LOCALBASE}/lib -L${X11BASE}/lib"
-
-MAKE_FLAGS =   CFLAGS="${CFLAGS}" CXXFLAGS="${CXXFLAGS} -std=c++11"
-
-AUTOCONF_VERSION = 2.69
-AUTOMAKE_VERSION = 1.16
-
-# Correct version number
-pre-configure:
-   sed -i 's,10.4i,${V},g' ${WRKSRC}/configure.ac
+CXXFLAGS +=-I${LOCALBASE}/include -I${X11BASE}/include \
+   -DHYPERPATH="\\\"${LOCALBASE}/share/hyperrogue/\\\""
+LDFLAGS += -L${LOCALBASE}/lib -L${X11BASE}/lib
+
+MAKE_ENV +=CFLAGS="${CFLAGS}" CXXFLAGS="${CXXFLAGS}" \
+   LDFLAGS="${LDFLAGS}" CXX="${CXX}" \
+   HYPERROGUE_USE_GLEW=1 \
+   HYPERROGUE_USE_PNG=1
+
+USE_GMAKE =Yes
+NO_TEST =  Yes
+ALL_TARGET =   hyperrogue
+
+# There is no install target
+do-install:
+   ${INSTALL_PROGRAM} ${WRKSRC}/hyperrogue ${PREFIX}/bin
+
+   ${INSTALL_DATA_DIR} ${PREFIX}/share/hyperrogue
+   ${INSTALL_DATA_DIR} ${PREFIX}/share/hyperrogue/sounds
+   ${INSTALL_DATA_DIR} ${PREFIX}/share/hyperrogue/music
+
+   ${INSTALL_DATA} ${WRKSRC}/*.ttf \
+   ${WRKSRC}/*.dat \
+   ${WRKSRC}/hyperrogue-music.txt \
+   ${PREFIX}/share/hyperrogue
+   ${INSTALL_DATA} ${WRKSRC}/sounds/* ${PREFIX}/share/hyperrogue/sounds
+   ${INSTALL_DATA} ${WRKSRC}/music/* ${PREFIX}/share/hyperrogue/music
+
+   ${INSTALL_DATA_DIR} ${PREFIX}/share/doc/hyperrogue
+   ${INSTALL_DATA} ${WRKSRC}/README.md ${PREFIX}/share/doc/hyperrogue
 
 .include 
Index: distinfo
===
RCS file: /cvs/ports/games/hyperrogue/distinfo,v
retrieving revision 1.13
diff -u -p -u -p -r1.13 distinfo
--- distinfo28 Jun 2020 02:37:25 -  1.13
+++ distinfo29 Jun 2021 19:24:58 -
@@ -1,2 +1,2 @@
-SHA256 (hyperrogue-11.3o.tar.gz) = y8luUVxGjuvwGEzu1SRLUoZLv5sOhEaUAaNime0kdQs=
-SIZE (hyperrogue-11.3o.tar.gz) = 69246164
+SHA256 (hyperrogue-12.0b.tar.gz) = v6el52DKYSfIwlOlUdpEoW+PQXabJF3B4YRVARavV5w=
+SIZE (hyperrogue-12.0b.tar.gz) = 79894443
Index: patches/patch-Makefile
===
RCS file: patches/patch-Makefile
diff -N patches/patch-Makefile
--- /dev/null   1 Jan 1970 00:00:00 -
+++ patches/patch-Makefile  29 Jun 2021 19:24:58 -
@@ -0,0 +1,94 @@
+$OpenBSD$
+
+Remove hardcoded optimisation flags and -Werror, add a consistent
+flag set for clang and ports-gcc
+
+Index: Makefile
+--- Makefile.orig
 Makefile
+@@ -42,7 +42,6 @@ endif
+ ## Begin customization points for each OS and TOOLCHAIN we support.
+ 
+ 
+-ifeq (${OS},linux)
+   CXXFLAGS_EARLY += -DLINUX
+   EXE_EXTENSION :=
+   LDFLAGS_GL := -lGL
+@@ -51,7 +50,6 @@ ifeq (${OS},linux)
+   LDFLAGS_SDL := -lSDL -lSDL_gfx -lSDL_mixer -lSDL_ttf -lpthread -lz
+   OBJ_EXTENSION := .o
+   hyper_RES :=
+-endif
+ 
+ ifeq (${OS},mingw)
+   CXXFLAGS_EARLY += -DWINDOWS -mwindows -D_A_VOLID=8
+@@ -79,28 +77,11 @@ ifeq (${OS},osx)
+   hyper_RES :=
+ endif
+ 
+-ifeq (${TOOLCHAIN},clang)
+-  CXXFLAGS_STD = -std=c++11
+-  CXXFLAGS_EARLY += -march=native -fPIC
+-  CXXFLAGS_EARLY += -W -Wall -Wextra -Werror -pedantic
+-  CXXFLAGS_EARLY += -Wno-unused-parameter -Wno-implicit-fallthrough 
-Wno-maybe-uniniti

Re: NEW: games/tic80

[Charlene Wendling]

On Tue, 29 Jun 2021 17:54:56 +
Brian Callahan  wrote:

> Hi ports --
> 
> Attached is a new port, TIC-80. TIC-80 is a fantasy computer to make, 
> play and share tiny games.
> For those of you familiar with PICO-8, TIC-80 is an open source 
> alternative.

[...] 

> TIC-80 works well for me on amd64. I am interested in hearing about
> big endian systems.

I've tested on macppc.

It builds but does not work. There is no sound, the welcome
screen is red, carts either segfault or display a white window.

Big endian support has been refused upstream [0], the code has
never been written to do so and would need a major refactoring to
make things maintainable. The PR code is interesting to fix
similar issues in other ports though.

Meanwhile, it builds and works fine on amd64, testing in an
empty chroot revealed no missing depends.

> OK?

OK cwen@, but as you probably guessed with NOT_FOR_ARCHS=${BE_ARCHS}

> ~Brian

[0] https://github.com/nesbox/TIC-80/pull/1057/

[update] games/hyperrogue 11.3o -> 12.0b

[Charlene Wendling]

Hi,

I heard that a new Hyperrogue version came out, so i wanted to see if
segfaults at build time on macppc were fixed. They're not, sadly.

The changelog is very long to read [0]. What impacts us the most is that
upstream changed their build system to a simple Makefile, but this
makes things harder for us. It also requires GNU make now.

Most of the port changes come straight from FreeBSD, adapted for our
infrastructure.

Port changes:

- Drop all autotools things, and move CONFIGURE_* to MAKE_*. I 
  preferred to put CXXFLAGS and LDFLAGS out of MAKE_ENV, mostly
  because CXXFLAGS is now too long to fit nicely.  
- Add HYPERPATH to CXXFLAGS, so resources are properly found.
- Bring back GLEW and PNG support, they're off by default.
- Add a patch from FreeBSD [1] to remove all hardcoded optimisations.
  They're shipping 12.0a so i had to remove more of them. Also
  add a consistent flags set for clang and ports-gcc.
- The Makefile has no install target, so i've made our own. PLIST
  diff seems happy with it, so this looks correct to me. Maybe
  there is a less naive way to do so :)

Testing: 

On amd64, it builds and works fine with amdgpu (RX560).

On macppc, langen still segfaults. Upstream's Makefile allows easily to
build it with custom CXXFLAGS, so i tried -O{0,1} to no avail.

Comments/feedback are welcome,

Charlène.


[0]
https://github.com/zenorogue/hyperrogue/blob/master/changelog.txt#L3997
[1]
https://github.com/freebsd/freebsd-ports/blob/main/games/hyperrogue/files/patch-Makefile



Index: Makefile
===
RCS file: /cvs/ports/games/hyperrogue/Makefile,v
retrieving revision 1.16
diff -u -p -u -p -r1.16 Makefile
--- Makefile28 Jun 2020 02:37:25 -  1.16
+++ Makefile29 Jun 2021 16:28:21 -
@@ -1,6 +1,6 @@
 # $OpenBSD: Makefile,v 1.16 2020/06/28 02:37:25 bcallah Exp $
 
-V =11.3o
+V =12.0b
 COMMENT =  roguelike game in a non-Euclidean world
 CATEGORIES =   games x11
 
@@ -20,26 +20,40 @@ WANTLIB += c m png z
 # C++11
 COMPILER = base-clang ports-gcc
 
-BUILD_DEPENDS =${MODGNU_AUTOCONF_DEPENDS} \
-   ${MODGNU_AUTOMAKE_DEPENDS}
-
 LIB_DEPENDS =  devel/sdl-gfx \
devel/sdl-mixer \
devel/sdl-ttf \
graphics/glew \
graphics/png
 
-CONFIGURE_STYLE =  autoreconf
-CONFIGURE_ENV =CPPFLAGS="${CPPFLAGS} -I${LOCALBASE}/include 
-I${X11BASE}/include" \
-   LDFLAGS="${LDFLAGS} -L${LOCALBASE}/lib -L${X11BASE}/lib"
-
-MAKE_FLAGS =   CFLAGS="${CFLAGS}" CXXFLAGS="${CXXFLAGS} -std=c++11"
-
-AUTOCONF_VERSION = 2.69
-AUTOMAKE_VERSION = 1.16
-
-# Correct version number
-pre-configure:
-   sed -i 's,10.4i,${V},g' ${WRKSRC}/configure.ac
+CXXFLAGS +=-I${LOCALBASE}/include -I${X11BASE}/include \
+   -DHYPERPATH="\\\"${LOCALBASE}/share/hyperrogue/\\\""
+LDFLAGS += -L${LOCALBASE}/lib -L${X11BASE}/lib
+
+MAKE_ENV +=CFLAGS="${CFLAGS}" CXXFLAGS="${CXXFLAGS}" \
+   LDFLAGS="${LDFLAGS}" CXX="${CXX}" \
+   HYPERROGUE_USE_GLEW=1 \
+   HYPERROGUE_USE_PNG=1
+
+USE_GMAKE =Yes
+ALL_TARGET =   hyperrogue
+
+# There is no install target
+do-install:
+   ${INSTALL_PROGRAM} ${WRKSRC}/hyperrogue ${PREFIX}/bin
+
+   ${INSTALL_DATA_DIR} ${PREFIX}/share/hyperrogue
+   ${INSTALL_DATA_DIR} ${PREFIX}/share/hyperrogue/sounds
+   ${INSTALL_DATA_DIR} ${PREFIX}/share/hyperrogue/music
+
+   ${INSTALL_DATA} ${WRKSRC}/*.ttf \
+   ${WRKSRC}/*.dat \
+   ${WRKSRC}/hyperrogue-music.txt \
+   ${PREFIX}/share/hyperrogue
+   ${INSTALL_DATA} ${WRKSRC}/sounds/* ${PREFIX}/share/hyperrogue/sounds
+   ${INSTALL_DATA} ${WRKSRC}/music/* ${PREFIX}/share/hyperrogue/music
+
+   ${INSTALL_DATA_DIR} ${PREFIX}/share/doc/hyperrogue
+   ${INSTALL_DATA} ${WRKSRC}/README.md ${PREFIX}/share/doc/hyperrogue
 
 .include 
Index: distinfo
===
RCS file: /cvs/ports/games/hyperrogue/distinfo,v
retrieving revision 1.13
diff -u -p -u -p -r1.13 distinfo
--- distinfo28 Jun 2020 02:37:25 -  1.13
+++ distinfo29 Jun 2021 16:28:21 -
@@ -1,2 +1,2 @@
-SHA256 (hyperrogue-11.3o.tar.gz) = y8luUVxGjuvwGEzu1SRLUoZLv5sOhEaUAaNime0kdQs=
-SIZE (hyperrogue-11.3o.tar.gz) = 69246164
+SHA256 (hyperrogue-12.0b.tar.gz) = v6el52DKYSfIwlOlUdpEoW+PQXabJF3B4YRVARavV5w=
+SIZE (hyperrogue-12.0b.tar.gz) = 79894443
Index: patches/patch-Makefile
===
RCS file: patches/patch-Makefile
diff -N patches/patch-Makefile
--- /dev/null   1 Jan 1970 00:00:00 -
+++ patches/patch-Makefile  29 Jun 2021 16:28:21 -
@@ -0,0 +1,94 @@
+$OpenBSD$
+
+Remove hardcoded optimisation flags and -Werror, add a consistent
+flag set for clang and ports-gcc
+
+Index: Makefile
+--- Make

[macppc, arm*] Unbreak misc/uniutils' unireverse

[Charlene Wendling]

Hi,

>From macppc bulk logs:

> unireverse.c:173:10: warning: result of comparison of constant -1
> with expression of type 'char' is always false

On macppc the unireverse runtime goes like this:

> $ printf "infinite\nloop" | unireverse
> etinifni
> pool
[hangs forever]

On macppc and arm* char is unsigned by default, so a looped getc(3)
call in unireverse never reaches EOF there. Using an int instead of a
char to store the return value of getc(3) solves that issue.

OK?

Charlène.


Index: Makefile
===
RCS file: /cvs/ports/misc/uniutils/Makefile,v
retrieving revision 1.8
diff -u -p -u -p -r1.8 Makefile
--- Makefile12 Jul 2019 20:47:54 -  1.8
+++ Makefile19 Jun 2021 19:23:22 -
@@ -3,7 +3,7 @@
 COMMENT=   Unicode utilities
 
 DISTNAME=  uniutils-2.27
-REVISION=  2
+REVISION=  3
 CATEGORIES=misc
 
 HOMEPAGE=  http://billposer.org/Software/unidesc.html
Index: patches/patch-unireverse_c
===
RCS file: patches/patch-unireverse_c
diff -N patches/patch-unireverse_c
--- /dev/null   1 Jan 1970 00:00:00 -
+++ patches/patch-unireverse_c  19 Jun 2021 19:23:22 -
@@ -0,0 +1,17 @@
+$OpenBSD$
+
+Fix an infinite loop on archs where char is unsigned by default (arm*
+and powerpc), because EOF (-1) could not be reached by getc(3).
+
+Index: unireverse.c
+--- unireverse.c.orig
 unireverse.c
+@@ -151,7 +151,7 @@ int main(int ac, char **av) {
+ 
+ char * GetLine(FILE *fp, int *LineLength)
+ {
+-  char c;
++  int c;
+   int Available;
+   int CharsRead;
+   char *Line;

Re: sparc64 bulk build report

[Charlene Wendling]

Hi,

On Thu, 17 Jun 2021 09:56:27 +0100
Stuart Henderson  wrote:

> PowerPC doesn't have those either

But it doesn't need to, because embree is not built on all 32 bit
arches [0] due to a lack of "consumers" [1] :)

You may be able to see that in the i386 build logs but here is the
powerpc one [2] (beware, heavy!).

[0] https://github.com/godotengine/godot/issues/48482
[1] https://github.com/godotengine/godot/pull/48483/commits/e53422c8
[2] https://bin.charlenew.xyz/godot.log

> -- 
>   Sent from a phone, apologies for poor formatting.
> On 15 June 2021 08:26:18 Omar Polo  wrote:
> 
> > k...@openbsd.org writes:
> >
> >> http://build-failures.rhaalovely.net/sparc64/2021-06-13/games/godot.log
> >
> > This is due to a newly added bundled deps: embree[0]; it needs AVX,
> > SSE or NEON support, something that I'm assuming sparc doesn't have
> > (and that probably rules out other arches too...)
> >
> > So, should we mark it as ONLY_FOR_ARCHS?  (the list is probably
> > incomplete, it's taken from a quick glance at where the 3.3 package
> > was built)
> >
> >
> > Thanks,
> >
> > Omar Polo
> >
> > [0] actually, a fork of it
> >https://github.com/lighttransport/embree-aarch64
> >in particular, the fail is from this block
> >
> > https://github.com/godotengine/godot/blob/master/thirdparty/embree/common/sys/sysinfo.h#L17
> >
> >
> > Index: Makefile
> > ===
> > RCS file: /home/cvs/ports/games/godot/Makefile,v
> > retrieving revision 1.20
> > diff -u -p -r1.20 Makefile
> > --- Makefile9 Jun 2021 19:50:07 -   1.20
> > +++ Makefile15 Jun 2021 07:19:53 -
> > @@ -1,5 +1,6 @@
> > # $OpenBSD: Makefile,v 1.20 2021/06/09 19:50:07 sthen Exp $
> >
> > +ONLY_FOR_ARCHS =   aarch64 amd64 arm i386 powerpc
> > BROKEN-powerpc =fails at runtime, the UI is totally blank
> >
> > COMMENT =   2D and 3D game engine
> 

net/tdlib does not support big endian arches

[Charlene Wendling]

Hi,

> http://build-failures.rhaalovely.net/sparc64/2021-06-13/net/tdlib.log
> (octeon and macppc are also impacted but no bulk report
> yet)

tdlib does not support big endian arches [0].

OK? 

Charlène.


[0] https://github.com/tdlib/td/issues/364


Index: Makefile
===
RCS file: /cvs/ports/net/tdlib/Makefile,v
retrieving revision 1.1.1.1
diff -u -p -u -p -r1.1.1.1 Makefile
--- Makefile9 Jun 2021 10:42:16 -   1.1.1.1
+++ Makefile15 Jun 2021 14:15:53 -
@@ -1,5 +1,9 @@
 # $OpenBSD: Makefile,v 1.1.1.1 2021/06/09 10:42:16 landry Exp $
 
+
+# See https://github.com/tdlib/td/issues/364
+NOT_FOR_ARCHS =${BE_ARCHS}
+
 COMMENT =  cross-platform library for building Telegram clients
 
 # upstream tags stable versions, but various clients using this library

Re: [fix, macppc] Unbreak devel/sdl{1,2} on Mac G3s

[Charlene Wendling]

On Sun, 13 Jun 2021 23:21:57 -0400
George Koehler  wrote:

> On Sat, 12 Jun 2021 12:03:01 +
> Charlene Wendling  wrote:
> 
> > You were right, SDL 1.x is impacted as well. I've updated the diff
> > below to include SDL 1.x. It has been tested successfully against
> > frozen-bubble and tutris. xmoto fails with:
> 
> ok gkoehler@
> 
> and thanks for trying so many packages!
> I didn't know which packages used SDL 1.x.

You can use show-reverse-deps(1) from the sqlports package to get them,
getting direct depends still requires using sqlite3 directly though.

> > Program received signal SIGILL, Illegal instruction.
> > 0x9830c5f0 in jsimd_idct_ifast_altivec ()
> > from /usr/local/lib/libjpeg.so.70.0
> >
> > So i know what's up next, but given a lot of stuff depend on
> > libjpeg, it should be way longer, since i may need to rebuild
> > almost all my current packages ;)
> 
> This libjpeg crash might be a false positive (from the no-ALTIVEC
> kernel bug), so it might go away with Matthieu Herrb's diff on bugs@.
> I am running no-ALTIVEC with the diff, but here xmoto crashes in
> libLLVM.so.5.0 because I don't have radeondrm.  (I know that OpenGL,
> on macppc without radeondrm, always crashes in libLLVM.)

I will lurk for a good snow iBook G3 and test from there,
that will avoid any risk of false positives.

> I have a PowerBook G4 with radeondrm, but I didn't use it for this
> no-ALTIVEC trial.  I also have a G3, but it doesn't power on.
>
> By doing $ objdump -dlr /usr/local/lib/libjpeg.so.70.0 |less
> and searching for '[, ]v[0-9]', it looks like libjpeg uses altivec
> only in its *_altivec() functions.  If it knows not to call
> *_altivec() on G3, it wouldn't crash.

Thanks, i'll keep that line around!

> --George
> 
> > 
> > Charlène.
> > 
> > 
> > Index: devel/sdl2/Makefile
> > ===
> > RCS file: /cvs/ports/devel/sdl2/Makefile,v
> > retrieving revision 1.34
> > diff -u -p -r1.34 Makefile
> > --- devel/sdl2/Makefile 23 Jan 2021 17:47:43 -
> > 1.34 +++ devel/sdl2/Makefile12 Jun 2021 12:01:33 -
> > @@ -5,7 +5,7 @@ COMMENT=cross-platform multimedia libra
> >  V= 2.0.14
> >  DISTNAME=  SDL2-${V}
> >  PKGNAME=   sdl2-${V}
> > -REVISION=  1
> > +REVISION=  2
> >  CATEGORIES=devel
> >  MASTER_SITES=  https://www.libsdl.org/release/
> >  
> > @@ -50,6 +50,13 @@ CONFIGURE_ENV+=  CPPFLAGS="-I${LOCALBASE}
> >  
> >  .if ${MACHINE_ARCH} == "hppa"
> >  CONFIGURE_ARGS+= --disable-atomic
> > +.endif
> > +
> > +# The assembly code contains AltiVec instructions, whose support
> > is not +# properly detected at runtime. Fix SIGILLs on macppc G3s,
> > while keeping +# it for powerpc64.
> > +.if ${MACHINE_ARCH} == "powerpc"
> > +CONFIGURE_ARGS+= --enable-altivec=no
> >  .endif
> >  
> >  # tests in test subdir, but interactive and not hooked up to build
> > Index: devel/sdl/Makefile
> > ===
> > RCS file: /cvs/ports/devel/sdl/Makefile,v
> > retrieving revision 1.105
> > diff -u -p -r1.105 Makefile
> > --- devel/sdl/Makefile  23 Jan 2021 17:52:00 -
> > 1.105 +++ devel/sdl/Makefile12 Jun 2021 12:01:33 -
> > @@ -5,7 +5,7 @@ COMMENT=cross-platform multimedia libra
> >  VERSION=   1.2.15
> >  DISTNAME=  SDL-${VERSION}
> >  PKGNAME=   ${DISTNAME:L}
> > -REVISION=  11
> > +REVISION=  12
> >  CATEGORIES=devel
> >  MAINTAINER=Thomas Frohwein 
> >  MASTER_SITES=  ${HOMEPAGE}release/
> > @@ -64,6 +64,13 @@ WANTLIB+=X11 Xau Xdmcp Xext Xrandr Xren
> >  .endif
> >  
> >  NO_TEST=   Yes
> > +
> > +# The assembly code contains AltiVec instructions, whose support
> > is not +# properly detected at runtime. Fix SIGILLs on macppc G3s,
> > while keeping +# it for powerpc64.
> > +.if ${MACHINE_ARCH} == "powerpc"
> > +CONFIGURE_ARGS+= --enable-altivec=no
> > +.endif
> >  
> >  pre-configure:
> > @mkdir -p ${WRKSRC}/src/audio/sndio
> > 
> > 
> 
> 
> -- 
> George Koehler 

Re: [fix, macppc] Unbreak devel/sdl{1,2} on Mac G3s

[Charlene Wendling]

On Fri, 11 Jun 2021 22:38:09 -0400
George Koehler  wrote:

> On Fri, 11 Jun 2021 19:13:46 +
> Charlene Wendling  wrote:
> 
> > Hi,
> > 
> > As reported in the fediverse, sdl2 generates SIGILL on Mac G3s [0].
> > 
> > sdl2 is built with AltiVec support, runtime detection is present and
> > valid (see $WRKSRC/src/cpuinfo/SDL_cpuinfo.c), yet failing. So the
> > simple solution is to remove AltiVec from the build for now.
> > 
> > I've tested this on my PowerBook G4 against ffplay, mpv,
> > openttd and scummvm, without issues.
> > 
> > I've also tested successfully these ports on my Mac Mini G4, but
> > this time, using a kernel without the ALTIVEC option.
> > 
> > OK?
> 
> ok gkoehler@
> 
> SDL-1.x (devel/sdl) might need the same --enable-altivec=no fix, but
> I didn't try it.
> 
> SDL was using ${CC} -maltivec, so it's altivec blitter can use
>  operations.  The problem is that ${CC} -maltivec can also
> optimize any *.c code with altivec.  (graphics/ffmpeg had the same
> problem when we tried to enable altivec.)
> 
> Later, one might try to enable altivec again by moving the altivec
> blitter to a separate file, and compiling only that file with
> -maltivec (as xenocara/lib/pixman does now).
> 
> --George
> 
> > Charlène.
> > 
> > 
> > [0] https://empty.cafe/@will/106329827929064693

You were right, SDL 1.x is impacted as well. I've updated the diff
below to include SDL 1.x. It has been tested successfully against
frozen-bubble and tutris. xmoto fails with:

Program received signal SIGILL, Illegal instruction.
0x9830c5f0 in jsimd_idct_ifast_altivec ()
from /usr/local/lib/libjpeg.so.70.0

So i know what's up next, but given a lot of stuff depend on libjpeg, it
should be way longer, since i may need to rebuild almost all my
current packages ;)

Charlène.


Index: devel/sdl2/Makefile
===
RCS file: /cvs/ports/devel/sdl2/Makefile,v
retrieving revision 1.34
diff -u -p -r1.34 Makefile
--- devel/sdl2/Makefile 23 Jan 2021 17:47:43 -  1.34
+++ devel/sdl2/Makefile 12 Jun 2021 12:01:33 -
@@ -5,7 +5,7 @@ COMMENT=cross-platform multimedia libra
 V= 2.0.14
 DISTNAME=  SDL2-${V}
 PKGNAME=   sdl2-${V}
-REVISION=  1
+REVISION=  2
 CATEGORIES=devel
 MASTER_SITES=  https://www.libsdl.org/release/
 
@@ -50,6 +50,13 @@ CONFIGURE_ENV+=  CPPFLAGS="-I${LOCALBASE}
 
 .if ${MACHINE_ARCH} == "hppa"
 CONFIGURE_ARGS+= --disable-atomic
+.endif
+
+# The assembly code contains AltiVec instructions, whose support is not
+# properly detected at runtime. Fix SIGILLs on macppc G3s, while keeping
+# it for powerpc64.
+.if ${MACHINE_ARCH} == "powerpc"
+CONFIGURE_ARGS+= --enable-altivec=no
 .endif
 
 # tests in test subdir, but interactive and not hooked up to build
Index: devel/sdl/Makefile
===
RCS file: /cvs/ports/devel/sdl/Makefile,v
retrieving revision 1.105
diff -u -p -r1.105 Makefile
--- devel/sdl/Makefile  23 Jan 2021 17:52:00 -  1.105
+++ devel/sdl/Makefile  12 Jun 2021 12:01:33 -
@@ -5,7 +5,7 @@ COMMENT=cross-platform multimedia libra
 VERSION=   1.2.15
 DISTNAME=  SDL-${VERSION}
 PKGNAME=   ${DISTNAME:L}
-REVISION=  11
+REVISION=  12
 CATEGORIES=devel
 MAINTAINER=Thomas Frohwein 
 MASTER_SITES=  ${HOMEPAGE}release/
@@ -64,6 +64,13 @@ WANTLIB+=X11 Xau Xdmcp Xext Xrandr Xren
 .endif
 
 NO_TEST=   Yes
+
+# The assembly code contains AltiVec instructions, whose support is not
+# properly detected at runtime. Fix SIGILLs on macppc G3s, while keeping
+# it for powerpc64.
+.if ${MACHINE_ARCH} == "powerpc"
+CONFIGURE_ARGS+= --enable-altivec=no
+.endif
 
 pre-configure:
@mkdir -p ${WRKSRC}/src/audio/sndio

[fix, macppc] Unbreak devel/sdl2 on Mac G3s

[Charlene Wendling]

Hi,

As reported in the fediverse, sdl2 generates SIGILL on Mac G3s [0].

sdl2 is built with AltiVec support, runtime detection is present and
valid (see $WRKSRC/src/cpuinfo/SDL_cpuinfo.c), yet failing. So the
simple solution is to remove AltiVec from the build for now.

I've tested this on my PowerBook G4 against ffplay, mpv,
openttd and scummvm, without issues.

I've also tested successfully these ports on my Mac Mini G4, but this
time, using a kernel without the ALTIVEC option.

OK?

Charlène.


[0] https://empty.cafe/@will/106329827929064693


Index: Makefile
===
RCS file: /cvs/ports/devel/sdl2/Makefile,v
retrieving revision 1.34
diff -u -p -u -p -r1.34 Makefile
--- Makefile23 Jan 2021 17:47:43 -  1.34
+++ Makefile11 Jun 2021 19:07:43 -
@@ -5,7 +5,7 @@ COMMENT=cross-platform multimedia libra
 V= 2.0.14
 DISTNAME=  SDL2-${V}
 PKGNAME=   sdl2-${V}
-REVISION=  1
+REVISION=  2
 CATEGORIES=devel
 MASTER_SITES=  https://www.libsdl.org/release/
 
@@ -50,6 +50,13 @@ CONFIGURE_ENV+=  CPPFLAGS="-I${LOCALBASE}
 
 .if ${MACHINE_ARCH} == "hppa"
 CONFIGURE_ARGS+= --disable-atomic
+.endif
+
+# The assembly code contains AltiVec instructions, whose support is not
+# properly detected at runtime. Fix SIGILLs on macppc G3s, while keeping
+# it for powerpc64.
+.if ${MACHINE_ARCH} == "powerpc"
+CONFIGURE_ARGS+= --enable-altivec=no
 .endif
 
 # tests in test subdir, but interactive and not hooked up to build

Re: www/p5-Dancer2: Update to 0.301004

[Charlene Wendling]

Hi,

On Thu, 10 Jun 2021 11:34:33 +
wen heping  wrote:

> Hi, ports@:
> 
>Here is a patch for www/p5-Dancer2:
>i) Update to 0.301004
>ii) Add sysutils/p5-File-Which as RUN_DEPENDS
>It build well and pass all tests on amd64-6.9 system.
>4 ports depends on it and all build and pass tests too.
> 
> 
> wen

OK cwen@ with 2 tweaks:

- sysutils/p5-Capture-Tiny is missing as TEST_DEPENDS
- we don't need gitignore and docker files


Index: Makefile
===
RCS file: /cvs/ports/www/p5-Dancer2/Makefile,v
retrieving revision 1.13
diff -u -p -u -p -r1.13 Makefile
--- Makefile17 May 2021 13:31:19 -  1.13
+++ Makefile11 Jun 2021 11:24:07 -
@@ -4,7 +4,7 @@ COMMENT =   lightweight yet powerful web 
 
 MODULES =  cpan
 PKG_ARCH = *
-DISTNAME = Dancer2-0.301002
+DISTNAME = Dancer2-0.301004
 CATEGORIES =   www
 
 # Perl
@@ -17,6 +17,7 @@ RUN_DEPENDS = devel/p5-CLI-Osprey \
devel/p5-Config-Any \
security/p5-Digest-SHA1 \
devel/p5-File-Share \
+   sysutils/p5-File-Which \
www/p5-HTTP-Body \
www/p5-HTTP-Cookies \
www/p5-HTTP-Date \
@@ -47,7 +48,8 @@ TEST_DEPENDS =devel/p5-AnyEvent \
devel/p5-Test-Memory-Cycle>=1.04 \
devel/p5-Test-MockTime \
devel/p5-Test-NoTabs \
-   devel/p5-Test-Perl-Critic
+   devel/p5-Test-Perl-Critic \
+   sysutils/p5-Capture-Tiny
 
 MAKE_ENV +=TEST_POD=yes AUTHOR_TESTING=Yes
 
Index: distinfo
===
RCS file: /cvs/ports/www/p5-Dancer2/distinfo,v
retrieving revision 1.8
diff -u -p -u -p -r1.8 distinfo
--- distinfo17 May 2021 13:31:19 -  1.8
+++ distinfo11 Jun 2021 11:24:07 -
@@ -1,2 +1,2 @@
-SHA256 (Dancer2-0.301002.tar.gz) = MFcbmuiATfxdWTTyPgMJ30nKvP61o+C0R+awXbW3oHs=
-SIZE (Dancer2-0.301002.tar.gz) = 396769
+SHA256 (Dancer2-0.301004.tar.gz) = CO4NnvvMGhogWdpaG5kEHsqKzuM7FOPQtMu9AQAJ3eU=
+SIZE (Dancer2-0.301004.tar.gz) = 400506
Index: pkg/PLIST
===
RCS file: /cvs/ports/www/p5-Dancer2/pkg/PLIST,v
retrieving revision 1.5
diff -u -p -u -p -r1.5 PLIST
--- pkg/PLIST   17 May 2021 13:31:19 -  1.5
+++ pkg/PLIST   11 Jun 2021 11:24:07 -
@@ -60,6 +60,7 @@ ${P5SITE}/Dancer2/Logger/Null.pm
 ${P5SITE}/Dancer2/Manual/
 ${P5SITE}/Dancer2/Manual.pod
 ${P5SITE}/Dancer2/Manual/Deployment.pod
+${P5SITE}/Dancer2/Manual/Keywords.pod
 ${P5SITE}/Dancer2/Manual/Migration.pod
 ${P5SITE}/Dancer2/Manual/Testing.pod
 ${P5SITE}/Dancer2/Plugin.pm
@@ -82,6 +83,9 @@ ${P5SITE}/Dancer2/Template/Tiny.pm
 ${P5SITE}/Dancer2/Test.pm
 ${P5SITE}/Dancer2/Tutorial.pod
 ${P5SITE}/auto/share/dist/Dancer2/
+@comment ${P5SITE}/auto/share/dist/Dancer2/.gitignore
+@comment ${P5SITE}/auto/share/dist/Dancer2/docker/
+@comment ${P5SITE}/auto/share/dist/Dancer2/docker/Dockerfile
 ${P5SITE}/auto/share/dist/Dancer2/skel/
 ${P5SITE}/auto/share/dist/Dancer2/skel/.dancer
 ${P5SITE}/auto/share/dist/Dancer2/skel/MANIFEST.SKIP
@@ -166,6 +170,7 @@ ${P5SITE}/auto/share/dist/Dancer2/skel/v
 @man man/man3p/Dancer2::Logger::Null.3p
 @man man/man3p/Dancer2::Manual.3p
 @man man/man3p/Dancer2::Manual::Deployment.3p
+@man man/man3p/Dancer2::Manual::Keywords.3p
 @man man/man3p/Dancer2::Manual::Migration.3p
 @man man/man3p/Dancer2::Manual::Testing.3p
 @man man/man3p/Dancer2::Plugin.3p

[update] games/chessx 1.4.6 -> 1.5.6

[Charlene Wendling]

Hi,

Here is update for chessx to 1.5.6. A 1.5.8 version is around, but
upstream made no formal release. You can find the changelog there [0].

Port-wise, that update allows to drop almost all patches. Sadly,
there is still no easy way to use our system quazip. 

It builds and works fine on amd64 (it's not built on macppc because
of rust being in the dependency chain).

Comments/feedback are welcome,

Charlène.


[0] https://github.com/Isarhamster/chessx/blob/master/ChangeLog#L20


Index: Makefile
===
RCS file: /cvs/ports/games/chessx/Makefile,v
retrieving revision 1.10
diff -u -p -u -p -r1.10 Makefile
--- Makefile26 Feb 2020 06:14:43 -  1.10
+++ Makefile10 Jun 2021 12:02:17 -
@@ -1,17 +1,16 @@
 # $OpenBSD: Makefile,v 1.10 2020/02/26 06:14:43 rsadowski Exp $
 
 COMMENT =  free chess database and analyzer
-DISTNAME = chessx-1.4.6
+DISTNAME = chessx-1.5.6
 CATEGORIES =   games
 HOMEPAGE = http://chessx.sourceforge.net/
-REVISION = 5
 
 # GPLv2+
 PERMIT_PACKAGE =   Yes
 
-WANTLIB =  GL Qt5Core Qt5Gui Qt5Multimedia Qt5Network Qt5PrintSupport
-WANTLIB += Qt5Svg Qt5Widgets Qt5Xml c m pthread quazip5 z
 WANTLIB += ${COMPILER_LIBCXX}
+WANTLIB += GL Qt5Core Qt5Gui Qt5Multimedia Qt5Network Qt5PrintSupport
+WANTLIB += Qt5Svg Qt5Widgets Qt5Xml c m quazip5 z
 
 MASTER_SITES = ${MASTER_SITE_SOURCEFORGE:=chessx/}
 EXTRACT_SUFX = .tgz
Index: distinfo
===
RCS file: /cvs/ports/games/chessx/distinfo,v
retrieving revision 1.2
diff -u -p -u -p -r1.2 distinfo
--- distinfo2 Jul 2017 00:51:50 -   1.2
+++ distinfo10 Jun 2021 12:02:17 -
@@ -1,2 +1,2 @@
-SHA256 (chessx-1.4.6.tgz) = rZUZEKO9U/VlX56FEB55abU2B4Rz4Joqfd7W7xcaaO0=
-SIZE (chessx-1.4.6.tgz) = 4223076
+SHA256 (chessx-1.5.6.tgz) = 0JpLU0qQnF8aOYxkBlBJov3zBJe8gcy98NlAQSNk0gU=
+SIZE (chessx-1.5.6.tgz) = 3853395
Index: patches/patch-chessx_pro
===
RCS file: /cvs/ports/games/chessx/patches/patch-chessx_pro,v
retrieving revision 1.2
diff -u -p -u -p -r1.2 patch-chessx_pro
--- patches/patch-chessx_pro2 Jul 2017 00:51:50 -   1.2
+++ patches/patch-chessx_pro10 Jun 2021 12:02:17 -
@@ -3,52 +3,46 @@ Use system quazip.
 Index: chessx.pro
 --- chessx.pro.orig
 +++ chessx.pro
-@@ -141,22 +141,6 @@ HEADERS += src/database/board.h \
- src/gui/engineoptionlist.h \
- src/gui/helpbrowser.h \
- src/database/downloadmanager.h \
--src/quazip/zip.h \
--src/quazip/unzip.h \
--src/quazip/quazipnewinfo.h \
--src/quazip/quazipfileinfo.h \
--src/quazip/quazipfile.h \
--src/quazip/quazipdir.h \
--src/quazip/quazip.h \
--src/quazip/quazip_global.h \
--src/quazip/quaziodevice.h \
--src/quazip/quagzipfile.h \
--src/quazip/quacrc32.h \
--src/quazip/quachecksum32.h \
--src/quazip/quaadler32.h \
--src/quazip/JlCompress.h \
--src/quazip/ioapi.h \
--src/quazip/crypt.h \
- src/gui/toolmainwindow.h \
- src/gui/playerlistwidget.h \
- src/gui/qled.h \
-@@ -299,18 +283,6 @@ SOURCES += src/database/board.cpp \
- src/gui/engineoptionlist.cpp \
- src/gui/helpbrowser.cpp \
- src/database/downloadmanager.cpp \
--src/quazip/quazipnewinfo.cpp \
--src/quazip/quazipfile.cpp \
--src/quazip/quazipdir.cpp \
--src/quazip/quazip.cpp \
--src/quazip/quaziodevice.cpp \
--src/quazip/quagzipfile.cpp \
--src/quazip/quacrc32.cpp \
--src/quazip/quaadler32.cpp \
--src/quazip/qioapi.cpp \
--src/quazip/JlCompress.cpp \
--src/quazip/zip.c \
--src/quazip/unzip.c \
- src/gui/toolmainwindow.cpp \
- src/gui/playerlistwidget.cpp \
- src/gui/qled.cpp \
-@@ -382,12 +354,48 @@ SOURCES += src/database/board.cpp \
- src/database/streamdatabase.cpp \
- src/database/refcount.cpp
+@@ -293,22 +293,6 @@ HEADERS += src/database/board.h \
+   src/gui/textedit.h \
+   src/gui/toolmainwindow.h \
+   src/gui/translatingslider.h \
+-  src/quazip/JlCompress.h \
+-  src/quazip/crypt.h \
+-  src/quazip/ioapi.h \
+-  src/quazip/quaadler32.h \
+-  src/quazip/quachecksum32.h \
+-  src/quazip/quacrc32.h \
+-  src/quazip/quagzipfile.h \
+-  src/quazip/quaziodevice.h \
+-  src/quazip/quazip.h \
+-  src/quazip/quazip_global.h \
+-  src/quazip/quazipdir.h \
+-  src/quazip/quazipfile.h \
+-  src/quazip/quazipfileinfo.h \
+-  src/quazip/quazipnewinfo.h \
+-  src/quazip/unzip.h \
+-  src/quazip/zip.h
  
+ SOURCES += \
+   src/database/analysis.cpp \
+@@ -452,25 +436,49 @@ SOURCES += \
+   src/gui/textedit.cpp \
+   src/gui/toolmainwindow.cpp \
+   src/gui/translatingslider.cpp \
+-  src/quazip/JlCompress.cpp \
+-  src/quazip/qioapi.cpp \
+-  src/quazip/quaadler32.cpp \
+-  src/quazip/quacrc32.cpp \
+-  src/quazip/quagzipfile.cpp \
+-  src/quazip/quaziodevice.cpp \
+-  src/quaz

graphics/mtpaint: move to gtk+3, remove llvm-10 -fno-common workaround

[Charlene Wendling]

Hi,

Recently, i brought an update for mtpaint, and fixing the build for
a -fno-common patched llvm-10, that needs to be removed now [0].

I'm proposing to move mtpaint to gtk+3 while here, since we are
encouraged to move to gtk+3 after the final gtk+2 update [1]. I
totally missed that new feature while updating.

I had some linking error because "`-lX11' cannot be found". mtpaint's
configure script uses pkg-config(1) to get gtk+ linker flags, showing
clearly the difference:

> $ pkg-config --libs gtk+-x11-2.0 
> -L/usr/local/lib -L/usr/X11R6/lib -Wl,-rpath-link,/usr/X11R6/lib
> -lgtk-x11-2.0 -lgdk-x11-2.0 -lpangocairo-1.0 -lXrender -lXinerama -lXi
> -lXrandr -lXcursor -lXcomposite -lXdamage -lXfixes -lX11 -lXext
> -latk-1.0 -lcairo -lgdk_pixbuf-2.0 -lgio-2.0 -lpangoft2-1.0
> -lpango-1.0 -lgobject-2.0 -lglib-2.0 -lintl -lharfbuzz -lfontconfig
> -lfreetype -lz
> 
> $ pkg-config --libs gtk+-x11-3.0 
> -L/usr/local/lib -Wl,-rpath-link,/usr/X11R6/lib -lgtk-3 -lgdk-3
> -lpangocairo-1.0 -lpango-1.0 -lharfbuzz -latk-1.0 -lcairo-gobject
> -lcairo -lgdk_pixbuf-2.0 -lgio-2.0 -lgobject-2.0 -lglib-2.0 -lintl

The `-rpath-link' linker flag seems silently ignored by ld.lld(1), as
seen in /usr/src/gnu/llvm/lld/ELF/Options.td .

On top of that, upstream's configure does not call pkg-config(1)
against X11, but adds it to LIBS as-is, which looks wrong to me [2].

I added LDFLAGS to work around this issue.

This has been successfully built and (lightly) run tested on macppc
and amd64.


Comments/feedback are welcome,

Charlène.


[0] https://marc.info/?l=openbsd-ports&m=161204914031044&w=2
[1] https://marc.info/?l=openbsd-ports-cvs&m=160855070502243&w=2
[2]
https://github.com/wjaguar/mtPaint/blob/b0b023ba7ae231415d556ea88b2239fecf693faf/configure#L755


Index: Makefile
===
RCS file: /cvs/ports/graphics/mtpaint/Makefile,v
retrieving revision 1.6
diff -u -p -u -p -r1.6 Makefile
--- Makefile2 Feb 2021 22:41:40 -   1.6
+++ Makefile4 Jun 2021 22:43:12 -
@@ -1,8 +1,9 @@
 # $OpenBSD: Makefile,v 1.6 2021/02/02 22:41:40 cwen Exp $
 
-COMMENT =  simple GTK+2 raster painting program
+COMMENT =  simple GTK+3 raster painting program
 
 DISTNAME = mtpaint-3.50
+REVISION = 0
 EXTRACT_SUFX = .tar.bz2
 
 CATEGORIES =   graphics
@@ -12,12 +13,10 @@ HOMEPAGE =  http://mtpaint.sourceforge.ne
 # GPLv3+
 PERMIT_PACKAGE =   Yes
 
-WANTLIB += X11 Xcomposite Xcursor Xdamage Xext Xfixes Xi Xinerama
-WANTLIB += Xrandr Xrender atk-1.0 c cairo fontconfig freetype
-WANTLIB += gdk-x11-2.0 gdk_pixbuf-2.0 gio-2.0 glib-2.0 gobject-2.0
-WANTLIB += gthread-2.0 gtk-x11-2.0 harfbuzz intl jasper jpeg lcms2
-WANTLIB += m pango-1.0 pangocairo-1.0 pangoft2-1.0 png pthread
-WANTLIB += tiff z
+WANTLIB += X11 atk-1.0 c cairo cairo-gobject gdk-3 gdk_pixbuf-2.0
+WANTLIB += gio-2.0 glib-2.0 gobject-2.0 gthread-2.0 gtk-3 harfbuzz
+WANTLIB += intl jasper jpeg lcms2 m pango-1.0 pangocairo-1.0 png
+WANTLIB += pthread tiff z
 
 MASTER_SITES = ${MASTER_SITE_SOURCEFORGE:=mtpaint/}
 
@@ -26,19 +25,16 @@ LIB_DEPENDS =   devel/gettext,-runtime \
graphics/jasper \
graphics/lcms2 \
graphics/tiff \
-   x11/gtk+2
+   x11/gtk+3
 RUN_DEPENDS =  devel/desktop-file-utils
 
 USE_GMAKE =Yes
 CONFIGURE_STYLE =  simple
 CONFIGURE_ARGS =   nogif jpeg jasper staticft tiff lcms \
-   lcms2 intl man thread cflags \
+   lcms2 intl man thread cflags gtk3 \
--mandir=${TRUEPREFIX}/man
+CONFIGURE_ENV +=   LDFLAGS="${LDFLAGS} -L${X11BASE}/lib"
 
 NO_TEST =  Yes
-
-# Remove after clang update to LLVM11. Meanwhile it avoids false positives with
-# a patched clang-10 (mtpaint uses that flag only with clang-11 and gcc-10)
-CFLAGS +=  -fcommon
 
 .include 
Index: pkg/DESCR
===
RCS file: /cvs/ports/graphics/mtpaint/pkg/DESCR,v
retrieving revision 1.1.1.1
diff -u -p -u -p -r1.1.1.1 DESCR
--- pkg/DESCR   25 Apr 2016 13:22:00 -  1.1.1.1
+++ pkg/DESCR   4 Jun 2021 22:43:12 -
@@ -1,4 +1,4 @@
-mtPaint is a simple GTK+1/2 painting program designed for creating icons and
+mtPaint is a simple GTK+3 painting program designed for creating icons and
 pixel based artwork. It can edit indexed palette or 24 bit RGB images and 
offers
 basic painting and palette manipulation tools. It also has several other more
 powerful features such as channels, layers and animation. Due to its simplicity

Re: [update] games/godot 3.3 -> 3.3.1, mark BROKEN-powerpc

[Charlene Wendling]

On Tue, 08 Jun 2021 17:56:55 +0200
Omar Polo  wrote:

> Charlene Wendling  writes:
> 
> > Hi,
> >
> > On Mon, 07 Jun 2021 18:44:45 +0200
> > Omar Polo  wrote:
> >
> >> friendly ping!
> >
> > It builds and works fine on amd64 with my AMD RX560. On macppc
> > it builds but the UI is totally blank with my ATI 9600.
> > I'm proposing to mark it BROKEN there.
> >
> > Obviously, ok cwen@
> 
> Thanks for testing!
> 
> I'm fine with marking it BROKEN for powerpc, I'm just surprised that a
> bugfix release broke it :/

I did not test 3.3 back in the days, so it may come from before than
that. Since powerpc moved to clang, i'm more carefree about potentially
dangerous updates.

Re: [update] games/godot 3.3 -> 3.3.1, mark BROKEN-powerpc

[Charlene Wendling]

Hi,

On Mon, 07 Jun 2021 18:44:45 +0200
Omar Polo  wrote:

> friendly ping!

It builds and works fine on amd64 with my AMD RX560. On macppc
it builds but the UI is totally blank with my ATI 9600.
I'm proposing to mark it BROKEN there.

Obviously, ok cwen@


Index: Makefile
===
RCS file: /cvs/ports/games/godot/Makefile,v
retrieving revision 1.18
diff -u -p -u -p -r1.18 Makefile
--- Makefile3 May 2021 19:10:24 -   1.18
+++ Makefile8 Jun 2021 15:40:08 -
@@ -1,8 +1,10 @@
 # $OpenBSD: Makefile,v 1.18 2021/05/03 19:10:24 thfr Exp $
 
+BROKEN-powerpc =   fails at runtime, the UI is totally blank
+
 COMMENT =  2D and 3D game engine
 
-V =3.3
+V =3.3.2
 DISTNAME = godot-${V}-stable
 PKGNAME =  godot-${V}
 CATEGORIES =   games
Index: distinfo
===
RCS file: /cvs/ports/games/godot/distinfo,v
retrieving revision 1.6
diff -u -p -u -p -r1.6 distinfo
--- distinfo3 May 2021 19:10:24 -   1.6
+++ distinfo8 Jun 2021 15:40:08 -
@@ -1,2 +1,2 @@
-SHA256 (godot-3.3-stable.tar.xz) = /LvGqqsWBZ5mIkgsM1jVgJjTTtUeCZyFTsCtefRm1VU=
-SIZE (godot-3.3-stable.tar.xz) = 20581028
+SHA256 (godot-3.3.2-stable.tar.xz) = 
dpPE71QbgV+lA18Taq91GHow8OqNFc9O8FQzwcJGwtg=
+SIZE (godot-3.3.2-stable.tar.xz) = 20619728
Index: patches/patch-platform_x11_detect_py
===
RCS file: /cvs/ports/games/godot/patches/patch-platform_x11_detect_py,v
retrieving revision 1.4
diff -u -p -u -p -r1.4 patch-platform_x11_detect_py
--- patches/patch-platform_x11_detect_py3 May 2021 19:10:25 -   
1.4
+++ patches/patch-platform_x11_detect_py8 Jun 2021 15:40:08 -
@@ -33,7 +33,7 @@ Index: platform/x11/detect.py
  elif env["target"] == "debug":
  env.Prepend(CCFLAGS=["-ggdb"])
  env.Prepend(CCFLAGS=["-g3"])
-@@ -318,6 +301,10 @@ def configure(env):
+@@ -325,6 +308,10 @@ def configure(env):
  env.Append(CPPDEFINES=["ALSA_ENABLED", "ALSAMIDI_ENABLED"])
  else:
  print("ALSA libraries not found, disabling driver")
Index: patches/patch-thirdparty_embree_common_math_math_h
===
RCS file: patches/patch-thirdparty_embree_common_math_math_h
diff -N patches/patch-thirdparty_embree_common_math_math_h
--- /dev/null   1 Jan 1970 00:00:00 -
+++ patches/patch-thirdparty_embree_common_math_math_h  8 Jun 2021 15:40:08 
-
@@ -0,0 +1,25 @@
+$OpenBSD$
+
+add an implementation for max for long (ssize_t are long typedef'ed).
+Otherwise it will fail in parallel_partition.h:185 due to an ambiguous
+call.
+
+Index: thirdparty/embree/common/math/math.h
+--- thirdparty/embree/common/math/math.h.orig
 thirdparty/embree/common/math/math.h
+@@ -242,6 +242,7 @@ namespace embree
+ __forceinline T twice(const T& a) { return a+a; }
+ 
+   __forceinline  int min(int  a, int  b) { return a __forceinline T mini(const T& a, const T& b, const T& 
c, const T& d, const T& e) { return mini(mini(mini(a,b),mini(c,d)),e); }
+ 
+   __forceinline  int max(int  a, int  b) { return a

Re: lang/crystal: unbreak, update to 1.0.0

[Charlene Wendling]

On Mon, 7 Jun 2021 08:11:31 -0500
joshua stein  wrote:

> I sent an earlier version to the maintainer and pinged them on IRC 
> but got no response.

I've built and run-tested that update with a basic hello world, and met
no issues. 

OK cwen@

[ld.bfd] Unbreak x11/gnome/seahorse

[Charlene Wendling]

Hi,

> http://build-failures.rhaalovely.net/sparc64/2021-05-22/x11/gnome/seahorse.log
(same on macppc, no mips64 bulk report available)

The `--as-needed' linker flag strikes again, it builds fine on macppc
once it is removed [0]. No REVISION bump is needed, since this version
has never been built on sparc64 and macppc.

OK?

Charlène.


[0] https://bin.charlenew.xyz/seahorse.log


Index: Makefile
===
RCS file: /cvs/ports/x11/gnome/seahorse/Makefile,v
retrieving revision 1.129
diff -u -p -u -p -r1.129 Makefile
--- Makefile14 May 2021 08:30:12 -  1.129
+++ Makefile6 Jun 2021 21:58:43 -
@@ -26,6 +26,11 @@ DEBUG_PACKAGES=  ${BUILD_PACKAGES}
 .if !${PROPERTIES:Mclang}
 CFLAGS +=  -std=gnu99
 .endif
+# disable meson's default of using "-Wl,--as-needed" on ld.bfd arches;
+# build failures due to undefined references are often seen.
+.if !${PROPERTIES:Mlld}
+CONFIGURE_ARGS +=  -Db_asneeded=false
+.endif
 
 LIB_DEPENDS=   databases/openldap \
devel/libsoup \

Re: Mark graphics/asymptote BROKEN-sparc64 or remove it?

[Charlene Wendling]

Hi,

On Sun, 23 May 2021 23:47:03 -0400
Kurt Mosiejczuk  wrote:

> I'm unsure whether to mark graphics/asymptote BROKEN-sparc64 or to
> nominate it for removal.
>
> It's marked NOT_FOR_ARCHES with CLANG_ARCHES and marked BROKEN-hppa. 
> 
> So I don't know if it actually compiles for any arches at this point.

> > https://marc.info/?l=openbsd-ports&m=157875638611191&w=2

We were discussing the removal of that port last year already, and
no one seems interested into fixing it, or needing it on clang arches
as well.

OK cwen@ to remove the port

> Anyone in favor of keeping it with it marked BROKEN-sparc64?
> 
> --Kurt
> 
> Index: Makefile
> ===
> RCS file: /cvs/ports/graphics/asymptote/Makefile,v
> retrieving revision 1.39
> diff -u -p -r1.39 Makefile
> --- Makefile  23 Feb 2021 19:39:24 -  1.39
> +++ Makefile  24 May 2021 03:43:10 -
> @@ -7,6 +7,7 @@ NOT_FOR_ARCHS=${CLANG_ARCHS} # hangs d
>  
>  COMMENT= powerful descriptive vector graphics language
>  BROKEN-hppa= ghostscript segfaults during build
> +BROKEN-sparc64=  undefined reference to
> GC_throw_bad_alloc 
>  DISTNAME=asymptote-2.41.src
>  PKGNAME= asymptote-2.41
> 

Re: [MAINTAINER UPDATE] x11/bemenu -> 0.6.0

[Charlene Wendling]

Hi,

It builds and works fine for me on macppc and amd64 (i tested the x11
and curses variants).

OK cwen@

On Sat, 22 May 2021 12:15:38 +
Dimitri Karamazov  wrote:

> Update x11/bemenu to 0.6.0
> 
> Dynamic export changes:
> added:
> bm_menu_add_item_at
> bm_menu_get_password
> bm_menu_set_monitor_name
> bm_menu_set_password
> 
> removed:
> bm_menu_add_items_at
> filter_dmenu_fun
> 
> PLT removed:
> filter_dmenu_fun
> 
> Build & Run tested on amd64
> 
> regards,
>   Dimitri

[macppc, !rust] cad/kicad* and cad/oce adjustments

[Charlene Wendling]

Hi,

I've removed BROKEN-powerpc lines in ports that now need rust to be
built, since this arch has no rust support. kicad is one (and the last)
of them:

> $ cd /usr/ports/cad/kicad
> $ make full-build-depends | grep rust
> rust-1.52.1

It requires changes for other ports, and also other !rust archs,
so i preferred to share the diff for this one.

This is very unlikely that kicad will be rust-free in the future, but
who knows :) Also, i mentioned the ridiculous oce build time and lack
of consumers as BROKEN-powerpc reason.

Comments/feedback are welcome,

Charlène. 


Index: oce/Makefile
===
RCS file: /cvs/ports/cad/oce/Makefile,v
retrieving revision 1.3
diff -u -p -r1.3 Makefile
--- oce/Makefile30 Apr 2020 21:05:03 -  1.3
+++ oce/Makefile16 May 2021 19:21:14 -
@@ -1,6 +1,6 @@
 # $OpenBSD: Makefile,v 1.3 2020/04/30 21:05:03 cwen Exp $
 
-BROKEN-powerpc =   kicad is broken, this is the only consumer (save bulk 
time)
+BROKEN-powerpc =   no consumers, saves 53 build hours
 
 COMMENT =  C++ 3d modeling library
 DPB_PROPERTIES = parallel
Index: kicad/Makefile
===
RCS file: /cvs/ports/cad/kicad/Makefile,v
retrieving revision 1.45
diff -u -p -r1.45 Makefile
--- kicad/Makefile  23 Feb 2021 19:39:10 -  1.45
+++ kicad/Makefile  16 May 2021 19:21:14 -
@@ -1,7 +1,7 @@
 # $OpenBSD: Makefile,v 1.45 2021/02/23 19:39:10 sthen Exp $
 
-# Sync cad/kicad-share/Makefile.inc with archs where kicad is broken
-BROKEN-powerpc =   segfaults when trying to run the PCB editor
+# Sync cad/kicad-share/Makefile.inc with archs where kicad is broken,
+# or if kicad has no more rust in its build dependencies chain again.
 
 V =5.1.7
 COMMENT =  schematic and PCB editing software
Index: kicad-share/Makefile.inc
===
RCS file: /cvs/ports/cad/kicad-share/Makefile.inc,v
retrieving revision 1.6
diff -u -p -r1.6 Makefile.inc
--- kicad-share/Makefile.inc19 Feb 2021 14:48:11 -  1.6
+++ kicad-share/Makefile.inc16 May 2021 19:21:14 -
@@ -1,6 +1,6 @@
 # $OpenBSD: Makefile.inc,v 1.6 2021/02/19 14:48:11 tracey Exp $
 
-NOT_FOR_ARCHS =powerpc
+ONLY_FOR_ARCHS =   ${RUST_ARCHS}
 
 PKG_ARCH ?=*
 

[macppc] Unbreak x11/qt5/qtwebkit

[Charlene Wendling]

Hi, 

qtwebkit does not build in the current macppc bulk, the first with
clang-11:

> ../Source/WTF/wtf/Atomics.cpp:63:9: error: definition of builtin function 
> '__sync_add_and_fetch_8'
> int64_t __sync_add_and_fetch_8(int64_t volatile* addend, int64_t value)
> ../Source/WTF/wtf/Atomics.cpp:68:9: error: definition of builtin function 
> '__sync_sub_and_fetch_8'
> int64_t __sync_sub_and_fetch_8(int64_t volatile* addend, int64_t value)

__GCC_HAVE_SYNC_COMPARE_AND_SWAP_8 is not defined on powerpc because
there is no instruction on that arch for that. Nonetheless we
can use 8 byte __sync_* operators [0].

This code addresses issues with old gcc versions that are not used on
macppc anymore. This has been removed from the latest qtwebkit
developments [1]. As such i've simply excluded powerpc from that.

With the below diff i can successfully build qtwebkit on macppc [2],
as expected it has no impact on amd64.

Comments/feedback are welcome,

Charlène.


[0] https://marc.info/?l=openbsd-cvs&m=159131001523468&w=2
[1] https://github.com/qtwebkit/qtwebkit/tree/qtwebkit-dev-wip/Source/WTF/wtf
[2] https://bin.charlenew.xyz/qtwebkit.log


Index: Makefile
===
RCS file: /cvs/ports/x11/qt5/qtwebkit/Makefile,v
retrieving revision 1.31
diff -u -p -u -p -r1.31 Makefile
--- Makefile8 May 2021 14:58:43 -   1.31
+++ Makefile10 May 2021 13:35:03 -
@@ -15,7 +15,7 @@ PKGNAME = qtwebkit-${VERSION}
 PKGSPEC =  qtwebkit->=${VERSION}v0
 
 EPOCH =0
-REVISION = 3
+REVISION = 4
 
 
 SHARED_LIBS +=  Qt5WebKit   3.1 # 5.9
Index: patches/patch-Source_WTF_wtf_Atomics_cpp
===
RCS file: patches/patch-Source_WTF_wtf_Atomics_cpp
diff -N patches/patch-Source_WTF_wtf_Atomics_cpp
--- /dev/null   1 Jan 1970 00:00:00 -
+++ patches/patch-Source_WTF_wtf_Atomics_cpp10 May 2021 13:35:03 -
@@ -0,0 +1,17 @@
+$OpenBSD$
+
+powerpc fix: don't redefine 64-bits __sync_* built-ins operators
+
+Index: Source/WTF/wtf/Atomics.cpp
+--- Source/WTF/wtf/Atomics.cpp.orig
 Source/WTF/wtf/Atomics.cpp
+@@ -32,7 +32,8 @@
+ // (http://gcc.gnu.org/bugzilla/show_bug.cgi?id=56296). GCC >= 4.8 will 
support __atomic_* builtin
+ // functions for this purpose for all the GCC targets, but for current 
compilers we have to include
+ // our own implementation.
+-#if COMPILER(GCC_OR_CLANG) && !defined(__GCC_HAVE_SYNC_COMPARE_AND_SWAP_8) && 
USE(PTHREADS)
++#if !defined(__powerpc__) \
++&& COMPILER(GCC_OR_CLANG) && !defined(__GCC_HAVE_SYNC_COMPARE_AND_SWAP_8) 
&& USE(PTHREADS)
+ 
+ #include "ThreadingPrimitives.h"
+ 

[update] devel/p5-Devel-Hide: update to 0.0014

[Charlene Wendling]

Hi,

Here is an update to Devel::Hide, to version 0.0014. There is no code
change, it only adds a version requirement for Test::More [0].

Port-wise, i'm dropping maintainership.

I've still tested direct depends [1], they all pass.

Comments/feedback are welcome,

Charlène. 


[0]
https://metacpan.org/diff/file?target=DCANTRELL/Devel-Hide-0.0014/&source=DCANTRELL%2FDevel-Hide-0.0013
[1] https://bin.charlenew.xyz/p5-Devel-Hide.tgz 


Index: Makefile
===
RCS file: /cvs/ports/devel/p5-Devel-Hide/Makefile,v
retrieving revision 1.4
diff -u -p -u -p -r1.4 Makefile
--- Makefile3 Jul 2020 21:44:49 -   1.4
+++ Makefile28 Apr 2021 15:52:19 -
@@ -2,11 +2,8 @@
 
 COMMENT =  forces the unavailability of specified Perl modules
 
-DISTNAME = Devel-Hide-0.0013
+DISTNAME = Devel-Hide-0.0014
 CATEGORIES =   devel
-REVISION = 0
-
-MAINTAINER =   Charlene Wendling 
 
 MODULES =  cpan
 PKG_ARCH = *
Index: distinfo
===
RCS file: /cvs/ports/devel/p5-Devel-Hide/distinfo,v
retrieving revision 1.2
diff -u -p -u -p -r1.2 distinfo
--- distinfo15 Jun 2020 11:43:22 -  1.2
+++ distinfo28 Apr 2021 15:52:19 -
@@ -1,2 +1,2 @@
-SHA256 (Devel-Hide-0.0013.tar.gz) = 
b8O7z08CU6blFoKWyQmTjBPIhpI6eZWslp5Hrfzwfss=
-SIZE (Devel-Hide-0.0013.tar.gz) = 9787
+SHA256 (Devel-Hide-0.0014.tar.gz) = 
ONRTnGna1gl3zXBaENYo8B9jMSLZlyogv0eHHzDpYmU=
+SIZE (Devel-Hide-0.0014.tar.gz) = 10036

Re: comms/sigrok/*: conflict between usbhid(3) and comms/libhidapi

[Charlene Wendling]

Ping. 

On Sun, 4 Apr 2021 00:14:56 +0200
Charlene Wendling  wrote:

> Hi,
> 
> > http://build-failures.rhaalovely.net/powerpc/2021-03-11/comms/sigrok/pulseview.log
> > http://build-failures.rhaalovely.net/powerpc/2021-03-11/comms/sigrok/sigrok-cli.log
> > http://build-failures.rhaalovely.net/sparc64/2021-03-31/comms/sigrok/pulseview.log
> > http://build-failures.rhaalovely.net/sparc64/2021-03-31/comms/sigrok/sigrok-cli.log
> 
> It looks like it happens because our usbhid defines hid_init(3),
> but the libhidapi's one must be used instead. As mentioned in
> comms/libhidapi/patches/patch-hidapi_hidapi_h, we should use
> hidapi_hid_init().
> 
> That diff fixes the concerned libsigrok [0] consumers' build [1] [2]
> on macppc, and amd64 is still fine.
> 
> As far as macppc go, i may not be able to apply this patch for the
> current bulk.
> 
> Comments/feedback are welcome,
> 
> Charlène.
> 
> 
> [0] https://bin.charlenew.xyz/libsigrok.log
> [1] https://bin.charlenew.xyz/sigrok-cli.log
> [2] https://bin.charlenew.xyz/pulseview.log


Index: Makefile
===
RCS file: /cvs/ports/comms/sigrok/libsigrok/Makefile,v
retrieving revision 1.11
diff -u -p -u -p -r1.11 Makefile
--- Makefile6 Mar 2021 22:42:43 -   1.11
+++ Makefile11 Apr 2021 13:49:35 -
@@ -5,7 +5,7 @@ COMMENT =   sigrok hardware driver librar
 SIGROK_PROJECT =   libsigrok
 SIGROK_VERSION =   0.5.2
 MODPY_EGG_VERSION =${SIGROK_VERSION}
-REVISION = 0
+REVISION = 1
 
 SHARED_LIBS +=  sigrok0.1 # 0.0
 SHARED_LIBS +=  sigrokcxx 0.1 # 4.0
Index: patches/patch-src_backend_c
===
RCS file: patches/patch-src_backend_c
diff -N patches/patch-src_backend_c
--- /dev/null   1 Jan 1970 00:00:00 -
+++ patches/patch-src_backend_c 11 Apr 2021 13:49:35 -
@@ -0,0 +1,22 @@
+$OpenBSD$
+
+Fix for:
+src/backend.c:613:6: warning: implicit declaration of function 'hid_init'
+is invalid in C99
+On ld.bfd arches, this led to undefined references at build time.
+
+It appears that comms/libhidapi redefines hid_init(3) to hidapi_hid_init().
+
+Index: src/backend.c
+--- src/backend.c.orig
 src/backend.c
+@@ -23,6 +23,9 @@
+ #ifdef _WIN32
+ #include 
+ #endif
++#ifdef __OpenBSD__
++#define hid_init hidapi_hid_init
++#endif
+ #include 
+ #include "libsigrok-internal.h"
+ 

comms/sigrok/*: conflict between usbhid(3) and comms/libhidapi

[Charlene Wendling]

Hi,

> http://build-failures.rhaalovely.net/powerpc/2021-03-11/comms/sigrok/pulseview.log
> http://build-failures.rhaalovely.net/powerpc/2021-03-11/comms/sigrok/sigrok-cli.log
> http://build-failures.rhaalovely.net/sparc64/2021-03-31/comms/sigrok/pulseview.log
> http://build-failures.rhaalovely.net/sparc64/2021-03-31/comms/sigrok/sigrok-cli.log

It looks like it happens because our usbhid defines hid_init(3),
but the libhidapi's one must be used instead. As mentioned in
comms/libhidapi/patches/patch-hidapi_hidapi_h, we should use
hidapi_hid_init().

That diff fixes the concerned libsigrok [0] consumers' build [1] [2]
on macppc, and amd64 is still fine.

As far as macppc go, i may not be able to apply this patch for the
current bulk.

Comments/feedback are welcome,

Charlène.


[0] https://bin.charlenew.xyz/libsigrok.log
[1] https://bin.charlenew.xyz/sigrok-cli.log
[2] https://bin.charlenew.xyz/pulseview.log


Index: Makefile
===
RCS file: /cvs/ports/comms/sigrok/libsigrok/Makefile,v
retrieving revision 1.11
diff -u -p -u -p -r1.11 Makefile
--- Makefile6 Mar 2021 22:42:43 -   1.11
+++ Makefile3 Apr 2021 20:57:49 -
@@ -5,7 +5,7 @@ COMMENT =   sigrok hardware driver librar
 SIGROK_PROJECT =   libsigrok
 SIGROK_VERSION =   0.5.2
 MODPY_EGG_VERSION =${SIGROK_VERSION}
-REVISION = 0
+REVISION = 1
 
 SHARED_LIBS +=  sigrok0.1 # 0.0
 SHARED_LIBS +=  sigrokcxx 0.1 # 4.0
Index: patches/patch-src_backend_c
===
RCS file: patches/patch-src_backend_c
diff -N patches/patch-src_backend_c
--- /dev/null   1 Jan 1970 00:00:00 -
+++ patches/patch-src_backend_c 3 Apr 2021 20:57:49 -
@@ -0,0 +1,23 @@
+$OpenBSD$
+
+Fix for:
+src/backend.c:613:6: warning: implicit declaration of function 'hid_init'
+is invalid in C99
+On ld.bfd arches, this led to undefined references at build time.
+
+It appears that comms/libhidapi redefines hid_init(3) to hidapi_hid_init()
+to avoid clashes with base usbhid(3).
+
+Index: src/backend.c
+--- src/backend.c.orig
 src/backend.c
+@@ -23,6 +23,9 @@
+ #ifdef _WIN32
+ #include 
+ #endif
++#ifdef __OpenBSD__
++#define hid_init hidapi_hid_init
++#endif
+ #include 
+ #include "libsigrok-internal.h"
+ 

[ld.bfd] Unbreak x11/gtk+4

[Charlene Wendling]

Hi,

> http://build-failures.rhaalovely.net/sparc64/2021-03-31/x11/gtk+4.log

I can reproduce that on macppc at home (the current bulk builds the
previous version).

ld.bfd often does not like "-Wl,--as-needed". The below diff allows
gtk+4 to be built on macppc [0]. This version has never been built
on ld.bfd archs, so no revision bump is needed.

OK ?

Charlène.


[0] https://bin.charlenew.xyz/gtk4.success.log


Index: Makefile
===
RCS file: /cvs/ports/x11/gtk+4/Makefile,v
retrieving revision 1.42
diff -u -p -u -p -r1.42 Makefile
--- Makefile31 Mar 2021 13:42:46 -  1.42
+++ Makefile3 Apr 2021 12:14:54 -
@@ -118,6 +118,12 @@ CONFIGURE_ARGS +=  -Dmedia-ffmpeg=disable
 CFLAGS +=  -Wno-error
 .endif
 
+# disable meson's default of using "-Wl,--as-needed" on ld.bfd arches;
+# build failures due to undefined references are often seen.
+.if !${PROPERTIES:Mlld}
+CONFIGURE_ARGS +=  -Db_asneeded=false
+.endif
+
 pre-configure:
${SUBST_CMD} ${WRKSRC}/modules/printbackends/meson.build
 

[macppc] Unbreak www/webkitgtk4

[Charlene Wendling]

Hi,

Building webkitgtk4 on macppc since 2.32 fails at the patching stage.

The powerpc specific patch has been upstreamed [0], so it's not needed
anymore. Also the ENABLE_GRAPHICS_CONTEXT_GL option is now
USE_OPENGL_OR_ES [1]

I needed an additional patch specific to archs where USE_OPENGL_OR_ES=OFF,
only powerpc for now, a recent commit breaks the build. See the patch
comment for more infos.

With the below diff, webkitgtk4 is built [3]. No REVISION bump is
needed since this version has never been built on powerpc and the
whole diff only impacts this arch.

As far as runtime goes, javascript runtime is mostly broken like it
used to be before 2.30. I tried to build it with OpenGL support,
but colors are still off.

Comments/feedback are welcome.

Charlène.


[0] https://trac.webkit.org/changeset/271220/webkit
[1] https://trac.webkit.org/changeset/273104/webkit
[2] https://trac.webkit.org/changeset/271171/webkit
[3] https://bin.charlenew.xyz/webkitgtk4.success.log


Index: Makefile
===
RCS file: /cvs/ports/www/webkitgtk4/Makefile,v
retrieving revision 1.139
diff -u -p -u -p -r1.139 Makefile
--- Makefile30 Mar 2021 14:52:14 -  1.139
+++ Makefile31 Mar 2021 13:55:01 -
@@ -124,9 +124,8 @@ LDFLAGS +=  -Wl,--no-keep-memory
 
 .if ${MACHINE_ARCH} == "powerpc"
 # XXX fix colors being off, it would be nice to fix the code
-CONFIGURE_ARGS +=  -DENABLE_GRAPHICS_CONTEXT_GL=OFF
+CONFIGURE_ARGS +=  -DUSE_OPENGL_OR_ES=OFF
 LDFLAGS += -Wl,--relax
-PATCH_LIST =   patch-* powerpc-patch-*
 .endif
 
 post-extract:
Index: 
patches/patch-Source_WebKit_WebProcess_WebPage_CoordinatedGraphics_LayerTreeHost_h
===
RCS file: 
patches/patch-Source_WebKit_WebProcess_WebPage_CoordinatedGraphics_LayerTreeHost_h
diff -N 
patches/patch-Source_WebKit_WebProcess_WebPage_CoordinatedGraphics_LayerTreeHost_h
--- /dev/null   1 Jan 1970 00:00:00 -
+++ 
patches/patch-Source_WebKit_WebProcess_WebPage_CoordinatedGraphics_LayerTreeHost_h
  31 Mar 2021 13:55:01 -
@@ -0,0 +1,27 @@
+$OpenBSD$
+
+Fix for:
+LayerTreeHost.h:216:28: error: out-of-line definition of 'forceRepaintAsync'
+does not match any declaration in 'WebKit::LayerTreeHost'
+LayerTreeHost.h:82:28: note: type of 1st parameter of member declaration does
+not match definition ('CompletionHandler &&' vs
+'CompletionHandler &')
+
+LayerTreeHost::forceRepaintAsync used to be a bool method, but is now void.
+It only impacts powerpc where OpenGL is disabled.
+
+That error has been introduced by:
+https://trac.webkit.org/changeset/271171/webkit/trunk/Source/WebKit/WebProcess/WebPage/CoordinatedGraphics/LayerTreeHost.cpp
+
+Index: Source/WebKit/WebProcess/WebPage/CoordinatedGraphics/LayerTreeHost.h
+--- Source/WebKit/WebProcess/WebPage/CoordinatedGraphics/LayerTreeHost.h.orig
 Source/WebKit/WebProcess/WebPage/CoordinatedGraphics/LayerTreeHost.h
+@@ -213,7 +213,7 @@ inline void LayerTreeHost::setRootCompositingLayer(Web
+ inline void LayerTreeHost::setViewOverlayRootLayer(WebCore::GraphicsLayer*) { 
}
+ inline void LayerTreeHost::scrollNonCompositedContents(const 
WebCore::IntRect&) { }
+ inline void LayerTreeHost::forceRepaint() { }
+-inline bool LayerTreeHost::forceRepaintAsync(CompletionHandler&) { 
return false; }
++inline void LayerTreeHost::forceRepaintAsync(CompletionHandler&&) { }
+ inline void LayerTreeHost::sizeDidChange(const WebCore::IntSize&) { }
+ inline void LayerTreeHost::pauseRendering() { }
+ inline void LayerTreeHost::resumeRendering() { }
Index: 
patches/powerpc-patch-Source_JavaScriptCore_llint_LowLevelInterpreter32_64_asm
===
RCS file: 
patches/powerpc-patch-Source_JavaScriptCore_llint_LowLevelInterpreter32_64_asm
diff -N 
patches/powerpc-patch-Source_JavaScriptCore_llint_LowLevelInterpreter32_64_asm
--- 
patches/powerpc-patch-Source_JavaScriptCore_llint_LowLevelInterpreter32_64_asm  
11 Nov 2020 21:27:03 -  1.1
+++ /dev/null   1 Jan 1970 00:00:00 -
@@ -1,58 +0,0 @@
-$OpenBSD: 
powerpc-patch-Source_JavaScriptCore_llint_LowLevelInterpreter32_64_asm,v 1.1 
2020/11/11 21:27:03 cwen Exp $
-
-Fix Javascript runtime on powerpc, from
-https://bugs.webkit.org/show_bug.cgi?id=211592
-
-Index: Source/JavaScriptCore/llint/LowLevelInterpreter32_64.asm
 Source/JavaScriptCore/llint/LowLevelInterpreter32_64.asm.orig
-+++ Source/JavaScriptCore/llint/LowLevelInterpreter32_64.asm
-@@ -1552,13 +1552,21 @@ llintOpWithMetadata(op_put_by_id, OpPutById, macro (si
- loadp StructureChain::m_vector[t3], t3
- assert(macro (ok) btpnz t3, ok end)
- 
--loadp Structure::m_prototype[t2], t2
-+if JSVALUE64
-+loadp Structure::m_prototype[t2], t2
-+else
-+loadp Structure::m_prototype + PayloadOffset[t2], t2
-+end
- btpz t2, .opPutByIdTransitionChainDone
- .opPutByIdTransitionChainLoop:
- l

[BE_ARCHS] Unbreak games/egoboo

[Charlene Wendling]

Hi !

> http://build-failures.rhaalovely.net/sparc64/2021-03-21/games/egoboo.log
> http://build-failures.rhaalovely.net/powerpc/2021-02-22/games/egoboo.log
> http://build-failures.rhaalovely.net/mips64/2021-03-21/games/egoboo.log

There is a typo in the big endian code there. It's 'SDL_Swap32'.

With the below diff, it builds and works fine on macppc, colors are
not off. I added a safety revision bump as well.

Comments/feedback are welcome,

Charlène.


[0] https://wiki.libsdl.org/SDL_Swap32


Index: Makefile
===
RCS file: /cvs/ports/games/egoboo/Makefile,v
retrieving revision 1.11
diff -u -p -u -p -r1.11 Makefile
--- Makefile8 Mar 2021 18:20:14 -   1.11
+++ Makefile26 Mar 2021 16:03:56 -
@@ -3,6 +3,7 @@
 COMMENT =  dungeon-crawling action-RPG adventure
 
 DISTNAME = egoboo-2.8.1
+REVISION = 0
 CATEGORIES =   games
 
 HOMEPAGE = http://egoboo.sourceforge.net
@@ -21,7 +22,8 @@ LIB_DEPENDS = devel/sdl-image \
 FIX_CRLF_FILES =   src/Makefile src/enet/Makefile src/game/Makefile \
src/game/extensions/ogl_texture.c \
src/game/game.c \
-   src/game/platform/file_linux.c
+   src/game/platform/file_linux.c \
+   src/game/md2.c
 
 USE_GMAKE =Yes
 WRKSRC =   ${WRKDIST}/src
Index: patches/patch-src_game_md2_c
===
RCS file: patches/patch-src_game_md2_c
diff -N patches/patch-src_game_md2_c
--- /dev/null   1 Jan 1970 00:00:00 -
+++ patches/patch-src_game_md2_c26 Mar 2021 16:03:56 -
@@ -0,0 +1,16 @@
+$OpenBSD$
+
+Fix the build on BE_ARCHS.
+
+Index: src/game/md2.c
+--- src/game/md2.c.orig
 src/game/md2.c
+@@ -511,7 +511,7 @@ MD2_Model_t* md2_load( const char * szFilename, MD2_Mo
+ int i;
+ for ( i = 0; i < cmd->command_count; i++ )
+ {
+-cmd->data[i].index = SDL_swap32( cmd->data[i].s );
++cmd->data[i].index = SDL_Swap32( cmd->data[i].s );
+ cmd->data[i].s = ENDIAN_FLOAT( cmd->data[i].s );
+ cmd->data[i].t = ENDIAN_FLOAT( cmd->data[i].t );
+ };

Re: WIP UPDATE: games/openttd

[Charlene Wendling]

Hi,

On Sun, 28 Feb 2021 23:09:24 -0700
"Anthony J. Bentley"  wrote:

> Hi,
> 
> Here's an update to openttd-1.11.0-beta2.
> 
> Upstream claims new OpenGL support, faster video, and 60fps.
> 
> The build system has been switched to CMake. This led to the GTK icons
> and .desktop file not being installed, a bug upstream claims will be
> fixed before release; see:
>   https://github.com/OpenTTD/OpenTTD/issues/8613
>   https://github.com/OpenTTD/OpenTTD/issues/8778
> 
> The endian patch no longer applies and seems to have been refactored
> although the original pull request was merged. I'd appreciate a
> test on PowerPC to make sure this is still correct.

I've built and tested that update on macppc, colors are still correct.

Re: x11/agar update + -fno-common fix

[Charlene Wendling]

Hi,

On Tue, 23 Feb 2021 02:28:47 -0700
"Anthony J. Bentley"  wrote:

> Hi,
> 
> Here's an update to agar-1.6.0 and agartest-1.6.0.
> 
> The update is actually still broken with -fno-common, but it's only
> one variable so it was easy enough to fix for real.
> 
> There was a lot of weird churn in this update. For instance, there's
> no point in enabling iconv because it will unfixably break the build.
> The end result seems to work though, as demonstrated by running the
> tests in agartest.
> 
> ok?

OK cwen@, but an extra patch is needed to make agar built on powerpc.
The 'rlwimi' powerpc instruction wants a 32-bit integer to work with.
It was already present in the 1.5.0 patches but in a different place.

Index: Makefile.inc
===
RCS file: /cvs/ports/x11/agar/Makefile.inc,v
retrieving revision 1.4
diff -u -p -u -p -r1.4 Makefile.inc
--- Makefile.inc9 Nov 2020 21:48:25 -   1.4
+++ Makefile.inc27 Feb 2021 08:51:04 -
@@ -1,6 +1,6 @@
 # $OpenBSD: Makefile.inc,v 1.4 2020/11/09 21:48:25 cwen Exp $
 
-V ?=   1.5.0
+V ?=   1.6.0
 DISTNAME ?=agar-$V
 
 CATEGORIES ?=  x11
@@ -10,7 +10,7 @@ HOMEPAGE ?=   https://libagar.org/
 # BSD
 PERMIT_PACKAGE ?=  Yes
 
-MASTER_SITES ?=http://stable.hypertriton.com/agar/
+MASTER_SITES ?=https://stable.hypertriton.com/agar/
 
 CONFIGURE_STYLE ?= simple
 
Index: agar/Makefile
===
RCS file: /cvs/ports/x11/agar/agar/Makefile,v
retrieving revision 1.7
diff -u -p -u -p -r1.7 Makefile
--- agar/Makefile   1 Aug 2020 18:47:33 -   1.7
+++ agar/Makefile   27 Feb 2021 08:51:04 -
@@ -3,31 +3,30 @@
 COMMENT =  cross-platform widget toolkit
 
 PKGNAME =  agar-$V
-REVISION = 4
 
-SHARED_LIBS +=  ag_au 0.0 # 5.0
-SHARED_LIBS +=  ag_core   0.0 # 5.0
-SHARED_LIBS +=  ag_dev0.0 # 5.0
-SHARED_LIBS +=  ag_gui0.0 # 5.0
-SHARED_LIBS +=  ag_math   0.0 # 5.0
-SHARED_LIBS +=  ag_vg 0.0 # 5.0
-
-WANTLIB += GL SDL X11 X11-xcb Xdamage Xext Xfixes Xinerama Xxf86vm
-WANTLIB += db drm expat fontconfig freetype glapi iconv jpeg m
-WANTLIB += png pthread sndio usbhid xcb xcb-dri2
-WANTLIB += xcb-glx z
+SHARED_LIBS +=  ag_au 0.1 # 5.0
+SHARED_LIBS +=  ag_core   1.0 # 5.0
+SHARED_LIBS +=  ag_gui1.0 # 5.0
+SHARED_LIBS +=  ag_map0.0 # 5.0
+SHARED_LIBS +=  ag_math   1.0 # 5.0
+SHARED_LIBS +=  ag_net0.0 # 5.0
+SHARED_LIBS +=  ag_sg 0.0 # 5.0
+SHARED_LIBS +=  ag_sk 0.0 # 5.0
+SHARED_LIBS +=  ag_vg 1.0 # 5.0
 
 LIB_DEPENDS =  audio/libsndfile \
audio/portaudio-svn \
-   converters/libiconv \
-   databases/db/v4 \
devel/sdl \
graphics/jpeg \
graphics/png
 
-CONFIGURE_ARGS +=  --with-gl=${X11BASE} \
+CONFIGURE_ARGS +=  --enable-au \
+   --enable-map \
+   --with-gl=${X11BASE} \
--with-x=${X11BASE} \
--with-portaudio=${LOCALBASE} \
--with-sndfile=${LOCALBASE}
+
+MAKE_FLAGS =   PKGCONFIG_LIBDIR=${PREFIX}/lib/pkgconfig
 
 .include 
Index: agar/distinfo
===
RCS file: /cvs/ports/x11/agar/agar/distinfo,v
retrieving revision 1.1.1.1
diff -u -p -u -p -r1.1.1.1 distinfo
--- agar/distinfo   10 Oct 2016 09:33:17 -  1.1.1.1
+++ agar/distinfo   27 Feb 2021 08:51:04 -
@@ -1,2 +1,2 @@
-SHA256 (agar-1.5.0.tar.gz) = gjQt7TQsV4FBmEvv6TGPPTdhduX0J64yePiYXyZmPAA=
-SIZE (agar-1.5.0.tar.gz) = 2427644
+SHA256 (agar-1.6.0.tar.gz) = dQAC8LNo+1guRWlqjfxT+0X1U2eSp/BSDNzCC0iMni4=
+SIZE (agar-1.6.0.tar.gz) = 22133371
Index: agar/patches/patch-Makefile
===
RCS file: /cvs/ports/x11/agar/agar/patches/patch-Makefile,v
retrieving revision 1.1.1.1
diff -u -p -u -p -r1.1.1.1 patch-Makefile
--- agar/patches/patch-Makefile 10 Oct 2016 09:33:17 -  1.1.1.1
+++ agar/patches/patch-Makefile 27 Feb 2021 08:51:04 -
@@ -1,15 +1,17 @@
 $OpenBSD: patch-Makefile,v 1.1.1.1 2016/10/10 09:33:17 bentley Exp $
 Makefile.orig  Mon Aug 24 03:08:39 2015
-+++ Makefile   Mon Aug 24 03:08:48 2015
-@@ -83,9 +83,9 @@ install-config:
-   ${SUDO} ${INSTALL_PROG} $$PROG ${DESTDIR}${BINDIR}; \
-   done
-   @echo "${INSTALL_DATA_DIR} ${PREFIX}/share/aclocal"
--  @${SUDO} ${INSTALL_DATA_DIR} ${DESTDIR}${PREFIX}/share/aclocal
-+  @${SUDO} ${INSTAL

Re: powerpc bulk build report

[Charlene Wendling]

Hi,

On Sun, 21 Feb 2021 13:29:53 -0700 (MST)
c...@openbsd.org wrote:

> Bulk build on macppc-0.ports.openbsd.org
> 
> Started : Thu Feb  4 17:42:22 MST 2021
> Finished: Sun Feb 21 13:29:25 MST 2021
> Duration: 16 Days 19 hours 47 minutes

> http://build-failures.rhaalovely.net/powerpc/2021-02-04/graphics/openvdb.log

This new port won't be built on powerpc, see
https://marc.info/?l=openbsd-ports-cvs&m=161387035817473&w=2

> http://build-failures.rhaalovely.net/powerpc/2021-02-04/shells/ksh93.log

Marked BROKEN, as seen on powerpc64.

Re: [update] sysutils/wmcube 0.98 -> 1.0.2 (+ -fno-common fix)

[Charlene Wendling]

Ping. 

On Fri, 5 Feb 2021 00:46:13 +0100
Charlene Wendling  wrote:

> 
> This is an update for wmcube, to 1.0.2. This adds color schemes,
> smooth fonts, more logos, a manual page and various fixes.
> 
> Port-wise:
> 
> - Move to dockapps.net, HOMEPAGE is not reachable
> - Fix an hardcoded path to logo files (introduced in 1.0.0,
>   so our current package is not affected)
> - Remove the Makefile patch, and override things with MAKE_FLAGS. That
>   also allows us to simplify the port, since we won't have to fix
> MSDOS line endings, and that's why i've attached the diff
> - Due to the previous change, don't use upstream's install target,
>   avoiding further patching
> - Drop the wmgeneral/list.h patch. Upstream "fixed" that in 1.0.2
> - Fix the build with -fno-common
> 
> Testing:
> 
> It has been built and tested on macppc and amd64 without issues.
> 
> OK? 
> 
> Charlène.


wmcube.diff
Description: Binary data

[macppc, all archs] net/echoping wants libm

[Charlene Wendling]

Hi,

I've spotted that echoping's build is broken in the current macppc bulk:

> util.o: In function `tvstddev':
> util.c:(.text+0x6f4): undefined reference to `sqrt'

I thought it was a ld.bfd issue but... 

Very surprisingly, i tried to reproduce on amd64 with USE_LLD=No to no
avail, and the sparc64 bulk reports no failures. I can't compare with
mips64 because no bulk is run currently.

The autoconf stuff really detects the need for`-lm' as seen in
config.log: 'ac_cv_lib_m_sqrt=yes', but is not used anywhere at all.

In fact, FreeBSD met the issue before us [0]; we need the `-lm' LDFLAG.
While here, i removed a trailing whitespace in DESCR that portcheck
found.

Then it builds fine on macppc [1], and still does on amd64. I tested
examples provided by the manpage without issues (there is no regression
tests).

Comments/feedback are welcome,

Charlène.


[0] https://svnweb.freebsd.org/ports?view=revision&revision=368948
[1] https://bin.charlenew.xyz/echoping.log


Index: Makefile
===
RCS file: /cvs/ports/net/echoping/Makefile,v
retrieving revision 1.15
diff -u -p -u -p -r1.15 Makefile
--- Makefile1 Feb 2021 19:47:08 -   1.15
+++ Makefile9 Feb 2021 23:50:36 -
@@ -5,7 +5,7 @@ COMMENT=test performance of remote TCP/
 VER=   6_0_2
 DISTNAME=  echoping-RELEASE_${VER}
 PKGNAME=   echoping-${VER:S/_/./g}
-REVISION=  0
+REVISION=  1
 CATEGORIES=net
 
 MASTER_SITES=  
https://framagit.org/bortzmeyer/echoping/-/archive/RELEASE_${VER}/
@@ -36,6 +36,10 @@ CONFIGURE_ARGS+= --enable-plugin="dns ra
--enable-smtp \
--enable-tos \
--without-libidn
+
+# Requires libm (uses sqrt(3)), detected by autoconf but not used anywhere else
+# in the build configuration
+CONFIGURE_ENV+=LDFLAGS="${LDFLAGS} -lm"
 
 WRKDIST=   ${WRKDIR}/${DISTNAME}/SRC
 
Index: pkg/DESCR
===
RCS file: /cvs/ports/net/echoping/pkg/DESCR,v
retrieving revision 1.3
diff -u -p -u -p -r1.3 DESCR
--- pkg/DESCR   1 Feb 2021 18:10:44 -   1.3
+++ pkg/DESCR   9 Feb 2021 23:50:36 -
@@ -1,3 +1,3 @@
-"echoping" is a small program to test (approximately) performance of a 
+"echoping" is a small program to test (approximately) performance of a
 remote host by sending it packets using various protocols (TCP/UDP echo,
 HTTP/HTTPS and some others).

Re: games/typespeed: fix with -fno-common

[Charlene Wendling]

On Mon, 8 Feb 2021 16:54:24 +
Stuart Henderson  wrote:

> On 2021/02/07 23:57, Charlene Wendling wrote:
> > Ping.
> > 
> > On Sat, 30 Jan 2021 17:20:44 +0100
> > Charlene Wendling  wrote:
> > 
> > > Hi,
> > > 
> > > This fix is taken from Gentoo [0]. Built and tested on amd64.
> > > 
> > > OK? 
> > > 
> > > Charlène. 
> > > 
> > > 
> > > [0]
> > > https://gitweb.gentoo.org/repo/gentoo.git/tree/games-misc/typespeed/files/typespeed-0.6.5-use-extern.patch
> > 
> > 
> > Index: Makefile
> > ===
> > RCS file: /cvs/ports/games/typespeed/Makefile,v
> > retrieving revision 1.6
> > diff -u -p -u -p -r1.6 Makefile
> > --- Makefile12 Jul 2019 20:46:26 -  1.6
> > +++ Makefile30 Jan 2021 16:16:29 -
> > @@ -3,7 +3,7 @@ COMMENT =   Test your typing speed and
> > ge
> 
> OK.
> 
> Would you mind fixing up COMMENT while there please? (new blank line
> before, s/T/t/)

I don't know how i haven't spotted it! I did so.

Re: games/typespeed: fix with -fno-common

[Charlene Wendling]

Ping.

On Sat, 30 Jan 2021 17:20:44 +0100
Charlene Wendling  wrote:

> Hi,
> 
> This fix is taken from Gentoo [0]. Built and tested on amd64.
> 
> OK? 
> 
> Charlène. 
> 
> 
> [0]
> https://gitweb.gentoo.org/repo/gentoo.git/tree/games-misc/typespeed/files/typespeed-0.6.5-use-extern.patch


Index: Makefile
===
RCS file: /cvs/ports/games/typespeed/Makefile,v
retrieving revision 1.6
diff -u -p -u -p -r1.6 Makefile
--- Makefile12 Jul 2019 20:46:26 -  1.6
+++ Makefile30 Jan 2021 16:16:29 -
@@ -3,7 +3,7 @@ COMMENT =   Test your typing speed and ge
 
 DISTNAME = typespeed-0.6.5
 CATEGORIES =   games
-REVISION = 3
+REVISION = 4
 
 HOMEPAGE = http://typespeed.sourceforge.net/
 
Index: patches/patch-src_file_c
===
RCS file: /cvs/ports/games/typespeed/patches/patch-src_file_c,v
retrieving revision 1.1.1.1
diff -u -p -u -p -r1.1.1.1 patch-src_file_c
--- patches/patch-src_file_c22 Jan 2016 15:24:30 -  1.1.1.1
+++ patches/patch-src_file_c30 Jan 2021 16:16:29 -
@@ -1,10 +1,22 @@
 $OpenBSD: patch-src_file_c,v 1.1.1.1 2016/01/22 15:24:30 landry Exp $
 
-Don't check if the userconf exists, inconditionally create
-the userdir in ~/.typespeed and the user scorefile.
+Hunk #1: Fix -fno-common build failure.
+Hunk #2: Don't check if the userconf exists, inconditionally create
+ the userdir in ~/.typespeed and the user scorefile.
 
 src/file.c.origWed Aug 13 17:08:09 2008
-+++ src/file.c Fri Jan 22 15:21:45 2016
+Index: src/file.c
+--- src/file.c.orig
 src/file.c
+@@ -142,6 +142,9 @@ extern int  hfd;
+ extern char   *rankki[11];
+ extern char   *typorank[12];
+ extern char   *usedwordfile;
++extern struct rules   rules;
++struct rawdatawords;
++struct optopt;
+ 
+ char ruledir[MAXPATHLEN];
+ char worddir[MAXPATHLEN];
 @@ -1125,7 +1125,7 @@ void
  readconfig(void)
  {
Index: patches/patch-src_menu_c
===
RCS file: patches/patch-src_menu_c
diff -N patches/patch-src_menu_c
--- /dev/null   1 Jan 1970 00:00:00 -
+++ patches/patch-src_menu_c30 Jan 2021 16:16:29 -
@@ -0,0 +1,17 @@
+$OpenBSD$
+
+Fix -fno-common build failure.
+
+Index: src/menu.c
+--- src/menu.c.orig
 src/menu.c
+@@ -76,6 +76,9 @@ void  tellstory(void);
+ 
+ extern int misses;
+ extern char   *rankki[11];
++extern struct stats   now;
++extern struct rules   rules;
++extern struct opt opt;
+ 
+ struct positions {
+   int rank;
Index: patches/patch-src_misc_c
===
RCS file: /cvs/ports/games/typespeed/patches/patch-src_misc_c,v
retrieving revision 1.1.1.1
diff -u -p -u -p -r1.1.1.1 patch-src_misc_c
--- patches/patch-src_misc_c22 Jan 2016 15:24:30 -  1.1.1.1
+++ patches/patch-src_misc_c30 Jan 2021 16:16:29 -
@@ -1,9 +1,19 @@
 $OpenBSD: patch-src_misc_c,v 1.1.1.1 2016/01/22 15:24:30 landry Exp $
 
-use arc4random()
+Hunk #1: Fix -fno-common build failure.
+Hunk #2: Use arc4random()
 
 src/misc.c.origFri Jan 22 14:58:04 2016
-+++ src/misc.c Fri Jan 22 14:58:24 2016
+Index: src/misc.c
+--- src/misc.c.orig
 src/misc.c
+@@ -75,6 +75,7 @@ char *xstrdup(char *);
+ void   xstrncpy(char *, char *, size_t);
+ 
+ extern char   *progname;
++extern struct opt opt;
+ 
+ static int graph = 0;
+ 
 @@ -224,7 +224,7 @@ r(int range)
  #ifdef WIN32
return (int)(rand() % range);
Index: patches/patch-src_network_c
===
RCS file: patches/patch-src_network_c
diff -N patches/patch-src_network_c
--- /dev/null   1 Jan 1970 00:00:00 -
+++ patches/patch-src_network_c 30 Jan 2021 16:16:29 -
@@ -0,0 +1,15 @@
+$OpenBSD$
+
+Fix -fno-common build failure.
+
+Index: src/network.c
+--- src/network.c.orig
 src/network.c
+@@ -116,6 +116,7 @@ void   setnoblock(void);
+ void  writenum(char *, uint32_t);
+ 
+ extern FILE   *netlogfile;
++extern struct opt opt;
+ 
+ static int ds = -1;
+ static int ss = -1;
Index: patches/patch-src_typespeed_c
===
RCS file: /cvs/ports/games/typespeed/patches/patch-src_typespeed_c,v
retrieving revision 1.1.1.1
diff -u -p -u -p -r1.1.1.1 patch-src_typespeed_c
--- patches/patch-src_typespeed_c   22 Jan 2016 15:24:30 -  1.1.1.1
+++ patches/patch-src_typespeed_c   30 Jan 2021 16:16:29 -
@@ -1,10 +1,30 @@
 $OpenBSD: patch-src_typespeed_c,v 1.1.1.1 2016/01/22 15:24:30 landry Exp $
 
-Disable the code opening the systemwide highscore file
-and dropping privileges, since we're removing the setgid bit
+Hunk #1: Fix -fno-common build failure
+Hunk #2: Disable the code opening the 

Re: x11/skippy: fix with -fno-common

[Charlene Wendling]

On Sun, 7 Feb 2021 16:39:49 +0100
Christian Weisgerber  wrote:

> Charlene Wendling:
> 
> > The below diff allows building skippy with -fno-common. While here:
> > 
> > - MASTER_SITES and HOMEPAGE are 404'd, drop them
> > - refresh WANTLIB
> > - don't hardcode PREFIX, don't build with -g
> > - add -fno-common fix, from Gentoo [0] (and add a missing RCS tag)
> 
> Hmm, a port without MASTER_SITES?  I guess there are precedence
> cases.

When i previously worked on some dockapps i've found out there were,
actually there are a few more: 

$ echo 'select count(*) from ports where master_sites is null ' \
   'and fullpkgpath not like "meta/%";' |
  sqlite3 /usr/local/share/sqlports

60

> ok naddy@
> 
> There is a fork on GitHub, but the last activity was in 2015.
> https://github.com/richardgv/skippy-xd
> 
> -- 
> Christian "naddy" Weisgerber
> na...@mips.inka.de

Re: powerpc bulk build report

[Charlene Wendling]

On Wed, 3 Feb 2021 23:12:13 -0700 (MST)
c...@openbsd.org wrote:

> Bulk build on macppc-0.ports.openbsd.org
> 
> Started : Mon Jan 18 10:28:09 MST 2021
> Finished: Wed Feb  3 23:11:47 MST 2021
> Duration: 16 Days 12 hours 44 minutes
> 
> Built using OpenBSD 6.8-current (GENERIC) #830: Sun Jan 17 08:10:13
> MST 2021

Some comments:

> http://build-failures.rhaalovely.net/powerpc/2021-01-18/games/valyriatear.log

It won't be built anymore on BE_ARCHS until upstream fix their code.

> http://build-failures.rhaalovely.net/powerpc/2021-01-18/x11/e17/elementary.log

It has been removed from CVS.

> http://build-failures.rhaalovely.net/powerpc/2021-01-18/net/ntopng.log
> http://build-failures.rhaalovely.net/powerpc/2021-01-18/net/pmacct,postgresql.log

It's a temporary failure due to a desync between the kernel used and the
ports tree, see https://github.com/openbsd/src/commit/ff78d6edd
They'll be back once the next bulk is finished.

x11/skippy: fix with -fno-common

[Charlene Wendling]

Hi,

The below diff allows building skippy with -fno-common. While here:

- MASTER_SITES and HOMEPAGE are 404'd, drop them
- refresh WANTLIB
- don't hardcode PREFIX, don't build with -g
- add -fno-common fix, from Gentoo [0] (and add a missing RCS tag)

This has been built and tested on amd64 and macppc.

OK? 

Charlène.


[0]
https://gitweb.gentoo.org/repo/gentoo.git/tree/x11-misc/skippy/files/skippy-0.5.0-fno-common.patch


Index: Makefile
===
RCS file: /cvs/ports/x11/skippy/Makefile,v
retrieving revision 1.21
diff -u -p -u -p -r1.21 Makefile
--- Makefile12 Jul 2019 20:51:21 -  1.21
+++ Makefile6 Feb 2021 10:46:16 -
@@ -3,24 +3,21 @@
 COMMENT=   full-screen X11 task-switcher like Apple's Expose
 
 DISTNAME=  skippy-0.5.0
-REVISION=  8
+REVISION=  9
 CATEGORIES=x11
 
-MASTER_SITES=  http://thegraveyard.org/files/
 EXTRACT_SUFX=  .tar.bz2
 
-HOMEPAGE=  http://thegraveyard.org/skippy.php
-
 # X11
-PERMIT_PACKAGE=Yes
+PERMIT_PACKAGE=Yes
 
-WANTLIB=   X11 Xau Xdmcp Xext Xft Xinerama Xrender c expat fontconfig \
-   freetype m z xcb Imlib2
+WANTLIB+= Imlib2 X11 Xext Xft Xinerama c m
 
+LIB_DEPENDS=   graphics/imlib2
 
-NO_TEST=   Yes
+FAKE_FLAGS+=   PREFIX="${PREFIX}"
 
-LIB_DEPENDS=   graphics/imlib2
+NO_TEST=   Yes
 
 post-install:
${INSTALL_DATA_DIR} ${PREFIX}/share/examples/skippy
Index: patches/patch-Makefile
===
RCS file: /cvs/ports/x11/skippy/patches/patch-Makefile,v
retrieving revision 1.2
diff -u -p -u -p -r1.2 patch-Makefile
--- patches/patch-Makefile  22 Jul 2005 19:32:17 -  1.2
+++ patches/patch-Makefile  6 Feb 2021 10:46:16 -
@@ -1,15 +1,19 @@
 $OpenBSD: patch-Makefile,v 1.2 2005/07/22 19:32:17 fgsch Exp $
 Makefile.orig  Wed May 19 18:23:50 2004
-+++ Makefile   Wed Jul 13 16:34:00 2005
-@@ -1,7 +1,7 @@
+Index: Makefile
+--- Makefile.orig
 Makefile
+@@ -1,9 +1,9 @@
  PREFIX = /usr/local
  BINDIR = ${PREFIX}/bin
  
 -X11PREFIX = /usr/X11R6
 +X11PREFIX = ${X11BASE}
  
- CFLAGS += -I${X11PREFIX}/include `imlib2-config --cflags` `pkg-config xft 
--cflags` -g -pedantic -Wall
+-CFLAGS += -I${X11PREFIX}/include `imlib2-config --cflags` `pkg-config xft 
--cflags` -g -pedantic -Wall
++CFLAGS += -I${X11PREFIX}/include `imlib2-config --cflags` `pkg-config xft 
--cflags` -pedantic -Wall
  LDFLAGS += -L${X11PREFIX}/lib -lX11 -lm `imlib2-config --libs` `pkg-config 
xft --libs`
+ 
+ # Comment these out to disable Xinerama support
 @@ -17,11 +17,15 @@ EXESUFFIX =
  
  SOURCES = skippy.c wm.c dlist.c mainwin.c clientwin.c layout.c focus.c 
config.c tooltip.c
Index: patches/patch-wm_c
===
RCS file: /cvs/ports/x11/skippy/patches/patch-wm_c,v
retrieving revision 1.3
diff -u -p -u -p -r1.3 patch-wm_c
--- patches/patch-wm_c  3 Jan 2009 23:02:57 -   1.3
+++ patches/patch-wm_c  6 Feb 2021 10:46:16 -
@@ -1,6 +1,59 @@
 wm.c.orig  Wed May 19 23:23:49 2004
-+++ wm.c   Sat Jan  3 23:56:53 2009
-@@ -258,7 +258,7 @@ wm_get_stack(Display *dpy)
+$OpenBSD$
+
+Hunk #1: fix the build with -fno-common
+
+Index: wm.c
+--- wm.c.orig
 wm.c
+@@ -43,7 +43,47 @@
+ #define WIN_STATE_FIXED_POSITION  (1<<8) /*window is fixed in position even*/
+ #define WIN_STATE_ARRANGE_IGNORE  (1<<9) /*ignore for auto arranging*/
+ 
++Atom
++  /* Generic atoms */
++  XA_WM_STATE,
++  WM_CLIENT_LEADER,
++  XA_UTF8_STRING,
+ 
++  /* Root pixmap / wallpaper atoms */
++  _XROOTPMAP_ID,
++  ESETROOT_PMAP_ID,
++
++  /* NetWM atoms */
++  _NET_SUPPORTING_WM_CHECK,
++  _NET_SUPPORTED,
++  _NET_NUMBER_OF_DESKTOPS,
++  _NET_CLIENT_LIST,
++  _NET_CLIENT_LIST_STACKING,
++  _NET_CURRENT_DESKTOP,
++  _NET_WM_DESKTOP,
++  _NET_WM_STATE,
++  _NET_WM_STATE_HIDDEN,
++  _NET_WM_STATE_SKIP_TASKBAR,
++  _NET_WM_STATE_SKIP_PAGER,
++  _NET_WM_STATE_FULLSCREEN,
++  _NET_WM_STATE_SHADED,
++  _NET_WM_STATE_ABOVE,
++  _NET_WM_STATE_STICKY,
++  _NET_WM_WINDOW_TYPE,
++  _NET_WM_WINDOW_TYPE_DESKTOP,
++  _NET_WM_WINDOW_TYPE_DOCK,
++  _NET_WM_VISIBLE_NAME,
++  _NET_WM_NAME,
++
++  /* Old gnome atoms */
++  _WIN_SUPPORTING_WM_CHECK,
++  _WIN_WORKSPACE,
++  _WIN_WORKSPACE_COUNT,
++  _WIN_PROTOCOLS,
++  _WIN_CLIENT_LIST,
++  _WIN_STATE,
++  _WIN_HINTS;
++
+ static int WM_PERSONALITY = WM_PERSONALITY_NETWM,
+NETWM_HAS_FULLSCREEN = 0,
+IGNORE_SKIP_TASKBAR = 0;
+@@ -258,7 +298,7 @@ wm_get_stack(Display *dpy)
return 0;

for(i = 0; i < items_read; i++)
@@ -9,7 +62,7 @@

XFree(data);

-@@ -293,10 +293,10 @@ wm_get_root_pmap(Display *d

[update] x11/oroborus 2.0.12 -> 2.0.20 (-fno-common fix)

[Charlene Wendling]

The below diff updates oroborus to 2.0.20. While HOMEPAGE is dead
(redirects to some VR stuff), Debian takes care of this WM, so 
it's not totally dead. There are some window focus fixes and more
themes have been added, except from that, it's mostly maintenance stuff.

Port wise: 

- Update license marker
- Take Gentoo's -fno-common fix [0]
- Fix a missing include
- Make spacing a bit more consistent. Given that at that point,
  most of the Makefile was rewritten, i ordered things according
  to Makefile.template.
- Also, while the distfile uses '_' as version separator, the
  extracted root folder has '-' instead, so i had to use WRKDIST.

It builds fine and has been lightly tested on amd64 and macppc.

Comments and feedback are welcome,

Charlène.


[0]
https://gitweb.gentoo.org/repo/gentoo.git/tree/x11-wm/oroborus/files/oroborus-2.0.20-fno-common.patch


Index: Makefile
===
RCS file: /cvs/ports/x11/oroborus/Makefile,v
retrieving revision 1.10
diff -u -p -u -p -r1.10 Makefile
--- Makefile12 Jul 2019 20:51:18 -  1.10
+++ Makefile5 Feb 2021 23:01:27 -
@@ -1,21 +1,26 @@
 # $OpenBSD: Makefile,v 1.10 2019/07/12 20:51:18 sthen Exp $
 
-COMMENT=   minimalistic X11 window manager
+COMMENT =  minimalistic X11 window manager
 
-CATEGORIES=x11
-DISTNAME=  oroborus-2.0.12
-DISTFILES= oroborus_2.0.12-1.tar.gz
-REVISION=  0
-MASTER_SITES=  http://www.mguillaud.net/ports/oroborus/ \
-   http://matt.peterson.org/FreeBSD/ports/
-HOMEPAGE=  http://www.oroborus.org/
-
-# GPL
-PERMIT_PACKAGE=Yes
-WANTLIB=   ICE SM X11 Xext Xpm c 
-
-CONFIGURE_STYLE=gnu
-USE_GMAKE= Yes
-SEPARATE_BUILD= Yes
+DISTNAME = oroborus_2.0.20
+PKGNAME =  ${DISTNAME:S/_/-/}
+
+CATEGORIES =   x11
+
+# GPLv2
+PERMIT_PACKAGE =   Yes
+
+WANTLIB += ICE SM X11 Xext Xpm c
+
+MASTER_SITES = ${MASTER_SITE_DEBIAN:=main/o/oroborus/} \
+   https://distfiles.gentoo.org/distfiles/
+
+SEPARATE_BUILD =   Yes
+
+USE_GMAKE =Yes
+
+CONFIGURE_STYLE =  gnu
+
+WRKDIST =  ${WRKDIR}/${PKGNAME}
 
 .include 
Index: distinfo
===
RCS file: /cvs/ports/x11/oroborus/distinfo,v
retrieving revision 1.4
diff -u -p -u -p -r1.4 distinfo
--- distinfo18 Jan 2015 03:15:59 -  1.4
+++ distinfo5 Feb 2021 23:01:27 -
@@ -1,2 +1,2 @@
-SHA256 (oroborus_2.0.12-1.tar.gz) = 
da6P0CzvF4e5o0iPudlw/iAyjbpb6pvLBy1ZgmRe8LE=
-SIZE (oroborus_2.0.12-1.tar.gz) = 236581
+SHA256 (oroborus_2.0.20.tar.gz) = UiClQLz2gSxnc/vlg9nqP1T0UsusVb7cqt66jDqae4k=
+SIZE (oroborus_2.0.20.tar.gz) = 335099
Index: patches/patch-src_hints_c
===
RCS file: patches/patch-src_hints_c
diff -N patches/patch-src_hints_c
--- /dev/null   1 Jan 1970 00:00:00 -
+++ patches/patch-src_hints_c   5 Feb 2021 23:01:27 -
@@ -0,0 +1,16 @@
+$OpenBSD$
+
+Fix the build with -fno-common
+
+Index: src/hints.c
+--- src/hints.c.orig
 src/hints.c
+@@ -43,7 +43,7 @@ Atom win_hints, win_state, win_client_list, win_layer,
+   win_workspace_count, win_desktop_button_proxy, win_supporting_wm_check,
+   gnome_panel_desktop_area;
+ Atom net_atoms[NET_ATOM_COUNT];
+-Display *dpy;
++extern Display *dpy;
+ 
+ void
+ initHints (Display * d)
Index: patches/patch-src_keyboard_c
===
RCS file: patches/patch-src_keyboard_c
diff -N patches/patch-src_keyboard_c
--- /dev/null   1 Jan 1970 00:00:00 -
+++ patches/patch-src_keyboard_c5 Feb 2021 23:01:27 -
@@ -0,0 +1,16 @@
+$OpenBSD$
+
+Fix the build with -fno-common
+
+Index: src/keyboard.c
+--- src/keyboard.c.orig
 src/keyboard.c
+@@ -35,6 +35,8 @@
+ #include 
+ #endif
+ 
++int NumLockMask, CapsLockMask, ScrollLockMask;
++
+ void
+ parseKeyString (Display * dpy, MyKey * key, char *str)
+ {
Index: patches/patch-src_keyboard_h
===
RCS file: patches/patch-src_keyboard_h
diff -N patches/patch-src_keyboard_h
--- /dev/null   1 Jan 1970 00:00:00 -
+++ patches/patch-src_keyboard_h5 Feb 2021 23:01:27 -
@@ -0,0 +1,16 @@
+$OpenBSD$
+
+Fix the build with -fno-common
+
+Index: src/keyboard.h
+--- src/keyboard.h.orig
 src/keyboard.h
+@@ -90,7 +90,7 @@ void grabKey (Display *, MyKey *, Window);
+ void ungrabKeys (Display *, Window);
+ void init_keyboard (Display * dpy);
+ 
+-int NumLockMask, CapsLockMask, ScrollLockMask;
++extern int NumLockMask, CapsLockMask, ScrollLockMask;
+ 
+ #endif /* __KEYBOARD_H */
+ 
Index: patches/patch-src_oroborus_c
===
RCS file: patches/patch-src_oroborus_c
diff -N patches/patch-src_oroborus_c
--- /dev/null   1 Jan 1970 00:00:00 -
+++ patches/patch-src_oroborus_c5 Feb 2021 23:01:27 -
@@ -0,0 +1

Re: [update] games/an 0.95 -> 1.2

[Charlene Wendling]

Ping; with a new diff though, to avoid patches.

On Fri, 25 Dec 2020 03:21:08 +0100
Charlene Wendling  wrote:

> Ping :) 
> 
> On Sat, 10 Oct 2020 14:14:20 +0200
> Charlene Wendling wrote:
> 
> > Hi,
> > 
> > Here is an update for an to 1.2. This is a rewrite with several
> > bugfixes, icu support and less bloat (the library has been
> > removed).
> > 
> > I don't have much to say port-wise, changes are trivial - i took the
> > opportunity to reorder the Makefile per Makefile.template since it
> > doesn't affect much the diff readability. Also, i've zapped some of
> > our hooks since upstream's Makefile has now friendlier targets.
> > 
> > This has been tested on amd64, macppc (with clang and gcc there)
> > successfully.
> > 
> > Comments/feedback are welcome,
> > 
> > Charlène.
> > 

Index: Makefile
===
RCS file: /cvs/ports/games/an/Makefile,v
retrieving revision 1.20
diff -u -p -u -p -r1.20 Makefile
--- Makefile12 Jul 2019 20:46:14 -  1.20
+++ Makefile5 Feb 2021 19:21:06 -
@@ -2,31 +2,36 @@
 
 COMMENT=   fast anagram generator
 
-V= 0.95
+V= 1.2
 DISTNAME=  an_${V}.orig
 PKGNAME=   an-${V}
 CATEGORIES=games
-MASTER_SITES=  ${MASTER_SITE_DEBIAN:=main/a/an/}
 
-# GPL - LGPL
+# GPLv2+
 PERMIT_PACKAGE=Yes
 
-WANTLIB=   c
+HOMEPAGE=  https://salsa.debian.org/pm/an
 
-WRKDIST=   ${WRKDIR}/an-${V}
+WANTLIB=   c icudata icui18n icuuc
+
+MASTER_SITES=  ${MASTER_SITE_DEBIAN:=main/a/an/}
+
+LIB_DEPENDS=   textproc/icu4c,-main
+
+MAKE_FLAGS+=   CC="${CC}" CFLAGS="${CFLAGS}"
+# Overriding DESTDIR is not enough, their Makefile subdirectories
+# don't respect our files hierarchy
+FAKE_FLAGS+=   MANDIR="${FROOT}/man/man6" INSTALLDIR="${FROOT}/bin"
 
 NO_TEST=   Yes
 
+USE_GMAKE= Yes
+
+WRKDIST=   ${WRKDIR}/an-${V}
+FROOT= ${WRKINST}${PREFIX}
+
 pre-configure:
@perl -pi -e 's,/usr/dict/words,/usr/share/dict/words,g' \
${WRKSRC}/an.6
-
-do-build:
-   cd ${WRKBUILD} && \
-   ${CC} ${CFLAGS} an.c gan_qsort.c -o an ${LDFLAGS}
-
-do-install:
-   ${INSTALL_PROGRAM} ${WRKBUILD}/an ${PREFIX}/bin
-   ${INSTALL_MAN} ${WRKBUILD}/an.6 ${PREFIX}/man/man6
 
 .include 
Index: distinfo
===
RCS file: /cvs/ports/games/an/distinfo,v
retrieving revision 1.5
diff -u -p -u -p -r1.5 distinfo
--- distinfo18 Jan 2015 03:13:55 -  1.5
+++ distinfo5 Feb 2021 19:21:06 -
@@ -1,2 +1,2 @@
-SHA256 (an_0.95.orig.tar.gz) = ESZt0eNHWU2Y+N/R4UHYtPvgZ3UlQtDpPEIsR4n5KVw=
-SIZE (an_0.95.orig.tar.gz) = 47829
+SHA256 (an_1.2.orig.tar.gz) = uBqNt0zf3znmyn3XUs9HIkNdL7f7OhuXHU9PrlFC2ys=
+SIZE (an_1.2.orig.tar.gz) = 14231
Index: pkg/PLIST
===
RCS file: /cvs/ports/games/an/pkg/PLIST,v
retrieving revision 1.3
diff -u -p -u -p -r1.3 PLIST
--- pkg/PLIST   15 Sep 2004 00:46:08 -  1.3
+++ pkg/PLIST   5 Feb 2021 19:21:06 -
@@ -1,3 +1,3 @@
 @comment $OpenBSD: PLIST,v 1.3 2004/09/15 00:46:08 espie Exp $
-bin/an
+@bin bin/an
 @man man/man6/an.6

Re: Fix games/jumpnbump -fno-common

[Charlene Wendling]

Hi,

On Sat, 30 Jan 2021 00:02:22 +
Brian Callahan  wrote:

> Hello ports --
> 
> Attached is a fix for games/jumpnbump for -fno-common.
> Taken from upstream (no new release since this commit):
> https://gitlab.com/LibreGames/jumpnbump/-/commit/b72b70a4233776bdaa6a683c89af2becefd53bd6.diff
> 
> Left off the Makefile portion, since it is a no-op for us.
> 
> OK?

OK cwen@

> ~Brian

Re: [update, fix] games/xmj 1.0 -> 1.16 (-fno-common)

[Charlene Wendling]

On Sat, 30 Jan 2021 20:40:58 +0100
Charlene Wendling  wrote:

> Hi,
> 
> On top of failing with `-fno-common', xmj does not work at all, here
> on my amd64 machine:
> 
> $ doas pkg_add xmj
> $ xmj
> Segmentation Fault
> 
> So, here is an update for the 20 years old xmj 1.0, to 2020's 1.16.
> There are major upstream changes, notably they dropped Imake (yay!),
> and xmj is now gtk+2 based. This fixes the -fno-common failure ootb.
> 
> Port changes:
> 
> - Update license marker, refresh WANTLIB (ports-lib-depends-check is
> happy)
> - Use up to date HOMEPAGE and MASTER_SITES (found out thanks to
> Macports and repology)
> - Requires gmake.
> - Install the manpages
> 
> 
> Testing:
> 
> - It builds fine on macppc (vanilla clang) and amd64 (modified with
>   naddy's patch)
> - The runtime is great on these 2 platforms. I tried networked games
>   without issues.
> 
> Comments/feedback are welcome,
> 
> Charlène.

^ edited; i was looking at my backlog and took into account what sthen
told me about MAKE_FLAGS, making the diff way smaller :)

I accidentally built xmj with gcc and met the infamous "C99 styled
declarations in for loops" error, so it will require gnu99 on base-gcc
archs. This still builds and works fine with clang on macppc and amd64,
with that flag, so i have not brought bsd.port.arch.mk.

Index: Makefile
===
RCS file: /cvs/ports/games/xmj/Makefile,v
retrieving revision 1.25
diff -u -p -u -p -r1.25 Makefile
--- Makefile12 Jul 2019 20:46:27 -  1.25
+++ Makefile5 Feb 2021 01:50:14 -
@@ -2,22 +2,37 @@
 
 COMMENT=   Mahjongg game for X11
 
-DISTNAME=   xmj
-PKGNAME=   xmj-1.0
-REVISION = 3
+DISTNAME=  mj-1.16-src
+PKGNAME=   x${DISTNAME:S/-src$//}
+
 CATEGORIES=games
 
-# BSD
+HOMEPAGE=  https://mahjong.julianbradfield.org/
+
+# GPLv2
 PERMIT_PACKAGE=Yes
 
-WANTLIB=   X11 Xau Xdmcp Xext c xcb
+WANTLIB += X11 Xcomposite Xcursor Xdamage Xext Xfixes Xi Xinerama
+WANTLIB += Xrandr Xrender atk-1.0 c cairo fontconfig freetype
+WANTLIB += gdk-x11-2.0 gdk_pixbuf-2.0 gio-2.0 glib-2.0 gobject-2.0
+WANTLIB += gtk-x11-2.0 harfbuzz intl m pango-1.0 pangocairo-1.0
+WANTLIB += pangoft2-1.0 z
+
+MASTER_SITES=  ${HOMEPAGE}/Source/
+
+LIB_DEPENDS=   x11/gtk+2
+
+MAKE_FLAGS+=   CC="${CC}" EXTRA_CFLAGS="${CFLAGS}" CDEBUGFLAGS=
+# The final '/' matters
+FAKE_FLAGS+=   DESTDIR="${WRKINST}${PREFIX}/"
+
+USE_GMAKE= Yes
 
-CONFIGURE_STYLE=imake noman
 NO_TEST=   Yes
 
-pre-configure:
-   @perl -pi -e 's,_height=,_height ,g;' \
-   -e 's,_width=,_width ,g' $i \
-   ${WRKSRC}/tiles/c*
+# base-gcc fix for "c99 initial declarations in for loops" error
+CFLAGS+=   -std=gnu99
+
+INSTALL_TARGET=install install.man
 
 .include 
Index: distinfo
===
RCS file: /cvs/ports/games/xmj/distinfo,v
retrieving revision 1.4
diff -u -p -u -p -r1.4 distinfo
--- distinfo18 Jan 2015 03:14:04 -  1.4
+++ distinfo5 Feb 2021 01:50:14 -
@@ -1,2 +1,2 @@
-SHA256 (xmj.tar.gz) = EGcUMVIKBYSp3BJHqnYPM988WHP7G7IVgN//NK7xhyg=
-SIZE (xmj.tar.gz) = 52255
+SHA256 (mj-1.16-src.tar.gz) = wl6YSPSfcGNeXjugy1l5nlvBSalfeGYhuZaWtmeHop0=
+SIZE (mj-1.16-src.tar.gz) = 403981
Index: patches/patch-Imakefile
===
RCS file: patches/patch-Imakefile
diff -N patches/patch-Imakefile
--- patches/patch-Imakefile 27 Oct 2007 13:32:05 -  1.1
+++ /dev/null   1 Jan 1970 00:00:00 -
@@ -1,9 +0,0 @@
-$OpenBSD: patch-Imakefile,v 1.1 2007/10/27 13:32:05 ajacoutot Exp $
 Imakefile.orig Fri May  5 16:46:04 1989
-+++ Imakefile  Sat Oct 27 15:15:14 2007
-@@ -4,4 +4,4 @@ LOCAL_LIBRARIES = $(XLIB)
-OBJS = bicons.o cicons.o cicons1.o cicons2.o cicons3.o cicons4.o \
- cicons5.o event.o icons.o mj.o nicons.o
- 
--ComplexProgramTarget(xmj)
-+ComplexProgramTargetNoMan(xmj)
Index: patches/patch-event_c
===
RCS file: patches/patch-event_c
diff -N patches/patch-event_c
--- patches/patch-event_c   19 Apr 2017 22:02:45 -  1.1
+++ /dev/null   1 Jan 1970 00:00:00 -
@@ -1,11 +0,0 @@
-$OpenBSD: patch-event_c,v 1.1 2017/04/19 22:02:45 espie Exp $
 event.c.orig   Thu Apr 20 00:00:59 2017
-+++ event.cThu Apr 20 00:01:22 2017
-@@ -599,6 +599,7 @@ void board_num_proc()
- place_tiles(FALSE);
- }
- 
-+void
- msg_event_proc(event)
- XEvent *event;
- {
Index: pkg/PLIST
===
RCS file: /cvs/ports/games/xmj/pkg/PLIST,v
retrieving revision 1.2
diff -u -p -u -p -r1.2 PLIST
--- pkg/PLIST   7 Apr 2002 01:54:4

[update] sysutils/wmcube 0.98 -> 1.0.2 (+ -fno-common fix)

[Charlene Wendling]

This is an update for wmcube, to 1.0.2. This adds color schemes, smooth
fonts, more logos, a manual page and various fixes.

Port-wise:

- Move to dockapps.net, HOMEPAGE is not reachable
- Fix an hardcoded path to logo files (introduced in 1.0.0,
  so our current package is not affected)
- Remove the Makefile patch, and override things with MAKE_FLAGS. That
  also allows us to simplify the port, since we won't have to fix MSDOS
  line endings, and that's why i've attached the diff
- Due to the previous change, don't use upstream's install target,
  avoiding further patching
- Drop the wmgeneral/list.h patch. Upstream "fixed" that in 1.0.2
- Fix the build with -fno-common

Testing:

It has been built and tested on macppc and amd64 without issues.

OK? 

Charlène.


wmcube.diff
Description: Binary data

Re: audio/wmmp3: -fno-common fix (and more)

[Charlene Wendling]

On Thu, 4 Feb 2021 23:05:05 +0100
Charlene Wendling  wrote:

> On Thu, 4 Feb 2021 22:07:08 +0100
> Charlene Wendling  wrote:
> 
> > 
> > The below diff:
> > 
> > - moves HOMEPAGE and MASTER_SITES to dockapps.net, since it was not
> >   reachable. The distfile checksum is the same.
> > - removes an hardcoded path to madplay; we could move to mpg123 but
> > i did not want to include a breaking change
> > - fixes missing includes spotted by clang
> > - fixes the build with -fno-common (from Gentoo [0])
> > 
> > It builds and works fine on macppc and amd64.
> > 
> > OK?
> > 
> > Charlène.
> > 
> > 
> > [0]
> > https://gitweb.gentoo.org/repo/gentoo.git/tree/x11-plugins/wmmp3/files/wmmp3-0.12-fno-common.patch
> 
> SUBST_CMD should be run in pre-configure, not in post-patch.

Typoed pre-configure, sorry for the noise.


Index: Makefile
===
RCS file: /cvs/ports/audio/wmmp3/Makefile,v
retrieving revision 1.24
diff -u -p -u -p -r1.24 Makefile
--- Makefile12 Jul 2019 20:43:42 -  1.24
+++ Makefile4 Feb 2021 22:22:47 -
@@ -3,17 +3,17 @@
 COMMENT=   wm-dockapp; front end for an mp3 player
 
 DISTNAME=  wmmp3-0.12
-REVISION=  4
+REVISION=  5
 CATEGORIES=audio x11 x11/windowmaker
 
-HOMEPAGE=  http://www.dotfiles.com/software/wmmp3/
+HOMEPAGE=  https://www.dockapps.net/wmmp3
 
 # GPL
 PERMIT_PACKAGE=Yes
 
 WANTLIB=   X11 Xext Xpm c
 
-MASTER_SITES=  ${HOMEPAGE}
+MASTER_SITES=  https://www.dockapps.net/download/
 
 RUN_DEPENDS=   audio/madplay
 
@@ -23,6 +23,9 @@ CONFIGURE_STYLE=  gnu
 
 SAMPLE_DIR=${PREFIX}/share/examples/wmmp3
 SUBST_VARS+=   SAMPLE_DIR
+
+pre-configure:
+   @${SUBST_CMD} ${WRKSRC}/main.c
 
 do-install:
${INSTALL_PROGRAM} ${WRKBUILD}/wmmp3 ${PREFIX}/bin
Index: patches/patch-main_c
===
RCS file: /cvs/ports/audio/wmmp3/patches/patch-main_c,v
retrieving revision 1.2
diff -u -p -u -p -r1.2 patch-main_c
--- patches/patch-main_c4 Aug 2003 13:31:28 -   1.2
+++ patches/patch-main_c4 Feb 2021 22:22:47 -
@@ -1,16 +1,17 @@
 $OpenBSD: patch-main_c,v 1.2 2003/08/04 13:31:28 wilfried Exp $
 main.c.orig2000-04-03 02:11:46.0 +0200
-+++ main.c 2003-08-04 15:05:48.0 +0200
+Index: main.c
+--- main.c.orig
 main.c
 @@ -125,7 +125,7 @@ void loadconfig()
  errno = 0;
  
  /* set defualts in case anything fails */
 -set_mpg123("/usr/local/bin/mpg123");
-+set_mpg123("/usr/local/bin/madplay");
++set_mpg123("${LOCALBASE}/bin/madplay");
  set_mp3ext(".mp3");
  set_playlistext(".m3u");
  
-@@ -230,7 +230,7 @@ int check_options(int argc, char *argv[]
+@@ -230,7 +230,7 @@ int check_options(int argc, char *argv[])
  return option_entered;
  }
  
Index: patches/patch-main_h
===
RCS file: patches/patch-main_h
diff -N patches/patch-main_h
--- /dev/null   1 Jan 1970 00:00:00 -
+++ patches/patch-main_h4 Feb 2021 22:22:47 -
@@ -0,0 +1,15 @@
+$OpenBSD$
+
+Add missing include
+
+Index: main.h
+--- main.h.orig
 main.h
+@@ -14,6 +14,7 @@
+ 
+ #define __MAIN_H__
+ 
++#include 
+ #include 
+ #include 
+ #include 
Index: patches/patch-mpg123ctl_h
===
RCS file: patches/patch-mpg123ctl_h
diff -N patches/patch-mpg123ctl_h
--- /dev/null   1 Jan 1970 00:00:00 -
+++ patches/patch-mpg123ctl_h   4 Feb 2021 22:22:47 -
@@ -0,0 +1,15 @@
+$OpenBSD$
+
+Add missing include
+
+Index: mpg123ctl.h
+--- mpg123ctl.h.orig
 mpg123ctl.h
+@@ -14,6 +14,7 @@
+ 
+ #define __MPG123CTL__
+ 
++#include 
+ #include 
+ #include 
+ #include 
Index: patches/patch-wmgeneral_c
===
RCS file: patches/patch-wmgeneral_c
diff -N patches/patch-wmgeneral_c
--- /dev/null   1 Jan 1970 00:00:00 -
+++ patches/patch-wmgeneral_c   4 Feb 2021 22:22:47 -
@@ -0,0 +1,19 @@
+$OpenBSD$
+
+Fix the build with -fno-common
+
+Index: wmgeneral.c
+--- wmgeneral.c.orig
 wmgeneral.c
+@@ -58,6 +58,11 @@
+  /* X11 Variables */
+ /*/
+ 
++Display *display;
++Window Root, iconwin, win;
++XpmIcon wmgen;
++XpmIcon wmfont;
++
+ int screen;
+ int x_fd;
+ int d_depth;
Index: patches/patch-wmgeneral_h
===
RCS file: patches/patch-wmgeneral_h
diff -N patches/patch-wmgeneral_h
--- /dev/null   1 Jan 1970 00:00:00 -
+++ patches/patch-wmgeneral_h   4 Feb 2021 22:22:47 -
@@ -0,0 +1,22 @@
+$OpenBSD$
+
+Fix the build with -f

Re: audio/wmmp3: -fno-common fix (and more)

[Charlene Wendling]

On Thu, 4 Feb 2021 22:07:08 +0100
Charlene Wendling  wrote:

> 
> The below diff:
> 
> - moves HOMEPAGE and MASTER_SITES to dockapps.net, since it was not
>   reachable. The distfile checksum is the same.
> - removes an hardcoded path to madplay; we could move to mpg123 but i
>   did not want to include a breaking change
> - fixes missing includes spotted by clang
> - fixes the build with -fno-common (from Gentoo [0])
> 
> It builds and works fine on macppc and amd64.
> 
> OK?
> 
> Charlène.
> 
> 
> [0]
> https://gitweb.gentoo.org/repo/gentoo.git/tree/x11-plugins/wmmp3/files/wmmp3-0.12-fno-common.patch

SUBST_CMD should be run in pre-configure, not in post-patch

Index: Makefile
===
RCS file: /cvs/ports/audio/wmmp3/Makefile,v
retrieving revision 1.24
diff -u -p -u -p -r1.24 Makefile
--- Makefile12 Jul 2019 20:43:42 -  1.24
+++ Makefile4 Feb 2021 22:02:43 -
@@ -3,17 +3,17 @@
 COMMENT=   wm-dockapp; front end for an mp3 player
 
 DISTNAME=  wmmp3-0.12
-REVISION=  4
+REVISION=  5
 CATEGORIES=audio x11 x11/windowmaker
 
-HOMEPAGE=  http://www.dotfiles.com/software/wmmp3/
+HOMEPAGE=  https://www.dockapps.net/wmmp3
 
 # GPL
 PERMIT_PACKAGE=Yes
 
 WANTLIB=   X11 Xext Xpm c
 
-MASTER_SITES=  ${HOMEPAGE}
+MASTER_SITES=  https://www.dockapps.net/download/
 
 RUN_DEPENDS=   audio/madplay
 
@@ -23,6 +23,9 @@ CONFIGURE_STYLE=  gnu
 
 SAMPLE_DIR=${PREFIX}/share/examples/wmmp3
 SUBST_VARS+=   SAMPLE_DIR
+
+pre_configure:
+   @${SUBST_CMD} ${WRKSRC}/main.c
 
 do-install:
${INSTALL_PROGRAM} ${WRKBUILD}/wmmp3 ${PREFIX}/bin
Index: patches/patch-main_c
===
RCS file: /cvs/ports/audio/wmmp3/patches/patch-main_c,v
retrieving revision 1.2
diff -u -p -u -p -r1.2 patch-main_c
--- patches/patch-main_c4 Aug 2003 13:31:28 -   1.2
+++ patches/patch-main_c4 Feb 2021 22:02:43 -
@@ -1,16 +1,17 @@
 $OpenBSD: patch-main_c,v 1.2 2003/08/04 13:31:28 wilfried Exp $
 main.c.orig2000-04-03 02:11:46.0 +0200
-+++ main.c 2003-08-04 15:05:48.0 +0200
+Index: main.c
+--- main.c.orig
 main.c
 @@ -125,7 +125,7 @@ void loadconfig()
  errno = 0;
  
  /* set defualts in case anything fails */
 -set_mpg123("/usr/local/bin/mpg123");
-+set_mpg123("/usr/local/bin/madplay");
++set_mpg123("${LOCALBASE}/bin/madplay");
  set_mp3ext(".mp3");
  set_playlistext(".m3u");
  
-@@ -230,7 +230,7 @@ int check_options(int argc, char *argv[]
+@@ -230,7 +230,7 @@ int check_options(int argc, char *argv[])
  return option_entered;
  }
  
Index: patches/patch-main_h
===
RCS file: patches/patch-main_h
diff -N patches/patch-main_h
--- /dev/null   1 Jan 1970 00:00:00 -
+++ patches/patch-main_h4 Feb 2021 22:02:43 -
@@ -0,0 +1,15 @@
+$OpenBSD$
+
+Add missing include
+
+Index: main.h
+--- main.h.orig
 main.h
+@@ -14,6 +14,7 @@
+ 
+ #define __MAIN_H__
+ 
++#include 
+ #include 
+ #include 
+ #include 
Index: patches/patch-mpg123ctl_h
===
RCS file: patches/patch-mpg123ctl_h
diff -N patches/patch-mpg123ctl_h
--- /dev/null   1 Jan 1970 00:00:00 -
+++ patches/patch-mpg123ctl_h   4 Feb 2021 22:02:43 -
@@ -0,0 +1,15 @@
+$OpenBSD$
+
+Add missing include
+
+Index: mpg123ctl.h
+--- mpg123ctl.h.orig
 mpg123ctl.h
+@@ -14,6 +14,7 @@
+ 
+ #define __MPG123CTL__
+ 
++#include 
+ #include 
+ #include 
+ #include 
Index: patches/patch-wmgeneral_c
===
RCS file: patches/patch-wmgeneral_c
diff -N patches/patch-wmgeneral_c
--- /dev/null   1 Jan 1970 00:00:00 -
+++ patches/patch-wmgeneral_c   4 Feb 2021 22:02:43 -
@@ -0,0 +1,19 @@
+$OpenBSD$
+
+Fix the build with -fno-common
+
+Index: wmgeneral.c
+--- wmgeneral.c.orig
 wmgeneral.c
+@@ -58,6 +58,11 @@
+  /* X11 Variables */
+ /*/
+ 
++Display *display;
++Window Root, iconwin, win;
++XpmIcon wmgen;
++XpmIcon wmfont;
++
+ int screen;
+ int x_fd;
+ int d_depth;
Index: patches/patch-wmgeneral_h
===
RCS file: patches/patch-wmgeneral_h
diff -N patches/patch-wmgeneral_h
--- /dev/null   1 Jan 1970 00:00:00 -
+++ patches/patch-wmgeneral_h   4 Feb 2021 22:02:43 -
@@ -0,0 +1,22 @@
+$OpenBSD$
+
+Fix the build with -fno-common
+
+Index: wmgeneral.h
+--- wmgeneral.h.orig
 wmgeneral.h
+@@ -41,10 +41,10 @@ typedef struct {
+  /* Global variable */
+ /***/
+ 
+-Display *display;
+-Window Root, iconwin, win;
+-XpmIcon wmgen;
+-XpmI

audio/wmmp3: -fno-common fix (and more)

[Charlene Wendling]

The below diff:

- moves HOMEPAGE and MASTER_SITES to dockapps.net, since it was not
  reachable. The distfile checksum is the same.
- removes an hardcoded path to madplay; we could move to mpg123 but i
  did not want to include a breaking change
- fixes missing includes spotted by clang
- fixes the build with -fno-common (from Gentoo [0])

It builds and works fine on macppc and amd64.

OK?

Charlène.


[0]
https://gitweb.gentoo.org/repo/gentoo.git/tree/x11-plugins/wmmp3/files/wmmp3-0.12-fno-common.patch


Index: Makefile
===
RCS file: /cvs/ports/audio/wmmp3/Makefile,v
retrieving revision 1.24
diff -u -p -u -p -r1.24 Makefile
--- Makefile12 Jul 2019 20:43:42 -  1.24
+++ Makefile4 Feb 2021 21:02:24 -
@@ -3,17 +3,17 @@
 COMMENT=   wm-dockapp; front end for an mp3 player
 
 DISTNAME=  wmmp3-0.12
-REVISION=  4
+REVISION=  5
 CATEGORIES=audio x11 x11/windowmaker
 
-HOMEPAGE=  http://www.dotfiles.com/software/wmmp3/
+HOMEPAGE=  https://www.dockapps.net/wmmp3
 
 # GPL
 PERMIT_PACKAGE=Yes
 
 WANTLIB=   X11 Xext Xpm c
 
-MASTER_SITES=  ${HOMEPAGE}
+MASTER_SITES=  https://www.dockapps.net/download/
 
 RUN_DEPENDS=   audio/madplay
 
@@ -23,6 +23,9 @@ CONFIGURE_STYLE=  gnu
 
 SAMPLE_DIR=${PREFIX}/share/examples/wmmp3
 SUBST_VARS+=   SAMPLE_DIR
+
+post-patch:
+   @${SUBST_CMD} ${WRKSRC}/main.c
 
 do-install:
${INSTALL_PROGRAM} ${WRKBUILD}/wmmp3 ${PREFIX}/bin
Index: patches/patch-main_c
===
RCS file: /cvs/ports/audio/wmmp3/patches/patch-main_c,v
retrieving revision 1.2
diff -u -p -u -p -r1.2 patch-main_c
--- patches/patch-main_c4 Aug 2003 13:31:28 -   1.2
+++ patches/patch-main_c4 Feb 2021 21:02:24 -
@@ -1,16 +1,17 @@
 $OpenBSD: patch-main_c,v 1.2 2003/08/04 13:31:28 wilfried Exp $
 main.c.orig2000-04-03 02:11:46.0 +0200
-+++ main.c 2003-08-04 15:05:48.0 +0200
+Index: main.c
+--- main.c.orig
 main.c
 @@ -125,7 +125,7 @@ void loadconfig()
  errno = 0;
  
  /* set defualts in case anything fails */
 -set_mpg123("/usr/local/bin/mpg123");
-+set_mpg123("/usr/local/bin/madplay");
++set_mpg123("${LOCALBASE}/bin/madplay");
  set_mp3ext(".mp3");
  set_playlistext(".m3u");
  
-@@ -230,7 +230,7 @@ int check_options(int argc, char *argv[]
+@@ -230,7 +230,7 @@ int check_options(int argc, char *argv[])
  return option_entered;
  }
  
Index: patches/patch-main_h
===
RCS file: patches/patch-main_h
diff -N patches/patch-main_h
--- /dev/null   1 Jan 1970 00:00:00 -
+++ patches/patch-main_h4 Feb 2021 21:02:24 -
@@ -0,0 +1,15 @@
+$OpenBSD$
+
+Add missing include
+
+Index: main.h
+--- main.h.orig
 main.h
+@@ -14,6 +14,7 @@
+ 
+ #define __MAIN_H__
+ 
++#include 
+ #include 
+ #include 
+ #include 
Index: patches/patch-mpg123ctl_h
===
RCS file: patches/patch-mpg123ctl_h
diff -N patches/patch-mpg123ctl_h
--- /dev/null   1 Jan 1970 00:00:00 -
+++ patches/patch-mpg123ctl_h   4 Feb 2021 21:02:24 -
@@ -0,0 +1,15 @@
+$OpenBSD$
+
+Add missing include
+
+Index: mpg123ctl.h
+--- mpg123ctl.h.orig
 mpg123ctl.h
+@@ -14,6 +14,7 @@
+ 
+ #define __MPG123CTL__
+ 
++#include 
+ #include 
+ #include 
+ #include 
Index: patches/patch-wmgeneral_c
===
RCS file: patches/patch-wmgeneral_c
diff -N patches/patch-wmgeneral_c
--- /dev/null   1 Jan 1970 00:00:00 -
+++ patches/patch-wmgeneral_c   4 Feb 2021 21:02:24 -
@@ -0,0 +1,19 @@
+$OpenBSD$
+
+Fix the build with -fno-common
+
+Index: wmgeneral.c
+--- wmgeneral.c.orig
 wmgeneral.c
+@@ -58,6 +58,11 @@
+  /* X11 Variables */
+ /*/
+ 
++Display *display;
++Window Root, iconwin, win;
++XpmIcon wmgen;
++XpmIcon wmfont;
++
+ int screen;
+ int x_fd;
+ int d_depth;
Index: patches/patch-wmgeneral_h
===
RCS file: patches/patch-wmgeneral_h
diff -N patches/patch-wmgeneral_h
--- /dev/null   1 Jan 1970 00:00:00 -
+++ patches/patch-wmgeneral_h   4 Feb 2021 21:02:24 -
@@ -0,0 +1,22 @@
+$OpenBSD$
+
+Fix the build with -fno-common
+
+Index: wmgeneral.h
+--- wmgeneral.h.orig
 wmgeneral.h
+@@ -41,10 +41,10 @@ typedef struct {
+  /* Global variable */
+ /***/
+ 
+-Display *display;
+-Window Root, iconwin, win;
+-XpmIcon wmgen;
+-XpmIcon wmfont;
++extern Display *display;
++extern Window Root, iconwin, win;
++extern XpmIcon wmgen;
++extern XpmIcon wmfont;
+ 
+ /***/
+  /* Function Prototypes */

[update] x11/wmtime 1.0 -> 1.4 (-fno-common fix)

[Charlene Wendling]

Here is an update for wmtime to 0.4. The main change is that it uses
libdockapp now.

Port-wise:

- drop wmgeneral stuff (it's in libdockapp)
- remove the upstream Makefile patch, explaining that ours expanded
- once again, i had to add libconv flags

Testing:

This builds and works fine on macppc and amd64.

OK? 

Charlène.


Index: Makefile
===
RCS file: /cvs/ports/x11/wmtime/Makefile,v
retrieving revision 1.19
diff -u -p -u -p -r1.19 Makefile
--- Makefile12 Jul 2019 21:02:30 -  1.19
+++ Makefile4 Feb 2021 19:38:09 -
@@ -2,23 +2,32 @@
 
 COMMENT=   time/date applet for WindowMaker
 
-DISTNAME=  wmtime-1.0b2
-REVISION=  2
+DISTNAME=  wmtime-1.4
 CATEGORIES=x11 x11/windowmaker
 
+HOMEPAGE=  https://www.dockapps.net/wmtime
+
 # GPL
 PERMIT_PACKAGE=Yes
 
-WANTLIB=   X11 Xext Xpm c m
+WANTLIB += X11 Xext Xpm c dockapp iconv m
 
-MASTER_SITES=  http://mirror2.unixfreunde.de/ \
-   http://web.cs.mun.ca/~gstarkes/wmaker/dockapps/files/
+MASTER_SITES=  https://www.dockapps.net/download/
 
-MAKE_FLAGS=CC="${CC}"
+LIB_DEPENDS=   converters/libiconv \
+   devel/libdockapp>=0.7.3
 
-WRKDIST=${WRKDIR}/wmtime.app
-WRKSRC=${WRKDIST}/wmtime
+MAKE_FLAGS+=   CC="${CC}" CFLAGS="${CFLAGS}" LDFLAGS="${LDFLAGS}" \
+   MANDIR="${PREFIX}/man/man1"
 
 NO_TEST=   Yes
+
+CFLAGS+=   -I${X11BASE}/include -I${LOCALBASE}/include
+
+LDFLAGS+=  -L${X11BASE}/lib -L${LOCALBASE}/lib -liconv
+
+WRKDIST=   ${WRKDIR}/dockapps-7d1ef92
+
+ALL_TARGET+=   wmtime
 
 .include 
Index: distinfo
===
RCS file: /cvs/ports/x11/wmtime/distinfo,v
retrieving revision 1.4
diff -u -p -u -p -r1.4 distinfo
--- distinfo18 Jan 2015 03:16:04 -  1.4
+++ distinfo4 Feb 2021 19:38:09 -
@@ -1,2 +1,2 @@
-SHA256 (wmtime-1.0b2.tar.gz) = +6R8IM7ej5d73TzB5Rs+1su1Tx1oAi2Q+cSnDBVynZY=
-SIZE (wmtime-1.0b2.tar.gz) = 20680
+SHA256 (wmtime-1.4.tar.gz) = lY/qJBAAlJ755JK950FQlVLdw6mbSyhd2e7XkDVTQ5Y=
+SIZE (wmtime-1.4.tar.gz) = 19045
Index: patches/patch-wmgeneral_list_h
===
RCS file: patches/patch-wmgeneral_list_h
diff -N patches/patch-wmgeneral_list_h
--- patches/patch-wmgeneral_list_h  5 May 2017 13:48:23 -   1.1
+++ /dev/null   1 Jan 1970 00:00:00 -
@@ -1,16 +0,0 @@
-$OpenBSD: patch-wmgeneral_list_h,v 1.1 2017/05/05 13:48:23 jca Exp $
-
-Index: wmgeneral/list.h
 wmgeneral/list.h.orig
-+++ wmgeneral/list.h
-@@ -29,9 +29,7 @@ Boston, MA 02111-1307, USA.  */
- #ifndef __LIST_H_
- #define __LIST_H_
- 
--#if defined(__GNUC__) && !defined(__STRICT_ANSI__)
--# define INLINE inline
--#else
-+#if 1
- # define INLINE
- #endif
- 
Index: patches/patch-wmtime_Makefile
===
RCS file: patches/patch-wmtime_Makefile
diff -N patches/patch-wmtime_Makefile
--- patches/patch-wmtime_Makefile   26 Dec 2000 18:22:14 -  1.1.1.1
+++ /dev/null   1 Jan 1970 00:00:00 -
@@ -1,30 +0,0 @@
-$OpenBSD: patch-wmtime_Makefile,v 1.1.1.1 2000/12/26 18:22:14 naddy Exp $
 wmtime/Makefile.orig   Tue May 19 23:04:53 1998
-+++ wmtime/MakefileTue Dec 26 19:02:32 2000
-@@ -1,18 +1,23 @@
--LIBDIR = -L/usr/X11R6/lib
-+LIBDIR = -L${X11BASE}/lib
- LIBS   = -lXpm -lXext -lX11 -lm
- OBJS =wmtime.o \
-   ../wmgeneral/wmgeneral.o \
-   ../wmgeneral/misc.o \
-   ../wmgeneral/list.o
- 
-+all: wmtime
-+
- .c.o:
--  cc -c -O2 -Wall $< -o $*.o
-+  $(CC) $(CFLAGS) -I$(X11BASE)/include -c $< -o $@
- 
- wmtime: $(OBJS)
--  cc -o wmtime $^ $(LIBDIR) $(LIBS)
-+  $(CC) -o wmtime $(OBJS) $(LIBDIR) $(LIBS)
- 
- clean::
-   for i in $(OBJS) ; do \
-   rm $$i; \
-   done
-   rm wmtime
-+
-+install:
-+  ${BSD_INSTALL_PROGRAM} wmtime ${PREFIX}/bin
Index: patches/patch-wmtime_wmtime_c
===
RCS file: /cvs/ports/x11/wmtime/patches/patch-wmtime_wmtime_c,v
retrieving revision 1.2
diff -u -p -u -p -r1.2 patch-wmtime_wmtime_c
--- patches/patch-wmtime_wmtime_c   5 May 2017 13:47:35 -   1.2
+++ patches/patch-wmtime_wmtime_c   4 Feb 2021 19:38:09 -
@@ -1,26 +1,15 @@
 $OpenBSD: patch-wmtime_wmtime_c,v 1.2 2017/05/05 13:47:35 jca Exp $
-Index: wmtime/wmtime.c
 wmtime/wmtime.c.orig
-+++ wmtime/wmtime.c
-@@ -98,7 +98,7 @@ void printversion(void);
- void wmtime_routine(int, char **);
- void get_lang();
+Index: wmtime.c
+--- wmtime.c.orig
 wmtime.c
+@@ -282,8 +282,8 @@ void wmtime_routine(int argc, char **argv) {
  
--void main(int argc, char *argv[]) {
-+int main(int argc, char *argv[]) {
- 
-   int i;
-   
-@@ -187,9 +187,9 @@ void wmtime_routine(int argc, char **argv) {
struct tm   *time_struct;

games/wmtictactoe: -fno-common fix

[Charlene Wendling]

Hi,

FreeBSD has been creative while fixing it [0]. This fixes the build
with -fno-common. Runtime has been tested successfully on amd64 and
macppc.

I've found the hard way that it requires permissions fixing while
issueing update-patches, and, while here, let's move upstream to
dockapps.net and refresh WANTLIB. The distfile checksum did not change.

OK?

Charlène.


[0]
https://svnweb.freebsd.org/ports/head/games/wmtictactoe/Makefile?view=markup&pathrev=549081#l22


Index: Makefile
===
RCS file: /cvs/ports/games/wmtictactoe/Makefile,v
retrieving revision 1.19
diff -u -p -u -p -r1.19 Makefile
--- Makefile12 Jul 2019 20:46:26 -  1.19
+++ Makefile4 Feb 2021 16:59:00 -
@@ -4,12 +4,18 @@ COMMENT=  wm-dockapp; TicTacToe game
 
 DISTNAME=  wmtictactoe-1.1-1
 PKGNAME=   wmtictactoe-1.1.1
-REVISION = 2
+REVISION=  3
+FIX_EXTRACT_PERMISSIONS=   Yes
+
 CATEGORIES=games x11 x11/windowmaker
 
+HOMEPAGE=  https://www.dockapps.net/wmtictactoe
+
 # GPL
 PERMIT_PACKAGE=Yes
-WANTLIB=   X11 Xext Xpm c xcb
+WANTLIB=   X11 Xext Xpm c
+
+MASTER_SITES=  https://www.dockapps.net/download/
 
 NO_TEST=   Yes
 
Index: patches/patch-wmgeneral_wmgeneral_c
===
RCS file: patches/patch-wmgeneral_wmgeneral_c
diff -N patches/patch-wmgeneral_wmgeneral_c
--- /dev/null   1 Jan 1970 00:00:00 -
+++ patches/patch-wmgeneral_wmgeneral_c 4 Feb 2021 16:59:00 -
@@ -0,0 +1,16 @@
+$OpenBSD$
+
+Fix the build with -fno-common
+
+Index: wmgeneral/wmgeneral.c
+--- wmgeneral/wmgeneral.c.orig
 wmgeneral/wmgeneral.c
+@@ -65,6 +65,8 @@ char *Geometry = "";
+ GCNormalGC;
+ XpmIcon   wmgen;
+ Pixmappixmask;
++Display *display;
++Window  Root, iconwin, win;
+ 
+   /*/
+  /* Mouse Regions */
Index: patches/patch-wmgeneral_wmgeneral_h
===
RCS file: patches/patch-wmgeneral_wmgeneral_h
diff -N patches/patch-wmgeneral_wmgeneral_h
--- /dev/null   1 Jan 1970 00:00:00 -
+++ patches/patch-wmgeneral_wmgeneral_h 4 Feb 2021 16:59:00 -
@@ -0,0 +1,18 @@
+$OpenBSD$
+
+Fix the build with -fno-common
+
+Index: wmgeneral/wmgeneral.h
+--- wmgeneral/wmgeneral.h.orig
 wmgeneral/wmgeneral.h
+@@ -36,8 +36,8 @@ typedef struct {
+  /* Global variable */
+ /***/
+ 
+-Display   *display;
+-Window  Root, iconwin, win;
++extern Display*display;
++extern Window  Root, iconwin, win;
+ 
+ 
+   /***/
Index: pkg/PLIST
===
RCS file: /cvs/ports/games/wmtictactoe/pkg/PLIST,v
retrieving revision 1.1.1.1
diff -u -p -u -p -r1.1.1.1 PLIST
--- pkg/PLIST   9 Feb 2001 13:02:38 -   1.1.1.1
+++ pkg/PLIST   4 Feb 2021 16:59:00 -
@@ -1,2 +1,2 @@
 @comment $OpenBSD: PLIST,v 1.1.1.1 2001/02/09 13:02:38 wilfried Exp $
-bin/wmtictactoe
+@bin bin/wmtictactoe

[update] x11/wmtz 0.7 -> 0.7.20150816 (-fno-common fix)

[Charlene Wendling]

Here is an update for wmtz. dockapps.net provides an old version, i
preferred to use what they have in their git head, which is already
provided by Gentoo, because this new version uses libdockapp, this
avoids duplicating patches in wmgeneral.

Port-wise:

- FIX_EXTRACT_PERMISSIONS is not needed anymore
- their install target is still not good, so keep our own

Testing:

It builds and runs fine on amd64 and macppc with the provided sample
configuration file.

Comments/feedback are welcome,

Charlène.


Index: Makefile
===
RCS file: /cvs/ports/x11/wmtz/Makefile,v
retrieving revision 1.24
diff -u -p -u -p -r1.24 Makefile
--- Makefile12 Jul 2019 20:51:24 -  1.24
+++ Makefile1 Feb 2021 13:24:26 -
@@ -2,22 +2,28 @@
 
 COMMENT=   wm-dockapp; displays the time in different time zones
 
-DISTNAME=  wmtz-0.7
-REVISION=  3
+DISTNAME=  wmtz-0.7_p20150816
+PKGNAME=   ${DISTNAME:S/_p/./}
+
 CATEGORIES=x11 x11/windowmaker
-FIX_EXTRACT_PERMISSIONS=Yes
 
 HOMEPAGE=  http://www.dockapps.net/wmtz
 
 # GPL
 PERMIT_PACKAGE=Yes
-WANTLIB=   X11 Xext Xpm c m
+WANTLIB=   X11 Xext Xpm c dockapp m
+
+# dockapps.net only provides an old tarball without libdockapp support.
+MASTER_SITES=  http://distfiles.gentoo.org/distfiles/
 
-MASTER_SITES=  http://www.dockapps.net/download/
+LIB_DEPENDS=   devel/libdockapp>=0.7.3
 
-MAKE_FLAGS=CC="${CC}"
+MAKE_ENV+= CC="${CC}" LDFLAGS="${LDFLAGS}" CFLAGS="${CFLAGS}"
 
 NO_TEST=   Yes
+
+CFLAGS+=   -I${X11BASE}/include -I${LOCALBASE}/include
+LDFLAGS+=  -L${X11BASE}/lib -L${LOCALBASE}/lib
 
 WRKSRC=${WRKDIST}/wmtz
 
Index: distinfo
===
RCS file: /cvs/ports/x11/wmtz/distinfo,v
retrieving revision 1.4
diff -u -p -u -p -r1.4 distinfo
--- distinfo18 Jan 2015 03:16:04 -  1.4
+++ distinfo1 Feb 2021 13:24:26 -
@@ -1,2 +1,2 @@
-SHA256 (wmtz-0.7.tar.gz) = ywnmSpgjs9px5ULXzGbmy9ATWAm+4eVl0YrNzwVlgjo=
-SIZE (wmtz-0.7.tar.gz) = 26350
+SHA256 (wmtz-0.7_p20150816.tar.gz) = 
Dvi7e3qRLm8vLOiOblv4CAzNlUv2JYikNTKzQzK6Lls=
+SIZE (wmtz-0.7_p20150816.tar.gz) = 19310
Index: patches/patch-wmgeneral_list_h
===
RCS file: patches/patch-wmgeneral_list_h
diff -N patches/patch-wmgeneral_list_h
--- patches/patch-wmgeneral_list_h  3 May 2017 19:29:45 -   1.1
+++ /dev/null   1 Jan 1970 00:00:00 -
@@ -1,14 +0,0 @@
-$OpenBSD: patch-wmgeneral_list_h,v 1.1 2017/05/03 19:29:45 jca Exp $
 wmgeneral/list.h.orig
-+++ wmgeneral/list.h
-@@ -29,9 +29,7 @@ Boston, MA 02111-1307, USA.  */
- #ifndef __LIST_H_
- #define __LIST_H_
- 
--#if defined(__GNUC__) && !defined(__STRICT_ANSI__)
--# define INLINE inline
--#else
-+#if 1
- # define INLINE
- #endif
- 
Index: patches/patch-wmtz_Makefile
===
RCS file: /cvs/ports/x11/wmtz/patches/patch-wmtz_Makefile,v
retrieving revision 1.2
diff -u -p -u -p -r1.2 patch-wmtz_Makefile
--- patches/patch-wmtz_Makefile 15 Jun 2001 09:59:34 -  1.2
+++ patches/patch-wmtz_Makefile 1 Feb 2021 13:24:26 -
@@ -1,26 +1,23 @@
 $OpenBSD: patch-wmtz_Makefile,v 1.2 2001/06/15 09:59:34 wilfried Exp $
 wmtz/Makefile.orig Sun Mar 11 14:57:59 2001
-+++ wmtz/Makefile  Fri Jun 15 11:49:09 2001
-@@ -1,8 +1,5 @@
+Index: wmtz/Makefile
+--- wmtz/Makefile.orig
 wmtz/Makefile
+@@ -1,15 +1,11 @@
 -CC = gcc
 -LIBDIR = -L/usr/X11R6/lib
-+LIBDIR = -L${X11BASE}/lib
- LIBS   = -lXpm -lXext -lX11 -lm
+ LIBS   = -lXpm -lXext -lX11 -lm -ldockapp
 -FLAGS = -O2
 -DESTDIR = /usr/local
- OBJS =wmtz.o \
-   ../wmgeneral/wmgeneral.o \
-   ../wmgeneral/misc.o \
-@@ -10,10 +7,10 @@ OBJS =  wmtz.o \
- 
+-OBJS =wmtz.o
++OBJS= wmtz.o
  
  .c.o:
 -  $(CC) -I/usr/X11R6/include $(FLAGS) -c -Wall $< -o $*.o
-+  $(CC) -I${X11BASE}/include ${CFLAGS} -c $< -o $@
++  $(CC) $(CFLAGS) -c $< -o $@
  
  wmtz: $(OBJS)
 -  $(CC) $(FLAGS) -o wmtz $^ -lXext $(LIBDIR) $(LIBS)
-+  $(CC) ${CFLAGS} -o $@ ${OBJS} $(LIBDIR) $(LIBS)
++  $(CC) -o $@ $(OBJS) $(LDFLAGS) $(LIBS)
  
  all:: wmtz
  
Index: patches/patch-wmtz_wmtz_c
===
RCS file: /cvs/ports/x11/wmtz/patches/patch-wmtz_wmtz_c,v
retrieving revision 1.3
diff -u -p -u -p -r1.3 patch-wmtz_wmtz_c
--- patches/patch-wmtz_wmtz_c   22 May 2010 15:15:52 -  1.3
+++ patches/patch-wmtz_wmtz_c   1 Feb 2021 13:24:26 -
@@ -1,38 +1,22 @@
 $OpenBSD: patch-wmtz_wmtz_c,v 1.3 2010/05/22 15:15:52 espie Exp $
 wmtz/wmtz.c.orig   Mon Apr 16 21:26:34 2001
-+++ wmtz/wmtz.cSat May 22 17:15:07 2010
-@@ -51,8 +51,7 @@
+Index: wmtz/wmtz.c

[update] astro/wmmoonclock 1.27 -> 1.30 (-fno-common fix)

[Charlene Wendling]

The below diff updates wmmoonclock to 1.30. This fixes the `-fno-common'
issue out of the box. Note that it uses a configure script instead of a
Makefile.

Port-wise:

- drop the whole Makefile patch, flags are good out of the box, and
  the new build system uses pkg-config, so everything is in place
- provide a .desktop file
- keep the legacy binary name, upstream provides an all lower cased
  binary, but a camel cased manpage
- move HOMEPAGE and MASTER_SITES to dockapps.net

Testing:

It has been build and run tested on amd64 and macppc successfully.

Comments/feedback are welcome,

Charlène.


Index: Makefile
===
RCS file: /cvs/ports/astro/wmmoonclock/Makefile,v
retrieving revision 1.34
diff -u -p -u -p -r1.34 Makefile
--- Makefile26 Jan 2020 11:14:29 -  1.34
+++ Makefile1 Feb 2021 14:47:55 -
@@ -2,22 +2,26 @@
 
 COMMENT=   wm-dockapp; shows the moon phase
 
-DISTNAME=  wmMoonClock-1.27
-PKGNAME=   ${DISTNAME:L}
-REVISION=  5
+DISTNAME=  wmmoonclock-1.30
 CATEGORIES=astro x11 x11/windowmaker
 
+HOMEPAGE=  https://www.dockapps.net/wmmoonclock
+
 # GPL
 PERMIT_PACKAGE=Yes
 
 WANTLIB=   X11 Xext Xpm c m
 
-MASTER_SITES=  https://distfiles.sigtrap.nl/
+MASTER_SITES=  https://www.dockapps.net/download/
+
+RUN_DEPENDS=   devel/desktop-file-utils
 
-MAKE_ENV=  CC="${CC}"
+CONFIGURE_STYLE=   gnu
 
 NO_TEST=   Yes
 
-WRKSRC=${WRKDIST}/Src
+# Keep the legacy binary name (and match with the manpage name)
+post-install:
+   cd ${PREFIX}/bin && mv wmmoonclock wmMoonClock
 
 .include 
Index: distinfo
===
RCS file: /cvs/ports/astro/wmmoonclock/distinfo,v
retrieving revision 1.4
diff -u -p -u -p -r1.4 distinfo
--- distinfo4 Apr 2013 16:06:27 -   1.4
+++ distinfo1 Feb 2021 14:47:55 -
@@ -1,2 +1,2 @@
-SHA256 (wmMoonClock-1.27.tar.gz) = Q50zCjDuHLCE4FEuiU9eGi6YaXRaInT72aVoMsFSfHc=
-SIZE (wmMoonClock-1.27.tar.gz) = 155197
+SHA256 (wmmoonclock-1.30.tar.gz) = i6fc6xDOP3eEgDUwOd5HQEU3rzUJLJQa5O4Ifp49Q6A=
+SIZE (wmmoonclock-1.30.tar.gz) = 238171
Index: patches/patch-Makefile
===
RCS file: patches/patch-Makefile
diff -N patches/patch-Makefile
--- patches/patch-Makefile  18 Dec 2007 10:39:47 -  1.2
+++ /dev/null   1 Jan 1970 00:00:00 -
@@ -1,41 +0,0 @@
-$OpenBSD: patch-Makefile,v 1.2 2007/12/18 10:39:47 ajacoutot Exp $
 Src/Makefile.orig  Tue Mar 23 03:20:32 1999
-+++ Src/Makefile   Tue Dec 18 10:16:26 2007
-@@ -1,8 +1,5 @@
--CC = gcc
--CFLAGS = -O2
--INCDIR = -I/usr/X11R6/include/X11
--DESTDIR= /usr/X11R6
--LIBDIR = -L/usr/X11R6/lib
-+CFLAGS += -I${X11BASE}/include
-+LIBDIR = -L${X11BASE}/lib
- 
- # for Linux
- LIBS   = -lXpm -lX11 -lXext
-@@ -13,16 +10,11 @@ LIBS   = -lXpm -lX11 -lXext
- OBJS   = wmMoonClock.o CalcEphem.o Moon.o MoonRise.o \
-  xutils.o
- 
--
--.c.o:
--  $(CC) $(CFLAGS) -D$(shell echo `uname -s`) -c $< -o $*.o $(INCDIR)
--
--
- all:  wmMoonClock.o wmMoonClock
- 
- wmMoonClock.o: wmMoonClock_master.xpm wmMoonClock_mask.xbm CalcEphem.h
- wmMoonClock:  $(OBJS) 
--  $(CC) $(COPTS) $(SYSTEM) -lm -o wmMoonClock $^ $(INCDIR) $(LIBDIR) 
$(LIBS)
-+  $(CC) $(COPTS) $(SYSTEM) -o wmMoonClock $(OBJS) $(INCDIR) $(LIBDIR) 
$(LIBS) -lm
- 
- clean:
-   for i in $(OBJS) ; do \
-@@ -31,6 +23,6 @@ clean:
-   rm -f wmMoonClock
- 
- install:: wmMoonClock
--  install -s -m 0755 wmMoonClock $(DESTDIR)/bin
--  install-m 0644 wmMoonClock.1 $(DESTDIR)/man/man1 
-+  $(BSD_INSTALL_PROGRAM) wmMoonClock $(PREFIX)/bin
-+  $(BSD_INSTALL_MAN) wmMoonClock.1 $(PREFIX)/man/man1 
- 
Index: patches/patch-wmMoonClock_c
===
RCS file: /cvs/ports/astro/wmmoonclock/patches/patch-wmMoonClock_c,v
retrieving revision 1.2
diff -u -p -u -p -r1.2 patch-wmMoonClock_c
--- patches/patch-wmMoonClock_c 18 Dec 2007 10:39:47 -  1.2
+++ patches/patch-wmMoonClock_c 1 Feb 2021 14:47:55 -
@@ -1,16 +1,17 @@
 $OpenBSD: patch-wmMoonClock_c,v 1.2 2007/12/18 10:39:47 ajacoutot Exp $
 Src/wmMoonClock.c.orig Mon Jun  7 16:27:18 1999
-+++ Src/wmMoonClock.c  Tue Dec 18 10:16:26 2007
-@@ -162,7 +162,7 @@ int main(int argc, char *argv[]) {
- int   i, n, k, j, ImageNumber;
- int   Year, Month, DayOfWeek, DayOfMonth;
- int   Hours, Mins, Secs, OldSecs, digit, xoff, xsize;
+Index: src/wmMoonClock.c
+--- src/wmMoonClock.c.orig
 src/wmMoonClock.c
+@@ -157,7 +157,7 @@ int main(int argc, char *argv[]) {
+ struct tm *GMTTime, *LocalTime;
+ XEventevent;
+ int   i, n, j, ImageNumber, Year, Month, D

x11/wmglobe: -fno-common fix

[Charlene Wendling]

There are 20+ duplicate symbols, and no patches around. I'm proposing
to use `-fcommon'. While here move dead HOMEPAGE and MASTER_SITES to
dockapps.net.

It has been tested successfully on macppc and amd64.

OK?

Charlène.


Index: Makefile
===
RCS file: /cvs/ports/astro/wmglobe/Makefile,v
retrieving revision 1.32
diff -u -p -u -p -r1.32 Makefile
--- Makefile12 Jul 2019 20:43:31 -  1.32
+++ Makefile1 Feb 2021 14:10:29 -
@@ -3,17 +3,17 @@
 COMMENT=   wm-dockapp; displays the earth on an icon
 
 DISTNAME=  wmglobe-1.3
-REVISION=  6
+REVISION=  7
 CATEGORIES=astro x11 x11/windowmaker
 
-HOMEPAGE=  http://hamete.org/static/wmg/
+HOMEPAGE=  https://www.dockapps.net/wmglobe
 
 # GPL
 PERMIT_PACKAGE=Yes
 
 WANTLIB=   X11 Xext Xpm c m z jpeg tiff png gif wraster
 
-MASTER_SITES=  ${HOMEPAGE}
+MASTER_SITES=  https://www.dockapps.net/download/
 
 LIB_DEPENDS=   graphics/jpeg \
graphics/tiff \
@@ -23,5 +23,8 @@ LIB_DEPENDS=  graphics/jpeg \
 
 MAKE_ENV=  CC="${CC}"
 NO_TEST=   Yes
+
+# 20+ duplicate symbols
+CFLAGS+=   -fcommon
 
 .include 
 

x11/wmweather: -fno-common fix

[Charlene Wendling]

I took this fix from Debian [0]. It builds and works fine on amd64 and
macppc.

OK?

Charlène.


[0]
https://sources.debian.org/patches/wmweather/2.4.7-3/957956_gcc10.diff/


Index: Makefile
===
RCS file: /cvs/ports/x11/wmweather/Makefile,v
retrieving revision 1.30
diff -u -p -u -p -r1.30 Makefile
--- Makefile12 Jul 2019 20:51:24 -  1.30
+++ Makefile1 Feb 2021 13:55:54 -
@@ -3,6 +3,7 @@
 COMMENT=   wm-dockapp; weather monitor
 
 DISTNAME=  wmweather-2.4.7
+REVISION=  0
 CATEGORIES=x11 x11/windowmaker
 
 HOMEPAGE=  http://people.debian.org/~godisch/wmweather/
Index: patches/patch-src_wmgeneral_c
===
RCS file: patches/patch-src_wmgeneral_c
diff -N patches/patch-src_wmgeneral_c
--- /dev/null   1 Jan 1970 00:00:00 -
+++ patches/patch-src_wmgeneral_c   1 Feb 2021 13:55:54 -
@@ -0,0 +1,15 @@
+$OpenBSD$
+
+Fix the build with -fno-common
+
+Index: src/wmgeneral.c
+--- src/wmgeneral.c.orig
 src/wmgeneral.c
+@@ -85,6 +85,7 @@ Window   iconwin, win;
+ GCNormalGC;
+ XpmIcon   wmgen;
+ Pixmappixmask;
++Display   *display;
+ 
+   /*/
+  /* Mouse Regions */
Index: patches/patch-src_wmgeneral_h
===
RCS file: patches/patch-src_wmgeneral_h
diff -N patches/patch-src_wmgeneral_h
--- /dev/null   1 Jan 1970 00:00:00 -
+++ patches/patch-src_wmgeneral_h   1 Feb 2021 13:55:54 -
@@ -0,0 +1,16 @@
+$OpenBSD$
+
+Fix the build with -fno-common
+
+Index: src/wmgeneral.h
+--- src/wmgeneral.h.orig
 src/wmgeneral.h
+@@ -36,7 +36,7 @@ typedef struct {
+  /* Global variable */
+ /***/
+ 
+-Display   *display;
++extern Display*display;
+ 
+   /***/
+  /* Function Prototypes */

[update] x11/wmitime 0.3 -> 0.5 (-fno-common fix)

[Charlene Wendling]

The below diff updates wmitime to 0.5. This dockapp now uses
libdockapp, so there is no further patch for -fno-common needed.

Port-wise:

- provide .desktop file, manpage and HOMEPAGE
- drop wmgeneral patch since it's dealt with by libdockapp
- add build flags for iconv, it has been forgotten by upstream

Testing: 

This builds and works fine on macppc and amd64.

Comments/feedback are welcome,

Charlène.


Index: Makefile
===
RCS file: /cvs/ports/x11/wmitime/Makefile,v
retrieving revision 1.19
diff -u -p -u -p -r1.19 Makefile
--- Makefile12 Jul 2019 20:51:24 -  1.19
+++ Makefile1 Feb 2021 13:36:07 -
@@ -2,25 +2,33 @@
 
 COMMENT=   wm-dockapp; shows the date, time, and Internet time
 
-DISTNAME=  wmitime-0.3
-REVISION=  2
+DISTNAME=  wmitime-0.5
 CATEGORIES=x11 x11/windowmaker
 
+HOMEPAGE=  https://www.dockapps.net/wmitime
+
 # GPL
 PERMIT_PACKAGE=Yes
-WANTLIB=   X11 Xext Xpm c m
 
-MASTER_SITES=  http://dockapps.org/files/22/29/ \
-   http://web.cs.mun.ca/~gstarkes/wmaker/dockapps/files/
+WANTLIB += X11 Xext Xpm c dockapp iconv m
+
+MASTER_SITES=  https://www.dockapps.net/download/
 
-MAKE_FLAGS=CC="${CC}"
+RUN_DEPENDS=   devel/desktop-file-utils
+LIB_DEPENDS=   converters/libiconv \
+   devel/libdockapp>=0.7.3
+
+MAKE_FLAGS=CC="${CC}" PREFIX="${PREFIX}" \
+   CFLAGS="${CFLAGS}" LDFLAGS="${LDFLAGS}"
 
 NO_TEST=   Yes
 
-WRKDIST=   ${WRKDIR}/wmitime
-WRKSRC=${WRKDIST}/wmitime
+CFLAGS+=   -I${LOCALBASE}/include -I${X11BASE}/include
+LDFLAGS+=  -L${LOCALBASE}/lib -L${X11BASE}/lib
+
+WRKDIST=   ${WRKDIR}/dockapps-26db447
 
-do-install:
-   ${INSTALL_PROGRAM} ${WRKBUILD}/wmitime ${PREFIX}/bin
+#do-install:
+#  ${INSTALL_PROGRAM} ${WRKBUILD}/wmitime ${PREFIX}/bin
 
 .include 
Index: distinfo
===
RCS file: /cvs/ports/x11/wmitime/distinfo,v
retrieving revision 1.4
diff -u -p -u -p -r1.4 distinfo
--- distinfo18 Jan 2015 03:16:04 -  1.4
+++ distinfo1 Feb 2021 13:36:07 -
@@ -1,2 +1,2 @@
-SHA256 (wmitime-0.3.tar.gz) = /OxEG6wYOsL7u/nVd8Lqr5ueApKmBBJW5aQoanbN3GE=
-SIZE (wmitime-0.3.tar.gz) = 21442
+SHA256 (wmitime-0.5.tar.gz) = oYPiCNi6v5fopA4356WoseE3HKVH14lvMU+AhRb0Xm0=
+SIZE (wmitime-0.5.tar.gz) = 15099
Index: patches/patch-Makefile
===
RCS file: /cvs/ports/x11/wmitime/patches/patch-Makefile,v
retrieving revision 1.2
diff -u -p -u -p -r1.2 patch-Makefile
--- patches/patch-Makefile  9 Mar 2001 14:29:58 -   1.2
+++ patches/patch-Makefile  1 Feb 2021 13:36:07 -
@@ -1,26 +1,27 @@
-$OpenBSD: patch-Makefile,v 1.2 2001/03/09 14:29:58 wilfried Exp $
 wmitime/Makefile.orig  Mon Mar 29 05:51:39 1999
-+++ wmitime/Makefile   Thu Mar  8 13:17:14 2001
-@@ -1,7 +1,6 @@
- #LANG = fr
--LIBDIR = -L/usr/X11R6/lib
-+LIBDIR = -L${X11BASE}/lib
- LIBS   = -lXpm -lXext -lX11 -lm
--FLAGS = -O2
- OBJS =wmitime.o \
-   ../wmgeneral/wmgeneral.o \
-   ../wmgeneral/misc.o \
-@@ -9,12 +8,12 @@ OBJS =   wmitime.o \
- 
+$OpenBSD$
+
+Index: Makefile
+--- Makefile.orig
 Makefile
+@@ -1,19 +1,18 @@
+-LIBS   = -lXpm -lXext -lX11 -lm -ldockapp
+-CFLAGS = -O2 -Wall
++LIBS   = -lXpm -lXext -lX11 -lm -ldockapp -liconv
+ OBJS = wmitime.o
+ INSTALL = install
+-PREFIX = /usr/local
++PREFIX ?= /usr/local
+ BINDIR = $(PREFIX)/bin
+-MANDIR = $(PREFIX)/share/man/man1
++MANDIR = $(PREFIX)/man/man1
+ DESKTOPDIR = $(PREFIX)/share/applications
  
  .c.o:
--  cc -I/usr/X11R6/share/include $(FLAGS) -D$(LANG) -c -Wall $< -o $*.o
-+  ${CC} -I${X11BASE}/include ${CFLAGS} -c $< -o $@
+   $(CC) $(CPPFLAGS) $(CFLAGS) -c $< -o $*.o
  
--wmtime: $(OBJS)
--  cc $(FLAGS) -D$(LANG) -o wmitime $^ -lXext $(LIBDIR) $(LIBS)
-+wmitime: $(OBJS)
-+  ${CC} ${CFLAGS} -o $@ $(OBJS) $(LIBDIR) $(LIBS)
+ wmitime: $(OBJS)
+-  $(CC) $(LDFLAGS) -o wmitime $^ $(LIBS)
++  $(CC) $(LDFLAGS) -o $@ $(OBJS) $(LIBDIR) $(LIBS)
  
 -all:: wmtime
 +all:: wmitime
Index: patches/patch-wmgeneral_list_h
===
RCS file: patches/patch-wmgeneral_list_h
diff -N patches/patch-wmgeneral_list_h
--- patches/patch-wmgeneral_list_h  3 May 2017 19:18:27 -   1.1
+++ /dev/null   1 Jan 1970 00:00:00 -
@@ -1,14 +0,0 @@
-$OpenBSD: patch-wmgeneral_list_h,v 1.1 2017/05/03 19:18:27 jca Exp $
 wmgeneral/list.h.orig
-+++ wmgeneral/list.h
-@@ -29,9 +29,7 @@ Boston, MA 02111-1307, USA.  */
- #ifndef __LIST_H_
- #define __LIST_H_
- 
--#if defined(__GNUC__) && !defined(__STRICT_ANSI__)
--# define INLINE inline
--#else
-+#if 1
- # define INLINE

Re: UPDATE: devel/boehm-gc 8.0.4

[Charlene Wendling]

On Tue, 2 Feb 2021 18:40:50 -0500 (EST)
k...@intricatesoftware.com wrote:

> Update to 8.0.4 with the following additional changes:
> - Use mmap via the configure argument instead of hard-coding it.
> - Define USE_MMAP_ANON for all OpenBSD arch to match upstream
>   code conventions.
> - Remove unused ELF_CLASS define for OpenBSD.
> - Remove old HAVE_DL_ITERATE_PHDR workaround for OpenBSD.
> - Don't use old GC_find_limit_openbsd() which is only needed for
>   uthreads.
> 
> I've tested this on i386, amd64, sparc64 and aarch64 using the
> port's regress tests and by using w3m. Note that on aarch64 the
> current boehm-gc port isn't stable and w3m will crash it pretty
> easily. This update has not fixed that (aarch64 is still unstable
> but the regress tests pass).
> 
> Additional testing on mips, powerpc, etc. would be appreciated.

I tested on macppc:

- all 17 tests are passing
- w3m's runtime is still fine

Charlène.

Re: [Update] textproc/p5-Lingua-Stem : Update to 2.31

[Charlene Wendling]

On Tue, 2 Feb 2021 09:26:53 +
wen heping  wrote:

> Hi, ports@:
> 
> Here is a patch for textproc/p5-Lingua-Stem:
> i) Update to 2.31
> ii) Add devel/p5-Test-Distribution to TEST_DEPENDS and
> add TEST_POD_COVERAGE=1 to MAKE_ENV
>
>It build well and pass all tests on amd64-6.8 system.
>Only one port depends on it: textproc/p5-Lingua-EN-Tagger, it
> build well and pass all tests with this patch.
> 
> 
> wen

OK cwen@, but with some changes:

- drop BUILD_DEPENDS = RUN_DEPENDS
- add TEST_POD=1, otherwise it won't pull the proper test depends for
  pod testing
- add version check for some depends, as stated in the meta files
- spacing and casing tweaks

Index: Makefile
===
RCS file: /cvs/ports/textproc/p5-Lingua-Stem/Makefile,v
retrieving revision 1.14
diff -u -p -u -p -r1.14 Makefile
--- Makefile3 Jul 2020 21:45:47 -   1.14
+++ Makefile2 Feb 2021 13:05:16 -
@@ -4,25 +4,26 @@ COMMENT=  stemming of words
 
 MODULES=   cpan
 PKG_ARCH=  *
-DISTNAME=  Lingua-Stem-0.84
+DISTNAME=  Lingua-Stem-2.31
 CATEGORIES=textproc
-REVISION=  0
 
-# perl
+# Perl
 PERMIT_PACKAGE=Yes
 
-CONFIGURE_STYLE=modbuild
+CONFIGURE_STYLE=   modbuild
 
-BUILD_DEPENDS= ${RUN_DEPENDS}
 RUN_DEPENDS=   textproc/p5-Lingua-PT-Stemmer \
-   textproc/p5-Lingua-Stem-Fr \
+   textproc/p5-Lingua-Stem-Fr>=0.02 \
textproc/p5-Lingua-Stem-It \
textproc/p5-Lingua-Stem-Ru \
-   textproc/p5-Lingua-Stem-Snowball-Da \
+   textproc/p5-Lingua-Stem-Snowball-Da>=1.01 \
textproc/p5-Lingua-Stem-Snowball-No \
textproc/p5-Lingua-Stem-Snowball-Se \
textproc/p5-Text-German
+TEST_DEPENDS=  devel/p5-Test-Distribution
 
 MODCPAN_EXAMPLES=  Yes
+
+MAKE_ENV +=TEST_POD_COVERAGE=1 TEST_POD=1
 
 .include 
Index: distinfo
===
RCS file: /cvs/ports/textproc/p5-Lingua-Stem/distinfo,v
retrieving revision 1.5
diff -u -p -u -p -r1.5 distinfo
--- distinfo18 Jan 2015 03:15:24 -  1.5
+++ distinfo2 Feb 2021 13:05:16 -
@@ -1,2 +1,2 @@
-SHA256 (Lingua-Stem-0.84.tar.gz) = j/VLbhOuloAEXThjanvz241D1lslZ6hrpfQwa6uAW4k=
-SIZE (Lingua-Stem-0.84.tar.gz) = 1358061
+SHA256 (Lingua-Stem-2.31.tar.gz) = qhqZMrZCflmCU+YajM0NBMxVn66dWNh3TCAncItjAmQ=
+SIZE (Lingua-Stem-2.31.tar.gz) = 1354267

Re: [Update] www/p5-Dancer2 : Update to 0.300005

[Charlene Wendling]

On Tue, 2 Feb 2021 09:12:42 +
wen heping  wrote:

> Hi, ports@:
> 
> Here is a patch for www/p5-Dancer2:
> i) Update to 0.35
> ii) Add p5-Test-CPAN-Meta as TEST_DEPENDS, also it is required
> by an author test.
> It build well and pass all tests on amd64-6.8 system.
> 
> 4 ports depends on it, all build well and pass all tests with
> this patch.
> 
> wen

OK cwen@

Re: graphics/tumble -fno-common

[Charlene Wendling]

On Mon, 1 Feb 2021 17:28:43 +1100
Jonathan Gray  wrote:

> no newer release or upstream fix

I've successfully build and run tested this diff on amd64.

OK cwen@

> Index: Makefile
> ===
> RCS file: /cvs/ports/graphics/tumble/Makefile,v
> retrieving revision 1.1.1.1
> diff -u -p -r1.1.1.1 Makefile
> --- Makefile  25 Apr 2020 14:02:14 -  1.1.1.1
> +++ Makefile  1 Feb 2021 06:03:42 -
> @@ -8,6 +8,7 @@ COMMENT = convert pictures into pdf book
>  GH_PROJECT = tumble
>  GH_ACCOUNT = brouhaha
>  GH_TAGNAME = v0.36
> +REVISION =   0
>  
>  CATEGORIES = graphics textproc
>  
> Index: patches/patch-semantics_c
> ===
> RCS file: patches/patch-semantics_c
> diff -N patches/patch-semantics_c
> --- /dev/null 1 Jan 1970 00:00:00 -
> +++ patches/patch-semantics_c 1 Feb 2021 06:03:42 -
> @@ -0,0 +1,16 @@
> +$OpenBSD$
> +
> +fix -fno-common build
> +
> +Index: semantics.c
> +--- semantics.c.orig
>  semantics.c
> +@@ -119,7 +119,7 @@ typedef struct output_page_t
> + #endif
> + 
> + 
> +-FILE *yyin;
> ++extern FILE *yyin;
> + int line;  /* line number in spec file */
> + 
> + int bookmark_level;
> Index: patches/patch-tumble_input_h
> ===
> RCS file: patches/patch-tumble_input_h
> diff -N patches/patch-tumble_input_h
> --- /dev/null 1 Jan 1970 00:00:00 -
> +++ patches/patch-tumble_input_h  1 Feb 2021 06:03:42 -
> @@ -0,0 +1,13 @@
> +$OpenBSD$
> +
> +fix -fno-common build
> +
> +Index: tumble_input.h
> +--- tumble_input.h.orig
>  tumble_input.h
> +@@ -71,4 +71,4 @@ void init_jpeg_handler (void);
> + void init_pbm_handler  (void);
> + void init_png_handler  (void);
> + 
> +-input_handler_t blank_handler;
> ++extern input_handler_t blank_handler;
> 

x11/wmdate: fix with the new devel/libdockapp (3/4)

[Charlene Wendling]

This has been tested successfully on macppc and amd64. If you don't use
windowmaker, you'll need to use the '-w' option.

A vintage patch got refreshed while here.

Comments/feedback are welcome,

Charlène.


Index: Makefile
===
RCS file: /cvs/ports/x11/wmdate/Makefile,v
retrieving revision 1.23
diff -u -p -u -p -r1.23 Makefile
--- Makefile12 Jul 2019 20:51:24 -  1.23
+++ Makefile1 Feb 2021 01:27:06 -
@@ -3,7 +3,7 @@
 COMMENT=   wm-dockapp; shows the current date
 
 DISTNAME=  wmdate-0.7
-REVISION=  3
+REVISION=  4
 CATEGORIES=x11 x11/windowmaker
 
 # BSD
@@ -11,7 +11,7 @@ PERMIT_PACKAGE=   Yes
 
 WANTLIB=   X11 Xau Xdmcp Xext Xpm c dockapp
 
-LIB_DEPENDS=   devel/libdockapp
+LIB_DEPENDS=   devel/libdockapp>=0.7.3
 
 NO_TEST=   Yes
 CONFIGURE_STYLE=   imake noman
Index: patches/patch-Imakefile
===
RCS file: /cvs/ports/x11/wmdate/patches/patch-Imakefile,v
retrieving revision 1.1.1.1
diff -u -p -u -p -r1.1.1.1 patch-Imakefile
--- patches/patch-Imakefile 16 Jan 2001 22:10:15 -  1.1.1.1
+++ patches/patch-Imakefile 1 Feb 2021 01:27:06 -
@@ -1,6 +1,7 @@
 $OpenBSD: patch-Imakefile,v 1.1.1.1 2001/01/16 22:10:15 naddy Exp $
 Imakefile.orig Mon May 17 22:36:35 1999
-+++ Imakefile  Tue Jan 16 16:41:43 2001
+Index: Imakefile
+--- Imakefile.orig
 Imakefile
 @@ -89,8 +89,7 @@ XCOMM
  
  MASTERXPM = xpm/$(MASTER_XPM).xpm
@@ -11,7 +12,7 @@ $OpenBSD: patch-Imakefile,v 1.1.1.1 2001
-L. \
-ldockapp \
  -lXpm -lXext -lX11
-@@ -132,7 +131,7 @@ XCOMM --
+@@ -132,7 +131,7 @@ XCOMM ---
  XCOMM Create Makefile
  XCOMM ---
  
Index: patches/patch-wmdate_c
===
RCS file: /cvs/ports/x11/wmdate/patches/patch-wmdate_c,v
retrieving revision 1.1.1.1
diff -u -p -u -p -r1.1.1.1 patch-wmdate_c
--- patches/patch-wmdate_c  16 Jan 2001 22:10:16 -  1.1.1.1
+++ patches/patch-wmdate_c  1 Feb 2021 01:27:06 -
@@ -1,9 +1,25 @@
 $OpenBSD: patch-wmdate_c,v 1.1.1.1 2001/01/16 22:10:16 naddy Exp $
 wmdate.c.orig  Mon May 17 22:25:23 1999
-+++ wmdate.c   Tue Jan 16 16:35:36 2001
-@@ -95,7 +95,7 @@ main (int argc, char *argv[])
+
+Hunk #1: fix with libdockapp>=0.7
+
+Index: wmdate.c
+--- wmdate.c.orig
 wmdate.c
+@@ -39,7 +39,7 @@ THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF S
+ #include 
+ #include 
+ 
+-#include 
++#include 
+ #include "config.h"
+ #include "geometry.h"
+ 
+@@ -93,9 +93,9 @@ main (int argc, char *argv[])
+Pixmap disp;   /* Pixmap that is actually displayed */
+unsigned long fore_pix, back_pix;
 int old_day;
-unsigned int w=183, h=84;
+-   unsigned int w=183, h=84;
++   unsigned short w=183, h=84;
  
 -   DAParseArguments(argc, argv, options, 4, DESCRIPTION, VERSION);
 +   DAParseArguments(argc, argv, options, 3, DESCRIPTION, VERSION);

net/wmnetload: fix with the new devel/libdockapp (2/4)

[Charlene Wendling]

This has been tested on amd64 and macppc. To potential testers, if
you're not using windowmaker, you'll need to use the '-w' option.

Comments/feedback are welcome,

Charlène.


Index: Makefile
===
RCS file: /cvs/ports/net/wmnetload/Makefile,v
retrieving revision 1.15
diff -u -p -u -p -r1.15 Makefile
--- Makefile12 Jul 2019 20:48:52 -  1.15
+++ Makefile1 Feb 2021 01:31:12 -
@@ -3,7 +3,7 @@
 COMMENT=   wm-dockapp; simple network interface monitoring tool
 
 DISTNAME=  wmnetload-1.3
-REVISION=  4
+REVISION=  5
 CATEGORIES=net x11 x11/windowmaker
 
 HOMEPAGE=  http://freshmeat.net/projects/wmnetload
@@ -14,7 +14,7 @@ WANTLIB=  X11 Xext Xpm c m dockapp
 
 MASTER_SITES=  ftp://truffula.com/pub/
 
-LIB_DEPENDS=   devel/libdockapp
+LIB_DEPENDS=   devel/libdockapp>=0.7.3
 
 SEPARATE_BUILD=Yes
 CONFIGURE_STYLE=   gnu
Index: patches/patch-configure
===
RCS file: /cvs/ports/net/wmnetload/patches/patch-configure,v
retrieving revision 1.2
diff -u -p -u -p -r1.2 patch-configure
--- patches/patch-configure 7 Dec 2015 14:58:54 -   1.2
+++ patches/patch-configure 1 Feb 2021 01:31:12 -
@@ -1,6 +1,10 @@
 $OpenBSD: patch-configure,v 1.2 2015/12/07 14:58:54 jasper Exp $
 configure.orig Sun Feb 23 10:08:00 2003
-+++ configure  Mon Dec  7 14:35:10 2015
+
+Hunk #2, 3: fix the build with libdockapp>=0.7
+
+Index: configure
+--- configure.orig
 configure
 @@ -3052,6 +3052,9 @@ case $host_os in
OS=freebsd
LIBRARY_RPATH="$LIBRARY_RPATH:/usr/X11R6/lib"
@@ -11,3 +15,21 @@ $OpenBSD: patch-configure,v 1.2 2015/12/
  *)
echo ""
echo "Sorry, $host_os is not supported yet."
+@@ -4713,7 +4716,7 @@ cat confdefs.h >>conftest.$ac_ext
+ cat >>conftest.$ac_ext <<_ACEOF
+ /* end confdefs.h.  */
+ $ac_includes_default
+-#include 
++#include 
+ _ACEOF
+ rm -f conftest.$ac_objext
+ if { (eval echo "$as_me:$LINENO: \"$ac_compile\"") >&5
+@@ -4748,7 +4751,7 @@ _ACEOF
+ cat confdefs.h >>conftest.$ac_ext
+ cat >>conftest.$ac_ext <<_ACEOF
+ /* end confdefs.h.  */
+-#include 
++#include 
+ _ACEOF
+ if { (eval echo "$as_me:$LINENO: \"$ac_cpp conftest.$ac_ext\"") >&5
+   (eval $ac_cpp conftest.$ac_ext) 2>conftest.er1
Index: patches/patch-src_wmnetload_c
===
RCS file: patches/patch-src_wmnetload_c
diff -N patches/patch-src_wmnetload_c
--- /dev/null   1 Jan 1970 00:00:00 -
+++ patches/patch-src_wmnetload_c   1 Feb 2021 01:31:12 -
@@ -0,0 +1,26 @@
+$OpenBSD$
+
+Hunk #1: Unbreak with libdockapp>=0.7
+Hunk #2: fix text display that was blank
+
+Index: src/wmnetload.c
+--- src/wmnetload.c.orig
 src/wmnetload.c
+@@ -37,7 +37,7 @@
+ #include 
+ #include 
+ #include 
+-#include 
++#include 
+ #ifdef HAVE_SYS_SOCKIO_H
+ #include 
+ #endif
+@@ -1051,7 +1051,7 @@ static void
+ setshape(void)
+ {
+   Pixmap mask, pixmap;
+-  unsigned int h, w;
++  unsigned short h, w;
+ 
+   if (DAMakePixmapFromData(backlight_off_xpm, &pixmap, &mask, &h, &w)) {
+   DASetShape(mask);

[update] devel/libdockapp 0.4 -> 0.7.3 (1/4)

[Charlene Wendling]

Hi,

Here is an update for libdockapp to 0.7.3. This is needed to update
some dockapps (i've already a diff for x11/wmitime) concerned by
-fno-common and that now uses this lib.

The major highlight is that the include file for  changed. All
consumers are impacted and no updates for them in this regard has been
provided, so there will be follow up diffs.

Port-wise: 

- symbols have been removed, bump major
- install fonts (a first for me, i took example from comms/x3270)
- move HOMEPAGE and MASTER_SITES to dockapps.net

This has been tested on amd64 and macppc successfully. 

Comments and feedback are welcome,

Charlène.


Index: Makefile
===
RCS file: /cvs/ports/devel/libdockapp/Makefile,v
retrieving revision 1.25
diff -u -p -u -p -r1.25 Makefile
--- Makefile12 Jul 2019 20:44:37 -  1.25
+++ Makefile1 Feb 2021 01:44:40 -
@@ -2,21 +2,24 @@
 
 COMMENT=   dockapp-making standard library for Window Maker
 
-DISTNAME=  libdockapp-0.4.0
-REVISION=  5
-SHARED_LIBS=   dockapp  1.0  # .1.0
+DISTNAME=  libdockapp-0.7.3
+SHARED_LIBS=   dockapp  2.0  # .1.0
 CATEGORIES=devel
 
+HOMEPAGE=  https://www.dockapps.net/libdockapp/
+
 # BSD
 PERMIT_PACKAGE=Yes
-WANTLIB=   X11 Xau Xdmcp Xext Xpm xcb
 
-MASTER_SITES=  ftp://shadowmere.student.utwente.nl/pub/WindowMaker/
+WANTLIB=   X11 Xext Xpm xcb
+
+MASTER_SITES=  https://www.dockapps.net/download/
 
 NO_TEST=   Yes
 SEPARATE_BUILD=Yes
 CONFIGURE_STYLE=   gnu
-CONFIGURE_ARGS+=   --enable-static
+CONFIGURE_ARGS+=   --enable-static \
+   --with-fontdir=${PREFIX}/share/fonts/libdockapps
 
 post-install:
${INSTALL_DATA_DIR} ${PREFIX}/share/doc/libdockapp
Index: distinfo
===
RCS file: /cvs/ports/devel/libdockapp/distinfo,v
retrieving revision 1.4
diff -u -p -u -p -r1.4 distinfo
--- distinfo18 Jan 2015 03:13:15 -  1.4
+++ distinfo1 Feb 2021 01:44:40 -
@@ -1,2 +1,2 @@
-SHA256 (libdockapp-0.4.0.tar.gz) = /X7Prvw8z2VCyKhmo3C85NUey4AozYmLCCNpgfiwzGU=
-SIZE (libdockapp-0.4.0.tar.gz) = 108566
+SHA256 (libdockapp-0.7.3.tar.gz) = FtmS9oTfzbGeNoKCLUlfRzYV+6bp4vp6EGMHvHZ7OJw=
+SIZE (libdockapp-0.7.3.tar.gz) = 387507
Index: patches/patch-src_Makefile_in
===
RCS file: /cvs/ports/devel/libdockapp/patches/patch-src_Makefile_in,v
retrieving revision 1.1
diff -u -p -u -p -r1.1 patch-src_Makefile_in
--- patches/patch-src_Makefile_in   2 Jan 2006 18:15:53 -   1.1
+++ patches/patch-src_Makefile_in   1 Feb 2021 01:44:40 -
@@ -1,12 +1,13 @@
 $OpenBSD: patch-src_Makefile_in,v 1.1 2006/01/02 18:15:53 bernd Exp $
 src/Makefile.in.orig   Mon Jan  2 18:19:24 2006
-+++ src/Makefile.inMon Jan  2 18:19:44 2006
-@@ -79,7 +79,7 @@ AUTOMAKE_OPTIONS = no-dependencies
- 
+Index: src/Makefile.in
+--- src/Makefile.in.orig
 src/Makefile.in
+@@ -335,7 +335,7 @@ xpm_CFLAGS = @xpm_CFLAGS@
+ xpm_LIBS = @xpm_LIBS@
+ AUTOMAKE_OPTIONS = no-dependencies
  lib_LTLIBRARIES = libdockapp.la
- 
--libdockapp_la_LDFLAGS = -version-info 1:0:0 @X_LIBS@
-+libdockapp_la_LDFLAGS = $(LIBdockapp_LTVERSION) @X_LIBS@
- 
- include_HEADERS = dockapp.h
- 
+-libdockapp_la_LDFLAGS = -version-info 3:1:0
++libdockapp_la_LDFLAGS = $(LIBdockapp_LTVERSION)
+ otherincludedir = $(includedir)/libdockapp
+ otherinclude_HEADERS = dockapp.h \
+ wmgeneral.h \
Index: pkg/PLIST
===
RCS file: /cvs/ports/devel/libdockapp/pkg/PLIST,v
retrieving revision 1.3
diff -u -p -u -p -r1.3 PLIST
--- pkg/PLIST   16 Mar 2015 18:07:41 -  1.3
+++ pkg/PLIST   1 Feb 2021 01:44:40 -
@@ -1,7 +1,26 @@
 @comment $OpenBSD: PLIST,v 1.3 2015/03/16 18:07:41 naddy Exp $
-include/dockapp.h
-lib/libdockapp.a
+include/libdockapp/
+include/libdockapp/dockapp.h
+include/libdockapp/list.h
+include/libdockapp/misc.h
+include/libdockapp/wmgeneral.h
+@static-lib lib/libdockapp.a
 lib/libdockapp.la
 @lib lib/libdockapp.so.${LIBdockapp_VERSION}
+lib/pkgconfig/dockapp.pc
 share/doc/libdockapp/
 share/doc/libdockapp/COPYING
+share/doc/libdockapp/README
+share/doc/libdockapp/basic/
+share/doc/libdockapp/basic/Makefile
+share/doc/libdockapp/basic/README
+share/doc/libdockapp/basic/ball_red.xpm
+share/doc/libdockapp/basic/basic.c
+share/doc/libdockapp/rectangles/
+share/doc/libdockapp/rectangles/Makefile
+share/doc/libdockapp/rectangles/rectangles.c
+share/fonts/
+@fontdir share/fonts/libdockapps/
+share/fonts/libdockapps/fonts.dir
+share/fonts/libdockapps/luxel-ascii-06x09.pcf.gz
+share/fonts/libdockapps/seg7-ascii-05x07.pcf.gz

Re: Use -fcommon with or remove security/siphon

[Charlene Wendling]

On Mon, 1 Feb 2021 02:34:53 +0100
Klemens Nanni  wrote:

> Upstream still ships the tarball, that's it,
> "primary distsite and homepage have gone away" in 2002 as per CVS log.
> 
> The only TCP port I've been able to detect (after producing traffing
> on a variety of them) is SSH -- and that only worked after enabling
> IPv4.
> 
> Siphon does not seem to support/detect IPv6 at all and it's OS
> fingerprints are extremely old;  with Gentoo we're the only ones still
> packaging it according to Repology.
> 
> I tend to simply remove it (ports tree house keeping), but here's a
> diff to "fix" the port in case anyone actually finds it useful.
> 
> OK either for removal or the diff?

OK cwen@ to remove.

> ---
> Set "-fcommon" for twelve duplicate symbols.
> 
> Regen patch and add "@bin" PLIST marker while here.
> 
> 
> Index: Makefile
> ===
> RCS file: /cvs/ports/security/siphon/Makefile,v
> retrieving revision 1.16
> diff -u -p -r1.16 Makefile
> --- Makefile  12 Jul 2019 20:49:36 -  1.16
> +++ Makefile  1 Feb 2021 01:13:10 -
> @@ -4,7 +4,7 @@ COMMENT=  passive network mapping tool
>  
>  DISTNAME=siphon-v.666
>  PKGNAME= siphon-0.666
> -REVISION=0
> +REVISION=1
>  CATEGORIES=  security net
>  
>  # BSD-style
> @@ -22,6 +22,9 @@ MAKE_FLAGS= CC="${CC}" \
>  MAKE_ENV=SYSCONFDIR="${SYSCONFDIR}"
>  
>  NO_TEST= Yes
> +
> +# twelve duplicate symbols
> +CFLAGS +=-fcommon
>  
>  do-install:
>   ${INSTALL_PROGRAM} ${WRKSRC}/siphon ${PREFIX}/bin
> Index: patches/patch-log_c
> ===
> RCS file: /cvs/ports/security/siphon/patches/patch-log_c,v
> retrieving revision 1.1.1.1
> diff -u -p -r1.1.1.1 patch-log_c
> --- patches/patch-log_c   16 Jun 2001 14:56:26 -
> 1.1.1.1 +++ patches/patch-log_c   1 Feb 2021 01:12:32 -
> @@ -1,7 +1,8 @@
>  $OpenBSD: patch-log_c,v 1.1.1.1 2001/06/16 14:56:26 lebel Exp $
>  log.c.orig   Fri Jun 15 18:36:47 2001
> -+++ log.cFri Jun 15 18:36:55 2001
> -@@ -137,7 +137,7 @@ char *oslookup(int window, int ttl, int 
> +Index: log.c
> +--- log.c.orig
>  log.c
> +@@ -137,7 +137,7 @@ char *oslookup(int window, int ttl, int df)
> FILE *osprints;
> static char line[80], *oswin, *osttl, *osdf, *os, hexed[10];
> static int check = 0;
> Index: pkg/PLIST
> ===
> RCS file: /cvs/ports/security/siphon/pkg/PLIST,v
> retrieving revision 1.4
> diff -u -p -r1.4 PLIST
> --- pkg/PLIST 24 Sep 2004 09:37:39 -  1.4
> +++ pkg/PLIST 1 Feb 2021 01:13:24 -
> @@ -1,5 +1,5 @@
>  @comment $OpenBSD: PLIST,v 1.4 2004/09/24 09:37:39 espie Exp $
> -bin/siphon
> +@bin bin/siphon
>  share/doc/siphon/
>  share/doc/siphon/LICENSE
>  share/doc/siphon/README
> 

misc/deco: -fno-common fix

[Charlene Wendling]

Hi,

This fix has been taken once again from FreeBSD [0]. This builds and
works fine on amd64.

OK? 

Charlène.


[0] https://github.com/freebsd/freebsd-ports/commit/4b6803fe


Index: Makefile
===
RCS file: /cvs/ports/misc/deco/Makefile,v
retrieving revision 1.27
diff -u -p -u -p -r1.27 Makefile
--- Makefile26 Jan 2020 11:14:32 -  1.27
+++ Makefile31 Jan 2021 21:39:05 -
@@ -4,7 +4,7 @@ COMMENT=Demos Commander, a free Norton 
 
 DISTNAME=  deco383
 PKGNAME=   deco-3.8.3
-REVISION=  1
+REVISION=  2
 CATEGORIES=misc
 MASTER_SITES=  https://distfiles.sigtrap.nl/
 EXTRACT_SUFX=  .tgz
Index: patches/patch-env_h
===
RCS file: patches/patch-env_h
diff -N patches/patch-env_h
--- /dev/null   1 Jan 1970 00:00:00 -
+++ patches/patch-env_h 31 Jan 2021 21:39:05 -
@@ -0,0 +1,13 @@
+$OpenBSD$
+
+Fix -fno-common build failure.
+
+Index: env.h
+--- env.h.orig
 env.h
+@@ -1,4 +1,4 @@
+-char **EnvVector;
++static char **EnvVector;
+ void EnvInit (char **env);
+ char *EnvGet (char *name);
+ void EnvPut (char *name, char *value);
Index: patches/patch-scr_c
===
RCS file: /cvs/ports/misc/deco/patches/patch-scr_c,v
retrieving revision 1.1
diff -u -p -u -p -r1.1 patch-scr_c
--- patches/patch-scr_c 21 Mar 2002 09:59:48 -  1.1
+++ patches/patch-scr_c 31 Jan 2021 21:39:05 -
@@ -1,6 +1,19 @@
 $OpenBSD: patch-scr_c,v 1.1 2002/03/21 09:59:48 form Exp $
 scr.c.orig Tue Dec 29 22:57:11 1998
-+++ scr.c  Thu Mar 21 15:31:17 2002
+
+Hunk #1: Fix -fno-common build failure.
+
+Index: scr.c
+--- scr.c.orig
 scr.c
+@@ -148,7 +148,7 @@ WINDOW VScreen;
+ int BlackWhite = 0;
+ int ColorMode = 1;
+ int GraphMode = 1;
+-int TtyUpperCase = 0;
++extern int TtyUpperCase;
+ 
+ static WINDOW curscr;
+ static scrool, rscrool;
 @@ -167,7 +167,7 @@ static char *KS, *KE;
  
  static char *CL, *CM, *SE, *SO, *TE, *TI, *VE, *VS,
Index: patches/patch-tty_c
===
RCS file: patches/patch-tty_c
diff -N patches/patch-tty_c
--- /dev/null   1 Jan 1970 00:00:00 -
+++ patches/patch-tty_c 31 Jan 2021 21:39:05 -
@@ -0,0 +1,16 @@
+$OpenBSD$
+
+Fix -fno-common build failure.
+
+Index: tty.c
+--- tty.c.orig
 tty.c
+@@ -82,7 +82,7 @@ static struct ltchars oldchars, newchars;
+ #   define OXTABS TAB3
+ #endif
+ 
+-int TtyUpperCase;
++int TtyUpperCase = 0;
+ 
+ #define NOCHAR 0
+ 

net/ncftp: update with -fno-common

[Charlene Wendling]

Hi,

The fix has been taken from FreeBSD [0]. With that i can build and run
ncftp without issues on amd64. While here, i moved HOMEPAGE to https.

OK?

Charlène.


[0] https://github.com/freebsd/freebsd-ports/commit/3f1dcac7


Index: Makefile
===
RCS file: /cvs/ports/net/ncftp/Makefile,v
retrieving revision 1.47
diff -u -p -u -p -r1.47 Makefile
--- Makefile10 Mar 2020 03:31:45 -  1.47
+++ Makefile31 Jan 2021 21:07:48 -
@@ -4,8 +4,8 @@ COMMENT=FTP client with advanced user i
 
 DISTNAME=  ncftp-3.2.6
 CATEGORIES=net
-HOMEPAGE=  http://www.ncftp.com/ncftp/
-REVISION=  1
+HOMEPAGE=  https://www.ncftp.com/ncftp/
+REVISION=  2
 
 # Artistic
 PERMIT_PACKAGE=Yes
Index: patches/patch-sh_util_gpshare_c
===
RCS file: patches/patch-sh_util_gpshare_c
diff -N patches/patch-sh_util_gpshare_c
--- /dev/null   1 Jan 1970 00:00:00 -
+++ patches/patch-sh_util_gpshare_c 31 Jan 2021 21:07:48 -
@@ -0,0 +1,16 @@
+$OpenBSD$
+
+Fix -fno-common build failure.
+
+Index: sh_util/gpshare.c
+--- sh_util/gpshare.c.orig
 sh_util/gpshare.c
+@@ -28,7 +28,7 @@
+ 
+ static int gIsAtty1 = 1, gIsAtty2 = 1;
+ extern int gLoadedBm, gBookmarkMatchMode;
+-Bookmark gBm;
++extern Bookmark gBm;
+ 
+ double
+ FileSize(double size, const char **uStr0, double *uMult0)

Re: games/ace: -fno-common fix

[Charlene Wendling]

On Mon, 1 Feb 2021 00:38:19 +1100
Jonathan Gray  wrote:

> On Sun, Jan 31, 2021 at 01:30:19PM +0100, Charlene Wendling wrote:
> > Hi,
> > 
> > This fix has been taken from FreeBSD [0]. check_sym reports no
> > dynamic change. This has been built and tested successfully on
> > macppc and amd64.
> > 
> > OK? 
> 
> Is a revision bump strictly needed for this?

As stated in the porting guide: « Anything that affects the binary
package implies a bump. This includes HOMEPAGE, MAINTAINER or
description changes, changes to patches or build flags. »

There are exceptions to this, but when in doubt, it's better to bump.

If you extract the p1 ace package and compare it to p2, you get the
following, showing we should bump revision (if someone has an easier
way to do this, i would be happy to try out): 

$ diff -u ace.{orig,new}/+CONTENTS | awk '/^.@(sha|lib|bin|name)/ || /^.[a-z]/' 
 
-@name ace-1.4p1
+@name ace-1.4p2
 @bin bin/canfield
 @sha FxJ+4ebeIYT29VAoweyphGjWSGQjSsHHEc4HlQ8ykHY=
 @bin bin/freecell
 @sha ieFkiDSBc2StI2uWXy+5viHg7mvbmLrUK6HJJZ4YWho=
 @bin bin/golf
 @sha 6A/XiMiKxLFpOBLWl1yYD8k5b2nh93aG/j3T5OWIxE8=
 @bin bin/mastermind
 @sha V080t6kHmhU64KWGM4LEgfSGcZ8vRW550CVDT1yoG/U=
 @bin bin/merlin
 @sha 29PgmSv41RETlKjuTXqU3HtO170AIAFYmOftXX9Q3Sw=
 @bin bin/minesweeper
 @sha b0fyUOEbbdQMvvxmdkdyzlf07bkxNA0/nAi9yQ3hIyE=
 @bin bin/pegged
 @sha xwBzJCT2ViSVcMZCAtCWgc2cOvPU3ERjMERB+sFkcIc=
 @bin bin/penguins
 @sha c05np8yPse2w039vvzCoC3ggjsL1ONg2HO+MRoBnLpk=
 @bin bin/solitaire
 @sha vcTyoA0fhyGxSATvK54rDQk81eS8yuv6pi5WSGvoLOU=
 @bin bin/spider
 @sha +8qcljp6IZl8o4aFoB6ArnkxIG9d/BkNu0NCiNmgbSQ=
 @bin bin/taipedit
-@sha VtXInNZJp3vcqOyRAQVEOZOFIZN+tgfGKV6XTOsuZb4=
+@sha mLXyO3hRYNcIjDrDuYcV4s//rI/mcKv1noFSJKqmL+g=
 @bin bin/taipei
-@sha /RCvEK3MYqLA9nfrlMZ+vAvytAhHXy1ua0GlyUbPT24=
+@sha CJg34MSvzRNUzp6ND1cNqGQW4RWSW5JBWl2U5Uc/ND4=
 @bin bin/thornq
 @sha xGPdJRHJWWlbfazQQnNfuQ1tPlF34tAOOIQqcGEz3II=
 lib/libcards.a
-@sha AZ9WtPI/9OEQZFbUxGH074A/L87sXmuBQ+JUuVvrPUQ=
+@sha ZW6cmJaV0GzmvHSKb9d0QJXovsKtoOzMMtCmP1ZhGTI=
 lib/libcards.la
 @sha KhETIzRiigcLGRlsXwVP42LzPFWmUkX/oADVC3zRO5s=
 @lib lib/libcards.so.0.0
-@sha hcmlxe2+4tg08anjTxpXzj6yqZ1cAxODVgMOPrLbup4=
+@sha Fk+bcQ5+DEtXH5hCmj+a3D21FNbBeR3uDkY9EKqojUc=


> ok jsg@
> 
> > 
> > Charlène. 
> > 
> > 
> > [0]  https://github.com/freebsd/freebsd-ports/commit/53cdc659
> > 
> > 
> > Index: Makefile
> > ===
> > RCS file: /cvs/ports/games/ace/Makefile,v
> > retrieving revision 1.5
> > diff -u -p -u -p -r1.5 Makefile
> > --- Makefile2 Oct 2019 21:11:08 -   1.5
> > +++ Makefile31 Jan 2021 12:18:00 -
> > @@ -3,7 +3,7 @@
> >  COMMENT =  solitaire games
> >  
> >  DISTNAME = ace-1.4
> > -REVISION = 1
> > +REVISION = 2
> >  
> >  SHARED_LIBS +=  cards 0.0 # 1.0
> >  
> > Index: patches/patch-lib_table_c
> > ===
> > RCS file: patches/patch-lib_table_c
> > diff -N patches/patch-lib_table_c
> > --- /dev/null   1 Jan 1970 00:00:00 -
> > +++ patches/patch-lib_table_c   31 Jan 2021 12:18:00 -
> > @@ -0,0 +1,16 @@
> > +$OpenBSD$
> > +
> > +Fix the build with -fno-common
> > +
> > +Index: lib/table.c
> > +--- lib/table.c.orig
> >  lib/table.c
> > +@@ -57,7 +57,7 @@ static int ex=0, ey=0, ew=0, eh=0;
> > + static int graphics_disabled = 1;
> > + 
> > + OptionDesc *app_options;
> > +-OptionDesc *xwin_options;
> > ++extern OptionDesc *xwin_options;
> > + static OptionDesc *options[5];
> > + 
> > + static OptionDesc ace_options[] = {
> > 
> > 
> 

Re: security/libsrtp -fno-common

[Charlene Wendling]

On Sun, 31 Jan 2021 12:52:54 +
Stuart Henderson  wrote:

> On 2021/01/31 13:15, Charlene Wendling wrote:
> > On Sun, 31 Jan 2021 12:03:39 +
> > Stuart Henderson  wrote:
> > 
> > > On 2021/01/31 12:58, Charlene Wendling wrote:
> > > > On Sun, 31 Jan 2021 16:38:02 +1100
> > > > Jonathan Gray  wrote:
> > > > 
> > > > > https://github.com/cisco/libsrtp/commit/716a73862b387a2107f37398c0fb7d9a754c0ccd.patch
> > > > > 
> > > > > /tmp/libsrtp2.so.1.0 --> /usr/local/lib/libsrtp2.so.2.0
> > > > > Dynamic export changes:
> > > > > removed:
> > > > >   bit_string
> > > > > 
> > > 
> > > Deps still work, OK with me.
> > > 
> > > > Tested on amd64 and macppc (vanilla clang there). It needs
> > > > itself to be installed for tests to run, i've tweaked your diff
> > > > to do so. 
> > > 
> > > Can you paste a log please, I don't see that?
> > 
> > 
> > Huh, weird. I have attached the log.
> 
> Very odd. Yours is linking libsrtp2.so.2.0 right after "ar cr
> libsrtp2.a ..." so the .so is present when the test programs are
> compiled/linked so they pick it up. I can't find a way to replicate
> that. Mine doesn't link libsrtp2.so.2.0 until after the tests have
> built so they pick up the static lib.
> 
> (not that this is related to the patch)
> 

Sorry for the noise, it appears that MAKE_JOBS>1 is responsible for
this.

games/ace: -fno-common fix

[Charlene Wendling]

Hi,

This fix has been taken from FreeBSD [0]. check_sym reports no dynamic
change. This has been built and tested successfully on macppc and amd64.

OK? 

Charlène. 


[0]  https://github.com/freebsd/freebsd-ports/commit/53cdc659


Index: Makefile
===
RCS file: /cvs/ports/games/ace/Makefile,v
retrieving revision 1.5
diff -u -p -u -p -r1.5 Makefile
--- Makefile2 Oct 2019 21:11:08 -   1.5
+++ Makefile31 Jan 2021 12:18:00 -
@@ -3,7 +3,7 @@
 COMMENT =  solitaire games
 
 DISTNAME = ace-1.4
-REVISION = 1
+REVISION = 2
 
 SHARED_LIBS +=  cards 0.0 # 1.0
 
Index: patches/patch-lib_table_c
===
RCS file: patches/patch-lib_table_c
diff -N patches/patch-lib_table_c
--- /dev/null   1 Jan 1970 00:00:00 -
+++ patches/patch-lib_table_c   31 Jan 2021 12:18:00 -
@@ -0,0 +1,16 @@
+$OpenBSD$
+
+Fix the build with -fno-common
+
+Index: lib/table.c
+--- lib/table.c.orig
 lib/table.c
+@@ -57,7 +57,7 @@ static int ex=0, ey=0, ew=0, eh=0;
+ static int graphics_disabled = 1;
+ 
+ OptionDesc *app_options;
+-OptionDesc *xwin_options;
++extern OptionDesc *xwin_options;
+ static OptionDesc *options[5];
+ 
+ static OptionDesc ace_options[] = {