Re: NEW: games/urbanterror and games/urbanterror-data
On Sat, Aug 20, 2016 at 04:25:03PM -0700, Ryan Freeman wrote: > On Sat, Aug 20, 2016 at 03:43:34PM -0700, Ryan Freeman wrote: > > On Sat, Aug 20, 2016 at 03:27:52PM -0700, Ryan Freeman wrote: > > > Hey, > > > > > > From their website: > > > - > > > Urban Terror(TM) is a free multiplayer first person shooter developed by > > > FrozenSand, that will run on any Quake III Arena compatible engine. It > > > is available for Windows, Linux and Macintosh. > > > > > > Urban Terror can be described as a Hollywood tactical shooter; somewhat > > > realism based, but the motto is "fun over realism". This results in a > > > very unique, enjoyable and addictive game. > > > - > > > > > > Now that this is merged: > > > https://github.com/Barbatos/ioq3-for-UrbanTerror-4/commit/27bd827f1ba04b5846c302e447adfdb556e66b7f > > > > > > I have created a port for Urban Terror 4.2.023. > > > > > > Big thanks to Charles Duprey (Barbatos) for merging in the patches > > > upstream! > > > > > > I am unsure if the license markers are correct enough, please advise. > > > > > > The package for data is quite large, 1.8GB, beware when trying to build > > > this. > > > I wanted to try and use MULTI_PACKAGES but I couldn't wrap my head around > > > it, > > > particularly that the engine is coming from github and the data from their > > > servers. Hopefully this is still acceptable! > > > > > > Cheers, > > > --ryan > > > > Very sorry, original post for this had ${ARCH} used in the engine Makefile, > > I had intended to fix that properly before posting and forgot once I had > > the data port done. :( > > > > Please use the updated attached tarball for engine, it removes the ${ARCH} > > usage in port Makefile by patching the project Makefile to remove the ARCH > > extensions. So not quite a patchless port now, but close! > > > > Is there a smart way we can override what I have patched using MAKE_FLAGS > > or MAKE_ENV? > > > > Cheers! > > -ryan > > And I just noticed this still needs the wxneeded treatment. I am not sure > that > USE_WXNEEDED has any effect here, nor does using MAKE_ENV to do: > LDFLAGS="${LDFLAGS} -Wl,-z,wxneeded" > > Is there a better method or might I have to patch this into the Makefile for > now? Attached an updated version with the wxneeded stuff patched right into the project Makefile under the OpenBSD section. I added USE_WXNEEDED = Yes to the top of the port Makefile as a marker. Also removed the MAKE_FLAGS line dealing with COPYDIR, as I just install the game binaries directly via do-install anyhow. ok now I hope? :-) -ryan urbanterror.tgz Description: application/tar-gz
Re: NEW: games/urbanterror and games/urbanterror-data
On Sat, Aug 20, 2016 at 03:43:34PM -0700, Ryan Freeman wrote: > On Sat, Aug 20, 2016 at 03:27:52PM -0700, Ryan Freeman wrote: > > Hey, > > > > From their website: > > - > > Urban Terror(TM) is a free multiplayer first person shooter developed by > > FrozenSand, that will run on any Quake III Arena compatible engine. It > > is available for Windows, Linux and Macintosh. > > > > Urban Terror can be described as a Hollywood tactical shooter; somewhat > > realism based, but the motto is "fun over realism". This results in a > > very unique, enjoyable and addictive game. > > - > > > > Now that this is merged: > > https://github.com/Barbatos/ioq3-for-UrbanTerror-4/commit/27bd827f1ba04b5846c302e447adfdb556e66b7f > > > > I have created a port for Urban Terror 4.2.023. > > > > Big thanks to Charles Duprey (Barbatos) for merging in the patches upstream! > > > > I am unsure if the license markers are correct enough, please advise. > > > > The package for data is quite large, 1.8GB, beware when trying to build > > this. > > I wanted to try and use MULTI_PACKAGES but I couldn't wrap my head around > > it, > > particularly that the engine is coming from github and the data from their > > servers. Hopefully this is still acceptable! > > > > Cheers, > > --ryan > > Very sorry, original post for this had ${ARCH} used in the engine Makefile, > I had intended to fix that properly before posting and forgot once I had > the data port done. :( > > Please use the updated attached tarball for engine, it removes the ${ARCH} > usage in port Makefile by patching the project Makefile to remove the ARCH > extensions. So not quite a patchless port now, but close! > > Is there a smart way we can override what I have patched using MAKE_FLAGS > or MAKE_ENV? > > Cheers! > -ryan And I just noticed this still needs the wxneeded treatment. I am not sure that USE_WXNEEDED has any effect here, nor does using MAKE_ENV to do: LDFLAGS="${LDFLAGS} -Wl,-z,wxneeded" Is there a better method or might I have to patch this into the Makefile for now?
NEW: games/urbanterror and games/urbanterror-data
Hey, >From their website: - Urban Terror(TM) is a free multiplayer first person shooter developed by FrozenSand, that will run on any Quake III Arena compatible engine. It is available for Windows, Linux and Macintosh. Urban Terror can be described as a Hollywood tactical shooter; somewhat realism based, but the motto is "fun over realism". This results in a very unique, enjoyable and addictive game. - Now that this is merged: https://github.com/Barbatos/ioq3-for-UrbanTerror-4/commit/27bd827f1ba04b5846c302e447adfdb556e66b7f I have created a port for Urban Terror 4.2.023. Big thanks to Charles Duprey (Barbatos) for merging in the patches upstream! I am unsure if the license markers are correct enough, please advise. The package for data is quite large, 1.8GB, beware when trying to build this. I wanted to try and use MULTI_PACKAGES but I couldn't wrap my head around it, particularly that the engine is coming from github and the data from their servers. Hopefully this is still acceptable! Cheers, --ryan urbanterror.tgz Description: application/tar-gz urbanterror-data.tgz Description: application/tar-gz
Re: NEW: games/urbanterror and games/urbanterror-data
On Sat, Aug 20, 2016 at 03:27:52PM -0700, Ryan Freeman wrote: > Hey, > > From their website: > - > Urban Terror(TM) is a free multiplayer first person shooter developed by > FrozenSand, that will run on any Quake III Arena compatible engine. It > is available for Windows, Linux and Macintosh. > > Urban Terror can be described as a Hollywood tactical shooter; somewhat > realism based, but the motto is "fun over realism". This results in a > very unique, enjoyable and addictive game. > - > > Now that this is merged: > https://github.com/Barbatos/ioq3-for-UrbanTerror-4/commit/27bd827f1ba04b5846c302e447adfdb556e66b7f > > I have created a port for Urban Terror 4.2.023. > > Big thanks to Charles Duprey (Barbatos) for merging in the patches upstream! > > I am unsure if the license markers are correct enough, please advise. > > The package for data is quite large, 1.8GB, beware when trying to build this. > I wanted to try and use MULTI_PACKAGES but I couldn't wrap my head around it, > particularly that the engine is coming from github and the data from their > servers. Hopefully this is still acceptable! > > Cheers, > --ryan Very sorry, original post for this had ${ARCH} used in the engine Makefile, I had intended to fix that properly before posting and forgot once I had the data port done. :( Please use the updated attached tarball for engine, it removes the ${ARCH} usage in port Makefile by patching the project Makefile to remove the ARCH extensions. So not quite a patchless port now, but close! Is there a smart way we can override what I have patched using MAKE_FLAGS or MAKE_ENV? Cheers! -ryan urbanterror.tgz Description: application/tar-gz
Re: PATCH: devel/qt-creator wxneeded
Caspar Schutijser writes: > qtcreator crashes when one selects the "Help" tab. Webkit strikes again. Committed, thanks. -- jca | PGP : 0x1524E7EE / 5135 92C1 AD36 5293 2BDF DDCC 0DFA 74AE 1524 E7EE
Re: wxneeded and gcc 4.9.3
Il 20 agosto 2016 22:17:36 CEST, Stuart Henderson ha scritto: >On 2016/08/20 17:29, Stuart Henderson wrote: >> On 2016/08/20 14:19, Adam Wolk wrote: >> > PS. I tried switching to USE_WXNEEDED but it seems to have no >effect on cmake. >> >> USE_WXNEEDED does work with cmake, but it's not working with gcc4 >> because it hardcodes /usr/bin/ld rather than searching the path >> (and removing the path from --with-ld in the gcc4 port's Makefile >> results in autoconf failing there). > >Aha - this seems to work (i.e. builds and uses ld from the path, so >the wrapper script is found). Any comments on whether it is desirable? I think this approach is good, ok with me. >(I'd like some simple way to avoid having to decipher the libreoffice >build :) > >Index: Makefile >=== >RCS file: /cvs/ports/lang/gcc/4.9/Makefile,v >retrieving revision 1.33 >diff -u -p -r1.33 Makefile >--- Makefile 14 Jul 2016 07:22:31 - 1.33 >+++ Makefile 20 Aug 2016 20:15:11 - >@@ -4,7 +4,7 @@ ONLY_FOR_ARCHS = amd64 arm hppa i386 mip > DPB_PROPERTIES = parallel > > V = 4.9.3 >-REVISION = 9 >+REVISION = 10 > FULL_VERSION = $V > FULL_PKGVERSION = $V > >@@ -137,7 +137,6 @@ CONFIGURE_ARGS += \ > --disable-libgomp \ > --disable-tls \ > --with-as=/usr/bin/as \ >- --with-ld=/usr/bin/ld \ > --with-gnu-ld \ > --with-gnu-as \ > --enable-threads=posix \
PATCH: devel/qt-creator wxneeded
qtcreator crashes when one selects the "Help" tab. Webkit strikes again. Best regards, Caspar Schutijser Index: Makefile === RCS file: /cvs/ports/devel/qt-creator/Makefile,v retrieving revision 1.16 diff -u -p -r1.16 Makefile --- Makefile18 Mar 2016 23:12:16 - 1.16 +++ Makefile20 Aug 2016 20:37:22 - @@ -1,12 +1,13 @@ # $OpenBSD: Makefile,v 1.16 2016/03/18 23:12:16 naddy Exp $ BROKEN-powerpc = qdoc3 goes into an infinite loop +USE_WXNEEDED = Yes COMMENT = cross-platform IDE for use with Qt V =1.3.1 DISTNAME = qt-creator-${V}-src PKGNAME = qt-creator-${V} -REVISION = 6 +REVISION = 7 EXTRACT_SUFX = .zip ONLY_FOR_ARCHS = ${GCC4_ARCHS}
Update print/poppler
Here comes the monthly poppler update. Only some bugfixes and the non-gnu-grep fix to configure.ac, no interface changes this time. I'll test and commit it when my builds finish (probably in a day or two), but in case anyone wants to run some special tests... Index: Makefile === RCS file: /cvs/ports/print/poppler/Makefile,v retrieving revision 1.109 diff -u -p -r1.109 Makefile --- Makefile1 Aug 2016 18:35:40 - 1.109 +++ Makefile20 Aug 2016 20:30:00 - @@ -6,7 +6,7 @@ COMMENT-qt4=qt4 interface to PDF render COMMENT-qt5= Qt5 interface to PDF rendering library COMMENT-utils= PDF conversion tools and utilities -V= 0.46.0 +V= 0.47.0 DISTNAME= poppler-$V CATEGORIES=print PKGNAME-main= poppler-$V Index: distinfo === RCS file: /cvs/ports/print/poppler/distinfo,v retrieving revision 1.56 diff -u -p -r1.56 distinfo --- distinfo1 Aug 2016 18:35:40 - 1.56 +++ distinfo20 Aug 2016 20:30:00 - @@ -1,4 +1,4 @@ SHA256 (poppler-0.16.7.tar.gz) = +rTory001BQqcYgZ/XR9QJMLce9SDuDmvMlOLETyK80= -SHA256 (poppler-0.46.0.tar.xz) = ln010T1h3uL+5la4Dv7544ip51K8ebcSPxW0nHdp5Ic= +SHA256 (poppler-0.47.0.tar.xz) = uHLnIo/DSnHOS0elrqKleuZ1KIGPqEbh4O2giTGb0kI= SIZE (poppler-0.16.7.tar.gz) = 1952179 -SIZE (poppler-0.46.0.tar.xz) = 1680940 +SIZE (poppler-0.47.0.tar.xz) = 1682184 Index: patches/patch-configure_ac === RCS file: /cvs/ports/print/poppler/patches/patch-configure_ac,v retrieving revision 1.20 diff -u -p -r1.20 patch-configure_ac --- patches/patch-configure_ac 1 Aug 2016 18:35:40 - 1.20 +++ patches/patch-configure_ac 20 Aug 2016 20:30:00 - @@ -26,15 +26,6 @@ $OpenBSD: patch-configure_ac,v 1.20 2016 dnl dnl Try Qt4 dnl -@@ -779,7 +797,7 @@ if test x$enable_poppler_qt5 = xyes; then - AC_CHECK_TOOL(MOCQT52, moc-qt5) - AC_MSG_CHECKING([for Qt5 moc-qt5]) - mocversion=`$MOCQT52 -v 2>&1` --mocversiongrep=`echo $mocversion | grep "Qt 5\|moc-qt5 5\|moc 5"` -+mocversiongrep=`echo $mocversion | grep -E "Qt 5|moc-qt5 5|moc 5"` - if test x"$mocversiongrep" != x"$mocversion"; then - AC_CHECK_TOOL(QTCHOOSER, qtchooser) - AC_MSG_CHECKING([for qtchooser]) @@ -997,6 +1015,9 @@ glib/reference/Makefile glib/reference/version.xml glib/demo/Makefile Ciao, Kili
[PATCH] geo/josm update to r.10786
usual and straight forward update to josm 10786. Tested with -current and latest java 1.8 (with the wxallowed changes). Regards Holger Index: Makefile === RCS file: /cvs/ports/geo/josm/Makefile,v retrieving revision 1.35 diff -u -p -u -r1.35 Makefile --- Makefile22 Jun 2016 06:20:19 - 1.35 +++ Makefile20 Aug 2016 19:50:06 - @@ -2,7 +2,7 @@ COMMENT= OpenStreetMap map editor -SVN_REV= 10448 +SVN_REV= 10786 DISTNAME= josm-snapshot-${SVN_REV} EXTRACT_SUFX = .jar EXTRACT_CASES+=*.jar) unzip -qa ${FULLDISTDIR}/$$archive -d ${WRKDIST};; Index: distinfo === RCS file: /cvs/ports/geo/josm/distinfo,v retrieving revision 1.19 diff -u -p -u -r1.19 distinfo --- distinfo22 Jun 2016 06:20:19 - 1.19 +++ distinfo20 Aug 2016 19:50:06 - @@ -1,2 +1,2 @@ -SHA256 (josm-snapshot-10448.jar) = qZSlAdOrH7jcNQvbm/Y08b9rLBQ2W03FVHIyl/0Xbd8= -SIZE (josm-snapshot-10448.jar) = 10909486 +SHA256 (josm-snapshot-10786.jar) = 9sr9W+donQDk7NH1KP5Qedqm9U63eByhvrFFxO0Yi7Y= +SIZE (josm-snapshot-10786.jar) = 10703670
wxneeded and gcc 4.9.3
On 2016/08/20 17:29, Stuart Henderson wrote: > On 2016/08/20 14:19, Adam Wolk wrote: > > PS. I tried switching to USE_WXNEEDED but it seems to have no effect on > > cmake. > > USE_WXNEEDED does work with cmake, but it's not working with gcc4 > because it hardcodes /usr/bin/ld rather than searching the path > (and removing the path from --with-ld in the gcc4 port's Makefile > results in autoconf failing there). Aha - this seems to work (i.e. builds and uses ld from the path, so the wrapper script is found). Any comments on whether it is desirable? (I'd like some simple way to avoid having to decipher the libreoffice build :) Index: Makefile === RCS file: /cvs/ports/lang/gcc/4.9/Makefile,v retrieving revision 1.33 diff -u -p -r1.33 Makefile --- Makefile14 Jul 2016 07:22:31 - 1.33 +++ Makefile20 Aug 2016 20:15:11 - @@ -4,7 +4,7 @@ ONLY_FOR_ARCHS = amd64 arm hppa i386 mip DPB_PROPERTIES = parallel V = 4.9.3 -REVISION = 9 +REVISION = 10 FULL_VERSION = $V FULL_PKGVERSION = $V @@ -137,7 +137,6 @@ CONFIGURE_ARGS += \ --disable-libgomp \ --disable-tls \ --with-as=/usr/bin/as \ - --with-ld=/usr/bin/ld \ --with-gnu-ld \ --with-gnu-as \ --enable-threads=posix \
USE_WXNEEDED for cmake
Hi! Simple USE_WXNEEDED support for cmake, and the related patch for trojita. Tested with trojita on amd64. Comments? OKs? Index: Makefile === RCS file: /cvs/ports/devel/cmake/Makefile,v retrieving revision 1.148 diff -u -p -r1.148 Makefile --- Makefile13 Aug 2016 15:01:13 - 1.148 +++ Makefile20 Aug 2016 18:20:50 - @@ -6,7 +6,7 @@ COMMENT = portable build system VER = 3.6.1 DISTNAME = cmake-${VER} -REVISION = 1 +REVISION = 2 CATEGORIES = devel HOMEPAGE = http://www.cmake.org/ Index: cmake.port.mk === RCS file: /cvs/ports/devel/cmake/cmake.port.mk,v retrieving revision 1.58 diff -u -p -r1.58 cmake.port.mk --- cmake.port.mk 29 Apr 2016 10:34:33 - 1.58 +++ cmake.port.mk 20 Aug 2016 18:20:50 - @@ -63,7 +63,8 @@ CONFIGURE_ENV += MODJAVA_VER=${MODJAVA_V MODTCL_LIBDIR=${MODTCL_LIBDIR} \ MODTK_LIBDIR=${MODTK_LIBDIR} \ MODTCL_LIB=${MODTCL_LIB} \ - MODTK_LIB=${MODTK_LIB} + MODTK_LIB=${MODTK_LIB} \ + USE_WXNEEDED=${USE_WXNEEDED} .if empty(CONFIGURE_STYLE) CONFIGURE_STYLE= cmake Index: patches/patch-Modules_Platform_OpenBSD_cmake === RCS file: /cvs/ports/devel/cmake/patches/patch-Modules_Platform_OpenBSD_cmake,v retrieving revision 1.14 diff -u -p -r1.14 patch-Modules_Platform_OpenBSD_cmake --- patches/patch-Modules_Platform_OpenBSD_cmake31 Dec 2014 11:06:02 - 1.14 +++ patches/patch-Modules_Platform_OpenBSD_cmake20 Aug 2016 18:20:50 - @@ -1,6 +1,6 @@ $OpenBSD: patch-Modules_Platform_OpenBSD_cmake,v 1.14 2014/12/31 11:06:02 dcoppa Exp $ Modules/Platform/OpenBSD.cmake.origMon Dec 15 14:07:43 2014 -+++ Modules/Platform/OpenBSD.cmake Thu Dec 18 05:37:12 2014 +--- Modules/Platform/OpenBSD.cmake.origFri Jul 22 15:50:22 2016 Modules/Platform/OpenBSD.cmake Sat Aug 20 11:12:08 2016 @@ -1,5 +1,7 @@ include(Platform/NetBSD) @@ -9,3 +9,14 @@ $OpenBSD: patch-Modules_Platform_OpenBSD # On OpenBSD, the compile time linker does not share it's configuration with # the runtime linker. This will extract the library search paths from the # system's ld.so.hints file which will allow CMake to set the appropriate +@@ -23,6 +25,10 @@ set_property(GLOBAL PROPERTY FIND_LIBRARY_USE_LIB64_PA + # OpenBSD policy requires that shared libraries be installed without + # executable permission. + set(CMAKE_INSTALL_SO_NO_EXE 1) ++ ++if("$ENV{USE_WXNEEDED}" STREQUAL "Yes") ++ set(CMAKE_EXE_LINKER_FLAGS_INIT "-Wl,-z,wxneeded") ++endif() + + if($ENV{LOCALBASE}) + set(OPENBSD_LOCALBASE $ENV{LOCALBASE}) ---8<--- Index: Makefile === RCS file: /cvs/ports/mail/trojita/Makefile,v retrieving revision 1.15 diff -u -p -r1.15 Makefile --- Makefile19 Aug 2016 10:15:24 - 1.15 +++ Makefile20 Aug 2016 18:21:52 - @@ -1,7 +1,5 @@ # $OpenBSD: Makefile,v 1.15 2016/08/19 10:15:24 sthen Exp $ -# ld wrapper doesn't actually fix build, but setting USE_WXNEEDED for -# documentation purposes. see CONFIGURE_ARGS. USE_WXNEEDED = Yes # QtWebKit COMMENT = fast Qt IMAP e-mail client @@ -36,8 +34,7 @@ CONFIGURE_ENV += QTDIR="${MODQT_LIBDIR}" CONFIGURE_ARGS += -DWITH_RAGEL=OFF \ -DWITH_QT5=OFF \ -DWITH_QTKEYCHAIN_PLUGIN=OFF \ - -DGIT_EXECUTABLE="" \ - -DCMAKE_EXE_LINKER_FLAGS="-Wl,-z,wxneeded" + -DGIT_EXECUTABLE="" TEST_IS_INTERACTIVE = X11
Re: www/otter-browser 0.9.11-dev129 => 0.9.11-dev135 + wxneeded + printf (null) patch
On 2016/08/20 14:19, Adam Wolk wrote: > PS. I tried switching to USE_WXNEEDED but it seems to have no effect on cmake. USE_WXNEEDED does work with cmake, but it's not working with gcc4 because it hardcodes /usr/bin/ld rather than searching the path (and removing the path from --with-ld in the gcc4 port's Makefile results in autoconf failing there). > +CONFIGURE_ENV += LDFLAGS="-Wl,-z,wxneeded" > +# USE_WXNEEDED = Yes Please would you uncomment the USE_WXNEEDED line - even though it doesn't work for gcc4, it's useful for annotation in sqlports. Otherwise ok.
Re: wxneeded for textproc/wkhtmltopdf
On 2016/08/20 17:21, Jeremie Courreges-Anglas wrote: > Frank Groeneveld writes: > > > Because of the bundled webkit, attached patch needs to be applied to > > textproc/wkhtmltopdf to be usable again. Could somebody commit this for > > me? > > Committed, bt with USE_WXNEEDED at the top, which seems to be the > pattern used elsewhere. Thanks jca - yes I had been doing that on purpose to make it obvious.
Re: wxneeded for textproc/wkhtmltopdf
Frank Groeneveld writes: > Because of the bundled webkit, attached patch needs to be applied to > textproc/wkhtmltopdf to be usable again. Could somebody commit this for > me? Committed, bt with USE_WXNEEDED at the top, which seems to be the pattern used elsewhere. Thanks, -- jca | PGP : 0x1524E7EE / 5135 92C1 AD36 5293 2BDF DDCC 0DFA 74AE 1524 E7EE
wxneeded for textproc/wkhtmltopdf
Because of the bundled webkit, attached patch needs to be applied to textproc/wkhtmltopdf to be usable again. Could somebody commit this for me? Cheers. Frank ? wkhtmltopdf-0.12.3.2p0.patch Index: Makefile === RCS file: /cvs/ports/textproc/wkhtmltopdf/Makefile,v retrieving revision 1.9 diff -u -p -r1.9 Makefile --- Makefile10 Aug 2016 20:53:27 - 1.9 +++ Makefile20 Aug 2016 14:56:48 - @@ -8,6 +8,7 @@ COMMENT = convert HTML to PDF using Web GH_ACCOUNT = wkhtmltopdf GH_PROJECT = wkhtmltopdf GH_TAGNAME = 0.12.3.2 +REVISION = 0 QT_COMMIT =8dae19a82231e87366d81c683bafcb025aea9c1d MASTER_SITES0 =https://github.com/wkhtmltopdf/qt/archive/ DISTFILES =${DISTNAME}.tar.gz \ @@ -35,6 +36,7 @@ LIB_DEPENDS = graphics/png \ textproc/icu4c USE_GMAKE =Yes +USE_WXNEEDED = Yes MAKE_FLAGS = LIBwkhtmltox_VERSION=${LIBwkhtmltox_VERSION} MAKE_ENV +=WRKBUILD=${WRKBUILD} FAKE_FLAGS = INSTALL_ROOT=${WRKINST}${TRUEPREFIX}
Re: Brave
On Sat, Aug 20, 2016 at 09:50:10AM -0500, Jordon wrote: > > > On Aug 20, 2016, at 8:19 AM, Adam Wolk wrote: > > > > On Fri, Aug 19, 2016 at 01:27:47PM -0500, Jordon wrote: > >> I was kind of surprised to see that there was no port for the Brave web > >> browser so I figured this would be a chance to take a stab at making a > >> port. > >> > >> The first thing I discovered about Brave is that it requires nodejs 6.1 or > >> greater and OpenBSD only has a port of the 4.x branch. Ok, lets start > >> there??? > >> > >> To build nodejs 6.4 (current latest), i need python 2.6 or 2.7. I > >> installed python 2.7.11 from pkg but running ./config for nodejs still > >> fails because it can???t find python. I think it is because there is no > >> ???python", but ???python2.7???. Perhaps a symlink is needed? > >> > > > > When you installed python it told you: > > If you want to use this package as your default system python, as root > > create symbolic links like so (overwriting any previous default): > > ln -sf /usr/local/bin/python2.7 /usr/local/bin/python > > ln -sf /usr/local/bin/python2.7-2to3 /usr/local/bin/2to3 > > ln -sf /usr/local/bin/python2.7-config /usr/local/bin/python-config > > ln -sf /usr/local/bin/pydoc2.7 /usr/local/bin/pydoc > > Yeah, i figured I forgot to read that output. Sorry... > > > > you can see that message yourself by running: pkg_info -q python > > Good to know, because I will probably make that mistake again! > > > > This wouldn't matter if you started porting actually using the ports > > infrastructure versus doing it out of tree. > > This is interesting. I thought it would be better to build outside of the > ports infrastructure first to prove it could work, and then ???port??? it to > the ports system. > > > >> So??? what is my next step? Should I contact the maintainer of node 4.3.0 > >> and request node 6.x? Should I attempt it myself? Is someone already > >> working on Brave and I am wasting my time? (although the experience of > >> doing this may still be worth it..) > >> > > > > You next step should be to step back & read the ports guide. > > > > https://www.openbsd.org/faq/ports/guide.html > > I did skim this a bit first, but yeah, I do need to give it a more thorough > reading. Also, in all fairness, the first line of the guide is "So you've > just compiled your favorite software package on your OpenBSD machine and you > want to share your effort by turning it into a standard port.???, which could > be interpreted to mean that building outside of the ports system is a > prerequisite! :) > > Seriously though, thanks for the feedback. I have some stuff going on today > but I will dig into this more in the near future. > > jordon > You may end-up building out of tree but you picked a pretty big project for a first port. The amount of stuff the ports system will configure & set for you (like proper GCC version, dependencies, env variables etc) is significant. In the worst case scenario it will make it easier to build out of tree as you will be able to look at what the ports system did to move that one step forward. Regards, Adam
Re: simple wxneeded setting from port Makefiles
On Sat, Aug 20, 2016 at 03:25:25PM +0200, Jeremie Courreges-Anglas wrote: > Pretty much any time you see this, you can assume that the kernel > returned ENOEXEC. This can happen with an invalid executable, > a partition mounted "noexec", or... a program annotated with wxneeded > but placed on a partition that isn't mounted with "wxallowed". Check > the partition that holds WRKOBJDIR (probably /home here). Great tip, thank you very much. The diff is now just a USE_WXNEEDED addition :-) Frank
Re: Brave
> On Aug 20, 2016, at 8:19 AM, Adam Wolk wrote: > > On Fri, Aug 19, 2016 at 01:27:47PM -0500, Jordon wrote: >> I was kind of surprised to see that there was no port for the Brave web >> browser so I figured this would be a chance to take a stab at making a port. >> >> The first thing I discovered about Brave is that it requires nodejs 6.1 or >> greater and OpenBSD only has a port of the 4.x branch. Ok, lets start >> there??? >> >> To build nodejs 6.4 (current latest), i need python 2.6 or 2.7. I installed >> python 2.7.11 from pkg but running ./config for nodejs still fails because >> it can???t find python. I think it is because there is no ???python", but >> ???python2.7???. Perhaps a symlink is needed? >> > > When you installed python it told you: > If you want to use this package as your default system python, as root > create symbolic links like so (overwriting any previous default): > ln -sf /usr/local/bin/python2.7 /usr/local/bin/python > ln -sf /usr/local/bin/python2.7-2to3 /usr/local/bin/2to3 > ln -sf /usr/local/bin/python2.7-config /usr/local/bin/python-config > ln -sf /usr/local/bin/pydoc2.7 /usr/local/bin/pydoc Yeah, i figured I forgot to read that output. Sorry... > you can see that message yourself by running: pkg_info -q python Good to know, because I will probably make that mistake again! > This wouldn't matter if you started porting actually using the ports > infrastructure versus doing it out of tree. This is interesting. I thought it would be better to build outside of the ports infrastructure first to prove it could work, and then ‘port’ it to the ports system. >> So??? what is my next step? Should I contact the maintainer of node 4.3.0 >> and request node 6.x? Should I attempt it myself? Is someone already >> working on Brave and I am wasting my time? (although the experience of >> doing this may still be worth it..) >> > > You next step should be to step back & read the ports guide. > > https://www.openbsd.org/faq/ports/guide.html I did skim this a bit first, but yeah, I do need to give it a more thorough reading. Also, in all fairness, the first line of the guide is "So you've just compiled your favorite software package on your OpenBSD machine and you want to share your effort by turning it into a standard port.”, which could be interpreted to mean that building outside of the ports system is a prerequisite! :) Seriously though, thanks for the feedback. I have some stuff going on today but I will dig into this more in the near future. jordon
Re: simple wxneeded setting from port Makefiles
Frank Groeneveld writes: > On Thu, Aug 18, 2016 at 11:49:04PM +0100, Stuart Henderson wrote: >> I think that would be prefereable, if it works. > > Thanks for making this simpler for us! I tried applying it to > wkhtmltopdf (because of the bundled Webkit) and think I've found a > build that doesn't work with USE_WXNEEDED (haven't gotten it to > work in other ways either). Attached the output. It seems the > qmake binary get corrupted somehow. Any hints on how to debug this? [...] > mkdir /home/pobj/wkhtmltopdf-0.12.3.2/build-amd64/qt/src/tools > mkdir /home/pobj/wkhtmltopdf-0.12.3.2/build-amd64/qt/src/tools/uic > Creating qmake. Please wait... > qbytearray.o: In function `qstrcpy(char*, char const*)': > /home/pobj/wkhtmltopdf-0.12.3.2/wkhtmltopdf-0.12.3.2/qt/src/corelib/tools/qbytearray.cpp:143: > warning: warning: strcpy() is almost always misused, please use strlcpy() > option.o: In function `qmake_version()': > /home/pobj/wkhtmltopdf-0.12.3.2/build-amd64/qt/qmake/option.cpp:695: warning: > warning: sprintf() is often misused, please use snprintf() > qglobal.o: In function `qrand()': > /home/pobj/wkhtmltopdf-0.12.3.2/wkhtmltopdf-0.12.3.2/qt/src/corelib/global/qglobal.cpp:2810: > warning: warning: rand() may return deterministic values, is that what you > want? > /home/pobj/wkhtmltopdf-0.12.3.2/build-amd64/qt/bin/qmake[1]: : not found > /home/pobj/wkhtmltopdf-0.12.3.2/build-amd64/qt/bin/qmake[1]: ELF: not found > /home/pobj/wkhtmltopdf-0.12.3.2/build-amd64/qt/bin/qmake[3]: ÿÿ%BXh: not found > /home/pobj/wkhtmltopdf-0.12.3.2/build-amd64/qt/bin/qmake[3]: @ÿÿ%BXh0ÿÿ%BXh > : not found > /home/pobj/wkhtmltopdf-0.12.3.2/build-amd64/qt/bin/qmake[3]: > CXhéÿÿ%CXhpÿÿ%BXhþÿ%rBXhéPÿÿ%jBXhé@ÿÿ%bBXhé0ÿÿ%ZBXhé: > not found > /home/pobj/wkhtmltopdf-0.12.3.2/build-amd64/qt/bin/qmake[5]: syntax error: > `(' unexpected Pretty much any time you see this, you can assume that the kernel returned ENOEXEC. This can happen with an invalid executable, a partition mounted "noexec", or... a program annotated with wxneeded but placed on a partition that isn't mounted with "wxallowed". Check the partition that holds WRKOBJDIR (probably /home here). -- jca | PGP : 0x1524E7EE / 5135 92C1 AD36 5293 2BDF DDCC 0DFA 74AE 1524 E7EE
Re: Brave
On Fri, Aug 19, 2016 at 01:27:47PM -0500, Jordon wrote: > I was kind of surprised to see that there was no port for the Brave web > browser so I figured this would be a chance to take a stab at making a port. > > The first thing I discovered about Brave is that it requires nodejs 6.1 or > greater and OpenBSD only has a port of the 4.x branch. Ok, lets start > there??? > > To build nodejs 6.4 (current latest), i need python 2.6 or 2.7. I installed > python 2.7.11 from pkg but running ./config for nodejs still fails because it > can???t find python. I think it is because there is no ???python", but > ???python2.7???. Perhaps a symlink is needed? > When you installed python it told you: If you want to use this package as your default system python, as root create symbolic links like so (overwriting any previous default): ln -sf /usr/local/bin/python2.7 /usr/local/bin/python ln -sf /usr/local/bin/python2.7-2to3 /usr/local/bin/2to3 ln -sf /usr/local/bin/python2.7-config /usr/local/bin/python-config ln -sf /usr/local/bin/pydoc2.7 /usr/local/bin/pydoc you can see that message yourself by running: pkg_info -q python This wouldn't matter if you started porting actually using the ports infrastructure versus doing it out of tree. > So??? what is my next step? Should I contact the maintainer of node 4.3.0 > and request node 6.x? Should I attempt it myself? Is someone already > working on Brave and I am wasting my time? (although the experience of doing > this may still be worth it..) > You next step should be to step back & read the ports guide. https://www.openbsd.org/faq/ports/guide.html Regards, Adam
Re: www/otter-browser 0.9.11-dev129 => 0.9.11-dev135 + wxneeded + printf (null) patch
On Tue, Aug 16, 2016 at 10:52:07PM +0200, Adam Wolk wrote: > On Mon, Aug 01, 2016 at 02:43:49AM +0200, Adam Wolk wrote: > > On Mon, Jul 18, 2016 at 12:58:17AM +0200, Adam Wolk wrote: > > > On Wed, Jul 06, 2016 at 02:28:08PM +0200, Adam Wolk wrote: > > > > Hi ports@ > > > > > > > > Trivial bump for otter-browser latest weekly. > > > > > > > > OK's? feedback? > > > > > > > > > > Here is a bump for the latest weekly. Obviously waiting for port unlock > > > but open > > > to testing before that happens. > > > > > > > Ping plus a bump to weekly 134 > > > Ping plus a bump to weekly 135 and CONFIGURE_ENV accounting for the wxneeded > by the qt5 JIT > in otter-browser. > Including a patch to handle fprintf calls trying to print NULL using a %s format specifier. Reported by bentley@, thx. PS. I tried switching to USE_WXNEEDED but it seems to have no effect on cmake. Upstream fix for fprintf handling: https://github.com/OtterBrowser/otter-browser/issues/1173 The fix: https://github.com/OtterBrowser/otter-browser/commit/9907933100b07b5d6754ff119c4abc5c87734ff0 The patch included in this diff is the same as the upstream change. Feedback? OK's? :) Index: Makefile === RCS file: /cvs/ports/www/otter-browser/Makefile,v retrieving revision 1.14 diff -u -p -r1.14 Makefile --- Makefile21 Jun 2016 09:03:45 - 1.14 +++ Makefile20 Aug 2016 12:08:26 - @@ -2,7 +2,7 @@ COMMENT = browser aiming to recreate classic Opera (12.x) UI using Qt5 MASTER_SITES = ${MASTER_SITE_SOURCEFORGE:=otter-browser/} -DISTNAME = otter-browser-0.9.11-dev129 +DISTNAME = otter-browser-0.9.11-dev135 PKGNAME = ${DISTNAME:S/-dev/pre/} EXTRACT_SUFX = .tar.bz2 @@ -27,6 +27,9 @@ LIB_DEPENDS = security/libgcrypt \ textproc/hunspell MODULES = devel/cmake x11/qt5 + +CONFIGURE_ENV += LDFLAGS="-Wl,-z,wxneeded" +# USE_WXNEEDED = Yes NO_TEST = Yes Index: distinfo === RCS file: /cvs/ports/www/otter-browser/distinfo,v retrieving revision 1.11 diff -u -p -r1.11 distinfo --- distinfo21 Jun 2016 09:03:45 - 1.11 +++ distinfo20 Aug 2016 12:08:26 - @@ -1,2 +1,2 @@ -SHA256 (otter-browser-0.9.11-dev129.tar.bz2) = mwfq2hjHHJgjT9Jrv2jCD6R1tJoSEFxp37DWQe2GluA= -SIZE (otter-browser-0.9.11-dev129.tar.bz2) = 2310570 +SHA256 (otter-browser-0.9.11-dev135.tar.bz2) = T6tHeUbCRjU1w0hb059Tn+mJyF1AsFBUZ67aocL4cDU= +SIZE (otter-browser-0.9.11-dev135.tar.bz2) = 2337334 Index: patches/patch-src_main_cpp === RCS file: patches/patch-src_main_cpp diff -N patches/patch-src_main_cpp --- /dev/null 1 Jan 1970 00:00:00 - +++ patches/patch-src_main_cpp 20 Aug 2016 12:08:26 - @@ -0,0 +1,54 @@ +$OpenBSD$ +--- src/main.cpp.orig Wed Jun 1 20:47:08 2016 src/main.cpp Sat Aug 20 13:45:49 2016 +@@ -28,6 +28,7 @@ + + using namespace Otter; + ++#if QT_VERSION >= 0x050400 + void otterMessageHander(QtMsgType type, const QMessageLogContext &context, const QString &message) + { + if (message.trimmed().startsWith(QLatin1String("OpenType support missing")) || message.startsWith(QLatin1String("libpng warning: iCCP:")) || message.startsWith(QLatin1String("OpenType support missing for script")) || message.startsWith(QLatin1String("QNetworkReplyImplPrivate::error: Internal problem, this method must only be called once")) || message.startsWith(QLatin1String("QBasicTimer::start: QBasicTimer can only be used with threads started with QThread")) || message.startsWith(QLatin1String("QNetworkReplyImpl::_q_startOperation was called more than once"))) +@@ -35,35 +36,20 @@ void otterMessageHander(QtMsgType type, const QMessage + return; + } + +- switch (type) +- { +- case QtDebugMsg: +- fprintf(stderr, "Debug: %s (%s:%u, %s)\n", message.toLocal8Bit().constData(), context.file, context.line, context.function); ++ fputs(qFormatLogMessage(type, context, message).toLocal8Bit().constData(), stderr); + +- break; +-#if QT_VERSION >= 0x050500 +- case QtInfoMsg: +- fprintf(stderr, "Info: %s (%s:%u, %s)\n", message.toLocal8Bit().constData(), context.file, context.line, context.function); +- +- break; +-#endif +- case QtWarningMsg: +- fprintf(stderr, "Warning: %s (%s:%u, %s)\n", message.toLocal8Bit().constData(), context.file, context.line, context.function); +- +- break; +- case QtCriticalMsg: +- fprintf(stderr, "Critical: %s (%s:%u, %s)\n", message.toLocal8Bit().constData(), context.file, context.line, context.function); +- +- break; +-
Re: fortran.port.mk diff
On Fri, Aug 19, 2016 at 06:55:13PM +0500, Alexandr Shadchin wrote: > On Fri, Aug 19, 2016 at 01:31:46PM +0200, David Coppa wrote: > > On Fri, 19 Aug 2016, David Coppa wrote: > > > > > For the lang/gfortran removal... > > > > > > OK? > > > > No way... Here's a working one: > > > > May be also cleanup gcc 2.95 ? Does port-modules(5) also need a tweak after these changes? > Index: fortran.port.mk > === > RCS file: /cvs/ports/infrastructure/mk/fortran.port.mk,v > retrieving revision 1.12 > diff -u -p -r1.12 fortran.port.mk > --- fortran.port.mk 20 Nov 2010 19:57:30 - 1.12 > +++ fortran.port.mk 19 Aug 2016 13:50:57 - > @@ -7,35 +7,18 @@ ERRORS += "Fatal: need to specify MODFOR > .endif > > .if ${MODFORTRAN_COMPILER:L} == "g77" > -. if ${COMPILER_VERSION:L:Mgcc[34]*} > -_MODFORTRAN_LIB_DEPENDS_G77 = devel/libf2c > -_MODFORTRAN_WANTLIB_G77 = g2c > -_MODFORTRAN_BUILD_DEPENDS_G77 = lang/g77 devel/libf2c > -. else > -_MODFORTRAN_LIB_DEPENDS_G77 = devel/libf2c-old > -_MODFORTRAN_WANTLIB_G77 += g2c > -_MODFORTRAN_BUILD_DEPENDS_G77 = lang/g77-old devel/libf2c-old > -. endif > -MODFORTRAN_LIB_DEPENDS += ${_MODFORTRAN_LIB_DEPENDS_G77} > -MODFORTRAN_WANTLIB += ${_MODFORTRAN_WANTLIB_G77} > -MODFORTRAN_BUILD_DEPENDS += ${_MODFORTRAN_BUILD_DEPENDS_G77} > +MODFORTRAN_LIB_DEPENDS += devel/libf2c > +MODFORTRAN_WANTLIB += g2c > +MODFORTRAN_BUILD_DEPENDS += lang/g77 devel/libf2c > MODFORTRAN_post-patch = \ > if test -e /usr/bin/g77 -o -e /usr/bin/f77; then \ > echo "Error: remove old fortran compiler /usr/bin/f77 /usr/bin/g77"; \ > exit 1; \ > fi > .elif ${MODFORTRAN_COMPILER:L} == "gfortran" > -. if ${COMPILER_VERSION:L:Mgcc4} > -_MODFORTRAN_LIB_DEPENDS_GFORTRAN = lang/gfortran,-lib > -_MODFORTRAN_WANTLIB_GFORTRAN = gfortran > -_MODFORTRAN_BUILD_DEPENDS_GFORTRAN = lang/gfortran > -. else > MODULES += gcc4 > +MODGCC4_ARCHS ?= * > MODGCC4_LANGS += fortran > -. endif > -MODFORTRAN_LIB_DEPENDS += ${_MODFORTRAN_LIB_DEPENDS_GFORTRAN} > -MODFORTRAN_WANTLIB += ${_MODFORTRAN_WANTLIB_GFORTRAN} > -MODFORTRAN_BUILD_DEPENDS += ${_MODFORTRAN_BUILD_DEPENDS_GFORTRAN} > .else > ERRORS += "Fatal: MODFORTRAN_COMPILER must be one of: g77 gfortran" > .endif > -- Antoine