Re: p5-Plack: update RUN_DEPENDS for FCGI
looks ok to me. tested in amd64. Thanks. On Fri, Dec 19, 2014 at 11:13 AM, Todd C. Miller wrote: > This updates RUN_DEPENDS to include modules needed for FCGI support. > I bumped the REVISION but if that's not right please let me know. > > Comments or OKs? > > - todd > > Index: www/p5-Plack/Makefile > === > RCS file: /cvs/ports/www/p5-Plack/Makefile,v > retrieving revision 1.6 > diff -u -r1.6 Makefile > --- www/p5-Plack/Makefile 10 Sep 2014 06:35:19 - 1.6 > +++ www/p5-Plack/Makefile 19 Dec 2014 17:12:21 - > @@ -3,7 +3,7 @@ > COMMENT = interface between perl web frameworks and web servers > > MODULES = cpan > -REVISION = 1 > +REVISION = 2 > DISTNAME = Plack-0.9980 > CATEGORIES = www > MAINTAINER = Abel Abraham Camarillo Ojeda > @@ -14,14 +14,18 @@ > BUILD_DEPENDS =${RUN_DEPENDS} > RUN_DEPENDS = devel/p5-Devel-StackTrace>=1.23 \ > devel/p5-Devel-StackTrace-AsHTML>=0.11 \ > + devel/p5-File-pushd \ > devel/p5-File-ShareDir>=1.00 \ > devel/p5-Hash-MultiValue>=0.05 \ > devel/p5-Test-Requires \ > devel/p5-Try-Tiny \ > net/p5-Test-TCP>=0.11 \ > sysutils/p5-Filesys-Notify-Simple \ > + www/fcgi \ > www/p5-libwww>=5.814 \ > www/p5-libapreq \ > + www/p5-CGI-Compile \ > + www/p5-FCGI-ProcManager \ > www/p5-HTTP-Body>=1.06 \ > www/p5-URI>=1.36 >
sshguard dies on boot + workaround
$ uname -mrsv OpenBSD 5.6 GENERIC#274 i386 # reboot /var/log/authlog: Dec 25 14:04:11 palm sshd[23898]: Server listening on 0.0.0.0 port 22. Dec 25 14:04:16 palm sshguard[18164]: Started successfully [(a,p,s)=(30, 3600, 6000)], now ready to scan. /etc/rc exits: Dec 25 14:04:41 palm sshguard[18164]: Got exit signal, flushing blocked addresses and exiting... root cronjob: @reboot sleep 5; /etc/rc.d/sshguard check || /etc/rc.d/sshguard start Dec 25 14:04:48 palm sshguard[13329]: Started successfully [(a,p,s)=(30, 3600, 6000)], now ready to scan. Dinnae ken if it dies due to the parent process (rc) exiting: /etc/rc reboot output: ... .. setting tty flags keyboard.encoding -> uk display.vblank -> on display.screen_off -> 30 display.msact -> off display.kbdact -> on display.outact -> off pf enabled starting network starting early daemons: syslogd pflogd nsd unbound ntpd. starting RPC daemons:. savecore: no core dump checking quotas: done. clearing /tmp kern.securelevel: 0 -> 1 creating runtime link editor directory cache. preserving editor files. starting network daemons: sshd dhcpd nginx spamd spamlogd. starting package daemons: sshguard greyscanner postfix viagrad. starting local daemons: cron. sshguard up Thu Dec 25 14:04:41 GMT 2014 <<< dies now $ tail -n 6 /etc/rc echo -n 'starting local daemons:' start_daemon apmd sensorsd hotplugd watchdogd cron wsmoused xdm echo '.' /etc/rc.d/sshguard check && print 'sshguard up' date exit 0 Whatever,... the cronjob hack gets it back up. Then my brain quits. Ta, -- The Story of Your Enslavement http://www.youtube.com/watch?v=Xbp6umQT58A&feature=related
Re: NEW: devel/greg
greg.tgz Description: proposed new port: devel/greg attila writes: > Hi ports@, > > Attached is a new port, devel/greg. Greg is a PEG-based parser > generator. Sorry, I forgot to attach the port in my first email. Here it is. > > pkg/DESCR: > greg is a recursive-descent parser generator based on Ian Piumarta's > peg/leg. It implements a formalism called Parsing Expression Grammars. > --- > Works well on i386. I'm working on getting all of the patches > upstream. Another port I submitted recently, textproc/multimarkdown, > depends on devel/greg to build. > > Thanks to bcallah@ for working with me on this. > > Testing and comments please! > > Pax, --A -- att...@stalphonsos.com | http://trac.haqistan.net/~attila
pulseaudio useradd warning on port install
Hi ports@ While installing the qt5 port I noticed a warning from the pulseaudio port about a missing /var/run/pulse. Is this a normal situation or should the port provide -m # pkg_add qt5 quirks-2.46 signed on 2014-12-24T16:44:44Z qt5-5.3.2p0:gperf-3.0.4p0: ok qt5-5.3.2p0:openal-1.15.1p0v0: ok qt5-5.3.2p0:mozjs17-17.0p1: ok qt5-5.3.2p0:polkit-0.112p7: ok qt5-5.3.2p0:consolekit-0.4.6p11: ok qt5-5.3.2p0:json-c-0.12: ok useradd: Warning: home directory `/var/run/pulse' doesn't exist, and -m was not specified qt5-5.3.2p0:pulseaudio-5.0p2: ok qt5-5.3.2p0:libxkbcommon-0.5.0: ok qt5-5.3.2p0:iodbc-3.52.9p0: ok qt5-5.3.2p0 (extracting)|* | 66% Regards, -- Adam Wolk adam.w...@koparo.com
Re: PostgreSQL 9.4.0
On Wed, Dec 24, 2014 at 12:20:14PM +0100, Pierre-Emmanuel André wrote: > On Wed, Dec 24, 2014 at 10:34:33AM +, Stuart Henderson wrote: > > On 2014/12/23 17:08, Pierre-Emmanuel André wrote: > > > Hi, > > > > > > Here is a diff for the latest version of PostgreSQL (9.4.0). > > > + Announcement: http://www.postgresql.org/about/news/1557/ > > > + Release infos: > > > http://www.postgresql.org/docs/9.4/static/release-9-4.html > > > > > > Tested on @amd64. > > > Comment, ok ? > > > > Has this been through a bulk build yet? If not I can add it to my next > > test build. > > > > Landry@ is doing a bulk build with the update. So postgresql-plv8 failed (but sthen@ has a diff to update/fix it), postgresql-pllua fails but bumping the requirement for RDEP fixes it: -RUN_DEPENDS = postgresql-server->=9.3,<9.4:databases/postgresql,-server +RUN_DEPENDS = postgresql-server->=9.4,<9.5:databases/postgresql,-server and databases/pg_statsinfo fails to build, even if i bump the RDEP requirement so probably needs update/api fix. other than that looks ok to me. Landry
Re: New diff for CMake-3.1 to test (Was: UPDATE: freetype-2.5.4)
On Tue, Dec 23, 2014 at 05:01:43PM +0100, David Coppa wrote: > On Fri, Dec 19, 2014 at 9:17 PM, Landry Breuil wrote: > > On Thu, Dec 18, 2014 at 09:07:40PM +0100, Landry Breuil wrote: > >> On Thu, Dec 18, 2014 at 11:18:59AM +0100, David Coppa wrote: > >> > On Mon, Dec 15, 2014 at 9:55 PM, Matthieu Herrb > >> > wrote: > >> > > On Mon, Dec 15, 2014 at 04:32:47AM -0700, David Coppa wrote: > >> > >> > >> > >> Hi all! > >> > >> > >> > >> An update to freetype-2.5.4, released on 2014-12-06. > >> > >> > >> > >> This release also contains a fix for vulnerability CVE-2014-2240 > >> > >> in the CFF driver. > >> > >> > >> > >> abi-compliance-checker reported 2.5.4 as being incompatible with > >> > >> 2.5.3, thus I've bumped shlib's major. > >> > >> > >> > >> Tested in a full xenocara build. > >> > >> It would be nice to have it in a ports bulk build... > >> > >> > >> > > Given that it doesn't break ports, ok matthieu@ > >> > > >> > So, please put it in a bulk build... > >> > >> Starting a cmake/freetype bulk in a few. > > > > Not many failures, but without x265 that takes a good chunk of the tree.. > > Here's a new revision of the cmake-3.1's diff. > > Kudos to Brad King, responsive and very fast! > http://public.kitware.com/Bug/view.php?id=15325 > > Also, since we're here, can you put the update to ninja-1.5.3 into the bulk? So, failures this time: devel/ipython: Could not import extension matplotlib.sphinxext.mathmpl (exception: No module named six) (that one looks totally unrelated to cmake, probably some py-six changes around..) graphics/asymptote: (that one also looks unrelated, and its always been failing randomly) ../base/plain_shipout.asy: 82.10: runtime: Stack overflow or segmentation fault: rerun with -nothreads Abort trap (core dumped) Makefile:41: recipe for target 'secondaryaxis.pdf' failed x11/kde4/kscd + net/ktorrent-kde4 + x11/kde4/marble share th same failure: CMake Error at /usr/local/share/cmake/Modules/FindPkgConfig.cmake:112 (elseif): given arguments: "VERSION_LESS" "3.1" Unknown arguments specified Call Stack (most recent call first): /usr/local/share/cmake/Modules/FindPkgConfig.cmake:501 (_pkgconfig_parse_options) /usr/local/share/cmake/Modules/FindLCMS.cmake:24 (pkg_check_modules) /usr/local/share/cmake/Modules/Qt4ConfigDependentSettings.cmake:37 (include) /usr/local/share/cmake/Modules/FindQt4.cmake:1198 (include) CMakeLists.txt:5 (find_package) x11/kde/kdeedu3: kalzium.all_cpp.o(.text+0x3a5f): In function `DetailedInfoDlg::addHTMLTab(QString const&, QString const&, QString const&)': : undefined reference to `KHTMLPart::KHTMLPart(QWidget*, char const*, QObject*, char const*, KHTMLPart::GUIProfile)' collect2: ld returned 1 exit status Makefile:598: recipe for target 'kalzium' failed (an unrelated fluke ?) editors/nvi: Scanning dependencies of target headers /usr/bin/make -f CMakeFiles/headers.dir/build.make CMakeFiles/headers.dir/build make: don't know how to make /usr/obj/mfs/nvi-2.1.2/nvi-2.1.2/cl/extern.h (prerequisite of: CMakeFiles/headers) Stop in . graphics/blender: /usr/obj/mfs/blender-2.72b/blender-2.72b/source/blender/blenfont/intern/blf_glyph.c: In function 'blf_ glyph_add': /usr/obj/mfs/blender-2.72b/blender-2.72b/source/blender/blenfont/intern/blf_glyph.c:265:19: error: con version to 'int' from 'unsigned int' may change the sign of the result [-Werror=sign-conversion] g->width = bitmap.width; ^ /usr/obj/mfs/blender-2.72b/blender-2.72b/source/blender/blenfont/intern/blf_glyph.c:266:20: error: conversion to 'int' from 'unsigned int' may change the sign of the result [-Werror=sign-conversion] g->height = bitmap.rows; ^ cc1: some warnings being treated as errors (cmake puts more warnings or -Werror ?) Landry
NEW port net/czmq
Follow a new port net/czmq. It is a high-level C bindings for zeromq. The port is for version 2.2 of czmq. When the new version (3.0) will be released, I will update it. Is this OK? czmq.tgz Description: GNU Zip compressed data
Re: maintainer net/zeromq
Follow the patch, with REVISION Index: Makefile === RCS file: /cvs/ports/net/zeromq/Makefile,v retrieving revision 1.13 diff -u -p -r1.13 Makefile --- Makefile15 Oct 2014 14:28:42 - 1.13 +++ Makefile25 Dec 2014 19:28:44 - @@ -13,6 +13,10 @@ SHARED_LIBS += zmq 4.0 CATEGORIES = net HOMEPAGE = http://zeromq.org/ +MAINTAINER =Rodrigo Mosconi + +REVISION = 1 + # LGPLv3 PERMIT_PACKAGE_CDROM = Yes 2014-12-24 5:46 GMT-02:00 Antoine Jacoutot : > On Wed, Dec 24, 2014 at 12:43:26AM -0200, Rodrigo Mosconi wrote: > > I`m taking the maintainership of net/zeromq > > I added you but you should start looking at how to maintain ports right > away then. > Because you missed a REVISION bump. > > > > Index: Makefile > > === > > RCS file: /cvs/ports/net/zeromq/Makefile,v > > retrieving revision 1.13 > > diff -u -p -r1.13 Makefile > > --- Makefile 15 Oct 2014 14:28:42 - 1.13 > > +++ Makefile 24 Dec 2014 02:41:13 - > > @@ -13,6 +13,8 @@ SHARED_LIBS += zmq 4.0 > > CATEGORIES = net > > HOMEPAGE = http://zeromq.org/ > > > > +MAINTAINER =Rodrigo Mosconi > > + > > # LGPLv3 > > PERMIT_PACKAGE_CDROM = Yes > > > > > -- > Antoine >
Re: New diff for CMake-3.1 to test (Was: UPDATE: freetype-2.5.4)
On Thu, Dec 25, 2014 at 7:46 PM, Landry Breuil wrote: > x11/kde4/kscd + net/ktorrent-kde4 + x11/kde4/marble share th same failure: > > CMake Error at /usr/local/share/cmake/Modules/FindPkgConfig.cmake:112 > (elseif): > given arguments: > > "VERSION_LESS" "3.1" > > Unknown arguments specified > Call Stack (most recent call first): > /usr/local/share/cmake/Modules/FindPkgConfig.cmake:501 > (_pkgconfig_parse_options) > /usr/local/share/cmake/Modules/FindLCMS.cmake:24 (pkg_check_modules) > /usr/local/share/cmake/Modules/Qt4ConfigDependentSettings.cmake:37 (include) > /usr/local/share/cmake/Modules/FindQt4.cmake:1198 (include) > CMakeLists.txt:5 (find_package) > > x11/kde/kdeedu3: > kalzium.all_cpp.o(.text+0x3a5f): In function > `DetailedInfoDlg::addHTMLTab(QString const&, QString const&, QString > const&)': : undefined reference to `KHTMLPart::KHTMLPart(QWidget*, char > const*, > QObject*, char const*, KHTMLPart::GUIProfile)' > collect2: ld returned 1 exit status > Makefile:598: recipe for target 'kalzium' failed > > (an unrelated fluke ?) > > editors/nvi: > Scanning dependencies of target headers > /usr/bin/make -f CMakeFiles/headers.dir/build.make > CMakeFiles/headers.dir/build > make: don't know how to make /usr/obj/mfs/nvi-2.1.2/nvi-2.1.2/cl/extern.h > (prerequisite of: CMakeFiles/headers) > Stop in . I've just now fixed nvi. > graphics/blender: > > /usr/obj/mfs/blender-2.72b/blender-2.72b/source/blender/blenfont/intern/blf_glyph.c: > In function 'blf_ > glyph_add': > /usr/obj/mfs/blender-2.72b/blender-2.72b/source/blender/blenfont/intern/blf_glyph.c:265:19: > error: con > version to 'int' from 'unsigned int' may change the sign of the result > [-Werror=sign-conversion] > g->width = bitmap.width; >^ > /usr/obj/mfs/blender-2.72b/blender-2.72b/source/blender/blenfont/intern/blf_glyph.c:266:20: > error: conversion to 'int' from 'unsigned int' may change the sign of the > result [-Werror=sign-conversion] > g->height = bitmap.rows; > ^ This is caused by freetype-2.5.4 like openimageio and openttd. I'll fix it asap... CCing Vadim about kde-related failures. Ciao! David -- "If you try a few times and give up, you'll never get there. But if you keep at it... There's a lot of problems in the world which can really be solved by applying two or three times the persistence that other people will." -- Stewart Nelson
qt5 testing - working on a port of Otter browser
Hi ports@ I'm CC'ing the KDE porting team as they are marked as the MAINTAINER of the qt5 port in the OpenBSD ports tree. I'm working on a port of the Otter Browser (http://otter-browser.org/) and have hit a bunch of issues that do not seem to be upstream (otter-browser) related. Details: - OpenBSD i386 running snapshots (from 23 Dec 2014) - otter-browser commit e6a48fd71c121f1d486dabcf21e5b48809497c3a from Thu Dec 25 16:28:02 2014 +0100 - ldd output at the end of the email I will provide as many details as necessary - just ask :) 0. CMake fails to find qt5 qt5 is not found unless I prefix it with: CMAKE_PREFIX_PATH=/usr/local/lib/qt5/cmake/ $ cmake ../ -- The C compiler identification is GNU 4.2.1 -- The CXX compiler identification is GNU 4.2.1 -- Check for working C compiler: /usr/bin/cc -- Check for working C compiler: /usr/bin/cc -- works -- Detecting C compiler ABI info -- Detecting C compiler ABI info - done -- Check for working CXX compiler: /usr/bin/c++ -- Check for working CXX compiler: /usr/bin/c++ -- works -- Detecting CXX compiler ABI info -- Detecting CXX compiler ABI info - done CMake Error at CMakeLists.txt:56 (find_package): By not providing "FindQt5Core.cmake" in CMAKE_MODULE_PATH this project has asked CMake to find a package configuration file provided by "Qt5Core", but CMake did not find one. Could not find a package configuration file provided by "Qt5Core" with any of the following names: Qt5CoreConfig.cmake qt5core-config.cmake Add the installation prefix of "Qt5Core" to CMAKE_PREFIX_PATH or set "Qt5Core_DIR" to a directory containing one of the above files. If "Qt5Core" provides a separate development package or SDK, be sure it has been installed. -- Configuring incomplete, errors occurred! See also "/home/mulander/otter-head/otter/build/CMakeFiles/CMakeOutput.log". 1. Otter browser uses CMake to find libraries - packaging forces me to install all qt5 libs Otter Browser only uses: find_package(Qt5Core REQUIRED) find_package(Qt5Network REQUIRED) find_package(Qt5PrintSupport REQUIRED) find_package(Qt5Script REQUIRED) find_package(Qt5Sql REQUIRED) find_package(Qt5Widgets REQUIRED) find_package(Qt5WebKit REQUIRED) find_package(Qt5WebKitWidgets REQUIRED) Yet the build fails unless I have the following installed: $ pkg_info -Q qt5 qt5-5.3.2p0 (installed) qt5-examples-5.3.2 qt5-mysql-5.3.2 (installed) qt5-postgresql-5.3.2 (installed) qt5-sqlite2-5.3.2 (installed) qt5-tds-5.3.2 (installed) In summary - CMake prevents the build unless I have tds, mysql & postgresql qt5 installed even though otter-browser only uses qt5-sqlite2. 2. gcc warnings after compilation Linking CXX executable otter-browser /usr/local/lib/qt5/libQt5Script.so.0.0: warning: random() may return deterministic values, is that what you want? /usr/local/lib/qt5/libQt5WebKit.so.0.0: warning: strcpy() is almost always misused, please use strlcpy() /usr/local/lib/qt5/libQt5Core.so.0.0: warning: rand() may return deterministic values, is that what you want? /usr/local/lib/qt5/libQt5Script.so.0.0: warning: sprintf() is often misused, please use snprintf() /usr/X11R6/lib/libX11.so.16.0: warning: strcat() is almost always misused, please use strlcat() /usr/local/lib/libglib-2.0.so.4200.0: warning: stpcpy() is dangerous; do not use it /usr/local/lib/libglib-2.0.so.4200.0: warning: vsprintf() is often misused, please use vsnprintf() 3. Crash on first run On first run I see the browser window rendered and a second later I hit the following crash $ ./otter-browser ./otter-browser:/usr/local/lib/libestdc++.so.16.0: /usr/lib/libstdc++.so.57.0 : WARNING: symbol(_ZN11__gnu_debug17_S_debug_messagesE) size mismatch, relink your program Warning: QGtkStyle could not resolve GTK. Make sure you have installed the proper libraries. (/usr/obj/ports/qt-everywhere-opensource-src-5.3.2/qt-everywhere-opensource-src-5.3.2/qtbase/src/widgets/styles/qgtkstyle_p.cpp:583, virtual void QGtkStylePrivate::initGtkWidgets() const) otter-browser(20064) in free(): error: bogus pointer (double free?) 0xdfdfdfdf Abort trap (core dumped) 4. Crash on consecutive runs On consecutive runs I see the 'start from scratch/blank page' dialog fine but when I hit open the following crash occurs $ ./otter-browser ./otter-browser:/usr/local/lib/libestdc++.so.16.0: /usr/lib/libstdc++.so.57.0 : WARNING: symbol(_ZN11__gnu_debug17_S_debug_messagesE) size mismatch, relink your program Warning: QGtkStyle could not resolve GTK. Make sure you have installed the proper libraries. (/usr/obj/ports/qt-everywhere-opensource-src-5.3.2/qt-everywhere-opensource-src-5.3.2/qtbase/src/widgets/styles/qgtkstyle_p.cpp:583, virtual void QGtkStylePrivate::initGtkWidgets() const) Warning: QCommonStyle::drawComplexControl: Control 1 not handled (/usr/obj/ports/qt-everywhere-opensource-src-5.3.2/qt-everywhere-opensource-src-5.3.2/qtbase/src/widgets/styles/qcommonstyle.cpp:3755, virtual void QCommonStyle::drawComplexControl(QStyle::ComplexC
Re: maintainer net/zeromq
On Thu, Dec 25, 2014 at 05:35:08PM -0200, Rodrigo Mosconi wrote: > Follow the patch, with REVISION REVISION starts at 0 Landry > Index: Makefile > === > RCS file: /cvs/ports/net/zeromq/Makefile,v > retrieving revision 1.13 > diff -u -p -r1.13 Makefile > --- Makefile15 Oct 2014 14:28:42 - 1.13 > +++ Makefile25 Dec 2014 19:28:44 - > @@ -13,6 +13,10 @@ SHARED_LIBS += zmq 4.0 > CATEGORIES = net > HOMEPAGE = http://zeromq.org/ > > +MAINTAINER =Rodrigo Mosconi > + > +REVISION = 1 > + > # LGPLv3 > PERMIT_PACKAGE_CDROM = Yes > > > 2014-12-24 5:46 GMT-02:00 Antoine Jacoutot : > > > On Wed, Dec 24, 2014 at 12:43:26AM -0200, Rodrigo Mosconi wrote: > > > I`m taking the maintainership of net/zeromq > > > > I added you but you should start looking at how to maintain ports right > > away then. > > Because you missed a REVISION bump. > > > > > > > Index: Makefile > > > === > > > RCS file: /cvs/ports/net/zeromq/Makefile,v > > > retrieving revision 1.13 > > > diff -u -p -r1.13 Makefile > > > --- Makefile 15 Oct 2014 14:28:42 - 1.13 > > > +++ Makefile 24 Dec 2014 02:41:13 - > > > @@ -13,6 +13,8 @@ SHARED_LIBS += zmq 4.0 > > > CATEGORIES = net > > > HOMEPAGE = http://zeromq.org/ > > > > > > +MAINTAINER =Rodrigo Mosconi > > > + > > > # LGPLv3 > > > PERMIT_PACKAGE_CDROM = Yes > > > > > > > > > -- > > Antoine > >
Re: qt5 testing - working on a port of Otter browser
On Thu, Dec 25, 2014, at 09:28 PM, Adam Wolk wrote: > Hi ports@ > > I'm CC'ing the KDE porting team as they are marked as the MAINTAINER of > the qt5 port in the > OpenBSD ports tree. > > I'm working on a port of the Otter Browser (http://otter-browser.org/) > and have hit a bunch of issues that > do not seem to be upstream (otter-browser) related. > > Details: > - OpenBSD i386 running snapshots (from 23 Dec 2014) > - otter-browser commit e6a48fd71c121f1d486dabcf21e5b48809497c3a from > Thu Dec 25 16:28:02 2014 +0100 > - ldd output at the end of the email Forgot to mention. I'm running with Xfce 4.10 - KDE & Gnome is not installed on this system. Regards, Adam
Re: New diff for CMake-3.1 to test (Was: UPDATE: freetype-2.5.4)
25 дек. 2014 г. 22:48 пользователь "David Coppa" написал: > > On Thu, Dec 25, 2014 at 7:46 PM, Landry Breuil wrote: > > > x11/kde4/kscd + net/ktorrent-kde4 + x11/kde4/marble share th same failure: > > > > CMake Error at /usr/local/share/cmake/Modules/FindPkgConfig.cmake:112 > > (elseif): > > given arguments: > > > > "VERSION_LESS" "3.1" > > > > Unknown arguments specified > > Call Stack (most recent call first): > > /usr/local/share/cmake/Modules/FindPkgConfig.cmake:501 (_pkgconfig_parse_options) > > /usr/local/share/cmake/Modules/FindLCMS.cmake:24 (pkg_check_modules) > > /usr/local/share/cmake/Modules/Qt4ConfigDependentSettings.cmake:37 (include) > > /usr/local/share/cmake/Modules/FindQt4.cmake:1198 (include) > > CMakeLists.txt:5 (find_package) > > > > x11/kde/kdeedu3: > > kalzium.all_cpp.o(.text+0x3a5f): In function > > `DetailedInfoDlg::addHTMLTab(QString const&, QString const&, QString > > const&)': : undefined reference to `KHTMLPart::KHTMLPart(QWidget*, char const*, > > QObject*, char const*, KHTMLPart::GUIProfile)' > > collect2: ld returned 1 exit status > > Makefile:598: recipe for target 'kalzium' failed > > > > (an unrelated fluke ?) > > > > editors/nvi: > > Scanning dependencies of target headers > > /usr/bin/make -f CMakeFiles/headers.dir/build.make CMakeFiles/headers.dir/build > > make: don't know how to make /usr/obj/mfs/nvi-2.1.2/nvi-2.1.2/cl/extern.h (prerequisite of: CMakeFiles/headers) > > Stop in . > > I've just now fixed nvi. > > > graphics/blender: > > > > /usr/obj/mfs/blender-2.72b/blender-2.72b/source/blender/blenfont/intern/blf_glyph.c: In function 'blf_ > > glyph_add': > > /usr/obj/mfs/blender-2.72b/blender-2.72b/source/blender/blenfont/intern/blf_glyph.c:265:19: error: con > > version to 'int' from 'unsigned int' may change the sign of the result [-Werror=sign-conversion] > > g->width = bitmap.width; > >^ > > /usr/obj/mfs/blender-2.72b/blender-2.72b/source/blender/blenfont/intern/blf_glyph.c:266:20: error: conversion to 'int' from 'unsigned int' may change the sign of the result [-Werror=sign-conversion] > > g->height = bitmap.rows; > > ^ > > This is caused by freetype-2.5.4 like openimageio and openttd. I'll > fix it asap... > > CCing Vadim about kde-related failures. Already looking there. Actually, there's a KDE 4.14.3 minor update pending, so I have an evil mix of KDE packages here... The problems look like being easily solvable, so there should a patch happen in a few hours. -- Vadim Zhukov
anyone using www/hs-snap*, lang/feldspar/*?
Hi, Is anyone using the www/hs-snap* and/or lang/feldspar/* ports? If not, I'll remove them next sunday, and all hs-ports only those ports depend on. Ciao, Kili
Re: maintainer net/zeromq
with revision=0 2014-12-25 18:32 GMT-02:00 Landry Breuil : > On Thu, Dec 25, 2014 at 05:35:08PM -0200, Rodrigo Mosconi wrote: > > Follow the patch, with REVISION > > REVISION starts at 0 > > Landry > > > Index: Makefile > > === > > RCS file: /cvs/ports/net/zeromq/Makefile,v > > retrieving revision 1.13 > > diff -u -p -r1.13 Makefile > > --- Makefile15 Oct 2014 14:28:42 - 1.13 > > +++ Makefile25 Dec 2014 19:28:44 - > > @@ -13,6 +13,10 @@ SHARED_LIBS += zmq 4.0 > > CATEGORIES = net > > HOMEPAGE = http://zeromq.org/ > > > > +MAINTAINER =Rodrigo Mosconi > > + > > +REVISION = 1 > > + > > # LGPLv3 > > PERMIT_PACKAGE_CDROM = Yes > > > > > > 2014-12-24 5:46 GMT-02:00 Antoine Jacoutot : > > > > > On Wed, Dec 24, 2014 at 12:43:26AM -0200, Rodrigo Mosconi wrote: > > > > I`m taking the maintainership of net/zeromq > > > > > > I added you but you should start looking at how to maintain ports right > > > away then. > > > Because you missed a REVISION bump. > > > > > > > > > > Index: Makefile > > > > === > > > > RCS file: /cvs/ports/net/zeromq/Makefile,v > > > > retrieving revision 1.13 > > > > diff -u -p -r1.13 Makefile > > > > --- Makefile 15 Oct 2014 14:28:42 - 1.13 > > > > +++ Makefile 24 Dec 2014 02:41:13 - > > > > @@ -13,6 +13,8 @@ SHARED_LIBS += zmq 4.0 > > > > CATEGORIES = net > > > > HOMEPAGE = http://zeromq.org/ > > > > > > > > +MAINTAINER =Rodrigo Mosconi > > > > + > > > > # LGPLv3 > > > > PERMIT_PACKAGE_CDROM = Yes > > > > > > > > > > > > > -- > > > Antoine > > > > > Index: Makefile === RCS file: /cvs/ports/net/zeromq/Makefile,v retrieving revision 1.13 diff -u -p -r1.13 Makefile --- Makefile 15 Oct 2014 14:28:42 - 1.13 +++ Makefile 25 Dec 2014 19:28:44 - @@ -13,6 +13,10 @@ SHARED_LIBS += zmq 4.0 CATEGORIES = net HOMEPAGE = http://zeromq.org/ +MAINTAINER =Rodrigo Mosconi + +REVISION = 0 + # LGPLv3 PERMIT_PACKAGE_CDROM = Yes
Re: GnuPGInterface, python-gnupg and duplicity (was: Re: Python GnuPG naming woes)
On Tue, Dec 23, 2014 at 1:07 AM, Jérémie Courrèges-Anglas wrote: > viq writes: > >> So, currently we have security/py-gnupg, which is >> http://py-gnupg.sourceforge.net/ and >> https://pypi.python.org/pypi/GnuPGInterface/ > > Yup. > >> Now I'm looking into porting >> https://pypi.python.org/pypi/python-gnupg/ and I have no idea how it >> should be named... Unless current one was renamed to >> py-gnupgpinterface or similiar and the new one took it's place. > > The current py-gnupg port could go away. It isn't used by any port, not > even duplicity. Rather than importing python-gnupg under a different > name, I propose a replacement. Yes the API is different, but I don't > think this is a concern. > > Here's a diff that: > - drops duplicity's dep on GnuPGInterface (it includes its own, improved > copy). > - replaces GnuPGInterface-0.3.2 with python-gnupg-0.3.7. DESCR stolen > from viq, the test that fails is due to files missing in the tarball. > > ok? If you're asking me, then yes, certainly. Though I assume you're not, in which case - bump. -- viq
security/gnupg update (+ maintainership)
Two changes in this release: - fix regression in --recv-keys if multiple keys are specified - bound the key sizes for automatic key generation While, here: - take maintainership - remove the unneeded custom do-test target Regress tests pass on amd64 and sparc64. ok? Index: Makefile === RCS file: /cvs/ports/security/gnupg/Makefile,v retrieving revision 1.95 diff -u -p -r1.95 Makefile --- Makefile13 Oct 2014 19:21:16 - 1.95 +++ Makefile25 Dec 2014 21:26:16 - @@ -2,8 +2,7 @@ COMMENT= GNU privacy guard - a free PGP replacement -DISTNAME= gnupg-1.4.17 -REVISION= 0 +DISTNAME= gnupg-1.4.18 CATEGORIES=security # restrict, not compatible with gnupg-2. @@ -13,6 +12,8 @@ MASTER_SITES= ${MASTER_SITES_GNUPG:=gnup HOMEPAGE= http://www.gnupg.org/ +MAINTAINER=Jeremie Courreges-Anglas + # GPLv3 PERMIT_PACKAGE_CDROM= Yes @@ -64,8 +65,5 @@ post-install: ${INSTALL_DATA_DIR} ${PREFIX}/share/doc/gnupg cd ${WRKSRC}; ${INSTALL_DATA} ${GNUPGDOCFILES} \ ${PREFIX}/share/doc/gnupg - -do-test: - @cd ${WRKSRC}/checks && ${MAKE_PROGRAM} check .include Index: distinfo === RCS file: /cvs/ports/security/gnupg/distinfo,v retrieving revision 1.27 diff -u -p -r1.27 distinfo --- distinfo26 Jun 2014 00:30:15 - 1.27 +++ distinfo25 Dec 2014 19:03:01 - @@ -1,2 +1,2 @@ -SHA256 (gnupg-1.4.17.tar.gz) = IvWAYMxXOSkOnPLJYPLas8ijNyZx57aLVr96EHdXONo= -SIZE (gnupg-1.4.17.tar.gz) = 5047785 +SHA256 (gnupg-1.4.18.tar.gz) = EKiTa3bMrbmFIVNbX0HMW0GzwVmmwpFfBivU3IrBLRI= +SIZE (gnupg-1.4.18.tar.gz) = 5047888 -- jca | PGP : 0x1524E7EE / 5135 92C1 AD36 5293 2BDF DDCC 0DFA 74AE 1524 E7EE
Re: qt5 testing - working on a port of Otter browser
On Thu, Dec 25, 2014, at 09:44 PM, Adam Wolk wrote: > On Thu, Dec 25, 2014, at 09:28 PM, Adam Wolk wrote: > > Hi ports@ > > > > I'm CC'ing the KDE porting team as they are marked as the MAINTAINER of > > the qt5 port in the > > OpenBSD ports tree. > > > > I'm working on a port of the Otter Browser (http://otter-browser.org/) > > and have hit a bunch of issues that > > do not seem to be upstream (otter-browser) related. > > > > Details: > > - OpenBSD i386 running snapshots (from 23 Dec 2014) > > - otter-browser commit e6a48fd71c121f1d486dabcf21e5b48809497c3a from > > Thu Dec 25 16:28:02 2014 +0100 > > - ldd output at the end of the email > > Forgot to mention. I'm running with Xfce 4.10 - KDE & Gnome is not > installed on this system. > > Regards, > Adam > I found the reason for the following error: "Warning: QGtkStyle could not resolve GTK. Make sure you have installed the proper libraries." In the qt5 source code the error comes from: https://qt.gitorious.org/qt/qt5-maemo5-qtbase/source/3e11afdea6edb753291ad15d963e8b44c9f5d33c:src/widgets/styles/qgtkstyle_p.cpp#L612 qWarning("QGtkStyle could not resolve GTK. Make sure you have installed the proper libraries."); which is based on a call to QGtkStylePrivate::gtk_init which in turn is dynamically resolved by a call to void QGtkStylePrivate::resolveGtk() const: https://qt.gitorious.org/qt/qt5-maemo5-qtbase/source/3e11afdea6edb753291ad15d963e8b44c9f5d33c:src/widgets/styles/qgtkstyle_p.cpp#L351 // enforce the "0" suffix, so we'll open libgtk-x11-2.0.so.0 QLibrary libgtk(QLS("gtk-x11-2.0"), 0, 0); as you can see it tries to load libgtk-x11-2.0.so.0 but on OpenBSD the .so file is named libgtk-x11-2.0.so.2400.0 leading to the library not being loaded and the warning displayed with an improperly initialized UI. I confirmed this by making a symlink from /usr/local/lib/libgtk-x11-2.0.so.0 to /usr/local/lib/libgtk-x11-2.0.so.2400.0 and running the application which got rid of the gtk style warning. Does anyone know why the .so name differs on OpenBSD? Should this be handled by the Qt5 port or altered and submitted upstream? Perhaps I should handle this somehow in the otter port I'm working on - any suggestions? The remaining issue is the crashes & cmake issues I mentioned before. I don't know yet if the crashes could be a result of the following message: ./otter-browser:/usr/local/lib/libestdc++.so.16.0: /usr/lib/libstdc++.so.57.0 : WARNING: symbol(_ZN11__gnu_debug17_S_debug_messagesE) size mismatch, relink your program full transcript of an attempt to run the app: $ ./otter-browser ./otter-browser:/usr/local/lib/libestdc++.so.16.0: /usr/lib/libstdc++.so.57.0 : WARNING: symbol(_ZN11__gnu_debug17_S_debug_messagesE) size mismatch, relink your program otter-browser(32731) in free(): error: bogus pointer (double free?) 0xdfdfdfdf Abort trap (core dumped) which were also present during compilation but I'm starting to believe that this could be the reason and could be caused by a mismatch between my packages/base (though they are only 1 day apart - base Dec 23, ports Dec 24) updated with pkg_add -uiv instead of manually recompiling. I would appreciate any input that could lead me closer to the root cause Regards, -- Adam Wolk adam.w...@koparo.com
Re: ntp-4.2.8
On Mon, 22 Dec 2014 21:23:41 +0100, j...@wxcvbn.org (Jérémie Courrèges-Anglas) wrote: > j...@wxcvbn.org (Jérémie Courrèges-Anglas) writes: > > > Here's an update for net/ntp, which should fix a few CVEs. > > > > http://www.kb.cert.org/vuls/id/852879 > > > > Basic tests on amd64: add a few minutes to the system clock then > > watch ntpd -n -dd sync it. > > > > ok? > > And... same diff for OPENBSD_5_6 and OPENBSD_5_5, but I don't have > stable boxes to test them. > > Tested on OpenBSD 5.6-stable amd64, builds fine and the basic test works properly. Cheers, Daniel
Re: maintainer net/zeromq
Rodrigo Mosconi writes: > with revision=0 And with REVISION kept close to DISTNAME, so that it doesn't get forgotten with the next zeromq update. Index: Makefile === RCS file: /cvs/ports/net/zeromq/Makefile,v retrieving revision 1.13 diff -u -p -p -u -r1.13 Makefile --- Makefile15 Oct 2014 14:28:42 - 1.13 +++ Makefile25 Dec 2014 22:39:03 - @@ -7,11 +7,14 @@ SHARED_ONLY = Yes COMMENT = open source message queue optimised for performance DISTNAME = zeromq-4.0.5 +REVISION = 0 SHARED_LIBS += zmq 4.0 # 4.0 CATEGORIES = net HOMEPAGE = http://zeromq.org/ + +MAINTAINER = Rodrigo Mosconi # LGPLv3 PERMIT_PACKAGE_CDROM = Yes -- jca | PGP : 0x1524E7EE / 5135 92C1 AD36 5293 2BDF DDCC 0DFA 74AE 1524 E7EE
Re: qt5 testing - working on a port of Otter browser
25 дек. 2014 г. 23:28 пользователь "Adam Wolk" написал: > > Hi ports@ > > I'm CC'ing the KDE porting team as they are marked as the MAINTAINER of > the qt5 port in the > OpenBSD ports tree. > > I'm working on a port of the Otter Browser (http://otter-browser.org/) > and have hit a bunch of issues that > do not seem to be upstream (otter-browser) related. > > Details: > - OpenBSD i386 running snapshots (from 23 Dec 2014) > - otter-browser commit e6a48fd71c121f1d486dabcf21e5b48809497c3a from > Thu Dec 25 16:28:02 2014 +0100 > - ldd output at the end of the email > > I will provide as many details as necessary - just ask :) > > 0. CMake fails to find qt5 > > qt5 is not found unless I prefix it with: > > CMAKE_PREFIX_PATH=/usr/local/lib/qt5/cmake/ > > $ cmake ../ > -- The C compiler identification is GNU 4.2.1 > -- The CXX compiler identification is GNU 4.2.1 > -- Check for working C compiler: /usr/bin/cc > -- Check for working C compiler: /usr/bin/cc -- works > -- Detecting C compiler ABI info > -- Detecting C compiler ABI info - done > -- Check for working CXX compiler: /usr/bin/c++ > -- Check for working CXX compiler: /usr/bin/c++ -- works > -- Detecting CXX compiler ABI info > -- Detecting CXX compiler ABI info - done > CMake Error at CMakeLists.txt:56 (find_package): > By not providing "FindQt5Core.cmake" in CMAKE_MODULE_PATH this project > has > asked CMake to find a package configuration file provided by > "Qt5Core", but > CMake did not find one. > > Could not find a package configuration file provided by "Qt5Core" with > any > of the following names: > > Qt5CoreConfig.cmake > qt5core-config.cmake > > Add the installation prefix of "Qt5Core" to CMAKE_PREFIX_PATH or set > "Qt5Core_DIR" to a directory containing one of the above files. If > "Qt5Core" provides a separate development package or SDK, be sure it > has > been installed. > > > -- Configuring incomplete, errors occurred! > See also > "/home/mulander/otter-head/otter/build/CMakeFiles/CMakeOutput.log". That's "normal" - I.e., things are the way Qt developers set them. They provide Qt5 CMake package config files, containing a lot of hardcoded relative paths there. So it's not easy to move .cmake files under some common prefix, outside of /usr/local/lib/qt5. N.B.: If you'd use x11/qt5 module, you'll get all the CMake setup for free, see /usr/ports/x11/qt5/qt5.port.mk (documentation is on the way). > 1. Otter browser uses CMake to find libraries - packaging forces me to > install all qt5 libs > > Otter Browser only uses: > > find_package(Qt5Core REQUIRED) > find_package(Qt5Network REQUIRED) > find_package(Qt5PrintSupport REQUIRED) > find_package(Qt5Script REQUIRED) > find_package(Qt5Sql REQUIRED) > find_package(Qt5Widgets REQUIRED) > find_package(Qt5WebKit REQUIRED) > find_package(Qt5WebKitWidgets REQUIRED) > > Yet the build fails unless I have the following installed: > > $ pkg_info -Q qt5 > qt5-5.3.2p0 (installed) > qt5-examples-5.3.2 > qt5-mysql-5.3.2 (installed) > qt5-postgresql-5.3.2 (installed) > qt5-sqlite2-5.3.2 (installed) > qt5-tds-5.3.2 (installed) > > In summary - CMake prevents the build unless I have tds, mysql & > postgresql qt5 installed even though otter-browser only uses > qt5-sqlite2. Yes, known problem. CMake config file for Qt5Sql talks about all backends always. To be redone after fixing other urgent stuff. BTW, can't the Otter use the SQLite3 backend instead of SQLite2? The former is bundled with Qt5 itself because SQLite3 is part of base system nowadays. > 2. gcc warnings after compilation > Linking CXX executable otter-browser > /usr/local/lib/qt5/libQt5Script.so.0.0: warning: random() may return > deterministic values, is that what you want? > /usr/local/lib/qt5/libQt5WebKit.so.0.0: warning: strcpy() is almost > always misused, please use strlcpy() > /usr/local/lib/qt5/libQt5Core.so.0.0: warning: rand() may return > deterministic values, is that what you want? > /usr/local/lib/qt5/libQt5Script.so.0.0: warning: sprintf() is often > misused, please use snprintf() > /usr/X11R6/lib/libX11.so.16.0: warning: strcat() is almost always > misused, please use strlcat() > /usr/local/lib/libglib-2.0.so.4200.0: warning: stpcpy() is dangerous; do > not use it > /usr/local/lib/libglib-2.0.so.4200.0: warning: vsprintf() is often > misused, please use vsnprintf() That's "normal"... Well, except strcpy() in Qt5WebKit - strange, I thought I removed them all. I'll check again, thanks for reporting. The sprintf() worths looking into as well. (other points to be answered in a separate mail) -- Vadim Zhukov
Re: qt5 testing - working on a port of Otter browser
On Fri, Dec 26, 2014, at 12:15 AM, Vadim Zhukov wrote: > > Yes, known problem. CMake config file for Qt5Sql talks about all > backends always. To be redone after fixing other urgent stuff. > > BTW, can't the Otter use the SQLite3 backend instead of SQLite2? The > former is bundled with Qt5 itself because SQLite3 is part of base > system nowadays. > I installed the sqlite2 backend as that's the only one that showed up on packages for me: $ pkg_info -Q qt5 qt5-5.3.2p0 (installed) qt5-examples-5.3.2 qt5-mysql-5.3.2 (installed) qt5-postgresql-5.3.2 (installed) qt5-sqlite2-5.3.2 (installed) qt5-tds-5.3.2 (installed) that's the complete list of qt5 packages I get for this query. > > 2. gcc warnings after compilation > > Linking CXX executable otter-browser > > /usr/local/lib/qt5/libQt5Script.so.0.0: warning: random() may return > > deterministic values, is that what you want? > > /usr/local/lib/qt5/libQt5WebKit.so.0.0: warning: strcpy() is almost > > always misused, please use strlcpy() > > /usr/local/lib/qt5/libQt5Core.so.0.0: warning: rand() may return > > deterministic values, is that what you want? > > /usr/local/lib/qt5/libQt5Script.so.0.0: warning: sprintf() is often > > misused, please use snprintf() > > /usr/X11R6/lib/libX11.so.16.0: warning: strcat() is almost always > > misused, please use strlcat() > > /usr/local/lib/libglib-2.0.so.4200.0: warning: stpcpy() is dangerous; do > > not use it > > /usr/local/lib/libglib-2.0.so.4200.0: warning: vsprintf() is often > > misused, please use vsnprintf() > > That's "normal"... Well, except strcpy() in Qt5WebKit - strange, I > thought I removed them all. I'll check again, thanks for reporting. > The sprintf() worths looking into as well. > > (other points to be answered in a separate mail) > > -- > Vadim Zhukov > First of all thanks for the reply. I'm trying my best to help out and I hope I'm not spamming. If I should remove the CC to KDE porters then please inform me. Here are some follow ups that I found after investigating more since my last email. 1. The double free crashes happen in 2 cases. a) trying to resize the window b) closing a modal dialog 2. While debugging one of the core dumps with gdb - gdb core dumped itself I have this core dump set aside in case it's meaningful to someone. 3. The browser itself works. I can load pages (ie. openbsd.org) open tabs, close tabs etc. Here are the backtraces: 1) GDB core dump while debugging a core dump (this is while trying to resize the whole window) #0 0x04a7e371 in kill () at :2 2 : No such file or directory. in (gdb) bt #0 0x04a7e371 in kill () at :2 #1 0x04abc836 in raise (s=6) at /usr/src/lib/libc/gen/raise.c:39 #2 0x04abc780 in abort () at /usr/src/lib/libc/stdlib/abort.c:53 #3 0x04ab8b87 in wrterror (msg=0x24a16956 "bogus pointer (double free?)", p=0xdfdfdfdf) at /usr/src/lib/libc/stdlib/malloc.c:281 #4 0x04aba099 in free (ptr=0xdfdfdfdf) at /usr/src/lib/libc/stdlib/malloc.c:1282 #5 0x0fa1c516 in QXcbShmImage::destroy () from /usr/local/lib/qt5/plugins/platforms/libqxcb.so #6 0xdfdfdfdf in ?? () #7 0x027f in ?? () #8 0x2f478040 in QCoreApplicationPrivate::checkInstance(char const*)::__PRETTY_FUNCTION__ () from /usr/local/lib/qt5/libQt5Core.so.0.0 #9 0x2a80b704 in ?? () from /usr/lib/libstdc++.so.57.0 #10 0x82f09740 in ?? () #11 0x2fa07b48 in ?? () from /usr/local/lib/qt5/plugins/platforms/libqxcb.so #12 0x8561f630 in ?? () #13 0x0fa1d4ac in QXcbBackingStore::resize () from /usr/local/lib/qt5/plugins/platforms/libqxcb.so #14 0x0145dca2 in QBackingStore::resize () from /usr/local/lib/qt5/libQt5Gui.so.0.0 #15 0x04fdf74c in QWidgetBackingStore::doSync () from /usr/local/lib/qt5/libQt5Widgets.so.0.0 #16 0x04fe083e in QWidgetBackingStore::sync () from /usr/local/lib/qt5/libQt5Widgets.so.0.0 Segmentation fault (core dumped) 2) backtrace from dialog close #0 0x0e212371 in kill () at :2 2 : No such file or directory. in (gdb) bt #0 0x0e212371 in kill () at :2 #1 0x0e250836 in raise (s=6) at /usr/src/lib/libc/gen/raise.c:39 #2 0x0e250780 in abort () at /usr/src/lib/libc/stdlib/abort.c:53 #3 0x0e24cb87 in wrterror (msg=0x2e1aa956 "bogus pointer (double free?)", p=0xdfdfdfdf) at /usr/src/lib/libc/stdlib/malloc.c:281 #4 0x0e24e099 in free (ptr=0xdfdfdfdf) at /usr/src/lib/libc/stdlib/malloc.c:1282 #5 0x0ac88516 in QXcbShmImage::destroy () from /usr/local/lib/qt5/plugins/platforms/libqxcb.so #6 0xdfdfdfdf in ?? () #7 0x852d7600 in ?? () #8 0xcfbd7cc4 in ?? () #9 0x0e24e0b5 in free (ptr=0x873cf990) at /usr/src/lib/libc/stdlib/malloc.c:1285 #10 0x0babb7a9 in QWidget::~QWidget () from /usr/local/lib/qt5/libQt5Widgets.so.0.0 #11 0x0bcba297 in QDialog::~QDialog () from /usr/local/lib/qt5/libQt5Widgets.so.0.0 #12 0x1910c16d in Otter::StartupDialog::~StartupDialog () from /home/mulander/otter-head/otter/build/otter-browser #13 0x18fd91d3 in main () from /home/mulander/otter-head/otter/build/otter-browser Current language: auto; currently asm
Re: qt5 testing - working on a port of Otter browser
2014-12-26 1:37 GMT+03:00 Adam Wolk : > > On Thu, Dec 25, 2014, at 09:44 PM, Adam Wolk wrote: >> On Thu, Dec 25, 2014, at 09:28 PM, Adam Wolk wrote: >> > Hi ports@ >> > >> > I'm CC'ing the KDE porting team as they are marked as the MAINTAINER of >> > the qt5 port in the >> > OpenBSD ports tree. >> > >> > I'm working on a port of the Otter Browser (http://otter-browser.org/) >> > and have hit a bunch of issues that >> > do not seem to be upstream (otter-browser) related. >> > >> > Details: >> > - OpenBSD i386 running snapshots (from 23 Dec 2014) >> > - otter-browser commit e6a48fd71c121f1d486dabcf21e5b48809497c3a from >> > Thu Dec 25 16:28:02 2014 +0100 >> > - ldd output at the end of the email >> >> Forgot to mention. I'm running with Xfce 4.10 - KDE & Gnome is not >> installed on this system. >> >> Regards, >> Adam >> > > I found the reason for the following error: > "Warning: QGtkStyle could not resolve GTK. Make sure you have installed > the proper libraries." > > In the qt5 source code the error comes from: > https://qt.gitorious.org/qt/qt5-maemo5-qtbase/source/3e11afdea6edb753291ad15d963e8b44c9f5d33c:src/widgets/styles/qgtkstyle_p.cpp#L612 > qWarning("QGtkStyle could not resolve GTK. Make sure you have installed > the proper libraries."); > > which is based on a call to QGtkStylePrivate::gtk_init which in turn is > dynamically resolved by a call to void QGtkStylePrivate::resolveGtk() > const: > https://qt.gitorious.org/qt/qt5-maemo5-qtbase/source/3e11afdea6edb753291ad15d963e8b44c9f5d33c:src/widgets/styles/qgtkstyle_p.cpp#L351 > > // enforce the "0" suffix, so we'll open libgtk-x11-2.0.so.0 > QLibrary libgtk(QLS("gtk-x11-2.0"), 0, 0); > > as you can see it tries to load libgtk-x11-2.0.so.0 but on OpenBSD the > .so file is named libgtk-x11-2.0.so.2400.0 leading to the library not > being loaded and the warning displayed with an improperly initialized > UI. > > I confirmed this by making a symlink from > /usr/local/lib/libgtk-x11-2.0.so.0 to > /usr/local/lib/libgtk-x11-2.0.so.2400.0 and running the application > which got rid of the gtk style warning. > > Does anyone know why the .so name differs on OpenBSD? Should this be > handled by the Qt5 port or altered and submitted upstream? Perhaps I > should handle this somehow in the otter port I'm working on - any > suggestions? Nice catch! Yes, OpenBSD shared library rules are quiet different - in particular, we override library versions due to the unfortunate fact that many upstream developers don't respect shared library ABI rules always. I've just committed fix to Qt5 port that avoids hardcoding shared library major versions. So this issue should be fixed as of now. You'll need to build Qt5 yourself (that will require some patience) or wait for new packages snapshot. I can't provide amd64 package as of now because in my tree I have quiet a few other patches to be polished (or discarded) for Qt. > The remaining issue is the crashes & cmake issues I mentioned before. > I don't know yet if the crashes could be a result of the following > message: > > ./otter-browser:/usr/local/lib/libestdc++.so.16.0: > /usr/lib/libstdc++.so.57.0 : WARNING: > symbol(_ZN11__gnu_debug17_S_debug_messagesE) size mismatch, relink your > program > > full transcript of an attempt to run the app: > $ ./otter-browser > ./otter-browser:/usr/local/lib/libestdc++.so.16.0: > /usr/lib/libstdc++.so.57.0 : WARNING: > symbol(_ZN11__gnu_debug17_S_debug_messagesE) size mismatch, relink your > program > otter-browser(32731) in free(): error: bogus pointer (double free?) > 0xdfdfdfdf > Abort trap (core dumped) Hm, looks like really packages mismatch in GTK land. I'm not an expert here but remember there were many problems due to the nature of GTK (or GDK?) stuff. The rule here is simple: things should be in sync. -- WBR, Vadim Zhukov
Re: qt5 testing - working on a port of Otter browser
2014-12-26 2:33 GMT+03:00 Adam Wolk : > On Fri, Dec 26, 2014, at 12:15 AM, Vadim Zhukov wrote: >> >> Yes, known problem. CMake config file for Qt5Sql talks about all >> backends always. To be redone after fixing other urgent stuff. >> >> BTW, can't the Otter use the SQLite3 backend instead of SQLite2? The >> former is bundled with Qt5 itself because SQLite3 is part of base >> system nowadays. >> > > I installed the sqlite2 backend as that's the only one that showed up > on packages for me: > > $ pkg_info -Q qt5 > qt5-5.3.2p0 (installed) > qt5-examples-5.3.2 > qt5-mysql-5.3.2 (installed) > qt5-postgresql-5.3.2 (installed) > qt5-sqlite2-5.3.2 (installed) > qt5-tds-5.3.2 (installed) > > that's the complete list of qt5 packages I get for this query. > >> > 2. gcc warnings after compilation >> > Linking CXX executable otter-browser >> > /usr/local/lib/qt5/libQt5Script.so.0.0: warning: random() may return >> > deterministic values, is that what you want? >> > /usr/local/lib/qt5/libQt5WebKit.so.0.0: warning: strcpy() is almost >> > always misused, please use strlcpy() >> > /usr/local/lib/qt5/libQt5Core.so.0.0: warning: rand() may return >> > deterministic values, is that what you want? >> > /usr/local/lib/qt5/libQt5Script.so.0.0: warning: sprintf() is often >> > misused, please use snprintf() >> > /usr/X11R6/lib/libX11.so.16.0: warning: strcat() is almost always >> > misused, please use strlcat() >> > /usr/local/lib/libglib-2.0.so.4200.0: warning: stpcpy() is dangerous; do >> > not use it >> > /usr/local/lib/libglib-2.0.so.4200.0: warning: vsprintf() is often >> > misused, please use vsnprintf() >> >> That's "normal"... Well, except strcpy() in Qt5WebKit - strange, I >> thought I removed them all. I'll check again, thanks for reporting. >> The sprintf() worths looking into as well. >> >> (other points to be answered in a separate mail) >> >> -- >> Vadim Zhukov >> > > First of all thanks for the reply. I'm trying my best to help out and I > hope I'm not spamming. > If I should remove the CC to KDE porters then please inform me. No, no, no, everyting is normal. The reality is that I'm 95% of "KDE development team" as of now, and also I'm responsible for Qt5 as well. The "KDE team" is just my hope that someone will pick up my work when I'll be hit by bus. :) > Here are some follow ups that I found after investigating more since my > last email. > > 1. The double free crashes happen in 2 cases. > a) trying to resize the window > b) closing a modal dialog > > 2. While debugging one of the core dumps with gdb - gdb core dumped > itself > I have this core dump set aside in case it's meaningful to someone. GDB can't debug real Qt5 apps. :( That comes from old binutils here in OpenBSD, at least, and that's not a thing I can fix, sorry. > 3. The browser itself works. I can load pages (ie. openbsd.org) open > tabs, close tabs etc. > > Here are the backtraces: > > 1) GDB core dump while debugging a core dump (this is while trying to > resize the whole window) > #0 0x04a7e371 in kill () at :2 > 2 : No such file or directory. > in > (gdb) bt > #0 0x04a7e371 in kill () at :2 > #1 0x04abc836 in raise (s=6) at /usr/src/lib/libc/gen/raise.c:39 > #2 0x04abc780 in abort () at /usr/src/lib/libc/stdlib/abort.c:53 > #3 0x04ab8b87 in wrterror (msg=0x24a16956 "bogus pointer (double > free?)", p=0xdfdfdfdf) at /usr/src/lib/libc/stdlib/malloc.c:281 > #4 0x04aba099 in free (ptr=0xdfdfdfdf) at > /usr/src/lib/libc/stdlib/malloc.c:1282 > #5 0x0fa1c516 in QXcbShmImage::destroy () from > /usr/local/lib/qt5/plugins/platforms/libqxcb.so > #6 0xdfdfdfdf in ?? () > #7 0x027f in ?? () > #8 0x2f478040 in QCoreApplicationPrivate::checkInstance(char > const*)::__PRETTY_FUNCTION__ () from > /usr/local/lib/qt5/libQt5Core.so.0.0 > #9 0x2a80b704 in ?? () from /usr/lib/libstdc++.so.57.0 > #10 0x82f09740 in ?? () > #11 0x2fa07b48 in ?? () from > /usr/local/lib/qt5/plugins/platforms/libqxcb.so > #12 0x8561f630 in ?? () > #13 0x0fa1d4ac in QXcbBackingStore::resize () from > /usr/local/lib/qt5/plugins/platforms/libqxcb.so > #14 0x0145dca2 in QBackingStore::resize () from > /usr/local/lib/qt5/libQt5Gui.so.0.0 > #15 0x04fdf74c in QWidgetBackingStore::doSync () from > /usr/local/lib/qt5/libQt5Widgets.so.0.0 > #16 0x04fe083e in QWidgetBackingStore::sync () from > /usr/local/lib/qt5/libQt5Widgets.so.0.0 > Segmentation fault (core dumped) > > 2) backtrace from dialog close > #0 0x0e212371 in kill () at :2 > 2 : No such file or directory. > in > (gdb) bt > #0 0x0e212371 in kill () at :2 > #1 0x0e250836 in raise (s=6) at /usr/src/lib/libc/gen/raise.c:39 > #2 0x0e250780 in abort () at /usr/src/lib/libc/stdlib/abort.c:53 > #3 0x0e24cb87 in wrterror (msg=0x2e1aa956 "bogus pointer (double > free?)", p=0xdfdfdfdf) at /usr/src/lib/libc/stdlib/malloc.c:281 > #4 0x0e24e099 in free (ptr=0xdfdfdfdf) at > /usr/src/lib/libc/stdlib/malloc.c:1282 > #5 0x0ac88516 in QXcbShmImage::destroy () from > /usr/local/lib/qt5/plugins/pl
Re: NEW port net/czmq
Hi Rodrigo, thanks for your submission. Rodrigo Mosconi writes: > Follow a new port net/czmq. It is a high-level C bindings for zeromq. > > The port is for version 2.2 of czmq. When the new version (3.0) will be > released, I will update it. > > Is this OK? Not quite. - please respect the order defined in /usr/ports/infrastructure/templates/Makefile.template - for an initial import, or an update to a newer upstream version, REVISION should not be present. - shared libs are an important subject, see the porter's handbook (http://www.openbsd.org/faq/ports/specialtopics.html#SharedLibs) Here the concern is that you shouldn't use upstream library versioning, but maintain your own versioning in parallel, inside the ports tree. So the version should start at 0.0. - the customary way to set WANTLIB is just to copy/paste the output of "make port-lib-depends-check". CONFIGURE_ARGS are usually split on several lines, one per argument. Watch out for trailing whitespace. COMMENT should not start with a capital letter. - it sounds weird to depend on settings added by --with-libzmq=${LOCALBASE} when other libraries (eg. libsodium) should also get picked up from there. In this case I suggest using CONFIGURE_ENV to ensure that CPPFLAGS and LDFLAGS are sane from the start. - libuuid from e2fsprogs, asciidoc and xmlto could be picked up in bulk builds. Either enable (*_DEPENDS) or disable them (see updated tarball). Note that there is fallback code for uuid generation but it isn't quite as optimal as what we have in libc. A nice proposal for upstream would be to use uuid functions from libc (present on most recent *BSD systems). http://www.openbsd.org/cgi-bin/man.cgi/OpenBSD-current/man3/uuid.3 Regarding the packaged files, now: - I think makecert should be renamed as czmq_makecert, at least to avoid a conflict with mono. - wtf are those "gsl" scripts doing in bin/? :) Shouldn't they go to /usr/local/share/czmq... instead? https://github.com/imatix/gsl - I'm wondering whether czmq_selftest should be present Updated tarball, to be unpacked from mystuff/net. czmq.tgz Description: Binary data -- jca | PGP : 0x1524E7EE / 5135 92C1 AD36 5293 2BDF DDCC 0DFA 74AE 1524 E7EE
Re: qt5 testing - working on a port of Otter browser
On Fri, Dec 26, 2014, at 12:45 AM, Vadim Zhukov wrote: > 2014-12-26 2:33 GMT+03:00 Adam Wolk : > > On Fri, Dec 26, 2014, at 12:15 AM, Vadim Zhukov wrote: > > First of all thanks for the reply. I'm trying my best to help out and I > > hope I'm not spamming. > > If I should remove the CC to KDE porters then please inform me. > > No, no, no, everyting is normal. The reality is that I'm 95% of "KDE > development team" as of now, and also I'm responsible for Qt5 as well. > The "KDE team" is just my hope that someone will pick up my work when > I'll be hit by bus. :) > > > Here are some follow ups that I found after investigating more since my > > last email. > > > > 1. The double free crashes happen in 2 cases. > > a) trying to resize the window > > b) closing a modal dialog > > > > 2. While debugging one of the core dumps with gdb - gdb core dumped > > itself > > I have this core dump set aside in case it's meaningful to someone. > > GDB can't debug real Qt5 apps. :( That comes from old binutils here in > OpenBSD, at least, and that's not a thing I can fix, sorry. > Not a problem, just pointing things out as I hit them :) If there is a soul still versed in gdb internals then that person can hit me up to get those core dumps and binaries that resulted in them. > > 3. The browser itself works. I can load pages (ie. openbsd.org) open > > tabs, close tabs etc. > > > > Here are the backtraces: > > > > 1) GDB core dump while debugging a core dump (this is while trying to > > resize the whole window) > > > > 2) backtrace from dialog close > > I have another report about bugs in Qt5WebKit. It's New Year coming > here in Russia; hope to have time to debug those things during > holidays. I understand those crashes are not pretty looking, but there > is some other urgent stuff in Qt land as well. :-\ > > -- > WBR, > Vadim Zhukov > You already helped :) All I wanted to know is if those crashes are stuff that I'm doing wrong or a known/unknown issue in the Qt port. I'll wait for new packages in the ports and will keep retrying / pin pointing the issue. Regards,
Re: New diff for CMake-3.1 to test (Was: UPDATE: freetype-2.5.4)
2014-12-26 0:01 GMT+03:00 Vadim Zhukov : > 25 дек. 2014 г. 22:48 пользователь "David Coppa" написал: > > >> >> On Thu, Dec 25, 2014 at 7:46 PM, Landry Breuil wrote: >> >> > x11/kde4/kscd + net/ktorrent-kde4 + x11/kde4/marble share th same >> > failure: >> > >> > CMake Error at /usr/local/share/cmake/Modules/FindPkgConfig.cmake:112 >> > (elseif): >> > given arguments: >> > >> > "VERSION_LESS" "3.1" >> > >> > Unknown arguments specified >> > Call Stack (most recent call first): >> > /usr/local/share/cmake/Modules/FindPkgConfig.cmake:501 >> > (_pkgconfig_parse_options) >> > /usr/local/share/cmake/Modules/FindLCMS.cmake:24 (pkg_check_modules) >> > /usr/local/share/cmake/Modules/Qt4ConfigDependentSettings.cmake:37 >> > (include) >> > /usr/local/share/cmake/Modules/FindQt4.cmake:1198 (include) >> > CMakeLists.txt:5 (find_package) >> > >> > x11/kde/kdeedu3: >> > kalzium.all_cpp.o(.text+0x3a5f): In function >> > `DetailedInfoDlg::addHTMLTab(QString const&, QString const&, QString >> > const&)': : undefined reference to `KHTMLPart::KHTMLPart(QWidget*, char >> > const*, >> > QObject*, char const*, KHTMLPart::GUIProfile)' >> > collect2: ld returned 1 exit status >> > Makefile:598: recipe for target 'kalzium' failed >> > >> > (an unrelated fluke ?) >> > >> > editors/nvi: >> > Scanning dependencies of target headers >> > /usr/bin/make -f CMakeFiles/headers.dir/build.make >> > CMakeFiles/headers.dir/build >> > make: don't know how to make >> > /usr/obj/mfs/nvi-2.1.2/nvi-2.1.2/cl/extern.h (prerequisite of: >> > CMakeFiles/headers) >> > Stop in . >> >> I've just now fixed nvi. >> >> > graphics/blender: >> > >> > >> > /usr/obj/mfs/blender-2.72b/blender-2.72b/source/blender/blenfont/intern/blf_glyph.c: >> > In function 'blf_ >> > glyph_add': >> > >> > /usr/obj/mfs/blender-2.72b/blender-2.72b/source/blender/blenfont/intern/blf_glyph.c:265:19: >> > error: con >> > version to 'int' from 'unsigned int' may change the sign of the result >> > [-Werror=sign-conversion] >> > g->width = bitmap.width; >> >^ >> > >> > /usr/obj/mfs/blender-2.72b/blender-2.72b/source/blender/blenfont/intern/blf_glyph.c:266:20: >> > error: conversion to 'int' from 'unsigned int' may change the sign of the >> > result [-Werror=sign-conversion] >> > g->height = bitmap.rows; >> > ^ >> >> This is caused by freetype-2.5.4 like openimageio and openttd. I'll >> fix it asap... >> >> CCing Vadim about kde-related failures. > > Already looking there. Actually, there's a KDE 4.14.3 minor update pending, > so I have an evil mix of KDE packages here... The problems look like being > easily solvable, so there should a patch happen in a few hours. Fixes for KDE4 for CMake 3.1+ are in. x11/kde/edu3 is interesting but totally unrelated to CMake. Can I have a log, please? -- Vadim Zhukov
UPDATE: www/lynx 2.8.9dev1 => 2.8.9dev2
Hi ports -- A new version of lynx was released a few days ago. Diff attached. Please test on your favorite arch. Runs well here on amd64; will test some others tomorrow. OK? ~Brian Index: Makefile === RCS file: /cvs/ports/www/lynx/Makefile,v retrieving revision 1.13 diff -u -p -r1.13 Makefile --- Makefile 21 Jul 2014 07:18:07 - 1.13 +++ Makefile 26 Dec 2014 05:41:50 - @@ -1,8 +1,7 @@ # $OpenBSD: Makefile,v 1.13 2014/07/21 07:18:07 landry Exp $ V = 2.8.9 -PL = 1 -REVISION = 0 +PL = 2 COMMENT = console Internet browser DISTNAME = lynx${V}dev.${PL} PKGNAME = lynx-${V}pl${PL} Index: distinfo === RCS file: /cvs/ports/www/lynx/distinfo,v retrieving revision 1.1 diff -u -p -r1.1 distinfo --- distinfo 17 Jul 2014 12:15:21 - 1.1 +++ distinfo 26 Dec 2014 05:41:50 - @@ -1,2 +1,2 @@ -SHA256 (lynx2.8.9dev.1.tar.bz2) = 6SeAAuNBdSQOmG25KmD2IU81mZ9RKHBWFJzcCFeMK/8= -SIZE (lynx2.8.9dev.1.tar.bz2) = 2587468 +SHA256 (lynx2.8.9dev.2.tar.bz2) = jtapc2PloKMRuWxJH3wAgpWvp2/RtGuRxQ4Imc3JAQQ= +SIZE (lynx2.8.9dev.2.tar.bz2) = 2584499