SHAREMODE defaults to 444, and gives us endless grief because one
cannot modify a file that is not writable.
So use _SHAREMODE that is 644, so that we can avoid all those CHMOD +w
in so many Makefiles.
Exp-run: antoine
Sponsored by: Absolight
Differential Revision: https://reviews.freebsd.org/D7623
As noted in PR 209813, some glib headers include libintl.h and then use
gettext in some macros. Programs that use this API (directly or indirectly)
can end up with calls to gettext and need to be linked with libintl.so.
glib-2.0.pc has -lintl in its Libs too.
PR: 209813
With hat: portmgr
http://mirrors.kernel.org/sourceware/ in MASTER_SITE_GCC and
MASTER_SITE_SOURCEWARE.
The original name sources.redhat.com for sourceware.org has been
deprecated many years ago.
This allows WITH_CCACHE_BUILD to be set and still allow modifying MAKE_ENV/CC,
etc, to achieve using the ccache wrappers located at LOCALBASE/libexec/ccache.
This kind of mechanism is needed to support ccache-memcached which has
many build dependencies. Building a version of that first then allows
building further ports with it without introducing cyclic dependencies.
Poudriere will grow support for using a static version of ccache for this
purpose.
With hat: portmgr
ftp://ftp.funet.fi/pub/mirrors/sourceware.org in MASTER_SITE_GCC and
MASTER_SITE_SOURCEWARE, sourceware.org having replaced sources.redhat.com
as the standard name of that host years ago.
This should not bring any practical change for now; it's really the same.
More candidates are likely to come before the official announcement (aka
the rush hour) scheduled on 2017-03-07. This one intended to pick up upstream
security fixes earlier while looking for downstream regressions.
Changes: https://www.mozilla.org/firefox/52.0/releasenotes/
Security: 96eca031-1313-4daf-9be2-9d6e1c4f1eb5
MFH: 2017Q1 (piling up, requires r431637 + r434531)
* Update Qt5 to 5.7.1
* Move Qt4 binaries to lib/qt4/bin
* Move Qt5 libraries to lib/qt5/lib
By moving the libraries we should finally be able to get rid of the inplace
upgrade bug (see ports bugs 194088, 195105 and 198720): when Qt5's libraries
were lying in /usr/local/lib, which would often get added by pkgconfig to the
linker paths via dependencies, the already installed libraries were linked
against, instead of the ones that were being built. This forced us to make
sure, that -L${WRKSRC}/lib was always coming before -L/usr/local/lib in the
linker flags. With this change this should no longer be the case.
* Rename some ports to match the rest (foo-qtX -> qtX-foo)
* Depend on new port misc/qtchooser [see UPDATING & CHANGES]
There are several new Qt5 ports which all have been created by Marie Loise Nolden
<nolden@kde.org>. Thanks again.
PR: 216797
Exp-Run by: antoine
Reviewed by: rakuco, mat, groot_kde.org
Approved by: rakuco (mentor)
Differential Revision: https://reviews.freebsd.org/D9213
This was causing excessive 'uname -s' executions during Poudriere dependency
calculations. The reasoning is due to HAVE_COMPAT_IA32_KERN having a newline
quoted at the end of it as '\ ' by bmake:
# make -f Mk/bsd.port.mk -V 'HAVE_COMPAT_IA32_KERN=${HAVE_COMPAT_IA32_KERN:Q}' | cat -v
HAVE_COMPAT_IA32_KERN=YES\
This resulted in the following being executed:
export HAVE_COMPAT_IA32_KERN="YES OPSYS=FreeBSD"
With hat: portmgr
The mistake was completely on my part, I somehow connected the dots the
wrong way in my head.
The only exceptions (for now) are archivers/zstd and ports-mgmt/synth
which were already picked up by new volunteers in the mean time.
Tested with a modified sysutils/biosfont in which the :debug option was added,
a file was installed into ${KERN_DEBUGDIR}/${KMODDIR} and that file was added
to pkg-plist.
PR: 216879 (inspired by)
Submitted by: Sergey Kozlov (kozlov.sergey.404@gmail.com)
No longer depend on (KERN_)DEBUGDIR being non-empty to decide whether to
create the empty KERN_DEBUGDIR/KMODDIR directory, instead use a new
argument 'debug' to indicate this.
Add KERN_DEBUGDIR to PLIST_SUB and MAKE_ENV so that the including ports
can use them.
Ports using kmod:debug do need to add @dir entries to their pkg-plist for
now to prevent 'make check-plist' errors, for example for default KMODDDIR:
@dir /%%KERN_DEBUGDIR%%/%%KMODDIR%%
@dir /%%KERN_DEBUGDIR%%/boot
@dir /%%KERN_DEBUGDIR%%
It would indeed be nice to automate this.
PR: 216879
Submitted by: Sergey Kozlov (kozlov.sergey.404@gmail.com)
from ports, remove the symlink dance with libGL/libEGL/libgles and the
nvidia-driver ports.
* Revert libGL/libEGL/libglesv2 ports to normal.
* Rename the conflicting libraries from the nvidia-driver ports with a
-NVIDIA prefix and install a libmap.conf file to map to these versions.
* Remove the special case from Mk/Scripts/qa.sh for libGL and libEGL.
* Sadly the symlink dance between nvidia-driver and xorg-server for glx.so
need to stay, due to xorg-server opening that file with dlopen(3).
Approved by: portmgr@ (mat@ (prio version))
Differential Revision: https://reviews.freebsd.org/D7848
bsd.default-versions.mk can rely on ARCH being defined.
- In bsd.port.mk move inclusion of bsd.default-versions.mk from the
pre-makefile section to the options section so the variables can be used
earlier. Also put the bit of code sitting between the options section and
the pre-makefile section into the options section.
- Remove last few cases where ports set WITH_OPENSSL_PORT. This variable is
handled in bsd.default-versions.mk and some ports were setting it after
including bsd.port.options.mk. After FreeBSD 9 EoL all but a few ports,
and then only when setting non-default options, work without setting that
variable.
PR: 215996
Exp-run by: antoine
Approved by: portmgr (antoine)
a last resort.
- Construct a list of DISTFILES for every architecture (amd64, i386).
Select one list for regular make and all lists for make makesum so linux
ports can have one combined distinfo file.
- Set DIST_SUBDIR?=centos so there's only one copy of common distfiles.
- Use MASTER_SITES and DISTFILES groups so make fetch is a bit smarter.
- Regenerate all distinfo files.
- For linux_base-c6 (just like linux_base-c7 already does) turn bin, lib,
lib64 and sbin into symbolic links to the same directory under usr.
This fixes the problem where some programs/libraries exist under / on
FreeBSD and /usr on Linux or vice versa and then depending on the order
of search paths Linux programs may run/load FreeBSD programs/libraries
and fail.
- Turn usr/share/icons into a symbolic link to LOCALBASE/share/icons so
Linux programs can find desktop theme icons. This eliminates the need
for x11-themes/linux*-hicolor-icon-theme.
- Rename pkg-plist.x86_64 to pkg-plist.amd64.
- Regenerate all pkg-plist files.
- Add @preexec to linux_base pkg-plist that moves existing files in bin,
lib, lib64, sbin and icons to the new destination before the directories
are turned into symbolic links.
- Remove several empty directories from linux_base ports.
- Sweep over all linux ports: use consistent style, remove old CONFLICTS,
remove unused pkg-descr files, add NLS option,...
- Remove old linux arts, esound and openssl-compat ports.
be doing at the time.
When fetching, the code was looking for a distfile in either its
DIST_SUBDIR or the main DISTDIR, which was ok. Until a port
(devel/cargo) moves one of its distfile
(cargo-nightly-x86_64-unknown-freebsd.tar.gz) into a subdirectory.
do-fetch would see the distfile in DISTDIR and say, ok, it's there, and
checksum would not see it in its subdirectory and fail.
1: phabricator's blame mode is really, really, great to unroll history.
PR: 216442
Submitted by: mat
Reported by: Bob Willcox, dhw (on ports)
Exp-run by: antoine
Sponsored by: Absolight
Differential Revision: https://reviews.freebsd.org/D9318
Change default GEM_NAME from ${PORTNAME}-${PORTVERSION} to ${DISTNAME} to avoid
unnecessary GEM_NAME setting in ports Makefile
Differential Revision: https://reviews.FreeBSD.org/D9250
PR: 216267
Exp-run by: antoine
packages there are no such packages for 7.3 yet so i386 support has been
marked IGNORE for now.
Change DIST_SUBDIR to just "centos" so distfiles that remain the same
between major versions don't have to be redownloaded.
Remove fmake hacks from linux.mk.
Submitted by: Piotr Kubaj <pkubaj@anongoth.pl>
Sometime, Perl ports names do not map to the modules they provide
easily, and it makes it impossible to check for the existence of those
modules in Perl core.
Sponsored by: Absolight
locale set by the user. Add LANG=C and LC_ALL=C at the beginning of
bsd.port.mk and export them so all commands are executed with the C locale.
LC_ALL=C overrides all other LC_* variables. LANG is used by setlocale(3)
as default value for LC_* variables, so normally it isn't used when LC_ALL
is set, but there's code out there that looks at LANG directly so it's safer
to set it as well. The only commands not captured by this are !=
assignments before any inclusion of bsd.port.*mk.
Introduce USE_LOCALE=<locale> that adds LANG=<locale> and LC_ALL=<locale> to
CONFIGURE_ENV and MAKE_ENV so upstream build systems can be executed with a
different locale (e.g. USE_LOCALE=en_US.UTF-8).
PR: 215882
Exp-run by: antoine
Approved by: portmgr (antoine)
Unfortunately, it's not uncommon for FreeBSD users to not update
their systems timely and thus end up using unsupported FreeBSD
release. These users continue to update ports tree as usual and
expect it to work, either unaware of the release EoL, or not clearly
understanding the consequences, which results in unexpected build
failures, bogus bug reports, attempts to bring back removed legacy
support bits and general discontent.
This change introduces system version check which makes ports refuse
to build anything on unsupported system. This makes users aware of
EoL of their system and makes it clear that no port is guaranteed to
build.
The error message tells how to override the check (by defining
ALLOW_UNSUPPORTED_SYSTEM, in which case it's turned into a simple
warning), additionally stressing that this configurartion is not
supported.
Currently outdated are OSVERSION < 1003000 (pre 10.3-RELEASE) and
1100000 <= OSVERSION < 1100122 (from 11-CURRENT'2013 to 11.0-PRERELEASE)
I expect these to be kept up to date with base system lifetimes,
be updated BEFORE removing any support for outdated release from
the tree and also serve as a reference of which OSVERSION checks
may be removed.
Approved by: portmgr (swills, mat)
Differential Revision: D9210
More candidates are likely to come before the official announcement (aka
the rush hour) scheduled on 2017-01-24. This one intended to pick up upstream
security fixes earlier while looking for downstream regressions.
Changes: https://www.mozilla.org/firefox/51.0/releasenotes/
Security: e60169c4-aa86-46b0-8ae2-0d81f683df09
MFH: 2017Q1 (piling up)
- client and server version 5.7.16-10
- pam-for-mysql: compatible with newer MySQL API (MySQL 5.7+)
Thanks to all who have worked on Percona, MariaDB, and MySQL ports. Your
tireless efforts have been instrumental in making this port happen.
Differential Revision: https://reviews.freebsd.org/D8785
- Add PYTHON_PYOEXTENSION and PYTHON_SUFFIX
- Add PYTHON2 and PYTHON3
- Respect PYTHON_VERSION
- Rename PYOEXTENSION to PYTHON_PYOEXTENSION
This change would help:
- Build databases/postgresql*-plpython with Python 3
(It has PLIST issue since bsd.python.mk to Uses/python.mk transition)
- Simplify Makefile
PR: 205807
Differential Revision: https://reviews.FreeBSD.org/D4758
Exp-run by: antoine
This is a development snapshot of upcoming MySQL 8.0.
PR: 215738
Submitted by: Mahdi Mokhtari <mokhi64@gmail.com>
Reviewed by: matthew, feld
Approved by: matthew (mentor), feld
Differential Revision: https://reviews.freebsd.org/D9031
- us.pear.php.net - dead, domain name does not resolve
- de.pear.php.net - not updated, does not contain new packages, creates problems for repackaged modules (see ports/208532)
Approved by: portmgr (mat)
Differential Revision: D8913
- Minimize the differences with editors/emacs.
- Do not require GCC to work around a clang bug. The OS versions for
which this was relevant are no longer supported.
- The check for X11 is handled with the X11 option and _IMPLIES, so
remove the redundant .if logic.
Approved by: swills (mentor) asish (maintainer)
Differential Revision: https://reviews.freebsd.org/D8796
This adds kde-applications to categories that can be used to setup
MASTER_SITES and such for kde ports.
With that we are now ready to update KDE4 and look at importing at least
some of the KF5 based aplications from the testing repo.
Reviewed by: rakuco
Approved by: rakuco (mentor)
Differential Revision: https://reviews.freebsd.org/D8893
Release notes: https://www.kde.org/announcements/kde-frameworks-5.29.0.php
New ports:
* textproc/kf5-syntax-highlighting (new in 5.28.0)
* graphics/kf5-prison (new in 5.29.0)
Fixes:
* kf5-kservice: Move applications.menu to kf5-applications.menu to stop conflicting with kdelibs4.
* Define conflicts in the kde4-l10n ports (debian does the same).
* kf5-kapidox: Fix with python3
Thanks to Matthew Rezny <matthew@reztek.cz> for looking at the conflicts in 214528,
and Dima Panov (fluffy@) for the python3 fix.
PR: 214528
Reviewed by: rakuco, mat
Approved by: rakuco (mentor)
Differential Revision: https://reviews.freebsd.org/D8776
As noticed in x11-toolkits/qt5-charts we never corrected the way in which qmake
sets the pkgconfig path for Qt5-modules. Instead there was a sed-call on the
generated Makefiles in bsd.qt.mk in the target qt-pre-install.
This diff modifies devel/qmake5 to set QMAKE_PKGCONFIG_DESTDIR to
../libdata/pkgconfig from the default pkgconfig.
As we split Qt-base into multiple ports, simply patching devel/qmake5 is not
enough, as the other base-ports will use the file from inside the tarball.
Therefore add it as an extrapatch for the Qt-base ports.
As this affects the same file extrapatch-libtool in devel/qt5 touched, rename
this one and add a hunk to do this change too.
Note: qt4 ports are unaffected, as they never were converted to used the
qt-preinstall target, and still contain the sed-call in their Makefiles.
Reviewed by: rakuco
Approved by: rakuco (mentor)
Differential Revision: https://reviews.freebsd.org/D8713
As at the moment QT_BINDIR and QT_LIBDIR are 'bin' respectively 'lib' depending
on the Qt version these subs ended up at many wrong places in plists.
So only export it if QT_DIST is set.
PR: 210227
Reviewed by: mat, rakuco
Approved by: portmgr (mat), rakuco (mentor)
Differential Revision: https://reviews.freebsd.org/D8378
Using ${GREP} was failing when grep was built with the WITHOUT_GNU_GREP_COMPAT
knob in src.conf. According to POSIX, | only works for alternation in extended
regular expressions (grep -E), although GNU grep disregards the difference.
The new syntax with ${EGREP} and no escaping of special characters works with
both BSD grep (without WITHOUT_GNU_GREP_COMPAT) as well as GNU grep.
Approved by: mat
Differential Revision: https://reviews.freebsd.org/D8663
- Instead of excluding the specific packages added to DISTFILES (but only for
FreeBSD 9+), exclude all files with a '.nupkg' extension.
- Move the nuget-extract command to be before post-extract, to allow the port's
post-extract command to move packages around
audio/libgpod-sharp: add port for mono bindings for libgpod
- audio/libgpod: [2]
- USE_GNOME: add missing dependency
- Remove mono files from pkg-plist (now in audio/libgpod-sharp)
- Allow relevant variables to be overridden
- Fix MONO options
- Patch source to allow building with modern mono
- Avoid post-install target for slave ports.
devel/monodevelop: update to 4.6.2.7
- Update nuget package Fantomas to 2.4.0 (latest at time of monodevelop's
release).
- Downgrade nuget package FSharp.Compiler.CodeDom (per paket's specification)
- LICENSE: Add license of LGPL21
- LICENSE_FILE: Add license file from source
- PORTSCOUT: Avoid checking alpha releases (aka x.y.0.z releases)
games/openra: simplify post-extract [3]
- USES=mono:nuget allow overridding of NUGET_PACKAGEDIR
- Set NUGET_PACKAGEDIR to correct location, fix patching accordingly
lang/fsharp: update to 4.6.2.7
ChangeLog:
- Fix#639: Problems with F# scripts on Mono
- Fix FSharp.Compiler.Tools targets for .NET Core usage
- Fix FSharp.Compiler.Tools package dependencies
lang/mono: update to 4.6.2.7
Mono 4.6.2 is the 1st service release to the 4.6 series.
Changes:
- LICENSE: Add license of MIT
- LICENSE_FILE: Add license file from source
- PORTSCOUT: Avoid checking alpha releases (aka x.y.0.z releases)
Bug fixes:
- #44708 - "TrustFailure (The authentication or decryption has failed.) ...
Invalid certificate received from server." with "Error code: 0x5"
or "Error code: 0xffffffff800b010f" when attempting to access
HTTPS servers on ports other than 443
- #39832 - SIGSEGV when running roslyn
multimedia/banshee: bump port revision.
- Update missing port dependencies
- Convert to use option helpers
- OPTIONS: Add APPLEDEV (missing dependency added)
- LICENSE: Add MIT license
- LICENSE_FILE: Add license file from source
- PORTSCOUT: Skip alpha release (2.9.1)
x11-toolkits/gnome-sharp20: fix port dependencies
- Add missing dependencies
- Convert post-patch to shebangfix
- LICENSE: add LGPL21 license
- LICENSE_FILE: use license file from source
- MKAE_JOBS_UNSAFE: remove, does appear to work
x11-toolkits/gtk-sharp20: update to 2.12.42.
PR: 214680 [2]
PR: 214681
Suggested by: jbeich [1]
Submitted by: jbeich [3]
Approved by: avilla (maintainer; timeout) [2]
As devel/qt5-help is the only consumer of textproc/clucene-qt5 and the latter
it is part of the same distfile we can simplify it, and get rid of clucene-qt5.
This was spotted by Marie Loise Nolden <nolden@kde.org> -- thanks :)
Reviewed by: rakuco
Approved by: rakuco (mentor)
Sponsored by: https://reviews.freebsd.org/D8638
The dependency code from kde.mk can add LIB_DEPENDS if the sense-file is a
library. Import this into bsd.qt.mk
PR: 214744
Reviewed by: rakuco, mat
Approved by: mat (mentor), rakuco (mentor)
Exp-run by: antoine
Differential Revision: https://reviews.freebsd.org/D8602
Collection as well as the lang/gcc port from GCC 4.8.5 to GCC 4.9.4!
See http://gcc.gnu.org/gcc-4.9/changes.html for an extensive list of
changes and http://gcc.gnu.org/gcc-4.9/porting_to.html for information
on how to port to that new version (if necessary).
files/java-patch-hier required adjustments, gcc/files/patch-arm-libcpp
is not needed any longer (merged upstream), and we're also loosing the
local Stack Protector patches/backports.
PR: 196712
Tested by: antoine (-exp runs)
Supported by: antoine, kwm, and others
of x11/qterminal).
LXQt is the Qt port of and the upcoming version of LXDE, the Lightweight
Desktop Environment. It is the product of a merge between the LXDE-Qt and
Razor-qt projects.
This work is based on the LXQt testing repo [1]. Importing is not complete
due to missing pieces in KDE Frameworks (r426118).
I would like to thank contributor and tester, Jesper Schmitz Mouridsen.
[1] https://subversion.assembla.com/svn/lxqt/tags/0.11
Reviewed by: antoine, rakuco
Obtained from: LXQt repository
FC after:
Differential Revision: https://reviews.freebsd.org/D5002
KDE Frameworks is a collection of libraries and software frameworks by KDE
that serve as technological foundation for KDE Plasma 5 and KDE Applications
distributed under the GNU Lesser General Public License (LGPL) [1].
The work is based on what we have in the KDE testing repo [2].
This is the next big step in updating the KDE Desktop and its Applications
to anything less dusty.
With this change, `USES=kde:5` is now a valid option. Ports that need to depend
on KDE Framework can now set:
USES=kde:5
USE_KDE=<framework1> <framework2> ... <frameworkX>
For example: www/qupzilla-qt5 can depend on sysutils/kf5-kwallet via:
KWALLET_USE= KDE=wallet
I would like to thank Raphael and Adriaan for reviewing the ports in the testing
repo :)
[1] https://en.wikipedia.org/wiki/KDE_Frameworks
[2] http://src.mouf.net/area51/log/branches/plasma5
Reviewed by: rakuco, mat, groot_kde.org
Approved by: rakuco (maintainer)
Differential Revision: https://reviews.freebsd.org/D8329
Update PyQt from 5.5.1 to 5.6.2 to match the Qt updates that have
already landed in the ports tree.
PR: 214355
Submitted by: Adriaan de Groot <groot@kde.org>
Reviewed by: rakuco, tcberner
Approved by: rakuco (mentor)
Exp-Run by: antoine
Differential Revision: https://reviews.freebsd.org/D8408
The main problem slowing down those macros was that it was ending up
calling chmod(1) once for each file and directory.
It was doing this because it was running find(1) in the source tree and
needed to chmod in the target tree, so it had to append the full path to
the chmod(1) calls, which made it impossible to use -exec +.
Rewrite the -exec calls to call sh and cd into it before running
chmod(1) and do so on as many files as we can using the -exec + construct.
While there, optimize a bit more after figuring out that it is possible
to only use one find(1) to run two -exec for different options.
Also switch to using the more powerfull modern regular expressions
instead of the usual basic regular expressions by using find -E. This
will allow the use of | which as no basic regular expressions equivalent
(as in, for example, "*.(html|txt)") as well as the + and ? keywords,
and back references.
About the speed gain, I did a few tests, using time's builtin from csh (table
with all the data available in the review), comparing the home baked versions
of COPYTREE_SHARE found in ports, it boils down to:
| version | www/sit (~2k) | www/mathjax (~30k) |
+------------+------------------+--------------------+
| home baked | 0.032u 0.104s | 0.946u 7.446s |
| current | 0.281u 0.649s | 5.446u 18.626s |
| new | 0.024u 0.081s | 0.770u 7.012s |
PR: 213614
Submitted by: mat
Exp-run by: antoine
Sponsored by: Absolight
Differential Revision: https://reviews.freebsd.org/D8283
