The reason is that modules2tuple is a tool useful to port maintainer to
generate contents for GH_TUPLE for ports written in Go.
PR: 237907
Approved by: Dmitri Goutnik <dg@syrec.org> (maintainer)
- Extend data model to allow for target command formatting if turned
on with the new -t flag
- General parser improvements
- Sort USE_PYQT like USE_QT
- Add preliminary support for WIP USES=cabal options helpers/variables
- Add a portedit command which instead of formatting the entire
file only formats edited portions of it. As a proof of concept
this for now only has a 'bump-revision' command to bump PORTREVISION
of a port, and a 'get' command to return unevaluated variable
values as parsed by portfmt.
Changes: ff04b7c3f4...5f6ab52521
Changes from pkg 1.10.99.12 to 1.10.99.13
- fix segfault when running pkg version <an_invalid_index_file>
- fix segfault that can occurs sometime when forcing the installation of a
localpackage: pkg install -f ./mypkg-X.txz
- fix abi detection when multiple elf notes are used in the binary like on
fedora
Reported by: tobik, sunpoet
From the ChangeLog: https://github.com/ppekala/ports-tools/releases/tag/1.8
* pfind:
* add -d option for searching deleted ports
* support searching for slaves in other categories then master port
* add information about already installed packages in search results
* upgrade-ports:
* don't use full dependencies list while sorting, this prevents out of
memory problems
* while sorting large list of ports
* fix uprade of security/sudo package when using sudo for upgrade
process (non-root user)
* inform user about lenghty backups
* don't attempt to big version upgrade of perl5, needs manual intervention
* make required by list more compact in need to be installed summary
and many small fixes.
Reported by: portscout
* Add check for sensible CMAKE settings to ports-mgmt/portlint [1]
* Do not warn about absolute paths in *_OLD_CMD (when USES=shebangfix is set) [2]
* If an option only has QMAKE_ON or QMAKE_OFF settings, it is flagged as no effect [3]
* Check for options defined per ARCH to make sure we don't have a lot
of false positives. [4]
(Note: [1] was already in PORTREVISION 2).
PR: 232948 [2]
237045 [3]
234888 [4]
Submitted by: adridg [3]
If a port sets CMAKE variables (CMAKE_ARGS, or the option helpers
like foo_CMAKE_BOOL) then it should have USES=cmake; otherwise
those variables don't make any sense.
This is the slightly simplified version of the patch: there
really shouldn't be options that switch (meta-)build systems.
PORTREVISION bumped because I'm not sure what the versioning
scheme is for portlint.
PR: 235650
Reported by: yuri
Approved by: marcus (maintainer timeout 2 weeks)
Differential Revision: https://reviews.freebsd.org/D19525
2019-03-25 chinese/reciteword: depends on audio/esound
2019-03-31 comms/conserver: Superseeded by comms/conserver-com
2019-03-24 databases/flamerobin: Abandonware since 2009, does not work with newer version of wxGTK
2019-04-01 databases/mariadb100-client: Upstream End-of-Life March 2019 (https://mariadb.org/about/maintenance-policy/)
2019-04-01 databases/mariadb100-server: Upstream End-of-Life March 2019 (https://mariadb.org/about/maintenance-policy/)
2019-03-12 devel/physfs-devel: is the same as devel/physfs, serving no purpose
2019-03-20 devel/py-construct28: Latest upstream version is in ports tree devel/py-construct
2019-03-31 devel/xtoolchain-llvm40: No supported release uses this version
2019-03-31 devel/xtoolchain-llvm50: No supported release uses this version
2019-03-31 finance/aqmoney: no longer maintained upstream
2019-03-31 finance/openhbci: no longer maintained upstream
2019-03-31 graphics/py-exiv2: py-exiv2 has been deprecated in favour of GExiv2
2019-04-04 mail/dbmail22: Please use mail/dbmail which uses latest version of 3.X
2019-03-31 net/rubygem-fog-google17: Use net/rubygem-fog-google instead
2019-03-31 ports-mgmt/bsdadminscripts: Upstream stale nearly 10 years, no longer works properly, use ports-mgmt/bsdadminscripts2 instead
2019-03-24 www/mod_md-devel: Use the module that is shipped with Apache 2.4
2019-03-25 x11/xbelld: Depends on audio/esound
Changes from pkg 1.10.99.11 to 1.10.99.12
- use /usr/bin/uname to determine of the ABI
- Add 3 new aliases:
* rall-depends: showing depenencies of a non installed package
* rcomment: showing only the comment of a non installed package
* rdesc: showing only the description of a non installed package
- switch to cirrus-ci
- improve portability on linux
- simplify code to create the repository
- close stdin when running scripts
- update bundled libucl to latest version
- remove backward compat (pkg_*) in periodic scripts
- always use anticongersion on periodic scripts
- Prevent SIGSEGV with 'pkg version' with packaged base
- Fix issue with recent lld
Portfmt is a tool for formatting FreeBSD Ports Collection Makefiles.
For the time being portfmt concentrates on formatting individual
variables as such it does not move variables to preferred positions.
WWW: https://github.com/t6/portfmt
- poudriere: Only pass MAKEOBJDIRPREFIX to 'jail'
- Document environment fully for poudriere/-jail/-ports.
- Revert progress indicator
- poudriere: Fix all environment to be space/quote safe.
- jail/ports -l: Avoid printing empty columns/trailing whitespace.
- jail -c -m null/tar/src: Again require -v if newvers.sh is not readable.
- Show proper errors for missing jail metadata.
- poudriere-jail.8: jail -m null does NOT null-mount during build.
- jail -c: Defer pwd_mkdb until bulk's jail_start() call.
- jail -cu: Stop modifying /etc/login.conf in the jail.
- Slightly speedup MOVED parsing
- Fix symlinked POUDRIERE_DATA unmount bug.
- testport: Fix ports-mgmt/pkg Latest link leaking into real package repository.
- bulk/status: Display the progress indicator extracted from build logs.
This will be 3.3.0.
Upstream has no updates for nearly 10 years, pkg_libchk command no longer
detects missing libraries correctly. Switch to ports-mgmt/bsdadminscripts2
instead. It is redeveloped new one written by the same author.
PR: 236001
Submitted by: myself
- Update sh from head r343981
- Add a builtin for simple string substitutions
- Fix some invalid memset(3)
- mapfile fixes and tests
- Micro-optimize some hash functions to save around 30%
- Inline some mapfile checks
- Manpage updates/fixes
2019-01-31 mail/dovecot-pigeonhole04: End of Life upstream, use mail/dovecot-pigeonhole instead
2019-01-31 multimedia/pyjama: Unmaintained upstream
2019-01-31 devel/py-omniorb-3: Uses legacy version of omniORB, consider using devel/py-omniorb
2019-01-31 mail/dovecot22: End of Life upstream, use mail/dovecot instead
2019-01-31 devel/hs-uuagc-bootstrap: No release since 2011
2019-01-31 sysutils/hs-angel: No releases since 2016
2019-01-31 devel/hs-uuagc: No release since 2015
2019-01-31 ports-mgmt/hs-porte: No updates since 2010
2019-02-01 net/pdb: Depends on expired net/py-pcs
2019-02-01 irc/iroffer: Abandoned upstream
2019-02-03 sysutils/fusefs-wdfs: Abandonware, functionally incomplete, has problems with caching
2018-12-19 net/py-pcs: Broken for more than 6 months
a symbol matches multiple clauses the last one takes precedence. If the
catch-all is last it captures everything. In the case of Qt5 libraries
this caused all symbols to have a Qt_5 label while some should have
Qt_5_PRIVATE_API. This only affects lld because GNU ld always gives the
catch-all lowest priority.
Older versions of Qt5Webengine exported some memory allocation symbols from
the bundled Chromium. Version 5.9 stopped exporting these [1] but the
symbols were kept as weak wrappers for the standard allocation functions to
maintain binary compatibility. [2][3] The problem is that the call to the
standard function in these weak wrappers is only resolved to the standard
function if there's a call to this standard function in other parts of
Qt5Webengine, because only then is there a non-weak symbol that takes
precedence over the weak one. If there's no such non-weak symbol the call
in the weak wrapper resolves to the weak wrapper itself creating an infinite
call loop that overflows the stack and causes a crash. Some of the
allocation functions are variants of C++ new and delete and it probably
depends on the compiler whether these variants are used in other parts of
Qt5Webengine.
Remove the weak wrappers (make them Linux specific). This isn't binary
compatible but we are already breaking that with the changes to the symbol
versions.
[1] 5c2cbfccf9
[2] 2ed5054e3a
[3] 009f5ebb4b
Bump all ports that depend on Qt5.
PR: 234070
Exp-run by: antoine
Approved by: kde (adridg)
- switch from GNU diff to BSD diff needs to be handled
PR: 234769, 234763
Reported by: Lapo Luchini <lapo@lapo.it>
Submitted by: alexander@wittig.name (maintainer)
