| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
| |
Signed-off-by: Ionen Wolkens <ionen@gentoo.org>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Few ebuild changes, but for some notes:
* Meant to use system's lua+sol2 (newly bundled), but it does not seem
to have a switch to use system's easily and not going to bother on
such a big package if not.
* The bits requiring Go have not landed in 14 and remain 9999-only.
* Dropped the comment about Qt's version given cmake files may
optionally require yet a newer one for specific features.
* There is a new switch for building docs, so can simplify a bit.
* Thought to try unbundling syntax-highlighting but after finding
out it uses the old slot opted to keep it like that for now.
May possibly have overlooked something given there is a lot of
noise in cmake files, please report a bug if anything is off.
Signed-off-by: Ionen Wolkens <ionen@gentoo.org>
|
|
|
|
| |
Signed-off-by: Ionen Wolkens <ionen@gentoo.org>
|
|
|
|
| |
Signed-off-by: Ionen Wolkens <ionen@gentoo.org>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Missed that a plugin (qtapplicationmanager) added a dependency on
yaml-cpp while comparing comparing qt-creator-12 -> 13 and is no
longer clang-only.
Feel it's not worth a USE for the plugin on such a big package
(not for minimalism), so depending unconditionally unless someone
really wants to avoid it (at best it could be interesting to avoid
:= rebuilds).
Technically needs a revbump, but qt-creator takes a while to build
and most users already were pulling it through IUSE=+clang.
Signed-off-by: Ionen Wolkens <ionen@gentoo.org>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Not tested much, but did build successfully once w/ Go bits.
May or may not consider making this optional when there is a release
after a closer look. Does not have a proper option, but it already
skips it if it does not find Go... so -DGO_BIN=no or so would likely
work.
Being off by default would also be less annoying for go rebuilds,
I doubt anyone wants to rebuild something massive like qt-creator
every go bump. Not that I really checked what this is used for yet.
For now not going to touch this too much given it may get changed
around upstream.
Closes: https://bugs.gentoo.org/935332
Signed-off-by: Ionen Wolkens <ionen@gentoo.org>
|
|
|
|
| |
Signed-off-by: Arthur Zamarin <arthurzam@gentoo.org>
|
|
|
|
|
| |
Bug: https://bugs.gentoo.org/934462
Signed-off-by: Ionen Wolkens <ionen@gentoo.org>
|
|
|
|
| |
Signed-off-by: Ionen Wolkens <ionen@gentoo.org>
|
|
|
|
| |
Signed-off-by: Ionen Wolkens <ionen@gentoo.org>
|
|
|
|
| |
Signed-off-by: Arthur Zamarin <arthurzam@gentoo.org>
|
|
|
|
| |
Signed-off-by: Ionen Wolkens <ionen@gentoo.org>
|
|
|
|
| |
Signed-off-by: Ionen Wolkens <ionen@gentoo.org>
|
|
|
|
|
|
|
|
|
|
| |
6.7.1 is (currently) due to release in ~2 days and it seems
unlikely qt-creator-13.0.1 will release before then.
Patch should be safe so doing straight-to-stable to spare
users from a revbump long rebuild.
Signed-off-by: Ionen Wolkens <ionen@gentoo.org>
|
|
|
|
| |
Signed-off-by: Ionen Wolkens <ionen@gentoo.org>
|
|
|
|
| |
Signed-off-by: Michał Górny <mgorny@gentoo.org>
|
|
|
|
| |
Signed-off-by: Ionen Wolkens <ionen@gentoo.org>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Does not seem to need real updates beside handling qtkeychain? But
may or may not have missed something given there is a lot of noise
in build system files changes.
Unfortunately upstream has not provided a switch (that I can see)
to allow using system qtkeychain[qt6] which would've been more
convenient than handling libsecret+dbus used by it.
For now opted to make dbus hard-required to simplify, doubtful
qt-creator users are minimalists in general.
Signed-off-by: Ionen Wolkens <ionen@gentoo.org>
|
|
|
|
| |
Signed-off-by: Michał Górny <mgorny@gentoo.org>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Tried to compile Qt Creator's HelloWorld plugin externally and it built
fine with the USE enabled.
Debated installing unconditionally, but it installs quite a lot of
extras. Unsure about the IUSE name choice but it should hopefully do
(did feel IUSE=devel would be a bit too generic sounding).
Went for a bit hacky solution (for now) given --component integrates
poorly with cmake.eclass and does not seem(?) other ebuilds have needed
to do this yet in ::gentoo for ideas.
Closes: https://bugs.gentoo.org/928423
Signed-off-by: Ionen Wolkens <ionen@gentoo.org>
|
|
|
|
| |
Signed-off-by: Ionen Wolkens <ionen@gentoo.org>
|
|
|
|
| |
Signed-off-by: Arthur Zamarin <arthurzam@gentoo.org>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This been set by default in the old ebuilds and opted to leave it like
that but, with Qt6 and the non-split qttools where designer is off by
default, this been annoying for users.
Default designer on qttools when most users don't need it does not
seem like the right move -- and believe users that actually want this
will know to set it themselves (perhaps even globally).
Not that I'm either a qt-creator nor designer user (just maintaining
regardless), so opinions from actual users for defaults are welcome.
(live-only to avoid long rebuilds, will propagate in next version)
Signed-off-by: Ionen Wolkens <ionen@gentoo.org>
|
|
|
|
|
|
|
|
|
| |
Currently ommited in the llvm-r1 example, but that's being changed
and >=llvm-18.1.0_rc3 will use $(ver_cut 1-2) as subslot.
No need for a revbump (slot 18 is still masked either way).
Signed-off-by: Ionen Wolkens <ionen@gentoo.org>
|
|
|
|
|
|
|
| |
Done together with the Qt6.6.2 bump which is going to trigger
a rebuild either way.
Signed-off-by: Ionen Wolkens <ionen@gentoo.org>
|
|
|
|
|
|
|
|
|
|
| |
Tested with 18.1.0_rc2 at same time, seems to be fine so may
as well do this now.
Will migrate 12.0.2 when add Qt6.6.2 given that will trigger
a rebuild either way.
Signed-off-by: Ionen Wolkens <ionen@gentoo.org>
|
|
|
|
| |
Signed-off-by: Ionen Wolkens <ionen@gentoo.org>
|
|
|
|
|
|
| |
Signed-off-by: Michał Górny <mgorny@gentoo.org>
Closes: https://github.com/gentoo/gentoo/pull/34792
Signed-off-by: Michał Górny <mgorny@gentoo.org>
|
|
|
|
|
|
| |
Signed-off-by: Michał Górny <mgorny@gentoo.org>
Closes: https://github.com/gentoo/gentoo/pull/34790
Signed-off-by: Michał Górny <mgorny@gentoo.org>
|
|
|
|
|
|
| |
Signed-off-by: Michał Górny <mgorny@gentoo.org>
Closes: https://github.com/gentoo/gentoo/pull/34786
Signed-off-by: Michał Górny <mgorny@gentoo.org>
|
|
|
|
| |
Signed-off-by: Arthur Zamarin <arthurzam@gentoo.org>
|
|
|
|
| |
Signed-off-by: Ionen Wolkens <ionen@gentoo.org>
|
|
|
|
| |
Signed-off-by: Ionen Wolkens <ionen@gentoo.org>
|
|
|
|
| |
Signed-off-by: Ionen Wolkens <ionen@gentoo.org>
|
|
|
|
| |
Signed-off-by: Ionen Wolkens <ionen@gentoo.org>
|
|
|
|
|
|
|
| |
Didn't mean to push that bit, it is now disabled by default
in qt-creator-12, but it is not *gone*.
Signed-off-by: Ionen Wolkens <ionen@gentoo.org>
|
|
|
|
|
|
|
|
|
|
| |
Not much to update, but next release is close and checking state.
Tests pass except tst_utils_deviceshell which is failing in upstream's
CI as well (unless QTC_USE_QPROCESS=false). The change causing this is
not included in 12.0.0rc1 though, so not going to worry about it.
Signed-off-by: Ionen Wolkens <ionen@gentoo.org>
|
|
|
|
|
|
|
| |
Forgot, albeit typically not an issue unless someone has an extra
llvm slot without clang.
Signed-off-by: Ionen Wolkens <ionen@gentoo.org>
|
|
|
|
| |
Signed-off-by: Ionen Wolkens <ionen@gentoo.org>
|
|
|
|
| |
Signed-off-by: Ionen Wolkens <ionen@gentoo.org>
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
I find dev-util/android-sdk-update-manager just confusing unless
*really* want to set this up for system-wide availability. Qt Creator
has a button to handle everything and open the sdk manager. Feel it
is better to remove than suggest it (not to mention that the package
is mostly unmaintained in Gentoo and not planning to pick it up).
On the other hand, jdk is required and it is simpler to use system's
(can be finnicky but is typically is autodetected).
Signed-off-by: Ionen Wolkens <ionen@gentoo.org>
|
|
|
|
| |
Signed-off-by: Ionen Wolkens <ionen@gentoo.org>
|
|
|
|
|
| |
Closes: https://bugs.gentoo.org/914732
Signed-off-by: Ionen Wolkens <ionen@gentoo.org>
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
With llvm-17.0.1 being released, keeping a -14 ebuild around
is a stretch and would be surprised if someone still relied
on this old ebuild. So let's just drop it now.
This loses keywords for arm and x86. As noted in bug #905166,
currently no intention to restore these unless someone actually
wants and requests it.
Closes: https://bugs.gentoo.org/905166
Signed-off-by: Ionen Wolkens <ionen@gentoo.org>
|
|
|
|
|
|
|
|
|
|
|
| |
It would probably be fine to pull that instead, albeit without
looking closer it may be safer to keep as-is until next version.
Note that while upstream's fix removes the private API usage for
that bit, it still uses QtQuick3D private API for other things,
so binding operator needs to stay.
Signed-off-by: Ionen Wolkens <ionen@gentoo.org>
|
|
|
|
|
|
|
|
|
|
|
| |
Mostly the same as 11.0.2 bare no longer any unexpected test
failures, minor adjustments due to newer litehtml, and allowing
clang-17.
May still be more that need changes but will give it a closer
look on release unless something comes up first.
Signed-off-by: Ionen Wolkens <ionen@gentoo.org>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Not going to list every changes given this ebuild is from scratch.
There are a few ebuilds for this going around, but they about all
have (IMO) overcustomization, with a lot of USE to juggle for both
users and maintainers (some handled through sed'ing cmake files) --
all while plugins can still be disabled at runtime either way.
So this is now mostly limited to options that introduce an extra
dependency (e.g. clang). "Some" of the longest to build plugins
do have dependencies, so this still allows to cut down build time
a decent amount (not that this is a short build still).
Other small notes:
* Default IUSEs match what Qt Creator enables by default at runtime
(albeit default USE=-tracing could be interesting for faster builds).
* Left gumbo unbundling out for now given ultimate goal is to unbundle
litehtml instead (which itself uses gumbo), but currently it requires
a large patch for 0.8 support (likewise to unbundle on qttools).
wrt bug #905166, delaying cleanup given restoring keywords is
non-trivial and users may need time to adjust for Qt5->Qt6, but
can be dropped anytime anyway if llvm:14 cleanup is happening
(do not consider this a blocker).
Bug: https://bugs.gentoo.org/905166
Closes: https://bugs.gentoo.org/835042
Closes: https://bugs.gentoo.org/905417
Signed-off-by: Ionen Wolkens <ionen@gentoo.org>
|
|
|
|
|
|
|
|
|
|
| |
xdg_pkg_postinst needs to be explicitly called since we define
our own pkg_postinst.
Closes: https://bugs.gentoo.org/879887
Signed-off-by: Peter Levine <plevine457@gmail.com>
Closes: https://github.com/gentoo/gentoo/pull/28149
Signed-off-by: Sam James <sam@gentoo.org>
|
|
|
|
|
|
| |
Signed-off-by: Peter Levine <plevine457@gmail.com>
Closes: https://github.com/gentoo/gentoo/pull/28047
Signed-off-by: Sam James <sam@gentoo.org>
|
|
|
|
|
| |
Signed-off-by: Peter Levine <plevine457@gmail.com>
Signed-off-by: Sam James <sam@gentoo.org>
|