| Commit message (Collapse) | Author | Age | Files | Lines |
| |
|
| |
|
| |
|
| |
|
|
|
|
|
|
|
|
| |
We handle KeyboardInterrupt instead of installing a SIGINT handler,
since exiting from signal handlers intermittently causes python to
ignore the SystemExit exception with a message like this:
Exception SystemExit: 130 in <function remove at 0x7fd2146c1320> ignored
|
|
|
|
| |
Display._set_root_columns() for consistency with Display._set_no_columns().
|
| |
|
| |
|
| |
|
| |
|
|
|
|
| |
available_eclasses(), eclass_path() and license_path() functions.
|
| |
|
|
|
|
|
|
| |
This fixes a regression like bug #403895, introduced in commit
b696337bf20fdc539ce7721df7a4b42b35999705, since functions have to
return True in order to be continuously scheduled.
|
|
|
|
|
|
| |
Move the 30 second default to the Scheduler class, since that's the
only place that it's currently needed (all other schedulers have
relatively short-running jobs).
|
|
|
|
|
| |
With no argument, removes a previous load limit (same behavior as
make).
|
|
|
|
|
|
| |
It isn't used externally anymore, since SchedulerInterface is used
directly in those places now. Many of the self.sched_iface references
updated here, it's more appropriate to use self._event_loop.
|
|
|
|
|
|
| |
SchedulerInterface suffices for all of these cases.
EventLoop(main=False) is used for thread safety where
API consumers may be using threads.
|
| |
|
|
|
|
|
| |
The PollScheduler class suffices here, if we just add a small amount of
logic to calculate the _background attribute from the emerge opts.
|
| |
|
| |
|
|
|
|
|
|
|
| |
These methods were aliases for the EventLoop io_add_watch and
source_remove methods. Migrating to the EventLoop method names allows
an EventLoop instance to substitute for a PollScheduler inside
subclasses of AbstractPollTask.
|
|
|
|
| |
This is variable is only needed by the Scheduler class.
|
|
|
|
| |
This allows the QueueScheduler class to be eliminated.
|
|
|
|
|
| |
Also add missing __slots__ to ForkProcess. TODO: Share code
between ForkProcess and MergeProcess.
|
|
|
|
|
|
|
|
|
|
| |
Update manifest files, and sign them if signing is enabled. This
supports parallelization if enabled via the --jobs option. The
--thin-manifests and --sign-manifests options may be used to manually
override layout.conf settings. There's also a new --strict-manifests
option that may be used to manually override the "strict" FEATURES
setting, a --gpg-key option to override PORTAGE_GPG_KEY, and a
--gpg-dir option to override PORTAGE_GPG_DIR.
|
| |
|
|
|
|
| |
See http://forums.gentoo.org/viewtopic-t-938022.html for example.
|
| |
|
| |
|
| |
|
| |
|
|
|
|
| |
This makes it easier to add new types, like HDEPEND.
|
| |
|
| |
|
|
|
|
| |
This allows them to be looked up outside of the USE_EXPAND loop.
|
|
|
|
|
|
| |
This will be useful for adding flags that behave specially in
experimental EAPIs, such as the targetroot/sysroot flag which is
planned for EAPI 5-hdepend.
|
|
|
|
|
|
|
|
| |
This only affects r'\w' with Python 2.x, since Python 3 already
defaults to re.UNICODE behavior when compiling unicode str objects
(unless re.ASCII is specified). If a repository wants to ban unicode
categore/package names then we can add a layout.conf setting for that,
as discussed in bug #435934.
|
|
|
|
|
|
|
|
|
| |
EAPI 5 supports automatic rebuilds via the slot-operator and sub-slots,
which makes preserve-libs much more user-friendly, since it makes
@preserved-rebuild unnecessary (also see bug #364425 for explanation of
@preserved-rebuild shortcomings). Therefore, enable preserve-libs for
the stable branch, but not by default. After EAPI 5 is widely adopted,
we can consider enabling preserve-libs by default.
|
|
|
|
|
| |
This will allow for the option to be available in the stable branch
without changing the default --depclean behavior.
|
| |
|
| |
|
|
|
|
|
| |
The fetch_symbol variable was always just a space, and no return value
was really needed.
|
|
|
|
|
| |
This is for reinstall due to slot or sub-slot change of a dependency
when the := slot operator is used.
|
|
|
|
|
| |
This should not change the behavior at all, but it makes the display
much easier to modify or extend.
|
|
|
|
|
|
| |
Also, make options like emerge --newuse ignore the state of USE=test,
since users typically don't want to trigger a bunch of rebuilds when
they enable or disable FEATURES=test.
|
|
|
|
|
|
|
|
|
| |
This is the default location used by current versions of eselect
profile, and will soon be the default location in stages. Since
eselect profile generates a warning if both /etc/make.profile and
/etc/portage/make.profile exist, the new default should be fairly
safe (although eselect profile currently prefers /etc/make.profile
when both exist).
|
|
|
|
| |
See bug #332217, comment #24.
|
| |
|
| |
|