| Age | Commit message (Collapse) | Author |
|
Signed-off-by: Matt Turner <mattst88@gentoo.org>
|
|
gentoolkit supports only Python 3.6+ now, so these are not used.
Signed-off-by: Matt Turner <mattst88@gentoo.org>
|
|
Regression from commit: 517581df206766
link: https://gitweb.gentoo.org/proj/gentoolkit.git/commit/?id=517581df206766fabf10273cde565e0a6dc62829
Gentoo bug: https://bugs.gentoo.org/748129
Signed-off-by: Brian Dolbec <dolsen@gentoo.org>
|
|
Closes: https://github.com/gentoo/gentoolkit/pull/11
Signed-off-by: Georgy Yakovlev <gyakovlev@gentoo.org>
|
|
Use separate variables for binary package and corresponding ebuild
EAPI values, since the ebuild EAPI may have changed. This avoids
a possible InvalidAtom exception as reported in bug 727596.
Bug: https://bugs.gentoo.org/727596
Signed-off-by: Zac Medico <zmedico@gentoo.org>
|
|
Signed-off-by: Michał Górny <mgorny@gentoo.org>
|
|
Signed-off-by: Michał Górny <mgorny@gentoo.org>
|
|
This also fixes 'all' to process all architectures using stable keywords
and not just these having stable profiles.
Signed-off-by: Michał Górny <mgorny@gentoo.org>
|
|
Signed-off-by: Michał Górny <mgorny@gentoo.org>
|
|
Signed-off-by: Matt Turner <mattst88@gentoo.org>
|
|
I found the original code to be nearly incomprehensible. Instead of
populating a dict of potential binpkgs to remove and then removing from
the to-be-removed list, just selectively add to-be-removed packages.
Signed-off-by: Matt Turner <mattst88@gentoo.org>
|
|
Prevent this Popen exception:
UnicodeEncodeError: 'ascii' codec can't encode character '\xe8' in position 20: ordinal not in range(128)
Bug: https://bugs.gentoo.org/709610#c0
Signed-off-by: Zac Medico <zmedico@gentoo.org>
|
|
Signed-off-by: Michał Górny <mgorny@gentoo.org>
|
|
Signed-off-by: Michał Górny <mgorny@gentoo.org>
|
|
Signed-off-by: Michał Górny <mgorny@gentoo.org>
|
|
Also print categories on the line with the package name for much simpler
consumption by external tools.
Signed-off-by: Matt Turner <mattst88@gentoo.org>
|
|
Signed-off-by: Matt Turner <mattst88@gentoo.org>
|
|
Closes: https://bugs.gentoo.org/671592
Signed-off-by: Matt Turner <mattst88@gentoo.org>
|
|
The boilerplate for calling this method was larger than what it actually
contained. Additionally I think this change will allow the loop to run
on a generator rather than a full list.
Signed-off-by: Matt Turner <mattst88@gentoo.org>
|
|
Closes: https://bugs.gentoo.org/698256
Signed-off-by: Michael Haubenwallner <haubi@gentoo.org>
|
|
Bug: https://bugs.gentoo.org/698256
Signed-off-by: Michael Haubenwallner <haubi@gentoo.org>
|
|
Signed-off-by: Zac Medico <zmedico@gentoo.org>
|
|
This copy of glsa-check is no longer need since it has
moved to the portage repository.
Bug: https://bugs.gentoo.org/463952
Signed-off-by: Zac Medico <zmedico@gentoo.org>
|
|
Closes: https://github.com/gentoo/gentoolkit/pull/4
Signed-off-by: Zac Medico <zmedico@gentoo.org>
|
|
Signed-off-by: Zac Medico <zmedico@gentoo.org>
|
|
Fix comparisons to use the result of reduce_flag.
Bug: https://bugs.gentoo.org/690786
Signed-off-by: Zac Medico <zmedico@gentoo.org>
|
|
Bug: https://bugs.gentoo.org/685538
Signed-off-by: Virgil Dupras <vdupras@gentoo.org>
|
|
Signed-off-by: Michał Górny <mgorny@gentoo.org>
|
|
Remove old "optimization". Slicing two strings can't possibly be faster
than checking if `.category` is `None`: the call to `.cp` earlier
ensures that we've already populated that property.
Also, add tests for it.
|
|
Avoid `Atom` instanciation in obvious non-matches (`.cp` not matching)
when search for revdeps. On my machine, it made `equery d -a
dev-python/pillow` go from `1m3s` to `0m19s`.
|
|
The code for it was already there, but commented out. Portage now
supports repo alright...
|
|
As of now, pylint checks return a metric ton of warnings, which tells us
that it hasn't been running lately. pylint is replaced by flake8 and its
continuous run will enventually be ensured by a CI running tox on this
project. The immediate goal, for now, is to have an easy command that
checks as much as possible, but that *passes* so that we don't litter
the project with more code that decrease quality.
|
|
Add config in tox to run flake8 checks. For now, only "F*" errors are
checked (errors reported by pyflakes). pep8 style checks are ignored for
now.
This is accompanied by changes in the code that are necessary to silence
these errors. Most of them were about unused imports, but also unsed
referenced and even undefined ones (meaning that this code path was
broken, gotta increase test coverage).
This setup will replace pylint which reports way too many errors to be
usable (see next commit).
|
|
It is not referenced anywhere and is likely broken according to
pyflakes's undefined references warnings.
|
|
The test suite (python setup.py test) was depending on a broken
interface from dev-python/snakeoil. This indicates that tests weren't
run in a while.
We don't actually need snakeoil to run tests. Replaced the test command
with a plain distutils Command.
Moreover, one of test Query tests wasn't in sync with the code. Changed
it to reflect current code.
|
|
This adds a small convenience for displaying the ebuilds in reverse
order. This helps people prune packages that grow quickly as the
most relevant ones are at the top.
|
|
Always group all Prefix keywords after other types of keywords. This
not only ensures that fbsd sorts first but more importantly stabilizes
the LHS output between regular and -P variant -- that is, -P always adds
additional keywords at the end.
|
|
Sort keywords by reversed component list rather than combined name,
in order to group Prefix keywords by system. Requested by Ulrich Müller.
|
|
To match the new ordering on Bugzilla, order arches with pure ~arch
keywords after those having stable keywords. The idea behind it is to
make it easier to determine which arches to CC on stablereqs.
The pure ~arch arch list is hardcoded for now to mips + *-* (fbsd,
prefix). However, in the future it will be replaced by status defined
by arches.desc once that GLEP is finalized.
|
|
Since dev profile status implies working towards stable status, and we
are currently working on improving the status of dev arches
(in particular preventing new depgraph breakages from being committed),
we have switched Bugzilla to display both statuses alongside. Update
eshowkw to use the same output ordering.
|
|
|
|
Add component count (implied by the count of '-' separators) to sort
criteria. This ensures that *-fbsd sorts after regular arches, much like
we list it in Bugzilla.
|
|
|
|
|
|
|
|
|
|
Encode the filename before calling os.lstat
X-Gentoo-Bug: 531636
X-Gentoo-Bug-URL: https://bugs.gentoo.org/531636
|
|
os.path.exists() returns false on broken symlinks. We don't actually
care if the symlink target exists.
X-Gentoo-Bug: 619472
X-Gentoo-Bug-URL: https://bugs.gentoo.org/619472
|
|
The portage.root variable is using late binding. This breaks in Python3.6
where the type is being checked before use in os.path.join(). This fix
creates a new variable of the correct type from the value of portage.root
instead of using portage.root directly.
X-Gentoo-bug: 617498
X-Gentoo-bug-url: https://bugs.gentoo.org/show_bug.cgi?id=617498
Acked-by: Brian Dolbec <dolsen@gentoo.org>
|
|
|