pkgcheck
pkgcheck copied to clipboard
pkgcore
pkgcore-based QA utility for ebuild repos
This is a proposal for a "soft" version of `NonsolvableDepsIn*`, which would detect when a package is forcing an old version of a dependency to be pulled in, even though...
### Is there an existing such new check request for this? - [X] I have searched the existing issues ### Explain Any use of `declare` is local scope unless a...
### Is there an existing such new check request for this? - [X] I have searched the existing issues ### Explain - [ ] existing package - [ ] each...
This is WIP, showing results across the repo, but with possible false positives (since we compare on `cat/pkg` level instead of `cat/pkg-ver` levels). Resolves: https://github.com/pkgcore/pkgcheck/issues/400
Tested at ::gentoo commit `71d0d48fe392b55b1ed9fdd3727d9dd4a9310ca2` in net-dns/unbound. ```ebuild $ head -n 45 unbound-1.17.0.ebuild # Copyright 1999-2023 Gentoo Authors # Distributed under the terms of the GNU General Public License v2...
It would be nice if `pkgcheck scan --commit` would report if the newly added package matches an existing `move` entry, e.g. if I add `app-text/pytextile` I should get a warning...
pkgcheck doesn't seem to emit a `PythonCompatUpdate` result for me on dev-python/jinga2_pluralize at commit `fd6b3fcdbc4106ca78401042f5e93e561263d458`: ```shell $ git checkout fd6b3fcdbc4106ca78401042f5e93e561263d458 $ pkgcheck scan [nothing] $ grep -rsn 12 jinja2_pluralize-0.3.0-r1.ebuild [nothing]...
Add support for an ebuild to declare a skip annotation for a class of **version** results. This skip annotation must be declared before any non-comment non-empty line (most of the...
I don't seem to get a `MatchingChksums` result when I expected one: 1. `git checkout 659cab9931ee5424de3e993d656d9d3b9cb6dca3~1` 2. `cd sec-keys/openpgp-keys-wget && cp openpgp-keys-wget-20220407.ebuild openpgp-keys-wget-20230313.ebuild` 3. `pkgdev manifest` 4. `pkgcheck scan` with:...
@mgorny and I were discussing how Gentoo handles last-rites and removed packages, and I noted that Fedora requires developers to announce on their mailing list that an ex-package is being...