Update to 1.2.5. JB#52079 #1
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
With much help from Martin Blanchard <tchaik@gmx.com> too, thanks to all. https://wiki.gnome.org/Initiatives/GnomeGoals/MesonPorting
The pisi package manager is no longer used as Pardus is Debian-based now.
Mageia 6 switched to DNF backend exclusively and Mageia 5 is EOL. ROSA also switched to rpm.org + libdnf.
Even macports doesn't carry it anymore.
This is only used in RHEL 7 which is not going to get rebased to this brand new PackageKit branch.
Fix typo
This restores the original build of the GIR with automake.
Previously aptcc would report some warnings as errors. This way, it will drop any warning if there is no error. This matches the behavior of python-apt. As a bonus, we also log the entire message as a warning to the journal.
This prevents an apparently infinite loop due to automatic retrying. We cannot be sure whether we had a GPG error and/or other errors, so raising an error that causes automatic retrying is a bad idea. This will cause PK_ERROR_ENUM_CANNOT_FETCH_SOURCES to be emitted by the caller of the function.
Currently packagekit exits on startup when the option was specified: packagekitd[18705]: Failed to initialize alpm: using /etc/PackageKit/alpm.d/pacman.conf: /etc/PackageKit/alpm.d/pacman.conf:8/etc/pacman.conf:41unrecognised directive 'DisableDownloadTimeout' This change adds support for DisableDownloadTimeout and fixes the error.
According to upstream doc [1], total being -1 is a signal for download initialized. We should avoid giving unnecessary assertion failures for every file download. [1] https://git.archlinux.org/pacman.git/tree/lib/libalpm/dload.c?id=e76ec94083235ddc5510ab57b7c2bc12a1d34e8a#n134
No matter whether you are using Tumbleweed or Leap, update-packages is the place where all the action happens.
In commit e1a6975, updating is handled in update-packages whether you are using Tumbleweed or not. So remove upgrade-system implementation.
Rather than having them run one after the other, it makes sense to have them run independently in parallel. This also makes it clearer in the job logs, since now each build is an independent CI job.
Use correct function call and add missing comma.
Previously, when users are installing multiple packages at the same time using "pkcon install", if one/some of those packages is/are already installed, the whole installing process terminates, withouting installing those uninstalled packages. Fix that by ignoring those already installed pacakge and proceeding to install those which are not yet installed.
packagekit/backends/alpm/meson.build:35:4: ERROR: Expecting rbracket got string.
packagekit/backends/alpm/meson.build:47:2: ERROR: Expecting rparen got string.
packagekit/backends/alpm/meson.build:3:0: ERROR: Tried to access unknown option "confdir".
the latest stable version of libalpm
Upstream commit 310c944 changed this to filter based on primary arch, but a) it's wasn't taking into use the configured arch, b) pkcon install -> pk_console_install_packages() uses enum_arch by default presumably to get compatible packages for which there are no better filters either. Just assuming pkcon knows better so it's possible to pkcon install i586 package for x86_64 device with i486 runtime. Like Jolla tablet.
…created to install local packages from file.
Squashed with: Move normal cache dir to /home too. JB#45137
Previously simulate would not be possible with dist-upgrade mode. Change it so that simulate can be run unless in extended download size mode.
Make it so upgrade kind complete with only download -flag is identical in behavior with upgrade kind minimal.
Maintain nemo::query-size: for backwards compability.
Send size details using details signal.
With all the cache switching make sure that our cache is up to date. Refresh is always run but without forcing the update repositories are removed but not refreshed as autorefresh option for the repositories is false. See zypp_refresh_cache() and comment "do as zypper does".
When first package in resolve query is special string 'ext::data:repo' packages emitted have repository as data field always. If the package is installed but doesn't have repository return special @System repository.
[PackageKit] Reset update mode along with upgrade during zypp reset. Fixes JB#50165 Zypp pool is reset when the cache changes, but update mode is not reset. That causes any pending package updates to be performed on any transaction.
If it gets synced during simulations, the repo data of the old version is lost and upgradeable packages will not resolve.
Given a package that is required by other package(s), the allow_deps argument is meant to control whether it may remove the package and everything requiring it or fail. The argument to zypp_perform_execution that was originally hardcoded as TRUE is forwarded to zypp::Resolver::setForceResolve. Handling allow_deps this way, although not ideal, is sufficient to fix our use case and shouldn't make things worse.
This feature has been implemented on zypp backend Added a new D-Bus API method to PackageKit: org.freedesktop.packagekit.import-pubkey
This feature has been implemented on zypp backend Added a new D-Bus API method to PackageKit: org.freedesktop.packagekit.remove-pubkey
When querying for the size of the upgrade, the space check was running before the numbers were returned, and an error on lack of download or installation space overrode the real space check. Move the simulation report above the error generation.
|
Just wondering with which libzypp version the upstream tested, is our libzypp version still fine? |
Thaodan
reviewed
Oct 5, 2022
| BuildRequires: gettext | ||
| BuildRequires: libarchive-devel | ||
| BuildRequires: fontconfig-devel |
There was a problem hiding this comment.
This was only needed to the gtk backend (just adding why it was removed).
Thaodan
requested changes
Oct 5, 2022
Weell, the actual upstream probably didn't test any version but rather the people doing contributions so that's probably what is there on some newish opensuse. Then again the changes don't generally seem that big or related to libzypp behavior changes. |
Thaodan
approved these changes
Oct 5, 2022
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Updated to new version following the README.sailfishos.
Some notable details:
Not much testing so far. Maybe the thing I'd be the most cautious is if the upstream zypp backend changes are all ok. Didn't go through those with thought yet.