- Pacman-dev - lists.archlinux.org

[Git][pacman/pacman][release/7.1.x] 2 commits: makepkg: fix generation of "--allsource" packages with VCS sources
by Allan McRae (＠allan) 12 Dec '25

12 Dec '25

Allan McRae pushed to branch release/7.1.x at Pacman / Pacman Commits: 3cd55f6f by Allan McRae at 2025-12-12T21:41:03+10:00 makepkg: fix generation of "--allsource" packages with VCS sources The globbing used to get a list of files to reproducibly include in a source package does not recurse through directory symlinks (by design). For any VCS source, recursively "copy" the source into the srclinks directory for inclusion in the final source package. Fixes #290 Signed-off-by: Allan McRae <allan(a)archlinux.org> (cherry picked from commit fd97639c07a4261a037b1bc427e29c60a7652119) - - - - - b9f7d4a5 by Allan McRae at 2025-12-12T21:41:12+10:00 Partially revert resolving paths in pacman.conf For the CacheDir and HookDir options, multiple paths can be specified on a single line (separated by a space). The resolving of paths does not handle multiple entries and causes failures. Additional, directories in CacheDir are created by libalpm if they do not exist, and this is not accounted for when realpath fails due to a non-exisiting directory. "Fixes" #289 and #292. Signed-off-by: Allan McRae <allan(a)archlinux.org> (cherry picked from commit de68378f9c8e60963b76abdc770900b0e16a6cf4) - - - - - 2 changed files: - scripts/makepkg.sh.in - src/pacman/conf.c View it on GitLab: https://gitlab.archlinux.org/pacman/pacman/-/compare/2d0d5d7b317daafe64ac54… -- View it on GitLab: https://gitlab.archlinux.org/pacman/pacman/-/compare/2d0d5d7b317daafe64ac54… You're receiving this email because of your account on gitlab.archlinux.org.

1 0

[Git][pacman/pacman][master] pacman.8: remove "%m" from --print-format
by Allan McRae (＠allan) 12 Dec '25

12 Dec '25

Allan McRae pushed to branch master at Pacman / Pacman Commits: 6cdca15c by Diego Viola at 2025-12-12T21:24:30+10:00 pacman.8: remove "%m" from --print-format --print-format no longer takes "%m" as argument after commit bdedf621. Update the documentation to match this change. Signed-off-by: Diego Viola <diego.viola(a)gmail.com> - - - - - 1 changed file: - doc/pacman.8.asciidoc View it on GitLab: https://gitlab.archlinux.org/pacman/pacman/-/commit/6cdca15cd9d325ecb8b465b… -- View it on GitLab: https://gitlab.archlinux.org/pacman/pacman/-/commit/6cdca15cd9d325ecb8b465b… You're receiving this email because of your account on gitlab.archlinux.org.

1 0

[Git][pacman/pacman][master] Partially revert resolving paths in pacman.conf
by Allan McRae (＠allan) 12 Dec '25

12 Dec '25

Allan McRae pushed to branch master at Pacman / Pacman Commits: de68378f by Allan McRae at 2025-12-12T21:07:47+10:00 Partially revert resolving paths in pacman.conf For the CacheDir and HookDir options, multiple paths can be specified on a single line (separated by a space). The resolving of paths does not handle multiple entries and causes failures. Additional, directories in CacheDir are created by libalpm if they do not exist, and this is not accounted for when realpath fails due to a non-exisiting directory. "Fixes" #289 and #292. Signed-off-by: Allan McRae <allan(a)archlinux.org> - - - - - 1 changed file: - src/pacman/conf.c View it on GitLab: https://gitlab.archlinux.org/pacman/pacman/-/commit/de68378f9c8e60963b76abd… -- View it on GitLab: https://gitlab.archlinux.org/pacman/pacman/-/commit/de68378f9c8e60963b76abd… You're receiving this email because of your account on gitlab.archlinux.org.

1 0

[Git][pacman/pacman][master] makepkg: fix generation of "--allsource" packages with VCS sources
by Allan McRae (＠allan) 12 Dec '25

12 Dec '25

Allan McRae pushed to branch master at Pacman / Pacman Commits: fd97639c by Allan McRae at 2025-12-12T00:08:15+10:00 makepkg: fix generation of "--allsource" packages with VCS sources The globbing used to get a list of files to reproducibly include in a source package does not recurse through directory symlinks (by design). For any VCS source, recursively "copy" the source into the srclinks directory for inclusion in the final source package. Fixes #290 Signed-off-by: Allan McRae <allan(a)archlinux.org> - - - - - 1 changed file: - scripts/makepkg.sh.in View it on GitLab: https://gitlab.archlinux.org/pacman/pacman/-/commit/fd97639c07a4261a037b1bc… -- View it on GitLab: https://gitlab.archlinux.org/pacman/pacman/-/commit/fd97639c07a4261a037b1bc… You're receiving this email because of your account on gitlab.archlinux.org.

1 0

[Git][pacman/pacman][release/7.1.x] 4 commits: Provide generic sandbox status functions
by Allan McRae (＠allan) 21 Nov '25

21 Nov '25

Allan McRae pushed to branch release/7.1.x at Pacman / Pacman Commits: 63bfb53a by Allan McRae at 2025-11-21T11:32:58+10:00 Provide generic sandbox status functions The function alpm_option_set_disable_sandbox() was removed in favour of the vairants that can enable/disable specific parts of the sandbox. This removal was undocumented in the release notes, and the continued inclusion of the more generic sandbox control functions is still useful. Readd alpm_option_set_disable_sandbox() which now is a shortcut for doing the combined alpm_option_set_disable_sandbox-{filesystem,syscalls}. The alpm_option_get_disable_sandbox() function was readded, but adjusted to return 0 if the sandbox is fully enabled, 1 if any component of the sandbox is disabled, and 2 if all components are disabled. Give the libalpm soname a minor level bump to indicated these interfaces have been added since release, as this commit will be backorted to the release branch. Signed-off-by: Allan McRae <allan(a)archlinux.org> (cherry picked from commit 8a573085714c40e68ed8c03d883c68f69a88c4b2) - - - - - fcdf7408 by Diego Viola at 2025-11-21T11:33:08+10:00 libmakepkg: fix spelling of Overridden Overriden -> Overridden Signed-off-by: Diego Viola <diego.viola(a)gmail.com> (cherry picked from commit a0282cbfea2d67732fc4361c9bbc78b3f9c9a637) - - - - - 883cd747 by Diego Viola at 2025-11-21T11:33:21+10:00 docs: remove PM_ERR_TRANS_COMMITTING This constant is no longer defined in the source code. Removing its reference cleans up the API documentation to match the actual code. Signed-off-by: Diego Viola <diego.viola(a)gmail.com> (cherry picked from commit a16d482b61918708111dde1aaa02d7c40d00f3de) - - - - - 2d0d5d7b by Allan McRae at 2025-11-21T11:33:42+10:00 Document that final hook match wins Signed-off-by: Allan McRae <allan(a)archlinux.org> (cherry picked from commit 8ebdccd4d66f64b3d00f4105beb064ecff2b6a2f) - - - - - 6 changed files: - README - doc/alpm-hooks.5.asciidoc - lib/libalpm/alpm.h - lib/libalpm/handle.c - meson.build - scripts/libmakepkg/lint_pkgbuild/arch.sh.in View it on GitLab: https://gitlab.archlinux.org/pacman/pacman/-/compare/2df7c407e852543e5c9e13… -- View it on GitLab: https://gitlab.archlinux.org/pacman/pacman/-/compare/2df7c407e852543e5c9e13… You're receiving this email because of your account on gitlab.archlinux.org.

1 0

[Git][pacman/pacman][master] Drop md5sum support from repo databases
by Allan McRae (＠allan) 21 Nov '25

21 Nov '25

Allan McRae pushed to branch master at Pacman / Pacman Commits: bdedf621 by Diego Viola at 2025-11-20T04:36:48-03:00 Drop md5sum support from repo databases repo-add removed md5sums in pacman-6.1.0. Remove reading support and the validation option in favor of sha256sums. Signed-off-by: Diego Viola <diego.viola(a)gmail.com> - - - - - 8 changed files: - lib/libalpm/alpm.h - lib/libalpm/be_local.c - lib/libalpm/be_package.c - lib/libalpm/be_sync.c - lib/libalpm/package.c - lib/libalpm/package.h - src/pacman/util.c - test/pacman/tests/remove-print-empty-replacements.py View it on GitLab: https://gitlab.archlinux.org/pacman/pacman/-/commit/bdedf621efc0950c6dc1752… -- View it on GitLab: https://gitlab.archlinux.org/pacman/pacman/-/commit/bdedf621efc0950c6dc1752… You're receiving this email because of your account on gitlab.archlinux.org.

1 0

[Git][pacman/pacman][master] Provide generic sandbox status functions
by Allan McRae (＠allan) 20 Nov '25

20 Nov '25

Allan McRae pushed to branch master at Pacman / Pacman Commits: 8a573085 by Allan McRae at 2025-11-20T16:41:35+10:00 Provide generic sandbox status functions The function alpm_option_set_disable_sandbox() was removed in favour of the vairants that can enable/disable specific parts of the sandbox. This removal was undocumented in the release notes, and the continued inclusion of the more generic sandbox control functions is still useful. Readd alpm_option_set_disable_sandbox() which now is a shortcut for doing the combined alpm_option_set_disable_sandbox-{filesystem,syscalls}. The alpm_option_get_disable_sandbox() function was readded, but adjusted to return 0 if the sandbox is fully enabled, 1 if any component of the sandbox is disabled, and 2 if all components are disabled. Give the libalpm soname a minor level bump to indicated these interfaces have been added since release, as this commit will be backorted to the release branch. Signed-off-by: Allan McRae <allan(a)archlinux.org> - - - - - 4 changed files: - README - lib/libalpm/alpm.h - lib/libalpm/handle.c - meson.build View it on GitLab: https://gitlab.archlinux.org/pacman/pacman/-/commit/8a573085714c40e68ed8c03… -- View it on GitLab: https://gitlab.archlinux.org/pacman/pacman/-/commit/8a573085714c40e68ed8c03… You're receiving this email because of your account on gitlab.archlinux.org.

1 0

[Git][pacman/pacman][master] libmakepkg: fix spelling of Overridden
by Allan McRae (＠allan) 20 Nov '25

20 Nov '25

Allan McRae pushed to branch master at Pacman / Pacman Commits: a0282cbf by Diego Viola at 2025-11-20T16:37:39+10:00 libmakepkg: fix spelling of Overridden Overriden -> Overridden Signed-off-by: Diego Viola <diego.viola(a)gmail.com> - - - - - 1 changed file: - scripts/libmakepkg/lint_pkgbuild/arch.sh.in View it on GitLab: https://gitlab.archlinux.org/pacman/pacman/-/commit/a0282cbfea2d67732fc4361… -- View it on GitLab: https://gitlab.archlinux.org/pacman/pacman/-/commit/a0282cbfea2d67732fc4361… You're receiving this email because of your account on gitlab.archlinux.org.

1 0

[Git][pacman/pacman][master] docs: remove PM_ERR_TRANS_COMMITTING
by Allan McRae (＠allan) 20 Nov '25

20 Nov '25

Allan McRae pushed to branch master at Pacman / Pacman Commits: a16d482b by Diego Viola at 2025-11-19T02:31:05-03:00 docs: remove PM_ERR_TRANS_COMMITTING This constant is no longer defined in the source code. Removing its reference cleans up the API documentation to match the actual code. Signed-off-by: Diego Viola <diego.viola(a)gmail.com> - - - - - 1 changed file: - README View it on GitLab: https://gitlab.archlinux.org/pacman/pacman/-/commit/a16d482b61918708111dde1… -- View it on GitLab: https://gitlab.archlinux.org/pacman/pacman/-/commit/a16d482b61918708111dde1… You're receiving this email because of your account on gitlab.archlinux.org.

1 0

Issues in man pages of pacman
by Helge Kreutzmann 16 Nov '25

16 Nov '25

Dear pacman maintainer, the manpage-l10n project[1] maintains a large number of translations of man pages both from a large variety of sources (including pacman) as well for a large variety of target languages. During their work translators notice different possible issues in the original (english) man pages. Sometimes this is a straightforward typo, sometimes a hard to read sentence, sometimes this is a convention not held up and sometimes we simply do not understand the original. We use several distributions as sources and update regularly (at least every 2 month). This means we are fairly recent (some distributions like archlinux also update frequently) but might miss your latest upstream version once in a while, so the error might be already fixed. We apologize and ask you to close the issue immediately if this should be the case, but given the huge volume of projects and the very limited number of volunteers we are not able to double check each and every issue. Secondly we translators see the manpages in the neutral po format, i.e. converted and harmonized, but not the original source (be it man, groff, xml or other). So we cannot provide a true patch (where possible), but only an approximation which you need to convert into your source format. Finally the issues I'm reporting have accumulated over time and are not always discovered by me, so sometimes my description of the problem my be a bit limited - do not hesitate to ask so we can clarify them. I'm now reporting the issues for your project. If future reports should use another channel, please let me know. [1] https://manpages-l10n-team.pages.debian.net/manpages-l10n/ Man page: libalpm.3 Issue: Fullstop ok here? (All the others don't have one) "Functions dealing with libalpm's dependency and conflict information\\&." "Functions to query and manipulate the database of libalpm\\&." -- Man page: libalpm.3 Issue: pacman → pacman(8) "libalpm is a package management library, primarily used by pacman\\&. For " "ease of access, the libalpm manual has been split up into several " "sections\\&." -- Man page: makepkg.8 Issue 1: makepkg → B<makepkg> Issue 2: possibly $srcdir → I<$srcdir>? "Clean build artifacts from previous runs of makepkg in the current directory " "by removing $srcdir before building the package\\&." -- Man page: makepkg.8 Issue 1: makepkg → B<makepkg> Issue 2: bash → B<bash>(1) "Setting to 0 disables PKGBUILD linting within makepkg\\&. Useful on systems " "with slow bash subshell operations, or on PKGBUILDs with extreme amounts of " "package splitting\\&." -- Man page: makepkg.conf.5 Issue: ld(1) → B<ld>(1) "Flags used for the linker\\&. Several options may be specified with common " "usage resembling \\(lq-Wl,--hash-style=gnu\\(rq\\&. Read ld(1) for more " "details on available linker flags\\&." -- Man page: makepkg.conf.5 Issue: gcc(1) → B<gcc>(1) "Additional compiler flags appended to CFLAGS for use in debugging\\&. " "Usually this would include: \\(lq-g\\(rq\\&. Read gcc(1) for more details on " "the wide variety of compiler flags available\\&." -- Man page: makepkg-template.1 Issue: PKGBUILD → I<PKGBUILD> "I<makepkg-template> is a script to ease the work of maintaining multiple " "similar PKGBUILDs\\&. It allows you to move most of the code from the " "PKGBUILD into a template file and uses markers to allow in-place updating of " "existing PKGBUILDs if the template has been changed\\&." "Template files can contain any code allowed in a PKGBUILD\\&. You can think " "of them like external files included with \"\\&.\" or \"source\", but they " "will be inlined into the PKGBUILD by I<makepkg-template> so you do not " "depend on the template file when building the package\\&." "Template files should be stored in one directory and filenames should be " "\"$template_name-$version\\&.template\" with a symlink " "\"$template_name\\&.template\" pointing to the most recent template\\&. If " "the version is not set in the marker, I<makepkg-template> will automatically " "use the target of \"$template_name\\&.template\", otherwise the specified " "version will be used\\&. This allows for easier verification of untrusted " "PKGBUILDs if the template is trusted\\&. You verify the non-template code " "and then use a command similar to this:" "Template files may also contain markers leading to nested templates in the " "resulting PKGBUILD\\&. If you use markers in a template, please set the " "version you used/tested with in the start/input marker so other people can " "properly recreate from templates\\&." -- Man page: pacman.8 Issue: NULL character → NUL? "Print each match in a machine readable output format\\&. The format is " "I<repository\\e0pkgname\\e0pkgver\\e0path\\en> with I<\\e0> being the NULL " "character and I<\\en> a linefeed\\&." -- Man page: pacman.conf.5 Issue: Where did "%o" come from? The XferCommand command usage does not have an %o "If set, an external program will be used to download all remote files\\&. " "All instances of %u will be replaced with the download URL\\&. If present, " "instances of %o will be replaced with the local filename, plus a \\" "(lq\\&.part\\(rq extension, which allows programs like wget to do file " "resumes properly\\&." -- Man page: pacman.conf.5 Issue: The part "will not be removed" reads strang - is there something missing? Who is not removed? The URL? The (non-)cahce servers? And what is the relation to database fields? "A full URL to a location where the packages, and signatures (if available) " "for this repository can be found\\&. Cache servers will be tried before any " "non-cache servers, will not be removed from the server pool for 404 download " "errors, and will not be used for database files\\&." -- Man page: pacman.conf.5 Issue: I<pacman-db-upgrade> → B<pacman-db-upgrade> "I<pacman-db-upgrade> [options]" "I<pacman-db-upgrade> is a script that upgrades the local database used by " "B<pacman>(8) to a newer format\\&." -- Man page: pacman.conf.5 Issue: This man page part is broken, check formatting? (Version from August 2022 looks better) (Both formatting and missing B<>/I<>) "*-h, \\e--help*::\n" "Show the built-in help message and exit\\&.\n" "*-V, \\e--version*::\n" "Show version information and exit\\&.\n" "*-d, \\e--dbpath* E<lt>pathE<gt>::\n" "Set an alternate database location\\&.\n" "*-r, \\e--root* E<lt>pathE<gt>::\n" "Set an alternate installation root\\&.\n" "*\\e--config* E<lt>pathE<gt>::\n" "Set an alternate configuration file\\&.\n" "*\\e--nocolor*::\n" "Remove color from output\\&.\n" -- Man page: PKGBUILD.5 Issue: PKGBUILD\\&.proto → I<PKGBUILD\\&.proto> "An example PKGBUILD, useful for reference, is located in I</usr/share/" "pacman> along with other example files such as an install script\\&. You can " "copy the provided PKGBUILD\\&.proto file to a new package build directory " "and make customizations to suit your needs\\&." -- Man page: repo-add.8 Issue: Missing full stop "Include package PGP signatures in the repository database (if available)" -- Man page: testpkg.8 Issue: I<testpkg> → B<testpkg> "I<testpkg> E<lt>package fileE<gt>" "I<testpkg> is a script used to make sure that a pacman package is valid\\&." -- Dr. Helge Kreutzmann debian(a)helgefjell.de Dipl.-Phys. http://www.helgefjell.de/debian.php 64bit GNU powered gpg signed mail preferred Help keep free software "libre": http://www.ffii.de/

2 2