Get Gentoo!
gentoo.org sites
gentoo.org Wiki Bugs Forums Packages
Planet Archives Sources
Infra Status
summaryrefslogtreecommitdiff
path: root/eclass
Commit message (Collapse)AuthorAgeFilesLines
* eclass: ignoring flags warning in findlib.eclassAlfredo Tupone2021-01-251-1/+2
| | | | Signed-off-by: Alfredo Tupone <tupone@gentoo.org>
* eclass: ignore CFLAGS warningsAlfredo Tupone2021-01-231-0/+3
| | | | Signed-off-by: Alfredo Tupone <tupone@gentoo.org>
* toolchain-funcs.eclass: new function tc-ld-force-bfd()Theo Anderson2021-01-231-6/+15
| | | | | | | | | | | | No functions currently force ld.bfd usage when ld.lld is active. This function forces ld.bfd when either ld.gold or ld.lld is active. tc-ld-disable-gold() now calls this new function only if ld.gold is active. Package-Manager: Portage-3.0.14, Repoman-3.0.2 Signed-off-by: Theo Anderson <telans@posteo.de> Closes: https://github.com/gentoo/gentoo/pull/19116 Signed-off-by: Sergei Trofimovich <slyfox@gentoo.org>
* eclass/lua-utils.eclass: Add support for test-runnersConrad Kostecki2021-01-201-0/+70
| | | | | | | | | | | | | | | | | | | | | | During migration of dev-lua/* ebuilds to slotted lua, I noticed, that many ebuilds use 'dev-lua/busted' for running tests. This change adds support for running a test-runner, at first only 'busted' for now. Also a non-color and plaintext output will be used for the test-runner 'busted'. This is basically a copy of the test-runner section, written by mgorny, which already exists in 'distutils-r1', but modified and adapted to lua. In order to use this feature, you can define 'lua_enable_tests busted' to setup everything needed for tests and run them. By default, 'dev-lua/busted' assumes, that tests are in the 'spec' folder. If this is not the case, you can add a second argument to specify a different folder. For example, if the folder is called 'foo', you can just run 'lua_enable_tests busted foo'. More test-runners can be added in future, if needed. Signed-off-by: Conrad Kostecki <conikost@gentoo.org>
* eclass/db.eclass: use get_libname for Darwin hosts (version 2)Fabian Groffen2021-01-191-13/+17
| | | | | | | | | Redo the original attempt of this patch (b158fba5eee07da21f8b80fd737b9bdd3865256b) this time ensuring there is absoletely no change in installed files. Closes: https://bugs.gentoo.org/673322 Signed-off-by: Fabian Groffen <grobian@gentoo.org>
* vala.eclass: make has_version aware of ROOT for EAPI 7David Michael2021-01-181-2/+2
| | | | | | | | | | | The vala dependencies are declared in BDEPEND since EAPI 7 so that the valac command is natively executable. With no arguments, the has_version function would look for a cross-compiled vala package in the target ROOT and always fail. Signed-off-by: David Michael <fedora.dm0@gmail.com> Closes: https://github.com/gentoo/gentoo/pull/18813 Signed-off-by: Matt Turner <mattst88@gentoo.org>
* mount-boot.eclass: Fix eclassdoc tag order and copyright dateMichał Górny2021-01-161-2/+2
| | | | Signed-off-by: Michał Górny <mgorny@gentoo.org>
* Revert "eclass/db.eclass: use get_libname for Darwin support"Mike Gilbert2021-01-161-13/+13
| | | | | | | | | This change somehow causes libdb.so to be installed. Bug: https://bugs.gentoo.org/673322 Closes: https://bugs.gentoo.org/765625 Reverts: b158fba5eee07da21f8b80fd737b9bdd3865256b Signed-off-by: Mike Gilbert <floppym@gentoo.org>
* kernel-install.eclass: Improve error message on /boot problemsMichał Górny2021-01-161-1/+1
| | | | | | | | Use the newly-introduced mount-boot.eclass nonfatal support to amend the error message with the instruction to run 'emerge --config' rather than rebuild the whole kernel. Signed-off-by: Michał Górny <mgorny@gentoo.org>
* dist-kernel-utils.eclass: Support dracut's uefi=yes optionMichał Górny2021-01-161-1/+30
| | | | | | | | | Support dracut's uefi=yes configuration option that creates a combined UEFI stub, kernel and initramfs in a single UEFI executable. If such an output is detected, install it in place of the actual kernel image and stub out the duplicate initrd to save space. Signed-off-by: Michał Górny <mgorny@gentoo.org>
* kernel-install.eclass: Improve failed install error messagesMichał Górny2021-01-162-15/+32
| | | | | | | Support and use nonfatal to provide a detailed error message when kernel postinst fails, in particular the correct 'emerge --config' command. Signed-off-by: Michał Górny <mgorny@gentoo.org>
* kernel-install.eclass: Include SLOT in --config suggestionMichał Górny2021-01-161-1/+1
| | | | Signed-off-by: Michał Górny <mgorny@gentoo.org>
* kernel-install.eclass: Update symlink before installingMichał Górny2021-01-161-2/+1
| | | | | | | | | Update symlink before actually installing the kernel. This primarily ensures that module rebuilds triggered by the upgrade will work correctly even if postinst fails. Besides, pkg_config() retries installing the kernel but does not update the symlink. Signed-off-by: Michał Górny <mgorny@gentoo.org>
* kernel-install.eclass: Move common to kernel-install_install_allMichał Górny2021-01-161-30/+32
| | | | | | | | | Move the code shared by pkg_postinst() and pkg_config() to a new kernel-install_install_all() function. After all, the purpose of pkg_config() is to repeat what pkg_postinst() does normally. Keeping it in a common function improves maintainability. Signed-off-by: Michał Górny <mgorny@gentoo.org>
* mount-boot.eclass: Support nonfatalMichał Górny2021-01-161-5/+9
| | | | | | | Support making mount-boot_check_status() nonfatal. This is useful to amend the error message with additional instructions. Signed-off-by: Michał Górny <mgorny@gentoo.org>
* mount-boot.eclass: Remove support for EAPI 4 and 5Michał Górny2021-01-161-1/+2
| | | | | | | The eclass is no longer used by any EAPI 4 or 5 ebuilds in ::gentoo. This removes the need for EAPI-conditional nonfatal support. Signed-off-by: Michał Górny <mgorny@gentoo.org>
* python-utils-r1.eclass: Disable python3_6Michał Górny2021-01-151-3/+3
| | | | Signed-off-by: Michał Górny <mgorny@gentoo.org>
* python-utils-r1.eclass: Inline _python_impl_supported()Michał Górny2021-01-152-54/+20
| | | | | | | | | | | | The _python_impl_supported() function is not used anymore in its original function. It is called only once, in order to die on incorrect targets in PYTHON_COMPAT. Let's inline the corresponding logic in _python_set_impls() and remove the function. While at it, add an extra check for outdated patterns. This also renders the relevant tests obsolete. Signed-off-by: Michał Górny <mgorny@gentoo.org>
* office-ext-r1.eclass: Switch echo to einfo, add missing || dieAndreas Sturmlechner2021-01-151-6/+6
| | | | Signed-off-by: Andreas Sturmlechner <asturm@gentoo.org>
* office-ext-r1.eclass: Drop openoffice support and functionsAndreas Sturmlechner2021-01-151-109/+14
| | | | Signed-off-by: Andreas Sturmlechner <asturm@gentoo.org>
* office-ext-r1.eclass: Drop EAPI-6 supportAndreas Sturmlechner2021-01-151-3/+3
| | | | Signed-off-by: Andreas Sturmlechner <asturm@gentoo.org>
* eclass/db.eclass: use get_libname for Darwin supportFabian Groffen2021-01-141-13/+13
| | | | | Closes: https://bugs.gentoo.org/673322 Signed-off-by: Fabian Groffen <grobian@gentoo.org>
* dist-kernel-utils.eclass: Add a function to rebuild initramfsMichał Górny2021-01-131-0/+36
| | | | Signed-off-by: Michał Górny <mgorny@gentoo.org>
* dist-kernel-utils.eclass: Introduce eclass for helper functionsMichał Górny2021-01-133-76/+108
| | | | | | | | | | | | Move some of the utility functions from kernel-install.eclass into dist-kernel-utils.eclass, in order to permit using them without having all kernel-install phases exported. This will be used in order to support rebuilding initramfs in sys-fs/zfs-kmod. Since the eclasses are used only by dist-kernel project eclasses and ebuilds, update the function prefix while moving them. Signed-off-by: Michał Górny <mgorny@gentoo.org>
* Revert "meson.eclass: fix machine files"Mike Gilbert2021-01-121-16/+6
| | | | | | | needs_exe_wrapper and pkg_config_libdir are not built-in options. Reverts: e3cc29cc79d4e94d0144c3ebc4d8cf00d4ee2e70 Signed-off-by: Mike Gilbert <floppym@gentoo.org>
* autotools.eclass: Rename gnuconfig variable to _gnuconfigLars Wendler2021-01-121-4/+4
| | | | Signed-off-by: Lars Wendler <polynomial-c@gentoo.org>
* optfeature.eclass: [QA] Fix unwanted globbingDavid Seifert2021-01-111-2/+5
| | | | | | | Closes: https://bugs.gentoo.org/498988 Reported-by: Mike Frysinger <vapier@gentoo.org> Suggested-by: Ulrich Müller <ulm@gentoo.org> Signed-off-by: David Seifert <soap@gentoo.org>
* meson.eclass: change '--host-route' to '--host-root'Mike Gilbert2021-01-111-1/+1
| | | | Signed-off-by: Mike Gilbert <floppym@gentoo.org>
* meson.eclass: fix machine filesWilliam Hubbs2021-01-111-6/+16
| | | | | | | | | Several options we were setting in the [properties] section of the machine files have been moved to the [built-in options] section in meson 0.56. Closes: https://bugs.gentoo.org/738710 Signed-off-by: William Hubbs <williamh@gentoo.org>
* autotools.eclass: Updated _LATEST_AUTOMAKE variableLars Wendler2021-01-111-1/+3
| | | | | Signed-off-by: Lars Wendler <polynomial-c@gentoo.org> Closes: https://github.com/gentoo/gentoo/pull/19034
* autotools.eclass: Introduce AT_NOEAUTOHEADER variableLars Wendler2021-01-111-2/+8
| | | | | | | | | | Sometimes packages get their headers being broken by autoheader (see media-libs/freetype for example) or get unwanted defines added (see media-libs/libsdl2). In that case we want to run eautoreconf without call to eautoheader. Bug: https://bugs.gentoo.org/764959 Signed-off-by: Lars Wendler <polynomial-c@gentoo.org>
* nvidia-driver.eclass: ModerniseDavid Seifert2021-01-101-128/+108
| | | | | | | | | | | | | * Prefer [[ ]] over [ ] * Add EAPI whitelist * Add inherit guard * sys-apps/pciutils needs to be present on CBUILD * Prefer https over http * Drop reference to eselect-opengl * document NV_KV_MAX_PLUS Closes: https://github.com/gentoo/gentoo/pull/19017 Signed-off-by: David Seifert <soap@gentoo.org>
* acct-user.eclass: allow opt-out of user modificationThomas Deutschmann2021-01-101-0/+27
| | | | | | | | | | | | | | | In some setups where users are changed/managed not only via ebuilds, for example through configuration management systems, it could be problematic if acct-user.eclass will restore user/group settings to values set in ebuild. Setting ACCT_USER_NO_MODIFY to a non-zero value will allow system administrator to disable modification of any existing user. Note: Lock/unlock when acct-* package will be installed/removed will still happen. Signed-off-by: Thomas Deutschmann <whissi@gentoo.org>
* acct-user.eclass: fix styleThomas Deutschmann2021-01-101-5/+5
| | | | Signed-off-by: Thomas Deutschmann <whissi@gentoo.org>
* qt5-build.eclass: Drop obsolete pre-5.15.2 quirkAndreas Sturmlechner2021-01-101-3/+0
| | | | Signed-off-by: Andreas Sturmlechner <asturm@gentoo.org>
* xorg-3.eclass: Call einstalldocs when not using XORG_MULTILIBDavid Michael2021-01-091-0/+1
| | | | | | | | | | DOCS only got installed from the multilib-minimal_src_install call, so they were missing from non-multilib packages. This corrects a behavior difference from xorg-2.eclass where autotools-utils.eclass installed DOCS for the non-multilib case. Signed-off-by: David Michael <fedora.dm0@gmail.com> Signed-off-by: Matt Turner <mattst88@gentoo.org>
* xorg-2.eclass: Explicitly disable static librariesMatt Turner2021-01-091-0/+6
| | | | Signed-off-by: Matt Turner <mattst88@gentoo.org>
* xorg-3.eclass: Explicitly disable static librariesMatt Turner2021-01-091-0/+6
| | | | Signed-off-by: Matt Turner <mattst88@gentoo.org>
* xorg-2.eclass: Remove XORG_STATICMatt Turner2021-01-091-22/+1
| | | | | | | Statically linking X libraries into your program is an extremely bad idea. Signed-off-by: Matt Turner <mattst88@gentoo.org>
* xorg-3.eclass: Remove XORG_STATICMatt Turner2021-01-091-22/+0
| | | | | | | Statically linking X libraries into your program is an extremely bad idea. Signed-off-by: Matt Turner <mattst88@gentoo.org>
* acct-group.eclass: Support ACCT_GROUP_ID overrideMichał Górny2021-01-091-10/+29
| | | | Signed-off-by: Michał Górny <mgorny@gentoo.org>
* acct-user.eclass: Support ACCT_USER_ID overrideMichał Górny2021-01-091-11/+20
| | | | Signed-off-by: Michał Górny <mgorny@gentoo.org>
* acct-user.eclass: Support var overrides for user propertiesMichał Górny2021-01-091-24/+60
| | | | | | | | | | | | | | | | | | | | | | | | | | | | Introduce a few variables to allow easy overrides of common user account proprerties, that is: - ACCT_USER_<username>_SHELL - ACCT_USER_<username>_HOME - ACCT_USER_<username>_HOME_OWNER - ACCT_USER_<username>_HOME_PERMS - ACCT_USER_<username>_GROUPS - ACCT_USER_<username>_GROUPS_ADD The first five variables override the respective ACCT_USER_* variables, with ACCT_USER_*_GROUPS being a space-separated list. The *_GROUPS_ADD variable appends to groups present in the ebuild, as this seems a common necessity. We do realize that the original requirement of overriding ebuilds in a local repository was inconvenient. This new logic should permit easy updates via make.conf. Additionally, it has the advantage of clearly reporting the changes made in the build logs. This does not preclude other solutions to the problem. However, this is probably the best one and it should become the current recommendation. Signed-off-by: Michał Górny <mgorny@gentoo.org>
* linux-mod.eclass: Add a prebuilt-kernel rebuild flagMichał Górny2021-01-071-3/+10
| | | | | | | | Add a 'prebuilt-kernel' USE flag that controls a bound-slot dependency on virtual/prebuilt-kernel. If the flag is enabled, it should ensure that the module will be rebuilt automatically on future kernel upgrades. Signed-off-by: Michał Górny <mgorny@gentoo.org>
* eclass/acct-group.eclass: drop x86-macosFabian Groffen2021-01-061-1/+1
| | | | Signed-off-by: Fabian Groffen <grobian@gentoo.org>
* eclass/acct-user: drop x86-macosFabian Groffen2021-01-061-1/+1
| | | | Signed-off-by: Fabian Groffen <grobian@gentoo.org>
* toolchain-funcs.eclass: fix or1k* tuple detectionSergei Trofimovich2021-01-052-2/+3
| | | | | | | | | | Before the change the only recognised CHOST was 'or1k'. After the change CHOSTs like 'or1k-linux-musl' are also recognised as 'openrisc'. Reported-by: adam@pimentel.space Bug: https://bugs.gentoo.org/763606 Signed-off-by: Sergei Trofimovich <slyfox@gentoo.org>
* Revert "systemd.eclass: remove SYSROOT from pkg-config output"Mike Gilbert2021-01-041-1/+0
| | | | | | | | | After discussion on the mailing list, it was determined that pkgconfig and pkgconf differ in behavior. Revert this eclass change until we have a more clear path forward. Reverts: ac07a890dd55813eee120b328a17ee413e5c3489. Signed-off-by: Mike Gilbert <floppym@gentoo.org>
* eclass/pam.eclass: remove deprecated functionalityMikle Kolyada2021-01-031-45/+2
| | | | Signed-off-by: Mikle Kolyada <zlogene@gentoo.org>
* systemd.eclass: remove SYSROOT from pkg-config outputMike Gilbert2021-01-021-0/+1
| | | | | | | | | | When cross-compiling, users will typically have PKG_CONFIG_SYSROOT=${SYSROOT} defined via pkg-config wrapper. When PKG_CONFIG_SYSROOT is set, all paths included in pkg-config output get prefixed with this value. Signed-off-by: Mike Gilbert <floppym@gentoo.org>