summaryrefslogtreecommitdiffstats
path: root/bin/ebuild.sh
Commit message (Collapse)AuthorAgeFilesLines
...
* debug-print-function: output like older portageZac Medico2010-10-271-1/+1
|
* debug-print: permission fixesZac Medico2010-10-271-4/+7
|
* Simplify retval checks in has_version() and best_version().Michał Górny2010-10-241-10/+4
|
* Simplify debug-print*() functions code.Michał Górny2010-10-241-20/+12
| | | | | Make debug-print() and friends a lot simpler. Thanks to Mike Frysinger for his suggestions.
* Fix default() and default_src_install() definitions to account forZac Medico2010-09-201-0/+10
| | | | the new EAPI 4 default src_install implementation.
* Revert "Add option -s to docompress command."Ulrich Mueller2010-09-201-7/+1
| | | | This reverts commit f2375609adc80ebe5395d84902af4045ecea2f73.
* Put more variables in PORTAGE_READONLY_VARS instead of usingZac Medico2010-09-171-4/+14
| | | | save_ebuild_env() to unset them.
* Rename READONLY_EBUILD_METADATA and READONLY_PORTAGE_VARS to startZac Medico2010-09-171-4/+5
| | | | with PORTAGE_, and make them both readonly.
* Add PORTAGE_BUILDDIR to READONLY_PORTAGE_VARS since it's extremelyZac Medico2010-09-171-1/+1
| | | | important now that ebuild-ipc relies on it.
* Add option -s to docompress command.Ulrich Mueller2010-09-141-1/+7
|
* Remove redundant local retval declaration in has_version().Zac Medico2010-09-101-1/+0
|
* Bug #336644 - Make ebuild-ipc.py use array.fromfile() to read picklesZac Medico2010-09-091-8/+7
| | | | | | | in single atomic non-blocking read() calls, similar to how AbstractPollTask._read_buf() does it. If the read fails, exit with status 2 so that best_version() and has_version() can detect the error and die when necessary.
* Filter the FEATURES variable out of the environment between phases,Zac Medico2010-09-071-2/+3
| | | | | | | so that it always matches the latest value from the config class and it can never be stale. With the ability to use package.env for per-package FEATURES settings, there's no need for bashrc to make persistent modifications to the FEATURES variable.
* Fix dyn_setup 'already setup' message to refer to setup instead ofZac Medico2010-09-071-1/+1
| | | | prepare.
* Move stale elog cleanup code from ebuild.sh to _doebuild_spawn(), soZac Medico2010-09-071-10/+11
| | | | that elog messages generated by bashrc are preserved when appropriate.
* Define EBUILD_MASTER_PID earlier in ebuild.sh, before $EBUILD isv2.2_rc76Zac Medico2010-09-061-4/+8
| | | | sourced.
* Bug #336285 - Add workaround to unpack() in order to prevent unzipZac Medico2010-09-061-0/+3
| | | | from interactively prompting under some error conditions.
* Remove all \a (alert/bell/beep) tokens (bug 336024)Sebastian Luther2010-09-061-1/+0
|
* print up-to-date copyright yearsFabian Groffen2010-09-051-1/+1
|
* Take the code which displays ebuild information before pkg_setup,Zac Medico2010-09-051-5/+0
| | | | | and move it from ebuild.sh to EbuildPhase. This way we can easily use python to extend it for bug #336019.
* Unify to different pieces of code that handle sandbox considerations forZac Medico2010-09-021-12/+12
| | | | TEMP, TMP, and TMPDIR variables.
* Prevent local x variable in ebuild_main() from showing up inZac Medico2010-09-021-2/+2
| | | | environment.bz2.
* Prevent the local filter_opts variable in filter_readonly_variables()Zac Medico2010-09-021-5/+6
| | | | from being saved in environment.bz2.
* Add a _SANDBOX_COMPAT_LEVEL constant which will be used to enableZac Medico2010-09-021-0/+9
| | | | a workaround for bug #288863 in the stable branch.
* When signaling successful exit from an ebuild phase via ebuild-ipc,Zac Medico2010-09-021-1/+23
| | | | | | first check for non-empty $SANDBOX_LOG and signal unsuccessful exit if necessary. This fixes an issue with phases being treated as successful even when there were sandbox violations.
* Add support to config.setcpv() for caching USE settings from builtZac Medico2010-09-011-2/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | packages, and use it for evaluating conditional USE deps in atoms passed via IPC to helpers like has_version and best_version. Since the pickled object that contains the IPC command will no longer contain USE, it should solve cases that have been reported in which large USE strings trigger poor handling of non-blocking read in pickle.load(): Traceback (most recent call last): File "/usr/bin/emerge", line 43, in <module> retval = emerge_main() File "/usr/lib64/portage/pym/_emerge/main.py", line 1649, in emerge_main myopts, myaction, myfiles, spinner) File "/usr/lib64/portage/pym/_emerge/actions.py", line 437, in action_build retval = mergetask.merge() File "/usr/lib64/portage/pym/_emerge/Scheduler.py", line 1033, in merge rval = self._merge() File "/usr/lib64/portage/pym/_emerge/Scheduler.py", line 1349, in _merge self._main_loop() File "/usr/lib64/portage/pym/_emerge/Scheduler.py", line 1498, in _main_loop self._poll_loop() File "/usr/lib64/portage/pym/_emerge/PollScheduler.py", line 139, in _poll_loop handler(f, event) File "/usr/lib64/portage/pym/_emerge/EbuildIpcDaemon.py", line 36, in _input_handler obj = pickle.load(self._files.pipe_in) File "/usr/lib64/python2.6/pickle.py", line 1370, in load return Unpickler(file).load() File "/usr/lib64/python2.6/pickle.py", line 858, in load dispatch[key](self) File "/usr/lib64/python2.6/pickle.py", line 1195, in load_setitem value = stack.pop() IndexError: pop from empty list
* Add a fallback PORTAGE_BZIP2_COMMAND setting in ebuild.sh, as aZac Medico2010-08-311-0/+6
| | | | workaround for portage-9999 users.
* Bug #335340 - Use PORTAGE_BUNZIP2_COMMAND and PORTAGE_BZIP2_COMMAND inZac Medico2010-08-301-9/+9
| | | | | | ebuild.sh where appropriate: * unpack() * environment.bz2 compression
* Invert _ebuild_helpers_path EAPI conditional in ebuild.sh.Zac Medico2010-08-301-3/+3
|
* Implement controllable compression and docompress, bug #273633.Ulrich Mueller2010-08-281-0/+28
|
* Handle SIGPIPE when unpack() extracts tar files through a pipe (bug #309001).Zac Medico2010-08-231-2/+2
| | | | | | | When checking ${PIPESTATUS[@]} for extraction of tar files in unpack(), use a new assert_sigpipe_ok() function which behaves the same as the existing assert() function except that it tolerates pipe writers being killed by SIGPIPE.
* Quote arrays in QA_PREBUILT handling.Zac Medico2010-08-211-1/+1
|
* Add support for a QA_PREBUILT variable should contain a list ofZac Medico2010-08-211-0/+20
| | | | | | pre-built binaries. Paths listed here will be appended to each of the other QA_* variables. Thanks to Diego Pettenò <flameeyes@g.o> for the suggestion.
* At some point we may want to stop exporting FEATURES to the ebuildZac Medico2010-08-201-1/+1
| | | | | | | environment, in order to prevent ebuilds from abusing it. In preparation for that, export it as PORTAGE_FEATURES so that bashrc users will be able to migrate any FEATURES conditional code to use this alternative variable.
* * In helpers_die arguments, display ${0##*/} instead of $0 since $0Zac Medico2010-08-181-4/+4
| | | | | | is needlessly long. * When calling helpers_die from a function, use ${FUNCNAME[0]} in arguments instead of $0.
* DoebuildSpawnTestCase: Invoke portage.package.ebuild.doebuild.spawn()Zac Medico2010-08-151-0/+2
| | | | | | with a minimal environment. This gives coverage to some of the ebuild execution internals, like ebuild.sh, EbuildSpawnProcess, and EbuildIpcDaemon.
* Disable EbuildIpcDaemon for now, since it seems to be triggeringZac Medico2010-08-141-1/+3
| | | | | intermittent build failures in my stage builds. For testing purposes set PORTAGE_IPC_DAEMON_ENABLE=1 to enable EbuildIpcDaemon.
* Pass $USE via IPC in order to ensure that we have the correct value forZac Medico2010-08-141-2/+2
| | | | | | built/installed packages, since the config class doesn't currently provide a way to access built/installed $USE that would work in all possible scenarios.
* Add a QueryCommand.settings attribute and use it so that $USEZac Medico2010-08-141-2/+2
| | | | | doesn't have to be passed to the daemon in has_version and best_version calls.
* Don't call exit after pkg_nofetch since that prevents the ebuild-ipcZac Medico2010-08-131-1/+0
| | | | callback from executing.
* Use /usr/bin/python when PORTAGE_PYTHON variable is empty.Arfrever Frehtes Taifersar Arahesis2010-08-141-4/+4
| | | | It avoids some errors during upgrade from older Portage versions.
* Ensure that all Python processes use the same version of Python.Arfrever Frehtes Taifersar Arahesis2010-08-141-14/+8
|
* Setup EBUILD_MASTER_PID at the bottom, for scripts that sourceZac Medico2010-08-131-0/+4
| | | | ebuild.sh.
* Bug #315615 - Use EbuildIpcDaemon to implement has_version and best_version.Zac Medico2010-08-131-0/+10
| | | | | This provides performance benefits and also avoids permissions issues with FEATURES=userpriv.
* Move EBUILD_MASTER_PID and associated SIGTERM trap into ebuild_main()Zac Medico2010-08-131-4/+5
| | | | | | | since it seems more reliable this way. This helps us avoid having EbuildIpcDaemon need to send a kill signal (and triggering the zombie message too). This is especially important for the die helper which is called by other helpers for EAPI 4.
* Use EbuildIpcDaemon to replace the functionality ofZac Medico2010-08-131-10/+4
| | | | EBUILD_EXIT_STATUS_FILE.
* Fix faulty $? handling in *into functions (from previous commit).Zac Medico2010-08-121-12/+16
|
* Make utilities die on failure in EAPI 4, bug 273651Dror Levin2010-08-121-4/+20
|
* Make sure we don't allow REQUIRED_USE in EAPIs that don't support itSebastian Luther2010-08-111-5/+0
|
* Don't cache REQUIRED_USE for EAPIs 1-3 which don't support it.Zac Medico2010-08-101-1/+5
|
generated by cgit v1.2.3-1-g7c22 (git 2.25.1) at 2024-07-01 02:11:35 +0000