summaryrefslogtreecommitdiffstats
path: root/pym/_emerge
Commit message (Collapse)AuthorAgeFilesLines
* Handle zlib.error when reading compressed build logs via GzipFile.Zac Medico2010-08-221-0/+4
|
* Allow more time for natural exit of the ebuild process after it hasZac Medico2010-08-221-4/+9
| | | | called the exit command via IPC.
* Remove deprecated destroot argument from dblink.isowner() call.Zac Medico2010-08-221-1/+1
|
* During the post src_install QA checks, log to a temporary fileZac Medico2010-08-222-4/+50
| | | | | | since the code we are running reads PORTAGE_LOG_FILE, and we want to avoid annoying "gzip: unexpected end of file" messages when FEATURES=compress-build-logs is enabled.
* Make Binpkg call prepare_build_dirs() before trying to clean the oldZac Medico2010-08-221-2/+2
| | | | log, so PORTAGE_LOG_FILE is initialized.
* Add missing scheduler attribute to BinpkgVerifier (broken since theZac Medico2010-08-223-3/+4
| | | | the addition of FEATURES=compress-build-logs support).
* Fix AttributeError for MetadataRegen._sched_iface.Zac Medico2010-08-221-1/+1
|
* Fix broken writemsg call in depgraph._resolve() that triggers this error:Zac Medico2010-08-211-1/+1
| | | | TypeError: writemsg() got an unexpected keyword argument 'file'
* Bug #324191 - Add support for FEATURES=compress-build-logs. The causesZac Medico2010-08-2112-155/+92
| | | | | | | all build logs to be compressed while they are being written. Log file names have an extension that is appropriate for the compression type. Currently, only gzip(1) compression is supported, so build logs will have a '.gz' extension when this feature is enabled.
* When accessing pkgsettings.features inside Scheduler._generate_digests(),Zac Medico2010-08-211-0/+8
| | | | | ensure that we are using global features settings rather than those from package.env.
* Add a wrapper around config.features that provides the followingZac Medico2010-08-211-4/+0
| | | | | | | | | | | | | | | | enhancements: * The FEATURES variable is automatically synchronized upon modification. * Modifications result in a permanent override that will cause the change to propagate to the incremental stacking mechanism in config.regenerate(). This eliminates the need to call config.backup_changes() when FEATURES is modified, since any overrides are guaranteed to persist despite calls to config.reset(). This allows cleanup of all code that overrides FEATURES, and also allows config.regenerate() to stack FEATURES such that special cases are not needed for package.env handling inside config.setcpv().
* --autounmask: If we can't do a change because of REQUIRED_USE, give at least ↵Sebastian Luther2010-08-201-11/+41
| | | | a good error message
* --autounmask: Don't do use changes if they violate REQUIRED_USESebastian Luther2010-08-201-19/+32
| | | | This also fixes some pkg.use.enabled / _pkg_use_enabled(pkg) bugs
* slot collision handler: Drop solutions that violate REQUIRED_USESebastian Luther2010-08-201-0/+18
|
* slot collision handler: Work correctly with --autounmaskSebastian Luther2010-08-202-29/+35
|
* circular dependency handler: Add support for REQUIRED_USESebastian Luther2010-08-201-13/+31
|
* circular dependency handler: Small code clean upsSebastian Luther2010-08-201-5/+6
|
* Add experimental EPREFIX/EROOT support to the config and vartreeZac Medico2010-08-191-4/+1
| | | | | | | | | classes, and use it in ResolverPlayground to emulate a prefix-like environment. This fixes ResolverPlayground so that it doesn't have to abuse the --root and --root-deps options in order to create a testing environment. Instead it simply creates a temporary EPREFIX. WARNING: EPREFIX/EROOT support is experimental and may be incomplete for cases in which EPREFIX is non-empty.
* Update docs, comments, and messages to refer to make.globals inZac Medico2010-08-191-3/+5
| | | | /usr/share/portage/config/ instead of /etc/make.globals.
* Rename the bdeps_root variable to depend_root, since it's used to controlZac Medico2010-08-191-4/+4
| | | | | DEPEND and in the future we may have separate build-time dep types for host and target.
* Eliminate the bdeps_optional variables inside depgraph._add_pkg_deps(),Zac Medico2010-08-191-4/+3
| | | | and use the pkg.built attribute instead.
* Move the clever parts of depgraph._show_circular_deps into its own module.Sebastian Luther2010-08-192-155/+240
| | | | This also fixes some bugs related to autounmask.
* cycle-finder: Don't try to change flags that are in use.{make,focrce}Sebastian Luther2010-08-191-16/+5
|
* Move code from EbuildProcess to EbuildPhase.Zac Medico2010-08-192-24/+24
|
* Remove duplicate attributes from __slots__.Zac Medico2010-08-191-2/+1
|
* Move the code from _prepare_env_file() to a new BinpkgEnvExtractorZac Medico2010-08-183-9/+99
| | | | class.
* Remove unnecessary call to _prepare_env_file(), since it only makesZac Medico2010-08-181-8/+1
| | | | sense for binary or installed packages.
* Move environment sanity check to the Scheduler and do it if thereZac Medico2010-08-181-0/+36
| | | | are any source packages in the merge list.
* When suggesting solutions inside _show_circular_deps(), ignore solutionsZac Medico2010-08-181-0/+16
| | | | that involve changes to use.mask or use.force settings.
* Fix deprecation warnings from paren_reduce.Zac Medico2010-08-181-2/+2
|
* Fix small bug in _show_circular_deps and add some testsSebastian Luther2010-08-181-2/+1
|
* _show_circular_deps: Ignore solution that violate use dpendencies specified ↵Sebastian Luther2010-08-181-6/+39
| | | | by parents. Warn the user if there are many cycles.
* _show_circular_deps: Stop prunning the graph to show (other) parents of the ↵Sebastian Luther2010-08-181-14/+0
| | | | involved packages in the merge list
* _show_circular_deps: Fix handling of 'None' returned by ↵Sebastian Luther2010-08-181-1/+2
| | | | extract_affecting_use and remove debug output
* _emerge.depgraph._show_circular_deps(): Find suggestions if an atom occures ↵Sebastian Luther2010-08-181-11/+66
| | | | more than once in a dep string
* Print suggestions for circular dependenciesSebastian Luther2010-08-181-8/+44
|
* _emerge.depgraph._show_circular_deps(): Display only the smallest cycleSebastian Luther2010-08-181-1/+31
|
* Use _prepare_env_file() to extract the binary package environment.bz2Zac Medico2010-08-181-1/+10
| | | | | | prior to the pkg_setup phase. This fixes a regression since EbuildPhase no longer calls doebuild(), and therefore _prepare_env_file() needs to be called now.
* * Replace config._iuse_implicit_re with a callable _iuse_implicit_matchZac Medico2010-08-182-13/+8
| | | | | | | object that simply returns boolean, and update consumers. * Make check_required_use() take a callable for the iuse parameter, and pass in Package.use.is_valid_flag or iuse.__contains__ as appropriate.
* Remove unused iuse_implicit attribute.Zac Medico2010-08-181-1/+0
|
* Temporarily disable portage.dep deprecation warnings for external apiZac Medico2010-08-171-0/+1
| | | | | consumers. Once the relevant api changes are in a portage release with stable keywords, make these warnings unconditional.
* Use binarytree._ensure_dir() to create the binpkg parent dir.Zac Medico2010-08-171-7/+1
|
* Fix breakage in REPLACING_VERSIONS support for emerge.Zac Medico2010-08-172-1/+15
|
* Use EbuildPhase instead of EbuildProcess, for things like die_hooksZac Medico2010-08-171-2/+2
| | | | support.
* Make EbuildBinpkg inherit from CompositeTask since the EbuildPhaseZac Medico2010-08-171-12/+23
| | | | class is capable of executing the 'package' phase.
* Make EbuildBinpkg inherit from EbuildProcess since the 'package'Zac Medico2010-08-171-18/+6
| | | | phase is already a member of actionmap.
* Pass actionmap keyword argument as keyword, rather than positional.Zac Medico2010-08-171-1/+1
|
* Unify ebuild spawing some more. All proper ebuild phases which executeZac Medico2010-08-171-24/+5
| | | | ebuild.sh are spawned via _doebuild_spawn() now. No exceptions.
* Fix _spawn_phase() to spawn phases without sandbox when necessary.Zac Medico2010-08-171-8/+2
|
* Make doebuild() use EbuildPhase to execute all phase (unless returnpidZac Medico2010-08-171-0/+10
| | | | is True, which is only for the "depend" phase now).