Zac Medico [Mon, 14 Feb 2011 03:28:07 +0000 (19:28 -0800)]
get_dep_chain: prefer parents in merge list
This will fix bug #354747.
Zac Medico [Sun, 13 Feb 2011 17:35:07 +0000 (09:35 -0800)]
ResolverDepthTestCase: add more cases
Zac Medico [Sun, 13 Feb 2011 14:18:00 +0000 (06:18 -0800)]
egencache: --preserve-comments allow missing file
This will fix bug #354687.
Zac Medico [Sun, 13 Feb 2011 13:55:18 +0000 (05:55 -0800)]
depgraph: fix and test depth control for virtuals
Zac Medico [Sun, 13 Feb 2011 13:00:28 +0000 (05:00 -0800)]
ResolverDepthTestCase: test new-style virtuals
Zac Medico [Sun, 13 Feb 2011 11:34:32 +0000 (03:34 -0800)]
ResolverDepthTestCase: add more cases
Zac Medico [Sun, 13 Feb 2011 11:13:34 +0000 (03:13 -0800)]
depgraph: tweak depth calculation for virtuals
Recursively traversed virtual dependencies, and their direct
dependencies, are considered to have the same depth as direct
dependencies.
Zac Medico [Sun, 13 Feb 2011 10:20:24 +0000 (02:20 -0800)]
depgraph: fix and test --deep control of depth
Control over recursion depth hasn't behaved properly since commit
6503980e0e3bcfce9fbaff85c33d87f616e955a9. Now it is fixed and tested.
Zac Medico [Sun, 13 Feb 2011 09:06:39 +0000 (01:06 -0800)]
depgraph: make empty param imply deep = True
This won't change the existing depgraph behavior, but it will allow
logic to be simplified in some cases such that solely the deep param
needs to be consulted (rather than needing to consult both empty and
deep params).
Zac Medico [Sun, 13 Feb 2011 08:56:42 +0000 (00:56 -0800)]
ebuild.1: update outdated BUILD_PREFIX docs
This will fix bug #354387.
Zac Medico [Sun, 13 Feb 2011 08:33:51 +0000 (00:33 -0800)]
egencache: don't ignore as much with --tolerant
Errors returned from --update-use-local-desc and --update-changelogs
are no longer ignored with --tolerant. This will fix part of
bug #354687.
Zac Medico [Sun, 13 Feb 2011 07:50:08 +0000 (23:50 -0800)]
repoman: variable.usedwithhelpers ignore comments
This will fix bug #354685.
Zac Medico [Sun, 13 Feb 2011 07:36:07 +0000 (23:36 -0800)]
depgraph: avoid atom hash collisions in dep_check
Atoms are stored in the graph as (atom, id(atom)) tuples since each
atom is considered to be a unique entity. For example, atoms that
appear identical may behave differently in USE matching, depending on
their unevaluated form. Also, specially generated virtual atoms may
appear identical while having different _orig_atom attributes.
Zac Medico [Sun, 13 Feb 2011 01:52:11 +0000 (17:52 -0800)]
depgraph: return virtual edges from select_atoms
Zac Medico [Sun, 13 Feb 2011 00:22:49 +0000 (16:22 -0800)]
emerge.1: cross-ref --newuse and --changed-use
This will fix bug #354437.
Zac Medico [Sat, 12 Feb 2011 06:09:30 +0000 (22:09 -0800)]
depgraph: don't distort edges for virtuals
In _add_pkg_dep_string(), _add_pkg() was called with generated
Dependency instances that created fake/distorted edges in the
graph for new-style virtuals (similar to those that were fixed
in commit
5b16e0d800f34dc90cdd8c505172ad16b97f0af4). Now there
won't be any such distorted edges.
Zac Medico [Sat, 12 Feb 2011 00:09:22 +0000 (16:09 -0800)]
depgraph: fix distorted display of virtuals
As mentioned in bug #353933, comment #4, special new-style virtual
atoms generated/distorted by _expand_new_virtuals() can prevent
extract_affecting_use() from working properly. This is fixed by
saving the original atoms so that the depgraph can map them back
into place at the appropriate step in dependency evaluation.
Zac Medico [Fri, 11 Feb 2011 08:57:11 +0000 (00:57 -0800)]
depgraph: avoid reinstall of identical binpkg
This will fix bug #354441.
Zac Medico [Tue, 8 Feb 2011 20:47:19 +0000 (12:47 -0800)]
emerge.1: update date in header
Zac Medico [Tue, 8 Feb 2011 19:52:03 +0000 (11:52 -0800)]
emerge.1: clarify docs for package arguments
This will fix bug #354137.
Zac Medico [Tue, 8 Feb 2011 19:39:41 +0000 (11:39 -0800)]
emerge: support ../ path arguments
This will fix bug #354141.
Zac Medico [Tue, 8 Feb 2011 18:54:51 +0000 (10:54 -0800)]
expand_new_virtuals: fix pkg_use_enabled call
AttributeError: 'tuple' object has no attribute 'use'
Zac Medico [Tue, 8 Feb 2011 10:01:13 +0000 (02:01 -0800)]
repoman.herdbase: tolerate expat ImportError
We tolerate global scope import failures for optional modules, so that
ImportModulesTestCase can succeed (or possibly alert us about
unexpected import failures).
Zac Medico [Tue, 8 Feb 2011 09:32:09 +0000 (01:32 -0800)]
portage.debug: handle threading ImportError
Zac Medico [Tue, 8 Feb 2011 09:09:03 +0000 (01:09 -0800)]
test_import_modules: exempt repoman.herdbase
The repoman.herdbase module requires python with xml support.
Zac Medico [Tue, 8 Feb 2011 06:35:14 +0000 (22:35 -0800)]
cache.ebuild_xattr: don't import ENODATA globally
ENODATA is undefined on FreeBSD. This will fix bug #354057.
Zac Medico [Tue, 8 Feb 2011 00:42:37 +0000 (16:42 -0800)]
expand_new_virtuals: handle None for myuse
Zac Medico [Mon, 7 Feb 2011 22:19:33 +0000 (14:19 -0800)]
expand_new_virtuals: fix last commit USE handling
Zac Medico [Mon, 7 Feb 2011 22:06:56 +0000 (14:06 -0800)]
Update dates in headers.
Zac Medico [Mon, 7 Feb 2011 21:59:23 +0000 (13:59 -0800)]
paren_reduce: allow parens in atoms, for EAPI 4
This will fix bug #354003.
Zac Medico [Mon, 7 Feb 2011 21:25:42 +0000 (13:25 -0800)]
ebuild.5: fix spelling of "defaults"
Zac Medico [Mon, 7 Feb 2011 11:43:06 +0000 (03:43 -0800)]
expand_new_virtuals: fix logic from prev commit
We need to evaluate conditionals for dependency matching purposes,
though the unevaluated atom will still be available later for other
purposes.
Zac Medico [Mon, 7 Feb 2011 11:11:08 +0000 (03:11 -0800)]
expand_new_virtuals: save atom unevaluated use
This leads to a better unsatisfied USE message than the one reported
in bug 353933, comment #3. TODO: Fix it so that these generated
atoms don't break depgraph._get_dep_chain(), due to not being
identical to the original atoms and therefore breaking
extract_affecting_use().
Zac Medico [Mon, 7 Feb 2011 00:13:22 +0000 (16:13 -0800)]
cache.sqlite: handle sqlite ImportError
This will fix bug #353836.
Zac Medico [Sun, 6 Feb 2011 23:00:47 +0000 (15:00 -0800)]
default_src_test: handle EAPI 4 emake failure
This will fix bug #353897.
Zac Medico [Sat, 5 Feb 2011 03:20:37 +0000 (19:20 -0800)]
get_dep_chain: add "set" node type
Zac Medico [Sat, 5 Feb 2011 00:56:14 +0000 (16:56 -0800)]
Revert "parse_metadata_use(): omit empty text tokens."
This reverts commit
52f6b55d51e2accaa446bc5d6920421959289baa.
Since split() never return empty tokens, this patch does nothing.
Zac Medico [Sat, 5 Feb 2011 00:27:25 +0000 (16:27 -0800)]
check_required_use: remove obsolete hunk
Zac Medico [Sat, 5 Feb 2011 00:16:15 +0000 (16:16 -0800)]
REQUIRED_USE: fix parens display and test more
Zac Medico [Fri, 4 Feb 2011 22:59:28 +0000 (14:59 -0800)]
REQUIRED_USE: fix single child conditionals disp
Zac Medico [Fri, 4 Feb 2011 22:31:32 +0000 (14:31 -0800)]
check_required_use: clarify operator logic
Zac Medico [Fri, 4 Feb 2011 21:07:52 +0000 (13:07 -0800)]
REQUIRED_USE: fix parens display and test more
Zac Medico [Fri, 4 Feb 2011 17:41:30 +0000 (09:41 -0800)]
depgraph: make downgrades trigger complete graph
It's common for downgrades to break dependencies, as in bug #353613, so
automatically trigger complete graph mode when a downgrade is found.
Zac Medico [Fri, 4 Feb 2011 17:23:12 +0000 (09:23 -0800)]
ebuild(1): skip REQUIRED_USE for some phases
Zac Medico [Fri, 4 Feb 2011 17:17:46 +0000 (09:17 -0800)]
ebuild(1): ensure REQUIRED_USE is satisfied
Zac Medico [Fri, 4 Feb 2011 15:41:45 +0000 (07:41 -0800)]
depgraph: don't drop deps in _add_dep
This will fix bug #353613. We remove a hunk of code that was meant
to limit recursion, but is obsolete an buggy. The recursion limiting
is now handled by _add_pkg and the _ignored_deps stack.
Zac Medico [Fri, 4 Feb 2011 06:28:35 +0000 (22:28 -0800)]
Remove from list by index, not search.
Zac Medico [Fri, 4 Feb 2011 06:03:07 +0000 (22:03 -0800)]
REQUIRED_USE: more display tests
Zac Medico [Fri, 4 Feb 2011 05:42:18 +0000 (21:42 -0800)]
REQUIRED_USE: fix parens display and test more
Michał Górny [Thu, 3 Feb 2011 20:49:31 +0000 (21:49 +0100)]
parse_metadata_use(): omit empty text tokens.
Zac Medico [Fri, 4 Feb 2011 01:38:32 +0000 (17:38 -0800)]
REQUIRED_USE: fix parens display and test
Zac Medico [Thu, 3 Feb 2011 23:29:50 +0000 (15:29 -0800)]
REQUIRED_USE: display unsatisfied part
This will fix bug #353234.
Zac Medico [Thu, 3 Feb 2011 08:14:49 +0000 (00:14 -0800)]
action_uninstall: allow repo with wildcard atom
Zac Medico [Thu, 3 Feb 2011 07:48:55 +0000 (23:48 -0800)]
bintree: cleanup/fix old binhost code more
Zac Medico [Thu, 3 Feb 2011 07:39:09 +0000 (23:39 -0800)]
depgraph: use catsplit where appropriate
Zac Medico [Thu, 3 Feb 2011 07:12:24 +0000 (23:12 -0800)]
bintree: clean up dict use in old binhost code
Zac Medico [Thu, 3 Feb 2011 07:05:28 +0000 (23:05 -0800)]
bintree: add missing del from previous commit
Zac Medico [Thu, 3 Feb 2011 05:51:19 +0000 (21:51 -0800)]
bintree: remove unused __remotepkgs
Zac Medico [Thu, 3 Feb 2011 04:30:37 +0000 (20:30 -0800)]
depgraph: remove obsolete gpg exception handlers
Zac Medico [Thu, 3 Feb 2011 04:18:47 +0000 (20:18 -0800)]
depgraph: skip backtrack for fatal errors
Zac Medico [Thu, 3 Feb 2011 03:41:10 +0000 (19:41 -0800)]
depgraph: rm obsolete AmbguousPackageName handler
Zac Medico [Thu, 3 Feb 2011 03:04:53 +0000 (19:04 -0800)]
repoman.1: new variable.usedwithhelpers check
Zac Medico [Thu, 3 Feb 2011 02:03:05 +0000 (18:03 -0800)]
REQUIRED_USE: no autounmask, so skip backtracking
Zac Medico [Wed, 2 Feb 2011 23:51:00 +0000 (15:51 -0800)]
REQUIRED_USE: handle unsatisfied like normal dep
Zac Medico [Wed, 2 Feb 2011 22:04:35 +0000 (14:04 -0800)]
Fix grammar in a comment.
Zac Medico [Wed, 2 Feb 2011 21:46:47 +0000 (13:46 -0800)]
REQUIRED_USE: add blank line after display
Zac Medico [Wed, 2 Feb 2011 21:29:42 +0000 (13:29 -0800)]
Make || choices independent of REQUIRED_USE.
Zac Medico [Wed, 2 Feb 2011 05:33:10 +0000 (21:33 -0800)]
_get_dep_chain: allow SetArg for first node
Zac Medico [Wed, 2 Feb 2011 04:09:50 +0000 (20:09 -0800)]
Show "missed update" message for REQUIRED_USE.
This will fix bug #353443.
Zac Medico [Tue, 1 Feb 2011 16:28:49 +0000 (08:28 -0800)]
portage.5: update date in header
Zac Medico [Tue, 1 Feb 2011 16:19:28 +0000 (08:19 -0800)]
portage.5: document /etc/portage/license_groups
Zac Medico [Tue, 1 Feb 2011 09:54:01 +0000 (01:54 -0800)]
doebuild: always return success for listonly mode
This compensates for changed fetch() return value in commit
394cd4a00d37fd078f8dd40f89f8c810355d7816, which serves to
trigger the pkg_nofetch phase inside doebuild().
Zac Medico [Tue, 1 Feb 2011 04:47:57 +0000 (20:47 -0800)]
_exec: use default SIGINT/TERM handlers
Avoid issues like bug #353239.
Zac Medico [Tue, 1 Feb 2011 04:42:46 +0000 (20:42 -0800)]
_MergeProcess: use default SIGINT/TERM handlers
Signal handlers inherited from the parent process are irrelevant here.
Zac Medico [Tue, 1 Feb 2011 04:33:48 +0000 (20:33 -0800)]
EbuildFetcher: use default SIGINT/TERM handlers
This seems to fix bug #353239.
Zac Medico [Tue, 1 Feb 2011 00:11:20 +0000 (16:11 -0800)]
emerge: spawn nofetch for --pretend --fetch
This is part of bug 353334.
Zac Medico [Mon, 31 Jan 2011 23:44:55 +0000 (15:44 -0800)]
depgraph: display 'searching for similar names'
It can take few seconds to search, so display a status message.
Sebastian Luther [Mon, 31 Jan 2011 22:35:40 +0000 (23:35 +0100)]
Print near matches for cmd line atoms without matching ebuild
Zac Medico [Mon, 31 Jan 2011 22:28:12 +0000 (14:28 -0800)]
ImportModulesTestCase: use iterator, not list
Zac Medico [Mon, 31 Jan 2011 22:10:14 +0000 (14:10 -0800)]
EbuildBuild: spawn nofetch for fetchonly
Arfrever Frehtes Taifersar Arahesis [Mon, 31 Jan 2011 21:48:11 +0000 (22:48 +0100)]
Fix a typo in a comment.
Sebastian Luther [Mon, 31 Jan 2011 17:40:08 +0000 (18:40 +0100)]
Allow relative paths (starting with ./) on the cmd line
bug 352208
Sebastian Luther [Mon, 31 Jan 2011 12:45:43 +0000 (13:45 +0100)]
slot collision suggestions: Avoid showing redundant suggestions.
http://forums.gentoo.org/viewtopic-t-862257.html?sid=
9ba8646a15044fe024a41154df351c6a
Zac Medico [Mon, 31 Jan 2011 07:46:45 +0000 (23:46 -0800)]
ebuild_xattr: lazily import xattr
Sebastian Luther [Mon, 31 Jan 2011 07:30:26 +0000 (08:30 +0100)]
test/lint/test_import_modules.py: Better error reporting and expected failures
Zac Medico [Sun, 30 Jan 2011 23:24:48 +0000 (15:24 -0800)]
die: include $EBUILD_PHASE in message
When a helper binary dies automatically in EAPI 4 and later, we don't
get a stack trace, so at least report the phase that failed.
Zac Medico [Sun, 30 Jan 2011 22:19:13 +0000 (14:19 -0800)]
isolated-functions.sh: fix syntax error for die
Fix die so that dump_trace is not called when the main script is a
helper binary. This prevents the following error message when a helper
dies automatically in EAPI 4:
isolated-functions.sh: line 61: p -= : syntax error: operand expected (error token is "-= ")
Fabian Groffen [Sun, 30 Jan 2011 21:21:03 +0000 (22:21 +0100)]
NoOffsetWithHelpers: fix invalid errors name
Fabian Groffen [Sun, 30 Jan 2011 20:52:50 +0000 (21:52 +0100)]
repoman: add check for using offset vars with helpers
New check: NoOffsetWithHelpers. Checks whether helper functions, such as
dodir and insinto, are used with offset variables either referring to
the installation image (D, ED), the alternate root location (ROOT,
EROOT), or the offset prefix (EPREFIX).
Zac Medico [Sun, 30 Jan 2011 04:11:00 +0000 (20:11 -0800)]
bintree: warn for missing binhost TIMESTAMP
This seems to be the case that's triggered in bug 353189 and bug 350139.
Zac Medico [Sat, 29 Jan 2011 08:02:34 +0000 (00:02 -0800)]
Fix selinux import error handling.
Zac Medico [Sat, 29 Jan 2011 07:54:50 +0000 (23:54 -0800)]
Test import of all modules.
Zac Medico [Sat, 29 Jan 2011 05:46:02 +0000 (21:46 -0800)]
depgraph: fix get_dep_chain unicode handling
Sebastian Luther [Fri, 28 Jan 2011 19:59:50 +0000 (20:59 +0100)]
autounmask USE change comment: ensure that the first parent requires the change
Fixes bug 350515.
Sebastian Luther [Fri, 28 Jan 2011 12:59:41 +0000 (13:59 +0100)]
Show use conditionals in old style dep chains
Zac Medico [Thu, 27 Jan 2011 02:48:47 +0000 (18:48 -0800)]
binarytree: fix BASE_URI for old-style binhost
Zac Medico [Thu, 27 Jan 2011 02:32:18 +0000 (18:32 -0800)]
binarytree: fix broken BASE_URI initialization
Zac Medico [Thu, 27 Jan 2011 02:04:07 +0000 (18:04 -0800)]
make.conf.5: doc multi PORTAGE_BINHOST support
David James [Wed, 26 Jan 2011 22:45:43 +0000 (14:45 -0800)]
Update Portage to support multiple binhosts.
BUG=chromium-os:11295
TEST=Build x86-mario with multiple binhosts (board binhost, preflight binhost)
and verify that binaries are picked up from both binhosts. Also verify
that if one of the binhosts fails, the other binhost is used correctly.
Review URL: http://codereview.chromium.org/
6329022
Change-Id: I095f28c82e3effbc25105af6ac89e42c335e6381
Zac Medico [Tue, 25 Jan 2011 20:40:54 +0000 (12:40 -0800)]
grab_updates: return all after first modified
This ensures that all relevant updates are returned for cases
in which the destination package of an earlier move corresponds to
the source package of a move that comes somewhere later in the entire
sequence of files.