| Commit message (Collapse) | Author | Age | Files | Lines |
... | |
|
|
|
|
|
| |
(trunk r12200)
svn path=/main/branches/2.1.6/; revision=12233
|
|
|
|
|
|
| |
(trunk r12199)
svn path=/main/branches/2.1.6/; revision=12232
|
|
|
|
|
|
| |
that a new PORTAGE_LOG_FILE is created. (trunk r12198)
svn path=/main/branches/2.1.6/; revision=12231
|
|
|
|
|
|
| |
distfiles are automatically renamed. (trunk r12197)
svn path=/main/branches/2.1.6/; revision=12230
|
|
|
|
|
|
|
| |
directing output to PORTAGE_LOG_FILE instead of emerge-fetch.log which is
only used for prefetchers now. (trunk r12196)
svn path=/main/branches/2.1.6/; revision=12229
|
|
|
|
|
|
| |
also verified in this case). (trunk r12195)
svn path=/main/branches/2.1.6/; revision=12228
|
|
|
|
|
|
| |
and injection. (trunk r12194)
svn path=/main/branches/2.1.6/; revision=12227
|
|
|
|
|
|
| |
after the package is verified and injected before returning. (trunk r12193)
svn path=/main/branches/2.1.6/; revision=12226
|
|
|
|
|
|
|
| |
requires the base url to have a trailing slash, so join PORTAGE_BINHOST
manually. Thanks to Lance Albertson <ramereth@g.o> for reporting. (trunk r12192)
svn path=/main/branches/2.1.6/; revision=12225
|
|
|
|
|
|
|
| |
a sanity check so that bash's returncode doesn't have to be trusted.
(trunk r12191)
svn path=/main/branches/2.1.6/; revision=12224
|
|
|
|
|
|
|
|
|
|
| |
files (such as partially downloaded files), trigger the fetch code if the file
exists and it's size doesn't match the current manifest entry. If there really
is a legitimate reason for the digest to change, `ebuild --force digest` can
be used to avoid triggering this code (or else the old digests can be manually
removed from the Manifest). (trunk r12185:12188)
svn path=/main/branches/2.1.6/; revision=12221
|
|
|
|
|
|
|
|
| |
display the final digraph state. This should help when troubleshooting
problems with blockers that don't resolve automatically, by making it easier
to see what pulled in unwanted packages. (trunk r12185)
svn path=/main/branches/2.1.6/; revision=12220
|
|
|
|
|
|
| |
(trunk r12183)
svn path=/main/branches/2.1.6/; revision=12218
|
|
|
|
|
|
| |
Jeremy Olexa <darkside@g.o> for the initial patch. (trunk r12181)
svn path=/main/branches/2.1.6/; revision=12217
|
|
|
|
|
|
|
| |
FEATURES=python-trace. Thanks to Fabian Groffen <grobian@g.o> for reporting.
(trunk r12180)
svn path=/main/branches/2.1.6/; revision=12216
|
|
|
|
|
|
|
| |
the EMakeParallelDisabled check. Thanks to Mike Auty <ikelos@g.o> for this
patch. (trunk r12179)
svn path=/main/branches/2.1.6/; revision=12215
|
|
|
|
|
|
| |
(trunk r12176)
svn path=/main/branches/2.1.6/; revision=12177
|
|
|
|
|
|
|
|
|
|
|
| |
PORTAGE_BUILDDIR doesn't exist, like when called by digestgen(), use
mkdtemp to create a private temporary directory so that pkg_nofetch
can be spawned (directory needed to satisfy safe $PWD requirement of
bug #239560). This is more user friendly since before the pkg_nofetch
phase would simply be skipped in this case. Thanks to Petteri Räty
<betelgeuse@g.o> for reporting. (trunk r12174)
svn path=/main/branches/2.1.6/; revision=12175
|
|
|
|
| |
svn path=/main/branches/2.1.6/; revision=12173
|
|
|
|
|
|
|
| |
KeyError due to a stale merge list that was for a different $ROOT setting.
Thanks to Ned Ludd <solar@g.o> for reporting. (trunk r12166)
svn path=/main/branches/2.1.6/; revision=12167
|
|
|
|
|
|
| |
which means 'binary update' was used). (trunk r12159)
svn path=/main/branches/2.1.6/; revision=12160
|
|
|
|
|
|
| |
that are modified by updates. (trunk r12155)
svn path=/main/branches/2.1.6/; revision=12158
|
|
|
|
|
|
| |
atom more than once. (trunk r12150)
svn path=/main/branches/2.1.6/; revision=12151
|
|
|
|
|
|
|
| |
a short explanation and suggest a course of action to solve the problem.
(trunk r12148)
svn path=/main/branches/2.1.6/; revision=12149
|
|
|
|
|
|
|
| |
Scheduler._calc_resume_list(), since the tasks are now saved for
display at the end. (trunk r12146)
svn path=/main/branches/2.1.6/; revision=12147
|
|
|
|
|
|
|
|
| |
them in the list of packgaes that "failed to build or install" that's
displayed at the end. Thanks to Jeremy Olexa <darkside@g.o> for the
suggestion. (trunk r12144)
svn path=/main/branches/2.1.6/; revision=12145
|
|
|
|
|
|
|
|
|
| |
that are masked. This is necessary for correct choices in || deps when we
want to ensure that installed masked packages are not preferred over other
available choices that are not masked. Thanks to Jan Kundrát <jkt@g.o> for
reporting this issue and testing the patch. (trunk r12138)
svn path=/main/branches/2.1.6/; revision=12139
|
|
|
|
|
|
|
|
| |
trigger "SLOT conflicts", show the specific atoms that triggered the
conflict. TODO: Distiguish between various possible causes and tailor
messages to suit them. (trunk r12125)
svn path=/main/branches/2.1.6/; revision=12128
|
|
|
|
|
|
|
| |
sets of packages pulled into a given slot. This will make the data easier
to analyze when implementing a fix for bug #249185. (trunk r12124)
svn path=/main/branches/2.1.6/; revision=12127
|
|
|
|
|
|
| |
of the same package are never processed more than once. (trunk r12122)
svn path=/main/branches/2.1.6/; revision=12123
|
|
|
|
| |
svn path=/main/branches/2.1.6/; revision=12118
|
|
|
|
|
|
| |
fetching. (trunk r12115)
svn path=/main/branches/2.1.6/; revision=12116
|
|
|
|
|
|
| |
(trunk r12113)
svn path=/main/branches/2.1.6/; revision=12114
|
|
|
|
|
|
| |
(trunk r12111)
svn path=/main/branches/2.1.6/; revision=12112
|
|
|
|
|
|
|
|
|
| |
using a
tuple of (md5, st_size) as a key to a list of hardlink candidates. Multiple candidates
are used in case some happen to be merged to separate devices. (trunk r12109)
svn path=/main/branches/2.1.6/; revision=12110
|
|
|
|
|
|
| |
(trunk r12103)
svn path=/main/branches/2.1.6/; revision=12104
|
|
|
|
|
|
| |
(trunk r12101)
svn path=/main/branches/2.1.6/; revision=12102
|
|
|
|
|
|
|
|
|
| |
since it's
not really a pressing matter and we still need to update docs and maybe add an emaint
command to generate missing entries for users who just want the warning to go away.
svn path=/main/branches/2.1.6/; revision=12100
|
|
|
|
|
|
|
|
| |
PORTAGE_TMPDIR
is not writable. (trunk r12092)
svn path=/main/branches/2.1.6/; revision=12093
|
|
|
|
| |
svn path=/main/branches/2.1.6/; revision=12088
|
|
|
|
|
|
|
|
| |
now that
they are sets rather than actions. (trunk r12086)
svn path=/main/branches/2.1.6/; revision=12087
|
|
|
|
| |
svn path=/main/branches/2.1.6/; revision=12085
|
|
|
|
| |
svn path=/main/branches/2.1.6/; revision=12080
|
|
|
|
|
|
| |
(trunk r12077)
svn path=/main/branches/2.1.6/; revision=12078
|
|
|
|
| |
svn path=/main/branches/2.1.6/; revision=12076
|
|
|
|
|
|
| |
2.1.6 branch doesn't have lib awareness.
svn path=/main/branches/2.1.6/; revision=12074
|
|
|
|
|
|
|
|
|
| |
profiles, and bail out if the profile contains an unsupported EAPI value
in any one of it's directories. We don't necessarily have to use this but
at least it gives us some way to make emerge bail out early if a profile
contains unsupported EAPI features. (trunk r12068)
svn path=/main/branches/2.1.6/; revision=12069
|
|
|
|
|
|
|
|
| |
--help, --info, --sync, and --version actions. When bailing out, suggest
to revert back to the previous profile configuration advise the user which
actions are allowed with an invalid profile. (trunk r12064:12066)
svn path=/main/branches/2.1.6/; revision=12067
|
|
|
|
|
|
| |
way to a root node, even when circular deps are encountered. (trunk r12051)
svn path=/main/branches/2.1.6/; revision=12052
|
|
|
|
|
|
| |
Thanks to Alec Warner <antarus@g.o>. (trunk r12049)
svn path=/main/branches/2.1.6/; revision=12050
|