Commit message (Collapse) | Author | Age | Files | Lines | |
---|---|---|---|---|---|
* | elog/mod_save: wrap IOException for bug #441948 | Zac Medico | 2012-11-13 | 1 | -6/+18 |
| | |||||
* | pkg_pretend: keep temp on failure for bug #442536 | Zac Medico | 2012-11-11 | 1 | -7/+12 |
| | |||||
* | Deprecate python-distutils-ng. | Michał Górny | 2012-10-31 | 1 | -0/+1 |
| | | | | python-r1 & distutils-r1 are almost drop-in replacements. | ||||
* | dblink.mergeme(): use stack instead of recursion | Zac Medico | 2012-10-30 | 1 | -11/+11 |
| | | | | | Maybe this helps reduce memory consumption when merging packages with lots of subdirectories like gentoo-sources (bug #320859). | ||||
* | validate_desktop_entry: handle df-utils-0.21 hint | Zac Medico | 2012-10-28 | 1 | -1/+3 |
| | |||||
* | emerge --info: show linker version, bug #436294v2.2.0_alpha142 | Zac Medico | 2012-10-26 | 1 | -0/+18 |
| | |||||
* | test_slot_conflict_rebuild: add a valid rebuild | Zac Medico | 2012-10-26 | 1 | -2/+29 |
| | |||||
* | depgraph: use _match_slot in multislot match | Zac Medico | 2012-10-26 | 1 | -2/+2 |
| | | | | This amkes sub-slot matching work correctly. | ||||
* | backtracking: discard extra rebuilds, bug #439688 | Zac Medico | 2012-10-26 | 3 | -8/+50 |
| | |||||
* | Add test case for bug #439688. | Zac Medico | 2012-10-25 | 1 | -0/+80 |
| | |||||
* | depgraph: trigger rebuild for unbuilt child | Zac Medico | 2012-10-25 | 2 | -4/+7 |
| | | | | | This will fix bug #439694, where built slot-operator deps failed to trigger rebuilds if the deps were initially broken. | ||||
* | depgraph: no multislot tweak for built slot op | Zac Medico | 2012-10-25 | 2 | -3/+4 |
| | | | | | | | | The multislot tweak from bug #220341 (see commit 65a421ae35acd5639ad1258e220754a85e55de6e) triggered erroneous matches for built slot-operator deps, causing the depgraph to treat unsatisfied built slot-operator deps as if they were satisfied (one of the issues uncovered in bug #439694). | ||||
* | Add test case for bug #439694. | Zac Medico | 2012-10-25 | 1 | -0/+67 |
| | |||||
* | merge-sync: handle unmerge corner cases | Zac Medico | 2012-10-25 | 1 | -4/+13 |
| | | | | | In dblink_merged_path we need to distinguish between existing and non-existing, since we need an existing path of syncfs. | ||||
* | merge-sync: syncfs on parent of dir for unmerge | Zac Medico | 2012-10-25 | 1 | -7/+34 |
| | | | | | | Note that we use a realpath and a regular stat here, since we want to follow any symlinks back to the real device where the real parent directory resides. | ||||
* | MergeProcess: cache syncfs libc library lookup | Zac Medico | 2012-10-25 | 2 | -14/+20 |
| | |||||
* | Add FEATURES=merge-sync, for bug #439584. | Zac Medico | 2012-10-25 | 2 | -1/+41 |
| | |||||
* | Add portage.util._ctypes module, for bug #439584. | Zac Medico | 2012-10-25 | 1 | -0/+47 |
| | |||||
* | dblink: add hooks to sync disk for bug #439584 | Zac Medico | 2012-10-25 | 1 | -1/+58 |
| | |||||
* | emaint logs --help: fix module-options typo | Brian Dolbec | 2012-10-21 | 1 | -1/+1 |
| | |||||
* | Test PopenProcess + PipeLogger. | Zac Medico | 2012-10-18 | 1 | -2/+42 |
| | |||||
* | PipeReader: unregister in _cancel | Zac Medico | 2012-10-18 | 1 | -1/+2 |
| | |||||
* | PipeLogger/Reader: remove redundant isAlive() | Zac Medico | 2012-10-18 | 2 | -6/+0 |
| | | | | Parent class AbstractPollTask has equivalent implementation. | ||||
* | SpawnProcess: split out a PipeLogger class | Zac Medico | 2012-10-18 | 2 | -131/+183 |
| | | | | | | The copyright dates for these classes begin in 2008, since SpawnProcess code is derived from the EbuildFetcherAsync class which was added in commit e4edadf5ae7063f375d76be151c6d0e949980ecf in 2008. | ||||
* | ManifestTask: use null stdin when checking sigv2.2.0_alpha141 | Zac Medico | 2012-10-17 | 1 | -1/+3 |
| | |||||
* | Handle InvalidData from _pkg_str. | Zac Medico | 2012-10-17 | 4 | -8/+18 |
| | |||||
* | Remove old binhost protocol for bug #438640. | Zac Medico | 2012-10-17 | 3 | -70/+42 |
| | | | | | The old protocol has been deprecated since portage-2.1.6 (2009), so it should be completely irrelevant now. | ||||
* | config.setcpv: use constructed _pkg_str for mycpv | Zac Medico | 2012-10-17 | 1 | -1/+2 |
| | |||||
* | emaint: sync MoveHandler check code with dbapi's | Zac Medico | 2012-10-17 | 1 | -4/+12 |
| | |||||
* | emaint: fix MoveHandler for EAPI 5 sub-slot | Zac Medico | 2012-10-17 | 1 | -3/+5 |
| | |||||
* | doebuild: fix REPLACING_VERSIONS EAPI 5 sub-slot | Zac Medico | 2012-10-17 | 1 | -3/+2 |
| | |||||
* | bintree/portree: fix getslot for EAPI 5 sub-slot | Zac Medico | 2012-10-17 | 2 | -8/+4 |
| | |||||
* | SecuritySet: fix for EAPI 5 sub-slot in SLOT | Zac Medico | 2012-10-17 | 1 | -2/+2 |
| | |||||
* | LibraryConsumerSet: fix for EAPI 5 sub-slot | Zac Medico | 2012-10-17 | 1 | -4/+3 |
| | |||||
* | StaticFileSet: fix for EAPI 5 sub-slot in SLOT | Zac Medico | 2012-10-17 | 1 | -2/+2 |
| | |||||
* | _sets/dbapi: fix for EAPI 5 sub-slot in SLOT | Zac Medico | 2012-10-17 | 1 | -14/+11 |
| | |||||
* | depgraph: fix for EAPI 5 sub-slot in SLOT | Zac Medico | 2012-10-17 | 1 | -16/+12 |
| | |||||
* | actions.py: use _pkg_str for EAPI 5 sub-slot | Zac Medico | 2012-10-17 | 1 | -12/+4 |
| | |||||
* | getmaskingreason: use _pkg_str | Zac Medico | 2012-10-17 | 1 | -11/+9 |
| | |||||
* | ManifestTask: stricter gpg key comparison | Zac Medico | 2012-10-17 | 1 | -1/+14 |
| | | | | Use normalization and == comparison instead of the "in" operator. | ||||
* | ManifestTask: use PopenProcess more | Zac Medico | 2012-10-17 | 1 | -3/+1 |
| | |||||
* | PopenProcess: child_watch_add if no pipe_reader | Zac Medico | 2012-10-17 | 1 | -4/+6 |
| | |||||
* | MaskManager: use _pkg_str | Zac Medico | 2012-10-17 | 1 | -11/+20 |
| | |||||
* | config.setcpv: remove unused slot variable | Zac Medico | 2012-10-17 | 1 | -1/+0 |
| | |||||
* | vartree.py: account for EAPI 5 sub-slot in SLOT | Zac Medico | 2012-10-17 | 1 | -16/+18 |
| | |||||
* | create_world_atom: account for EAPI 5 sub-slot | Zac Medico | 2012-10-17 | 1 | -7/+5 |
| | |||||
* | glsa: account for EAPI 5 sub-slot in SLOT | Zac Medico | 2012-10-17 | 1 | -1/+1 |
| | |||||
* | glsa.py: use try/finally for urlopen | Zac Medico | 2012-10-17 | 1 | -1/+6 |
| | | | | | | Fixes this error reported with python 2.7.3: AttributeError: addinfourl instance has no attribute '__exit__' | ||||
* | Scheduler: tweak job_delay calc for bug #438650v2.2.0_alpha140 | Zac Medico | 2012-10-16 | 1 | -5/+16 |
| | | | | | | | | * Max delay is reduced to 5 seconds (was 10). * Calculated delay is now proportional to current/desired load ratio. * When a delay is triggered, an event loop timeout is used to ensure that scheduling occurs as soon as the delay expires (rather than waiting for the _loadavg_latency timeout to expire). | ||||
* | ManifestTask: improve gpg key parsing | Zac Medico | 2012-10-16 | 1 | -7/+10 |
| | | | | | | | This fixes it to correctly parse longer key IDs, which do not fit on the first line of gpg output. Without this fix, failure to parse the key results in manifest being re-signed even though they already have a signature with the correct key. |