| Age | Commit message (Collapse) | Author | |
|---|---|---|---|
| 2014-03-07 | rename $_ for better style... lots more to be done | Marc Espie | |
| 2014-03-07 | the IO::Uncompress framework can auto-detect compression type, turn | Marc Espie | |
| that on to make sure things are okay all the time. | |||
| 2014-03-05 | shush. signify is quiet now, but we want it to stay that way | Marc Espie | |
| 2014-03-05 | refactor solve_from_ports a bit. | Marc Espie | |
| do the write to cache thru a temp file, so the end result is atomic, to be used to enable a global depends cache for parallel building in the near future. | |||
| 2014-02-13 | this one is non obvious enough it deserves a comment | Marc Espie | |
| 2014-02-13 | still need the plists for kept handles... because of OldLibs. | Marc Espie | |
| Problem noticed by henning@ | |||
| 2014-02-11 | s/anyways/anyway/, ok espie@ | Stuart Henderson | |
| 2014-02-10 | pattern for "our" temporary files is not random... get it out of the | Marc Espie | |
| normal "unknown files" routine and allow to remove them beforehand. | |||
| 2014-02-10 | assume tweaking package status always work if $state->{not}. | Marc Espie | |
| Problem noticed by Mikolaj Kucharski | |||
| 2014-02-10 | cvstags are a bit fickle, there's no need to always-update if they change. | Marc Espie | |
| from Mikolaj Kucharski | |||
| 2014-02-09 | forgot about that small issue. Don't take crypto sigs into account for | Marc Espie | |
| "always-update" comparisons (among other things, the crypto sig contains a timestamp which will always change). Put things in line with other "package changed" comparison. Reminded by Mikolaj Kucharski | |||
| 2014-02-09 | wipe stuff during wipe_info, deref should keep unput correct. | Marc Espie | |
| 2014-02-09 | grr... this should be kept across derefs | Marc Espie | |
| 2014-02-09 | do a better job in wipe_info: close files and reap children. | Marc Espie | |
| also wipe every state variable... fixes reget update_info in fringe cases (this should fix nigel@'s problems...) | |||
| 2014-02-08 | be less verbose for pkg_delete -X | Marc Espie | |
| 2014-02-08 | scrape update_info when we have full plists. | Marc Espie | |
| 2014-02-08 | style | Marc Espie | |
| 2014-02-08 | don't allow unsigned packages by default. | Marc Espie | |
| as kettenis@ says "test it now", always possible to revert if we don't want it for the release. | |||
| 2014-02-08 | also limit the number of currently opened packages locally: each uncompress | Marc Espie | |
| gobbles memory. (this was probably already an issue with external gunzip, but not THAT visible because of separate processes). | |||
| 2014-02-08 | fudge verbose state so that makewhatis is only verbose if pkg_add/delete -vv | Marc Espie | |
| or more. | |||
| 2014-02-07 | reclaim: we don't need to keep tracking temporary files/directories we | Marc Espie | |
| removed. | |||
| 2014-02-07 | don't store full terminal, we just need a few control sequences | Marc Espie | |
| 2014-02-06 | http 404 is definitive, so treat it like 550 ftp. | Marc Espie | |
| 2014-02-06 | style | Marc Espie | |
| 2014-02-06 | don't store full directory objects if they're not "special", just knowing | Marc Espie | |
| there's a directory that should stay around is enough. shaves on the number of shared directories for shared items quite a lot. | |||
| 2014-02-04 | sort list of stuff we can't update for user display (otherwise, it's a list | Marc Espie | |
| of keys, so it isn't in any particularly interesting order) | |||
| 2014-02-04 | no functional change (for now). Make things more debuggable by subclassing | Marc Espie | |
| PackingList into types that describe what info they've been limited to. Long term (post 5.5) some operations will be restricted, such as e.g., only committing lists to installation if they're "full", or only checking signatures for same. | |||
| 2014-02-04 | prevent updates from starting with a lone ":" | Marc Espie | |
| Some UpdateSets will only contain kept packages, BUT require the update of some dependencies. just spew out the kept_names instead of newer_names in that case. | |||
| 2014-02-04 | kill very old compatibility code (2009/04... those exceptions are fully dead) | Marc Espie | |
| 2014-02-04 | check for errors before dependencies, anyways. | Marc Espie | |
| split the part for arch, since it requires full plists. avoid creating dummy locations for handles in errors. | |||
| 2014-02-03 | perl is actually VERY BAD at scraping captured variables until the end. | Marc Espie | |
| Avoid creating a gratuitous back reference to that packing-list. Prevents us from keeping shas around... :( | |||
| 2014-02-03 | don't remove this test, indicate why it can trigger | Marc Espie | |
| 2014-02-03 | gc old stuff that never fixed the problem it intended to fix :( | Marc Espie | |
| 2014-02-03 | store compact form of manpages. | Marc Espie | |
| fix a long standing stupid bug: can't store both deleted and added manpages in the same hash, otherwise the first deletion is going to scrape already added stuff. Ouchie. | |||
| 2014-02-03 | remove extra cache copy in memory... | Marc Espie | |
| to be revisited, the setting of info dir so late is slightly broken and will reopen packages more times than needed. | |||
| 2014-02-03 | zap duplicate test, spotted by Arto Jonsson | Marc Espie | |
| 2014-02-03 | don't really need exactstem, the initial grep is good enough | Marc Espie | |
| 2014-02-03 | don't keep tied information around ! | Marc Espie | |
| 2014-02-02 | don't create empty lists that won't be used | Marc Espie | |
| 2014-02-02 | gc really old code | Marc Espie | |
| 2014-02-02 | pseudo handle needs dependency_info too, forgot that one. | Marc Espie | |
| 2014-02-02 | simplify how "kept" packages are handled. First, we never need their | Marc Espie | |
| full plist, just their update_info. Also, the routine for move_kept is pretty much always the same, so fold it. Finally, delay full plists in create_old. | |||
| 2014-02-02 | now that things are clean enough, we just need to figure out kept packages | Marc Espie | |
| before we solve depends. Avoids keeping full packing-lists all along the dependency chain... | |||
| 2014-02-02 | make sure kept handles are complete... not really optimized so far, but will | Marc Espie | |
| fix issues seen by nigel@ | |||
| 2014-02-01 | style | Marc Espie | |
| 2014-02-01 | sort out the code to figuring out kept packages. | Marc Espie | |
| Fix a (formerly) silent bug: kept packages *must* be completed somewhere. (older solve_depends would create dummy plists, new one errors out on them!) | |||
| 2014-02-01 | let solve_depends work as soon as we have update_info | Marc Espie | |
| 2014-02-01 | handling of critical section was wrong. Interrupting an update once we're | Marc Espie | |
| "committed" is a bad idea. Make it possible by ^C^C still, but try to get out only when the db is in a safe state. | |||
| 2014-02-01 | don't bother caching, it's no longer useful | Marc Espie | |
| 2014-02-01 | use conflict_list directly from handle. | Marc Espie | |
| (only user left of plist->conflict_list is PkgCfl::register thru OldLibs stub_list) | |||
 |
index : src | |
| OpenBSD base system |
| summaryrefslogtreecommitdiff |