summaryrefslogtreecommitdiff
path: root/CmdLine.hs
Commit message (Collapse)AuthorAge
* fix failure count memory leakGravatar Joey Hess2012-02-15
| | | | | | | This is the last memory leak that prevents git-annex from running in constant space, as far as I can see. I can now run git annex find dummied up to repeatedly find the same file over and over, on millions olf files, and memory stays entirely constant.
* set oneshot mode on a per-command basisGravatar Joey Hess2012-02-14
| | | | Avoids ugly (and test suite failing) hack in Command.Version
* force state strictlyGravatar Joey Hess2012-02-13
| | | | | | When converting to the strict state monad, I missed this place where thunks to the state could be built up, possibly. This seems to make it run in some percentage less memory.
* IO exception reworkGravatar Joey Hess2012-02-03
| | | | | | ghc 7.4 comaplains about use of System.IO.Error to catch exceptions. Ok, use Control.Exception, with variants specialized to only catch IO exceptions.
* Avoid repeated location log commits when a remote is receiving files.Gravatar Joey Hess2012-01-28
| | | | | | | | | Done by adding a oneshot mode, in which location log changes are written to the journal, but not committed. Taking advantage of git-annex's existing ability to recover in this situation. This is used by git-annex-shell and other places where changes are made to a remote's location log.
* ssh connection cachingGravatar Joey Hess2012-01-20
| | | | | | | | | | | Ssh connection caching is now enabled automatically by git-annex. Only one ssh connection is made to each host per git-annex run, which can speed some things up a lot, as well as avoiding repeated password prompts. Concurrent git-annex processes also share ssh connections. Cached ssh connections are shut down when git-annex exits. Note: The rsync special remote does not yet participate in the ssh connection caching.
* tweaksGravatar Joey Hess2012-01-11
|
* per-command optionsGravatar Joey Hess2012-01-05
| | | | | | Finally commands can define their own options. Moved --format and --print0 to be options only of find.
* Command data structure tweakingGravatar Joey Hess2012-01-05
|
* more partial function removalGravatar Joey Hess2011-12-15
| | | | | Left a few Prelude.head's in where it was checked not null and too hard to remove, etc.
* split out Git/Command.hsGravatar Joey Hess2011-12-14
|
* hslintGravatar Joey Hess2011-12-09
|
* When not run in a git repository, git-annex can still display a usage ↵Gravatar Joey Hess2011-11-16
| | | | | | | message, and "git annex version" even works. Things that sound simple, but are made hard by the Annex monad being built with the assumption that there will always be a git repo.
* cleanupGravatar Joey Hess2011-10-31
|
* add command name to some outputGravatar Joey Hess2011-10-30
|
* refactoring and cleanupGravatar Joey Hess2011-10-30
| | | | No code changes.
* clean up check selection codeGravatar Joey Hess2011-10-29
| | | | | | | | | This new approach allows filtering out checks from the default set that are not appropriate for a command, rather than having to list every check that is appropriate. It also reduces some boilerplate. Haskell does not define Eq for functions, so I had to go a long way around with each check having a unique id. Meh.
* refactored and generalized pre-command sanity checkingGravatar Joey Hess2011-10-27
|
* renameGravatar Joey Hess2011-10-05
|
* renameGravatar Joey Hess2011-10-04
|
* factor out common importsGravatar Joey Hess2011-10-03
| | | | no code changes
* when all you have is a zombie, everything looks like a shotgunGravatar Joey Hess2011-10-02
| | | | | | Actually, let's do a targeted fix of the actual forkProcess that was not waited on. The global reap is moved back to the end, after the long-running git processes actually exit.
* Contain the zombie hordes.aGravatar Joey Hess2011-10-02
| | | | | Specifically, when using gpg, a zombie is forked for each file, so waiting until shutdown to reap won't do.
* Improve display of newlines around error and warning messages.Gravatar Joey Hess2011-09-06
|
* optimise initialized checkGravatar Joey Hess2011-08-17
| | | | Avoid running external command if annex.version is set.
* when reading configs of local repos, first initializeSafeGravatar Joey Hess2011-08-17
| | | | This auto-generates a uuid if the local repo does not already have one.
* Now "git annex init" only has to be run onceGravatar Joey Hess2011-08-17
| | | | | | | when a git repository is first being created. Clones will automatically notice that git-annex is in use and automatically perform a basic initalization. It's still recommended to run "git annex init" in any clones, to describe them.
* hlint tweaksGravatar Joey Hess2011-07-15
| | | | Did all sources except Remotes/* and Command/*
* remove unused backend machineryGravatar Joey Hess2011-07-05
| | | | | | | | | | | | | The only remaining vestiage of backends is different types of keys. These are still called "backends", mostly to avoid needing to change user interface and configuration. But everything to do with storing keys in different backends was gone; instead different types of remotes are used. In the refactoring, lots of code was moved out of odd corners like Backend.File, to closer to where it's used, like Command.Drop and Command.Fsck. Quite a lot of dead code was removed. Several data structures became simpler, which may result in better runtime efficiency. There should be no user-visible changes.
* better display of thrown errorsGravatar Joey Hess2011-07-05
|
* renamed GitRepo to GitGravatar Joey Hess2011-06-30
| | | | It was always imported qualified as Git anyway
* commit git-annex branch when copying to a remote (locally)Gravatar Joey Hess2011-06-22
| | | | | | | Otherwise, the location log changes are only staged in its index, and this can confuse matters if pulling or cloning from the remote. The test suite was failing because this wasn't done.
* store trust.log and remote.log in the git-annex branchGravatar Joey Hess2011-06-22
| | | | .. and I think that's everything that will use the branch
* commit changes to git-annex branch on shutdownGravatar Joey Hess2011-06-22
|
* tweakGravatar Joey Hess2011-06-02
|
* Periodically flush git command queue, to avoid boating memory usage too much.Gravatar Joey Hess2011-04-07
| | | | | | | Since the queue is flushed in between subcommand actions being run, there should be no issues with actions that expect to queue up some stuff and have it run after they do other stuff. So I didn't have to audit for such assumptions.
* avoid version check before running version and upgrade commandsGravatar Joey Hess2011-03-19
| | | | | There are two types of commands; those that access the repository and those that don't. Sorted.
* No longer auto-upgrade to repository format 2, to avoid accidental upgrades, ↵Gravatar Joey Hess2011-03-19
| | | | etc. Use git-annex upgrade when you're ready to run this version.
* use queue when upgrading, flushing every so oftenGravatar Joey Hess2011-03-16
| | | | | | | | | | | | | | | | | Added a cheap way to query the size of a queue. runQueueAt is not the default yet only because there may be some code that expects to be able to queue some suff, do something else, and run the whole queue at the end. 10240 is an arbitrary size for the queue. If we assume annexed filenames are between 10 and 255 characters long, then the queue will build up between 100kb and 2550kb long commands. The max command line length on linux is somewhere above 20k, so this is a fairly good balance -- the queue will buffer only a few megabytes of stuff and a minimal number of commands will be run by xargs. Also, insert queue items strictly, this should save memory.
* Rethink filename encoding handling for display. Since filename encoding may ↵0.23Gravatar Joey Hess2011-03-12
| | | | or may not match locale settings, any attempt to decode filenames will fail for some files. So instead, do all output in binary mode.
* put in utf8 forcing workaroundGravatar Joey Hess2011-03-08
| | | | | | | | | Haskell's IO layer crashes on characters > 255 when in a non-unicode (latin1) locale. Until Haskell gets better behavior, put in an admittedly ugly workaround for that: git-annex forces utf8 output mode no matter what locale is selected. So if you use a non-utf8 locale, your filenames with characters > 127 will not be displayed as you'd expect. But at least it won't crash.
* Fix test suite to reap zombies.Gravatar Joey Hess2011-02-13
| | | | | | | | I had not taken into account that the code was written to run git and leave zombies, for performance/laziness reasons, when I wrote the test suite. So rather than the typical 1 zombie process that git-annex develops, test developed dozens. Caused problems on system with low process limits. Added a reap function to GitRepo, that waits for any zombie child processes.
* more pure code refactoringGravatar Joey Hess2011-01-29
|
* shutdown no longer a special caseGravatar Joey Hess2011-01-29
|
* cleanupGravatar Joey Hess2011-01-29
|
* Improved temp file handlingGravatar Joey Hess2011-01-28
| | | | | | | * Improved temp file handling. Transfers of content can now be resumed from temp files later; the resume does not have to be the immediate next git-annex run. * unused: Include partially transferred content in the list.
* better directory handlingGravatar Joey Hess2011-01-27
| | | | | | | | | | Rename Locations functions for better consitency, and make their values more consistent too. Used </> rather than manually building paths. There are still more places that manually do so, but are tricky, due to the behavior of </> when the second FilePath is absolute. So I only changed places where it obviously was relative.
* better warnings displayGravatar Joey Hess2011-01-26
|
* rework config storageGravatar Joey Hess2011-01-26
| | | | | | | Moved away from a map of flags to storing config directly in the AnnexState structure. Got rid of most accessor functions in Annex. This allowed supporting multiple --exclude flags.
* successfully split Annex and AnnexState out of TypeInternalsGravatar Joey Hess2011-01-25
|