path: root/doc/git-annex.mdwn

# NAME

git-annex - manage files with git, without checking their contents in

# SYNOPSIS

git annex command [params ...]

# DESCRIPTION

git-annex allows managing files with git, without checking the file
contents into git. While that may seem paradoxical, it is useful when
dealing with files larger than git can currently easily handle, whether due
to limitations in memory, checksumming time, or disk space.

Even without file content tracking, being able to manage files with git,
move files around and delete files with versioned directory trees, and use
branches and distributed clones, are all very handy reasons to use git. And
annexed files can co-exist in the same git repository with regularly
versioned files, which is convenient for maintaining documents, Makefiles,
etc that are associated with annexed files but that benefit from full
revision control.

When a file is annexed, its content is moved into a key-value store, and
a symlink is made that points to the content. These symlinks are checked into
git and versioned like regular files. You can move them around, delete 
them, and so on. Pushing to another git repository will make git-annex
there aware of the annexed file, and it can be used to retrieve its
content from the key-value store.

# EXAMPLES

	# git annex get video/hackity_hack_and_kaxxt.mov
	get video/_why_hackity_hack_and_kaxxt.mov (not available)
	  I was unable to access these remotes: server
	  Try making some of these repositories available:
	  	5863d8c0-d9a9-11df-adb2-af51e6559a49  -- my home file server
	   	58d84e8a-d9ae-11df-a1aa-ab9aa8c00826  -- portable USB drive
	   	ca20064c-dbb5-11df-b2fe-002170d25c55  -- backup SATA drive
	failed
	# sudo mount /media/usb
	# git remote add usbdrive /media/usb
	# git annex get video/hackity_hack_and_kaxxt.mov
	get video/hackity_hack_and_kaxxt.mov (from usbdrive...) ok
	
	# git annex add iso
	add iso/Debian_5.0.iso ok
	
	# git annex drop iso/Debian_4.0.iso
	drop iso/Debian_4.0.iso ok
	
	# git annex move iso --to=usbdrive
	move iso/Debian_5.0.iso (moving to usbdrive...) ok

# COMMONLY USED COMMANDS

Like many git commands, git-annex can be passed a path that 
is either a file or a directory. In the latter case it acts on all relevant
files in the directory. When no path is specified, most git-annex commands
default to acting on all relevant files in the current directory (and
subdirectories).

* `add [path ...]`

  Adds files in the path to the annex. Files that are already checked into
  git, or that git has been configured to ignore will be silently skipped.
  (Use `--force` to add ignored files.) Dotfiles are skipped unless explicitly
  listed.

* `get [path ...]`

  Makes the content of annexed files available in this repository. This
  will involve copying them from another repository, or downloading them,
  or transferring them from some kind of key-value store.

  Normally git-annex will choose which repository to copy the content from,
  but you can override this using the `--from` option.

* `drop [path ...]`

  Drops the content of annexed files from this repository. 

  git-annex will refuse to drop content if it cannot verify it is
  safe to do so. This can be overridden with the `--force` switch.

  To drop content from a remote, specify `--from`.

* `move [path ...]`
  
  When used with the `--from` option, moves the content of annexed files
  from the specified repository to the current one.

  When used with the `--to` option, moves the content of annexed files from
  the current repository to the specified one.

* `copy [path ...]`

  When used with the `--from` option, copies the content of annexed files
  from the specified repository to the current one.

  When used with the `--to` option, copies the content of annexed files from
  the current repository to the specified one.

  To avoid contacting the remote to check if it has every file
  when copying --to the repository, specify `--fast`

  To force checking the remote for every file when copying --from the
  repository, specify `--force`.

* `status [path ...]`

  Similar to `git status --short`, displays the status of the files in the
  working tree. Shows files that are not checked into git, files that
  have been deleted, and files that have been modified. 
  Particulary useful in direct mode.

* `unlock [path ...]`

  Normally, the content of annexed files is protected from being changed.
  Unlocking a annexed file allows it to be modified. This replaces the
  symlink for each specified file with a copy of the file's content.
  You can then modify it and `git annex add` (or `git commit`) to inject
  it back into the annex.

* `edit [path ...]`

  This is an alias for the unlock command. May be easier to remember,
  if you think of this as allowing you to edit an annexed file.

* `lock [path ...]`

  Use this to undo an unlock command if you don't want to modify
  the files, or have made modifications you want to discard.

* `sync [remote ...]`

  Use this command when you want to synchronize the local repository with
  one or more of its remotes. You can specify the remotes to sync with;
  the default is to sync with all remotes. Or specify `--fast` to sync with
  the remotes with the lowest annex-cost value.

  The sync process involves first committing all local changes
  then fetching and merging the `synced/master` and the `git-annex` branch
  from the remote repositories and finally pushing the changes back to
  those branches on the remote repositories. You can use standard git
  commands to do each of those steps by hand, or if you don't want to
  worry about the details, you can use sync.

  Merge conflicts are automatically resolved by sync. When two conflicting
  versions of a file have been committed, both will be added to the tree,
  under different filenames. For example, file "foo" would be replaced
  with "foo.somekey" and "foo.otherkey".

  Note that syncing with a remote will not update the remote's working
  tree with changes made to the local repository. However, those changes
  are pushed to the remote, so can be merged into its working tree
  by running "git annex sync" on the remote.

  Note that sync does not transfer any file contents from or to the remote
  repositories.

* `merge`

  This performs the same merging that is done by the sync command, but
  without pushing or pulling any data.

  One way to use this is to put `git annex merge` into a repository's
  post-receive hook. Then any syncs to the repository will update its working
  copy automatically.

* `mirror [path ...]`

  This causes a destination repository to mirror a source repository.

  To use the local repository as the source repository,
  specify mirror `--to` remote.

  To use a remote as the source repository, specify mirror `--from` remote.

  Each specified file in the source repository is mirrored to the destination
  repository. If a file's content is present in the source repository, it is
  copied to the destination repository. If a file's content is not present in
  the source repository, it will be dropped from the destination repository
  when the numcopies setting allows.

  Note that mirror does not sync the git repository, but only the file
  contents.

  Also, --all may be specified to mirror all objects stored in the git
  annex, not only objects used by currently existing files. However, this
  bypasses checking the .gitattributes annex.numcopies setting when
  dropping files.

* `addurl [url ...]`

  Downloads each url to its own file, which is added to the annex.

  To avoid immediately downloading the url, specify `--fast`.

  To avoid storing the size of the url's content, and accept whatever
  is there at a future point, specify `--relaxed`. (Implies `--fast`.)

  Normally the filename is based on the full url, so will look like
  "www.example.com_dir_subdir_bigfile". For a shorter filename, specify
  `--pathdepth=N`. For example, `--pathdepth=1` will use "dir/subdir/bigfile",
  while `--pathdepth=3` will use "bigfile". It can also be negative;
  `--pathdepth=-2` will use the last two parts of the url.

  Or, to directly specify what file the url is added to, specify `--file`.
  This changes the behavior; now all the specified urls are recorded as
  alternate locations from which the file can be downloaded. In this mode,
  addurl can be used both to add new files, or to add urls to existing files.

  When quvi is installed, urls are automatically tested to see if they
  are on a video hosting site, and the video is downloaded instead.

* `rmurl file url`

  Record that the file is no longer available at the url.

* `import [path ...]`

  Moves files from somewhere outside the git working copy, and adds them to
  the annex. Individual files to import can be specified. 
  If a directory is specified, the entire directory is imported.

        	git annex import /media/camera/DCIM/*

  By default, importing two files with the same contents from two different
  locations will result in both files being added to the repository.
  (With all checksumming backends, including the default SHA256E,
  only one copy of the data will be stored.)

  To not delete files from the import location, use the
  `--duplicate` option. This could allow importing the same files repeatedly
  to different locations in a repository. More likely, it could be used to
  import the same files to a number of different branches or separate git
  repositories.

  To only import files whose content has not been seen before by git-annex,
  use the `--deduplicate` option. Duplicate files will be deleted from the
  import location.

  To only import files whose content has not been seen before by git-annex,
  but avoid deleting duplicate files, use the `--skip-duplicates` option.

  The `--clean-duplicates` option does not import any new files, but any files
  found in the import location that are duplicates of content in the annex
  are deleted.

  (Note that using `--deduplicate` or `--clean-duplicates` with the WORM
  backend does not look at file content, but filename and mtime.)

* `importfeed [url ...]`

  Imports the contents of podcast feeds. Only downloads files whose
  urls have not already been added to the repository before, so you can
  delete, rename, etc the resulting files and repeated runs won't duplicate
  them. (Use `--force` to force downloading urls it's seen before.)

  Use `--template` to control where the files are stored.
  The default template is '${feedtitle}/${itemtitle}${extension}'
  (Other available variables: feedauthor, itemauthor, itemsummary, itemdescription, itemrights, itemid)

  The `--relaxed` and `--fast` options behave the same as they do in addurl.
  
  When quvi is installed, links in the feed are tested to see if they
  are on a video hosting site, and the video is downloaded. This allows
  importing eg, youtube playlists.

* `watch`

  Watches for changes to files in the current directory and its subdirectories,
  and takes care of automatically adding new files, as well as dealing with
  deleted, copied, and moved files. With this running as a daemon in the
  background, you no longer need to manually run git commands when
  manipulating your files.

  By default, all files in the directory will be added to the repository.
  (Including dotfiles.) To block some files from being added, use
  `.gitignore` files.

  By default, all files that are added are added to the annex, the same
  as when you run `git annex add`. If you configure annex.largefiles,
  files that it does not match will instead be added with `git add`.

  To not daemonize, run with `--foreground` ; to stop a running daemon,
  run with `--stop`

* `assistant`

  Like watch, but also automatically syncs changes to other remotes.
  Typically started at boot, or when you log in.

  With the `--autostart` option, the assistant is started in any repositories
  it has created. These are listed in `~/.config/git-annex/autostart`

* `webapp`

  Opens a web app, that allows easy setup of a git-annex repository,
  and control of the git-annex assistant.

  By default, the webapp can only be accessed from localhost, and running
  it opens a browser window.

  With the `--listen=address[:port]` option, the webapp can be made to listen
  for connections on the specified address. This disables running a
  local web browser, and outputs the url you can use to open the webapp
  from a remote computer.
  Note that this does not yet use HTTPS for security, so use with caution!

# REPOSITORY SETUP COMMANDS

* `init [description]`

  Until a repository (or one of its remotes) has been initialized,
  git-annex will refuse to operate on it, to avoid accidentally
  using it in a repository that was not intended to have an annex.

  It's useful, but not mandatory, to initialize each new clone
  of a repository with its own description. If you don't provide one,
  one will be generated.

* `describe repository description`

  Changes the description of a repository.

  The repository to describe can be specified by git remote name or
  by uuid. To change the description of the current repository, use
  "here".

* `initremote name [param=value ...]`

  Creates a new special remote, and adds it to `.git/config`. 

  The remote's configuration is specified by the parameters. Different
  types of special remotes need different configuration values. The
  command will prompt for parameters as needed.

  All special remotes support encryption. You can either specify
  `encryption=none` to disable encryption, or specify
  `encryption=hybrid keyid=$keyid ...` to specify a GPG key id (or an email
  address associated with a key.)

  There are actually three schemes that can be used for management of the
  encryption keys. When using the encryption=hybrid scheme, additional
  GPG keys can be given access to the encrypted special remote easily
  (without re-encrypting everything). When using encryption=shared,
  a shared key is generated and stored in the git repository, allowing
  anyone who can clone the git repository to access it. Finally, when using
  encryption=pubkey, content in the special remote is directly encrypted
  to the specified GPG keys, and additional ones cannot easily be given
  access.

  Note that with encryption enabled, a cryptographic key is created.
  This requires sufficient entropy. If initremote seems to hang or take
  a long time while generating the key, you may want to Ctrl-c it and
  re-run with `--fast`, which causes it to use a lower-quality source of
  randomness.

  Example Amazon S3 remote:

    	git annex initremote mys3 type=S3 encryption=hybrid keyid=me@example.com datacenter=EU

* `enableremote name [param=value ...]`

  Enables use of an existing special remote in the current repository,
  which may be a different repository than the one in which it was
  originally created with the initremote command.

  The name of the remote is the same name used when originally 
  creating that remote with "initremote". Run "git annex enableremote"
  with no parameters to get a list of special remote names.

  Some special remotes may need parameters to be specified every time.
  For example, the directory special remote requires a directory= parameter.

  This command can also be used to modify the configuration of an existing
  special remote, by specifying new values for parameters that were
  originally set when using initremote. (However, some settings such as
  the as the encryption scheme cannot be changed once a special remote
  has been created.)

  The GPG keys that an encrypted special remote is encrypted to can be
  changed using the keyid+= and keyid-= parameters. These respectively
  add and remove keys from the list. However, note that removing a key
  does NOT necessarily prevent the key's owner from accessing data
  in the encrypted special remote
  (which is by design impossible, short of deleting the remote).

  One use-case of keyid-= is to replace a revoked key with
  a new key:

        	git annex enableremote mys3 keyid-=revokedkey keyid+=newkey

  Also, note that for encrypted special remotes using plain public-key
  encryption (encryption=pubkey), adding or removing a key has NO effect
  on files that have already been copied to the remote. Hence using
  keyid+= and keyid-= with such remotes should be used with care, and
  make little sense except in cases like the revoked key example above.

* `trust [repository ...]`

  Records that a repository is trusted to not unexpectedly lose
  content. Use with care.

  To trust the current repository, use "here".

* `untrust [repository ...]`

  Records that a repository is not trusted and could lose content
  at any time.

* `semitrust [repository ...]`

  Returns a repository to the default semi trusted state.

* `dead [repository ...]`

  Indicates that the repository has been irretrevably lost.
  (To undo, use semitrust.)

* `group repository groupname`

  Adds a repository to a group, such as "archival", "enduser", or "transfer".
  The groupname must be a single word.

* `ungroup repository groupname`

  Removes a repository from a group.

* `wanted repository [expression]`

  When run with an expression, configures the content that is preferred
  to be held in the archive. See PREFERRED CONTENT below.

  For example:

        	git annex wanted . "include=*.mp3 or include=*.ogg"

  Without an expression, displays the current preferred content setting
  of the repository.

* `schedule repository [expression]`

  When run with an expression, configures scheduled jobs to run at a
  particular time. This can be used to make the assistant periodically run
  incremental fscks. See SCHEDULED JOBS below.

* `vicfg`

  Opens EDITOR on a temp file containing most of the above configuration
  settings, and when it exits, stores any changes made back to the git-annex
  branch.

* `direct`

  Switches a repository to use direct mode, where rather than symlinks to
  files, the files are directly present in the repository.

  As part of the switch to direct mode, any changed files will be committed.

  Note that git commands that operate on the work tree are often unsafe to
  use in direct mode repositories, and can result in data loss or other
  bad behavior.

* `indirect`

  Switches a repository back from direct mode to the default, indirect mode.

  As part of the switch from direct mode, any changed files will be committed.

# REPOSITORY MAINTENANCE COMMANDS

* `fsck [path ...]`

  With no parameters, this command checks the whole annex for consistency,
  and warns about or fixes any problems found. This is a good compliment to
  `git fsck`.

  With parameters, only the specified files are checked.

  To check a remote to fsck, specify `--from`.

  To avoid expensive checksum calculations (and expensive transfers when
  fscking a remote), specify `--fast`.

  To start a new incremental fsck, specify `--incremental`. Then
  the next time you fsck, you can specify `--more` to skip over
  files that have already been checked, and continue where it left off.

  The `--incremental-schedule` option makes a new incremental fsck be
  started a configurable time after the last incremental fsck was started.
  Once the current incremental fsck has completely finished, it causes
  a new one to start.

  Maybe you'd like to run a fsck for 5 hours at night, picking up each
  night where it left off. You'd like this to continue until all files
  have been fscked. And once it's done, you'd like a new fsck pass to start,
  but no more often than once a month. Then put this in a nightly cron job:

        	git annex fsck --incremental-schedule 30d --time-limit 5h

  To verify data integrity only while disregarding required number of copies,
  use `--numcopies=1`.

* `unused`

  Checks the annex for data that does not correspond to any files present
  in any tag or branch, and prints a numbered list of the data.

  To only show unused temp and bad files, specify `--fast`.

  To check for annexed data on a remote, specify `--from`.

  After running this command, you can use the `--unused` option to
  operate on all the unused data that was found. For example, to
  move all unused data to origin:

        	git annex unused; git annex move --unused --to origin

* `dropunused [number|range ...]`

  Drops the data corresponding to the numbers, as listed by the last
  `git annex unused`

  You can also specify ranges of numbers, such as "1-1000".
  Or, specify "all" to drop all unused data.

  To drop the data from a remote, specify `--from.`

* `addunused [number|range ...]`

  Adds back files for the content corresponding to the numbers or ranges,
  as listed by the last `git annex unused`. The files will have names
  starting with "unused."

* `fix [path ...]`

  Fixes up symlinks that have become broken to again point to annexed content.
  This is useful to run if you have been moving the symlinks around,
  but is done automatically when committing a change with git too.

* `upgrade`

  Upgrades the repository to current layout.

* `forget`

  Causes the git-annex branch to be rewritten, throwing away historical
  data about past locations of files. The resulting branch will use less
  space, but `git annex log` will not be able to show where
  files used to be located.

  To also prune references to repositories that have been marked as dead,
  specify `--drop-dead`.

  When this rewritten branch is merged into other clones of
  the repository, `git-annex` will automatically perform the same rewriting
  to their local `git-annex` branches. So the forgetfulness will automatically
  propagate out from its starting point until all repositories running
  git-annex have forgotten their old history. (You may need to force
  git to push the branch to any git repositories not running git-annex.)

* `repair`

  This can repair many of the problems with git repositories that `git fsck`
  detects, but does not itself fix. It's useful if a repository has become
  badly damaged. One way this can happen is if a repository used by git-annex
  is on a removable drive that gets unplugged at the wrong time.

  This command can actually be used inside git repositories that do not
  use git-annex at all; when used in a repository using git-annex, it
  does additional repairs of the git-annex branch.

  It works by deleting any corrupt objects from the git repository, and
  retrieving all missing objects it can from the remotes of the repository.

  If that is not sufficient to fully recover the repository, it can also
  reset branches back to commits before the corruption happened, delete
  branches that are no longer available due to the lost data, and remove any
  missing files from the index. It will only do this if run with the
  `--force` option, since that rewrites history and throws out missing data.
  Note that the `--force` option never touches tags, even if they are no
  longer usable due to missing data.

  After running this command, you will probably want to run `git fsck` to
  verify it fixed the repository. Note that fsck may still complain about
  objects referenced by the reflog, or the stash, if they were unable to be
  recovered. This command does not try to clean up either the reflog or the
  stash.

  It is also a good idea to run `git annex fsck --fast` after this command,
  to make sure that the git-annex branch reflects reality.

# QUERY COMMANDS

* `find [path ...]`

  Outputs a list of annexed files in the specified path. With no path,
  finds files in the current directory and its subdirectories.

  By default, only lists annexed files whose content is currently present.
  This can be changed by specifying file matching options. To list all
  annexed files, present or not, specify `--include "*"`. To list all
  annexed files whose content is not present, specify `--not --in=here`

  To output filenames terminated with nulls, for use with xargs -0,
  specify `--print0`. Or, a custom output formatting can be specified using
  `--format`. The default output format is the same as `--format='${file}\\n'`

  These variables are available for use in formats: file, key, backend,
  bytesize, humansize, keyname, hashdirlower, hashdirmixed, mtime.

* `whereis [path ...]`

  Displays a information about where the contents of files are located.

* `list [path ...]`

  Displays a table of remotes that contain the contents of the specified
  files. This is similar to whereis but a more compact display. Only
  configured remotes are shown by default; specify --allrepos to list
  all repositories.

* `log [path ...]`

  Displays the location log for the specified file or files,
  showing each repository they were added to ("+") and removed from ("-").

  To limit how far back to search for location log changes, the options
  `--since`, `--after`, `--until`, `--before`, and `--max-count` can be specified.
  They are passed through to git log. For example, `--since "1 month ago"`

  To generate output suitable for the gource visualisation program,
  specify `--gource`.

* `info [directory ...]`

  Displays some statistics and other information, including how much data
  is in the annex and a list of all known repositories.

  To only show the data that can be gathered quickly, use `--fast`.

  When a directory is specified, shows a differently formatted info
  display for that directory. In this mode, all of the file matching
  options can be used to filter the files that will be included in
  the information.

  For example, suppose you want to run "git annex get .", but
  would first like to see how much disk space that will use.
  Then run:

        	git annex info --fast . --not --in here
* `version`

  Shows the version of git-annex, as well as repository version information.

* `map`

  Helps you keep track of your repositories, and the connections between them,
  by going out and looking at all the ones it can get to, and generating a
  Graphviz file displaying it all. If the `dot` command is available, it is
  used to display the file to your screen (using x11 backend). (To disable
  this display, specify `--fast`)

  This command only connects to hosts that the host it's run on can
  directly connect to. It does not try to tunnel through intermediate hosts.
  So it might not show all connections between the repositories in the network.

  Also, if connecting to a host requires a password, you might have to enter
  it several times as the map is being built.

  Note that this subcommand can be used to graph any git repository; it
  is not limited to git-annex repositories.

# UTILITY COMMANDS

* `migrate [path ...]`

  Changes the specified annexed files to use the default key-value backend
  (or the one specified with `--backend`). Only files whose content
  is currently available are migrated.

  Note that the content is also still available using the old key after
  migration. Use `git annex unused` to find and remove the old key.

  Normally, nothing will be done to files already using the new backend.
  However, if a backend changes the information it uses to construct a key,
  this can also be used to migrate files to use the new key format.

* `reinject src dest`

  Moves the src file into the annex as the content of the dest file.
  This can be useful if you have obtained the content of a file from
  elsewhere and want to put it in the local annex.

  Automatically runs fsck on dest to check that the expected content was
  provided.

  Example:

        	git annex reinject /tmp/foo.iso foo.iso

* `unannex [path ...]`

  Use this to undo an accidental `git annex add` command. It puts the
  file back how it was before the add.

  Note that for safety, the content of the file remains in the annex,
  until you use `git annex unused` and `git annex dropunused`.

  This is not the command you should use if you intentionally annexed a
  file and don't want its contents any more. In that case you should use
  `git annex drop` instead, and you can also `git rm` the file.

  Normally this does a slow copy of the file. In `--fast` mode, it
  instead makes a hard link from the file to the content in the annex.
  But use --fast mode with caution, because editing the file will
  change the content in the annex.

* `uninit`

  Use this to stop using git annex. It will unannex every file in the
  repository, and remove all of git-annex's other data, leaving you with a
  git repository plus the previously annexed files.

# PLUMBING COMMANDS

* `pre-commit [path ...]`

  This is meant to be called from git's pre-commit hook. `git annex init`
  automatically creates a pre-commit hook using this.

  Fixes up symlinks that are staged as part of a commit, to ensure they
  point to annexed content. Also handles injecting changes to unlocked
  files into the annex.

* `lookupkey [file ...]`

  This plumbing-level command looks up the key used for a file in the
  index. The key is output to stdout. If there is no key (because
  the file is not present in the index, or is not a git-annex managed file),
  nothing is output, and it exits nonzero.

* `examinekey [key ...]`

  This plumbing-level command is given a key, and prints information
  that can be determined purely by looking at the key.

  To specify what information to print, use `--format`. Or use `--json`
  to get all available information in JSON format.

  The same variables can be used in the format string as can be used in
  the format string of git annex find (except there is no file option
  here).

  For example, the location a key's value is stored (in indirect mode)
  can be looked up by running:
  
        	git annex examinekey --format='.git/annex/objects/${hashdirmixed}${key}/${key}'

* `fromkey key file`

  This plumbing-level command can be used to manually set up a file
  in the git repository to link to a specified key.

* `dropkey [key ...]`

  This plumbing-level command drops the annexed data for the specified
  keys from this repository.

  This can be used to drop content for arbitrary keys, which do not need
  to have a file in the git repository pointing at them.

  Example:

        	git annex dropkey SHA1-s10-7da006579dd64330eb2456001fd01948430572f2

* `transferkey`

  This plumbing-level command is used to request a single key be
  transferred. Either the --from or the --to option can be used to specify
  the remote to use. A --file option can be used to hint at the file
  associated with the key.

* `transferkeys`

  This plumbing-level command is used by the assistant to transfer data.
  It is fed instructions about the keys to transfer using an internal
  stdio protocol, which is intentionally not documented (as it may change
  at any time).

* `rekey [file key ...]`

  This plumbing-level command is similar to migrate, but you specify
  both the file, and the new key to use for it.

  With `--force`, even files whose content is not currently available will
  be rekeyed. Use with caution.

* `test`

  This runs git-annex's built-in test suite.

* `xmppgit`

  This command is used internally to perform git pulls over XMPP.

# OPTIONS

* `--force`

  Force unsafe actions, such as dropping a file's content when no other
  source of it can be verified to still exist, or adding ignored files.
  Use with care.

* `--fast`

  Enables less expensive, but also less thorough versions of some commands.
  What is avoided depends on the command.

* `--auto`

  Enables automatic mode. Commands that get, drop, or move file contents
  will only do so when needed to help satisfy the setting of annex.numcopies,
  and preferred content configuration.

* `--all`

  Operate on all data that has been stored in the git annex,
  including old versions of files. This is the default behavior when
  running git-annex in a bare repository; in a non-bare repository the
  normal behavior is to only operate on specified files in the working
  tree.

* `--unused`

  Operate on all data that has been determined to be unused by
  a previous run of `git-annex unused`.

* `--quiet`

  Avoid the default verbose display of what is done; only show errors
  and progress displays.

* `--verbose`

  Enable verbose display.

* `--json`

  Rather than the normal output, generate JSON. This is intended to be
  parsed by programs that use git-annex. Each line of output is a JSON
  object. Note that json output is only usable with some git-annex commands,
  like info and find.

* `--debug`

  Show debug messages.

* `--no-debug`

  Disable debug messages.

* `--from=repository`

  Specifies a repository that content will be retrieved from, or that
  should otherwise be acted on.

  It should be specified using the name of a configured remote.

* `--to=repository`

  Specifies a repository that content will be sent to.

  It should be specified using the name of a configured remote.

* `--numcopies=n`

  Overrides the `annex.numcopies` setting, forcing git-annex to ensure the
  specified number of copies exist. 

  Note that setting numcopies to 0 is very unsafe.

* `--time-limit=time`

  Limits how long a git-annex command runs. The time can be something
  like "5h", or "30m" or even "45s" or "10d".

  Note that git-annex may continue running a little past the specified
  time limit, in order to finish processing a file.

  Also, note that if the time limit prevents git-annex from doing all it 
  was asked to, it will exit with a special code, 101.

* `--trust=repository`
* `--semitrust=repository`
* `--untrust=repository`

  Overrides trust settings for a repository. May be specified more than once.

  The repository should be specified using the name of a configured remote,
  or the UUID or description of a repository.

* `--trust-glacier-inventory`

  Amazon Glacier inventories take hours to retrieve, and may not represent
  the current state of a repository. So git-annex does not trust that
  files that the inventory claims are in Glacier are really there.
  This switch can be used to allow it to trust the inventory.

  Be careful using this, especially if you or someone else might have recently
  removed a file from Glacier. If you try to drop the only other copy of the
  file, and this switch is enabled, you could lose data!  

* `--backend=name`

  Specifies which key-value backend to use. This can be used when
  adding a file to the annex, or migrating a file. Once files
  are in the annex, their backend is known and this option is not
  necessary.

* `--format=value`

  Specifies a custom output format. The value is a format string, 
  in which '${var}' is expanded to the value of a variable. To right-justify
  a variable with whitespace, use '${var;width}' ; to left-justify
  a variable, use '${var;-width}'; to escape unusual characters in a variable,
  use '${escaped_var}'

  Also, '\\n' is a newline, '\\000' is a NULL, etc.

* `--user-agent=value`

  Overrides the User-Agent to use when downloading files from the web.

* `-c name=value`

  Used to override git configuration settings. May be specified multiple times.

# FILE MATCHING OPTIONS

These options can all be specified multiple times, and can be combined to
limit which files git-annex acts on.

Arbitrarily complicated expressions can be built using these options.
For example:

	--exclude '*.mp3' --and --not -( --in=usbdrive --or --in=archive -)

The above example prevents git-annex from working on mp3 files whose
file contents are present at either of two repositories.

* `--exclude=glob`

  Skips files matching the glob pattern. The glob is matched relative to
  the current directory. For example:

        	--exclude='*.mp3' --exclude='subdir/*'

* `--include=glob`

  Skips files not matching the glob pattern.  (Same as `--not --exclude`.)
  For example, to include only mp3 and ogg files:

        	--include='*.mp3' --or --include='*.ogg'

* `--in=repository`

  Matches only files that git-annex believes have their contents present
  in a repository. Note that it does not check the repository to verify
  that it still has the content.

  The repository should be specified using the name of a configured remote,
  or the UUID or description of a repository. For the current repository,
  use `--in=here`

* `--copies=number`

  Matches only files that git-annex believes to have the specified number
  of copies, or more. Note that it does not check remotes to verify that
  the copies still exist.

* `--copies=trustlevel:number`

  Matches only files that git-annex believes have the specified number of
  copies, on remotes with the specified trust level. For example,
  `--copies=trusted:2`

  To match any trust level at or higher than a given level,
  use 'trustlevel+'. For example, `--copies=semitrusted+:2`

* `--copies=groupname:number`

  Matches only files that git-annex believes have the specified number of
  copies, on remotes in the specified group. For example,
  `--copies=archive:2`

* `--inbackend=name`

  Matches only files whose content is stored using the specified key-value
  backend.

* `--inallgroup=groupname`

  Matches only files that git-annex believes are present in all repositories
  in the specified group.

* `--smallerthan=size`
* `--largerthan=size`

  Matches only files whose content is smaller than, or larger than the
  specified size.

  The size can be specified with any commonly used units, for example,
  "0.5 gb" or "100 KiloBytes"

* `--want-get`

  Matches files that the preferred content settings for the repository
  make it want to get. Note that this will match even files that are
  already present, unless limited with eg, `--not --in .`

* `--want-drop`

  Matches files that the preferred content settings for the repository
  make it want to drop. Note that this will match even files that have
  already been dropped, unless limited with eg, `--in .`

* `--not`

  Inverts the next file matching option. For example, to only act on
  files with less than 3 copies, use `--not --copies=3`

* `--and`

  Requires that both the previous and the next file matching option matches.
  The default.

* `--or`

  Requires that either the previous, or the next file matching option matches.

* `-(`

  Opens a group of file matching options.

* `-)`

  Closes a group of file matching options.

# PREFERRED CONTENT

Each repository has a preferred content setting, which specifies content
that the repository wants to have present. These settings can be configured
using `git annex vicfg` or `git annex wanted`.
They are used by the `--auto` option, and by the git-annex assistant.

The preferred content settings are similar, but not identical to
the file matching options specified above, just without the dashes.
For example:

	exclude=archive/* and (include=*.mp3 or smallerthan=1mb)

The main differences are that `exclude=` and `include=` always
match relative to the top of the git repository, and that there is
no equivilant to `--in`.

When a repository is in one of the standard predefined groups, like "backup"
and "client", setting its preferred content to "standard" will use a
built-in preferred content expression developed for that group.

# SCHEDULED JOBS

The git-annex assistant daemon can be configured to run scheduled jobs.
This is similar to cron and anacron (and you can use them if you prefer),
but has the advantage of being integrated into git-annex, and so being able
to eg, fsck a repository on a removable drive when the drive gets
connected.

The scheduled jobs can be configured using `git annex vicfg` or
`git annex schedule`. 

These actions are available: "fsck self", "fsck UUID" (where UUID
is the UUID of a remote to fsck). After the action comes the duration
to allow the action to run, and finally the schedule of when to run it.

To schedule multiple jobs, separate them with "; ".

Some examples:

	fsck self 30m every day at any time
	fsck self 1h every month at 3 AM
	fsck self 1h on day 1 of every month at any time
	fsck self 1h every week divisible by 2 at any time

# CONFIGURATION VIA .git/config

Like other git commands, git-annex is configured via `.git/config`.
Here are all the supported configuration settings.

* `annex.uuid`

  A unique UUID for this repository (automatically set).

* `annex.numcopies`

  Number of copies of files to keep across all repositories. (default: 1)

  Note that setting numcopies to 0 is very unsafe.

* `annex.backends`

  Space-separated list of names of the key-value backends to use.
  The first listed is used to store new files by default.

* `annex.diskreserve`

  Amount of disk space to reserve. Disk space is checked when transferring
  content to avoid running out, and additional free space can be reserved
  via this option, to make space for more important content (such as git
  commit logs). Can be specified with any commonly used units, for example,
  "0.5 gb", "500M", or "100 KiloBytes"

  The default reserve is 1 megabyte.

* `annex.largefiles`

  Allows configuring which files `git annex add` and the assistant consider
  to be large enough to need to be added to the annex. By default,
  all files are added to the annex.

  The value is a preferred content expression. See PREFERRED CONTENT
  for details.

  Example:

        	annex.largefiles = largerthan=100kb and not (include=*.c or include=*.h)

* `annex.queuesize`

  git-annex builds a queue of git commands, in order to combine similar
  commands for speed. By default the size of the queue is limited to
  10240 commands; this can be used to change the size. If you have plenty
  of memory and are working with very large numbers of files, increasing
  the queue size can speed it up.

* `annex.bloomcapacity`

  The `git annex unused` command uses a bloom filter to determine
  what data is no longer used. The default bloom filter is sized to handle
  up to 500000 keys. If your repository is larger than that,
  you can adjust this to avoid `git annex unused` not noticing some unused
  data files. Increasing this will make `git-annex unused` consume more memory;
  run `git annex info` for memory usage numbers.

* `annex.bloomaccuracy`

  Adjusts the accuracy of the bloom filter used by
  `git annex unused`. The default accuracy is 1000 -- 
  1 unused file out of 1000 will be missed by `git annex unused`. Increasing
  the accuracy will make `git annex unused` consume more memory;
  run `git annex info` for memory usage numbers.

* `annex.sshcaching`

  By default, git-annex caches ssh connections
  (if built using a new enough ssh). To disable this, set to `false`.

* `annex.alwayscommit`

  By default, git-annex automatically commits data to the git-annex branch
  after each command is run. To disable these commits,
  set to `false`. Then data will only be committed when
  running `git annex merge` (or by automatic merges) or `git annex sync`.

* `annex.delayadd`

  Makes the watch and assistant commands delay for the specified number of
  seconds before adding a newly created file to the annex. Normally this
  is not needed, because they already wait for all writers of the file
  to close it. On Mac OSX, when not using direct mode this defaults to
  1 second, to work around a bad interaction with software there.

* `annex.fscknudge`

  When set to false, prevents the webapp from reminding you when using
  repositories that lack consistency checks.

* `annex.autoupgrade`

  When set to ask (the default), the webapp will check for new versions
  and prompt if they should be upgraded to. When set to true, automatically
  upgrades without prompting (on some supported platforms). When set to
  false, disables any upgrade checking.

  Note that upgrade checking is only done when git-annex is installed
  from one of the prebuilt images from its website. This does not
  bypass e.g., a Linux distribution's own upgrade handling code.

  This setting also controls whether to restart the git-annex assistant
  when the git-annex binary is detected to have changed. That is useful
  no matter how you installed git-annex.

* `annex.autocommit`

  Set to false to prevent the git-annex assistant from automatically
  committing changes to files in the repository.

* `annex.debug`

  Set to true to enable debug logging by default.

* `annex.version`

  Automatically maintained, and used to automate upgrades between versions.

* `annex.direct`

  Set to true when the repository is in direct mode. Should not be set
  manually; use the "git annex direct" and "git annex indirect" commands
  instead.

* `annex.crippledfilesystem`

  Set to true if the repository is on a crippled filesystem, such as FAT,
  which does not support symbolic links, or hard links, or unix permissions.
  This is automatically probed by "git annex init".

* `remote.<name>.annex-cost`

  When determining which repository to
  transfer annexed files from or to, ones with lower costs are preferred.
  The default cost is 100 for local repositories, and 200 for remote
  repositories.

* `remote.<name>.annex-cost-command`

  If set, the command is run, and the number it outputs is used as the cost.
  This allows varying the cost based on eg, the current network. The
  cost-command can be any shell command line.

* `remote.<name>.annex-start-command`

  A command to run when git-annex begins to use the remote. This can
  be used to, for example, mount the directory containing the remote.

  The command may be run repeatedly when multiple git-annex processes
  are running concurrently.

* `remote.<name>.annex-stop-command`

  A command to run when git-annex is done using the remote.

  The command will only be run once *all* running git-annex processes
  are finished using the remote.

* `remote.<name>.annex-ignore`

  If set to `true`, prevents git-annex
  from storing file contents on this remote by default.
  (You can still request it be used by the `--from` and `--to` options.)

  This is, for example, useful if the remote is located somewhere
  without git-annex-shell. (For example, if it's on GitHub).
  Or, it could be used if the network connection between two
  repositories is too slow to be used normally.

  This does not prevent git-annex sync (or the git-annex assistant) from
  syncing the git repository to the remote.

* `remote.<name>.annex-sync`

  If set to `false`, prevents git-annex sync (and the git-annex assistant)
  from syncing with this remote.

* `remote.<name>.annex-readonly`

  If set to `true`, prevents git-annex from making changes to a remote.
  This both prevents git-annex sync from pushing changes, and prevents
  storing or removing files from read-only remote.

* `remote.<name>.annexUrl`

  Can be used to specify a different url than the regular `remote.<name>.url`
  for git-annex to use when talking with the remote. Similar to the `pushUrl`
  used by git-push.

* `remote.<name>.annex-uuid`

  git-annex caches UUIDs of remote repositories here.

* `remote.<name>.annex-trustlevel`

  Configures a local trust level for the remote. This overrides the value
  configured by the trust and untrust commands. The value can be any of
  "trusted", "semitrusted" or "untrusted".

* `remote.<name>.annex-ssh-options`

  Options to use when using ssh to talk to this remote.

* `remote.<name>.annex-rsync-options`

  Options to use when using rsync
  to or from this remote. For example, to force ipv6, and limit
  the bandwidth to 100Kbyte/s, set it to `-6 --bwlimit 100`

* `remote.<name>.annex-rsync-transport`

  The remote shell to use to connect to the rsync remote. Possible
  values are `ssh` (the default) and `rsh`, together with their
  arguments, for instance `ssh -p 2222 -c blowfish`; Note that the
  remote hostname should not appear there, see rsync(1) for details.
  When the transport used is `ssh`, connections are automatically cached
  unless `annex.sshcaching` is unset.

* `remote.<name>.annex-bup-split-options`

  Options to pass to bup split when storing content in this remote.
  For example, to limit the bandwidth to 100Kbyte/s, set it to `--bwlimit 100k`
  (There is no corresponding option for bup join.)

* `remote.<name>.annex-gnupg-options`

  Options to pass to GnuPG for symmetric encryption. For instance, to
  use the AES cipher with a 256 bits key and disable compression, set it
  to `--cipher-algo AES256 --compress-algo none`. (These options take
  precedence over the default GnuPG configuration, which is otherwise
  used.)

* `annex.ssh-options`, `annex.rsync-options`, `annex.bup-split-options`,
  `annex.gnupg-options`

  Default ssh, rsync, wget/curl, bup, and GnuPG options to use if a
  remote does not have specific options.

* `annex.web-options`

  Options to use when using wget or curl to download a file from the web.
  (wget is always used in preference to curl if available.)
  For example, to force ipv4 only, set it to "-4"

* `annex.quvi-options`

  Options to pass to quvi when using it to find the url to download for a
  video.

* `annex.http-headers`

  HTTP headers to send when downloading from the web. Multiple lines of
  this option can be set, one per header.

* `annex.http-headers-command`

  If set, the command is run and each line of its output is used as a HTTP
  header. This overrides annex.http-headers.

* `annex.web-download-command`

  Use to specify a command to run to download a file from the web.
  (The default is to use wget or curl.)

  In the command line, %url is replaced with the url to download,
  and %file is replaced with the file that it should be saved to.
  Note that both these values will automatically be quoted, since
  the command is run in a shell.

* `remote.<name>.rsyncurl`

  Used by rsync special remotes, this configures
  the location of the rsync repository to use. Normally this is automatically
  set up by `git annex initremote`, but you can change it if needed.

* `remote.<name>.buprepo`

  Used by bup special remotes, this configures
  the location of the bup repository to use. Normally this is automatically
  set up by `git annex initremote`, but you can change it if needed.

* `remote.<name>.directory`

  Used by directory special remotes, this configures
  the location of the directory where annexed files are stored for this
  remote. Normally this is automatically set up by `git annex initremote`,
  but you can change it if needed.

* `remote.<name>.s3`

  Used to identify Amazon S3 special remotes.
  Normally this is automatically set up by `git annex initremote`.

* `remote.<name>.glacier`

  Used to identify Amazon Glacier special remotes.
  Normally this is automatically set up by `git annex initremote`.

* `remote.<name>.webdav`

  Used to identify webdav special remotes.
  Normally this is automatically set up by `git annex initremote`.

* `remote.<name>.annex-xmppaddress`

  Used to identify the XMPP address of a Jabber buddy.
  Normally this is set up by the git-annex assistant when pairing over XMPP.

* `remote.<name>.gcrypt`

  Used to identify gcrypt special remotes.
  Normally this is automatically set up by `git annex initremote`.

  It is set to "true" if this is a gcrypt remote.
  If the gcrypt remote is accessible over ssh and has git-annex-shell
  available to manage it, it's set to "shell"

* `remote.<name>.hooktype`, `remote.<name>.externaltype`

  Used by hook special remotes and external special remotes to record
  the type of the remote.

# CONFIGURATION VIA .gitattributes

The key-value backend used when adding a new file to the annex can be
configured on a per-file-type basis via `.gitattributes` files. In the file,
the `annex.backend` attribute can be set to the name of the backend to
use. For example, this here's how to use the WORM backend by default,
but the SHA256E backend for ogg files:

	* annex.backend=WORM
	*.ogg annex.backend=SHA256E

The numcopies setting can also be configured on a per-file-type basis via
the `annex.numcopies` attribute in `.gitattributes` files. This overrides
any value set using `annex.numcopies` in `.git/config`.
For example, this makes two copies be needed for wav files:

	*.wav annex.numcopies=2

Note that setting numcopies to 0 is very unsafe.

# FILES

These files are used by git-annex:

`.git/annex/objects/` in your git repository contains the annexed file
contents that are currently available. Annexed files in your git
repository symlink to that content.

`.git/annex/` in your git repository contains other run-time information
used by git-annex.

`~/.config/git-annex/autostart` is a list of git repositories
to start the git-annex assistant in.

# SEE ALSO

Most of git-annex's documentation is available on its web site, 
<http://git-annex.branchable.com/>

If git-annex is installed from a package, a copy of its documentation
should be included, in, for example, `/usr/share/doc/git-annex/`

# AUTHOR

Joey Hess <joey@kitenet.net>

<http://git-annex.branchable.com/>

Warning: Automatically converted into a man page by mdwn2man. Edit with care