Documentation/gitrepository-layout.txt

   1 gitrepository-layout(5)
   2 =======================
   3
   4 NAME
   5 ----
   6 gitrepository-layout - Git Repository Layout
   7
   8 SYNOPSIS
   9 --------
  10 $GIT_DIR/*
  11
  12 DESCRIPTION
  13 -----------
  14
  15 A Git repository comes in two different flavours:
  16
  17  * a `.git` directory at the root of the working tree;
  18
  19  * a `<project>.git` directory that is a 'bare' repository
  20    (i.e. without its own working tree), that is typically used for
  21    exchanging histories with others by pushing into it and fetching
  22    from it.
  23
  24 *Note*: Also you can have a plain text file `.git` at the root of
  25 your working tree, containing `gitdir: <path>` to point at the real
  26 directory that has the repository.  This mechanism is often used for
  27 a working tree of a submodule checkout, to allow you in the
  28 containing superproject to `git checkout` a branch that does not
  29 have the submodule.  The `checkout` has to remove the entire
  30 submodule working tree, without losing the submodule repository.
  31
  32 These things may exist in a Git repository.
  33
  34 objects::
  35         Object store associated with this repository.  Usually
  36         an object store is self sufficient (i.e. all the objects
  37         that are referred to by an object found in it are also
  38         found in it), but there are a few ways to violate it.
  39 +
  40 . You could have an incomplete but locally usable repository
  41 by creating a shallow clone.  See linkgit:git-clone[1].
  42 . You could be using the `objects/info/alternates` or
  43 `$GIT_ALTERNATE_OBJECT_DIRECTORIES` mechanisms to 'borrow'
  44 objects from other object stores.  A repository with this kind
  45 of incomplete object store is not suitable to be published for
  46 use with dumb transports but otherwise is OK as long as
  47 `objects/info/alternates` points at the object stores it
  48 borrows from.
  49 +
  50 This directory is ignored if $GIT_COMMON_DIR is set and
  51 "$GIT_COMMON_DIR/objects" will be used instead.
  52
  53 objects/[0-9a-f][0-9a-f]::
  54         A newly created object is stored in its own file.
  55         The objects are splayed over 256 subdirectories using
  56         the first two characters of the sha1 object name to
  57         keep the number of directory entries in `objects`
  58         itself to a manageable number. Objects found
  59         here are often called 'unpacked' (or 'loose') objects.
  60
  61 objects/pack::
  62         Packs (files that store many object in compressed form,
  63         along with index files to allow them to be randomly
  64         accessed) are found in this directory.
  65
  66 objects/info::
  67         Additional information about the object store is
  68         recorded in this directory.
  69
  70 objects/info/packs::
  71         This file is to help dumb transports discover what packs
  72         are available in this object store.  Whenever a pack is
  73         added or removed, `git update-server-info` should be run
  74         to keep this file up to date if the repository is
  75         published for dumb transports.  'git repack' does this
  76         by default.
  77
  78 objects/info/alternates::
  79         This file records paths to alternate object stores that
  80         this object store borrows objects from, one pathname per
  81         line. Note that not only native Git tools use it locally,
  82         but the HTTP fetcher also tries to use it remotely; this
  83         will usually work if you have relative paths (relative
  84         to the object database, not to the repository!) in your
  85         alternates file, but it will not work if you use absolute
  86         paths unless the absolute path in filesystem and web URL
  87         is the same. See also 'objects/info/http-alternates'.
  88
  89 objects/info/http-alternates::
  90         This file records URLs to alternate object stores that
  91         this object store borrows objects from, to be used when
  92         the repository is fetched over HTTP.
  93
  94 refs::
  95         References are stored in subdirectories of this
  96         directory.  The 'git prune' command knows to preserve
  97         objects reachable from refs found in this directory and
  98         its subdirectories. This directory is ignored if $GIT_COMMON_DIR
  99         is set and "$GIT_COMMON_DIR/refs" will be used instead.
 100
 101 refs/heads/`name`::
 102         records tip-of-the-tree commit objects of branch `name`
 103
 104 refs/tags/`name`::
 105         records any object name (not necessarily a commit
 106         object, or a tag object that points at a commit object).
 107
 108 refs/remotes/`name`::
 109         records tip-of-the-tree commit objects of branches copied
 110         from a remote repository.
 111
 112 refs/replace/`<obj-sha1>`::
 113         records the SHA-1 of the object that replaces `<obj-sha1>`.
 114         This is similar to info/grafts and is internally used and
 115         maintained by linkgit:git-replace[1]. Such refs can be exchanged
 116         between repositories while grafts are not.
 117
 118 packed-refs::
 119         records the same information as refs/heads/, refs/tags/,
 120         and friends record in a more efficient way.  See
 121         linkgit:git-pack-refs[1]. This file is ignored if $GIT_COMMON_DIR
 122         is set and "$GIT_COMMON_DIR/packed-refs" will be used instead.
 123
 124 HEAD::
 125         A symref (see glossary) to the `refs/heads/` namespace
 126         describing the currently active branch.  It does not mean
 127         much if the repository is not associated with any working tree
 128         (i.e. a 'bare' repository), but a valid Git repository
 129         *must* have the HEAD file; some porcelains may use it to
 130         guess the designated "default" branch of the repository
 131         (usually 'master').  It is legal if the named branch
 132         'name' does not (yet) exist.  In some legacy setups, it is
 133         a symbolic link instead of a symref that points at the current
 134         branch.
 135 +
 136 HEAD can also record a specific commit directly, instead of
 137 being a symref to point at the current branch.  Such a state
 138 is often called 'detached HEAD.'  See linkgit:git-checkout[1]
 139 for details.
 140
 141 config::
 142         Repository specific configuration file. This file is ignored
 143         if $GIT_COMMON_DIR is set and "$GIT_COMMON_DIR/config" will be
 144         used instead.
 145
 146 branches::
 147         A slightly deprecated way to store shorthands to be used
 148         to specify a URL to 'git fetch', 'git pull' and 'git push'.
 149         A file can be stored as `branches/<name>` and then
 150         'name' can be given to these commands in place of
 151         'repository' argument.  See the REMOTES section in
 152         linkgit:git-fetch[1] for details.  This mechanism is legacy
 153         and not likely to be found in modern repositories. This
 154         directory is ignored if $GIT_COMMON_DIR is set and
 155         "$GIT_COMMON_DIR/branches" will be used instead.
 156
 157
 158 hooks::
 159         Hooks are customization scripts used by various Git
 160         commands.  A handful of sample hooks are installed when
 161         'git init' is run, but all of them are disabled by
 162         default.  To enable, the `.sample` suffix has to be
 163         removed from the filename by renaming.
 164         Read linkgit:githooks[5] for more details about
 165         each hook. This directory is ignored if $GIT_COMMON_DIR is set
 166         and "$GIT_COMMON_DIR/hooks" will be used instead.
 167
 168
 169 index::
 170         The current index file for the repository.  It is
 171         usually not found in a bare repository.
 172
 173 sharedindex.<SHA-1>::
 174         The shared index part, to be referenced by $GIT_DIR/index and
 175         other temporary index files. Only valid in split index mode.
 176
 177 info::
 178         Additional information about the repository is recorded
 179         in this directory. This directory is ignored if $GIT_COMMON_DIR
 180         is set and "$GIT_COMMON_DIR/info" will be used instead.
 181
 182 info/refs::
 183         This file helps dumb transports discover what refs are
 184         available in this repository.  If the repository is
 185         published for dumb transports, this file should be
 186         regenerated by 'git update-server-info' every time a tag
 187         or branch is created or modified.  This is normally done
 188         from the `hooks/update` hook, which is run by the
 189         'git-receive-pack' command when you 'git push' into the
 190         repository.
 191
 192 info/grafts::
 193         This file records fake commit ancestry information, to
 194         pretend the set of parents a commit has is different
 195         from how the commit was actually created.  One record
 196         per line describes a commit and its fake parents by
 197         listing their 40-byte hexadecimal object names separated
 198         by a space and terminated by a newline.
 199 +
 200 Note that the grafts mechanism is outdated and can lead to problems
 201 transferring objects between repositories; see linkgit:git-replace[1]
 202 for a more flexible and robust system to do the same thing.
 203
 204 info/exclude::
 205         This file, by convention among Porcelains, stores the
 206         exclude pattern list. `.gitignore` is the per-directory
 207         ignore file.  'git status', 'git add', 'git rm' and
 208         'git clean' look at it but the core Git commands do not look
 209         at it.  See also: linkgit:gitignore[5].
 210
 211 info/attributes::
 212         Defines which attributes to assign to a path, similar to per-directory
 213         `.gitattributes` files.   See also: linkgit:gitattributes[5].
 214
 215 info/sparse-checkout::
 216         This file stores sparse checkout patterns.
 217         See also: linkgit:git-read-tree[1].
 218
 219 remotes::
 220         Stores shorthands for URL and default refnames for use
 221         when interacting with remote repositories via 'git fetch',
 222         'git pull' and 'git push' commands.  See the REMOTES section
 223         in linkgit:git-fetch[1] for details.  This mechanism is legacy
 224         and not likely to be found in modern repositories. This
 225         directory is ignored if $GIT_COMMON_DIR is set and
 226         "$GIT_COMMON_DIR/remotes" will be used instead.
 227
 228 logs::
 229         Records of changes made to refs are stored in this directory.
 230         See linkgit:git-update-ref[1] for more information. This
 231         directory is ignored if $GIT_COMMON_DIR is set and
 232         "$GIT_COMMON_DIR/logs" will be used instead.
 233
 234 logs/refs/heads/`name`::
 235         Records all changes made to the branch tip named `name`.
 236
 237 logs/refs/tags/`name`::
 238         Records all changes made to the tag named `name`.
 239
 240 shallow::
 241         This is similar to `info/grafts` but is internally used
 242         and maintained by shallow clone mechanism.  See `--depth`
 243         option to linkgit:git-clone[1] and linkgit:git-fetch[1]. This
 244         file is ignored if $GIT_COMMON_DIR is set and
 245         "$GIT_COMMON_DIR/shallow" will be used instead.
 246
 247 commondir::
 248         If this file exists, $GIT_COMMON_DIR (see linkgit:git[1]) will
 249         be set to the path specified in this file if it is not
 250         explicitly set. If the specified path is relative, it is
 251         relative to $GIT_DIR. The repository with commondir is
 252         incomplete without the repository pointed by "commondir".
 253
 254 modules::
 255         Contains the git-repositories of the submodules.
 256
 257 worktrees::
 258         Contains administrative data for linked
 259         working trees. Each subdirectory contains the working tree-related
 260         part of a linked working tree. This directory is ignored if
 261         $GIT_COMMON_DIR is set, in which case
 262         "$GIT_COMMON_DIR/worktrees" will be used instead.
 263
 264 worktrees/<id>/gitdir::
 265         A text file containing the absolute path back to the .git file
 266         that points to here. This is used to check if the linked
 267         repository has been manually removed and there is no need to
 268         keep this directory any more. The mtime of this file should be
 269         updated every time the linked repository is accessed.
 270
 271 worktrees/<id>/locked::
 272         If this file exists, the linked working tree may be on a
 273         portable device and not available. The presence of this file
 274         prevents `worktrees/<id>` from being pruned either automatically
 275         or manually by `git worktree prune`. The file may contain a string
 276         explaining why the repository is locked.
 277
 278 worktrees/<id>/link::
 279         If this file exists, it is a hard link to the linked .git
 280         file. It is used to detect if the linked repository is
 281         manually removed.
 282
 283 SEE ALSO
 284 --------
 285 linkgit:git-init[1],
 286 linkgit:git-clone[1],
 287 linkgit:git-fetch[1],
 288 linkgit:git-pack-refs[1],
 289 linkgit:git-gc[1],
 290 linkgit:git-checkout[1],
 291 linkgit:gitglossary[7],
 292 link:user-manual.html[The Git User's Manual]
 293
 294 GIT
 295 ---
 296 Part of the linkgit:git[1] suite