Documentation / git-cat-file.txton commit Merge branch 'js/win32-mmap' (f9dd741)
   1git-cat-file(1)
   2===============
   3
   4NAME
   5----
   6git-cat-file - Provide content or type and size information for repository objects
   7
   8
   9SYNOPSIS
  10--------
  11[verse]
  12'git cat-file' (-t [--allow-unknown-type]| -s [--allow-unknown-type]| -e | -p | <type> | --textconv ) <object>
  13'git cat-file' (--batch | --batch-check) [--follow-symlinks]
  14
  15DESCRIPTION
  16-----------
  17In its first form, the command provides the content or the type of an object in
  18the repository. The type is required unless '-t' or '-p' is used to find the
  19object type, or '-s' is used to find the object size, or '--textconv' is used
  20(which implies type "blob").
  21
  22In the second form, a list of objects (separated by linefeeds) is provided on
  23stdin, and the SHA-1, type, and size of each object is printed on stdout.
  24
  25OPTIONS
  26-------
  27<object>::
  28        The name of the object to show.
  29        For a more complete list of ways to spell object names, see
  30        the "SPECIFYING REVISIONS" section in linkgit:gitrevisions[7].
  31
  32-t::
  33        Instead of the content, show the object type identified by
  34        <object>.
  35
  36-s::
  37        Instead of the content, show the object size identified by
  38        <object>.
  39
  40-e::
  41        Suppress all output; instead exit with zero status if <object>
  42        exists and is a valid object.
  43
  44-p::
  45        Pretty-print the contents of <object> based on its type.
  46
  47<type>::
  48        Typically this matches the real type of <object> but asking
  49        for a type that can trivially be dereferenced from the given
  50        <object> is also permitted.  An example is to ask for a
  51        "tree" with <object> being a commit object that contains it,
  52        or to ask for a "blob" with <object> being a tag object that
  53        points at it.
  54
  55--textconv::
  56        Show the content as transformed by a textconv filter. In this case,
  57        <object> has be of the form <tree-ish>:<path>, or :<path> in order
  58        to apply the filter to the content recorded in the index at <path>.
  59
  60--batch::
  61--batch=<format>::
  62        Print object information and contents for each object provided
  63        on stdin.  May not be combined with any other options or arguments.
  64        See the section `BATCH OUTPUT` below for details.
  65
  66--batch-check::
  67--batch-check=<format>::
  68        Print object information for each object provided on stdin.  May
  69        not be combined with any other options or arguments.  See the
  70        section `BATCH OUTPUT` below for details.
  71
  72--batch-all-objects::
  73        Instead of reading a list of objects on stdin, perform the
  74        requested batch operation on all objects in the repository and
  75        any alternate object stores (not just reachable objects).
  76        Requires `--batch` or `--batch-check` be specified. Note that
  77        the objects are visited in order sorted by their hashes.
  78
  79--buffer::
  80        Normally batch output is flushed after each object is output, so
  81        that a process can interactively read and write from
  82        `cat-file`. With this option, the output uses normal stdio
  83        buffering; this is much more efficient when invoking
  84        `--batch-check` on a large number of objects.
  85
  86--allow-unknown-type::
  87        Allow -s or -t to query broken/corrupt objects of unknown type.
  88
  89--follow-symlinks::
  90        With --batch or --batch-check, follow symlinks inside the
  91        repository when requesting objects with extended SHA-1
  92        expressions of the form tree-ish:path-in-tree.  Instead of
  93        providing output about the link itself, provide output about
  94        the linked-to object.  If a symlink points outside the
  95        tree-ish (e.g. a link to /foo or a root-level link to ../foo),
  96        the portion of the link which is outside the tree will be
  97        printed.
  98+
  99This option does not (currently) work correctly when an object in the
 100index is specified (e.g. `:link` instead of `HEAD:link`) rather than
 101one in the tree.
 102+
 103This option cannot (currently) be used unless `--batch` or
 104`--batch-check` is used.
 105+
 106For example, consider a git repository containing:
 107+
 108--
 109        f: a file containing "hello\n"
 110        link: a symlink to f
 111        dir/link: a symlink to ../f
 112        plink: a symlink to ../f
 113        alink: a symlink to /etc/passwd
 114--
 115+
 116For a regular file `f`, `echo HEAD:f | git cat-file --batch` would print
 117+
 118--
 119        ce013625030ba8dba906f756967f9e9ca394464a blob 6
 120--
 121+
 122And `echo HEAD:link | git cat-file --batch --follow-symlinks` would
 123print the same thing, as would `HEAD:dir/link`, as they both point at
 124`HEAD:f`.
 125+
 126Without `--follow-symlinks`, these would print data about the symlink
 127itself.  In the case of `HEAD:link`, you would see
 128+
 129--
 130        4d1ae35ba2c8ec712fa2a379db44ad639ca277bd blob 1
 131--
 132+
 133Both `plink` and `alink` point outside the tree, so they would
 134respectively print:
 135+
 136--
 137        symlink 4
 138        ../f
 139
 140        symlink 11
 141        /etc/passwd
 142--
 143
 144
 145OUTPUT
 146------
 147If '-t' is specified, one of the <type>.
 148
 149If '-s' is specified, the size of the <object> in bytes.
 150
 151If '-e' is specified, no output.
 152
 153If '-p' is specified, the contents of <object> are pretty-printed.
 154
 155If <type> is specified, the raw (though uncompressed) contents of the <object>
 156will be returned.
 157
 158BATCH OUTPUT
 159------------
 160
 161If `--batch` or `--batch-check` is given, `cat-file` will read objects
 162from stdin, one per line, and print information about them. By default,
 163the whole line is considered as an object, as if it were fed to
 164linkgit:git-rev-parse[1].
 165
 166You can specify the information shown for each object by using a custom
 167`<format>`. The `<format>` is copied literally to stdout for each
 168object, with placeholders of the form `%(atom)` expanded, followed by a
 169newline. The available atoms are:
 170
 171`objectname`::
 172        The 40-hex object name of the object.
 173
 174`objecttype`::
 175        The type of of the object (the same as `cat-file -t` reports).
 176
 177`objectsize`::
 178        The size, in bytes, of the object (the same as `cat-file -s`
 179        reports).
 180
 181`objectsize:disk`::
 182        The size, in bytes, that the object takes up on disk. See the
 183        note about on-disk sizes in the `CAVEATS` section below.
 184
 185`deltabase`::
 186        If the object is stored as a delta on-disk, this expands to the
 187        40-hex sha1 of the delta base object. Otherwise, expands to the
 188        null sha1 (40 zeroes). See `CAVEATS` below.
 189
 190`rest`::
 191        If this atom is used in the output string, input lines are split
 192        at the first whitespace boundary. All characters before that
 193        whitespace are considered to be the object name; characters
 194        after that first run of whitespace (i.e., the "rest" of the
 195        line) are output in place of the `%(rest)` atom.
 196
 197If no format is specified, the default format is `%(objectname)
 198%(objecttype) %(objectsize)`.
 199
 200If `--batch` is specified, the object information is followed by the
 201object contents (consisting of `%(objectsize)` bytes), followed by a
 202newline.
 203
 204For example, `--batch` without a custom format would produce:
 205
 206------------
 207<sha1> SP <type> SP <size> LF
 208<contents> LF
 209------------
 210
 211Whereas `--batch-check='%(objectname) %(objecttype)'` would produce:
 212
 213------------
 214<sha1> SP <type> LF
 215------------
 216
 217If a name is specified on stdin that cannot be resolved to an object in
 218the repository, then `cat-file` will ignore any custom format and print:
 219
 220------------
 221<object> SP missing LF
 222------------
 223
 224If --follow-symlinks is used, and a symlink in the repository points
 225outside the repository, then `cat-file` will ignore any custom format
 226and print:
 227
 228------------
 229symlink SP <size> LF
 230<symlink> LF
 231------------
 232
 233The symlink will either be absolute (beginning with a /), or relative
 234to the tree root.  For instance, if dir/link points to ../../foo, then
 235<symlink> will be ../foo.  <size> is the size of the symlink in bytes.
 236
 237If --follow-symlinks is used, the following error messages will be
 238displayed:
 239
 240------------
 241<object> SP missing LF
 242------------
 243is printed when the initial symlink requested does not exist.
 244
 245------------
 246dangling SP <size> LF
 247<object> LF
 248------------
 249is printed when the initial symlink exists, but something that
 250it (transitive-of) points to does not.
 251
 252------------
 253loop SP <size> LF
 254<object> LF
 255------------
 256is printed for symlink loops (or any symlinks that
 257require more than 40 link resolutions to resolve).
 258
 259------------
 260notdir SP <size> LF
 261<object> LF
 262------------
 263is printed when, during symlink resolution, a file is used as a
 264directory name.
 265
 266CAVEATS
 267-------
 268
 269Note that the sizes of objects on disk are reported accurately, but care
 270should be taken in drawing conclusions about which refs or objects are
 271responsible for disk usage. The size of a packed non-delta object may be
 272much larger than the size of objects which delta against it, but the
 273choice of which object is the base and which is the delta is arbitrary
 274and is subject to change during a repack.
 275
 276Note also that multiple copies of an object may be present in the object
 277database; in this case, it is undefined which copy's size or delta base
 278will be reported.
 279
 280GIT
 281---
 282Part of the linkgit:git[1] suite