bd8a6e5c5ed92532dd3c4d1ef0e228e763349dde
   1git-diff-cache(1)
   2=================
   3v0.1, May 2005
   4
   5NAME
   6----
   7git-diff-cache - Compares content and mode of blobs between the cache and repository
   8
   9
  10SYNOPSIS
  11--------
  12'git-diff-cache' [-p] [-r] [-z] [-m] [-M] [-R] [--cached] <tree-ish>
  13
  14DESCRIPTION
  15-----------
  16Compares the content and mode of the blobs found via a tree object
  17with the content of the current cache and, optionally ignoring the
  18stat state of the file on disk.
  19
  20OPTIONS
  21-------
  22<tree-ish>::
  23        The id of a tree object to diff against.
  24
  25-p::
  26        Generate patch (see section on generating patches)
  27
  28-r::
  29        This flag does not mean anything.  It is there only to match
  30        "git-diff-tree".  Unlike "git-diff-tree", "git-diff-cache"
  31        always looks at all the subdirectories.
  32
  33-z::
  34        \0 line termination on output
  35
  36-M::
  37        Detect renames; implies -p.
  38
  39-R::
  40        Output diff in reverse.
  41
  42--cached::
  43        do not consider the on-disk file at all
  44
  45-m::
  46        By default, files recorded in the index but not checked
  47        out are reported as deleted.  This flag makes
  48        "git-diff-cache" say that all non-checked-out files are up
  49        to date.
  50
  51Output format
  52-------------
  53include::diff-format.txt[]
  54
  55Operating Modes
  56---------------
  57You can choose whether you want to trust the index file entirely
  58(using the '--cached' flag) or ask the diff logic to show any files
  59that don't match the stat state as being "tentatively changed".  Both
  60of these operations are very useful indeed.
  61
  62Cached Mode
  63-----------
  64If '--cached' is specified, it allows you to ask:
  65
  66        show me the differences between HEAD and the current index
  67        contents (the ones I'd write with a "git-write-tree")
  68
  69For example, let's say that you have worked on your index file, and are
  70ready to commit. You want to see eactly *what* you are going to commit is
  71without having to write a new tree object and compare it that way, and to
  72do that, you just do
  73
  74        git-diff-cache --cached $(cat .git/HEAD)
  75
  76Example: let's say I had renamed `commit.c` to `git-commit.c`, and I had
  77done an "git-update-cache" to make that effective in the index file.
  78"git-diff-files" wouldn't show anything at all, since the index file
  79matches my working directory. But doing a "git-diff-cache" does:
  80
  81  torvalds@ppc970:~/git> git-diff-cache --cached $(cat .git/HEAD)
  82  -100644 blob    4161aecc6700a2eb579e842af0b7f22b98443f74        commit.c
  83  +100644 blob    4161aecc6700a2eb579e842af0b7f22b98443f74        git-commit.c
  84
  85You can trivially see that the above is a rename.
  86
  87In fact, "git-diff-cache --cached" *should* always be entirely equivalent to
  88actually doing a "git-write-tree" and comparing that. Except this one is much
  89nicer for the case where you just want to check where you are.
  90
  91So doing a "git-diff-cache --cached" is basically very useful when you are 
  92asking yourself "what have I already marked for being committed, and 
  93what's the difference to a previous tree".
  94
  95Non-cached Mode
  96---------------
  97The "non-cached" mode takes a different approach, and is potentially
  98the more useful of the two in that what it does can't be emulated with
  99a "git-write-tree" + "git-diff-tree". Thus that's the default mode.
 100The non-cached version asks the question:
 101
 102   show me the differences between HEAD and the currently checked out
 103   tree - index contents _and_ files that aren't up-to-date
 104
 105which is obviously a very useful question too, since that tells you what
 106you *could* commit. Again, the output matches the "git-diff-tree -r"
 107output to a tee, but with a twist.
 108
 109The twist is that if some file doesn't match the cache, we don't have
 110a backing store thing for it, and we use the magic "all-zero" sha1 to
 111show that. So let's say that you have edited `kernel/sched.c`, but
 112have not actually done a "git-update-cache" on it yet - there is no
 113"object" associated with the new state, and you get:
 114
 115  torvalds@ppc970:~/v2.6/linux> git-diff-cache $(cat .git/HEAD )
 116  *100644->100664 blob    7476bb......->000000......      kernel/sched.c
 117
 118ie it shows that the tree has changed, and that `kernel/sched.c` has is
 119not up-to-date and may contain new stuff. The all-zero sha1 means that to
 120get the real diff, you need to look at the object in the working directory
 121directly rather than do an object-to-object diff.
 122
 123NOTE! As with other commands of this type, "git-diff-cache" does not
 124actually look at the contents of the file at all. So maybe
 125`kernel/sched.c` hasn't actually changed, and it's just that you
 126touched it. In either case, it's a note that you need to
 127"git-upate-cache" it to make the cache be in sync.
 128
 129NOTE 2! You can have a mixture of files show up as "has been updated"
 130and "is still dirty in the working directory" together. You can always
 131tell which file is in which state, since the "has been updated" ones
 132show a valid sha1, and the "not in sync with the index" ones will
 133always have the special all-zero sha1.
 134
 135
 136Author
 137------
 138Written by Linus Torvalds <torvalds@osdl.org>
 139
 140Documentation
 141--------------
 142Documentation by David Greaves, Junio C Hamano and the git-list <git@vger.kernel.org>.
 143
 144GIT
 145---
 146Part of the link:git.html[git] suite
 147