Documentation / git-read-tree.txton commit Merge branch 'sb/notes-parse-opt' (3b37d9c)
   1git-read-tree(1)
   2================
   3
   4NAME
   5----
   6git-read-tree - Reads tree information into the index
   7
   8
   9SYNOPSIS
  10--------
  11'git read-tree' [[-m [--trivial] [--aggressive] | --reset | --prefix=<prefix>]
  12                [-u [--exclude-per-directory=<gitignore>] | -i]]
  13                [--index-output=<file>] [--no-sparse-checkout]
  14                <tree-ish1> [<tree-ish2> [<tree-ish3>]]
  15
  16
  17DESCRIPTION
  18-----------
  19Reads the tree information given by <tree-ish> into the index,
  20but does not actually *update* any of the files it "caches". (see:
  21linkgit:git-checkout-index[1])
  22
  23Optionally, it can merge a tree into the index, perform a
  24fast-forward (i.e. 2-way) merge, or a 3-way merge, with the `-m`
  25flag.  When used with `-m`, the `-u` flag causes it to also update
  26the files in the work tree with the result of the merge.
  27
  28Trivial merges are done by 'git read-tree' itself.  Only conflicting paths
  29will be in unmerged state when 'git read-tree' returns.
  30
  31OPTIONS
  32-------
  33-m::
  34        Perform a merge, not just a read.  The command will
  35        refuse to run if your index file has unmerged entries,
  36        indicating that you have not finished previous merge you
  37        started.
  38
  39--reset::
  40        Same as -m, except that unmerged entries are discarded
  41        instead of failing.
  42
  43-u::
  44        After a successful merge, update the files in the work
  45        tree with the result of the merge.
  46
  47-i::
  48        Usually a merge requires the index file as well as the
  49        files in the working tree are up to date with the
  50        current head commit, in order not to lose local
  51        changes.  This flag disables the check with the working
  52        tree and is meant to be used when creating a merge of
  53        trees that are not directly related to the current
  54        working tree status into a temporary index file.
  55
  56-v::
  57        Show the progress of checking files out.
  58
  59--trivial::
  60        Restrict three-way merge by 'git read-tree' to happen
  61        only if there is no file-level merging required, instead
  62        of resolving merge for trivial cases and leaving
  63        conflicting files unresolved in the index.
  64
  65--aggressive::
  66        Usually a three-way merge by 'git read-tree' resolves
  67        the merge for really trivial cases and leaves other
  68        cases unresolved in the index, so that Porcelains can
  69        implement different merge policies.  This flag makes the
  70        command to resolve a few more cases internally:
  71+
  72* when one side removes a path and the other side leaves the path
  73  unmodified.  The resolution is to remove that path.
  74* when both sides remove a path.  The resolution is to remove that path.
  75* when both sides adds a path identically.  The resolution
  76  is to add that path.
  77
  78--prefix=<prefix>/::
  79        Keep the current index contents, and read the contents
  80        of named tree-ish under directory at `<prefix>`.  The
  81        original index file cannot have anything at the path
  82        `<prefix>` itself, and have nothing in `<prefix>/`
  83        directory.  Note that the `<prefix>/` value must end
  84        with a slash.
  85
  86--exclude-per-directory=<gitignore>::
  87        When running the command with `-u` and `-m` options, the
  88        merge result may need to overwrite paths that are not
  89        tracked in the current branch.  The command usually
  90        refuses to proceed with the merge to avoid losing such a
  91        path.  However this safety valve sometimes gets in the
  92        way.  For example, it often happens that the other
  93        branch added a file that used to be a generated file in
  94        your branch, and the safety valve triggers when you try
  95        to switch to that branch after you ran `make` but before
  96        running `make clean` to remove the generated file.  This
  97        option tells the command to read per-directory exclude
  98        file (usually '.gitignore') and allows such an untracked
  99        but explicitly ignored file to be overwritten.
 100
 101--index-output=<file>::
 102        Instead of writing the results out to `$GIT_INDEX_FILE`,
 103        write the resulting index in the named file.  While the
 104        command is operating, the original index file is locked
 105        with the same mechanism as usual.  The file must allow
 106        to be rename(2)ed into from a temporary file that is
 107        created next to the usual index file; typically this
 108        means it needs to be on the same filesystem as the index
 109        file itself, and you need write permission to the
 110        directories the index file and index output file are
 111        located in.
 112
 113--no-sparse-checkout::
 114        Disable sparse checkout support even if `core.sparseCheckout`
 115        is true.
 116
 117<tree-ish#>::
 118        The id of the tree object(s) to be read/merged.
 119
 120
 121Merging
 122-------
 123If `-m` is specified, 'git read-tree' can perform 3 kinds of
 124merge, a single tree merge if only 1 tree is given, a
 125fast-forward merge with 2 trees, or a 3-way merge if 3 trees are
 126provided.
 127
 128
 129Single Tree Merge
 130~~~~~~~~~~~~~~~~~
 131If only 1 tree is specified, 'git read-tree' operates as if the user did not
 132specify `-m`, except that if the original index has an entry for a
 133given pathname, and the contents of the path match with the tree
 134being read, the stat info from the index is used. (In other words, the
 135index's stat()s take precedence over the merged tree's).
 136
 137That means that if you do a `git read-tree -m <newtree>` followed by a
 138`git checkout-index -f -u -a`, the 'git checkout-index' only checks out
 139the stuff that really changed.
 140
 141This is used to avoid unnecessary false hits when 'git diff-files' is
 142run after 'git read-tree'.
 143
 144
 145Two Tree Merge
 146~~~~~~~~~~~~~~
 147
 148Typically, this is invoked as `git read-tree -m $H $M`, where $H
 149is the head commit of the current repository, and $M is the head
 150of a foreign tree, which is simply ahead of $H (i.e. we are in a
 151fast-forward situation).
 152
 153When two trees are specified, the user is telling 'git read-tree'
 154the following:
 155
 156     1. The current index and work tree is derived from $H, but
 157        the user may have local changes in them since $H.
 158
 159     2. The user wants to fast-forward to $M.
 160
 161In this case, the `git read-tree -m $H $M` command makes sure
 162that no local change is lost as the result of this "merge".
 163Here are the "carry forward" rules, where "I" denotes the index,
 164"clean" means that index and work tree coincide, and "exists"/"nothing"
 165refer to the presence of a path in the specified commit:
 166
 167        I                   H        M        Result
 168       -------------------------------------------------------
 169     0  nothing             nothing  nothing  (does not happen)
 170     1  nothing             nothing  exists   use M
 171     2  nothing             exists   nothing  remove path from index
 172     3  nothing             exists   exists,  use M if "initial checkout",
 173                                     H == M   keep index otherwise
 174                                     exists,  fail
 175                                     H != M
 176
 177        clean I==H  I==M
 178       ------------------
 179     4  yes   N/A   N/A     nothing  nothing  keep index
 180     5  no    N/A   N/A     nothing  nothing  keep index
 181
 182     6  yes   N/A   yes     nothing  exists   keep index
 183     7  no    N/A   yes     nothing  exists   keep index
 184     8  yes   N/A   no      nothing  exists   fail
 185     9  no    N/A   no      nothing  exists   fail
 186
 187     10 yes   yes   N/A     exists   nothing  remove path from index
 188     11 no    yes   N/A     exists   nothing  fail
 189     12 yes   no    N/A     exists   nothing  fail
 190     13 no    no    N/A     exists   nothing  fail
 191
 192        clean (H==M)
 193       ------
 194     14 yes                 exists   exists   keep index
 195     15 no                  exists   exists   keep index
 196
 197        clean I==H  I==M (H!=M)
 198       ------------------
 199     16 yes   no    no      exists   exists   fail
 200     17 no    no    no      exists   exists   fail
 201     18 yes   no    yes     exists   exists   keep index
 202     19 no    no    yes     exists   exists   keep index
 203     20 yes   yes   no      exists   exists   use M
 204     21 no    yes   no      exists   exists   fail
 205
 206In all "keep index" cases, the index entry stays as in the
 207original index file.  If the entry is not up to date,
 208'git read-tree' keeps the copy in the work tree intact when
 209operating under the -u flag.
 210
 211When this form of 'git read-tree' returns successfully, you can
 212see which of the "local changes" that you made were carried forward by running
 213`git diff-index --cached $M`.  Note that this does not
 214necessarily match what `git diff-index --cached $H` would have
 215produced before such a two tree merge.  This is because of cases
 21618 and 19 --- if you already had the changes in $M (e.g. maybe
 217you picked it up via e-mail in a patch form), `git diff-index
 218--cached $H` would have told you about the change before this
 219merge, but it would not show in `git diff-index --cached $M`
 220output after the two-tree merge.
 221
 222Case 3 is slightly tricky and needs explanation.  The result from this
 223rule logically should be to remove the path if the user staged the removal
 224of the path and then switching to a new branch.  That however will prevent
 225the initial checkout from happening, so the rule is modified to use M (new
 226tree) only when the content of the index is empty.  Otherwise the removal
 227of the path is kept as long as $H and $M are the same.
 228
 2293-Way Merge
 230~~~~~~~~~~~
 231Each "index" entry has two bits worth of "stage" state. stage 0 is the
 232normal one, and is the only one you'd see in any kind of normal use.
 233
 234However, when you do 'git read-tree' with three trees, the "stage"
 235starts out at 1.
 236
 237This means that you can do
 238
 239----------------
 240$ git read-tree -m <tree1> <tree2> <tree3>
 241----------------
 242
 243and you will end up with an index with all of the <tree1> entries in
 244"stage1", all of the <tree2> entries in "stage2" and all of the
 245<tree3> entries in "stage3".  When performing a merge of another
 246branch into the current branch, we use the common ancestor tree
 247as <tree1>, the current branch head as <tree2>, and the other
 248branch head as <tree3>.
 249
 250Furthermore, 'git read-tree' has special-case logic that says: if you see
 251a file that matches in all respects in the following states, it
 252"collapses" back to "stage0":
 253
 254   - stage 2 and 3 are the same; take one or the other (it makes no
 255     difference - the same work has been done on our branch in
 256     stage 2 and their branch in stage 3)
 257
 258   - stage 1 and stage 2 are the same and stage 3 is different; take
 259     stage 3 (our branch in stage 2 did not do anything since the
 260     ancestor in stage 1 while their branch in stage 3 worked on
 261     it)
 262
 263   - stage 1 and stage 3 are the same and stage 2 is different take
 264     stage 2 (we did something while they did nothing)
 265
 266The 'git write-tree' command refuses to write a nonsensical tree, and it
 267will complain about unmerged entries if it sees a single entry that is not
 268stage 0.
 269
 270OK, this all sounds like a collection of totally nonsensical rules,
 271but it's actually exactly what you want in order to do a fast
 272merge. The different stages represent the "result tree" (stage 0, aka
 273"merged"), the original tree (stage 1, aka "orig"), and the two trees
 274you are trying to merge (stage 2 and 3 respectively).
 275
 276The order of stages 1, 2 and 3 (hence the order of three
 277<tree-ish> command line arguments) are significant when you
 278start a 3-way merge with an index file that is already
 279populated.  Here is an outline of how the algorithm works:
 280
 281- if a file exists in identical format in all three trees, it will
 282  automatically collapse to "merged" state by 'git read-tree'.
 283
 284- a file that has _any_ difference what-so-ever in the three trees
 285  will stay as separate entries in the index. It's up to "porcelain
 286  policy" to determine how to remove the non-0 stages, and insert a
 287  merged version.
 288
 289- the index file saves and restores with all this information, so you
 290  can merge things incrementally, but as long as it has entries in
 291  stages 1/2/3 (i.e., "unmerged entries") you can't write the result. So
 292  now the merge algorithm ends up being really simple:
 293
 294  * you walk the index in order, and ignore all entries of stage 0,
 295    since they've already been done.
 296
 297  * if you find a "stage1", but no matching "stage2" or "stage3", you
 298    know it's been removed from both trees (it only existed in the
 299    original tree), and you remove that entry.
 300
 301  * if you find a matching "stage2" and "stage3" tree, you remove one
 302    of them, and turn the other into a "stage0" entry. Remove any
 303    matching "stage1" entry if it exists too.  .. all the normal
 304    trivial rules ..
 305
 306You would normally use 'git merge-index' with supplied
 307'git merge-one-file' to do this last step.  The script updates
 308the files in the working tree as it merges each path and at the
 309end of a successful merge.
 310
 311When you start a 3-way merge with an index file that is already
 312populated, it is assumed that it represents the state of the
 313files in your work tree, and you can even have files with
 314changes unrecorded in the index file.  It is further assumed
 315that this state is "derived" from the stage 2 tree.  The 3-way
 316merge refuses to run if it finds an entry in the original index
 317file that does not match stage 2.
 318
 319This is done to prevent you from losing your work-in-progress
 320changes, and mixing your random changes in an unrelated merge
 321commit.  To illustrate, suppose you start from what has been
 322committed last to your repository:
 323
 324----------------
 325$ JC=`git rev-parse --verify "HEAD^0"`
 326$ git checkout-index -f -u -a $JC
 327----------------
 328
 329You do random edits, without running 'git update-index'.  And then
 330you notice that the tip of your "upstream" tree has advanced
 331since you pulled from him:
 332
 333----------------
 334$ git fetch git://.... linus
 335$ LT=`cat .git/FETCH_HEAD`
 336----------------
 337
 338Your work tree is still based on your HEAD ($JC), but you have
 339some edits since.  Three-way merge makes sure that you have not
 340added or modified index entries since $JC, and if you haven't,
 341then does the right thing.  So with the following sequence:
 342
 343----------------
 344$ git read-tree -m -u `git merge-base $JC $LT` $JC $LT
 345$ git merge-index git-merge-one-file -a
 346$ echo "Merge with Linus" | \
 347  git commit-tree `git write-tree` -p $JC -p $LT
 348----------------
 349
 350what you would commit is a pure merge between $JC and $LT without
 351your work-in-progress changes, and your work tree would be
 352updated to the result of the merge.
 353
 354However, if you have local changes in the working tree that
 355would be overwritten by this merge, 'git read-tree' will refuse
 356to run to prevent your changes from being lost.
 357
 358In other words, there is no need to worry about what exists only
 359in the working tree.  When you have local changes in a part of
 360the project that is not involved in the merge, your changes do
 361not interfere with the merge, and are kept intact.  When they
 362*do* interfere, the merge does not even start ('git read-tree'
 363complains loudly and fails without modifying anything).  In such
 364a case, you can simply continue doing what you were in the
 365middle of doing, and when your working tree is ready (i.e. you
 366have finished your work-in-progress), attempt the merge again.
 367
 368
 369Sparse checkout
 370---------------
 371
 372"Sparse checkout" allows to sparsely populate working directory.
 373It uses skip-worktree bit (see linkgit:git-update-index[1]) to tell
 374Git whether a file on working directory is worth looking at.
 375
 376"git read-tree" and other merge-based commands ("git merge", "git
 377checkout"...) can help maintaining skip-worktree bitmap and working
 378directory update. `$GIT_DIR/info/sparse-checkout` is used to
 379define the skip-worktree reference bitmap. When "git read-tree" needs
 380to update working directory, it will reset skip-worktree bit in index
 381based on this file, which uses the same syntax as .gitignore files.
 382If an entry matches a pattern in this file, skip-worktree will be
 383set on that entry. Otherwise, skip-worktree will be unset.
 384
 385Then it compares the new skip-worktree value with the previous one. If
 386skip-worktree turns from unset to set, it will add the corresponding
 387file back. If it turns from set to unset, that file will be removed.
 388
 389While `$GIT_DIR/info/sparse-checkout` is usually used to specify what
 390files are in. You can also specify what files are _not_ in, using
 391negate patterns. For example, to remove file "unwanted":
 392
 393----------------
 394*
 395!unwanted
 396----------------
 397
 398Another tricky thing is fully repopulating working directory when you
 399no longer want sparse checkout. You cannot just disable "sparse
 400checkout" because skip-worktree are still in the index and you working
 401directory is still sparsely populated. You should re-populate working
 402directory with the `$GIT_DIR/info/sparse-checkout` file content as
 403follows:
 404
 405----------------
 406*
 407----------------
 408
 409Then you can disable sparse checkout. Sparse checkout support in "git
 410read-tree" and similar commands is disabled by default. You need to
 411turn `core.sparseCheckout` on in order to have sparse checkout
 412support.
 413
 414
 415SEE ALSO
 416--------
 417linkgit:git-write-tree[1]; linkgit:git-ls-files[1];
 418linkgit:gitignore[5]
 419
 420
 421Author
 422------
 423Written by Linus Torvalds <torvalds@osdl.org>
 424
 425Documentation
 426--------------
 427Documentation by David Greaves, Junio C Hamano and the git-list <git@vger.kernel.org>.
 428
 429GIT
 430---
 431Part of the linkgit:git[1] suite