git rebase

GIT-REBASE(1)                                                        Git Manual                                                       GIT-REBASE(1)






NAME
       git-rebase - Forward-port local commits to the updated upstream head


SYNOPSIS
       git rebase [-i | --interactive] [options] [--onto ]
               [] []
       git rebase [-i | --interactive] [options] --onto
               --root []
       git rebase --continue | --skip | --abort

git-rebase命令主要用在从上游分支获取最新commit信息，并有机的将当前分支和上游分支进行合并。


DESCRIPTION
       If is specified, git rebase will perform an automatic git checkout before doing anything else. Otherwise it remains on the
       current branch.  


       If is not specified, the upstream configured in branch..remote and branch..merge options will be used; see git-
       config(1) for details. If you are currently not on any branch or if the current branch does not have a configured upstream, the rebase will
       abort.


       All changes made by commits in the current branch but that are not in are saved to a temporary area. This is the same set of
       commits that would be shown by git log ..HEAD (or git log HEAD, if --root is specified).


       The current branch is reset to , or if the --onto option was supplied. This has the exact same effect as git reset
       --hard (or ). ORIG_HEAD is set to point at the tip of the branch before the reset.


       The commits that were previously saved into the temporary area are then reapplied to the current branch, one by one, in order. Note that any
       commits in HEAD which introduce the same textual changes as a commit in HEAD.. are omitted (i.e., a patch already accepted
       upstream with a different commit message or timestamp will be skipped).


       It is possible that a merge failure will prevent this process from being completely automatic. You will have to resolve any such merge
       failure and run git rebase --continue. Another option is to bypass the commit that caused the merge failure with git rebase --skip. To check
       out the original and remove the .git/rebase-apply working files, use the command git rebase --abort instead.


       Assume the following history exists and the current branch is "topic":


                     A---B---C topic
                    /
               D---E---F---G master




       From this point, the result of either of the following commands:


           git rebase master
           git rebase master topic


       would be:


                             A'--B'--C' topic
                            /
               D---E---F---G master




       NOTE: The latter form is just a short-hand of git checkout topic followed by git rebase master. When rebase exits topic will remain the
       checked-out branch.


       If the upstream branch already contains a change you have made (e.g., because you mailed a patch which was applied upstream), then that
       commit will be skipped. For example, running ‘git rebase master` on the following history (in which A’ and A introduce the same set of
       changes, but have different committer information):


                     A---B---C topic
                    /
               D---E---A'---F master




       will result in:


                              B'---C' topic
                             /
               D---E---A'---F master




       Here is how you would transplant a topic branch based on one branch to another, to pretend that you forked the topic branch from the latter
       branch, using rebase --onto.


       First let’s assume your topic is based on branch next. For example, a feature developed in topic depends on some functionality which is
       found in next.


               o---o---o---o---o  master
                    \
                     o---o---o---o---o  next
                                      \
                                       o---o---o  topic




       We want to make topic forked from branch master; for example, because the functionality on which topic depends was merged into the more
       stable master branch. We want our tree to look like this:


               o---o---o---o---o  master
                   |            \
                   |             o'--o'--o'  topic
                    \
                     o---o---o---o---o  next




       We can get this using the following command:


           git rebase --onto master next topic


       Another example of --onto option is to rebase part of a branch. If we have the following situation:


                                       H---I---J topicB
                                      /
                             E---F---G  topicA
                            /
               A---B---C---D  master




       then the command


           git rebase --onto master topicA topicB


       would result in:


                            H'--I'--J'  topicB
                           /
                           | E---F---G  topicA
                           |/
               A---B---C---D  master




       This is useful when topicB does not depend on topicA.


       A range of commits could also be removed with rebase. If we have the following situation:


               E---F---G---H---I---J  topicA




       then the command


           git rebase --onto topicA~5 topicA~3 topicA


       would result in the removal of commits F and G:


               E---H'---I'---J'  topicA




       This is useful if F and G were flawed in some way, or should not be part of topicA. Note that the argument to --onto and the
       parameter can be any valid commit-ish.


       In case of conflict, git rebase will stop at the first problematic commit and leave conflict markers in the tree. You can use git diff to
       locate the markers (<<<<<<) and make edits to resolve the conflict. For each file you edit, you need to tell git that the conflict has been
       resolved, typically this would be done with


           git add


       After resolving the conflict manually and updating the index with the desired resolution, you can continue the rebasing process with


           git rebase --continue


       Alternatively, you can undo the git rebase with


           git rebase --abort


CONFIGURATION
       rebase.stat
           Whether to show a diffstat of what changed upstream since the last rebase. False by default.


       rebase.autosquash
           If set to true enable --autosquash option by default.


OPTIONS
       
           Starting point at which to create the new commits. If the --onto option is not specified, the starting point is . May be any
           valid commit, and not just an existing branch name.


           As a special case, you may use "A...B" as a shortcut for the merge base of A and B if there is exactly one merge base. You can leave out
           at most one of A and B, in which case it defaults to HEAD.


       
           Upstream branch to compare against. May be any valid commit, not just an existing branch name. Defaults to the configured upstream for
           the current branch.


       
           Working branch; defaults to HEAD.


       --continue
           Restart the rebasing process after having resolved a merge conflict.


       --abort
           Abort the rebase operation and reset HEAD to the original branch. If was provided when the rebase operation was started, then
           HEAD will be reset to . Otherwise HEAD will be reset to where it was when the rebase operation was started.


       --skip
           Restart the rebasing process by skipping the current patch.


       -m, --merge
           Use merging strategies to rebase. When the recursive (default) merge strategy is used, this allows rebase to be aware of renames on the
           upstream side.


           Note that a rebase merge works by replaying each commit from the working branch on top of the branch. Because of this, when a
           merge conflict happens, the side reported as ours is the so-far rebased series, starting with , and theirs is the working
           branch. In other words, the sides are swapped.


       -s , --strategy=
           Use the given merge strategy. If there is no -s option git merge-recursive is used instead. This implies --merge.


           Because git rebase replays each commit from the working branch on top of the branch using the given strategy, using the ours
           strategy simply discards all patches from the , which makes little sense.


       -X , --strategy-option=
           Pass the through to the merge strategy. This implies --merge and, if no strategy has been specified, -s recursive.
           Note the reversal of ours and theirs as noted in above for the -m option.


       -q, --quiet
           Be quiet. Implies --no-stat.


       -v, --verbose
           Be verbose. Implies --stat.


       --stat
           Show a diffstat of what changed upstream since the last rebase. The diffstat is also controlled by the configuration option rebase.stat.


       -n, --no-stat
           Do not show a diffstat as part of the rebase process.


       --no-verify
           This option bypasses the pre-rebase hook. See also githooks(5).


       --verify
           Allows the pre-rebase hook to run, which is the default. This option can be used to override --no-verify. See also githooks(5).


       -C
           Ensure at least lines of surrounding context match before and after each change. When fewer lines of surrounding context exist they
           all must match. By default no context is ever ignored.


       -f, --force-rebase
           Force the rebase even if the current branch is a descendant of the commit you are rebasing onto. Normally non-interactive rebase will
           exit with the message "Current branch is up to date" in such a situation. Incompatible with the --interactive option.


           You may find this (or --no-ff with an interactive rebase) helpful after reverting a topic branch merge, as this option recreates the
           topic branch with fresh commits so it can be remerged successfully without needing to "revert the reversion" (see the
           revert-a-faulty-merge How-To[1] for details).


       --ignore-whitespace, --whitespace=
           These flag are passed to the git apply program (see git-apply(1)) that applies the patch. Incompatible with the --interactive option.


       --committer-date-is-author-date, --ignore-date
           These flags are passed to git am to easily change the dates of the rebased commits (see git-am(1)). Incompatible with the --interactive
           option.


       -i, --interactive
           Make a list of the commits which are about to be rebased. Let the user edit that list before rebasing. This mode can also be used to
           split commits (see SPLITTING COMMITS below).


       -p, --preserve-merges
           Instead of ignoring merges, try to recreate them.


           This uses the --interactive machinery internally, but combining it with the --interactive option explicitly is generally not a good idea
           unless you know what you are doing (see BUGS below).


       --root
           Rebase all commits reachable from , instead of limiting them with an . This allows you to rebase the root commit(s) on
           a branch. Must be used with --onto, and will skip changes already contained in (instead of ). When used together
           with --preserve-merges, all root commits will be rewritten to have as parent instead.


       --autosquash, --no-autosquash
           When the commit log message begins with "squash! ..." (or "fixup! ..."), and there is a commit whose title begins with the same ...,
           automatically modify the todo list of rebase -i so that the commit marked for squashing comes right after the commit to be modified, and
           change the action of the moved commit from pick to squash (or fixup).


           This option is only valid when the --interactive option is used.


           If the --autosquash option is enabled by default using the configuration variable rebase.autosquash, this option can be used to override
           and disable this setting.


       --no-ff
           With --interactive, cherry-pick all rebased commits instead of fast-forwarding over the unchanged ones. This ensures that the entire
           history of the rebased branch is composed of new commits.


           Without --interactive, this is a synonym for --force-rebase.


           You may find this helpful after reverting a topic branch merge, as this option recreates the topic branch with fresh commits so it can
           be remerged successfully without needing to "revert the reversion" (see the revert-a-faulty-merge How-To[1] for details).


MERGE STRATEGIES
       The merge mechanism (git-merge and git-pull commands) allows the backend merge strategies to be chosen with -s option. Some strategies can
       also take their own options, which can be passed by giving -X arguments to git-merge and/or git-pull.


       resolve
           This can only resolve two heads (i.e. the current branch and another branch you pulled from) using a 3-way merge algorithm. It tries to
           carefully detect criss-cross merge ambiguities and is considered generally safe and fast.


       recursive
           This can only resolve two heads using a 3-way merge algorithm. When there is more than one common ancestor that can be used for 3-way
           merge, it creates a merged tree of the common ancestors and uses that as the reference tree for the 3-way merge. This has been reported
           to result in fewer merge conflicts without causing mis-merges by tests done on actual merge commits taken from Linux 2.6 kernel
           development history. Additionally this can detect and handle merges involving renames. This is the default merge strategy when pulling
           or merging one branch.


           The recursive strategy can take the following options:


           ours
               This option forces conflicting hunks to be auto-resolved cleanly by favoring our version. Changes from the other tree that do not
               conflict with our side are reflected to the merge result.


               This should not be confused with the ours merge strategy, which does not even look at what the other tree contains at all. It
               discards everything the other tree did, declaring our history contains all that happened in it.


           theirs
               This is opposite of ours.


           patience
               With this option, merge-recursive spends a little extra time to avoid mismerges that sometimes occur due to unimportant matching
               lines (e.g., braces from distinct functions). Use this when the branches to be merged have diverged wildly. See also git-diff(1)
               --patience.


           ignore-space-change, ignore-all-space, ignore-space-at-eol
               Treats lines with the indicated type of whitespace change as unchanged for the sake of a three-way merge. Whitespace changes mixed
               with other changes to a line are not ignored. See also git-diff(1) -b, -w, and --ignore-space-at-eol.


               ·   If their version only introduces whitespace changes to a line, our version is used;


               ·   If our version introduces whitespace changes but their version includes a substantial change, their version is used;


               ·   Otherwise, the merge proceeds in the usual way.


           renormalize
               This runs a virtual check-out and check-in of all three stages of a file when resolving a three-way merge. This option is meant to
               be used when merging branches with different clean filters or end-of-line normalization rules. See "Merging branches with differing
               checkin/checkout attributes" in gitattributes(5) for details.


           no-renormalize
               Disables the renormalize option. This overrides the merge.renormalize configuration variable.


           rename-threshold=
               Controls the similarity threshold used for rename detection. See also git-diff(1) -M.


           subtree[=]
               This option is a more advanced form of subtree strategy, where the strategy makes a guess on how two trees must be shifted to match
               with each other when merging. Instead, the specified path is prefixed (or stripped from the beginning) to make the shape of two
               trees to match.


       octopus
           This resolves cases with more than two heads, but refuses to do a complex merge that needs manual resolution. It is primarily meant to
           be used for bundling topic branch heads together. This is the default merge strategy when pulling or merging more than one branch.


       ours
           This resolves any number of heads, but the resulting tree of the merge is always that of the current branch head, effectively ignoring
           all changes from all other branches. It is meant to be used to supersede old development history of side branches. Note that this is
           different from the -Xours option to the recursive merge strategy.


       subtree
           This is a modified recursive strategy. When merging trees A and B, if B corresponds to a subtree of A, B is first adjusted to match the
           tree structure of A, instead of reading the trees at the same level. This adjustment is also done to the common ancestor tree.


NOTES
       You should understand the implications of using git rebase on a repository that you share. See also RECOVERING FROM UPSTREAM REBASE below.


       When the git-rebase command is run, it will first execute a "pre-rebase" hook if one exists. You can use this hook to do sanity checks and
       reject the rebase if it isn’t appropriate. Please see the template pre-rebase hook script for an example.


       Upon completion, will be the current branch.


INTERACTIVE MODE
       Rebasing interactively means that you have a chance to edit the commits which are rebased. You can reorder the commits, and you can remove
       them (weeding out bad or otherwise unwanted patches).


       The interactive mode is meant for this type of workflow:


        1. have a wonderful idea


        2. hack on the code


        3. prepare a series for submission


        4. submit


       where point 2. consists of several instances of


       a) regular use


        1. finish something worthy of a commit


        2. commit


       b) independent fixup


        1. realize that something does not work


        2. fix that


        3. commit it


       Sometimes the thing fixed in b.2. cannot be amended to the not-quite perfect commit it fixes, because that commit is buried deeply in a
       patch series. That is exactly what interactive rebase is for: use it after plenty of "a"s and "b"s, by rearranging and editing commits, and
       squashing multiple commits into one.


       Start it with the last commit you want to retain as-is:


           git rebase -i


       An editor will be fired up with all the commits in your current branch (ignoring merge commits), which come after the given commit. You can
       reorder the commits in this list to your heart’s content, and you can remove them. The list looks more or less like this:


           pick deadbee The oneline of this commit
           pick fa1afe1 The oneline of the next commit
           ...




       The oneline descriptions are purely for your pleasure; git rebase will not look at them but at the commit names ("deadbee" and "fa1afe1" in
       this example), so do not delete or edit the names.


       By replacing the command "pick" with the command "edit", you can tell git rebase to stop after applying that commit, so that you can edit
       the files and/or the commit message, amend the commit, and continue rebasing.


       If you just want to edit the commit message for a commit, replace the command "pick" with the command "reword".


       If you want to fold two or more commits into one, replace the command "pick" for the second and subsequent commits with "squash" or "fixup".
       If the commits had different authors, the folded commit will be attributed to the author of the first commit. The suggested commit message
       for the folded commit is the concatenation of the commit messages of the first commit and of those with the "squash" command, but omits the
       commit messages of commits with the "fixup" command.


       git rebase will stop when "pick" has been replaced with "edit" or when a command fails due to merge errors. When you are done editing and/or
       resolving conflicts you can continue with git rebase --continue.


       For example, if you want to reorder the last 5 commits, such that what was HEAD~4 becomes the new HEAD. To achieve that, you would call git
       rebase like this:


           $ git rebase -i HEAD~5




       And move the first patch to the end of the list.


       You might want to preserve merges, if you have a history like this:


                      X
                       \
                    A---M---B
                   /
           ---o---O---P---Q




       Suppose you want to rebase the side branch starting at "A" to "Q". Make sure that the current HEAD is "B", and call


           $ git rebase -i -p --onto Q O




       Reordering and editing commits usually creates untested intermediate steps. You may want to check that your history editing did not break
       anything by running a test, or at least recompiling at intermediate points in history by using the "exec" command (shortcut "x"). You may do
       so by creating a todo list like this one:


           pick deadbee Implement feature XXX
           fixup f1a5c00 Fix to feature XXX
           exec make
           pick c0ffeee The oneline of the next commit
           edit deadbab The oneline of the commit after
           exec cd subdir; make test
           ...




       The interactive rebase will stop when a command fails (i.e. exits with non-0 status) to give you an opportunity to fix the problem. You can
       continue with git rebase --continue.


       The "exec" command launches the command in a shell (the one specified in $SHELL, or the default shell if $SHELL is not set), so you can use
       shell features (like "cd", ">", ";" ...). The command is run from the root of the working tree.


SPLITTING COMMITS
       In interactive mode, you can mark commits with the action "edit". However, this does not necessarily mean that git rebase expects the result
       of this edit to be exactly one commit. Indeed, you can undo the commit, or you can add other commits. This can be used to split a commit
       into two:


       ·   Start an interactive rebase with git rebase -i ^, where is the commit you want to split. In fact, any commit range will
           do, as long as it contains that commit.


       ·   Mark the commit you want to split with the action "edit".


       ·   When it comes to editing that commit, execute git reset HEAD^. The effect is that the HEAD is rewound by one, and the index follows
           suit. However, the working tree stays the same.


       ·   Now add the changes to the index that you want to have in the first commit. You can use git add (possibly interactively) or git gui (or
           both) to do that.


       ·   Commit the now-current index with whatever commit message is appropriate now.


       ·   Repeat the last two steps until your working tree is clean.


       ·   Continue the rebase with git rebase --continue.


       If you are not absolutely sure that the intermediate revisions are consistent (they compile, pass the testsuite, etc.) you should use git
       stash to stash away the not-yet-committed changes after each commit, test, and amend the commit if fixes are necessary.


RECOVERING FROM UPSTREAM REBASE
       Rebasing (or any other form of rewriting) a branch that others have based work on is a bad idea: anyone downstream of it is forced to
       manually fix their history. This section explains how to do the fix from the downstream’s point of view. The real fix, however, would be to
       avoid rebasing the upstream in the first place.


       To illustrate, suppose you are in a situation where someone develops a subsystem branch, and you are working on a topic that is dependent on
       this subsystem. You might end up with a history like the following:


               o---o---o---o---o---o---o---o---o  master
                    \
                     o---o---o---o---o  subsystem
                                      \
                                       *---*---*  topic




       If subsystem is rebased against master, the following happens:


               o---o---o---o---o---o---o---o  master
                    \                       \
                     o---o---o---o---o       o'--o'--o'--o'--o'  subsystem
                                      \
                                       *---*---*  topic




       If you now continue development as usual, and eventually merge topic to subsystem, the commits from subsystem will remain duplicated
       forever:


               o---o---o---o---o---o---o---o  master
                    \                       \
                     o---o---o---o---o       o'--o'--o'--o'--o'--M  subsystem
                                      \                         /
                                       *---*---*-..........-*--*  topic




       Such duplicates are generally frowned upon because they clutter up history, making it harder to follow. To clean things up, you need to
       transplant the commits on topic to the new subsystem tip, i.e., rebase topic. This becomes a ripple effect: anyone downstream from topic is
       forced to rebase too, and so on!


       There are two kinds of fixes, discussed in the following subsections:


       Easy case: The changes are literally the same.
           This happens if the subsystem rebase was a simple rebase and had no conflicts.


       Hard case: The changes are not the same.
           This happens if the subsystem rebase had conflicts, or used --interactive to omit, edit, squash, or fixup commits; or if the upstream
           used one of commit --amend, reset, or filter-branch.


   The easy case
       Only works if the changes (patch IDs based on the diff contents) on subsystem are literally the same before and after the rebase subsystem
       did.


       In that case, the fix is easy because git rebase knows to skip changes that are already present in the new upstream. So if you say (assuming
       you’re on topic)


               $ git rebase subsystem




       you will end up with the fixed history


               o---o---o---o---o---o---o---o  master
                                            \
                                             o'--o'--o'--o'--o'  subsystem
                                                              \
                                                               *---*---*  topic




   The hard case
       Things get more complicated if the subsystem changes do not exactly correspond to the ones before the rebase.


           Note
           While an "easy case recovery" sometimes appears to be successful even in the hard case, it may have unintended consequences. For
           example, a commit that was removed via git rebase --interactive will be resurrected!


       The idea is to manually tell git rebase "where the old subsystem ended and your topic began", that is, what the old merge-base between them
       was. You will have to find a way to name the last commit of the old subsystem, for example:


       ·   With the subsystem reflog: after git fetch, the old tip of subsystem is at subsystem@{1}. Subsequent fetches will increase the number.
           (See git-reflog(1).)


       ·   Relative to the tip of topic: knowing that your topic has three commits, the old tip of subsystem must be topic~3.


       You can then transplant the old subsystem..topic to the new tip by saying (for the reflog case, and assuming you are on topic already):


               $ git rebase --onto subsystem subsystem@{1}




       The ripple effect of a "hard case" recovery is especially bad: everyone downstream from topic will now have to perform a "hard case"
       recovery too!


BUGS
       The todo list presented by --preserve-merges --interactive does not represent the topology of the revision graph. Editing commits and
       rewording their commit messages should work fine, but attempts to reorder commits tend to produce counterintuitive results.


       For example, an attempt to rearrange


           1 --- 2 --- 3 --- 4 --- 5




       to


           1 --- 2 --- 4 --- 3 --- 5




       by moving the "pick 4" line will result in the following history:


                   3
                  /
           1 --- 2 --- 4 --- 5




GIT
       Part of the git(1) suite


NOTES
        1. revert-a-faulty-merge How-To
           file:///usr/share/doc/git/html/howto/revert-a-faulty-merge.txt






Git 1.7.9.5                                                          04/11/2012                                                       GIT-REBASE(1)