Branches - Git Recipes - page 194

to display all branches: $ git log --oneline --graph --decorate --
all .

How It Works

The recipe uses two important commands:

$ git cherry-pick doc
$ git reset --hard HEAD ∼

The first command— cherry-pick —makes a copy of a given revision in a cur-
rent branch. It needs a single parameter that points to the revision to be copied. If you
pass a name of the branch to this command, it will copy the revision pointed to by the
branch; that is, the tip of the branch. You can use other means to identify a revision, as
well, for example:

$ git cherry-pick 7c9bc41 # shortened SHA-1 name

$ git cherry-pick info ∼ # last but one commit in

info branch
$ git cherry-pick info ∼ 3 # branch name and ancestor

reference

The term to copy a revision can be a little misleading here. Every revision you cre-
ate has a unique name. There are no exemptions from this rule. You can only:

• Add new revisions to the git's database
• Remove existing revisions from the git's database

Git doesn't implement modification of objects stored in a database. The name of the
commit is a SHA-1 hash, which was produced using your name, current timestamp, the
snapshot of your files, and the name of the parent revision. When we copy a revision
from one place in the history to another we change at least one of these parameters: the
name of the parent revision. As a result there is no way to preserve the old SHA-1 for a
copied revision. Thus the operation realized by cherry-pick command can be bet-
ter described as applying the changes introduced by a given revision on top of your
current branch. The command creates a completely new revision that reproduces the