Git Reference
Git Reference (http://gitref.org/index.html)
Getting and Creating Projects
In order to do anything in Git, you have to have a Git repository. This is where Git stores the data for the snapshots you are saving.
There are two main ways to get a Git repository. One way is to simply initialize a new one from an existing directory, such as a new project or a project new to source control. The second way is to clone one from a public Git repository, as you would do if you wanted a copy or wanted to work with someone on a project. We will cover both of these here.
In this case, we can use it to unstage something that you have accidentally staged. Let's say that you have modified two files and want to record them into two different commits. You should stage and commit one, then stage and commit the other. If you accidentally stage both of them, how do you un-stage one? You do it with
So, let's see what it looks like to unstage something. Here we have two files that have been modified since our last commit. We will stage both, then unstage one of them.
There are two main ways to get a Git repository. One way is to simply initialize a new one from an existing directory, such as a new project or a project new to source control. The second way is to clone one from a public Git repository, as you would do if you wanted a copy or wanted to work with someone on a project. We will cover both of these here.
git init initializes a directory as a Git repository
To create a repository from an existing directory of files, you can simply rungit init in that directory. For example, let's say we have a directory with a few files in it, like this: $ cd konichiwa $ ls README hello.rbThis is a project where we are writing examples of the "Hello World" program in every language. So far, we just have Ruby, but hey, it's a start. To start version controlling this with Git, we can simply run
git init. $ git init Initialized empty Git repository in /opt/konichiwa/.git/Now you can see that there is a
.git subdirectory in your project. This is your Git repository where all the data of your project snapshots are stored. $ ls -a . .. .git README hello.rbCongratulations, you now have a skeleton Git repository and can start snapshotting your project.
In a nutshell, you use
git init to make an existing directory of content into a new Git repository. You can do this in any directory at any time, completely locally.git clone copy a git repository so you can add to it
If you need to collaborate with someone on a project, or if you want to get a copy of a project so you can look at or use the code, you will clone it. You simply run the
git clone [url] command with the URL of the project you want to copy. $ git clone git://github.com/schacon/simplegit.git
Initialized empty Git repository in /private/tmp/simplegit/.git/
remote: Counting objects: 100, done.
remote: Compressing objects: 100% (86/86), done.
remote: Total 100 (delta 35), reused 0 (delta 0)
Receiving objects: 100% (100/100), 9.51 KiB, done.
Resolving deltas: 100% (35/35), done.
$ cd simplegit/
$ ls
README Rakefile libThis will copy the entire history of that project so you have it locally and it will give you a working directory of the main branch of that project so you can look at the code or start editing it. If you change into the new directory, you can see the .git subdirectory - that is where all the project data is. $ ls -a . .. .git README Rakefile lib $ cd .git $ ls HEAD description info packed-refs branches hooks logs refs config index objectsBy default, Git will create a directory that is the same name as the project in the URL you give it - basically whatever is after the last slash of the URL. If you want something different, you can just put it at the end of the command, after the URL.
In a nutshell, you use
git clone to get a local copy of a Git repository so you can look at it or start modifying it.Basic Snapshotting
Git is all about composing and saving snapshots of your project and then working with and comparing those snapshots. This section will explain the commands needed to compose and commit snapshots of your project.
An important concept here is that Git has an 'index', which acts as sort of a staging area for your snapshot. This allows you to build up a series of well composed snapshots from changed files in your working directory, rather than having to commit all of the file changes at once.
An important concept here is that Git has an 'index', which acts as sort of a staging area for your snapshot. This allows you to build up a series of well composed snapshots from changed files in your working directory, rather than having to commit all of the file changes at once.
In a nutshell, you will use
git add to start tracking new files and also to stage changes to already tracked files, then git status and git diff to see what has been modified and staged and finally git commit to record your snapshot into your history. This will be the basic workflow that you use most of the time. git add adds file contents to the staging area
In Git, you have to add file contents to your staging area before you can commit them. If the file is new, you can run
Going back to our Hello World example, once we've initiated the project, we would now start adding our files to it and we would do that with
git add to initially add the file to your staging area, but even if the file is already "tracked" - ie, it was in your last commit - you still need to call git add to add new modifications to your staging area. Let's see a few examples of this. Going back to our Hello World example, once we've initiated the project, we would now start adding our files to it and we would do that with
git add. We can use git status to see what the state of our project is. $ git status -s ?? README ?? hello.rbSo right now we have two untracked files. We can now add them.
$ git add README hello.rbNow if we run
git status again, we'll see that they've been added. $ git status -s A README A hello.rb
It is also common to recursively add all files in a new project by specifying the current working directory like this:
OK, so now if we edit one of these files and run git add .. Since Git will recursively add all files under a directory you give it, if you give it the current working directory, it will simply start tracking every file there. In this case, a git add . would have done the same thing as a git add README hello.rb, or for that matter so would git add *, but that's only because we don't have subdirectories which the * would not recurse into. git status again, we will see something odd.$ vim README $ git status -s AM README A hello.rbThe 'AM' status means that the file has been modified on disk since we last added it. This means that if we commit our snapshot right now, we will be recording the version of the file when we last ran
git add, not the version that is on our disk. Git does not assume that what the file looks like on disk is necessarily what you want to snapshot - you have to tell Git with the git add command. In a nutshell, you run
For a very interesting example of using this flexibility to stage only parts of modified files at a time, see the '-p' option to git add on a file when you want to include whatever changes you've made to it in your next commit snapshot. Anything you've changed that is not added will not be included - this means you can craft your snapshots with a bit more precision than most other SCM systems.git add in the Pro Git book.git status view the status of your files in the working directory and staging area
As you saw in the
Git will also tell you about files that were deleted since your last commit or files that were modified or staged since your last commit.
git add section, in order to see what the status of your staging area is compared to the code in your working directory, you can run the git status command. I demonstrated using it with the -s option, which gives you short output. Without that flag, the git status command will give you more context and hints. Here is the same status output with and without the -s. The short output looks like this: $ git status -s AM README A hello.rbWhere the same status with the long output looks like this:
$ git status # On branch master # # Initial commit # # Changes to be committed: # (use "git rm --cached ..." to unstage) # # new file: README # new file: hello.rb # # Changed but not updated: # (use "git add ..." to update what will be committed) # (use "git checkout -- ..." to discard changes in working directory) # # modified: README #You can easily see how much more compact the short output is, but the long output has useful tips and hints as to what commands you may want to use next.
Git will also tell you about files that were deleted since your last commit or files that were modified or staged since your last commit.
$ git status -s M README D hello.rbYou can see there are two columns in the short status output. The first column is for the staging area, the second is for the working directory. So for example, if you have the README file staged and then you modify it again without running
git add a second time, you'll see this: $ git status -s MM README D hello.rb
In a nutshell, you run
git status to see if anything has been modified and/or staged since your last commit so you can decide if you want to commit a new snapshot and what will be recorded in it. git diff shows diff of what is staged and what is modified but unstaged
There are two main uses of the
git diff command. One use we will describe here, the other we will describe later in the "Inspection and Comparison" section. The way we're going to use it here is to describe the changes that are staged or modified on disk but unstaged.git diff show diff of unstaged changes
Without any extra arguments, a simplegit diff will display in unified diff format (a patch) what code or content you've changed in your project since the last commit that are not yet staged for the next commit snapshot. $ vim hello.rb $ git status -s M hello.rb $ git diff diff --git a/hello.rb b/hello.rb index d62ac43..8d15d50 100644 --- a/hello.rb +++ b/hello.rb @@ -1,7 +1,7 @@ class HelloWorld def self.hello - puts "hello world" + puts "hola mundo" end endSo where
git status will show you what files have changed and/or been staged since your last commit, git diff will show you what those changes actually are, line by line. It's generally a good follow-up command to git status git diff --cached show diff of staged changes
Thegit diff --cached command will show you what contents have been staged. That is, this will show you the changes that will currently go into the next commit snapshot. So, if you were to stage the change to hello.rb in the example above, git diff by itself won't show you any output because it will only show you what is not yet staged. $ git status -s M hello.rb $ git add hello.rb $ git status -s M hello.rb $ git diff $If you want to see the staged changes, you can run
git diff --cached instead.$ git status -s M hello.rb $ git diff $ $ git diff --cached diff --git a/hello.rb b/hello.rb index d62ac43..8d15d50 100644 --- a/hello.rb +++ b/hello.rb @@ -1,7 +1,7 @@ class HelloWorld def self.hello - puts "hello world" + puts "hola mundo" end end
git diff HEAD show diff of all staged or unstaged changes
If you want to see both staged and unstaged changes together, you can rungit diff HEAD - this basically means you want to see the difference between your working directory and the last commit, ignoring the staging area. If we make another change to our hello.rb file then we'll have some changes staged and some changes unstaged. Here are what all three diff commands will show you:$ vim hello.rb $ git diff diff --git a/hello.rb b/hello.rb index 4f40006..2ae9ba4 100644 --- a/hello.rb +++ b/hello.rb @@ -1,7 +1,7 @@ class HelloWorld + # says hello def self.hello puts "hola mundo" end end $ git diff --cached diff --git a/hello.rb b/hello.rb index 2aabb6e..4f40006 100644 --- a/hello.rb +++ b/hello.rb @@ -1,7 +1,7 @@ class HelloWorld def self.hello - puts "hello world" + puts "hola mundo" end end $ git diff HEAD diff --git a/hello.rb b/hello.rb index 2aabb6e..2ae9ba4 100644 --- a/hello.rb +++ b/hello.rb @@ -1,7 +1,8 @@ class HelloWorld + # says hello def self.hello - puts "hello world" + puts "hola mundo" end end
git diff --stat show summary of changes instead of a full diff
If we don't want the full diff output, but we want more than thegit status output, we can use the --stat option, which will give us a summary of changes instead. Here is the same example as above, but using the --stat option instead. $ git status -s MM hello.rb $ git diff --stat hello.rb | 1 + 1 files changed, 1 insertions(+), 0 deletions(-) $ git diff --cached --stat hello.rb | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-) $ git diff HEAD --stat hello.rb | 3 ++- 1 files changed, 2 insertions(+), 1 deletions(-)You can also provide a file path at the end of any of these options to limit the
diff output to a specific file or subdirectory. In a nutshell, you run
git diff to see details of the git status command - how files have been modified or staged on a line by line basis. git commit records a snapshot of the staging area
Now that you have staged the content you want to snapshot with the
In general, it's very important to write a good commit message. For open source projects, it's generally a rule to write your message more or less in this format:
This now lets you complete the entire snapshotting workflow - you make changes to your files, then use
git add command, you run git commit to actually record the snapshot. Git records your name and email address with every commit you make, so the first step is to tell Git what these are. $ git config --global user.name 'Your Name' $ git config --global user.email you@somedomain.comLet's stage and commit all the changes to our
hello.rb file. In this first example, we'll use the -m option to provide the commit message on the command line. $ git add hello.rb
$ git status -s
M hello.rb
$ git commit -m 'my hola mundo changes'
[master 68aa034] my hola mundo changes
1 files changed, 2 insertions(+), 1 deletions(-)Now we have recorded the snapshot. If we run git status again, we will see that we have a "clean working directory", which means that we have not made any changes since our last commit - there is no un-snapshotted work in our checkout.$ git status # On branch master nothing to commit (working directory clean)If you leave off the
-m option, Git will try to open a text editor for you to write your commit message. In vim, which it will default to if it can find nothing else in your settings, the screen might look something like this: # Please enter the commit message for your changes. Lines starting # with '#' will be ignored, and an empty message aborts the commit. # On branch master # Changes to be committed: # (use "git reset HEAD ..." to unstage) # # modified: hello.rb # ~ ~ ".git/COMMIT_EDITMSG" 9L, 257CAt this point you add your actual commit message at the top of the document. Any lines starting with '#' will be ignored - Git will put the output of the
git status command in there for you as a reminder of what you have modified and staged.In general, it's very important to write a good commit message. For open source projects, it's generally a rule to write your message more or less in this format:
Short (50 chars or less) summary of changes More detailed explanatory text, if necessary. Wrap it to about 72 characters or so. In some contexts, the first line is treated as the subject of an email and the rest of the text as the body. The blank line separating the summary from the body is critical (unless you omit the body entirely); some git tools can get confused if you run the two together. Further paragraphs come after blank lines. - Bullet points are okay, too - Typically a hyphen or asterisk is used for the bullet, preceded by a single space, with blank lines in between, but conventions vary here # Please enter the commit message for your changes. Lines starting # with '#' will be ignored, and an empty message aborts the commit. # On branch master # Changes to be committed: # (use "git reset HEAD ..." to unstage) # # modified: hello.rb # ~ ~ ~ ".git/COMMIT_EDITMSG" 25L, 884C written
The commit message is very important. Since much of the power of Git is this flexibility in carefully crafting commits locally and then sharing them later, it is very powerful to be able to write three or four commits of logically separate changes so that your work may be more easily peer reviewed. Since there is a separation between committing and pushing those changes, do take the time to make it easier for the people you are working with to see what you've done by putting each logically separate change in a separate commit with a nice commit message so it is easier for them to see what you are doing and why.
git commit -a automatically stage all tracked, modified files before the commit
If you think thegit add stage of the workflow is too cumbersome, Git allows you to skip that part with the -a option. This basically tells Git to run git add on any file that is "tracked" - that is, any file that was in your last commit and has been modified. This allows you to do a more Subversion style workflow if you want, simply editing files and then running git commit -a when you want to snapshot everything that has been changed. You still need to run git add to start tracking new files, though, just like Subversion. $ vim hello.rb $ git status -s M hello.rb $ git commit -m 'changes to hello file' # On branch master # Changed but not updated: # (use "git add ..." to update what will be committed) # (use "git checkout -- ..." to discard changes in working directory) # # modified: hello.rb # no changes added to commit (use "git add" and/or "git commit -a") $ git commit -am 'changes to hello file' [master 78b2670] changes to hello file 1 files changed, 2 insertions(+), 1 deletions(-)Notice how if you don't stage any changes and then run
git commit, Git will simply give you the output of the git status command, reminding you that nothing is staged. I've highlighted the important part of that message, saying that nothing is added to be committed. If you use -a, it will add and commit everything at once. This now lets you complete the entire snapshotting workflow - you make changes to your files, then use
git add to stage files you want to change, git status and git diff to see what you've changed, and then finally git commit to actually record the snapshot forever.In a nutshell, you run
git commit to record the snapshot of your staged content. This snapshot can then be compared, shared and reverted to if you need to. git reset HEAD unstage changes that you have staged
git reset is probably the most confusing command written by humans. I've been using Git for years, even wrote a book on it and I still get confused by what it is going to do at times. So, I'll just tell you the three specific invocations of it that are generally helpful and ask you to blindly use it as I do - because it can be very useful. In this case, we can use it to unstage something that you have accidentally staged. Let's say that you have modified two files and want to record them into two different commits. You should stage and commit one, then stage and commit the other. If you accidentally stage both of them, how do you un-stage one? You do it with
git reset HEAD -- file. Technically here you don't have to add the -- - it is used to tell Git when you have stopped listing options and are now listing file paths, but it's probably good to get into the habit of using it to separate options from paths even if you don't need to. So, let's see what it looks like to unstage something. Here we have two files that have been modified since our last commit. We will stage both, then unstage one of them.
$ git status -s M README M hello.rb $ git add . $ git status -s M README M hello.rb $ git reset HEAD -- hello.rb Unstaged changes after reset: M hello.rb $ git status -s M README M hello.rbNow you can run a
git commit which will just record the changes to the README file, not the now unstaged hello.rb. In case you're curious, what it's actually doing here is it is resetting the checksum of the entry for that file in the "index" to be what it was in the last commit. Since
git add checksums a file and adds it to the "index", git reset HEAD overwrites that with what it was before, thereby effectively unstaging it. If you want to be able to just run
If you forget the command to unstage something, Git is helpful in reminding you in the output of the normal git unstage, you can easily setup an alias in Git. Just run git config --global alias.unstage "reset HEAD". Once you have run that, you can then just run git unstage [file] instead. git status command. For example, if you run git status without the -s when you have staged files, it will tell you how to unstage them:$ git status # On branch master # Changes to be committed: # (use "git reset HEAD ..." to unstage) # # modified: README # modified: hello.rb #
In a nutshell, you run
git reset HEAD to unstage files that you previously ran git add on and wish to not include in the next commit snapshotBranching and Merging
Branching in Git is one of my favorite features. If you have used other version control systems, it's probably helpful to forget most of what you think about branches - in fact, it may be more helpful to think of them practically as contexts since that is how you will most often be using them. When you checkout different branches, you change contexts that you are working in and you can quickly context-switch back and forth between several different branches.
In a nutshell you can create a branch with
git branch (branchname), switch into that context with git checkout (branchname), record commit snapshots while in that context, then can switch back and forth easily. When you switch branches, Git replaces your working directory with the snapshot of the latest commit on that branch so you don't have to have multiple directories for multiple branches. You merge branches together with git merge. You can easily merge multiple times from the same branch over time, or alternately you can choose to delete a branch immediately after merging it. git branch list, create and manage working contexts
git checkout switch to a new branch context
The
If you start on work it is very useful to always start it in a branch (because it's fast and easy to do) and then merge it in and delete the branch when you're done. That way if what you're working on doesn't work out you can easily discard it and if you're forced to switch back to a more stable context your work in progress is easy to put aside and then come back to.
git branch command is a general branch management tool for Git and can do several different things. We'll cover the basic ones that you'll use most - listing branches, creating branches and deleting branches. We will also cover basic git checkout here which switches you between your branches. git branch list your available branches
Without arguments,git branch will list out the local branches that you have. The branch that you are currently working on will have a star next to it and if you have coloring turned on, will show the current branch in green. $ git branch
* masterThis means that we have a 'master' branch and we are currently on it. When you run git init it will automatically create a 'master' branch for you by default, however there is nothing special about the name - you don't actually have to have a 'master' branch but since it's the default that is created, most projects do. git branch (branchname) create a new branch
So let's start by creating a new branch and switching to it. You can do that by runninggit branch (branchname). $ git branch testing
$ git branch
* master
testingNow we can see that we have a new branch. When you create a branch this way it creates the branch at your last commit so if you record some commits at this point and then switch to 'testing', it will revert your working directory context back to when you created the branch in the first place - you can think of it like a bookmark for where you currently are. Let's see this in action - we use git checkout (branch) to switch the branch we're currently on. $ ls README hello.rb $ echo 'test content' > test.txt $ echo 'more content' > more.txt $ git add *.txt $ git commit -m 'added two files' [master 8bd6d8b] added two files 2 files changed, 2 insertions(+), 0 deletions(-) create mode 100644 more.txt create mode 100644 test.txt $ ls README hello.rb more.txt test.txt $ git checkout testing Switched to branch 'testing' $ ls README hello.rbSo now we can see that when we switch to the 'testing' branch, our new files were removed. We could switch back to the 'master' branch and see them re-appear.
$ ls README hello.rb $ git checkout master Switched to branch 'master' $ ls README hello.rb more.txt test.txt
git checkout -b (branchname) create and immediately switch to a branch
In most cases you will be wanting to switch to the branch immediately, so you can do work in it and then merging into a branch that only contains stable work (such as 'master') at a later point when the work in your new context branch is stable. You can do this pretty easily withgit branch newbranch; git checkout newbranch, but Git gives you a shortcut for this: git checkout -b newbranch. $ git branch * master $ ls README hello.rb more.txt test.txt $ git checkout -b removals Switched to a new branch 'removals' $ git rm more.txt rm 'more.txt' $ git rm test.txt rm 'test.txt' $ ls README hello.rb $ git commit -am 'removed useless files' [removals 8f7c949] removed useless files 2 files changed, 0 insertions(+), 2 deletions(-) delete mode 100644 more.txt delete mode 100644 test.txt $ git checkout master Switched to branch 'master' $ ls README hello.rb more.txt test.txtYou can see there how we created a branch, removed some of our files while in the context of that branch, then switched back to our main branch and we see the files return. Branching safely isolates work that we do into contexts we can switch between.
If you start on work it is very useful to always start it in a branch (because it's fast and easy to do) and then merge it in and delete the branch when you're done. That way if what you're working on doesn't work out you can easily discard it and if you're forced to switch back to a more stable context your work in progress is easy to put aside and then come back to.
git branch -d (branchname) delete a branch
If we want to delete a branch (such as the 'testing' branch in the previous example, since there is no unique work on it), we can rungit branch -d (branch) to remove it. $ git branch * master testing $ git branch -d testing Deleted branch testing (was 78b2670). $ git branch * master
In a nutshell you use
git branch to list your current branches, create new branches and delete unnecessary or already merged branches. git merge merge a branch context into your current one
Once you have work isolated in a branch, you will eventually want to incorporate it into your main branch. You can merge any branch into your current branch with the
git merge command. Let's take as a simple example the 'removals' branch from above. If we create a branch and remove files in it and commit our removals to that branch, it is isolated from our main ('master', in this case) branch. To include those deletions in your 'master' branch, you can just merge in the 'removals' branch. $ git branch * master removals $ ls README hello.rb more.txt test.txt $ git merge removals Updating 8bd6d8b..8f7c949 Fast-forward more.txt | 1 - test.txt | 1 - 2 files changed, 0 insertions(+), 2 deletions(-) delete mode 100644 more.txt delete mode 100644 test.txt $ ls README hello.rb
more complex merges
Of course, this doesn't just work for simple file additions and deletions. Git will merge file modifications as well - in fact, it's very good at it. For example, let's see what happens when we edit a file in one branch and in another branch we rename it and then edit it and then merge these branches together. Chaos, you say? Let's see.$ git branch
* master
$ cat hello.rb
class HelloWorld
def self.hello
puts "Hello World"
end
end
HelloWorld.helloSo first we're going to create a new branch named 'change_class' and switch to it so your class renaming changes are isolated. I'm going to change each instance of 'HelloWorld' to 'HiWorld'.$ git checkout -b change_class M hello.rb Switched to a new branch 'change_class' $ vim hello.rb $ head -1 hello.rb class HiWorld $ git commit -am 'changed the class name' [change_class 3467b0a] changed the class name 1 files changed, 2 insertions(+), 4 deletions(-)So now I've committed the class renaming changes to the 'change_class' branch. If I now switch back to the 'master' branch my class name will revert to what it was before I switched branches. Here I can change something different (in this case the printed output) and at the same time rename the file from
hello.rb to ruby.rb. $ git checkout master Switched to branch 'master' $ git mv hello.rb ruby.rb $ vim ruby.rb $ git diff diff --git a/ruby.rb b/ruby.rb index 2aabb6e..bf64b17 100644 --- a/ruby.rb +++ b/ruby.rb @@ -1,7 +1,7 @@ class HelloWorld def self.hello - puts "Hello World" + puts "Hello World from Ruby" end end $ git commit -am 'added from ruby' [master b7ae93b] added from ruby 1 files changed, 1 insertions(+), 1 deletions(-) rename hello.rb => ruby.rb (65%)Now those changes are recorded in my 'master' branch. Notice that the class name is back to 'HelloWorld', not 'HiWorld'. Now I want to incorporate the 'HiWorld' change so I can just merge in my 'change_class' branch. However, I've changed the name of the file since I branched, what will Git do?
$ git branch
change_class
* master
$ git merge change_class
Renaming hello.rb => ruby.rb
Auto-merging ruby.rb
Merge made by recursive.
ruby.rb | 6 ++----
1 files changed, 2 insertions(+), 4 deletions(-)
$ cat ruby.rb
class HiWorld
def self.hello
puts "Hello World from Ruby"
end
end
HiWorld.helloWell, it will just figure it out. Notice that I had no merge conflicts and the file that had been renamed now has the 'HiWorld' class name change that was done in the other branch. Pretty cool.merge conflicts
So, Git merges are magical, we never ever have to deal with merge conflicts again, right? Not quite. In situations where the same block of code is edited in different branches there is no way for a computer to figure it out, so it's up to us. Let's see another example of changing the same line in two branches.$ git branch * master $ git checkout -b fix_readme Switched to a new branch 'fix_readme' $ vim README $ git commit -am 'fixed readme title' [fix_readme 3ac015d] fixed readme title 1 files changed, 1 insertions(+), 1 deletions(-)Now we have committed a change to one line in our README file in a branch. Now let's change the same line in a different way back on our 'master' branch.
$ git checkout master Switched to branch 'master' $ vim README $ git commit -am 'fixed readme title differently' [master 3cbb6aa] fixed readme title differently 1 files changed, 1 insertions(+), 1 deletions(-)Now is the fun part - we will merge the first branch into our master branch, causing a merge conflict.
$ git merge fix_readme Auto-merging README CONFLICT (content): Merge conflict in README Automatic merge failed; fix conflicts and then commit the result. $ cat README <<<<<<< HEAD Many Hello World Examples ======= Hello World Lang Examples >>>>>>> fix_readme This project has examples of hello world in nearly every programming language.You can see that Git inserts standard merge conflict markers, much like Subversion, into files when it gets a merge conflict. Now it's up to us to resolve them. We will do it manually here, but check out git mergetool if you want Git to fire up a graphical mergetool (like kdiff3, emerge, p4merge, etc) instead.
$ vim README # here I'm fixing the conflict $ git diff diff --cc README index 9103e27,69cad1a..0000000 --- a/README +++ b/README @@@ -1,4 -1,4 +1,4 @@@ - Many Hello World Examples -Hello World Lang Examples ++Many Hello World Lang Examples This project has examples of hello world inA cool tip in doing merge conflict resolution in Git is that if you run
git diff, it will show you both sides of the conflict and how you've resolved it as I've shown here. Now it's time to mark the file as resolved. In Git we do that with git add - to tell Git the file has been resolved, you have to stage it.$ git status -s UU README $ git add README $ git status -s M README $ git commit [master 8d585ea] Merge branch 'fix_readme'And now we've successfully resolved our merge conflict and committed the result.
In a nutshell you use
git merge to combine another branch context into your current branch. It automatically figures out how to best combine the different snapshots into a new snapshot with the unique work of both. git log show commit history of a branch
So far we have been committing snapshots of your project and switching between different isolated contexts, but what if we've forgotten how we've got to where we are? Or what if we want to know how one branch differs from another? Git provides a tool that shows you all the commit messages that have lead up to the snapshot you are currently on, which is called
To understand the log command, you have to understand what information is stored when you run the
To see a chronological list of the parents of any branch, you can run
We can also use it to see when the history was branched and merged with the very helpful
First we'll create a new branch to add the Erlang programming language Hello World example - we want to do this in a branch so that we don't muddy up our stable branch with code that may not work for a while so we can cleanly switch in and out of it.
In this case if we are interested in merging in the 'erlang' branch we want to see what commits are going to effect our snapshot when we do that merge. The way we tell Git that is by putting a
git log.To understand the log command, you have to understand what information is stored when you run the
git commit command to store a snapshot. In addition to the manifest of files and commit message and information about the person who committed it, Git also stores the commit that you based this snapshot on. That is, if you clone a project, what was the snapshot that you modified to get to the snapshot that you saved? This is helpful to give context to how the project got to where it is and allows Git to figure out who changed what. If Git has the snapshot you save and the one you based it on, then it can automatically figure out what you changed. The commit that a new commit was based on is called the "parent". To see a chronological list of the parents of any branch, you can run
git log when you are in that branch. For example, if we run git log in the Hello World project that we have been working on in this section, we'll see all the commit messages that we've done. $ git log commit 8d585ea6faf99facd39b55d6f6a3b3f481ad0d3d Merge: 3cbb6aa 3ac015d Author: Scott ChaconTo see a more compact version of the same history, we can use theDate: Fri Jun 4 12:59:47 2010 +0200 Merge branch 'fix_readme' Conflicts: README commit 3cbb6aae5c0cbd711c098e113ae436801371c95e Author: Scott Chacon Date: Fri Jun 4 12:58:53 2010 +0200 fixed readme title differently commit 3ac015da8ade34d4c7ebeffa2053fcac33fb495b Author: Scott Chacon Date: Fri Jun 4 12:58:36 2010 +0200 fixed readme title commit 558151a95567ba4181bab5746bc8f34bd87143d6 Merge: b7ae93b 3467b0a Author: Scott Chacon Date: Fri Jun 4 12:37:05 2010 +0200 Merge branch 'change_class' ...
--oneline option.$ git log --oneline 8d585ea Merge branch 'fix_readme' 3cbb6aa fixed readme title differently 3ac015d fixed readme title 558151a Merge branch 'change_class' b7ae93b added from ruby 3467b0a changed the class name 17f4acf first commitWhat this is telling us is that this is the history of the development of this project. If the commit messages are descriptive, this can inform us as to what all changes have been applied or have influenced the current state of the snapshot and thus what is in it.
We can also use it to see when the history was branched and merged with the very helpful
--graph option. Here is the same command but with the topology graph turned on:$ git log --oneline --graph * 8d585ea Merge branch 'fix_readme' |\ | * 3ac015d fixed readme title * | 3cbb6aa fixed readme title differently |/ * 558151a Merge branch 'change_class' |\ | * 3467b0a changed the class name * | b7ae93b added from ruby |/ * 17f4acf first commitNow we can more clearly see when effort diverged and then was merged back together. This is very nice for seeing what has happened or what changes are applied, but it is also incredibly useful for managing your branches. Let's create a new branch, do some work in it and then switch back and do some work in our master branch, then see how the
log command can help us figure out what is happening on each.First we'll create a new branch to add the Erlang programming language Hello World example - we want to do this in a branch so that we don't muddy up our stable branch with code that may not work for a while so we can cleanly switch in and out of it.
$ git checkout -b erlang Switched to a new branch 'erlang' $ vim erlang_hw.erl $ git add erlang_hw.erl $ git commit -m 'added erlang' [erlang ab5ab4c] added erlang 1 files changed, 5 insertions(+), 0 deletions(-) create mode 100644 erlang_hw.erlSince we're having fun playing in functional programming languages we get caught up in it and also add a Haskell example program while still in the branch named 'erlang'.
$ vim haskell.hs $ git add haskell.hs $ git commit -m 'added haskell' [erlang 1834130] added haskell 1 files changed, 4 insertions(+), 0 deletions(-) create mode 100644 haskell.hsFinally, we decide that we want to change the class name of our Ruby program back to the way it was. So, we can go back to the master branch and change that and we decide to just commit it directly in the master branch instead of creating another branch.
$ git checkout master Switched to branch 'master' $ ls README ruby.rb $ vim ruby.rb $ git commit -am 'reverted to old class name' [master 594f90b] reverted to old class name 1 files changed, 2 insertions(+), 2 deletions(-)So, now say we don't work on the project for a while, we have other things to do. When we come back we want to know what the 'erlang' branch is all about and where we've left off on the master branch. Just by looking at the branch name, we can't know that we made Haskell changes in there, but using
git log we easily can. If you give Git a branch name, it will show you just the commits that are "reachable" in the history of that branch, that is the commits that influenced the final snapshot.$ git log --oneline erlang
1834130 added haskell
ab5ab4c added erlang
8d585ea Merge branch 'fix_readme'
3cbb6aa fixed readme title differently
3ac015d fixed readme title
558151a Merge branch 'change_class'
b7ae93b added from ruby
3467b0a changed the class name
17f4acf first commitThis way, it's pretty easy to see that we have Haskell code included in the branch (as I've highlighted). What is even cooler is that we can easily tell Git that we only are interested in the commits that are reachable in one branch that are not reachable in another, in other words which commits are unique to a branch in comparison to another. In this case if we are interested in merging in the 'erlang' branch we want to see what commits are going to effect our snapshot when we do that merge. The way we tell Git that is by putting a
^ in front of the branch that we don't want to see. For instance, if we want to see the commits that are in the 'erlang' branch that are not in the 'master' branch, we can do erlang ^master, or vice versa. $ git log --oneline erlang ^master 1834130 added haskell ab5ab4c added erlang $ git log --oneline master ^erlang 594f90b reverted to old class nameThis gives us a nice, simple branch management tool. It allows us to easily see what commits are unique to which branches so we know what we're missing and what we would be merging in if we were to do a merge.
In a nutshell you use
git log to list out the commit history or list of changes people have made that have lead to the snapshot at the tip of the branch. This allows you to see how the project in that context got to the state that it is currently in. git tag tag a point in history as important
If you get to a point that is important and you want to forever remember that specific commit snapshot, you can tag it with
Let's say we want to release our Hello World project as version "1.0". We can tag the last commit (
Now, notice when we run
We don't have to tag the commit that we're on, however. If we forgot to tag a commit that we released, we can retroactively tag it by running the same command, but with the commit SHA at the end. For example, say we had released commit
git tag. The tag command will basically put a permanent bookmark at a specific commit so you can use it to compare to other commits in the future. This is often done when you cut a release or ship something. Let's say we want to release our Hello World project as version "1.0". We can tag the last commit (
HEAD) as "v1.0" by running git tag -a v1.0. The -a means "make an annotated tag", which allows you to add a tag message to it, which is what you almost always want to do. Running this without the -a works too, but it doesn't record when it was tagged, who tagged it, or let you add a tag message. I would recommend always creating annotated tags.$ git tag -a v1.0When you run the
git tag -a command, Git will open your editor and have you write a tag message, just like you would write a commit message.Now, notice when we run
git log --decorate, we can see our tag there.$ git log --oneline --decorate --graph
* 594f90b (HEAD, tag: v1.0, master) reverted to old class name
* 8d585ea Merge branch 'fix_readme'
|\
| * 3ac015d (fix_readme) fixed readme title
* | 3cbb6aa fixed readme title differently
|/
* 558151a Merge branch 'change_class'
|\
| * 3467b0a changed the class name
* | b7ae93b added from ruby
|/
* 17f4acf first commitIf we do more commits, the tag will stay right at that commit, so we have that specific snapshot tagged forever and can always compare future snapshots to it.We don't have to tag the commit that we're on, however. If we forgot to tag a commit that we released, we can retroactively tag it by running the same command, but with the commit SHA at the end. For example, say we had released commit
558151a (several commits back) but forgot to tag it at the time. We can just tag it now:$ git tag -a v0.9 558151a
$ git log --oneline --decorate --graph
* 594f90b (HEAD, tag: v1.0, master) reverted to old class name
* 8d585ea Merge branch 'fix_readme'
|\
| * 3ac015d (fix_readme) fixed readme title
* | 3cbb6aa fixed readme title differently
|/
* 558151a (tag: v0.9) Merge branch 'change_class'
|\
| * 3467b0a changed the class name
* | b7ae93b added from ruby
|/
* 17f4acf first commitSharing and Updating Projects
Git doesn't have a central server like Subversion. All of the commands so far have been done locally, just updating a local database. To collaborate with other developers in Git, you have to put all that data on a server that the other developers have access to. The way Git does this is to synchronize your data with another repository. There is no real difference between a server and a client - a Git repository is a Git repository and you can synchronize between any two easily.
Once you have a Git repository, either one that you set up on your own server, or one hosted someplace like GitHub, you can tell Git to either push any data that you have that is not in the remote repository up, or you can ask Git to fetch differences down from the other repo.
You can do this any time you are online, it does not have to correspond with a
Once you have a Git repository, either one that you set up on your own server, or one hosted someplace like GitHub, you can tell Git to either push any data that you have that is not in the remote repository up, or you can ask Git to fetch differences down from the other repo.
You can do this any time you are online, it does not have to correspond with a
commit or anything else. Generally you will do a number of commits locally, then fetch data from the online shared repository you cloned the project from to get up to date, merge any new work into the stuff you did, then push your changes back up.In a nutshell you can update your project with
git fetch and share your changes with git push. You can manage your remote repositories with git remote. git remote list, add and delete remote repository aliases
Unlike centralized version control systems that have a client that is very different from a server, Git repositories are all basically equal and you simply synchronize between them. This makes it easy to have more than one remote repository - you can have some that you have read-only access to and others that you can write to as well.
So that you don't have to use the full URL of a remote repository every time you want to synchronize with it, Git stores an alias or nickname for each remote repository URL you are interested in. You use the
For example, if we want to share our Hello World program with the world, we can create a new repository on a server (I'll use GitHub as an example), which should give you a URL, in this case "git@github.com:schacon/hw.git". To add that to our project so we can push to it and fetch updates from it we would do this:
So that you don't have to use the full URL of a remote repository every time you want to synchronize with it, Git stores an alias or nickname for each remote repository URL you are interested in. You use the
git remote command to manage this list of remote repos that you care about.git remote list your remote aliases
Without any arguments, Git will simply show you the remote repository aliases that it has stored. By default, if you cloned the project (as opposed to creating a new one locally), Git will automatically add the URL of the repository that you cloned from under the name 'origin'. If you run the command with the-v option, you can see the actual URL for each alias.$ git remote origin $ git remote -v origin git@github.com:schacon/git-reference.git (fetch) origin git@github.com:schacon/git-reference.git (push)You see the URL there twice because Git allows you to have different push and fetch URLs for each remote in case you want to use different protocols for reads and writes.
git remote add add a new remote repository of your project
If you want to share a locally created repository, or you want to take contributions from someone elses repository - if you want to interact in any way with a new repository, it's generally easiest to add it as a remote. You do that by runninggit remote add [alias] [url]. That adds [url] under a local remote named [alias].For example, if we want to share our Hello World program with the world, we can create a new repository on a server (I'll use GitHub as an example), which should give you a URL, in this case "git@github.com:schacon/hw.git". To add that to our project so we can push to it and fetch updates from it we would do this:
$ git remote $ git remote add github git@github.com:schacon/hw.git $ git remote -v github git@github.com:schacon/hw.git (fetch) github git@github.com:schacon/hw.git (push)Like the branch naming, remote alias names are arbitrary - just as 'master' has no special meaning but is widely used because
git init sets it up by default, 'origin' is often used as a remote name because git clone sets it up by default as the cloned-from URL. In this case I've decided to name my remote 'github', but I could have really named it just about anything. git remote rm removing an existing remote alias
Git addeth and Git taketh away. If you need to remove a remote - you are not using it anymore, the project is gone, etc - you can remove it withgit remote rm [alias].$ git remote -v github git@github.com:schacon/hw.git (fetch) github git@github.com:schacon/hw.git (push) $ git remote add origin git://github.com/pjhyett/hw.git $ git remote -v github git@github.com:schacon/hw.git (fetch) github git@github.com:schacon/hw.git (push) origin git://github.com/pjhyett/hw.git (fetch) origin git://github.com/pjhyett/hw.git (push) $ git remote rm origin $ git remote -v github git@github.com:schacon/hw.git (fetch) github git@github.com:schacon/hw.git (push)
In a nutshell with
git remote you can list our remote repositories and whatever URL that repository is using. You can use git remote add to add new remotes and git remote rm to delete existing ones. git fetch download new branches and data from a remote repository
git pull fetch from a remote repo and try to merge into the current branch
Git has two commands to update itself from a remote repository.
The second command that will fetch down new data from a remote server is
Assuming you have a remote all set up and you want to pull in updates, you would first run
You can see the mapping that Git makes. The 'master' branch on the remote repository becomes a branch named 'github/master' locally. That way now I can merge the 'master' branch on that remote into my local 'master' branch by running
If you have more than one remote repository, you can either fetch from specific ones by running
git fetch will synchronize you with another repo, pulling down any data that you do not have locally and giving you bookmarks to where each branch on that remote was when you synchronized. These are called "remote branches" and are identical to local branches except that Git will not allow you to check them out - however, you can merge from them, diff them to other branches, run history logs on them, etc. You do all of that stuff locally after you synchronize. The second command that will fetch down new data from a remote server is
git pull. This command will basically run a git fetch immediately followed by a git merge of the branch on that remote that is tracked by whatever branch you are currently in. I personally don't much like this command - I prefer running fetch and merge seperately. Less magic, less problems. However, if you like this idea, you can read about it in more detail in the official docs. Assuming you have a remote all set up and you want to pull in updates, you would first run
git fetch [alias] to tell Git to fetch down all the data it has that you do not, then you would run git merge [alias]/[branch] to merge into your current branch anything new you see on the server (like if someone else has pushed in the meantime). So, if I were working on my Hello World project with several other people and I wanted to bring in any changes that had been pushed since I last connected, I would do something like this:$ git fetch github remote: Counting objects: 4006, done. remote: Compressing objects: 100% (1322/1322), done. remote: Total 2783 (delta 1526), reused 2587 (delta 1387) Receiving objects: 100% (2783/2783), 1.23 MiB | 10 KiB/s, done. Resolving deltas: 100% (1526/1526), completed with 387 local objects. From github.com:schacon/hw 8e29b09..c7c5a10 master -> github/master 0709fdc..d4ccf73 c-langs -> github/c-langs 6684f82..ae06d2b java -> github/java * [new branch] ada -> github/ada * [new branch] lisp -> github/lispI can see that since the last time I synchronized with this remote, five branches have been added or updated. The 'ada' and 'lisp' branches are new, where the 'master', 'c-langs' and 'java' branches have been updated. In this case, my team is pushing proposed updates to remote branches for review before they're merged into 'master'.
You can see the mapping that Git makes. The 'master' branch on the remote repository becomes a branch named 'github/master' locally. That way now I can merge the 'master' branch on that remote into my local 'master' branch by running
git merge github/master. Or, I can see what new commits are on that branch by running git log github/master ^master. If your remote is named 'origin' it would be origin/master instead. Almost any command you would run using local branches you can use remote branches with too. If you have more than one remote repository, you can either fetch from specific ones by running
git fetch [alias] or you can tell Git to synchronize with all of your remotes by running git fetch --all. In a nutshell you run
git fetch [alias] to synchronize your repository with a remote repository, fetching all the data it has that you do not into branch references locally for merging and whatnot. git push push your new branches and data to a remote repository
To share the cool commits you've done with others, you need to push your changes to the remote repository. To do this, you run
What if I have a topic branch like the 'erlang' branch we created earlier and I just want to share that? You can just push that branch instead.
The last major issue you run into with pushing to remote branches is the case of someone pushing in the meantime. If you and another developer clone at the same time, you both do commits, then she pushes and then you try to push, Git will by default not allow you to overwrite her changes. Instead, it basically runs
This is what happens when you try to push a branch to a remote branch that has been updated in the meantime:
git push [alias] [branch] which will attempt to make your [branch] the new [branch] on the [alias] remote. Let's try it by initially pushing our 'master' branch to the new 'github' remote we created earlier.$ git push github master Counting objects: 25, done. Delta compression using up to 2 threads. Compressing objects: 100% (25/25), done. Writing objects: 100% (25/25), 2.43 KiB, done. Total 25 (delta 4), reused 0 (delta 0) To git@github.com:schacon/hw.git * [new branch] master -> masterPretty easy. Now if someone clones that repository they will get exactly what I have committed and all of its history.
What if I have a topic branch like the 'erlang' branch we created earlier and I just want to share that? You can just push that branch instead.
$ git push github erlang Counting objects: 7, done. Delta compression using up to 2 threads. Compressing objects: 100% (6/6), done. Writing objects: 100% (6/6), 652 bytes, done. Total 6 (delta 1), reused 0 (delta 0) To git@github.com:schacon/hw.git * [new branch] erlang -> erlangNow when people clone or fetch from that repository, they'll get an 'erlang' branch they can look at and merge from. You can push any branch to any remote repository that you have write access to in this way. If your branch is already on the server, it will try to update it, if it is not, Git will add it.
The last major issue you run into with pushing to remote branches is the case of someone pushing in the meantime. If you and another developer clone at the same time, you both do commits, then she pushes and then you try to push, Git will by default not allow you to overwrite her changes. Instead, it basically runs
git log on the branch you're trying to push and makes sure it can see the current tip of the server's branch in your push's history. If it can't see what is on the server in your history, it concludes that you are out of date and will reject your push. You will rightly have to fetch, merge then push again - which makes sure you take her changes into account.This is what happens when you try to push a branch to a remote branch that has been updated in the meantime:
$ git push github master To git@github.com:schacon/hw.git ! [rejected] master -> master (non-fast-forward) error: failed to push some refs to 'git@github.com:schacon/hw.git' To prevent you from losing history, non-fast-forward updates were rejected Merge the remote changes before pushing again. See the 'Note about fast-forwards' section of 'git push --help' for details.You can fix this by running
git fetch github; git merge github/master and then pushing again. In a nutshell you run
git push [alias] [branch] to update a remote repository with the changes you've made locally. It will take what your [branch] looks like and push it to be [branch] on the remote, if possible. If someone else has pushed since you last fetched and merged, the Git server will deny your push until you are up to date. Inspection and Comparison
So now you have a bunch of branches that you are using for short lived topics, long lived features and what not. How do you keep track of them? Git has a couple of tools to help you figure out where work was done, what the difference between two branches are and more.
In a nutshell you can use
git log to find specific commits in your project history - by author, date, content or history. You can use git diff to compare two different points in your history - generally to see how two branches differ or what has changed from one version of your software to another. git log filter your commit history
We've already seen how to use
If I look for the commit messages with 'p4 depo' in them, I get these three commits:
git log to compare branches, by looking at the commits on one branch that are not reachable from another. (If you don't remember, it looks like this: git log branchA ^branchB). However, you can also use git log to look for specific commits. Here we'll be looking at some of the more commonly used git log options, but there are many. Take a look at the official docs for the whole list. git log --author look for only commits from a specific author
To filter your commit history to only the ones done by a specific author, you can use the--author option. For example, let's say we're looking for the commits in the Git source code done by Linus. We would type something like git log --author=Linus. The search is case sensitive and also will search the email address. I'll do the example using the -[number] option, which will limit the results to the last [number] commits. $ git log --author=Linus --oneline -5 81b50f3 Move 'builtin-*' into a 'builtin/' subdirectory 3bb7256 make "index-pack" a built-in 377d027 make "git pack-redundant" a built-in b532581 make "git unpack-file" a built-in 112dd51 make "mktag" a built-in
git log --since --before filter commits by date authored
If you want to specify a date range that you're interested in filtering your commits down to, you can use a number of options - I use--since and --before, but you can also use --until and --after. For example, if I wanted to see all the commits in the Git project before 3 weeks ago but after April 18th, I could run this (I'm also going to use --no-merges to remove merge commits): $ git log --oneline --before={3.weeks.ago} --after={2010-04-18} --no-merges
5469e2d Git 1.7.1-rc2
d43427d Documentation/remote-helpers: Fix typos and improve language
272a36b Fixup: Second argument may be any arbitrary string
b6c8d2d Documentation/remote-helpers: Add invocation section
5ce4f4e Documentation/urls: Rewrite to accomodate transport::address
00b84e9 Documentation/remote-helpers: Rewrite description
03aa87e Documentation: Describe other situations where -z affects git diff
77bc694 rebase-interactive: silence warning when no commits rewritten
636db2c t3301: add tests to use --format="%N"git log --grep filter commits by commit message
You may also want to look for commits with a certain phrase in the commit message. You can use--grep for that. Let's say I knew there was a commit that dealt with using the P4EDITOR environment variable and I wanted to remember what that change looked like - I could find the commit with --grep. $ git log --grep=P4EDITOR --no-merges
commit 82cea9ffb1c4677155e3e2996d76542502611370
Author: Shawn Bohrer
Date: Wed Mar 12 19:03:24 2008 -0500
git-p4: Use P4EDITOR environment variable when set
Perforce allows you to set the P4EDITOR environment variable to your
preferred editor for use in perforce. Since we are displaying a
perforce changelog to the user we should use it when it is defined.
Signed-off-by: Shawn Bohrer
Signed-off-by: Simon Hausmann Git will logically OR all --grep and --author arguments. If you want to use --grep and --author to see commits that were authored by someone AND have a specific message content, you have to add the --all-match option. In these examples, I'm going to use the --format option, so we can see who the author of each commit was. If I look for the commit messages with 'p4 depo' in them, I get these three commits:
$ git log --grep="p4 depo" --format="%h %an %s" ee4fd1a Junio C Hamano Merge branch 'master' of git://repo.or.cz/git/fastimport da4a660 Benjamin Sergeant git-p4 fails when cloning a p4 depo. 1cd5738 Simon Hausmann Make incremental imports easier to use by storing the p4 dIf I add a
--author=Hausmann argument, instead of further filtering it down to the one commit by Simon, it instead will show me all commits by Simon OR commits with "p4 depo" in the message:$ git log --grep="p4 depo" --format="%h %an %s" --author="Hausmann" cdc7e38 Simon Hausmann Make it possible to abort the submission of a change to Pe f5f7e4a Simon Hausmann Clean up the git-p4 documentation 30b5940 Simon Hausmann git-p4: Fix import of changesets with file deletions 4c750c0 Simon Hausmann git-p4: git-p4 submit cleanups. 0e36f2d Simon Hausmann git-p4: Removed git-p4 submit --direct. edae1e2 Simon Hausmann git-p4: Clean up git-p4 submit's log message handling. 4b61b5c Simon Hausmann git-p4: Remove --log-substitutions feature. 36ee4ee Simon Hausmann git-p4: Ensure the working directory and the index are cle e96e400 Simon Hausmann git-p4: Fix submit user-interface. 38f9f5e Simon Hausmann git-p4: Fix direct import from perforce after fetching cha 2094714 Simon Hausmann git-p4: When skipping a patch as part of "git-p4 submit" m 1ca3d71 Simon Hausmann git-p4: Added support for automatically importing newly ap ...However, if I add a
--all-match, I get the results I'm looking for:$ git log --grep="p4 depo" --format="%h %an %s" --author="Hausmann" --all-match 1cd5738 Simon Hausmann Make incremental imports easier to use by storing the p4 d
git log -S filter by introduced diff
What if you write really horrible commit messages? Or, what if you are looking for when a function was introduced, or where variables started to be used? You can also tell Git to look through the diff of each commit for a string. For example, if we wanted to find which commits modified anything that looked like the function name 'userformat_find_requirements', we would run this: (note there is no '=' between the '-S' and what you are searching for)$ git log -Suserformat_find_requirements
commit 5b16360330822527eac1fa84131d185ff784c9fb
Author: Johannes Gilger
Date: Tue Apr 13 22:31:12 2010 +0200
pretty: Initialize notes if %N is used
When using git log --pretty='%N' without an explicit --show-notes, git
would segfault. This patches fixes this behaviour by loading the needed
notes datastructures if --pretty is used and the format contains %N.
When --pretty='%N' is used together with --no-notes, %N won't be
expanded.
This is an extension to a proposed patch by Jeff King.
Signed-off-by: Johannes Gilger
Signed-off-by: Junio C Hamanogit log -p show patch introduced at each commit
Each commit is a snapshot of the project, but since each commit records the snapshot it was based off of, Git can always calculate the difference and show it to you as a patch. That means for any commit you can get the patch that commit introduced to the project. You can either do this by runninggit show [SHA] with a specific commit SHA, or you can run git log -p, which tells Git to put the patch after each commit. It is a great way to summarize what has happened on a branch or between commits. $ git log -p --no-merges -2 commit 594f90bdee4faf063ad07a4a6f503fdead3ef606 Author: Scott Chacon schacon@gmail.com Date: Fri Jun 4 15:46:55 2010 +0200 reverted to old class name diff --git a/ruby.rb b/ruby.rb index bb86f00..192151c 100644 --- a/ruby.rb +++ b/ruby.rb @@ -1,7 +1,7 @@ -class HiWorld +class HelloWorld def self.hello puts "Hello World from Ruby" end end -HiWorld.hello +HelloWorld.hello commit 3cbb6aae5c0cbd711c098e113ae436801371c95e Author: Scott Chacon schacon@gmail.com Date: Fri Jun 4 12:58:53 2010 +0200 fixed readme title differently diff --git a/README b/README index d053cc8..9103e27 100644 --- a/README +++ b/README @@ -1,4 +1,4 @@ -Hello World Examples +Many Hello World Examples ====================== This project has examples of hello world inThis is a really nice way of summarizing changes or reviewing a series of commits before merging them or releasing something.
git log --stat show diffstat of changes introduced at each commit
If the-p option is too verbose for you, you can summarize the changes with --stat instead. Here is the same log output with --stat instead of -p$ git log --stat --no-merges -2 commit 594f90bdee4faf063ad07a4a6f503fdead3ef606 Author: Scott Chacon schacon@gmail.com Date: Fri Jun 4 15:46:55 2010 +0200 reverted to old class name ruby.rb | 4 ++-- 1 files changed, 2 insertions(+), 2 deletions(-) commit 3cbb6aae5c0cbd711c098e113ae436801371c95e Author: Scott Chacon schacon@gmail.com Date: Fri Jun 4 12:58:53 2010 +0200 fixed readme title differently README | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-)Same basic information, but a little more compact - it still lets you see relative changes and which files were modified.
git diff
Finally, to see the absolute changes between any two commit snapshots, you can use the
To see what has changed since the last release, you can simply run
As a bit of an aside, you can also have git manually calculate the merge-base (first common ancestor commit) of any two commits would be with the
git diff command. This is largely used in two main situations - seeing how two branches differ from one another and seeing what has changed since a release or some other older point in history. Let's look at both of these situations.To see what has changed since the last release, you can simply run
git diff [version] (or whatever you tagged the release). For example, if we want to see what has changed in our project since the v0.9 release, we can run git diff v0.9. $ git diff v0.9 diff --git a/README b/README index d053cc8..d4173d5 100644 --- a/README +++ b/README @@ -1,4 +1,4 @@ -Hello World Examples +Many Hello World Lang Examples ====================== This project has examples of hello world in diff --git a/ruby.rb b/ruby.rb index bb86f00..192151c 100644 --- a/ruby.rb +++ b/ruby.rb @@ -1,7 +1,7 @@ -class HiWorld +class HelloWorld def self.hello puts "Hello World from Ruby" end end -HiWorld.hello +HelloWorld.helloJust like
git log, you can use the --stat option with it.$ git diff v0.9 --stat README | 2 +- ruby.rb | 4 ++-- 2 files changed, 3 insertions(+), 3 deletions(-)To compare two divergent branches, however, you can run something like
git diff branchA branchB but the problem is that it will do exactly what you are asking - it will basically give you a patch file that would turn the snapshot at the tip of branchA into the snapshot at the tip of branchB. This means if the two branches have diverged - gone in different directions - it will remove all the work that was introduced into branchA and then add everything that was introduced into branchB. This is probably not what you want - you want the changes added to branchB that are not in branchA, so you really want the difference between where the two branches diverged and the tip of branchB. So, if our history looks like this:$ git log --graph --oneline --decorate --all * 594f90b (HEAD, tag: v1.0, master) reverted to old class name | * 1834130 (erlang) added haskell | * ab5ab4c added erlang |/ * 8d585ea Merge branch 'fix_readme' ...And we want to see what is on the "erlang" branch compared to the "master" branch, running
git diff master erlang will give us the wrong thing.$ git diff --stat master erlang erlang_hw.erl | 5 +++++ haskell.hs | 4 ++++ ruby.rb | 4 ++-- 3 files changed, 11 insertions(+), 2 deletions(-)You see that it adds the erlang and haskell files, which is what we did in that branch, but then the output also reverts the changes to the ruby file that we did in the master branch. What we really want to see is just the changes that happened in the "erlang" branch (adding the two files). We can get the desired result by doing the diff from the common commit they diverged from:
$ git diff --stat 8d585ea erlang erlang_hw.erl | 5 +++++ haskell.hs | 4 ++++ 2 files changed, 9 insertions(+), 0 deletions(-)That's what we're looking for, but we don't want to have to figure out what commit the two branches diverged from every time. Luckily, Git has a shortcut for this. If you run
git diff master...erlang (with three dots in between the branch names), Git will automatically figure out what the common commit (otherwise known as the "merge base") of the two commit is and do the diff off of that.$ git diff --stat master erlang erlang_hw.erl | 5 +++++ haskell.hs | 4 ++++ ruby.rb | 4 ++-- 3 files changed, 11 insertions(+), 2 deletions(-) $ git diff --stat master...erlang erlang_hw.erl | 5 +++++ haskell.hs | 4 ++++ 2 files changed, 9 insertions(+), 0 deletions(-)Nearly every time you want to compare two branches, you'll want to use the triple-dot syntax, because it will almost always give you what you want.
As a bit of an aside, you can also have git manually calculate the merge-base (first common ancestor commit) of any two commits would be with the
git merge-base command:$ git merge-base master erlang 8d585ea6faf99facd39b55d6f6a3b3f481ad0d3dSo, you can do the equivalent of
git diff master...erlang by running this:$ git diff --stat $(git merge-base master erlang) erlang erlang_hw.erl | 5 +++++ haskell.hs | 4 ++++ 2 files changed, 9 insertions(+), 0 deletions(-)I would of course recommend using the easier syntax, though.
In a nutshell you can use
git diff to see how a project has changed since a known point in the past or to see what unique work is in one branch since it diverged from another. Always use git diff branchA...branchB to inspect branchB relative to branchA to make things easier. 
留言