Technologist

From time to time I get questions about how to check differences between two Git commits or two branches, I will answer those questions in this post.

How to check differences between two branches:

Example:
you have a dev branch and a master branch. You develop in the dev branch and would like to know if you were to merge the dev changes to master, what will those changes be:
Please note the order of the branch this states differences that will be added to master from dev.

$ git diff master dev

How to check differences between two commits:

Example:
When you do git log you will see your different commits over time, you need to check the differences between two commits.

commit eab54d0ec21a1d7e351fee4c67139ada740e7e6b
Author: John
Date: Wed Jul 8 22:18:05 2015 -0400

Add feature 2

commit b722f1650b9fb33e0990beec027c097526c61478
Author: John
Date: Wed Jul 8 22:31:18 2015 -0400

Add feature 1

// Remember the order the first argument is the point in time, and the second item is what is added to the first argument’s commit

$ git diff b722f1650b9fb33e0990beec027c097526c61478  eab54d0ec21a1d7e351fee4c67139ada740e7e6b
diff --git a/test/script.sh b/test/script.sh
index 33f9e16..cc40434 100755
--- a//test/script.sh
+++ b//test/script.sh
@@ -14,7 +14,7 @@ fi

...
-blah1
+blah2
...

How to check differences between the HEAD (current state/latest revision) and a previous commit:

$ git diff HEAD b722f1650b9fb33e0990beec027c097526c61478 
diff --git a/test/script.sh b/test/script.sh
index 33f9e16..cc40434 100755
--- a//test/script.sh
+++ b//test/script.sh
@@ -14,7 +14,7 @@ fi

...
-blah1
+blah2
...

Or you can specify the last commit or a number of commits:

// Last commit

$ git diff HEAD~1
diff --git a/test/script.sh b/test/script.sh
index 33f9e16..cc40434 100755
--- a//test/script.sh
+++ b//test/script.sh
@@ -14,7 +14,7 @@ fi

...
-blah1
+blah2
...

//Last 2 commits

$ git diff HEAD~2
diff --git a/test/script2.pp b/test/script2.pp
index 8def87e..be66b71 100644
--- a/test/script2.pp
+++ b/test/script2.pp
@@ -4,11 +4,10 @@

 class test {
-  stage { "first": before => Stage["main"] }

-  class { "repos": stage => "first" }
+  class { "test": stage => "first" }

-  class repos {
+  class test {
...
diff --git a/test/script.sh b/test/script.sh
index 33f9e16..cc40434 100755
--- a//test/script.sh
+++ b//test/script.sh
@@ -14,7 +14,7 @@ fi

...
-blah1
+blah2
...

How to check what changed since yesterday or a specific date:

git diff @{yesterday}
diff --git a/.gitignore b/.gitignore
index 4327d2e..8fa1531 100644
--- a/.gitignore
+++ b/.gitignore
@@ -1,2 +1,3 @@
-.svn
-.project
+.com_puppetlabs_geppetto_pptp_target/
+.metadata/
+Geppetto.AutoFileSystemLinked/

// OR a specific date

git diff @{2015-05-01}
diff --git a/.gitignore b/.gitignore
index 4327d2e..8fa1531 100644
--- a/.gitignore
+++ b/.gitignore
@@ -1,2 +1,3 @@
-.svn
-.project
+.com_puppetlabs_geppetto_pptp_target/
+.metadata/
+Geppetto.AutoFileSystemLinked/

How to check all changes to a specific file:

$ git log --follow /path/to/file
commit eab54d0ec21a1d7e351fee4c67139ada740e7e6b
Author: John 
Date:   Wed Jul 8 22:18:05 2015 -0400

    Add feature 2

commit b722f1650b9fb33e0990beec027c097526c61478
Author: John 
Date:   Wed Jul 8 22:31:18 2015 -0400

    Add feature 1

How to check size differences between two trees

I found this one in stackoverflow and has been very useful:

Reference Source: http://stackoverflow.com/questions/10845051/git-show-total-file-size-difference-between-two-commits/10847242#10847242

“git cat-file -s will output the size in bytes of an object in git. git diff-tree can tell you the differences between one tree and another. Putting this together into a script called git-file-size-diff located somewhere on your PATH will give you the ability to call git file-size-diff . Putting this together we can try something like the following” – Stackoverflow

$ cat /usr/local/bin/git-file-size-diff
#!/bin/sh
. git sh-setup
args=$(git rev-parse --sq "$@")
eval "git diff-tree -r $args" | {
  total=0
  while read A B C D M P
  do
    case $M in
      M) bytes=$(( $(git cat-file -s $D) - $(git cat-file -s $C) )) ;;
      A) bytes=$(git cat-file -s $D) ;;
      D) bytes=-$(git cat-file -s $C) ;;
      *)
        echo >&2 warning: unhandled mode $M in \"$A $B $C $D $M $P\"
        continue
        ;;
    esac
    total=$(( $total + $bytes ))
    printf '%d\t%s\n' $bytes "$P"
  done
  echo total $total Bytes
  echo total $(($total/1000)) KBytes
}

// Using it to check size differences between two branches:

$ git file-size-diff master dev
66	.gitignore
239	modules/test/Modulefile
171	modules/test/README
167	modules/test/files/file.txt
1425	modules/test/manifests/init.pp
total 2068 Bytes
total 2 KBytes

// Using it to check size differences between the local master branch and the remote master branch, this is useful to know how much data will be downloaded when doing a ‘git pull’:

$ git file-size-diff master origin/master
633	modules/test/manifests/config.pp
326	modules/test/manifests/service.pp
60	modules/test/manifests/params.pp
573	modules/test/manifests/install.pp
13	modules/test/manifests/init.pp
86	modules/mod/manifests/params.pp
total 1691 Bytes
total 1 KBytes

“Git is a free & open source, distributed version control system designed to handle everything from small to very large projects with speed and efficiency.” –http://git-scm.com/

In this guide I will walk you through setting up a Git server and accessing it from a Git client over SSH.

Git Server
I am using CentOS 5.3 as my server.
Hostname: gitserver.example.com
SSH Port: 22444 (As opposed to the default port 22, for increased security)

Install Git

yum install git

Create a folder where you will keep the repositories

mkdir /opt/git

Create a git user and change the owership and permissions of the previously created folder to the new user

useradd -c “Git Repository” git
chown git:git git:git /opt/git/
chmod 770 /opt/git

Create an empty project, I am calling my project ‘myproject’

mkdir /opt/git/myproject.git

Initialize the repository using –bare to only include objects at the server side

cd /opt/git/myproject.git
git –bare init

You are done with the Git server, let’s take a look at the client

Git Client
Install Git on your client.

Debian/Ubuntu:
sudo apt-get install git-core

Red Hat/Centos:
yum install git (I am using the rpmforge repo)

Now it’s time to version-control your project.

Create a new folder to put your code (unless it exists already)

mkdir /home/john/myproject

Now it is time to add your project to Git (locally):

cd /home/john/myproject
git init

Add all the files in the current directory to be source controlled:
git add *

Commit the changes:
git commit -m “Myproject first commit”

Tell the Git client where to find the server and remote repository

git remote add origin ssh://git@gitserver.example.com:22444/opt/git/myproject.git

Push your project to the server

If you have shared SSH keys:
git push origin master

OR If you dont have configured SSH keys:
git push ssh://git@gitserver.example.com:22444/opt/git/myproject.git master

Now your project is under Git source control

Pull/Clone project:
The below will download the project folder in the current folder:

git clone ssh://git@gitserver.example.com:22444/opt/git/myproject.git

When you make changes to your project, you need to tell Git about it and commit the changes:

cd /home/john/myproject
git add *
git status (Check status)
git commit -a -m “A comment describing the change”