Department of Electronic Systems

Subversion (aka. SVN) is a version control system like CVS. SVN is intended to replace CVS, so if you are familiar with CVS, you will feel right at home with Subversion. It has some new features which is missing in CVS e.g. rename of files and handling of binary files. More information about SVN is avaliable on

A complete book for SVN is avaliable here:

Creating a repository

A repository is a shared place where SVN can store a database for the verion control information and the data.

To create a repository it is important that you select the correct database format or you will get into trouble because you are using a network filesystem on KOM.

  1. Goto your Private folder in your group directory e.g. /afs/
  2. Use svnadmin to create the repository (Further information of how the repository is avaliable can be read in the SVN book):
    svnadmin create --fs-type fsfs repos
  3. Done

Checkout a working copy

When the repository is created all members have to checkout a working copy of the repository.

To do this you may choose a graphical client e.g. TortoiseSVN

or the commandline interface:

  1. Goto the directory where you what the svn checkout dir to be
  2. Type:
     svn co file:///afs/[groupname]/svn/
    where [groupname] is your group directory (file:///// in Windows)
  3. This will create a dir called svn and you are now ready to use SVN.  You should goto and read the manual howto use SVN in daily use.

Where to put your working copy

Sometimes it is not obvious where to checkout your working copy. Above we suggest that you do it in your home directory, but is that always a good idea?

On a laptop that may not always be connected to a network, it might be a bad idea for you to only have a working copy in your AFS home directory, because you lose the ability to work off-line. In that case you might prefer to checkout a working copy on the local disk (whereever you prefer; it makes no difference except to you).

If, however, you have a stationary computer that is always connected to the network, you might as well always run an AFS client and have your working copy in your AFS home directory. The best reason to do otherwise would be to not lose the ability to work off-line in the case of a network failure, but even then AFS allows you to keep working from your cache.

Fortunately, this doesn't have to be an either-or proposition: you can create as many working copies as you like, whereever you like. You only need to remember to keep them updated before you start working on them.