It's no secret that Subversion was developed to replace CVS, and it's for this reason that both applications are so similar to use. For the client, many of the commands are identical, and you can often get away with replacing cvs with svn on the command line. It's the background processing, the server and the network protocols that are different, and that's largly transparent to the typical user. Here is a list of some of the commands that are slightly different:
cvs admin svnadmin Access the administrative front end cvs annotate svn blame Show revision information for files cvs co -j svn merge Merge two sources into a working copy cvs history svn log Outputs the log messages cvs init svnadmin create Create a new repository cvs rdiff svn diff Generate list of differences for making a patch cvs remove svn delete Removes an entry from the repository cvs rtag svn copy Copies an item including its revision history
The best way of getting to grips with Subversion is through its excellent online documentation. Rather than relying on the traditional man of help commands, svn provides an excellent overview of all its internal commands by executing svn help from the command line. This provides a list of svn's internal commands which can themselves be queried by adding them to the end of the help command:
$ svn help co
There is also an excellent online book entitled 'Version Control with Subversion' that can be either browsed online, or downloaded as a PDF or HTML document: http://svnbook.red-bean.com/
documented on: 2007-09-22
The following sections teach you how to use different commands in Subversion.
To create a new Subversion archive, type the following:
$ cd ~/your-project # go to your source directory $ svn import http://localhost/repos your-project \ project-name -m "initial project import"
This creates a directory named project-name in your Subversion repository which contains your project files. Look at http://localhost/repos/ to see if it's there.
Working with project-y using Subversion:
$ cd # move to the work area
$ svn co http://localhost/repos/project-y # Check out sources
$ cd project-y
... do some work ...
$ svn diff # similar to diff -u repository/ local/
$ svn revert modified_file # undo changes to a file
$ svn ci -m "Describe changes" # check in your changes to the repository
$ vi newfile_added
$ svn add newfile_added
$ svn add new_dir # recursively add all files in new_dir
$ svn add -N new_dir2 # nonrecursively add the directory
$ svn ci -m "Added newfile_added, new_dir, new_dir2"
$ svn up # merge in latest version from repository
$ svn log # shows all changes committed
$ svn copy http://localhost/repos/project-y \
http://localhost/repos/project-y-branch \
-m "creating my branch of project-y" # branching project-y
$ svn copy http://localhost/repos/project-y \
http://localhost/repos/proj-y_release1.0 \
-m "project-y 1.0 release" # added release tag
... note that branching and tagging are the same. The only difference
... is that branches get committed whereas tags do not.
... make changes to branch ...
$ # merge branched copy back to main copy $ svn merge http://localhost/repos/project-y \ http://localhost/repos/project-y-branch $ svn co -r 4 http://localhost/repos/project-y # get revision 4
documented on: 2007-09-21
April 17th, 2007 mysurface
The aim of this tutorial is to guide beginners for using svn command line with simple examples.
This post is not going to focus on svn installation, as the installation is available anywhere, let me list some links for you in case you are actually looking for installation.
Subversion Installation References
Installing Subversion (svn) on Linux (Debian Stable) Setting up Subversion and websvn on Debian How To Configure Web Access To Subversion Repositories Using Apache Install SVN with Web Access on Ubuntu
If you are looking for svn reference in man pages, you have gone to the wrong place. To check the references of svn commands, simple do this:
svn help
This will make svn list all the available functions, to get the function reference, let say checkout
svn help checkout
The same thing goes to other svn related commands, such as svnadmin
svnadmin help
First of all what is repository? It is a core file for svn, or you can call it a centralized svn backup database. After created it, it is just a directory with its files. IMPORTANT! Do NOT try to modify or add something into the repository, unless you know what are you doing.
To create a svn repo, let say I wanna create a repo to store all my programming codes, I do this
svnadmin create /home/mysurface/repo/programming_repo
Remember try to use absolute path for everything, sometimes the relative path is not going to work.
svn import /home/mysurface/programming file:///home/mysurface/repo/programming_repo -m "Initial import"
-m stand for log message, the first revision was created with log as "Initial import". You need to specified URL for the repo, URL is the standard argument for svn. Therefore for local file, you need to specified with file://
svn list file:///home/mysurface/repo/programming_repo
Another way of listing all the files and folder in the tree view, I use svnlook
svnlook tree programming_repo
The difference between svn list and svnlook tree is one expect URL another one do not.
Part 2 will covers how to checkout, track changes, commit, add or delete files and message logs.
This is the most critical part of svn and also the most common part of svn command line. A lots of open source development projects provided the way for user to check out their latest code through the internet.
You need to check out in order to commit the changes to svn repo later. Refers back to the previous post, where I import entire directory /home/mysurface/programming to programming_repo. I am going to checkout to the same folder. If you are skeptical of doing this, you may want to backup the directory first.
mv programming programming-bk
Now checkout to programming, mkdir is not needed, as svn will create the directory for you if it is doesn't exist.
svn co file:///home/mysurface/repo/programming_repo programming
co is the shortform of checkout.
Okay, lets just compare both folder with diff and store the result into a file comp.diff
diff programming programming-bk > comp.diff
Diff will list the folder in common, and also the differences. Check comp.diff, as it tracks the additional folder .svn that only exist in programming/. Again, do NOT modified or delete this folder.
Are you convinced to remove your programming-bk/ ? Make sure you keep the repo safe and you can check out the same data anytime, at any place.
You can even checkout only a specific folder from your repo. e.g.
svn co file:///home/mysurface/repo/programming_repo/c/curses
This will only check out a folder at current directory.
Single file can't be checkout like directories, but you can extract them from repository by svn export
svn export file:///home/mysurface/repo/programming_repo/c/curses/matrix.cc
First of all, you track what files had changed,
svn status
It will list files which have changed, with some attributes besides the filename. Common attributes are M, ?, A … M is modified, A is newly added (how to add refers later section), ? indicate the file is added into local directory but not added into repo.
Secondly, you want to track the differences between the previous revision and the working one. Lets assume color.c has changed,
svn diff color.c
I really don't like svn diff 's result. Fortunately, I found a simple bash script what makes vimdiff as the compare tool. The script was written by Erik C. Thauvin, you can get it from here. http://vc.thauvin.net/svn/linux/svndiff/svndiff.sh?view=markup
I name it as svndiff and place it at /usr/bin, change the mode to executable.
chmod +x /usr/bin/svndiff
Now, I can simply do this,
svndiff color.c
To close the vimdiff, type :qa
You can commit with -m to place your log message if it is short. But if it is long, I suggest you to make use of your default editor. I am a vim user, therefore I add a line into my ~/.bashrc
EDITOR=vim
Now I can commit with this:
svn ci
ci is the shortform of checkin as in "check in", or commit. Write the log message and close save vim :x , I am done. The same way as checkout, you can choose to commit one file or any folder.
The file won't be committed if you don't add it into repo. Therefore you need to add it manually if you want it to goes into your repo. Let say you wanna add a new file color2.cc
svn add color2.cc
Delete does the same way, if you only delete file at your working directory, it won't reflects the changes to our repo.
The simplest way is doing just,
svn log
It will list all logs, start from latest revision. That is really irritating! You can limit it to 3 latest revision log by doing this
svn log --limit 3
If you wanna check for specific revision, specified with -r,
svn log -r 3
I find something awkward, let say I have done svn delete at revision 3 (latest), and revision 2 is the changes of the deleted file at revision 3. When I do svn log, by right it should show all 3 logs, but It only shows for revision 1. It means the svn log will only shows the log if the file is exist, bear in mind.
svn update
Update to specific revision?
svn update -r 3
I think thats all for normal use of svn commands, further reading at http://svnbook.red-bean.com/.
*Tags*: svn tutorial, subversion, svn howto, source version control
documented on: 2007-09-21
From http://lena.franken.de/linux/subversion.html By Bavaria.
Edited by xpt on 2007-09-21.
Subversion is a file revision control system for developers.
Repository is the place where the versioned data lives in a database. Go to the computer where the repository lives. Be the user who is repsonsible for the repository.
$ svnadmin create /data/subversion
develop$ svn import findus4 file:///data/subversion/findus4
develop$ svn import findus4 http://svn.rosi13.de/svn/findus4
Creating a trunk-directory, initial import, checkout:
develop$ svn mkdir http://svn.rosi13.de/svn/findus4/trunk develop$ cd findus4 develop$ svn import -m "initial import" http://svn.rosi13.de/svn/findus4/trunk develop$ mv findus4 findus4.old develop$ mkdir findus4 develop$ svn checkout http://svn.rosi13.de/svn/findus4/trunk findus4 develop$ rm rf findus4.old
Checkout, getting out files from repository
svn checkout file:///data/subversion/findus4 svn checkout http://subversion.rosi13.de/subversion/test
Checkout, getting out files from repository when you are *behind a proxy-server*
So, you are behind a proxyserver and want to access the repository through this proxyserver.
Configure svn to use a proxy in the file ~/.subversion/servers which then should look like this:
[global] http-proxy-host = 127.0.0.1 http-proxy-port = 3128 http-proxy-compression = no
Configure your Squidserver to allow some commands with this line in /etc/squid.conf:
extension_methods REPORT MERGE MKACTIVITY CHECKOUT
Do you initial checkout:
svn checkout http://subversion.rosi13.de/subversion/test test
Putting changed file to repository
You don't need the place of the repository anymore, svn remembers it.
svn commit
Which files have changed? Are there new files?
svn status
What has changed?
svn diff
New file in tree
svn add MyNewFile.pm (schedules file to next update) svn commit (puts diffs and this new file to repos)
Ignore files in a directory
Before this do an 'svn commit'
Do
svn propedit svn:ignore .
Every file to be ignored in a new line
Do
svn commit -m "ignoring this and that"
Moving projects from a common used repository to several single repositories
Dump complete subversion-repository to a file:
$ svnadmin dump /data/subversion >complete.svn.dump Create an extract from your first project: $ cat complete.svn.dump | svndumpfilter \ --drop-empty-revs \ --renumber-revs \ include project007 \ >project007.svn.dump
Create new repository:
$ svnadmin create /data/subversion/project007
Import complete history to new repository
$svnadmin load /data/subversion/project007 <project007.svn.dump
Activate creating keywords
svn propset svn:keywords "Revision Date Id HeadURL Author" myfile.pm
Install these package in debian-linux on your subversion-server: libsvn-notify-perl - Subversion activity notification.
% apt-get install libsvn-notify-perl
Check:
% svnnotify Missing required "--repos-path" parameter
If you get this error-message then you are ready to run this script when commiting code. To do this write a command into the shell-script post-commit, which is already there in your repository-directory:
cd /data/svn/findus/hooks/ ls -l
cat > post-commit
#!/bin/sh
svnnotify --repos-path "$1" --revision "$2" \
--to one@dontcare.com,two@dontcare.com \
--reply-to responsible-for-everything@somewhere.com \
--language de
You get a mail like this after commit code:
Revision: 329
Author: horshack
Date: 2006-01-31 22:27:05 +0100 (Tue, 31 Jan 2006)
Log Message:
-----------
this is the commit-message bla bla
Modified Paths:
--------------
trunk/findus_mini.ini
Your ~/.ssh/authorized_keys file has to look like this (everything must be in a single line):
command="/usr/bin/svnserve --tunnel --root=/var/svn",no-port-forwarding,no-agent-forwarding,no-X11-forwarding,no-pty ssh-dss AAAAB3N...
This makes that whatever program the user tries to run via ssh, he will always run
/usr/bin/svnserver --tunnel...
The user can only access svn-repositories in /var/svn.
When you have a repository /data/svn/findus/trunk he has to do this:
svn checkout svn+ssh://username@svn.juhei.org/data/svn/findus/trunk findus
documented on: 2007-09-21
