svn and nano Issues

Question

I'm trying to configure svn to run locally on my machine (Ubuntu) and am having some problems. Following the instructions - verbatim - from the official svn-redbooks tutorial is causing me headaches.

This is just a plain-jane VCS setup, with svn and svnadmin installed locally at /user/lib and my intended working copy (though I have not specified/imported it as a working copy) located at /home/myUser/myProj. My svn repository will be located at /opt/sandbox/svn

I want my repository to be capable of storing multiple projects, each of them implementing the standard "trunk, branches, tags" directories.

Also, I don't need any security/auth config since this is just version control for my own projects on my local machine.

So I'm guessing the logical flow of setting this up would be:
(1) Create the repository at /opt/sandbox/svn
(2) Add the myProj project directory into the repository
(3) Add trunk/, branches/ and tags/ directories inside of myProj
(4) Specify /home/myUser/myProj to be the working copy of /opt/sandbox/svn/myProj/trunk
(5) Add /home/myUser/myProj into the /trunk

On a side note, don't worry about the logic of setting this up for the trunk vs. branches, etc. I'm just looking for someone to demo how source control is set up for a project in svn, and I'll worry about my internal "version control process" (pushing/pulling from whichever directory, etc.) later on!

For the life of me I can't figure out how to do this. Plus, whenever I run any commands using file:/// (I have to since it's a local svn), instead of the command executing the terminal just opens up an instance of nano like it wants me to edit something.

So can someone shed some light as to:
(1) Whether or not the logical flow I specified above is correct, and if not what needs to change; and
(2) What terminal/svn commands I need to run to implement the correct logical flow; and
(3) What is going on with that nano business?

Thanks!!!

Are you writing commit messages when you commit things? I've been prompted with a nano window to write a commit message when I try to commit without one before. — girasquid
How far along in your series of 5 steps did you get? What command did you run that caused nano to start? — slowdog
repository data does not belong in /opt, which is for add-on software. Consider putting it in /var/svn/sandbox or in your home directory instead. — Wim Coenen

vinnyjames vinnyjames · Accepted Answer · 2011-01-05T18:36:49

First off, I wouldn't call the repository sandbox. This term is usually synonymous with working copy. Maybe /opt/repository/svn The less typing the better for me so I'd do /opt/repo

To avoid launching the editor use:

svn commit -m "my commit message" somefiles.*

or to commit everything in shorthand

cd working_copy
svn ci -m ""

1) logical flow is good

2) inital layout commands

svn co file:///opt/repository/svn
cd svn
mkdir -p trunk/myProj branches/myProj tags/myProj
svn add *
svn ci -m "initial layout"

wipeout top level checkout

rm -rf trunk branches tags

create your project

svn co file:///opt/repository/svn/trunk/myProj
cd myProj
svn add *
svn ci -m "initial project layout"

3) nano fix = commit -m

svn and nano Issues

1 Answers