The main steps for setting up a new project are:

• Create a virtual host
• Add it to /etc/hosts
• Enable the virtual host
• Import the new project to the SVN repository
• Checkout the project to /var/www
• Create a TRAC environment for the project
• Restart Apache

After these steps I have http://projectName.lh/ which points to /var/www/projectName/public/, SVN repo under http://localhost/repos/projectName/ and the TRAC environment under http://localhost/trac/projectName/.

As I have this ability to forget things, I always forget a step or 2 of this process. Thus, I wrote pywst (note, this is a txt file, to use it, save it to your HDD and rename it to pywst.py). It’s not the best and nicest Python script ever wrote, but gets the job done. All that is need to be done to setup a project with pywst is:

sudo ./pywst.py projectName

2 things are required: to run it with sudo powers and to provide a name for the project.

Future improvements

The first, and the most important is to finish the rollback() method. Now, it only exits pywst when an error occurs, but it should undo all the steps made prior to the error.

Second, to make it work on other distros, not only on Ubuntu. That would require for me getting those other distros, set them up, look where they store Apache and stuff, where’s the default document root, etc. Hmm… This will take a while :)

Third, support PHP frameworks – Zend Framework, CodeIgniter and CakePHP — ZF is a must :P Under support I mean to create the basic file structure for them automagically.

Cheers!

What is Trac?

As said on the Trac homepage:

Trac is an enhanced wiki and issue tracking system for software development projects.

It’s free, it’s open source, it comes under the BSD license and it’s really awesome. You can write a wiki with it, have a ticket system, connect it with SVN, so you can browse the sources from the browser and see all the commit messages, when was something changed, added… It can support one project, it can support multiple projects. It can be viewable/editable by anyone, or you can close it down for your little team…

Trac is big. It has lots of plug-ins, so you can extend and customize your Trac. I haven’t played with them yet, but as soon as I will, you’ll get notified ;)

It’s written in Python. It can run on it’s own server, or it can run under Apache (where there are also several options). It can use SQlite, PostrgeSQL or MySQL databases. Currently it can connect only to SVN.

I’ll show you how to setup a basic Trac 0.11-dot-something-dot-something. It will run under Apache with mod_wsgi, use a SQlite database, connect to the SVN repository and require user authentication.

Installing Trac

Before anything, I want to say that my machine where I installed Trac has LAMP and SVN configured like this. So, this post is kinda the next part of that post.

First, I installed a Python tool, called Easy Install. It’s here to make our installation process easier. Lovely. Go to http://pypi.python.org/pypi/setuptools/, scroll down to the downloads section and choose a Python egg to download (match it to your currently installed Python version — I have Python 2.5 so I downloaded “setuptools-0.6c9-py2.5.egg”).

Fire up a console and type:

sudo sh setuptools-0.6c9-py2.5.egg

Of course, you need to match this to your own setuptools file.

Next, type:

sudo easy_install Trac

EasyInstall will now locate Trac and it’s dependencies, download and install them.

Download the mod_wsgi:

sudo apt-get install libapache2-mod-wsgi

It will install and enable mod_wsgi. And, in my case, it only tried to restart Apache, but for an unknown reason it fails to do so. If that happens, just do a quick:

sudo /etc/init.d/apache2 restart

If you want Subversion with your Trac, you’ll need the python-subversion package:

sudo apt-get install python-subversion

If you have it already, it’ll just skip it. If you want SVN, but you don’t have this package, later on it will show an error message like: Unsupported version control system “svn”.

Now to make a folder for Trac, where it will keep all the Trac projects and stuff.

sudo mkdir /var/trac /var/trac/sites /var/trac/eggs /var/trac/apache
sudo chown -R www-data /var/trac

Under /var/trac/sites will be the files for Trac projects. The /var/trac/eggs folder will be used as a cache folder for Python eggs. /var/trac/apache will hold a wsgi script file.

The wsgi script is actually a Python script, but with the .wsgi extension, used by mod_wsgi. With this script, Trac will be able to run as a WSGI application.
File: /var/trac/apache/trac.wsgi

import sys
sys.stdout = sys.stderr

import os
os.environ['TRAC_ENV_PARENT_DIR'] = '/var/trac/sites'
os.environ['PYTHON_EGG_CACHE'] = '/var/trac/eggs'

import trac.web.main

application = trac.web.main.dispatch_request

With this kind of script, one single Trac installation will be able to manage multiple projects (you can see here some other scripts).

Configure Apache, add this to your httpd.conf file:

WSGIScriptAlias /trac /var/trac/apache/trac.wsgi

<Directory /var/trac/apache>
    WSGIApplicationGroup %{GLOBAL}
    Order deny,allow
    Allow from all
</Directory>

Restart Apache:

sudo /etc/init.d/apache2 restart

If you go to http://localhost/trac/ in your browser, you should see an empty list of Available Projects. It’s empty, cause we haven’t added any project yet.

Now, let’s asume that we have a project called “testProject” with it’s source located in /var/www/testProject and a SVN repo located in /var/svn/repos/testProject. I’ll show how to add that project to Trac.

In console type:

sudo trac-admin /var/trac/sites/testProject initenv

Note that you need to provide the full path to /var/trac/sites, cause it will create a Trac project in the current folder you’re in.

It will ask you now a few things:

• Project Name — the name of the project, e.g. “Trac testing project”
• Database connection string — leave it empty, and it will use SQlite
• Repository type — leave it empty, and it will use SVN
• Path to repository — path to the project repo, e.g. /var/svn/repos/testProject

It will start to print out a bunch of lines, about what is it doing. In the end you’ll get a message like “Project environment for ‘testProject’ created.” and a few more lines. One more thing. We need to add the whole project to www-data user, so it can manage the files:

sudo chown -R www-data /var/trac/sites/testProject

If you direct your browser again to http://localhost/trac/, you will now see a link for the testProject. Click it. There, a fully working basic Trac environment for your project. A wiki, a ticket/bug tracking system, a repo browser in only a few minutes. How cool is that? Very.

This Trac environment can now be accessible by everyone. If we do not want that, we need to add this to the httdp.conf file:

<Location /trac>
    AuthType Basic
    AuthName "Trac login"
    AuthUserFile /var/trac/.htpasswd
    Require valid-user
</Location>

Create the .htpasswd file:

sudo htpasswd -bcm /var/trac/.htpasswd your_username your_password

All set. You’ll now have to login to Trac to be able to work on it. As I’m the big boss on my localhost, I gave myself some super-power privileges for Trac: TRAC_ADMIN. It’s like root on *NIX.

sudo trac-admin /var/trac/sites/testProject permission add robert TRAC_ADMIN

Read more about privileges.

That would be it. With this kind of setup, for now, it’s working perfectly for me. For Trac that’s available from the whole Internet, more security measures are needed, but this is only on localhost, so this is enough for me.

Comments, thoughts, ideas?

Happy hacking!

Ubuntu 8.10 was released bout a month ago and today I wasn’t in the mood of doing any coding so I decided to try out the new Ubuntu. Once again, I’m installing it under VirtualBox (VB), cause it seems that they still haven’t fixed the bug related to the rtl8187 chipset. Oh well…

Be sure to use VB v2.x.x. (v2.0.6. is the latest now), cause it’s recognizing the correct screen resolution, not like VB v.1.6.4, whit which I had to configure manually the xorg.conf file…

Setting up LAMP

Here are the commands:

sudo apt-get install apache2
sudo apt-get install php5 libapache2-mod-php5
sudo /etc/init.d/apache2 restart
sudo apt-get install mysql-server
sudo apt-get install libapache2-mod-auth-mysql php5-mysql phpmyadmin
sudo /etc/init.d/apache2 restart
sudo a2enmod rewrite
sudo /etc/init.d/apache2 restart

If mod_rewrite doesn’t work, do the following:

sudo gvim /etc/apache2/sites-available/default

And change AllowOverride None to AllowOverride All.

Setting up SVN

I’m not gonna explain how SVN works or the terms, this is just how to set it up. If you are not familiar with versioning and Subversion, read this book: Version Control with Subversion. It’s free, available for download and contains probably everything you need to know about SVN. Be sure to learn the commands like commit, import, export, checkout, add, info, etc…

There are 2 ways for setting up SVN: as an Apache module or to use svnserve which is designed for SVN. As I already have Apache installed, the best solution is to use Apache for SVN. It’s using a module called mod_dav_svn.

The setup presented here is very basic, it has no authentication and probably is insecure, but it’s good for my needs on localhost.

The commands:

sudo apt-get install subversion
sudo a2enmod dav
sudo /etc/init.d/apache2 restart
sudo apt-get install libapache2-svn
sudo /etc/init.d/apache2 restart

Now we have all packages installed, only the configuration left.

First, I create a folder called svn under the var folder:

sudo mkdir /var/svn

Now I need to create a folder under the svn folder where all my repositories will be:

sudo svnadmin create /var/svn/repos

We use the svnadmin create command to create the repository; mkdir is not good for this.

Next, open up the httpd.conf file and add the following lines to it:

<Location /repos>
    DAV svn
    SVNPath /var/svn/repos
</Location>

I’ve seen people creating a new user and group for SVN. I think (I haven’t looked into it detailed) that’s for the authentication stuff. I did a much simpler thing: I added the ownership over /var/svn to www-data (Apache user):

sudo chown -R www-data /var/svn

This is probably a big security hole, but again: I use it only on localhost so I can live with that.

We are now ready to import a project into SVN, i.e. to add a project to the repository:

svn import -m "First import to SVN" /import/from/here/project file:///var/svn/repos/project/trunk

To start working on that project we need to checkout it:

svn checkout http://localhost/repos/project/trunk /var/www/project

Now the “project” is under SVN which should ease the development process. Since I’m using SVN I have no more backups of projects all over the place; if something goes wrong I know it’s under SVN and I can revert to any older working version of my project.

Cheers!

I’m installing Ubuntu 8.04 on VirtualBox, with windows xp as the host machine. I must do it this way, because my wireless card is having some problems with Linux, something with the drivers. The possible solution includes kernel compiling — thanks, but no thanks.

Anyway… The installation itself is no trouble, so I’ll skip that. I always keep the apt-cache from previous installations, sparing hours of updating the system… On the host I have a folder that I share between the host OS and the client OS and first I need to reach that folder, to get from it the apt-cache.

First, need to install the Guest Additions. In Virtualbox go to Devices —> Install Guest Additions. In the console run:

sudo /media/cdrom/VBoxLinuxAdditions.run

After it’s finished, we need to mount the shared folder:

sudo mount -t vboxsf name_of_the_sharing_folder /path/to/mount_point

Now, for me, this command shows some error. Here’s what I have to do:

sudo modprobe vboxfs
sudo mount -t vboxsf name_of_the_sharing_folder /path/to/mount_point

Something with some modules not being loaded into the kernel, not bothered with it really… Now I can copy the apt-cache to where it needs to be:

sudo cp -r /path/to/mount_point/apt-cache /var/cache/apt/archives

Now do the system update. If the system update includes a kernel update, you’ll have to install Guest Additions once more…

Next installing the LAMP:

sudo apt-get install apache2
sudo apt-get install php5 libapache2-mod-php5
sudo /etc/init.d/apache2 restart
sudo apt-get install mysql-server
sudo apt-get install libapache2-mod-auth-mysql php5-mysql phpmyadmin
sudo /etc/init.d/apache2 restart
sudo a2enmod rewrite
sudo /etc/init.d/apache2 restart

That should do it. But hey! mod_rewrite still doesn’t work!

sudo gvim /etc/apache2/sites-available/default

And change AllowOverride None to AllowOverride All.

There. I have a basic LAMP on Ubuntu under VirtualBox. I made a few snapshots of the VirtualBox image, in case I trash it (which probably will happen soon), so I don’t need to reinstall over again.

Now, I’m of to setup SVN…