This is the source code for the Boston Ruby Group website, bostonrb.org.
It is licensed under the MIT License, so if you would like to fork and customize for your own user group, you are more than welcome to.
Getting up and running
bostonrb.org is built and managed primarily the following:
- Ruby 1.8.x
- Rubygems 1.3.5
- Ruby on Rails 2.3.9
You will need these installed before you can start developing and running the application. For help getting this environment running, see these excellent installation notes
(You can try Google for documentation on MySQL installation for your operating system.)
Configure RVM and project gemset (RVM is optional, but recommended)
If necessary install Ruby 1.8.7
rvm install 1.8.7-p249
Create a gemset to minimize gem conflicts
rvm --create 1.8.7-p249@bostonrb
Make sure you use Ruby 1.8.7 and the bostonrb gemset in every terminal
rvm use 1.8.7-p249@bostonrb
Since we use some gems that are only on github (ie feedzirra), this needs to be run, if you haven't before:
gem sources -a http://gems.github.com
We're using GitHub for our git hosting. To download the source, use the following command:
git clone git://github.com/bostonrb/bostonrb.git
Before anything else, change into the newly created
Copy the sample database config file and edit as needed.
cp config/database.yml.sample config/database.yml
gem install rails -v 2.3.9
Install a couple gems to satisfy plugin requirements
gem install chronic -v 0.2.3 gem install haml -v 2.2.20 gem install factory_girl -v 1.2.2 gem install fakeweb -v 1.2.0
We need to make sure all the gem dependencies are installed. This may take several minutes:
Now we need to create an initial database. We currently assume you have MySQL installed, and that you can login as 'root' without a password.
Next migrate the database to the latest version:
At last, we can run the server:
This will keep running until you hit Control-C, but now you can open a browser and go to http://localhost:3000 to see the app running.
- If you get odd MySQL errors when you run the server, you may have a mismatch between the version of this gemset's mysql gem and your installation of MySQL. Running "rake gems:install" installs the most recent version of each gem (unless specified otherwise in config/environment.rb). If you use MySQL databases successfully in another project, check that project's version of the mysql gem. Try uninstalling the version installed in this gemset by "rake gems:install", and install the version that works for you elsewhere, e.g. "gem install mysql -v 2.7".
You'll notice the site is kind of... empty. You can generate a bit of seed data with:
We need to make sure all the gem dependencies for the test environment are installed. This may take several minutes:
rake gems:install RAILS_ENV=test
Also, you'll need to vendor and build your gems:
rake gems:unpack RAILS_ENV=test rake gems:build RAILS_ENV=test
We are using shoulda for our testing. Before running them, you need to run the following to create the test database:
rake db:create RAILS_ENV=test
At this point, you can run unit tests with:
We are also using cucumber for user stories / integration testing. Install the gem dependencies:
rake gems:install RAILS_ENV=cucumber
Vendor and build your gems:
rake gems:unpack RAILS_ENV=cucumber rake gems:build RAILS_ENV=cucumber
Remove an extraneous file, if you have it:
You can now run your cucumber tests with:
Regular scheduled tasks
Tweets and blog posts are regularly updated, based on the users in the system. You can run the following to make them update in development:
rake get_tweets rake get_blogs
For our EY deployment, there's a custom wrapper to handle logging output at /engineyard/custom/cron_wrapper, which among other things, adds logging to /data/bostonrb/current/logs. It takes four arguments: the application to run for (always will be bostonrb), the name of the log file, the rails environment, and lastly the command to run
The current crontab entries look like:
*/5 * * * * /engineyard/custom/cron_wrapper bostonrb get_tweets production 'rake get_tweets' */5 * * * * /engineyard/custom/cron_wrapper bostonrb get_blogs production 'rake get_blogs'
Tools and libraries used
Here is a list, in no particular order, of the libraries and other tools used, and how or why we're using them.
Library for generating HTML from textile markup. We use this so we don't have to have users enter HTML for descriptions of jobs, events, etc
Library for generating HTML from markdown markup. We use this so we don't have to have users enter HTML for descriptions of jobs, events, etc
A library by Josh Nichols and Dan Croak to access the GitHub API. Users enter their GitHub username when setting up an account, and we use le-git to aggregate them.
A library by Dan Croak to access the Twitter Search API. Users enter their Twitter usernames when setting up an account, and we use twitter-search to aggregate them.
A test library that lets you intercept network connections setup by [net/http], and setup canned responses for them.
We use this to make sure no network connections happen during testing.
A library by Josh Nichols that lets you easily setup add feed aggregation to a Rails app.
We use this to let users and companies enter the URL of their blog feed, so we can aggregate it.
A library for interacting with gravatar.
We use this to display user gravatars on the front page.
An extension to ActionView written by Harold Gimenez to truncate
An extension to ActiveRecord to allow for easily pagination of queries.
We usually use this on index pages, where we want to show several smaller pages of results, rather than one extremely long one.
An extension to ActiveRecord written by Harold Gimenez to provide a database asgnostic named_scope to get a random record.
We use this to
A library which can geocode an address into a latitude and longitude. We use this geocode the addresses of our events, so we can display a Google map to them in combination with Google Maps
A plugin to add an ActiveRecord validation method for emails. This is used to validate a user's email address.
A plugin to add an ActiveRecord validation method for URLs. This is used to validate a user's website URL.
You should go through the steps lined out in 'Getting up and running'.
For bugs and features requests, we use GitHub Issues. You can see what is open here: http://github.com/bostonrb/bostonrb/issues
To contribute, you should fork on GitHub, make your changes, and then either comment on an existing issue pointing at your changes, or create a new issue if it's a new bug or feature.
Dan Croak, Josh Nichols, Harold Gimenez, Dan Pickett, Rebecca Frankel, Joe Ferris.
Design by Angelo Simeoni.