TeamZoom by Ryan Kruizenga & Michael Lungo

---

Co-Founder, Chief Technology Officer, & Architect: Michael Lungo Cur1yJ mike@michaellungo.com

Requirements:

ruby 1.9.3
rvm
bundler
homebrew
rails 3.2.13
amazon ec2/s3 instances
wowza media services
capistrano
therubyracer
devise
cancan
postgresql (pg)
simple_form
nested_form
activemerchant
will_paginate
event_calendar
watu_table_builder
paperclip
aws-sdk
remotipart
validates_timeliness
client_side_validations
twitter-bootstrap

Repository Notes:

Add this to your ~/.gitconfig file

[branch]
  autosetuprebase = always

Standard working procedure:

1. Always maintain fresh copy of master

git fetch origin

2. Create a new branch per (features/chores/bugs)

git branch features/<feature_name>
git checkout features/<feature_name>

3. On Commit messages

Always add the PivotalTracker Story ID in the commit message [#ID]
$ git commit -m "This is my update [#41703835]"

4. Push the feature branch to github

git push origin features/<feature_name>

5. Login to github, and make a pull request from the repository to master

6. If the merge will be automatically merge with out issues, confirm the merge

7. If the merge will not do clean merge, checkout the repository to local machine and do a merge and push it directly to master

Deployment using capistrano

live is the main branch for deployment

git checkout live
git merge master
git push origin live
git checkout master
cap deploy

If there's a new migration

cap production deploy:migrate
cap production deploy:restart

Setup local development

# Installing the requirements

# RVM & Gemset
$ rvm get head && rvm reload
$ rvm install 1.9.3
$ rvm use ruby-1.9.3-p327@teamzoom --create

# Tunnels for SSL redirects to HTTP(80)
$ rvm reload
$ gem install tunnels --no-ri --no-rdoc

# RMagick
$ brew update
$ brew upgrade
$ brew doctor
$ brew remove imagemagick
$ brew install imagemagick --disable-openmp
$ cd /usr/local/Cellar/imagemagick/<latest_version>/lib
$ ln -s libMagick++-Q16.7.dylib   libMagick++.dylib
$ ln -s libMagickCore-Q16.7.dylib libMagickCore.dylib
$ ln -s libMagickWand-Q16.7.dylib libMagickWand.dylib
$ gem install rmagick --no-ri --no-rdoc

$ cp config/examples/app_config.yml config/
$ cp config/examples/database.yml config/
$ bundle install
$ rake db:create
$ rake db:migrate
$ rake db:seed
$ rake db:seed:trunklines    

Tasks Overview:

local database

We recommend using PG for your local database because that is what Heroku uses, but due to the flexibility of Rails database configurations, a SQLite, MySQL or other database would work as well. If you are using a Postgres database, your config/database.yml file might look something like this:

	development:
	  adapter: postgresql
	  host: localhost
	  database: teamzoom_dev
	  port: 5432
	  pool: 5
	  timeout: 5000

	production:
	  <<: *login
	  database: teamzoom_pro
	
	test:
	  <<: *login
	  database: teamzoom_test

For initial installation, a suggested method of loading the database schema is to do it directly instead of running through all the migrations which have different revisions over time. To create the database and load the schema in a single step, use this command:

rake db:create db:schema:load

Alternatively, you could directly import the most recent SQL file located in db/dumps directly to the database using a GUI such as Navicat

authentication/roles

The gem CanCan built by Ryan Bates controls our permissions and user authorization. User roles are defined in the models/user.rb file. In the backend these roles are stored as integers, higher numbers representing higher roles (with Admin role being the highest). It is very important to maintain the data abstraction here so that adding more roles in the future is easy and there are no conflicts in storage and representation of roles.

staging & production server

We are using Heroku to deploy to our staging environment, and Amazon EC2 to deploy to our production environment. The Heroku platform uses a Postgres database. EC2 has an instance for Posgres, Wowza, & Rails. We deploy to production with Capistrano. This required heavy customization in our config/deploy.rb script file. We are in the process of writing custom deploy scripts to each environment.

testing

Tests are written with RSpec and model objects are easily built using FactoryGirl.

seed

rake db:seed:trunklines # import new seeded trunklines from YAML file