To migrate from the data in the old site run
$ ./bin/migrate
This script assumes the following files will existing. These are not shared publicly at the moment.
lib
├── data
│ ├── db-person
│ ├── db-sponsor
│ ├── db.sql.erb
│ ├── migrate.sql
│ ├── migration.rb
│ └── photos
./bin/migrate drops the current development database, sets it up, loads the
old data from a sql file. The sql file is processed by ERB to dynamically add
data for columns with null constraints. After the db is loaded, using rails r,
the images are loaded from db-person, db-sponsor, and photos. This script
looks for a record with the corresponding id as the file name and updates the
record. On save, the image is processed by paperclip and stored in
public/system.
The app can also be deployed to a VM on anyone of the cloud providers using Capistrano. The easiest way to setup a VM is to follow the GoRails Guide. If you use CentOS or RHEL, you'll need to find a way to workaround SELinux. Passenger doesn't play nice with it.
Add gem 'aws-sdk-v1 to the gemfile and add the following snippet to the
paperclip configuration block in config/environments/production.rb:
:storage => :s3,
:s3_credentials => {
:bucket => ENV['S3_BUCKET_NAME'],
:access_key_id => ENV['AWS_ACCESS_KEY_ID'],
:secret_access_key => ENV['AWS_SECRET_ACCESS_KEY']
}
Don't forget to remove ':storage => :filesytem'. You can only have one entry
Running
$ ./bin/heroku-sync
copies the development database to Heroku and syncs the images in
public/system with AWS. The AWS sync requires you to have the awscli
installed and configured with your bucket:
S3_BUCKET_NAME
AWS_ACCESS_KEY_ID
AWS_SECRET_ACCESS_KEY