README

Non-Implemented Models

Mapper

• topic.rb
• variable.rb
• map.rb
• link.rb
• dataset.rb
• instruments_datasets.rb

Model Relationship Diagram

Run rails generate_erd to regenerate (must have graphvis).

Configuration

• Ruby: 2.2.3
• Rails: >=5.0.0
• Postgres: >=9.4
• Redis: 3.x

Deployment

Currently Archivist has only been designed to be deployed to Heroku, but rolling your own deployment should not be too difficult.

Some basic steps

1. git clone -b master --single-branch https://github.com/CLOSER-Cohorts/archivist.git
2. cd archivist
3. gem install bundler
4. bundler install
5. Set the environment variable RAILS_ENV=production (e.g. export RAILS_ENV=production)
6. Copy config/application.yml.dist to config/application.yml
7. Set the mailer, CDN and database password in the above file
8. rails db:migrate
9. Start server: bundle exec puma -C config/puma.rb

This could cause the webserver to start...

Testing

To run the test suite just call rails test. Currently 196 tests and 263 assertions.

Stats

Name	Lines	LOC	Classes	Methods	M/C	LOC/M
Controllers	605	358	27	28	1	10
Jobs	41	38	3	3	1	10
Models	712	614	27	50	1	10
Mailers	12	11	2	1	0	9
Javascripts	2677	2296	0	305	0	5
Libraries	2001	1734	8	88	11	17
Tasks	209	163	0	1	0	161
Controller tests	835	684	18	93	5	5
Model tests	664	488	27	104	3	2
Mailer tests	11	5	2	0	0	0
Total	7767	6391	114	673	5	7

• Code LOC: 5214
• Test LOC: 1177
• Code to Test Ratio: 1:0.2

Archivist Realtime

Archivist is both a module within Archivist and an entirely separate webapp. Archivist-realtime is very slim Node.js app that provides a communications channel between the Redis cache and all active users via websockets (Socket.io). The purpose of this is to be able to lock objects while editing them and to update a users screen as changes are made to an instrument.