A restful web service for the Saccharomyces Genome Database (SGD) NEX 2, as well as an authenticated curation interface.

SGD API documentation can be found at https://github.com/yeastgenome/SGDBackend-Nex2/blob/master/docs/webservice.MD.

Prerequisites: node.js 6.0.0+, Python 3 (to simplify Python setup, virtualenv is highly suggested), and redis.

Make sure you have the needed environmental variables configured in dev_variables.sh, then

$ make build

If npm install fails with:

npm ERR! ERESOLVE unable to resolve dependency tree

and your npm install is newer (this happens at least for 7.18.1) then running again with npm install --legacy-peer-deps should work.

Lastly, npm run build may emit error messages, but everything will continue fine. Ticket for existing errors here: https://redmine.yeastgenome.org/issues/6040

To run locally you must have a running instance of Redis. On Ubuntu style Linux systems after installing with apt-get install redis a redis service should have been started.

This can be turned on and off with $ service redis start|stop. Running a redis instance directly is also possible by $ redis-server as long as the redis service is stopped first.

Ensure also that dev_variables.sh is present in the Makefile directory.

$ make run

Running locally will start the backend web service which will point to dev databases. In order to access data in the database with some endpoints (like /annotations which rely on being logged in) you'll need a user in the databases.

Navigating a browser to http://localhost:6543/ will show the home login screen. If you have a database user you can then log in.

Not all endpoints require credentials, however. As a test to prove you're getting real data, you can go to http://localhost:6543/complex/CPX-863 which should display JSON.

Be sure to have a test_variables.sh file to configure test environemntal variables.

$ make tests

This command runs just the Python tests. To run just the JavaScript tests use

$ npm test

or

$ make npm-tests

Caching uses varnish. To rebuild the cache, run

$ source /data/envs/sgd/bin/activate && source prod_variables.sh && python src/loading/scrapy/pages/spiders/pages_spider.py

using environmental variable CACHE_URLS, a comma-sepratated list of varshish host URLs (or a single one) such as http://locahost:5000.