FIXME

• install Clojure

brew install clojure/tools/clojure

• install Leiningen https://leiningen.org/#install

• clone this repo

git clone git+ssh://[email protected]/precisely/pia-server.git```

• configure secrets:
  • copy .env.sample to .env, update any auth tokens & secrets
  • NOTE: the pia-developer user in our AWS dev-precisely account has minimal permissions for retrieving the longterm library from our S3 bucket
  • env.edn is in .gitignore

brew install postgresql

createdb pia_runstore
createdb test_pia_runstore
createdb hl7
createdb test_hl7

Migrations should run automatically on PIA server startup, but can also manually run like this:

lein run -m scripts/create-db

• Install Cursive
• Create Configuration "Clojure REPL" (if it doesn't exist)
  • Type: nREPL, How to Run it: Run with Leiningen
• Create a REPL command: Tools > REPL > Add New REPL Command
  • Name it "Run All Tests" and add the following code

(letfn [(list-sources [test-dir]
          (map #(clojure.string/replace (second (re-matches (re-pattern (str test-dir "/(.*)\\.clj?$")) (.getPath %))) "/" ".")
               (filter #(.isFile %) (file-seq (java.io.File. test-dir)))))]

  (run! #(require (symbol %))
        (list-sources "test"))
    (clojure.test/run-all-tests #"pia-server.*test.*"))

Recommended for development. Allows hot reloading code with your development environment.

(start)

With this approach, you will need to stop and restart the server if you change any files in pia-server, but you avoid having to start a REPL. This might be useful if you want to start pia-server from a script.

lein run

Point your browser at http://localhost:8080 for a peak at the Swagger interface.

The API is available at /api

POST /api/runs/start/{flow} Body is a JSON array of arguments Returns a run

POST /api/runs/continue/{uuid} Body is a JSON object with input and permit keys Returns a run

GET /api/runs/{uuid} Returns a run

GET /api/runs/find Returns an array of Runs * Query one or more keys /api/runs/find?state=running - active runs * Limit results /api/runs/find?state=running&limit=3 * Query nested keys using dot-separated keys. /api/runs/find?index.patient-id=123 - all runs involving patient 123 * Test for text in a JSON array by appending $ to the key name. /api/runs/find?index.roles$=doctor&index.patient-id=123&state=running - the current doctor activities for patient 123

lein test

lein do clean, ring uberjar
java -jar target/server.jar

lein ring uberwar