Closed mgramigna closed 1 year ago
St.:grey_question: |
Category | Percentage | Covered / Total |
---|---|---|---|
🟢 | Statements | 85.71% | 1938/2261 |
🟡 | Branches | 73.75% | 1711/2320 |
🟢 | Functions | 88.72% | 346/390 |
🟢 | Lines | 85.96% | 1868/2173 |
349 tests passing in 28 suites.
Report generated by 🧪jest coverage report action from c90865d6cd1723bde510ed086c37dfb48e43ec8f
One thing I just realized is that re-building all the test data will generate new UUIDs on the bundles even if the underlying CQL has not changed.
I am going to look into ensuring that bundles are re-generated only if their cql has changed
Resolved in https://github.com/projecttacoma/fqm-execution/pull/175/commits/d4b84f1a9a34a940b1920be3ec78014bb080157f
Summary
Adds necessary scripts and setup for the new integration testing approach. Gets rid of old approach and sets up structure for new one.
New behavior
None in the actual codebase.
Main new behavior to call out is new scripts:
npm run build:test-data
: Builds all measure bundles intest/integration
npm run test:integration
: Runs all the.test.ts
files inside thetest/integration
directory/subdirectoriesCode changes
test
totest/unit
to keep things clean (updated import relative paths as well)test/integration
directory with a placeholder forproportion-boolean
measureecqm-bundler
integration
step of github actions yml to reflect new scriptTesting guidance
Check out the instructions in the integration test README and make sure it's logical. You can even try following the steps for creating a new one and seeing if it works.
Try out
npm run build:test-data
andnpm run test:integration
Visually inspect everything in
test/integration