You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Our current tests are written differently. This requires a contributor to know the 3 different interfaces, separate utilities, and why they'd write one over the other, ...
The reason you write each kind is also blurred eg. in tmp dir? programmatically builds the test case code? requires fixture on disk? requires npm packages? console.log mocks? stdout/stderr assertions? network mocking? module mocking? remote API access? separate process interop?
We should find a common abstraction for each test with a common interface, common output, common utilities and explicit configuration for what the test needs (requires module mocking? requires network mocking? requires actual remote API?) that allows us to script when+how a test runs
As a starter point, the common interface should look something like the WranglerE2ETestHelper class
The text was updated successfully, but these errors were encountered:
Our current tests are written differently. This requires a contributor to know the 3 different interfaces, separate utilities, and why they'd write one over the other, ...
The reason you write each kind is also blurred eg. in tmp dir? programmatically builds the test case code? requires fixture on disk? requires npm packages? console.log mocks? stdout/stderr assertions? network mocking? module mocking? remote API access? separate process interop?
We should find a common abstraction for each test with a common interface, common output, common utilities and explicit configuration for what the test needs (requires module mocking? requires network mocking? requires actual remote API?) that allows us to script when+how a test runs
As a starter point, the common interface should look something like the
WranglerE2ETestHelper
classThe text was updated successfully, but these errors were encountered: