Testing Patterns (Architecture Angle)

ExtremePackaging-specific testing concerns: per-source-target test targets and cross-target test doubles. Full Swift Testing rules live in testing.md (the top-level rule). Both files state the source-target/test-target pairing rule by deliberate cross-context dual-homing.

Full Swift Testing rules live in testing.md. Two ExtremePackaging-specific points:

Per-source-target test target. Every source target ships with a matching test target. Run a single target's tests in isolation:

cd <active-package-root>
swift test --filter <PackageName>Tests

Folder layout: Sources/<Package>/ and Tests/<Package>Tests/ in package repos; Packages/Sources/<Package>/ and Packages/Tests/<Package>Tests/ in project monorepos.

Cross-target test doubles. Mocks live in the test target, not the source target. Public protocols can be defined in the source target; their mock implementations stay in Tests/<Target>Tests/Mocks/ or Packages/Tests/<Target>Tests/Mocks/, matching repository shape. Never publish mocks from Sources/ (they leak into production binaries), except in a declared test-support target that the gate checks (testing.md, section "Exception: test-support targets").

// Sources/ApiClient/APIClientProtocol.swift
public protocol APIClientProtocol {
    func login(email: String, password: String) async throws -> User
}

// Tests/ApiClientTests/Mocks/MockAPIClient.swift
public struct MockAPIClient: APIClientProtocol { /* ... */ }

Acceptance check

Compare every source target in Package.swift with the test targets: each source target MUST have exactly one <SourceTarget>Tests partner that depends on it. Grep production Sources/ for test-double names and confirm mocks live only under the matching test target. Run each new target's filtered tests once, then run the full suite and cite the total count.

1.9 KiB Raw Blame History

Testing Patterns (Architecture Angle)

Acceptance check

1.9 KiB

Raw Blame History