Unit Test Generation

Purpose

Generate comprehensive test suites validating correctness, edge cases, and iterator contracts across Java, C++, and Rust.

Test Strategy

1. Contract Tests: Verify iterator protocol (hasNext/next consistency)
2. Correctness Tests: Validate output matches expected for known inputs
3. Edge Cases: Empty, single element, duplicates, large inputs
4. Property Tests: Invariants hold for random inputs (fuzzing)
5. Performance Tests: Sanity check (not benchmarks, just "fast enough")

Framework Usage

Java (JUnit 5)

@ParameterizedTest
@MethodSource("provideTestCases")
void testMerge(List<List<Integer>> inputs, List<Integer> expected) {
    // Test implementation
}

@Test
void testEmptyIterators() {
    // Edge case
}

C++ (Catch2)

TEST_CASE("Merge iterator combines sorted sequences", "[merge]") {
    SECTION("empty input") {
        // Edge case
    }

    SECTION("single iterator") {
        // Edge case
    }
}

Rust (Built-in + proptest)

#[test]
fn merge_empty_iterators() {
    // Edge case
}

#[test]
fn merge_maintains_order() {
    // Correctness
}

proptest! {
    #[test]
    fn merge_property(inputs: Vec<Vec<i32>>) {
        // Property-based test
    }
}

Test Categories

1. Iterator Contract Tests

• hasNext() true ⟹ next() succeeds
• hasNext() false ⟹ next() throws/panics
• Multiple hasNext() calls without next() consistent

2. Correctness Tests

• Merge [1,3,5] + [2,4,6] → [1,2,3,4,5,6]
• Flat map [1,2] → [[10,11], [20,21]] → [10,11,20,21]
• Filter even [1,2,3,4] → [2,4]

3. Edge Cases

• Empty input: [] → []
• Single element: [1] → [1]
• All duplicates: [5,5,5] → [5,5,5]
• Large inputs: 1M elements

4. Property Tests (Fuzzing)

• Output size = sum of input sizes
• Output sorted if inputs sorted
• All input elements appear in output

Cross-Skill Integration

Requires: multi_language_codegen (code to test) Feeds into: self_consistency_checker (validation)