feat: AI-powered test case generation - #194

Merged

cweill merged 32 commits into

feature/ai-test-generation

Oct 23, 2025

Commits on Oct 21, 2025

feat: add AI test generation foundation (WIP)

cweill
and
claude
committed
d0873d4
feat: complete AI test case generation integration
cweill
committed
a4ef187
fix: correct validation logic for error-returning functions
cweill
committed
48be4f6
refactor: switch from JSON to Go code generation for AI
cweill
committed
9d55ae2
feat: add AI golden test files using qwen2.5-coder:0.5b
cweill
committed
e47504c
refactor: LLM generates complete test functions instead of just test cases

cweill
and
claude
committed
474b022

Commits on Oct 22, 2025

test: add 6 more AI-generated golden files

cweill
and
claude
committed
b160aa2
feat: add realistic test fixtures with meaningful implementations

cweill
and
claude
committed
ae7f97e
test: add comprehensive unit tests and documentation for AI feature

cweill
and
claude
committed
94540bb
feat: add comprehensive testing and security improvements for AI feature

cweill
and
claude
committed
53a27fe
fix: address PR review feedback - validation, timeouts, and privacy docs

cweill
and
claude
committed
5aef9fa
fix: add context cancellation checks and runtime warning (required changes)

cweill
and
claude
committed
258c967
Merge remote-tracking branch 'origin/develop' into feature/ai-test-generation
cweill
committed
bf76351
fix: update render test signatures after merge from develop

cweill
and
claude
committed
e05222b
refactor: improve code coverage to 85.6% and address code review feedback

cweill
and
claude
committed
5ac64f2
fix: add AI integration tests and fix missing timeout defaults

cweill
and
claude
committed
67a7277
feat: add E2E tests with real Ollama validating against golden files

cweill
and
claude
committed
08f64db
fix: resolve import cycle in E2E tests

cweill
and
claude
committed
c2efd9d
fix: correct E2E test API usage for goparser and models

cweill
and
claude
committed
1a07297
fix: remove problematic Foo8 E2E test case

cweill
and
claude
committed
b070d1e
feat: add coverage collection for E2E AI tests

cweill
and
claude
committed
09bfbae
refactor: remove unused GenerateTestCasesWithScaffold method

cweill
and
claude
committed
5f3e554
test: add 5 E2E test cases to improve parser_go.go coverage

cweill
and
claude
committed
5492369

Commits on Oct 23, 2025

fix: implement strict golden file validation and regenerate all goldens

cweill
and
claude
committed
e07ade5
fix: improve AI prompt to prevent duplicate test cases

cweill
and
claude
committed
068d2cc
fix: improve test case names and clean up unused golden files

cweill
and
claude
committed
b28c959
feat: add retry logic to E2E tests for non-deterministic LLM output

cweill
and
claude
committed
9dcbfc4
feat: increase E2E test retry count to 10 attempts

cweill
and
claude
committed
5050e2a
fix(ai): improve E2E test determinism and use natural language test names

cweill
and
claude
committed
7c352c8
test(ai): temporarily disable non-deterministic calculator E2E tests

cweill
and
claude
committed
c7c4657
test(ai): disable 2 more non-deterministic E2E tests (4 total disabled)

cweill
and
claude
committed
58337c3
feat(ai): add min/max range for AI test case generation

cweill
and
claude
committed
40895e9