feat: Add comprehensive unit test expansion for API endpoints

backend/TEST_DOCUMENTATION.md

@@ -0,0 +1,267 @@
+# Unit Tests Documentation
+
+## Overview
+This directory contains comprehensive unit tests for the EduAid backend API endpoints.
+
+## Test Files
+
+### 1. `test_server.py` - Integration Tests
+**Purpose:** End-to-end testing of all API endpoints against a running server
+
+**Test Coverage:**
+- ✅ All question generation endpoints (MCQ, Boolean, Short Q)
+- ✅ LLM-based endpoints (Qwen3 models)
+- ✅ Combined endpoint (`/get_problems`)
+- ✅ Answer prediction endpoints
+- ✅ Error handling and validation
+- ✅ Edge cases (empty input, oversized input, invalid types)
+- ✅ Response format validation
+
+**Running Tests:**
+```bash
+# Start the Flask server first
+python server.py
+
+# In another terminal, run tests
+python test_server.py
+
+# Or with pytest
+pytest test_server.py -v
+```
+
+**Test Categories in test_server.py:**
+- **Basic Tests (7):** Test all major endpoints with valid inputs
+- **LLM Tests (4):** Test Qwen3-0.6B model-based endpoints
+- **Error Validation Tests (13):** Test input validation and error handling
+- **Response Format Tests (2):** Validate response structure
+
+### 2. `test_endpoints.py` - Unit Tests (pytest)
+**Purpose:** Isolated unit tests using pytest fixtures and mocking
+
+**Test Coverage:**
+- ✅ 40+ test cases organized by endpoint
+- ✅ Success scenarios
+- ✅ Error scenarios
+- ✅ Edge cases (empty lists, mismatched data, etc.)
+- ✅ Parametrized tests for multiple invalid inputs
+
+**Test Classes:**
+- `TestMCQEndpoint` - 8 tests
+- `TestBoolQEndpoint` - 3 tests
+- `TestShortQEndpoint` - 2 tests
+- `TestMCQAnswerEndpoint` - 5 tests
+- `TestShortQAnswerEndpoint` - 3 tests
+- `TestBooleanAnswerEndpoint` - 2 tests
+- `TestUploadEndpoint` - 3 tests
+- `TestGetContentEndpoint` - 2 tests
+- `TestErrorResponses` - 1 test
+- `TestInputValidation` - 3 parametrized tests
+
+**Running Tests:**
+```bash
+# Run all tests
+pytest test_endpoints.py -v
+
+# Run specific test class
+pytest test_endpoints.py::TestMCQEndpoint -v
+
+# Run with coverage
+pytest test_endpoints.py --cov=server --cov-report=html
+
+# Run with detailed output
+pytest test_endpoints.py -v --tb=short
+```
+
+## What's Tested
+
+### Error Handling ✅
+- Empty input text rejection
+- Oversized input text rejection (>50,000 chars)
+- Invalid max_questions values (0, 100, "abc", 3.14)
+- Invalid use_mediawiki values (2, -1, etc.)
+- Invalid URL formats
+- Missing required fields
+- Mismatched question/option counts
+- Whitespace-only inputs
+
+### Validation ✅
+- Type validation (string, integer, list)
+- Range validation (questions: 1-50)
+- Length validation (text: 1-50,000 chars)
+- Enum validation (use_mediawiki: 0 or 1)
+- URL format validation (http/https)
+- Required field checking
+
+### Response Formats ✅
+- Standardized error format: `{"error": "message"}`
+- Success format: `{"output": [...]}`
+- Proper HTTP status codes (200, 400, 500)
+- Datetime formats and content validation
+
+## Test Statistics
+
+| Category | Count | Status |
+|----------|-------|--------|
+| Total Tests | 50+ | ✅ Ready |
+| Error Validation Tests | 13 | ✅ Ready |
+| Success Case Tests | 7 | ✅ Ready |
+| LLM Integration Tests | 4 | ✅ Ready |
+| Format Validation Tests | 2 | ✅ Ready |
+| Pytest Unit Tests | 40+ | ✅ Ready |
+
+## Running All Tests
+
+### Option 1: Integration Tests (test_server.py)
+```bash
+cd backend
+python server.py  # Terminal 1
+
+# Terminal 2
+python test_server.py
+```
+
+### Option 2: Unit Tests (test_endpoints.py)
+```bash
+cd backend
+pytest test_endpoints.py -v
+```
+
+### Option 3: Both (requires pytest)
+```bash
+cd backend
+
+# Run integration tests
+python test_server.py
+
+# Run unit tests
+pytest test_endpoints.py -v
+
+# Generate coverage report
+pytest test_endpoints.py --cov=server --cov-report=html
+```
+
+## Expected Output
+
+### Successful Run (test_server.py)
+```
+============================================================
+Running Basic Endpoint Tests
+============================================================
+
+SUCCESS TESTS:
+✓ Root endpoint working
+✓ MCQ generation successful
+✓ Boolean Q generation successful
+✓ Short Q generation successful
+✓ All problem types generation successful
+
+============================================================
+Running LLM Integration Tests
+============================================================
+
+LLM TESTS:
+✓ Short Q LLM generation successful
+✓ MCQ LLM generation successful
+✓ Boolean Q LLM generation successful
+✓ Mixed LLM generation successful
+
+============================================================
+Running Error/Edge Case Tests
+============================================================
+
+ERROR VALIDATION TESTS:
+✓ MCQ with empty input correctly rejected: input_text cannot be empty...
+✓ MCQ with max_questions=100 correctly rejected: max_questions cannot exceed 50...
+✓ MCQ with invalid max_questions type correctly rejected: max_questions must be an integer...
+[... more validation tests ...]
+
+All tests completed!
+```
+
+## Test Case Examples
+
+### Valid Request
+```python
+# Should succeed
+response = client.post('/get_mcq', json={
+    "input_text": "Valid educational content...",
+    "max_questions": 5,
+    "use_mediawiki": 0
+})
+assert response.status_code == 200
+assert "output" in response.json()
+```
+
+### Invalid Request - Empty Input
+```python
+# Should return 400 error
+response = client.post('/get_mcq', json={
+    "input_text": "",
+    "max_questions": 5
+})
+assert response.status_code == 400
+assert "error" in response.json()
+assert "empty" in response.json()["error"].lower()
+```
+
+### Invalid Request - Exceeded Limit
+```python
+# Should return 400 error
+response = client.post('/get_mcq', json={
+    "input_text": "Valid content",
+    "max_questions": 100  # Exceeds limit of 50
+})
+assert response.status_code == 400
+assert "error" in response.json()
+assert "exceed" in response.json()["error"].lower()
+```
+
+## Continuous Integration
+
+To integrate with CI/CD (GitHub Actions, etc.):
+
+```yaml
+- name: Run API Tests
+  run: |
+    cd backend
+    python -m pytest test_endpoints.py -v --tb=short
+
+- name: Run Integration Tests
+  run: |
+    cd backend
+    python server.py &
+    sleep 5
+    python test_server.py
+```
+
+## Troubleshooting
+
+### Issue: Tests fail with "Connection refused"
+**Solution:** Make sure Flask server is running: `python server.py`
+
+### Issue: "ModuleNotFoundError" for validators
+**Solution:** Ensure you're in the `backend` directory and validators.py exists with InputValidator class
+
+### Issue: Tests timeout
+**Solution:** Flask server may be slow to load models. Increase timeout or wait longer before running tests.
+
+### Issue: Some tests skipped
+**Solution:** Tests may skip if models aren't downloaded. Run `python server.py` once to download models, then run tests again.
+
+## Future Enhancements
+
+- [ ] Add more edge case tests
+- [ ] Add performance benchmarking tests
+- [ ] Add concurrency stress tests
+- [ ] Add security penetration tests
+- [ ] Add response schema validation tests
+- [ ] Add database integration tests (when applicable)
+
+## Contributing
+
+When adding new endpoints:
+1. Add success case test
+2. Add error case tests (empty input, invalid types, missing fields)
+3. Add edge case tests
+4. Update this documentation
+5. Run all tests before submitting PR