test_evaluate_python.py

#!/usr/bin/env python3
"""
Test suite for Python program evaluation function using unittest framework.

This test validates that the evaluate function in synthesizer_python.py correctly
handles various types of errors and scenarios:
1. Syntax errors
2. Indentation errors
3. Runtime errors
4. Timeout errors
5. Partial test failures
6. Complete success
"""

import unittest
import sys
import os
from dataset import ProgramSynthesisDatapoint
from synthesizer_python import PythonProgramSynthesizer


class TestPythonEvaluation(unittest.TestCase):
    """Test cases for Python program evaluation."""
    
    def setUp(self):
        """Set up test fixtures before each test method."""
        self.synthesizer = PythonProgramSynthesizer()
        self.simple_datapoint = self._create_simple_datapoint()
    
    def _create_simple_datapoint(self):
        """Create a simple test datapoint for testing."""
        return ProgramSynthesisDatapoint({
            "description": "Sum all numbers in a list",
            "input_from": "standard input",
            "output_to": "standard output", 
            "time_limit": 2.0,
            "memory_limit": "256 megabytes",
            "input_spec": "First line contains n, then n integers on the next line",
            "output_spec": "Print the sum of all integers",
            "notes": "Simple test case",
            "sample_inputs": ["1 2 3", "10 20", "5"],  # Just the numbers, no count
            "sample_outputs": ["6", "30", "5"],
            "tags": ["implementation", "math"],
            "src_uid": "test_001",
            "difficulty": 100
        })
    
    def test_syntax_error(self):
        """Test evaluation with syntax error."""
        # Program with syntax error (missing colon)
        bad_program = """
numbers = list(map(int, input().split()))
sum_val = sum(numbers)
print(sum_val)  # Missing colon after if statement
if sum_val > 0
    print("Positive")
"""
        
        report = self.synthesizer.evaluate(self.simple_datapoint, bad_program)
        
        # Validate results
        self.assertFalse(report.compiles, "Should not compile due to syntax error")
        self.assertFalse(report.executes, "Should not execute if compilation fails")
        self.assertTrue(report.has_syntax_errors, "Should have syntax errors")
        self.assertGreater(len(report.compiler_errors), 0, "Should have compiler errors")
        self.assertEqual(report.overall_status, "failed", "Overall status should be failed")
        
        print(f"✅ Syntax error test passed - {len(report.compiler_errors)} compiler errors")
    
    def test_indentation_error(self):
        """Test evaluation with indentation error."""
        # Program with indentation error
        bad_program = """
numbers = list(map(int, input().split()))
sum_val = sum(numbers)
print(sum_val)
    if sum_val > 0:  # Incorrect indentation
        print("Positive")
"""
        
        report = self.synthesizer.evaluate(self.simple_datapoint, bad_program)
        
        # Validate results
        self.assertFalse(report.compiles, "Should not compile due to indentation error")
        self.assertFalse(report.executes, "Should not execute if compilation fails")
        self.assertTrue(report.has_syntax_errors, "Should have syntax errors")
        self.assertGreater(len(report.compiler_errors), 0, "Should have compiler errors")
        self.assertEqual(report.overall_status, "failed", "Overall status should be failed")
        
        print(f"✅ Indentation error test passed - {len(report.compiler_errors)} compiler errors")
    
    def test_runtime_error(self):
        """Test evaluation with runtime error."""
        # Program that compiles but has runtime error (division by zero)
        bad_program = """
numbers = list(map(int, input().split()))
sum_val = sum(numbers)
result = sum_val / 0  # Division by zero will cause runtime error
print(result)
"""
        
        report = self.synthesizer.evaluate(self.simple_datapoint, bad_program)
        
        # Validate results
        self.assertTrue(report.compiles, "Should compile successfully")
        self.assertTrue(report.executes, "Should attempt execution")
        self.assertTrue(report.has_runtime_errors, "Should have runtime errors")
        self.assertGreater(len(report.runtime_errors), 0, "Should have runtime errors")
        self.assertIn(report.overall_status, ["failed", "partial"], "Overall status should be failed or partial")
        
        print(f"✅ Runtime error test passed - {len(report.runtime_errors)} runtime errors, {report.passed_tests}/{report.total_tests} tests passed")
    
    def test_timeout(self):
        """Test evaluation with timeout."""
        # Create a datapoint with very short timeout
        timeout_datapoint = ProgramSynthesisDatapoint({
            "description": "Sum all numbers in a list",
            "input_from": "standard input",
            "output_to": "standard output", 
            "time_limit": 0.1,  # Very short timeout
            "memory_limit": "256 megabytes",
            "input_spec": "First line contains n, then n integers on the next line",
            "output_spec": "Print the sum of all integers",
            "notes": "Simple test case",
            "sample_inputs": ["1 2 3"],
            "sample_outputs": ["6"],
            "tags": ["implementation", "math"],
            "src_uid": "test_001",
            "difficulty": 100
        })
        
        # Program that will timeout (infinite loop)
        bad_program = """
numbers = list(map(int, input().split()))
sum_val = sum(numbers)
while True:  # Infinite loop to cause timeout
    pass
print(sum_val)
"""
        
        report = self.synthesizer.evaluate(timeout_datapoint, bad_program)
        
        # Validate results
        self.assertTrue(report.compiles, "Should compile successfully")
        self.assertTrue(report.executes, "Should attempt execution")
        self.assertGreater(len(report.runtime_errors), 0, "Should have timeout errors")
        self.assertTrue(any("timed out" in error.lower() for error in report.runtime_errors), "Should have timeout error")
        self.assertIn(report.overall_status, ["failed", "partial"], "Overall status should be failed or partial")
        
        print(f"✅ Timeout test passed - {len(report.runtime_errors)} runtime errors, {report.passed_tests}/{report.total_tests} tests passed")
    
    def test_partial_failure(self):
        """Test evaluation with partial test failure."""
        # Program that works for some inputs but not others (only works for even-length arrays)
        bad_program = """
numbers = list(map(int, input().split()))
sum_val = sum(numbers)
if len(numbers) % 2 == 0:
    print(sum_val)
else:
    print(sum_val + 1)  # Add 1 for odd-length arrays
"""
        
        report = self.synthesizer.evaluate(self.simple_datapoint, bad_program)
        
        # Validate results
        self.assertTrue(report.compiles, "Should compile successfully")
        self.assertTrue(report.executes, "Should execute successfully")
        self.assertFalse(report.has_syntax_errors, "Should not have syntax errors")
        self.assertFalse(report.has_runtime_errors, "Should not have runtime errors")
        self.assertLess(report.passed_tests, report.total_tests, "Should have some failed tests")
        self.assertGreater(report.passed_tests, 0, "Should have some passed tests")
        self.assertEqual(report.overall_status, "partial", "Overall status should be partial")
        self.assertGreater(report.success_rate, 0, "Success rate should be greater than 0")
        self.assertLess(report.success_rate, 1.0, "Success rate should be less than 1.0")
        
        print(f"✅ Partial failure test passed - {report.passed_tests}/{report.total_tests} tests passed ({report.success_rate:.2%})")
        
        # Print test details
        for i, test in enumerate(report.test_results):
            status_icon = "✅" if test["status"] == "passed" else "❌"
            print(f"   Test {i+1}: {status_icon} Input: {test['input']} | Expected: {test['expected_output']} | Got: {test['actual_output']}")
    
    def test_success(self):
        """Test evaluation with complete success."""
        # Correct program
        good_program = """
numbers = list(map(int, input().split()))
sum_val = sum(numbers)
print(sum_val)
"""
        
        report = self.synthesizer.evaluate(self.simple_datapoint, good_program)
        
        # Validate results
        self.assertTrue(report.compiles, "Should compile successfully")
        self.assertTrue(report.executes, "Should execute successfully")
        self.assertFalse(report.has_syntax_errors, "Should not have syntax errors")
        self.assertFalse(report.has_runtime_errors, "Should not have runtime errors")
        self.assertEqual(report.passed_tests, report.total_tests, "Should pass all tests")
        self.assertEqual(report.overall_status, "success", "Overall status should be success")
        self.assertEqual(report.success_rate, 1.0, "Success rate should be 100%")
        
        print(f"✅ Complete success test passed - {report.passed_tests}/{report.total_tests} tests passed ({report.success_rate:.2%})")
    
    def test_name_error(self):
        """Test evaluation with NameError (undefined variable)."""
        # Program with undefined variable
        bad_program = """
numbers = list(map(int, input().split()))
sum_val = sum(numbers)
print(undefined_variable)  # NameError: name 'undefined_variable' is not defined
"""
        
        report = self.synthesizer.evaluate(self.simple_datapoint, bad_program)
        
        # Validate results
        self.assertTrue(report.compiles, "Should compile successfully (syntax is valid)")
        self.assertTrue(report.executes, "Should attempt execution")
        self.assertTrue(report.has_runtime_errors, "Should have runtime errors")
        self.assertGreater(len(report.runtime_errors), 0, "Should have runtime errors")
        self.assertIn(report.overall_status, ["failed", "partial"], "Overall status should be failed or partial")
        
        print(f"✅ NameError test passed - {len(report.runtime_errors)} runtime errors, {report.passed_tests}/{report.total_tests} tests passed")
    
    def test_type_error(self):
        """Test evaluation with TypeError."""
        # Program with type error
        bad_program = """
numbers = list(map(int, input().split()))
sum_val = sum(numbers)
result = sum_val + "invalid"  # TypeError: unsupported operand type(s)
print(result)
"""
        
        report = self.synthesizer.evaluate(self.simple_datapoint, bad_program)
        
        # Validate results
        self.assertTrue(report.compiles, "Should compile successfully (syntax is valid)")
        self.assertTrue(report.executes, "Should attempt execution")
        self.assertTrue(report.has_runtime_errors, "Should have runtime errors")
        self.assertGreater(len(report.runtime_errors), 0, "Should have runtime errors")
        self.assertIn(report.overall_status, ["failed", "partial"], "Overall status should be failed or partial")
        
        print(f"✅ TypeError test passed - {len(report.runtime_errors)} runtime errors, {report.passed_tests}/{report.total_tests} tests passed")
    
    def test_value_error(self):
        """Test evaluation with ValueError."""
        # Program with value error (invalid input parsing)
        bad_program = """
numbers = list(map(int, input().split()))
sum_val = sum(numbers)
result = int("not_a_number")  # ValueError: invalid literal for int()
print(result)
"""
        
        report = self.synthesizer.evaluate(self.simple_datapoint, bad_program)
        
        # Validate results
        self.assertTrue(report.compiles, "Should compile successfully (syntax is valid)")
        self.assertTrue(report.executes, "Should attempt execution")
        self.assertTrue(report.has_runtime_errors, "Should have runtime errors")
        self.assertGreater(len(report.runtime_errors), 0, "Should have runtime errors")
        self.assertIn(report.overall_status, ["failed", "partial"], "Overall status should be failed or partial")
        
        print(f"✅ ValueError test passed - {len(report.runtime_errors)} runtime errors, {report.passed_tests}/{report.total_tests} tests passed")


if __name__ == "__main__":
    loader = unittest.TestLoader()
    suite = loader.loadTestsFromTestCase(TestPythonEvaluation)
    runner = unittest.TextTestRunner(verbosity=2, stream=sys.stdout)
    result = runner.run(suite)