Update core/safe_executor.py

core/safe_executor.py

@@ -1,131 +1,156 @@
 # algoforge_prime/core/safe_executor.py
 # WARNING: THIS IS A SIMULATED SAFE EXECUTOR. DO NOT USE FOR REAL UNTRUSTED CODE.
-# A real implementation would require robust sandboxing (Docker, nsjail, Firecracker, WASM, etc.)
-# This simulation is purely for demonstrating the application flow.
+# A real implementation would require robust sandboxing.
+# This simulation is purely for demonstrating the application flow and data structures.
 
 import time
 import random
-import traceback
+import traceback # For capturing simulated exception details
+
+class TestResult:
+    def __init__(self, test_string: str, passed: bool, error_message: str = None, output:str = None):
+        self.test_string = test_string    # The original assert statement
+        self.passed = passed
+        self.error_message = error_message # e.g., AssertionError message
+        self.output = output # Any stdout specifically for this test (advanced)
 
 class ExecutionResult:
-    def __init__(self, success: bool, output: str = None, error: str = None, execution_time: float = 0.0, passed_tests: int = 0, total_tests: int = 0):
-        self.success = success          # True if code ran without crashing (not necessarily if tests passed)
-        self.output = output            # Stdout from the execution
-        self.error = error              # Stderr or exception message
+    def __init__(self,
+                 success: bool, # Did the overall code snippet run without crashing the sandbox?
+                 stdout: str = None,
+                 stderr: str = None, # For general stderr, not specific test failures
+                 execution_time: float = 0.0,
+                 individual_test_results: list[TestResult] = None, # List of TestResult objects
+                 compilation_error: str = None, # If code couldn't even compile/parse
+                 timeout_error: bool = False
+                ):
+        self.success = success
+        self.stdout = stdout if stdout else ""
+        self.stderr = stderr if stderr else ""
         self.execution_time = execution_time
-        self.passed_tests = passed_tests  # Number of 'assert' statements that passed
-        self.total_tests = total_tests    # Total 'assert' statements found and attempted
+        self.individual_test_results = individual_test_results if individual_test_results else []
+        self.compilation_error = compilation_error
+        self.timeout_error = timeout_error
+
+    @property
+    def passed_tests(self) -> int:
+        return sum(1 for tr in self.individual_test_results if tr.passed)
+
+    @property
+    def total_tests(self) -> int:
+        return len(self.individual_test_results)
+
+    @property
+    def overall_error_summary(self) -> str:
+        if self.compilation_error: return f"Compilation Error: {self.compilation_error}"
+        if self.timeout_error: return "Execution Timed Out."
+        if self.stderr and not self.individual_test_results: return f"Runtime Error: {self.stderr}" # General stderr if no specific test failures yet
+        # If there are test results, specific test errors are more informative
+        failed_tests = [tr for tr in self.individual_test_results if not tr.passed and tr.error_message]
+        if failed_tests:
+            return f"Failed Tests: {'; '.join([f'`{ft.test_string[:50]}...` ({ft.error_message[:50]}...)' for ft in failed_tests[:2]])}" # Show first 2
+        if self.stderr: return f"General Stderr: {self.stderr}" # If tests passed but still stderr
+        return None # No obvious top-level error
 
     def __str__(self):
-        if self.success:
-            test_summary = f"Tests: {self.passed_tests}/{self.total_tests} passed. " if self.total_tests > 0 else ""
-            return f"Execution OK. {test_summary}Time: {self.execution_time:.4f}s. Output: {str(self.output)[:200]}"
-        else:
-            return f"Execution FAILED. Error: {str(self.error)[:300]}. Time: {self.execution_time:.4f}s"
+        status = "SUCCESS (All tests passed)" if self.success and self.total_tests > 0 and self.passed_tests == self.total_tests else \
+                 ("PARTIAL SUCCESS (Some tests passed)" if self.success and self.total_tests > 0 and self.passed_tests > 0 else \
+                  ("SUCCESS (No tests run or all passed if no asserts)" if self.success else "FAILED"))
+        
+        error_info = self.overall_error_summary
+        test_summary = f"Tests: {self.passed_tests}/{self.total_tests} passed." if self.total_tests > 0 else "No assert-based tests run."
+
+        return (f"Execution {status}. {test_summary} Time: {self.execution_time:.4f}s. "
+                f"{('Error: ' + error_info) if error_info else ''} "
+                f"{('Stdout: ' + self.stdout[:100] + '...') if self.stdout else ''}")
 
-def execute_python_code_with_tests(code_string: str, tests_string: str, timeout_seconds: int = 5) -> ExecutionResult:
+
+def execute_python_code_with_tests(code_string: str, tests_string: str, timeout_seconds: int = 7) -> ExecutionResult:
     """
     SIMULATES sandboxed execution of Python code against a set of assert-based tests.
-    In a real system, this function would be a complex interface to a secure sandbox.
     """
-    print(f"SIMULATOR: Attempting to 'execute' code with tests. Timeout: {timeout_seconds}s")
+    print(f"SIMULATOR: Attempting to 'execute' code. Timeout: {timeout_seconds}s")
+    print(f"  Code (first 100 chars): {code_string[:100].strip()}...")
+    print(f"  Tests (first 100 chars): {tests_string[:100].strip()}...")
     start_time = time.time()
+    
+    individual_results = []
+    sim_stdout_lines = []
+    sim_stderr_lines = []
+    sim_compilation_error = None
+    sim_timeout = False
 
     if not code_string.strip():
-        return ExecutionResult(success=False, error="No code provided to execute.", execution_time=time.time() - start_time)
+        return ExecutionResult(success=False, compilation_error="No code provided.", execution_time=time.time() - start_time)
 
-    # Basic simulation:
-    # 1. Check for obvious syntax issues (very naive)
-    if "def " not in code_string and "class " not in code_string and not any(op in code_string for op in ["=", "+", "-", "*", "/"]):
-         # If it doesn't look like defining something or doing operations, maybe it's just a malformed snippet
-        if len(code_string) < 50 and "print" not in code_string : # very arbitrary
-            return ExecutionResult(success=False, error="Simulated: Code appears incomplete or malformed for execution.", execution_time=time.time()-start_time)
+    # Simulate compilation/syntax check (very naive)
+    if "def foo bar" in code_string or "syntax error" in code_string.lower(): # Bad syntax example
+        sim_compilation_error = "Simulated: Invalid syntax detected."
+        return ExecutionResult(success=False, compilation_error=sim_compilation_error, execution_time=time.time()-start_time)
+
+    # Simulate timeout
+    if "while True:" in code_string and "break" not in code_string and "sleep" not in code_string:
+        sim_timeout = True
+        # time.sleep(timeout_seconds + 0.1) # Actual sleep for simulation
+        return ExecutionResult(success=False, timeout_error=True, execution_time=min(timeout_seconds, time.time() - start_time), individual_test_results=individual_results)
+
+    # Simulate runtime errors not caught by tests
+    if "1/0" in code_string or "zerodivisionerror" in code_string.lower():
+        sim_stderr_lines.append("Traceback (most recent call last):\n  File \"<string>\", line X, in <module>\nZeroDivisionError: division by zero")
+        return ExecutionResult(success=False, stderr="\n".join(sim_stderr_lines), execution_time=time.time()-start_time, individual_test_results=individual_results)
+    if "undefined_variable_xyz" in code_string:
+        sim_stderr_lines.append("Traceback (most recent call last):\n  File \"<string>\", line Y, in <module>\nNameError: name 'undefined_variable_xyz' is not defined")
+        return ExecutionResult(success=False, stderr="\n".join(sim_stderr_lines), execution_time=time.time()-start_time, individual_test_results=individual_results)
+
+    # Simulate print statements in the main code
+    if "print('Setup complete')" in code_string:
+        sim_stdout_lines.append("Setup complete")
 
 
-    # 2. Simulate test parsing and execution
     test_lines = [line.strip() for line in tests_string.splitlines() if line.strip().startswith("assert")]
-    total_tests = len(test_lines)
-    passed_tests = 0
-    simulated_stdout = []
-    simulated_stderr = None
-
-    if total_tests == 0 and tests_string.strip(): # If tests were provided but no asserts found
-        simulated_stderr = "Simulated: Test string provided, but no 'assert' statements found."
-        # We can still "run" the code itself without asserts
     
-    # Simulate code "running"
-    try:
-        if "loop forever" in code_string.lower() or "while True:" in code_string and "break" not in code_string: # Timeout simulation
-            time.sleep(timeout_seconds + 0.1) # Exceed timeout
-            raise TimeoutError("Simulated: Code execution timed out.")
-        
-        # Simulate print statements
-        if "print('hello world')" in code_string:
-            simulated_stdout.append("hello world")
-        if "print(1+1)" in code_string:
-            simulated_stdout.append("2")
-
-        # Simulate errors
-        if "1/0" in code_string or "zerodivisionerror" in code_string:
-            raise ZeroDivisionError("Simulated: division by zero")
-        if "undefined_variable" in code_string:
-            raise NameError("Simulated: name 'undefined_variable' is not defined")
-
-        # Simulate test passing/failing (very naively)
-        if total_tests > 0:
-            for i, test_line in enumerate(test_lines):
-                # This is extremely basic and not real evaluation.
-                # A real system would execute each assert in the context of the provided code.
-                if "==" in test_line:
-                    if "True" in test_line or "120" in test_line or "correct" in test_line.lower(): # Naive pass conditions
-                        if random.random() > 0.1: # 90% chance of passing "good" tests
-                            passed_tests += 1
-                            simulated_stdout.append(f"Simulated Test {i+1} ({test_line[:30]}...): PASSED")
-                        else:
-                            simulated_stdout.append(f"Simulated Test {i+1} ({test_line[:30]}...): FAILED (Random Sim)")
-                            if not simulated_stderr: simulated_stderr = f"Simulated: Assertion failed on test {i+1}"
-                    elif "False" in test_line or "wrong" in test_line.lower(): # Naive fail conditions
-                         if random.random() > 0.1: # 90% chance of failing "bad" tests
-                            simulated_stdout.append(f"Simulated Test {i+1} ({test_line[:30]}...): FAILED (Expected by Sim)")
-                            if not simulated_stderr: simulated_stderr = f"Simulated: Assertion failed on test {i+1} (Expected by Sim)"
-                         else: # 10% chance of unexpectedly passing
-                            passed_tests +=1
-                            simulated_stdout.append(f"Simulated Test {i+1} ({test_line[:30]}...): PASSED (Unexpected by Sim)")
-                    else: # Other asserts
-                        if random.random() > 0.5: # 50/50
-                            passed_tests += 1
-                            simulated_stdout.append(f"Simulated Test {i+1} ({test_line[:30]}...): PASSED (Random Sim)")
-                        else:
-                            simulated_stdout.append(f"Simulated Test {i+1} ({test_line[:30]}...): FAILED (Random Sim)")
-                            if not simulated_stderr: simulated_stderr = f"Simulated: Assertion failed on test {i+1} (Random Sim)"
-                else: # Non-equality asserts, just pass some randomly
-                    if random.random() > 0.3: passed_tests +=1
+    for test_str in test_lines:
+        passed_this_test = True
+        error_this_test = None
+        # Super naive simulation based on keywords in the test or code
+        # This needs to be replaced by actual execution in a sandbox.
+        if "None" in test_str and "TypeError" in code_string and "raise TypeError" in code_string:
+            if "reverse_string(None)" in test_str: # Specific to the example
+                passed_this_test = True # Simulating a try-except test for TypeError
+                sim_stdout_lines.append(f"SimTest '{test_str[:30]}...': PASSED (TypeError correctly raised and caught - simulated)")
+            else: # General "None" test might fail if code doesn't handle it right
+                passed_this_test = random.choice([True, False, False]) # Make None checks less likely to pass randomly
+        elif "==" in test_str:
+            parts = test_str.split("==")
+            actual_call_sim = parts[0].replace("assert", "").strip()
+            expected_sim = parts[1].strip()
+            
+            # Simulate based on a few known patterns from the example
+            if "reverse_string(\"hello\")" == actual_call_sim and "\"olleh\"" == expected_sim:
+                passed_this_test = True if "s[::-1]" in code_string or "char + reversed_string" in code_string else False
+            elif "reverse_string(\"\")" == actual_call_sim and "\"\"" == expected_sim:
+                passed_this_test = True # Empty string case usually handled
+            elif "reverse_string(123)" == actual_call_sim and "\"321\"" == expected_sim:
+                passed_this_test = True if "str(s)" in code_string and ("s[::-1]" in code_string or "char + reversed_string" in code_string) else False
+            else: # Randomize other equality tests
+                passed_this_test = random.random() > 0.3 # 70% chance of passing random asserts
+
+        if not passed_this_test:
+            error_this_test = f"Simulated AssertionError: {test_str} evaluated to False"
+            sim_stderr_lines.append(f"FAIL: {test_str}\n  {error_this_test}")
         
-        # If no explicit error, but not all tests passed
-        if total_tests > 0 and passed_tests < total_tests and not simulated_stderr:
-            simulated_stderr = f"Simulated: {total_tests - passed_tests} out of {total_tests} assertions failed."
-
-        execution_time = time.time() - start_time
-        if execution_time > timeout_seconds: # Check after simulated work
-            raise TimeoutError("Simulated: Code execution exceeded timeout.")
-
-        return ExecutionResult(
-            success=True if not simulated_stderr or (total_tests > 0 and passed_tests == total_tests) else False, # Success if no errors or all tests pass
-            output="\n".join(simulated_stdout) if simulated_stdout else None,
-            error=simulated_stderr,
-            execution_time=execution_time,
-            passed_tests=passed_tests,
-            total_tests=total_tests
-        )
-
-    except Exception as e:
-        execution_time = time.time() - start_time
-        tb_str = traceback.format_exc()
-        print(f"SIMULATOR EXCEPTION: {e}\n{tb_str}")
-        return ExecutionResult(
-            success=False,
-            error=f"Simulated Exception: {type(e).__name__} - {str(e)}",
-            execution_time=execution_time,
-            passed_tests=passed_tests, # Could be some if error happened mid-tests
-            total_tests=total_tests
-        )
+        individual_results.append(TestResult(test_string=test_str, passed=passed_this_test, error_message=error_this_test))
+
+    final_success = not sim_compilation_error and not sim_timeout and not any(not tr.passed for tr in individual_results) and not (sim_stderr_lines and not individual_results)
+
+    return ExecutionResult(
+        success=final_success,
+        stdout="\n".join(sim_stdout_lines) if sim_stdout_lines else None,
+        stderr="\n".join(sim_stderr_lines) if sim_stderr_lines else None,
+        execution_time=min(time.time() - start_time, timeout_seconds), # Cap time for simulation
+        individual_test_results=individual_results,
+        compilation_error=sim_compilation_error,
+        timeout_error=sim_timeout
+    )
+
+print("DEBUG: core.safe_executor (SIMULATED) - Module defined.")