Spaces:
Sleeping
Sleeping
Update core/safe_executor.py
Browse files- core/safe_executor.py +133 -108
core/safe_executor.py
CHANGED
@@ -1,131 +1,156 @@
|
|
1 |
# algoforge_prime/core/safe_executor.py
|
2 |
# WARNING: THIS IS A SIMULATED SAFE EXECUTOR. DO NOT USE FOR REAL UNTRUSTED CODE.
|
3 |
-
# A real implementation would require robust sandboxing
|
4 |
-
# This simulation is purely for demonstrating the application flow.
|
5 |
|
6 |
import time
|
7 |
import random
|
8 |
-
import traceback
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
9 |
|
10 |
class ExecutionResult:
|
11 |
-
def __init__(self,
|
12 |
-
|
13 |
-
|
14 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
15 |
self.execution_time = execution_time
|
16 |
-
self.
|
17 |
-
self.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
18 |
|
19 |
def __str__(self):
|
20 |
-
if self.success
|
21 |
-
|
22 |
-
|
23 |
-
|
24 |
-
|
|
|
|
|
|
|
|
|
|
|
25 |
|
26 |
-
|
|
|
27 |
"""
|
28 |
SIMULATES sandboxed execution of Python code against a set of assert-based tests.
|
29 |
-
In a real system, this function would be a complex interface to a secure sandbox.
|
30 |
"""
|
31 |
-
print(f"SIMULATOR: Attempting to 'execute' code
|
|
|
|
|
32 |
start_time = time.time()
|
|
|
|
|
|
|
|
|
|
|
|
|
33 |
|
34 |
if not code_string.strip():
|
35 |
-
return ExecutionResult(success=False,
|
36 |
|
37 |
-
#
|
38 |
-
|
39 |
-
|
40 |
-
|
41 |
-
|
42 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
43 |
|
44 |
|
45 |
-
# 2. Simulate test parsing and execution
|
46 |
test_lines = [line.strip() for line in tests_string.splitlines() if line.strip().startswith("assert")]
|
47 |
-
total_tests = len(test_lines)
|
48 |
-
passed_tests = 0
|
49 |
-
simulated_stdout = []
|
50 |
-
simulated_stderr = None
|
51 |
-
|
52 |
-
if total_tests == 0 and tests_string.strip(): # If tests were provided but no asserts found
|
53 |
-
simulated_stderr = "Simulated: Test string provided, but no 'assert' statements found."
|
54 |
-
# We can still "run" the code itself without asserts
|
55 |
|
56 |
-
|
57 |
-
|
58 |
-
|
59 |
-
|
60 |
-
|
61 |
-
|
62 |
-
|
63 |
-
|
64 |
-
|
65 |
-
|
66 |
-
|
67 |
-
|
68 |
-
|
69 |
-
|
70 |
-
|
71 |
-
|
72 |
-
|
73 |
-
|
74 |
-
|
75 |
-
|
76 |
-
|
77 |
-
|
78 |
-
|
79 |
-
|
80 |
-
|
81 |
-
|
82 |
-
|
83 |
-
|
84 |
-
|
85 |
-
simulated_stdout.append(f"Simulated Test {i+1} ({test_line[:30]}...): FAILED (Random Sim)")
|
86 |
-
if not simulated_stderr: simulated_stderr = f"Simulated: Assertion failed on test {i+1}"
|
87 |
-
elif "False" in test_line or "wrong" in test_line.lower(): # Naive fail conditions
|
88 |
-
if random.random() > 0.1: # 90% chance of failing "bad" tests
|
89 |
-
simulated_stdout.append(f"Simulated Test {i+1} ({test_line[:30]}...): FAILED (Expected by Sim)")
|
90 |
-
if not simulated_stderr: simulated_stderr = f"Simulated: Assertion failed on test {i+1} (Expected by Sim)"
|
91 |
-
else: # 10% chance of unexpectedly passing
|
92 |
-
passed_tests +=1
|
93 |
-
simulated_stdout.append(f"Simulated Test {i+1} ({test_line[:30]}...): PASSED (Unexpected by Sim)")
|
94 |
-
else: # Other asserts
|
95 |
-
if random.random() > 0.5: # 50/50
|
96 |
-
passed_tests += 1
|
97 |
-
simulated_stdout.append(f"Simulated Test {i+1} ({test_line[:30]}...): PASSED (Random Sim)")
|
98 |
-
else:
|
99 |
-
simulated_stdout.append(f"Simulated Test {i+1} ({test_line[:30]}...): FAILED (Random Sim)")
|
100 |
-
if not simulated_stderr: simulated_stderr = f"Simulated: Assertion failed on test {i+1} (Random Sim)"
|
101 |
-
else: # Non-equality asserts, just pass some randomly
|
102 |
-
if random.random() > 0.3: passed_tests +=1
|
103 |
|
104 |
-
|
105 |
-
|
106 |
-
|
107 |
-
|
108 |
-
|
109 |
-
|
110 |
-
|
111 |
-
|
112 |
-
|
113 |
-
|
114 |
-
|
115 |
-
|
116 |
-
|
117 |
-
|
118 |
-
|
119 |
-
)
|
120 |
-
|
121 |
-
except Exception as e:
|
122 |
-
execution_time = time.time() - start_time
|
123 |
-
tb_str = traceback.format_exc()
|
124 |
-
print(f"SIMULATOR EXCEPTION: {e}\n{tb_str}")
|
125 |
-
return ExecutionResult(
|
126 |
-
success=False,
|
127 |
-
error=f"Simulated Exception: {type(e).__name__} - {str(e)}",
|
128 |
-
execution_time=execution_time,
|
129 |
-
passed_tests=passed_tests, # Could be some if error happened mid-tests
|
130 |
-
total_tests=total_tests
|
131 |
-
)
|
|
|
1 |
# algoforge_prime/core/safe_executor.py
|
2 |
# WARNING: THIS IS A SIMULATED SAFE EXECUTOR. DO NOT USE FOR REAL UNTRUSTED CODE.
|
3 |
+
# A real implementation would require robust sandboxing.
|
4 |
+
# This simulation is purely for demonstrating the application flow and data structures.
|
5 |
|
6 |
import time
|
7 |
import random
|
8 |
+
import traceback # For capturing simulated exception details
|
9 |
+
|
10 |
+
class TestResult:
|
11 |
+
def __init__(self, test_string: str, passed: bool, error_message: str = None, output:str = None):
|
12 |
+
self.test_string = test_string # The original assert statement
|
13 |
+
self.passed = passed
|
14 |
+
self.error_message = error_message # e.g., AssertionError message
|
15 |
+
self.output = output # Any stdout specifically for this test (advanced)
|
16 |
|
17 |
class ExecutionResult:
|
18 |
+
def __init__(self,
|
19 |
+
success: bool, # Did the overall code snippet run without crashing the sandbox?
|
20 |
+
stdout: str = None,
|
21 |
+
stderr: str = None, # For general stderr, not specific test failures
|
22 |
+
execution_time: float = 0.0,
|
23 |
+
individual_test_results: list[TestResult] = None, # List of TestResult objects
|
24 |
+
compilation_error: str = None, # If code couldn't even compile/parse
|
25 |
+
timeout_error: bool = False
|
26 |
+
):
|
27 |
+
self.success = success
|
28 |
+
self.stdout = stdout if stdout else ""
|
29 |
+
self.stderr = stderr if stderr else ""
|
30 |
self.execution_time = execution_time
|
31 |
+
self.individual_test_results = individual_test_results if individual_test_results else []
|
32 |
+
self.compilation_error = compilation_error
|
33 |
+
self.timeout_error = timeout_error
|
34 |
+
|
35 |
+
@property
|
36 |
+
def passed_tests(self) -> int:
|
37 |
+
return sum(1 for tr in self.individual_test_results if tr.passed)
|
38 |
+
|
39 |
+
@property
|
40 |
+
def total_tests(self) -> int:
|
41 |
+
return len(self.individual_test_results)
|
42 |
+
|
43 |
+
@property
|
44 |
+
def overall_error_summary(self) -> str:
|
45 |
+
if self.compilation_error: return f"Compilation Error: {self.compilation_error}"
|
46 |
+
if self.timeout_error: return "Execution Timed Out."
|
47 |
+
if self.stderr and not self.individual_test_results: return f"Runtime Error: {self.stderr}" # General stderr if no specific test failures yet
|
48 |
+
# If there are test results, specific test errors are more informative
|
49 |
+
failed_tests = [tr for tr in self.individual_test_results if not tr.passed and tr.error_message]
|
50 |
+
if failed_tests:
|
51 |
+
return f"Failed Tests: {'; '.join([f'`{ft.test_string[:50]}...` ({ft.error_message[:50]}...)' for ft in failed_tests[:2]])}" # Show first 2
|
52 |
+
if self.stderr: return f"General Stderr: {self.stderr}" # If tests passed but still stderr
|
53 |
+
return None # No obvious top-level error
|
54 |
|
55 |
def __str__(self):
|
56 |
+
status = "SUCCESS (All tests passed)" if self.success and self.total_tests > 0 and self.passed_tests == self.total_tests else \
|
57 |
+
("PARTIAL SUCCESS (Some tests passed)" if self.success and self.total_tests > 0 and self.passed_tests > 0 else \
|
58 |
+
("SUCCESS (No tests run or all passed if no asserts)" if self.success else "FAILED"))
|
59 |
+
|
60 |
+
error_info = self.overall_error_summary
|
61 |
+
test_summary = f"Tests: {self.passed_tests}/{self.total_tests} passed." if self.total_tests > 0 else "No assert-based tests run."
|
62 |
+
|
63 |
+
return (f"Execution {status}. {test_summary} Time: {self.execution_time:.4f}s. "
|
64 |
+
f"{('Error: ' + error_info) if error_info else ''} "
|
65 |
+
f"{('Stdout: ' + self.stdout[:100] + '...') if self.stdout else ''}")
|
66 |
|
67 |
+
|
68 |
+
def execute_python_code_with_tests(code_string: str, tests_string: str, timeout_seconds: int = 7) -> ExecutionResult:
|
69 |
"""
|
70 |
SIMULATES sandboxed execution of Python code against a set of assert-based tests.
|
|
|
71 |
"""
|
72 |
+
print(f"SIMULATOR: Attempting to 'execute' code. Timeout: {timeout_seconds}s")
|
73 |
+
print(f" Code (first 100 chars): {code_string[:100].strip()}...")
|
74 |
+
print(f" Tests (first 100 chars): {tests_string[:100].strip()}...")
|
75 |
start_time = time.time()
|
76 |
+
|
77 |
+
individual_results = []
|
78 |
+
sim_stdout_lines = []
|
79 |
+
sim_stderr_lines = []
|
80 |
+
sim_compilation_error = None
|
81 |
+
sim_timeout = False
|
82 |
|
83 |
if not code_string.strip():
|
84 |
+
return ExecutionResult(success=False, compilation_error="No code provided.", execution_time=time.time() - start_time)
|
85 |
|
86 |
+
# Simulate compilation/syntax check (very naive)
|
87 |
+
if "def foo bar" in code_string or "syntax error" in code_string.lower(): # Bad syntax example
|
88 |
+
sim_compilation_error = "Simulated: Invalid syntax detected."
|
89 |
+
return ExecutionResult(success=False, compilation_error=sim_compilation_error, execution_time=time.time()-start_time)
|
90 |
+
|
91 |
+
# Simulate timeout
|
92 |
+
if "while True:" in code_string and "break" not in code_string and "sleep" not in code_string:
|
93 |
+
sim_timeout = True
|
94 |
+
# time.sleep(timeout_seconds + 0.1) # Actual sleep for simulation
|
95 |
+
return ExecutionResult(success=False, timeout_error=True, execution_time=min(timeout_seconds, time.time() - start_time), individual_test_results=individual_results)
|
96 |
+
|
97 |
+
# Simulate runtime errors not caught by tests
|
98 |
+
if "1/0" in code_string or "zerodivisionerror" in code_string.lower():
|
99 |
+
sim_stderr_lines.append("Traceback (most recent call last):\n File \"<string>\", line X, in <module>\nZeroDivisionError: division by zero")
|
100 |
+
return ExecutionResult(success=False, stderr="\n".join(sim_stderr_lines), execution_time=time.time()-start_time, individual_test_results=individual_results)
|
101 |
+
if "undefined_variable_xyz" in code_string:
|
102 |
+
sim_stderr_lines.append("Traceback (most recent call last):\n File \"<string>\", line Y, in <module>\nNameError: name 'undefined_variable_xyz' is not defined")
|
103 |
+
return ExecutionResult(success=False, stderr="\n".join(sim_stderr_lines), execution_time=time.time()-start_time, individual_test_results=individual_results)
|
104 |
+
|
105 |
+
# Simulate print statements in the main code
|
106 |
+
if "print('Setup complete')" in code_string:
|
107 |
+
sim_stdout_lines.append("Setup complete")
|
108 |
|
109 |
|
|
|
110 |
test_lines = [line.strip() for line in tests_string.splitlines() if line.strip().startswith("assert")]
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
111 |
|
112 |
+
for test_str in test_lines:
|
113 |
+
passed_this_test = True
|
114 |
+
error_this_test = None
|
115 |
+
# Super naive simulation based on keywords in the test or code
|
116 |
+
# This needs to be replaced by actual execution in a sandbox.
|
117 |
+
if "None" in test_str and "TypeError" in code_string and "raise TypeError" in code_string:
|
118 |
+
if "reverse_string(None)" in test_str: # Specific to the example
|
119 |
+
passed_this_test = True # Simulating a try-except test for TypeError
|
120 |
+
sim_stdout_lines.append(f"SimTest '{test_str[:30]}...': PASSED (TypeError correctly raised and caught - simulated)")
|
121 |
+
else: # General "None" test might fail if code doesn't handle it right
|
122 |
+
passed_this_test = random.choice([True, False, False]) # Make None checks less likely to pass randomly
|
123 |
+
elif "==" in test_str:
|
124 |
+
parts = test_str.split("==")
|
125 |
+
actual_call_sim = parts[0].replace("assert", "").strip()
|
126 |
+
expected_sim = parts[1].strip()
|
127 |
+
|
128 |
+
# Simulate based on a few known patterns from the example
|
129 |
+
if "reverse_string(\"hello\")" == actual_call_sim and "\"olleh\"" == expected_sim:
|
130 |
+
passed_this_test = True if "s[::-1]" in code_string or "char + reversed_string" in code_string else False
|
131 |
+
elif "reverse_string(\"\")" == actual_call_sim and "\"\"" == expected_sim:
|
132 |
+
passed_this_test = True # Empty string case usually handled
|
133 |
+
elif "reverse_string(123)" == actual_call_sim and "\"321\"" == expected_sim:
|
134 |
+
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
|
135 |
+
else: # Randomize other equality tests
|
136 |
+
passed_this_test = random.random() > 0.3 # 70% chance of passing random asserts
|
137 |
+
|
138 |
+
if not passed_this_test:
|
139 |
+
error_this_test = f"Simulated AssertionError: {test_str} evaluated to False"
|
140 |
+
sim_stderr_lines.append(f"FAIL: {test_str}\n {error_this_test}")
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
141 |
|
142 |
+
individual_results.append(TestResult(test_string=test_str, passed=passed_this_test, error_message=error_this_test))
|
143 |
+
|
144 |
+
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)
|
145 |
+
|
146 |
+
return ExecutionResult(
|
147 |
+
success=final_success,
|
148 |
+
stdout="\n".join(sim_stdout_lines) if sim_stdout_lines else None,
|
149 |
+
stderr="\n".join(sim_stderr_lines) if sim_stderr_lines else None,
|
150 |
+
execution_time=min(time.time() - start_time, timeout_seconds), # Cap time for simulation
|
151 |
+
individual_test_results=individual_results,
|
152 |
+
compilation_error=sim_compilation_error,
|
153 |
+
timeout_error=sim_timeout
|
154 |
+
)
|
155 |
+
|
156 |
+
print("DEBUG: core.safe_executor (SIMULATED) - Module defined.")
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|