fix(examples): escape paths in circle packing subprocess scripts

examples/circle_packing/evaluator.py

@@ -79,41 +79,36 @@ def validate_packing(centers, radii):
     return True
 
 
-def run_with_timeout(program_path, timeout_seconds=20):
+def _subprocess_script(program_path, results_path):
     """
-    Run the program in a separate process with timeout
-    using a simple subprocess approach
+    Build the runner executed in the child process.
 
-    Args:
-        program_path: Path to the program file
-        timeout_seconds: Maximum execution time in seconds
-
-    Returns:
-        centers, radii, sum_radii tuple from the program
+    Paths are inserted with repr(). A quoted Windows path such as
+    'C:\\Users\\...' is not valid Python: \\U starts a unicode escape and the
+    child exits with SyntaxError before the program runs.
     """
-    # Create a temporary file to execute
-    with tempfile.NamedTemporaryFile(suffix=".py", delete=False) as temp_file:
-        # Write a script that executes the program and saves results
-        script = f"""
+    program_lit = repr(os.fspath(program_path))
+    results_lit = repr(os.fspath(results_path))
+    return f"""
 import sys
 import numpy as np
 import os
 import pickle
 import traceback
 
 # Add the directory to sys.path
-sys.path.insert(0, os.path.dirname('{program_path}'))
+sys.path.insert(0, os.path.dirname({program_lit}))
 
 # Debugging info
 print(f"Running in subprocess, Python version: {{sys.version}}")
-print(f"Program path: {program_path}")
+print("Program path: " + {program_lit})
 
 try:
     # Import the program
-    spec = __import__('importlib.util').util.spec_from_file_location("program", '{program_path}')
+    spec = __import__('importlib.util').util.spec_from_file_location("program", {program_lit})
     program = __import__('importlib.util').util.module_from_spec(spec)
     spec.loader.exec_module(program)
-    
+
     # Run the packing function
     print("Calling run_packing()...")
     centers, radii, sum_radii = program.run_packing()
@@ -126,22 +121,39 @@ def run_with_timeout(program_path, timeout_seconds=20):
         'sum_radii': sum_radii
     }}
 
-    with open('{temp_file.name}.results', 'wb') as f:
+    with open({results_lit}, 'wb') as f:
         pickle.dump(results, f)
-    print(f"Results saved to {temp_file.name}.results")
-    
+    print("Results saved to " + {results_lit})
+
 except Exception as e:
     # If an error occurs, save the error instead
     print(f"Error in subprocess: {{str(e)}}")
     traceback.print_exc()
-    with open('{temp_file.name}.results', 'wb') as f:
+    with open({results_lit}, 'wb') as f:
         pickle.dump({{'error': str(e)}}, f)
-    print(f"Error saved to {temp_file.name}.results")
+    print("Error saved to " + {results_lit})
 """
-        temp_file.write(script.encode())
-        temp_file_path = temp_file.name
 
-    results_path = f"{temp_file_path}.results"
+
+def run_with_timeout(program_path, timeout_seconds=20):
+    """
+    Run the program in a separate process with timeout
+    using a simple subprocess approach
+
+    Args:
+        program_path: Path to the program file
+        timeout_seconds: Maximum execution time in seconds
+
+    Returns:
+        centers, radii, sum_radii tuple from the program
+    """
+    # Create a temporary file to execute
+    with tempfile.NamedTemporaryFile(suffix=".py", delete=False) as temp_file:
+        temp_file_path = temp_file.name
+        results_path = f"{temp_file_path}.results"
+        # Write a script that executes the program and saves results
+        script = _subprocess_script(program_path, results_path)
+        temp_file.write(script.encode())
 
     try:
         # Run the script with timeout

examples/circle_packing_with_artifacts/evaluator.py

@@ -94,41 +94,36 @@ def validate_packing(centers, radii):
     return is_valid, validation_details
 
 
-def run_with_timeout(program_path, timeout_seconds=20):
+def _subprocess_script(program_path, results_path):
     """
-    Run the program in a separate process with timeout
-    using a simple subprocess approach
+    Build the runner executed in the child process.
 
-    Args:
-        program_path: Path to the program file
-        timeout_seconds: Maximum execution time in seconds
-
-    Returns:
-        centers, radii, sum_radii tuple from the program
+    Paths are inserted with repr(). A quoted Windows path such as
+    'C:\\Users\\...' is not valid Python: \\U starts a unicode escape and the
+    child exits with SyntaxError before the program runs.
     """
-    # Create a temporary file to execute
-    with tempfile.NamedTemporaryFile(suffix=".py", delete=False) as temp_file:
-        # Write a script that executes the program and saves results
-        script = f"""
+    program_lit = repr(os.fspath(program_path))
+    results_lit = repr(os.fspath(results_path))
+    return f"""
 import sys
 import numpy as np
 import os
 import pickle
 import traceback
 
 # Add the directory to sys.path
-sys.path.insert(0, os.path.dirname('{program_path}'))
+sys.path.insert(0, os.path.dirname({program_lit}))
 
 # Debugging info
 print(f"Running in subprocess, Python version: {{sys.version}}")
-print(f"Program path: {program_path}")
+print("Program path: " + {program_lit})
 
 try:
     # Import the program
-    spec = __import__('importlib.util').util.spec_from_file_location("program", '{program_path}')
+    spec = __import__('importlib.util').util.spec_from_file_location("program", {program_lit})
     program = __import__('importlib.util').util.module_from_spec(spec)
     spec.loader.exec_module(program)
-    
+
     # Run the packing function
     print("Calling run_packing()...")
     centers, radii, sum_radii = program.run_packing()
@@ -141,22 +136,39 @@ def run_with_timeout(program_path, timeout_seconds=20):
         'sum_radii': sum_radii
     }}
 
-    with open('{temp_file.name}.results', 'wb') as f:
+    with open({results_lit}, 'wb') as f:
         pickle.dump(results, f)
-    print(f"Results saved to {temp_file.name}.results")
-    
+    print("Results saved to " + {results_lit})
+
 except Exception as e:
     # If an error occurs, save the error instead
     print(f"Error in subprocess: {{str(e)}}")
     traceback.print_exc()
-    with open('{temp_file.name}.results', 'wb') as f:
+    with open({results_lit}, 'wb') as f:
         pickle.dump({{'error': str(e)}}, f)
-    print(f"Error saved to {temp_file.name}.results")
+    print("Error saved to " + {results_lit})
 """
-        temp_file.write(script.encode())
-        temp_file_path = temp_file.name
 
-    results_path = f"{temp_file_path}.results"
+
+def run_with_timeout(program_path, timeout_seconds=20):
+    """
+    Run the program in a separate process with timeout
+    using a simple subprocess approach
+
+    Args:
+        program_path: Path to the program file
+        timeout_seconds: Maximum execution time in seconds
+
+    Returns:
+        centers, radii, sum_radii tuple from the program
+    """
+    # Create a temporary file to execute
+    with tempfile.NamedTemporaryFile(suffix=".py", delete=False) as temp_file:
+        temp_file_path = temp_file.name
+        results_path = f"{temp_file_path}.results"
+        # Write a script that executes the program and saves results
+        script = _subprocess_script(program_path, results_path)
+        temp_file.write(script.encode())
 
     try:
         # Run the script with timeout

tests/test_circle_packing_subprocess_paths.py

@@ -0,0 +1,85 @@
+"""
+Regression test for circle-packing evaluators embedding paths in generated Python.
+
+A Windows temp path contains \\Users. Putting that path inside single quotes makes
+\\U a unicode escape, so the child process dies with SyntaxError and every
+evaluation is scored as a failure.
+"""
+
+import importlib.util
+import os
+import tempfile
+import textwrap
+import unittest
+
+REPO_ROOT = os.path.abspath(os.path.join(os.path.dirname(__file__), ".."))
+EVALUATOR_PATHS = [
+    os.path.join(REPO_ROOT, "examples", "circle_packing", "evaluator.py"),
+    os.path.join(REPO_ROOT, "examples", "circle_packing_with_artifacts", "evaluator.py"),
+]
+
+
+def load_evaluator(path):
+    """Load an example evaluator by path (examples are not part of the package)"""
+    name = "circle_packing_eval_" + os.path.splitext(os.path.basename(os.path.dirname(path)))[0]
+    spec = importlib.util.spec_from_file_location(name, path)
+    evaluator = importlib.util.module_from_spec(spec)
+    spec.loader.exec_module(evaluator)
+    return evaluator
+
+
+class TestCirclePackingSubprocessPaths(unittest.TestCase):
+    """Generated runner source must be valid Python for Windows paths"""
+
+    @classmethod
+    def setUpClass(cls):
+        cls.evaluators = [load_evaluator(path) for path in EVALUATOR_PATHS]
+
+    def test_windows_user_temp_path_compiles(self):
+        """C:\\Users\\... must not be parsed as a unicode escape"""
+        program_path = r"C:\Users\alice\AppData\Local\Temp\prog.py"
+        results_path = program_path + ".results"
+        for evaluator in self.evaluators:
+            script = evaluator._subprocess_script(program_path, results_path)
+            compile(script, "<circle-packing-runner>", "exec")
+            self.assertIn(repr(program_path), script)
+            self.assertIn(repr(results_path), script)
+
+    def test_apostrophe_in_path_compiles(self):
+        """A quote in a directory name must stay inside the string literal"""
+        program_path = r"C:\Users\o'brien\prog.py"
+        results_path = program_path + ".results"
+        for evaluator in self.evaluators:
+            script = evaluator._subprocess_script(program_path, results_path)
+            compile(script, "<circle-packing-runner>", "exec")
+
+    def test_run_with_timeout_executes_program_from_user_temp(self):
+        """The documented subprocess path must return the program's packing"""
+        program = textwrap.dedent("""
+            import numpy as np
+
+            def run_packing():
+                centers = np.array([[0.5, 0.5]])
+                radii = np.array([0.1])
+                return centers, radii, float(radii.sum())
+            """)
+        with tempfile.NamedTemporaryFile(
+            mode="w", suffix=".py", delete=False, encoding="utf-8"
+        ) as handle:
+            handle.write(program)
+            program_path = handle.name
+
+        try:
+            for evaluator in self.evaluators:
+                centers, radii, reported_sum = evaluator.run_with_timeout(
+                    program_path, timeout_seconds=30
+                )
+                self.assertEqual(centers.shape, (1, 2))
+                self.assertEqual(float(radii[0]), 0.1)
+                self.assertAlmostEqual(float(reported_sum), 0.1)
+        finally:
+            os.unlink(program_path)
+
+
+if __name__ == "__main__":
+    unittest.main()