first commit

backend/app/dataset/rules/python/null_pointer.py

@@ -0,0 +1,65 @@
+"""AST-level null-pointer / None-reference injection for Python."""
+
+import ast
+from typing import Optional
+
+from app.dataset.rules.base import Mutation, MutationRule
+
+
+class NoneReferenceRule(MutationRule):
+    """Replace a checked variable access with an unchecked None dereference.
+
+    This rule uses the standard library `ast` module to precisely locate a
+    variable that is used after an `if x is not None:` guard, then removes the
+    guard. The mutation position is derived from AST line numbers so it is
+    exact and reproducible.
+    """
+
+    name = "none_reference"
+    language = "python"
+    defect_type = "null_pointer"
+
+    def detect_and_mutate(self, source: str, filename: str = "") -> Optional[Mutation]:
+        try:
+            tree = ast.parse(source)
+        except SyntaxError:
+            return None
+
+        for node in ast.walk(tree):
+            if not isinstance(node, ast.If):
+                continue
+            test = node.test
+            # Match: if x is not None:
+            if (
+                isinstance(test, ast.Compare)
+                and isinstance(test.left, ast.Name)
+                and len(test.ops) == 1
+                and isinstance(test.ops[0], ast.IsNot)
+                and len(test.comparators) == 1
+                and isinstance(test.comparators[0], ast.Constant)
+                and test.comparators[0].value is None
+            ):
+                var_name = test.left.id
+                lines = source.splitlines(keepends=True)
+                start = node.lineno
+                end = getattr(node, "end_lineno", node.lineno) or node.lineno
+                body_lines = lines[start:end]
+                dedented = []
+                for line in body_lines[1:]:
+                    if line.startswith("    "):
+                        dedented.append(line[4:])
+                    elif line.startswith("\t"):
+                        dedented.append(line[1:])
+                    else:
+                        dedented.append(line)
+                mutated = "".join(lines[: start - 1] + dedented + lines[end:])
+                return Mutation(
+                    defect_type=self.defect_type,
+                    language=self.language,
+                    line_start=start,
+                    line_end=end,
+                    mutated_source=mutated,
+                    reference_fix=f"Add `if {var_name} is not None:` guard before use.",
+                    description=f"Removed None-check guard for variable '{var_name}'.",
+                )
+        return None