⚡ Bolt: Avoid eager list allocation in recursive AST traversal

.jules/bolt.md

@@ -0,0 +1,3 @@
+## 2024-03-24 - AST Iteration Allocation Overhead
+**Learning:** In the static analysis tools processing large ASTs recursively, eagerly materializing `ast.iter_child_nodes()` or `func_node.body` into lists (e.g., `list(ast.iter_child_nodes(node))`) introduces significant unnecessary memory allocation and garbage collection overhead. Since the traversal functions `_collect_return_paths` and `_assignment_callee` only iterate through the children exactly once, this eager list conversion is completely avoidable.
+**Action:** Always accept `Iterable[ast.AST]` in recursive AST walker functions and pass the generator directly (`ast.iter_child_nodes()`) to avoid O(N) list allocation overhead at every node depth.

src/wardline/scanner/taint/variable_level.py

@@ -33,7 +33,7 @@
 from wardline.core.taints import _PROVENANCE_CLASH, RAW_ZONE, TRUST_RANK, TaintState, combine
 
 if TYPE_CHECKING:
-    from collections.abc import Iterator
+    from collections.abc import Iterable, Iterator
 
 # Serialisation sinks — calls that cross the representation boundary. Their
 # output sheds validation provenance (raw bytes/str), so → UNKNOWN_RAW. This is
@@ -2414,7 +2414,7 @@ def compute_return_taint(
     (the function's anchored *body* taint, pinned to its declaration).
     """
     returns: list[tuple[TaintState, str | None, ast.expr]] = []
-    _collect_return_paths(list(func_node.body), function_taint, taint_map, var_taints, returns, return_snapshots)
+    _collect_return_paths(func_node.body, function_taint, taint_map, var_taints, returns, return_snapshots)
     if not returns:
         return None
     result = returns[0][0]
@@ -2454,7 +2454,7 @@ def compute_return_callee(
     beyond one hop stay ``None`` (the N-hop walk lives in the Loomweave stored-fact path).
     """
     returns: list[tuple[TaintState, str | None, ast.expr]] = []
-    _collect_return_paths(list(func_node.body), function_taint, taint_map, var_taints, returns, return_snapshots)
+    _collect_return_paths(func_node.body, function_taint, taint_map, var_taints, returns, return_snapshots)
     if not returns:
         return None
     worst = returns[0][0]
@@ -2468,14 +2468,14 @@ def compute_return_callee(
     #    a direct call. Provenance only — never changes a fire/no-fire decision.
     for taint, callee, node in returns:
         if taint == worst and callee is None and isinstance(node, ast.Name):
-            indirect = _assignment_callee(list(func_node.body), node.id, worst, function_taint, taint_map, var_taints)
+            indirect = _assignment_callee(func_node.body, node.id, worst, function_taint, taint_map, var_taints)
             if indirect is not None:
                 return indirect
     return None
 
 
 def _assignment_callee(
-    nodes: list[ast.AST],
+    nodes: Iterable[ast.AST],
     name: str,
     worst: TaintState,
     function_taint: TaintState,
@@ -2508,9 +2508,7 @@ def _assignment_callee(
                 and _resolve_expr(node.value, function_taint, taint_map, var_taints) == worst
             ):
                 result = callee
-        nested = _assignment_callee(
-            list(ast.iter_child_nodes(node)), name, worst, function_taint, taint_map, var_taints
-        )
+        nested = _assignment_callee(ast.iter_child_nodes(node), name, worst, function_taint, taint_map, var_taints)
         if nested is not None:
             result = nested
     return result
@@ -2527,7 +2525,7 @@ def _return_callee(node: ast.expr) -> str | None:
 
 
 def _collect_return_paths(
-    nodes: list[ast.AST],
+    nodes: Iterable[ast.AST],
     function_taint: TaintState,
     taint_map: dict[str, TaintState],
     var_taints: dict[str, TaintState],
@@ -2558,7 +2556,7 @@ def _collect_return_paths(
             taint = _resolve_expr(node.value, function_taint, taint_map, dict(snapshot or var_taints))
             out.append((taint, _return_callee(node.value), node.value))
         _collect_return_paths(
-            list(ast.iter_child_nodes(node)),
+            ast.iter_child_nodes(node),
             function_taint,
             taint_map,
             var_taints,