fix(extraction): land upstream declaration initializer walks (#1511) (#1802)

Squash danusha2345's PR #1511 at d282f9e8 onto main 8c9c4761,
preserving its nine non-merge commits and main's existing Unreleased notes.
Calls in Kotlin, Java, TS/JS, Scala, Rust and Python declaration initializers
now retain the owner established by the upstream regression expectations.
Include the upstream CFML, dynamic-dispatch summary and viewer follow-ups.

Linux fail-to-pass validation (Node 22.19.0, rebuilt dist and native kernel):
- Before: TS load belonged to file:app.ts; Python/Kotlin/Scala/Rust calls
  vanished; Java lost the field-lambda, anonymous override and eager calls.
- After: all six languages PASS; 12 native/WASM LF/CRLF parity checks PASS.
- Focused initializer regressions: 10 passed with CODEGRAPH_KERNEL=0 and
  10 passed with the kernel enabled; Kotlin's grammar fallback is recorded.
- Related regression suites: 879 passed, 1 skipped across 15 test files.
- Evidence: /workspace/cg-1510-repro/before and /workspace/cg-1510-repro/after
  (combined test output: after/vitest.log).

Fixes #1510
Supersedes #1511

Co-authored-by: Colby McHenry <colbymchenry@users.noreply.github.com>
Co-authored-by: danusha2345 <ewidusoc498@gmail.com>

__tests__/extraction.test.ts

@@ -1094,6 +1094,42 @@ const token = getTokenMp();
     );
     expect(call).toBeDefined();
   });
+
+  describe('initializer walk is scoped to the declared symbol (#693 for TS/JS)', () => {
+    const code = `
+const eager = load();
+const obj = { handler: () => target(), plain: target() };
+const list = [() => target()];
+export const exported = { handler: () => target() };
+`;
+    const callersOf = (name: string) => {
+      const result = extractFromSource('app.ts', code);
+      const byId = new Map(result.nodes.map((n) => [n.id, n]));
+      return result.unresolvedReferences
+        .filter((u) => u.referenceKind === 'calls' && u.referenceName === name)
+        .map((u) => byId.get(u.fromNodeId))
+        .map((n) => (n ? `${n.kind}:${n.name}` : '?'))
+        .sort();
+    };
+
+    it("a plain call initializer names the CONSTANT as caller, not the file", () => {
+      // The walk ran with only the file on the stack, so `load` recorded the
+      // file as its caller — useless for callers/impact.
+      expect(callersOf('load')).toEqual(['constant:eager']);
+    });
+
+    it('a non-exported object literal contributes calls (it was skipped outright)', () => {
+      // `exported`'s members are minted as their own function nodes, so its
+      // arrow's call comes from `handler`; the non-exported ones attribute to
+      // the declared constant.
+      expect(callersOf('target')).toEqual([
+        'constant:list',
+        'constant:obj',
+        'constant:obj',
+        'function:handler',
+      ]);
+    });
+  });
 });
 
 describe('File Node Extraction', () => {
@@ -1182,6 +1218,42 @@ class UserService:
     expect(classNode).toBeDefined();
     expect(classNode?.name).toBe('UserService');
   });
+
+  it('walks a module-level assignment initializer scoped to the name (#693 for Python)', () => {
+    // The assignment minted a node and stopped, so everything a module builds
+    // at import time — `app = FastAPI()`, `ENGINE = create_engine(url)` — was
+    // missing from the graph. A tuple target mints no symbol, so its
+    // right-hand side attributes to the enclosing scope instead of vanishing.
+    const code = `
+def target(): pass
+def compute(): return 1
+
+APP = compute()
+handler = lambda: target()
+MAPPING = {"a": compute()}
+first, second = compute(), target()
+
+class K:
+    ATTR = compute()
+`;
+    const result = extractFromSource('app.py', code);
+    const byId = new Map(result.nodes.map((n) => [n.id, n]));
+    const owners = result.unresolvedReferences
+      .filter((u) => u.referenceKind === 'calls')
+      .map((u) => {
+        const n = byId.get(u.fromNodeId);
+        return `${u.referenceName}<-${n ? `${n.kind}:${n.name}` : '?'}`;
+      })
+      .sort();
+    expect(owners).toEqual([
+      'compute<-class:K', // a class attribute still rides the class (no node of its own)
+      'compute<-file:app.py', // the tuple target mints nothing
+      'compute<-variable:APP',
+      'compute<-variable:MAPPING',
+      'target<-file:app.py',
+      'target<-variable:handler',
+    ]);
+  });
 });
 
 describe('Go Extraction', () => {
@@ -1507,6 +1579,26 @@ impl Counter {
     expect(implRefs).toHaveLength(0);
   });
 
+  it('walks a const/static initializer scoped to the declared symbol (#693 for Rust)', () => {
+    // The declaration minted a node and stopped, so a handler table, a
+    // lazily-built singleton or any computed const linked to nothing.
+    const code = `
+const LEN: usize = compute_len();
+static REGISTRY: Lazy<Cfg> = Lazy::new(|| build_cfg());
+`;
+    const result = extractFromSource('lib.rs', code);
+    const byId = new Map(result.nodes.map((n) => [n.id, n]));
+    const owner = (name: string) => {
+      const u = result.unresolvedReferences.find(
+        (r) => r.referenceKind === 'calls' && r.referenceName === name
+      );
+      const n = u ? byId.get(u.fromNodeId) : undefined;
+      return n ? `${n.kind}:${n.name}` : undefined;
+    };
+    expect(owner('compute_len')).toBe('variable:LEN');
+    expect(owner('build_cfg')).toBe('variable:REGISTRY');
+  });
+
   it('should extract union declarations and their impl edges', () => {
     const code = `
 pub union Reg {
@@ -1714,6 +1806,37 @@ public class Splitter {
     );
     expect(sepStart, 'override inside the lambda-returned anon class should be a method node').toBeDefined();
   });
+
+  it('walks a field initializer scoped to the field (#693 for Java)', () => {
+    // The dispatcher only scanned a field_declaration for function-as-value
+    // candidates, so a lambda or anonymous class holding the work — the
+    // Android listener idiom — contributed no call edge and `target` looked
+    // callerless.
+    const code = `
+package p;
+class T {
+    private final Runnable fieldLambda = () -> target();
+    private final Runnable anonClass = new Runnable() {
+        public void run() { target(); }
+    };
+    private final int eager = compute();
+    void directCall() { target(); }
+    private void target() {}
+    private static int compute() { return 1; }
+}
+`;
+    const result = extractFromSource('T.java', code);
+    const byId = new Map(result.nodes.map((n) => [n.id, n]));
+    const callersOf = (name: string) =>
+      result.unresolvedReferences
+        .filter((u) => u.referenceKind === 'calls' && u.referenceName === name)
+        .map((u) => byId.get(u.fromNodeId)?.name)
+        .sort();
+
+    // `run` is the anonymous class's override, itself extracted under the field.
+    expect(callersOf('target')).toEqual(['directCall', 'fieldLambda', 'run']);
+    expect(callersOf('compute')).toEqual(['eager']);
+  });
 });
 
 describe('C# Extraction', () => {
@@ -2317,6 +2440,120 @@ class Bar {
     const cls = result.nodes.find((n) => n.kind === 'class' && n.name === 'Bar');
     expect(cls?.qualifiedName).toBe('Bar');
   });
+
+  describe('property initializers are walked, attributed to the property (#693 for Kotlin)', () => {
+    // The property hook consumes the whole property_declaration subtree, so
+    // before this the initializer was only scanned for function-as-value
+    // candidates and every call inside it vanished from the graph. Android/MSDK
+    // callbacks are declared exactly this way (`private val l = Listener { … }`),
+    // so anything reached only through one looked like it had no callers at all.
+    const code = `
+package repro
+
+class Repro {
+    private val fieldLambda: () -> Unit = { target() }
+    private val samField = Runnable { target() }
+    private val plain = target()
+    private val delegated by lazy { target() }
+    private val anonObject = object : Runnable { override fun run() { target() } }
+
+    fun directCall() { target() }
+    fun lambdaInMethod() { run { target() } }
+
+    private fun target() {}
+}
+
+object Holder {
+    val topLevelLambda: () -> Unit = { hit() }
+    private fun hit() {}
+}
+`;
+    const callersOf = (target: string) => {
+      const result = extractFromSource('Repro.kt', code);
+      const byId = new Map(result.nodes.map((n) => [n.id, n]));
+      return result.unresolvedReferences
+        .filter((u) => u.referenceKind === 'calls' && u.referenceName === target)
+        .map((u) => byId.get(u.fromNodeId)?.name)
+        .sort();
+    };
+
+    it('a lambda / SAM / plain / delegated / object initializer calls FROM the property', () => {
+      // `run` is the anonymous object's override, extracted as its own node
+      // under `anonObject` — the same shape Go's initializer walk produces.
+      expect(callersOf('target')).toEqual([
+        'delegated',
+        'directCall',
+        'fieldLambda',
+        'lambdaInMethod',
+        'plain',
+        'run',
+        'samField',
+      ]);
+    });
+
+    it('a property in an `object` singleton is a caller too', () => {
+      expect(callersOf('hit')).toEqual(['topLevelLambda']);
+    });
+
+    it('an accessor body belongs to its property, written on either line', () => {
+      // `val x: T get() = …` nests the accessor UNDER the declaration; written
+      // on its own line the grammar makes it a following SIBLING instead. Both
+      // used to lose their calls (the nested one) or hand them to the enclosing
+      // class (the sibling); both now attribute to the property.
+      const src = `
+package p
+
+class C {
+    val sameLine: Int get() = compute()
+    val nextLine: Int
+        get() = compute()
+    var written: Int = 0
+        set(v) { store(v) }
+    private fun compute(): Int = 1
+    private fun store(v: Int) {}
+}
+`;
+      const result = extractFromSource('C.kt', src);
+      const byId = new Map(result.nodes.map((n) => [n.id, n]));
+      const ownersOf = (name: string) =>
+        result.unresolvedReferences
+          .filter((u) => u.referenceKind === 'calls' && u.referenceName === name)
+          .map((u) => {
+            const n = byId.get(u.fromNodeId);
+            return n ? `${n.kind}:${n.name}` : '?';
+          })
+          .sort();
+      expect(ownersOf('compute')).toEqual(['field:nextLine', 'field:sameLine']);
+      expect(ownersOf('store')).toEqual(['field:written']);
+    });
+
+    it('an `init` block and a destructuring RHS no longer vanish', () => {
+      // Both mint no symbol of their own, so the hook consumed them and their
+      // code disappeared entirely; they now attribute to the enclosing scope.
+      const src = `
+package p
+
+class C {
+    init { val q = initCall() }
+    val (a, b) = makePair()
+}
+
+val (t1, t2) = topMakePair()
+`;
+      const result = extractFromSource('C.kt', src);
+      const byId = new Map(result.nodes.map((n) => [n.id, n]));
+      const owner = (name: string) => {
+        const u = result.unresolvedReferences.find(
+          (r) => r.referenceKind === 'calls' && r.referenceName === name
+        );
+        const n = u ? byId.get(u.fromNodeId) : undefined;
+        return n ? `${n.kind}:${n.name}` : undefined;
+      };
+      expect(owner('initCall')).toBe('class:C');
+      expect(owner('makePair')).toBe('class:C');
+      expect(owner('topMakePair')).toBe('namespace:p');
+    });
+  });
 });
 
 describe('Dart Extraction', () => {
@@ -8294,6 +8531,35 @@ def processData(): Unit = {
       const calls = result.unresolvedReferences.filter((r) => r.referenceKind === 'calls');
       expect(calls.length).toBeGreaterThan(0);
     });
+
+    it('walks a val/var initializer scoped to the declared symbol (#693 for Scala)', () => {
+      // The val/var hook minted the node and returned true, so the dispatcher
+      // only scanned the subtree for function-as-value candidates — every call
+      // in an initializer was dropped, which on a `val`-heavy codebase
+      // (SpinalHDL, Akka wiring) is most of the wiring.
+      const code = `
+class C {
+  val fieldLambda: () => Unit = () => target()
+  val direct = target()
+  lazy val lazily = target()
+  private def target(): Unit = {}
+}
+
+object O {
+  val topLambda = () => hit()
+  def hit(): Unit = {}
+}
+`;
+      const result = extractFromSource('C.scala', code);
+      const byId = new Map(result.nodes.map((n) => [n.id, n]));
+      const callersOf = (name: string) =>
+        result.unresolvedReferences
+          .filter((u) => u.referenceKind === 'calls' && u.referenceName === name)
+          .map((u) => byId.get(u.fromNodeId)?.name)
+          .sort();
+      expect(callersOf('target')).toEqual(['direct', 'fieldLambda', 'lazily']);
+      expect(callersOf('hit')).toEqual(['topLambda']);
+    });
   });
 });
 

__tests__/fixtures/kernel-parity/Torture.java

@@ -22,6 +22,15 @@ public class TortureService extends BaseService implements Runnable, AutoCloseab
   protected int count = 0;
   private final List<String> names;
   int packagePrivate, secondDeclarator;
+  /** Field initializers — walked scoped to the field (#693). */
+  private final Runnable fieldLambda = () -> helper(RETRY_LIMITS);
+  private final Runnable fieldAnonClass = new Runnable() {
+    @Override
+    public void run() {
+      helper(RETRY_LIMITS);
+    }
+  };
+  private final Runnable fieldMethodRef = TortureService::compute;
 
   /** Ctor javadoc. */
   public TortureService(List<String> names) {

__tests__/fixtures/kernel-parity/torture.js

@@ -74,6 +74,11 @@ export default {
   },
 };
 
+// Initializer walks attributed to the declared symbol (#693). A plain call
+// leaked to the FILE node; a non-exported object literal was skipped outright.
+const eagerConfig = loadConfig();
+const handlerMap = { onSave: () => persist(eagerConfig), onLoad: loadConfig() };
+const lazyList = [() => persist(eagerConfig)];
 // --- CommonJS export assignments (#1675) -----------------------------------
 exports.getItems = async (req, res) => { res.json(await findItems()); };
 module.exports.deleteItem = function (req, res) { removeItem(req.params.id); res.end(); };

__tests__/fixtures/kernel-parity/torture.kt

@@ -50,6 +50,13 @@ val topDelegated by lazy { WidgetK(1) }
 val (destA, destB) = makePair()
 val withGetter: Int
     get() = 42
+val initLambda: () -> Unit = { caller() }
+val initSam = Runnable { caller() }
+val initObject = object : Runnable {
+    override fun run() {
+        caller()
+    }
+}
 
 class WidgetK(val size: Int, private var name: String = defaultName()) {
     val area: Int = size * size
@@ -265,3 +272,20 @@ fun labeledLambda() {
 }
 
 fun whereClause(): Int where Int : Comparable<Int> = 1
+
+class AccessorK {
+    val sameLineGetter: Int get() = compute()
+    var sameLinePair: Int get() = compute()
+        set(v) { draw(v) }
+}
+
+class SiblingAccessorK {
+    var nextLine: Int = 0
+        get() = compute()
+        set(v) { draw(v) }
+    val (localA, localB) = makePair()
+    init {
+        val fromInit = compute()
+        register(fromInit)
+    }
+}

__tests__/fixtures/kernel-parity/torture.py

@@ -48,6 +48,11 @@ def shadowed():
 handlers = {"recv": target_cb}
 callbacks = [target_cb, view]
 
+# Initializer walks attributed to the assigned name (#693).
+INIT_EAGER = helper()
+INIT_LAMBDA = lambda: target_cb()
+INIT_MAP = {"a": helper()}
+init_a, init_b = helper(), view()
 
 # --- call receivers (#1683) ---------------------------------------------------
 def bucket_chains(d, k, v):

__tests__/fixtures/kernel-parity/torture.rs

@@ -285,6 +285,11 @@ fn mount() {
 
 routes![top_level_h];
 
+// Initializer walks attributed to the declared symbol (#693).
+const INIT_CONST: usize = compute_len();
+static INIT_LAZY: Lazy<Cfg> = Lazy::new(|| build_cfg());
+static INIT_ALIAS: fn() = free_fn;
+
 pub union Reg {
     pub raw: u32,
     pub halves: [u16; 2],

__tests__/fixtures/kernel-parity/torture.scala

@@ -175,3 +175,10 @@ package object utilpkg {
   def pkgHelper(): Int = 1
   val pkgShared = 2
 }
+
+class InitWalk {
+  val initLambda: () => Unit = () => helperCall()
+  val initDirect = helperCall()
+  lazy val initLazy = process(1)
+  val initAnon = new Runnable { def run(): Unit = helperCall() }
+}

__tests__/function-ref.test.ts

@@ -795,8 +795,10 @@ describe('Function-as-value capture (#756)', () => {
 
       // The DRF wiring: get_serializer_class → the imported serializer class,
       // via `return` — the issue's headline gap. The module-level registry
-      // dict rides the file node.
+      // dict rides BOTH the assigned name (the initializer walk, #693) and the
+      // file node (the dispatcher's own scan, which runs either way).
       expect(sourceNames(cg, fnRefEdgesInto(cg, 'OrgSerializerFull'))).toEqual([
+        'SERIALIZER_REGISTRY',
         'get_serializer_class',
         'views.py',
       ]);

__tests__/kernel-kotlin-parity.test.ts

@@ -5,7 +5,9 @@
  * compiled from the vendored fwcd 0.3.8 C sources, the arc's first
  * vendored-grammar-C language) produces the SAME ExtractionResult as the
  * wasm TreeSitterExtractor over the checked-in torture fixture (torture.kt:
- * the property hook's scope classification, extension-function receiver QNs
+ * the property hook's scope classification and its initializer walk (a
+ * lambda / SAM / anonymous-object RHS attributing its calls to the property),
+ * extension-function receiver QNs
  * (`WidgetK::extend`, the qualified `com::qext` bug) + the owner-contains
  * fallback, expect/actual → node DECORATORS (the KMP synthesizer feed),
  * the bodiless-vs-bodied class header asymmetry, comment-glued