Merge pull request #2597 from magyargergo/fix/2564-record-newexpr-callgraph

fix(java): record container node + new-expression chained call receiver typing (#2564)

gitnexus/bench/scope-capture/baselines.json

@@ -89,7 +89,7 @@
     "_rebaselined": "#1919 review CF3 fix: extended kotlin-local-property-owner (init/accessor destructuring) + new dart-accessor-owner fixture (getter/setter ownership). Fingerprint-only corpus drift; scaling ~1.0."
   },
   "java": {
-    "fingerprint": "975b68aaac6d06094260fb0c67f9b1bc03692ba7220669d192aca9dccd5fc0ca",
+    "fingerprint": "85fc7af9c3c1bceac76cb4f27214410b04967682a2eaa7e468e26efd1f4e2537",
     "scaling_budget": 1.5,
     "_rebaselined_callable_flow_2522_review": "PR #2522 review hardening: callable operands retain expression/qualified identity and formals retain signature metadata; same-name lexical regions use an O(ancestor-depth) ID-set lookup. Prior d5c59d7dc9e206637515d5aea1163f7c1cdd76410c38c5fe6143d13d19677d6a -> 004a3592998dca1193bd1429a8284513725de7764f2a3eceedaaa984cfd763b4; scaling 0.992 < 1.5.",
     "_rebaselined_callable_flow_2522_followup": "PR #2522 follow-up: Java method-reference/SAM callable flow facts with invocation-result suppression. Prior 062d754764aaa8a6772fb90875c710502a63e3e7a300e633942381ed914faada -> d5c59d7dc9e206637515d5aea1163f7c1cdd76410c38c5fe6143d13d19677d6a; scaling 1.074 < 1.5.",
@@ -97,7 +97,8 @@
     "_note": "#1928 / #2045: F35 adds qualified + qualified-generic constructor query captures (`new pkg.Foo()`, `new a.b.Foo()`, `new pkg.Box<T>()`); F38 synthesizes `@reference.call.constructor` on `super(...)`/`this(...)` explicit_constructor_invocation nodes; F41 generic-aware stripQualifier in interpret (type-binding normalization). + java-qualified-constructor and java-explicit-constructor fixtures. Pure capture-additive + fixture-corpus drift; scaling stays linear (~1.06).",
     "_rebaselined_2522_review_fixes": "PR #2522 review fixes: get/test dropped from callableProtocolMethods. Prior 004a3592998dca1193bd1429a8284513725de7764f2a3eceedaaa984cfd763b4 -> f3b4f4b6610e07c3ac90deb1c53d3572b6ad55a36e5d7134984876d30031ff67; scaling ratio re-verified within budget.",
     "_rebaselined_2550_instance_model": "PR #2549 (#2550): anonymous class bodies emit synthesized @declaration.class/@declaration.name (Worker$N), an @reference.inherits to the constructed type, and receiver @type-binding.* captures; six new java-* fixtures joined the corpus. Prior f3b4f4b6610e07c3ac90deb1c53d3572b6ad55a36e5d7134984876d30031ff67 -> d79c3b92acfc866094981499b977388ca14f90839bca0c040342ab1cec00aa90; scaling 1.058 < 1.5.",
-    "_rebaselined_2555_enum_constant_bodies": "PR for #2555: enum constant bodies emit synthesized E$N classes + @reference.inherits to the host enum; anonymous naming follows JLS 13.1 immediately-enclosing-type chains INCLUDING anonymous enclosing types (NestHost$1$1, N$1$1); six new java-* fixtures joined the corpus. Prior d79c3b92acfc866094981499b977388ca14f90839bca0c040342ab1cec00aa90 -> 975b68aaac6d06094260fb0c67f9b1bc03692ba7220669d192aca9dccd5fc0ca; scaling 1.05 < 1.5."
+    "_rebaselined_2555_enum_constant_bodies": "PR for #2555: enum constant bodies emit synthesized E$N classes + @reference.inherits to the host enum; anonymous naming follows JLS 13.1 immediately-enclosing-type chains INCLUDING anonymous enclosing types (NestHost$1$1, N$1$1); six new java-* fixtures joined the corpus. Prior d79c3b92acfc866094981499b977388ca14f90839bca0c040342ab1cec00aa90 -> 975b68aaac6d06094260fb0c67f9b1bc03692ba7220669d192aca9dccd5fc0ca; scaling 1.05 < 1.5.",
+    "_rebaselined_2564_record_capture": "PR for #2564: JAVA_QUERIES gained a (record_declaration name: (identifier) @name) @definition.record capture, previously entirely missing (record_declaration had no structure-phase capture at all, unlike class/interface/enum) - a record's methods existed as ownerless Method nodes with no HAS_METHOD edge. Two new java-* fixtures (java-record-methods, java-new-expr-chain-call) joined the corpus. Prior 975b68aaac6d06094260fb0c67f9b1bc03692ba7220669d192aca9dccd5fc0ca -> 85fc7af9c3c1bceac76cb4f27214410b04967682a2eaa7e468e26efd1f4e2537; scaling 1.059 < 1.5."
   },
   "typescript": {
     "fingerprint": "3280b13d3f9378ab23eee31c2edc779b5a9ae1e7bb510c23a24855b44406d2f4",

gitnexus/src/core/ingestion/languages/java/captures.ts

@@ -150,6 +150,29 @@ export function emitJavaScopeCaptures(
       continue;
     }
 
+    // Normalize a `new`-expression receiver to its constructed type's simple
+    // name: `new Local().inner()` binds the WHOLE `object_creation_expression`
+    // as `@reference.receiver`, so its raw text is `"new Local()"` — a string
+    // that can never match a scope binding, so the compound-receiver resolver
+    // silently falls through to name-only fallback resolution and picks the
+    // wrong same-named method on a collision (#2564). Rewriting the text to
+    // just `Local` lets Case 2 (class-name / static receiver) in
+    // receiver-bound-calls.ts resolve it via its normal MRO walk. Mirrors the
+    // established `normalizePhpReceiver` precedent (php/captures.ts) — a
+    // language-local capture rewrite, no shared-pipeline change.
+    if (grouped['@reference.receiver'] !== undefined) {
+      const receiverNode = nodeIfType(nodeMap['@reference.receiver'], 'object_creation_expression');
+      const typeNode = receiverNode?.childForFieldName('type');
+      const simpleName = typeNode ? javaBaseSimpleNameOf(typeNode) : undefined;
+      if (simpleName !== undefined) {
+        grouped['@reference.receiver'] = syntheticCapture(
+          '@reference.receiver',
+          receiverNode!,
+          simpleName,
+        );
+      }
+    }
+
     // Filter read.member when it's a child of method_invocation or assignment.
     // `@reference.read.member` is captured directly on the `field_access` node.
     if (grouped['@reference.read.member'] !== undefined) {

gitnexus/src/core/ingestion/tree-sitter-queries.ts

@@ -743,10 +743,11 @@ export const PYTHON_QUERIES = `
 
 // Java queries - works with tree-sitter-java
 export const JAVA_QUERIES = `
-; Classes, Interfaces, Enums, Annotations
+; Classes, Interfaces, Enums, Records, Annotations
 (class_declaration name: (identifier) @name) @definition.class
 (interface_declaration name: (identifier) @name) @definition.interface
 (enum_declaration name: (identifier) @name) @definition.enum
+(record_declaration name: (identifier) @name) @definition.record
 (annotation_type_declaration name: (identifier) @name) @definition.annotation
 
 ; Anonymous class bodies: new Runnable() { ... } — no @name capture; the

gitnexus/src/storage/repo-manager.ts

@@ -429,8 +429,16 @@ export interface RepoMeta {
  * `E.hook` to `E$1.hook`, and nested-host anonymous names re-key
  * (`EnumWrap$1` → `EnumWrap$Mode$1`). Same contract as v8: identities move
  * on unchanged files; force a full re-analyze.
+ * v10: Java `record_declaration` now emits a first-class `Record` graph node
+ * (#2564): a record's container node was previously never created (JAVA_QUERIES
+ * had no capture for it), so its methods existed as ownerless Method nodes
+ * with no `HAS_METHOD` edge. The incremental write set only covers changed
+ * files — a top-up against a pre-v10 index would keep silently omitting the
+ * `Record` node and its `HAS_METHOD` edges for every unchanged record file
+ * (same v7 contract: new nodes/edges the incremental path would otherwise
+ * never backfill); force a full re-analyze instead.
  */
-export const INCREMENTAL_SCHEMA_VERSION = 9;
+export const INCREMENTAL_SCHEMA_VERSION = 10;
 
 export interface IndexedRepo {
   repoPath: string;

gitnexus/test/fixtures/lang-resolution/java-new-expr-chain-call/LocalChain.java

@@ -0,0 +1,12 @@
+package probe;
+
+public class LocalChain {
+    void m() {
+        class Local {
+            void inner() {
+                System.out.println("right target");
+            }
+        }
+        new Local().inner();
+    }
+}

gitnexus/test/fixtures/lang-resolution/java-new-expr-chain-call/Other.java

@@ -0,0 +1,7 @@
+package probe;
+
+class Other {
+    void inner() {
+        System.out.println("wrong target");
+    }
+}

gitnexus/test/fixtures/lang-resolution/java-record-methods/Point.java

@@ -0,0 +1,11 @@
+package probe;
+
+public record Point(int x, int y) {
+    public int sum() {
+        return x + y;
+    }
+
+    public int scaled(int factor) {
+        return sum() * factor;
+    }
+}

gitnexus/test/integration/resolvers/java.test.ts

@@ -1174,6 +1174,75 @@ describe('Java chained method call resolution', () => {
   });
 });
 
+// ---------------------------------------------------------------------------
+// Chained call on a new-expression receiver: new Local().inner()
+// The receiver of inner() is an object_creation_expression, not a variable.
+// Regression test for #2564: without treating `new Local()` as a typed
+// receiver, the call falls back to name-only resolution and can pick an
+// unrelated same-named method (Other.inner) instead of Local.inner.
+// ---------------------------------------------------------------------------
+
+describe('Java chained call on a new-expression receiver (#2564)', () => {
+  let result: PipelineResult;
+
+  beforeAll(async () => {
+    result = await runPipelineFromRepo(path.join(FIXTURES, 'java-new-expr-chain-call'), () => {});
+  }, 60000);
+
+  it('detects LocalChain and Other classes', () => {
+    const classes = getNodesByLabel(result, 'Class');
+    expect(classes).toContain('LocalChain');
+    expect(classes).toContain('Other');
+  });
+
+  it('resolves new Local().inner() to the local Local#inner, NOT Other#inner', () => {
+    const calls = getRelationships(result, 'CALLS');
+    const localInner = calls.find(
+      (c) =>
+        c.target === 'inner' && c.source === 'm' && c.targetFilePath.includes('LocalChain.java'),
+    );
+    const otherInner = calls.find(
+      (c) => c.target === 'inner' && c.source === 'm' && c.targetFilePath.includes('Other.java'),
+    );
+    expect(localInner).toBeDefined();
+    expect(otherInner).toBeUndefined();
+  });
+});
+
+// ---------------------------------------------------------------------------
+// Java record: container node + HAS_METHOD edges
+// Regression test for #2564: JAVA_QUERIES previously had no @definition.record
+// capture, so a record never got a Class/Record graph node — its methods
+// existed as ownerless orphans with no HAS_METHOD edge.
+// ---------------------------------------------------------------------------
+
+describe('Java record method resolution (#2564)', () => {
+  let result: PipelineResult;
+
+  beforeAll(async () => {
+    result = await runPipelineFromRepo(path.join(FIXTURES, 'java-record-methods'), () => {});
+  }, 60000);
+
+  it('detects a Record node for Point', () => {
+    const records = getNodesByLabel(result, 'Record');
+    expect(records).toContain('Point');
+  });
+
+  it('emits HAS_METHOD edges linking sum and scaled to Point', () => {
+    const hasMethod = getRelationships(result, 'HAS_METHOD');
+    const sumEdge = hasMethod.find((e) => e.source === 'Point' && e.target === 'sum');
+    const scaledEdge = hasMethod.find((e) => e.source === 'Point' && e.target === 'scaled');
+    expect(sumEdge).toBeDefined();
+    expect(scaledEdge).toBeDefined();
+  });
+
+  it('resolves scaled() calling sum() via a CALLS edge', () => {
+    const calls = getRelationships(result, 'CALLS');
+    const sumCall = calls.find((c) => c.target === 'sum' && c.source === 'scaled');
+    expect(sumCall).toBeDefined();
+  });
+});
+
 // ---------------------------------------------------------------------------
 // Java 16+ instanceof pattern variable: `if (obj instanceof User user)`
 // Phase 5.2: extractPatternBinding on instanceof_expression binds user → User.

gitnexus/test/unit/call-summary-schema-version.test.ts

@@ -73,8 +73,8 @@ describe('CALL_SUMMARY relation-type exclusion (U-C1)', () => {
 });
 
 describe('CALL_SUMMARY incremental reuse gate (U-C5)', () => {
-  it('INCREMENTAL_SCHEMA_VERSION is bumped to 9 (Java enum-constant-body + JLS-naming re-index window)', () => {
-    expect(INCREMENTAL_SCHEMA_VERSION).toBe(9);
+  it('INCREMENTAL_SCHEMA_VERSION is bumped to 10 (Java record container-node re-index window)', () => {
+    expect(INCREMENTAL_SCHEMA_VERSION).toBe(10);
   });
 
   it('a pre-current stamp fails the `=== INCREMENTAL_SCHEMA_VERSION` reuse gate → forces full re-analyze', () => {
@@ -108,7 +108,11 @@ describe('CALL_SUMMARY incremental reuse gate (U-C5)', () => {
     // topmost-anchored `EnumWrap$1`-style ids would be stranded alongside
     // the re-keyed ones on unchanged files → must NOT reuse.
     expect(passesReuseGate(8)).toBe(false);
+    // A pre-v10 (v9) index predates the Java record container-node fix
+    // (#2564) — a record's methods would keep being ownerless Method nodes
+    // with no HAS_METHOD edge on unchanged files → must NOT reuse.
+    expect(passesReuseGate(9)).toBe(false);
     // A current-version stamp passes the gate (incremental top-up eligible).
-    expect(passesReuseGate(9)).toBe(true);
+    expect(passesReuseGate(10)).toBe(true);
   });
 });