fix(android): use $name when a Java package's global name is taken by NathanWalker · Pull Request #86 · NativeScript/runtimes

NathanWalker · 2026-10-24T17:14:16Z

Important

Squash merge this PR. The first commit (f44d92e) describes a read-only-global fallback that review replaced with an existence check. Use this as the squash commit message (title + body), replacing GitHub's default list of commit messages:

fix(android): use $name when a Java package's global name is taken

If the global object already has a top-level Java package's name, expose
the package as `$name` and log a warning, in both the jsi and napi trees,
so an existing global is never overwritten and the package stays reachable.

Addresses the "Expo-specific special-casing in core runtime code" point in the #77 review.

Summary

Top-level Java package globals no longer overwrite, or depend on, a global the host already defined.

CreateTopLevelNamespaces
  for each top-level Java package
    name = package name, or "$name" for a JS keyword      (unchanged)
    if global already has `name`
      warn "Java package 'expo' is exposed as '$expo': the global 'expo' already exists."
      name = "$name"
    set global[name]

jsi (ffi/jni/jsi) used to try the assignment and fall back to $name only if it threw. It now checks for the name first.
napi (ffi/jni/napi) had no fallback, so a taken name made the package unreachable. It now runs the same check.
The warning goes to logcat (TNS.Runtime).
The rule is documented in the README PR (Android section).

Evidence

Both trees compile for arm64-v8a: ./gradlew :runtime:externalNativeBuildDebug -PonlyArm64 -PbindingLayer=napi, then the same with -PbindingLayer=jsi.
Nothing the runtime defines before CreateTopLevelNamespaces matches a Java package name. That set is global, self, Worker, and on workers postMessage, close, terminate and __ns__worker. A standalone app therefore gets exactly the names it got before.
Not verified at runtime: the fallback path. It needs a host that defines a matching global first, such as Expo's expo.

Merge Danger

Door: two-way

Blast Radius: top-level Java package globals

The only change is for a package whose name is already a global. Before, that package overwrote a writable global, or was unreachable on napi when the global was read-only. Now it appears as $name.

A host can own a top-level package name as a read-only global; Expo defines `expo`. The jsi runtime already exposed such a package as `$expo`, but did so silently. It now logs a warning naming both globals. The napi runtime had no fallback: depending on the engine, the assignment either failed or was ignored, and the package was not reachable at all. It now gets the same `$name` fallback and warning.

coderabbitai · 2026-10-24T17:14:23Z

Important

Review skipped

Auto reviews are disabled on this repository. Please check the settings in the CodeRabbit UI or the .coderabbit.yaml file in this repository. To trigger a single review, invoke the @coderabbitai review command.

Run configuration

Source mode: Auto
Repository: Unavailable
Branch: Unavailable
File: No configuration file from this repository applied
Commit: Unavailable
Layers: Organization UI
Review profile: CHILL
Plan: Advanced
Run ID: 3ce7bc9c-c72c-4b01-8155-40c58b60750e

You can disable this status message by setting the reviews.review_status to false in the CodeRabbit configuration file.

Use the checkbox below for a quick retry:

🔍 Trigger review

Autofix · Keep fixing CodeRabbit findings and required CI, and resolving merge conflicts

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

_{Comment @coderabbitai help to get the list of available commands.}

ammarahm-ed · 2026-10-25T10:21:52Z

@@ -1120,7 +1142,15 @@ void MetadataNode::CreateTopLevelNamespaces(napi_env env) {
            if (IsJavascriptKeyword(nameSpace)) {
                nameSpace = "$" + nameSpace;
            }
-            NAPI_GUARD(napi_set_named_property(env, global, nameSpace.c_str(), packageObj)) {}
+            if (!SetGlobalIfWritable(env, global, nameSpace, packageObj)) {
+                // The host already owns the name as a read-only global (Expo's
+                // `expo`), so the package takes the same $ prefix as keywords.
+                string fallback = "$" + nameSpace;
+                __android_log_print(ANDROID_LOG_WARN, "TNS.Runtime",
+                                    "Java package '%s' is exposed as '%s': the global '%s' is read-only.",
+                                    nameSpace.c_str(), fallback.c_str(), nameSpace.c_str());
+                NAPI_GUARD(napi_set_named_property(env, global, fallback.c_str(), packageObj)) {}
+            }


The SetGlobalIfWritable function is used only once hence move the function body inside CreateTopLevelNamespaces.

Done. The helper is gone and the check is inline in CreateTopLevelNamespaces.

ammarahm-ed · 2026-10-25T10:22:19Z

            }
-            NAPI_GUARD(napi_set_named_property(env, global, nameSpace.c_str(), packageObj)) {}
+            if (!SetGlobalIfWritable(env, global, nameSpace, packageObj)) {
+                // The host already owns the name as a read-only global (Expo's


Remove mention of expo or Expo from the comments.

Done. No comment mentions Expo now.

ammarahm-ed · 2026-10-25T10:25:36Z

+    if (napi_set_named_property(env, global, name.c_str(), value) == napi_ok) {
+        napi_value assigned;
+        bool same = false;
+        if (napi_get_named_property(env, global, name.c_str(), &assigned) == napi_ok &&
+            napi_strict_equals(env, assigned, value, &same) == napi_ok && same) {
+            return true;
+        }
+    }
+    bool pending = false;
+    napi_is_exception_pending(env, &pending);
+    if (pending) {
+        napi_value error;
+        napi_get_and_clear_last_exception(env, &error);
+    }
+    return false;
+}


All of this is probably not needed because we do not want to silently overwrite a value set on a global object even if it was not read-only I think. So A simple napi_has_named_property is enough.

Agreed. It now does a plain napi_has_named_property: if the global already has the name, the package goes on $name and a warning is logged, so nothing is overwritten. 0948f32

ammarahm-ed · 2026-10-25T10:28:07Z

            try {
                global.setProperty(rt, nameSpace.c_str(), packageObj);
            } catch (JsError &) {
-                // A host can already own the name as a non-writable global (Expo
-                // defines `expo`), so fall back to the same $ prefix as keywords.
-                global.setProperty(rt, ("$" + nameSpace).c_str(), packageObj);
+                // The host already owns the name as a read-only global (Expo's
+                // `expo`), so the package takes the same $ prefix as keywords.
+                string fallback = "$" + nameSpace;
+                __android_log_print(ANDROID_LOG_WARN, "TNS.Runtime",
+                                    "Java package '%s' is exposed as '%s': the global '%s' is read-only.",
+                                    nameSpace.c_str(), fallback.c_str(), nameSpace.c_str());
+                global.setProperty(rt, fallback.c_str(), packageObj);


Do the same here as comments for the napi runtime. Check if the property already exists, if it does prefix with $. also remove the comment.

Done. The jsi tree checks global.hasProperty first, takes the same $name fallback, and the comment is removed.

Check whether the global already has the package's name and, if it does, expose the package as `$name`, logging a warning. Both the jsi and napi trees now do the same check, so an existing global is never overwritten and no engine-specific read-only handling is needed.

This was referenced Oct 24, 2026

docs(react-native): rewrite README #90

Open

fix(react-native): interop ng-native broad #77

Closed

ammarahm-ed requested changes Oct 25, 2026

View reviewed changes

NathanWalker changed the title ~~fix(android): warn when a Java package global falls back to $name~~ fix(android): use $name when a Java package's global name is taken Oct 25, 2026

Navigation Menu

Repository navigation

fix(android): use $name when a Java package's global name is taken - #86

fix(android): use $name when a Java package's global name is taken#86
NathanWalker wants to merge 2 commits into
android-react-nativefrom
fix/android-readonly-package-global

NathanWalker commented Oct 24, 2026 •

edited

Loading

Uh oh!

coderabbitai Bot commented Oct 24, 2026 •

edited

Loading

Review skipped

Uh oh!

ammarahm-ed Oct 25, 2026

Uh oh!

NathanWalker Oct 25, 2026

Uh oh!

ammarahm-ed Oct 25, 2026

Uh oh!

NathanWalker Oct 25, 2026

Uh oh!

ammarahm-ed Oct 25, 2026

Uh oh!

NathanWalker Oct 25, 2026

Uh oh!

ammarahm-ed Oct 25, 2026

Uh oh!

NathanWalker Oct 25, 2026

Uh oh!

This branch has not been deployed

Reviewers

Uh oh!

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

Conversation

NathanWalker commented Oct 24, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Summary

Evidence

Merge Danger

Uh oh!

coderabbitai Bot commented Oct 24, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Review skipped

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

This branch has not been deployed

Reviewers

Uh oh!

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

NathanWalker commented Oct 24, 2026 •

edited

Loading

coderabbitai Bot commented Oct 24, 2026 •

edited

Loading