module: correctly detect top-level await in ambiguous contexts

Fixes: #58331

Author: islandryu

src/node_contextify.cc

@@ -1647,6 +1647,12 @@ static const auto throws_only_in_cjs_error_messages =
         "await is only valid in async functions and "
         "the top level bodies of modules"};
 
+static std::vector<std::string_view> maybe_top_level_await_errors = {
+    // example: `func(await 1);`
+    "missing ) after argument list",
+    // example: `if(await 1)`
+    "SyntaxError: Unexpected"};
+
 // If cached_data is provided, it would be used for the compilation and
 // the on-disk compilation cache from NODE_COMPILE_CACHE (if configured)
 // would be ignored.
@@ -1877,6 +1883,16 @@ bool ShouldRetryAsESM(Realm* realm,
       break;
     }
   }
+
+  for (const auto& error_message : maybe_top_level_await_errors) {
+    if (message_view.find(error_message) != std::string_view::npos) {
+      // If the error message is related to top-level await, we can try to
+      // compile it as ESM.
+      maybe_valid_in_esm = true;
+      break;
+    }
+  }
+
   if (!maybe_valid_in_esm) {
     return false;
   }

test/es-module/test-esm-detect-ambiguous.mjs

@@ -427,22 +427,75 @@ describe('when working with Worker threads', () => {
   });
 });
 
-describe('cjs & esm ambiguous syntax case', () => {
-  it('should throw an ambiguous syntax error when using top-level await with require', async () => {
-    const { stderr, code, signal } = await spawnPromisified(
-      process.execPath,
-      [
-        '--input-type=module',
-        '--eval',
-        `await 1;\nconst fs = require('fs');`,
-      ]
-    );
+describe('maybe top-level await syntax errors that are not recognized as top-level await errors', () => {
+  const expressions = [
+    // string
+    { expression: '""' },
+    // number
+    { expression: '0' },
+    // boolean
+    { expression: 'true' },
+    // null
+    { expression: 'null' },
+    // undefined
+    { expression: 'undefined' },
+    // object
+    { expression: '{}' },
+    // array
+    { expression: '[]' },
+    // new
+    { expression: 'new Date()' },
+    // identifier
+    { initialize: 'const a = 2;', expression: 'a' },
+  ];
+  it('should not crash the process', async () => {
+    for (const { expression, initialize } of expressions) {
+      const wrapperExpressions = [
+        `function callAwait() {}; callAwait(await ${expression});`,
+        `if (await ${expression}) {}`,
+        `{ key: await ${expression} }`,
+        `[await ${expression}]`,
+        `(await ${expression})`,
+      ];
+      for (const wrapperExpression of wrapperExpressions) {
+        const { code, signal, stdout, stderr } = await spawnPromisified(process.execPath, [
+          '--eval',
+          `
+          ${initialize || ''}
+          ${wrapperExpression}
+          `,
+        ]);
+
+        strictEqual(stderr, '');
+        strictEqual(stdout, '');
+        strictEqual(code, 0);
+        strictEqual(signal, null);
+      }
+    }
+  });
 
-    match(
-      stderr,
-      /ReferenceError: Cannot determine intended module format because both require\(\) and top-level await are present\. If the code is intended to be CommonJS, wrap await in an async function\. If the code is intended to be an ES module, replace require\(\) with import\./
-    );
+  it('should crash when the expression is not valid', async () => {
+    let { code, signal, stdout, stderr } = await spawnPromisified(process.execPath, [
+      '--eval',
+      `
+      function callAwait() {}
+      callAwait(await "" ""); 
+      `,
+    ]);
+    match(stderr, /SyntaxError: missing \) after argument list/);
+    strictEqual(stdout, '');
+    strictEqual(code, 1);
+    strictEqual(signal, null);
 
+    ({ code, signal, stdout, stderr } = await spawnPromisified(process.execPath, [
+      '--eval',
+      `
+      function callAwait() {}
+      if (a "") {}
+      `,
+    ]));
+    match(stderr, /SyntaxError: Unexpected string/);
+    strictEqual(stdout, '');
     strictEqual(code, 1);
     strictEqual(signal, null);
   });