[Bug] Java syntax highlighting does not handle unicode escape sequences

Reproducible in vscode.dev or in VS Code Desktop?

• Not reproducible in vscode.dev or VS Code Desktop

Reproducible in the monaco editor playground?

• Not reproducible in the monaco editor playground

Monaco Editor Playground Code

monaco.editor.create(document.getElementById('container'), {
	value: '// \\u000a is a line break in Java unicode escapes\n// This comment \\u000a int[] x = {0}; // is actually code\nclass Test {\n    // \\u0048\\u0065\\u006C\\u006C\\u006F\n    public static void main(String[] args) {}\n}',
	language: 'java'
});

Description

Java processes unicode escape sequences (\uXXXX) at a very early stage — before tokenisation. This means \u000a inside a comment is actually a line break, and code after it is executable. The syntax highlighter doesn't account for this, so what appears to be a comment can hide real code.

This is a known Java "feature" that can be used to hide malicious code: https://wh0.github.io/2019/11/16/easter-egg-inspection.html

Ideally the Java tokeniser would process \uXXXX sequences the same way javac does, or at minimum flag them visually.

Cross-reference: https://github.com/compiler-explorer/compiler-explorer/issues/4223

Start with the Monaco editor playground reproduction and inspect the Java tokenizer's handling of unicode escapes, using the linked compiler-explorer issue for comparison. Done means comments containing Java escapes such as \u000a are tokenized according to Java's preprocessing behavior, or the sequences are visibly flagged as requested.