[facebook/react] Bug: React Compiler does not preserve HTML entity

### ROOT CAUSE The issue arises because the React Compiler's JSX parser incorrectly processes HTML entities (e.g., `&`) within string literals. It mistakenly interprets them as part of the JSX syntax, causing the entities to be unescaped or altered during transformation. This is due to the parser's failure to properly distinguish between valid JSX syntax and escaped HTML characters within string literals. ### CODE FIX ```javascript // Modify the JSX parser to preserve entities in string literals // Example fix in a hypothetical JSX transformer (e.g., using Acorn): const acorn = require('acorn'); const jsx = require('acorn-jsx'); function transform(code) { const ast = acorn.parse(code, { ecmaVersion: 2018, jsx: true }); // Custom visitor to preserve entities in string literals ast.walk((node) => { if (node.type === 'Literal' && typeof node.value === 'string') { // Check if the string contains entities if (node.value.includes('&') && !node.value.includes('<')) { node.value = node.value.replace(/&/g, '&'); // Preserve entities } } }); return jsx.generate(ast); } ``` This fix ensures entities like `&` are preserved by scanning string literals and re-escaping any `&` characters not part of HTML tags.