amandoti/cindermail

const ENTITIES: Record<string, string> = {
  nbsp: " ",
  amp: "&",
  lt: "<",
  gt: ">",
  quot: '"',
  apos: "'",
  "#39": "'",
  rsquo: "’",
  lsquo: "‘",
  rdquo: "”",
  ldquo: "“",
  mdash: "—",
  ndash: "–",
  hellip: "…",
  copy: "©",
  reg: "®",
  trade: "™",
  // Same invisible-preheader-padding characters INVISIBLE_CHARS strips, but
  // some senders write them as named entities instead of raw bytes. Without
  // these mapped, decodeEntities falls through to returning the entity text
  // unchanged (there's no "shy"/"zwnj" case to match), so they'd survive as
  // literal "&shy;"/"&zwnj;" garbage instead of becoming strippable.
  shy: "\u00AD",
  zwnj: "\u200C",
  zwj: "\u200D",
  lrm: "\u200E",
  rlm: "\u200F",
};

function decodeEntities(text: string): string {
  return text.replace(/&(#x[0-9a-f]+|#\d+|[a-z0-9]+);/gi, (match, entity: string) => {
    if (entity[0] === "#") {
      const code = entity[1] === "x" || entity[1] === "X" ? parseInt(entity.slice(2), 16) : parseInt(entity.slice(1), 10);
      return Number.isFinite(code) ? String.fromCodePoint(code) : match;
    }
    return ENTITIES[entity.toLowerCase()] ?? match;
  });
}