/**
 * Extract the FIRST previewable http(s) URL from a message's plain text.
 *
 * Telegram/WhatsApp rule: at most one preview per message — we surface the
 * first link only. Returns `null` when there's nothing to unfurl.
 *
 * Deliberately conservative: only absolute http/https URLs (no bare
 * `example.com`, no mailto/ftp), trailing punctuation trimmed, and a length
 * cap to avoid pathological inputs.
 */

// http(s) URL up to the first whitespace or closing bracket/quote.
const URL_RE = /\bhttps?:\/\/[^\s<>"'`)\]}]+/i;

/** Trailing punctuation that's almost always sentence/markup, not the URL. */
const TRAILING = /[.,;:!?)\]}'"»]+$/;

export function extractFirstUrl(text: string | undefined | null): string | null {
  if (!text) return null;
  // Skip fenced/inline code-heavy content cheaply: still match, but the cap
  // below guards huge blobs.
  if (text.length > 50_000) return null;

  const match = URL_RE.exec(text);
  if (!match) return null;

  let url = match[0].replace(TRAILING, '');

  // Balance a trailing ")" that belongs to a markdown link `](url)` wrapper:
  // if the URL has more ")" than "(", drop the extras.
  const opens = (url.match(/\(/g) ?? []).length;
  const closes = (url.match(/\)/g) ?? []).length;
  if (closes > opens) {
    url = url.replace(/\)+$/, (m) => m.slice(0, Math.max(0, m.length - (closes - opens))));
  }

  // Validate — a malformed match resolves to no preview.
  try {
    const parsed = new URL(url);
    if (parsed.protocol !== 'http:' && parsed.protocol !== 'https:') return null;
    return parsed.toString();
  } catch {
    return null;
  }
}