src/lib/web-search.ts

export interface WebSearchResult {
  title: string;
  snippet: string;
  url: string;
  updatedAt?: string;
}

export interface WebSearchResponse {
  provider: 'wikipedia';
  query: string;
  language: 'en' | 'ru';
  results: WebSearchResult[];
}

function plainText(value: unknown): string {
  return String(value ?? '')
    .replace(/<[^>]*>/gu, '')
    .replaceAll('&quot;', '"')
    .replaceAll('&#039;', "'")
    .replaceAll('&amp;', '&')
    .replaceAll('&lt;', '<')
    .replaceAll('&gt;', '>')
    .replace(/\s+/gu, ' ')
    .trim()
    .slice(0, 1200);
}

export async function searchWikipedia(
  query: string,
  requestedLanguage?: string,
  requestedLimit?: number,
  signal?: AbortSignal,
  fetchImpl: typeof fetch = fetch,
): Promise<WebSearchResponse> {
  const normalizedQuery = query.trim().slice(0, 500);
  if (!normalizedQuery) throw new Error('web_search.query must be a non-empty string');
  const language: 'en' | 'ru' = requestedLanguage === 'ru'
    || (requestedLanguage !== 'en' && /\p{Script=Cyrillic}/u.test(normalizedQuery))
    ? 'ru'
    : 'en';
  const limit = Math.min(8, Math.max(1, Math.round(requestedLimit ?? 5)));
  const params = new URLSearchParams({
    action: 'query',
    list: 'search',
    srsearch: normalizedQuery,
    srlimit: String(limit),
    srprop: 'snippet|timestamp',
    format: 'json',
    formatversion: '2',
    origin: '*',
  });
  const controller = new AbortController();
  const abort = () => controller.abort(signal?.reason);
  signal?.addEventListener('abort', abort, { once: true });
  const timer = setTimeout(() => controller.abort(new DOMException('web_search timed out', 'TimeoutError')), 8000);
  try {
    const response = await fetchImpl(`https://${language}.wikipedia.org/w/api.php?${params}`, {
      method: 'GET',
      credentials: 'omit',
      referrerPolicy: 'no-referrer',
      signal: controller.signal,
    });
    if (!response.ok) throw new Error(`web_search provider returned HTTP ${response.status}`);
    const payload = await response.json() as {
      query?: { search?: Array<{ title?: unknown; snippet?: unknown; timestamp?: unknown }> };
      error?: { info?: unknown };
    };
    if (payload.error) throw new Error(`web_search provider error: ${String(payload.error.info ?? 'unknown error')}`);
    const results = (payload.query?.search ?? []).slice(0, limit).map((item) => {
      const title = plainText(item.title);
      return {
        title,
        snippet: plainText(item.snippet),
        url: `https://${language}.wikipedia.org/wiki/${encodeURIComponent(title.replaceAll(' ', '_'))}`,
        ...(typeof item.timestamp === 'string' ? { updatedAt: item.timestamp } : {}),
      };
    });
    return { provider: 'wikipedia', query: normalizedQuery, language, results };
  } finally {
    clearTimeout(timer);
    signal?.removeEventListener('abort', abort);
  }
}