middleware.ts

/**
 * Vercel Edge Middleware — tracks AI bot visits
 */

const AI_BOTS: [string, string][] = [
  ['GPTBot', 'OpenAI'],
  ['ChatGPT-User', 'ChatGPT'],
  ['Google-Extended', 'Gemini'],
  ['ClaudeBot', 'Claude'],
  ['anthropic-ai', 'Anthropic'],
  ['PerplexityBot', 'Perplexity'],
  ['Bytespider', 'ByteDance'],
  ['Applebot-Extended', 'Apple'],
  ['cohere-ai', 'Cohere'],
  ['CCBot', 'CommonCrawl'],
]

export default function middleware(request: Request) {
  const ua = request.headers.get('user-agent') || ''
  const url = new URL(request.url)

  if (url.pathname.startsWith('/assets/') || url.pathname.startsWith('/api/') || url.pathname.includes('.')) {
    return
  }

  // Auto-redirect non-Chinese browsers to /en/ version (only for key pages without /en/ prefix)
  if (!url.pathname.startsWith('/en/') && url.pathname !== '/en') {
    const lang = request.headers.get('accept-language') || ''
    const isZh = lang.startsWith('zh') || lang.includes('zh-TW') || lang.includes('zh-CN')
    if (!isZh && (url.pathname === '/probe' || url.pathname === '/probe/')) {
      return new Response(null, {
        status: 302,
        headers: { Location: `/en${url.pathname}${url.search}` },
      })
    }
  }

  for (const [sig, name] of AI_BOTS) {
    if (ua.includes(sig)) {
      try {
        fetch(`${url.origin}/api/probe-collect-email`, {
          method: 'POST',
          headers: { 'Content-Type': 'application/json' },
          body: JSON.stringify({
            action: 'bot-visit',
            bot: name,
            path: url.pathname,
            timestamp: new Date().toISOString(),
          }),
        }).catch(() => {})
      } catch {}
      break
    }
  }
}

export const config = {
  matcher: ['/((?!api|_next|assets|favicon|logo|og-|sitemap|robots|llms).*)'],
}