feat: add WeChat media analysis adapters

wechat-media.mjs

@@ -9,12 +9,19 @@ const __dirname = path.dirname(fileURLToPath(import.meta.url));
 
 const DEFAULT_WECHAT_MEDIA_URL = 'https://api.weixin.qq.com/cgi-bin/media/get';
 const DEFAULT_MAX_IMAGE_BYTES = 10 * 1024 * 1024;
+const DEFAULT_MAX_ATTACHMENT_BYTES = 30 * 1024 * 1024;
 const ALLOWED_IMAGE_MIME_TYPES = new Map([
   ['image/jpeg', 'jpg'],
   ['image/png', 'png'],
   ['image/webp', 'webp'],
   ['image/gif', 'gif'],
 ]);
+const ALLOWED_ATTACHMENT_EXTENSIONS = new Map([
+  ['.doc', 'application/msword'],
+  ['.docx', 'application/vnd.openxmlformats-officedocument.wordprocessingml.document'],
+  ['.xls', 'application/vnd.ms-excel'],
+  ['.xlsx', 'application/vnd.openxmlformats-officedocument.spreadsheetml.sheet'],
+]);
 
 function resolveImageExtension(contentType = '', fallbackUrl = '') {
   const normalized = String(contentType ?? '')
@@ -44,6 +51,32 @@ function ensureImageWithinLimit(buffer, maxBytes) {
   }
 }
 
+function sanitizeAttachmentFilename(filename = '') {
+  const basename = path.basename(String(filename ?? '').trim()).replace(/[\u0000-\u001f\u007f]/g, '');
+  if (!basename || basename === '.' || basename === '..') {
+    throw new Error('微信文件缺少有效文件名');
+  }
+  const extension = path.extname(basename).toLowerCase();
+  const mimeType = ALLOWED_ATTACHMENT_EXTENSIONS.get(extension);
+  if (!mimeType) {
+    throw new Error('当前服务号文件仅支持 Word（doc/docx）和 Excel（xls/xlsx）');
+  }
+  return {
+    filename: basename.slice(0, 160),
+    extension,
+    mimeType,
+  };
+}
+
+function ensureAttachmentWithinLimit(buffer, maxBytes) {
+  if (!Buffer.isBuffer(buffer) || buffer.length === 0) {
+    throw new Error('微信文件内容为空');
+  }
+  if (buffer.length > maxBytes) {
+    throw new Error(`文件超过大小限制（${maxBytes} bytes）`);
+  }
+}
+
 export async function downloadTemporaryMedia(accessToken, mediaId, { wechatFetch = undiciFetch } = {}) {
   if (!accessToken) throw new Error('缺少微信 access_token');
   if (!mediaId) throw new Error('缺少微信 mediaId');
@@ -154,3 +187,58 @@ export async function persistWechatImage(
     source,
   };
 }
+
+export async function persistWechatAttachment(
+  {
+    userId,
+    appId,
+    openid,
+    msgId,
+    mediaId,
+    filename,
+    publicBaseUrl,
+    maxFileBytes = DEFAULT_MAX_ATTACHMENT_BYTES,
+  },
+  {
+    wechatFetch = undiciFetch,
+    accessToken,
+    h5Root = __dirname,
+  } = {},
+) {
+  if (!userId) throw new Error('缺少 userId');
+  const resolved = sanitizeAttachmentFilename(filename);
+  const downloaded = await downloadTemporaryMedia(accessToken, mediaId, { wechatFetch });
+  ensureAttachmentWithinLimit(downloaded.buffer, maxFileBytes);
+
+  const publishDir = path.join(h5Root, PUBLISH_ROOT_DIR, String(userId), PUBLIC_ZONE_DIR, 'wechat-mp');
+  fs.mkdirSync(publishDir, { recursive: true });
+
+  const timestamp = Date.now();
+  const hash = crypto.createHash('sha1').update(downloaded.buffer).digest('hex').slice(0, 12);
+  const originalStem = path.basename(resolved.filename, resolved.extension)
+    .replace(/[^\p{L}\p{N}._-]+/gu, '_')
+    .replace(/^_+|_+$/g, '')
+    .slice(0, 80) || 'attachment';
+  const identity = [appId || 'wx', openid || 'openid', msgId || timestamp, mediaId || hash]
+    .filter(Boolean)
+    .join('-')
+    .replace(/[^a-zA-Z0-9._-]+/g, '_')
+    .slice(0, 100);
+  const publicFilename = `${originalStem}-${identity}-${hash}${resolved.extension}`;
+  const absolutePath = path.join(publishDir, publicFilename);
+  fs.writeFileSync(absolutePath, downloaded.buffer);
+
+  return {
+    absolutePath,
+    bytes: downloaded.buffer.length,
+    contentType: resolved.mimeType,
+    filename: resolved.filename,
+    publicFilename,
+    publicUrl: buildWechatImagePublicUrl({
+      publicBaseUrl,
+      publishKey: String(userId),
+      filename: publicFilename,
+    }),
+    source: 'wechat_media',
+  };
+}