feat: 完善微信小程序与红包消息解析

- 新增小程序和红包结构化消息及专用卡片 - 修复嵌套类型、拍一拍和表情包误判 - 补充常见 AppMsg 类型分类
2026-08-17 19:47:08 +08:00 · 2026-07-28 03:20:11 +08:00
parent 7db845ac7e
commit 5ccabb3898
10 changed files with 502 additions and 27 deletions
@@ -16,6 +16,22 @@ type ShareContent = {
  appname?: string
  typeVal?: string
 }
+type MiniProgramContent = {
+  type: 'miniProgram'
+  title: string
+  description?: string
+  appName?: string
+  iconUrl?: string
+  thumbMd5?: string
+  thumbDatName?: string
+  thumbDataUrl?: string
+}
+type RedPacketContent = {
+  type: 'redPacket'
+  title: string
+  description?: string
+  url?: string
+}
 type VoipContent = { type: 'voip'; duration?: number; status: string; roomType?: number }
 type ImageContent = {
  type: 'image'
@@ -74,6 +90,8 @@ export type ParsedContent =
  | LocationContent
  | CardContent
  | ShareContent
+  | MiniProgramContent
+  | RedPacketContent
  | VoipContent
  | ImageContent
  | VideoContent
@@ -413,6 +431,31 @@ function parseShareMessage(content: string): ParsedContent {
    }
  }

+  if (appMsgType === '33' || appMsgType === '36') {
+    return {
+      type: 'miniProgram',
+      title: extractXmlValue(content, 'title') || '小程序',
+      description: extractXmlValue(content, 'des') || undefined,
+      appName:
+        extractXmlValue(content, 'sourcedisplayname') ||
+        extractXmlValue(content, 'appname') ||
+        '小程序',
+      iconUrl: decodeXmlUrl(extractXmlValue(content, 'weappiconurl')) || undefined,
+      thumbMd5: normalizeMd5(
+        extractXmlValue(content, 'cdnthumbmd5') || extractXmlValue(content, 'md5')
+      )
+    }
+  }
+
+  if (appMsgType === '2001') {
+    return {
+      type: 'redPacket',
+      title: extractXmlValue(content, 'title') || '微信红包',
+      description: extractXmlValue(content, 'des') || '恭喜发财，大吉大利',
+      url: decodeXmlUrl(extractXmlValue(content, 'url')) || undefined
+    }
+  }
+
  const title = extractXmlValue(content, 'title') || ''
  const des = extractXmlValue(content, 'des') || extractXmlValue(content, 'desc') || ''
  const url = extractXmlValue(content, 'url') || ''
@@ -485,6 +528,10 @@ function extractAppMsgType(content: string): string {
  const inner = appmsgMatch[1]
    .replace(/<refermsg[\s\S]*?<\/refermsg>/gi, '')
    .replace(/<patMsg[\s\S]*?<\/patMsg>/gi, '')
+    .replace(/<weappinfo[\s\S]*?<\/weappinfo>/gi, '')
+    .replace(/<appattach[\s\S]*?<\/appattach>/gi, '')
+    .replace(/<wcpayinfo[\s\S]*?<\/wcpayinfo>/gi, '')
+    .replace(/<findernamecard[\s\S]*?<\/findernamecard>/gi, '')
  const typeMatch = /<type>([\s\S]*?)<\/type>/i.exec(inner)
  return typeMatch?.[1]?.trim() || ''
 }
@@ -666,6 +713,71 @@ export function parseImageDatNameFromRow(row: Record<string, unknown>): string |
  return hexMatch?.[1]?.toLowerCase()
 }

+export function parseImageBufferDataUrlFromRow(
+  row: Record<string, unknown>
+): string | undefined {
+  const raw = pickRowString(row, [
+    'ImgBuf',
+    'imgBuf',
+    'img_buf',
+    'imageBuffer',
+    'image_buffer',
+    'thumbBuffer',
+    'thumb_buffer',
+    'WCDB_CT_img_buf',
+    'WCDB_CT_ImgBuf'
+  ])
+  const buffer = decodeInlineImageBuffer(raw)
+  if (!buffer || buffer.length === 0) return undefined
+  const mime = detectImageMime(buffer)
+  return mime ? `data:${mime};base64,${buffer.toString('base64')}` : undefined
+}
+
+function decodeInlineImageBuffer(raw: unknown): Buffer | null {
+  if (!raw) return null
+  if (Buffer.isBuffer(raw)) return raw
+  if (raw instanceof Uint8Array) return Buffer.from(raw)
+  if (Array.isArray(raw)) return Buffer.from(raw)
+  if (typeof raw === 'object') {
+    const record = raw as { buffer?: unknown; data?: unknown }
+    return decodeInlineImageBuffer(record.buffer ?? record.data)
+  }
+  if (typeof raw !== 'string') return null
+  const value = raw.trim()
+  const dataUrl = /^data:image\/[a-z0-9.+-]+;base64,(.+)$/i.exec(value)
+  const encoded = dataUrl?.[1] || value
+  if (!/^[a-z0-9+/]+={0,2}$/i.test(encoded)) return null
+  try {
+    return Buffer.from(encoded, 'base64')
+  } catch {
+    return null
+  }
+}
+
+function detectImageMime(buffer: Buffer): string | undefined {
+  if (buffer.length >= 3 && buffer[0] === 0xff && buffer[1] === 0xd8 && buffer[2] === 0xff) {
+    return 'image/jpeg'
+  }
+  if (
+    buffer.length >= 8 &&
+    buffer[0] === 0x89 &&
+    buffer.subarray(1, 4).toString('ascii') === 'PNG'
+  ) {
+    return 'image/png'
+  }
+  if (buffer.length >= 6 && /^GIF8[79]a$/.test(buffer.subarray(0, 6).toString('ascii'))) {
+    return 'image/gif'
+  }
+  if (
+    buffer.length >= 12 &&
+    buffer.subarray(0, 4).toString('ascii') === 'RIFF' &&
+    buffer.subarray(8, 12).toString('ascii') === 'WEBP'
+  ) {
+    return 'image/webp'
+  }
+  return undefined
+}
+
 function pickRowString(row: Record<string, unknown>, keys: string[]): unknown {
  for (const key of keys) {
    if (Object.prototype.hasOwnProperty.call(row, key)) return row[key]
@@ -155,6 +155,27 @@ function cachedMessageIdentity(message: Message): string {
  return `id:${message.id}`
 }

+function containsLegacyMisparsedAppMessage(items: Message[]): boolean {
+  return items.some((message) => {
+    const content = message.contentData
+    if (content?.type === 'system' && content.raw) {
+      return (
+        /<weappinfo\b/i.test(content.raw) &&
+        /<type>\s*(?:33|36|2001)\s*<\/type>/i.test(content.raw)
+      )
+    }
+    if (
+      content?.type === 'share' &&
+      ((content.typeVal === '3' && !content.url) || content.typeVal === '2001')
+    ) {
+      return true
+    }
+    if (content?.type !== 'sticker') return false
+    const url = String(content.url || content.thumbUrl || '')
+    return /wxapp\.tenpay\.com\/mmpayhb|mp\.weixin\.qq\.com\/mp\/waerrpage/i.test(url)
+  })
+}
+
 function pruneMessageBuckets(messages: Record<string, CachedMessageBucket>): void {
  const entries = Object.entries(messages)
  if (entries.length <= MAX_MESSAGE_BUCKETS) return
@@ -315,7 +336,7 @@ export function getCachedMessagePage(
    }
  }
  return {
-    hit: Boolean(bucket),
+    hit: Boolean(bucket) && !containsLegacyMisparsedAppMessage(bucket?.items || []),
    messages: bucket?.items || [],
    groupSnapshot: cache?.groupSnapshots?.[userMd5]?.snapshot
  }
@@ -1,5 +1,6 @@
 import { WechatDb, WechatMessage } from '../wechat-db'
 import {
+  parseImageBufferDataUrlFromRow,
  parseImageDatNameFromRow,
  parseMessageContent,
  parseStickerMessageFromRow
@@ -235,7 +236,10 @@ function listSourceMessages(

    let contentData: ReturnType<typeof parseMessageContent> | undefined
    let displayType = MSG_TYPE_DICT[msgType] || msg.messageType
-    const isPatMessage = /<patMsg\b|拍了拍|拍一拍/i.test(String(content || ''))
+    const rawContent = String(content || '')
+    const isPatMessage =
+      /<patinfo\b|<type>\s*62\s*<\/type>/i.test(rawContent) ||
+      ([10000, 10002].includes(msgType) && /拍了拍/i.test(rawContent))
    if (isPatMessage) {
      const system = parseMessageContent(content, 10000)
      const patContent =
@@ -254,27 +258,40 @@ function listSourceMessages(
    if (!isPatMessage && [3, 42, 43, 47, 48, 49, 50, 10000, 10002].includes(inferredMsgType)) {
      try {
        const isQuotePayload = /<refermsg\b/i.test(content)
-        const hasStickerHints =
+        const hasStickerPayload =
          /<(?:emoji|sticker|emoticon)\b/i.test(content) ||
-          [
-            'emoji_md5',
-            'emojiMd5',
-            'emoji_cdn_url',
-            'emojiCdnUrl',
-            'packed_info_data',
-            'packed_info',
-            'reserved0',
-            'Reserved0',
-            'WCDB_CT_reserved0'
-          ].some((key) => Boolean(msg[key]))
+          /<type>\s*47\s*<\/type>/i.test(content)
        const rowSticker =
-          inferredMsgType === 47 || (inferredMsgType === 49 && !isQuotePayload && hasStickerHints)
+          inferredMsgType === 47 ||
+          (inferredMsgType === 49 && !isQuotePayload && hasStickerPayload)
            ? parseStickerMessageFromRow(msg, content)
            : undefined
-        const parsed =
-          rowSticker?.type === 'sticker'
-            ? rowSticker
-            : parseMessageContent(content, inferredMsgType)
+        const parsedContent = parseMessageContent(content, inferredMsgType)
+        const rowStickerUrl = rowSticker?.type === 'sticker' ? String(rowSticker.url || '') : ''
+        const parsedShareUrl = parsedContent.type === 'share' ? parsedContent.url : ''
+        const redPacketUrl = rowStickerUrl || parsedShareUrl
+        const isRedPacketFallback =
+          (parsedContent.type === 'share' && parsedContent.typeVal === '2001') ||
+          /wxapp\.tenpay\.com\/mmpayhb/i.test(redPacketUrl)
+        const parsed: ReturnType<typeof parseMessageContent> =
+          parsedContent.type === 'miniProgram' || parsedContent.type === 'redPacket'
+            ? parsedContent
+            : isRedPacketFallback
+              ? {
+                  type: 'redPacket',
+                  title:
+                    parsedContent.type === 'share' && parsedContent.title
+                      ? parsedContent.title
+                      : '微信红包',
+                  description:
+                    parsedContent.type === 'share' && parsedContent.des
+                      ? parsedContent.des
+                      : '恭喜发财，大吉大利',
+                  url: redPacketUrl || undefined
+                }
+              : rowSticker?.type === 'sticker'
+                ? rowSticker
+                : parsedContent
        if (parsed.type === 'system') {
          content = parsed.content
          contentData = parsed
@@ -284,6 +301,13 @@ function listSourceMessages(
        if (parsed.type === 'image') {
          const imageDatName = parseImageDatNameFromRow(msg)
          contentData = { ...parsed, datName: parsed.datName || imageDatName }
+        } else if (parsed.type === 'miniProgram') {
+          contentData = {
+            ...parsed,
+            thumbDatName: parsed.thumbDatName || parseImageDatNameFromRow(msg),
+            thumbDataUrl:
+              parsed.thumbDataUrl || parseImageBufferDataUrlFromRow(msg.raw || msg)
+          }
        } else if (parsed.type !== 'system') {
          if (parsed.type === 'sticker' && !parsed.url && parsed.md5) {
            parsed.url = wcdb4Client.resolveEmoticonCdnUrl(parsed.md5)
@@ -295,6 +319,15 @@ function listSourceMessages(
        }
        if (parsed.type === 'quote') displayType = '引用消息'
        if (parsed.type === 'sticker') displayType = '表情包'
+        if (parsed.type === 'miniProgram') displayType = '小程序'
+        if (parsed.type === 'redPacket') displayType = '微信红包'
+        if (parsed.type === 'share') {
+          if (parsed.typeVal === '5') displayType = '公众号链接'
+          if (parsed.typeVal === '6') displayType = '文件'
+          if (parsed.typeVal === '74') displayType = '文件发送中'
+          if (parsed.typeVal === '51') displayType = '视频号'
+          if (parsed.typeVal === '2000') displayType = '转账'
+        }
      } catch {
        // ignore parse errors
      }