fix(realtime): 修复服务号实时同步写入解密库并优化桌面端运行时兼容性

服务号实时同步改为直接读取 live biz_message 库，避免 gh_* 会话走空的 wcdb_get_messages。

将同步结果写回 output/databases 解密库，并补充异常降级路径。

修复桌面端日志处理与 wcdb_api.dll 的运行时定位问题。

README.md

@@ -6,6 +6,7 @@
     <h1>WeChatDataAnalysis - 微信数据库解密与分析工具</h1>
     <p>微信4.x数据解密并生成年度总结，高仿微信，支持实时更新，导出聊天记录，朋友圈等大量便捷功能</p>
     <p><b>特别致谢</b>：<a href="https://github.com/H3CoF6">H3CoF6</a>（密钥与朋友圈等核心内容的技术支持）、<a href="https://github.com/ycccccccy/echotrace">echotrace</a>、<a href="https://github.com/hicccc77/WeFlow">WeFlow</a>（本项目大量功能参考其实现）</p>
+    <p>如需定制功能，请联系 QQ：2977094657。</p>
     <img src="https://img.shields.io/github/v/tag/LifeArchiveProject/WeChatDataAnalysis" alt="Version" />
     <img src="https://img.shields.io/github/stars/LifeArchiveProject/WeChatDataAnalysis" alt="Stars" />
     <img src="https://gh-down-badges.linkof.link/LifeArchiveProject/WeChatDataAnalysis" alt="Downloads" />
@@ -192,16 +193,6 @@ npm run dist
 3. **数据隐私**: 解密后的数据包含个人隐私信息，请谨慎处理
 4. **合法使用**: 请遵守相关法律法规，不得用于非法目的
 
-## 修改消息
-
-支持在聊天页对单条消息进行本地修改（如修改消息文本/字段、修复为我发送、反转本地气泡方向），并在“修改记录”页查看原始与当前对比，支持单条恢复或按会话一键恢复。
-
-该功能只修改本机本地数据库（`db_storage` 与解密副本），不会调用远端回写接口。
-
-<p align="center">
-  <img src="frontend/public/edit.gif" alt="本地消息修改" width="800" />
-</p>
-
 ## 致谢
 
 本项目的开发过程中参考了以下优秀的开源项目和资源：

frontend/assets/css/chat.css

@@ -1139,6 +1139,132 @@
   flex-shrink: 0;
 }
 
+.wechat-link-card-finder {
+  width: 135px;
+  min-width: 135px;
+  max-width: 135px;
+  border: none;
+  box-shadow: none;
+  outline: none;
+  cursor: pointer;
+  text-decoration: none;
+}
+
+.wechat-link-card-finder.wechat-link-card--disabled {
+  cursor: default;
+}
+
+.wechat-link-finder-cover {
+  width: 135px;
+  height: 185px;
+  position: relative;
+  overflow: hidden;
+  border-radius: 4px;
+  background: var(--app-surface-muted);
+}
+
+.wechat-link-finder-cover--empty {
+  background: linear-gradient(180deg, #37cc6a 0%, #118f42 100%);
+}
+
+.wechat-link-finder-cover-img {
+  width: 100%;
+  height: 100%;
+  object-fit: cover;
+  object-position: center;
+  display: block;
+}
+
+.wechat-link-finder-cover-placeholder {
+  position: absolute;
+  inset: 0;
+  display: flex;
+  align-items: center;
+  justify-content: center;
+  color: rgba(255, 255, 255, 0.92);
+}
+
+.wechat-link-finder-cover-placeholder svg {
+  width: 34px;
+  height: 34px;
+}
+
+.wechat-link-finder-cover-shade {
+  position: absolute;
+  inset: 0;
+  background: linear-gradient(180deg, rgba(0, 0, 0, 0.04) 0%, rgba(0, 0, 0, 0.12) 42%, rgba(0, 0, 0, 0.68) 100%);
+}
+
+.wechat-link-finder-play {
+  position: absolute;
+  left: 50%;
+  top: 50%;
+  transform: translate(-50%, -66%);
+  width: 40px;
+  height: 40px;
+  border-radius: 50%;
+  background: rgba(0, 0, 0, 0.42);
+  display: flex;
+  align-items: center;
+  justify-content: center;
+  color: #fff;
+  box-shadow: 0 8px 24px rgba(0, 0, 0, 0.18);
+}
+
+.wechat-link-finder-play svg {
+  width: 20px;
+  height: 20px;
+  margin-left: 2px;
+}
+
+.wechat-link-finder-meta {
+  position: absolute;
+  left: 8px;
+  right: 8px;
+  bottom: 8px;
+  display: flex;
+  flex-direction: column;
+  gap: 0;
+}
+
+.wechat-link-finder-author {
+  display: flex;
+  align-items: center;
+  gap: 5px;
+  min-width: 0;
+  padding: 5px 7px;
+  border-radius: 999px;
+  background: rgba(0, 0, 0, 0.28);
+  backdrop-filter: blur(6px);
+}
+
+.wechat-link-finder-author-avatar {
+  width: 18px;
+  height: 18px;
+  flex-shrink: 0;
+  display: flex;
+  align-items: center;
+  justify-content: center;
+}
+
+.wechat-link-finder-author-avatar-img {
+  width: 100%;
+  height: 100%;
+  object-fit: contain;
+  display: block;
+}
+
+.wechat-link-finder-author-name {
+  min-width: 0;
+  flex: 1 1 auto;
+  font-size: 10px;
+  color: rgba(255, 255, 255, 0.96);
+  overflow: hidden;
+  text-overflow: ellipsis;
+  white-space: nowrap;
+  text-shadow: 0 1px 2px rgba(0, 0, 0, 0.28);
+}
+
 /* 隐私模式模糊效果 */
 .privacy-blur {
   filter: blur(9px);

frontend/assets/css/tailwind.css

@@ -1437,14 +1437,20 @@
 
   .session-list-item-name {
     color: var(--session-list-name);
+    font-weight: 400;
+    font-synthesis: none;
   }
 
   .session-list-item-time {
     color: var(--session-list-meta);
+    font-weight: 400;
+    font-synthesis: none;
   }
 
   .session-list-item-preview {
     color: var(--session-list-preview);
+    font-weight: 400;
+    font-synthesis: none;
   }
 
   .contact-search-wrapper {

frontend/components/BizMessages.vue

@@ -0,0 +1,394 @@
+<template>
+  <div class="biz-page h-full min-h-0 flex overflow-hidden" style="background-color: var(--app-shell-bg)">
+
+    <div :class="['w-[300px] lg:w-[320px] border-r flex flex-col flex-shrink-0 z-10', isDark ? 'bg-[#1e1e1e] border-[#333]' : 'bg-white border-gray-200']">
+      <div class="p-3 border-b" :class="isDark ? 'border-[#333]' : 'border-gray-200'" style="background-color: var(--app-surface-muted)">
+        <div class="contact-search-wrapper flex-1">
+          <input
+              v-model="searchQuery"
+              type="text"
+              class="contact-search-input"
+              placeholder="搜索服务号"
+          />
+        </div>
+      </div>
+
+      <div class="flex-1 overflow-y-auto min-h-0">
+        <div v-if="loadingAccounts" class="flex justify-center py-4">
+          <span class="text-sm" :class="isDark ? 'text-gray-500' : 'text-gray-400'">加载中...</span>
+        </div>
+        <div v-else class="pb-4">
+          <div
+              v-for="item in filteredAccounts"
+              :key="item.username"
+              @click="selectAccount(item)"
+              class="flex items-center gap-3 px-4 py-3 cursor-pointer transition-colors border-b"
+              :class="[
+                isDark ? 'border-[#333]' : 'border-gray-50',
+                selectedBizAccount?.username === item.username
+                  ? (isDark ? 'bg-[#333]' : 'bg-[#E5E5E5]') // 选中状态
+                  : item.username === 'gh_3dfda90e39d6'
+                    ? (isDark ? 'bg-[#2a2a2a] hover:bg-[#333]' : 'bg-[#F2F2F2] hover:bg-[#EAEAEA]') // 微信支付专门的底色
+                    : (isDark ? 'hover:bg-[#252525]' : 'hover:bg-gray-50') // 普通悬浮色
+              ]"
+          >
+            <img v-if="item.avatar" :src="api.getBizProxyImageUrl(item.avatar)" :class="['w-10 h-10 rounded-md object-cover flex-shrink-0', isDark ? 'bg-[#333]' : 'bg-gray-200']" alt=""/>
+            <div v-else class="w-10 h-10 rounded-md bg-[#03C160] text-white flex items-center justify-center text-lg font-medium flex-shrink-0 shadow-sm">
+              {{ (item.name || item.username).charAt(0).toUpperCase() }}
+            </div>
+
+            <div class="flex-1 min-w-0 flex flex-col justify-center gap-0.5">
+              <div class="flex justify-between items-center">
+                <h3 class="text-sm truncate" :class="isDark ? 'text-gray-100' : 'text-gray-900'">{{ item.name || item.username }}</h3>
+                <span v-if="item.formatted_last_time" class="text-[11px] flex-shrink-0 ml-2" :class="isDark ? 'text-gray-500' : 'text-gray-400'">
+                  {{ item.formatted_last_time }}
+                </span>
+              </div>
+
+              <div
+                  class="text-[10px] px-1.5 py-0.5 rounded w-max mt-0.5"
+                  :class="[
+                      item.type === 1 ? (isDark ? 'text-[#03C160] bg-[#03C160]/20' : 'text-[#03C160] bg-[#03C160]/10') : // 服务号
+                      item.type === 0 ? (isDark ? 'text-blue-400 bg-blue-900/40' : 'text-blue-500 bg-blue-50') :         // 公众号
+                      item.type === 2 ? (isDark ? 'text-orange-400 bg-orange-900/40' : 'text-orange-500 bg-orange-50') : // 企业号
+                                        (isDark ? 'text-gray-400 bg-gray-700/50' : 'text-gray-400 bg-gray-100')          // 未知
+                  ]"
+              >
+                {{ {1: '服务号', 0: '公众号', 2: '企业号', 3: '未知'}[item.type] || '未知' }}
+              </div>
+            </div>
+          </div>
+        </div>
+      </div>
+    </div>
+
+    <div class="flex-1 flex flex-col min-h-0 min-w-0" :class="isDark ? 'bg-[#121212]' : 'bg-[#F5F5F5]'">
+      <div v-if="selectedBizAccount" class="flex-1 flex flex-col min-h-0 relative">
+        <div class="h-14 border-b flex items-center px-5 shrink-0 z-10" :class="isDark ? 'bg-[#121212] border-[#333]' : 'bg-[#F5F5F5] border-gray-200'">
+          <h2 class="text-base" :class="isDark ? 'text-gray-100' : 'text-gray-900'">{{ selectedBizAccount.name }}</h2>
+        </div>
+
+        <div class="flex-1 overflow-y-auto px-4 py-6 flex flex-col-reverse" @scroll="handleScroll" ref="messageListRef">
+          <div class="h-4 shrink-0" aria-hidden="true"></div>
+          <div v-if="!hasMore" class="text-center text-xs py-4 w-full" :class="isDark ? 'text-gray-500' : 'text-gray-400'">没有更多消息了</div>
+          <div v-if="loadingMessages" class="text-center text-xs py-4 w-full" :class="isDark ? 'text-gray-500' : 'text-gray-400'">正在加载...</div>
+
+          <div class="w-full max-w-[400px] mx-auto flex flex-col-reverse gap-6">
+            <div v-for="msg in messages" :key="msg.local_id" class="w-full">
+
+              <div v-if="selectedBizAccount.username === 'gh_3dfda90e39d6'" class="rounded-xl shadow-sm p-5 border" :class="isDark ? 'bg-[#1e1e1e] border-[#333]' : 'bg-white border-gray-100'">
+                <div class="flex items-center text-sm mb-5" :class="isDark ? 'text-gray-400' : 'text-gray-500'">
+                  <img v-if="msg.merchant_icon" :src="api.getBizProxyImageUrl(msg.merchant_icon)" class="w-6 h-6 rounded-full mr-2 object-cover"  alt=""/>
+                  <div v-else class="w-6 h-6 rounded-full mr-2 flex items-center justify-center" :class="isDark ? 'bg-green-900/40 text-green-400' : 'bg-green-100 text-green-600'">¥</div>
+                  <span>{{ msg.merchant_name || '微信支付' }}</span>
+                </div>
+                <div class="text-center mb-6">
+                  <h3 class="text-[22px] font-medium mb-1" :class="isDark ? 'text-gray-100' : 'text-gray-900'">{{ msg.title }}</h3>
+                </div>
+                <div class="text-[13px] whitespace-pre-wrap leading-relaxed" :class="isDark ? 'text-gray-400' : 'text-gray-500'">
+                  {{ msg.description }}
+                </div>
+                <div class="mt-4 pt-3 border-t text-[12px] text-right" :class="isDark ? 'border-[#333] text-gray-500' : 'border-gray-100 text-gray-400'">
+                  {{ msg.formatted_time }}
+                </div>
+              </div>
+
+              <div v-else class="rounded-xl shadow-sm overflow-hidden border" :class="isDark ? 'bg-[#1e1e1e] border-[#333]' : 'bg-white border-gray-100'">
+                <a :href="msg.url" target="_blank" class="block relative group cursor-pointer">
+                  <img :src="msg.cover ? api.getBizProxyImageUrl(msg.cover) : defaultImage" :class="['w-full h-[180px] object-cover', isDark ? 'bg-[#333]' : 'bg-gray-100']"  alt=""/>
+                  <div class="absolute bottom-0 left-0 right-0 bg-gradient-to-t from-black/80 to-transparent p-3 pt-8">
+                    <h3 class="text-white text-[15px] font-medium leading-snug line-clamp-2 group-hover:underline">
+                      {{ msg.title }}
+                    </h3>
+                  </div>
+                </a>
+
+                <div v-if="msg.des" class="px-4 py-3 text-[13px] border-b" :class="isDark ? 'text-gray-400 border-[#333]' : 'text-gray-500 border-gray-50'">
+                  {{ msg.des }}
+                </div>
+
+                <div v-if="msg.content_list && msg.content_list.length > 1" class="flex flex-col">
+                  <a
+                      v-for="(item, idx) in msg.content_list.slice(1)"
+                      :key="idx"
+                      :href="item.url"
+                      target="_blank"
+                      class="flex items-center justify-between p-3 border-t hover:bg-opacity-50 cursor-pointer group"
+                      :class="isDark ? 'border-[#333] hover:bg-[#252525]' : 'border-gray-100 hover:bg-gray-50'"
+                  >
+                    <span class="text-[14px] leading-snug line-clamp-2 pr-3 group-hover:underline" :class="isDark ? 'text-gray-200' : 'text-gray-800'">
+                      {{ item.title }}
+                    </span>
+                    <img :src="item.cover ? api.getBizProxyImageUrl(item.cover) : defaultImage" :class="['w-12 h-12 rounded object-cover flex-shrink-0 border', isDark ? 'bg-[#333] border-[#444]' : 'bg-gray-100 border-gray-100']"  alt=""/>
+                  </a>
+                </div>
+              </div>
+
+            </div>
+          </div>
+        </div>
+      </div>
+
+      <div v-else class="flex-1 flex items-center justify-center">
+        <div class="text-center">
+          <div class="w-20 h-20 mx-auto mb-5 rounded-2xl flex items-center justify-center" :class="isDark ? 'bg-[#2a2a2a]' : 'bg-gray-200/50'">
+            <svg class="w-10 h-10" :class="isDark ? 'text-gray-600' : 'text-gray-400'" fill="none" viewBox="0 0 24 24" stroke="currentColor">
+              <path stroke-linecap="round" stroke-linejoin="round" stroke-width="1.5" d="M19 20H5a2 2 0 01-2-2V6a2 2 0 012-2h10a2 2 0 012 2v1m2 13a2 2 0 01-2-2V7m2 13a2 2 0 002-2V9.5L18.5 7H20" />
+            </svg>
+          </div>
+          <p class="text-sm" :class="isDark ? 'text-gray-500' : 'text-gray-400'">请选择一个服务号查看消息</p>
+        </div>
+      </div>
+    </div>
+  </div>
+</template>
+
+<script setup>
+import { ref, computed, onMounted, watch } from 'vue'
+
+import { useApi } from '~/composables/useApi'
+const api = useApi()
+
+import { storeToRefs } from 'pinia'
+import { useThemeStore } from '~/stores/theme'
+import { useChatAccountsStore } from '~/stores/chatAccounts'
+import { useChatRealtimeStore } from '~/stores/chatRealtime'
+
+const accounts = ref([])
+const loadingAccounts = ref(false)
+const searchQuery = ref('')
+const selectedBizAccount = ref(null)
+
+const themeStore = useThemeStore()
+const chatAccountsStore = useChatAccountsStore()
+const realtimeStore = useChatRealtimeStore()
+const { isDark } = storeToRefs(themeStore)
+const { selectedAccount: selectedDbAccount } = storeToRefs(chatAccountsStore)
+const { enabled: realtimeEnabled, changeSeq } = storeToRefs(realtimeStore)
+
+const messages = ref([])
+const loadingMessages = ref(false)
+const offset = ref(0)
+const limit = 20
+const hasMore = ref(true)
+
+const messageListRef = ref(null)
+let realtimeRefreshFuture = null
+let realtimeRefreshQueued = false
+
+// 默认占位图
+// const defaultAvatar = 'data:image/svg+xml;base64,PHN2ZyB4bWxucz0iaHR0cDovL3d3dy53My5vcmcvMjAwMC9zdmciIHdpZHRoPSI0MCIgaGVpZ2h0PSI0MCIgdmlld0JveD0iMCAwIDQwIDQwIj48cmVjdCB3aWR0aD0iNDAiIGhlaWdodD0iNDAiIGZpbGw9IiNlNWU3ZWIiLz48L3N2Zz4='
+const defaultImage = 'data:image/svg+xml;base64,PHN2ZyB4bWxucz0iaHR0cDovL3d3dy53My5vcmcvMjAwMC9zdmciIHdpZHRoPSI0MDAiIGhlaWdodD0iMTgwIj48cmVjdCB3aWR0aD0iNDAwIiBoZWlnaHQ9IjE4MCIgZmlsbD0iI2Y1ZjVmNSIvPjwvc3ZnPg=='
+
+const getCurrentAccountParam = () => {
+  const account = String(selectedDbAccount.value || '').trim()
+  return account || undefined
+}
+
+const resetMessagesState = () => {
+  messages.value = []
+  offset.value = 0
+  hasMore.value = true
+}
+
+const fetchAccounts = async ({ preserveSelection = true } = {}) => {
+  loadingAccounts.value = true
+  const previousUsername = preserveSelection ? String(selectedBizAccount.value?.username || '').trim() : ''
+  try {
+    const res = await api.listBizAccounts({ account: getCurrentAccountParam() })
+    const nextAccounts = Array.isArray(res?.data) ? res.data : []
+    accounts.value = nextAccounts
+
+    if (previousUsername) {
+      selectedBizAccount.value = nextAccounts.find(item => item.username === previousUsername) || null
+    } else if (!selectedBizAccount.value?.username) {
+      selectedBizAccount.value = null
+    }
+  } catch (err) {
+    accounts.value = []
+    selectedBizAccount.value = null
+    console.error('获取服务号失败:', err)
+  } finally {
+    loadingAccounts.value = false
+  }
+}
+
+// 搜索过滤
+const filteredAccounts = computed(() => {
+  if (!searchQuery.value) return accounts.value
+  const q = searchQuery.value.toLowerCase()
+  return accounts.value.filter(a =>
+      (a.name && a.name.toLowerCase().includes(q)) ||
+      (a.username && a.username.toLowerCase().includes(q))
+  )
+})
+
+// 点击选择服务号
+const selectAccount = async (account) => {
+  if (selectedBizAccount.value?.username === account.username) return
+  selectedBizAccount.value = account
+
+  // 重置消息状态
+  resetMessagesState()
+
+  await loadMessages()
+}
+
+// 加载消息
+const loadMessages = async () => {
+  if (loadingMessages.value || !hasMore.value || !selectedBizAccount.value) return
+
+  loadingMessages.value = true
+  try {
+    const username = selectedBizAccount.value.username
+    const params = {
+      account: getCurrentAccountParam(),
+      username,
+      offset: offset.value,
+      limit,
+    }
+
+    let res
+    if (username === 'gh_3dfda90e39d6') {
+      res = await api.listBizPayRecords(params)
+    } else {
+      res = await api.listBizMessages(params)
+    }
+
+    if (res && res.data) {
+      if (res.data.length < limit) {
+        hasMore.value = false
+      }
+      // 追加数据
+      messages.value.push(...res.data)
+      offset.value += limit
+    }
+  } catch (err) {
+    console.error('加载消息失败:', err)
+  } finally {
+    loadingMessages.value = false
+  }
+}
+
+const reloadSelectedMessages = async () => {
+  if (!selectedBizAccount.value) return
+  resetMessagesState()
+  await loadMessages()
+}
+
+const syncAllBizRealtime = async ({ forceReload = false } = {}) => {
+  const priorityUsername = String(selectedBizAccount.value?.username || '').trim()
+  if (!realtimeEnabled.value) {
+    if (forceReload) {
+      await reloadSelectedMessages()
+    }
+    return
+  }
+
+  try {
+    const result = await api.syncChatRealtimeAll({
+      account: getCurrentAccountParam(),
+      max_scan: 200,
+      priority_username: priorityUsername,
+      priority_max_scan: 400,
+      include_hidden: true,
+      include_official: true,
+      only_official: true,
+      backfill_limit: 0,
+    })
+    const hasDelta = Number(result?.insertedTotal || 0) > 0 || Number(result?.sessionsUpdated || 0) > 0
+    await fetchAccounts({ preserveSelection: true })
+    if (selectedBizAccount.value?.username) {
+      if (hasDelta || forceReload) {
+        await reloadSelectedMessages()
+      }
+    } else if (forceReload) {
+      resetMessagesState()
+    }
+  } catch (err) {
+    console.error('实时同步服务号失败:', err)
+    if (forceReload) {
+      await fetchAccounts({ preserveSelection: true })
+      await reloadSelectedMessages()
+    }
+  }
+}
+
+const queueRealtimeBizRefresh = () => {
+  if (!realtimeEnabled.value) return
+  if (realtimeRefreshFuture) {
+    realtimeRefreshQueued = true
+    return
+  }
+
+  realtimeRefreshFuture = syncAllBizRealtime().finally(() => {
+    realtimeRefreshFuture = null
+    if (realtimeRefreshQueued) {
+      realtimeRefreshQueued = false
+      queueRealtimeBizRefresh()
+    }
+  })
+}
+
+// 向上滚动加载逻辑
+// 因为容器设置了 flex-col-reverse，所以 scrollTop 越靠近负值(或0取决于浏览器)越是到了历史消息端
+// 但比较通用兼容的做法是监听 scroll，距离顶部或底部小于阈值时触发
+const handleScroll = (e) => {
+  const target = e.target
+  // 针对 flex-col-reverse: 滚动到底部实际上是视觉上的最上方(历史消息)
+  // 当 scrollHeight - Math.abs(scrollTop) - clientHeight < 50 时加载
+  if (target.scrollHeight - Math.abs(target.scrollTop) - target.clientHeight < 50) {
+    loadMessages()
+  }
+}
+
+watch(selectedDbAccount, async (next, prev) => {
+  if (String(next || '').trim() === String(prev || '').trim()) return
+  selectedBizAccount.value = null
+  resetMessagesState()
+  searchQuery.value = ''
+
+  if (!String(next || '').trim()) {
+    accounts.value = []
+    return
+  }
+  await fetchAccounts({ preserveSelection: false })
+  if (realtimeEnabled.value) {
+    await syncAllBizRealtime({ forceReload: true })
+  }
+})
+
+watch(changeSeq, (next, prev) => {
+  if (!realtimeEnabled.value) return
+  if (next === prev) return
+  queueRealtimeBizRefresh()
+})
+
+watch(realtimeEnabled, async (enabled, wasEnabled) => {
+  if (enabled && !wasEnabled) {
+    await syncAllBizRealtime({ forceReload: true })
+  }
+})
+
+onMounted(async () => {
+  await chatAccountsStore.ensureLoaded()
+  await fetchAccounts({ preserveSelection: false })
+  if (realtimeEnabled.value) {
+    await syncAllBizRealtime({ forceReload: true })
+  }
+})
+</script>
+
+<style scoped>
+/* 隐藏滚动条但允许滚动（可选） */
+.overflow-y-auto::-webkit-scrollbar {
+  width: 6px;
+}
+.overflow-y-auto::-webkit-scrollbar-track {
+  background: transparent;
+}
+.overflow-y-auto::-webkit-scrollbar-thumb {
+  background-color: rgba(0,0,0,0.1);
+  border-radius: 10px;
+}
+</style>

frontend/components/SidebarRail.vue

@@ -101,6 +101,21 @@
         </div>
       </div>
 
+      <div
+          class="sidebar-rail-action w-full h-[var(--sidebar-rail-step)] flex items-center justify-center cursor-pointer group"
+          title="服务号"
+          @click="goBiz"
+      >
+        <div class="sidebar-rail-plate w-[var(--sidebar-rail-btn)] h-[var(--sidebar-rail-btn)] rounded-md flex items-center justify-center transition-colors bg-transparent">
+          <div class="sidebar-rail-icon w-[var(--sidebar-rail-icon)] h-[var(--sidebar-rail-icon)]" :class="{ 'sidebar-rail-icon-active': isBizRoute }">
+            <svg class="w-full h-full" viewBox="0 0 24 24" fill="none" stroke="currentColor" stroke-width="1.8" stroke-linecap="round" stroke-linejoin="round" aria-hidden="true">
+              <path d="M11 5L6 9H2v6h4l5 4V5z"></path>
+              <path d="M19.07 4.93a10 10 0 0 1 0 14.14M15.54 8.46a5 5 0 0 1 0 7.07"></path>
+            </svg>
+          </div>
+        </div>
+      </div>
+
       <!-- Wrapped -->
       <div
         class="sidebar-rail-action w-full h-[var(--sidebar-rail-step)] flex items-center justify-center cursor-pointer group"
@@ -479,34 +494,17 @@ const isChatRoute = computed(() => route.path?.startsWith('/chat'))
 const isEditsRoute = computed(() => route.path?.startsWith('/edits'))
 const isSnsRoute = computed(() => route.path?.startsWith('/sns'))
 const isContactsRoute = computed(() => route.path?.startsWith('/contacts'))
+const isBizRoute = computed(() => route.path?.startsWith('/biz')) // 新增
 const isWrappedRoute = computed(() => route.path?.startsWith('/wrapped'))
-const goChat = async () => {
-  await navigateTo('/chat')
-}
 
-const goEdits = async () => {
-  await navigateTo('/edits')
-}
-
-const goSns = async () => {
-  await navigateTo('/sns')
-}
-
-const goContacts = async () => {
-  await navigateTo('/contacts')
-}
-
-const goWrapped = async () => {
-  await navigateTo('/wrapped')
-}
-
-const goGuide = async () => {
-  await navigateTo('/')
-}
-
-const goSettings = () => {
-  openSettingsDialog()
-}
+const goChat = async () => { await navigateTo('/chat') }
+const goEdits = async () => { await navigateTo('/edits') }
+const goSns = async () => { await navigateTo('/sns') }
+const goContacts = async () => { await navigateTo('/contacts') }
+const goBiz = async () => { await navigateTo('/biz') }
+const goWrapped = async () => { await navigateTo('/wrapped') }
+const goGuide = async () => { await navigateTo('/') }
+const goSettings = () => { openSettingsDialog() }
 
 const onWindowKeydown = (event) => {
   if (event?.key !== 'Escape') return

frontend/components/chat/LinkCard.vue

@@ -2,6 +2,8 @@
 import { defineComponent, h, ref, watch } from 'vue'
 import miniProgramIconUrl from '~/assets/images/wechat/mini-program.svg'
 
+const finderLogoUrl = '/assets/images/wechat/channels-logo.svg'
+
 export default defineComponent({
   name: 'LinkCard',
   props: {
@@ -51,7 +53,11 @@ export default defineComponent({
         return text ? (Array.from(text)[0] || '') : ''
       })()
       const fromAvatarUrl = String(props.fromAvatar || '').trim()
+      const headingText = String(props.heading || href || '').trim()
+      let abstractText = String(props.abstract || '').trim()
+      if (abstractText && headingText && abstractText === headingText) abstractText = ''
       const isMiniProgram = String(props.linkType || '').trim() === 'mini_program'
+      const isFinder = String(props.linkType || '').trim() === 'finder'
       const isCoverVariant = !isMiniProgram && String(props.variant || '').trim() === 'cover'
       const Tag = canNavigate ? 'a' : 'div'
 
@@ -140,9 +146,68 @@ export default defineComponent({
         )
       }
 
-      const headingText = String(props.heading || href || '').trim()
-      let abstractText = String(props.abstract || '').trim()
-      if (abstractText && headingText && abstractText === headingText) abstractText = ''
+      if (isFinder) {
+        return h(
+          Tag,
+          {
+            ...(canNavigate ? { href, target: '_blank', rel: 'noreferrer' } : { role: 'group', 'aria-disabled': 'true' }),
+            class: [
+              'wechat-link-card-finder',
+              !canNavigate ? 'wechat-link-card--disabled' : '',
+              'wechat-special-card',
+              'msg-radius',
+              props.isSent ? 'wechat-special-sent-side' : ''
+            ].filter(Boolean).join(' '),
+            style: {
+              width: '135px',
+              minWidth: '135px',
+              maxWidth: '135px',
+              display: 'flex',
+              flexDirection: 'column',
+              boxSizing: 'border-box',
+              flex: '0 0 auto',
+              border: 'none',
+              boxShadow: 'none',
+              textDecoration: 'none',
+              outline: 'none'
+            }
+          },
+          [
+            h('div', { class: ['wechat-link-finder-cover', !props.preview ? 'wechat-link-finder-cover--empty' : ''].filter(Boolean).join(' ') }, [
+              props.preview
+                ? h('img', {
+                    src: props.preview,
+                    alt: props.heading || '视频号封面',
+                    class: 'wechat-link-finder-cover-img',
+                    referrerpolicy: 'no-referrer'
+                  })
+                : h('div', { class: 'wechat-link-finder-cover-placeholder', 'aria-hidden': 'true' }, [
+                    h('svg', { viewBox: '0 0 24 24', fill: 'currentColor' }, [
+                      h('path', { d: 'M8 5v14l11-7z' })
+                    ])
+                  ]),
+              h('div', { class: 'wechat-link-finder-cover-shade', 'aria-hidden': 'true' }),
+              h('div', { class: 'wechat-link-finder-play', 'aria-hidden': 'true' }, [
+                h('svg', { viewBox: '0 0 24 24', fill: 'currentColor' }, [
+                  h('path', { d: 'M8 5v14l11-7z' })
+                ])
+              ]),
+              h('div', { class: 'wechat-link-finder-meta' }, [
+                h('div', { class: 'wechat-link-finder-author' }, [
+                  h('div', { class: 'wechat-link-finder-author-avatar', 'aria-hidden': 'true' }, [
+                    h('img', {
+                      src: finderLogoUrl,
+                      alt: '',
+                      class: 'wechat-link-finder-author-avatar-img'
+                    })
+                  ]),
+                  h('div', { class: 'wechat-link-finder-author-name' }, fromText || '视频号')
+                ])
+              ])
+            ])
+          ]
+        )
+      }
 
       if (isMiniProgram) {
         return h(

frontend/components/chat/SessionListPanel.vue

@@ -69,7 +69,7 @@
           <div v-else-if="contacts.length === 0" class="session-list-status px-3 py-2 text-sm">
             暂无会话
           </div>
-          <template v-else>
+          <div v-else class="pb-4">
             <div v-for="contact in filteredContacts" :key="contact.id"
               class="session-list-item px-3 cursor-pointer transition-colors duration-150 h-[calc(80px/var(--dpr))] flex items-center"
               :class="{
@@ -98,7 +98,7 @@
                 <!-- 联系人信息 -->
                 <div class="flex-1 min-w-0">
                   <div class="flex items-center justify-between">
-                    <h3 class="session-list-item-name text-sm font-medium truncate" :class="{ 'privacy-blur': privacyMode }">{{ contact.name }}</h3>
+                    <h3 class="session-list-item-name text-sm truncate" :class="{ 'privacy-blur': privacyMode }">{{ contact.name }}</h3>
                     <div class="flex items-center flex-shrink-0 ml-2">
                       <span class="session-list-item-time text-xs">{{ contact.lastMessageTime }}</span>
                     </div>
@@ -118,7 +118,7 @@
                 </div>
               </div>
             </div>
-          </template>
+          </div>
         </div>
       </div>
 

frontend/composables/useApi.js

@@ -205,6 +205,8 @@ export const useApi = () => {
     if (params && params.priority_max_scan != null) query.set('priority_max_scan', String(params.priority_max_scan))
     if (params && params.include_hidden != null) query.set('include_hidden', String(!!params.include_hidden))
     if (params && params.include_official != null) query.set('include_official', String(!!params.include_official))
+    if (params && params.only_official != null) query.set('only_official', String(!!params.only_official))
+    if (params && params.backfill_limit != null) query.set('backfill_limit', String(params.backfill_limit))
     const url = '/chat/realtime/sync_all' + (query.toString() ? `?${query.toString()}` : '')
     return await request(url, { method: 'POST' })
   }
@@ -561,6 +563,44 @@ export const useApi = () => {
     return await request('/get_image_key')
   }
 
+  // 枚举服务号信息
+  const listBizAccounts = async (params = {}) => {
+    const query = new URLSearchParams()
+    if (params && params.account) query.set('account', params.account)
+    const url = '/biz/list' + (query.toString() ? `?${query.toString()}` : '')
+    return await request(url)
+  }
+
+  // 获取普通服务号消息
+  const listBizMessages = async (params = {}) => {
+    const query = new URLSearchParams()
+    if (params && params.account) query.set('account', params.account)
+    if (params && params.username) query.set('username', params.username)
+    if (params && params.limit != null) query.set('limit', String(params.limit))
+    if (params && params.offset != null) query.set('offset', String(params.offset))
+    const url = '/biz/messages' + (query.toString() ? `?${query.toString()}` : '')
+    return await request(url)
+  }
+
+  // 获取微信支付记录
+  const listBizPayRecords = async (params = {}) => {
+    const query = new URLSearchParams()
+    if (params && params.account) query.set('account', params.account)
+    if (params && params.limit != null) query.set('limit', String(params.limit))
+    if (params && params.offset != null) query.set('offset', String(params.offset))
+    const url = '/biz/pay_records' + (query.toString() ? `?${query.toString()}` : '')
+    return await request(url)
+  }
+
+  const getBizProxyImageUrl = (url) => {
+    if (!url) return ''
+    if (url.startsWith('data:')) return url // 如果已经是 base64，不处理
+    const query = new URLSearchParams()
+    query.set('url', url)
+    const base = baseURL ? baseURL.replace(/\/$/, '') : ''
+    return `${base}/biz/proxy_image?${query.toString()}`
+  }
+
   return {
     detectWechat,
     detectCurrentAccount,
@@ -616,5 +656,9 @@ export const useApi = () => {
     getKeys,
     getImageKey,
     getWxStatus,
+    listBizAccounts,
+    listBizMessages,
+    listBizPayRecords,
+    getBizProxyImageUrl,
   }
 }

frontend/pages/biz.vue

@@ -0,0 +1,16 @@
+<template>
+  <div class="h-full min-h-0 flex overflow-hidden bg-white">
+    <div class="flex-1 min-w-0">
+      <BizMessages />
+    </div>
+  </div>
+</template>
+
+<script setup>
+
+import BizMessages from "../components/BizMessages.vue";
+
+useHead({
+  title: '服务号消息 - WeChatDataAnalysis'
+})
+</script>

frontend/pages/decrypt.vue

@@ -73,9 +73,6 @@
                 </svg>
                 点击按钮将自动获取【数据库】与【图片】双重密钥。您也可以手动输入已知的64位密钥（使用<a href="https://github.com/ycccccccy/wx_key" target="_blank" class="text-[#07C160] hover:text-[#06AD56]">wx_key</a>等工具获取）。
               </p>
-              <div class="mt-3 rounded-lg border border-amber-200 bg-amber-50 px-3 py-2 text-xs leading-5 text-amber-900">
-                提示：数据库密钥跟随“账号 + 设备”下发。同一账号在另一台电脑生成的聊天记录，复制到当前设备后，通常无法在当前设备重新获取原设备对应的密钥，因此也无法直接解密。
-              </div>
             </div>
             
             <!-- 数据库路径输入 -->

frontend/public/assets/images/wechat/channels-logo.svg

@@ -0,0 +1,5 @@
+<?xml version="1.0" standalone="no"?>
+<!DOCTYPE svg PUBLIC "-//W3C//DTD SVG 1.1//EN" "http://www.w3.org/Graphics/SVG/1.1/DTD/svg11.dtd">
+<svg t="1774499781741" class="icon" viewBox="0 0 1024 1024" version="1.1" xmlns="http://www.w3.org/2000/svg" p-id="7897" xmlns:xlink="http://www.w3.org/1999/xlink" width="200" height="200">
+<path d="M118.81813333 106.3936c87.27893333-26.2144 192.03413333 75.1616 358.46826667 346.9312 18.70506667 30.5152 34.7136 55.46666667 35.60106667 55.5008 0.88746667 0 17.74933333-26.4192 37.41013333-58.70933333 165.51253333-271.53066667 270.336-371.3024 359.35573333-342.08426667 56.55893333 18.56853333 80.41813333 73.18186667 80.21333334 183.63733333-0.4096 214.86933333-103.69706667 551.59466667-188.0064 612.89813334-69.4272 50.44906667-173.1584-13.07306667-269.85813334-165.30773334-9.59146667-15.1552-18.36373333-27.57973333-19.42186666-27.61386666-1.05813333 0-9.59146667 12.32213333-18.944 27.4432-50.96106667 82.39786667-113.4592 146.26133333-167.04853334 170.7008-26.04373333 11.8784-71.33866667 13.5168-90.45333333 3.24266666-52.08746667-27.98933333-110.72853333-149.504-156.16-323.72053333C7.3728 310.95466667 21.504 135.68 118.81813333 106.3936zM848.31573333 217.088c-55.26186667 42.93973333-126.49813333 138.58133333-230.8096 309.93066667l-49.2544 80.82773333 16.86186667 30.17386667c42.35946667 75.94666667 91.30666667 139.81013333 130.79893333 170.66666666 26.76053333 20.95786667 35.60106667 16.55466667 58.9824-29.4912 73.5232-144.55466667 136.192-440.7296 115.712-547.19146666-6.144-32.0512-15.80373333-35.46453333-42.2912-14.91626667zM143.73546667 207.9744c-19.72906667 19.49013333-14.60906667 145.8176 10.99093333 271.90613333 30.89066667 152.23466667 95.91466667 329.3184 124.5184 339.2512 27.81866667 9.65973333 104.31146667-77.824 164.38613333-188.0064l13.14133334-24.13226666-42.15466667-69.18826667c-112.98133333-185.344-186.64106667-284.3648-240.8448-323.72053333-16.65706667-12.0832-22.7328-13.312-30.03733333-6.10986667z" fill="#FF9908" p-id="7898"></path>
+</svg>

src/wechat_decrypt_tool/api.py

@@ -36,6 +36,7 @@ from .routers.wrapped import router as _wrapped_router
 from .request_logging import log_server_errors_middleware
 from .sns_stage_timing import add_sns_stage_timing_headers
 from .wcdb_realtime import WCDB_REALTIME, shutdown as _wcdb_shutdown
+from .routers.biz import router as _biz_router
 
 app = FastAPI(
     title="微信数据库解密工具",
@@ -96,6 +97,7 @@ app.include_router(_chat_media_router)
 app.include_router(_sns_router)
 app.include_router(_sns_export_router)
 app.include_router(_wrapped_router)
+app.include_router(_biz_router)
 
 
 class _SPAStaticFiles(StaticFiles):

src/wechat_decrypt_tool/chat_helpers.py

@@ -1220,6 +1220,70 @@ def _parse_app_message(text: str) -> dict[str, Any]:
             "linkStyle": link_style,
         }
 
+    if app_type == 51:
+        # 视频号分享（Finder / Channels）
+        # 常见特征：
+        # - title 是「当前版本不支持展示该内容，请升级至最新版本。」
+        # - 真正标题在 <finderFeed><desc> 或其它 finder 节点里
+        finder_feed = _extract_xml_tag_text(text, "finderFeed")
+        finder_desc = (
+            (_extract_xml_tag_text(finder_feed, "desc") if finder_feed else "")
+            or _extract_xml_tag_text(text, "finderdesc")
+            or des
+        )
+        finder_nickname = (
+            _extract_xml_tag_text(text, "findernickname")
+            or _extract_xml_tag_text(text, "finder_nickname")
+            or (_extract_xml_tag_text(finder_feed, "nickname") if finder_feed else "")
+            or (_extract_xml_tag_text(finder_feed, "findernickname") if finder_feed else "")
+        )
+        finder_username = (
+            _extract_xml_tag_text(text, "finderusername")
+            or _extract_xml_tag_text(text, "finder_username")
+            or (_extract_xml_tag_text(finder_feed, "username") if finder_feed else "")
+            or (_extract_xml_tag_text(finder_feed, "finderusername") if finder_feed else "")
+        )
+
+        thumb_url = _normalize_xml_url(
+            _extract_xml_tag_or_attr(text, "thumburl")
+            or _extract_xml_tag_or_attr(text, "cdnthumburl")
+            or _extract_xml_tag_or_attr(text, "coverurl")
+            or _extract_xml_tag_or_attr(text, "cover")
+            or (_extract_xml_tag_or_attr(finder_feed, "thumbUrl") if finder_feed else "")
+            or (_extract_xml_tag_or_attr(finder_feed, "thumburl") if finder_feed else "")
+            or (_extract_xml_tag_or_attr(finder_feed, "coverUrl") if finder_feed else "")
+            or (_extract_xml_tag_or_attr(finder_feed, "coverurl") if finder_feed else "")
+        )
+
+        finder_url = url or _normalize_xml_url(
+            (_extract_xml_tag_text(finder_feed, "url") if finder_feed else "")
+            or (_extract_xml_tag_text(text, "playurl"))
+            or (_extract_xml_tag_text(text, "dataurl"))
+        )
+
+        display_title = str(title or "").strip()
+        if (not display_title) or ("不支持" in display_title):
+            display_title = str(finder_desc or "").strip()
+        if not display_title:
+            display_title = str(des or "").strip()
+        display_title = display_title or "[视频号]"
+
+        summary_text = str(finder_desc or "").strip() or display_title
+        from_display = str(finder_nickname or source_display_name or "").strip() or "视频号"
+        from_u = str(finder_username or source_username or "").strip()
+
+        return {
+            "renderType": "link",
+            "content": summary_text,
+            "title": display_title,
+            "url": finder_url or "",
+            "thumbUrl": thumb_url or "",
+            "from": from_display,
+            "fromUsername": from_u,
+            "linkType": "finder",
+            "linkStyle": "finder",
+        }
+
     if app_type in (33, 36):
         # 小程序分享（WeChat v4 常见：local_type = 49 + (33<<32) / 49 + (36<<32)）
         # 注：部分 payload 的 <url> 为空；前端会按需渲染为不可点击卡片。

src/wechat_decrypt_tool/logging_config.py

@@ -41,6 +41,22 @@ class ColoredFormatter(logging.Formatter):
         return formatted
 
 
+def _can_use_logging_stream(stream) -> bool:
+    try:
+        if stream is None or getattr(stream, "closed", False):
+            return False
+    except Exception:
+        return False
+
+    try:
+        stream.write("")
+        stream.flush()
+    except Exception:
+        return False
+
+    return True
+
+
 class WeChatLogger:
     """微信解密工具统一日志管理器"""
     
@@ -64,6 +80,12 @@ class WeChatLogger:
         if env_level:
             log_level = env_level
 
+        console_logging_env = str(os.environ.get("WECHAT_TOOL_ENABLE_CONSOLE_LOG", "") or "").strip().lower()
+        console_logging_forced = console_logging_env in {"1", "true", "yes", "on"}
+        console_logging_disabled = console_logging_env in {"0", "false", "no", "off"}
+
+        level = getattr(logging, str(log_level or "INFO").upper(), logging.INFO)
+
         # 创建日志目录
         now = datetime.now()
         from .app_paths import get_output_dir
@@ -73,10 +95,41 @@ class WeChatLogger:
         
         # 设置日志文件名
         date_str = now.strftime("%d")
-        self.log_file = log_dir / f"{date_str}_wechat_tool.log"
+        desired_log_file = log_dir / f"{date_str}_wechat_tool.log"
+
+        root_logger = logging.getLogger()
+        wants_console_handler = _can_use_logging_stream(sys.stdout)
+        if getattr(sys, "frozen", False) and not console_logging_forced:
+            wants_console_handler = False
+        if console_logging_disabled:
+            wants_console_handler = False
+
+        if WeChatLogger._initialized:
+            current_log_file = Path(getattr(self, "log_file", desired_log_file))
+            has_expected_file_handler = False
+            has_stream_handler = False
+            for handler in root_logger.handlers:
+                if isinstance(handler, logging.FileHandler):
+                    try:
+                        if Path(handler.baseFilename).resolve() == desired_log_file.resolve():
+                            has_expected_file_handler = True
+                    except Exception:
+                        if Path(handler.baseFilename) == desired_log_file:
+                            has_expected_file_handler = True
+                elif isinstance(handler, logging.StreamHandler):
+                    has_stream_handler = True
+            if (
+                current_log_file == desired_log_file
+                and root_logger.level == level
+                and has_expected_file_handler
+                and (has_stream_handler or not wants_console_handler)
+            ):
+                self.log_file = desired_log_file
+                return self.log_file
+
+        self.log_file = desired_log_file
         
         # 清除现有的处理器
-        root_logger = logging.getLogger()
         for handler in root_logger.handlers[:]:
             root_logger.removeHandler(handler)
             try:
@@ -100,18 +153,20 @@ class WeChatLogger:
         # 文件处理器
         file_handler = logging.FileHandler(self.log_file, encoding='utf-8')
         file_handler.setFormatter(file_formatter)
-        level = getattr(logging, str(log_level or "INFO").upper(), logging.INFO)
         file_handler.setLevel(level)
 
         # 控制台处理器
-        console_handler = logging.StreamHandler(sys.stdout)
-        console_handler.setFormatter(console_formatter)
-        console_handler.setLevel(level)
+        console_handler = None
+        if wants_console_handler:
+            console_handler = logging.StreamHandler(sys.stdout)
+            console_handler.setFormatter(console_formatter)
+            console_handler.setLevel(level)
         
         # 配置根日志器
         root_logger.setLevel(level)
         root_logger.addHandler(file_handler)
-        root_logger.addHandler(console_handler)
+        if console_handler is not None:
+            root_logger.addHandler(console_handler)
         
         # 只为uvicorn日志器添加文件处理器，保持其原有的控制台处理器（带颜色）
         uvicorn_logger = logging.getLogger("uvicorn")
@@ -158,7 +213,8 @@ class WeChatLogger:
             except Exception:
                 pass
         fastapi_logger.addHandler(file_handler)
-        fastapi_logger.addHandler(console_handler)
+        if console_handler is not None:
+            fastapi_logger.addHandler(console_handler)
         fastapi_logger.setLevel(level)
         
         # 记录初始化信息

src/wechat_decrypt_tool/routers/biz.py

@@ -0,0 +1,370 @@
+import hashlib
+import sqlite3
+import time
+import xml.etree.ElementTree as ET
+from pathlib import Path
+from typing import Optional, Any, Dict, List
+import urllib
+from fastapi import APIRouter, HTTPException, Response
+from pydantic import BaseModel
+
+from ..chat_helpers import _resolve_account_dir
+from ..path_fix import PathFixRoute
+from ..logging_config import get_logger
+
+try:
+    import zstandard as zstd
+except Exception:
+    zstd = None
+
+logger = get_logger(__name__)
+router = APIRouter(route_class=PathFixRoute)
+
+
+def decompress_zstd_content(data: bytes, source_id: str, local_id: int) -> Optional[bytes]:
+    """Zstandard 解压逻辑"""
+    if not data or not data.startswith(b'\x28\xb5\x2f\xfd'):
+        return None
+    try:
+        if zstd:
+            dctx = zstd.ZstdDecompressor()
+            return dctx.decompress(data, max_output_size=10 * 1024 * 1024)
+    except Exception as e:
+        error_msg = f"❌ [解压失败] 服务号id: {source_id}, local_id: {local_id} -> {e}"
+        print(error_msg)
+        logger.error(error_msg)
+    return None
+
+
+def extract_xml_from_db_content(content: Any, source_id: str, local_id: int) -> str:
+    """提取并解压数据库内容"""
+    if not content:
+        return ""
+
+    if isinstance(content, memoryview):
+        content = content.tobytes()
+    elif isinstance(content, str):
+        content = content.encode('utf-8', errors='ignore')
+
+    if isinstance(content, bytes):
+        decompressed = decompress_zstd_content(content, source_id, local_id)
+        if decompressed:
+            return decompressed.decode('utf-8', errors='ignore')
+
+        # 若不是 zstd 压缩或解压失败，尝试直接 decode
+        try:
+            return content.decode('utf-8', errors='ignore')
+        except Exception:
+            return ""
+    return ""
+
+
+def parse_wechat_xml_to_struct(xml_str: str, source_id: str, local_id: int) -> Optional[Dict[str, Any]]:
+    """解析微信服务号 XML 到 Dict"""
+    if not xml_str.strip():
+        return None
+    try:
+        root = ET.fromstring(xml_str)
+
+        def get_tag_text(element, path, default=""):
+            node = element.find(path)
+            return node.text if node is not None and node.text else default
+
+        main_cover = get_tag_text(root, ".//appmsg/thumburl")
+        if not main_cover:
+            main_cover = get_tag_text(root, ".//topnew/cover")
+
+        result = {
+            "title": get_tag_text(root, ".//appmsg/title"),
+            "des": get_tag_text(root, ".//appmsg/des"),
+            "url": get_tag_text(root, ".//appmsg/url"),
+            "cover": main_cover,
+            "content_list": []
+        }
+
+        items = root.findall(".//mmreader/category/item")
+        for item in items:
+            item_struct = {
+                "title": get_tag_text(item, "title"),
+                "url": get_tag_text(item, "url"),
+                "cover": get_tag_text(item, "cover"),
+                "summary": get_tag_text(item, "summary")
+            }
+            if item_struct["title"]:
+                result["content_list"].append(item_struct)
+
+        return result
+    except Exception as e:
+        error_msg = f"❌ [解析XML失败] 服务号id: {source_id}, local_id: {local_id} -> {e}"
+        print(error_msg)
+        logger.error(error_msg)
+        return None
+
+
+def parse_pay_xml(xml_str: str, local_id: int) -> Optional[Dict[str, Any]]:
+    """解析微信支付 XML"""
+    if not xml_str.strip():
+        return None
+    try:
+        root = ET.fromstring(xml_str)
+
+        def get_text(path):
+            node = root.find(path)
+            return node.text if node is not None else ""
+
+        record = {
+            "title": get_text(".//appmsg/title"),
+            "description": get_text(".//appmsg/des"),
+            "merchant_name": get_text(".//template_header/display_name"),
+            "merchant_icon": get_text(".//template_header/icon_url"),
+            "timestamp": int(get_text(".//pub_time") or 0),
+            "formatted_time": ""
+        }
+        return record
+    except Exception as e:
+        error_msg = f"❌ [解析微信支付XML失败] 支付id: gh_3dfda90e39d6, local_id: {local_id} -> {e}"
+        print(error_msg)
+        logger.error(error_msg)
+        return None
+
+@router.get("/api/biz/proxy_image", summary="代理请求微信服务号图片")
+def proxy_biz_image(url: str):
+    if not url:
+        return Response(status_code=400)
+    try:
+        req = urllib.request.Request(url, headers={
+            'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/120.0.0.0 Safari/537.36'
+        })
+        with urllib.request.urlopen(req, timeout=10) as response:
+            content = response.read()
+            content_type = response.headers.get('Content-Type', 'image/jpeg')
+            return Response(content=content, media_type=content_type)
+    except Exception as e:
+        logger.error(f"[biz] 代理图片失败: {url} -> {e}")
+        return Response(status_code=500)
+
+# 接口 1：获取全部的服务号/公众号的信息
+@router.get("/api/biz/list", summary="获取全部服务号/公众号列表")
+def get_biz_account_list(account: Optional[str] = None):
+    account_dir = _resolve_account_dir(account)
+
+    biz_ids = set()
+    biz_latest_time = {}
+
+    # 1. 遍历 biz_message_*.db
+    for db_file in account_dir.glob("biz_message*.db"):
+        try:
+            conn = sqlite3.connect(str(db_file))
+            cursor = conn.cursor()
+
+            cursor.execute("PRAGMA table_info(Name2Id)")
+            cols = [row[1].lower() for row in cursor.fetchall()]
+            user_col = "username" if "username" in cols else "user_name" if "user_name" in cols else ""
+
+            if user_col:
+                rows = cursor.execute(f"SELECT {user_col} FROM Name2Id").fetchall()
+                for r in rows:
+                    if r[0]:
+                        uname = r[0]
+                        biz_ids.add(uname)
+
+                        # 顺便查询该号的最后一条消息时间
+                        md5_id = hashlib.md5(uname.encode('utf-8')).hexdigest().lower()
+                        table_name = f"Msg_{md5_id}"
+                        try:
+                            time_res = conn.execute(f"SELECT MAX(create_time) FROM {table_name}").fetchone()
+                            if time_res and time_res[0]:
+                                current_max = biz_latest_time.get(uname, 0)
+                                biz_latest_time[uname] = max(current_max, time_res[0])
+                        except Exception:
+                            pass
+            conn.close()
+        except Exception as e:
+            logger.warning(f"读取 Name2Id 失败 {db_file}: {e}")
+
+    contact_db_path = account_dir / "contact.db"
+    contact_info = {}
+    if contact_db_path.exists() and biz_ids:
+        try:
+            conn = sqlite3.connect(str(contact_db_path))
+            cursor = conn.cursor()
+            placeholders = ",".join(["?"] * len(biz_ids))
+
+            # 先查 contact 表
+            query_contact = f"SELECT username, remark, nick_name, alias, big_head_url FROM contact WHERE username IN ({placeholders})"
+            rows_contact = cursor.execute(query_contact, list(biz_ids)).fetchall()
+
+            for r in rows_contact:
+                uname = r[0]
+                name = r[1] or r[2] or r[3] or uname
+                contact_info[uname] = {
+                    "username": uname,
+                    "name": name,
+                    "avatar": r[4],
+                    "type": 3  # 默认给个 3（未知）
+                }
+
+            # 再查 biz_info 表获取类型
+            try:
+                query_biz = f"SELECT username, type FROM biz_info WHERE username IN ({placeholders})"
+                rows_biz = cursor.execute(query_biz, list(biz_ids)).fetchall()
+                for r in rows_biz:
+                    uname = r[0]
+                    biz_type = r[1]
+                    # 如果查到了且是 0, 1, 2，就更新进去，否则保留 3
+                    if uname in contact_info:
+                        if biz_type in (0, 1, 2):
+                            contact_info[uname]["type"] = biz_type
+                        else:
+                            contact_info[uname]["type"] = 3
+            except Exception as e:
+                logger.warning(f"读取 biz_info 失败: {e}")
+
+            conn.close()
+        except Exception as e:
+            logger.warning(f"读取 contact.db 失败: {e}")
+
+    # 3. 组装结果（不在 contact_info 里的直接丢弃）
+    result = []
+    for uid in biz_ids:
+        if uid in contact_info:
+            info = contact_info[uid]
+            info["last_time"] = biz_latest_time.get(uid, 0)
+            if info["last_time"]:
+                # 格式化日期给前端展示用
+                info["formatted_last_time"] = time.strftime("%Y-%m-%d", time.localtime(info["last_time"]))
+            else:
+                info["formatted_last_time"] = ""
+            result.append(info)
+
+    # 4. 按最后一条消息的时间降序排列
+    result.sort(key=lambda x: x.get("last_time", 0), reverse=True)
+
+    return {"status": "success", "total": len(result), "data": result}
+
+
+# 接口 2：获取普通服务号/公众号的 json 消息 (已修复表名比对 bug)
+@router.get("/api/biz/messages", summary="获取指定服务号的消息")
+def get_biz_messages(username: str, account: Optional[str] = None, limit: int = 50, offset: int = 0):
+    if username == "gh_3dfda90e39d6":
+        raise HTTPException(status_code=400, detail="微信支付记录请请求 /api/biz/pay_records 接口")
+
+    account_dir = _resolve_account_dir(account)
+    md5_id = hashlib.md5(username.encode('utf-8')).hexdigest().lower()
+    table_name = f"Msg_{md5_id}"
+
+    target_db = None
+    for db_file in account_dir.glob("biz_message*.db"):
+        conn = sqlite3.connect(str(db_file))
+        try:
+            # 必须用 table_name.lower()，否则永远匹配不上
+            res = conn.execute("SELECT name FROM sqlite_master WHERE type='table' AND lower(name)=?",
+                               (table_name.lower(),)).fetchone()
+            if res:
+                target_db = db_file
+                break
+        except Exception:
+            pass
+        finally:
+            conn.close()
+
+    if not target_db:
+        return {"status": "success", "data": [], "message": f"未找到 {username} 的消息历史"}
+
+    # ... (后续数据库查询逻辑保持不变) ...
+    messages = []
+    try:
+        conn = sqlite3.connect(str(target_db))
+        cursor = conn.cursor()
+
+        query = f"""
+            SELECT local_id, create_time, message_content 
+            FROM [{table_name}] 
+            WHERE local_type != 1 
+            ORDER BY create_time DESC 
+            LIMIT ? OFFSET ?
+        """
+        rows = cursor.execute(query, (limit, offset)).fetchall()
+
+        for local_id, c_time, content in rows:
+            raw_xml = extract_xml_from_db_content(content, username, local_id)
+            if not raw_xml:
+                continue
+
+            struct_data = parse_wechat_xml_to_struct(raw_xml, username, local_id)
+            if struct_data:
+                struct_data["local_id"] = local_id
+                struct_data["create_time"] = c_time
+                messages.append(struct_data)
+
+        conn.close()
+    except Exception as e:
+        logger.error(f"[biz] 数据库查询出错: {e}")
+        return {"status": "error", "message": str(e)}
+
+    return {"status": "success", "data": messages}
+
+
+# 接口 3：返回微信支付的 json 消息 (已修复表名比对 bug)
+@router.get("/api/biz/pay_records", summary="获取微信支付记录")
+def get_wechat_pay_records(account: Optional[str] = None, limit: int = 50, offset: int = 0):
+    username = "gh_3dfda90e39d6"
+    account_dir = _resolve_account_dir(account)
+    md5_id = hashlib.md5(username.encode('utf-8')).hexdigest().lower()
+    table_name = f"Msg_{md5_id}"
+
+    target_db = None
+    for db_file in account_dir.glob("biz_message*.db"):
+        conn = sqlite3.connect(str(db_file))
+        try:
+            # 必须用 table_name.lower()
+            res = conn.execute("SELECT name FROM sqlite_master WHERE type='table' AND lower(name)=?",
+                               (table_name.lower(),)).fetchone()
+            if res:
+                target_db = db_file
+                break
+        except Exception:
+            pass
+        finally:
+            conn.close()
+
+    if not target_db:
+        return {"status": "success", "data": [], "message": "未找到微信支付的消息历史"}
+
+    messages = []
+    try:
+        conn = sqlite3.connect(str(target_db))
+        cursor = conn.cursor()
+
+        query = f"""
+            SELECT local_id, create_time, message_content 
+            FROM [{table_name}] 
+            WHERE local_type = 21474836529 OR local_type != 1 
+            ORDER BY create_time DESC 
+            LIMIT ? OFFSET ?
+        """
+        rows = cursor.execute(query, (limit, offset)).fetchall()
+
+        for local_id, c_time, content in rows:
+            raw_xml = extract_xml_from_db_content(content, username, local_id)
+            if not raw_xml:
+                continue
+
+            parsed_data = parse_pay_xml(raw_xml, local_id)
+            if parsed_data:
+                parsed_data["local_id"] = local_id
+                parsed_data["create_time"] = c_time
+                if not parsed_data["timestamp"]:
+                    parsed_data["timestamp"] = c_time
+
+                parsed_data["formatted_time"] = time.strftime(
+                    "%Y-%m-%d %H:%M:%S", time.localtime(parsed_data["timestamp"])
+                )
+                messages.append(parsed_data)
+
+        conn.close()
+    except Exception as e:
+        logger.error(f"[biz] 查询微信支付数据库出错: {e}")
+        return {"status": "error", "message": str(e)}
+
+    return {"status": "success", "data": messages}

src/wechat_decrypt_tool/routers/chat.py

@@ -1391,6 +1391,299 @@ def _load_contact_top_flags(contact_db_path: Path, usernames: list[str]) -> dict
         conn.close()
 
 
+def _coerce_realtime_blobish_value(value: Any) -> Any:
+    if value is None:
+        return None
+    if isinstance(value, memoryview):
+        value = value.tobytes()
+    if isinstance(value, bytearray):
+        return bytes(value)
+    if isinstance(value, bytes):
+        try:
+            s = value.decode("ascii").strip()
+        except Exception:
+            return value
+        if not s:
+            return value
+        b = _hex_to_bytes(s)
+        if b is not None:
+            return b
+        if (len(s) % 2 == 0) and (_HEX_RE.fullmatch(s) is not None):
+            try:
+                return bytes.fromhex(s)
+            except Exception:
+                return value
+        return value
+    if isinstance(value, str):
+        s = value.strip()
+        if not s:
+            return value
+        b = _hex_to_bytes(s)
+        if b is not None:
+            return b
+        if (len(s) % 2 == 0) and (_HEX_RE.fullmatch(s) is not None):
+            try:
+                return bytes.fromhex(s)
+            except Exception:
+                return value
+        return value
+    return value
+
+
+def _normalize_realtime_message_item(item: dict[str, Any]) -> dict[str, Any]:
+    def _pick(*keys: str) -> Any:
+        return _pick_case_insensitive_value(item, *keys)
+
+    message_content = _coerce_realtime_blobish_value(
+        _pick("message_content", "messageContent", "MessageContent")
+    )
+    if message_content is None:
+        message_content = ""
+
+    return {
+        "local_id": int(_pick("local_id", "localId") or 0),
+        "server_id": int(_pick("server_id", "serverId", "MsgSvrID") or 0),
+        "local_type": int(_pick("local_type", "localType", "Type", "type") or 0),
+        "sort_seq": int(_pick("sort_seq", "sortSeq", "SortSeq") or 0),
+        "real_sender_id": int(_pick("real_sender_id", "realSenderId") or 0),
+        "create_time": int(_pick("create_time", "createTime", "CreateTime") or 0),
+        "message_content": message_content,
+        "compress_content": _coerce_realtime_blobish_value(
+            _pick("compress_content", "compressContent", "CompressContent")
+        ),
+        "packed_info_data": _coerce_realtime_blobish_value(
+            _pick("packed_info_data", "packedInfoData", "PackedInfoData")
+        ),
+        "sender_username": str(
+            _pick("sender_username", "senderUsername", "sender", "SenderUsername") or ""
+        ).strip(),
+    }
+
+
+def _collect_realtime_rows_for_session(
+    *,
+    trace_id: Optional[str],
+    account_name: str,
+    rt_conn: Any,
+    username: str,
+    msg_db_path_real: Path,
+    table_name: str,
+    max_local_id: int,
+    max_scan: int,
+    backfill_limit: int,
+) -> dict[str, Any]:
+    label = f"[{trace_id}]" if trace_id else "[realtime]"
+    log_fn = logger.info if trace_id else logger.debug
+    uname = str(username or "").strip()
+    use_biz_exec_query = uname.startswith("gh_") and ("biz_message" in str(msg_db_path_real.name).lower())
+
+    if use_biz_exec_query:
+        try:
+            quoted_table = _quote_ident(table_name)
+            select_cols = (
+                "local_id",
+                "server_id",
+                "local_type",
+                "sort_seq",
+                "real_sender_id",
+                "create_time",
+                "message_content",
+                "compress_content",
+                "packed_info_data",
+            )
+            select_sql = ", ".join([_quote_ident(col) for col in select_cols])
+
+            if int(max_local_id) > 0:
+                sql_new = (
+                    f"SELECT {select_sql} FROM {quoted_table} "
+                    f"WHERE local_id > {int(max_local_id)} "
+                    f"ORDER BY local_id ASC LIMIT {int(max_scan)}"
+                )
+            else:
+                sql_new = f"SELECT {select_sql} FROM {quoted_table} ORDER BY local_id DESC LIMIT {int(max_scan)}"
+
+            log_fn(
+                "%s wcdb_exec_query biz account=%s username=%s mode=new_rows max_local_id=%s limit=%s",
+                label,
+                account_name,
+                uname,
+                int(max_local_id),
+                int(max_scan),
+            )
+            wcdb_t0 = time.perf_counter()
+            with rt_conn.lock:
+                raw_new_rows = _wcdb_exec_query(rt_conn.handle, kind="message", path=str(msg_db_path_real), sql=sql_new)
+            wcdb_ms = (time.perf_counter() - wcdb_t0) * 1000.0
+            logger.info(
+                "%s wcdb_exec_query biz done account=%s username=%s mode=new_rows rows=%s ms=%.1f",
+                label,
+                account_name,
+                uname,
+                len(raw_new_rows or []),
+                wcdb_ms,
+            )
+            if wcdb_ms > 2000:
+                logger.warning(
+                    "%s wcdb_exec_query biz slow account=%s username=%s mode=new_rows ms=%.1f",
+                    label,
+                    account_name,
+                    uname,
+                    wcdb_ms,
+                )
+
+            normalized_new_rows: list[dict[str, Any]] = []
+            for item in raw_new_rows or []:
+                if not isinstance(item, dict):
+                    continue
+                norm = _normalize_realtime_message_item(item)
+                if int(norm.get("local_id") or 0) <= 0:
+                    continue
+                normalized_new_rows.append(norm)
+
+            if int(max_local_id) > 0:
+                new_rows = list(reversed(normalized_new_rows))
+            else:
+                new_rows = normalized_new_rows
+
+            backfill_rows: list[dict[str, Any]] = []
+            scanned = len(raw_new_rows or [])
+            if int(backfill_limit) > 0 and int(max_local_id) > 0:
+                sql_backfill = (
+                    f"SELECT {select_sql} FROM {quoted_table} "
+                    f"WHERE local_id <= {int(max_local_id)} "
+                    f"ORDER BY local_id DESC LIMIT {int(backfill_limit)}"
+                )
+                log_fn(
+                    "%s wcdb_exec_query biz account=%s username=%s mode=backfill limit=%s",
+                    label,
+                    account_name,
+                    uname,
+                    int(backfill_limit),
+                )
+                backfill_t0 = time.perf_counter()
+                with rt_conn.lock:
+                    raw_backfill_rows = _wcdb_exec_query(
+                        rt_conn.handle,
+                        kind="message",
+                        path=str(msg_db_path_real),
+                        sql=sql_backfill,
+                    )
+                backfill_ms = (time.perf_counter() - backfill_t0) * 1000.0
+                logger.info(
+                    "%s wcdb_exec_query biz done account=%s username=%s mode=backfill rows=%s ms=%.1f",
+                    label,
+                    account_name,
+                    uname,
+                    len(raw_backfill_rows or []),
+                    backfill_ms,
+                )
+                if backfill_ms > 2000:
+                    logger.warning(
+                        "%s wcdb_exec_query biz slow account=%s username=%s mode=backfill ms=%.1f",
+                        label,
+                        account_name,
+                        uname,
+                        backfill_ms,
+                    )
+                scanned += len(raw_backfill_rows or [])
+                for item in raw_backfill_rows or []:
+                    if not isinstance(item, dict):
+                        continue
+                    norm = _normalize_realtime_message_item(item)
+                    if int(norm.get("local_id") or 0) <= 0:
+                        continue
+                    backfill_rows.append(norm)
+
+            return {
+                "fetchMode": "biz_exec_query",
+                "scanned": int(scanned),
+                "new_rows": new_rows,
+                "backfill_rows": backfill_rows,
+            }
+        except Exception as e:
+            logger.warning(
+                "%s wcdb_exec_query biz failed account=%s username=%s err=%s fallback=wcdb_get_messages",
+                label,
+                account_name,
+                uname,
+                str(e),
+            )
+
+    batch_size = 200
+    scanned = 0
+    offset = 0
+    new_rows: list[dict[str, Any]] = []
+    backfill_rows: list[dict[str, Any]] = []
+    reached_existing = False
+    stop = False
+
+    while scanned < int(max_scan):
+        take = min(batch_size, int(max_scan) - scanned)
+        log_fn(
+            "%s wcdb_get_messages account=%s username=%s take=%s offset=%s",
+            label,
+            account_name,
+            uname,
+            int(take),
+            int(offset),
+        )
+        wcdb_t0 = time.perf_counter()
+        with rt_conn.lock:
+            raw_rows = _wcdb_get_messages(rt_conn.handle, uname, limit=take, offset=offset)
+        wcdb_ms = (time.perf_counter() - wcdb_t0) * 1000.0
+        log_fn(
+            "%s wcdb_get_messages done account=%s username=%s rows=%s ms=%.1f",
+            label,
+            account_name,
+            uname,
+            len(raw_rows or []),
+            wcdb_ms,
+        )
+        if wcdb_ms > 2000:
+            logger.warning(
+                "%s wcdb_get_messages slow account=%s username=%s ms=%.1f",
+                label,
+                account_name,
+                uname,
+                wcdb_ms,
+            )
+        if not raw_rows:
+            break
+
+        scanned += len(raw_rows)
+        offset += len(raw_rows)
+
+        for item in raw_rows:
+            if not isinstance(item, dict):
+                continue
+            norm = _normalize_realtime_message_item(item)
+            lid = int(norm.get("local_id") or 0)
+            if lid <= 0:
+                continue
+            if (not reached_existing) and lid > int(max_local_id):
+                new_rows.append(norm)
+                continue
+
+            reached_existing = True
+            if int(backfill_limit) <= 0:
+                stop = True
+                break
+            backfill_rows.append(norm)
+            if len(backfill_rows) >= int(backfill_limit):
+                stop = True
+                break
+
+        if stop or len(raw_rows) < take:
+            break
+
+    return {
+        "fetchMode": "wcdb_get_messages",
+        "scanned": int(scanned),
+        "new_rows": new_rows,
+        "backfill_rows": backfill_rows,
+    }
+
+
 @router.post("/api/chat/realtime/sync", summary="实时消息同步到解密库（按会话增量）")
 def sync_chat_realtime_messages(
     request: Request,
@@ -1511,127 +1804,30 @@ def sync_chat_realtime_messages(
 
             placeholders = ",".join(["?"] * len(insert_cols))
             insert_sql = f"INSERT OR IGNORE INTO {quoted_table} ({','.join(insert_cols)}) VALUES ({placeholders})"
-
-            def pick(item: dict[str, Any], *keys: str) -> Any:
-                for k in keys:
-                    if k in item and item[k] is not None:
-                        return item[k]
-                    lk = k.lower()
-                    for kk in item.keys():
-                        if str(kk).lower() == lk and item[kk] is not None:
-                            return item[kk]
-                return None
-
-            def normalize_blob(value: Any) -> Optional[bytes]:
-                if value is None:
-                    return None
-                if isinstance(value, memoryview):
-                    return value.tobytes()
-                if isinstance(value, (bytes, bytearray)):
-                    return bytes(value)
-                if isinstance(value, str):
-                    s = value.strip()
-                    if s.lower().startswith("0x"):
-                        s = s[2:]
-                    if s and re.fullmatch(r"[0-9a-fA-F]+", s) and (len(s) % 2 == 0):
-                        try:
-                            return bytes.fromhex(s)
-                        except Exception:
-                            return None
-                    return s.encode("utf-8", errors="ignore")
-                return None
-
-            def normalize(item: dict[str, Any]) -> dict[str, Any]:
-                return {
-                    "local_id": int(pick(item, "local_id", "localId") or 0),
-                    "server_id": int(pick(item, "server_id", "serverId", "MsgSvrID") or 0),
-                    "local_type": int(pick(item, "local_type", "localType", "Type", "type") or 0),
-                    "sort_seq": int(pick(item, "sort_seq", "sortSeq", "SortSeq") or 0),
-                    "real_sender_id": int(pick(item, "real_sender_id", "realSenderId") or 0),
-                    "create_time": int(pick(item, "create_time", "createTime", "CreateTime") or 0),
-                    "message_content": pick(item, "message_content", "messageContent", "MessageContent") or "",
-                    "compress_content": pick(item, "compress_content", "compressContent", "CompressContent"),
-                    "packed_info_data": normalize_blob(pick(item, "packed_info_data", "packedInfoData")),
-                    "sender_username": str(
-                        pick(item, "sender_username", "senderUsername", "sender", "SenderUsername") or ""
-                    ).strip(),
-                }
-
-            batch_size = 200
-            scanned = 0
-            offset = 0
-            new_rows: list[dict[str, Any]] = []
-            backfill_rows: list[dict[str, Any]] = []
-            reached_existing = False
-            stop = False
-
-            while scanned < int(max_scan):
-                take = min(batch_size, int(max_scan) - scanned)
-                logger.info(
-                    "[%s] wcdb_get_messages account=%s username=%s take=%s offset=%s",
-                    trace_id,
-                    account_dir.name,
-                    username,
-                    int(take),
-                    int(offset),
-                )
-                wcdb_t0 = time.perf_counter()
-                with rt_conn.lock:
-                    raw_rows = _wcdb_get_messages(rt_conn.handle, username, limit=take, offset=offset)
-                wcdb_ms = (time.perf_counter() - wcdb_t0) * 1000.0
-                logger.info(
-                    "[%s] wcdb_get_messages done account=%s username=%s rows=%s ms=%.1f",
-                    trace_id,
-                    account_dir.name,
-                    username,
-                    len(raw_rows or []),
-                    wcdb_ms,
-                )
-                if wcdb_ms > 2000:
-                    logger.warning(
-                        "[%s] wcdb_get_messages slow account=%s username=%s ms=%.1f",
-                        trace_id,
-                        account_dir.name,
-                        username,
-                        wcdb_ms,
-                    )
-                if not raw_rows:
-                    break
-
-                scanned += len(raw_rows)
-                offset += len(raw_rows)
-
-                for item in raw_rows:
-                    if not isinstance(item, dict):
-                        continue
-                    norm = normalize(item)
-                    lid = int(norm.get("local_id") or 0)
-                    if lid <= 0:
-                        continue
-                    if (not reached_existing) and lid > max_local_id:
-                        new_rows.append(norm)
-                        continue
-
-                    reached_existing = True
-                    if int(backfill_limit) <= 0:
-                        stop = True
-                        break
-                    backfill_rows.append(norm)
-                    if len(backfill_rows) >= int(backfill_limit):
-                        stop = True
-                        break
-
-                if stop or len(raw_rows) < take:
-                    break
+            fetch_result = _collect_realtime_rows_for_session(
+                trace_id=trace_id,
+                account_name=account_dir.name,
+                rt_conn=rt_conn,
+                username=username,
+                msg_db_path_real=msg_db_path_real,
+                table_name=table_name,
+                max_local_id=max_local_id,
+                max_scan=int(max_scan),
+                backfill_limit=int(backfill_limit),
+            )
+            scanned = int(fetch_result.get("scanned") or 0)
+            new_rows = list(fetch_result.get("new_rows") or [])
+            backfill_rows = list(fetch_result.get("backfill_rows") or [])
 
             inserted = 0
             backfilled = 0
-            if new_rows and (not name2id_synced):
-                _best_effort_upsert_output_name2id_rows(
-                    msg_conn,
-                    account_name=account_dir.name,
-                    rows=new_rows,
-                )
+            if new_rows:
+                if not name2id_synced:
+                    _best_effort_upsert_output_name2id_rows(
+                        msg_conn,
+                        account_name=account_dir.name,
+                        rows=new_rows,
+                    )
 
                 # Insert older -> newer to keep sqlite btree locality similar to existing data.
                 values = [tuple(r.get(c) for c in insert_cols) for r in reversed(new_rows)]
@@ -1879,124 +2075,30 @@ def _sync_chat_realtime_messages_for_table(
 
         placeholders = ",".join(["?"] * len(insert_cols))
         insert_sql = f"INSERT OR IGNORE INTO {quoted_table} ({','.join(insert_cols)}) VALUES ({placeholders})"
-
-        def pick(item: dict[str, Any], *keys: str) -> Any:
-            for k in keys:
-                if k in item and item[k] is not None:
-                    return item[k]
-                lk = k.lower()
-                for kk in item.keys():
-                    if str(kk).lower() == lk and item[kk] is not None:
-                        return item[kk]
-            return None
-
-        def normalize_blob(value: Any) -> Optional[bytes]:
-            if value is None:
-                return None
-            if isinstance(value, memoryview):
-                return value.tobytes()
-            if isinstance(value, (bytes, bytearray)):
-                return bytes(value)
-            if isinstance(value, str):
-                s = value.strip()
-                if s.lower().startswith("0x"):
-                    s = s[2:]
-                if s and re.fullmatch(r"[0-9a-fA-F]+", s) and (len(s) % 2 == 0):
-                    try:
-                        return bytes.fromhex(s)
-                    except Exception:
-                        return None
-                return s.encode("utf-8", errors="ignore")
-            return None
-
-        def normalize(item: dict[str, Any]) -> dict[str, Any]:
-            return {
-                "local_id": int(pick(item, "local_id", "localId") or 0),
-                "server_id": int(pick(item, "server_id", "serverId", "MsgSvrID") or 0),
-                "local_type": int(pick(item, "local_type", "localType", "Type", "type") or 0),
-                "sort_seq": int(pick(item, "sort_seq", "sortSeq", "SortSeq") or 0),
-                "real_sender_id": int(pick(item, "real_sender_id", "realSenderId") or 0),
-                "create_time": int(pick(item, "create_time", "createTime", "CreateTime") or 0),
-                "message_content": pick(item, "message_content", "messageContent", "MessageContent") or "",
-                "compress_content": pick(item, "compress_content", "compressContent", "CompressContent"),
-                "packed_info_data": normalize_blob(pick(item, "packed_info_data", "packedInfoData")),
-                "sender_username": str(
-                    pick(item, "sender_username", "senderUsername", "sender", "SenderUsername") or ""
-                ).strip(),
-            }
-
-        batch_size = 200
-        scanned = 0
-        offset = 0
-        new_rows: list[dict[str, Any]] = []
-        backfill_rows: list[dict[str, Any]] = []
-        reached_existing = False
-        stop = False
-
-        while scanned < int(max_scan):
-            take = min(batch_size, int(max_scan) - scanned)
-            logger.debug(
-                "[realtime] wcdb_get_messages account=%s username=%s take=%s offset=%s",
-                account_dir.name,
-                username,
-                int(take),
-                int(offset),
-            )
-            wcdb_t0 = time.perf_counter()
-            with rt_conn.lock:
-                raw_rows = _wcdb_get_messages(rt_conn.handle, username, limit=take, offset=offset)
-            wcdb_ms = (time.perf_counter() - wcdb_t0) * 1000.0
-            logger.debug(
-                "[realtime] wcdb_get_messages done account=%s username=%s rows=%s ms=%.1f",
-                account_dir.name,
-                username,
-                len(raw_rows or []),
-                wcdb_ms,
-            )
-            if wcdb_ms > 2000:
-                logger.warning(
-                    "[realtime] wcdb_get_messages slow account=%s username=%s ms=%.1f",
-                    account_dir.name,
-                    username,
-                    wcdb_ms,
-                )
-            if not raw_rows:
-                break
-
-            scanned += len(raw_rows)
-            offset += len(raw_rows)
-
-            for item in raw_rows:
-                if not isinstance(item, dict):
-                    continue
-                norm = normalize(item)
-                lid = int(norm.get("local_id") or 0)
-                if lid <= 0:
-                    continue
-                if (not reached_existing) and lid > max_local_id:
-                    new_rows.append(norm)
-                    continue
-
-                reached_existing = True
-                if int(backfill_limit) <= 0:
-                    stop = True
-                    break
-                backfill_rows.append(norm)
-                if len(backfill_rows) >= int(backfill_limit):
-                    stop = True
-                    break
-
-            if stop or len(raw_rows) < take:
-                break
+        fetch_result = _collect_realtime_rows_for_session(
+            trace_id=None,
+            account_name=account_dir.name,
+            rt_conn=rt_conn,
+            username=username,
+            msg_db_path_real=msg_db_path_real,
+            table_name=table_name,
+            max_local_id=max_local_id,
+            max_scan=int(max_scan),
+            backfill_limit=int(backfill_limit),
+        )
+        scanned = int(fetch_result.get("scanned") or 0)
+        new_rows = list(fetch_result.get("new_rows") or [])
+        backfill_rows = list(fetch_result.get("backfill_rows") or [])
 
         inserted = 0
         backfilled = 0
-        if new_rows and (not name2id_synced):
-            _best_effort_upsert_output_name2id_rows(
-                msg_conn,
-                account_name=account_dir.name,
-                rows=new_rows,
-            )
+        if new_rows:
+            if not name2id_synced:
+                _best_effort_upsert_output_name2id_rows(
+                    msg_conn,
+                    account_name=account_dir.name,
+                    rows=new_rows,
+                )
 
             values = [tuple(r.get(c) for c in insert_cols) for r in reversed(new_rows)]
             insert_t0 = time.perf_counter()
@@ -2161,6 +2263,7 @@ def sync_chat_realtime_messages_all(
     priority_max_scan: int = 600,
     include_hidden: bool = True,
     include_official: bool = True,
+    only_official: bool = False,
     backfill_limit: int = 200,
 ):
     """
@@ -2171,13 +2274,14 @@ def sync_chat_realtime_messages_all(
     account_dir = _resolve_account_dir(account)
     trace_id = f"rt-syncall-{int(time.time() * 1000)}-{threading.get_ident()}"
     logger.info(
-        "[%s] realtime sync_all start account=%s max_scan=%s priority=%s include_hidden=%s include_official=%s",
+        "[%s] realtime sync_all start account=%s max_scan=%s priority=%s include_hidden=%s include_official=%s only_official=%s",
         trace_id,
         account_dir.name,
         int(max_scan),
         str(priority_username or "").strip(),
         bool(include_hidden),
         bool(include_official),
+        bool(only_official),
     )
 
     if max_scan < 20:
@@ -2239,6 +2343,8 @@ def sync_chat_realtime_messages_all(
                 hidden_val = 0
             if not include_hidden and hidden_val == 1:
                 continue
+            if only_official and not uname.startswith("gh_"):
+                continue
             if not _should_keep_session(uname, include_official=include_official):
                 continue
 

src/wechat_decrypt_tool/routers/decrypt.py

@@ -14,12 +14,7 @@ from ..app_paths import get_output_databases_dir
 from ..logging_config import get_logger
 from ..path_fix import PathFixRoute
 from ..key_store import upsert_account_keys_in_store
-from ..wechat_decrypt import (
-    WeChatDatabaseDecryptor,
-    build_decrypt_result_message,
-    decrypt_wechat_databases,
-    scan_account_databases_from_path,
-)
+from ..wechat_decrypt import WeChatDatabaseDecryptor, decrypt_wechat_databases, scan_account_databases_from_path
 
 logger = get_logger(__name__)
 
@@ -81,7 +76,6 @@ async def decrypt_databases(request: DecryptRequest):
             "message": results["message"],
             "processed_files": results["processed_files"],
             "failed_files": results["failed_files"],
-            "failure_details": results.get("failure_details", []),
             "account_results": results.get("account_results", {}),
         }
 
@@ -165,7 +159,6 @@ async def decrypt_databases_stream(
         fail_count = 0
         processed_files: list[str] = []
         failed_files: list[str] = []
-        failure_details: list[dict] = []
         account_results: dict = {}
         overall_current = 0
 
@@ -188,7 +181,6 @@ async def decrypt_databases_stream(
             account_success = 0
             account_processed: list[str] = []
             account_failed: list[str] = []
-            account_failure_details: list[dict] = []
 
             for db_info in dbs:
                 if await request.is_disconnected():
@@ -240,20 +232,11 @@ async def decrypt_databases_stream(
                     status = "success"
                     msg = "解密成功"
                 else:
-                    failure_detail = {
-                        "account": account,
-                        "file": db_path,
-                        "name": db_name,
-                        "code": str(decryptor.last_error_code or "").strip(),
-                        "reason": str(decryptor.last_error_message or "").strip() or "解密失败",
-                    }
                     account_failed.append(db_path)
-                    account_failure_details.append(failure_detail)
                     failed_files.append(db_path)
-                    failure_details.append(failure_detail)
                     fail_count += 1
                     status = "fail"
-                    msg = failure_detail["reason"]
+                    msg = "解密失败"
 
                 yield _sse(
                     {
@@ -278,7 +261,6 @@ async def decrypt_databases_stream(
                 "output_dir": str(account_output_dir),
                 "processed_files": account_processed,
                 "failed_files": account_failed,
-                "failure_details": account_failure_details,
             }
 
             # Build cache table (keep behavior consistent with the POST endpoint).
@@ -325,15 +307,9 @@ async def decrypt_databases_stream(
             "success_count": success_count,
             "failure_count": total_databases - success_count,
             "output_directory": str(base_output_dir.absolute()),
-            "message": build_decrypt_result_message(
-                total_databases=total_databases,
-                success_count=success_count,
-                failed_count=total_databases - success_count,
-                failure_details=failure_details,
-            ),
+            "message": f"解密完成: 成功 {success_count}/{total_databases}",
             "processed_files": processed_files,
             "failed_files": failed_files,
-            "failure_details": failure_details,
             "account_results": account_results,
         }
 

src/wechat_decrypt_tool/wcdb_realtime.py

@@ -26,6 +26,32 @@ _DEFAULT_WCDB_API_DLL = _NATIVE_DIR / "wcdb_api.dll"
 _WCDB_API_DLL_SELECTED: Optional[Path] = None
 
 
+def _iter_runtime_wcdb_api_dll_paths() -> tuple[Path, ...]:
+    candidates: list[Path] = []
+    seen: set[str] = set()
+
+    def add_anchor(anchor: str | Path | None) -> None:
+        if not anchor:
+            return
+        try:
+            base = Path(anchor).resolve()
+        except Exception:
+            base = Path(anchor)
+        candidate = base / "native" / "wcdb_api.dll"
+        key = str(candidate).replace("/", "\\").rstrip("\\").lower()
+        if key in seen:
+            return
+        seen.add(key)
+        candidates.append(candidate)
+
+    add_anchor(os.environ.get("WECHAT_TOOL_DATA_DIR", "").strip())
+    add_anchor(Path.cwd())
+    if getattr(sys, "frozen", False):
+        add_anchor(Path(sys.executable).resolve().parent)
+
+    return tuple(candidates)
+
+
 def _is_project_wcdb_api_dll_path(path: Path) -> bool:
     try:
         resolved = path.resolve(strict=False)
@@ -40,6 +66,14 @@ def _is_project_wcdb_api_dll_path(path: Path) -> bool:
     if resolved == default_resolved:
         return True
 
+    for candidate in _iter_runtime_wcdb_api_dll_paths():
+        try:
+            if resolved == candidate.resolve(strict=False):
+                return True
+        except Exception:
+            if resolved == candidate:
+                return True
+
     parts = tuple(str(part).lower() for part in resolved.parts)
     allowed_suffixes = (
         ("backend", "native", "wcdb_api.dll"),

src/wechat_decrypt_tool/wechat_decrypt.py

@@ -13,12 +13,12 @@ import hashlib
 import hmac
 import os
 import json
-import shutil
-import tempfile
 from pathlib import Path
 
 from cryptography.hazmat.backends import default_backend
+from cryptography.hazmat.primitives import hashes
 from cryptography.hazmat.primitives.ciphers import Cipher, algorithms, modes
+from cryptography.hazmat.primitives.kdf.pbkdf2 import PBKDF2HMAC
 
 from .app_paths import get_output_databases_dir
 
@@ -26,117 +26,6 @@ from .app_paths import get_output_databases_dir
 
 # SQLite文件头
 SQLITE_HEADER = b"SQLite format 3\x00"
-PAGE_SIZE = 4096
-KEY_SIZE = 32
-SALT_SIZE = 16
-IV_SIZE = 16
-HMAC_SIZE = 64
-RESERVE_SIZE = 80
-KEY_MISMATCH_GUIDANCE = (
-    "请在当前设备登录该账号后重新获取密钥；"
-    "如果聊天记录是从另一台设备复制过来的，当前设备通常无法获取原设备对应的密钥。"
-)
-
-
-def _derive_mac_key(raw_key: bytes, salt: bytes) -> bytes:
-    mac_salt = bytes(b ^ 0x3A for b in salt)
-    return hashlib.pbkdf2_hmac("sha512", raw_key, mac_salt, 2, dklen=KEY_SIZE)
-
-
-def _derive_sqlcipher_enc_key(key_material: bytes, salt: bytes) -> bytes:
-    return hashlib.pbkdf2_hmac("sha512", key_material, salt, 256000, dklen=KEY_SIZE)
-
-
-def _resolve_page1_key_material(key_material: bytes, page1: bytes) -> tuple[bytes, bytes, str] | None:
-    salt = page1[:SALT_SIZE]
-    stored_page1_hmac = page1[PAGE_SIZE - HMAC_SIZE : PAGE_SIZE]
-
-    candidates = [
-        ("raw_enc_key", key_material, _derive_mac_key(key_material, salt)),
-    ]
-
-    derived_key = _derive_sqlcipher_enc_key(key_material, salt)
-    candidates.append(("sqlcipher_passphrase", derived_key, _derive_mac_key(derived_key, salt)))
-
-    for mode, enc_key, mac_key in candidates:
-        expected_page1_hmac = _compute_page_hmac(mac_key, page1, 1)
-        if stored_page1_hmac == expected_page1_hmac:
-            return enc_key, mac_key, mode
-
-    return None
-
-
-def _compute_page_hmac(mac_key: bytes, page: bytes, page_num: int) -> bytes:
-    offset = SALT_SIZE if page_num == 1 else 0
-    data_end = PAGE_SIZE - RESERVE_SIZE + IV_SIZE
-    mac = hmac.new(mac_key, digestmod=hashlib.sha512)
-    mac.update(page[offset:data_end])
-    mac.update(page_num.to_bytes(4, "little"))
-    return mac.digest()
-
-
-def _decrypt_page(raw_key: bytes, page: bytes, page_num: int) -> bytes:
-    iv = page[PAGE_SIZE - RESERVE_SIZE : PAGE_SIZE - RESERVE_SIZE + IV_SIZE]
-    offset = SALT_SIZE if page_num == 1 else 0
-    encrypted = page[offset : PAGE_SIZE - RESERVE_SIZE]
-
-    cipher = Cipher(
-        algorithms.AES(raw_key),
-        modes.CBC(iv),
-        backend=default_backend(),
-    )
-    decryptor = cipher.decryptor()
-    decrypted = decryptor.update(encrypted) + decryptor.finalize()
-
-    if page_num == 1:
-        return SQLITE_HEADER + decrypted + (b"\x00" * RESERVE_SIZE)
-    return decrypted + (b"\x00" * RESERVE_SIZE)
-
-
-def _failure_matches_key_mismatch(detail: dict | None) -> bool:
-    if not isinstance(detail, dict):
-        return False
-    code = str(detail.get("code") or "").strip().lower()
-    reason = str(detail.get("reason") or "").strip()
-    if code == "key_mismatch":
-        return True
-    return ("密钥" in reason and "不匹配" in reason) or ("当前数据库密钥不正确" in reason)
-
-
-def build_decrypt_result_message(
-    total_databases: int,
-    success_count: int,
-    failed_count: int,
-    failure_details: list[dict] | None = None,
-) -> str:
-    total = max(int(total_databases or 0), 0)
-    success = max(int(success_count or 0), 0)
-    failed = max(int(failed_count or 0), 0)
-    details = list(failure_details or [])
-
-    if total == 0:
-        return "未找到可解密的数据库文件"
-
-    if failed == 0:
-        return f"解密完成: 成功 {success}/{total}"
-
-    key_mismatch_count = sum(1 for item in details if _failure_matches_key_mismatch(item))
-
-    if success == 0 and failed == total:
-        if key_mismatch_count == failed:
-            return (
-                f"解密失败：当前数据库密钥不正确，或该密钥不属于当前账号/当前设备（0/{total} 成功）。"
-                + KEY_MISMATCH_GUIDANCE
-            )
-        return f"解密失败：0/{total} 个数据库解密成功，请检查密钥、账号与数据库路径是否匹配。"
-
-    if key_mismatch_count > 0:
-        return (
-            f"解密完成：成功 {success}/{total}，失败 {failed}/{total}。"
-            "失败文件中包含密钥不匹配的数据库，请确认使用的是当前账号在当前设备上的密钥。"
-        )
-
-    return f"解密完成：成功 {success}/{total}，失败 {failed}/{total}。"
 
 
 def _normalize_account_name(name: str) -> str:
@@ -332,123 +221,153 @@ class WeChatDatabaseDecryptor:
             self.key_bytes = bytes.fromhex(key_hex)
         except ValueError:
             raise ValueError("密钥必须是有效的十六进制字符串")
-        self.last_error_code = ""
-        self.last_error_message = ""
-
-    def _set_last_error(self, code: str, message: str) -> None:
-        self.last_error_code = str(code or "").strip()
-        self.last_error_message = str(message or "").strip()
-
-    def _clear_last_error(self) -> None:
-        self.last_error_code = ""
-        self.last_error_message = ""
     
     def decrypt_database(self, db_path: str, output_path: str) -> bool:
         """解密微信4.x版本数据库
 
-        兼容两种输入形态：
-        - raw enc_key（部分内存扫描/工具直接返回）
-        - SQLCipher 口令/基础 key（需先用数据库 salt 做一轮 PBKDF2）
+        使用SQLCipher 4.0参数:
+        - PBKDF2-SHA512, 256000轮迭代
+        - AES-256-CBC加密
+        - HMAC-SHA512验证
+        - 页面大小4096字节
         """
         from .logging_config import get_logger
         logger = get_logger(__name__)
 
         logger.info(f"开始解密数据库: {db_path}")
-
-        tmp_output_path = ""
-        self._clear_last_error()
+        
         try:
-            file_size = os.path.getsize(db_path)
-            logger.info(f"读取文件大小: {file_size} bytes")
+            with open(db_path, 'rb') as f:
+                encrypted_data = f.read()
+            
+            logger.info(f"读取文件大小: {len(encrypted_data)} bytes")
 
-            if file_size < PAGE_SIZE:
-                message = f"数据库文件过小，无法解密: {db_path}"
-                self._set_last_error("file_too_small", message)
-                logger.warning(message)
-                return False
-
-            output_dir = Path(output_path).parent
-            output_dir.mkdir(parents=True, exist_ok=True)
-
-            with open(db_path, "rb") as source:
-                page1 = source.read(PAGE_SIZE)
-
-            if len(page1) < PAGE_SIZE:
-                message = f"数据库首页大小不足，无法解密: {db_path}"
-                self._set_last_error("page_too_small", message)
-                logger.warning(message)
+            if len(encrypted_data) < 4096:
+                logger.warning(f"文件太小，跳过解密: {db_path}")
                 return False
 
             # 检查是否已经是解密的数据库
-            if page1.startswith(SQLITE_HEADER):
+            if encrypted_data.startswith(SQLITE_HEADER):
                 logger.info(f"文件已是SQLite格式，直接复制: {db_path}")
-                fd, tmp_output_path = tempfile.mkstemp(
-                    prefix=f".{Path(output_path).name}.",
-                    suffix=".tmp",
-                    dir=str(output_dir),
-                )
-                os.close(fd)
-                with open(db_path, "rb") as src, open(tmp_output_path, "wb") as dst:
-                    shutil.copyfileobj(src, dst, length=1024 * 1024)
-                os.replace(tmp_output_path, output_path)
-                tmp_output_path = ""
+                with open(output_path, 'wb') as f:
+                    f.write(encrypted_data)
                 return True
-
-            resolved_key_material = _resolve_page1_key_material(self.key_bytes, page1)
-            if resolved_key_material is None:
-                message = f"当前数据库密钥不正确，或该密钥不属于当前账号/当前设备: {db_path}"
-                self._set_last_error("key_mismatch", message)
-                logger.error(f"页面 1 HMAC验证失败，密钥与数据库不匹配: {db_path}")
-                return False
-            enc_key, mac_key, key_mode = resolved_key_material
-            logger.info(f"页面 1 HMAC验证通过: mode={key_mode} path={db_path}")
-
-            total_pages = (file_size + PAGE_SIZE - 1) // PAGE_SIZE
-            successful_pages = 0
-            fd, tmp_output_path = tempfile.mkstemp(
-                prefix=f".{Path(output_path).name}.",
-                suffix=".tmp",
-                dir=str(output_dir),
+            
+            # 提取salt (前16字节)
+            salt = encrypted_data[:16]
+            
+            # 计算mac_salt (salt XOR 0x3a)
+            mac_salt = bytes(b ^ 0x3a for b in salt)
+            
+            # 使用PBKDF2-SHA512派生密钥
+            kdf = PBKDF2HMAC(
+                algorithm=hashes.SHA512(),
+                length=32,
+                salt=salt,
+                iterations=256000,
+                backend=default_backend()
             )
-            os.close(fd)
-
-            with open(db_path, "rb") as source, open(tmp_output_path, "wb") as target:
-                for page_num in range(1, total_pages + 1):
-                    page = source.read(PAGE_SIZE)
-                    if not page:
-                        break
-                    if len(page) < PAGE_SIZE:
-                        logger.warning(f"页面 {page_num} 大小不足: {len(page)} bytes，自动补齐到 {PAGE_SIZE} bytes")
-                        page = page + (b"\x00" * (PAGE_SIZE - len(page)))
-
-                    stored_hmac = page[PAGE_SIZE - HMAC_SIZE : PAGE_SIZE]
-                    expected_hmac = _compute_page_hmac(mac_key, page, page_num)
-                    if stored_hmac != expected_hmac:
-                        message = f"数据库校验失败，文件可能损坏或密钥不匹配: {db_path}"
-                        self._set_last_error("page_hmac_mismatch", message)
-                        logger.error(f"页面 {page_num} HMAC验证失败，终止解密: {db_path}")
-                        return False
-
-                    target.write(_decrypt_page(enc_key, page, page_num))
+            derived_key = kdf.derive(self.key_bytes)
+            
+            # 派生MAC密钥
+            mac_kdf = PBKDF2HMAC(
+                algorithm=hashes.SHA512(),
+                length=32,
+                salt=mac_salt,
+                iterations=2,
+                backend=default_backend()
+            )
+            mac_key = mac_kdf.derive(derived_key)
+            
+            # 解密数据
+            decrypted_data = bytearray()
+            decrypted_data.extend(SQLITE_HEADER)
+            
+            page_size = 4096
+            iv_size = 16
+            hmac_size = 64  # SHA512的HMAC是64字节
+            
+            # 计算保留区域大小 (对齐到AES块大小)
+            reserve_size = iv_size + hmac_size
+            if reserve_size % 16 != 0:
+                reserve_size = ((reserve_size // 16) + 1) * 16
+            
+            total_pages = len(encrypted_data) // page_size
+            successful_pages = 0
+            failed_pages = 0
+            
+            # 逐页解密
+            for cur_page in range(total_pages):
+                start = cur_page * page_size
+                end = start + page_size
+                page = encrypted_data[start:end]
+                
+                page_num = cur_page + 1  # 页面编号从1开始
+                
+                if len(page) < page_size:
+                    logger.warning(f"页面 {page_num} 大小不足: {len(page)} bytes")
+                    break
+                
+                # 确定偏移量：第一页(cur_page == 0)需要跳过salt
+                offset = 16 if cur_page == 0 else 0  # SALT_SIZE = 16
+                
+                # 提取存储的HMAC
+                hmac_start = page_size - reserve_size + iv_size
+                hmac_end = hmac_start + hmac_size
+                stored_hmac = page[hmac_start:hmac_end]
+                
+                # 按照wechat-dump-rs的方式验证HMAC
+                data_end = page_size - reserve_size + iv_size
+                hmac_data = page[offset:data_end]
+                
+                # 分步计算HMAC：先更新数据，再更新页面编号
+                mac = hmac.new(mac_key, digestmod=hashlib.sha512)
+                mac.update(hmac_data)  # 包含加密数据+IV
+                mac.update(page_num.to_bytes(4, 'little'))  # 页面编号(小端序)
+                expected_hmac = mac.digest()
+                
+                if stored_hmac != expected_hmac:
+                    logger.warning(f"页面 {page_num} HMAC验证失败")
+                    failed_pages += 1
+                    continue
+                
+                # 提取IV和加密数据用于AES解密
+                iv = page[page_size - reserve_size:page_size - reserve_size + iv_size]
+                encrypted_page = page[offset:page_size - reserve_size]
+                
+                # AES-CBC解密
+                try:
+                    cipher = Cipher(
+                        algorithms.AES(derived_key),
+                        modes.CBC(iv),
+                        backend=default_backend()
+                    )
+                    decryptor = cipher.decryptor()
+                    decrypted_page = decryptor.update(encrypted_page) + decryptor.finalize()
+                    
+                    # 按照wechat-dump-rs的方式重组页面数据
+                    decrypted_data.extend(decrypted_page)
+                    decrypted_data.extend(page[page_size - reserve_size:])  # 保留区域
+                    
                     successful_pages += 1
+                
+                except Exception as e:
+                    logger.error(f"页面 {page_num} AES解密失败: {e}")
+                    failed_pages += 1
+                    continue
 
-            logger.info(f"解密完成: 成功 {successful_pages} 页, 失败 0 页")
-            os.replace(tmp_output_path, output_path)
-            tmp_output_path = ""
-            logger.info(f"解密文件大小: {os.path.getsize(output_path)} bytes")
-            self._clear_last_error()
+            logger.info(f"解密完成: 成功 {successful_pages} 页, 失败 {failed_pages} 页")
+
+            # 写入解密后的文件
+            with open(output_path, 'wb') as f:
+                f.write(decrypted_data)
+
+            logger.info(f"解密文件大小: {len(decrypted_data)} bytes")
             return True
 
         except Exception as e:
-            self._set_last_error("exception", f"解密过程中发生异常: {e}")
             logger.error(f"解密失败: {db_path}, 错误: {e}")
             return False
-        finally:
-            if tmp_output_path:
-                try:
-                    os.remove(tmp_output_path)
-                except OSError:
-                    pass
 
 def decrypt_wechat_databases(db_storage_path: str = None, key: str = None) -> dict:
     """
@@ -573,7 +492,6 @@ def decrypt_wechat_databases(db_storage_path: str = None, key: str = None) -> di
     success_count = 0
     processed_files = []
     failed_files = []
-    failure_details = []
     account_results = {}
 
     for account_name, databases in account_databases.items():
@@ -605,7 +523,6 @@ def decrypt_wechat_databases(db_storage_path: str = None, key: str = None) -> di
         account_success = 0
         account_processed = []
         account_failed = []
-        account_failure_details = []
 
         for db_info in databases:
             db_path = db_info['path']
@@ -625,16 +542,7 @@ def decrypt_wechat_databases(db_storage_path: str = None, key: str = None) -> di
             else:
                 account_failed.append(db_path)
                 failed_files.append(db_path)
-                failure_detail = {
-                    "account": account_name,
-                    "file": db_path,
-                    "name": db_name,
-                    "code": str(decryptor.last_error_code or "").strip(),
-                    "reason": str(decryptor.last_error_message or "").strip() or "解密失败",
-                }
-                account_failure_details.append(failure_detail)
-                failure_details.append(failure_detail)
-                logger.error(f"解密失败: {account_name}/{db_name} reason={failure_detail['reason']}")
+                logger.error(f"解密失败: {account_name}/{db_name}")
 
         # 记录账号解密结果
         account_results[account_name] = {
@@ -643,8 +551,7 @@ def decrypt_wechat_databases(db_storage_path: str = None, key: str = None) -> di
             "failed": len(databases) - account_success,
             "output_dir": str(account_output_dir),
             "processed_files": account_processed,
-            "failed_files": account_failed,
-            "failure_details": account_failure_details,
+            "failed_files": account_failed
         }
 
         # 构建“会话最后一条消息”缓存表：把耗时挪到解密阶段，后续会话列表直接查表
@@ -668,23 +575,15 @@ def decrypt_wechat_databases(db_storage_path: str = None, key: str = None) -> di
         logger.info(f"账号 {account_name} 解密完成: 成功 {account_success}/{len(databases)}")
 
     # 返回结果
-    failed_count = total_databases - success_count
-    message = build_decrypt_result_message(
-        total_databases=total_databases,
-        success_count=success_count,
-        failed_count=failed_count,
-        failure_details=failure_details,
-    )
     result = {
         "status": "success" if success_count > 0 else "error",
-        "message": message,
+        "message": f"解密完成: 成功 {success_count}/{total_databases}",
         "total_databases": total_databases,
         "successful_count": success_count,
-        "failed_count": failed_count,
+        "failed_count": total_databases - success_count,
         "output_directory": str(base_output_dir.absolute()),
         "processed_files": processed_files,
         "failed_files": failed_files,
-        "failure_details": failure_details,
         "account_results": account_results,  # 新增：按账号的详细结果
         "detected_accounts": detected_accounts,
     }
@@ -692,9 +591,8 @@ def decrypt_wechat_databases(db_storage_path: str = None, key: str = None) -> di
     logger.info("=" * 60)
     logger.info("解密任务完成!")
     logger.info(f"成功: {success_count}/{total_databases}")
-    logger.info(f"失败: {failed_count}/{total_databases}")
+    logger.info(f"失败: {total_databases - success_count}/{total_databases}")
     logger.info(f"输出目录: {base_output_dir.absolute()}")
-    logger.info(f"结果说明: {message}")
     logger.info("=" * 60)
 
     return result

tests/test_chat_realtime_name2id_sync.py

@@ -112,6 +112,125 @@ class TestChatRealtimeName2IdSync(unittest.TestCase):
                 ],
             )
 
+    def test_sync_still_inserts_new_messages_when_name2id_is_up_to_date(self):
+        with TemporaryDirectory() as td:
+            account_dir = Path(td) / "acc"
+            account_dir.mkdir(parents=True, exist_ok=True)
+
+            username = "wxid_friend"
+            table_name = f"Msg_{hashlib.md5(username.encode('utf-8')).hexdigest()}"
+            msg_db_path = account_dir / "message_0.db"
+
+            conn = sqlite3.connect(str(msg_db_path))
+            try:
+                conn.execute("CREATE TABLE Name2Id (user_name TEXT, is_session INTEGER DEFAULT 1)")
+                conn.execute(
+                    """
+                    CREATE TABLE "{table_name}" (
+                        local_id INTEGER PRIMARY KEY,
+                        server_id INTEGER,
+                        local_type INTEGER,
+                        sort_seq INTEGER,
+                        real_sender_id INTEGER,
+                        create_time INTEGER,
+                        message_content TEXT,
+                        compress_content BLOB,
+                        packed_info_data BLOB
+                    )
+                    """.format(table_name=table_name)
+                )
+                conn.execute("INSERT INTO Name2Id(rowid, user_name, is_session) VALUES (1, ?, 1)", ("acc",))
+                conn.execute("INSERT INTO Name2Id(rowid, user_name, is_session) VALUES (2, ?, 1)", (username,))
+                conn.execute(
+                    f'INSERT INTO "{table_name}" '
+                    "(local_id, server_id, local_type, sort_seq, real_sender_id, create_time, message_content, compress_content, packed_info_data) "
+                    "VALUES (?, ?, ?, ?, ?, ?, ?, ?, ?)",
+                    (10, 10010, 1, 10, 2, 1710000010, "old", None, None),
+                )
+                conn.commit()
+            finally:
+                conn.close()
+
+            session_conn = sqlite3.connect(str(account_dir / "session.db"))
+            try:
+                session_conn.execute(
+                    """
+                    CREATE TABLE SessionTable (
+                        username TEXT PRIMARY KEY,
+                        summary TEXT DEFAULT '',
+                        last_timestamp INTEGER DEFAULT 0,
+                        sort_timestamp INTEGER DEFAULT 0,
+                        last_msg_locald_id INTEGER DEFAULT 0,
+                        last_msg_type INTEGER DEFAULT 0,
+                        last_msg_sub_type INTEGER DEFAULT 0,
+                        last_msg_sender TEXT DEFAULT ''
+                    )
+                    """
+                )
+                session_conn.commit()
+            finally:
+                session_conn.close()
+
+            def _fake_exec_query(_handle, *, kind, path, sql):
+                self.assertEqual(kind, "message")
+                self.assertTrue(str(path).endswith("message_0.db"))
+                if "COUNT(1)" in sql:
+                    return [{"c": 2, "mx": 2}]
+                raise AssertionError(f"Unexpected SQL: {sql}")
+
+            live_messages = [
+                {
+                    "local_id": 11,
+                    "server_id": 10011,
+                    "local_type": 1,
+                    "sort_seq": 11,
+                    "real_sender_id": 2,
+                    "create_time": 1710000011,
+                    "message_content": "new message",
+                    "compress_content": None,
+                    "sender_username": username,
+                }
+            ]
+
+            with (
+                patch.object(
+                    chat_router,
+                    "_resolve_db_storage_message_paths",
+                    return_value=(Path(td) / "live_message_0.db", Path(td) / "message_resource.db"),
+                ),
+                patch.object(chat_router, "_wcdb_exec_query", side_effect=_fake_exec_query),
+                patch.object(chat_router, "_wcdb_get_messages", side_effect=[list(live_messages)]),
+                patch.object(chat_router, "_best_effort_upsert_output_name2id_rows") as mock_upsert_name2id,
+            ):
+                result = chat_router._sync_chat_realtime_messages_for_table(
+                    account_dir=account_dir,
+                    rt_conn=_DummyConn(),
+                    username=username,
+                    msg_db_path=msg_db_path,
+                    table_name=table_name,
+                    max_scan=50,
+                    backfill_limit=0,
+                )
+
+            self.assertEqual(result.get("inserted"), 1)
+            mock_upsert_name2id.assert_not_called()
+
+            conn = sqlite3.connect(str(msg_db_path))
+            try:
+                rows = conn.execute(
+                    f'SELECT local_id, server_id, real_sender_id, create_time, message_content FROM "{table_name}" ORDER BY local_id ASC'
+                ).fetchall()
+            finally:
+                conn.close()
+
+            self.assertEqual(
+                rows,
+                [
+                    (10, 10010, 2, 1710000010, "old"),
+                    (11, 10011, 2, 1710000011, "new message"),
+                ],
+            )
+
 
 if __name__ == "__main__":
     unittest.main()

tests/test_decrypt_stream_sse.py

@@ -3,44 +3,14 @@ import os
 import sys
 import unittest
 import importlib
-import hashlib
-import hmac
 from pathlib import Path
 from tempfile import TemporaryDirectory
 
-from cryptography.hazmat.backends import default_backend
-from cryptography.hazmat.primitives.ciphers import Cipher, algorithms, modes
-
 
 ROOT = Path(__file__).resolve().parents[1]
 sys.path.insert(0, str(ROOT / "src"))
 
 
-def _encrypt_page(raw_key: bytes, plain_page: bytes, page_num: int, salt: bytes, iv: bytes) -> bytes:
-    from wechat_decrypt_tool.wechat_decrypt import PAGE_SIZE, RESERVE_SIZE, SALT_SIZE, _derive_mac_key
-
-    if page_num == 1:
-        encrypted_input = plain_page[SALT_SIZE : PAGE_SIZE - RESERVE_SIZE]
-        prefix = salt
-    else:
-        encrypted_input = plain_page[: PAGE_SIZE - RESERVE_SIZE]
-        prefix = b""
-
-    cipher = Cipher(
-        algorithms.AES(raw_key),
-        modes.CBC(iv),
-        backend=default_backend(),
-    )
-    encryptor = cipher.encryptor()
-    encrypted = encryptor.update(encrypted_input) + encryptor.finalize()
-
-    page_without_hmac = prefix + encrypted + iv
-    mac = hmac.new(_derive_mac_key(raw_key, salt), digestmod=hashlib.sha512)
-    mac.update(page_without_hmac[SALT_SIZE if page_num == 1 else 0 :])
-    mac.update(page_num.to_bytes(4, "little"))
-    return page_without_hmac + mac.digest()
-
-
 class TestDecryptStreamSSE(unittest.TestCase):
     def test_decrypt_stream_reports_progress(self):
         from fastapi import FastAPI
@@ -115,76 +85,6 @@ class TestDecryptStreamSSE(unittest.TestCase):
                 else:
                     os.environ["WECHAT_TOOL_BUILD_SESSION_LAST_MESSAGE"] = prev_build_cache
 
-    def test_decrypt_stream_reports_key_scope_error_for_wrong_key(self):
-        from fastapi import FastAPI
-        from fastapi.testclient import TestClient
-
-        from wechat_decrypt_tool.wechat_decrypt import PAGE_SIZE, RESERVE_SIZE, SQLITE_HEADER
-
-        good_key = bytes.fromhex("00112233445566778899aabbccddeefffedcba98765432100123456789abcdef")
-        bad_key = "ffeeddccbbaa998877665544332211000123456789abcdeffedcba9876543210"
-        salt = bytes.fromhex("11223344556677889900aabbccddeeff")
-        iv1 = bytes.fromhex("0102030405060708090a0b0c0d0e0f10")
-        plain_page = SQLITE_HEADER + (b"A" * (PAGE_SIZE - RESERVE_SIZE - len(SQLITE_HEADER))) + (b"\x00" * RESERVE_SIZE)
-        encrypted_db = _encrypt_page(good_key, plain_page, 1, salt, iv1)
-
-        with TemporaryDirectory() as td:
-            root = Path(td)
-
-            prev_data_dir = os.environ.get("WECHAT_TOOL_DATA_DIR")
-            prev_build_cache = os.environ.get("WECHAT_TOOL_BUILD_SESSION_LAST_MESSAGE")
-            try:
-                os.environ["WECHAT_TOOL_DATA_DIR"] = str(root)
-                os.environ["WECHAT_TOOL_BUILD_SESSION_LAST_MESSAGE"] = "0"
-
-                import wechat_decrypt_tool.app_paths as app_paths
-                import wechat_decrypt_tool.routers.decrypt as decrypt_router
-
-                importlib.reload(app_paths)
-                importlib.reload(decrypt_router)
-
-                db_storage = root / "xwechat_files" / "wxid_wrong_key_user" / "db_storage"
-                db_storage.mkdir(parents=True, exist_ok=True)
-                (db_storage / "MSG0.db").write_bytes(encrypted_db)
-
-                app = FastAPI()
-                app.include_router(decrypt_router.router)
-                client = TestClient(app)
-
-                events: list[dict] = []
-                with client.stream(
-                    "GET",
-                    "/api/decrypt_stream",
-                    params={"key": bad_key, "db_storage_path": str(db_storage)},
-                ) as resp:
-                    self.assertEqual(resp.status_code, 200)
-                    for line in resp.iter_lines():
-                        if not line:
-                            continue
-                        if isinstance(line, bytes):
-                            line = line.decode("utf-8", errors="ignore")
-                        line = str(line)
-                        if line.startswith(":") or not line.startswith("data: "):
-                            continue
-                        payload = json.loads(line[len("data: ") :])
-                        events.append(payload)
-                        if payload.get("type") in {"complete", "error"}:
-                            break
-
-                self.assertEqual(events[-1].get("type"), "complete")
-                self.assertEqual(events[-1].get("status"), "failed")
-                self.assertIn("当前数据库密钥不正确", events[-1].get("message", ""))
-                self.assertIn("另一台设备复制", events[-1].get("message", ""))
-            finally:
-                if prev_data_dir is None:
-                    os.environ.pop("WECHAT_TOOL_DATA_DIR", None)
-                else:
-                    os.environ["WECHAT_TOOL_DATA_DIR"] = prev_data_dir
-                if prev_build_cache is None:
-                    os.environ.pop("WECHAT_TOOL_BUILD_SESSION_LAST_MESSAGE", None)
-                else:
-                    os.environ["WECHAT_TOOL_BUILD_SESSION_LAST_MESSAGE"] = prev_build_cache
-
 
 if __name__ == "__main__":
     unittest.main()

tests/test_parse_app_message.py

@@ -118,6 +118,36 @@ class TestParseAppMessage(unittest.TestCase):
         self.assertEqual(parsed.get("linkType"), "official_article")
         self.assertEqual(parsed.get("linkStyle"), "cover")
 
+    def test_finder_type_51_uses_nested_desc_and_cover(self):
+        raw_text = (
+            '<msg><appmsg appid="" sdkver="0">'
+            '<title>当前版本不支持展示该内容，请升级至最新版本。</title>'
+            '<des></des>'
+            '<type>51</type>'
+            '<url></url>'
+            '<finderFeed>'
+            '<nickname><![CDATA[央视新闻]]></nickname>'
+            '<username><![CDATA[finder_cctv_news]]></username>'
+            '<desc><![CDATA[微信视频号全金融行业今公布发布]]></desc>'
+            '<mediaList><media>'
+            '<coverUrl><![CDATA[https://finder.video.qq.com/cover.jpg]]></coverUrl>'
+            '<url><![CDATA[https://channels.weixin.qq.com/web/pages/feed?feedid=abc]]></url>'
+            '</media></mediaList>'
+            '</finderFeed>'
+            '</appmsg></msg>'
+        )
+
+        parsed = _parse_app_message(raw_text)
+
+        self.assertEqual(parsed.get("renderType"), "link")
+        self.assertEqual(parsed.get("linkType"), "finder")
+        self.assertEqual(parsed.get("title"), "微信视频号全金融行业今公布发布")
+        self.assertEqual(parsed.get("content"), "微信视频号全金融行业今公布发布")
+        self.assertEqual(parsed.get("from"), "央视新闻")
+        self.assertEqual(parsed.get("fromUsername"), "finder_cctv_news")
+        self.assertEqual(parsed.get("thumbUrl"), "https://finder.video.qq.com/cover.jpg")
+        self.assertEqual(parsed.get("url"), "https://channels.weixin.qq.com/web/pages/feed?feedid=abc")
+
     def test_quote_type_5_nested_xml_refermsg_uses_inner_title(self):
         raw_text = (
             '<msg><appmsg appid="" sdkver="0">'

tests/test_wechat_decrypt_raw_key.py

@@ -1,172 +0,0 @@
-import hashlib
-import hmac
-import os
-import sys
-import tempfile
-import unittest
-from pathlib import Path
-
-from cryptography.hazmat.backends import default_backend
-from cryptography.hazmat.primitives.ciphers import Cipher, algorithms, modes
-
-ROOT = Path(__file__).resolve().parents[1]
-sys.path.insert(0, str(ROOT / "src"))
-
-from wechat_decrypt_tool.wechat_decrypt import (
-    PAGE_SIZE,
-    RESERVE_SIZE,
-    SALT_SIZE,
-    SQLITE_HEADER,
-    WeChatDatabaseDecryptor,
-    _derive_mac_key,
-    _derive_sqlcipher_enc_key,
-    decrypt_wechat_databases,
-)
-
-
-def _encrypt_page(
-    raw_key: bytes,
-    plain_page: bytes,
-    page_num: int,
-    salt: bytes,
-    iv: bytes,
-    *,
-    sqlcipher_passphrase: bool = False,
-) -> bytes:
-    enc_key = _derive_sqlcipher_enc_key(raw_key, salt) if sqlcipher_passphrase else raw_key
-
-    if page_num == 1:
-        encrypted_input = plain_page[SALT_SIZE : PAGE_SIZE - RESERVE_SIZE]
-        prefix = salt
-    else:
-        encrypted_input = plain_page[: PAGE_SIZE - RESERVE_SIZE]
-        prefix = b""
-
-    cipher = Cipher(
-        algorithms.AES(enc_key),
-        modes.CBC(iv),
-        backend=default_backend(),
-    )
-    encryptor = cipher.encryptor()
-    encrypted = encryptor.update(encrypted_input) + encryptor.finalize()
-
-    page_without_hmac = prefix + encrypted + iv
-    mac = hmac.new(_derive_mac_key(enc_key, salt), digestmod=hashlib.sha512)
-    mac.update(page_without_hmac[SALT_SIZE if page_num == 1 else 0 :])
-    mac.update(page_num.to_bytes(4, "little"))
-    return page_without_hmac + mac.digest()
-
-
-def _build_plain_page(body_byte: int, *, first_page: bool) -> bytes:
-    if first_page:
-        payload = SQLITE_HEADER + bytes([body_byte]) * (PAGE_SIZE - RESERVE_SIZE - len(SQLITE_HEADER))
-    else:
-        payload = bytes([body_byte]) * (PAGE_SIZE - RESERVE_SIZE)
-    return payload + (b"\x00" * RESERVE_SIZE)
-
-
-class WeChatDecryptRawKeyTests(unittest.TestCase):
-    def test_decrypt_database_uses_raw_enc_key(self):
-        raw_key = bytes.fromhex("00112233445566778899aabbccddeefffedcba98765432100123456789abcdef")
-        salt = bytes.fromhex("11223344556677889900aabbccddeeff")
-        iv1 = bytes.fromhex("0102030405060708090a0b0c0d0e0f10")
-        iv2 = bytes.fromhex("1112131415161718191a1b1c1d1e1f20")
-
-        page1 = _build_plain_page(0x41, first_page=True)
-        page2 = _build_plain_page(0x42, first_page=False)
-        encrypted_db = _encrypt_page(raw_key, page1, 1, salt, iv1) + _encrypt_page(raw_key, page2, 2, salt, iv2)
-
-        with tempfile.TemporaryDirectory() as tmpdir:
-            src = Path(tmpdir) / "source.db"
-            dst = Path(tmpdir) / "out.db"
-            src.write_bytes(encrypted_db)
-
-            decryptor = WeChatDatabaseDecryptor(raw_key.hex())
-            self.assertTrue(decryptor.decrypt_database(str(src), str(dst)))
-            self.assertEqual(dst.read_bytes(), page1 + page2)
-
-    def test_decrypt_database_falls_back_to_sqlcipher_passphrase_mode(self):
-        passphrase_key = bytes.fromhex("9f5dd0d3b6d0477ea5045c9e380ee272e53927993eb548dd98a022e842d5f7bd")
-        salt = bytes.fromhex("50f4090ef6897e146f94109f13743e34")
-        iv1 = bytes.fromhex("0102030405060708090a0b0c0d0e0f10")
-        iv2 = bytes.fromhex("1112131415161718191a1b1c1d1e1f20")
-
-        page1 = _build_plain_page(0x41, first_page=True)
-        page2 = _build_plain_page(0x42, first_page=False)
-        encrypted_db = _encrypt_page(
-            passphrase_key,
-            page1,
-            1,
-            salt,
-            iv1,
-            sqlcipher_passphrase=True,
-        ) + _encrypt_page(
-            passphrase_key,
-            page2,
-            2,
-            salt,
-            iv2,
-            sqlcipher_passphrase=True,
-        )
-
-        with tempfile.TemporaryDirectory() as tmpdir:
-            src = Path(tmpdir) / "source.db"
-            dst = Path(tmpdir) / "out.db"
-            src.write_bytes(encrypted_db)
-
-            decryptor = WeChatDatabaseDecryptor(passphrase_key.hex())
-            self.assertTrue(decryptor.decrypt_database(str(src), str(dst)))
-            self.assertEqual(dst.read_bytes(), page1 + page2)
-
-    def test_decrypt_database_keeps_existing_output_on_hmac_failure(self):
-        good_key = bytes.fromhex("00112233445566778899aabbccddeefffedcba98765432100123456789abcdef")
-        bad_key_hex = "ffeeddccbbaa998877665544332211000123456789abcdeffedcba9876543210"
-        salt = bytes.fromhex("11223344556677889900aabbccddeeff")
-        iv1 = bytes.fromhex("0102030405060708090a0b0c0d0e0f10")
-
-        page1 = _build_plain_page(0x41, first_page=True)
-        encrypted_db = _encrypt_page(good_key, page1, 1, salt, iv1)
-
-        with tempfile.TemporaryDirectory() as tmpdir:
-            src = Path(tmpdir) / "source.db"
-            dst = Path(tmpdir) / "out.db"
-            src.write_bytes(encrypted_db)
-            dst.write_bytes(b"keep-existing-output")
-
-            decryptor = WeChatDatabaseDecryptor(bad_key_hex)
-            self.assertFalse(decryptor.decrypt_database(str(src), str(dst)))
-            self.assertEqual(dst.read_bytes(), b"keep-existing-output")
-
-    def test_decrypt_wechat_databases_reports_key_scope_message(self):
-        good_key = bytes.fromhex("00112233445566778899aabbccddeefffedcba98765432100123456789abcdef")
-        bad_key_hex = "ffeeddccbbaa998877665544332211000123456789abcdeffedcba9876543210"
-        salt = bytes.fromhex("11223344556677889900aabbccddeeff")
-        iv1 = bytes.fromhex("0102030405060708090a0b0c0d0e0f10")
-
-        page1 = _build_plain_page(0x41, first_page=True)
-        encrypted_db = _encrypt_page(good_key, page1, 1, salt, iv1)
-
-        with tempfile.TemporaryDirectory() as tmpdir:
-            root = Path(tmpdir)
-            db_storage = root / "xwechat_files" / "wxid_scope_user" / "db_storage"
-            db_storage.mkdir(parents=True, exist_ok=True)
-            (db_storage / "MSG0.db").write_bytes(encrypted_db)
-
-            prev_data_dir = os.environ.get("WECHAT_TOOL_DATA_DIR")
-            try:
-                os.environ["WECHAT_TOOL_DATA_DIR"] = str(root)
-                result = decrypt_wechat_databases(str(db_storage), bad_key_hex)
-            finally:
-                if prev_data_dir is None:
-                    os.environ.pop("WECHAT_TOOL_DATA_DIR", None)
-                else:
-                    os.environ["WECHAT_TOOL_DATA_DIR"] = prev_data_dir
-
-            self.assertEqual(result["status"], "error")
-            self.assertIn("当前数据库密钥不正确", result["message"])
-            self.assertIn("账号/当前设备", result["message"])
-            self.assertIn("另一台设备复制", result["message"])
-
-
-if __name__ == "__main__":
-    unittest.main()