feat: 缺失文件扫描支持清理并发布 v2.1.1

src/_test/electron/library-notes.integration.js

@@ -357,6 +357,66 @@ app.whenReady().then(async () => {
         .every((title) => allTitles.includes(title))
       && (await js("document.getElementById('libStatus').textContent")) === '共 3 条'
   );
+
+  const rescanMissingFile = path.join(TMP, 'rescan-missing.txt');
+  fs.writeFileSync(rescanMissingFile, 'rescan missing fixture');
+  const rescanMissingShelf = library.addShelf({ name: 'Rescan Missing Shelf' });
+  const rescanMissingBook = library.add({
+    title: 'Rescan Missing Book',
+    shelfId: rescanMissingShelf.id,
+    tags: ['Rescan Missing Tag'],
+    files: [{ path: rescanMissingFile, name: 'rescan-missing.txt', format: 'TXT' }]
+  });
+  await pollJs(
+    '缺失扫描夹具显示在书库',
+    `!!document.querySelector('#libGrid .card[data-id="${rescanMissingBook.id}"]')`
+  );
+  fs.unlinkSync(rescanMissingFile);
+  await js("document.getElementById('rescanBtn').click()");
+  await waitForModal('发现文件缺失');
+  check(
+    '重新扫描报告缺失数量并说明保留阅读资料',
+    await js(`(() => {
+      const text = document.getElementById('modalBody').textContent;
+      return text.includes('1 条书库内容的文件均已缺失')
+        && text.includes('对应分类和标签也会一并清理')
+        && text.includes('笔记、书签、进度和标注会保留');
+    })()`)
+  );
+  await js("document.getElementById('modalCancel').click()");
+  await pollJs(
+    '取消缺失清理后恢复扫描按钮',
+    "!document.getElementById('rescanBtn').disabled",
+    5000
+  );
+  check(
+    '取消清理保留缺失书库条目',
+    !!library.get(rescanMissingBook.id)
+      && library.get(rescanMissingBook.id).missing
+      && library.listShelves().some((shelf) => shelf.id === rescanMissingShelf.id)
+      && library.listTags().some((tag) => tag.name === 'Rescan Missing Tag')
+  );
+
+  await js("document.getElementById('rescanBtn').click()");
+  await waitForModal('发现文件缺失');
+  await submitModal();
+  await poll(
+    '确认后清理缺失书库条目',
+    () => Promise.resolve(!library.get(rescanMissingBook.id))
+  );
+  await pollJs(
+    '缺失条目清理后书库恢复',
+    `!document.querySelector('#libGrid .card[data-id="${rescanMissingBook.id}"]')
+      && document.querySelectorAll('#libGrid .card').length === 3`
+  );
+  check(
+    '缺失清理移除对应空分类和标签且不影响其余内容',
+    library.list().length === 3
+      && library.list().every((item) => item.id !== rescanMissingBook.id)
+      && !library.listShelves().some((shelf) => shelf.id === rescanMissingShelf.id)
+      && !library.listTags().some((tag) => tag.name === 'Rescan Missing Tag')
+  );
+
   await js(`(() => {
     document.getElementById('librarySearchInput').value = 'Rsrch';
     document.getElementById('librarySearchBtn').click();

src/_test/library.test.js

@@ -335,6 +335,61 @@ test('批量移除只写一次索引，可选删除库内文件', () => {
   assert.deepStrictEqual(store.removeMany(['missing'], false), { removed: 0 });
 });
 
+test('清理缺失条目时一并清理对应的空书架和标签', () => {
+  const root = freshRoot('remove-missing');
+  const emptyShelf = store.addShelf({ name: '仅缺失书籍分类' });
+  const sharedShelf = store.addShelf({ name: '仍在使用分类' });
+  const unrelatedShelf = store.addShelf({ name: '无关空分类' });
+  store.addTag({ name: '仅缺失书籍标签' });
+  store.addTag({ name: '仍在使用标签' });
+  store.addTag({ name: '无关空标签' });
+
+  const missingA = store.add({
+    title: '缺失 A',
+    shelfId: emptyShelf.id,
+    tags: ['仅缺失书籍标签', '仍在使用标签'],
+    files: [{ path: path.join(root, 'missing-a.pdf') }]
+  });
+  const missingB = store.add({
+    title: '缺失 B',
+    shelfId: sharedShelf.id,
+    tags: ['仍在使用标签'],
+    files: [{ path: path.join(root, 'missing-b.pdf') }]
+  });
+  const existingPath = path.join(root, 'existing.pdf');
+  fs.writeFileSync(existingPath, '%PDF-1.4\n');
+  const existing = store.add({
+    title: '仍存在',
+    shelfId: sharedShelf.id,
+    tags: ['仍在使用标签'],
+    files: [{ path: existingPath }]
+  });
+  const noFiles = store.add({ title: '无文件元数据', files: [] });
+
+  assert.deepStrictEqual(store.removeMissing(), {
+    removed: 2,
+    shelvesRemoved: 1,
+    tagsRemoved: 1
+  });
+  assert.strictEqual(store.get(missingA.id), null);
+  assert.strictEqual(store.get(missingB.id), null);
+  assert.ok(store.get(existing.id));
+  assert.ok(store.get(noFiles.id), '没有文件的元数据条目不应按文件缺失清理');
+  assert.deepStrictEqual(
+    store.listShelves().map((shelf) => shelf.name).sort(),
+    [sharedShelf.name, unrelatedShelf.name].sort()
+  );
+  assert.deepStrictEqual(
+    store.listTags().map((tag) => tag.name).sort(),
+    ['仍在使用标签', '无关空标签'].sort()
+  );
+  assert.deepStrictEqual(store.removeMissing(), {
+    removed: 0,
+    shelvesRemoved: 0,
+    tagsRemoved: 0
+  });
+});
+
 test('explicit tag creation enforces the existing catalog limit', () => {
   const root = freshRoot('limit');
   const now = Date.now();

src/_test/main.test.js

@@ -73,7 +73,7 @@ test('下载校验协议，拒绝 file:// 等非 http(s)', () => {
 test('下载请求使用可识别且带项目地址的 User-Agent', () => {
   assert.match(mainSrc, /UA:\s*DL_UA[\s\S]+require\('\.\/src\/sources\/http'\)/);
   const httpSrc = fs.readFileSync(path.join(__dirname, '..', 'sources', 'http.js'), 'utf8');
-  assert.match(httpSrc, /PeopleLib\/2\.1\.0 \(\+https:\/\/github\.com\/lofyer\/peoplelib\)/);
+  assert.match(httpSrc, /PeopleLib\/2\.1\.1 \(\+https:\/\/github\.com\/lofyer\/peoplelib\)/);
 });
 
 test('移除书籍默认保留阅读资料，仅显式勾选时清理', () => {
@@ -108,6 +108,18 @@ test('批量整理与移除走单次索引提交，不按条目循环 IPC', () =
     '渲染层不应再逐条发 IPC');
 });
 
+test('重新扫描清理缺失条目前在主进程重新核对磁盘状态', () => {
+  const start = mainSrc.indexOf("ipcMain.handle('library:removeMissing'");
+  assert.ok(start > 0, '缺少文件缺失清理通道');
+  const end = mainSrc.indexOf('ipcMain.handle(', start + 20);
+  const segment = mainSrc.slice(start, end < 0 ? undefined : end);
+  assert.match(segment, /wrap\(async \(\) =>/);
+  assert.match(segment, /library\.list\(\)[\s\S]*\.filter\(\(item\) => item\.missing\)/);
+  assert.match(segment, /await requestReaderPurge\(id\)/);
+  assert.match(segment, /library\.removeMissing\(\)/);
+  assert.doesNotMatch(segment, /readerStore\.forget|annotations\.forget|aiSessions\.forget/);
+});
+
 test('孤立阅读资料只报告不自动删除，清理需分别勾选', () => {
   const start = mainSrc.indexOf("ipcMain.handle('reader:orphanReport'");
   assert.ok(start > 0, '缺少孤立资料对账通道');

src/_test/sources.test.js

@@ -455,7 +455,7 @@ test('wikisource: 搜索使用整数偏移并携带可识别 User-Agent', async
   assert.strictEqual(result.items[0].subtitle, '中文');
   assert.strictEqual(result.maxPage, 500, 'MediaWiki 搜索最多允许偏移到 10000 条');
   assert.ok(request.url.includes('sroffset=20'), request.url);
-  assert.match(request.options.headers['User-Agent'], /PeopleLib\/2\.1\.0/);
+  assert.match(request.options.headers['User-Agent'], /PeopleLib\/2\.1\.1/);
 });
 
 test('wikisource: 浏览按 continuation 令牌翻页', async () => {

src/_test/ui.test.js

@@ -645,6 +645,17 @@ test('书库多选提供全选、批量整理与批量移除', () => {
   assert.match(library, /window\.api\.library\.updateMany\(patches\)/);
 });
 
+test('重新扫描发现文件缺失后提示清理书库条目', () => {
+  const library = fs.readFileSync(libFile, 'utf8');
+  const preload = fs.readFileSync(path.join(__dirname, '..', '..', 'preload.js'), 'utf8');
+  assert.match(library, /const missing = Math\.max\(0, Number\(\(r\.data \|\| \{\}\)\.missing\)/);
+  assert.match(library, /confirmModal\(\s*'发现文件缺失'/);
+  assert.match(library, /对应分类和标签也会一并清理/);
+  assert.match(library, /笔记、书签、进度和标注会保留/);
+  assert.match(library, /window\.api\.library\.removeMissing\(\)/);
+  assert.match(preload, /removeMissing: \(\) => ipcRenderer\.invoke\('library:removeMissing'\)/);
+});
+
 test('卡片定位上下文不随多选状态消失', () => {
   const css = fs.readFileSync(path.join(__dirname, '..', 'ui', 'style.css'), 'utf8');
   // 退出多选是同步移除 class，重绘要等 IPC；若定位上下文只在 select-mode 下建立，

src/library/store.js

@@ -978,6 +978,38 @@ function removeMany(ids, deleteFiles) {
   return { removed: doomed.length };
 }
 
+function removeMissing() {
+  load();
+  const doomed = items.filter((item) => {
+    const files = item.files || [];
+    return files.length > 0 && files.every((file) => !fs.existsSync(toAbsolute(file.path)));
+  });
+  if (!doomed.length) return { removed: 0, shelvesRemoved: 0, tagsRemoved: 0 };
+
+  const ids = new Set(doomed.map((item) => item.id));
+  const candidateShelves = new Set(doomed.map((item) => item.shelfId).filter(Boolean));
+  const candidateTags = new Set(doomed.flatMap((item) => normalizeTags(item.tags).map(tagKey)));
+  const nextItems = items.filter((item) => !ids.has(item.id));
+  const usedShelves = new Set(nextItems.map((item) => item.shelfId).filter(Boolean));
+  const usedTags = new Set(nextItems.flatMap((item) => normalizeTags(item.tags).map(tagKey)));
+  const nextShelves = shelves.filter((shelf) => (
+    !candidateShelves.has(shelf.id) || usedShelves.has(shelf.id)
+  ));
+  const nextTags = tags.filter((tag) => (
+    !candidateTags.has(tagKey(tag.name)) || usedTags.has(tagKey(tag.name))
+  ));
+  const shelvesRemoved = shelves.length - nextShelves.length;
+  const tagsRemoved = tags.length - nextTags.length;
+
+  commit(nextItems, true, nextShelves, nextTags);
+  for (const item of doomed) removeCoverFile(item.id);
+  return {
+    removed: doomed.length,
+    shelvesRemoved,
+    tagsRemoved
+  };
+}
+
 function remove(id, deleteFiles) {
   load();
   const it = items.find((x) => x.id === id);
@@ -1311,7 +1343,7 @@ function importLegacy(legacyDir) {
 module.exports = {
   init, getRoot, filesDir, allocFilePath, sanitize,
   list, get, findBySource, listShelves, listTags,
-  add, importLocal, update, updateMany, remove, removeMany, attachFile,
+  add, importLocal, update, updateMany, remove, removeMany, removeMissing, attachFile,
   addShelf, updateShelf, removeShelf,
   addTag, updateTag, removeTag,
   scan, migrateTo, finalizeMigration, rollbackMigration, importLegacy,

src/sources/http.js

@@ -1,4 +1,4 @@
-const UA = 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/120.0 Safari/537.36 PeopleLib/2.1.0 (+https://github.com/lofyer/peoplelib)';
+const UA = 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/120.0 Safari/537.36 PeopleLib/2.1.1 (+https://github.com/lofyer/peoplelib)';
 const { fetch: undiciFetch, ProxyAgent } = require('undici');
 
 // 代理配置：默认直连（空字符串）。一旦设置，所有 HTTP 请求统一走该代理。

src/ui/views/library.js

@@ -84,9 +84,39 @@ const Library = (() => {
     btn.disabled = true;
     btn.textContent = '扫描中...';
     const r = await window.api.library.scan();
-    btn.textContent = r.ok && r.data && r.data.added ? `新增 ${r.data.added} 本 ✓` : '已是最新 ✓';
+    if (!r || !r.ok) {
+      btn.textContent = '扫描失败';
+      await confirmModal('扫描失败', (r && r.error) || '未知错误');
+      setTimeout(() => { btn.textContent = '重新扫描'; btn.disabled = false; }, 2000);
+      return;
+    }
+    btn.textContent = r.data && r.data.added ? `新增 ${r.data.added} 本 ✓` : '已是最新 ✓';
     dirty = true;
     await refresh(true);
+    const missing = Math.max(0, Number((r.data || {}).missing) || 0);
+    if (missing) {
+      const clean = await confirmModal(
+        '发现文件缺失',
+        `扫描发现 ${missing} 条书库内容的文件均已缺失，是否从书库中清理这些条目？不再被其他书籍使用的对应分类和标签也会一并清理；笔记、书签、进度和标注会保留。`
+      );
+      if (clean) {
+        const removed = await window.api.library.removeMissing();
+        if (!removed || !removed.ok) {
+          await confirmModal('清理失败', (removed && removed.error) || '未知错误');
+        } else {
+          const count = Math.max(0, Number((removed.data || {}).removed) || 0);
+          const shelvesRemoved = Math.max(0, Number((removed.data || {}).shelvesRemoved) || 0);
+          const tagsRemoved = Math.max(0, Number((removed.data || {}).tagsRemoved) || 0);
+          const extras = [
+            shelvesRemoved ? `${shelvesRemoved} 个分类` : '',
+            tagsRemoved ? `${tagsRemoved} 个标签` : ''
+          ].filter(Boolean);
+          btn.textContent = `已清理 ${count} 条${extras.length ? `、${extras.join('、')}` : ''} ✓`;
+          dirty = true;
+          await refresh(true);
+        }
+      }
+    }
     setTimeout(() => { btn.textContent = '重新扫描'; btn.disabled = false; }, 2000);
   }