Improve LightRAG knowledge search locator alignment

scripts/reasonix-acp-wrapper.mjs

@@ -37,6 +37,7 @@ function debugLog(message) {
 }
 
 const toolContextStorage = new AsyncLocalStorage();
+const MNOTE_UI_CITATION_QUEUE = [];
 
 const MNOTE_TOOL_NAMES = [
   'mnote.skill.read',
@@ -408,6 +409,8 @@ let nextId = 1;
 const pendingReqs = new Map(); // id → { resolve, reject }
 const requestHandlers = new Map(); // method → handler
 const notificationHandlers = new Map(); // method → handler
+let rpcServerReady = false;
+const queuedRpcLines = [];
 
 function sendMessage(msg) {
   const line = JSON.stringify(msg) + '\n';
@@ -434,9 +437,7 @@ function onNotification(method, handler) {
   notificationHandlers.set(method, handler);
 }
 
-// Start reading NDJSON from stdin
-const rl = createInterface({ input: stdin, terminal: false });
-rl.on('line', async (raw) => {
+async function handleRpcLine(raw) {
   const trimmed = raw.trim();
   if (!trimmed) return;
 
@@ -480,6 +481,16 @@ rl.on('line', async (raw) => {
     const handler = notificationHandlers.get(msg.method);
     if (handler) handler(msg.params);
   }
+}
+
+// 先接住 stdio 输入，但等全部 handler 注册完成后再处理，避免 initialize 抢跑。
+const rl = createInterface({ input: stdin, terminal: false });
+rl.on('line', (raw) => {
+  if (!rpcServerReady) {
+    queuedRpcLines.push(raw);
+    return;
+  }
+  void handleRpcLine(raw);
 });
 
 // ── Helper: send ACP session/update (camelCase per protocol spec) ──
@@ -566,7 +577,129 @@ async function callMnoteTool(toolName, args) {
     const text = await response.text().catch(() => '');
     throw new Error(`mnote tool ${toolName} failed: HTTP ${response.status} ${text}`);
   }
-  return response.json();
+  const result = compactMnoteToolResultForReasonix(toolName, await response.json());
+  const citations = collectUiCitationMarkdowns(result).slice(0, 8);
+  if (citations.length) MNOTE_UI_CITATION_QUEUE.push(citations);
+  return result;
+}
+
+function compactMnoteToolResultForReasonix(toolName, payload) {
+  if (toolName !== 'mnote.knowledge_rag.query') return payload;
+  const result = payload?.result && typeof payload.result === 'object' ? payload.result : payload;
+  const citationMarkdowns = collectUiCitationMarkdowns(result).slice(0, 8);
+  const citations = citationMarkdowns
+    .map((citationMarkdown) => ({ citationMarkdown }))
+    .slice(0, 8);
+  if (!citations.length || !result || typeof result !== 'object') return payload;
+  return {
+    ok: payload?.ok !== false,
+    schema: result.schema || 'mnote.knowledge_rag.agent_query_result.v1',
+    uiCitations: citations,
+    citationRendering: 'MNote UI renders uiCitations after the answer as clickable source locators. Do not copy citationMarkdown into the final answer and do not hand-write /documents links.',
+    answerCitationPolicy: 'Answer the substance in plain text. Mention source titles only if useful; leave clickable citation insertion to MNote UI.',
+    answerGuidance: result.answerGuidance || '',
+    references: Array.isArray(result.references) ? result.references.slice(0, 8) : [],
+    citations: citationMarkdowns,
+    sourceScope: result.sourceScope || [],
+    sourceScopeMode: result.sourceScopeMode || '',
+    rawScopeFiltered: Boolean(result.rawScopeFiltered),
+  };
+}
+
+function collectUiCitationMarkdowns(value) {
+  const references = Array.isArray(value?.references) ? value.references : [];
+  const referenceCitations = [];
+  if (references.length) {
+    const hasPrecise = references.some((reference) =>
+      typeof reference?.citationMarkdown === 'string' &&
+      reference.citationMarkdown.trim() &&
+      reference.locatorDegraded !== true
+    );
+    const seen = new Set();
+    for (const reference of references) {
+      const citation = String(reference?.citationMarkdown || '').trim();
+      if (!citation || seen.has(citation)) continue;
+      if (hasPrecise && reference?.locatorDegraded === true) continue;
+      seen.add(citation);
+      referenceCitations.push(citation);
+    }
+    if (referenceCitations.length) return referenceCitations;
+  }
+  const citations = collectCitationMarkdowns(value);
+  const hasPrecise = citations.some((item) => !isDegradedCitationMarkdown(item));
+  return citations.filter((citation) => {
+    return !hasPrecise || !isDegradedCitationMarkdown(citation);
+  });
+}
+
+function isDegradedCitationMarkdown(value) {
+  const text = String(value || '').toLowerCase();
+  return text.includes('来源定位降级') || text.includes('locator degraded');
+}
+
+function collectCitationMarkdowns(value) {
+  const out = [];
+  const seen = new Set();
+  function add(text) {
+    const value = String(text || '').trim();
+    if (!value || seen.has(value)) return;
+    seen.add(value);
+    out.push(value);
+  }
+  function visit(node) {
+    if (!node) return;
+    if (typeof node === 'string') {
+      const trimmed = node.trim();
+      if (trimmed.startsWith('{') || trimmed.startsWith('[')) {
+        try {
+          visit(JSON.parse(trimmed));
+        } catch {}
+      }
+      return;
+    }
+    if (Array.isArray(node)) {
+      node.forEach(visit);
+      return;
+    }
+    if (typeof node !== 'object') return;
+    if (typeof node.citationMarkdown === 'string') add(node.citationMarkdown);
+    Object.values(node).forEach(visit);
+  }
+  visit(value);
+  return out;
+}
+
+function toolResultTextWithUiCitations(rawText) {
+  const text = String(rawText || '');
+  let citationMarkdowns = collectUiCitationMarkdownsFromText(text).slice(0, 8);
+  if (!citationMarkdowns.length && MNOTE_UI_CITATION_QUEUE.length) {
+    citationMarkdowns = MNOTE_UI_CITATION_QUEUE.shift();
+  }
+  const citations = citationMarkdowns.map((citationMarkdown) => ({ citationMarkdown })).slice(0, 8);
+  if (!citations.length) return text.slice(0, 8000);
+  const prefix = JSON.stringify({
+    schema: 'mnote.acp.tool_result_ui_citations.v1',
+    uiCitations: citations,
+    citationRendering: 'MNote UI renders these citations after the answer; the model must not hand-write local citation links.',
+  });
+  const budget = Math.max(0, 8000 - prefix.length - 2);
+  return `${prefix}\n${text.slice(0, budget)}`;
+}
+
+function collectUiCitationMarkdownsFromText(text) {
+  const trimmed = String(text || '').trim();
+  if (trimmed.startsWith('{') || trimmed.startsWith('[')) {
+    try {
+      return collectUiCitationMarkdowns(JSON.parse(trimmed));
+    } catch {}
+    const firstLine = trimmed.split('\n')[0]?.trim();
+    if (firstLine && firstLine !== trimmed && (firstLine.startsWith('{') || firstLine.startsWith('['))) {
+      try {
+        return collectUiCitationMarkdowns(JSON.parse(firstLine));
+      } catch {}
+    }
+  }
+  return collectCitationMarkdowns(trimmed);
 }
 
 // ── Register Tools ───────────────────────────────────
@@ -637,7 +770,7 @@ function fallbackMnoteToolSpecs() {
     {
       mnoteToolName: 'mnote.knowledge_rag.query',
       name: 'mnote_knowledge_rag_query',
-      description: '向 LightRAG 资料库提问，返回 provider 原始结果和经 MNote registry 映射、sourcePaths 后过滤的 references。回答必须引用返回来源，不要引用 raw chunks。',
+      description: '向 LightRAG 资料库提问，返回 provider 原始结果和经 MNote registry 映射、sourcePaths 后过滤的 references/citations。用户要求链接、来源、引用或证据时优先调用。不要在最终回答中手写 citationMarkdown、/documents、mnote:// 或搜索引擎包装链接；MNote 前端会把 uiCitations/citationMarkdown 自动追加成可点击来源。不要引用 raw chunks。',
       parameters: {
         type: 'object',
         properties: {
@@ -775,7 +908,7 @@ onRequest('session/new', async (params) => {
     '<available-skills>',
     'Use mnote_skill_read to load the full content of any skill listed by the current prompt capabilities.',
     '- mnote-current-page — Read the current MNote Markdown page when the task needs page content.',
-    '- mnote-knowledge-rag — Ask the LightRAG-backed knowledge library across books, papers, PDFs, Office files, images, and attachments. Use mnote_knowledge_rag_query for knowledge-library questions and cite only returned references; sourcePaths filters returned references after provider retrieval, so do not cite raw chunks. If locatorDegraded is true, say the source location is degraded rather than inventing page/bbox.',
+    '- mnote-knowledge-rag — Ask the LightRAG-backed knowledge library across books, papers, PDFs, Office files, images, and attachments. Use mnote_knowledge_rag_query for knowledge-library questions and answers that require links/sources/citations. Returned uiCitations/citationMarkdown values are MNote clickable source locators and are rendered by the MNote UI after the final answer. Do not copy citationMarkdown into the answer, do not hand-write /documents or mnote:// links, and never wrap local citation URLs with search engines. Mention source titles in plain text only when useful; sourcePaths filters returned references after provider retrieval, so do not cite raw chunks. If locatorDegraded is true, say the source location is degraded rather than inventing page/bbox.',
     '- mnote-local-file — Resolve MNote targets and then use native file tools inside allowed roots.',
     '- mnote-chat-only — Reply conversationally without MNote file/page tools.',
     '</available-skills>',
@@ -841,6 +974,7 @@ onRequest('session/prompt', async (params) => {
   const announcedToolKeys = new Set();
   const preparingToolCallIds = [];
   const inflightToolCallIds = [];
+  MNOTE_UI_CITATION_QUEUE.length = 0;
 
   function nextToolCallId() {
     return `tc_${nextToolCallSeq++}`;
@@ -933,7 +1067,7 @@ onRequest('session/prompt', async (params) => {
         }
         case 'tool': {
           hasToolCall = true;
-          const resultText = String(ev.content || '').slice(0, 8000);
+          const resultText = toolResultTextWithUiCitations(ev.content);
           emitToolResult(
             session.id,
             inflightToolCallIds.shift() || ev.callId || nextToolCallId(),
@@ -990,4 +1124,8 @@ onNotification('session/cancel', (params) => {
 
 // ── Start ────────────────────────────────────────────
 
+rpcServerReady = true;
+for (const raw of queuedRpcLines.splice(0)) {
+  await handleRpcLine(raw);
+}
 process.stderr.write(`[reasonix-acp-mnote] ready (mnote=${MNOTE_WEB_URL})\n`);

scripts/task538-knowledge-rag-source-scope-api-smoke.js

@@ -112,6 +112,21 @@ async function query(context, queryText, sourcePaths) {
   return result.payload;
 }
 
+async function search(context, queryText, sourcePaths) {
+  const result = await apiJson(context, "POST", `${BASE_URL}/api/knowledge-rag/search`, {
+    rootUri: ROOT_URI,
+    workspaceId: WORKSPACE_ID,
+    query: queryText,
+    mode: "mix",
+    topK: 12,
+    chunkTopK: 12,
+    includeChunkContent: true,
+    sourcePaths,
+  });
+  assert(result.ok, `knowledge-rag search 失败: ${result.status} ${result.text.slice(0, 1000)}`);
+  return result.payload;
+}
+
 async function deleteSource(context, sourcePath) {
   const result = await apiJson(context, "POST", `${BASE_URL}/api/knowledge-rag/delete-source`, {
     rootUri: ROOT_URI,
@@ -146,6 +161,23 @@ async function main() {
       `sourcePaths scope 不应返回非 alpha 来源: ${JSON.stringify(references, null, 2).slice(0, 3000)}`,
     );
     assert(scoped.raw, "HTTP API 仍应保留 raw 供调试调用方使用");
+    const searched = await search(context, marker, [sourceA]);
+    const searchResults = Array.isArray(searched.results) ? searched.results : [];
+    assert.equal(searched.schema, "mnote.knowledge_rag.search_results.v1", `search schema 不正确: ${JSON.stringify(searched, null, 2).slice(0, 3000)}`);
+    assert.equal(searched.sourceScopeMode, "post_filter_mapped_references", `search sourceScopeMode 不正确: ${JSON.stringify(searched, null, 2).slice(0, 3000)}`);
+    assert(searchResults.length > 0, `资料库 search 应至少返回 alpha: ${JSON.stringify(searched, null, 2).slice(0, 3000)}`);
+    assert(
+      searchResults.every((item) => item.provider === "lightrag" && item.matchSource === "lightrag_reference"),
+      `资料库 search 结果必须来自 LightRAG references: ${JSON.stringify(searchResults, null, 2).slice(0, 3000)}`,
+    );
+    assert(
+      searchResults.every((item) => item.path === sourceA),
+      `资料库 search sourcePaths scope 不应返回非 alpha 来源: ${JSON.stringify(searchResults, null, 2).slice(0, 3000)}`,
+    );
+    assert(
+      searchResults.some((item) => item.citationUrl || item.locator || item.openAction),
+      `资料库 search 结果必须可打开来源: ${JSON.stringify(searchResults, null, 2).slice(0, 3000)}`,
+    );
 
     await deleteSource(context, sourceA);
     assert(fs.existsSync(path.join(ROOT_PATH, sourceA)), "delete-source 不应删除用户原始 source 文件");
@@ -164,6 +196,7 @@ async function main() {
       sourceScopeMode: scoped.sourceScopeMode,
       rawScopeFiltered: scoped.rawScopeFiltered,
       scopedReferenceCount: references.length,
+      searchResultCount: searchResults.length,
       alphaDocId: entryFor(indexedStatus, sourceA)?.lightRagDocId || null,
       betaDocId: entryFor(indexedStatus, sourceB)?.lightRagDocId || null,
       sourceAExistsAfterDelete: fs.existsSync(path.join(ROOT_PATH, sourceA)),

scripts/task540-knowledge-rag-office-result-open-locator-smoke.js

@@ -0,0 +1,179 @@
+#!/usr/bin/env node
+"use strict";
+
+const assert = require("node:assert");
+const fs = require("node:fs");
+const path = require("node:path");
+const { chromium } = require("playwright");
+
+const BASE_URL = (process.env.MNOTE_WEB_SMOKE_BASE_URL || "http://127.0.0.1:3000").replace(/\/+$/, "");
+const ROOT_URI = process.env.MNOTE_KNOWLEDGE_RAG_ROOT_URI || "file:///mnt/Data1T/Mnote_data/users/mnote-e2e/workspaces/my-space";
+const WORKSPACE_ID = process.env.MNOTE_KNOWLEDGE_RAG_WORKSPACE_ID || "local-ws:mnote-e2e:my-space";
+const QUERY = process.env.MNOTE_KNOWLEDGE_RAG_OFFICE_QUERY || "三乙基硅";
+const EXPECTED_RESOURCE = process.env.MNOTE_KNOWLEDGE_RAG_EXPECTED_RESOURCE
+  || "有机合成中的保护基/[OCR]_有机合成中的保护基-酚羰基羧基巯基的保护_20250201_1908.layered_删减-2025-02-04 18-59-42.docx";
+const EXPECTED_BLOCK_ID = process.env.MNOTE_KNOWLEDGE_RAG_EXPECTED_BLOCK_ID || "0ace5daa070f0911e09d8ab37c64eeea";
+const EXPECTED_HIGHLIGHT = process.env.MNOTE_KNOWLEDGE_RAG_EXPECTED_HIGHLIGHT || "三乙基硅酯";
+const OUTPUT_DIR = path.join(process.cwd(), "tmp", "task540-knowledge-rag-office-result-open-locator-smoke");
+const RESULT_PATH = path.join(OUTPUT_DIR, "result.json");
+const SEARCH_SCREENSHOT_PATH = path.join(OUTPUT_DIR, "office-search-result.png");
+const OPEN_SCREENSHOT_PATH = path.join(OUTPUT_DIR, "office-open-locator.png");
+const UI_TIMEOUT_MS = Number(process.env.MNOTE_SMOKE_UI_TIMEOUT_MS || 30_000);
+const CHROMIUM_EXECUTABLE_PATH = process.env.PLAYWRIGHT_CHROMIUM_EXECUTABLE_PATH
+  || ["/usr/bin/google-chrome-stable", "/usr/bin/google-chrome", "/snap/bin/chromium"]
+    .find((candidate) => fs.existsSync(candidate));
+
+async function main() {
+  fs.mkdirSync(OUTPUT_DIR, { recursive: true });
+  const browser = await chromium.launch({ headless: true, executablePath: CHROMIUM_EXECUTABLE_PATH });
+  const context = await browser.newContext({ viewport: { width: 1360, height: 900 } });
+  try {
+    const auth = await context.request.post(`${BASE_URL}/api/auth`, {
+      data: {
+        action: "auth:signIn",
+        args: {
+          provider: "password",
+          params: {
+            email: "mnote.e2e@example.com",
+            password: process.env.MNOTE_E2E_PASSWORD || "MnoteE2E123!",
+            flow: "signIn",
+          },
+        },
+      },
+    });
+    assert(auth.ok(), `登录失败: ${auth.status()} ${await auth.text()}`);
+
+    const apiSearch = await context.request.post(`${BASE_URL}/api/knowledge-rag/search`, {
+      data: {
+        rootUri: ROOT_URI,
+        workspaceId: WORKSPACE_ID,
+        query: QUERY,
+        mode: "mix",
+        topK: 8,
+        chunkTopK: 8,
+        includeChunkContent: true,
+        sourcePaths: [EXPECTED_RESOURCE],
+      },
+    });
+    assert(apiSearch.ok(), `knowledge-rag search 失败: ${apiSearch.status()} ${await apiSearch.text()}`);
+    const apiPayload = await apiSearch.json();
+    const apiFirst = apiPayload.results?.[0];
+    assert(apiFirst, `API 未返回资料库结果: ${JSON.stringify(apiPayload, null, 2).slice(0, 3000)}`);
+    assert.equal(apiFirst.locator?.blockId, EXPECTED_BLOCK_ID, `API locator blockId 不匹配: ${JSON.stringify(apiFirst, null, 2)}`);
+    assert(String(apiFirst.snippet || "").includes(EXPECTED_HIGHLIGHT), `API snippet 未命中 ${EXPECTED_HIGHLIGHT}: ${apiFirst.snippet}`);
+
+    const page = await context.newPage();
+    const pageErrors = [];
+    const consoleErrors = [];
+    page.on("pageerror", (error) => pageErrors.push(String(error?.stack || error?.message || error)));
+    page.on("console", (message) => {
+      if (["error", "warning"].includes(message.type())) consoleErrors.push(`${message.type()}: ${message.text()}`);
+    });
+    await page.goto(`${BASE_URL}/`, { waitUntil: "domcontentloaded", timeout: UI_TIMEOUT_MS });
+    await page.evaluate(() => {
+      window.localStorage?.setItem("mnote.search.collapseSourcesDefault.v1", "0");
+    });
+    await page.waitForSelector('[data-mnote-action="open-search-modal"]', { timeout: UI_TIMEOUT_MS });
+    await page.click('[data-mnote-action="open-search-modal"]');
+    await page.waitForSelector('[data-testid="wolai-search-input"]', { timeout: UI_TIMEOUT_MS });
+    await page.fill('[data-testid="wolai-search-input"]', QUERY);
+    await page.evaluate(() => {
+      const button = document.querySelector('[data-search-switch="knowledge"]');
+      if (button instanceof HTMLElement) {
+        button.setAttribute("aria-checked", "true");
+        button.classList.add("is-on");
+      }
+      document.querySelector('[data-testid="wolai-search-input"]')
+        ?.dispatchEvent(new Event("input", { bubbles: true }));
+    });
+    await page.waitForFunction(
+      (expected) => {
+        const first = document.querySelector('[data-testid="wolai-search-result-row"]');
+        return first && first.textContent.includes(expected);
+      },
+      EXPECTED_HIGHLIGHT,
+      { timeout: UI_TIMEOUT_MS },
+    );
+    await page.screenshot({ path: SEARCH_SCREENSHOT_PATH, fullPage: true });
+    await page.click('[data-testid="wolai-search-result-row"]');
+    await page.waitForFunction(
+      (expectedResource) => document.querySelector(`[data-mnote-resource-tab-panel][data-resource-path="${CSS.escape(expectedResource)}"]`),
+      EXPECTED_RESOURCE,
+      { timeout: UI_TIMEOUT_MS },
+    );
+    await page.waitForTimeout(8_000);
+
+    const state = await page.evaluate((expectedResource) => {
+      const panel = document.querySelector(`[data-mnote-resource-tab-panel][data-resource-path="${CSS.escape(expectedResource)}"]`);
+      const frame = panel?.querySelector("iframe.mnote-resource-tab-frame");
+      const result = {
+        url: location.href,
+        panelVisible: !!panel && !panel.hidden,
+        panelResourcePath: panel ? panel.getAttribute("data-resource-path") : "",
+        panelBlockId: panel ? panel.getAttribute("data-mnote-evidence-block-id") : "",
+        panelEvidenceText: panel ? panel.getAttribute("data-mnote-evidence-text") : "",
+        frameSrc: frame ? frame.getAttribute("src") : "",
+      };
+      if (frame?.contentDocument) {
+        const doc = frame.contentDocument;
+        const viewer = doc.querySelector("#mnote-office-viewer");
+        const highlighted = doc.querySelector('[data-mnote-office-evidence-target="true"]');
+        result.iframe = {
+          readyState: doc.readyState,
+          status: doc.documentElement.getAttribute("data-mnote-office-preview-status") || "",
+          applied: doc.documentElement.getAttribute("data-mnote-office-evidence-applied") || "",
+          bodyEvidenceText: doc.body?.dataset?.evidenceText || "",
+          textHasQuery: (viewer?.textContent || "").includes("三乙基硅"),
+          scrollY: frame.contentWindow?.scrollY || 0,
+          highlightedText: highlighted?.textContent?.slice(0, 160) || "",
+          highlightedTop: highlighted ? Math.round(highlighted.getBoundingClientRect().top) : null,
+        };
+      }
+      return result;
+    }, EXPECTED_RESOURCE);
+    await page.screenshot({ path: OPEN_SCREENSHOT_PATH, fullPage: true });
+
+    assert.equal(state.panelVisible, true, `搜索结果未打开资源标签: ${JSON.stringify(state, null, 2)}`);
+    assert.equal(state.panelResourcePath, EXPECTED_RESOURCE, `资源路径不匹配: ${JSON.stringify(state, null, 2)}`);
+    assert.equal(state.panelBlockId, EXPECTED_BLOCK_ID, `资源标签 blockId 不匹配: ${JSON.stringify(state, null, 2)}`);
+    assert.equal(state.iframe?.applied, "true", `Office preview 未应用 evidence locator: ${JSON.stringify(state, null, 2)}`);
+    assert(String(state.iframe?.highlightedText || "").includes(EXPECTED_HIGHLIGHT), `Office preview 未高亮 ${EXPECTED_HIGHLIGHT}: ${JSON.stringify(state, null, 2)}`);
+
+    const result = {
+      ok: true,
+      baseUrl: BASE_URL,
+      rootUri: ROOT_URI,
+      workspaceId: WORKSPACE_ID,
+      query: QUERY,
+      expectedResource: EXPECTED_RESOURCE,
+      expectedBlockId: EXPECTED_BLOCK_ID,
+      apiFirst: {
+        snippet: apiFirst.snippet,
+        blockId: apiFirst.locator?.blockId,
+        citationUrl: apiFirst.citationUrl,
+      },
+      state,
+      pageErrors,
+      consoleErrors,
+      screenshots: {
+        search: SEARCH_SCREENSHOT_PATH,
+        open: OPEN_SCREENSHOT_PATH,
+      },
+    };
+    fs.writeFileSync(RESULT_PATH, `${JSON.stringify(result, null, 2)}\n`, "utf8");
+    console.log(JSON.stringify(result, null, 2));
+  } finally {
+    await browser.close().catch(() => undefined);
+  }
+}
+
+main().catch((error) => {
+  fs.mkdirSync(OUTPUT_DIR, { recursive: true });
+  fs.writeFileSync(
+    path.join(OUTPUT_DIR, "failure.json"),
+    `${JSON.stringify({ ok: false, error: error.stack || error.message || String(error) }, null, 2)}\n`,
+    "utf8",
+  );
+  console.error(error.stack || error.message || String(error));
+  process.exit(1);
+});

scripts/task541-knowledge-rag-search-panel-office-reposition-smoke.js

@@ -0,0 +1,233 @@
+#!/usr/bin/env node
+"use strict";
+
+const assert = require("node:assert");
+const fs = require("node:fs");
+const path = require("node:path");
+const { chromium } = require("playwright");
+
+const BASE_URL = (process.env.MNOTE_WEB_SMOKE_BASE_URL || "http://127.0.0.1:3000").replace(/\/+$/, "");
+const QUERY = process.env.MNOTE_KNOWLEDGE_RAG_REPOSITION_QUERY || "三甲基硅";
+const EXPECTED_RESOURCE = process.env.MNOTE_KNOWLEDGE_RAG_EXPECTED_RESOURCE
+  || "有机合成中的保护基/[OCR]_有机合成中的保护基-酚羰基羧基巯基的保护_20250201_1908.layered_删减-2025-02-04 18-59-42.docx";
+const OUTPUT_DIR = path.join(process.cwd(), "tmp", "task541-knowledge-rag-search-panel-office-reposition-smoke");
+const RESULT_PATH = path.join(OUTPUT_DIR, "result.json");
+const SEARCH_SCREENSHOT_PATH = path.join(OUTPUT_DIR, "search-restored.png");
+const OPEN_SCREENSHOT_PATH = path.join(OUTPUT_DIR, "office-reposition.png");
+const UI_TIMEOUT_MS = Number(process.env.MNOTE_SMOKE_UI_TIMEOUT_MS || 30_000);
+const CHROMIUM_EXECUTABLE_PATH = process.env.PLAYWRIGHT_CHROMIUM_EXECUTABLE_PATH
+  || ["/usr/bin/google-chrome-stable", "/usr/bin/google-chrome", "/snap/bin/chromium"]
+    .find((candidate) => fs.existsSync(candidate));
+
+async function openKnowledgeSearch(page, query) {
+  await page.click('[data-mnote-action="open-search-modal"]');
+  await page.waitForSelector('[data-testid="wolai-search-input"]', { timeout: UI_TIMEOUT_MS });
+  await page.evaluate(() => {
+    const button = document.querySelector('[data-search-switch="knowledge"]');
+    if (button instanceof HTMLElement) {
+      button.setAttribute("aria-checked", "true");
+      button.classList.add("is-on");
+    }
+  });
+  await page.fill('[data-testid="wolai-search-input"]', query);
+  await page.evaluate(() => {
+    document.querySelector('[data-testid="wolai-search-input"]')
+      ?.dispatchEvent(new Event("input", { bubbles: true }));
+  });
+}
+
+async function waitForOfficeRows(page) {
+  await page.waitForFunction(
+    ({ query, expectedResource }) => {
+      const rows = Array.from(document.querySelectorAll('[data-testid="wolai-search-result-row"]'));
+      return rows.some((row) => {
+        const locator = JSON.parse(row.getAttribute("data-evidence-locator") || "null");
+        const path = String(locator?.resourcePath || locator?.resource_path || "");
+        return path === expectedResource && row.textContent.includes(query);
+      });
+    },
+    { query: QUERY, expectedResource: EXPECTED_RESOURCE },
+    { timeout: UI_TIMEOUT_MS },
+  );
+}
+
+async function pickOfficeResult(page, excludeBlockId = "") {
+  return page.evaluate(({ expectedResource, excludeBlockId }) => {
+    const rows = Array.from(document.querySelectorAll('[data-testid="wolai-search-result-row"]'));
+    for (const row of rows) {
+      const locator = JSON.parse(row.getAttribute("data-evidence-locator") || "null");
+      const resourcePath = String(locator?.resourcePath || locator?.resource_path || "");
+      const blockId = String(locator?.blockId || locator?.block_id || "");
+      if (resourcePath === expectedResource && blockId && blockId !== excludeBlockId) {
+        return {
+          index: Number(row.getAttribute("data-search-result-index") || -1),
+          blockId,
+          text: row.textContent,
+        };
+      }
+    }
+    return null;
+  }, { expectedResource: EXPECTED_RESOURCE, excludeBlockId });
+}
+
+async function clickSearchResultByIndex(page, index) {
+  await page.locator(`[data-testid="wolai-search-result-row"][data-search-result-index="${index}"]`).click();
+}
+
+async function waitForOfficeLocator(page, expectedBlockId) {
+  await page.waitForFunction(
+    ({ expectedResource, expectedBlockId }) => {
+      const panel = document.querySelector(`[data-mnote-resource-tab-panel][data-resource-path="${CSS.escape(expectedResource)}"]`);
+      return panel && panel.getAttribute("data-mnote-evidence-block-id") === expectedBlockId;
+    },
+    { expectedResource: EXPECTED_RESOURCE, expectedBlockId },
+    { timeout: UI_TIMEOUT_MS },
+  );
+  await page.waitForTimeout(1_500);
+}
+
+async function readOfficeState(page) {
+  return page.evaluate((expectedResource) => {
+    const panel = document.querySelector(`[data-mnote-resource-tab-panel][data-resource-path="${CSS.escape(expectedResource)}"]`);
+    const frame = panel?.querySelector("iframe.mnote-resource-tab-frame");
+    const result = {
+      searchOpen: !document.querySelector('[data-testid="wolai-search-modal"]')?.hidden,
+      panelBlockId: panel?.getAttribute("data-mnote-evidence-block-id") || "",
+      frameSrc: frame?.getAttribute("src") || "",
+      loadCount: window.__mnoteTask541LoadCount || 0,
+    };
+    if (frame?.contentDocument) {
+      const doc = frame.contentDocument;
+      const target = doc.querySelector('[data-mnote-office-evidence-target="true"]');
+      const marker = doc.querySelector('[data-mnote-office-evidence-marker="true"]');
+      result.iframe = {
+        status: doc.documentElement.getAttribute("data-mnote-office-preview-status") || "",
+        applied: doc.documentElement.getAttribute("data-mnote-office-evidence-applied") || "",
+        highlightedText: target?.textContent?.slice(0, 200) || "",
+        markerMode: marker?.getAttribute("data-mnote-office-evidence-marker-mode") || "",
+        markerTargetText: marker?.getAttribute("data-mnote-office-evidence-target-text") || "",
+        markerHeight: marker ? Math.round(marker.getBoundingClientRect().height) : 0,
+      };
+    }
+    return result;
+  }, EXPECTED_RESOURCE);
+}
+
+async function main() {
+  fs.mkdirSync(OUTPUT_DIR, { recursive: true });
+  const browser = await chromium.launch({ headless: true, executablePath: CHROMIUM_EXECUTABLE_PATH });
+  const context = await browser.newContext({ viewport: { width: 1360, height: 900 } });
+  try {
+    const auth = await context.request.post(`${BASE_URL}/api/auth`, {
+      data: {
+        action: "auth:signIn",
+        args: {
+          provider: "password",
+          params: {
+            email: "mnote.e2e@example.com",
+            password: process.env.MNOTE_E2E_PASSWORD || "MnoteE2E123!",
+            flow: "signIn",
+          },
+        },
+      },
+    });
+    assert(auth.ok(), `登录失败: ${auth.status()} ${await auth.text()}`);
+
+    const page = await context.newPage();
+    const pageErrors = [];
+    const consoleErrors = [];
+    page.on("pageerror", (error) => pageErrors.push(String(error?.stack || error?.message || error)));
+    page.on("console", (message) => {
+      if (["error", "warning"].includes(message.type())) consoleErrors.push(`${message.type()}: ${message.text()}`);
+    });
+
+    await page.goto(`${BASE_URL}/`, { waitUntil: "domcontentloaded", timeout: UI_TIMEOUT_MS });
+    await page.evaluate(() => {
+      window.localStorage?.setItem("mnote.search.collapseSourcesDefault.v1", "0");
+    });
+    await page.waitForSelector('[data-mnote-action="open-search-modal"]', { timeout: UI_TIMEOUT_MS });
+    await openKnowledgeSearch(page, QUERY);
+    await waitForOfficeRows(page);
+    const beforeClose = await page.evaluate(() => ({
+      query: document.querySelector('[data-testid="wolai-search-input"]')?.value || "",
+      count: document.querySelectorAll('[data-testid="wolai-search-result-row"]').length,
+      meta: document.querySelector('[data-testid="wolai-search-result-meta"]')?.textContent || "",
+    }));
+    await page.click('[data-testid="wolai-search-close"]');
+    await page.click('[data-mnote-action="open-search-modal"]');
+    await page.waitForSelector('[data-testid="wolai-search-input"]', { timeout: UI_TIMEOUT_MS });
+    const afterReopen = await page.evaluate(() => ({
+      query: document.querySelector('[data-testid="wolai-search-input"]')?.value || "",
+      count: document.querySelectorAll('[data-testid="wolai-search-result-row"]').length,
+      meta: document.querySelector('[data-testid="wolai-search-result-meta"]')?.textContent || "",
+    }));
+    await page.screenshot({ path: SEARCH_SCREENSHOT_PATH, fullPage: true });
+    assert.equal(afterReopen.query, QUERY, `搜索面板未恢复 query: ${JSON.stringify({ beforeClose, afterReopen })}`);
+    assert(afterReopen.count >= beforeClose.count, `搜索面板重新打开后结果丢失: ${JSON.stringify({ beforeClose, afterReopen })}`);
+
+    const first = await pickOfficeResult(page);
+    assert(first, `没有找到可打开的 Office 搜索结果`);
+    await clickSearchResultByIndex(page, first.index);
+    await waitForOfficeLocator(page, first.blockId);
+    const firstState = await readOfficeState(page);
+    assert.equal(firstState.searchOpen, false, `点击搜索结果后搜索面板应关闭但保留状态: ${JSON.stringify(firstState, null, 2)}`);
+    await page.evaluate((expectedResource) => {
+      const panel = document.querySelector(`[data-mnote-resource-tab-panel][data-resource-path="${CSS.escape(expectedResource)}"]`);
+      const frame = panel?.querySelector("iframe.mnote-resource-tab-frame");
+      window.__mnoteTask541LoadCount = 0;
+      if (frame) frame.addEventListener("load", () => {
+        window.__mnoteTask541LoadCount = (window.__mnoteTask541LoadCount || 0) + 1;
+      });
+    }, EXPECTED_RESOURCE);
+
+    await page.click('[data-mnote-action="open-search-modal"]');
+    await waitForOfficeRows(page);
+    const second = await pickOfficeResult(page, first.blockId);
+    assert(second, `没有找到同一文件的第二个不同 block 搜索结果: ${JSON.stringify(first)}`);
+    await clickSearchResultByIndex(page, second.index);
+    await waitForOfficeLocator(page, second.blockId);
+    const secondState = await readOfficeState(page);
+    await page.screenshot({ path: OPEN_SCREENSHOT_PATH, fullPage: true });
+
+    assert.equal(secondState.loadCount, 0, `同一 Office 文件重定位触发了 iframe reload: ${JSON.stringify({ firstState, secondState }, null, 2)}`);
+    assert.equal(secondState.iframe?.applied, "true", `Office preview 未应用第二次定位: ${JSON.stringify(secondState, null, 2)}`);
+    const highlighted = `${secondState.iframe?.highlightedText || ""} ${secondState.iframe?.markerTargetText || ""}`;
+    assert(highlighted.includes(QUERY), `Office preview 高亮未包含搜索词: ${JSON.stringify(secondState, null, 2)}`);
+    assert((secondState.iframe?.markerTargetText || "").length <= 160, `Office preview range marker 文本过长: ${JSON.stringify(secondState, null, 2)}`);
+    assert((secondState.iframe?.markerHeight || 0) <= 140, `Office preview range marker 框选过高: ${JSON.stringify(secondState, null, 2)}`);
+
+    const result = {
+      ok: true,
+      baseUrl: BASE_URL,
+      query: QUERY,
+      expectedResource: EXPECTED_RESOURCE,
+      beforeClose,
+      afterReopen,
+      first,
+      second,
+      firstState,
+      secondState,
+      pageErrors,
+      consoleErrors,
+      screenshots: {
+        search: SEARCH_SCREENSHOT_PATH,
+        open: OPEN_SCREENSHOT_PATH,
+      },
+    };
+    fs.writeFileSync(RESULT_PATH, `${JSON.stringify(result, null, 2)}\n`, "utf8");
+    console.log(JSON.stringify(result, null, 2));
+  } finally {
+    await browser.close().catch(() => undefined);
+  }
+}
+
+main().catch((error) => {
+  fs.mkdirSync(OUTPUT_DIR, { recursive: true });
+  fs.writeFileSync(
+    path.join(OUTPUT_DIR, "failure.json"),
+    `${JSON.stringify({ ok: false, error: error.stack || error.message || String(error) }, null, 2)}\n`,
+    "utf8",
+  );
+  console.error(error.stack || error.message || String(error));
+  process.exit(1);
+});

scripts/task542-knowledge-rag-search-grouping-and-short-query-smoke.js

@@ -0,0 +1,231 @@
+#!/usr/bin/env node
+"use strict";
+
+const assert = require("node:assert");
+const fs = require("node:fs");
+const path = require("node:path");
+const { chromium } = require("playwright");
+
+const BASE_URL = (process.env.MNOTE_WEB_SMOKE_BASE_URL || "http://127.0.0.1:3000").replace(/\/+$/, "");
+const ROOT_URI = process.env.MNOTE_KNOWLEDGE_RAG_ROOT_URI || "file:///mnt/Data1T/Mnote_data/users/mnote-e2e/workspaces/my-space";
+const WORKSPACE_ID = process.env.MNOTE_KNOWLEDGE_RAG_WORKSPACE_ID || "local-ws:mnote-e2e:my-space";
+const GROUP_QUERY = process.env.MNOTE_KNOWLEDGE_RAG_GROUP_QUERY || "三甲基硅";
+const SHORT_QUERY = process.env.MNOTE_KNOWLEDGE_RAG_SHORT_QUERY || "吗啉";
+const TOO_SHORT_QUERY = process.env.MNOTE_KNOWLEDGE_RAG_TOO_SHORT_QUERY || "吗";
+const OUTPUT_DIR = path.join(process.cwd(), "tmp", "task542-knowledge-rag-search-grouping-and-short-query-smoke");
+const RESULT_PATH = path.join(OUTPUT_DIR, "result.json");
+const GROUP_SCREENSHOT_PATH = path.join(OUTPUT_DIR, "grouped-collapsed.png");
+const EXPANDED_SCREENSHOT_PATH = path.join(OUTPUT_DIR, "grouped-expanded.png");
+const UI_TIMEOUT_MS = Number(process.env.MNOTE_SMOKE_UI_TIMEOUT_MS || 30_000);
+const CHROMIUM_EXECUTABLE_PATH = process.env.PLAYWRIGHT_CHROMIUM_EXECUTABLE_PATH
+  || ["/usr/bin/google-chrome-stable", "/usr/bin/google-chrome", "/snap/bin/chromium"]
+    .find((candidate) => fs.existsSync(candidate));
+
+async function main() {
+  fs.mkdirSync(OUTPUT_DIR, { recursive: true });
+  const browser = await chromium.launch({ headless: true, executablePath: CHROMIUM_EXECUTABLE_PATH });
+  const context = await browser.newContext({ viewport: { width: 1360, height: 900 } });
+  try {
+    const auth = await context.request.post(`${BASE_URL}/api/auth`, {
+      data: {
+        action: "auth:signIn",
+        args: {
+          provider: "password",
+          params: {
+            email: "mnote.e2e@example.com",
+            password: process.env.MNOTE_E2E_PASSWORD || "MnoteE2E123!",
+            flow: "signIn",
+          },
+        },
+      },
+    });
+    assert(auth.ok(), `登录失败: ${auth.status()} ${await auth.text()}`);
+
+    const localSearch = await context.request.post(`${BASE_URL}/api/search/documents`, {
+      data: {
+        workspaceId: WORKSPACE_ID,
+        sourceKind: "local_folder",
+        rootUri: ROOT_URI,
+        query: SHORT_QUERY,
+        limit: 30,
+        filters: {
+          includeOcr: false,
+          titleOnly: false,
+          exact: false,
+          onlyCurrentPage: false,
+          timeRange: "any",
+          timeField: "updated",
+        },
+      },
+    });
+    assert(localSearch.ok(), `本地搜索失败: ${localSearch.status()} ${await localSearch.text()}`);
+    const localPayload = await localSearch.json();
+    assert.equal(localPayload.results?.length || 0, 0, `本地文档搜索不应把两字 CJK query 匹配到无关文档: ${JSON.stringify(localPayload.results?.slice(0, 3), null, 2)}`);
+
+    const knowledgeSearch = await context.request.post(`${BASE_URL}/api/knowledge-rag/search`, {
+      data: {
+        workspaceId: WORKSPACE_ID,
+        rootUri: ROOT_URI,
+        query: SHORT_QUERY,
+        mode: "mix",
+        topK: 12,
+        chunkTopK: 24,
+        includeChunkContent: true,
+      },
+    });
+    assert(knowledgeSearch.ok(), `资料库短 query 检索失败: ${knowledgeSearch.status()} ${await knowledgeSearch.text()}`);
+    const knowledgePayload = await knowledgeSearch.json();
+    assert((knowledgePayload.results?.length || 0) >= 8, `资料库 2 字 query 应返回段落去重后的 LightRAG 检索命中: ${JSON.stringify(knowledgePayload, null, 2).slice(0, 2000)}`);
+    assert(String(knowledgePayload.results?.[0]?.snippet || "").includes(SHORT_QUERY), `资料库结果未包含 ${SHORT_QUERY}: ${JSON.stringify(knowledgePayload.results?.[0], null, 2)}`);
+    assert(knowledgePayload.references?.some((reference) => reference.matchSource === "lightrag_search"), `2 字 query 应走 LightRAG search provider: ${JSON.stringify(knowledgePayload.references?.slice(0, 3), null, 2)}`);
+
+    const tooShortSearch = await context.request.post(`${BASE_URL}/api/knowledge-rag/search`, {
+      data: {
+        workspaceId: WORKSPACE_ID,
+        rootUri: ROOT_URI,
+        query: TOO_SHORT_QUERY,
+        mode: "mix",
+        topK: 12,
+        chunkTopK: 24,
+        includeChunkContent: true,
+      },
+    });
+    assert.equal(tooShortSearch.status(), 400, `资料库 1 字 query 应返回 400: ${tooShortSearch.status()} ${await tooShortSearch.text()}`);
+    const tooShortPayload = await tooShortSearch.json();
+    assert.equal(tooShortPayload.code, "knowledge_rag_search_query_too_short", `1 字 query 错误码不符合预期: ${JSON.stringify(tooShortPayload)}`);
+    assert(String(tooShortPayload.message || "").includes("至少 2 个字"), `1 字 query 提示不符合预期: ${JSON.stringify(tooShortPayload)}`);
+
+    const page = await context.newPage();
+    const pageErrors = [];
+    const consoleErrors = [];
+    page.on("pageerror", (error) => pageErrors.push(String(error?.stack || error?.message || error)));
+    page.on("console", (message) => {
+      if (["error", "warning"].includes(message.type())) consoleErrors.push(`${message.type()}: ${message.text()}`);
+    });
+    await page.goto(`${BASE_URL}/`, { waitUntil: "domcontentloaded", timeout: UI_TIMEOUT_MS });
+    await page.evaluate(() => {
+      window.localStorage?.setItem("mnote.search.collapseSourcesDefault.v1", "1");
+    });
+    await page.waitForSelector('[data-mnote-action="open-search-modal"]', { timeout: UI_TIMEOUT_MS });
+    await page.click('[data-mnote-action="open-search-modal"]');
+    await page.waitForSelector('[data-testid="wolai-search-input"]', { timeout: UI_TIMEOUT_MS });
+    await page.evaluate(() => {
+      const knowledge = document.querySelector('[data-search-switch="knowledge"]');
+      if (knowledge instanceof HTMLElement) {
+        knowledge.setAttribute("aria-checked", "true");
+        knowledge.classList.add("is-on");
+      }
+      const collapse = document.querySelector('[data-search-switch="collapseSource"]');
+      if (collapse instanceof HTMLElement) {
+        collapse.setAttribute("aria-checked", "true");
+        collapse.classList.add("is-on");
+      }
+    });
+    await page.fill('[data-testid="wolai-search-input"]', TOO_SHORT_QUERY);
+    await page.evaluate(() => {
+      document.querySelector('[data-testid="wolai-search-input"]')
+        ?.dispatchEvent(new Event("input", { bubbles: true }));
+    });
+    await page.waitForSelector('[data-search-empty="true"]', { timeout: UI_TIMEOUT_MS });
+    const tooShortUi = await page.evaluate(() => ({
+      meta: document.querySelector('[data-testid="wolai-search-result-meta"]')?.textContent || "",
+      empty: document.querySelector('[data-search-empty="true"]')?.textContent || "",
+      rows: document.querySelectorAll('[data-testid="wolai-search-result-row"]').length,
+    }));
+    assert(tooShortUi.empty.includes("至少 2 个字"), `1 字 query UI 未提示至少 2 个字: ${JSON.stringify(tooShortUi)}`);
+    assert.equal(tooShortUi.rows, 0, `1 字 query 不应发起并渲染结果: ${JSON.stringify(tooShortUi)}`);
+
+    await page.fill('[data-testid="wolai-search-input"]', GROUP_QUERY);
+    await page.evaluate(() => {
+      document.querySelector('[data-testid="wolai-search-input"]')
+        ?.dispatchEvent(new Event("input", { bubbles: true }));
+    });
+    await page.waitForSelector(".wolai-search-source-group", { timeout: UI_TIMEOUT_MS });
+    const collapsed = await page.evaluate(() => ({
+      switchOn: document.querySelector('[data-search-switch="collapseSource"]')?.getAttribute("aria-checked") === "true",
+      groups: document.querySelectorAll(".wolai-search-source-group").length,
+      visibleRows: Array.from(document.querySelectorAll('[data-testid="wolai-search-result-row"]'))
+        .filter((row) => row.offsetParent !== null).length,
+      meta: document.querySelector('[data-testid="wolai-search-result-meta"]')?.textContent || "",
+      firstHeader: document.querySelector(".wolai-search-source-header")?.textContent || "",
+    }));
+    await page.screenshot({ path: GROUP_SCREENSHOT_PATH, fullPage: true });
+    assert.equal(collapsed.switchOn, true, `折叠同来源开关未默认打开: ${JSON.stringify(collapsed)}`);
+    assert(collapsed.groups >= 1, `未按来源分组: ${JSON.stringify(collapsed)}`);
+    assert.equal(collapsed.visibleRows, 0, `默认折叠时不应展示重复来源子结果: ${JSON.stringify(collapsed)}`);
+    assert(
+      !String(collapsed.firstHeader || "").includes("有机合成中的保护基/[OCR]"),
+      `来源组标题不应再在标题下重复显示完整来源路径: ${JSON.stringify(collapsed)}`
+    );
+
+    await page.click(".wolai-search-source-header");
+    await page.waitForFunction(() => {
+      return Array.from(document.querySelectorAll('[data-testid="wolai-search-result-row"]'))
+        .some((row) => row.offsetParent !== null);
+    }, null, { timeout: UI_TIMEOUT_MS });
+    const expanded = await page.evaluate(() => ({
+      expanded: document.querySelector(".wolai-search-source-header")?.getAttribute("aria-expanded") === "true",
+      visibleRows: Array.from(document.querySelectorAll('[data-testid="wolai-search-result-row"]'))
+        .filter((row) => row.offsetParent !== null).length,
+      firstRows: Array.from(document.querySelectorAll('[data-testid="wolai-search-result-row"]'))
+        .slice(0, 3)
+        .map((row) => row.textContent),
+    }));
+    await page.screenshot({ path: EXPANDED_SCREENSHOT_PATH, fullPage: true });
+    assert.equal(expanded.expanded, true, `来源组未展开: ${JSON.stringify(expanded)}`);
+    assert(expanded.visibleRows > 0, `展开后未展示来源下结果: ${JSON.stringify(expanded)}`);
+    assert(
+      expanded.firstRows.every((text) => !String(text || "").includes("有机合成中的保护基/[OCR]")),
+      `折叠同来源展开后，子结果不应重复显示完整来源路径: ${JSON.stringify(expanded.firstRows)}`
+    );
+    assert(
+      expanded.firstRows.every((text) => !String(text || "").trim().endsWith("docx")),
+      `折叠同来源展开后，子结果不应重复显示来源类型: ${JSON.stringify(expanded.firstRows)}`
+    );
+    assert(
+      expanded.firstRows.every((text) => !/(?:<\/?e(?:q(?:uation)?)?\b|<\/?drawing\b|format=["']?latex|\blatex\b)/i.test(String(text || ""))),
+      `资料库搜索结果仍暴露原始公式/绘图标记: ${JSON.stringify(expanded.firstRows)}`
+    );
+
+    const result = {
+      ok: true,
+      baseUrl: BASE_URL,
+      rootUri: ROOT_URI,
+      workspaceId: WORKSPACE_ID,
+      shortQuery: SHORT_QUERY,
+      tooShortQuery: TOO_SHORT_QUERY,
+      groupQuery: GROUP_QUERY,
+      localCount: localPayload.results?.length || 0,
+      knowledgeCount: knowledgePayload.results?.length || 0,
+      knowledgeFirst: {
+        title: knowledgePayload.results?.[0]?.title,
+        snippet: knowledgePayload.results?.[0]?.snippet,
+        blockId: knowledgePayload.results?.[0]?.locator?.blockId,
+      },
+      tooShortUi,
+      collapsed,
+      expanded,
+      pageErrors,
+      consoleErrors,
+      screenshots: {
+        collapsed: GROUP_SCREENSHOT_PATH,
+        expanded: EXPANDED_SCREENSHOT_PATH,
+      },
+    };
+    fs.writeFileSync(RESULT_PATH, `${JSON.stringify(result, null, 2)}\n`, "utf8");
+    console.log(JSON.stringify(result, null, 2));
+  } finally {
+    await browser.close().catch(() => undefined);
+  }
+}
+
+main().catch((error) => {
+  fs.mkdirSync(OUTPUT_DIR, { recursive: true });
+  fs.writeFileSync(
+    path.join(OUTPUT_DIR, "failure.json"),
+    `${JSON.stringify({ ok: false, error: error.stack || error.message || String(error) }, null, 2)}\n`,
+    "utf8",
+  );
+  console.error(error.stack || error.message || String(error));
+  process.exit(1);
+});

scripts/task543-knowledge-rag-pyrrolidine-top5-locator-smoke.js

@@ -0,0 +1,253 @@
+#!/usr/bin/env node
+"use strict";
+
+const assert = require("node:assert");
+const fs = require("node:fs");
+const path = require("node:path");
+const { chromium } = require("playwright");
+
+const BASE_URL = (process.env.MNOTE_WEB_SMOKE_BASE_URL || "http://127.0.0.1:3000").replace(/\/+$/, "");
+const ROOT_URI = process.env.MNOTE_KNOWLEDGE_RAG_ROOT_URI || "file:///mnt/Data1T/Mnote_data/users/mnote-e2e/workspaces/my-space";
+const WORKSPACE_ID = process.env.MNOTE_KNOWLEDGE_RAG_WORKSPACE_ID || "local-ws:mnote-e2e:my-space";
+const QUERY = process.env.MNOTE_KNOWLEDGE_RAG_PYRROLIDINE_QUERY || "吡咯烷";
+const EXPECTED_RESOURCE = process.env.MNOTE_KNOWLEDGE_RAG_EXPECTED_RESOURCE
+  || "有机合成中的保护基/[OCR]_有机合成中的保护基-酚羰基羧基巯基的保护_20250201_1908.layered_删减-2025-02-04 18-59-42.docx";
+const TOP_N = Number(process.env.MNOTE_KNOWLEDGE_RAG_TOP_N || 5);
+const OUTPUT_DIR = path.join(process.cwd(), "tmp", "task543-knowledge-rag-pyrrolidine-top5-locator-smoke");
+const RESULT_PATH = path.join(OUTPUT_DIR, "result.json");
+const UI_TIMEOUT_MS = Number(process.env.MNOTE_SMOKE_UI_TIMEOUT_MS || 30_000);
+const CHROMIUM_EXECUTABLE_PATH = process.env.PLAYWRIGHT_CHROMIUM_EXECUTABLE_PATH
+  || ["/usr/bin/google-chrome-stable", "/usr/bin/google-chrome", "/snap/bin/chromium"]
+    .find((candidate) => fs.existsSync(candidate));
+
+function normalizeText(value) {
+  return String(value || "")
+    .replace(/<[^>]+>/g, " ")
+    .replace(/[#*_`~>\[\](){}]+/g, " ")
+    .replace(/[\u200B-\u200D\uFEFF]/g, "")
+    .replace(/\s+/g, " ")
+    .trim();
+}
+
+function compactText(value) {
+  return normalizeText(value).replace(/[0-9０-９]+/g, "").replace(/[\s\p{P}\p{S}]+/gu, "");
+}
+
+function longestCommonSubstringLength(left, right) {
+  const a = compactText(left);
+  const b = compactText(right);
+  if (!a || !b) return 0;
+  const shorter = a.length <= b.length ? a : b;
+  const longer = a.length <= b.length ? b : a;
+  for (let len = Math.min(80, shorter.length); len >= 6; len -= 1) {
+    for (let start = 0; start + len <= shorter.length; start += 1) {
+      if (longer.includes(shorter.slice(start, start + len))) return len;
+    }
+  }
+  return 0;
+}
+
+async function signIn(context) {
+  const auth = await context.request.post(`${BASE_URL}/api/auth`, {
+    data: {
+      action: "auth:signIn",
+      args: {
+        provider: "password",
+        params: {
+          email: "mnote.e2e@example.com",
+          password: process.env.MNOTE_E2E_PASSWORD || "MnoteE2E123!",
+          flow: "signIn",
+        },
+      },
+    },
+  });
+  assert(auth.ok(), `登录失败: ${auth.status()} ${await auth.text()}`);
+}
+
+async function apiSearch(context) {
+  const response = await context.request.post(`${BASE_URL}/api/knowledge-rag/search`, {
+    data: {
+      rootUri: ROOT_URI,
+      workspaceId: WORKSPACE_ID,
+      query: QUERY,
+      mode: "mix",
+      topK: 20,
+      chunkTopK: 20,
+      includeChunkContent: true,
+      sourcePaths: [EXPECTED_RESOURCE],
+    },
+  });
+  assert(response.ok(), `knowledge-rag search 失败: ${response.status()} ${await response.text()}`);
+  const payload = await response.json();
+  const results = Array.isArray(payload.results) ? payload.results : [];
+  assert(results.length >= TOP_N, `资料库结果不足 ${TOP_N} 条: ${JSON.stringify(payload, null, 2).slice(0, 3000)}`);
+  const blockIds = results.map((item) => item?.locator?.blockId).filter(Boolean);
+  assert.equal(new Set(blockIds).size, blockIds.length, `搜索结果仍有同段落重复: ${JSON.stringify(blockIds)}`);
+  return results.slice(0, TOP_N);
+}
+
+async function openSearchPanel(page, navigate = false) {
+  if (navigate) {
+    await page.goto(`${BASE_URL}/`, { waitUntil: "domcontentloaded", timeout: UI_TIMEOUT_MS });
+    await page.evaluate(() => {
+      window.localStorage?.setItem("mnote.search.collapseSourcesDefault.v1", "0");
+    });
+  }
+  const inputVisible = await page.locator('[data-testid="wolai-search-input"]').isVisible().catch(() => false);
+  if (!inputVisible) {
+    await page.waitForSelector('[data-mnote-action="open-search-modal"]', { timeout: UI_TIMEOUT_MS });
+    await page.click('[data-mnote-action="open-search-modal"]');
+  }
+  await page.waitForSelector('[data-testid="wolai-search-input"]', { timeout: UI_TIMEOUT_MS });
+  await page.fill('[data-testid="wolai-search-input"]', QUERY);
+  await page.evaluate(() => {
+    const button = document.querySelector('[data-search-switch="knowledge"]');
+    if (button instanceof HTMLElement) {
+      button.setAttribute("aria-checked", "true");
+      button.classList.add("is-on");
+    }
+    document.querySelector('[data-testid="wolai-search-input"]')
+      ?.dispatchEvent(new Event("input", { bubbles: true }));
+  });
+  await page.waitForFunction(
+    (topN) => document.querySelectorAll('[data-testid="wolai-search-result-row"]').length >= topN,
+    TOP_N,
+    { timeout: UI_TIMEOUT_MS },
+  );
+}
+
+async function clickAndAuditRow(page, index, apiResult) {
+  const rows = await page.$$('[data-testid="wolai-search-result-row"]');
+  assert(rows[index], `缺少第 ${index + 1} 条搜索结果`);
+  const rowText = await rows[index].evaluate((node) => node.textContent || "");
+  assert(
+    !/(?:<\/?equation\b|format=["']?latex|<\/?drawing\b|\blatex\b)/i.test(rowText),
+    `第 ${index + 1} 条搜索结果仍暴露原始公式/绘图标记: ${rowText}`,
+  );
+  const locator = await rows[index].evaluate((node) => {
+    const raw = node.getAttribute("data-evidence-locator") || "";
+    try {
+      return raw ? JSON.parse(raw) : null;
+    } catch (_) {
+      return null;
+    }
+  });
+  assert(locator?.blockId, `第 ${index + 1} 条缺少 locator blockId: ${rowText}`);
+  assert.equal(locator.blockId, apiResult?.locator?.blockId, `第 ${index + 1} 条 UI/API blockId 不一致`);
+
+  await rows[index].click();
+  await page.waitForFunction(
+    ({ expectedResource, expectedBlockId }) => {
+      const panel = document.querySelector(`[data-mnote-resource-tab-panel][data-resource-path="${CSS.escape(expectedResource)}"]`);
+      return panel && !panel.hidden && panel.getAttribute("data-mnote-evidence-block-id") === expectedBlockId;
+    },
+    { expectedResource: EXPECTED_RESOURCE, expectedBlockId: locator.blockId },
+    { timeout: UI_TIMEOUT_MS },
+  );
+  await page.waitForTimeout(1200);
+  await page.waitForFunction(
+    (expectedResource) => {
+      const panel = document.querySelector(`[data-mnote-resource-tab-panel][data-resource-path="${CSS.escape(expectedResource)}"]`);
+      const frame = panel?.querySelector("iframe.mnote-resource-tab-frame");
+      const doc = frame?.contentDocument;
+      return doc?.documentElement?.getAttribute("data-mnote-office-evidence-applied") === "true";
+    },
+    EXPECTED_RESOURCE,
+    { timeout: UI_TIMEOUT_MS },
+  );
+
+  const state = await page.evaluate((expectedResource) => {
+    const panel = document.querySelector(`[data-mnote-resource-tab-panel][data-resource-path="${CSS.escape(expectedResource)}"]`);
+    const frame = panel?.querySelector("iframe.mnote-resource-tab-frame");
+    const doc = frame?.contentDocument;
+    const highlighted = doc?.querySelector('[data-mnote-office-evidence-target="true"]');
+    return {
+      panelBlockId: panel?.getAttribute("data-mnote-evidence-block-id") || "",
+      panelEvidenceText: panel?.getAttribute("data-mnote-evidence-text") || "",
+      iframeApplied: doc?.documentElement?.getAttribute("data-mnote-office-evidence-applied") || "",
+      highlightedText: highlighted?.textContent || "",
+      highlightedTag: highlighted?.tagName || "",
+      highlightedTop: highlighted ? Math.round(highlighted.getBoundingClientRect().top) : null,
+    };
+  }, EXPECTED_RESOURCE);
+
+  const evidenceText = locator?.openAction?.params?.evidenceText || apiResult?.locator?.openAction?.params?.evidenceText || apiResult?.quote || rowText;
+  const overlap = longestCommonSubstringLength(state.highlightedText, evidenceText);
+  assert.equal(state.panelBlockId, locator.blockId, `第 ${index + 1} 条 panel blockId 不一致: ${JSON.stringify(state, null, 2)}`);
+  assert(state.highlightedText.includes(QUERY), `第 ${index + 1} 条定位高亮未包含 query: ${JSON.stringify(state, null, 2)}`);
+  assert(
+    normalizeText(state.highlightedText).length >= 24 || state.highlightedTag === "P",
+    `第 ${index + 1} 条仍是短词高亮，不是段落级高亮: ${JSON.stringify(state, null, 2)}`,
+  );
+  assert(
+    overlap >= 10 || (
+      state.highlightedTag === "P"
+      && compactText(state.highlightedText).includes(compactText(QUERY))
+      && compactText(state.highlightedText).length >= 6
+    ),
+    `第 ${index + 1} 条搜索结果与实际定位上下文不一致: ${JSON.stringify({ rowText, evidenceText, overlap, state }, null, 2)}`,
+  );
+
+  return {
+    index: index + 1,
+    rowText: normalizeText(rowText).slice(0, 220),
+    apiBlockId: apiResult?.locator?.blockId || "",
+    uiBlockId: locator.blockId,
+    panelBlockId: state.panelBlockId,
+    highlightedText: normalizeText(state.highlightedText).slice(0, 220),
+    highlightedTag: state.highlightedTag,
+    overlap,
+  };
+}
+
+async function main() {
+  fs.mkdirSync(OUTPUT_DIR, { recursive: true });
+  const browser = await chromium.launch({ headless: true, executablePath: CHROMIUM_EXECUTABLE_PATH });
+  const context = await browser.newContext({ viewport: { width: 1360, height: 900 } });
+  try {
+    await signIn(context);
+    const topResults = await apiSearch(context);
+    const page = await context.newPage();
+    const pageErrors = [];
+    const consoleErrors = [];
+    page.on("pageerror", (error) => pageErrors.push(String(error?.stack || error?.message || error)));
+    page.on("console", (message) => {
+      if (["error", "warning"].includes(message.type())) consoleErrors.push(`${message.type()}: ${message.text()}`);
+    });
+    await openSearchPanel(page, true);
+
+    const audits = [];
+    for (let index = 0; index < TOP_N; index += 1) {
+      await openSearchPanel(page, false);
+      audits.push(await clickAndAuditRow(page, index, topResults[index]));
+    }
+    await page.screenshot({ path: path.join(OUTPUT_DIR, "top5-after-last-click.png"), fullPage: true });
+
+    const result = {
+      ok: true,
+      baseUrl: BASE_URL,
+      rootUri: ROOT_URI,
+      workspaceId: WORKSPACE_ID,
+      query: QUERY,
+      topN: TOP_N,
+      audits,
+      pageErrors,
+      consoleErrors,
+    };
+    fs.writeFileSync(RESULT_PATH, `${JSON.stringify(result, null, 2)}\n`, "utf8");
+    console.log(JSON.stringify(result, null, 2));
+  } finally {
+    await browser.close().catch(() => undefined);
+  }
+}
+
+main().catch((error) => {
+  fs.mkdirSync(OUTPUT_DIR, { recursive: true });
+  fs.writeFileSync(
+    path.join(OUTPUT_DIR, "failure.json"),
+    `${JSON.stringify({ ok: false, error: error.stack || error.message || String(error) }, null, 2)}\n`,
+    "utf8",
+  );
+  console.error(error.stack || error.message || String(error));
+  process.exit(1);
+});