viepilot 2.48.0 → 2.49.0

package/CHANGELOG.md

@@ -7,6 +7,28 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 ## [Unreleased]
 
+## [2.49.0] - 2026-05-17
+
+### Added
+- **ENH-083** `vp-intake` — SharePoint sharing link + xlsx direct download support (Phase 124):
+  - **SharePoint anonymous sharing link** adapter: URLs matching `/:x:/g/` or `/:x:/r/` are
+    resolved via the SharePoint WOPI viewer page — `FileGetUrl` (temp-auth token) is scraped
+    from the HTML and used to download the xlsx file without OAuth credentials
+  - **xlsx parser** using SheetJS (`xlsx` npm package): downloaded xlsx Buffer is parsed to
+    `rows[][]` matching the Graph API response format; falls back to first sheet when the
+    configured sheet name is not found
+  - **`sharing_url` field** in `channels.json` (`excel_m365` type): optional alternative to
+    `workbook_id` for SharePoint "Anyone with the link" files; no credentials required
+  - **Redirect-following HTTP client** (`httpsGetRaw`): browser User-Agent, follows 301/302/307
+    chains needed for SharePoint sharing link flow
+  - **Routing logic in `readExcelM365`**: `sharing_url` present → anonymous WOPI flow;
+    `workbook_id` present → Graph API (OAuth); neither present → `AuthRequiredError` with guide
+  - **BUG_KEYWORDS extended**: `performance`, `slow`, `timeout`, `hang`, `freeze`,
+    `unresponsive`, `chậm`, `treo`, `tắc`, `không phản hồi` — "performance issue" now correctly
+    classifies as BUG instead of UNCLEAR
+  - **19 new tests** (1897 total): keyword expansion (8 tests), `isSharingLink` detection (5
+    tests), routing logic (3 tests), `parseXlsxBuffer` (3 tests)
+
 ## [2.48.0] - 2026-05-17
 
 ### Added

package/lib/intake/adapters/excel-m365.cjs

@@ -1,6 +1,7 @@
 'use strict';
 
 const fs    = require('fs');
+const os    = require('os');
 const path  = require('path');
 const https = require('https');
 
@@ -16,6 +17,9 @@ Microsoft 365 / Excel setup:
   5. In channels.json, set "workbook_id" to the file's drive item ID
      (visible in the file's URL on SharePoint/OneDrive)
   6. Re-run vp-intake
+
+For SharePoint sharing links ("Anyone with the link"), use "sharing_url" in channels.json
+instead of "workbook_id" — no credentials required.
 `;
 
 class AuthRequiredError extends Error {
@@ -35,6 +39,8 @@ function colLetterToIndex(letter) {
   return index - 1;
 }
 
+// ─── HTTP helpers ─────────────────────────────────────────────────────────────
+
 function httpsPost(url, body, headers) {
   return new Promise((resolve, reject) => {
     const payload = typeof body === 'string' ? body : JSON.stringify(body);
@@ -73,12 +79,46 @@ function httpsGet(url, headers) {
   });
 }
 
+/** Follow redirects and return {status, headers, body} */
+function httpsGetRaw(url, headers = {}, maxRedirects = 6) {
+  return new Promise((resolve, reject) => {
+    function doGet(currentUrl, remaining) {
+      const urlObj = new URL(currentUrl);
+      const req = https.request({
+        hostname: urlObj.hostname,
+        path: urlObj.pathname + urlObj.search,
+        method: 'GET',
+        headers: {
+          'User-Agent': 'Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/124.0.0.0 Safari/537.36',
+          ...headers,
+        },
+      }, (res) => {
+        if ([301, 302, 303, 307, 308].includes(res.statusCode) && res.headers.location && remaining > 0) {
+          let nextUrl = res.headers.location;
+          if (!nextUrl.startsWith('http')) {
+            nextUrl = `${urlObj.protocol}//${urlObj.host}${nextUrl}`;
+          }
+          return doGet(nextUrl, remaining - 1);
+        }
+        const chunks = [];
+        res.on('data', (chunk) => chunks.push(chunk));
+        res.on('end', () => resolve({ status: res.statusCode, headers: res.headers, body: Buffer.concat(chunks) }));
+      });
+      req.on('error', reject);
+      req.end();
+    }
+    doGet(url, maxRedirects);
+  });
+}
+
 function encodeFormData(obj) {
   return Object.entries(obj)
     .map(([k, v]) => `${encodeURIComponent(k)}=${encodeURIComponent(v)}`)
     .join('&');
 }
 
+// ─── Graph API (client credentials) ──────────────────────────────────────────
+
 async function getAccessToken(creds) {
   const now = Math.floor(Date.now() / 1000);
   if (_tokenCache && _tokenCache.expires_at > now + 60) {
@@ -101,7 +141,7 @@ async function getAccessToken(creds) {
   return res.access_token;
 }
 
-async function readExcelM365(channel, projectRoot) {
+async function readViaGraphApi(channel, projectRoot) {
   const root = projectRoot || process.cwd();
   const credPath = path.join(root, CREDENTIALS_REL);
 
@@ -126,12 +166,104 @@ async function readExcelM365(channel, projectRoot) {
   const response = await httpsGet(url, { Authorization: `Bearer ${token}` });
 
   if (!response.values || response.values.length === 0) return [];
+  return parseRowsWithColumnMap(response.values, channel.column_map);
+}
+
+// ─── SharePoint sharing link (anonymous WOPI) ─────────────────────────────────
+
+/** Detect sharing link type: /:x:/g/ or /:x:/r/ paths */
+function isSharingLink(url) {
+  return /\/:x:\/[gr]\//.test(url);
+}
+
+/**
+ * Resolve a SharePoint anonymous sharing link to a direct download URL.
+ * Technique: load WOPI viewer page → scrape FileGetUrl (temp auth token embedded in HTML).
+ */
+async function resolveSharePointDownloadUrl(sharingUrl) {
+  // First request: follow redirects to the WOPI viewer page
+  const { status, body } = await httpsGetRaw(sharingUrl);
+  if (status !== 200) {
+    throw new Error(`SharePoint sharing link returned HTTP ${status}. The link may be expired or restricted.`);
+  }
+
+  const html = body.toString('utf8');
+
+  // Scrape FileGetUrl — SharePoint embeds a temp-auth download URL in the viewer HTML
+  const match = html.match(/"FileGetUrl"\s*:\s*"([^"]+)"/);
+  if (!match) {
+    throw new Error('Could not extract download URL from SharePoint viewer page. The file may require sign-in.');
+  }
+
+  // Unescape JSON unicode escapes (\\u0026 → &)
+  return match[1].replace(/\\u([\da-fA-F]{4})/g, (_, hex) => String.fromCharCode(parseInt(hex, 16)));
+}
+
+/** Download xlsx bytes from a URL (no auth, temp token already embedded) */
+async function downloadXlsx(downloadUrl) {
+  const { status, body } = await httpsGetRaw(downloadUrl, {
+    Accept: 'application/vnd.openxmlformats-officedocument.spreadsheetml.sheet,*/*',
+  });
+  if (status !== 200) {
+    throw new Error(`Failed to download xlsx: HTTP ${status}`);
+  }
+  return body; // Buffer
+}
 
-  const colMap = channel.column_map;
+// ─── xlsx parser (SheetJS) ────────────────────────────────────────────────────
+
+/**
+ * Parse xlsx Buffer → array of row arrays (like response.values from Graph API).
+ * Uses the `xlsx` npm package (SheetJS).
+ */
+function parseXlsxBuffer(buffer, sheetName) {
+  let XLSX;
+  try {
+    XLSX = require('xlsx');
+  } catch {
+    throw new Error(
+      'The "xlsx" package is required to parse xlsx files from sharing links.\n' +
+      'Install it: npm install xlsx'
+    );
+  }
+
+  const workbook = XLSX.read(buffer, { type: 'buffer' });
+
+  let sheet;
+  if (sheetName && workbook.SheetNames.includes(sheetName)) {
+    sheet = workbook.Sheets[sheetName];
+  } else {
+    // Default to first sheet; warn if requested sheet not found
+    if (sheetName) {
+      process.stderr.write(`[vp-intake] Sheet "${sheetName}" not found. Using "${workbook.SheetNames[0]}" instead.\n`);
+    }
+    sheet = workbook.Sheets[workbook.SheetNames[0]];
+  }
+
+  if (!sheet) return [];
+
+  // Convert to array of arrays (raw values)
+  const rows = XLSX.utils.sheet_to_json(sheet, { header: 1, defval: '' });
+  return rows;
+}
+
+async function readViaSharingLink(channel) {
+  const sharingUrl = channel.sharing_url;
+  const downloadUrl = await resolveSharePointDownloadUrl(sharingUrl);
+  const xlsxBuffer = await downloadXlsx(downloadUrl);
+  const rows = parseXlsxBuffer(xlsxBuffer, channel.sheet_name || null);
+
+  if (!rows || rows.length === 0) return [];
+  return parseRowsWithColumnMap(rows, channel.column_map);
+}
+
+// ─── Common row → ticket mapper ───────────────────────────────────────────────
+
+function parseRowsWithColumnMap(values, colMap) {
   const tickets = [];
 
-  for (let i = 1; i < response.values.length; i++) {
-    const row = response.values[i];
+  for (let i = 1; i < values.length; i++) {
+    const row = values[i];
     const get = (field) => {
       const col = colMap[field];
       if (!col) return '';
@@ -147,7 +279,7 @@ async function readExcelM365(channel, projectRoot) {
       date:        get('date'),
       status:      get('status'),
       _source_row: i,
-      _channel_id: channel.id,
+      _channel_id: channel ? channel.id : undefined,
     };
 
     if (!ticket.title && !ticket.description) continue;
@@ -157,6 +289,40 @@ async function readExcelM365(channel, projectRoot) {
   return tickets;
 }
 
+// ─── Public entry point ───────────────────────────────────────────────────────
+
+async function readExcelM365(channel, projectRoot) {
+  // Sharing link mode: no credentials required
+  if (channel.sharing_url) {
+    if (isSharingLink(channel.sharing_url)) {
+      return readViaSharingLink(channel);
+    }
+    throw new Error(
+      `sharing_url "${channel.sharing_url}" is not a SharePoint sharing link.\n` +
+      'Sharing links look like: https://xxx.sharepoint.com/:x:/g/personal/xxx/EdXXX...\n' +
+      'For direct document links (Doc.aspx?sourcedoc=...), use workbook_id + credentials instead.'
+    );
+  }
+
+  // Graph API mode: requires workbook_id + credentials
+  if (!channel.workbook_id) {
+    throw new AuthRequiredError(
+      'excel_m365 channel requires either "sharing_url" (for SharePoint sharing links) or ' +
+      '"workbook_id" (for Graph API).' + SETUP_GUIDE
+    );
+  }
+
+  return readViaGraphApi(channel, projectRoot);
+}
+
 function clearTokenCache() { _tokenCache = null; }
 
-module.exports = { readExcelM365, AuthRequiredError, clearTokenCache };
+module.exports = {
+  readExcelM365,
+  AuthRequiredError,
+  clearTokenCache,
+  // exported for testing
+  isSharingLink,
+  resolveSharePointDownloadUrl,
+  parseXlsxBuffer,
+};

package/lib/intake/classifier.cjs

@@ -3,8 +3,10 @@
 const BUG_KEYWORDS = [
   'bug', 'error', 'fail', 'crash', 'broken', 'fix', 'exception', 'regression',
   'wrong', 'incorrect', 'unexpected', 'null', 'undefined', '500', '404',
+  'performance', 'slow', 'timeout', 'hang', 'freeze', 'unresponsive',
   // Vietnamese
   'lỗi', 'sự cố', 'không hoạt động', 'hỏng', 'sai', 'vá lỗi', 'bị lỗi',
+  'chậm', 'treo', 'tắc', 'không phản hồi',
 ];
 
 const ENH_KEYWORDS = [

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "viepilot",
-  "version": "2.48.0",
+  "version": "2.49.0",
   "description": "**Autonomous Vibe Coding Framework / Bộ khung phát triển tự động có kiểm soát**",
   "main": "index.js",
   "bin": {
@@ -74,7 +74,8 @@
   },
   "dependencies": {
     "docx": "^9.0.0",
-    "pptxgenjs": "^3.12.0"
+    "pptxgenjs": "^3.12.0",
+    "xlsx": "^0.18.5"
   },
   "optionalDependencies": {
     "@googleapis/slides": "^1.0.0"