koishi-plugin-bns-blacklist 1.0.4 → 1.0.5

package/lib/index.js

@@ -183,83 +183,127 @@ ${itemsHtml}
 </body></html>`;
 }
 // ─── Data Scraping ─────────────────────────────────────────
-async function scrapeSheet(page, url) {
-    await page.goto(url, { waitUntil: 'networkidle2', timeout: 60000 });
-    // Wait for the sheet to render
-    await new Promise(r => setTimeout(r, 5000));
-    // Try to extract data from Tencent Docs rendered cells
-    // Approach: use the Tencent Docs internal data API via JS
-    const data = await page.evaluate(() => {
-        // Try multiple approaches to extract cell data
-        const results = [];
-        // Approach 1: Look for the sheet's data model in the app's store
-        const app = window.__app || window.app || window.__SHEET_APP__;
-        if (app?.store?.getState) {
-            const state = app.store.getState();
-            // Navigate through common state paths
-            const sheetData = state?.sheet?.data || state?.sheetData || state?.data;
-            if (sheetData?.rows) {
-                for (const row of sheetData.rows) {
-                    const rowData = [];
-                    for (const cell of row.cells || row) {
-                        if (typeof cell === 'string')
-                            rowData.push(cell);
-                        else if (cell?.text)
-                            rowData.push(cell.text);
-                        else if (cell?.value)
-                            rowData.push(String(cell.value));
-                        else
-                            rowData.push('');
-                    }
-                    if (rowData.some(c => c))
-                        results.push(rowData);
+async function scrapeSheet(page, url, logger) {
+    // Collect network responses that might contain sheet data
+    const capturedData = [];
+    page.on('response', async (response) => {
+        const respUrl = response.url();
+        // Tencent Docs sheet data APIs
+        if (respUrl.includes('/api/') || respUrl.includes('/dop-api/') || respUrl.includes('opendoc')) {
+            try {
+                const body = await response.text();
+                // Only capture JSON responses that look like sheet data
+                if (body.length > 100 && body.length < 500000) {
+                    capturedData.push({ url: respUrl, body });
                 }
-                return results;
             }
+            catch (_) { /* ignore */ }
         }
-        // Approach 2: Fall back to DOM scraping - find all cell elements
-        const cellElements = document.querySelectorAll('[class*="cell"]:not([class*="cell-"])');
-        if (cellElements.length > 0) {
-            const cellMap = new Map();
-            cellElements.forEach(el => {
-                const row = el.getAttribute('data-row') || el.getAttribute('data-y');
-                const col = el.getAttribute('data-col') || el.getAttribute('data-x');
-                if (row && col) {
-                    cellMap.set(`${row},${col}`, el.innerText?.trim() || '');
-                }
-            });
-            if (cellMap.size > 0) {
-                const maxRow = Math.max(...Array.from(cellMap.keys()).map(k => parseInt(k.split(',')[0])));
-                const maxCol = Math.max(...Array.from(cellMap.keys()).map(k => parseInt(k.split(',')[1])));
-                for (let r = 0; r <= maxRow; r++) {
-                    const row = [];
-                    for (let c = 0; c <= maxCol; c++) {
-                        row.push(cellMap.get(`${r},${c}`) || '');
-                    }
-                    if (row.some(c => c))
-                        results.push(row);
-                }
-                return results;
+    });
+    await page.goto(url, { waitUntil: 'networkidle2', timeout: 60000 });
+    // Wait additional time for async data loading
+    await new Promise(r => setTimeout(r, 8000));
+    // Try to find the sheet data in captured API responses
+    for (const item of capturedData) {
+        try {
+            const json = JSON.parse(item.body);
+            const rows = extractRowsFromJson(json);
+            if (rows.length > 0) {
+                if (logger)
+                    logger.info(`Found data via API: ${item.url} (${rows.length} rows)`);
+                return parseSheetData(rows);
+            }
+        }
+        catch (_) { /* try next */ }
+    }
+    // Fallback: try to extract from page context
+    const fallbackData = await page.evaluate(() => {
+        const results = [];
+        // Try to find Tencent Docs internal data stores
+        const allKeys = Object.keys(window).filter(k => {
+            try {
+                return window[k] && typeof window[k] === 'object';
+            }
+            catch (_) {
+                return false;
             }
+        });
+        // Look for data in common global stores
+        for (const key of ['__app', 'app', '__SHEET_APP__', '__INITIAL_STATE__', '__NUXT__', '__NEXT_DATA__']) {
+            const obj = window[key];
+            if (!obj)
+                continue;
+            const jsonStr = JSON.stringify(obj);
+            const found = findSheetRowsInJson(jsonStr);
+            if (found.length > 0)
+                return found;
         }
-        // Approach 3: Get all visible text from the active sheet
-        const sheetView = document.querySelector('.sheet-view, [class*="sheet"], .spreadsheet-container');
-        if (sheetView) {
-            const text = sheetView.innerText;
-            // Parse tab-separated text
-            const lines = text.split('\n').filter(l => l.trim());
+        // Last resort: all visible text
+        const body = document.body.innerText;
+        if (body.length > 100) {
+            const lines = body.split('\n').filter((l) => l.trim());
             for (const line of lines) {
                 const cells = line.split('\t');
-                if (cells.length > 1)
+                if (cells.length >= 3)
                     results.push(cells);
             }
-            if (results.length > 1)
-                return results;
         }
         return results;
     });
-    // Parse extracted data into structured records
-    return parseSheetData(data);
+    return parseSheetData(fallbackData);
+}
+// Recursively search JSON for array-of-arrays that looks like sheet data
+function extractRowsFromJson(obj, depth = 0) {
+    if (depth > 8)
+        return [];
+    if (!obj || typeof obj !== 'object')
+        return [];
+    // Direct array of arrays (most common sheet data format)
+    if (Array.isArray(obj) && obj.length > 0 && Array.isArray(obj[0])) {
+        const rows = obj.map((row) => row.map(cell => {
+            if (cell === null || cell === undefined)
+                return '';
+            if (typeof cell === 'object') {
+                return cell.text || cell.value || cell.v || cell.formattedValue || '';
+            }
+            return String(cell);
+        }));
+        if (rows.length >= 2 && rows[0].length >= 2)
+            return rows;
+    }
+    // Search nested objects
+    if (Array.isArray(obj)) {
+        for (const item of obj) {
+            const found = extractRowsFromJson(item, depth + 1);
+            if (found.length > 0)
+                return found;
+        }
+    }
+    else {
+        for (const key of Object.keys(obj)) {
+            // Skip non-data keys
+            if (['config', 'style', 'meta', 'theme', 'permission'].some(s => key.toLowerCase().includes(s)))
+                continue;
+            const found = extractRowsFromJson(obj[key], depth + 1);
+            if (found.length > 0)
+                return found;
+        }
+    }
+    return [];
+}
+// Search a JSON string for array patterns that look like sheet rows
+function findSheetRowsInJson(jsonStr) {
+    // Try to find array patterns that could be sheet data
+    const results = [];
+    // Look for patterns like ["cell1","cell2",...] repeated
+    const rowRegex = /\[((?:"[^"]*"\s*,?\s*)+)\]/g;
+    let match;
+    while ((match = rowRegex.exec(jsonStr)) !== null) {
+        const cells = match[1].split(/"\s*,\s*"/).map(s => s.replace(/^"|"$/g, ''));
+        if (cells.length >= 3)
+            results.push(cells);
+    }
+    return results;
 }
 function parseSheetData(rows) {
     if (rows.length === 0)
@@ -333,7 +377,7 @@ function apply(ctx, config) {
             ctx.logger('bns-blacklist').info('Refreshing data from Tencent Docs...');
             page = await puppeteer.page();
             await page.setViewport({ width: 1920, height: 1080 });
-            const records = await scrapeSheet(page, config.sheetUrl);
+            const records = await scrapeSheet(page, config.sheetUrl, ctx.logger('bns-blacklist'));
             if (records.length > 0) {
                 blacklistCache = records;
                 saveCache();

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "koishi-plugin-bns-blacklist",
-  "version": "1.0.4",
+  "version": "1.0.5",
   "description": "剑灵怀旧服副本黑名单查询 - 从腾讯文档实时拉取数据，输入角色名即可生成黑名单截图",
   "main": "lib/index.js",
   "typings": "lib/index.d.ts",