koishi-plugin-bns-blacklist 1.0.4 → 1.0.6

package/lib/index.js

@@ -183,83 +183,134 @@ ${itemsHtml}
 </body></html>`;
 }
 // ─── Data Scraping ─────────────────────────────────────────
-async function scrapeSheet(page, url) {
+async function scrapeSheet(page, url, logger) {
     await page.goto(url, { waitUntil: 'networkidle2', timeout: 60000 });
-    // Wait for the sheet to render
-    await new Promise(r => setTimeout(r, 5000));
-    // Try to extract data from Tencent Docs rendered cells
-    // Approach: use the Tencent Docs internal data API via JS
-    const data = await page.evaluate(() => {
-        // Try multiple approaches to extract cell data
+    // Wait for the sheet to fully render
+    await new Promise(r => setTimeout(r, 10000));
+    // Log page info for debugging
+    if (logger) {
+        const title = await page.title();
+        const bodyLen = await page.evaluate(() => document.body.innerText.length);
+        logger.info(`Page title: "${title}", body text length: ${bodyLen}`);
+    }
+    // Strategy 1: Use the page's own fetch to call Tencent Docs internal API
+    const apiData = await page.evaluate(async () => {
         const results = [];
-        // Approach 1: Look for the sheet's data model in the app's store
-        const app = window.__app || window.app || window.__SHEET_APP__;
-        if (app?.store?.getState) {
-            const state = app.store.getState();
-            // Navigate through common state paths
-            const sheetData = state?.sheet?.data || state?.sheetData || state?.data;
-            if (sheetData?.rows) {
-                for (const row of sheetData.rows) {
-                    const rowData = [];
-                    for (const cell of row.cells || row) {
-                        if (typeof cell === 'string')
-                            rowData.push(cell);
-                        else if (cell?.text)
-                            rowData.push(cell.text);
-                        else if (cell?.value)
-                            rowData.push(String(cell.value));
-                        else
-                            rowData.push('');
-                    }
-                    if (rowData.some(c => c))
-                        results.push(rowData);
+        // Try common Tencent Docs API endpoints from within the authenticated page
+        const baseUrl = 'https://docs.qq.com';
+        const sheetId = 'DZXRuRmtTS21mWU9V';
+        const endpoints = [
+            `/dop-api/sheet/data?id=${sheetId}`,
+            `/dop-api/sheet/get?id=${sheetId}`,
+            `/dop-api/spreadsheet/${sheetId}`,
+            `/api/sheet/data?id=${sheetId}`,
+            `/api/sheet/${sheetId}/values`,
+            `/openapi/v1/spreadsheet/${sheetId}/values/Sheet1`,
+            `/v2/sheet/${sheetId}/export?format=json`,
+            `/cgi-bin/online_docs/sheet_get?docId=${sheetId}`,
+        ];
+        for (const endpoint of endpoints) {
+            try {
+                const resp = await fetch(baseUrl + endpoint, { credentials: 'include' });
+                const text = await resp.text();
+                if (text.length > 50 && !text.includes('Not Found') && !text.includes('<html')) {
+                    results.push({ endpoint, text, status: resp.status });
                 }
-                return results;
             }
+            catch (_) { /* skip */ }
         }
-        // Approach 2: Fall back to DOM scraping - find all cell elements
-        const cellElements = document.querySelectorAll('[class*="cell"]:not([class*="cell-"])');
-        if (cellElements.length > 0) {
-            const cellMap = new Map();
-            cellElements.forEach(el => {
-                const row = el.getAttribute('data-row') || el.getAttribute('data-y');
-                const col = el.getAttribute('data-col') || el.getAttribute('data-x');
-                if (row && col) {
-                    cellMap.set(`${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;
+        return results;
+    });
+    if (logger)
+        logger.info(`API endpoints tried: ${apiData.length} responded`);
+    for (const item of apiData) {
+        if (logger)
+            logger.info(`  ${item.endpoint} -> status=${item.status}, len=${item.body?.length || item.text?.length}`);
+        try {
+            const json = JSON.parse(item.text || item.body);
+            const rows = extractRowsFromJson(json);
+            if (rows.length > 0) {
+                if (logger)
+                    logger.info(`Found data: ${rows.length} rows from ${item.endpoint}`);
+                return parseSheetData(rows);
             }
         }
-        // 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());
-            for (const line of lines) {
-                const cells = line.split('\t');
-                if (cells.length > 1)
-                    results.push(cells);
+        catch (_) { /* try next */ }
+    }
+    // Strategy 2: Dump page text and try to parse
+    const pageText = await page.evaluate(() => document.body.innerText);
+    if (pageText.length > 100) {
+        const lines = pageText.split('\n').filter((l) => l.trim());
+        // Look for data lines that have at least 3 tab-separated fields
+        const tabLines = lines.filter((l) => l.split('\t').length >= 3);
+        if (tabLines.length >= 2) {
+            const rows = tabLines.map((l) => l.split('\t'));
+            if (logger)
+                logger.info(`Parsed ${rows.length} rows from page text`);
+            return parseSheetData(rows);
+        }
+        // If no tabs, maybe data is in a different format - log first few lines
+        if (logger) {
+            logger.info(`Page text first 5 lines (${lines.length} total):`);
+            for (let i = 0; i < Math.min(5, lines.length); i++) {
+                logger.info(`  [${i}] ${lines[i].substring(0, 200)}`);
             }
-            if (results.length > 1)
-                return results;
         }
-        return results;
-    });
-    // Parse extracted data into structured records
-    return parseSheetData(data);
+    }
+    return [];
+}
+// 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 +384,7 @@ 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": "1.0.6",
   "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",