koishi-plugin-bns-blacklist 1.0.5 → 1.0.7

package/lib/index.js

@@ -184,73 +184,88 @@ ${itemsHtml}
 }
 // ─── Data Scraping ─────────────────────────────────────────
 async function scrapeSheet(page, url, logger) {
-    // Collect network responses that might contain sheet data
-    const capturedData = [];
-    page.on('response', async (response) => {
-        const respUrl = response.url();
-        // Tencent Docs sheet data APIs
-        if (respUrl.includes('/api/') || respUrl.includes('/dop-api/') || respUrl.includes('opendoc')) {
+    await page.goto(url, { waitUntil: 'networkidle2', timeout: 60000 });
+    // Wait for the sheet to fully render
+    await new Promise(r => setTimeout(r, 10000));
+    // Log page info for debugging
+    if (logger) {
+        const title = await page.title();
+        const bodyLen = await page.evaluate(() => document.body.innerText.length);
+        logger.info(`Page title: "${title}", body text length: ${bodyLen}`);
+    }
+    // Strategy 1: Use the page's own fetch to call Tencent Docs internal API
+    const apiData = await page.evaluate(async () => {
+        const results = [];
+        // Try common Tencent Docs API endpoints from within the authenticated page
+        const baseUrl = 'https://docs.qq.com';
+        const sheetId = 'DZXRuRmtTS21mWU9V';
+        const endpoints = [
+            `/dop-api/sheet/data?id=${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 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 });
+                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 });
                 }
             }
-            catch (_) { /* ignore */ }
+            catch (_) { /* skip */ }
         }
+        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) {
+    if (logger)
+        logger.info(`API endpoints tried: ${apiData.length} responded`);
+    for (const item of apiData) {
+        const text = item.text || item.body;
+        if (logger)
+            logger.info(`  ${item.endpoint} -> status=${item.status}, len=${text?.length}`);
         try {
-            const json = JSON.parse(item.body);
+            const json = JSON.parse(text);
+            // Log sample of JSON structure for debugging
+            if (logger && text.length > 100) {
+                const keys = Object.keys(json);
+                logger.info(`  JSON keys: [${keys.join(', ')}]`);
+                // Log first 500 chars
+                logger.info(`  JSON sample: ${text.substring(0, 500)}`);
+            }
             const rows = extractRowsFromJson(json);
             if (rows.length > 0) {
                 if (logger)
-                    logger.info(`Found data via API: ${item.url} (${rows.length} rows)`);
+                    logger.info(`Found data: ${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;
+    // 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 ${rows.length} rows from page text`);
+            return parseSheetData(rows);
         }
-        // Last resort: all visible text
-        const body = document.body.innerText;
-        if (body.length > 100) {
-            const lines = body.split('\n').filter((l) => l.trim());
-            for (const line of lines) {
-                const cells = line.split('\t');
-                if (cells.length >= 3)
-                    results.push(cells);
+        // If no tabs, maybe data is in a different format - log first few lines
+        if (logger) {
+            logger.info(`Page text first 5 lines (${lines.length} total):`);
+            for (let i = 0; i < Math.min(5, lines.length); i++) {
+                logger.info(`  [${i}] ${lines[i].substring(0, 200)}`);
             }
         }
-        return results;
-    });
-    return parseSheetData(fallbackData);
+    }
+    return [];
 }
 // Recursively search JSON for array-of-arrays that looks like sheet data
 function extractRowsFromJson(obj, depth = 0) {

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "koishi-plugin-bns-blacklist",
-  "version": "1.0.5",
+  "version": "1.0.7",
   "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",