1

.gitignore

@@ -0,0 +1,4 @@
+server/node_modules/*
+server/puppeteer_profile/*
+.env
+chrome-win/*

mv2_simple_crx/src/actions/amazon.js

@@ -86,7 +86,7 @@ export function injected_amazon_switch_language(params) {
 
 export function injected_amazon_search_list(params) {
   const start_url = params && params.url ? String(params.url) : location.href;
-  const category_keyword = params && params.category_keyword ? String(params.category_keyword).trim() : '';
+              const category_keyword = params && params.category_keyword ? String(params.category_keyword).trim() : '';
   const sort_by = params && params.sort_by ? String(params.sort_by).trim() : '';
 
   function pick_number(text) {
@@ -106,15 +106,15 @@ export function injected_amazon_search_list(params) {
     if (mm) return Math.round(parseFloat(mm[1]) * 1000000);
     const digits = raw.replace(/[^\d]/g, '');
     return digits ? Number(digits) : null;
-  }
+              }
 
-  function abs_url(href) {
-    try {
-      return new URL(href, location.origin).toString();
-    } catch (_) {
-      return href;
-    }
-  }
+              function abs_url(href) {
+                try {
+                  return new URL(href, location.origin).toString();
+                } catch (_) {
+                  return href;
+                }
+              }
 
   function parse_asin_from_url(url) {
     if (!url || typeof url !== 'string') return null;
@@ -161,20 +161,20 @@ export function injected_amazon_search_list(params) {
         const n = pick_int(review_count_text);
         return Number.isFinite(n) ? n : null;
       })();
-      items.push({
+                  items.push({
         index: idx + 1,
         asin: asin || parse_asin_from_url(item_url),
-        title,
+                    title,
         url: item_url,
-        price,
+                    price,
         rating,
         rating_text,
         review_count,
         review_count_text,
-      });
-    });
-    return items;
-  }
+                  });
+                });
+                return items;
+              }
 
   function pick_next_url() {
     const a = document.querySelector('a.s-pagination-next');
@@ -191,7 +191,7 @@ export function injected_amazon_search_list(params) {
   return {
     start_url,
     href: location.href,
-    category_keyword,
+                  category_keyword,
     sort_by,
     total: items.length,
     items,
@@ -273,18 +273,90 @@ export function injected_amazon_product_detail() {
     if (t) bullets.push(t);
   });
 
-  const variants = {};
-  document.querySelectorAll('[id^="variation_"]').forEach((block) => {
-    const key = block.id.replace(/^variation_/, '') || block.id;
-    const sel =
-      block.querySelector('.selection') ||
-      block.querySelector('.a-button-selected .a-button-text') ||
-      block.querySelector('[class*="dropdown"]');
-    if (sel) {
-      const v = norm(sel.textContent);
-      if (v) variants[key] = v;
+  /** 变体 id 后缀是否为颜色 / 尺寸（仅提取这两项，不收集其它维度） */
+  function is_sku_color_key(k) {
+    const x = String(k).toLowerCase();
+    return x === 'color' || x === 'color_name' || x.endsWith('_color_name');
+  }
+  function is_sku_size_key(k) {
+    const x = String(k).toLowerCase();
+    return x === 'size' || x === 'size_name' || x.endsWith('_size_name');
+  }
+
+  /** 变体维度：颜色 / 尺寸 各为选项列表 */
+  const sku = { color: [], size: [] };
+
+  const twister_plus_root = document.querySelector('#twister-plus-desktop-twister-container');
+
+  if (twister_plus_root) {
+    const color_row = twister_plus_root.querySelector('#inline-twister-row-color_name');
+    if (color_row) {
+      const seen_c = new Set();
+      color_row.querySelectorAll('li').forEach((li) => {
+        const img = li.querySelector('img[alt]');
+        if (!img) return;
+        const v = norm(img.getAttribute('alt'));
+        if (v && !seen_c.has(v)) {
+          seen_c.add(v);
+          sku.color.push(v);
+        }
+      });
     }
-  });
+    if (!sku.color.length) {
+      const dim = twister_plus_root.querySelector('#inline-twister-expanded-dimension-text-color_name');
+      const v = dim && norm(dim.textContent);
+      if (v) sku.color.push(v);
+    }
+    const size_row = twister_plus_root.querySelector('#inline-twister-row-size_name');
+    if (size_row) {
+      const seen_s = new Set();
+      size_row.querySelectorAll('li').forEach((li) => {
+        const el = li.querySelector('.swatch-title-text-display, .swatch-title-text-single-line');
+        const v = el ? norm(el.textContent) : null;
+        if (v && !seen_s.has(v)) {
+          seen_s.add(v);
+          sku.size.push(v);
+        }
+      });
+    }
+    if (!sku.size.length) {
+      const dim = twister_plus_root.querySelector('#inline-twister-expanded-dimension-text-size_name');
+      const v = dim && norm(dim.textContent);
+      if (v) sku.size.push(v);
+    }
+  } else {
+    let cur_color = null;
+    let cur_size = null;
+    document.querySelectorAll('[id^="variation_"]').forEach((block) => {
+      const key = block.id.replace(/^variation_/, '') || block.id;
+      if (!is_sku_color_key(key) && !is_sku_size_key(key)) return;
+      const sel =
+        block.querySelector('.selection') ||
+        block.querySelector('.a-button-selected .a-button-text') ||
+        block.querySelector('[class*="dropdown"]');
+      if (!sel) return;
+      const v = norm(sel.textContent);
+      if (!v) return;
+      if (is_sku_color_key(key) && !cur_color) cur_color = v;
+      if (is_sku_size_key(key) && !cur_size) cur_size = v;
+    });
+    document.querySelectorAll('div.inline-twister-row[id^="inline-twister-row-"]').forEach((row) => {
+      const id = row.id || '';
+      const key = id.replace(/^inline-twister-row-/, '') || id;
+      if (!is_sku_color_key(key) && !is_sku_size_key(key)) return;
+      const selected =
+        row.querySelector('.a-button-selected .swatch-title-text-display') ||
+        row.querySelector('.a-button-selected .a-button-text') ||
+        row.querySelector('.a-button-selected');
+      if (!selected) return;
+      const v = norm(selected.textContent);
+      if (!v) return;
+      if (is_sku_color_key(key) && !cur_color) cur_color = v;
+      if (is_sku_size_key(key) && !cur_size) cur_size = v;
+    });
+    if (cur_color) sku.color.push(cur_color);
+    if (cur_size) sku.size.push(cur_size);
+  }
 
   let delivery_hint = null;
   const del = document.querySelector(
@@ -292,11 +364,6 @@ export function injected_amazon_product_detail() {
   );
   if (del) delivery_hint = norm(del.innerText).slice(0, 500);
 
-  let sku = null;
-  Object.keys(product_info).forEach((k) => {
-    if (/^sku$/i.test(k) || /item model|型号|part number|制造商型号/i.test(k)) sku = product_info[k];
-  });
-
   const images = [];
   const seen_img = new Set();
   function add_img(u) {
@@ -338,7 +405,6 @@ export function injected_amazon_product_detail() {
     product_info,
     detail_extra_lines,
     bullets,
-    variants,
     delivery_hint,
     sku,
     images,
@@ -492,8 +558,8 @@ export function amazon_search_list(data, sendResponse) {
             chrome.tabs.update(tab.id, { url: next_url, active: true }, () => {
               if (chrome.runtime.lastError) return reject_nav(new Error(chrome.runtime.lastError.message));
               resolve_nav(true);
-            });
-          });
+                });
+              });
           await wait_tab_complete(tab.id);
         }
         const injected_result_list = await tab.execute_script(
@@ -518,10 +584,10 @@ export function amazon_search_list(data, sendResponse) {
         total: unique_map.size,
         items: Array.from(unique_map.values()).slice(0, limit),
       };
-      const result = {
-        code: 0,
-        status: true,
-        message: 'ok',
+            const result = {
+              code: 0,
+              status: true,
+              message: 'ok',
         data: { tab_id: tab.id, url, category_keyword, sort_by: sort_by || 'featured', limit, result: list_result },
       };
       send_action('amazon_search_list', result);
@@ -610,7 +676,7 @@ export function amazon_set_language(data, sendResponse) {
         data: null,
         documentURI: AMAZON_HOME_FOR_LANG,
       });
-      reject(err);
+            reject(err);
     }
   });
 }
@@ -686,7 +752,8 @@ export function amazon_product_detail(data, sendResponse) {
   return run_pdp_action(data && data.product_url, injected_amazon_product_detail, [], 'amazon_product_detail', sendResponse);
 }
 
-amazon_product_detail.desc = 'Amazon 商品详情（标题、价格、品牌、SKU、要点、变体、配送摘要等）';
+amazon_product_detail.desc =
+  'Amazon 商品详情（标题、价格、品牌、sku{color[],size[]}、要点、配送摘要等）';
 amazon_product_detail.params = {
   product_url: {
     type: 'string',

mv2_simple_crx/src/bridge/bridge.html

@@ -0,0 +1,11 @@
+<!doctype html>
+<html lang="zh-CN">
+  <head>
+    <meta charset="utf-8" />
+    <title>server_bridge</title>
+  </head>
+  <body>
+    <!-- Puppeteer 打开此页后 evaluate 调用 window.server_bridge_invoke，与 background 同协议 -->
+    <script src="bridge.js"></script>
+  </body>
+</html>

mv2_simple_crx/src/bridge/bridge.js

@@ -0,0 +1,32 @@
+/**
+ * 服务端 Puppeteer 通过此页与 background 通讯（等同 UI 发 chrome.runtime.sendMessage）
+ * 页面内若需 Web Worker 做重计算，可在此 postMessage；当前直连 background 即可满足指令/结果
+ */
+(function () {
+  function server_bridge_invoke(action, data) {
+    return new Promise(function (resolve, reject) {
+      if (!action) {
+        reject(new Error('缺少 action'));
+        return;
+      }
+      chrome.runtime.sendMessage({ action: action, data: data || {} }, function (res) {
+        var err = chrome.runtime.lastError;
+        if (err) {
+          reject(new Error(err.message));
+          return;
+        }
+        if (!res) {
+          reject(new Error('background 无响应'));
+          return;
+        }
+        if (res.ok) {
+          resolve(res.data);
+        } else {
+          reject(new Error(res.error || 'action 失败'));
+        }
+      });
+    });
+  }
+
+  window.server_bridge_invoke = server_bridge_invoke;
+})();

server/.env.example

@@ -0,0 +1,15 @@
+# MySQL
+MYSQL_HOST=127.0.0.1
+MYSQL_PORT=3306
+MYSQL_USER=root
+MYSQL_PASSWORD=
+MYSQL_DATABASE=ecom_crawl
+
+# 扩展目录（未打包，含 manifest.json）
+CRX_SRC_PATH=d:/项目/电商抓取项目/mv2_simple_crx/src
+
+SERVER_PORT=38080
+ACTION_TIMEOUT_MS=300000
+PUPPETEER_HEADLESS=false
+# 可选：指定浏览器路径（不填默认用 ../chrome-win/chrome.exe）
+CHROME_EXECUTABLE_PATH=

server/README.md

@@ -0,0 +1,40 @@
+## server（Koa + Sequelize + MySQL）
+
+### 功能
+- `POST /api/crawl/run_action`：服务端调用扩展 action，返回结果，并按 `stage` 自动入库
+- **定时任务写死配置**：`config/cron_tasks.js`（不走数据库）
+
+### 运行
+1. 安装依赖
+
+```bash
+cd server
+npm install
+```
+
+2. 配置环境变量
+
+```bash
+copy .env.example .env
+```
+
+3. 初始化/更新表结构
+
+```bash
+node scripts/db_sync.js
+```
+
+4. 启动
+
+```bash
+node app.js
+```
+
+### 定时任务
+编辑 `config/cron_tasks.js` 的 `cron_task_list`，重启服务即可生效。
+
+### 落库表
+- `crawl_run_record`：所有 action 调用的原始请求/响应
+- `amazon_product`：`stage=detail` 详情
+- `amazon_search_item`：`stage=list` 列表 item
+- `amazon_review`：`stage=reviews` 评论

server/app.js

@@ -0,0 +1,29 @@
+import dotenv from 'dotenv';
+import Koa from 'koa';
+import body_parser from 'koa-bodyparser';
+
+import { sequelize } from './models/index.js';
+import { crawl_router } from './routes/crawl.js';
+import { start_all_cron_tasks } from './services/schedule_loader.js';
+
+dotenv.config();
+
+const app = new Koa();
+app.use(body_parser({ jsonLimit: '10mb' }));
+
+app.use(crawl_router.routes()).use(crawl_router.allowedMethods());
+
+app.use(async (ctx) => {
+  ctx.status = 404;
+  ctx.body = { ok: false, error: 'not_found' };
+});
+
+const port = Number(process.env.SERVER_PORT || 38080);
+
+await sequelize.authenticate();
+await sequelize.sync();
+start_all_cron_tasks();
+
+app.listen(port);
+// eslint-disable-next-line no-console
+console.log(`server listening on ${port}`);

server/config/cron_tasks.js

@@ -0,0 +1,14 @@
+/**
+ * 写死定时任务配置（不走数据库）
+ * cron_expression 参考 node-cron
+ */
+
+export const cron_task_list = [
+  // 示例：每 6 小时跑一次列表抓取
+  // {
+  //   name: 'amazon_search_list_every_6h',
+  //   cron_expression: '0 */6 * * *',
+  //   action_name: 'amazon_search_list',
+  //   action_payload: { keyword: '午餐包', limit: 100 }
+  // }
+];

server/config/database.js

@@ -0,0 +1,22 @@
+import dotenv from 'dotenv';
+
+dotenv.config();
+
+export function get_sequelize_options() {
+  return {
+    host: process.env.MYSQL_HOST || '127.0.0.1',
+    port: Number(process.env.MYSQL_PORT || 3306),
+    username: process.env.MYSQL_USER || 'root',
+    password: process.env.MYSQL_PASSWORD || '',
+    database: process.env.MYSQL_DATABASE || 'ecom_crawl',
+    dialect: 'mysql',
+    logging: false,
+    define: {
+      underscored: true,
+      timestamps: true,
+      createdAt: 'created_at',
+      updatedAt: 'updated_at'
+    },
+    timezone: '+08:00'
+  };
+}

server/models/amazon_product.js

@@ -0,0 +1,39 @@
+import { DataTypes } from 'sequelize';
+
+export function define_amazon_product(sequelize) {
+  return sequelize.define(
+    'amazon_product',
+    {
+      id: { type: DataTypes.BIGINT.UNSIGNED, primaryKey: true, autoIncrement: true },
+      asin: { type: DataTypes.STRING(32), allowNull: false },
+      url: { type: DataTypes.TEXT, allowNull: false },
+      title: { type: DataTypes.TEXT, allowNull: true },
+      price: { type: DataTypes.STRING(64), allowNull: true },
+      sku: { type: DataTypes.STRING(256), allowNull: true },
+      sku_color: { type: DataTypes.STRING(128), allowNull: true },
+      sku_size: { type: DataTypes.STRING(128), allowNull: true },
+      brand_line: { type: DataTypes.TEXT, allowNull: true },
+      brand_store_url: { type: DataTypes.TEXT, allowNull: true },
+      ac_badge: { type: DataTypes.STRING(128), allowNull: true },
+      bestseller_hint: { type: DataTypes.TEXT, allowNull: true },
+      delivery_hint: { type: DataTypes.TEXT, allowNull: true },
+      social_proof: { type: DataTypes.TEXT, allowNull: true },
+      sustainability_hint: { type: DataTypes.TEXT, allowNull: true },
+      rating_stars: { type: DataTypes.STRING(64), allowNull: true },
+      review_count_text: { type: DataTypes.STRING(64), allowNull: true },
+      main_image: { type: DataTypes.TEXT, allowNull: true },
+      images_json: { type: DataTypes.TEXT('long'), allowNull: true },
+      bullets_json: { type: DataTypes.TEXT('long'), allowNull: true },
+      product_info_json: { type: DataTypes.TEXT('long'), allowNull: true },
+      detail_extra_lines_json: { type: DataTypes.TEXT('long'), allowNull: true }
+    },
+    {
+      tableName: 'amazon_product',
+      indexes: [
+        { unique: true, fields: ['asin'] },
+        { fields: ['created_at'] },
+        { fields: ['updated_at'] }
+      ]
+    }
+  );
+}

server/models/amazon_review.js

@@ -0,0 +1,31 @@
+import { DataTypes } from 'sequelize';
+
+export function define_amazon_review(sequelize) {
+  return sequelize.define(
+    'amazon_review',
+    {
+      id: { type: DataTypes.BIGINT.UNSIGNED, primaryKey: true, autoIncrement: true },
+      asin: { type: DataTypes.STRING(32), allowNull: true },
+      url: { type: DataTypes.TEXT, allowNull: false },
+      review_id: { type: DataTypes.STRING(64), allowNull: false },
+      author: { type: DataTypes.STRING(256), allowNull: true },
+      title: { type: DataTypes.TEXT, allowNull: true },
+      body: { type: DataTypes.TEXT('long'), allowNull: true },
+      rating_text: { type: DataTypes.STRING(64), allowNull: true },
+      review_date: { type: DataTypes.STRING(128), allowNull: true },
+      review_index: { type: DataTypes.INTEGER, allowNull: true },
+      batch_key: { type: DataTypes.STRING(64), allowNull: false },
+      batch_total: { type: DataTypes.INTEGER, allowNull: true },
+      batch_limit: { type: DataTypes.INTEGER, allowNull: true }
+    },
+    {
+      tableName: 'amazon_review',
+      indexes: [
+        { unique: true, fields: ['review_id'] },
+        { fields: ['asin'] },
+        { fields: ['batch_key'] },
+        { fields: ['created_at'] }
+      ]
+    }
+  );
+}

server/models/amazon_search_item.js

@@ -0,0 +1,30 @@
+import { DataTypes } from 'sequelize';
+
+export function define_amazon_search_item(sequelize) {
+  return sequelize.define(
+    'amazon_search_item',
+    {
+      id: { type: DataTypes.BIGINT.UNSIGNED, primaryKey: true, autoIncrement: true },
+      asin: { type: DataTypes.STRING(32), allowNull: false },
+      url: { type: DataTypes.TEXT, allowNull: false },
+      title: { type: DataTypes.TEXT, allowNull: true },
+      price: { type: DataTypes.STRING(64), allowNull: true },
+      rating: { type: DataTypes.FLOAT, allowNull: true },
+      rating_text: { type: DataTypes.STRING(64), allowNull: true },
+      review_count: { type: DataTypes.INTEGER, allowNull: true },
+      review_count_text: { type: DataTypes.STRING(64), allowNull: true },
+      rank_index: { type: DataTypes.INTEGER, allowNull: true, comment: '列表中的 index 字段' },
+      batch_key: { type: DataTypes.STRING(64), allowNull: false, comment: '一次列表抓取的批次 key' },
+      batch_total: { type: DataTypes.INTEGER, allowNull: true },
+      batch_limit: { type: DataTypes.INTEGER, allowNull: true }
+    },
+    {
+      tableName: 'amazon_search_item',
+      indexes: [
+        { fields: ['asin'] },
+        { fields: ['batch_key'] },
+        { fields: ['created_at'] }
+      ]
+    }
+  );
+}

server/models/crawl_run_record.js

@@ -0,0 +1,25 @@
+import { DataTypes } from 'sequelize';
+
+export function define_crawl_run_record(sequelize) {
+  return sequelize.define(
+    'crawl_run_record',
+    {
+      id: { type: DataTypes.BIGINT.UNSIGNED, primaryKey: true, autoIncrement: true },
+      action_name: { type: DataTypes.STRING(128), allowNull: false },
+      request_payload: { type: DataTypes.TEXT, allowNull: true, comment: 'JSON 请求体' },
+      ok: { type: DataTypes.BOOLEAN, allowNull: false },
+      result_payload: { type: DataTypes.TEXT('long'), allowNull: true, comment: 'JSON 结果' },
+      error_message: { type: DataTypes.TEXT, allowNull: true },
+      source: {
+        type: DataTypes.STRING(32),
+        allowNull: false,
+        defaultValue: 'api',
+        comment: 'api | cron'
+      }
+    },
+    {
+      tableName: 'crawl_run_record',
+      indexes: [{ fields: ['action_name'] }, { fields: ['created_at'] }]
+    }
+  );
+}

server/models/index.js

@@ -0,0 +1,17 @@
+import { Sequelize } from 'sequelize';
+import { get_sequelize_options } from '../config/database.js';
+import { define_crawl_run_record } from './crawl_run_record.js';
+import { define_amazon_product } from './amazon_product.js';
+import { define_amazon_search_item } from './amazon_search_item.js';
+import { define_amazon_review } from './amazon_review.js';
+
+const sequelize_options = get_sequelize_options();
+const { database, username, password, ...rest } = sequelize_options;
+
+export const sequelize = new Sequelize(database, username, password, rest);
+
+export const crawl_run_record = define_crawl_run_record(sequelize);
+
+export const amazon_product = define_amazon_product(sequelize);
+export const amazon_search_item = define_amazon_search_item(sequelize);
+export const amazon_review = define_amazon_review(sequelize);

server/models/schedule_task.js

@@ -0,0 +1 @@
+// 已废弃：按需求改为写死定时任务（见 config/cron_tasks.js）

server/package.json

@@ -0,0 +1,20 @@
+{
+  "name": "ecom_crawl_server",
+  "version": "1.0.0",
+  "private": true,
+  "type": "module",
+  "scripts": {
+    "start": "node app.js",
+    "db_sync": "node scripts/db_sync.js"
+  },
+  "dependencies": {
+    "@koa/router": "^12.0.1",
+    "dotenv": "^16.4.5",
+    "koa": "^2.15.3",
+    "koa-bodyparser": "^4.4.1",
+    "mysql2": "^3.11.0",
+    "node-cron": "^3.0.3",
+    "puppeteer": "^23.4.1",
+    "sequelize": "^6.37.3"
+  }
+}

server/routes/crawl.js

@@ -0,0 +1,26 @@
+import Router from '@koa/router';
+import { execute_action_and_record } from '../services/task_executor.js';
+
+export const crawl_router = new Router();
+
+crawl_router.post('/api/crawl/run_action', async (ctx) => {
+  const { action_name, action_payload } = ctx.request.body || {};
+
+  if (!action_name) {
+    ctx.status = 400;
+    ctx.body = { ok: false, error: '缺少 action_name' };
+    return;
+  }
+
+  try {
+    const data = await execute_action_and_record({
+      action_name,
+      action_payload: action_payload || {},
+      source: 'api'
+    });
+    ctx.body = { ok: true, data };
+  } catch (err) {
+    ctx.status = 500;
+    ctx.body = { ok: false, error: (err && err.message) || String(err) };
+  }
+});

server/routes/schedule_task.js

@@ -0,0 +1 @@
+// 已废弃：按需求改为写死定时任务（见 config/cron_tasks.js）

server/scripts/db_sync.js

@@ -0,0 +1,6 @@
+import { sequelize } from '../models/index.js';
+
+await sequelize.sync({ alter: true });
+// eslint-disable-next-line no-console
+console.log('sync ok');
+await sequelize.close();

server/services/amazon_persist.js

@@ -0,0 +1,110 @@
+import {
+  amazon_product,
+  amazon_search_item,
+  amazon_review
+} from '../models/index.js';
+import { safe_json_stringify } from './json_utils.js';
+
+function build_batch_key(prefix) {
+  return `${prefix}_${Date.now()}_${Math.random().toString().slice(2, 8)}`;
+}
+
+function pick_asin_from_url(url) {
+  if (!url) return null;
+  const m = String(url).match(/\/dp\/([A-Z0-9]{8,16})/i);
+  return m && m[1] ? m[1].toUpperCase() : null;
+}
+
+export async function persist_amazon_result(result) {
+  if (!result || !result.stage) {
+    return;
+  }
+
+  if (result.stage === 'detail') {
+    const asin = result.asin || pick_asin_from_url(result.url);
+    if (!asin) {
+      return;
+    }
+
+    await amazon_product.upsert({
+      asin,
+      url: result.url || '',
+      title: result.title || null,
+      price: result.price || null,
+      sku: result.sku || null,
+      sku_color: result.sku_color || null,
+      sku_size: result.sku_size || null,
+      brand_line: result.brand_line || null,
+      brand_store_url: result.brand_store_url || null,
+      ac_badge: result.ac_badge || null,
+      bestseller_hint: result.bestseller_hint || null,
+      delivery_hint: result.delivery_hint || null,
+      social_proof: result.social_proof || null,
+      sustainability_hint: result.sustainability_hint || null,
+      rating_stars: result.rating_stars || null,
+      review_count_text: result.review_count_text || null,
+      main_image: result.main_image || null,
+      images_json: safe_json_stringify(result.images || []),
+      bullets_json: safe_json_stringify(result.bullets || []),
+      product_info_json: safe_json_stringify(result.product_info || {}),
+      detail_extra_lines_json: safe_json_stringify(result.detail_extra_lines || [])
+    });
+
+    return;
+  }
+
+  if (result.stage === 'list') {
+    const batch_key = build_batch_key('list');
+    const items = Array.isArray(result.items) ? result.items : [];
+
+    for (const it of items) {
+      const asin = it.asin || pick_asin_from_url(it.url);
+      if (!asin || !it.url) continue;
+
+      await amazon_search_item.create({
+        asin,
+        url: it.url,
+        title: it.title || null,
+        price: it.price || null,
+        rating: typeof it.rating === 'number' ? it.rating : null,
+        rating_text: it.rating_text || null,
+        review_count: typeof it.review_count === 'number' ? it.review_count : null,
+        review_count_text: it.review_count_text || null,
+        rank_index: typeof it.index === 'number' ? it.index : null,
+        batch_key,
+        batch_total: typeof result.total === 'number' ? result.total : null,
+        batch_limit: typeof result.limit === 'number' ? result.limit : null
+      });
+    }
+
+    return;
+  }
+
+  if (result.stage === 'reviews') {
+    const batch_key = build_batch_key('reviews');
+    const asin = pick_asin_from_url(result.url);
+    const items = Array.isArray(result.items) ? result.items : [];
+
+    for (const it of items) {
+      const review_id = it.review_id;
+      if (!review_id) continue;
+
+      const asin_value = asin || pick_asin_from_url(it.url) || pick_asin_from_url(result.url);
+
+      await amazon_review.upsert({
+        asin: asin_value || null,
+        url: result.url || '',
+        review_id,
+        author: it.author || null,
+        title: it.title || null,
+        body: it.body || null,
+        rating_text: it.rating_text || null,
+        review_date: it.date || null,
+        review_index: typeof it.index === 'number' ? it.index : null,
+        batch_key,
+        batch_total: typeof result.total === 'number' ? result.total : null,
+        batch_limit: typeof result.limit === 'number' ? result.limit : null
+      });
+    }
+  }
+}

server/services/cron_manager.js

@@ -0,0 +1 @@
+// 已废弃：按需求改为写死定时任务（见 config/cron_tasks.js）

server/services/json_utils.js

@@ -0,0 +1,18 @@
+export function safe_json_stringify(value) {
+  try {
+    return JSON.stringify(value);
+  } catch (err) {
+    return JSON.stringify({ error: 'json_stringify_failed', message: String(err) });
+  }
+}
+
+export function safe_json_parse(text) {
+  if (text === null || text === undefined || text === '') {
+    return null;
+  }
+  try {
+    return JSON.parse(text);
+  } catch (err) {
+    return null;
+  }
+}

server/services/puppeteer_runner.js

@@ -0,0 +1,158 @@
+import dotenv from 'dotenv';
+import fs from 'node:fs';
+import path from 'node:path';
+import puppeteer from 'puppeteer';
+
+dotenv.config();
+
+let browser_singleton = null;
+
+function get_action_timeout_ms() {
+  return Number(process.env.ACTION_TIMEOUT_MS || 300000);
+}
+
+function get_crx_src_path() {
+  const crx_src_path = process.env.CRX_SRC_PATH;
+  if (!crx_src_path) {
+    throw new Error('缺少环境变量 CRX_SRC_PATH');
+  }
+  return crx_src_path;
+}
+
+function get_extension_id_from_targets(targets) {
+  for (const target of targets) {
+    const url = target.url();
+    if (!url) continue;
+    if (url.startsWith('chrome-extension://')) {
+      const match = url.match(/^chrome-extension:\/\/([^/]+)\//);
+      if (match && match[1]) return match[1];
+    }
+  }
+  return null;
+}
+
+async function wait_for_extension_id(browser, timeout_ms) {
+  const existing = get_extension_id_from_targets(browser.targets());
+  if (existing) {
+    return existing;
+  }
+
+  const target = await browser
+    .waitForTarget((t) => {
+      const url = t.url();
+      return typeof url === 'string' && url.startsWith('chrome-extension://');
+    }, { timeout: timeout_ms })
+    .catch(() => null);
+
+  if (!target) {
+    return null;
+  }
+
+  return get_extension_id_from_targets([target]);
+}
+
+function get_chrome_executable_path() {
+  // 优先环境变量，方便你后续切换版本
+  const from_env = process.env.CHROME_EXECUTABLE_PATH;
+  if (from_env) {
+    return path.resolve(from_env);
+  }
+
+  // 默认使用项目根目录的 chrome-win/chrome.exe
+  // 当前进程 cwd 通常是 server/，所以回到上一级
+  return path.resolve(process.cwd(), '../chrome-win/chrome.exe');
+}
+
+export async function get_or_create_browser() {
+  if (browser_singleton) {
+    return browser_singleton;
+  }
+
+  const chrome_executable_path = get_chrome_executable_path();
+  if (!fs.existsSync(chrome_executable_path)) {
+    throw new Error(`Chrome 不存在: ${chrome_executable_path}`);
+  }
+
+  const raw_extension_path = path.resolve(get_crx_src_path());
+  const manifest_path = path.resolve(raw_extension_path, 'manifest.json');
+  if (!fs.existsSync(manifest_path)) {
+    throw new Error(`扩展 manifest.json 不存在: ${manifest_path}`);
+  }
+
+  const extension_path = raw_extension_path.replace(/\\/g, '/');
+  const headless = String(process.env.PUPPETEER_HEADLESS || 'false') === 'true';
+  const user_data_dir = path.resolve(process.cwd(), 'puppeteer_profile');
+
+  browser_singleton = await puppeteer.launch({
+    executablePath: chrome_executable_path,
+    headless,
+    args: [
+      `--user-data-dir=${user_data_dir}`,
+      '--enable-extensions',
+      `--disable-extensions-except=${extension_path}`,
+      `--load-extension=${extension_path}`,
+      '--no-default-browser-check',
+      '--disable-popup-blocking',
+      '--disable-dev-shm-usage'
+    ]
+  });
+
+  return browser_singleton;
+}
+
+export async function invoke_extension_action(action_name, action_payload) {
+  const browser = await get_or_create_browser();
+
+  const page = await browser.newPage();
+  await page.goto('about:blank');
+
+  // 尝试先打开 chrome://extensions 触发扩展初始化（某些环境下扩展 target 不会立刻出现）
+  try {
+    await page.goto('chrome://extensions/', { waitUntil: 'domcontentloaded' });
+  } catch (err) {
+    // ignore
+  }
+
+  const extension_id = await wait_for_extension_id(browser, 15000);
+  if (!extension_id) {
+    await page.close();
+    throw new Error(
+      '未找到扩展 extension_id：Chrome 未加载扩展（常见原因：MV2 被禁用/企业策略未生效/CRX_SRC_PATH 不正确/使用了 headless）'
+    );
+  }
+
+  const bridge_url = `chrome-extension://${extension_id}/bridge/bridge.html`;
+  await page.goto(bridge_url, { waitUntil: 'domcontentloaded' });
+
+  const timeout_ms = get_action_timeout_ms();
+  const action_res = await page.evaluate(
+    async (action, payload, timeout) => {
+      function with_timeout(promise, timeout_ms_inner) {
+        return new Promise((resolve, reject) => {
+          const timer = setTimeout(() => reject(new Error('action_timeout')), timeout_ms_inner);
+          promise
+            .then((v) => {
+              clearTimeout(timer);
+              resolve(v);
+            })
+            .catch((e) => {
+              clearTimeout(timer);
+              reject(e);
+            });
+        });
+      }
+
+      if (!window.server_bridge_invoke) {
+        throw new Error('bridge 未注入 window.server_bridge_invoke');
+      }
+
+      return await with_timeout(window.server_bridge_invoke(action, payload), timeout);
+    },
+    action_name,
+    action_payload || {},
+    timeout_ms
+  );
+
+  await page.close();
+  return action_res;
+}

server/services/schedule_loader.js

@@ -0,0 +1,30 @@
+import cron from 'node-cron';
+import { cron_task_list } from '../config/cron_tasks.js';
+import { execute_action_and_record } from './task_executor.js';
+
+const cron_jobs = [];
+
+export function start_all_cron_tasks() {
+  for (const task of cron_task_list) {
+    const job = cron.schedule(task.cron_expression, async () => {
+      try {
+        await execute_action_and_record({
+          action_name: task.action_name,
+          action_payload: task.action_payload || {},
+          source: 'cron'
+        });
+      } catch (err) {
+        // 失败会在 crawl_run_record 落库
+      }
+    });
+
+    cron_jobs.push(job);
+  }
+}
+
+export function stop_all_cron_tasks() {
+  for (const job of cron_jobs) {
+    job.stop();
+  }
+  cron_jobs.length = 0;
+}

server/services/task_executor.js

@@ -0,0 +1,39 @@
+import { crawl_run_record } from '../models/index.js';
+import { safe_json_stringify } from './json_utils.js';
+import { invoke_extension_action } from './puppeteer_runner.js';
+import { persist_amazon_result } from './amazon_persist.js';
+
+export async function execute_action_and_record(params) {
+  const { action_name, action_payload, source } = params;
+
+  const request_payload = safe_json_stringify(action_payload || {});
+
+  let ok = false;
+  let result_payload = null;
+  let error_message = null;
+
+  try {
+    const result = await invoke_extension_action(action_name, action_payload || {});
+
+    ok = true;
+    result_payload = safe_json_stringify(result);
+
+    // 按 stage 自动入库（不影响原始 run_record 记录）
+    await persist_amazon_result(result);
+
+    return result;
+  } catch (err) {
+    ok = false;
+    error_message = (err && err.message) || String(err);
+    throw err;
+  } finally {
+    await crawl_run_record.create({
+      action_name,
+      request_payload,
+      ok,
+      result_payload,
+      error_message,
+      source
+    });
+  }
+}