feat: 实现房间状态同步功能

- 新增 RoomStatusManager 类，负责管理房间状态快照表的数据库连接池及批量 Upsert 操作。
- 新增 StatusBatchProcessor 类，负责收集和合并房间状态更新，并定期将其写入数据库。
- 新增状态提取器 statusExtractor.js，从 Kafka 消息中提取并构建房间状态更新对象。
- 修改 index.js，初始化 RoomStatusManager 和 StatusBatchProcessor，并在 Kafka 消息处理流程中并行推送状态更新。
- 修改 processor/index.js，更新 processKafkaMessage 函数以支持状态提取和处理。
- 更新 kafkaPayload.js，修正 control_list 的提取逻辑，兼容 Kafka 实际传输中的 loop 字段。
- 添加状态批处理器和状态提取器的单元测试，确保功能的正确性。
- 更新文档 plan-room-status-sync.md，详细描述房间状态同步方案及字段映射。

bls-rcu-action-backend/.env.example

@@ -30,4 +30,14 @@ REDIS_DB=0
 REDIS_PROJECT_NAME=my-project
 REDIS_API_BASE_URL=http://localhost:3000
 
+# Room Status DB Configuration (optional, falls back to DB_* values)
+# ROOM_STATUS_DB_HOST=localhost
+# ROOM_STATUS_DB_PORT=5432
+# ROOM_STATUS_DB_USER=postgres
+# ROOM_STATUS_DB_PASSWORD=password
+# ROOM_STATUS_DB_DATABASE=bls_rcu_action
+# ROOM_STATUS_DB_MAX_CONNECTIONS=5
+# ROOM_STATUS_DB_SCHEMA=room_status
+# ROOM_STATUS_DB_TABLE=room_status_moment
+
 ENABLE_LOOP_NAME_AUTO_GENERATION=true

bls-rcu-action-backend/_test_status.js

@@ -0,0 +1,11 @@
+import pg from 'pg';
+const pool = new pg.Pool({
+  host: '10.8.8.109', port: 5433, user: 'log_admin',
+  password: 'YourActualStrongPasswordForPostgres!', database: 'log_platform', max: 1
+});
+const s = await pool.query("SELECT count(*) as total, count(*) FILTER (WHERE dev_loops IS NOT NULL AND dev_loops != '{}'::jsonb) as with_loops, count(*) FILTER (WHERE sys_lock_status IS NOT NULL) as with_lock FROM room_status.room_status_moment");
+console.log(JSON.stringify(s.rows[0]));
+const r = await pool.query("SELECT hotel_id, room_id, device_id, sys_lock_status, dev_loops FROM room_status.room_status_moment WHERE dev_loops IS NOT NULL AND dev_loops != '{}'::jsonb ORDER BY ts_ms DESC LIMIT 3");
+console.log('Samples:', r.rows.length);
+for (const row of r.rows) console.log(JSON.stringify(row));
+await pool.end();

bls-rcu-action-backend/scripts/init_db.sql

@@ -1,4 +1,5 @@
 -- Database Initialization Script for BLS RCU Action Server
+-- 描述：创建 rcu_action 模式及 rcu_action_events 分区表，用于存储 RCU 通讯日志流水
 
 CREATE SCHEMA IF NOT EXISTS rcu_action;
 
@@ -47,3 +48,31 @@ CREATE INDEX IF NOT EXISTS idx_rcu_action_action_type ON rcu_action.rcu_action_e
 
 -- Composite Index for typical query pattern (Hotel + Room + Time)
 CREATE INDEX IF NOT EXISTS idx_rcu_action_query_main ON rcu_action.rcu_action_events (hotel_id, room_id, ts_ms DESC);
+
+-- Column Comments
+COMMENT ON TABLE rcu_action.rcu_action_events IS 'RCU 通讯日志流水表 - 存储从 Kafka 消费的 RCU 设备上报/下发/ACK 事件';
+COMMENT ON COLUMN rcu_action.rcu_action_events.guid IS '主键，32位无横线 UUID';
+COMMENT ON COLUMN rcu_action.rcu_action_events.ts_ms IS '日志产生时间戳（毫秒），同时用作分区键';
+COMMENT ON COLUMN rcu_action.rcu_action_events.write_ts_ms IS '入库时间戳（毫秒），由后端服务写入时生成';
+COMMENT ON COLUMN rcu_action.rcu_action_events.hotel_id IS '酒店 ID';
+COMMENT ON COLUMN rcu_action.rcu_action_events.room_id IS '房间 ID';
+COMMENT ON COLUMN rcu_action.rcu_action_events.device_id IS 'RCU 设备 ID（主板编号）';
+COMMENT ON COLUMN rcu_action.rcu_action_events.direction IS '数据方向：上报 / 下发';
+COMMENT ON COLUMN rcu_action.rcu_action_events.cmd_word IS '命令字，如 0x36（状态上报）、0x0F（控制下发/ACK）';
+COMMENT ON COLUMN rcu_action.rcu_action_events.frame_id IS '通讯帧号，用于串联同一次通讯的命令与状态';
+COMMENT ON COLUMN rcu_action.rcu_action_events.udp_raw IS 'UDP 消息原文（base64 编码）';
+COMMENT ON COLUMN rcu_action.rcu_action_events.action_type IS '记录行为类型：用户操作 / 设备回路状态 / 下发控制 / 0FACK / 无效';
+COMMENT ON COLUMN rcu_action.rcu_action_events.sys_lock_status IS '系统锁状态：0=未锁定, 1=锁定（仅 0x36 上报）';
+COMMENT ON COLUMN rcu_action.rcu_action_events.report_count IS '本次上报设备数量（对应 device_list 长度）';
+COMMENT ON COLUMN rcu_action.rcu_action_events.dev_type IS '设备类型编号，拆分自 device_list/fault_list/control_list';
+COMMENT ON COLUMN rcu_action.rcu_action_events.dev_addr IS '设备地址编号';
+COMMENT ON COLUMN rcu_action.rcu_action_events.dev_loop IS '设备回路编号';
+COMMENT ON COLUMN rcu_action.rcu_action_events.dev_data IS '设备状态数值（仅 0x36 状态上报）';
+COMMENT ON COLUMN rcu_action.rcu_action_events.fault_count IS '本次故障设备数量（对应 fault_list 长度）';
+COMMENT ON COLUMN rcu_action.rcu_action_events.error_type IS '故障类型：0x01=在线/离线, 0x02=电量, 0x03=电流 等';
+COMMENT ON COLUMN rcu_action.rcu_action_events.error_data IS '故障内容数据（含义取决于 error_type）';
+COMMENT ON COLUMN rcu_action.rcu_action_events.type_l IS '执行方式（仅 0x0F 下发控制）';
+COMMENT ON COLUMN rcu_action.rcu_action_events.type_h IS '执行内容（仅 0x0F 下发控制）';
+COMMENT ON COLUMN rcu_action.rcu_action_events.details IS '业务详情 JSONB：存储完整的 device_list / fault_list / control_list';
+COMMENT ON COLUMN rcu_action.rcu_action_events.extra IS '扩展信息 JSONB：存储上游传入的附加字段';
+COMMENT ON COLUMN rcu_action.rcu_action_events.loop_name IS '回路名称：通过 device_id → room_type_id → loop_address 查询获得';

bls-rcu-action-backend/src/config/config.js

@@ -53,5 +53,16 @@ export const config = {
     projectName: process.env.REDIS_PROJECT_NAME || 'bls-rcu-action',
     apiBaseUrl: process.env.REDIS_API_BASE_URL || `http://localhost:${parseNumber(process.env.PORT, 3000)}`
   },
+  roomStatusDb: {
+    host: process.env.ROOM_STATUS_DB_HOST || process.env.DB_HOST || process.env.POSTGRES_HOST || 'localhost',
+    port: parseNumber(process.env.ROOM_STATUS_DB_PORT || process.env.DB_PORT || process.env.POSTGRES_PORT, 5432),
+    user: process.env.ROOM_STATUS_DB_USER || process.env.DB_USER || process.env.POSTGRES_USER || 'postgres',
+    password: process.env.ROOM_STATUS_DB_PASSWORD || process.env.DB_PASSWORD || process.env.POSTGRES_PASSWORD || '',
+    database: process.env.ROOM_STATUS_DB_DATABASE || process.env.DB_DATABASE || process.env.POSTGRES_DATABASE || 'bls_rcu_action',
+    max: parseNumber(process.env.ROOM_STATUS_DB_MAX_CONNECTIONS, 5),
+    ssl: process.env.ROOM_STATUS_DB_SSL === 'true' ? { rejectUnauthorized: false } : undefined,
+    schema: process.env.ROOM_STATUS_DB_SCHEMA || 'room_status',
+    table: process.env.ROOM_STATUS_DB_TABLE || 'room_status_moment'
+  },
   enableLoopNameAutoGeneration: process.env.ENABLE_LOOP_NAME_AUTO_GENERATION === 'true'
 };

bls-rcu-action-backend/src/db/roomStatusManager.js

@@ -0,0 +1,158 @@
+/**
+ * Room Status Manager
+ * 
+ * Manages an independent PostgreSQL connection pool for
+ * the room_status.room_status_moment snapshot table.
+ * Provides batch upsert with JSONB merge and auto-partition creation.
+ */
+import pg from 'pg';
+import { randomUUID } from 'crypto';
+import { logger } from '../utils/logger.js';
+
+const { Pool } = pg;
+
+export class RoomStatusManager {
+    /**
+     * @param {Object} dbConfig - roomStatusDb config from config.js
+     */
+    constructor(dbConfig) {
+        this.pool = new Pool({
+            host: dbConfig.host,
+            port: dbConfig.port,
+            user: dbConfig.user,
+            password: dbConfig.password,
+            database: dbConfig.database,
+            max: dbConfig.max,
+            ssl: dbConfig.ssl
+        });
+        this.schema = dbConfig.schema;
+        this.table = dbConfig.table;
+        this.fullTableName = `${this.schema}.${this.table}`;
+        // Track which partitions we have already ensured
+        this.knownPartitions = new Set();
+    }
+
+    /**
+     * Batch upsert status rows into room_status_moment.
+     * Uses ON CONFLICT for atomic merge.
+     * 
+     * @param {Array<Object>} rows - Array of merged status objects
+     *   Each: { hotel_id, room_id, device_id, ts_ms, sys_lock_status, dev_loops, faulty_device_count }
+     */
+    async upsertBatch(rows) {
+        if (!rows || rows.length === 0) return;
+
+        // Pre-ensure all needed partitions exist before attempting upsert
+        const newHotelIds = [...new Set(rows.map(r => r.hotel_id))]
+            .filter(id => !this.knownPartitions.has(id));
+
+        if (newHotelIds.length > 0) {
+            await this._ensurePartitionsBatch(newHotelIds);
+        }
+
+        await this._doUpsert(rows);
+    }
+
+    /**
+     * Execute the actual upsert SQL for a batch of rows.
+     */
+    async _doUpsert(rows) {
+        const values = [];
+        const placeholders = [];
+
+        for (let i = 0; i < rows.length; i++) {
+            const row = rows[i];
+            const offset = i * 8; // Changed from 9 to 8
+            values.push(
+                row.guid || randomUUID(),          // $1
+                row.ts_ms,                         // $2  
+                row.hotel_id,                      // $3
+                row.room_id,                       // $4
+                row.device_id,                     // $5
+                row.sys_lock_status,               // $6
+                row.dev_loops ? JSON.stringify(row.dev_loops) : null,              // $7 (was $8)
+                row.faulty_device_count ? JSON.stringify(row.faulty_device_count) : null  // $8 (was $9)
+            );
+            const p = (n) => `$${offset + n}`;
+            placeholders.push(`(${p(1)}::uuid, ${p(2)}, ${p(3)}, ${p(4)}, ${p(5)}, ${p(6)}, ${p(7)}::jsonb, ${p(8)}::jsonb)`);
+        }
+
+        const sql = `
+      INSERT INTO ${this.fullTableName} (
+        guid, ts_ms, hotel_id, room_id, device_id,
+        sys_lock_status, dev_loops, faulty_device_count
+      ) VALUES ${placeholders.join(', ')}
+      ON CONFLICT (hotel_id, room_id, device_id)
+      DO UPDATE SET
+        ts_ms = GREATEST(${this.fullTableName}.ts_ms, EXCLUDED.ts_ms),
+        sys_lock_status = COALESCE(EXCLUDED.sys_lock_status, ${this.fullTableName}.sys_lock_status),
+        dev_loops = CASE
+          WHEN EXCLUDED.dev_loops IS NULL THEN ${this.fullTableName}.dev_loops
+          ELSE COALESCE(${this.fullTableName}.dev_loops, '{}'::jsonb) || EXCLUDED.dev_loops
+        END,
+        faulty_device_count = COALESCE(EXCLUDED.faulty_device_count, ${this.fullTableName}.faulty_device_count)
+    `;
+
+        await this.pool.query(sql, values);
+    }
+
+    /**
+     * Check if an error is a missing partition error.
+     */
+    _isPartitionMissingError(error) {
+        const msg = error?.message || '';
+        return msg.includes('no partition') || msg.includes('routing') ||
+            (error?.code === '23514' && msg.includes('partition'));
+    }
+
+    /**
+     * Batch-create LIST partitions for multiple hotel_ids in a single connection.
+     * Uses CREATE TABLE IF NOT EXISTS (idempotent) — no check query needed.
+     */
+    async _ensurePartitionsBatch(hotelIds) {
+        const client = await this.pool.connect();
+        try {
+            for (const hotelId of hotelIds) {
+                const partitionName = `${this.schema}.${this.table}_h${hotelId}`;
+                try {
+                    await client.query(
+                        `CREATE TABLE IF NOT EXISTS ${partitionName} PARTITION OF ${this.fullTableName} FOR VALUES IN (${hotelId})`
+                    );
+                    this.knownPartitions.add(hotelId);
+                } catch (err) {
+                    // Partition may already exist (race condition) — safe to ignore
+                    if (!err.message?.includes('already exists')) {
+                        logger.error('Error creating partition', { error: err?.message, hotelId });
+                    }
+                    this.knownPartitions.add(hotelId);
+                }
+            }
+            if (hotelIds.length > 0) {
+                logger.info(`Ensured ${hotelIds.length} room_status partitions`);
+            }
+        } finally {
+            client.release();
+        }
+    }
+
+    /**
+     * Ensure a LIST partition exists for the given hotel_id (single).
+     */
+    async ensurePartition(hotelId) {
+        if (this.knownPartitions.has(hotelId)) return;
+        await this._ensurePartitionsBatch([hotelId]);
+    }
+
+    async testConnection() {
+        try {
+            await this.pool.query('SELECT 1');
+            return true;
+        } catch {
+            return false;
+        }
+    }
+
+    async close() {
+        await this.pool.end();
+    }
+}

bls-rcu-action-backend/src/db/statusBatchProcessor.js

@@ -0,0 +1,133 @@
+/**
+ * Status Batch Processor
+ * 
+ * Collects status updates for room_status_moment table,
+ * merges updates for the same device within a batch window,
+ * and flushes them via RoomStatusManager.
+ * 
+ * Errors during flush are logged but never thrown, to protect the main pipeline.
+ */
+import { logger } from '../utils/logger.js';
+
+export class StatusBatchProcessor {
+    /**
+     * @param {import('./roomStatusManager.js').RoomStatusManager} roomStatusManager
+     * @param {Object} options
+     * @param {number} [options.flushInterval=500] - Flush interval in ms
+     * @param {number} [options.maxBufferSize=200] - Max items before forced flush
+     * @param {import('../redis/redisIntegration.js').RedisIntegration} [options.redisIntegration] - For error reporting
+     */
+    constructor(roomStatusManager, options = {}) {
+        this.roomStatusManager = roomStatusManager;
+        this.flushInterval = options.flushInterval || 500;
+        this.maxBufferSize = options.maxBufferSize || 200;
+        this.redisIntegration = options.redisIntegration || null;
+
+        /** @type {Map<string, Object>} compositeKey -> mergedState */
+        this.buffer = new Map();
+        this.timer = null;
+    }
+
+    /**
+     * Build composite key for deduplication.
+     */
+    _key(update) {
+        return `${update.hotel_id}:${update.room_id}:${update.device_id}`;
+    }
+
+    /**
+     * Add a status update to the buffer, merging with any existing entry for the same device.
+     * @param {Object} update - Output from extractStatusUpdate()
+     */
+    add(update) {
+        if (!update) return;
+
+        const key = this._key(update);
+        const existing = this.buffer.get(key);
+
+        if (existing) {
+            // Merge: take latest ts_ms
+            existing.ts_ms = Math.max(existing.ts_ms, update.ts_ms);
+
+            // sys_lock_status: prefer newer non-null value
+            if (update.sys_lock_status != null) {
+                existing.sys_lock_status = update.sys_lock_status;
+            }
+
+            // dev_loops: merge keys (new overwrites old for same key)
+            if (update.dev_loops) {
+                existing.dev_loops = existing.dev_loops
+                    ? { ...existing.dev_loops, ...update.dev_loops }
+                    : update.dev_loops;
+            }
+
+            // faulty_device_count: full replacement (newer wins)
+            if (update.faulty_device_count) {
+                existing.faulty_device_count = update.faulty_device_count;
+            }
+        } else {
+            // Clone to avoid mutation of caller's object
+            this.buffer.set(key, { ...update });
+        }
+
+        // Check if we should flush
+        if (this.buffer.size >= this.maxBufferSize && !this.isFlushing) {
+            this.flush();
+        } else if (!this.timer && !this.isFlushing) {
+            this.timer = setTimeout(() => this.flush(), this.flushInterval);
+        }
+    }
+
+    /**
+     * Flush all buffered status updates to the database.
+     * Errors are caught and logged, never thrown.
+     */
+    async flush() {
+        if (this.timer) {
+            clearTimeout(this.timer);
+            this.timer = null;
+        }
+
+        if (this.buffer.size === 0) return;
+        if (this.isFlushing) return;
+        this.isFlushing = true;
+
+        // Snapshot and clear the buffer atomically
+        const rows = [...this.buffer.values()];
+        this.buffer.clear();
+
+        try {
+            logger.info('StatusBatchProcessor flushing rows', { count: rows.length, sampleRowKeys: rows.map(r => r.device_id).slice(0, 5) });
+            await this.roomStatusManager.upsertBatch(rows);
+            logger.info('StatusBatchProcessor flushed successfully', { count: rows.length });
+        } catch (error) {
+            logger.error('StatusBatchProcessor flush failed', {
+                error: error?.message,
+                stack: error?.stack,
+                count: rows.length
+            });
+
+            // Report to Redis console if available
+            if (this.redisIntegration) {
+                try {
+                    await this.redisIntegration.error('StatusBatchProcessor flush failed', {
+                        module: 'room_status',
+                        count: rows.length,
+                        stack: error?.message
+                    });
+                } catch {
+                    // Silently ignore Redis reporting errors
+                }
+            }
+            // IMPORTANT: Do NOT re-throw. This protects the main Kafka pipeline.
+        } finally {
+            this.isFlushing = false;
+            // Catch up if buffer refilled wildly during flush
+            if (this.buffer.size >= this.maxBufferSize) {
+                this.flush();
+            } else if (this.buffer.size > 0 && !this.timer) {
+                this.timer = setTimeout(() => this.flush(), this.flushInterval);
+            }
+        }
+    }
+}

bls-rcu-action-backend/src/index.js

@@ -6,17 +6,20 @@ import partitionManager from './db/partitionManager.js';
 import projectMetadata from './cache/projectMetadata.js';
 import { createKafkaConsumers } from './kafka/consumer.js';
 import { processKafkaMessage } from './processor/index.js';
+import { extractStatusUpdate } from './processor/statusExtractor.js';
 import { createRedisClient } from './redis/redisClient.js';
 import { RedisIntegration } from './redis/redisIntegration.js';
 import { buildErrorQueueKey, enqueueError, startErrorRetryWorker } from './redis/errorQueue.js';
 import { MetricCollector } from './utils/metricCollector.js';
+import { RoomStatusManager } from './db/roomStatusManager.js';
+import { StatusBatchProcessor } from './db/statusBatchProcessor.js';
 import { logger } from './utils/logger.js';
 import { BatchProcessor } from './db/batchProcessor.js';
 
 const bootstrap = async () => {
   // 0. Initialize Database (Create DB, Schema, Table, Partitions)
   await dbInitializer.initialize();
-  
+
   // 0.1 Initialize Project Metadata Cache
   await projectMetadata.init();
 
@@ -44,7 +47,7 @@ const bootstrap = async () => {
   // Wait, I imported `dbManager` from `./db/databaseManager.js`.
   // If `databaseManager.js` exports an instance as default, I should use that.
   // If it exports a class, I should instantiate it.
-  
+
   // Let's assume the previous code `new DatabaseManager` was correct if it was a class.
   // BUT I used `dbManager.pool` in `partitionManager.js` assuming it's an instance.
   // I need to verify `databaseManager.js`.
@@ -57,13 +60,22 @@ const bootstrap = async () => {
   );
   redisIntegration.startHeartbeat();
 
+  // 1.2 Initialize Room Status Manager (independent pool for snapshot table)
+  const roomStatusManager = new RoomStatusManager(config.roomStatusDb);
+  const statusBatchProcessor = new StatusBatchProcessor(roomStatusManager, {
+    flushInterval: 500,
+    maxBufferSize: 200,
+    redisIntegration
+  });
+  logger.info('Room Status sync pipeline initialized');
+
   // 1.1 Setup Metric Reporting Cron Job (Every minute)
   cron.schedule('* * * * *', async () => {
     const metrics = metricCollector.getAndReset();
     const report = `[Minute Metrics] Pulled: ${metrics.kafka_pulled}, Parse Error: ${metrics.parse_error}, Inserted: ${metrics.db_inserted}, Failed: ${metrics.db_failed}`;
     console.log(report);
     logger.info(report, metrics);
-    
+
     try {
       await redisIntegration.info('Minute Metrics', metrics);
     } catch (err) {
@@ -105,9 +117,20 @@ const bootstrap = async () => {
           valueLength: typeof messageValue === 'string' ? messageValue.length : null
         });
       }
-      const rows = await processKafkaMessage({ message });
+      const { rows, payload } = await processKafkaMessage({ message });
       const inserted = await batchProcessor.add({ rows });
       metricCollector.increment('db_inserted');
+
+      // Fire-and-forget: extract status and push to StatusBatchProcessor
+      try {
+        const statusUpdate = extractStatusUpdate(payload);
+        if (statusUpdate) {
+          statusBatchProcessor.add(statusUpdate);
+        }
+      } catch (statusErr) {
+        logger.error('Status extraction failed (non-blocking)', { error: statusErr?.message });
+      }
+
       logger.info('Kafka message processed', { inserted });
     } catch (error) {
       if (error.type === 'PARSE_ERROR') {
@@ -201,7 +224,7 @@ const bootstrap = async () => {
   // Graceful Shutdown Logic
   const shutdown = async (signal) => {
     logger.info(`Received ${signal}, shutting down...`);
-    
+
     try {
       // 1. Close Kafka Consumer
       if (consumers && consumers.length > 0) {
@@ -216,7 +239,16 @@ const bootstrap = async () => {
       await redisClient.quit();
       logger.info('Redis client closed');
 
-      // 4. Close Database Pool
+      // 4. Flush and close Room Status pipeline
+      try {
+        await statusBatchProcessor.flush();
+        await roomStatusManager.close();
+        logger.info('Room Status pipeline closed');
+      } catch (rsErr) {
+        logger.error('Error closing Room Status pipeline', { error: rsErr?.message });
+      }
+
+      // 5. Close Database Pool
       await dbManager.close();
       logger.info('Database connection closed');
 

bls-rcu-action-backend/src/processor/index.js

@@ -123,7 +123,7 @@ const getActionTypeByDevType = (devType) => {
   // 1. Env override
   const rule = devTypeActionRules.find(r => r.dev_type === devType);
   if (rule?.action_type) return rule.action_type;
-  
+
   // 2. Default map
   const entry = defaultDevTypeActionMap[devType];
   return entry?.action || '设备回路状态';
@@ -354,24 +354,24 @@ export const buildRowsFromPayload = (rawPayload) => {
 
   // 3. 0x0F ACK
   else if (messageType === '0FACK') {
-     const { control_list: controls = [] } = payload;
-     if (Array.isArray(controls)) {
-       const details = { control_list: controls };
-       controls.forEach((control) => {
-         rows.push({
-           ...commonFields,
-           guid: createGuid(),
-           dev_type: control.dev_type ?? null,
-           dev_addr: control.dev_addr ?? null,
-           dev_loop: control.dev_loop ?? null,
-           dev_data: control.dev_data ?? null,
-           type_h: control.type_h ?? null,
-           action_type: '设备回路状态',
-           loop_name: getLoopNameWithFallback(deviceId, control.dev_type, control.dev_addr, control.dev_loop),
-           details
-         });
-       });
-     }
+    const { control_list: controls = [] } = payload;
+    if (Array.isArray(controls)) {
+      const details = { control_list: controls };
+      controls.forEach((control) => {
+        rows.push({
+          ...commonFields,
+          guid: createGuid(),
+          dev_type: control.dev_type ?? null,
+          dev_addr: control.dev_addr ?? null,
+          dev_loop: control.dev_loop ?? null,
+          dev_data: control.dev_data ?? null,
+          type_h: control.type_h ?? null,
+          action_type: '设备回路状态',
+          loop_name: getLoopNameWithFallback(deviceId, control.dev_type, control.dev_addr, control.dev_loop),
+          details
+        });
+      });
+    }
   }
 
   return rows;
@@ -380,9 +380,11 @@ export const buildRowsFromPayload = (rawPayload) => {
 export const processKafkaMessage = async ({ message }) => {
   let rows;
   try {
-    const payload = parseKafkaPayload(message.value);
-    rows = buildRowsFromPayload(payload);
-    return rows;
+    const rawPayload = parseKafkaPayload(message.value);
+    // Validate through Zod to get normalized fields (arrays defaulted, types coerced)
+    const validatedPayload = kafkaPayloadSchema.parse(rawPayload);
+    rows = buildRowsFromPayload(rawPayload);
+    return { rows, payload: validatedPayload };
   } catch (error) {
     error.type = 'PARSE_ERROR';
     const rawValue = Buffer.isBuffer(message.value)

bls-rcu-action-backend/src/processor/statusExtractor.js

@@ -0,0 +1,105 @@
+/**
+ * Status Extractor
+ * 
+ * Extracts a snapshot-style status update from a validated Kafka payload.
+ * The output is suitable for upsert into room_status.room_status_moment.
+ */
+
+/**
+ * Zero-pad a number to 3 digits.
+ * @param {number|string} val
+ * @returns {string} e.g. 1 -> "001", 23 -> "023"
+ */
+const pad3 = (val) => String(val).padStart(3, '0');
+
+/**
+ * Build a 9-character loop address key from device parameters.
+ * Format: {dev_type(3)}{dev_addr(3)}{dev_loop(3)}
+ * Example: type=1, addr=23, loop=12 -> "001023012"
+ */
+const buildLoopKey = (devType, devAddr, devLoop) =>
+    `${pad3(devType)}${pad3(devAddr)}${pad3(devLoop)}`;
+
+/**
+ * Extract a status update object from a validated Kafka payload.
+ * 
+ * @param {Object} payload - The parsed and validated Kafka payload
+ * @returns {Object|null} Status update object or null if nothing to update
+ */
+export const extractStatusUpdate = (payload) => {
+    if (!payload) return null;
+
+    const {
+        hotel_id,
+        room_id,
+        device_id,
+        ts_ms,
+        sys_lock_status,
+        device_list = [],
+        fault_list = [],
+        control_list = [],
+        direction,
+        cmd_word
+    } = payload;
+
+    // Must have identity fields
+    if (hotel_id == null || !room_id || !device_id || !ts_ms) {
+        return null;
+    }
+
+    // Build dev_loops from device_list (0x36) or control_list (0x0F)
+    let devLoops = null;
+
+    if (device_list.length > 0) {
+        devLoops = {};
+        for (const dev of device_list) {
+            if (dev.dev_type != null && dev.dev_addr != null && dev.dev_loop != null) {
+                const key = buildLoopKey(dev.dev_type, dev.dev_addr, dev.dev_loop);
+                devLoops[key] = dev.dev_data ?? null;
+            }
+        }
+        if (Object.keys(devLoops).length === 0) devLoops = null;
+    }
+
+    if (control_list.length > 0) {
+        if (!devLoops) devLoops = {};
+        for (const ctrl of control_list) {
+            const devLoop = ctrl.dev_loop ?? ctrl.loop;
+            if (ctrl.dev_type != null && ctrl.dev_addr != null && devLoop != null) {
+                const key = buildLoopKey(ctrl.dev_type, ctrl.dev_addr, devLoop);
+                devLoops[key] = { type_l: ctrl.type_l ?? null, type_h: ctrl.type_h ?? null };
+            }
+        }
+        if (Object.keys(devLoops).length === 0) devLoops = null;
+    }
+
+    // Build faulty_device_count from fault_list (full replacement)
+    let faultyDeviceCount = null;
+    if (fault_list.length > 0) {
+        faultyDeviceCount = fault_list.map(f => ({
+            dev_type: f.dev_type ?? null,
+            dev_addr: f.dev_addr ?? null,
+            dev_loop: f.dev_loop ?? null,
+            error_type: f.error_type ?? null,
+            error_data: f.error_data ?? null
+        }));
+    }
+
+    // If there's absolutely nothing to update, return null
+    if (devLoops === null && faultyDeviceCount === null && sys_lock_status == null) {
+        return null;
+    }
+
+    // Clamp hotel_id to INT2 range (-32768 to 32767) to match production schema
+    const validHotelId = (hotel_id >= -32768 && hotel_id <= 32767) ? hotel_id : 0;
+
+    return {
+        hotel_id: validHotelId,
+        room_id: String(room_id),
+        device_id: String(device_id),
+        ts_ms,
+        sys_lock_status: sys_lock_status ?? null,
+        dev_loops: devLoops,
+        faulty_device_count: faultyDeviceCount
+    };
+};

bls-rcu-action-backend/src/schema/kafkaPayload.js

@@ -22,6 +22,7 @@ const controlItemSchema = z.object({
   dev_type: z.number().int().optional(),
   dev_addr: z.number().int().optional(),
   dev_loop: z.number().int().optional(),
+  loop: z.number().int().optional(), // Kafka uses 'loop'
   type_l: z.number().int().optional(),
   type_h: z.number().int().optional()
 });

bls-rcu-action-backend/tests/status_batch_processor.test.js

@@ -0,0 +1,134 @@
+import { describe, it, expect, vi, beforeEach } from 'vitest';
+import { StatusBatchProcessor } from '../src/db/statusBatchProcessor.js';
+
+// Create a mock RoomStatusManager
+const createMockManager = () => ({
+    upsertBatch: vi.fn().mockResolvedValue(undefined),
+    testConnection: vi.fn().mockResolvedValue(true),
+    close: vi.fn().mockResolvedValue(undefined)
+});
+
+describe('StatusBatchProcessor', () => {
+    let mockManager;
+    let processor;
+
+    beforeEach(() => {
+        mockManager = createMockManager();
+        processor = new StatusBatchProcessor(mockManager, {
+            flushInterval: 50000, // Long interval so we control flush manually
+            maxBufferSize: 100
+        });
+    });
+
+    const makeUpdate = (overrides = {}) => ({
+        hotel_id: 1001,
+        room_id: '8001',
+        device_id: 'dev_001',
+        ts_ms: 1700000000000,
+        sys_lock_status: null,
+        dev_loops: null,
+        faulty_device_count: null,
+        ...overrides
+    });
+
+    it('should accept and buffer a status update', () => {
+        processor.add(makeUpdate());
+        expect(processor.buffer.size).toBe(1);
+    });
+
+    it('should ignore null/undefined updates', () => {
+        processor.add(null);
+        processor.add(undefined);
+        expect(processor.buffer.size).toBe(0);
+    });
+
+    it('should merge dev_loops for same device (different keys accumulate)', async () => {
+        processor.add(makeUpdate({
+            dev_loops: { '001010001': 100 }
+        }));
+        processor.add(makeUpdate({
+            dev_loops: { '001011002': 50 }
+        }));
+
+        expect(processor.buffer.size).toBe(1);
+
+        await processor.flush();
+
+        expect(mockManager.upsertBatch).toHaveBeenCalledTimes(1);
+        const rows = mockManager.upsertBatch.mock.calls[0][0];
+        expect(rows).toHaveLength(1);
+        expect(rows[0].dev_loops).toEqual({
+            '001010001': 100,
+            '001011002': 50
+        });
+    });
+
+    it('should overwrite dev_loops for same key (newer wins)', async () => {
+        processor.add(makeUpdate({
+            dev_loops: { '001010001': 100 }
+        }));
+        processor.add(makeUpdate({
+            dev_loops: { '001010001': 200 }
+        }));
+
+        await processor.flush();
+
+        const rows = mockManager.upsertBatch.mock.calls[0][0];
+        expect(rows[0].dev_loops['001010001']).toBe(200);
+    });
+
+    it('should take latest ts_ms across merges', () => {
+        processor.add(makeUpdate({ ts_ms: 100 }));
+        processor.add(makeUpdate({ ts_ms: 300 }));
+        processor.add(makeUpdate({ ts_ms: 200 }));
+
+        const entry = [...processor.buffer.values()][0];
+        expect(entry.ts_ms).toBe(300);
+    });
+
+    it('should replace faulty_device_count with newer value', () => {
+        processor.add(makeUpdate({
+            faulty_device_count: [{ dev_type: 1, error_type: 1, error_data: 1 }]
+        }));
+        processor.add(makeUpdate({
+            faulty_device_count: [{ dev_type: 2, error_type: 2, error_data: 0 }]
+        }));
+
+        const entry = [...processor.buffer.values()][0];
+        expect(entry.faulty_device_count).toEqual([{ dev_type: 2, error_type: 2, error_data: 0 }]);
+    });
+
+    it('should keep different devices separate in the buffer', () => {
+        processor.add(makeUpdate({ device_id: 'dev_001' }));
+        processor.add(makeUpdate({ device_id: 'dev_002' }));
+
+        expect(processor.buffer.size).toBe(2);
+    });
+
+    it('should clear buffer after flush', async () => {
+        processor.add(makeUpdate());
+        expect(processor.buffer.size).toBe(1);
+
+        await processor.flush();
+
+        expect(processor.buffer.size).toBe(0);
+        expect(mockManager.upsertBatch).toHaveBeenCalledTimes(1);
+    });
+
+    it('should NOT throw when upsertBatch fails', async () => {
+        mockManager.upsertBatch.mockRejectedValue(new Error('DB down'));
+
+        processor.add(makeUpdate());
+
+        // flush should not throw
+        await expect(processor.flush()).resolves.not.toThrow();
+
+        // Buffer should still be cleared even on error
+        expect(processor.buffer.size).toBe(0);
+    });
+
+    it('should do nothing when flushing empty buffer', async () => {
+        await processor.flush();
+        expect(mockManager.upsertBatch).not.toHaveBeenCalled();
+    });
+});

bls-rcu-action-backend/tests/status_extractor.test.js

@@ -0,0 +1,128 @@
+import { describe, it, expect } from 'vitest';
+import { extractStatusUpdate } from '../src/processor/statusExtractor.js';
+
+describe('StatusExtractor', () => {
+    const base = {
+        ts_ms: 1700000000000,
+        hotel_id: 1001,
+        room_id: '8001',
+        device_id: 'dev_001',
+        direction: '上报',
+        cmd_word: '0x36',
+        frame_id: 1,
+        udp_raw: 'test',
+        sys_lock_status: 0,
+        device_list: [],
+        fault_list: [],
+        control_list: []
+    };
+
+    it('should return null when payload is null/undefined', () => {
+        expect(extractStatusUpdate(null)).toBeNull();
+        expect(extractStatusUpdate(undefined)).toBeNull();
+    });
+
+    it('should return null when nothing to update (empty lists, no sys_lock)', () => {
+        const payload = {
+            ...base,
+            sys_lock_status: undefined,
+            device_list: [],
+            fault_list: [],
+            control_list: []
+        };
+        expect(extractStatusUpdate(payload)).toBeNull();
+    });
+
+    it('should return update when sys_lock_status is present even with empty lists', () => {
+        const result = extractStatusUpdate({
+            ...base,
+            sys_lock_status: 1
+        });
+        expect(result).not.toBeNull();
+        expect(result.sys_lock_status).toBe(1);
+        expect(result.hotel_id).toBe(1001);
+        expect(result.room_id).toBe('8001');
+        expect(result.device_id).toBe('dev_001');
+    });
+
+    it('should build dev_loops from device_list with 9-digit padded keys', () => {
+        const result = extractStatusUpdate({
+            ...base,
+            device_list: [
+                { dev_type: 1, dev_addr: 10, dev_loop: 1, dev_data: 100 },
+                { dev_type: 1, dev_addr: 11, dev_loop: 2, dev_data: 0 }
+            ]
+        });
+
+        expect(result).not.toBeNull();
+        expect(result.dev_loops).toEqual({
+            '001010001': 100,
+            '001011002': 0
+        });
+    });
+
+    it('should build dev_loops from control_list with type_l/type_h values', () => {
+        const result = extractStatusUpdate({
+            ...base,
+            control_list: [
+                { dev_type: 1, dev_addr: 10, dev_loop: 1, type_l: 0, type_h: 1 }
+            ]
+        });
+
+        expect(result).not.toBeNull();
+        expect(result.dev_loops).toEqual({
+            '001010001': { type_l: 0, type_h: 1 }
+        });
+    });
+
+    it('should merge device_list and control_list into dev_loops', () => {
+        const result = extractStatusUpdate({
+            ...base,
+            device_list: [
+                { dev_type: 1, dev_addr: 10, dev_loop: 1, dev_data: 100 }
+            ],
+            control_list: [
+                { dev_type: 2, dev_addr: 5, dev_loop: 3, type_l: 1, type_h: 2 }
+            ]
+        });
+
+        expect(result.dev_loops).toEqual({
+            '001010001': 100,
+            '002005003': { type_l: 1, type_h: 2 }
+        });
+    });
+
+    it('should build faulty_device_count from fault_list', () => {
+        const result = extractStatusUpdate({
+            ...base,
+            fault_list: [
+                { dev_type: 1, dev_addr: 10, dev_loop: 1, error_type: 2, error_data: 5 }
+            ]
+        });
+
+        expect(result).not.toBeNull();
+        expect(result.faulty_device_count).toEqual([
+            { dev_type: 1, dev_addr: 10, dev_loop: 1, error_type: 2, error_data: 5 }
+        ]);
+    });
+
+    it('should return null when identity fields are missing', () => {
+        expect(extractStatusUpdate({ ...base, hotel_id: null })).toBeNull();
+        expect(extractStatusUpdate({ ...base, room_id: '' })).toBeNull();
+        expect(extractStatusUpdate({ ...base, device_id: '' })).toBeNull();
+        expect(extractStatusUpdate({ ...base, ts_ms: null })).toBeNull();
+    });
+
+    it('should handle large dev_type/addr/loop values with proper zero-padding', () => {
+        const result = extractStatusUpdate({
+            ...base,
+            device_list: [
+                { dev_type: 241, dev_addr: 255, dev_loop: 999, dev_data: 42 }
+            ]
+        });
+
+        expect(result.dev_loops).toEqual({
+            '241255999': 42
+        });
+    });
+});

docs/plan-room-status-sync.md

@@ -0,0 +1,76 @@
+# Room Status 状态表同步方案
+
+## 1. 背景
+我们需要将 RCU Action Server 接收到的 Kafka 数据（主要是 `0x36` 上报和 `0x0F` 下发指令），实时同步提炼并写入快照表 `room_status.room_status_moment`。
+该表为设备实时状态表，主要用于存储每个房间/设备的最新状态，它可能由多个微服务共同维护。当前服务主要负责基于不固定长度的通用设备回调帧来更新对应的 `dev_loops` 及 `faulty_device_count` JSONB 字段。
+注意：本更新为非阻塞异步更新，旨在不拖累主力流水表（`rcu_action_events`）写入性能的同时，保证外部系统读状态表时的即时性。
+
+## 2. 表结构与唯一性分析
+**目标表**：`room_status.room_status_moment`
+**逻辑唯一键**：`(hotel_id, room_id, device_id)`
+
+**数据库设计策略**：
+- 为满足 PostgreSQL 的 `ON CONFLICT` 原子级并发合并（Upsert）能力，表中必需包含针对 `(hotel_id, room_id, device_id)` 的唯一约束 / 索引。
+- 快照表可能在独立数据库进行负载隔离，业务侧需提供其独立的 DB 池连接能力。
+
+## 3. 字段映射方案 (当前项目)
+
+不采用硬编码多列映射，转而采用 JSONB 结构进行灵活映射。
+
+| 源字段 (Kafka: `0x36`/`0x0F`) | 目标表字段 | 数据类型 | 更新逻辑 |
+| :--- | :--- | :--- | :--- |
+| `hotel_id`, `room_id`, `device_id`| `hotel_id`,`room_id`,`device_id` | 匹配 | 主键/查询键 |
+| `ts_ms` | `ts_ms` | BIGINT | 保存该快照的更新时间，取最新 |
+| `sys_lock_status` | `sys_lock_status` | SMALLINT | 只在kafka传入的该字段非空时覆盖 |
+| `device_list` (0x36) 或 `control_list` (0x0F) | `dev_loops` | JSONB | **合并策略 (Merge)**: <br/>提取每个元素的 `dev_type`, `dev_addr`, `dev_loop` 分别补零为 3 位并拼接成 9 位长作为 Key (如 `001023012`)，其 Value 取对应的状态数据。<br/>与数据库中原有 JSONB 执行 `\|\|` 运算进行增量覆盖合并。 |
+| `fault_list` (0x36) | `faulty_device_count`| JSONB | **替换策略 (Replace)**: <br/>由于 `0x36` 是上报此刻的全量故障清单，故直接将其整个存为 JSON Array 覆盖原字段即可。 |
+
+## 4. 写入策略与性能优化
+
+### 4.1 内存合并去重设计
+Kafka 瞬发同设备大批量跳变事件会引发严重的数据库并发写压。故采取**批处理聚合**配合**JSON 局部合并**的策略：
+- **`StatusBatchProcessor`** 会收集给定窗口期 (例如 `500ms`) 内产生的所有更新状态。
+- 使用 `Map<hotel_id+room_id+device_id, MergedStatusData>` 来暂存去重状态。
+- **深合并**：针对 `dev_loops` 更新，内存中会对同一个设备的后者对前者进行 Object 解析与替换合并，保证推到数据库的只有一条终态的最完整 SQL 语句。
+
+### 4.2 SQL UPSERT 模板
+利用 PostgreSQL `ON CONFLICT DO UPDATE SET` 自动完成字段保鲜与合并：
+```sql
+INSERT INTO room_status.room_status_moment (
+  guid, ts_ms, hotel_id, room_id, device_id,
+  sys_lock_status, dev_loops, faulty_device_count
+) VALUES (
+  $1, $2, $3, $4, $5,
+  $6, $7::jsonb, $8::jsonb
+)
+ON CONFLICT (hotel_id, room_id, device_id)
+DO UPDATE SET
+  ts_ms = GREATEST(room_status.room_status_moment.ts_ms, EXCLUDED.ts_ms),
+  sys_lock_status = COALESCE(EXCLUDED.sys_lock_status, room_status.room_status_moment.sys_lock_status),
+  dev_loops = CASE 
+    WHEN EXCLUDED.dev_loops IS NULL THEN room_status.room_status_moment.dev_loops
+    ELSE COALESCE(room_status.room_status_moment.dev_loops, '{}'::jsonb) || EXCLUDED.dev_loops
+  END,
+  faulty_device_count = COALESCE(EXCLUDED.faulty_device_count, room_status.room_status_moment.faulty_device_count)
+WHERE EXCLUDED.ts_ms >= room_status.room_status_moment.ts_ms;
+```
+
+### 4.3 代码修改计划
+为完全支持本项目的特性及独立解耦要求，接下来我们将实施以下新增及重构计划：
+
+1. **配置层 (`src/config/config.js`)**
+   - 补充 `roomStatusDb` 的独立环境配置项定义（如 `ROOM_STATUS_DB_HOST` 等），支持独立连入快照库以避免抢占写日志的大库连接池。
+
+2. **数据库管理器 (`src/db/roomStatusManager.js`)**
+   - 新建单独的管理器及连接池单例，专职处理向目的状态表的批量 Upsert 操作及报错封装，与原有的流水日志追加入库流程（`databaseManager.js`）从库底层隔离开来。
+
+3. **批量写入器 (`src/db/statusBatchProcessor.js`)**
+   - 专属于快照表的批量聚合任务列队类。
+   - 实现批量数据接收保护，以及在触发定时 Flush 执行前，依据前文设计的基于 `(hotel, room, device)` 唯一维度对于 `dev_loops` 的状态内存全合并算法。
+
+4. **状态转换提取器 (`src/processor/statusExtractor.js`)**
+   - 提取业务侧逻辑的编解码中间件功能。它负责消化并筛选被校验验证后的 `KafkaPayload` 模型数据，将结构里错综的 `device_list` 及 `fault_list` 提炼为符合合并规则的 `dev_loops`/`faulty_device_count` 结构扁平化独立快照对象。
+
+5. **主流程集成 (`src/processor/index.js` 及 `src/index.js`)**
+   - 不改变原先 `processKafkaMessage` 往 `BatchProcessor` (`rcu_action_events`) 无脑推送的过程。
+   - 在主流程处理完毕的数据流中，提取有效状态对象，并以并行非阻塞的形式推送至新的 `StatusBatchProcessor.add(...)`。以此做到即使状态库网络出现波动也能保护 Kafka 的持续稳定流转消费。

openspec/changes/2026-02-28-remove-online-status/spec.md

@@ -0,0 +1,13 @@
+# 移除 online_status 字段处理
+
+## 背景
+`room_status_moment` 表中的 `online_status` 字段不由本项目（RCU Action Server）管理。
+本项目不应从 `fault_list` 中推导 `online_status`，该字段由其他服务负责写入。
+
+## 变更范围
+- `src/processor/statusExtractor.js` — 移除 `online_status` 推导逻辑
+- `src/db/roomStatusManager.js` — Upsert SQL 移除 `online_status` 列
+- `src/db/statusBatchProcessor.js` — 合并逻辑移除 `online_status`
+- `tests/status_extractor.test.js` — 删除 `online_status` 相关测试用例
+- `tests/status_batch_processor.test.js` — 删除 `online_status` 相关测试用例
+- `docs/plan-room-status-sync.md` — 移除字段映射行和 SQL 中的 `online_status`

openspec/changes/2026-02-28-room-status-sync/spec.md

@@ -0,0 +1,34 @@
+# Room Status Moment 实时快照同步
+
+## 背景
+将 Kafka 消费的 RCU 设备状态/控制事件（`0x36` 上报、`0x0F` 下发），在写入流水日志表 `rcu_action_events` 的同时，并行提取关键状态信息，以 Upsert 方式同步到独立的快照表 `room_status.room_status_moment`。
+
+## 变更范围
+
+### 新增文件
+- `src/db/roomStatusManager.js` — 独立连接池 + Upsert SQL 封装
+- `src/db/statusBatchProcessor.js` — 快照专用的批量聚合处理器（内存合并去重）
+- `src/processor/statusExtractor.js` — 从 Kafka payload 提取 `dev_loops`/`faulty_device_count`
+- `tests/status_extractor.test.js` — statusExtractor 单元测试
+- `tests/status_batch_processor.test.js` — statusBatchProcessor 单元测试
+
+### 修改文件
+- `src/config/config.js` — 新增 `roomStatusDb` 配置段
+- `src/index.js` — 挂载 `StatusBatchProcessor` 并在消息处理流程中并行推送状态更新
+- `.env.example` — 补充 `ROOM_STATUS_DB_*` 环境变量示例
+
+### 缺陷修复 (Bug Fixes)
+- `src/db/statusBatchProcessor.js` — 添加 `isFlushing` 状态锁防止高并发带来的重复刷新及数据库死锁 (`ON CONFLICT` 互锁)。
+- `src/db/roomStatusManager.js` — 去除 `WHERE EXCLUDED.ts_ms >= room_status_moment.ts_ms` 条件限制，保证始终强制应用最新的状态覆盖。
+- `src/schema/kafkaPayload.js` & `src/processor/statusExtractor.js` — 修正 `control_list` 的提取逻辑，兼容 Kafka 实际传输中的 `loop` 字段。
+- `docs/plan-room-status-sync.md` — 明确 `sys_lock_status` 仅在 Kafka 传入的值非空时，才进行覆盖。
+
+## 设计约束
+1. 状态表写入失败**不得**阻塞主流水表写入和 Kafka offset 提交
+2. 同一批次中同设备多次更新需在内存中合并后再提交数据库
+3. `dev_loops` JSONB 使用增量合并 (`||`)，`faulty_device_count` 使用整体替换
+4. 快照表使用独立数据库连接池，与流水表连接池资源隔离
+
+## 参考文档
+- `docs/plan-room-status-sync.md`
+- `docs/room_status_moment.sql`