From 2aeef3e9bfff859b06952b1dc15fc07efbae8b8e Mon Sep 17 00:00:00 2001
From: Milky0217 <m943790568@163.com>
Date: Fri, 17 Apr 2026 23:03:27 +0800
Subject: [PATCH] =?UTF-8?q?docs:=20=E6=B7=BB=E5=8A=A0=E7=94=9F=E4=BA=A7?=
 =?UTF-8?q?=E7=8E=AF=E5=A2=83=E6=95=B0=E6=8D=AE=E5=BA=93=E8=BF=81=E7=A7=BB?=
 =?UTF-8?q?=E6=AD=A5=E9=AA=A4?=
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

- 新增二、生产环境数据库迁移章节
- 记录测试/生产环境数据量对比
- 添加备份和迁移操作步骤
- 添加 memberships/invitation_codes 建表语句
- 添加回滚方案
---
 IMPROVEMENTS.md | 108 ++++++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 108 insertions(+)

diff --git a/IMPROVEMENTS.md b/IMPROVEMENTS.md
index 797de9e..c6b63a8 100644
--- a/IMPROVEMENTS.md
+++ b/IMPROVEMENTS.md
@@ -57,6 +57,114 @@ interface WeatherData {
 
 ---
 
+## 二、生产环境数据库迁移
+
+### 环境信息
+
+| 环境 | 数据库名 | 数据量 |
+|------|---------|--------|
+| 测试 | `milkydata_dev` | users: 1, weather_data: 4, payment_orders: 6 |
+| 生产 | `milkydata` | users: 2358, weather_data: 6563, payment_orders: 0 |
+
+### 连接方式
+
+```bash
+ssh root@1panel-server
+docker exec -it 1Panel-postgresql-FtMo psql -U milky -d milkydata_dev  # 测试环境
+docker exec -it 1Panel-postgresql-FtMo psql -U milky -d milkydata     # 生产环境
+```
+
+### 当前 schema 差异
+
+| 表 | 测试环境 | 生产环境 | 差异 |
+|---|---------|---------|------|
+| users | ✅ | ✅ | 无 |
+| payment_orders | ✅ | ✅ | 无 |
+| weather_data | ✅ | ✅ | **生产缺少 `is_favorite` 列** |
+
+### 迁移步骤
+
+#### ⚠️ 重要：执行前必须备份
+
+```bash
+ssh root@1panel-server
+# 备份生产数据库
+docker exec 1Panel-postgresql-FtMo pg_dump -U milky milkydata > /tmp/milkydata_backup_$(date +%Y%m%d_%H%M%S).sql
+
+# 验证备份成功
+ls -la /tmp/milkydata_backup_*.sql
+```
+
+#### 步骤 1：添加缺失列
+
+生产环境执行（PostgreSQL 11+ 几乎无影响）：
+
+```sql
+-- 添加 is_favorite 列到 weather_data 表
+ALTER TABLE weather_data ADD COLUMN is_favorite BOOLEAN NOT NULL DEFAULT false;
+```
+
+#### 步骤 2：验证
+
+```sql
+-- 检查列是否添加成功
+SELECT column_name, data_type, is_nullable, column_default
+FROM information_schema.columns
+WHERE table_name = 'weather_data' AND column_name = 'is_favorite';
+```
+
+#### 步骤 3：未来新增 memberships 表
+
+当需要会员系统时，在生产环境执行：
+
+```sql
+-- 1. 创建 memberships 表
+CREATE TABLE memberships (
+    id SERIAL PRIMARY KEY,
+    user_id INTEGER UNIQUE REFERENCES users(id),
+    is_paid BOOLEAN DEFAULT false,
+    paid_expires_at TIMESTAMPTZ,
+    created_at TIMESTAMPTZ DEFAULT NOW(),
+    updated_at TIMESTAMPTZ DEFAULT NOW()
+);
+
+CREATE INDEX idx_memberships_user_id ON memberships(user_id);
+
+-- 2. 从 users 表同步现有数据
+INSERT INTO memberships (user_id, is_paid, paid_expires_at)
+SELECT id, is_paid, paid_expires_at FROM users
+ON CONFLICT (user_id) DO NOTHING;
+
+-- 3. 创建 invitation_codes 表
+CREATE TABLE invitation_codes (
+    id SERIAL PRIMARY KEY,
+    code VARCHAR(32) UNIQUE NOT NULL,
+    package_type VARCHAR(20) NOT NULL,
+    paid_days INTEGER NOT NULL,
+    source VARCHAR(20) DEFAULT 'invitation',
+    used_by INTEGER REFERENCES users(id),
+    used_at TIMESTAMPTZ,
+    created_at TIMESTAMPTZ DEFAULT NOW(),
+    expires_at TIMESTAMPTZ,
+    is_used BOOLEAN DEFAULT FALSE
+);
+
+CREATE INDEX idx_invitation_codes_code ON invitation_codes(code);
+```
+
+### 回滚方案
+
+如果需要回滚：
+
+```sql
+-- 删除新增的列/表
+ALTER TABLE weather_data DROP COLUMN is_favorite;
+DROP TABLE IF EXISTS memberships;
+DROP TABLE IF EXISTS invitation_codes;
+```
+
+---
+
 ## 一、代码质量与架构改进
 
 ### 1.1 代码组织问题