使用dg库快速生成关系型数据CSV文件

dg是一个Go语言库，专门用于快速生成关系型数据的CSV文件。它特别适合测试数据生成、模拟数据场景等用途。下面我将详细介绍如何使用dg库。

安装dg库

首先需要安装dg库：

go get github.com/darwayne/dg

基本使用方法

1. 创建简单CSV文件

package main

import (
	"github.com/darwayne/dg"
	"os"
)

func main() {
	// 创建CSV生成器
	generator := dg.NewCSVGenerator(os.Stdout)
	
	// 定义表结构
	generator.AddTable("users", []dg.Column{
		{Name: "id", Type: dg.Int, AutoIncrement: true},
		{Name: "name", Type: dg.String},
		{Name: "email", Type: dg.String},
		{Name: "created_at", Type: dg.Time},
	})
	
	// 生成5条记录
	generator.Generate("users", 5)
}

2. 生成关联数据

dg的强大之处在于可以生成关联数据：

func main() {
	generator := dg.NewCSVGenerator(os.Stdout)
	
	// 定义用户表
	generator.AddTable("users", []dg.Column{
		{Name: "id", Type: dg.Int, AutoIncrement: true},
		{Name: "name", Type: dg.String},
		{Name: "email", Type: dg.String},
	})
	
	// 定义订单表，关联用户
	generator.AddTable("orders", []dg.Column{
		{Name: "id", Type: dg.Int, AutoIncrement: true},
		{Name: "user_id", Type: dg.Int, ForeignKey: "users.id"},
		{Name: "amount", Type: dg.Float, Min: 10.0, Max: 1000.0},
		{Name: "created_at", Type: dg.Time},
	})
	
	// 生成10个用户，每个用户有3-5个订单
	generator.GenerateWithRelationships(map[string]int{
		"users": 10,
		"orders": dg.PerParent(3, 5),
	})
}

3. 自定义数据生成器

你可以为特定列定义自定义生成逻辑：

func main() {
	generator := dg.NewCSVGenerator(os.Stdout)
	
	generator.AddTable("products", []dg.Column{
		{Name: "id", Type: dg.Int, AutoIncrement: true},
		{Name: "name", Type: dg.String, Generator: func() interface{} {
			// 自定义产品名称生成器
			products := []string{"Laptop", "Phone", "Tablet", "Monitor", "Keyboard"}
			return products[rand.Intn(len(products))] + " " + strconv.Itoa(rand.Intn(1000))
		}},
		{Name: "price", Type: dg.Float, Min: 50.0, Max: 2000.0},
	})
	
	generator.Generate("products", 20)
}

高级功能

1. 数据分布控制

generator.AddTable("users", []dg.Column{
	{Name: "status", Type: dg.String, Distribution: map[interface{}]float64{
		"active": 0.7,
		"inactive": 0.2,
		"banned": 0.1,
	}},
})

2. 唯一约束

generator.AddTable("users", []dg.Column{
	{Name: "username", Type: dg.String, Unique: true},
})

3. 生成到文件

func main() {
	file, err := os.Create("output.csv")
	if err != nil {
		panic(err)
	}
	defer file.Close()
	
	generator := dg.NewCSVGenerator(file)
	// ... 配置表结构和生成数据
}

完整示例

package main

import (
	"github.com/darwayne/dg"
	"os"
	"math/rand"
	"time"
)

func main() {
	rand.Seed(time.Now().UnixNano())
	
	file, err := os.Create("ecommerce_data.csv")
	if err != nil {
		panic(err)
	}
	defer file.Close()
	
	generator := dg.NewCSVGenerator(file)
	
	// 用户表
	generator.AddTable("customers", []dg.Column{
		{Name: "customer_id", Type: dg.Int, AutoIncrement: true},
		{Name: "first_name", Type: dg.String, MinLen: 3, MaxLen: 10},
		{Name: "last_name", Type: dg.String, MinLen: 3, MaxLen: 15},
		{Name: "email", Type: dg.String, Format: "{first_name}.{last_name}@example.com"},
		{Name: "join_date", Type: dg.Time, Min: time.Now().AddDate(-2, 0, 0), Max: time.Now()},
		{Name: "status", Type: dg.String, Distribution: map[interface{}]float64{
			"active": 0.8,
			"inactive": 0.2,
		}},
	})
	
	// 产品表
	generator.AddTable("products", []dg.Column{
		{Name: "product_id", Type: dg.Int, AutoIncrement: true},
		{Name: "name", Type: dg.String, Generator: productNameGenerator},
		{Name: "category", Type: dg.String, Values: []interface{}{"Electronics", "Clothing", "Home", "Books"}},
		{Name: "price", Type: dg.Float, Min: 5.0, Max: 500.0},
		{Name: "stock", Type: dg.Int, Min: 0, Max: 1000},
	})
	
	// 订单表
	generator.AddTable("orders", []dg.Column{
		{Name: "order_id", Type: dg.Int, AutoIncrement: true},
		{Name: "customer_id", Type: dg.Int, ForeignKey: "customers.customer_id"},
		{Name: "order_date", Type: dg.Time, Min: time.Now().AddDate(-1, 0, 0), Max: time.Now()},
		{Name: "status", Type: dg.String, Values: []interface{}{"pending", "shipped", "delivered", "cancelled"}},
	})
	
	// 订单详情表
	generator.AddTable("order_items", []dg.Column{
		{Name: "item_id", Type: dg.Int, AutoIncrement: true},
		{Name: "order_id", Type: dg.Int, ForeignKey: "orders.order_id"},
		{Name: "product_id", Type: dg.Int, ForeignKey: "products.product_id"},
		{Name: "quantity", Type: dg.Int, Min: 1, Max: 5},
		{Name: "unit_price", Type: dg.Float, ForeignCol: "products.price"},
	})
	
	// 生成数据
	generator.GenerateWithRelationships(map[string]int{
		"customers": 50,
		"products": 30,
		"orders": dg.PerParent(3, 10), // 每个客户3-10个订单
		"order_items": dg.PerParent(1, 5), // 每个订单1-5个商品
	})
}

func productNameGenerator() interface{} {
	prefixes := []string{"Pro", "Super", "Mega", "Ultra", "Smart"}
	products := []string{"Laptop", "Phone", "Tablet", "Watch", "Camera", "Speaker"}
	suffixes := []string{"X", "Pro", "Max", "Plus", "2023"}
	
	return prefixes[rand.Intn(len(prefixes))] + " " + 
		products[rand.Intn(len(products))] + " " + 
		suffixes[rand.Intn(len(suffixes))]
}

这个示例会生成一个完整的电子商务数据集，包含客户、产品、订单和订单详情表，所有表之间都有正确的关联关系。

dg库提供了丰富的数据生成选项，可以满足大多数测试数据生成需求。通过合理配置，你可以快速生成符合业务逻辑的关联数据。