最近看到一個小眾的套件有點意思，就測試了一下 叫做 Facet.Net ，這套件有一些方便的地方
但是也有一些限制，這邊就簡單介紹一下，我覺得在製作一些 API 的時候會是有用且方便的..
這套件主要是在用於 "扁平化" 你的資料模型，當你的模型是巢狀的時候他會進行把巢狀的物件拉到第一層來
直接看案例，這邊是我原本的模型
<![CDATA[
public class Order
{
public int Id { get; set; }
public Customer Customer { get; set; } = default!;
public List<orderline> Lines { get; set; } = new();
public DateTime CreatedAt { get; set; }
}
public class Customer
{
public int Id { get; set; }
public string Name { get; set; } = "";
public Address Address { get; set; } = default!;
}
public class Address
{
public string City { get; set; } = "";
public string Country { get; set; } = "";
}
public class OrderLine
{
public string ProductName { get; set; }
public decimal Price { get; set; }
public int Quantity { get; set; }
}
</orderline>]]>
這邊我先給出，將這資料序列畫成 JSON 之後的長相
<![CDATA[
[
{
"Id": 1,
"Customer": {
"Id": 1,
"Name": "許當麻",
"Address": {
"City": "台北市",
"Country": "台灣"
}
},
"Lines": [
{
"ProductName": "筆記型電腦",
"Price": 28000,
"Quantity": 1
},
{
"ProductName": "無線滑鼠",
"Price": 1200,
"Quantity": 1
}
],
"CreatedAt": "2026-01-14T11:59:31.505265+08:00"
},
{
"Id": 2,
"Customer": {
"Id": 2,
"Name": "林怡君",
"Address": {
"City": "新北市",
"Country": "台灣"
}
},
"Lines": [
{
"ProductName": "27吋螢幕",
"Price": 9500,
"Quantity": 1
}
],
"CreatedAt": "2026-01-13T11:59:31.5133105+08:00"
}
]
]]>
<![CDATA[
[Flatten(typeof(Order))]
public partial class OrderFlatten
{
}
]]>
測試轉換結果
<![CDATA[
var flattenData = FakeOrderData.FakeData1
.AsQueryable()
.Where(x => x.Id <= 2)
.Select(OrderFlatten.Projection)
.ToList();
Console.WriteLine(JsonConvert.SerializeObject(flattenData));
]]>
輸出結果
<![CDATA[
[
{
"Id": 1,
"CustomerId": 1,
"CustomerName": "許當麻",
"CustomerAddressCity": "台北市",
"CustomerAddressCountry": "台灣",
"CreatedAt": "2026-01-14T11:59:31.505265+08:00"
},
{
"Id": 2,
"CustomerId": 2,
"CustomerName": "林怡君",
"CustomerAddressCity": "新北市",
"CustomerAddressCountry": "台灣",
"CreatedAt": "2026-01-13T11:59:31.5133105+08:00"
}
]
]]>
再來他還有一些其他常見的用法，如果你在進行關聯你可以透過調整 Arrtibute 把子項目的 Id 不要顯示出來
<![CDATA[
[Flatten(typeof(Order), IgnoreNestedIds = true)]
public partial class OrderFlattenIgonreIds
{
}
//使用方法
var flattenIgonreNestedIdsData = FakeOrderData.FakeData1
.AsQueryable()
.Where(x => x.Id <= 2)
.Select(OrderFlattenIgonreIds.Projection)
.ToList();
Console.WriteLine(JsonConvert.SerializeObject(flattenIgonreNestedIdsData));
]]>
輸出結果
<![CDATA[
[
{
"Id": 1,
"CustomerName": "許當麻",
"CustomerAddressCity": "台北市",
"CustomerAddressCountry": "台灣",
"CreatedAt": "2026-01-14T11:59:31.505265+08:00"
},
{
"Id": 2,
"CustomerName": "林怡君",
"CustomerAddressCity": "新北市",
"CustomerAddressCountry": "台灣",
"CreatedAt": "2026-01-13T11:59:31.5133105+08:00"
}
]
]]>
不過，如果如果你是要比較複雜要多一個 多出來的屬性，還是得要自己作 mapping 自己寫一個 Projection
<![CDATA[
[Flatten(typeof(Order))]
public partial class OrderFlattenWithTotalAmount
{
public decimal TotalAmount { get; set; }
public static Expression<func orderflattenwithtotalamount="" rder="">> WithTotalAmount =>
o => new OrderFlattenWithTotalAmount
{
// Flatten 對應的基本欄位（你要「照名填」）
Id = o.Id,
CustomerId = o.Customer.Id,
CustomerName = o.Customer.Name,
CustomerAddressCity = o.Customer.Address.City,
CustomerAddressCountry = o.Customer.Address.Country,
CreatedAt = o.CreatedAt,
TotalAmount = o.Lines.Sum(l => l.Price * l.Quantity)
};
}
//使用方法
var customerData = FakeOrderData.FakeData1
.AsQueryable()
.Where(x => x.Id <= 2)
.Select(OrderFlattenWithTotalAmount.WithTotalAmount)
.ToList();
Console.WriteLine(JsonConvert.SerializeObject(customerData));
</func>]]>
輸出結果
<![CDATA[
[
{
"TotalAmount": 29200,
"Id": 1,
"CustomerId": 1,
"CustomerName": "許當麻",
"CustomerAddressCity": "台北市",
"CustomerAddressCountry": "台灣",
"CreatedAt": "2026-01-14T11:59:31.505265+08:00"
},
{
"TotalAmount": 9500,
"Id": 2,
"CustomerId": 2,
"CustomerName": "林怡君",
"CustomerAddressCity": "新北市",
"CustomerAddressCountry": "台灣",
"CreatedAt": "2026-01-13T11:59:31.5133105+08:00"
}
]
]]>
做個結論，Facet.Net 解決的是"結構型轉換"，而不是"查詢邏輯" 這兩件事一開始就該分開看，在快速處理一些 API 輸出時，你有扁平化的需求，他是一個好東西，但是他遇到是 List<object> ，或是陣列型的屬性時
他會略過不處理，所以在一些 "圖方便" 的狀況他是一個不錯的選擇，但是如果要需要客制化還是得寫 expreesion ，你說他雞肋嗎
倒也不至於，只是在一些狀況下的確可以快速產生一個需要的物件去輸出，但是現實雖然常常都是比較複雜的..
--
The bug existed in all possible states.
Until I ran the code.
如果這篇文章有幫助到您幫我分享一下，讓我有寫下去的動力...
