How do VB.NET anonymous types without key fields differ from C# anonymous types when compared?

Question

I'm scratching my head about this as I cannot understand why the following happens the way it does:

'//VB.NET
Dim product1 = New With {.Name = "paperclips", .Price = 1.29}
Dim product2 = New With {.Name = "paperclips", .Price = 1.29}

'compare product1 and product2 and you get false returned.

Dim product3 = New With {Key .Name = "paperclips", Key .Price = 1.29}
Dim product4 = New With {Key .Name = "paperclips", Key .Price = 1.29}

'compare product3 and product4 and you get true returned.

'//C#
var product5 = new {Name = "paperclips", Price = 1.29};
var product6 = new {Name = "paperclips", Price = 1.29};

//compare products 5 and 6 and you get true.

What is happening with products 1 and 2 that makes them not behave like products 5 and 6?

Answer 1

In C#, all properties of anonymous types behave as if they have the Key modifier in VB: the properties are read-only, and they're included in equality and hash code evaluation.

In VB, properties without the Key modifier are mutable, and are not used in the Equals/GetHashCode implementations.

From the Anonymous Type Definition documentation:

If an anonymous type declaration contains at least one key property, the type definition overrides three members inherited from Object: Equals, GetHashCode, and ToString. If no key properties are declared, only ToString is overridden. The overrides provide the following functionality:

• Equals returns True if two anonymous type instances are the same instance, or if they meet the following conditions:

  • They have the same number of properties.
  • The properties are declared in the same order, with the same names and the same inferred types. Name comparisons are not case-sensitive.
  • At least one of the properties is a key property, and the Key keyword is applied to the same properties.
  • Comparison of each corresponding pair of key properties returns True.

• GetHashcode provides an appropriately unique GetHashCode algorithm. The algorithm uses only the key properties to compute the hash code.

• ToString returns a string of concatenated property values, as shown in the following example. Both key and non-key properties are included.