What's the usecase for mutable records in C#9?

Marius_1585 0
Viewed 1116

C# 9 adds records, finally! I've been waiting forever for this, it's supergreat! But I'm wondering what the point of having a record with { get; set; } is? It seems like records should be immutable and use { get; init; } for all their properties. Maybe I'm just too used to working with immutable data, but I don't get the point of mutable records. It just seems like it increases the likelihood of bugs in the code. Am I missing a really obvious and useful usecase?

record c#
1 Answers

To start with: a record is just a quick, short notation to declare a class with a couple of useful features (a copy constructor + cloning + hashing + comparison/equality) automatically added for free. But the end result is still a class, like any other. It is syntactic sugar, with actually two flavours.

The extra-short positional syntax makes all properties init-only:

record Person(string FirstName, string LastName);

The less-short nominal syntax makes the developer responsible to specify set or init for each property:

record Person
{
    public string FirstName { get; set; }
    public string LastName { get; init; }
}

The reason behind offering both notations is that there is no 'one-size-fits-all'. The more verbose nominal syntax allows to create a class with still very little code, with custom (im)mutability that the positional syntax does not offer, while still getting the copy constructor + cloning + hashing + comparison/equality for free.

user_12205500
Related