Added: LDM notes for April 6th and April 8th, 2020

[agocke]

April 8, 2020

C# Language Design Notes for April 8, 2020

1. e is dynamic pure null check
2. Target typing ?:
3. Inferred type of an or pattern
4. Module initializers

April 6, 2020

C# Language Design Notes for April 6, 2020

1. Record Monday: Init-only members

[jnm2]

@RikkiGibson The prototype link in https://github.com/RikkiGibson/csharplang/blob/module-initializers/proposals/module-initializers.md is broken. It should be https://github.com/jnm2/roslyn/tree/module_initializer.

[RikkiGibson]

Thanks, updated!

[jnm2]

https://github.com/dotnet/csharplang/blob/master/meetings/2020/LDM-2020-04-08.md#module-initializers

Conclusion

Let's let any static method be a module initializer, and mark that method using a well-known attribute. We'll also allow multiple module initializer methods, and they will each be called in a reserved, but deterministic order.

I like this a lot.

[SingleAccretion]

What would be the appropriate issue to comment on init syntax?

[iam3yal]

@SingleAccretion Don't know whether there is a specific discussion about it but you're free to create one, worst-case scenario it would be a dup.

[MgSam]

I didn't see it addressed in the design notes or the proposal- for properties, why has the design moved from init being used in place of the word set to being a modifier on set? I prefer the terser version where its used in place of set.

Otherwise, I love the init proposal. It is a syntax-light way of adding substantial new functionality with broad usage scenarios to C#. I think it likely is more broadly useful than even records are.

[SingleAccretion]

@MgSam Presumably to keep it a "modifier" keyword, as it can also be applied to fields.

To be honest, I am skeptical about init (not the concept, but the word itself). It feels out of place for C#. I think I am going to open an issue about it as soon as I have enough time to properly analyze the alternatives.

[qrli]

On init syntax, I feel reusing readonly can be an option:

    public readonly int P { get; set; }

It feels similar to readonly fields, which can be set at initialization, but not later.

[SingleAccretion]

@qrli This would not work for fields (which enable init _value; Value { init set; } pattern).

[333fred]

Also, readonly can already be applied to getters and setters and means something completely different.