[Proposal]: Collection Expressions Next (C#13 and beyond) #7913
Comments
|
I'd like to see some way of calling the constructor of a type and/or setting some property during initialization. The first thing that came to mind is the comparer of a dictionary, but I'm sure there are other use-cases. Either way, collection expressions are amazing, and support for natural types and inline expressions would make them that little bit better! |
|
@KennethHoff that's in the list, as part of hte dictionary-expression exploration work. Thanks :) |
|
I'm not sure if the following proposal is too crazy, so I will describe it here quickly, as it's related to this topic: Imagine that I have a Ideally I would like to spread the array of ints into the array of strings while also calling If But since there is no such conversion, I would like to write something like Spread operator as a first-class citizen.So basically the spread operator The spread unary operator may be applied to any other expression, resulting in a spread_expression. When that happens, an implicit The spread_expression then behaves, for the purposes of member-lookup and overload-resolution, as an expression whose type is the element-type of the original enumerable. Because of that, one may invoke any members the element type might have, as well use the spread_expression on a method that takes element-type as argument. These invocations will then be inserted inside the invisible The result of a member invocation performed on, or taking the spread_expression as an argument, is also itself a spread_expression, whose element-type is the return type of the invoked member, if not Finally, one will want to capture the result of all these method invocations on each element of the original collection. Therefore, any spread_expression can be used regularly as a spread_element in a collection expression, with the existing rules. Problems:
|
|
C# already has a query comprehension syntax, LINQ. |
True, but, that's not really an argument. If you would argue against all proposals saying "it can already be done in some way", then none would ever be accepted. Lists could already be created with LINQ or with initializers, yet collection expressions were introduced. And they have the added benefits of duck typing/nice syntax/good performance. LINQ, on the other hand, is interface-based and makes use of delegates and anonymous objects. If one could perform some simple transformations through the use of spread operator, everything would be inserted directly in the caller method, with no delegates or closures. There isn't any optimization better than that, LINQ would almost be obsolete. |
|
See: #7634 |
That's where supporting extension methods would help. As you could write:
|
I assume you could also do this? [ ..[1, 2, 3].Select(i => i.ToString()) ] |
|
@KennethHoff yes. |
|
@KennethHoff Yes, or |
|
I suggest widening support for collection builders to accept other buildable collections. For example I can choose to wrap a HashSet or a List or an Array or whatever. It feels so weird to accept a Span for such types. |
|
@En3Tho Can you give an example? |
public class ArrayWrapperBuilder
{
// This works
public static ArrayWrapper<T> Create<T>(ReadOnlySpan<T> values)
{
return new(values.ToArray());
}
// I want this to work instead. Array is already a creatable collection, so just let compiler create it and use directly here
// Imagine if it was a List<T> or HashSet<T> wrapper. Even more allocations while compiler is perfectly able to create this kind of collection directly.
public static ArrayWrapper<T> Create<T>(T[] values)
{
return new(values);
}
}
[CollectionBuilder(typeof(ArrayWrapperBuilder), nameof(ArrayWrapperBuilder.Create))]
public readonly struct ArrayWrapper<T>(T[] array) : IEnumerable<T>
{
public T[] Array => array;
public IEnumerator<T> GetEnumerator()
{
throw new NotImplementedException();
}
IEnumerator IEnumerable.GetEnumerator()
{
return GetEnumerator();
}
}
public static class ArrayWrapperCreator
{
public static ArrayWrapper<T> GetWrapper<T>() => [default, default, default];
} |
|
I'm glad the lack of Something like foreach (int i in [1, 2, 3])was literally the first thing I tried with collection literals and I was surprised why it was failing to compile with CS9176 saying there was no target type. It basically provides exactly the same amount of information as foreach (var i in (IEnumerable<int>)[1, 2, 3])which happily compiles today (though I would rather use some old-school array initialization instead of the bulky cast). I think this could be added even independently from natural types. |
|
One addition that we would like to see is support for multi-dimensional collection literals. These are important when working with tensor libraries, e.g. here's how simple tensors are defined using the pytorch library: data = [[1, 2],[3, 4]]
x_data = torch.tensor(data)The above form isn't feasible using today's C# collection literals, so supporting it without some kind of language support seems unlikely. One possibility is that we could reuse nested collection literal syntax to construct multi-dimensional collections. TL;DR it should be possible to extend the collection builder pattern to recognize factory methods such as public static T[,] Create<T>(ReadOnlySpan<T> values, int nDim, int mDim);and then be able to define 2-D arrays like so T[,] values = [[1, 0], [0, 1]];In principle, it should be possible for the compiler to infer the rank and dimensions and detect shape mismatches (e.g. something like What's more interesting though is that by reusing nested collection syntax there is no inherent limit on the supported number of dimensions, and the number of dimensions doesn't need to be fixed for a given type. We could for instance support builder methods of shape public static Tensor<T> Create<T>(ReadOnlySpan<T> values, params ReadOnlySpan<int> dimensions);Which should let you specify the following 2x2x2 tensor: Tensor<int> tensor = [[[0, 0], [0, 0]], [[1, 0], [0, 1]]]; |
|
Thanks @eiriktsarpalis . The working group is discussing this. I'll add you to that. |
|
Having read the (hundreds) of comments on the original Collection Expression proposal, and noting the large number of natural type immutable versus mutable comments, what about if the type was immutable if the Collection Expression consisted only of immutable literals? but tt[1] = "z"; // fine |
|
@NetMage Why would |
|
Greetings to everyone. Sorry in advance if the topic I am asking about was already discussed somewhere. In that case I would be very grateful if you share the link to such discussion with me. I usually do not participate in the proposal discussions, this is a first one for me. So, please excuse me if I don't follow some workflow for proposal discussions. I decided to reach out after I worked a bit with collection expressions in C# 12, investigated the generated code for them, and found two cases that I personally found confusing. Here is the first one: Infinite generators. Suppose that you have some kind of a collection generator that can generate an infinite private static IEnumerable<int> OddNumbers()
{
int current = 1;
while (true)
{
yield return current;
current+=2;
}
}This can be used with LINQ: var oddNumbers = OddNumbers();
IEnumerable<int> oddNumberWithMinus1 = [-1, .. oddNumbers];
// we will never get hereUnexpectedly, this program hangs. The investigation of the generated code shows that the compiler generates the code that internally fully materializes the collection in memory: IEnumerable<int> ints1 = Program.OddNumbers();
List<int> items = new List<int>();
items.Add(-1);
foreach (int num in ints1)
items.Add(num);
IEnumerable<int> ints2 = (IEnumerable<int>) new \u003C\u003Ez__ReadOnlyList<int>(items);Because of this the program hangs - it's impossible to materialize an infinite collection. So my problems are:
However, the term collection was used for I think, it will be difficult if even possible to write an analyzer to warn about such cases, but at least the documentation should be very clear about them.
Could this scenario receive some extra support in C# 13? Because the extra allocation here could be removed with standard LINQ methods like var oddNumbers = OddNumbers();
IEnumerable<int> x = oddNumbers.Prepend(-1);
Console.WriteLine(string.Join(" ", x.Take(3))); // prints -1 1 3Sorry if such proposal was already considered and thanks in advance. |
|
Collection expressions aren't a comprehension language or intended as an alternative to LINQ, they always fully materialize spreads. I do agree that the documentation should probably call this out more clearly. |
Yes. It's a core part of the design. Linq is already there for comprehensions. Collection expressions exist intentionally to produce fully materialized collections. |
Thanks for the feedback, but this was an intentional decision. We do not want these collections to be lazy (and then have expensive enumeration semantics, or have them redo computation each time you enumerate). We have query comprehensions for that already. These collections are intended to be fully materialized, so you know that the final collection you get is cheap, efficient and finite. |
|
The second scenario I have encountered is somewhat artificial and definitely does not follow best design practices. Misuse of the "Count" property with duck typing The scenario involves the collection expression like this: public static IEnumerable<int> PrependOne(<Some integer collection type> s) => [1, ..s];The generated code for the method looks like this: int num1 = 1;
<collection type> myCollection = s;
int index1 = 0;
int[] items = new int[1 + myCollection.Count];
items[index1] = num1;
int index2 = index1 + 1;
foreach (int num2 in myCollection)
{
items[index2] = num2;
++index2;
}
return (IEnumerable<int>) new <>z__ReadOnlyArray<int>(items);The C# compiler is very clever, it attempts to generate the most optimized code relying on the However, while C# compiler cheats while using this size calculation. Everything is OK when the collection explicitly states that it can provide Now consider this example. Suppose I have the following custom collection: public class MyCollection : IEnumerable<int>, IEnumerable<string>
{
private readonly int[] _ints;
private readonly string[] _strings;
internal int Count => _strings.Length;
public MyCollection(int[] ints, string[] strings) => (_ints, _strings) = (ints, strings);
public IEnumerator<int> GetEnumerator() => ((IEnumerable<int>)_ints).GetEnumerator();
IEnumerator IEnumerable.GetEnumerator() => _ints.GetEnumerator();
IEnumerator<string> IEnumerable<string>.GetEnumerator() => ((IEnumerable<string>)_strings).GetEnumerator();
}The example is artificial and it's clearly not the best design. But similar code appears sometimes, or can live in an old legacy code base. Now, suppose I'm using this custom collection in a collection expression: public static IEnumerable<int> PrependOne(MyCollection s) => [1, ..s];
//...
var myCollection = new MyCollection([1, 2, 3, 4], ["string"]);
IEnumerable<int> modified = PrependOne(myCollection);And unexpectedly I received This looks confusing to me, no written code has any access by index. It requires for developer to know what code is generated by the compiler behind the nice syntax sugar. I can't say that I really like this compiler trick even if it is legal and probably brings some performance benefits and less allocations in case of value types. The compiler assumes a contract in a place where it does not exists (the whole idea of duck typing). I know that C# already did this before, for example with |
|
@HaloFour , @CyrusNajmabadi thank you for your response! You both very clearly confirmed that the feature was designed to materialize collections and works only on finite collections. |
We made explicit choices with collection expressions to assume that people write well-behaved and sensible types. We want the optimizations to hit the broadest set of cases. It is understood that a non-well-behaved type may then have problems. But we're optimizing the lang design for the literal 99.999% case, at the cost of these strange outliers. Our recommendation if you do have types like these is to write analyzers to block them off with collection-exprs. |
|
@CyrusNajmabadi that is understandable. I do understand at least part of the reasons why the duck typing was used. For example, you just introduced a new way to classify types - well-behaved. Is there a formal definition for a well behaved type? I may consider the collection from my example to be well-behaved, why not? It does not break any contracts it explicitly states, only some implicit contracts that were later imposed by the new version of compiler. |
Sure. Feel free to file doc bugs. :)
Yes. If you are a collection type (defined in our spec), and you supply a .Count, then enumerating you should produce the same number of elements. Similarly, if you have an indexer, and you index into the type from |
Because the Count and GetEnumerator refer to totally different sequences. Again, this is vastly out of the normal case for collection use in reality. :) We are being pragmatic here. The 99.999% case is well behaved collections. Sacrificing the value we get on the normal case for the ecosystem for strange types like this would be cutting off our nose to spite our face. |
And there is nothing in the documentation that clarifies this and prevents the feature from being misused. Moreover, collection expressions definitely overlap with
I definitely will. I have read more thoroughly the documentation and feature specs and I'm sorry to say this but I feel that the current documentation is in an awful state. A lot of things like collection materialization or duck typing are implicit or not mentioned at all. Many are described in "feature specs" which contain too much details about implementation and at the same time explicitly state that the actual implementation may be different which undermines their value. |
Feel free to contribute doc fixes or file issues. This is all open source :-) This is the repo for the design and specification of the feature. Docs are not done here. Feedback on that should go to the docs team. Thanks! |
|
You can always be consistent and put the type on the RHS, if that's the goal. |
|
If we won't get natural type in C# 13 (I hope we will), can at least this be supported in C# 13?: foreach (var number in [1, 2, 3])
{
Console.WriteLine(number);
}Here the compile could decide if it wants to use stackalloc and Span or Array |
I would be happy even if I had to specify |
|
I wonder if natural type enables slightly weird EF var results =
from x in items
where ["cat", "dog"].Contains(x.name)
select x; |
If they decide that natural type is |
|
Why is it intuitive to be an array? Why not |
|
As expected, not only me confused why there is not a natural type. |
|
|
What would the corresponding natural type be for a dictionary literal? |
So here's the thing. That's your intuition, and it's not universal. I think that Why? The performance difference between Performance is an important metric - but it's far from the only one, and in many situations not the most important by far. Aside: I've achieved many significant performance increases by rewriting array laden code to use At the risk of misquoting the aphorism:
My point here is not that the natural type should be Your intuition says array. |
Isn't that better? more readable? |
var b = [1, 2, 3, ..]; // List<int>This syntax simply involves lying to the reader. Good code doesn't require the reader to overcome deception to understand it. |
Most people would think |
|
I assume that is only an illustrative example and not shorthand for |
I don't disagree but I also see another interpretation that could be bad, which is to think that the array is just an infinite sequence and we are using a shorthand there to mean var b = [1, 2, 3, 4, 5, 6, 7, 8, 9, 10 .....;Yes, I know the symbol is not an ellipsis but still... just sharing this potential interpretation that just came to my mind when looking at it. You could even make a relation to how Excel generates values based on a previous sequence. |
|
@julealgon earlier in this thread it was already established that infinite sequences are not supported by collection expressions. They always materialize collections and this is by design: |
I never implied infinite sequence expressions were supported. I was just making a point that, from the outside, one could look into that syntax and think it was an infinite sequence. I was making the argument for readability only, in the sense that it could be misleading potentially. |
|
In C++ there is the std::initializer_list which is the resulting type of an expression like this: I can imagine that due to the static type system a similar approach could also work for C#. |
|
This might not fit into how C# natural types normally work, but here's a wild suggestion: How about making the natural type of a collection expression into just "a collection expression of the element type" in the sense that it is the type that it's used in. Here's an example: var coll = [1,2,3]; // Currently typed as simply "collection expression of ints" - basically just a recipe.
PrintNumbers(coll); // From this point on, because it was used as a `int[]` it's now an `int[]`.
void PrintNumbers(int[] numbers) { ... }The previous example is identical to the following, except you also get to keep the reference in the PrintNumbers([1,2,3]);
void PrintNumbers(int[] number) { ... }var coll = [1,2,3]; // Also just an "collection expression of ints"
foreach (var num in coll) // First used in a foreach loop, so we'll use the most efficient type here, which is `ReadOnlySpan<int>`.
{
...
}var coll = [1,2,3]; // Never used, so this never materializes.This natural type would be decided at compile time, so you can always hover over it in the IDE to see what it chose. I'm not sure what it means for the last example though; An unused local. |
|
@KennethHoff what if you never pass it to another method where there is a target type? What happens when you declare the I don't think your proposal would work well because you don't always have an obvious target-type to base the decision when coding. |
|
@KennethHoff That's an approach we've considered, and it still might come in handy. However, it brings up questions of teleporting the materialization to a specific type when there are multiple usages, which can be hard to reason about. |
I don't understand this case. Could you give an example? Assuming you mean "what if you never pass it to another method" then it's the last example in my original comment; It's a noop similar to how primary constructors work for non-records.
If you don't use it anywhere it doesn't have a type and therefore doesn't have any members. If you do use it elsewhere then it has the type of whatever "elsewhere" is, and you could use those members. var coll1 = [1,2,3];
var coll2 = [2,4,6];
var coll3 = [3,6,9];
coll1.ToString(); // Compile error. Cannot call ToString on an object of type "collection expression of ints"
coll2.ToString(); // Calls ToString on List<int>.
coll3.ToString(); // Calls ToString on IEnumerable<int>, which forwards to the synthesized type for IEnumerable<int> that currently exists. Partially UB.
PrintNumbers1(coll2); // As this is the first time coll2 is referenced after its declaration, coll2 is now retroactively typed as List<int>.
PrintNumbers2(coll3); // As this is the first time coll3 is referenced after its declaration, coll3 is now retroactively typed as IEnumerable<int>.
PrintNumbers1(coll3); // Compile error. Cannot implicitly convert IEnumerable<int> to List<int>
void PrintNumbers1(List<int> numbers) { ... }
void PrintNumbers2(IEnumerable<int> numbers) { ... }This does suffer from the ol' "spooky action at a distance" problem where changing something can break something seemingly unrelated. Say you changed the This interpretation also naturally adds some optimizations that you maybe otherwise couldn't've done, like simply never materializing it if it's not needed: var coll = [1,2,3];
foreach (var item in coll)
{
Console.WriteLine(item);
}Because this was only used inside a foreach loop, we can simply compile this as this: Console.WriteLine(1);
Console.WriteLine(2);
Console.WriteLine(3); |
var v = [1, 2, 3];
v.Add(4); |
Compile error. Cannot call Add on "Collection expression of ints". This however would work: List<int> DoThing()
{
var v = [1,2,3];
v.Add(4);
return v; // By virtue of being the return value - which is clearly defined - `v` is retroactively typed as List<int>
}I'm not saying this is particularly clear semantics - especially for human readers - but I do think it's at least consistent; First unambiguous reference defines the type, retroactively. That would also mean that changing the previous example to any of the following would indeed make it no longer compile: List<int> DoThing()
{
var v = [1,2,3];
v.Add(4);
DoThingWithArray(v);
return v; // int[] is not implicitly convertible to List<int>
}
void DoThingWithArray(int[] numbers) { ... }IEnumerable<int> DoThing()
{
var v = [1,2,3];
v.Add(4); // There is no `Add` on `IEnumerable<int>`
return v;
} |
This would lead to terrible dev experience. I would be ok if the compiler used context to decide the type, but there should be a type right away for when context is not known upfront. If someone is just typing code, like this: var stuff = [1, 2, 3];
stuff.They should get intellisense for something. Sometimes, context will just be completely insufficient: var stuff = [1, 2, 3];
var something = stuff.ToString();What is the type of You are making assumptions that there will always be some target collection type that can be inferred from usage, which is really not true at all. |
|
@KennethHoff: Would this work in your model? void Imply(this List<int> a, List<int> b) {} // Public extension method visible to DoThing
List<int> DoThing()
{
var a = [1];
var b = [2];
b.Imply(a); // Method not known yet
var c = [3];
c.Imply(b); // Method not known yet
return c; // Implies c: List<int>
// -> c.Imply(b) implies b: List<int>
// -> b.Imply(a) implies a: List<int>
}As much as I like type inference, this just feels like a half-assed solution that feels bad for both the Hindley-Milner crowd and the Grug brained devs. |
|
Just want to say; I do not think my suggestion is good. It has terrible DX. When it comes to @nuiva's question: void Imply(this List<int> a, List<int> b) {} // Public extension method visible to DoThing
List<int> DoThing()
{
var a = [1]; // a is unknown.
var b = [2]; // b is unknown.
b.Imply(a); // b is unknown. a will be List<int> if b turns out to be List<int>.
var c = [3]; // c is unknown.
c.Imply(b); // c is unknown. b will be List<int> if c turns out to be List<int>.
return c; // c is List<int>, so b is List<int>, so a has to be List<int>.
}So yes, you understood my (way too implicit/magical) thought experiment correctly :s |
You're (presumably) calling System.Object.ToString(), which implicitly involves a conversion to System.Object. Because of that, it's really the same question as asking what this does: object stuff = [1, 2, 3];
// stuff.GetType(), stuff.ToString(), etcIf the type is determined later and the only thing you're doing later is I'm not terribly bothered by My hope is for reasonable defaults, and then in cases where you want something else, you say what you want just like with every other var-declared local. |
|
@jnm2 My main concern is with the bad experience of zero intelissense until the IDE/compiler can figure out what the type will be. It will behave the same as if the type was Not a big fan. The only way I would support something like this was if it was used to improve the selected type. This would mean it would start with a simple native type, say |
|
@julealgon Another thing we explored was defining basic members for the " |
and others
Collection Expressions Next
Summary
This issue is intended to be the umbrella tracking item for all collection expression designs and work following the core design (#5354) that shipped in C#12.
As this is likely to be a large item with many constituent parts, it will link out to respective discussions and designs as they occur.
Roughly, here are the items we would like to consider, as well as early notes on the topic: https://github.com/dotnet/csharplang/blob/main/meetings/working-groups/collection-literals/CL-2024-01-23.md
["a", .. b ? ["c"] : []]andforeach (bool? b in [true, false, null]). Collection expressions: inline collections in spreads and foreach #7864Memory<T>,ArraySegment<T>etc.)IEnumerable, etc.)Design meetings
https://github.com/dotnet/csharplang/blob/main/meetings/2024/LDM-2024-01-08.md - Iteration types of collections
https://github.com/dotnet/csharplang/blob/main/meetings/working-groups/collection-literals/CL-2024-01-23.md - WG meetup
https://github.com/dotnet/csharplang/blob/main/meetings/2024/LDM-2024-01-10.md - Conversions vs construction
https://github.com/dotnet/csharplang/blob/main/meetings/2024/LDM-2024-02-05.md#collection-expressions-inline-collections
https://github.com/dotnet/csharplang/blob/main/meetings/2024/LDM-2024-09-06.md#collection-expressions-next-c13-and-beyond
The text was updated successfully, but these errors were encountered: