What is the Enum.GetName equivalent for F# union member? - f#

I want to get the equivalent of Enum.GetName for an F# discriminated union member. Calling ToString() gives me TypeName+MemberName, which isn't exactly what I want. I could substring it, of course, but is it safe? Or perhaps there's a better way?

You need to use the classes in the Microsoft.FSharp.Reflection namespace so:
open Microsoft.FSharp.Reflection
///Returns the case name of the object with union type 'ty.
let GetUnionCaseName (x:'a) =
match FSharpValue.GetUnionFields(x, typeof<'a>) with
| case, _ -> case.Name
///Returns the case names of union type 'ty.
let GetUnionCaseNames <'ty> () =
FSharpType.GetUnionCases(typeof<'ty>) |> Array.map (fun info -> info.Name)
// Example
type Beverage =
| Coffee
| Tea
let t = Tea
> val t : Beverage = Tea
GetUnionCaseName(t)
> val it : string = "Tea"
GetUnionCaseNames<Beverage>()
> val it : string array = [|"Coffee"; "Tea"|]

#DanielAsher's answer works, but to make it more elegant (and fast? because of the lack of reflection for one of the methods), I would do it this way:
type Beverage =
| Coffee
| Tea
static member ToStrings() =
Microsoft.FSharp.Reflection.FSharpType.GetUnionCases(typeof<Beverage>)
|> Array.map (fun info -> info.Name)
override self.ToString() =
sprintf "%A" self
(Inspired by this and this.)

This answer supplies additional information and solutions to the top answer.
I just now had a case where the top answer did not work. The problem was that the value was behind an interface, and then I would sometimes get the case name (Coffee or Tea), but mostly only the type name (Beverage). I don't understand why. I'm on .NET 5.0.
I changed the function to this, and then it worked as expected on my interfaced DU, always giving me the case name.
open FSharp.Reflection
let GetUnionCaseName (x: obj) =
match FSharpValue.GetUnionFields(x, x.GetType()) with
| case, _ -> case.Name
I am aware that this is similar to other answers here, but this is not a member function, and so I guess should work on any DU, whether behind interfaces or not. I haven't tested what happens if used on a non-DU type.
type IMessage = interface end
type Beverage = Coffee | Tea
type Car =
| Tesla of model:string
| Ford
interface IMessage
type MySingleCase = MySingleCase of string
type SingleCase2 = SingleCase2 of string interface IMessage
let m1: Beverage = Coffee
let m2: IMessage = (Tesla "Model 3") :> IMessage
let m3 = MySingleCase "x"
let m4 = SingleCase2 "x" :> IMessage
printfn "%s" (GetUnionCaseName m1) // Coffee
printfn "%s" (GetUnionCaseName m2) // Tesla
printfn "%s" (GetUnionCaseName m3) // MySingleCase
printfn "%s" (GetUnionCaseName m4) // SingleCase2

I would like to propose something even more concise:
open Microsoft.FSharp.Reflection
type Coffee = { Country: string; Intensity: int }
type Beverage =
| Tea
| Coffee of Coffee
member x.GetName() =
match FSharpValue.GetUnionFields(x, x.GetType()) with
| (case, _) -> case.Name
When union case is simple, GetName() may bring the same as ToString():
> let tea = Tea
val tea : Beverage = Tea
> tea.GetName()
val it : string = "Tea"
> tea.ToString()
val it : string = "Tea"
However, if union case is fancier, there will be a difference:.
> let coffee = Coffee ({ Country = "Kenya"; Intensity = 42 })
val coffee : Beverage = Coffee {Country = "Kenya"; Intensity = 42;}
> coffee.GetName()
val it : string = "Coffee"
> coffee.ToString()
val it : string = "Coffee {Country = "Kenya"; Intensity = 42;}"

Related

Is there a way to get Record fields by string in F#?

I would like to get the value of a field in a Record by looking it up with a string.
type Test = { example : string }
let test = { example = "this is the value" }
let getByName (s:string) =
???? //something like test.GetByName(s)
Standard .net reflection should be working fine for such scenario. Record fields are exposed as properties, so you can just query the type with reflection API.
It could look like this:
let getByName (s:string) =
match typeof<Test>.GetProperties() |> Array.tryFind (fun t -> t.Name = s)
with
| Some pi -> Some(pi.GetValue(test))
| None -> None

How do I condense this repetitive F# code?

EDIT: possible solutions at bottom
I'm doing some data work where I need to be very careful about string lengths that will eventually be sent in fixed width text output, stored in limited size nvarchar fields, etc. I want to have good strict typing for these rather than naked System.String types.
Suppose I've got some code like this to represent these, with a few useful module functions that play nicely with Result.map, Option.map, etc.
module String40 =
let private MaxLength = 40
type T = private T of string
let create (s:string) = checkStringLength MaxLength s |> Result.map T
let trustCreate (s:string) = checkStringLength MaxLength s |> Result.okVal |> T
let truncateCreate (s:string) = truncateStringToLength MaxLength s |> T
let toString (T s) = s
type T with
member this.AsString = this |> toString
module String100 =
let private MaxLength = 100
type T = private T of string
let create (s:string) = checkStringLength MaxLength s |> Result.map T
let trustCreate (s:string) = checkStringLength MaxLength s |> Result.okVal |> T
let truncateCreate (s:string) = truncateStringToLength MaxLength s |> T
let toString (T s) = s
type T with
member this.AsString = this |> toString
Obviously these are almost entirely repetitive with only the module name and max length different in each block.
What options are available to try and cut down on the repetitiveness here? I would love to have something like this:
type String40 = LengthLimitedString<40>
type String100 = LengthLimitedString<100>
tryToRetrieveString () // returns Result<string, ERRType>
|> Result.bind String40.create
T4 code generation doesn't seem to be an option for F# projects
Type providers seem like overkill for this kind of simple templating, and as best as I can tell they can only produce classes, not modules.
I'm aware of Scott Wlaschin's constrained strings page, but I end up with roughly the same level of repetitive code in the 'Create a type', 'Implement IWrappedString', 'create a public constructor' steps he lists.
These code blocks are fairly short and it wouldn't be the end of the world to just copy/paste a dozen times for the different field lengths. But I feel like I'm missing a simpler way to do this.
UPDATE:
One other note is that it's important that records using these types give information about what type they're carrying:
type MyRecord =
{
FirstName: String40;
LastName: String100;
}
and not be something like
type MyRecord =
{
FirstName: LimitedString;
LastName: LimitedString;
}
Tomas' answer, using the Depended Type Provider nuget library is a pretty good one, and would be a good solution for a lot of people who are fine with its behavior as-is. I felt like it would be a little tricky to extend and customize unless I wanted to maintain my own copy of a type provider which I was hoping to avoid.
Marcelo's suggestion of static parameter constraints was a fairly productive path of research. They give me basically what I was looking for -- a generic argument that is basically an 'interface' for a static methods. However the kicker is that they require inline functions to operate and I don't have the time to evaluate how much that would or would not matter in my code base.
But I took that and altered it to use regular generic constraints. It's a bit goofy to have to instantiate an object to get a max-length value, and fsharp type/generic code is just gross to look at, but from the module users's perspective it's clean, and I can easily extend this however I want.
type IMaxLengthProvider = abstract member GetMaxLength: unit -> int
type MaxLength3 () = interface IMaxLengthProvider with member this.GetMaxLength () = 3
type MaxLength4 () = interface IMaxLengthProvider with member this.GetMaxLength () = 4
module LimitedString =
type T< 'a when 'a :> IMaxLengthProvider> = private T of string
let create< 't when 't :> IMaxLengthProvider and 't : (new:unit -> 't)> (s:string) =
let len = (new 't()).GetMaxLength()
match checkStringLength len s with
| Ok s ->
let x : T< 't> = s |> T
x |> Ok
| Error e -> Error e
let trustCreate< 't when 't :> IMaxLengthProvider and 't : (new:unit -> 't)> (s:string) =
let len = (new 't()).GetMaxLength()
match checkStringLength len s with
| Ok s ->
let x : T< 't> = s |> T
x
| Error e ->
let msg = e |> formErrorMessage
failwith msg
let truncateCreate< 't when 't :> IMaxLengthProvider and 't : (new:unit -> 't)> (s:string) =
let len = (new 't()).GetMaxLength()
let s = truncateStringToLength len s
let x : T< 't> = s |> T
x
let toString (T s) = s
type T< 'a when 'a :> IMaxLengthProvider> with
member this.AsString = this |> toString
module test =
let dotest () =
let getString () = "asd" |> Ok
let mystr =
getString ()
|> Result.bind LimitedString.create<MaxLength3>
|> Result.okVal
|> LimitedString.toString
sprintf "it is %s" mystr
I think the BoundedString type provider from the Dependent type provider project lets you do exactly what you need. Using the example from the project documentation, you can do e.g.:
type ProductDescription = BoundedString<10, 2000>
type ProductName = BoundedString<5, 50>
type Product = { Name : ProductName; Description : ProductDescription }
let newProduct (name : string) (description : string) : Product option =
match ProductName.TryCreate(name), ProductDescription.TryCreate(description) with
| Some n, Some d -> { Name = n; Description = d }
| _ -> None
I don't know how many people use this project in practice, but it seems quite simple and it does exactly what you were asking for, so it might be worth a try.
Using a touch of careful reflection magic, we can achieve a lot and get some really nice types. How about something like this?
module Strings =
type [<AbstractClass>] Length(value: int) =
member this.Value = value
let getLengthInst<'L when 'L :> Length> : 'L =
downcast typeof<'L>.GetConstructor([||]).Invoke([||])
type LimitedString<'Length when 'Length :> Length> =
private | LimitedString of maxLength: 'Length * value: string
member this.Value =
let (LimitedString(_, value)) = this in value
member this.MaxLength =
let (LimitedString(maxLength, _)) = this in maxLength.Value
module LimitedString =
let checkStringLength<'L when 'L :> Length> (str: string) =
let maxLength = getLengthInst<'L>.Value
if str.Length <= maxLength then Ok str
else Error (sprintf "String of length %d exceeded max length of %d" str.Length maxLength)
let create<'L when 'L :> Length> (str: string) =
checkStringLength<'L> str
|> Result.map (fun str -> LimitedString (getLengthInst<'L>, str))
open Strings
// === Usage ===
type Len5() = inherit Length(5)
type Len1() = inherit Length(1)
// Ok
LimitedString.create<Len5> "Hello"
// Error
LimitedString.create<Len1> "world"
One option might be to use a single module for limited-length strings that uses curried parameters for the length-limit and the string itself, then just partially apply the limit parameter. An implementation might look like this:
module LimitedString =
type T = private T of string
let create length (s:string) = checkStringLength length s |> Result.map T
let trustCreate length (s:string) = checkStringLength length s |> Result.okVal |> T
let truncateCreate length (s:string) = truncateStringToLength length s |> T
let toString (T s) = s
type T with
member this.AsString = this |> toString
Then, your modules for each length would still be required, but wouldn't have all the boilerplate:
module String100 =
let create = LimitedString.create 100
let trustCreate = LimitedString.trustCreate 100
let truncateCreate = LimitedString.truncateCreate 100
EDIT
After reading the comment and the update to the original post, I would change my suggestion a bit. Instead of defining the T type inside each module, I would have a specific struct-type single-case union for each string length at the top level. Then, I would move to toString to the individual string modules. Finally, I would add one more parameter to the LimitedString module to allow us to partially apply both the length and the specific single-case union type:
[<Struct>] type String40 = private String40 of string
[<Struct>] type String100 = private String100 of string
module LimitedString =
let create length ctor (s:string) = checkStringLength length s |> Result.map ctor
let trustCreate length ctor (s:string) = checkStringLength length s |> Result.okVal |> ctor
let truncateCreate length ctor (s:string) = truncateStringToLength length s |> ctor
module String40 =
let create = LimitedString.create 40 String40
let trustCreate = LimitedString.trustCreate 40 String40
let truncateCreate = LimitedString.truncateCreate 40 String40
let toString (String40 s) = s
module String100 =
let create = LimitedString.create 100 String100
let trustCreate = LimitedString.trustCreate 100 String100
let truncateCreate = LimitedString.truncateCreate 100 String100
let toString (String100 s) = s
type MyRecord =
{
FirstName: String40
LastName: String100
}
There's still a fair amount of boilerplate here, but I think this is in the ballpark for a solution using single-case unions and modules. A Type Provider might be possible, but you would have to consider whether the added complexity outweighs the boilerplate.
Can't you use Static Parameters, as shown on the F#.Data package's HtmlProvider example?

Why isn't this generic function supported in my example?

I'm trying to learn F# and have put together the following code:
open System.Collections.Generic
type Version = Version of int
type AggregateId = AggregateId of int
type IEventdata = interface end
type EventMeta = EventMeta of AggregateId * Version
type Event = Event of EventMeta * IEventdata
type ICommandData = interface end
type CommandMeta = CommandMeta of AggregateId * Version
type Command = CommandMeta * ICommandData
type EventStore = {
GetEvents : AggregateId -> Event seq;
Insert : Event seq -> unit
}
let commandDispatcherFactory (mapping: (Command -> Event seq -> Event seq), es: EventStore) =
(
fun (command: Command) ->
match command with
| (CommandMeta(aggregateId, version), commandData) ->
let events = es.GetEvents (aggregateId)
let resultingEvents = mapping command events
(es.Insert(resultingEvents))
)
type DummyCommand =
| Command1
interface ICommandData
type DummyCommand2 =
| Command2
interface ICommandData
type UnsupportedCommand =
| Command3
let dummyMapping = (fun ((command, commandData):Command) (events: Event seq) ->
match commandData with
| :? DummyCommand as dummyCommand ->
(printfn "Handling Command: %A" dummyCommand)
Seq.empty<Event>
| :? DummyCommand2 as dummyCommand ->
(printfn "Handling Command: %A" dummyCommand)
Seq.empty<Event>
| _ ->
(printfn "Unsupported command: %A" commandData)
Seq.empty<Event>
)
let dummyEs = {
GetEvents = (fun (aggregateId) -> Seq.empty<Event>);
Insert = (fun (events:Event seq) -> ())
}
dummyMapping (CommandMeta(AggregateId(1),Version(1)), Command1) Seq.empty<Event>
dummyMapping (CommandMeta(AggregateId(1),Version(1)), Command2) Seq.empty<Event> // Why does this work?
let commandDispatcher = commandDispatcherFactory(dummyMapping, dummyEs)
commandDispatcher (CommandMeta(AggregateId(1),Version(1)), Command1)
commandDispatcher (CommandMeta(AggregateId(1),Version(1)), Command2) // this doesn't work
I know it might be a wall of text, but I didn't manage to narrow it down. It's the last call to my commandDispatcher that doesn't work, and I can't figure out why. I know there probably are other things to do better, but that is something I'm working on :)
Here's a cut down repro:
let f x (y:System.IComparable) = ()
let g = f 1
g 3
g "test" // not okay
let g' : System.IComparable -> unit = f 1
g' 3
g' "test" // okay
let g'' x = f 1 x
g'' 3
g'' "test" // okay
What happens is that the compiler infers a generic type for g (g:('a -> unit) when 'a :> System.IComparable in the case of my example). However, g is a syntactic value and therefor can't be generic (see questions about the value restriction), so the compiler needs to pick a specific type. If you don't use g at all, you'll get a compiler error. If you use it at exactly one concrete type, the compiler will pick that type. If you use it at more than one type, the compiler will specialize based on the first type used and throw an error after that.
Two workarounds are to either annotate with a non-generic type or to eta-expand so that g isn't a syntactic value.
The compiler is specializing the type of commandDispatcher based on the initial usage. You can fix it with a cast:
commandDispatcher (CommandMeta(AggregateId(1),Version(1)), Command1 :> ICommandData)
commandDispatcher (CommandMeta(AggregateId(1),Version(1)), Command2) //works

Avoiding nested pattern matching (possibly with maybe monad)

How could nested pattern matching, such as the following example, be re-written so that None is specified only once? I think the Maybe monad solves this problem. Is there something similar in the F# core library? Or, is there an alternative approach?
match a with
| Some b ->
let c = b.SomeProperty
match c with
| Some d ->
let e = d.SomeProperty
//and so on...
| None -> ()
| None -> ()
you can solve this using built-in capabilities: Option.bind
type A =
member this.X : B option = Unchecked.defaultof<_>
and B =
member this.Y : С option = Unchecked.defaultof<_>
and С =
member this.Z : string option = Unchecked.defaultof<_>
let a : A = Unchecked.defaultof<_>
let v =
match
a.X
|> Option.bind (fun v -> v.Y)
|> Option.bind (fun v -> v.Z) with
| Some s -> s
| None -> "<none>"
Frankly, I doubt that introducing full-fledged 'maybe' implementation (via computation expressions) here can shorten the code.
EDIT: Dream mode - on
I think that version with Option.bind can be made smaller if F# has more lightweight syntax for the special case: lambda that refer to some member of its argument:
"123" |> fun s -> s.Length // current version
"123" |> #.Length // hypothetical syntax
This is how the sample can be rewritten in Nemerle that already has such capabilities:
using System;
using Nemerle.Utility; // for Accessor macro : generates property for given field
variant Option[T]
{
| Some {value : T}
| None
}
module OptionExtensions
{
public Bind[T, U](this o : Option[T], f : T -> Option[U]) : Option[U]
{
match(o)
{
| Option.Some(value) => f(value)
| Option.None => Option.None()
}
}
}
[Record] // Record macro: checks existing fields and creates constructor for its initialization
class A
{
[Accessor]
value : Option[A];
}
def print(_)
{
// shortened syntax for functions with body -> match over arguments
| Option.Some(_) => Console.WriteLine("value");
| Option.None => Console.WriteLine("none");
}
def x = A(Option.Some(A(Option.Some(A(Option.None())))));
print(x.Value.Bind(_.Value)); // "value"
print(x.Value.Bind(_.Value).Bind(_.Value)); // "none"
I like desco's answer; one should always favor built-in constructs. But FWIW, here's what a workflow version might look like (if I understand the problem correctly):
type CE () =
member this.Bind (v,f) =
match v with
| Some(x) -> f x
| None -> None
member this.Return v = v
type A (p:A option) =
member this.P
with get() = p
let f (aIn:A option) = CE () {
let! a = aIn
let! b = a.P
let! c = b.P
return c.P }
let x = f (Some(A(None)))
let y = f (Some(A(Some(A(Some(A(Some(A(None)))))))))
printfn "Your breakpoint here."
I don't suggest this, but you can also solve it with exception handling:
try
<code that just keeps dotting into option.Value with impunity>
with
| :? System.NullReferenceException -> "None"
I just wanted to point out the rough equivalence of exception-handling to the Maybe/Either monads or Option.bind. Typically prefer one of them to throwing and catching exceptions.
Using Option.maybe from FSharpx:
open FSharpx
type Pet = { Name: string; PreviousOwner: option<string> }
type Person = { Name: string; Pet: option<Pet> }
let pers = { Name = "Bob"; Pet = Some {Name = "Mr Burns"; PreviousOwner = Some "Susan"} }
Option.maybe {
let! pet = pers.Pet
let! prevOwner = pet.PreviousOwner
do printfn "%s was the previous owner of %s." prevOwner pet.Name
}
Output:
Susan was the previous owner of Mr Burns.
But, e.g. with this person instead there is just no output:
let pers = { Name = "Bob"; Pet = None }

Working with Nullable<'T> in F#

I'm wondering what others have come up with for dealing with Nullable<'T> in F#. I want to use Nullable<'T> on data types so that serialization works properly (i.e., doesn't write out F# option type to XML). But, I don't want my code stuck dealing with the ugliness of dealing with Nullable<'T> directly. Any suggestions?
Is it better to use active patterns to match directly on Nullable, or just a converter to option and use Some/None matching?
Additionally, I'd love to hear ideas on dealing with nullable references in a nice manner too. If I use, say "string option", then I end up with the F# option type wrapping things. If I don't then I can't distinguish between truly optional strings and strings that shouldn't be null.
Any chance .NET 4 will take on an Option<'T> to help out? (If it's part of the BCL, then we might see better support for it...)
As active patterns as options plays nicely with pattern matching, but is seems by using active patterns (i.e. typeof and ??) your code will eat more ticks.
The base question is how you will deal with your nullable references?
In case your code is long chained computations it's nice to use monadic syntax:
type Maybe<'a> = (unit -> 'a option)
let succeed x : Maybe<'a> = fun () -> Some(x)
let fail : Maybe<'a> = fun () -> None
let run (a: Maybe<'a>) = a()
let bind p rest = match run p with None -> fail | Some r -> (rest r)
let delay f = fun () -> run (f ())
type MaybeBuilder() =
member this.Return(x) = succeed x
member this.Let(p,rest) = rest p
member this.Bind(p,rest) = bind p rest
member this.Delay(f) = delay f
let maybe = new MaybeBuilder()
let add (a:'a) (b:'a) =
maybe {
match TryGetNumericAssociation<'a>() with
| Some v -> return (v.Add(a,b))
| _ -> return! fail
}
let add3 (a:'a) (b:'a) (c:'a) =
maybe {
let! ab = add a b
let! abc = add ab c
return abc
}
> let r1 = add 1 2;;
val r1 : (unit -> int option)
> r1();;
val it : int option = Some 3
> let r2 = add "1" "2";;
val r2 : (unit -> string option)
> r2();;
val it : string option = None
> let r3 = add3 "one" "two" "three";;
val r3 : (unit -> string option)
> r3();;
val it : string option = None

Resources