I am just starting to learn F#, and impressed by the type inference I thought I would try a function that gets the first record from a table (using query expressions, Linq style):
let getfirst data =
let result = query { for n in data do take 1 }
result |> Seq.head
This works, the type is IQueryable<'a> -> 'a.
But why doesn't this version work?
let getfirst2 data =
query { for n in data do head }
Shouldn't for n in data do head give a scalar 'a just like last time? Can someone explain why the second version doesn't work, and how to make it work without using Seq.head?
The reason is that the query builder has a somewhat hacky overloaded Run method for running queries, with the following overloads:
QueryBuilder.Run : Quotations.Expr<'t> -> 't
QueryBuilder.Run : Quotations.Expr<Linq.QuerySource<'t, IEnumerable>> -> seq<'t>
QueryBuilder.Run : Quotations.Expr<Linq.QuerySource<'t, IQueryable>> -> IQueryable<'t>
In your case, any of the overloads could apply, given a suitable type for data (though QuerySource<_,_> is a type which isn't ever meant to be used by user code, so two of the overloads are quite unlikely). Unfortunately, due to the strange way these overloads are defined (the first and second are actually extension methods defined in separate modules), the third one wins the overload resolution battle.
I don't know why, but when you hover over the data argument in getfirst2 you see it's of type System.Linq.IQueryable<Linq.QuerySource<'a, System.Linq.IQueryable>> when it really should be System.Linq.IQueryable<'a>.
You can "fix" it by adding type annotations:
open System.Linq
let getfirst2 (data : IQueryable<'a>) : 'a = query {
for item in data do
head
}
Then it works like you have expected:
[1 .. 10]
|> System.Linq.Queryable.AsQueryable
|> getfirst2
|> printfn "%d" // Prints 1.
Maybe someone else can shed some light on why the compiler infers the types it does.
Related
this is a basic question but i could not find the simple answer reading the tutorial
suppose i have this simple frame
type Person =
{ Name:string; Age:int; Countries:string list; }
let peopleRecds =
[ { Name = "Joe"; Age = 51; Countries = [ "UK"; "US"; "UK"] }
{ Name = "Tomas"; Age = 28; Countries = [ "CZ"; "UK"; "US"; "CZ" ] }
{ Name = "Eve"; Age = 2; Countries = [ "FR" ] }
{ Name = "Suzanne"; Age = 15; Countries = [ "US" ] } ]
// Turn the list of records into data frame
let peopleList = Frame.ofRecords peopleRecds
// Use the 'Name' column as a key (of type string)
let people = peopleList |> Frame.indexRowsString "Name"
How do i access the value the row for Joe ? (as a record, tuple or whatever format)
i tried this
getRow "Joe" people;;
Stopped due to error System.Exception: Operation could not be
completed due to earlier error Value restriction. The value 'it' has
been inferred to have generic type
val it : Series Either define 'it' as a simple data term, make it a function with explicit arguments or, if you do
not intend for it to be generic, add a type annotation. at 3,0
EDIT: thanks for the answer, still i would like to know why my syntax is incorrect because i think i respected the signature
val it :
('a -> Frame<'a,'b> -> Series<'b,'c>) when 'a : equality and 'b : equality
I'll answer the second half of your question, why you got a "value restriction" error. If you search for [f#] value restriction on Stack Overflow you'll find lots of answers, which may or may not confuse you. But the really short version is: F# is built on top of the .Net framework, and .Net imposes certain limitations. Specifically, functions are allowed to be generic, but values cannot be generic. So you can do this:
let f<'TData> (a:'TData) = printfn "%A" a
but you cannot do this:
let (a:'TData) = Unchecked.defaultof<'TData>
The function definition is fine, because the underlying .Net framework knows how to handle generic functions. But you're not allowed to have generic values in .Net; any value must be a specific type.
(Note: I wrote the <'TData> in the f definition explicitly, but I didn't have to: I could have just written let f (a:'TData) = printfn "%A" a and the genericness of f would have still been understood. I could even have just written let f a = printfn "%A" a, and it would have done the same thing).
Now let's look at the error you got: "the value "it" has been inferred to have generic type val it : Series<string,obj>". If you look at the function signature of getRow that you posted, it looks like this:
('a -> Frame<'a,'b> -> Series<'b,'c>)
When you called it as getRow "Joe" people, the F# compiler was able to infer that the type 'a was string (because the parameter "Joe" is a string). And because the second argument people is a Frame<string,string>, the F# compiler was able to infer that the type 'b was also string. But the result of that function call is a Series<'b,'c>, and so far the F# compiler doesn't know anything about what 'c will be. And since you ran getRow "Joe" people at the F# interactive REPL, it tried to store the result of what you typed as the value of the name it (the F# interactive REPL always provides the value of the previous expression as it) -- but since the only type it knew so far was Series<string,'c>, F# couldn't figure out what specific type to assign to the value it. I know from looking at your code that the type 'c was the Person record, but the F# compiler couldn't know that from just that one call to getRow, because of how the getRow function is typed.
There are two ways you could have solved this value restriction error:
One way to solve this would have been to pipe the result of getRow into another function, which would have allowed the F# compiler to infer the specific type of its result. Unfortunately, since I don't know Deedle that well, I can't give you a good example here. Maybe someone else will come up with one and comment on this answer, and I'll edit it in. It would look like:
getRow "Joe" people |> (some Deedle function)
But I don't know which Deedle function to use in my example: it would have to be a function that takes a Series and does some specific calculation with it, in a way that would allow F# to infer that this is a Series<string,Person>. Sorry this isn't a great example, but I'll leave it in anyway in case it helps.
The second way you could have solved the error would have been to specify the type of the value you were getting. In F#, you do that with the : (type) syntax, e.g.:
getRow "Joe" people : Series<string,Person>
Or, since the F# compiler has enough information to infer the string part of that type, you could also have written:
getRow "Joe" people : Series<_,Person>
When you write _ in a type signature, you're telling the F# compiler "You figure out what type this is". This only works when the F# compiler has enough information to infer that type correctly, but it's often a handy shorthand when type signatures would be large and unwieldy.
Both of these approaches would have solved your immediate problem, gotten rid of the "value restriction" error, and allowed you to continue working.
I hope this answer helps you. If it hopelessly confuses you instead, let me know and I'll see if I can explain whatever has you confused.
EDIT: In the comments, Soldalma asks whether the F# compiler (which is a one-pass compiler that works top to bottom and left to right) can infer the type from a forward pipe. The answer is yes, because the expression isn't finished yet. As long as an expression isn't finished, F#'s type inference (which is based on the Hindley-Milner type system*) is fine with carrying around a set of not-yet-resolved types. And if the types are resolved before the expression is complete, then the expression can resolve to a specific value (or a specific function). If the types are not yet resolved when the expression is complete, then it has to resolve to a generic value or function. And generic functions are allowed in .Net, but not generic values, hence the "value restriction" error.
To see this in practice, let's look at some example code. Copy and paste the following code into an F# editor that lets you hover over a variable (or function) name to see its type. I recommend VS Code with the Ionide-fsharp extension since it's cross-platform, but Visual Studio will work just as well.
open System.Collections.Generic
let mkDict (key:'K) = new Dictionary<'K,'V>() // Legal
let getValueOrDefault (key:'a) (defaultVal:'b) (dict:Dictionary<'a,'b>) =
match dict.TryGetValue key with
| true,v -> v
| false,_ -> defaultVal
let d = mkDict "foo" // Error: value restriction
let bar = mkDict "foo" |> getValueOrDefault "foo" "bar" // Legal: type string
let five = mkDict "foo" |> getValueOrDefault "foo" 5 // Legal: type int
Go ahead and hover your cursor over each function and variable name to see its type, or else hit Alt+Enter to send each function or variable declaration to F# Interactive. (And once you've seen that the let d line gives a "value restriction" error, comment it out so the rest of the code will compile).
What's happening here is a good demonstration of how this all works. The mkDict function has two unresolved types, 'K and 'V, so it has to be generic. But that's fine, because .Net has no problem with generic functions. (mkDict isn't actually very useful, since it actually "throws away" the data of its argument and does nothing to it. But it's supposed to be a trivial example, so just ignore the fact that it's kind of useless.) Likewise, getValueOrDefault has two unresolved types, 'a and 'b, so it's also a generic function.
However, let d = mkDict "foo" is not legal. Here, the generic type 'K has been resolved to be the specific type string, but 'V has not yet been resolved by the time the expression is complete so d would have to be generic (it would look like d<'V> in explicitly-generic syntax). But d is not a function (since it has no parameters), it's the name of a value, and .Net doesn't allow generic values.
But in the next two lines, the expression is not complete by the time the compiler has parsed mkDict "foo", so it doesn't yet have to "lock in" the unknown types. It can quite happily carry the unresolved type 'V into the next part of the expression. And there, the getValueOrDefault function has two specific types, string and string in the first line, and string and int in the second line. Because its 'b type corresponds to the 'V type from mkDict, therefore F# can resolve 'V in both lines. And so bar has type string, and five has type int.
* Scott Wlaschin says that it should "more accurately ... be called "Damas-Milner's Algorithm W" ". Since I haven't studied it in detail myself, I'll take his word for it -- but if you're interested in learning more, the Wikipedia link I provided is probably a halfway decent starting point.
I'll be very short, promoting my comment to an answer.
You need to use syntax reverse to the one you have tried:
people.Rows.["Joe"]
This question already has an answer here:
How do I deal with IEnumerable in F#?
(1 answer)
Closed 6 years ago.
There are still a few things around in the .NET standard libraries that only expose the old school IEnumerable.GetEnumerator() iterator to the outside world, which is not very friendly to the F# seq-processing style. I was doing a quick google on how to get the resulting groups of a Regex.Match(...) into a list I could process, and didn't find anything.
I had this:
open System.Text.RegularExpressions
let input = "args=(hello, world, foo, bar)"
let mtc = Regex.Match( input, "args=\(([\w\s,]+)\)" )
What I'd like is to get access to mtc.Groups as a seq or as a list, but it doesn't allow this because it's a ye olde ICollection, which only exposes a GetEnumerator() method. So while you can do
mtc.Groups.[1].Value
you cannot do
mtc.Groups |> Seq.skip 1 // <=== THIS QUESTION IS ABOUT HOW TO ACHIEVE THIS
as this results in
error FS0001: The type 'Text.RegularExpressions.GroupCollection' is not compatible with the type 'seq<'a>
(For clarity, GroupCollection implements ICollection, which is a sub-interface of IEnumerable.)
So the question is: how do I neatly turn a GetEnumerator() into a seq?
The answer is really nothing complicated, it's just here for the next person who is googling for a quick answer. The idea is to wrap the horrid imperativeness in a seq {...} expression and to then cast the resulting seq<obj> to whatever you happen to know the results to be.
seq { let i = mtc.Groups.GetEnumerator() in while i.MoveNext() do yield i.Current }
|> Seq.cast<Text.RegularExpressions.Group>
|> Seq.map (fun m -> m.Value)
|> List.ofSeq
when run on the input described above, this produces the required output:
val input : string = "args=(hello, world, foo, bar)"
val mtc : Match = args=(hello, world, foo, bar)
val it : string list = ["args=(hello, world, foo, bar)"; "hello, world, foo, bar"]
As I said, I'm putting it here for the next googler of the answer so improvements, suggestions, downvotes, dupe flags all welcome.
EDIT: as per the suggestion in the first comment, Seq.cast is clever enough to eat IEnumerables directly. So the seq-expression is simply unnecessary and the answer to this is just Seq.cast<Text.RegularExpressions.Group>! Let me know if I should just remove this question.
For the Froto project (Google Protobuf in F#), I am trying to update the deserialization code from using 'a ref objects to passing values byref<'a>, for performance.
However, the code below fails on the hydrator &element field line:
type Field = TypeA | TypeB | Etc
let hydrateRepeated
(hydrator:byref<'a> -> Field -> unit)
(result:byref<'a list>)
(field:Field) =
let mutable element = Unchecked.defaultof<'a>
hydrator &element field
result <- element :: result
error FS0421: The address of the variable 'element' cannot be used at this point
Is there anything I can do to get this code to work without changing the signature of the hydrator parameter?
I'm very aware that I could use hydrator:'a ref -> Field -> unit and get things to work. However, the goal is to support deserializing into record types without needing to create a bunch of ref objects on the heap every time a record is deserialize.
Note that the following code is perfectly legal and has the same signature as the hydrator function declaration, above, so I'm unclear on what the problem is.
let assign (result:byref<'a>) (x:'a) =
result <- x
let thisWorks() =
let mutable v = Unchecked.defaultof<int>
assign &v 5
printfn "%A" v
I'll try to clarify what I was saying in my comments. You're right that your definition of assign is perfectly fine, and it appears to have the signature byref<'a> -> 'a -> unit. However, if you look at the resulting assembly, you'll find that the way it's compiled at the .NET representation level is:
Void assign[a](a ByRef, a)
(that is, it's a method that takes two arguments and doesn't return anything, not a function value that takes one argument and returns a function that takes the next argument and returns a value of type unit - the compiler uses some additional metadata to determine how the method was actually declared).
The same is true of function definitions that don't involve byref. For instance, assume you've got the following definition:
let someFunc (x:int) (y:string) = ()
Then the compiler actually creates a method with the signature
Void someFunc(Int32, System.String)
The compiler is smart enough to do the right thing when you try to use a function like someFunc as a first class value - if you use it in a context where it isn't applied to any arguments, the compiler will generate a subtype of int -> string -> unit (which is FSharpFunc<int, FSharpFunc<string, unit>> at the .NET representation level), and everything works seamlessly.
However, if you try to do the same thing with assign, it won't work (or shouldn't work, but there are several compiler bugs that may make it seem like certain variations work when really they don't - you might not get a compiler error but you may get an output assembly that is malformed instead) - it's not legal for .NET type instantiations to use byref types as generic type arguments, so FSharpFunc<int byref, FSharpFunc<int, unit>> is not a valid .NET type. The fundamental way that F# represents function values just doesn't work when there are byref arguments.
So the workaround is to create your own type with a method taking a byref argument and then create subtypes/instances that have the behavior you want, sort of like doing manually what the compiler does automatically in the non-byref case. You could do this with a named type
type MyByrefFunc2<'a,'b> =
abstract Invoke : 'a byref * 'b -> unit
let assign = {
new MyByrefFunc2<_,_> with
member this.Invoke(result, x) =
result <- x }
or with a delegate type
type MyByrefDelegate2<'a,'b> = delegate of 'a byref * 'b -> unit
let assign = MyByrefDelegate2(fun result x -> result <- x)
Note that when calling methods like Invoke on the delegate or nominal type, no actual tuple is created, so you shouldn't be concerned about any extra overhead there (it's a .NET method that takes two arguments and is treated as such by the compiler). There is the cost of a virtual method call or delegate call, but in most cases similar costs exist when using function values in a first class way too. And in general, if you're worried about performance then you should set a target and measure against it rather than trying to optimize prematurely.
This is a pretty simple question, and I just wanted to check that what I'm doing and how I'm interpreting the F# makes sense. If I have the statement
let printRandom =
x = MyApplication.getRandom()
printfn "%d" x
x
Instead of creating printRandom as a function, F# runs it once and then assigns it a value. So, now, when I call printRandom, instead of getting a new random value and printing it, I simply get whatever was returned the first time. I can get around this my defining it as such:
let printRandom() =
x = MyApplication.getRandom()
printfn "%d" x
x
Is this the proper way to draw this distinction between parameter-less functions and values? This seems less than ideal to me. Does it have consequences in currying, composition, etc?
The right way to look at this is that F# has no such thing as parameter-less functions. All functions have to take a parameter, but sometimes you don't care what it is, so you use () (the singleton value of type unit). You could also make a function like this:
let printRandom unused =
x = MyApplication.getRandom()
printfn "%d" x
x
or this:
let printRandom _ =
x = MyApplication.getRandom()
printfn "%d" x
x
But () is the default way to express that you don't use the parameter. It expresses that fact to the caller, because the type is unit -> int not 'a -> int; as well as to the reader, because the call site is printRandom () not printRandom "unused".
Currying and composition do in fact rely on the fact that all functions take one parameter and return one value.
The most common way to write calls with unit, by the way, is with a space, especially in the non .NET relatives of F# like Caml, SML and Haskell. That's because () is a singleton value, not a syntactic thing like it is in C#.
Your analysis is correct.
The first instance defines a value and not a function. I admit this caught me a few times when I started with F# as well. Coming from C# it seems very natural that an assignment expression which contains multiple statements must be a lambda and hence delay evaluated.
This is just not the case in F#. Statements can be almost arbitrarily nested (and it rocks for having locally scoped functions and values). Once you get comfortable with this you start to see it as an advantage as you can create functions and continuations which are inaccessible to the rest of the function.
The second approach is the standard way for creating a function which logically takes no arguments. I don't know the precise terminology the F# team would use for this declaration though (perhaps a function taking a single argument of type unit). So I can't really comment on how it would affect currying.
Is this the proper way to draw this
distinction between parameter-less
functions and values? This seems less
than ideal to me. Does it have
consequences in currying, composition,
etc?
Yes, what you describe is correct.
For what its worth, it has a very interesting consequence able to partially evaluate functions on declaration. Compare these two functions:
// val contains : string -> bool
let contains =
let people = set ["Juliet"; "Joe"; "Bob"; "Jack"]
fun person -> people.Contains(person)
// val contains2 : string -> bool
let contains2 person =
let people = set ["Juliet"; "Joe"; "Bob"; "Jack"]
people.Contains(person)
Both functions produce identical results, contains creates its people set on declaration and reuses it, whereas contains2 creates its people set everytime you call the function. End result: contains is slightly faster. So knowing the distinction here can help you write faster code.
Assignment bodies looking like function bodies have cought a few programmers unaware. You can make things even more interesting by having the assignment return a function:
let foo =
printfn "This runs at startup"
(fun () -> printfn "This runs every time you call foo ()")
I just wrote a blog post about it at http://blog.wezeku.com/2010/08/23/values-functions-and-a-bit-of-both/.
Why aren't option types like "int option" compatible with nullable types like "Nullable"?
I assume there is some semantic reason for the difference, but I can't figure what that is.
An option in F# is used when a value may or may not exist. An option has an underlying type and may either hold a value of that type or it may not have a value.
http://msdn.microsoft.com/en-us/library/dd233245%28VS.100%29.aspx
That sure sounds like the Nullable structure.
Because of the runtime representation choice for System.Nullable<'T>.
Nullable tries to represent the absent of values by the null pointer, and present values by pointers to those values.
(new System.Nullable<int>() :> obj) = null
|> printfn "%b" // true
(new System.Nullable<int>(1) :> obj).GetType().Name
|> printfn "%s" // Int32
Now consider strings. Unfortunately, strings are nullable. So this is valid:
null : string
But now a null runtime value is ambiguous - it can refer to either the absence of a value or a presence of a null value. For this reason, .NET does not allow constructing a System.Nullable<string>.
Contrast this with:
(Some (null : string) :> obj).GetType().Name
|> printfn "%s" // Option`1
That being said, one can define a bijection:
let optionOfNullable (a : System.Nullable<'T>) =
if a.HasValue then
Some a.Value
else
None
let nullableOfOption = function
| None -> new System.Nullable<_>()
| Some x -> new System.Nullable<_>(x)
If you observe the types, these functions constrain 'T to be a structure and have a zero-argument constructor. So perhaps F# compiler could expose .NET functions receiving/returning Nullable<'T> by substituting it for an Option<'T where 'T : struct and 'T : (new : unit -> 'T)>, and inserting the conversion functions where necessary..
The two have different semantics. Just to name one, Nullable is an idempotent data constructor that only works on value types, whereas option is a normal generic type. So you can't have a
Nullable<Nullable<int>>
but you can have an
option<option<int>>
Generally, though there are some overlapping scenarios, there are also things you can do with one but not the other.
Key difference is that must test the option type to see if it has a value. See this question for a good description of its semantics: How does the option type work in F#
Again, this is from my limited understanding, but the problem probably lies in how each gets rendered in the IL. The "nullable" structure probably gets handled slightly different from the option type.
You will find that the interactions between various .Net languages really boils down to how the IL gets rendered. Mostof the time it works just fine but on occasion, it causes issues. (check out this). Just when you thought it was safe to trust the level of abstraction. :)