Using Crystal 0.25.1 the type of the instance variable #foo in this example is inferred to be nil-able even though it can never be nil:
class Foo
end
class Bar
def initialize(foo : Foo? = nil)
#foo = foo || Foo.new
end
end
p typeof(Bar.new.#foo) # => (Foo | Nil)
I know that I can declare #foo : Foo at the class level, to work around the issue, but why can't the Crystal compiler infer that #foo can never be nil on its own?
Related
There are already several Q&As on this "X does not implement Y (... method has a pointer receiver)" thing, but to me, they seems to be talking about different things, and not applying to my specific case.
So, instead of making the question very specific, I'm making it broad and abstract -- Seems like there are several different cases that can make this error happen, can someone summary it up please?
I.e., how to avoid the problem, and if it occurs, what are the possibilities? Thx.
This compile-time error arises when you try to assign or pass (or convert) a concrete type to an interface type; and the type itself does not implement the interface, only a pointer to the type.
Short summary: An assignment to a variable of interface type is valid if the value being assigned implements the interface it is assigned to. It implements it if its method set is a superset of the interface. The method set of pointer types includes methods with both pointer and non-pointer receiver. The method set of non-pointer types only includes methods with non-pointer receiver.
Let's see an example:
type Stringer interface {
String() string
}
type MyType struct {
value string
}
func (m *MyType) String() string { return m.value }
The Stringer interface type has one method only: String(). Any value that is stored in an interface value Stringer must have this method. We also created a MyType, and we created a method MyType.String() with pointer receiver. This means the String() method is in the method set of the *MyType type, but not in that of MyType.
When we attempt to assign a value of MyType to a variable of type Stringer, we get the error in question:
m := MyType{value: "something"}
var s Stringer
s = m // cannot use m (type MyType) as type Stringer in assignment:
// MyType does not implement Stringer (String method has pointer receiver)
But everything is ok if we try to assign a value of type *MyType to Stringer:
s = &m
fmt.Println(s)
And we get the expected outcome (try it on the Go Playground):
something
So the requirements to get this compile-time error:
A value of non-pointer concrete type being assigned (or passed or converted)
An interface type being assigned to (or passed to, or converted to)
The concrete type has the required method of the interface, but with a pointer receiver
Possibilities to resolve the issue:
A pointer to the value must be used, whose method set will include the method with the pointer receiver
Or the receiver type must be changed to non-pointer, so the method set of the non-pointer concrete type will also contain the method (and thus satisfy the interface). This may or may not be viable, as if the method has to modify the value, a non-pointer receiver is not an option.
Structs and embedding
When using structs and embedding, often it's not "you" that implement an interface (provide a method implementation), but a type you embed in your struct. Like in this example:
type MyType2 struct {
MyType
}
m := MyType{value: "something"}
m2 := MyType2{MyType: m}
var s Stringer
s = m2 // Compile-time error again
Again, compile-time error, because the method set of MyType2 does not contain the String() method of the embedded MyType, only the method set of *MyType2, so the following works (try it on the Go Playground):
var s Stringer
s = &m2
We can also make it work, if we embed *MyType and using only a non-pointer MyType2 (try it on the Go Playground):
type MyType2 struct {
*MyType
}
m := MyType{value: "something"}
m2 := MyType2{MyType: &m}
var s Stringer
s = m2
Also, whatever we embed (either MyType or *MyType), if we use a pointer *MyType2, it will always work (try it on the Go Playground):
type MyType2 struct {
*MyType
}
m := MyType{value: "something"}
m2 := MyType2{MyType: &m}
var s Stringer
s = &m2
Relevant section from the spec (from section Struct types):
Given a struct type S and a type named T, promoted methods are included in the method set of the struct as follows:
If S contains an anonymous field T, the method sets of S and *S both include promoted methods with receiver T. The method set of *S also includes promoted methods with receiver *T.
If S contains an anonymous field *T, the method sets of S and *S both include promoted methods with receiver T or *T.
So in other words: if we embed a non-pointer type, the method set of the non-pointer embedder only gets the methods with non-pointer receivers (from the embedded type).
If we embed a pointer type, the method set of the non-pointer embedder gets methods with both pointer and non-pointer receivers (from the embedded type).
If we use a pointer value to the embedder, regardless of whether the embedded type is pointer or not, the method set of the pointer to the embedder always gets methods with both the pointer and non-pointer receivers (from the embedded type).
Note:
There is a very similar case, namely when you have an interface value which wraps a value of MyType, and you try to type assert another interface value from it, Stringer. In this case the assertion will not hold for the reasons described above, but we get a slightly different runtime-error:
m := MyType{value: "something"}
var i interface{} = m
fmt.Println(i.(Stringer))
Runtime panic (try it on the Go Playground):
panic: interface conversion: main.MyType is not main.Stringer:
missing method String
Attempting to convert instead of type assert, we get the compile-time error we're talking about:
m := MyType{value: "something"}
fmt.Println(Stringer(m))
To keep it short and simple, let say you have a Loader interface and a WebLoader that implements this interface.
package main
import "fmt"
// Loader defines a content loader
type Loader interface {
load(src string) string
}
// WebLoader is a web content loader
type WebLoader struct{}
// load loads the content of a page
func (w *WebLoader) load(src string) string {
return fmt.Sprintf("I loaded this page %s", src)
}
func main() {
webLoader := WebLoader{}
loadContent(webLoader)
}
func loadContent(loader Loader) {
loader.load("google.com")
}
The above code will give you this compile time error
./main.go:20:13: cannot use webLoader (type WebLoader) as type Loader
in argument to loadContent:
WebLoader does not implement Loader (Load method has pointer receiver)
To fix it you only need to change webLoader := WebLoader{} to following:
webLoader := &WebLoader{}
Why this will fix the issue? Because you defined this function func (w *WebLoader) Load to accept a pointer receiver. For more explanation please read #icza and #karora answers
Another case when I have seen this kind of thing happening is if I want to create an interface where some methods will modify an internal value and others will not.
type GetterSetter interface {
GetVal() int
SetVal(x int) int
}
Something that then implements this interface could be like:
type MyTypeA struct {
a int
}
func (m MyTypeA) GetVal() int {
return a
}
func (m *MyTypeA) SetVal(newVal int) int {
int oldVal = m.a
m.a = newVal
return oldVal
}
So the implementing type will likely have some methods which are pointer receivers and some which are not and since I have quite a variety of these various things that are GetterSetters I'd like to check in my tests that they are all doing the expected.
If I were to do something like this:
myTypeInstance := MyType{ 7 }
... maybe some code doing other stuff ...
var f interface{} = myTypeInstance
_, ok := f.(GetterSetter)
if !ok {
t.Fail()
}
Then I won't get the aforementioned "X does not implement Y (Z method has pointer receiver)" error (since it is a compile-time error) but I will have a bad day chasing down exactly why my test is failing...
Instead I have to make sure I do the type check using a pointer, such as:
var f interface{} = new(&MyTypeA)
...
Or:
myTypeInstance := MyType{ 7 }
var f interface{} = &myTypeInstance
...
Then all is happy with the tests!
But wait! In my code, perhaps I have methods which accept a GetterSetter somewhere:
func SomeStuff(g GetterSetter, x int) int {
if x > 10 {
return g.GetVal() + 1
}
return g.GetVal()
}
If I call these methods from inside another type method, this will generate the error:
func (m MyTypeA) OtherThing(x int) {
SomeStuff(m, x)
}
Either of the following calls will work:
func (m *MyTypeA) OtherThing(x int) {
SomeStuff(m, x)
}
func (m MyTypeA) OtherThing(x int) {
SomeStuff(&m, x)
}
Extend from above answers (Thanks for all of your answers)
I think it would be more instinctive to show all the methods of pointer / non pointer struct.
Here is the playground code.
https://play.golang.org/p/jkYrqF4KyIf
To summarize all the example.
Pointer struct type would include all non pointer / pointer receiver methods
Non pointer struct type would only include non pointer receiver methods.
For embedded struct
non pointer outer struct + non pointer embedded struct => only non pointer receiver methods.
non pointer outer struct + pointer embedded struct / pointer outer struct + non pointer embedded struct / pointer outer struct + pointer embedded struct => all embedded methods
If I have a function which accepts an argument of multiple types, how can I enforce that the return must match the value of the input?
This comes up particularly when I want a method to work with any children of a parent type. For demonstration, consider something that is "barlike":
abstract struct Barlike
property bar: Int32
def initialize(#bar); end
end
abstract def make_clang_sound
abstract def serve_drinks
end
Now any struct can implement those two methods, and store that value
struct Bar1 < Barlike
def make_clang_sound
puts "bing bang bong"
end
def serve_drinks
puts "your drink sir"
end
end
struct Bar2 < Barlike
def make_clang_sound
puts "kling klang"
end
def serve_drinks
puts "here are your drinks"
end
end
Now what if I have a method that wants to use the bar and return a new one with an updated value (these are structs afterall):
def foo(arg : Barlike)
new_bar = arg.bar + 2
arg.class.new(new_bar)
end
this will return a Bar1 if a Bar1 is passed in and a Bar2 if that is passed in but it's not guaranteed:
def foo(arg : Barlike)
"howdy"
end
I'm going to be putting my foo into an abstract structure as well, so I need to guarantee that implementers of foo return the same type of Barlike that was given.
I tried
def foo(arg : Barlike) : arg.class
end
But that's a compile time error (arg cannot be used there like that)
I also tried
def foo(arg : Barlike) : typeof(arg)
end
which passes, but typeof here is just Barlike whereas I really need it to be only the thing that was passed in, only Bar1 or Bar2 and so on.
Can macros help?
The tool for this are free variables. That's essentially generics scoped to a single method.
# This method returns the same type as its argument
def foo(arg : T) : T forall T
arg
end
This would already solve the main part of your question.
However, it is currently not possible to apply type restrictions to free variables, for example restricting T to Barlike.
There are workarounds, though:
Use a macro to validate the argument type:
def foo(arg : T) : T forall T
{% raise "arg must implement Barlike" unless T < Barlike %}
arg
end
Delegate to another method with a type restriction:
def foo(arg : T) : T forall T
foo_impl(arg)
end
private def foo_impl(arg : Barlike)
arg
end
Both workarounds affect the implementation of the method. There is no way to specify such a type restriction for a abstract def. Number 2 might be feasible, if you make foo_impl abstract and require inheriting classes to implement this one, instead of foo.
But it's probably also fine to just go with the initial example using free variables, without the Barlike restriction. In practice, you probably don't gain much.
Here is something that works:
{% for sub in Barlike.subclasses %}
struct {{sub}}
def foo() : {{sub}}
{{sub}}.new(#bar+1)
end
end
{% end %}
full example
But it feels like this is trying to solve the wrong problem. It uses the #subclasses macro to generate a foo for all of the child structs.
You could also declare these as self methods inside the abstract class: example.
Basically I've created a type with three attributes, lets call it "Foo"
type Foo = Foo Att1 Att2 Att3
Each attribute is one of 3 values, I declared each attribute with this format:
data Att1 = A | B | C
Now, after a little guesswork about the difference between when to use "data" or "type" I've tried to make a 2D list of Foo's:
type fooList = fooList [[Foo A A A, Foo B A C], [Foo C A B, Foo B B A]]
EDIT: I also get the same errors when I assemble this list with
data fooList = fooList [[Foo A A A, Foo B A C], [Foo C A B, Foo B B A]]
This is causing two main errors upon compilation:
Not in scope: type constructor or class ‘A’
A data constructor of that name is in scope; did you mean DataKinds?
and
Illegal Type ["Copy of foolist"]. Perhaps you intended to use DataKinds
Where am I going wrong here? I've more or less tried to follow learnyouahaskell to a tee but no matter how many ways I've tried this, I keep getting similar errors.
First of all the definition of Foo is incorrect. Since it here contains multiple values, and Foo is not an already defined type, you need to construct it like:
data Foo = Foo Att1 Att2 Att3
If you define type Foo = Foo Att1 Att2 Att3 then you define a type alias, but define a type alias to the same type (so recursively) does not make much sense here.
Furthermore Att2 and Att3 are probably just Att1s:
data Foo = Foo Att1 Att1 Att1
Furthermore you do not need type or data to declarate a value. You can create a 2d list by specifying it as:
fooList = [[Foo A A A, Foo B A C], [Foo C A B, Foo B B A]]
So here we declare a variable that stores values: a list of lists and each element is a Foo item.
So right now I have two classes of the same class type, e.g.
class foo : foo_type = object ... end
class bar : foo_type = object ... end
I'd like to have a third class that inherits from either foo or bar at runtime. E.g. (pseudo-syntax)
class baz (some_parent_class : foo_type) = object
inherit some_parent_class
...
end
Is this possible in OCaml?
Use case: I'm using objects for constructing AST visitors, and I'd like to be able to combine these visitors based on a set of runtime criteria so that they only do one combined traversal over the AST.
Edit: I think I have found a way to create the desired class using first-class modules:
class type visitor_type = object end
module type VMod = sig
class visitor : visitor_type
end
let mk_visitor m =
let module M = (val m : VMod) in
(module struct
class visitor = object
inherit M.visitor
end
end : VMod)
However, it seems a little roundabout to have to wrap a class in a module in order to make it "first-class". If there's a more straightforward way please let me know.
This is just a cleaner implementation of what you already suggested, but I would do it like this:
module type Foo = sig
class c : object
method x : int
end
end
module A = struct
class c = object method x = 4 end
end
module B = struct
class c = object method x = 5 end
end
let condition = true
module M = (val if condition then (module A) else (module B) : Foo)
class d = object (self)
inherit M.c
method y = self#x + 2
end
Not contrary to what Jeffrey said, but you can achieve this using first-class modules. Also, I'm not sure that you really need to create classes on runtime, maybe creating object would be enough. If what you want to get is different behavior, then this will be enough.
OCaml has a static type system. You can't do anything at runtime that would affect the type of something. Inheritance affects the types of things, so it can't happen at runtime.
(Your code also confuses types with values, which is understandable.)
There is probably a way to get close to what you want while retaining the desirable properties of static typing. What you actually need may be more like function composition, perhaps.
Referring to the answer of the following question:
List.empty vs. List() vs. new List()
How did the developers of Scala map the method apply[A](xs: A*) defined in the List object, to be usable as List[A](cs: A*)?
Also how did they translate aListInstance(n: Int) to the method apply(n: Int) (which returns the n'th element of the list) defined in the List class?
In both cases I'm calling the apply() methods without writing .apply() in my code. How does that work?
It works because the Scala Language Specification says so.
foo(bar)
is translated to
foo.apply(bar)
just like
foo.bar = baz
is translated to
foo.bar_=(baz)
and
foo(bar) = baz
is translated to
foo.update(bar, baz)
and
foo bar baz
is translated to
foo.bar(baz)
and
foo bar_: baz
is translated to
baz.bar_:(foo)
and so on and so forth.