I'm making my own Serializable trait, in the context of a client / server system.
My idea was that the messages sent by the system is an enum made by the user of this system, so it can be customize as needed.
Too ease implementing the trait on the enum, I would like to use the #[derive(Serializable)] method, as implementing it is always the same thing.
Here is the trait :
pub trait NetworkSerializable {
fn id(&self) -> usize;
fn size(&self) -> usize;
fn serialize(self) -> Vec<u8>;
fn deserialize(id: usize, data: Vec<u8>) -> Self;
}
Now, I've tried to look at the book (this one too) and this example to try to wrap my head around derive macros, but I'm really struggling to understand them and how to implement them. I've read about token streams and abstract trees, and I think I understand the basics.
Let's take the example of the id() method : it should gives a unique id for each variant of the enum, to allow headers of messages to tell which message is incoming.
let's say I have this enum as a message system :
enum NetworkMessages {
ErrorMessage,
SpawnPlayer(usize, bool, Transform), // player id, is_mine, position
MovePlayer(usize, Transform), // player id, new_position
DestroyPlayer(usize) // player_id
}
Then, the id() function should look like this :
fn id(&self) -> usize {
match &self {
&ErrorMessage => 0,
&SpawnPlayer => 1,
&MovePlayer => 2,
&DestroyPlayer => 3,
}
}
Here was my go with writting this using a derive macro :
#[proc_macro_derive(NetworkSerializable)]
pub fn network_serializable_derive(input: TokenStream) -> TokenStream {
// Construct a representation of Rust code as a syntax tree
// that we can manipulate
let ast = syn::parse(input).unwrap();
// Build the trait implementation
impl_network_serializable_macro(&ast)
}
fn impl_network_serializable_macro(ast: &syn::DeriveInput) -> TokenStream {
// get enum name
let ref name = ast.ident;
let ref data = ast.data;
let (id_func, size_func, serialize_func, deserialize_func) = match data {
// Only if data is an enum, we do parsing
Data::Enum(data_enum) => {
// Iterate over enum variants
let mut id_func_internal = TokenStream2::new();
let mut variant_id: usize = 0;
for variant in &data_enum.variants {
// add the branch for the variant
id_func_internal.extend(quote_spanned!{
variant.span() => &variant_id,
});
variant_id += 1;
}
(id_func_internal, (), (), ())
}
_ => {(TokenStream2::new(), (), (), ())},
};
let expanded = quote! {
impl NetworkSerializable for #name {
// variant_checker_functions gets replaced by all the functions
// that were constructed above
fn size(&self) -> usize {
match &self {
#id_func
}
}
/*
#size_func
#serialize_func
#deserialize_func
*/
}
};
expanded.into()
}
So this is generating quite a lot of errors, with the "proc macro NetworkSerializable not expanded: no proc macro dylib present" being first. So I'm guessing there a lot of misunderstaning from my part in here.
If I have variables like this:
let a: u32 = ...;
let b: Option<u32> = ...;
let c: u32 = ...;
, what is the shortest way to make a vector of those values, so that b is only included if it's Some?
In other words, is there something simpler than this:
let v = match b {
None => vec![a, c],
Some(x) => vec![a, x, c],
};
P.S. I would prefer a solution where we don't need to use the variables more than once. Consider this example:
let some_person: String = ...;
let best_man: Option<String> = ...;
let a_third_person: &str = ...;
let another_opt: Option<String> = ...;
...
As can be seen, we might have to use longer variable names, more than one Option (None), expressions (like a_third_person.to_string()), etc.
Yours is fine, but here's a sophisticated one:
[Some(a), b, Some(c)].into_iter().flatten().collect::<Vec<_>>()
This works since Option impls IntoIterator.
If it depends on just one variable:
b.map(|b| vec![a, b, c]).unwrap_or_else(|| vec![a, c]);
Playground
After some thinking and investigating, I've come with the following crazy thing.
The end goal is to have a macro, optional_vec![], that you can pass it either T or Option<T> and it should behave like described in the question. However, I decided on a strong restriction: it should have the best performance possible. So, you write:
optional_vec![a, b, c]
And get at least the performance of hand-written match, if not more. This forbids the use of the simple [Some(a), b, Some(c)].into_iter().flatten().collect::<Vec<_>>(), suggested in my other answer (though even this solution needs some way to differentiate between Option<T> and just T, which, like we'll see, is not an easy problem at all).
I will first warn that I've not found a way to make my macro work with Option. That is, if you want to build a vector of Option<T> from Option<T> and Option<Option<T>>, it will not work.
When a design a complex macro, I like to think first how the expanded code will look like. And in this macro, we have several hard problems to solve.
First, the macro take plain expressions. But somehow, it needs to switch on their type being T or Option<T>. How should such thing be done?
The feature we use to do such things is specialization.
#![feature(specialization)]
pub trait Optional {
fn some_method(self);
}
impl<T> Optional for T {
default fn some_method(self) {
// Just T
}
}
impl<T> Optional for Option<T> {
fn some_method(self) {
// Option<T>
}
}
Like you probably noticed, now we have two problems: first, specialization is unstable, and I'd like to stay with stable. Second, what should be inside the trait? The second problem is easier to solve, so let's begin with it.
Turns out that the most performant way to do the pushing to the vector is to pre-allocate capacity (Vec::with_capacity), write to the vector by using pointers (don't push(), it optimizes badly!) then set the length (Vec::set_len()).
We can get a pointer to the internal buffer of the vector using Vec::as_mut_ptr(), and advance the pointer via <*mut T>::add(1).
So, we need two methods: one to hint us about the capacity (can be zero for None or one for Some() and non-Option elements), and a write_and_advance() method:
pub trait Optional {
type Item;
fn len(&self) -> usize;
unsafe fn write_and_advance(self, place: &mut *mut Self::Item);
}
impl<T> Optional for T {
default type Item = Self;
default fn len(&self) -> usize { 1 }
default unsafe fn write_and_advance(self, place: &mut *mut Self) {
place.write(self);
*place = place.add(1);
}
}
impl<T> Optional<T> for Option<T> {
type Item = T;
fn len(&self) -> usize { self.is_some() as usize }
unsafe fn write_and_advance(self, place: &mut *mut T) {
if let Some(value) = self {
place.write(value);
*place = place.add(1);
}
}
}
It doesn't even compile! For the why, see Mismatch between associated type and type parameter only when impl is marked `default`. Luckily for us, the trick we'll use to workaround specialization not being stable does work in this situation. But for now, let's assume it works. How will the code using this trait look like?
match (a, b, c) { // The match is here because it's the best binding for liftimes: see https://stackoverflow.com/a/54855986/7884305
(a, b, c) => {
let len = Optional::len(&a) + Optional::len(&b) + Optional::len(&c);
let mut result = ::std::vec::Vec::with_capacity(len);
let mut next_element = result.as_mut_ptr();
unsafe {
Optional::write_and_advance(a, &mut next_element);
Optional::write_and_advance(b, &mut next_element);
Optional::write_and_advance(c, &mut next_element);
result.set_len(len);
}
result
}
}
And it works! Except that it does not, because the specialization does not compile as I said, and we also want to not repeat all of this boilerplate but insert it into a macro.
So, how do we solve the problems with specialization: being unstable and not working?
dtonlay has a very cool trick he calls autoref specialization (BTW, all of this repo is a very recommended reading!). This is a trick that can be used to emulate specialization. It works only in macros, but we're in a macro so this is fine.
I will not elaborate about the trick here (I recommend to read his post; he also used this trick in the excellent and very widely used anyhow crate). In short, the idea is to trick the typechecker by implementing a trait for T under certain conditions (the specialized impl) and other trait for &T for the general case (this could be inherent impl if not coherence). Since Rust performs automatic referencing during method resolution, that is take reference to the receiver as needed, this will work - the typechecker will autoref if needed, and will stop in the first applicable impl - i.e. the specialized impl if it matches, or the general impl otherwise.
Here's an example:
use std::fmt;
pub trait Display {
fn foo(&self);
}
// Level 1
impl<T: fmt::Display> Display for T {
fn foo(&self) { println!("Display({}), {}", std::any::type_name::<T>(), self); }
}
pub trait Debug {
fn foo(&self);
}
// Level 2
impl<T: fmt::Debug> Debug for &T {
fn foo(&self) { println!("Debug({}), {:?}", std::any::type_name::<T>(), self); }
}
macro_rules! foo {
($e:expr) => ((&$e).foo());
}
Playground.
We can use this trick in our case:
#[doc(hidden)]
pub mod autoref_specialization {
#[derive(Copy, Clone)]
pub struct OptionTag;
pub trait OptionKind {
fn optional_kind(&self) -> OptionTag;
}
impl<T> OptionKind for Option<T> {
#[inline(always)]
fn optional_kind(&self) -> OptionTag { OptionTag }
}
impl OptionTag {
#[inline(always)]
pub fn len<T>(self, this: &Option<T>) -> usize { this.is_some() as usize }
#[inline(always)]
pub unsafe fn write_and_advance<T>(self, this: Option<T>, place: &mut *mut T) {
if let Some(value) = this {
place.write(value);
*place = place.add(1);
}
}
}
#[derive(Copy, Clone)]
pub struct DefaultTag;
pub trait DefaultKind {
fn optional_kind(&self) -> DefaultTag;
}
impl<T> DefaultKind for &'_ T {
#[inline(always)]
fn optional_kind(&self) -> DefaultTag { DefaultTag }
}
impl DefaultTag {
#[inline(always)]
pub fn len<T>(self, _this: &T) -> usize { 1 }
#[inline(always)]
pub unsafe fn write_and_advance<T>(self, this: T, place: &mut *mut T) {
place.write(this);
*place = place.add(1);
}
}
}
And the expanded code will look like:
use autoref_specialization::{DefaultKind as _, OptionKind as _};
match (a, b, c) {
(a, b, c) => {
let (a_tag, b_tag, c_tag) = (
(&a).optional_kind(),
(&b).optional_kind(),
(&c).optional_kind(),
);
let len = a_tag.len(&a) + b_tag.len(&b) + c_tag.len(&c);
let mut result = ::std::vec::Vec::with_capacity(len);
let mut next_element = result.as_mut_ptr();
unsafe {
a_tag.write_and_advance(a, &mut next_element);
b_tag.write_and_advance(b, &mut next_element);
c_tag.write_and_advance(c, &mut next_element);
result.set_len(len);
}
result
}
}
It may be tempting to try to convert this immediately into a macro, but we still have one unsolved problem: our macro need to generate identifiers. This may not be obvious, but what if we pass optional_vec![1, Some(2), 3]? We need to generate the bindings for the match (in our case, (a, b, c) => ...) and the tag names ((a_tag, b_tag, c_tag)).
Unfortunately, generating names is not something macro_rules! can do in today's Rust. Fortunately, there is an excellent crate paste (another one from dtonlay!) that is a small proc-macro that allows you to do that. It is even available on the playground!
However, we need a series of identifiers. That can be done with tt-munching, by repeatedly adding some letter (I used a), so you get a, aa, aaa, ... you get the idea.
#[doc(hidden)]
pub mod reexports {
pub use std::vec::Vec;
pub use paste::paste;
}
#[macro_export]
macro_rules! optional_vec {
// Empty case
{ #generate_idents
exprs = []
processed_exprs = [$($e:expr,)*]
match_bindings = [$($binding:ident)*]
tags = [$($tag:ident)*]
} => {{
use $crate::autoref_specialization::{DefaultKind as _, OptionKind as _};
match ($($e,)*) {
($($binding,)*) => {
let ($($tag,)*) = (
$((&$binding).optional_kind(),)*
);
let len = 0 $(+ $tag.len(&$binding))*;
let mut result = $crate::reexports::Vec::with_capacity(len);
let mut next_element = result.as_mut_ptr();
unsafe {
$($tag.write_and_advance($binding, &mut next_element);)*
result.set_len(len);
}
result
}
}
}};
{ #generate_idents
exprs = [$e:expr, $($rest:expr,)*]
processed_exprs = [$($processed_exprs:tt)*]
match_bindings = [$first_binding:ident $($bindings:ident)*]
tags = [$($tags:ident)*]
} => {
$crate::reexports::paste! {
$crate::optional_vec! { #generate_idents
exprs = [$($rest,)*]
processed_exprs = [$($processed_exprs)* $e,]
match_bindings = [
[< $first_binding a >]
$first_binding
$($bindings)*
]
tags = [
[< $first_binding a_tag >]
$($tags)*
]
}
}
};
// Entry
[$e:expr $(, $exprs:expr)* $(,)?] => {
$crate::optional_vec! { #generate_idents
exprs = [$($exprs,)+]
processed_exprs = [$e,]
match_bindings = [__optional_vec_a]
tags = [__optional_vec_a_tag]
}
};
}
Playground.
I can also personally recommend
let mut v = vec![a, c];
v.extend(b);
Short and clear.
Sometime the straight forward solution is the best:
fn jim_power(a: u32, b: Option<u32>, c: u32) -> Vec<u32> {
let mut acc = Vec::with_capacity(3);
acc.push(a);
if let Some(b) = b {
acc.push(b);
}
acc.push(c);
acc
}
fn ys_iii(
some_person: String,
best_man: Option<String>,
a_third_person: String,
another_opt: Option<String>,
) -> Vec<String> {
let mut acc = Vec::with_capacity(4);
acc.push(some_person);
best_man.map(|x| acc.push(x));
acc.push(a_third_person);
another_opt.map(|x| acc.push(x));
acc
}
If you don't care about the order of the values, another option is
Iterator::chain(
[a, c].into_iter(),
[b].into_iter().flatten()
).collect()
Playground
I've been trying to implement a Strategy pattern in rust, but I'm having trouble understanding how to make it work.
So let's imagine we have a trait Adder and Element:
pub trait Element {
fn to_string(&self) -> String;
}
pub trait Adder {
type E: Element;
fn add (&self, a: &Self::E, b: &Self::E) -> Self::E;
}
And we have two implementations StringAdder with StringElements and UsizeAdder with UsizeElements:
// usize
pub struct UsizeElement {
pub value: usize
}
impl Element for UsizeElement {
fn to_string(&self) -> String {
self.value.to_string()
}
}
pub struct UsizeAdder {
}
impl Adder for UsizeAdder{
type E = UsizeElement;
fn add(&self, a: &UsizeElement, b: &UsizeElement) -> UsizeElement{
UsizeElement { value: a.value + b.value }
}
}
// String
pub struct StringElement {
pub value: String
}
impl Element for StringElement {
fn to_string(&self) -> String {
self.value.to_string()
}
}
pub struct StringAdder {
}
impl Adder for StringAdder {
type E = StringElement;
fn add(&self, a: &StringElement, b: &StringElement) -> StringElement {
let a: usize = a.value.parse().unwrap();
let b: usize = b.value.parse().unwrap();
StringElement {
value: (a + b).to_string()
}
}
}
And I want to write a code that uses trait methods from Adder trait and it's corresponding elements without knowing at compile time which strategy is going to be used.
fn main() {
let policy = "usize";
let element = "1";
let adder = get_adder(&policy);
let element_a = get_element(&policy, element);
let result = adder.add(element_a, element_a);
}
To simplify I'm going to assign a string to policy and element but normally that would be read from a file.
Is the only way to implement get_adder and get_element using dynamic dispatch? And by extension should I define Adder and Element traits to use trait objects and or the Any trait?
Edit: Here is what I managed to figure out so far.
An example of possible implementation is using match to help define concrete types for the compiler.
fn main() {
let policy = "string";
let element = "1";
let secret_key = "5";
let result = cesar(policy, element, secret_key);
dbg!(result.to_string());
}
fn cesar(policy: &str, element: &str, secret_key: &str) -> Box<dyn Element>{
match policy {
"usize" => {
let adder = UsizeAdder{};
let element = UsizeElement{ value: element.parse().unwrap() };
let secret_key = UsizeElement{ value: secret_key.parse().unwrap() };
Box::new(cesar_impl(&adder, &element, &secret_key))
}
"string" => {
let adder = StringAdder{};
let element = StringElement{ value: element.to_string() };
let secret_key = StringElement{ value: secret_key.to_string() };
Box::new(cesar_impl(&adder, &element, &secret_key))
}
_ => {
panic!("Policy not supported!")
}
}
}
fn cesar_impl<A>(adder: &A, element: &A::E, secret_key: &A::E) -> A::E where A: Adder, A::E : Element {
adder.add(&element, &secret_key)
}
However the issue is that I have to wrap every function I want to implement using a match function to determine the concrete type, and also case for every policy available.
It does not seem like the proper way of implementing it as it will bloat the code, make it more error prone and less maintainable unless I end up using macros.
Edit 2: Here you can find an example using dynamic dispatch. However I'm not convinced it's the proper way to implement the solution.
Example using dynamic dispatch
Thank you for your help :)
How to parse some string to most appropriate type?
I know there is .parse::<>() method, but you need to specify type in advance like this:
fn main() {
let val = String::from("54");
assert_eq!(val.parse::<i32>().unwrap(), 54i32);
let val = String::from("3.14159");
assert_eq!(val.parse::<f32>().unwrap(), 3.14159f32);
let val = String::from("Hello!");
assert_eq!(val.parse::<String>().unwrap(), "Hello!".to_string());
}
But I need something like this:
fn main() {
let val = String::from("54");
assert_eq!(val.generic_parse().unwrap(), 54i32); // or 54i16 or 54 u32 or etc ...
let val = String::from("3.14159");
assert_eq!(val.generic_parse().unwrap(), 3.14159f32);
let val = String::from("Hello!");
assert_eq!(val.generic_parse().unwrap(), "Hello!".to_string());
}
Is there an appropriate crate for something like this? I don't want to re-invent the wheel for the umpteenth time.
EDIT
This is what I actually want to do:
struct MyStruct<T> {
generic_val: T,
}
fn main() {
let val = String::from("54");
let b = MyStruct {generic_val: val.parse().unwrap()};
let val = String::from("3.14159");
let b = MyStruct {generic_val: val.parse().unwrap()};
}
Error:
error[E0282]: type annotations needed for `MyStruct<T>`
--> src/main.rs:7:13
|
7 | let b = MyStruct {generic_val: val.parse().unwrap()};
| - ^^^^^^^^ cannot infer type for type parameter `T` declared on the struct `MyStruct`
| |
| consider giving `b` the explicit type `MyStruct<T>`, where the type parameter `T` is specified
You need to base things on the right Enum type and implement FromStr for it. Like this.
#[derive(PartialEq, Debug)]
enum Val {
Isize(isize),
F64(f64),
}
impl core::str::FromStr for Val {
type Err = & 'static str;
fn from_str(s: &str) -> Result<Self, Self::Err> {
match (s.parse::<isize>(), s.parse::<f64>()) {
(Ok(i),_) => Ok(Val::Isize(i)),
(Err(_), Ok(f)) => Ok(Val::F64(f)),
(Err(_), Err(_)) => Err("neither parser worked"),
}
}
}
fn main() {
assert_eq!("34".parse(), Ok(Val::Isize(34)));
assert_eq!("12.3".parse(), Ok(Val::F64(12.3)));
assert!("wrong".parse::<Val>().is_err());
}
Rust is a statically typed language. This means that the compiler needs to know the type of variables at compile time. There are three ways things can go from there:
If your strings are known at compile-time, then you might as well replace them with literal values in your code (eg. "54" → 54).
If you have some other way of knowing at compile time what type a given string should parse to, then you can specify the appropriate type when parsing the string: let a = "54".parse::<i32>().unwrap()
If your strings are only known at run-time and you want to autodetect the type, then you need to use some kind of enumerated value that will store the type alongside the value in your program:
use std::str::FromStr;
enum Value {
I32 (i32),
F32 (f32),
String (String),
}
impl Value {
fn new (s: &str) -> Value {
if let Ok (v) = s.parse::<i32>() {
Value::I32 (v)
} else if let Ok (v) = s.parse::<f32>() {
Value::F32 (v)
} else {
Value::String (s.into())
}
}
}
That way, the rest of your code will have a way of knowing what type was detected and to adjust its processing accordingly.
Below is the example given by the mod documentation of syn::parse.
enum Item {
Struct(ItemStruct),
Enum(ItemEnum),
}
struct ItemStruct {
struct_token: Token![struct],
ident: Ident,
brace_token: token::Brace,
fields: Punctuated<Field, Token![,]>,
}
impl Parse for Item {
fn parse(input: ParseStream) -> Result<Self> {
let lookahead = input.lookahead1();
if lookahead.peek(Token![struct]) {
input.parse().map(Item::Struct) // <-- here
} else if lookahead.peek(Token![enum]) {
input.parse().map(Item::Enum) // <-- and here
} else {
Err(lookahead.error())
}
}
}
Is input.parse().map(Item::Struct) a valid normal Rust syntax (appears not as Item::Struct is not a function), or is it a kind of special syntax for proc_macro libs? If the latter is the case, is there a documentation of the proc_macro specific syntax rules?
This syntax is standard Rust syntax. You can use tuple struct or tuple struct-like enum variants as functions. See this small example:
enum Color {
Str(String),
Rgb(u8, u8, u8),
}
struct Foo(bool);
// Use as function pointers (type annotations not necessary)
let f: fn(String) -> Color = Color::Str;
let g: fn(u8, u8, u8) -> Color = Color::Rgb;
let h: fn(bool) -> Foo = Foo;
In the next example, those functions are directly passed to another function (like Option::map) (Playground):
// A function which takes a function
fn string_fn<O, F>(f: F) -> O
where
F: FnOnce(String) -> O,
{
f("peter".to_string())
}
string_fn(|s| println!("{}", s)); // using a clojure
string_fn(std::mem::drop); // using a function pointer
// Using the enum variant as function
let _: Color = string_fn(Color::Str);
You can find out more about this feature, in this chapter of the book.