Understanding Ok with error propagation operator? - rust

I've seen this pattern on more than one occasion:
fn f() -> Result<..., ...> {
...
Ok(expression()?)
}
Specifically, I find the order Ok(expression()?) confusing. What's the return type of expression()? and is there an interplay between Ok and this type? It seems, the function Ok must do more than just capturing the value. But how can it force a return with an error, if it's the last expression wrapping the return type of expression()?.

The ? operator either returns (as in the return statement) the error variant, or extracts the value from the Ok variant. Thus Ok(expression()?) is more or less equal to:
// This pattern is so common in rust, that the language designers
// introduced the `?` to reduce the amount of boilerplate needed
let result = expression();
let r = match result{
Err(e) => return Err(e.into()), // tries to convert the error into the required type if necessary (and if possible)
Ok(r) => r
}
Ok(r)
So basically Ok(expression()?) can be simplified to just
expression() if its Err variant matches the one from the function definition.
Historical context:
Extracting the Ok() variant and propagating the Err variant is a very common pattern in rust. In order to reduce the amount of boilerplate code, the rust team introduced the try!() macro; But it's so clumsy and does not really work well with method chaining. The rust team introduced the ? operator, which is doing absolutely the same thing, except for it's not so intrusive as try!()

? just leverages the error returned by expression if it exists or unwraps the value from within the returned Ok if not. Ok(expression()?) could be just expresion() and clippy would warn you about that:
warning: question mark operator is useless here
--> src/lib.rs:8:5
|
8 | Ok(expression()?)
| ^^^^^^^^^^^^^^^^^ help: try: `expression()`
|
= note: `#[warn(clippy::needless_question_mark)]` on by default
= help: for further information visit https://rust-lang.github.io/rust-clippy/master/index.html#needless_question_mark
Playground
You can follow the documentation for further information
As per the comments. Notice that ? is useful if the error types differ but you have an implementation of From between error types or you use Box<dyn Error> as return type.
In which case you can either, bubble up the error if it is an error, otherwise rewrap the value with an Ok
Ok(expression()?)
or, map your error type into the correct error type with map_err
expression.map_err(|e| ...)

Related

Confusing error message on ? (unwrap operator)

I have this rust method error message:
error[E0277]: the `?` operator can only be used on `Option`s,
not `Result`s, in an async function that returns `Option`
I must admit, I often encounter Rust error messages which appear confusing to me, while to most other coders they make absolute sense.
So, I apologize in advance for posting this question.
First of all: what does that second comma in the error message mean? Should I read it as the following:
"If an async function call [within another function] returns an enum of the type Result then the ? operator can only be applied if, and only if, the respective [other] function also returns an enum of type Result and not an enum of type Option"
Pardon my verbose language. I hope I got my point across.
What also got me confused was the error message with that very same Reference i.e. error[E0277] , which is listed in the official rust error codes index, states:
"You tried to use a type which doesn't implement some trait in a place which expected that trait."
In which universe do these two error messages have anything in common, except for the identical reference number?
And here's the entire error block, which Rust produced:
error[E0277]: the `?` operator can only be used on `Option`s, not `Result`s, in an async function that returns `Option`
--> src/utils/tokenizer.rs:72:73
|
70 | pub async fn clear(&self) -> Option<String> {
| _________________________________________________-
71 | | let mut conn = self.pool.get().await.unwrap();
72 | | let mut iter: redis::AsyncIter<i32> = conn.sscan("my_set").await?;
| | ^ use `.ok()?` if you want to discard the `Result<Infallible, Red
Error>` error information
73 | | while let Some(element) = iter.next_item().await {
... |
79 | | Some(String::from("A"))
80 | | }
| |_____- this function returns an `Option`
|
= help: the trait `FromResidual<Result<Infallible, RedisError>>` is not implemented for `std::option::Option<std::string::String>`
= help: the following other types implement trait `FromResidual<R>`:
<std::option::Option<T> as FromResidual<Yeet<()>>>
<std::option::Option<T> as FromResidual>
For more information about this error, try `rustc --explain E0277`.
What is the canonical error message, the one from the error code index page or the one, the compiler produces?
Your function clear returns Option<String> but the function conn.sscan("my_set").await?; returns a Result. You cannot use ? to propagate the result because that doesn't match the return type of clear(&self) -> Option<String>.
You will need explicitly handle the Result or convert it to an Option<String>. You could also try adding .ok()? per the compiler hint and see what that gets you.
Whatever you pick the important thing is just to make sure that the return type matches what your ? operator is unwrapping.
Yes, your interpretation is correct, although to be pedantic the error says the opposite:
"If an async function returns Option then the ? operator can only be applied to expressions that has the type Option, and not Result".
And if you ask why on earth this has error code E0277? Well, this is because in order to be propagated using ? a type has to implement the (experimental) Try trait with some combination of type parameters and associated types corresponding to some other combination of the return type of the function. So, in essence, an incompatible type in ? boils down to an unsatisfied trait bound. But unless you're interested in the inner workings of ?, this is not really important.

Rust - the trait `StdError` is not implemented for `OsString`

I'm writing some Rust code which uses the ? operator. Here is a few lines of that code:
fn files() -> Result<Vec<std::string::String>, Box<Error>> {
let mut file_paths: Vec<std::string::String> = Vec::new();
...
file_paths.push(pathbuf.path().into_os_string().into_string()?);
...
Ok(file_paths)
}
However, even though I'm using ? on a Result it is giving me the following error:
`the trait `StdError` is not implemented for `OsString`.
This is contrary to the Rust documentation here, which states that:
The ? is shorthand for the entire match statements we wrote earlier. In other words, ? applies to a Result value, and if it was an Ok, it unwraps it and gives the inner value. If it was an Err, it returns from the function you're currently in.
I've confirmed that pathbuf.path().into_os_string().into_string() is of type Result, because when I remove the ?, I get the following compiler error:
expected struct `std::string::String`, found enum `std::result::Result`
(since file_paths is a Vector of strings, not Results).
Is this a bug with the Rust language or documentation?
In fact I tried this without pushing to the Vector, but simply initializing a variable with the value of pathbuf.path().into_os_string().into_string()?, and I got the same error.
The function OsString::into_string is a little unusual. It returns a Result<String, OsString> - so the Err variant is actually not an error.
In the event that the OsString cannot be converted into a regular string, then the Err variant is returned, containing the original string.
Unfortunately this means you cannot use the ? operator directly. However, you can use map_err to map the error variant into an actual error, like this:
file_paths.push(
pathbuf.path()
.into_os_string()
.into_string().
.map_err(|e| InvalidPathError::new(e))?
);
In the above example, InvalidPathError might be your own error type. You could also use an error type from the std library.

Why does Rust allow code with the wrong return type, but only with a trailing semicolon?

Consider the following Rust code:
fn f() -> i32 {
loop {
println!("Infinite loop!");
}
println!("Unreachable");
}
This compiles (with a warning) and runs, despite the fact that the return type is wrong.
It would seem that the compiler is OK with the return type of () in the last line because it detects that this code is unreachable.
However, if we remove the last semicolon:
fn f() -> i32 {
loop {
println!("Infinite loop!");
}
println!("Unreachable")
}
Then the code no longer compiles, giving a type error:
error[E0308]: mismatched types
--> src/main.rs:14:5
|
14 | println!("Unreachable")
| ^^^^^^^^^^^^^^^^^^^^^^^ expected `i32`, found `()`
|
= note: this error originates in a macro outside of the current crate (in Nightly builds, run with -Z external-macro-backtrace for more info)
Why is this? Isn't the return type the same, (), in both of these code snipets?
Note: I'm interested in understanding why the Rust compiler behaves differently on these two examples, i.e. how the Rust compiler is implemented. I did not mean to ask a philosophical question about how it "should" behave, from the perspective of language design (I understand that such a question would probably be off-topic).
The return type in the first code block is actually ! (called never) because you have a loop that never exits (so rust gives you a warning saying it's unreachable). The full type would be:
fn f() -> !
I suspect ! is more like the 'bottom' type in Rust than anything else. In the second case, your function likely errors out in an earlier stage during type checking because of the mismatch between i32 and () before the compiler gets to the 'unreachability' analysis, like it does in the first example.
edit: as suggested, here is the relevant part of the rust book https://doc.rust-lang.org/book/ch19-04-advanced-types.html#the-never-type-that-never-returns
(Converting Sven's first comment into an answer)
The Rust compiler needs to infer a type for the function body. In the first case, there is no return expression, and apparently the compiler infers ! as the return type because of the infinite loop, which makes sense. In the second case, there's a return expression, so the type inference solver uses that to infer the type, which also makes sense.
I don't think this is specified in the language reference, nor do I think it matters in any way – just omit the unreachable statement and you'll be fine.

Why does pattern matching on &Option<T> yield something of type Some(&T)?

I have a tiny playground example here
fn main() {
let l = Some(3);
match &l {
None => {}
Some(_x) => {} // x is of type &i32
}
}
I'm pattern matching on &Option and if I use Some(x) as a branch, why is x of type &i32?
The type of the expression &l you match against is &Option<i32>, so if we are strict the patterns should be &None and &Some(x), and if we use these patterns, the type of x indeed is i32. If we omit the ampersand in the patterns, as you did in your code, it first looks like the patterns should not be able to match at all, and the compiler should throw an error similar to "expected Option, found reference", and indeed this is what the compiler did before Rust version 1.26.
Current versions of Rust support "match ergonomics" introduced by RFC 2005, and matching a reference to an enum against a pattern without the ampersand is now allowed. In general, if your match expression is only a reference, you can't move any members out of the enum, so matching a reference against Some(x) is equivalent to matching against the pattern &Some(ref x), i.e. x becomes a reference to the inner value of the Option. In your particular case, the inner value is an i32, which is Copy, so you would be allowed to match against &Some(x) and get an i32, but this is not possible for general types.
The idea of the RFC is to make it easier to get the ampersands and refs in patterns right, but I'm not completely convinced whether the new rules actually simplified things, or whether they added to the confusion by making things magically work in some cases, thereby making it more difficult for people to get a true understanding of the underlying logic. (This opinion is controversial – see the comments.)

Returning from inside for loop causes type mismatch

I am attempting to return a function pointer, which is located inside a for loop, from a function located in an impl of a struct.
fn locate_func(&self, string: &str) -> fn() -> bool {
let mut func;
for alt in &self.alts {
return alt.func;
}
}
There will be an if statement inside the for loop in the future, but as I am testing things at the very moment, it looks rather generic, and somewhat illogical.
The above code in my mind, is supposed to return the pointer to alt.func(), which clearly is a pointer, as it tells me so should I remove the return and semicolon of that line.
error[E0308]: mismatched types
--> src\main.rs:42:3
|
42 | for alt in &self.alts
| ^ expected fn pointer, found ()
|
= note: expected type `fn() -> bool`
= note: found type `()`
Above is the error that is caused upon running locate_func(). I am clearly missing something as the aforementioned code is not working properly. Any hints?
Your for-loop is the last expression inside the function. The compiler expects the last expression to evaluate to the return type. But all loops evaluate to () (unit or void), so the compiler has a classic type mismatch there.
The correct question to ask yourself is: what would happen if the return inside of the loop wouldn't be executed (for example, because the loop isn't executed at all, because self.alts is empty)? This would lead to problems, wouldn't it?
So you have to return a valid object after the for-loop to cover that case. But if you are certain that the spot after the loop will never be reached you can use unreachable!(); to tell the compiler what you already know. However, if the program will reach this spot, it will panic! So better make sure, you know for certain how the program behaves.

Resources