Call me a weirdo but the more errors a compilers give me the happier (albeit a bit frustrated) I am. That stuff generally surfaces in a way or another… and I prefer at compile time 🙂
That said I haven’t spent quality time with Rust yet… so not sure if there are a lot of nitpicks (ala go) or these are valgrind-level of “holy s*** I am so grateful to this tool” 😃
A friend told me about rust around 8 years ago and this was very much my first experience (at least with &str and lifetimes and borrow errors).
Learning C, that smasher would never have stopped.
It would also be emitting significantly less helpful messages.
“What messages?”
Lol yes
Segmentation fault?
The weird part of rust is replacing straight forward semicolons from other languages with the more verbose
.unwrap();
.Just kidding, don’t lecture me about it.
You can also use a shorter version .clone();
?
But then someone will have to deal with it somewhere, better just unwrap it under the carpet.
expr
The amount of people on the internet seriously complaining that both Rust error handling sucks and that
.unwrap();
is too verbose is just staggering.I’ll be honest, when I was learning to program in Java I mostly just wrapped errors in an empty try catch to shut them up, with no regard for actually handling them.
I assume most other learners do that too.
Java requiring you to write every exception that can happen in your code isn’t helpful.
Explicit error types are great, but Java managed to make them on a way where you get almost none of the upside and is so full of downsides that indoctrinated a generation into thinking knowing your errors is bad.
I think the problem is that many introductory examples use
unwrap
, so many beginner programmers don’t get exposed to alternatives likeunwrap_or
and the likes.Yeah, we onboarded some folks into a Rust project last year and a few months in, they were genuinely surprised when I told them that unwrapping is pretty bad. Granted, they probably did read about it at some point and just forgot, but that isn’t helped by lots of code using
.unwrap()
either.
Me, every time I try searching a Rust question.
That’s easy. Just do:
fn is_second_num_positive() -> bool { let input = "123,-45"; let is_positive = input.split(',') .collect::<Vec<&str>>() .last() .unwrap() .parse::<i32>() .unwrap() .is_positive(); is_positive }
Can’t resist pointing out how you should actually write the function in a “real” scenario (but still not handling errors properly), in case anyone wants to know.
If the list is guaranteed to have exactly two elements:
fn is_second_num_positive_exact(input: &str) -> bool { let (_, n) = input.split_once(',').unwrap(); n.parse::<i32>().unwrap() > 0 }
If you want to test the last element:
fn is_last_num_positive(input: &str) -> bool { let n = input.split(',').next_back().unwrap(); n.parse::<i32>().unwrap() > 0 }
If you want to test the 2nd (1-indexed) element:
fn is_second_num_positive(input: &str) -> bool { let n = input.split(',').nth(1).unwrap(); n.parse::<i32>().unwrap() > 0 }
Even better to use
expect
with a short message of what the assumption is: “the string should contain a comma” if it ever panics you’ll know exactly why.Can still use
.is_positive()
, though…
This is my experience every time I return to learning rust. I’m guessing that if I used it more often than once a quarter with hobby projects I’d stop falling into the same traps.
I find that the error messages themselves are a great tool for learning when it comes to Rust.
Eh, I’m not entirely sold on that idea.
I think they do a good job of pointing out “this is a behavior/feature of Rust you need to understand.” However they can send you down the wrong path of correction.
The compiler error mentioning static lifetime specifiers of
&str
demonstrates both. It indicates to the developer that ownership and scopes will play a significant role when defining and accessing data. The error though will guide them towards researching how to define static lifetimes and possibly believe that they will need to set this in their functions and structs. Each time you look at questions about this error in places like Stack Overflow with example code you’ll find suggested solutions explaining that a manually-defined static lifetime isn’t necessary to resolve the problem.
Yeah, these become a lot less relevant with routine.
-
Avoiding the main-thread panicking is mostly just a matter of not using
.unwrap()
and.expect()
. -
String
vs.&str
can mostly be solved by generally using owned datatypes (String
) for storing in structs and using references (&str
) for passing into function parameters. It does still happen that you forget the&
at times, but that’s then trivial to solve (by just adding the&
). -
“temporary value dropped while borrowed” can generally be avoided by not passing references outside of your scope/function. You want to pass the owned value outside. Clone, if you have to.
-
“missing lifetime specifier” is also largely solved by not storing references in structs.
The last two points are the kind of design advice I need to see. I’m probably so used to the C/C++ concept of passing by reference to prevent copies of complex data being generated that I forget how Rust’s definition of a reference is different.
-
Skill Issue.
For reals though adopting a functional style of programming makes rust extremely pleasant . It’s only when people program in object oriented styles that this gets annoying.
No loops, and no state change make rust devs happy devs.
I’m a OOP programmer.
I wrap everything within
Arc<Mutex<>>
.I’m a happy dev.
I mean yah. That’s what it takes. But like when I try to write code around Arc<_> the performance just tanks in highly concurrent work. Maybe it’s an OOP rust skill issue on my end. Lol.
Avoiding this leads, for me at least, to happiness and fearless, performant, concurrent work.
I’m not a huge fan of go-lang but I think they got it right with the don’t communicate by sharing memory thing.
You mean mutex? Arc allows synchronous read only access by multiple threads, so it’s not a performance bottleneck. Locking a mutex would be one.
I mean it could be Mutex, or Rwlock or anything atomic. It’s just when I have to put stuff into an Arc<> to pass around I know trouble is coming.
I just started learning rust like two days ago and I haven’t had too many issues with OOP so far… is it going to get considerably worse as the complexity of my projects increases?
The thing with OOP, particularly how it’s used in GCed languages, is that it’s all about handing references out to wherever and then dealing with the complexity of not knowing who has access to your fields via getters & setters, or by cloning memory whenever it’s modified in asynchronous code.
Rust has quite the opposite mindset. It’s all about tracking where references go. It pushes your code to be very tree-shaped, i.e. references typically¹ only exist between a function and the functions it calls underneath. This is what allows asynchronous code to be safe in Rust, and I would also argue that the tree shape makes code easier to understand, too.
But yeah, some of the patterns you might know from OOP will not work in Rust for that reason. You will likely need to get into a different mindset over time.
Also just in case: We are talking OOP in the sense of the paradigm, i.e. object-oriented.
Just using objects, i.e. data with associated functions/methods, that works completely normal in Rust.¹) If you genuinely need references that reach outside the tree shape, which is mostly going to be the case, if you work with multiple threads, then you can do so by wrapping your data structures in
Arc<Mutex<_>>
or similar. But yeah, when learning, you should try to solve your problems without these. Most programs don’t need them.It will become more complex when you start needing circular references in your datastructures.
You’ll be fine. You will learn the lifetime stuff and all will work out. It’s not that bad to be honest.
Worse in the sense of more errors, sure, but as you go you’ll pick up more of the rust patterns of thinking and imo it’s very worth it. It’s an odd blend and can be a bit verbose but I definitely prefer it to a pure OO or pure functional style language personally