Rust: baby step -- a Fibonacci sequence function.
I have spent the last twenty (20) odd days learning Rust :) We should apply what we have learned, so that the knowledge becomes ours. We are looking at a simple “traditional” recursive function which returns the nth Fibonacci number.
Rust: baby step – a Fibonacci sequence function. |
Installing Rust on Windows 10 and Ubuntu 22.10 is fairly simple and should take no times at all.
For the tutorial, I went through all twenty (20) chapters of “the book” – it felt like university all over again 😂 I’ve been warned that the learning curve is very steep, but I was not able to anticipate how steep it is!
I appreciate “the book”, I think it covers all important aspects of Rust. I certainly don’t remember all the things I’ve learned, but I should know what to look for when I run into problems.
In my mind, the concept of ownership in Rust is what makes it hard. I think once we have a hang of it, it should be a pleasurable language to use. I like the fact that the compiler does a lot to mitigate run-time problems as much as possible.
Rust aims to be a low level language. And it can be used to program for bare metal microcontrollers, too. I have positive opinions of Rust. I’ve been exposed to C and C++, I feel that Rust is not any harder than the two mentioned. Even though Rust has a different set of terminologies, some of Rust features I can relate to Delphi’s, which is the language I’ve used the longest.
With NodeJs and Python, within the first hours of learning, I was able to connect to different databases and running queries happily. With Rust, after twenty odd days, I feel like I know nothing, perhaps because, even just to do some meaningful programming exercise in Rust, we need to be fluent in a lot of Rust features? So I’ve decided to take simple steps to absorb Rust. Hence this post, a documentation of my Rust learning process.
For this starting exercise, I am going back to my very first programming book, which I bought on March 12, 1991: Elliot B. Koffman and Bruce R. Maxim, Turbo Pascal, 2nd Edition Covers 4.0/5.0, Addison-Wesley Publishing Company, Inc., May 1989, U.S.A., and translate the Fibonacci recursive function into Rust. I’m reprinting the function of the above book verbatim:
Figure 15.13 Recursive Function Fibonacci, page 608:
function Fibonacci( N: Integer ) : Integer;
{
Computes the Nth Fibonacci number.
Pre: N is defined and N > 0
Post: Returns the Nth Fibonacci number.
}
begin {Fibonacci}
if (N = 1) or (N = 2) then
Fibonacci := 1
else
Fibonacci := Fibonacci(N-2) + Fibonacci(N-1)
end; {Fibonacci}
(I don’t think Turbo Pascal ever had the function Result
variable, it was introduced during the Borland Delphi versions?)
– For more information on Fibonacci sequence.
I fancy a library package named maths
, while inside the
/rust
sub-directory, we run the following command to
create it:
$ cargo new maths --lib
On Windows 10, the source files for this library package looks like:
F:\rust>tree maths /F /A
F:\RUST\MATHS
| .gitignore
| Cargo.lock
| Cargo.toml
|
+---src
| lib.rs
| main.rs
...
And on Ubuntu 22.10:
behai@hp-pavilion-15:~/rust$ tree maths/
maths/
├── Cargo.lock
├── Cargo.toml
├── src
│ ├── lib.rs
│ └── main.rs
...
– Please note: main.rs
was created manually.
Content of /home/behai/rust/maths/src/lib.rs:
/// Computes the nth Fibonacci number.
///
/// # Examples
///
/// ```
/// use maths::fibonacci;
///
/// let n: u32 = 11;
/// let result: (bool, u32) = fibonacci(n);
/// println!("{n}th Fibonacci number: {:?}.\n", result);
///
/// assert_eq!(true, result.0);
/// assert_eq!(89, result.1);
/// ```
///
/// Pre: n is defined and n > 0.
///
/// Post: Returns a ``tuple`` of two (2) elements. The first element is a ``bool``,
/// ``true`` indicates a success, ``false`` a failure. The second element is the nth
/// Fibonacci number if the first element is ``true``.
pub fn fibonacci(n: u32) -> (bool, u32) {
if n == 0 {
return (false, 0);
}
if (n == 1) | (n == 2 ) {
(true, 1)
}
else {
(true, fibonacci(n-2).1 + fibonacci(n-1).1)
}
}
#[cfg(test)]
mod tests {
use super::*;
#[test]
fn fibonacci_valid() {
let result: (bool, u32) = fibonacci(6);
assert_eq!(result.0, true);
assert_eq!(result.1, 8);
let result: (bool, u32) = fibonacci(1);
assert_eq!(result.0, true);
assert_eq!(result.1, 1);
let result: (bool, u32) = fibonacci(2);
assert_eq!(result.0, true);
assert_eq!(result.1, 1);
}
#[test]
fn fibonacci_invalid() {
let result: (bool, u32) = fibonacci(0);
assert_eq!(result.0, false);
}
}
An u32
can only have values from 0 onward, so
the condition if n == 0
suffices – the compiler
will not let us pass in a literal negative integer nor any
type other than u32
in place of n
.
The original Turbo Pascal implementation does not handle
N <= 0
; I’m handling invalid n
,
for simplicity, I’m using a tuple
to return
the function result.
Rust also include formal documentation support.
I include a proper documentation for the function,
please note an example section in the documentation,
cargo
does test for the validity of this example
code.
– In Python, we could also do this with Sphinx.
Documentation
lines start with ///
, and using the
Markdown syntax. I find
this very pleasing. I was struggling with
Sphinx before, this
time, I find it easy to understand.
Another great Rust feature is the built in test mechanism,
in this implementation, I’m just using the default structure
that cargo
generated. The command:
$ cargo test
runs both normal tests and also example codes included in the documentations.
Please see the screenshots for the output of the cargo test
command in both Windows 10 and Ubuntu 22.10:
The command:
$ cargo doc --open
generates the document and opens the starting page via the default browser for us. Please see the following screenshots, where the document was displayed in both Windows 10 and Ubuntu 22.10:
The document files are on disk. But this’s as far as I look at it, I imagine that for proper production-ready libraries, we’ll just upload the document files to make the documentation publicly available.
We don’t exactly need main.rs
for the purpose of
this exercise, but I have it just for the shake of demonstration.
Content of /home/behai/rust/maths/src/main.rs:
/*
Demonstrating how to use function(s) in the maths crate.
*/
use maths::fibonacci;
fn main() {
let n: u32 = 9;
let (result, number) = fibonacci(n);
println!("\n{n}th Fibonacci result: {}.", result);
println!("{n}th Fibonacci number: {}.\n", number);
let n: u32 = 11;
let result: (bool, u32) = fibonacci(n);
println!("{n}th Fibonacci number: {:?}.\n", result);
assert_eq!(true, result.0);
assert_eq!(89, result.1);
let n: u32 = 18;
let result: (bool, u32) = fibonacci(n);
println!("{n}th Fibonacci number: {:?}.\n", result);
let n: u32 = 0;
let result: (bool, u32) = fibonacci(n);
println!("{n}th Fibonacci number: {:?}.\n", result);
if result.0 {
println!("{n}th Fibonacci number is successful, number is {}.", result.1);
}
else {
println!("Failed to calculate {n}th Fibonacci number.");
}
}
The fn main()
function is Rust entry-point for
an executable. This function just calls fibonacci(u32)
with different values for u32
and prints out results.
We can run it using the cargo run
command:
behai@hp-pavilion-15:~/rust/maths$ cargo run
behai@hp-pavilion-15:~/rust/maths$ cargo run
Compiling maths v0.1.0 (/home/behai/rust/maths)
Finished dev [unoptimized + debuginfo] target(s) in 10.94s
Running `target/debug/maths`
9th Fibonacci result: true.
9th Fibonacci number: 34.
11th Fibonacci number: (true, 89).
18th Fibonacci number: (true, 2584).
0th Fibonacci number: (false, 0).
Failed to calculate 0th Fibonacci number.
behai@hp-pavilion-15:~/rust/maths$
I hope you find this useful. Thank you for reading and stay safe as always.
✿✿✿
Feature image source: