Feature/internal math

[TonalidadeHidrica]

Since there seems to be no problem that directly requires this part of library in the practice contest, so I will submit it as it is, without writing examples. If I should add some tests, I will do it, so please let me know.

I tried to translate the code so that the original C++ code correspond to this Rust code as much as possible. I'm not sure how to write some of them in a "Rust-ish" way. Any suggestion is welcome.

(BTW I'm writing this issue comment in English because the original library is intended to be used internationally, but of course you can just comment in Japanese.)

[TonalidadeHidrica]

This pull request should be merged after #1 and #2 are merged.

[qryxip]

I have a question. How do you intend to expand internal_math.rs? Hand-by-hand copy-pasting?

As you might already know, namespaces in C++ differ to mods in Rust.

mod foo {}
mod foo {}

$ cargo clippy
    Checking a v0.1.0 (/home/ryo/src/local/hage/a)
error[E0428]: the name `foo` is defined multiple times
 --> src/main.rs:4:1
  |
3 | mod foo {}
  | ------- previous definition of the module `foo` here
4 | mod foo {}
  | ^^^^^^^ `foo` redefined here
  |
  = note: `foo` must be defined only once in the type namespace of this module

error: aborting due to previous error

For more information about this error, try `rustc --explain E0428`.
error: could not compile `a`.

To learn more, run the command again with --verbose.

My approval was a mistake.

[TonalidadeHidrica]

@qryxip I realized that I shouldn't wrap with namespace, so I will fix it.

[TonalidadeHidrica]

I'm worried that there may be some other points that the original C++ code relies on its C++ behavior...

[qryxip]

BTW, how do we suppress clippy::many_single_char_names? Adding #![allow] to each module, or the crate root?

#![allow(clippy::many_single_char_names)]

[TonalidadeHidrica]

@qryxip IMHO it is inevitable to use single-character variables in competitive programming, so it may be good idea to suppress module-wide.

[kenkoooo]

It's no longer constexpr.

[TonalidadeHidrica]

Do you think it's better change the name accordingly? (Yes, it's indeed confusing, I agree)

[kenkoooo]

Let's implement the same functionality, not a pure copy.

[TonalidadeHidrica]

By the way, why do they add underscores to some fields in structs (as in _m in this PR)? I guess it is to avoid name collision. However in Rust we do not have to worry about it, so it maybe better to replace with names without underscores.

[koba-e964]

Do I have to handle with cargo clippy?

I'd rather you did.
Adding #[allow(clippy::many_single_char_names)] to lib.rs will suffice.

[TonalidadeHidrica]

@koba-e964 I tried adding it at the first line in lib.rs, but the warnings did not disappear. When I added it to each function, it disappeared.

[TonalidadeHidrica]

By the way what should the visibility of each items be? I'm not familiar with it so much so please somebody tell me.

[TonalidadeHidrica]

I've added some tests and think that all the code is working properly.

[qryxip]

By the way what should the visibility of each items be? I'm not familiar with it so much so please somebody tell me.

pub(crate), I think.

Visibility and Privacy - The Rust Reference

[koba-e964]

@koba-e964 I tried adding it at the first line in lib.rs, but the warnings did not disappear. When I added it to each function, it disappeared.

I mean not the first line, but the line containing pub(crate) mod internal_math;.
But adding the annotation to each function is also fine by me.

[koba-e964]

Once you resolve all clippy warnings, LGTM.

[TonalidadeHidrica]

How can I get rid of "function/struct/method is not used" warnings?

[qryxip]

I've forgotten dead_code warnings. Why not make the visibilities pub instead of pub(crate) as a workaround.

[qryxip]

For clippy::unreadable_literal, Maybe we should add #![allow(〃)] at the beginning of mod tests.

[qryxip]

I've forgotten dead_code warnings. Why not make the visibilities pub instead of pub(crate) as a workaround.

Or adding #![allow(dead_code)] to the beginning of internal_math.rs.

[TonalidadeHidrica]

Adding #![allow(dead_code)] seemed reasonable for me so I did so.

[TonalidadeHidrica]

test_pow_mod is pointed out to have a large cognitive complexity. Should I fix it (such as splitting into multiple test functions or use for loop instead) or simply ignore the warning (by adding #![allow(...)])?

[qryxip]

test_pow_mod is pointed out to have a large cognitive complexity.

Reffering to clippy::cognitive_complexity? But it does not seem to be triggered on 1.42.0. If so, We may #[allow] it, as it is allowed by default on 1.46.0.

[TonalidadeHidrica]

Unfortunately it is triggered ( https://github.com/rust-lang-ja/ac-library-rs/pull/3/checks?check_run_id=1094713411#step:5:8 ). I'll simply allow it.

[qryxip]

The let's merge this PR.