Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

impl (From<u32>, Hash, Ord, Product) for Scalar #52

Open
wants to merge 3 commits into
base: main
Choose a base branch
from

Conversation

A-Manning
Copy link
Contributor

No description provided.

@A-Manning A-Manning changed the title impl (From<u32>, Hash, Ord) for Scalar impl (From<u32>, Hash, Ord, Product) for Scalar Jan 7, 2021
Copy link
Member

@str4d str4d left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Thanks for the PR!

src/scalar.rs Outdated Show resolved Hide resolved
src/scalar.rs Show resolved Hide resolved
Comment on lines +40 to +45
impl From<u32> for Scalar {
fn from(val: u32) -> Scalar {
Scalar([val as u64, 0, 0, 0]) * R2
}
}

Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

As is visible from other changes in the PR, this decreases usability by requiring the caller to specify the source type (e.g. 5u64.into() instead of 5.into()). I suspect this would be more usable if impl From<u64> for Scalar was replaced with impl From<T: Into<u64>> for Scalar (and it would then also work for any u* type).

Comment on lines +68 to +82
impl Ord for Scalar {
fn cmp(&self, other: &Self) -> core::cmp::Ordering {
let mut self_bytes = self.to_bytes();
let mut other_bytes = other.to_bytes();
&self_bytes.reverse();
&other_bytes.reverse();
self_bytes.cmp(&other_bytes)
}
}

impl PartialOrd for Scalar {
fn partial_cmp(&self, other: &Self) -> Option<core::cmp::Ordering> {
Some(self.cmp(other))
}
}
Copy link
Contributor

@ebfull ebfull Sep 1, 2021

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Cryptographic structures, in particular finite fields, do not have a total ordering (in the "well acktually" mathematical sense) so I disagree with exposing Ord/PartialOrd when they are in the form of Scalar since they're expected to behave like finite fields. However, obviously we can totally order them when they're represented in serialized form (as this implementation leverages) assuming they're serialized canonically.

If you need Ord for things like a BTreeMap (for example) then the end user should convert it into its serialized form and depend on the total ordering of that instead.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

4 participants