FSB hash function#256
Conversation
| .idea | ||
| .DS_Store |
There was a problem hiding this comment.
Speaking as a fellow IntelliJ user...
I'd suggest adding the following to your personal ~/.gitconfig instead:
[core]
excludesfile = /home/myuser/.gitignore
...and then putting these kinds of directives in there.
That avoids every other project having to add .gitignore directives for every editor.
| @@ -0,0 +1,16 @@ | |||
| [package] | |||
| name = "fsb_rust" | |||
There was a problem hiding this comment.
RustCrypto owns the fsb name. You can use that.
@newpavlov can you add rustcrypto:hashes as an owner?
|
Looks like a promising start! @iquerejeta can you rebase? You'll probably want to |
| #[allow(dead_code)] | ||
| #[macro_use] | ||
| mod macros; | ||
| mod pi; |
There was a problem hiding this comment.
GitHub won't even let me leave comments on that module, but... wow 😮
There was a problem hiding this comment.
Not my work! Copied that file from the reference implementation
|
@iquerejeta it looks like you need to bump clippy MSRV in Also can copy/modify a README.md from one of the other crates? (e.g. |
|
I've rebased, hopefully correctly 😄
qq, should I squash all commits to have a cleaner history? |
|
We use GitHub's "Squash and merge" feature so you don't need to worry about squashing it yourself |
|
Bumped version. I got a security audit failing in the repo, but can't really tell what the problems are. Also I got a bunch of PRs from the bot, requesting to bump dependencies on other crates (e.g. sha1 or sha2). Should I go ahead and merge all these PRs? |
|
Tests appear to be passing now. |
newpavlov
left a comment
newpavlov
left a comment
There was a problem hiding this comment.
I haven't checked the implementation itself, so I have only several surface-level comments for now.
| - uses: actions-rs/toolchain@v1 | ||
| with: | ||
| toolchain: 1.41.0 # MSRV | ||
| toolchain: 1.47.0 # MSRV |
There was a problem hiding this comment.
I think we can remove the "MSRV" comment from here, since it's not about MSRV, but about pinning clippy version.
|
|
||
| [[package]] | ||
| name = "fsb_rust" | ||
| version = "0.1.0" |
There was a problem hiding this comment.
Since you've already changed the crate name, re-generate the lock file, so it will use fsb instead of fsb_rust? I have shared ownership of the fsb crate with the hashes group, so we can publish it immediately after merge.
| categories = ["cryptography", "no-std"] | ||
|
|
||
| [dependencies] | ||
| whirlpool = "0.9" |
There was a problem hiding this comment.
I suggest using whirlpool = { version = "0.9", path = "../whirlpool", default-features = false } here.
| [![crate][crate-image]][crate-link] | ||
| [![Docs][docs-image]][docs-link] | ||
| [![Build Status][build-image]][build-link] | ||
| --> |
There was a problem hiding this comment.
Don't forget to uncomment these lines?
| @@ -0,0 +1,21 @@ | |||
| #![allow(non_snake_case)] | |||
| #[allow(dead_code)] | |||
| #[macro_use] | |||
There was a problem hiding this comment.
Is it truly necessary to allow dead code? It also would be nice to add crate-level docs, no_std and other attributes (see other crates for reference).
There was a problem hiding this comment.
added no_std, and I depend on alloc now for dynamically sized arrays. We use them here.
| assert_eq!( | ||
| result[..], | ||
| hex!("6e8ce7998e4c46a4ca7c5e8f6498a5778140d14b")[..] | ||
| ); |
There was a problem hiding this comment.
You can simply use the following code here:
assert_eq!(
Fsb160::digest(msg_1)[..],
hex!("6e8ce7998e4c46a4ca7c5e8f6498a5778140d14b")[..],
);used for dynamically sized arrays
|
@iquerejeta thank you! |
|
Pleasure! I'm happy to change the tests using the |
3 participants
As an exercise, I've implemented the FSB hash function taking as a reference the paper, so I might have missed out on some efficiency tricks in the reference implementation.
If you'd be happy to have this code in the repo, I'm happy to maintain it and make the necessary changes to take it to the repo quality standards!