[nemynm]

Adding support for ANSI X9.63 Key Derivation Function (ANSI-X9.63-KDF).

ANSI-X9.63-KDF is defined in SEC.1 and is the recommended KDF to be used for CMS with ECC - as per RFC 5753.

This would close issue #101.

Overview:
This pull request:

• adds ansi-x963-kdf crate with:
  • ansi_x963_kdf::derive_key_into function
  • tests for SHA254, SHA256 and SHA512 hash functions
• updates repository README.md with ansi-x963-kdf

Notes:

• As concat-kdf and ANSI-X9.63-KDF are close and essentially differ in the order in which the data is concatenated, most of the code for this ansi-x963-kdf implementation has been adapted from concat-kdf.
• I did not find public test vectors, so the test cases have been generated using a Java-based implementation of ANSI-X9.63-KDF
• I added one "unsafe" test case to test the input overflow case. As I am not an "unsafe" specialist, carefull review of this one is particularly appreciated.

[newpavlov]

I have several comments, but otherwise looks good!

Please also add a CI job to test this crate (see the .github/workflows folder).

[newpavlov]

You can update the Clippy version to 1.81 in the workspace.yml config to fix the failing CI.

[nemynm]

I have several comments, but otherwise looks good!

Please also add a CI job to test this crate (see the .github/workflows folder).

Thanks for reviewing! I tried and addressed your comments. The MSRV bump does trigger a failing clippy at workspace level though.

[newpavlov]

Thank you!

[nemynm]

Thanks for the review and the tweaks!

• you probably have noted it, but just in case, on README.md, it seems that the badge MSRV 1.81 and crate on crates.io are not-rendering/missing.
• I think some of these changes could be applied to concat-kdf (e.g. doctests, MSRV, std vs alloc)

In the meantime, I'll close the issue.

[newpavlov]

the badge MSRV 1.81 and crate on crates.io are not-rendering/missing.

The MSRV badge (rustc 1.81+) renders fine for me. Docs and crates.io badges require publishing to crates.io, which we will do after releasing digest v0.11.

I think some of these changes could be applied to concat-kdf (e.g. doctests, MSRV, std vs alloc)

Yes. I plan to do it a bit later.