aes-kw: no_std support + other cleanups
#7
Conversation
- Makes the `Kek::{wrap, unwrap}` functions take an output buffer rather
than allocating a `Vec<u8>`, allowing them to be used in heapless
`no_std` contexts.
- Adds an `alloc` feature along with `Kek::{wrap_vec, unwrap_vec}`
functions which provided the previous API.
- Adds convenience `KekAes128`/`KekAes192`/`KekAes256` type aliases.
- Makes the `IV` and `IV_LEN` constants public.
- Removes the use of `hex` for test vectors replacing it with
`hex-literal` instead.
- Improves README.md and rustdoc documentation.
aes-kw/src/error.rs
Outdated
| /// Input data length invalid. | ||
| InvalidDataLength, | ||
| /// Invalid kek size. | ||
| InvalidKekSize(usize), | ||
| /// Output buffer size invalid. | ||
| InvalidOutputSize { | ||
| /// Expected size in bytes. | ||
| expected: usize, | ||
| }, | ||
| /// Integrity check did not pass. | ||
| IntegrityCheckFailed, |
There was a problem hiding this comment.
Seems like these could be a little more consistent:
*Lengthvs*Size- Tuple variant vs one with named fields
There was a problem hiding this comment.
I'd probably suggest settling on *Size (it's what we tend to use in other crates, and also shorter).
I'd also suggest changing InvalidKekSize { size: usize } for consistency.
There was a problem hiding this comment.
yeah, makes sense to standardize on size
| @@ -8,8 +9,41 @@ | |||
| #![forbid(unsafe_code)] | |||
| #![warn(missing_docs, rust_2018_idioms)] | |||
|
|
|||
| //! # Usage | |||
There was a problem hiding this comment.
Had to move the usage out of README.md in order to be able to feature-gate the doctests, unfortunately.
Otherwise they fail unless the correct features are available.
tarcieri
force-pushed
the
aes-kw/no_std+other-fixups
branch
2 times, most recently
from
ed4ba84
to
4415799
Compare
tarcieri
force-pushed
the
aes-kw/no_std+other-fixups
branch
from
4415799
to
1ff6729
Compare
|
I think this is ready to release then? @dignifiedquire anything else? Should I cut a release or do you want to? |
|
Functionally, looks good. Just upd Deno PR to use latest commit, and is now passing 100% of WPT (WebCrypto/wrapKey_unwrapKey) and JOSE (AxxxKW) tests! Great work, guys. Thanks a lot for your effort and patience ;) |
Kek::{wrap, unwrap}functions take an output buffer rather than allocating aVec<u8>, allowing them to be used in heaplessno_stdcontexts.allocfeature along withKek::{wrap_vec, unwrap_vec}functions which provided the previous API.KekAes128/KekAes192/KekAes256type aliases.IVandIV_LENconstants public.hexfor test vectors replacing it withhex-literalinstead.