cleanup: use better-maintained sha-1 crate instead of sha1 crate #495
Conversation
durin42
requested review from
Dylan-DPC-zz,
kinggoesgaming and
KodrAus
as code owners
|
sha1 has a higher MSRV than UUID currently does. |
|
|
|
It looks like the original concern came from the weight of those dependencies? I had a look through them and they didn't seem too weighty to me. I'd be in favour of also bundling |
|
It was mostly I noticed the two implementations of SHA1 as I was importing crates into a monorepo and figured I should try and consolidate down onto the maintained-looking one. If you're raising your MSRV I'm happy to handle both sha1 and md5, but if you want to keep it low please feel encouraged to reject this PR. Thanks! |
|
We've actually already raised it, so please feel free to roll in both! |
|
@durin42 I just pushed a quick merge commit to your branch. Hope you don't mind! |
|
Not at all! I'll try and get the md5 upgrade done too over the weekend. Right now my keyboard time is somewhat limited. :) |
This removes a "duplicate" dependency in the tree of pyoxidizer.
I feel like it should be easier to go from the GenericArray to the [u8;16] here, but I can't figure out what it should look like.
|
I've rebased and added a commit for upgrading md5 to md-5, but I got stuck figuring out how to convert the GenericArray into a [u8;16] - please let me know if there's a trick I'm missing! |
|
I'm not sure what's up with the |
| optional = true | ||
| version = "0.7" | ||
| version = "0.9.1" |
There was a problem hiding this comment.
It could be worth to start with v0.8 versions of md-5 and sha-1 to avoid bumping MSRV. Updating versions can be done later in a separate PR.
There was a problem hiding this comment.
@newpavlov We've actually already bumped the MSRV so there's no reason not to track the latest versions here.
|
|
||
| let mut builder = crate::Builder::from_bytes(bytes); | ||
| let mut builder = crate::Builder::from_slice(&computed[..16]).unwrap(); |
There was a problem hiding this comment.
IIUC you don't need the slicing here, i.e. you can replace &computed[..16] with &computed. Same goes for SHA-1.
| @@ -20,15 +20,14 @@ impl Uuid { | |||
| /// [`NAMESPACE_URL`]: #associatedconstant.NAMESPACE_URL | |||
| /// [`NAMESPACE_X500`]: #associatedconstant.NAMESPACE_X500 | |||
| pub fn new_v3(namespace: &Uuid, name: &[u8]) -> Uuid { | |||
| let mut context = md5::Context::new(); | |||
| let mut hasher = md5::Md5::new(); | |||
There was a problem hiding this comment.
Use use md5::{Md5, Digest}; import to avoid duplication of md5 here? Same goes for SHA-1.
|
It looks like it's trying to read configuration files for dependencies and is failing on one of them: We can probably just remove the |
|
I'm not familiar with |
This removes a "duplicate" dependency in the tree of pyoxidizer.