Upgrade issues from sha-1 0.9.8 to sha1 0.10.x

[burkematthew]

Hi there!

First, thanks for all the hard work that has gone into building and maintaining this crate.

Secondly, we're running into some issues when upgrading from sha-1 0.9.8 to sha1 0.10.x. I've reviewed the PRs associated with the release of 0.10.x and looked through the docs, but nothing seems to indicate a recommended path to address.

Here is a snippet of our previously working source:

use base64::encode;
use hmacsha::HmacSha;
use sha1::Sha1;
use tracing::{debug, warn};

async fn valid_signature(signature: &str, uri: &str) -> bool {
  let signing_key = std::env::var("TOKEN").unwrap_or_else(|_| "".to_string());
  if signing_key.is_empty() {
    warn!(target: crate::EVENT_TARGET_LOG_STREAM, "No signing key found.  Failing validation.");
    return false;
  }

  let mut hasher = HmacSha::new(signing_key.as_bytes(), uri.as_bytes(), Sha1::default());
  let buf = hasher.compute_digest();
  let result = buf.as_slice();
  let encoded_digest = encode(result);

  debug!(target: crate::EVENT_TARGET_LOG_STREAM, "Encoded Digest: {}", &encoded_digest);
  encoded_digest == signature
}

However, when upgrading to 0.10.x, the following line is failing because trait bounds are not satisfied:

  let mut hasher = HmacSha::new(signing_key.as_bytes(), uri.as_bytes(), Sha1::default());

In particular, some of the trait bounds that are not satisfied are:

digest::Update
digest::FixedOutput
digest::Reset

This seems like some breaking changes to me, but nothing to assist in how to remediate these breaks. Anything that can be offered to assist would be greatly appreciated!

[tarcieri]

Whatever hmacsha is, it looks like it needs to be updated to digest v0.10 as well

[newpavlov]

Also, if hmacsha uses the hmac crate under the hood, then it also has to be updated to v0.12.