project: apply cargo fmt universally, verify in CI.

[cpu]

Self-explanatory. Applies cargo fmt universally, adds a CI job to ensure consistent formatting going forward. No functional changes.

[djc]

To be explicit, here's why I think this is a good idea:

1. The default cargo fmt settings represent idiomatic style, that is used across most Rust projects
2. Automatic formatting tooling makes life easier for contributors
3. Style enforcement in CI avoids (some) additional review roundtrips due to stylistic concerns

[cpu]

To be explicit, here's why I think this is a good idea

Thanks for adding motivating context. I should have done that in my PR description 👍 Your points echo the feelings I expressed in this comment.

[codecov]

Codecov Report

Merging #146 (772795e) into main (390e3ad) will decrease coverage by 1.32%.
The diff coverage is 66.36%.

@@            Coverage Diff             @@
##             main     #146      +/-   ##
==========================================
- Coverage   71.60%   70.28%   -1.32%     
==========================================
  Files           7        7              
  Lines        1729     1898     +169     
==========================================
+ Hits         1238     1334      +96     
- Misses        491      564      +73     

Files Changed	Coverage Δ
src/error.rs	1.92% <0.00%> (-0.86%)	⬇️
src/main.rs	0.00% <0.00%> (ø)
src/sign_algo.rs	49.03% <61.53%> (-0.48%)	⬇️
src/csr.rs	72.54% <62.50%> (+0.80%)	⬆️
src/lib.rs	72.99% <71.08%> (-0.72%)	⬇️
src/key_pair.rs	69.23% <75.51%> (-0.25%)	⬇️
src/crl.rs	93.13% <84.00%> (-1.95%)	⬇️

📣 We’re building smart automated test selection to slash your CI/CD build times. Learn more

[est31]

Btw, would it be possible to remove these coverage comments? they are put on every PR, even ones that don't touch anything related to tests. I think it's cool to have coverage reports you can browse online and look at graphs, but I don't think one needs to have a comment for each PR on how it changes the doc coverage.

As for autoformatters, I think there are some options that make cargo fmt nicer, I will make some experiments and will then propose something.

[cpu]

Btw, would it be possible to remove these coverage comments?

I'll look into that.

As for autoformatters, I think there are some options that make cargo fmt nicer, I will make some experiments and will then propose something.

@est31 Does this PR need to wait for that adjustment? The benefit of machine formatting is that if you change the rules of the formatter, you can run it again.

[cpu]

Btw, would it be possible to remove these coverage comments?

I'll look into that.

Let's try this first: #149

[djc]

As for autoformatters, I think there are some options that make cargo fmt nicer, I will make some experiments and will then propose something.

@est31 Does this PR need to wait for that adjustment? The benefit of machine formatting is that if you change the rules of the formatter, you can run it again.

@est31 friendly ping on this? Would be nice for further changes to get the formatting done first.

[est31]

It's a pity that overflow_delimited_expr isn't stable. Idk about you but I like former way more:

-		assert_eq!(uris, &[
-			"http://example.com/crl.der",
-			"http://crls.example.com/1234",
-			"ldap://example.com/crl.der"
-		]);
+		assert_eq!(
+			uris,
+			&[
+				"http://example.com/crl.der",
+				"http://crls.example.com/1234",
+				"ldap://example.com/crl.der"
+			]
+		);

[iamjpotts]

It's a pity that overflow_delimited_expr isn't stable. Idk about you but I like former way more:

-		assert_eq!(uris, &[
-			"http://example.com/crl.der",
-			"http://crls.example.com/1234",
-			"ldap://example.com/crl.der"
-		]);
+		assert_eq!(
+			uris,
+			&[
+				"http://example.com/crl.der",
+				"http://crls.example.com/1234",
+				"ldap://example.com/crl.der"
+			]
+		);

I like the latter way, but I'm also glad that rustfmt isn't black

[cpu]

Thanks 👍