Name iterator

[djc]

This makes the abstractions a lot more intuitive/idiomatic in my opinion.

I think it has the potential to bring list_cert_dns_names() out of the alloc domain (see #46), but that would be at the cost of making it yield an impl Iterator<Item = Result<GeneralName<'_>, Error>> rather than a Result<impl Iterator<Item = GeneralName<'_>>, Error> which might be undesirable?

[codecov]

Codecov Report

Merging #140 (9550d23) into main (205f5d7) will decrease coverage by 0.28%.
The diff coverage is 89.11%.

@@            Coverage Diff             @@
##             main     #140      +/-   ##
==========================================
- Coverage   96.28%   96.01%   -0.28%     
==========================================
  Files          16       16              
  Lines        4070     4061       -9     
==========================================
- Hits         3919     3899      -20     
- Misses        151      162      +11     

Files Changed	Coverage Δ
src/subject_name/verify.rs	92.09% <89.11%> (-4.48%)	⬇️

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

[djc]

One downside of the stricter iterator chaining is that each of the NameIterator users have to deal with error values separately (rather than iterate_names() propagating them in one place). However, since the only thing they have to do is yield them to the find_map() iterator, that doesn't seem like a big issue.

[ctz]

feels a bit less readable because we have this continue in a two-deep loop (though the comment gives a hint of which loop it applies to if the reader missed the inner loop), and the loop bodies are quite long now - with load-bearing code that has to run after the inner loop completes. perhaps hoist fn general_subtree out of the loop body? or drop the previous DirectoryName commit?

[ctz]

not a hard objection either way, it's a bit subjective.

[djc]

I hoisted general_subtree() out of the loop body, hopefully that helps a bit.

I think there's more that can be done here, but probably still makes sense to merge this first.

[cpu]

This looks good to me 👍 Thanks djc