tap: allow sharding of fonts in Homebrew/cask

[p-linnane]

• Have you followed the guidelines in our Contributing document?
• Have you checked to ensure there aren't other open Pull Requests for the same change?
• Have you added an explanation of what your changes do and why you'd like us to include them?
• Have you written new tests for your changes? Here's an example.
• Have you successfully run brew style with your changes locally?
• Have you successfully run brew typecheck with your changes locally?
• Have you successfully run brew tests with your changes locally?

---

We want to migrate Homebrew/cask-fonts into the main Homebrew/cask repo. There are ~2,200 font casks that need to be put into sharded subdirectories. This PR allows for paths such as Casks/fonts/font-a/font-apple.rb.

[krehel]

Throwing a comment here to poll the group. I made similar changes that is more code, but don't need to rename Casks. Not sure I like it though.

Effectively a new def in tap.rb for font paths specifc rather than using the existing new_cask_path def. This can be shortened for brevity.

  def new_font_path(token)
    font_first_letter = token.split("-").second[0].to_s
    cask_subdir = "fonts/font-#{token.split("-").second[0].to_s}"
    cask_dir/cask_subdir/"#{token.downcase}.rb"
  end

Then in cask/audit.rb, updated the path audit to check explicitly for font artifacts:

 def audit_cask_path
  return unless cask.tap.core_cask_tap?

  expected_path = if cask.artifacts.none? { |k| k.is_a?(Artifact::Font) }
                    cask.tap.new_cask_path(cask.token)
                  else
                    cask.tap.new_font_path(cask.token)
                  end

  return if cask.sourcefile_path.to_s.end_with?(expected_path)

  add_error "Cask should be located in '#{expected_path}'"
end

If there's a better way I could've shared these changes please let me know! I didn't want to open a competing PR and I didn't want to clobber the existing PR.

[p-linnane]

@krehel Can you benchmark your proposal? Want to see how it affects performance.

[krehel]

I did locally with hyperfine. I saw no difference.

Edit, fwiw:

My changes:

> hyperfine --warmup 5 "brew audit --only cask_path visual-studio-code"
Benchmark 1: brew audit --only cask_path visual-studio-code
  Time (mean ± σ):      2.535 s ±  0.016 s    [User: 1.896 s, System: 0.515 s]
  Range (min … max):    2.518 s …  2.570 s    10 runs


> hyperfine --warmup 5 "brew audit --only cask_path font-3270"
Benchmark 1: brew audit --only cask_path font-3270
  Time (mean ± σ):      2.562 s ±  0.031 s    [User: 1.915 s, System: 0.520 s]
  Range (min … max):    2.538 s …  2.644 s    10 runs

This PR:

> hyperfine --warmup 5 "brew audit --only cask_path visual-studio-code"
Benchmark 1: brew audit --only cask_path visual-studio-code
  Time (mean ± σ):      2.564 s ±  0.031 s    [User: 1.912 s, System: 0.519 s]
  Range (min … max):    2.531 s …  2.611 s    10 runs

> hyperfine --warmup 5 "brew audit --only cask_path font-3270"
Benchmark 1: brew audit --only cask_path font-3270
  Time (mean ± σ):      2.585 s ±  0.075 s    [User: 1.932 s, System: 0.518 s]
  Range (min … max):    2.506 s …  2.748 s    10 runs

[p-linnane]

@carlocab @MikeMcQuaid If you can weigh in on the above it would be appreciated. I want to make sure we're using the right approach here.

[krehel]

Indeed. Feel free to tell me it's terrible or not 😄 (also: I am genuinely curious which is preferred)

[reitermarkus]

As I said in Homebrew/homebrew-cask#173366 (comment), I'm against disallowing normal casks to start with font-, so I think basing this on the actual artifacts rather than the cask name, as @krehel suggested, is the better approach.

[p-linnane]

Sounds good. @krehel I'm going to close this and the cask PR. Can you open a PR with your proposed change? Thanks!

[krehel]

@p-linnane will do. Thanks