tests: Use full constant names sometimes (for Sorbet rule 5001 compat)

[issyl0]

---

• 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? Performance claims (e.g. "this is faster") must include Hyperfine benchmarks.
• Have you written new tests (excluding integration tests) for your changes? Here's an example.
• Have you successfully run brew lgtm (style, typechecking and tests) with your changes locally?

---

• AI was used to generate or assist with generating this PR.

I got OpenAI Codex 5.3 Medium to bump all the specs to typed: true (because Spoom wouldn't do it automatically), then run brew tc for only code 5001 errors. There were 120 of them.

---

• After Move away from RSpec's described_class #22307, I looked into what was needed to move the rest of the tests to Sorbet typed: true or higher.
• Running spoom in dry-run mode found that several files only had one error: using the new klass or subject variable as a dynamic constant, instead of the full constant name (that is, https://sorbet.org/docs/error-reference#5001).
• For each of the 120 errors upon bumping files, I replaced klass or subject with the full constant name. Now 30 more tests are typechecked to true, and we have less far to go on the others.

[Copilot]

Pull request overview

This PR reduces Sorbet error 5001 (“dynamic constant reference”) in the test suite to enable more specs to be bumped to # typed: true and successfully typechecked. It follows up on the earlier migration away from described_class by replacing klass::CONST / subject::CONST usages with fully qualified constant names where needed.

Changes:

• Bump a set of spec files from # typed: false to # typed: true.
• Replace dynamic constant references in specs with fully qualified constants (e.g., Homebrew::...::CONST, MacOSVersion::NULL, etc.) to satisfy Sorbet 5001.
• Adjust a small number of expectations/instantiations to avoid dynamic constant usage in test code.

Show a summary per file

File	Description
Library/Homebrew/test/version_spec.rb	Use Version::NULL_TOKEN instead of a dynamic klass::NULL_TOKEN reference.
Library/Homebrew/test/os/mac/sdk_spec.rb	Avoid dynamic exception constant reference when asserting SDK lookup failures.
Library/Homebrew/test/os/mac/reinstall_spec.rb	Bump to typed: true and use fully qualified Homebrew::Reinstall::InstallationContext.
Library/Homebrew/test/os/linux/libstdcxx_spec.rb	Use fully qualified OS::Linux::Libstdcxx constants for SONAME/SOVERSION.
Library/Homebrew/test/macos_version_spec.rb	Replace klass::... constant references with MacOSVersion::... constants.
Library/Homebrew/test/livecheck/strategy/launchpad_spec.rb	Use fully qualified Homebrew::Livecheck::Strategy::Launchpad::DEFAULT_REGEX.
Library/Homebrew/test/livecheck/strategy/github_releases_spec.rb	Bump to typed: true and use fully qualified ...GithubReleases::DEFAULT_REGEX.
Library/Homebrew/test/livecheck/strategy/extract_plist_spec.rb	Use fully qualified ExtractPlist::Item constant instead of dynamic constant access.
Library/Homebrew/test/livecheck/strategy/crate_spec.rb	Bump to typed: true and use fully qualified Crate::DEFAULT_REGEX.
Library/Homebrew/test/livecheck/strategy_spec.rb	Bump to typed: true and use fully qualified Strategy::INVALID_BLOCK_RETURN_VALUE_MSG.
Library/Homebrew/test/keg_relocate/text_spec.rb	Use Keg::Relocation instead of klass::Relocation.
Library/Homebrew/test/keg_relocate/grep_spec.rb	Bump to typed: true.
Library/Homebrew/test/install_steps_spec.rb	Use fully qualified Homebrew::InstallSteps::{DSL,Runner} rather than described_class::....
Library/Homebrew/test/github_runner_matrix_spec.rb	Use GitHubRunnerMatrix::... constants directly to avoid dynamic constant access.
Library/Homebrew/test/formulary_spec.rb	Replace dynamic loader class references with Formulary::From*Loader constants.
Library/Homebrew/test/env_config_spec.rb	Bump to typed: true and use Homebrew::EnvConfig::ENVS.
Library/Homebrew/test/dependency_expansion_spec.rb	Use Dependable::{PRUNE,SKIP,KEEP_BUT_PRUNE_RECURSIVE_DEPS} constants directly.
Library/Homebrew/test/completions_spec.rb	Bump to typed: true and use Homebrew::Completions::SHELLS.
Library/Homebrew/test/cmd/which-formula_spec.rb	Use fully qualified Homebrew::Cmd::WhichFormula::{ENDPOINT,DATABASE_FILE} constants.
Library/Homebrew/test/cmd/upgrade_spec.rb	Use fully qualified nested constants under Homebrew::Cmd::UpgradeCmd.
Library/Homebrew/test/cmd/services/restart_subcommand_spec.rb	Bump to typed: true and use fully qualified RestartSubcommand::TRIGGERS.
Library/Homebrew/test/cmd/services/list_subcommand_spec.rb	Bump to typed: true and use fully qualified ListSubcommand::{TRIGGERS,JSON_FIELDS}.
Library/Homebrew/test/cmd/services/info_subcommand_spec.rb	Bump to typed: true and use fully qualified InfoSubcommand::TRIGGERS.
Library/Homebrew/test/cmd/services/cleanup_subcommand_spec.rb	Bump to typed: true and use fully qualified CleanupSubcommand::TRIGGERS.
Library/Homebrew/test/cmd/bundle_spec.rb	Use fully qualified Homebrew::Cmd::Bundle::BUNDLE_EXTENSIONS.
Library/Homebrew/test/cask/list_spec.rb	Use Cask::List::TAP_AND_NAME_COMPARISON directly instead of klass::....
Library/Homebrew/test/bundle/winget_spec.rb	Use fully qualified Homebrew::Bundle::Winget::App constants.
Library/Homebrew/test/bundle/mac_app_store_spec.rb	Use fully qualified Homebrew::Bundle::MacAppStore::App constants.
Library/Homebrew/test/bundle/brew_spec.rb	Use fully qualified Homebrew::Bundle::Brew::Topo constant for mocking.
Library/Homebrew/test/attestation_spec.rb	Use fully qualified Homebrew::Attestation constants and error classes.
Library/Homebrew/test/api/cask_struct_spec.rb	Bump to typed: true and use fully qualified Homebrew::API::CaskStruct constants.

Copilot's findings

Tip

Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

• Files reviewed: 31/31 changed files
• Comments generated: 1

[MikeMcQuaid]

Thanks! I do wonder if we're doing this if it's worth just doing it everywhere instead of klass at all 🤷🏻

[issyl0]

Thanks! I do wonder if we're doing this if it's worth just doing it everywhere instead of klass at all 🤷🏻

I'll gently and with a smile remind you of #22307 (review). We could do, though.

[MikeMcQuaid]

@issyl0 To be fair, I did say "if we can use klass everywhere 😄. I think if we have to be inconsistent it's probably not worth it but don't feel strongly! 🤷🏻