Issy Long 45978435e7
rubocop: Use Sorbet/StrictSigil as it's better than comments
- Previously I thought that comments were fine to discourage people from
  wasting their time trying to bump things that used `undef` that Sorbet
  didn't support. But RuboCop is better at this since it'll complain if
  the comments are unnecessary.

- Suggested in https://github.com/Homebrew/brew/pull/18018#issuecomment-2283369501.

- I've gone for a mixture of `rubocop:disable` for the files that can't
  be `typed: strict` (use of undef, required before everything else, etc)
  and `rubocop:todo` for everything else that should be tried to make
  strictly typed. There's no functional difference between the two as
  `rubocop:todo` is `rubocop:disable` with a different name.

- And I entirely disabled the cop for the docs/ directory since
  `typed: strict` isn't going to gain us anything for some Markdown
  linting config files.

- This means that now it's easier to track what needs to be done rather
  than relying on checklists of files in our big Sorbet issue:

```shell
$ git grep 'typed: true # rubocop:todo Sorbet/StrictSigil' | wc -l
    268
```

- And this is confirmed working for new files:

```shell
$ git status
On branch use-rubocop-for-sorbet-strict-sigils
Untracked files:
  (use "git add <file>..." to include in what will be committed)
        Library/Homebrew/bad.rb
        Library/Homebrew/good.rb

nothing added to commit but untracked files present (use "git add" to track)

$ brew style
Offenses:

bad.rb:1:1: C: Sorbet/StrictSigil: Sorbet sigil should be at least strict got true.
^^^^^^^^^^^^^

1340 files inspected, 1 offense detected
```
2024-08-12 15:24:27 +01:00

198 lines
5.1 KiB
Ruby

# typed: true # rubocop:todo Sorbet/StrictSigil
# frozen_string_literal: true
module Cask
class DSL
# Class corresponding to the `version` stanza.
class Version < ::String
DIVIDERS = {
"." => :dots,
"-" => :hyphens,
"_" => :underscores,
}.freeze
DIVIDER_REGEX = /(#{DIVIDERS.keys.map { |v| Regexp.quote(v) }.join("|")})/
MAJOR_MINOR_PATCH_REGEX = /^([^.,:]+)(?:.([^.,:]+)(?:.([^.,:]+))?)?/
INVALID_CHARACTERS = /[^0-9a-zA-Z.,:\-_+ ]/
class << self
private
def define_divider_methods(divider)
define_divider_deletion_method(divider)
define_divider_conversion_methods(divider)
end
def define_divider_deletion_method(divider)
method_name = deletion_method_name(divider)
define_method(method_name) do
T.bind(self, Version)
version { delete(divider) }
end
end
def deletion_method_name(divider)
"no_#{DIVIDERS[divider]}"
end
def define_divider_conversion_methods(left_divider)
(DIVIDERS.keys - [left_divider]).each do |right_divider|
define_divider_conversion_method(left_divider, right_divider)
end
end
def define_divider_conversion_method(left_divider, right_divider)
method_name = conversion_method_name(left_divider, right_divider)
define_method(method_name) do
T.bind(self, Version)
version { gsub(left_divider, right_divider) }
end
end
def conversion_method_name(left_divider, right_divider)
"#{DIVIDERS[left_divider]}_to_#{DIVIDERS[right_divider]}"
end
end
DIVIDERS.each_key do |divider|
define_divider_methods(divider)
end
attr_reader :raw_version
sig { params(raw_version: T.nilable(T.any(String, Symbol))).void }
def initialize(raw_version)
@raw_version = raw_version
super(raw_version.to_s)
invalid = invalid_characters
raise TypeError, "#{raw_version} contains invalid characters: #{invalid.uniq.join}!" if invalid.present?
end
def invalid_characters
return [] if raw_version.blank? || latest?
raw_version.scan(INVALID_CHARACTERS)
end
sig { returns(T::Boolean) }
def unstable?
return false if latest?
s = downcase.delete(".").gsub(/[^a-z\d]+/, "-")
return true if s.match?(/(\d+|\b)(alpha|beta|preview|rc|dev|canary|snapshot)(\d+|\b)/i)
return true if s.match?(/\A[a-z\d]+(-\d+)*-?(a|b|pre)(\d+|\b)/i)
false
end
sig { returns(T::Boolean) }
def latest?
to_s == "latest"
end
# The major version.
#
# @api public
sig { returns(T.self_type) }
def major
version { slice(MAJOR_MINOR_PATCH_REGEX, 1) }
end
# The minor version.
#
# @api public
sig { returns(T.self_type) }
def minor
version { slice(MAJOR_MINOR_PATCH_REGEX, 2) }
end
# The patch version.
#
# @api public
sig { returns(T.self_type) }
def patch
version { slice(MAJOR_MINOR_PATCH_REGEX, 3) }
end
# The major and minor version.
#
# @api public
sig { returns(T.self_type) }
def major_minor
version { [major, minor].reject(&:empty?).join(".") }
end
# The major, minor and patch version.
#
# @api public
sig { returns(T.self_type) }
def major_minor_patch
version { [major, minor, patch].reject(&:empty?).join(".") }
end
# The minor and patch version.
#
# @api public
sig { returns(T.self_type) }
def minor_patch
version { [minor, patch].reject(&:empty?).join(".") }
end
# The comma separated values of the version as array.
#
# @api public
sig { returns(T::Array[Version]) } # Only top-level T.self_type is supported https://sorbet.org/docs/self-type
def csv
split(",").map { self.class.new(_1) }
end
# The version part before the first comma.
#
# @api public
sig { returns(T.self_type) }
def before_comma
version { split(",", 2).first }
end
# The version part after the first comma.
#
# @api public
sig { returns(T.self_type) }
def after_comma
version { split(",", 2).second }
end
# The version without any dividers.
#
# @see DIVIDER_REGEX
# @api public
sig { returns(T.self_type) }
def no_dividers
version { gsub(DIVIDER_REGEX, "") }
end
# The version with the given record separator removed from the end.
#
# @see String#chomp
# @api public
sig { params(separator: String).returns(T.self_type) }
def chomp(separator = T.unsafe(nil))
version { to_s.chomp(separator) }
end
private
sig { returns(T.self_type) }
def version
return self if empty? || latest?
self.class.new(yield)
end
end
end
end