Begin refactoring classes used by add-grammar

script/add-grammar

@@ -3,27 +3,102 @@
 require "optparse"
 require "open3"
 
-ROOT = File.expand_path("../../", __FILE__)
+class GrammarRepo
+
+  # Whitelist of trusted hosting providers
+  HOSTS = Regexp.union %w[github.com bitbucket.org gitlab.com]
+
+  # Public: Define a repository source by upstream URL.
+  #
+  # url - an HTTPS, HTTP, or SSH address accepted by git-remote(1)
+  #       Only domains listed in HOSTS are accepted; unrecognised
+  #       hostnames or invalid URLs will raise an ArgumentError.
+  #
+  #       Assumption: Repo URLs will never include subdomains.
+  #       We only check for a possible `www`, nothing else.
+  #
+  # module_path - path of submodule as registered in `.gitmodules`
+  #               Omit this unless grammar is being replaced.
+  def initialize(url, module_path = nil)
+    if https? url
+      @host = $1.downcase
+      @user = $2
+      @repo = $3.sub /\.git$/, ""
+    elsif ssh?(url) || shorthand?(url)
+      @host = $1.downcase
+      @user = $2
+      @repo = $3
+    elsif implicit_shorthand? url
+      @host = "github.com"
+      @user = $1
+      @repo = $2
+    else
+      raise ArgumentError, "Unsupported URL: #{url}"
+    end
+  end
+
+  # Match a well-formed HTTP or HTTPS address
+  def https?(url)
+    nil unless url =~ /
+      ^  (?<protocol>   https? ://)?
+         (?<userauth>   [^@.]+ @  )?
+         (?<subdomain>  www    \. )?
+         (?<hostname>   #{HOSTS}  )
+      \/ (?<user>       [^\/]+    )
+      \/ (?<repo>       [^\/]+    ) /xi
+  end
+
+  # Match an SSH address starting with `git@`
+  def ssh?(url)
+    nil unless url =~ /
+      ^ git@
+        (?<hostname> #{HOSTS}) :
+        (?<user>     [^\/]+)  \/
+        (?<repo>     [^\/]+)  \.git $/xi
+  end
+
+  # Match `provider:user/repo`
+  def shorthand?(url)
+    nil unless url =~ /
+      ^ (?<hostname> #{HOSTS}) : \/?
+        (?<user>     [^\/]+) \/
+        (?<repo>     [^\/]+) \/? $ /xi
+  end
+
+  # Match `user/repo` shorthand, assumed to be GitHub
+  def implicit_shorthand?(url)
+    nil unless url =~ /
+      ^ \/? (?<user>[^\/]+)
+        \/  (?<repo>[^\/]+)
+        \/? $/xi
+  end
+end
 
 
-# Break a repository URL into its separate components
-def parse_url(input)
-  hosts = "github\.com|bitbucket\.org|gitlab\.com"
+class GrammarGuardian
 
-  # HTTPS/HTTP link pointing to recognised hosts
-  if input =~ /^(?:https?:\/\/)?(?:[^.@]+@)?(?:www\.)?(#{hosts})\/([^\/]+)\/([^\/]+)/i
-    { host: $1.downcase(), user: $2, repo: $3.sub(/\.git$/, "") }
-  # SSH
-  elsif input =~ /^git@(#{hosts}):([^\/]+)\/([^\/]+)\.git$/i
-    { host: $1.downcase(), user: $2, repo: $3 }
-  # provider:user/repo
-  elsif input =~ /^(github|bitbucket|gitlab):\/?([^\/]+)\/([^\/]+)\/?$/i
-    { host: $1.downcase(), user: $2, repo: $3 }
-  # user/repo - Common GitHub shorthand
-  elsif input =~ /^\/?([^\/]+)\/([^\/]+)\/?$/
-    { host: "github.com", user: $1, repo: $2 }
-  else
-    raise "Unsupported URL: #{input}"
+  ROOT = File.expand_path("../../", __FILE__)
+
+  def initialize
+    # Track each change so we can roll back after a failed command
+    @changes = Hash.new
+  end
+
+  # Print debugging feedback to STDOUT if running with --verbose
+  def log(msg)
+    puts msg if $verbose
+  end
+
+  def command(*args)
+    log "$ #{args.join(' ')}"
+    output, status = Open3.capture2e(*args)
+    if !status.success?
+      output.each_line do |line|
+        log "  > #{line}"
+      end
+      warn "Command failed. Aborting."
+      exit 1
+    end
   end
 end
 
@@ -38,22 +113,6 @@ def parse_submodule(name)
   path
 end
 
-# Print debugging feedback to STDOUT if running with --verbose
-def log(msg)
-  puts msg if $verbose
-end
-
-def command(*args)
-  log "$ #{args.join(' ')}"
-  output, status = Open3.capture2e(*args)
-  if !status.success?
-    output.each_line do |line|
-      log "  > #{line}"
-    end
-    warn "Command failed. Aborting."
-    exit 1
-  end
-end
 
 usage = """Usage:
   #{$0} [-v|--verbose] [--replace grammar] url