linguist/script/add-grammar

#!/usr/bin/env ruby

require "optparse"
require "open3"

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


class GrammarGuardian

  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

# Isolate the vendor-name component of a submodule path
def parse_submodule(name)
  name =~ /^(?:.*(?:vendor\/)?grammars\/)?([^\/]+)/i
  path = "vendor/grammars/#{$1}"
  unless File.exist?("#{ROOT}/" + path)
    warn "Submodule '#{path}' does not exist. Aborting."
    exit 1
  end
  path
end


usage = """Usage:
  #{$0} [-v|--verbose] [--replace grammar] url
Examples:
  #{$0} https://github.com/Alhadis/language-roff
  #{$0} --replace sublime-apl https://github.com/Alhadis/language-apl
"""

$replace = nil
$verbose = true
$compile = false

OptionParser.new do |opts|
  opts.banner = usage
  opts.on("-q", "--quiet", "Do not print output unless there's a failure") do
    $verbose = false
  end
  opts.on("-rSUBMODULE", "--replace=SUBMODDULE", "Replace an existing grammar submodule.") do |name|
    $replace = name
  end
  opts.on("-C", "--compile", "Compile grammar using the new grammar-compiler.") do
    $compile = true
  end
end.parse!


$url = ARGV[0]

# No URL? Print a usage message and bail.
unless $url
  warn usage
  exit 1;
end

# Ensure the given URL is an HTTPS link
parts    = parse_url $url
https    = "https://#{parts[:host]}/#{parts[:user]}/#{parts[:repo]}"
repo_new = "vendor/grammars/#{parts[:repo]}"
repo_old = parse_submodule($replace) if $replace

Dir.chdir(ROOT)

if repo_old
  log "Deregistering: #{repo_old}"
  command('git', 'submodule', 'deinit', repo_old)
  command('git', 'rm', '-rf', repo_old)
  command('script/grammar-compiler', 'update', '-f') if $compile
end

log "Registering new submodule: #{repo_new}"
command('git', 'submodule', 'add', '-f', https, repo_new)
command('script/grammar-compiler', 'add', repo_new) if $compile

log "Confirming license"
if repo_old
  command('script/licensed')
else
  repo_new = File.absolute_path(repo_new)
  command('script/licensed', '--module', repo_new)
end

log "Updating grammar documentation in vendor/README.md"
command('bundle', 'exec', 'rake', 'samples')
command('script/sort-submodules')
command('script/list-grammars')