#!/usr/bin/env ruby

require 'bundler/setup'
require 'yaml'
require 'pry'

header = <<-EOF
# Defines all Languages known to GitHub.
#
# type              - Either data, programming, markup, prose, or nil
# aliases           - An Array of additional aliases (implicitly
#                     includes name.downcase)
# ace_mode          - A String name of the Ace Mode used for highlighting whenever
#                     a file is edited. This must match one of the filenames in http://git.io/3XO_Cg.
#                     Use "text" if a mode does not exist.
# codemirror_mode   - A String name of the CodeMirror Mode used for highlighting whenever a file is edited.
#                     This must match a mode from https://git.io/vi9Fx
# wrap              - Boolean wrap to enable line wrapping (default: false)
# extensions        - An Array of associated extensions (the first one is
#                     considered the primary extension, the others should be
#                     listed alphabetically)
# interpreters      - An Array of associated interpreters
# searchable        - Boolean flag to enable searching (defaults to true)
# language_id       - Integer used as a language-name-independent indexed field so that we can rename
#                     languages in Linguist without reindexing all the code on GitHub. Must not be
#                     changed for existing languages without the explicit permission of GitHub staff.
# color             - CSS hex color to represent the language. Only used if type is "programming" or "prose".
# tm_scope          - The TextMate scope that represents this programming
#                     language. This should match one of the scopes listed in
#                     the grammars.yml file. Use "none" if there is no grammar
#                     for this language.
# group             - Name of the parent language. Languages in a group are counted
#                     in the statistics as the parent language.
#
# Any additions or modifications (even trivial) should have corresponding
# test changes in `test/test_blob.rb`.
#
# Please keep this list alphabetized. Capitalization comes before lowercase.

EOF
require 'digest'

generated = true if ARGV[0] == "--force"
update = true if ARGV[0] == "--update"

def generate_language_id(language)
  Digest::SHA256.hexdigest(language).to_i(16) % (2**30 - 1)
end

if generated
  puts "You're regenerating all of the language_id attributes for all Linguist "
  puts "languages defined in languages.yml. This is almost certainly NOT what"
  puts "you meant to do!"

  languages = YAML.load(File.read("lib/linguist/languages.yml"))
  languages.each do |name, vals|
    vals.merge!('language_id' => generate_language_id(name))
  end

  File.write("lib/linguist/languages.yml", header + YAML.dump(languages))
elsif update
  puts "Adding new language_id attributes to languages.yml that don't have one set"
  languages = YAML.load(File.read("lib/linguist/languages.yml"))

  missing_count = 0

  languages.each do |name, vals|
    unless vals.has_key?('language_id')
      missing_count += 1
      vals.merge!('language_id' => generate_language_id(name))
    end
  end

  File.write("lib/linguist/languages.yml", header + YAML.dump(languages))
  puts "Updated language_id attributes for #{missing_count} languages"
else
  puts "Whatever you want me to do, I can't figure it out. Giving up..."
end