Generate language_id (#3284)

* Generate language_id from language names

The language_id is generated from the SHA256 hash of the language's name

* Test the validity of language ids

All languages should have a positive 32bit integer as an id

* Update languages.yml header in set-language-ids

script/set-language-ids

@@ -11,6 +11,8 @@ header = <<-EOF
 # 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
@@ -20,9 +22,9 @@ header = <<-EOF
 # search_term       - Deprecated: Some languages may be indexed under a
 #                     different alias. Avoid defining new exceptions.
 # 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 
+#                     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.
+# 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
@@ -36,21 +38,23 @@ header = <<-EOF
 # 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!"
 
-  language_index = 0
-
   languages = YAML.load(File.read("lib/linguist/languages.yml"))
   languages.each do |name, vals|
-    vals.merge!('language_id' => language_index)
-    language_index += 1
+    vals.merge!('language_id' => generate_language_id(name))
   end
 
   File.write("lib/linguist/languages.yml", header + YAML.dump(languages))
@@ -58,20 +62,12 @@ 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"))
 
-  # First grab the maximum language_id
-  language_ids = []
-  languages.each { |name, vals| language_ids << vals['language_id'] if vals.has_key?('language_id')}
-  max_language_id = language_ids.max
-  puts "Current maximum language_id is #{max_language_id}"
-
   missing_count = 0
-  language_index = max_language_id
 
   languages.each do |name, vals|
     unless vals.has_key?('language_id')
-      language_index += 1
       missing_count += 1
-      vals.merge!('language_id' => language_index)
+      vals.merge!('language_id' => generate_language_id(name))
     end
   end
 

test/test_language.rb

@@ -427,6 +427,14 @@ class TestLanguage < Minitest::Test
     assert missing.empty?, message
   end
 
+  def test_all_languages_have_a_valid_id
+    invalid = Language.all.select { |language| language.language_id < 0 || language.language_id >= (2**31 - 1) }
+
+    message = "The following languages do not have a valid language_id. Please use script/set-language-ids --update as per the contribution guidelines.\n"
+    invalid.each { |language| message << "#{language.name}\n" }
+    assert invalid.empty?, message
+  end
+
   def test_all_language_id_are_unique
     duplicates = Language.all.group_by{ |language| language.language_id }.select { |k, v| v.size > 1 }.map(&:first)