Add language groups

lib/linguist/language.rb

@@ -197,6 +197,21 @@ module Linguist
       @popular    = attributes.key?(:popular)    ? attributes[:popular]    : false
       @common     = attributes.key?(:common)     ? attributes[:common]     : false
       @searchable = attributes.key?(:searchable) ? attributes[:searchable] : true
+
+      # If group name is set, save the name so we can lazy load it later
+      if attributes[:group_name]
+        if common?
+          warn "#{name} is a major langauage, it should not be grouped with #{attributes[:group_name]}"
+        end
+
+        @group = nil
+        @group_name = attributes[:group_name]
+
+      # Otherwise we can set it to self now
+      else
+        @group = self
+      end
+
     end
 
     # Public: Get proper name
@@ -261,6 +276,19 @@ module Linguist
       name.downcase.gsub(/\s/, '-')
     end
 
+    # Public: Get Language group
+    #
+    # Minor languages maybe grouped with major languages for
+    # accounting purposes. For an example, JSP files are grouped as
+    # Java.
+    #
+    # For major languages, group should always return self.
+    #
+    # Returns a Language
+    def group
+      @group ||= Language.find_by_name(@group_name)
+    end
+
     # Public: Is it popular?
     #
     # Returns true or false
@@ -337,6 +365,7 @@ module Linguist
       :name        => name,
       :aliases     => options['aliases'],
       :lexer       => options['lexer'],
+      :group_name  => options['group'],
       :searchable  => options.key?('searchable') ? options['searchable'] : true,
       :search_term => options['search_term'],
       :extensions  => options['extensions'],

lib/linguist/languages.yml

@@ -65,6 +65,7 @@ Assembly:
   - .asm
 
 Batchfile:
+  group: Shell
   search_term: bat
   aliases:
   - bat
@@ -88,6 +89,7 @@ Brainfuck:
   extensions:
   - .b
   - .bf
+
 C:
   extensions:
   - .c
@@ -169,6 +171,7 @@ Cucumber:
   - .feature
 
 Cython:
+  group: Python
   extensions:
   - .pyx
   - .pxd
@@ -244,6 +247,7 @@ Factor:
   - .factor
 
 GAS:
+  group: Assembly
   extensions:
   - .s
   - .S
@@ -253,11 +257,13 @@ Genshi:
   - .kid
 
 Gentoo Ebuild:
+  group: Shell
   lexer: Bash
   extensions:
   - .ebuild
 
 Gentoo Eclass:
+  group: Shell
   lexer: Bash
   extensions:
   - .eclass
@@ -299,17 +305,20 @@ HTML:
   - .xslt
 
 HTML+Django:
+  group: HTML
   lexer: HTML+Django/Jinja
   extensions:
   - .mustache
 
 HTML+ERB:
+  group: HTML
   lexer: RHTML
   extensions:
   - .erb
   - .html.erb
 
 HTML+PHP:
+  group: HTML
   extensions:
   - .phtml
 
@@ -350,6 +359,7 @@ Io:
   - .io
 
 JSON:
+  group: JavaScript
   lexer: JavaScript
   search_term: javascript
   extensions:
@@ -361,6 +371,7 @@ Java:
   - .pde
 
 Java Server Pages:
+  group: Java
   lexer: Java Server Page
   search_term: jsp
   aliases:
@@ -387,6 +398,7 @@ LLVM:
   - .ll
 
 Literate Haskell:
+  group: Haskell
   search_term: lhs
   aliases:
   - lhs
@@ -460,6 +472,7 @@ Nu:
   - Nukefile
 
 NumPy:
+  group: Python
   extensions:
   - .numpy
   - .numsc
@@ -488,6 +501,7 @@ Objective-J:
   - .sj
 
 OpenCL:
+  group: C
   lexer: C
   extensions:
   - .cl
@@ -532,6 +546,7 @@ Python:
   - .pyw
 
 Python traceback:
+  group: Python
   lexer: Python Traceback
   search_term: pytb
   aliases:
@@ -546,6 +561,7 @@ R:
   - .R
 
 RHTML:
+  group: HTML
   extensions:
   - .rhtml
 

test/test_language.rb

@@ -196,6 +196,28 @@ class TestLanguage < Test::Unit::TestCase
     assert_equal Language['reStructuredText'], Language.find_by_alias('rst')
   end
 
+  def test_common_groups
+    Language.all.each do |language|
+      if language.common?
+        assert_equal language, language.group
+      end
+    end
+  end
+
+  def test_groups
+    assert_equal Language['Assembly'], Language['GAS'].group
+    assert_equal Language['C'], Language['OpenCL'].group
+    assert_equal Language['Haskell'], Language['Literate Haskell'].group
+    assert_equal Language['Java'], Language['Java Server Pages'].group
+    assert_equal Language['JavaScript'], Language['JSON'].group
+    assert_equal Language['Python'], Language['Cython'].group
+    assert_equal Language['Python'], Language['NumPy'].group
+    assert_equal Language['Python'], Language['Python traceback'].group
+    assert_equal Language['Shell'], Language['Batchfile'].group
+    assert_equal Language['Shell'], Language['Gentoo Ebuild'].group
+    assert_equal Language['Shell'], Language['Gentoo Eclass'].group
+  end
+
   # Used for code search indexing. Changing any of these values may
   # require reindexing repositories.
   def test_search_term