Merge pull request #1731 from github/multiple-ext-segments

Support for multiple file extension segments

lib/linguist/file_blob.rb

@@ -57,14 +57,20 @@ module Linguist
     #
     # Returns a String.
     def extension
-      # File.extname returns nil if the filename is an extension.
-      extension = File.extname(name)
-      basename = File.basename(name)
-      # Checks if the filename is an extension.
-      if extension.empty? && basename[0] == "."
-        basename
-      else
-        extension
+      extensions.last || ""
+    end
+
+    # Public: Return an array of the file extensions
+    #
+    #     >> Linguist::FileBlob.new("app/views/things/index.html.erb").extensions
+    #     => [".html.erb", ".erb"]
+    #
+    # Returns an Array
+    def extensions
+      basename, *segments = File.basename(name).split(".")
+
+      segments.map.with_index do |segment, index|
+        "." + segments[index..-1].join(".")
       end
     end
   end

lib/linguist/language.rb

@@ -106,8 +106,8 @@ module Linguist
       # A bit of an elegant hack. If the file is executable but extensionless,
       # append a "magic" extension so it can be classified with other
       # languages that have shebang scripts.
-      extension = FileBlob.new(name).extension
-      if extension.empty? && blob.mode && (blob.mode.to_i(8) & 05) == 05
+      extensions = FileBlob.new(name).extensions
+      if extensions.empty? && blob.mode && (blob.mode.to_i(8) & 05) == 05
         name += ".script!"
       end
 
@@ -190,8 +190,13 @@ module Linguist
     # Returns all matching Languages or [] if none were found.
     def self.find_by_filename(filename)
       basename = File.basename(filename)
-      extname = FileBlob.new(filename).extension
-      (@filename_index[basename] + find_by_extension(extname)).compact.uniq
+
+      # find the first extension with language definitions
+      extname = FileBlob.new(filename).extensions.detect do |e|
+        !@extension_index[e].empty?
+      end
+
+      (@filename_index[basename] + @extension_index[extname]).compact.uniq
     end
 
     # Public: Look up Languages by file extension.

test/test_file_blob.rb

@@ -0,0 +1,10 @@
+require 'linguist/file_blob'
+require 'test/unit'
+
+class TestFileBlob < Test::Unit::TestCase
+  def test_extensions
+    assert_equal [".gitignore"], Linguist::FileBlob.new(".gitignore").extensions
+    assert_equal [".xml"],  Linguist::FileBlob.new("build.xml").extensions
+    assert_equal [".html.erb", ".erb"],  Linguist::FileBlob.new("dotted.dir/index.html.erb").extensions
+  end
+end

test/test_language.rb

@@ -223,6 +223,7 @@ class TestLanguage < Test::Unit::TestCase
     assert_equal [Language['Clojure']], Language.find_by_filename('riemann.config')
     assert_equal [Language['HTML+Django']], Language.find_by_filename('index.jinja')
     assert_equal [Language['Chapel']], Language.find_by_filename('examples/hello.chpl')
+    assert_equal [Language['Ant Build System']], Language.find_by_filename('build.ant.xml')
   end
 
   def test_find_by_shebang