Merge branch 'master' into 950

Conflicts:
	lib/linguist/languages.yml

.gitignore

@@ -1 +1,3 @@
 Gemfile.lock
+.bundle/
+vendor/

.travis.yml

@@ -2,10 +2,8 @@ before_install:
   - sudo apt-get install libicu-dev -y
   - gem update --system 2.1.11
 rvm:
-  - 1.8.7
-  - 1.9.2
   - 1.9.3
   - 2.0.0
-  - ree
+  - 2.1.1
 notifications:
   disabled: true

Gemfile

@@ -1,7 +1,2 @@
 source 'https://rubygems.org'
 gemspec
-
-if RUBY_VERSION < "1.9.3"
-  # escape_utils 1.0.0 requires 1.9.3 and above
-  gem "escape_utils", "0.3.2"
-end

README.md

@@ -106,6 +106,29 @@ To update the `samples.json` after adding new files to [`samples/`](https://gith
 
     bundle exec rake samples
 
+### A note on language extensions
+
+Linguist has a number of methods available to it for identifying the language of a particular file. The initial lookup is based upon the extension of the file, possible file extensions are defined in an array called `extensions`. Take a look at this example for example for `Perl`:
+
+```
+Perl:
+  type: programming
+  ace_mode: perl
+  color: "#0298c3"
+  extensions:
+  - .pl
+  - .PL
+  - .perl
+  - .ph
+  - .plx
+  - .pm
+  - .pod
+  - .psgi
+  interpreters:
+  - perl
+```
+Any of the extensions defined are valid but the first in this array should be the most popular.
+
 ### Testing
 
 Sometimes getting the tests running can be too much work, especially if you don't have much Ruby experience. It's okay: be lazy and let our build bot [Travis](http://travis-ci.org/#!/github/linguist) run the tests for you. Just open a pull request and the bot will start cranking away.

github-linguist.gemspec

@@ -1,6 +1,8 @@
+require File.expand_path('../lib/linguist/version', __FILE__)
+
 Gem::Specification.new do |s|
   s.name    = 'github-linguist'
-  s.version = '2.10.11'
+  s.version = Linguist::VERSION
   s.summary = "GitHub Language detection"
   s.description = 'We use this library at GitHub to detect blob languages, highlight code, ignore binary files, suppress generated files in diffs, and generate language breakdown graphs.'
 
@@ -12,7 +14,7 @@ Gem::Specification.new do |s|
   s.executables << 'linguist'
 
   s.add_dependency 'charlock_holmes', '~> 0.6.6'
-  s.add_dependency 'escape_utils',    '>= 0.3.1'
+  s.add_dependency 'escape_utils',    '~> 1.0.1'
   s.add_dependency 'mime-types',      '~> 1.19'
   s.add_dependency 'pygments.rb',     '~> 0.5.4'
 

lib/linguist.rb

@@ -4,3 +4,4 @@
 require 'linguist/language'
 require 'linguist/repository'
 require 'linguist/samples'
+require 'linguist/version'

lib/linguist/blob_helper.rb

@@ -291,7 +291,7 @@ def language
         data = lambda { (binary_mime_type? || binary?) ? "" : self.data }
       end
 
-      @language = binary? ? nil : Language.detect(name.to_s, data, mode)
+      @language = Language.detect(name.to_s, data, mode)
     end
 
     # Internal: Get the lexer of the blob.

lib/linguist/generated.rb

@@ -62,7 +62,8 @@ def generated?
         generated_protocol_buffer? ||
         generated_jni_header? ||
         composer_lock? ||
-        node_modules?
+        node_modules? ||
+        vcr_cassette?
     end
 
     # Internal: Is the blob an XCode project file?
@@ -235,5 +236,15 @@ def node_modules?
     def composer_lock?
       !!name.match(/composer.lock/)
     end
+
+    # Is the blob a VCR Cassette file?
+    #
+    # Returns true or false
+    def vcr_cassette?
+      return false unless extname == '.yml'
+      return false unless lines.count > 2
+      # VCR Cassettes have "recorded_with: VCR" in the second last line.
+      return lines[-2].include?("recorded_with: VCR")
+    end
   end
 end

lib/linguist/heuristics.rb

@@ -28,6 +28,9 @@ def self.find_by_heuristics(data, languages)
         if languages.all? { |l| ["Common Lisp", "OpenCL"].include?(l) }
           disambiguate_cl(data, languages)
         end
+        if languages.all? { |l| ["Rebol", "R"].include?(l) }
+          disambiguate_r(data, languages)
+        end
       end
     end
 
@@ -73,6 +76,13 @@ def self.disambiguate_cl(data, languages)
       matches
     end
 
+    def self.disambiguate_r(data, languages)
+      matches = []
+      matches << Language["Rebol"] if /\bRebol\b/i.match(data)
+      matches << Language["R"] if data.include?("<-")
+      matches
+    end
+
     def self.active?
       !!ACTIVE
     end

lib/linguist/language.rb

@@ -24,7 +24,6 @@ class Language
     @extension_index          = Hash.new { |h,k| h[k] = [] }
     @interpreter_index        = Hash.new { |h,k| h[k] = [] }
     @filename_index           = Hash.new { |h,k| h[k] = [] }
-    @primary_extension_index  = {}
 
     # Valid Languages types
     TYPES = [:data, :markup, :programming, :prose]
@@ -80,12 +79,6 @@ def self.create(attributes = {})
         @extension_index[extension] << language
       end
 
-      if @primary_extension_index.key?(language.primary_extension)
-        raise ArgumentError, "Duplicate primary extension: #{language.primary_extension}"
-      end
-
-      @primary_extension_index[language.primary_extension] = language
-
       language.interpreters.each do |interpreter|
         @interpreter_index[interpreter] << language
       end
@@ -191,8 +184,7 @@ def self.find_by_alias(name)
     # Returns all matching Languages or [] if none were found.
     def self.find_by_filename(filename)
       basename, extname = File.basename(filename), File.extname(filename)
-      langs = [@primary_extension_index[extname]] +
-              @filename_index[basename] +
+      langs = @filename_index[basename] +
               @extension_index[extname]
       langs.compact.uniq
     end
@@ -299,15 +291,6 @@ def initialize(attributes = {})
       @interpreters = attributes[:interpreters]   || []
       @filenames  = attributes[:filenames]  || []
 
-      unless @primary_extension = attributes[:primary_extension]
-        raise ArgumentError, "#{@name} is missing primary extension"
-      end
-
-      # Prepend primary extension unless its already included
-      if primary_extension && !extensions.include?(primary_extension)
-        @extensions = [primary_extension] + extensions
-      end
-
       # Set popular, and searchable flags
       @popular    = attributes.key?(:popular)    ? attributes[:popular]    : false
       @searchable = attributes.key?(:searchable) ? attributes[:searchable] : true
@@ -395,20 +378,6 @@ def initialize(attributes = {})
     # Returns the extensions Array
     attr_reader :extensions
 
-    # Deprecated: Get primary extension
-    #
-    # Defaults to the first extension but can be overridden
-    # in the languages.yml.
-    #
-    # The primary extension can not be nil. Tests should verify this.
-    #
-    # This attribute is only used by app/helpers/gists_helper.rb for
-    # creating the language dropdown. It really should be using `name`
-    # instead. Would like to drop primary extension.
-    #
-    # Returns the extension String.
-    attr_reader :primary_extension
-
     # Public: Get interpreters
     #
     # Examples
@@ -426,6 +395,27 @@ def initialize(attributes = {})
     #
     # Returns the extensions Array
     attr_reader :filenames
+
+    # Public: Return all possible extensions for language
+    def all_extensions
+      (extensions + [primary_extension]).uniq
+    end
+
+    # Deprecated: Get primary extension
+    #
+    # Defaults to the first extension but can be overridden
+    # in the languages.yml.
+    #
+    # The primary extension can not be nil. Tests should verify this.
+    #
+    # This method is only used by app/helpers/gists_helper.rb for creating
+    # the language dropdown. It really should be using `name` instead.
+    # Would like to drop primary extension.
+    #
+    # Returns the extension String.
+    def primary_extension
+      extensions.first
+    end
 
     # Public: Get URL escaped name.
     #
@@ -485,7 +475,7 @@ def searchable?
     #
     # Returns html String
     def colorize(text, options = {})
-      lexer.highlight(text, options = {})
+      lexer.highlight(text, options)
     end
 
     # Public: Return name as String representation
@@ -568,9 +558,8 @@ def inspect
       :group_name        => options['group'],
       :searchable        => options.key?('searchable') ? options['searchable'] : true,
       :search_term       => options['search_term'],
-      :extensions        => options['extensions'].sort,
+      :extensions        => [options['extensions'].first] + options['extensions'][1..-1].sort,
       :interpreters      => options['interpreters'].sort,
-      :primary_extension => options['primary_extension'],
       :filenames         => options['filenames'],
       :popular           => popular.include?(name)
     )