Merge branch 'master' into 1036-local

Conflicts:
	lib/linguist/heuristics.rb
	lib/linguist/samples.json

.gitignore

@@ -1,3 +1,4 @@
 Gemfile.lock
 .bundle/
-vendor/
+benchmark/
+lib/linguist/samples.json

.travis.yml

@@ -1,8 +1,8 @@
 before_install:
   - git fetch origin master:master
   - git fetch origin v2.0.0:v2.0.0
+  - git fetch origin test/attributes:test/attributes
   - sudo apt-get install libicu-dev -y
-  - gem update --system 2.1.11
 rvm:
   - 1.9.3
   - 2.0.0

README.md

@@ -32,33 +32,57 @@ The Language stats bar that you see on every repository is built by aggregating
 
 The repository stats API, accessed through `#languages`, can be used on a directory:
 
+***API UPDATE***
+
+Since [Version 3.0.0](https://github.com/github/linguist/releases/tag/v3.0.0) Linguist expects a git repository (in the form of a [Rugged::Repository](https://github.com/libgit2/rugged#repositories)) to be passed when initializing `Linguist::Repository`.
+
+
 ```ruby
-project = Linguist::Repository.from_directory(".")
+require 'rugged'
-project.language.name  #=> "Ruby"
+require 'linguist'
-project.languages      #=> { "Ruby" => 0.98, "Shell" => 0.02 }
+
+repo = Rugged::Repository.new('.')
+project = Linguist::Repository.new(repo, repo.head.target_id)
+project.language       #=> "Ruby"
+project.languages      #=> { "Ruby" => 119387 }
 ```
 
 These stats are also printed out by the `linguist` binary. You can use the
 `--breakdown` flag, and the binary will also output the breakdown of files by language.
 
-You can try running `linguist` on the `lib/` directory in this repository itself:
+You can try running `linguist` on the root directory in this repository itself:
 
-    $ bundle exec linguist lib/ --breakdown
+    $ bundle exec linguist --breakdown
 
     100.00% Ruby
 
     Ruby:
-    linguist/blob_helper.rb
+    Gemfile
-    linguist/classifier.rb
+    Rakefile
-    linguist/file_blob.rb
+    bin/linguist
-    linguist/generated.rb
+    github-linguist.gemspec
-    linguist/heuristics.rb
+    lib/linguist.rb
-    linguist/language.rb
+    lib/linguist/blob_helper.rb
-    linguist/md5.rb
+    lib/linguist/classifier.rb
-    linguist/repository.rb
+    lib/linguist/file_blob.rb
-    linguist/samples.rb
+    lib/linguist/generated.rb
-    linguist/tokenizer.rb
+    lib/linguist/heuristics.rb
-    linguist.rb
+    lib/linguist/language.rb
+    lib/linguist/lazy_blob.rb
+    lib/linguist/md5.rb
+    lib/linguist/repository.rb
+    lib/linguist/samples.rb
+    lib/linguist/tokenizer.rb
+    lib/linguist/version.rb
+    test/test_blob.rb
+    test/test_classifier.rb
+    test/test_heuristics.rb
+    test/test_language.rb
+    test/test_md5.rb
+    test/test_pedantic.rb
+    test/test_repository.rb
+    test/test_samples.rb
+    test/test_tokenizer.rb
 
 #### Ignore vendored files
 
@@ -80,9 +104,32 @@ Linguist::FileBlob.new("underscore.min.js").generated? # => true
 
 See [Linguist::Generated#generated?](https://github.com/github/linguist/blob/master/lib/linguist/generated.rb).
 
+## Overrides
+
+Linguist supports custom overrides for language definitions and vendored paths. Add a `.gitattributes` file to your project using the keys `linguist-language` and `linguist-vendored` with the standard git-style path matchers for the files you want to override.
+
+```
+$ cat .gitattributes
+*.rb linguist-language=Java
+
+$ linguist --breakdown
+100.00% Java
+
+Java:
+ruby_file.rb
+```
+
+By default, Linguist treats all of the paths defined in [lib/linguist/vendor.yml](https://github.com/github/linguist/blob/master/lib/linguist/vendor.yml) as vendored and therefore doesn't include them in the language statistics for a repository. Use the `linguist-vendored` attribute to vendor or un-vendor paths.
+
+```
+$ cat .gitattributes
+special-vendored-path/* linguist-vendored
+jquery.js linguist-vendored=false
+```
+
 ## Installation
 
-github.com is usually running the latest version of the `github-linguist` gem that is released on [RubyGems.org](http://rubygems.org/gems/github-linguist).
+Github.com is usually running the latest version of the `github-linguist` gem that is released on [RubyGems.org](http://rubygems.org/gems/github-linguist).
 
 But for development you are going to want to checkout out the source. To get it, clone the repo and run [Bundler](http://gembundler.com/) to install its dependencies.
 
@@ -102,10 +149,6 @@ We try to only add languages once they have some usage on GitHub, so please note
 
 Almost all bug fixes or new language additions should come with some additional code samples. Just drop them under [`samples/`](https://github.com/github/linguist/tree/master/samples) in the correct subdirectory and our test suite will automatically test them. In most cases you shouldn't need to add any new assertions.
 
-To update the `samples.json` after adding new files to [`samples/`](https://github.com/github/linguist/tree/master/samples):
-
-    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`:
@@ -145,7 +188,7 @@ If you are the current maintainer of this gem:
  0. Ensure that tests are green: `bundle exec rake test`
  0. Bump gem version in `lib/linguist/version.rb`.  For example, [like this](https://github.com/github/linguist/commit/8d2ea90a5ba3b2fe6e1508b7155aa4632eea2985).
  0. Make a PR to github/linguist.  For example, [#1238](https://github.com/github/linguist/pull/1238).
- 0. Build a local gem: `gem build github-linguist.gemspec`
+ 0. Build a local gem: `bundle exec rake build_gem`
  0. Testing:
    0. Bump the Gemfile and Gemfile.lock versions for an app which relies on this gem
    0. Install the new gem locally

Rakefile

@@ -1,3 +1,4 @@
+require 'bundler/setup'
 require 'json'
 require 'rake/clean'
 require 'rake/testtask'
@@ -7,6 +8,16 @@ task :default => :test
 
 Rake::TestTask.new
 
+# Extend test task to check for samples
+task :test => :check_samples
+
+desc "Check that we have samples.json generated"
+task :check_samples do
+  unless File.exist?('lib/linguist/samples.json')
+    Rake::Task[:samples].invoke
+  end
+end
+
 task :samples do
   require 'linguist/samples'
   require 'yajl'
@@ -15,13 +26,74 @@ task :samples do
   File.open('lib/linguist/samples.json', 'w') { |io| io.write json }
 end
 
-task :build_gem do
+task :build_gem => :samples do
   languages = YAML.load_file("lib/linguist/languages.yml")
   File.write("lib/linguist/languages.json", JSON.dump(languages))
   `gem build github-linguist.gemspec`
   File.delete("lib/linguist/languages.json")
 end
 
+namespace :benchmark do
+  benchmark_path = "benchmark/results"
+
+  # $ bundle exec rake benchmark:generate CORPUS=path/to/samples
+  desc "Generate results for"
+  task :generate do
+    ref = `git rev-parse HEAD`.strip[0,8]
+
+    corpus = File.expand_path(ENV["CORPUS"] || "samples")
+
+    require 'linguist/language'
+
+    results = Hash.new
+    Dir.glob("#{corpus}/**/*").each do |file|
+      next unless File.file?(file)
+      filename = file.gsub("#{corpus}/", "")
+      results[filename] = Linguist::FileBlob.new(file).language
+    end
+
+    # Ensure results directory exists
+    FileUtils.mkdir_p("benchmark/results")
+
+    # Write results
+    if `git status`.include?('working directory clean')
+      result_filename = "benchmark/results/#{File.basename(corpus)}-#{ref}.json"
+    else
+      result_filename = "benchmark/results/#{File.basename(corpus)}-#{ref}-unstaged.json"
+    end
+
+    File.write(result_filename, results.to_json)
+    puts "wrote #{result_filename}"
+  end
+
+  # $ bundle exec rake benchmark:compare REFERENCE=path/to/reference.json CANDIDATE=path/to/candidate.json
+  desc "Compare results"
+  task :compare do
+    reference_file = ENV["REFERENCE"]
+    candidate_file = ENV["CANDIDATE"]
+
+    reference = JSON.parse(File.read(reference_file))
+    reference_counts = Hash.new(0)
+    reference.each { |filename, language| reference_counts[language] += 1 }
+
+    candidate = JSON.parse(File.read(candidate_file))
+    candidate_counts = Hash.new(0)
+    candidate.each { |filename, language| candidate_counts[language] += 1 }
+
+    changes = diff(reference_counts, candidate_counts)
+
+    if changes.any?
+      changes.each do |language, (before, after)|
+        before_percent = 100 * before / reference.size.to_f
+        after_percent = 100 * after / candidate.size.to_f
+        puts "%s changed from %.1f%% to %.1f%%" % [language || 'unknown', before_percent, after_percent]
+      end
+    else
+      puts "No changes"
+    end
+  end
+end
+
 namespace :classifier do
   LIMIT = 1_000
 
@@ -37,7 +109,7 @@ namespace :classifier do
       next if file_language.nil? || file_language == 'Text'
       begin
         data = open(file_url).read
-        guessed_language, score = Linguist::Classifier.classify(Linguist::Samples::DATA, data).first
+        guessed_language, score = Linguist::Classifier.classify(Linguist::Samples.cache, data).first
 
         total += 1
         guessed_language == file_language ? correct += 1 : incorrect += 1
@@ -71,3 +143,10 @@ namespace :classifier do
     end
   end
 end
+
+
+def diff(a, b)
+  (a.keys | b.keys).each_with_object({}) do |key, diff|
+    diff[key] = [a[key], b[key]] unless a[key] == b[key]
+  end
+end

github-linguist.gemspec

@@ -17,10 +17,11 @@ Gem::Specification.new do |s|
   s.add_dependency 'escape_utils',    '~> 1.0.1'
   s.add_dependency 'mime-types',      '~> 1.19'
   s.add_dependency 'pygments.rb',     '~> 0.6.0'
-  s.add_dependency 'rugged',          '~> 0.21.0'
+  s.add_dependency 'rugged',          '~> 0.21.1b2'
 
   s.add_development_dependency 'json'
   s.add_development_dependency 'mocha'
+  s.add_development_dependency 'pry'
   s.add_development_dependency 'rake'
   s.add_development_dependency 'yajl-ruby'
 end

lib/linguist/blob_helper.rb

@@ -321,6 +321,11 @@ module Linguist
       language ? language.lexer : Pygments::Lexer.find_by_name('Text only')
     end
 
+    # Internal: Get the TextMate compatible scope for the blob
+    def tm_scope
+      language && language.tm_scope
+    end
+
     # Public: Highlight syntax of blob
     #
     # options - A Hash of options (defaults to {})

lib/linguist/generated.rb

@@ -63,6 +63,7 @@ module Linguist
         generated_jni_header? ||
         composer_lock? ||
         node_modules? ||
+        godeps? ||
         vcr_cassette? ||
         generated_by_zephir?
     end
@@ -231,6 +232,14 @@ module Linguist
       !!name.match(/node_modules\//)
     end
 
+    # Internal: Is the blob part of Godeps/,
+    # which are not meant for humans in pull requests.
+    #
+    # Returns true or false.
+    def godeps?
+      !!name.match(/Godeps\//)
+    end
+
     # Internal: Is the blob a generated php composer lock file?
     #
     # Returns true or false.

lib/linguist/heuristics.rb

@@ -14,27 +14,25 @@ module Linguist
     def self.find_by_heuristics(data, languages)
       if active?
         if languages.all? { |l| ["Objective-C", "C++", "C"].include?(l) }
-          disambiguate_c(data, languages)
+          result = disambiguate_c(data, languages)
         end
         if languages.all? { |l| ["Perl", "Prolog"].include?(l) }
-          disambiguate_pl(data, languages)
+          result = disambiguate_pl(data, languages)
         end
         if languages.all? { |l| ["ECL", "Prolog"].include?(l) }
-          disambiguate_ecl(data, languages)
+          result = disambiguate_ecl(data, languages)
         end
-        if languages.all? { |l| ["TypeScript", "XML"].include?(l) }
+        if languages.all? { |l| ["IDL", "Prolog"].include?(l) }
-          disambiguate_ts(data, languages)
+          result = disambiguate_pro(data, languages)
         end
         if languages.all? { |l| ["Common Lisp", "OpenCL"].include?(l) }
-          disambiguate_cl(data, languages)
+          result = disambiguate_cl(data, languages)
-        end
-        if languages.all? { |l| ["Rebol", "R"].include?(l) }
-          disambiguate_r(data, languages)
         end
+        return result
       end
     end
 
-    # .h extensions are ambigious between C, C++, and Objective-C.
+    # .h extensions are ambiguous between C, C++, and Objective-C.
     # We want to shortcut look for Objective-C _and_ now C++ too!
     #
     # Returns an array of Languages or []
@@ -64,6 +62,16 @@ module Linguist
       matches
     end
 
+    def self.disambiguate_pro(data, languages)
+      matches = []
+      if (data.include?(":-"))
+        matches << Language["Prolog"]
+      else
+        matches << Language["IDL"]
+      end
+      matches
+    end
+
     def self.disambiguate_ts(data, languages)
       matches = []
       if (data.include?("</translation>"))

lib/linguist/language.rb

@@ -135,8 +135,8 @@ module Linguist
         # No shebang. Still more work to do. Try to find it with our heuristics.
         elsif (determined = Heuristics.find_by_heuristics(data, possible_language_names)) && !determined.empty?
           determined.first
-        # Lastly, fall back to the probablistic classifier.
+        # Lastly, fall back to the probabilistic classifier.
-        elsif classified = Classifier.classify(Samples::DATA, data, possible_language_names ).first
+        elsif classified = Classifier.classify(Samples.cache, data, possible_language_names).first
           # Return the actual Language object based of the string language name (i.e., first element of `#classify`)
           Language[classified[0]]
         end
@@ -290,6 +290,16 @@ module Linguist
       @lexer = Pygments::Lexer.find_by_name(attributes[:lexer] || name) ||
         raise(ArgumentError, "#{@name} is missing lexer")
 
+      @tm_scope = attributes[:tm_scope] || begin
+        context = case @type
+                  when :data, :markup, :prose
+                    'text'
+                  when :programming, nil
+                    'source'
+                  end
+        "#{context}.#{@name.downcase}"
+      end
+
       @ace_mode = attributes[:ace_mode]
       @wrap = attributes[:wrap] || false
 
@@ -363,6 +373,11 @@ module Linguist
     # Returns the Lexer
     attr_reader :lexer
 
+    # Public: Get the name of a TextMate-compatible scope
+    #
+    # Returns the scope
+    attr_reader :tm_scope
+
     # Public: Get Ace mode
     #
     # Examples
@@ -510,9 +525,9 @@ module Linguist
     end
   end
 
-  extensions = Samples::DATA['extnames']
+  extensions = Samples.cache['extnames']
-  interpreters = Samples::DATA['interpreters']
+  interpreters = Samples.cache['interpreters']
-  filenames = Samples::DATA['filenames']
+  filenames = Samples.cache['filenames']
   popular = YAML.load_file(File.expand_path("../popular.yml", __FILE__))
 
   languages_yml = File.expand_path("../languages.yml", __FILE__)
@@ -564,6 +579,7 @@ module Linguist
       :type              => options['type'],
       :aliases           => options['aliases'],
       :lexer             => options['lexer'],
+      :tm_scope          => options['tm_scope'],
       :ace_mode          => options['ace_mode'],
       :wrap              => options['wrap'],
       :group_name        => options['group'],

lib/linguist/languages.yml

@@ -83,6 +83,7 @@ ATS:
 ActionScript:
   type: programming
   lexer: ActionScript 3
+  tm_scope: source.actionscript.3
   color: "#e3491a"
   search_term: as3
   aliases:
@@ -119,7 +120,7 @@ ApacheConf:
 
 Apex:
   type: programming
-  lexer: Text only
+  lexer: Java
   extensions:
   - .cls
 
@@ -173,6 +174,7 @@ Assembly:
   - nasm
   extensions:
   - .asm
+  - .a51
 
 Augeas:
   type: programming
@@ -284,8 +286,9 @@ C:
 C#:
   type: programming
   ace_mode: csharp
+  tm_scope: source.cs
   search_term: csharp
-  color: "#5a25a2"
+  color: "#178600"
   aliases:
   - csharp
   extensions:
@@ -411,6 +414,7 @@ Clojure:
 
 CoffeeScript:
   type: programming
+  tm_scope: source.coffee
   ace_mode: coffee
   color: "#244776"
   aliases:
@@ -453,6 +457,7 @@ ColdFusion CFC:
 
 Common Lisp:
   type: programming
+  tm_scope: source.lisp
   color: "#3fb68b"
   aliases:
   - lisp
@@ -648,6 +653,7 @@ Elm:
 Emacs Lisp:
   type: programming
   lexer: Common Lisp
+  tm_scope: source.lisp
   color: "#c065db"
   aliases:
   - elisp
@@ -748,6 +754,7 @@ Forth:
   - .fth
   - .4th
   - .forth
+  - .frt
 
 Frege:
   type: programming
@@ -756,6 +763,14 @@ Frege:
   extensions:
   - .fr
 
+G-code:
+  type: data
+  lexer: Text only
+  extensions:
+  - .g
+  - .gco
+  - .gcode
+
 Game Maker Language:
   type: programming
   color: "#8ad353"
@@ -785,6 +800,12 @@ GAS:
   - .s
   - .S
 
+GDScript:
+  type: programming
+  lexer: Text only
+  extensions:
+  - .gd
+
 GLSL:
   group: C
   type: programming
@@ -877,6 +898,12 @@ Grammatical Framework:
   searchable: true
   color: "#ff0000"
 
+Graph Modeling Language:
+  type: data
+  lexer: Text only
+  extensions:
+  - .gml
+
 Groff:
   extensions:
   - .man
@@ -911,6 +938,7 @@ Groovy Server Pages:
 
 HTML:
   type: markup
+  tm_scope: text.html.basic
   ace_mode: html
   aliases:
   - xhtml
@@ -922,6 +950,7 @@ HTML:
 
 HTML+Django:
   type: markup
+  tm_scope: text.html.django
   group: HTML
   lexer: HTML+Django/Jinja
   extensions:
@@ -930,6 +959,7 @@ HTML+Django:
 
 HTML+ERB:
   type: markup
+  tm_scope: text.html.ruby
   group: HTML
   lexer: RHTML
   aliases:
@@ -940,6 +970,7 @@ HTML+ERB:
 
 HTML+PHP:
   type: markup
+  tm_scope: text.html.php
   group: HTML
   extensions:
   - .phtml
@@ -959,6 +990,8 @@ Haml:
 Handlebars:
   type: markup
   lexer: Handlebars
+  aliases:
+  - hbs
   extensions:
   - .handlebars
   - .hbs
@@ -1075,6 +1108,7 @@ J:
 
 JSON:
   type: data
+  tm_scope: source.json
   group: JavaScript
   ace_mode: json
   searchable: false
@@ -1137,6 +1171,7 @@ Java Server Pages:
 
 JavaScript:
   type: programming
+  tm_scope: source.js
   ace_mode: javascript
   color: "#f1e05a"
   aliases:
@@ -1149,6 +1184,7 @@ JavaScript:
   - .es6
   - .frag
   - .jake
+  - .jsb
   - .jsfl
   - .jsm
   - .jss
@@ -1203,6 +1239,16 @@ LLVM:
   extensions:
   - .ll
 
+LSL:
+  type: programming
+  lexer: LSL
+  ace_mode: lsl
+  extensions:
+  - .lsl
+  interpreters:
+  - lsl
+  color: '#3d9970'
+
 LabVIEW:
   type: programming
   lexer: Text only
@@ -1254,6 +1300,7 @@ Literate Agda:
 
 Literate CoffeeScript:
   type: programming
+  tm_scope: source.litcoffee
   group: CoffeeScript
   lexer: Text only
   ace_mode: markdown
@@ -1310,6 +1357,7 @@ Lua:
   color: "#fa1fa1"
   extensions:
   - .lua
+  - .fcgi
   - .nse
   - .pd_lua
   - .rbxs
@@ -1537,6 +1585,7 @@ ObjDump:
 
 Objective-C:
   type: programming
+  tm_scope: source.objc
   color: "#438eff"
   aliases:
   - obj-c
@@ -1547,6 +1596,7 @@ Objective-C:
 
 Objective-C++:
   type: programming
+  tm_scope: source.objc++
   color: "#4886FC"
   aliases:
   - obj-c++
@@ -1637,12 +1687,14 @@ PAWN:
 
 PHP:
   type: programming
+  tm_scope: text.html.php
   ace_mode: php
   color: "#4F5D95"
   extensions:
   - .php
   - .aw
   - .ctp
+  - .fcgi
   - .module
   - .php3
   - .php4
@@ -1694,6 +1746,7 @@ Pascal:
   - .dfm
   - .dpr
   - .lpr
+  - .pp
 
 Perl:
   type: programming
@@ -1782,11 +1835,13 @@ Processing:
 
 Prolog:
   type: programming
+  lexer: Logtalk
   color: "#74283c"
   extensions:
-  - .prolog
-  - .ecl
   - .pl
+  - .ecl
+  - .pro
+  - .prolog
 
 Propeller Spin:
   type: programming
@@ -1831,6 +1886,8 @@ Python:
   color: "#3581ba"
   extensions:
   - .py
+  - .cgi
+  - .fcgi
   - .gyp
   - .lmi
   - .pyde
@@ -1992,6 +2049,7 @@ Ruby:
   extensions:
   - .rb
   - .builder
+  - .fcgi
   - .gemspec
   - .god
   - .irbrc
@@ -2039,6 +2097,7 @@ SAS:
 
 SCSS:
   type: markup
+  tm_scope: source.scss
   group: CSS
   ace_mode: scss
   extensions:
@@ -2054,6 +2113,7 @@ SQF:
 
 SQL:
   type: data
+  tm_scope: source.sql
   ace_mode: sql
   extensions:
   - .sql
@@ -2078,6 +2138,7 @@ Sage:
 
 Sass:
   type: markup
+  tm_scope: source.sass
   group: CSS
   extensions:
   - .sass
@@ -2140,6 +2201,8 @@ Shell:
   - .sh
   - .bash
   - .bats
+  - .cgi
+  - .fcgi
   - .tmux
   - .zsh
   interpreters:
@@ -2271,6 +2334,9 @@ Tcl:
   - .tcl
   - .adp
   - .tm
+  interpreters:
+  - tclsh
+  - wish
 
 Tcsh:
   type: programming
@@ -2402,6 +2468,7 @@ VimL:
   - .vim
   filenames:
   - .vimrc
+  - _vimrc
   - vimrc
   - gvimrc
 
@@ -2551,6 +2618,7 @@ Xtend:
 
 YAML:
   type: data
+  tm_scope: source.yaml
   aliases:
   - yml
   extensions:

lib/linguist/lazy_blob.rb

@@ -1,8 +1,13 @@
 require 'linguist/blob_helper'
+require 'linguist/language'
 require 'rugged'
 
 module Linguist
   class LazyBlob
+    GIT_ATTR = ['linguist-language', 'linguist-vendored']
+    GIT_ATTR_OPTS = { :priority => [:index], :skip_system => true }
+    GIT_ATTR_FLAGS = Rugged::Repository::Attributes.parse_opts(GIT_ATTR_OPTS)
+
     include BlobHelper
 
     MAX_SIZE = 128 * 1024
@@ -19,6 +24,29 @@ module Linguist
       @mode = mode
     end
 
+    def git_attributes
+      @git_attributes ||= repository.fetch_attributes(
+        name, GIT_ATTR, GIT_ATTR_FLAGS)
+    end
+
+    def vendored?
+      if attr = git_attributes['linguist-vendored']
+        return boolean_attribute(attr)
+      else
+        return super
+      end
+    end
+
+    def language
+      return @language if defined?(@language)
+
+      @language = if lang = git_attributes['linguist-language']
+        Language.find_by_name(lang)
+      else
+        super
+      end
+    end
+
     def data
       load_blob!
       @data
@@ -30,6 +58,12 @@ module Linguist
     end
 
     protected
+
+    # Returns true if the attribute is present and not the string "false".
+    def boolean_attribute(attr)
+      attr != "false"
+    end
+
     def load_blob!
       @data, @size = Rugged::Blob.to_buffer(repository, oid, MAX_SIZE) if @data.nil?
     end

lib/linguist/repository.rb

@@ -110,18 +110,37 @@ module Linguist
         if @old_commit_oid == @commit_oid
           @old_stats
         else
-          compute_stats(@old_commit_oid, @commit_oid, @old_stats)
+          compute_stats(@old_commit_oid, @old_stats)
         end
       end
     end
 
+    def read_index
+      attr_index = Rugged::Index.new
+      attr_index.read_tree(current_tree)
+      repository.index = attr_index
+    end
+
+    def current_tree
+      @tree ||= Rugged::Commit.lookup(repository, @commit_oid).tree
+    end
+
     protected
-    def compute_stats(old_commit_oid, commit_oid, cache = nil)
-      file_map = cache ? cache.dup : {}
-      old_tree = old_commit_oid && Rugged::Commit.lookup(repository, old_commit_oid).tree
-      new_tree = Rugged::Commit.lookup(repository, commit_oid).tree
 
-      diff = Rugged::Tree.diff(repository, old_tree, new_tree)
+    def compute_stats(old_commit_oid, cache = nil)
+      old_tree = old_commit_oid && Rugged::Commit.lookup(repository, old_commit_oid).tree
+
+      read_index
+
+      diff = Rugged::Tree.diff(repository, old_tree, current_tree)
+
+      # Clear file map and fetch full diff if any .gitattributes files are changed
+      if cache && diff.each_delta.any? { |delta| File.basename(delta.new_file[:path]) == ".gitattributes" }
+        diff = Rugged::Tree.diff(repository, old_tree = nil, current_tree)
+        file_map = {}
+      else
+        file_map = cache ? cache.dup : {}
+      end
 
       diff.each_delta do |delta|
         old = delta.old_file[:path]

lib/linguist/samples.rb

@@ -17,9 +17,11 @@ module Linguist
     PATH = File.expand_path('../samples.json', __FILE__)
 
     # Hash of serialized samples object
-    if File.exist?(PATH)
+    def self.cache
+      @cache ||= begin
         serializer = defined?(JSON) ? JSON : YAML
-      DATA = serializer.load(File.read(PATH))
+        serializer.load(File.read(PATH))
+      end
     end
 
     # Public: Iterate over each sample.

lib/linguist/vendor.yml

@@ -33,6 +33,9 @@
 # Erlang bundles
 - ^rebar$
 
+# Go dependencies
+- Godeps/_workspace/
+
 # Bootstrap minified css and js
 - (^|/)bootstrap([^.]*)(\.min)?\.(js|css)$
 
@@ -235,3 +238,7 @@
 - octicons.css
 - octicons.min.css
 - sprockets-octicons.scss
+
+# Typesafe Activator
+- (^|/)activator$
+- (^|/)activator\.bat$

lib/linguist/version.rb

@@ -1,3 +1,3 @@
 module Linguist
-  VERSION = "3.1.2"
+  VERSION = "3.4.0"
 end

samples/Assembly/External Interrupt.a51

@@ -0,0 +1,66 @@
+	ORG	0000h
+	SJMP	START
+	ORG	0003h
+	LCALL	INT0_ISR
+	RETI
+	ORG	000Bh
+	LCALL	T0_ISR
+	RETI
+	ORG	0013h
+	LCALL	INT1_ISR
+	RETI
+	ORG	001Bh
+	LCALL	T1_ISR
+	RETI
+	ORG	0023h
+	LCALL	UART_ISR
+	RETI
+	ORG	0030h
+START:
+	MOV	A,#11111110b
+	SETB	IT0	; Set External Interrupt 0 to be falling edge triggered
+	SETB	EX0	; Enable External Interrut 0
+	SETB	EA	; Enable Interrupt
+LEFT:			
+	CJNE	A,#01111111b,LOOP1
+	JMP	RIGHT
+LOOP1:
+	MOV	P1,A
+	RL	A	
+	LCALL	DELAY
+	SJMP	LEFT	
+RIGHT:
+	CJNE	A,#11111110b,LOOP2
+	JMP	LEFT
+LOOP2:
+	MOV	P1,A
+	RR	A	
+	LCALL	DELAY
+	SJMP	RIGHT
+	
+INT0_ISR:
+	MOV	R1,#3
+FLASH:
+	MOV	P1,#00h
+	LCALL	DELAY
+	MOV	P1,#0FFh
+	LCALL	DELAY
+	DJNZ	R1,FLASH
+	RET
+T0_ISR:
+	RET
+INT1_ISR:
+	RET
+T1_ISR:
+	RET
+UART_ISR:
+	RET
+
+DELAY:	MOV	R5,#20	;R5*20 mS
+D1:     MOV	R6,#40
+D2:     MOV	R7,#249
+	DJNZ	R7,$
+ 	DJNZ	R6,D2
+  	DJNZ	R5,D1
+   	RET
+	END

samples/Ceylon/Foo.ceylon

@@ -1,13 +1,13 @@
-doc "Test function for Ceylon"
+"Test function for Ceylon"
-by  "Enrique"
+by ("Enrique")
 shared void test() {
     print("test");
 }
 
-doc "Test class for Ceylon"
+"Test class for Ceylon"
 shared class Test(name) satisfies Comparable<Test> {
     shared String name;
-    shared actual String string = "Test " name ".";
+    shared actual String string = "Test ``name``.";
 
     shared actual Comparison compare(Test other) {
         return name<=>other.name;

samples/Clean/GenHylo.dcl

@@ -0,0 +1,11 @@
+definition module GenHylo
+
+import StdGeneric, GenMap
+
+:: Fix f = In (f .(Fix f))
+Out :: !u:(Fix v:a) -> v:(a w:(Fix v:a)), [u <= w]
+
+hylo :: ((.f .b) -> .b) (.a -> (.f .a)) -> (.a -> .b) | gMap{|*->*|} f
+cata :: (u:(f .a) -> .a) -> (Fix u:f) -> .a | gMap{|*->*|} f
+ana :: (.a -> u:(f .a)) -> .a -> (Fix u:f) | gMap{|*->*|} f
+

samples/Clean/GenMap.dcl

@@ -0,0 +1,9 @@
+definition module GenMap
+
+import StdGeneric
+
+generic gMap a b :: .a -> .b
+derive gMap c, UNIT, PAIR, EITHER, CONS, FIELD, OBJECT, {}, {!} 
+
+derive gMap [], (,), (,,),  (,,,), (,,,,), (,,,,,), (,,,,,,), (,,,,,,,)
+

samples/Clean/GenMap.icl

@@ -0,0 +1,19 @@
+implementation module GenMap
+
+import StdClass, StdArray, StdInt, StdFunc
+import StdGeneric, _Array
+
+generic gMap a b :: .a -> .b
+gMap{|c|} x 					= x
+gMap{|UNIT|} x 					= x
+gMap{|PAIR|} fx fy (PAIR x y) 	= PAIR (fx x) (fy y) 
+gMap{|EITHER|} fl fr (LEFT x) 	= LEFT (fl x)
+gMap{|EITHER|} fl fr (RIGHT x) 	= RIGHT (fr x)
+gMap{|CONS|} f (CONS x) 		= CONS (f x)
+gMap{|FIELD|} f (FIELD x) 		= FIELD (f x)
+gMap{|OBJECT|} f (OBJECT x) 	= OBJECT (f x)
+gMap{|{}|} f xs 				= mapArray f xs
+gMap{|{!}|} f xs				= mapArray f xs
+
+derive gMap [], (,), (,,),  (,,,), (,,,,), (,,,,,), (,,,,,,), (,,,,,,,)
+

samples/Clean/fsieve.icl

@@ -0,0 +1,54 @@
+module fsieve
+
+/*
+The Fast Sieve of Eratosthenes.
+
+A sequential and optimized version of the sieve of Eratosthenes.
+The program calculates a list of the first NrOfPrime primes.
+The result of the program is the NrOfPrimes'th prime.
+
+Strictness annotations have been added because the strictness analyser
+is not able to deduce all strictness information. Removal of these !'s
+will make the program about 20% slower.
+
+On a machine without a math coprocessor the execution of this
+program might take a (very) long time. Set NrOfPrimes to a smaller value.
+*/
+
+import StdClass; // RWS
+import StdInt, StdReal
+     
+NrOfPrimes :== 3000 
+	
+//	The sieve algorithm: generate an infinite list of all primes.
+
+Primes::[Int]
+Primes = pr where pr = [5 : Sieve 7 4 pr]
+
+Sieve::Int !Int [Int] -> [Int]
+Sieve g i prs
+	| IsPrime prs g (toInt (sqrt (toReal g)))	=  [g : Sieve` g i prs]
+												=  Sieve (g + i) (6 - i) prs
+
+Sieve`::Int Int [Int] -> [Int]
+Sieve` g i prs =  Sieve (g + i) (6 - i) prs
+
+IsPrime::[Int] !Int Int -> Bool
+IsPrime [f:r] pr bd | f>bd 			=  True
+					| pr rem f==0	=  False
+									=  IsPrime r pr bd
+								  
+//	Select is used to get the NrOfPrimes'th prime from the infinite list.
+
+Select::[x] Int -> x
+Select [f:r] 1 =  f
+Select [f:r] n =  Select r (n - 1)
+
+
+/*	The Start rule: Select the NrOfPrimes'th prime from the list of primes
+	generated by Primes.
+*/
+
+Start::Int
+Start = Select [2, 3 : Primes] NrOfPrimes
+

samples/Clean/sem.icl

@@ -0,0 +1,99 @@
+module monadicSemantics
+
+import StdEnv, StdGeneric, GenMap, GenHylo
+
+/* For fun I implemented the recursive datastructre Exp and Stm as fixpoints
+   This helps us define recursive functions on them (only a little bit though)
+   However deriving gMap for Fix did not works out of the box
+   I had to remove some uniqueness typing in GenMap and GenHylo */
+:: Op      = Plus | Minus | Times | Rem | Equal | LessThan
+:: Var     :== String
+
+:: ExpP a  = Int Int | Var Var | Op Op a a
+:: Exp     :== Fix ExpP
+
+:: StmP a  = Assign Var Exp | If Exp a a | While Exp a | Seq a a | Cont
+:: Stm     :== Fix StmP
+
+derive gMap ExpP, StmP, Fix
+
+// Environment. Semantics is basically Env -> Env
+:: Env :== Var -> Int
+:: Sem :== Env -> (Int, Env)
+empty = \v . 0
+
+// return
+rtn :: Int -> Sem
+rtn i = \e. (i, e)
+
+// the usual bind
+(>>=) infixl 1 :: Sem (Int->Sem) -> Sem
+(>>=) x y = \e. (\(i,e2).y i e2) (x e)
+(>>|) infixl 1 :: Sem Sem -> Sem
+(>>|) x y = x >>= \_. y
+
+// read variable from environment
+read :: Var -> Sem
+read v = \e. (e v, e)
+
+// assign value to give variable in environment
+write :: Var Int -> Sem
+write v i = \e. (i, \w. if (w==v) i (e w))
+
+// semantics
+class sem a :: a -> Sem
+
+operator :: Op -> Int -> Int -> Int
+operator Plus     = (+)
+operator Minus    = (-)
+operator Times    = (*)
+operator Rem      = rem
+operator Equal    = \x y . if (x==y) 1 0
+operator LessThan = \x y . if (x< y)  1 0
+
+// semantics of expressions
+instance sem Exp where
+	sem x = cata phi x where
+		phi (Int n)     = rtn n
+		phi (Var v)     = read v
+		phi (Op op x y) = x >>= \v1. y >>= return o (operator op v1)
+
+// semantics of statments
+// NOTE: while will always return 0, as it might not even be executed
+instance sem Stm where
+	sem x = cata phi x where
+		phi (Assign v e)     = sem e >>= write v
+		phi (If e s1 s2)     = sem e >>= \b . if (b<>0) s1 s2
+		phi stm=:(While e s) = sem e >>= \b . if (b<>0) (s >>| phi stm) (phi Cont)
+		phi (Seq s1 s2)      = s1 >>| s2    // Here the cata *finally* pays off :D
+		phi Cont             = rtn 0
+
+// convenience functions
+int    = In o Int
+var    = In o Var
+op o   = In o2 (Op o)
+assign = In o2 Assign
+ifte e = In o2 (If e)
+while  = In o2 While
+seq    = In o2 Seq
+cont   = In Cont
+
+// test case, also testing the new operator <
+pEuclides =
+	while (op LessThan (int 0) (var "b"))(
+		seq (assign "r" (op Rem (var "a") (var "b")))
+		(seq (assign "a" (var "b"))
+		( (assign "b" (var "r")))
+		)
+	)
+
+Start = fst (program start) where
+	program = sem pEuclides >>| read "a"
+	start "a" = 9
+	start "b" = 12
+	start _ = 0
+
+// Helper
+(o2) infixr 9
+(o2) f g x :== f o (g x)
+

samples/Clean/stack.dcl

@@ -0,0 +1,14 @@
+definition module stack
+
+:: Stack a
+
+newStack :: (Stack a)
+push :: a (Stack a) -> Stack a
+pushes :: [a] (Stack a) -> Stack a
+pop :: (Stack a) -> Stack a
+popn :: Int (Stack a) -> Stack a
+top :: (Stack a) -> a
+topn :: Int (Stack a) -> [a]
+elements :: (Stack a) -> [a]
+count :: (Stack a) -> Int
+

samples/Clean/stack.icl

@@ -0,0 +1,33 @@
+implementation module stack
+import StdEnv
+
+:: Stack a :== [a]
+
+newStack :: (Stack a)
+newStack = []
+
+push :: a (Stack a) -> Stack a
+push x s = [x:s]
+
+pushes :: [a] (Stack a) -> Stack a
+pushes x s = x ++ s
+
+pop :: (Stack a) -> Stack a
+pop [] = abort "Cannot use pop on an empty stack"
+pop [e:s] = s
+
+popn :: Int (Stack a) -> Stack a
+popn n s  = drop n s
+
+top :: (Stack a) -> a
+top [] = abort "Cannot use top on an empty stack"
+top [e:s] = e
+
+topn :: Int (Stack a) -> [a]
+topn n s = take n s
+elements :: (Stack a) -> [a]
+elements s = s
+
+count :: (Stack a) -> Int
+count s = length s
+

samples/Clean/streams.dcl

@@ -0,0 +1,16 @@
+definition module streams
+
+import StdEnv
+
+instance zero [Real]
+instance one [Real]
+instance + [Real]        
+instance - [Real]
+instance * [Real]
+instance / [Real]
+
+X :: [Real]
+invert :: [Real] -> [Real]
+pow :: [Real] Int -> [Real]
+(shuffle) infixl 7 :: [Real] [Real] -> [Real]
+

samples/Clean/streams.icl

@@ -0,0 +1,49 @@
+implementation module streams
+
+import StdEnv
+
+instance zero [Real]
+where
+        zero = [] //Infinite row of zeroes represented as empty list to ease computation
+
+instance one [Real]
+where
+        one = [1.0:zero]
+
+instance + [Real]
+where
+        (+) [s:s`] [t:t`] = [s+t:s`+t`]
+        (+) [s:s`] [] = [s:s`]
+        (+) [] [t:t`] = [t:t`]
+        (+) [] [] = []
+        
+instance - [Real]
+where
+        (-) [s:s`] [t:t`] = [s-t:s`-t`]
+        (-) [s:s`] [] = [s:s`]
+        (-) [] [t:t`] = [-1.0] * [t:t`]
+        (-) [] [] = []
+
+instance * [Real]
+where
+        (*) [s:s`] [t:t`] = [s*t:s`*[t:t`]+[s]*t`]
+        (*) _ _ = []
+
+instance / [Real]
+where
+        (/) s t = s * (invert t)
+
+X :: [Real]
+X = [0.0:one]
+
+invert :: [Real] -> [Real]
+invert [s:s`] = [1.0/s:(invert [s:s`]) * s` * [-1.0/s]]
+
+pow :: [Real] Int -> [Real]
+pow s 0 = one
+pow s n = s * pow s (n-1)
+
+(shuffle) infixl 7 :: [Real] [Real] -> [Real]
+(shuffle) [s:s`] [t:t`] = [s*t:s` shuffle [t:t`] + [s:s`] shuffle t`]
+(shuffle) _ _ = []
+

samples/Forth/bitmap.frt

@@ -0,0 +1,8 @@
+\ Bit arrays.
+: bits ( u1 -- u2 ) 7 + 3 rshift ;
+: bitmap ( u "name" -- ) create bits here over erase allot
+   does> ( u -- a x ) over 3 rshift +  1 rot 7 and lshift ;
+: bit@ ( a x -- f ) swap c@ and ;
+: 1bit ( a x -- ) over c@ or swap c! ;
+: 0bit ( a x -- ) invert over c@ and swap c! ;
+: bit! ( f a x -- ) rot if 1bit else 0bit then ;

samples/Forth/enum.frt

@@ -0,0 +1,7 @@
+\ Implements ENUM.
+
+\ Double DOES>!
+: enum   create 0 ,  does> create dup @ 1 rot +! ,  does> @ ;
+
+\ But this is simpler.
+: enum   create 0 ,  does> dup @ constant 1 swap +! ;

samples/Forth/macros.frt

@@ -0,0 +1,8 @@
+\ Simplifies compiling words.
+
+: [[     ; immediate
+: '<>    >in @ ' swap >in ! <> ;
+: (]])   begin dup '<> while postpone postpone repeat drop ;
+: ]]     ['] [[ (]]) ; immediate
+
+( Usage:   : foo ]] dup * [[ ; immediate   : bar 42 foo . ; )

samples/G-code/duettest.g

@@ -0,0 +1,57 @@
+; RepRapPro Ormerod
+; Board test GCodes
+M111 S1; Debug on
+G21 ; mm
+G90 ; Absolute positioning
+M83 ; Extrusion relative
+M906 X800 Y800 Z800 E800 ; Motor currents (mA)
+T0 ; Extruder 0
+G1 X50 F500
+G1 X0
+G4 P500
+G1 Y50 F500
+G1 Y0
+G4 P500
+G1 Z20 F200
+G1 Z0
+G4 P500
+G1 E20 F200
+G1 E-20
+G4 P500
+M106 S255
+G4 P500
+M106 S0
+G4 P500
+M105
+G10 P0 S100
+T0
+M140 S100
+G4 P5000
+M105
+G4 P5000
+M105
+G4 P5000
+M105
+G4 P5000
+M105
+G4 P5000
+M105
+G4 P5000
+M105
+G4 P5000
+M105
+G4 P5000
+M105
+G4 P5000
+M105
+G4 P5000
+M105
+G4 P5000
+M105
+G4 P5000
+M105
+M0
+
+
+
+

samples/G-code/square.g

@@ -0,0 +1,13 @@
+G28 X0 Y0
+G1 X55 Y5 F2000
+G1 Y180
+G1 X180
+G1 Y5
+G1 X55
+G1 Y180
+G1 X180
+G1 Y5
+G1 X55
+M0
+
+

samples/GDScript/example.gd

@@ -0,0 +1,57 @@
+# Taken from https://github.com/okamstudio/godot/wiki/gdscript
+# a file is a class!
+
+# inheritance
+
+extends BaseClass
+
+# member variables
+
+var a = 5 
+var s = "Hello"
+var arr = [1, 2, 3]
+var dict = {"key":"value", 2:3}
+
+# constants
+
+const answer = 42
+const thename = "Charly"
+
+# built-in vector types
+
+var v2 = Vector2(1, 2)
+var v3 = Vector3(1, 2, 3)
+
+# function
+
+func some_function(param1, param2):
+    var local_var = 5
+
+    if param1 < local_var:
+        print(param1)
+    elif param2 > 5:
+        print(param2)
+    else:
+        print("fail!")
+
+    for i in range(20):
+        print(i)
+
+    while(param2 != 0):
+        param2 -= 1
+
+    var local_var2 = param1+3
+    return local_var2
+
+
+# subclass
+
+class Something:
+    var a = 10
+
+# constructor
+
+func _init():
+    print("constructed!")
+    var lv = Something.new()
+    print(lv.a)

samples/GDScript/grid.gd

@@ -0,0 +1,216 @@
+
+
+extends Control
+
+# Simple Tetris-like demo, (c) 2012 Juan Linietsky
+# Implemented by using a regular Control and drawing on it during the _draw() callback.
+# The drawing surface is updated only when changes happen (by calling update())
+
+
+var score = 0
+var score_label=null
+
+const MAX_SHAPES = 7
+
+var block = preload("block.png")
+
+var block_colors=[
+	Color(1,0.5,0.5),
+	Color(0.5,1,0.5),
+	Color(0.5,0.5,1),
+	Color(0.8,0.4,0.8),
+	Color(0.8,0.8,0.4),
+	Color(0.4,0.8,0.8),
+	Color(0.7,0.7,0.7)]
+
+var	block_shapes=[
+	[ Vector2(0,-1),Vector2(0,0),Vector2(0,1),Vector2(0,2) ], # I
+	[ Vector2(0,0),Vector2(1,0),Vector2(1,1),Vector2(0,1) ], # O
+	[ Vector2(-1,1),Vector2(0,1),Vector2(0,0),Vector2(1,0) ], # S
+	[ Vector2(1,1),Vector2(0,1),Vector2(0,0),Vector2(-1,0) ], # Z
+	[ Vector2(-1,1),Vector2(-1,0),Vector2(0,0),Vector2(1,0) ], # L
+	[ Vector2(1,1),Vector2(1,0),Vector2(0,0),Vector2(-1,0) ], # J
+	[ Vector2(0,1),Vector2(1,0),Vector2(0,0),Vector2(-1,0) ]] # T
+	
+
+var block_rotations=[
+	Matrix32( Vector2(1,0),Vector2(0,1), Vector2() ),
+	Matrix32( Vector2(0,1),Vector2(-1,0), Vector2() ),
+	Matrix32( Vector2(-1,0),Vector2(0,-1), Vector2() ),
+	Matrix32( Vector2(0,-1),Vector2(1,0), Vector2() )
+]
+	
+
+var width=0
+var height=0
+
+var cells={}
+
+var piece_active=false
+var piece_shape=0
+var piece_pos=Vector2()
+var piece_rot=0
+
+
+func piece_cell_xform(p,er=0):
+	var r = (4+er+piece_rot)%4
+	return piece_pos+block_rotations[r].xform(p)
+
+func _draw():
+
+	var sb = get_stylebox("bg","Tree") # use line edit bg
+	draw_style_box(sb,Rect2(Vector2(),get_size()).grow(3))
+	
+	var bs = block.get_size()
+	for y in range(height):
+		for x in range(width):
+			if (Vector2(x,y) in cells):
+				draw_texture_rect(block,Rect2(Vector2(x,y)*bs,bs),false,block_colors[cells[Vector2(x,y)]])
+				
+	if (piece_active):
+		
+		for c in block_shapes[piece_shape]:
+			draw_texture_rect(block,Rect2(piece_cell_xform(c)*bs,bs),false,block_colors[piece_shape])
+			
+
+func piece_check_fit(ofs,er=0):
+
+	for c in block_shapes[piece_shape]:
+		var pos = piece_cell_xform(c,er)+ofs
+		if (pos.x < 0):
+			return false
+		if (pos.y < 0):
+			return false
+		if (pos.x >= width):
+			return false
+		if (pos.y >= height):
+			return false
+		if (pos in cells):
+			return false
+	
+	return true	
+
+func new_piece():
+
+	piece_shape = randi() % MAX_SHAPES	
+	piece_pos = Vector2(width/2,0)
+	piece_active=true
+	piece_rot=0
+	if (piece_shape==0):
+		piece_pos.y+=1
+		
+	if (not piece_check_fit(Vector2())):
+		#game over
+		#print("GAME OVER!")
+		game_over()
+		
+	update()
+		
+	
+func test_collapse_rows():
+	var accum_down=0
+	for i in range(height):
+		var y = height - i - 1
+		var collapse = true
+		for x in range(width):
+			if (Vector2(x,y) in cells):
+				if (accum_down):
+					cells[ Vector2(x,y+accum_down) ] = cells[Vector2(x,y)]
+			else:
+				collapse=false
+				if (accum_down):
+					cells.erase( Vector2(x,y+accum_down) )
+						
+		if (collapse):
+			accum_down+=1
+		
+			
+	score+=accum_down*100
+	score_label.set_text(str(score))
+			
+		
+func game_over():
+
+		piece_active=false
+		get_node("gameover").set_text("Game Over")		
+		update()
+				
+		
+func restart_pressed():
+
+		score=0
+		score_label.set_text("0")
+		cells.clear()
+		get_node("gameover").set_text("")		
+		piece_active=true
+		update()
+		
+		
+
+func piece_move_down():
+
+	if (!piece_active):
+		return
+	if (piece_check_fit(Vector2(0,1))):
+		piece_pos.y+=1
+		update()		
+	else:
+
+		for c in block_shapes[piece_shape]:
+			var pos = piece_cell_xform(c)
+			cells[pos]=piece_shape
+		test_collapse_rows()
+		new_piece()
+		
+
+func piece_rotate():
+
+	var adv = 1
+	if (not piece_check_fit(Vector2(),1)):
+		return
+	piece_rot = (piece_rot + adv) % 4
+	update()
+	
+	
+
+func _input(ie):
+
+
+	if (not piece_active):
+		return
+	if (!ie.is_pressed()):
+		return
+
+	if (ie.is_action("move_left")):
+		if (piece_check_fit(Vector2(-1,0))):
+			piece_pos.x-=1
+			update()
+	elif (ie.is_action("move_right")):
+		if (piece_check_fit(Vector2(1,0))):
+			piece_pos.x+=1
+			update()
+	elif (ie.is_action("move_down")):
+		piece_move_down()
+	elif (ie.is_action("rotate")):
+		piece_rotate()
+		
+		
+func setup(w,h):
+	width=w
+	height=h
+	set_size( Vector2(w,h)*block.get_size() )
+	new_piece()
+	get_node("timer").start()
+	
+
+func _ready():
+	# Initalization here
+
+	setup(10,20)
+	score_label = get_node("../score")
+
+	set_process_input(true)
+
+
+
+

samples/GDScript/player.gd

@@ -0,0 +1,243 @@
+
+extends RigidBody
+
+# member variables here, example:
+# var a=2
+# var b="textvar"
+
+#var dir=Vector3()
+
+const ANIM_FLOOR = 0
+const ANIM_AIR_UP = 1
+const ANIM_AIR_DOWN = 2
+
+const SHOOT_TIME = 1.5
+const SHOOT_SCALE = 2
+
+const CHAR_SCALE = Vector3(0.3,0.3,0.3)
+
+var facing_dir = Vector3(1, 0, 0)
+var movement_dir = Vector3()
+
+var jumping=false
+
+var turn_speed=40
+var keep_jump_inertia = true
+var air_idle_deaccel = false
+var accel=19.0
+var deaccel=14.0
+var sharp_turn_threshhold = 140
+
+var max_speed=3.1
+var on_floor = false
+
+var prev_shoot = false
+
+var last_floor_velocity = Vector3()
+
+var shoot_blend = 0
+
+func adjust_facing(p_facing, p_target,p_step, p_adjust_rate,current_gn):
+
+	var n = p_target # normal
+	var t = n.cross(current_gn).normalized()
+	
+	var x = n.dot(p_facing)
+	var y = t.dot(p_facing)
+	
+	var ang = atan2(y,x)
+	
+	if (abs(ang)<0.001): # too small
+		return p_facing
+	
+	var s = sign(ang)
+	ang = ang * s
+	var turn = ang * p_adjust_rate * p_step
+	var a
+	if (ang<turn):
+		a=ang
+	else:
+		a=turn
+	ang = (ang - a) * s
+	
+	return ((n * cos(ang)) + (t * sin(ang))) * p_facing.length()
+
+
+
+func _integrate_forces( state ):
+
+	var lv = state.get_linear_velocity() # linear velocity
+	var g = state.get_total_gravity()
+	var delta = state.get_step()
+	var d = 1.0 - delta*state.get_total_density()
+	if (d<0):
+		d=0
+	lv += g * delta #apply gravity
+
+	var anim = ANIM_FLOOR
+
+	var up = -g.normalized() # (up is against gravity)
+	var vv = up.dot(lv) # vertical velocity
+	var hv = lv - (up*vv) # horizontal velocity
+
+
+
+	var hdir = hv.normalized() # horizontal direction
+	var hspeed = hv.length()	#horizontal speed
+
+	var floor_velocity
+	var onfloor = false
+
+	if (state.get_contact_count() == 0):
+		floor_velocity = last_floor_velocity
+	else:
+		for i in range(state.get_contact_count()):
+			if (state.get_contact_local_shape(i) != 1):
+				continue
+			
+			onfloor = true
+			floor_velocity = state.get_contact_collider_velocity_at_pos(i)
+			break
+		
+
+	var dir = Vector3() #where does the player intend to walk to
+	var cam_xform = get_node("target/camera").get_global_transform()
+	
+	if (Input.is_action_pressed("move_forward")):
+		dir+=-cam_xform.basis[2] 
+	if (Input.is_action_pressed("move_backwards")):
+		dir+=cam_xform.basis[2] 
+	if (Input.is_action_pressed("move_left")):
+		dir+=-cam_xform.basis[0] 
+	if (Input.is_action_pressed("move_right")):
+		dir+=cam_xform.basis[0] 
+		
+	var jump_attempt = Input.is_action_pressed("jump")
+	var shoot_attempt = Input.is_action_pressed("shoot")
+		
+	var target_dir = (dir - up*dir.dot(up)).normalized()
+	
+	if (onfloor):
+
+		var sharp_turn = hspeed > 0.1 and rad2deg(acos(target_dir.dot(hdir))) > sharp_turn_threshhold
+
+		if (dir.length()>0.1 and !sharp_turn) :
+			if (hspeed > 0.001) :
+
+				#linear_dir = linear_h_velocity/linear_vel
+				#if (linear_vel > brake_velocity_limit and linear_dir.dot(ctarget_dir)<-cos(Math::deg2rad(brake_angular_limit)))
+				#	brake=true
+				#else
+				hdir = adjust_facing(hdir,target_dir,delta,1.0/hspeed*turn_speed,up)
+				facing_dir = hdir
+			else:
+
+				hdir = target_dir
+			
+			if (hspeed<max_speed):
+				hspeed+=accel*delta
+
+		else:
+			hspeed-=deaccel*delta
+			if (hspeed<0):
+				hspeed=0
+		
+		hv = hdir*hspeed
+		
+		var mesh_xform = get_node("Armature").get_transform() 
+		var facing_mesh=-mesh_xform.basis[0].normalized()
+		facing_mesh = (facing_mesh - up*facing_mesh.dot(up)).normalized()
+		facing_mesh = adjust_facing(facing_mesh,target_dir,delta,1.0/hspeed*turn_speed,up)
+		var m3 = Matrix3(-facing_mesh,up,-facing_mesh.cross(up).normalized()).scaled( CHAR_SCALE )
+		
+		get_node("Armature").set_transform(Transform(m3,mesh_xform.origin))
+				
+		if (not jumping and jump_attempt):
+			vv = 7.0
+			jumping = true		
+			get_node("sfx").play("jump")
+	else:
+
+		if (vv>0):
+			anim=ANIM_AIR_UP
+		else:
+			anim=ANIM_AIR_DOWN
+			
+		var hs
+		if (dir.length()>0.1):
+
+			hv += target_dir * (accel * 0.2) * delta
+			if (hv.length() > max_speed):
+				hv = hv.normalized() * max_speed
+
+		else:
+
+			if (air_idle_deaccel):
+				hspeed = hspeed - (deaccel * 0.2) * delta
+				if (hspeed<0):
+					hspeed=0
+
+				hv = hdir*hspeed
+			
+		
+	if (jumping and vv < 0):
+		jumping=false
+
+	lv = hv+up*vv
+	
+	
+
+	if (onfloor):
+
+		movement_dir = lv
+		#lv += floor_velocity
+		last_floor_velocity = floor_velocity
+	else:
+
+		if (on_floor) :
+
+			#if (keep_jump_inertia):
+			#	lv += last_floor_velocity
+			pass
+		
+		last_floor_velocity = Vector3()
+		movement_dir = lv
+	
+	on_floor = onfloor
+
+	state.set_linear_velocity(lv)
+	
+	if (shoot_blend>0):
+		shoot_blend -= delta * SHOOT_SCALE
+		if (shoot_blend<0):
+			shoot_blend=0
+	
+	if (shoot_attempt and not prev_shoot):
+		shoot_blend = SHOOT_TIME		
+		var bullet = preload("res://bullet.scn").instance()
+		bullet.set_transform( get_node("Armature/bullet").get_global_transform().orthonormalized() )
+		get_parent().add_child( bullet )
+		bullet.set_linear_velocity( get_node("Armature/bullet").get_global_transform().basis[2].normalized() * 20 )
+		PS.body_add_collision_exception( bullet.get_rid(), get_rid() ) #add it to bullet
+		get_node("sfx").play("shoot")
+		
+	prev_shoot = shoot_attempt
+	
+	if (onfloor):
+		get_node("AnimationTreePlayer").blend2_node_set_amount("walk",hspeed / max_speed)
+		
+	get_node("AnimationTreePlayer").transition_node_set_current("state",anim)
+	get_node("AnimationTreePlayer").blend2_node_set_amount("gun",min(shoot_blend,1.0))
+#	state.set_angular_velocity(Vector3())	
+	
+	
+
+
+func _ready():
+
+
+	# Initalization here
+	get_node("AnimationTreePlayer").set_active(true)
+	pass
+
+

samples/GDScript/pong.gd

@@ -0,0 +1,73 @@
+
+extends Node2D
+
+# member variables here, example:
+# var a=2
+# var b="textvar"
+const INITIAL_BALL_SPEED = 80
+var ball_speed = INITIAL_BALL_SPEED
+var screen_size = Vector2(640,400)
+#default ball direction
+var direction = Vector2(-1,0)
+var pad_size = Vector2(8,32)
+const PAD_SPEED = 150
+
+
+func _process(delta):
+
+
+	# get ball positio and pad rectangles
+	var ball_pos = get_node("ball").get_pos()
+	var left_rect = Rect2( get_node("left").get_pos() - pad_size*0.5, pad_size )
+	var right_rect = Rect2( get_node("right").get_pos() - pad_size*0.5, pad_size )
+	
+	#integrate new ball postion
+	ball_pos+=direction*ball_speed*delta
+	
+	#flip when touching roof or floor
+	if ( (ball_pos.y<0 and direction.y <0) or (ball_pos.y>screen_size.y and direction.y>0)):
+		direction.y = -direction.y
+		
+	#flip, change direction and increase speed when touching pads	
+	if ( (left_rect.has_point(ball_pos) and direction.x < 0) or (right_rect.has_point(ball_pos) and direction.x > 0)):
+		direction.x=-direction.x
+		ball_speed*=1.1
+		direction.y=randf()*2.0-1
+		direction = direction.normalized()
+
+	#check gameover
+	if (ball_pos.x<0 or ball_pos.x>screen_size.x):
+		ball_pos=screen_size*0.5
+		ball_speed=INITIAL_BALL_SPEED
+		direction=Vector2(-1,0)
+			
+						
+	get_node("ball").set_pos(ball_pos)
+
+	#move left pad	
+	var left_pos = get_node("left").get_pos()
+	
+	if (left_pos.y > 0 and Input.is_action_pressed("left_move_up")):
+		left_pos.y+=-PAD_SPEED*delta
+	if (left_pos.y < screen_size.y and Input.is_action_pressed("left_move_down")):
+		left_pos.y+=PAD_SPEED*delta
+		
+	get_node("left").set_pos(left_pos)
+		
+	#move right pad	
+	var right_pos = get_node("right").get_pos()
+	
+	if (right_pos.y > 0 and Input.is_action_pressed("right_move_up")):
+		right_pos.y+=-PAD_SPEED*delta
+	if (right_pos.y < screen_size.y and Input.is_action_pressed("right_move_down")):
+		right_pos.y+=PAD_SPEED*delta
+		
+	get_node("right").set_pos(right_pos)
+	
+	 
+
+func _ready():
+	screen_size = get_viewport_rect().size # get actual size
+	pad_size = get_node("left").get_texture().get_size()
+	set_process(true)
+

samples/Graph Modeling Language/sample.gml

@@ -0,0 +1,21 @@
+graph
+[
+  directed 0
+  node
+  [
+    id 0
+    label "Node 1"
+    value 100
+  ]
+  node
+  [
+    id 1
+    label "Node 2"
+    value 200
+  ]
+  edge
+  [
+    source 1
+    target 0
+  ]
+]

samples/Groff/sample.4

@@ -0,0 +1,13 @@
+.TH FOO 1
+.SH NAME
+foo \- bar
+.SH SYNOPSIS
+.B foo
+.I bar
+.SH DESCRIPTION
+Foo bar
+.BR baz
+quux.
+.PP
+.B Foo
+bar baz.

samples/JavaScript/jsbuild.jsb

@@ -0,0 +1,12 @@
+jsb.library('mylibrary', jsb.STATIC_LIBRARY, function(libObject) {
+    libObject.outputName = 'mylibrary';
+    libObject.cflags = [ '-Wall' ];
+    libObject.ldflags = [ '-pthread' ];
+    libObject.includePaths = [ 'src/include' ];
+    libObject.sources = [ 
+        'src/main.cpp',
+        'src/app.cpp'
+    ];
+});
+
+jsb.build();

samples/LSL/LSL.lsl

@@ -0,0 +1,74 @@
+/*
+    Testing syntax highlighting
+    for the Linden Scripting Language
+*/
+
+integer someIntNormal       = 3672;
+integer someIntHex          = 0x00000000;
+integer someIntMath         = PI_BY_TWO;
+
+integer event               = 5673;// 'event' is invalid.illegal
+
+key someKeyTexture          = TEXTURE_DEFAULT;
+string someStringSpecial    = EOF;
+
+some_user_defined_function_without_return_type(string inputAsString)
+{
+    llSay(PUBLIC_CHANNEL, inputAsString);
+}
+
+string user_defined_function_returning_a_string(key inputAsKey)
+{
+    return (string)inputAsKey;
+}
+
+default
+{
+    state_entry()
+    {
+        key someKey = NULL_KEY;
+        someKey = llGetOwner();
+
+        string someString = user_defined_function_returning_a_string(someKey);
+
+        some_user_defined_function_without_return_type(someString);
+    }
+
+    touch_start(integer num_detected)
+    {
+        list agentsInRegion = llGetAgentList(AGENT_LIST_REGION, []);
+        integer numOfAgents = llGetListLength(agentsInRegion);
+
+        integer index;                                                          // defaults to 0
+        for (; index <= numOfAgents - 1; index++)                               // for each agent in region
+        {
+            llRegionSayTo(llList2Key(agentsInRegion, index), PUBLIC_CHANNEL, "Hello, Avatar!");
+        }
+    }
+
+    touch_end(integer num_detected)
+    {
+        someIntNormal       = 3672;
+        someIntHex          = 0x00000000;
+        someIntMath         = PI_BY_TWO;
+
+        event               = 5673;// 'event' is invalid.illegal
+
+        someKeyTexture      = TEXTURE_DEFAULT;
+        someStringSpecial   = EOF;
+
+        llSetInventoryPermMask("some item", MASK_NEXT, PERM_ALL);// 'llSetInventoryPermMask' is reserved.godmode
+
+        llWhisper(PUBLIC_CHANNEL, "Leaving \"default\" now...");
+        state other;
+    }
+}
+
+state other
+{
+    state_entry()
+    {
+        llWhisper(PUBLIC_CHANNEL, "Entered \"state other\", returning to \"default\" again...");
+        state default;
+    }
+}

samples/Lua/wsapi.fcgi

@@ -0,0 +1,28 @@
+#!/usr/bin/lua
+
+-- Generic WSAPI FastCGI launcher, extracts application to launch
+-- from SCRIPT_FILENAME/PATH_TRANSLATED, each application (defined
+-- by its script entry point) gets an isolated Lua VM; sequential
+-- requests to the same application go to the same VM
+
+pcall(require,"luarocks.require")
+
+local common = require "wsapi.common"
+local fastcgi = require "wsapi.fastcgi"
+
+local ONE_HOUR = 60 * 60
+local ONE_DAY = 24 * ONE_HOUR
+
+local wsapi_loader = common.make_loader{
+  isolated = true,         -- isolate each script in its own Lua state
+  filename = nil,          -- if you want to force the launch of a single script
+  launcher = "wsapi.fcgi", -- the name of this script
+  reload = false,          -- if you want to reload the application on every request
+  period = ONE_HOUR,       -- frequency of Lua state staleness checks
+  ttl = ONE_DAY,           -- time-to-live for Lua states
+  vars =                   -- order of checking for the path of the script
+   { "SCRIPT_FILENAME",
+     "PATH_TRANSLATED" } 
+}
+
+fastcgi.run(wsapi_loader)

samples/PHP/prefix.fcgi

@@ -0,0 +1,3 @@
+<?php
+echo $_SERVER[$_GET["var"]];
+?>

samples/Pascal/custforms.pp

@@ -0,0 +1,193 @@
+unit custforms;
+
+{$mode objfpc}{$H+}
+
+interface
+
+uses
+  Classes, SysUtils, Forms;
+
+Type
+
+  { TCustomFormDescr }
+
+  TCustomFormDescr = Class
+  private
+    FAuthor: String;
+    FCaption: String;
+    FCategory: String;
+    FDescription: String;
+    FFormClass: TFormClass;
+    FLazPackage: String;
+    FUnitName: String;
+  public
+    Constructor Create(AFormClass : TFormClass; const APackage: string);
+    Constructor Create(AFormClass : TFormClass; Const ACaption,ADescription,AUnit,APackage : String);
+    Property FormClass : TFormClass Read FFormClass Write FFormClass;
+    Property Caption : String Read FCaption Write FCaption;
+    Property Description : String Read FDescription Write FDescription;
+    Property UnitName : String Read FUnitName Write FUnitName;
+    Property Category : String Read FCategory Write FCategory;
+    Property Author : String Read FAuthor Write FAuthor;
+    Property LazPackage : String Read FLazPackage Write FLazPackage;
+  end;
+
+Procedure RegisterCustomForm(Descr : TCustomFormDescr);
+Procedure RegisterCustomForm(AFormClass : TFormClass; const APackage: string);
+Procedure RegisterCustomForm(AFormClass : TFormClass; Const AUnitName, APackage : String);
+
+Procedure Register;
+
+implementation
+
+uses ProjectIntf,NewItemIntf,contnrs;
+
+Const
+  SAppFrameWork = 'Custom forms';
+  SInstanceOf   = 'Create a new instance of %s';
+
+{ TCustomFormDescr }
+
+constructor TCustomFormDescr.Create(AFormClass: TFormClass;
+  const APackage: string);
+
+Var
+  N,U : String;
+
+begin
+  N:=AFormClass.ClassName;
+  U:=N;
+  If (Upcase(U[1])='T') then
+    Delete(U,1,1);
+  Create(AFormClass,N,Format(SInstanceOf,[N]),U,APackage);
+end;
+
+constructor TCustomFormDescr.Create(AFormClass: TFormClass;
+  const ACaption, ADescription, AUnit, APackage: String);
+begin
+  FFormClass:=AFormClass;
+  FCaption:=ACaption;
+  FDescription:=ADescription;
+  FUnitName:=AUnit;
+  FCategory:=SAppFrameWork;
+  FLazPackage:=APackage;
+end;
+
+// Registration code.
+
+Type
+  { TCustomFormFileDescriptor }
+  TCustomFormFileDescriptor = Class(TFileDescPascalUnitWithResource)
+  private
+    FFormDescr: TCustomFormDescr;
+  Public
+    Constructor Create(ADescr : TCustomFormDescr);
+    Property FormDescr : TCustomFormDescr Read FFormDescr;
+    Function GetLocalizedName : String; override;
+    Function GetLocalizedDescription : String; override;
+    Function GetInterfaceUsesSection : String; override;
+  end;
+
+{ TCustomFormFileDescriptor }
+
+constructor TCustomFormFileDescriptor.Create(ADescr: TCustomFormDescr);
+begin
+  Inherited Create;
+  FFormDescr:=ADescr;
+  ResourceClass:=FFormDescr.FFormClass;
+  Name:=FFormDescr.Caption;
+  RequiredPackages:=ADescr.LazPackage;
+  //Writeln('TCustomFormFileDescriptor.Create RequiredPackages=',RequiredPackages);
+end;
+
+function TCustomFormFileDescriptor.GetLocalizedName: String;
+begin
+  Result:=FFormDescr.Caption;
+end;
+
+function TCustomFormFileDescriptor.GetLocalizedDescription: String;
+begin
+  Result:=FFormDescr.Description;
+  If (FFormDescr.Author<>'') then
+    Result:=Result+LineEnding+'By '+FFormDescr.Author;
+end;
+
+function TCustomFormFileDescriptor.GetInterfaceUsesSection: String;
+begin
+  Result:=inherited GetInterfaceUsesSection;
+  Result:=Result+',Forms,'+FFormDescr.UnitName;
+end;
+
+Var
+  CustomFormList : TObjectList;
+
+Procedure RegisterCustomForm(Descr : TCustomFormDescr);
+
+begin
+  CustomFormList.Add(Descr);
+end;
+
+Procedure RegisterCustomForm(AFormClass : TFormClass; const APackage: string);
+
+begin
+  RegisterCustomForm(TCustomFormDescr.Create(AFormClass,APackage));
+end;
+
+Procedure RegisterCustomForm(AFormClass : TFormClass; Const AUnitName, APackage : String);
+
+Var
+  D : TCustomFormDescr;
+
+begin
+  D:=TCustomFormDescr.Create(AFormClass,APackage);
+  D.UnitName:=AUnitName;
+  RegisterCustomForm(D);
+end;
+
+
+Procedure Register;
+
+Var
+  L : TStringList;
+  I : Integer;
+  D : TCustomFormDescr;
+  
+begin
+  L:=TStringList.Create;
+  Try
+    L.Sorted:=True;
+    L.Duplicates:=dupIgnore;
+    For I:=0 to CustomFormList.Count-1 do
+      L.Add(TCustomFormDescr(CustomFormList[i]).Category);
+    For I:=0 to L.Count-1 do
+    begin
+      RegisterNewItemCategory(TNewIDEItemCategory.Create(L[i]));
+    end;
+  Finally
+    L.Free;
+  end;
+  For I:=0 to CustomFormList.Count-1 do
+    begin
+    D:=TCustomFormDescr(CustomFormList[i]);
+    RegisterProjectFileDescriptor(TCustomFormFileDescriptor.Create(D),D.Category);
+    end;
+end;
+
+Procedure InitCustomForms;
+
+begin
+  CustomFormList:=TObjectList.Create;
+end;
+
+Procedure DoneCustomForms;
+
+begin
+  FreeAndNil(CustomFormList);
+end;
+
+Initialization
+  InitCustomForms;
+Finalization
+  DoneCustomForms;
+end.
+

samples/Pascal/gtkextra.pp

@@ -0,0 +1,51 @@
+{ $Id$ }
+{
+ ---------------------------------------------------------------------------
+ gtkextra.pp  -  GTK(2) widgetset - additional gdk/gtk functions
+ ---------------------------------------------------------------------------
+
+ This unit contains missing gdk/gtk functions and defines for certain 
+ versions of gtk or fpc.
+
+ ---------------------------------------------------------------------------
+
+ @created(Sun Jan 28th WET 2006)
+ @lastmod($Date$)
+ @author(Marc Weustink <marc@@dommelstein.nl>)
+
+ *****************************************************************************
+  This file is part of the Lazarus Component Library (LCL)
+
+  See the file COPYING.modifiedLGPL.txt, included in this distribution,
+  for details about the license.
+ *****************************************************************************
+ }
+
+unit GtkExtra;
+
+{$mode objfpc}{$H+}
+
+interface
+
+{$I gtkdefines.inc}
+
+{$ifdef gtk1}
+{$I gtk1extrah.inc}
+{$endif}
+
+{$ifdef gtk2}
+{$I gtk2extrah.inc}
+{$endif}
+
+
+implementation
+
+{$ifdef gtk1}
+{$I gtk1extra.inc}
+{$endif}
+
+{$ifdef gtk2}
+{$I gtk2extra.inc}
+{$endif}
+
+end.

samples/Prolog/dleak-report.script!

@@ -0,0 +1,11 @@
+#!/usr/bin/env swipl
+
+:- set_prolog_flag(verbose, silent).
+:- use_module(dleak).
+
+:- initialization
+	main, halt.
+
+main :-
+	current_prolog_flag(argv, [File]),
+	dleak(File).

samples/Prolog/ex6.pl

@@ -0,0 +1,5 @@
+%6.8
+subset(Set, Subset) :-
+	append(L1, Subset, Set).
+powerset(Set, Subset) :-
+	bagof(Subset, subset(Set, Subset), Subset).

samples/Prolog/logic-problem.pro

@@ -0,0 +1,68 @@
+/**
+ * Question 1.1
+ * combiner(+Buddies, -Pairs)
+ */
+combiner([], []).
+combiner([First|Buddies], Pairs):-
+	make_pairs(First, Buddies, Pairs1),
+	combiner(Buddies, Pairs2),
+	concat(Pairs1, Pairs2, Pairs).
+
+/**
+ * make_pairs(+Buddy, +Buddies, -Pairs)
+ */
+make_pairs(Buddy, [], []).
+make_pairs(Buddy, [First|Buddies], [(Buddy, First)|Pairs]):-
+	make_pairs(Buddy, Buddies, Pairs).
+
+/**
+ * concat(+X, +Y, ?T)
+ */
+concat([], Y, Y).
+concat([P|R], Y, [P|T]):-
+	concat(R, Y, T).
+
+
+/**
+ * Question 1.2
+ * extraire(+AllPossiblePairs, +NbPairs, -Tp, -RemainingPairs)
+ */
+extraire(AllPossiblePairs, 0, [], AllPossiblePairs).
+extraire([PossiblePair|AllPossiblePairs], NbPairs, [PossiblePair|Tp], NewRemainingPairs):-
+	NbPairs > 0,
+	NewNbPairs is NbPairs - 1,
+	extraire(AllPossiblePairs, NewNbPairs, Tp, RemainingPairs),
+	not(pair_in_array(PossiblePair, Tp)),
+	delete_pair(RemainingPairs, PossiblePair, NewRemainingPairs).
+extraire([PossiblePair|AllPossiblePairs], NbPairs, Tp, [PossiblePair|RemainingPairs]):-
+	NbPairs > 0,
+	extraire(AllPossiblePairs, NbPairs, Tp, RemainingPairs),
+	pair_in_array(PossiblePair, Tp).
+
+/**
+ * delete_pair(+Pairs, +Pair, -PairsWithoutPair)
+ */
+delete_pair([], _, []).
+delete_pair([Pair|Pairs], Pair, Pairs):-!.
+delete_pair([FirstPair|Pairs], Pair, [FirstPair|PairsWithoutPair]):-
+	delete_pair(Pairs, Pair, PairsWithoutPair).
+
+/**
+ * pair_in_array(+Pair, +Pairs)
+ */
+pair_in_array((A, B), [(C, D)|Pairs]):-
+	(A == C ; B == D ; A == D ; B == C),
+	!.
+pair_in_array(Pair, [FirstPair|Pairs]):-
+	pair_in_array(Pair, Pairs).
+
+
+/**
+ * Question 1.3
+ * les_tps(+Buddies, -Tps)
+ */
+les_tps(Buddies, Tps):-
+	combiner(Buddies, PossiblePairs),
+	length(Buddies, NbBuddies),
+	NbPairs is integer(NbBuddies / 2),
+	findall(Tp, extraire(PossiblePairs, NbPairs, Tp, _), Tps).

samples/Puppet/expiringhost.pp

@@ -0,0 +1,26 @@
+define example::expiringhost($ip, $timestamp) {
+
+  # Calculate the age of this resource by comparing 'now' against $timestamp
+  $age = inline_template("<%= require 'time'; Time.now - Time.parse(timestamp) %>")
+
+  # Max age, in seconds.
+  $maxage = 60
+
+  if $age > $maxage {
+    $expired = true
+    notice("Expiring resource $class[$name] due to age > $maxage (actual: $age)")
+  } else {
+    $expired = false
+    notice("Found recently-active $class[$name] (age: $age)")
+  }
+
+  # I set target to a /tmp path so you can run this example as non-root.
+  # In production, you probabyl won't set target as it defaults to /etc/hosts
+  # (or wherever puppet thinks your platform wants it)
+  host {
+    $name:
+      ip => $ip,
+      target => "/tmp/expiring-hosts-example-output",
+      ensure => $expired ? { true => absent, false => present };
+  }
+}

samples/Puppet/stages-example.pp

@@ -0,0 +1,26 @@
+class foo {
+  notify {
+    "foo": ;
+  }
+}
+
+class bar {
+  notify {
+    "bar": ;
+  }
+}
+
+
+node default {
+  stage {
+    "one": ;
+    "two": ;
+  }
+
+  class {
+    "foo": stage => "one";
+    "bar": stage => "two";
+  }
+
+  Stage["one"] -> Stage["two"]
+}

samples/Puppet/unmanaged-notify-puppet25.pp

@@ -0,0 +1,22 @@
+# Manually manage /tmp/original
+# Each puppet run will copy it to /tmp/flag if there's a change and notify
+# the exec when it changes.
+# 
+# The idea here is you might need (in some case) to manually manage a file outside
+# of puppet (in this case, "/tmp/original"). Using this example, you can make puppet
+# signal other parts of your catalog based on changes to that file.
+
+file {
+  # This will, when different, copy /tmp/original to /tmp/flag and notify our
+  # exec.
+  "/tmp/flag":
+    source => "file:///tmp/original",
+    notify => Exec["hello world"];
+}
+
+exec {
+  "hello world":
+    command => "/bin/echo hello world",
+    refreshonly => true;
+}
+

samples/Python/action.cgi

@@ -0,0 +1,82 @@
+#!/usr/bin/python
+
+from model import Feed
+import session
+import datetime
+import sys
+
+argv = session.argv()
+
+feed = Feed.get(guid=argv[1])
+action = argv[2]
+
+if action == 'done':
+    when = feed.notify_interval * feed.notify_unit
+elif action == 'snooze':
+    if len(argv) > 3:
+        when = int(argv[3])
+    else:
+        when = 3600
+else:
+    print '''Status: 400 Bad request
+Content-type: text/html
+
+Unknown action %s''' % action
+    sys.exit(1)
+
+feed.notify_next = datetime.datetime.utcnow() + datetime.timedelta(seconds=when)
+feed.save()
+
+response = '''Content-type: text/html
+
+<html><head><title>Alarm reset</title>
+<link rel="stylesheet" href="{base_url}/style.css">
+</head>
+<body>
+
+<div class="container">
+<h1>Alarm reset</h1>
+<div>
+<p id="reset">Alarm "<span class="name">{name}</span>" has been reset. You won't be notified for another <span class="duration">{duration}</span>.</p>
+
+<p>Actions:</p>
+<ul>
+<li><a href="{edit_url}?feed={guid}">Edit this reminder</a></li>
+<li><a href="{edit_url}">Create another reminder</a></li>
+<li><a href="{base_url}">Visit the Reminder Me site</a></li>
+</ul>
+</div>
+</div>
+
+<p class="back"><a href=".">Reminder Me</a></p>
+
+</body></html>'''
+
+when_left = when
+duration_list = []
+for (label,period) in [('month',86400*365/12),
+                       ('week',86400*7),
+                       ('day',86400),
+                       ('hour',3600),
+                       ('minute',60),
+                       ('second',1)]:
+    if when == period:
+        duration_list = [label]
+        break
+    
+    val = when_left/period
+    if val:
+        duration_list.append("%d %s%s" % (
+            val,
+            label,
+            val > 1 and 's' or ''))
+        when_left -= val*period
+
+basedir=session.request_script_dir()
+
+print response.format(guid=feed.guid,
+                      name=feed.name,
+                      edit_url="%s/edit.cgi" % basedir,
+                      base_url=basedir,
+                      duration=', '.join(duration_list))
+                      

samples/Python/backstage.fcgi

@@ -0,0 +1,120 @@
+#!/usr/bin/env python
+
+import sqlite
+import urllib2
+import csv
+import cgi
+import simplejson
+import jsontemplate
+import time
+
+log = open('log.txt', 'a')
+
+def urldecode(query):
+   d = {}
+   a = query.split('&')
+   for s in a:
+      if s.find('='):
+         k,v = map(urllib2.unquote, s.split('='))
+         try:
+            d[k].append(v)
+         except KeyError:
+            d[k] = [v]
+ 
+   return d
+
+def load_table(uri, cur):
+    table = uri.split('/')[-1]
+    table = table.split('.')[0]
+
+    contents = urllib2.urlopen(uri)
+    fields = ""
+    for field in contents.readline().strip().split(','):
+      fields += field
+      fields += ","
+    fields = fields.rstrip(',')
+    
+    cur.execute("SELECT name FROM sqlite_master WHERE type='table' \
+      AND name='%s';" % (table))
+    if cur.fetchone() == None:
+#      cur.execute("DROP TABLE %s;" % (table))
+      cur.execute("CREATE TABLE %s (%s);" % (table, fields))
+      for line in contents:
+        values = line.strip()
+        values = "','".join([val.strip() for val in values.split(",")])
+        values = "'" + values + "'"
+        sql = "INSERT INTO %s (%s) VALUES (%s);" % (table, fields, values)
+        cur.execute(sql)
+    return table
+
+def build_structure(headings, allresults):
+  results = []
+  for result in allresults:
+    results.append(dict(zip(headings, result)))
+  results = { "query" : results }
+  return results
+
+def build_json(headings, allresults, callback):
+  results = build_structure(headings, allresults)
+  return_str = simplejson.dumps(results)
+  if callback != None:
+    return_str = callback + "(" + return_str + ");";
+  return return_str
+
+def load_template(templatefile):
+  return "".join(urllib2.urlopen(templatefile).readlines())
+
+def build_template(headings, allresults, template_str):
+  results = build_structure(headings, allresults)
+  return jsontemplate.expand(template_str, results)
+  return ""
+
+def myapp(environ, start_response):
+    args = cgi.parse_qs(environ['QUERY_STRING'])
+    
+    query = args['query'][0]
+    uri = args['uri'][0]
+    callback = None
+    if 'callback' in args:
+      callback = args['callback'][0]
+    label = "no label"
+    if 'label' in args:
+      label = args['label'][0]
+    templatefile = None
+    if 'templatefile' in args:
+      templatefile = args['templatefile'][0]
+
+    con = sqlite.connect('mydatabase.db')
+    cur = con.cursor()
+    table_uris = uri.split(',')
+    tables = [load_table(uri, cur) for uri in table_uris]
+    con.commit() 
+    before = time.time()
+    cur.execute(query)
+    allresults = cur.fetchall()
+    after = time.time()
+    log.write("%s: query time %f\n" % (label, after - before))
+
+    headings = [name[0] for name in cur.description]
+    return_str = ""
+    if templatefile != None:
+      start_response('200 OK', [('Content-Type', 'text/html')])
+      before = time.time()
+      template_str = load_template(templatefile)
+      after = time.time()
+      log.write("%s: template loading time %f\n" % (label, after - before))
+      before = time.time()
+      return_str = build_template(headings, allresults, template_str)
+      after = time.time()
+      log.write("%s: template rendering time %f\n" % (label, after - before))
+    else:
+      start_response('200 OK', [('Content-Type', 'text/plain')])
+      before = time.time()
+      return_str = build_json(headings, allresults, callback)
+      after = time.time()
+      log.write("%s: json-making time %f\n" % (label, after - before))
+    return return_str
+
+if __name__ == '__main__':
+    from fcgi import WSGIServer
+    WSGIServer(myapp).run()

samples/Ruby/mdata_server.fcgi

@@ -0,0 +1,68 @@
+#!/usr/bin/env ruby
+require "xmlrpc/server"
+
+# NOTE: force the usage of the pure-ruby version of fcgi. 
+# - this is required by the workaround to get fcgi+xmlrpc working together
+FCGI_PURE_RUBY=true
+require 'fcgi'
+
+require File.join(File.dirname(__FILE__), '../bt_cast/mdata_echo_server/bt_cast_mdata_server_t.rb')
+
+################################################################################
+################################################################################
+#			CGI handling for xmlrpc
+################################################################################
+################################################################################
+# - for basic xmlrpc via CGI example
+#   - see http://www.ntecs.de/projects/xmlrpc4r/server.html#label-19
+
+# create the directory needed for Neoip::Cast_mdata_server_t
+Neoip::Cast_mdata_server_t.create_dir_ifneeded();
+
+# init the cgi_server
+cgi_server	= XMLRPC::CGIServer.new     
+# register all the xmlrpc function
+cgi_server.add_handler("set_cast_mdata_pull") do |web2srv_str, cast_name, cast_privtext, cast_id, 
+						port_lview, port_pview, uri_pathquery|
+	Neoip::Cast_mdata_server_t.set_cast_mdata_pull(web2srv_str, cast_name, cast_privtext, cast_id,
+						port_lview, port_pview, uri_pathquery, ENV['REMOTE_ADDR']);
+end 
+cgi_server.add_handler("set_cast_mdata_push") do |web2srv_str, cast_name, cast_privtext, cast_mdata|
+	Neoip::Cast_mdata_server_t.set_cast_mdata_push(web2srv_str, cast_name, cast_privtext, cast_mdata);
+end 
+cgi_server.add_handler("get_cast_mdata") do |cast_name, cast_privhash|
+	Neoip::Cast_mdata_server_t.get_cast_mdata(cast_name, cast_privhash);
+end 
+cgi_server.add_handler("del_cast_mdata") do |cast_name, cast_privtext|
+	Neoip::Cast_mdata_server_t.del_cast_mdata(cast_name, cast_privtext);
+end
+
+# handle the unknown/bad formered calls
+cgi_server.set_default_handler do |name, *args|
+	raise XMLRPC::FaultException.new(-99, "Method #{name} missing" +
+                                   " or wrong number of parameters!")
+end
+
+# server the cgi_server
+#cgi_server.serve
+#exit
+  
+# experiment at using fast-cgi
+FCGI.each_request do |request|
+	# XMLRPC::CGIServer expect some value in ENV[] but FCGI doesnt provides them
+	# - so working around by copying them by hand... dirty 
+	ENV['REMOTE_ADDR']	= request.env['REMOTE_ADDR'];
+	ENV['REQUEST_METHOD']	= request.env['REQUEST_METHOD'];
+	ENV['CONTENT_TYPE']	= "text/xml";
+	ENV['CONTENT_LENGTH']	= "#{request.in.length}";
+
+	# copy the request in/out into the stdin/stdout to act as a CGI
+	$stdin	= request.in
+	$stdout	= request.out
+
+	# process the cgi itself
+	cgi_server.serve
+
+	# mark the request as finished
+	request.finish
+end

samples/Shell/php.fcgi

@@ -0,0 +1,16 @@
+#!/bin/sh
+# you can change the PHP version here.
+version="RB_PHP_VERSION_X_Y_Z"
+
+# php.ini file location
+PHPRC=/usr/local/php/phpfarm/inst/php-${version}/lib/php.ini
+export PHPRC
+
+PHP_FCGI_CHILDREN=3
+export PHP_FCGI_CHILDREN
+
+PHP_FCGI_MAX_REQUESTS=5000
+export PHP_FCGI_MAX_REQUESTS
+
+# which php-cgi binary to execute
+exec /usr/local/php/inst/php-${version}/bin/php-cgi

samples/Shell/settime.cgi

@@ -0,0 +1,27 @@
+#!/bin/bash
+echo "Content-type: text/html"
+day=`echo "$QUERY_STRING" | sed -n 's/^.*day=\([^&]*\).*$/\1/p' | sed "s/%20/ /g"`
+month=`echo "$QUERY_STRING" | sed -n 's/^.*month=\([^&]*\).*$/\1/p' | sed "s/%20/ /g"`
+year=`echo "$QUERY_STRING" | sed -n 's/^.*year=\([^&]*\).*$/\1/p' | sed "s/%20/ /g"`
+hour=`echo "$QUERY_STRING" | sed -n 's/^.*hour=\([^&]*\).*$/\1/p' | sed "s/%20/ /g"`
+minute=`echo "$QUERY_STRING" | sed -n 's/^.*minute=\([^&]*\).*$/\1/p' | sed "s/%20/ /g"`
+second=`echo "$QUERY_STRING" | sed -n 's/^.*second=\([^&]*\).*$/\1/p' | sed "s/%20/ /g"`
+echo ""
+echo "<html><body>"
+
+echo "<pre> $(killall ems) </pre>"
+
+
+
+echo "<pre> $(date $month$day$hour$minute$year.$second) </pre>"
+
+echo "<pre> $(/sbin/hwclock -w>/dev/null & /sbin/reboot) </pre>"
+
+echo "<pre> $(/sbin/reboot) </pre>"
+
+
+
+
+
+
+echo "</body></html>"

samples/Tcl/filenames/owh

@@ -0,0 +1,63 @@
+#!/usr/bin/env tclsh
+# http://wiki.tcl.tk/906
+
+if {[llength $argv] < 1} {
+   puts "usage: owh ?init? body ?exit?
+   performs body (in Tcl) for each line (\$0) from stdin
+   owh: Ousterhout - Welch - Hobbs, to name a few"
+   exit -1
+}
+
+proc awksplit {text {split default}} {
+    set no 0
+    if {$split eq "default"} {
+        set t {}
+        foreach string [split $text] {
+            if {$string ne {}} {
+                lappend t $string
+            }
+        }
+    } else {
+        set t [list $text $split]
+    }
+    uplevel 1 [list set NF [llength $t]]
+    foreach i $t {uplevel 1 [list set [incr no] $i]}
+    uplevel 1 {set 0 {};trace variable 0 ru 0}
+}
+proc 0 {_name index op} {
+    switch $op {
+        r {
+            uplevel {
+                set 0 {}
+                for {set i 1} {$i <= $NF} {incr i} {lappend 0 [set $i]}
+                set 0 [join $0 $OFS]
+            }
+        }
+        u {rename 0 {} ;# leave no traces of the trace..}
+    }
+}
+
+proc print s {if {[catch {puts $s}]} exit} ;# good for broken pipe
+
+set FS default
+set OFS { }
+
+if {[llength $argv] > 1} {
+   eval [lindex $argv 0]
+   set _body [lindex $argv 1] ;# strip outer braces
+   set _exit [lindex $argv 2]
+} else {
+   set _body [lindex $argv 0] ;# strip outer braces
+   set _exit {}
+}
+
+set NR 1
+while 1 {
+   gets stdin line
+   if {[eof stdin]} break
+   awksplit $line $FS
+   eval $_body
+   incr NR
+}
+set res [eval $_exit]
+if {[string length $res]} {puts $res}

samples/Tcl/filenames/starfield

@@ -0,0 +1,28 @@
+#!/usr/bin/env wish
+# http://wiki.tcl.tk/14140
+
+proc stars'go {c factor} {
+    set w [winfo width $c]
+    set h [winfo height $c] 
+    $c scale all [expr {$w/2}] [expr {$h/2}] $factor $factor 
+    foreach item [$c find all] {
+        if {[llength [$c bbox $item]] == 0} {$c delete $item; continue} ;# (1)
+        foreach {x0 y0 x1 y1} [$c bbox $item] break
+        if {$x1<0 || $x0>$w || $y1<0 || $y0>$h} {$c delete $item}
+    }
+    time {
+        set x [expr {rand()*$w}]
+        set y [expr {rand()*$h}]
+        set col [lpick {white yellow beige bisque cyan}]
+        $c create oval $x $y [expr {$x+1}] [expr {$y+1}] -fill $col \
+                -outline $col
+    } 10
+    after $::ms [info level 0]
+}
+proc lpick list {lindex $list [expr {int(rand()*[llength $list])}]}
+#-- Let's go!
+pack [canvas .c -bg black] -fill both -expand 1
+set ms 40
+bind . <Up> {incr ms -5}
+bind . <Down> {incr ms 5}
+stars'go .c 1.05

samples/VimL/filenames/_vimrc

@@ -0,0 +1,8 @@
+set nocompatible
+set ignorecase
+set incsearch
+set smartcase
+set showmatch
+set showcmd
+
+syntax on

script/cibuild

@@ -0,0 +1,17 @@
+#!/bin/sh
+if [ -d /usr/share/rbenv/shims ]; then
+  export PATH=/usr/share/rbenv/shims:$PATH
+  export RBENV_VERSION=2.1.2-github
+  export RUBY_VERSION=2.1.2-github
+fi
+
+set -x
+git log -n 1 HEAD | cat
+ruby -v
+bundle -v
+set +x
+
+mkdir -p ./vendor/gems
+
+bundle install --local --path ./vendor/gems
+bundle exec rake

test/test_blob.rb

@@ -262,6 +262,10 @@ class TestBlob < Test::Unit::TestCase
 
 
     assert Linguist::Generated.generated?("node_modules/grunt/lib/grunt.js", nil)
+
+    # Godep saved dependencies
+    assert blob("Godeps/Godeps.json").generated?
+    assert blob("Godeps/_workspace/src/github.com/kr/s3/sign.go").generated?
   end
 
   def test_vendored
@@ -279,6 +283,10 @@ class TestBlob < Test::Unit::TestCase
     assert blob("app/bower_components/custom/custom.js").vendored?
     assert blob("vendor/assets/bower_components/custom/custom.js").vendored?
 
+    # Go dependencies
+    assert !blob("Godeps/Godeps.json").vendored?
+    assert blob("Godeps/_workspace/src/github.com/kr/s3/sign.go").vendored?
+
     # Rails vendor/
     assert blob("vendor/plugins/will_paginate/lib/will_paginate.rb").vendored?
 
@@ -437,6 +445,12 @@ class TestBlob < Test::Unit::TestCase
     assert blob("octicons.css").vendored?
     assert blob("public/octicons.min.css").vendored?
     assert blob("public/octicons/sprockets-octicons.scss").vendored?
+
+    # Typesafe Activator
+    assert blob("activator").vendored?
+    assert blob("activator.bat").vendored?
+    assert blob("subproject/activator").vendored?
+    assert blob("subproject/activator.bat").vendored?
   end
 
   def test_language

test/test_classifier.rb

@@ -44,12 +44,12 @@ class TestClassifier < Test::Unit::TestCase
   end
 
   def test_instance_classify_empty
-    results = Classifier.classify(Samples::DATA, "")
+    results = Classifier.classify(Samples.cache, "")
     assert results.first[1] < 0.5, results.first.inspect
   end
 
   def test_instance_classify_nil
-    assert_equal [], Classifier.classify(Samples::DATA, nil)
+    assert_equal [], Classifier.classify(Samples.cache, nil)
   end
 
   def test_classify_ambiguous_languages
@@ -58,7 +58,7 @@ class TestClassifier < Test::Unit::TestCase
       languages = Language.find_by_filename(sample[:path]).map(&:name)
       next unless languages.length > 1
 
-      results = Classifier.classify(Samples::DATA, File.read(sample[:path]), languages)
+      results = Classifier.classify(Samples.cache, File.read(sample[:path]), languages)
       assert_equal language.name, results.first[0], "#{sample[:path]}\n#{results.inspect}"
     end
   end

test/test_heuristics.rb

@@ -74,6 +74,18 @@ class TestHeuristcs < Test::Unit::TestCase
     assert_equal Language["ECL"], results.first
   end
 
+  def test_pro_prolog_by_heuristics
+    languages = ["IDL", "Prolog"]
+    results = Heuristics.disambiguate_pro(fixture("Prolog/logic-problem.pro"), languages)
+    assert_equal Language["Prolog"], results.first
+  end
+
+  def test_pro_idl_by_heuristics
+    languages = ["IDL", "Prolog"]
+    results = Heuristics.disambiguate_pro(fixture("IDL/mg_acosh.pro"), languages)
+    assert_equal Language["IDL"], results.first
+  end
+
   def test_ts_typescript_by_heuristics
     languages = ["TypeScript", "XML"]
     results = Heuristics.disambiguate_ts(fixture("TypeScript/classes.ts"), languages)

test/test_language.rb

@@ -33,6 +33,7 @@ class TestLanguage < Test::Unit::TestCase
     assert_equal Lexer['Java'], Language['ChucK'].lexer
     assert_equal Lexer['Java'], Language['Java'].lexer
     assert_equal Lexer['JavaScript'], Language['JavaScript'].lexer
+    assert_equal Lexer['LSL'], Language['LSL'].lexer
     assert_equal Lexer['MOOCode'], Language['Moocode'].lexer
     assert_equal Lexer['MuPAD'], Language['mupad'].lexer
     assert_equal Lexer['NASM'], Language['Assembly'].lexer
@@ -186,6 +187,7 @@ class TestLanguage < Test::Unit::TestCase
 
   def test_programming
     assert_equal :programming, Language['JavaScript'].type
+    assert_equal :programming, Language['LSL'].type
     assert_equal :programming, Language['Perl'].type
     assert_equal :programming, Language['PowerShell'].type
     assert_equal :programming, Language['Python'].type
@@ -326,6 +328,7 @@ class TestLanguage < Test::Unit::TestCase
     assert_equal '#3581ba', Language['Python'].color
     assert_equal '#f1e05a', Language['JavaScript'].color
     assert_equal '#31859c', Language['TypeScript'].color
+    assert_equal '#3d9970', Language['LSL'].color
   end
 
   def test_colors
@@ -338,6 +341,7 @@ class TestLanguage < Test::Unit::TestCase
     assert_equal 'coffee', Language['CoffeeScript'].ace_mode
     assert_equal 'csharp', Language['C#'].ace_mode
     assert_equal 'css', Language['CSS'].ace_mode
+    assert_equal 'lsl', Language['LSL'].ace_mode
     assert_equal 'javascript', Language['JavaScript'].ace_mode
   end
 
@@ -352,6 +356,7 @@ class TestLanguage < Test::Unit::TestCase
   end
 
   def test_extensions
+    assert Language['LSL'].extensions.include?('.lsl')
     assert Language['Perl'].extensions.include?('.pl')
     assert Language['Python'].extensions.include?('.py')
     assert Language['Ruby'].extensions.include?('.rb')

test/test_repository.rb

@@ -1,5 +1,5 @@
 require 'linguist/repository'
-
+require 'linguist/lazy_blob'
 require 'test/unit'
 
 class TestRepository < Test::Unit::TestCase
@@ -47,4 +47,58 @@ class TestRepository < Test::Unit::TestCase
 
     assert_equal linguist_repo.cache, new_repo.cache
   end
+
+  def test_repo_git_attributes
+    # See https://github.com/github/linguist/blob/351c1cc8fd57340839bdb400d7812332af80e9bd/.gitattributes
+    #
+    # It looks like this:
+    # Gemfile linguist-vendored=true
+    # lib/linguist.rb linguist-language=Java
+    # test/*.rb linguist-language=Java
+    # Rakefile linguist-generated
+    # test/fixtures/* linguist-vendored=false
+
+    attr_commit = '351c1cc8fd57340839bdb400d7812332af80e9bd'
+    repo = linguist_repo(attr_commit)
+
+    assert repo.breakdown_by_file.has_key?("Java")
+    assert repo.breakdown_by_file["Java"].include?("lib/linguist.rb")
+
+    assert repo.breakdown_by_file.has_key?("Ruby")
+    assert !repo.breakdown_by_file["Ruby"].empty?
+  end
+
+  def test_commit_with_git_attributes_data
+    # Before we had any .gitattributes data
+    old_commit = '4a017d9033f91b2776eb85275463f9613cc371ef'
+    old_repo = linguist_repo(old_commit)
+
+    # With some .gitattributes data
+    attr_commit = '7ee006cbcb2d7261f9e648510a684ee9ac64126b'
+    # It's incremental but should bust the cache
+    new_repo = Linguist::Repository.incremental(rugged_repository, attr_commit, old_commit, old_repo.cache)
+
+    assert new_repo.breakdown_by_file["Java"].include?("lib/linguist.rb")
+  end
+
+  def test_linguist_override_vendored?
+    attr_commit = '351c1cc8fd57340839bdb400d7812332af80e9bd'
+    repo = linguist_repo(attr_commit).read_index
+
+    override_vendored = Linguist::LazyBlob.new(rugged_repository, attr_commit, 'Gemfile')
+
+    # overridden .gitattributes
+    assert override_vendored.vendored?
+  end
+
+  def test_linguist_override_unvendored?
+    attr_commit = '351c1cc8fd57340839bdb400d7812332af80e9bd'
+    repo = linguist_repo(attr_commit).read_index
+
+    # lib/linguist/vendor.yml defines this as vendored.
+    override_unvendored = Linguist::LazyBlob.new(rugged_repository, attr_commit, 'test/fixtures/foo.rb')
+
+    # overridden .gitattributes
+    assert !override_unvendored.vendored?
+  end
 end

test/test_samples.rb

@@ -8,7 +8,7 @@ class TestSamples < Test::Unit::TestCase
   include Linguist
 
   def test_up_to_date
-    assert serialized = Samples::DATA
+    assert serialized = Samples.cache
     assert latest = Samples.data
 
     # Just warn, it shouldn't scare people off by breaking the build.
@@ -29,7 +29,7 @@ class TestSamples < Test::Unit::TestCase
   end
 
   def test_verify
-    assert data = Samples::DATA
+    assert data = Samples.cache
 
     assert_equal data['languages_total'], data['languages'].inject(0) { |n, (_, c)| n += c }
     assert_equal data['tokens_total'], data['language_tokens'].inject(0) { |n, (_, c)| n += c }
@@ -38,7 +38,7 @@ class TestSamples < Test::Unit::TestCase
 
   # Check that there aren't samples with extensions that aren't explicitly defined in languages.yml
   def test_parity
-    extensions = Samples::DATA['extnames']
+    extensions = Samples.cache['extnames']
     languages_yml = File.expand_path("../../lib/linguist/languages.yml", __FILE__)
     languages = YAML.load_file(languages_yml)