From c4260ae6814e0b60fa97db1359d1044057df46aa Mon Sep 17 00:00:00 2001
From: Vicent Marti <tanoku@gmail.com>
Date: Mon, 16 Jun 2014 15:19:56 +0200
Subject: [PATCH 01/20] Use Rugged when computing Repository stats

---
 Gemfile                     |   2 +
 lib/linguist/blob_helper.rb |  10 +--
 lib/linguist/file_blob.rb   |   4 +-
 lib/linguist/language.rb    |  14 ++--
 lib/linguist/repository.rb  | 128 ++++++++++++++++++++----------------
 test/test_heuristics.rb     |   5 --
 test/test_repository.rb     |  11 +---
 7 files changed, 82 insertions(+), 92 deletions(-)

diff --git a/Gemfile b/Gemfile
index 851fabc2..3ac3ffa4 100644
--- a/Gemfile
+++ b/Gemfile
@@ -1,2 +1,4 @@
 source 'https://rubygems.org'
+
 gemspec
+gem 'rugged', :git => 'https://github.com/libgit2/rugged.git', branch: 'development', submodules: true
diff --git a/lib/linguist/blob_helper.rb b/lib/linguist/blob_helper.rb
index 15ab2d9f..27c4b3dd 100644
--- a/lib/linguist/blob_helper.rb
+++ b/lib/linguist/blob_helper.rb
@@ -313,15 +313,7 @@ module Linguist
     #
     # Returns a Language or nil if none is detected
     def language
-      return @language if defined? @language
-
-      if defined?(@data) && @data.is_a?(String)
-        data = @data
-      else
-        data = lambda { (binary_mime_type? || binary?) ? "" : self.data }
-      end
-
-      @language = Language.detect(name.to_s, data, mode)
+      @language ||= Language.detect(self)
     end
 
     # Internal: Get the lexer of the blob.
diff --git a/lib/linguist/file_blob.rb b/lib/linguist/file_blob.rb
index 7e7f1acd..e84e9b61 100644
--- a/lib/linguist/file_blob.rb
+++ b/lib/linguist/file_blob.rb
@@ -34,9 +34,9 @@ module Linguist
 
     # Public: Read file permissions
     #
-    # Returns a String like '100644'
+    # Returns an Int like 0100644
     def mode
-      File.stat(@path).mode.to_s(8)
+      File.stat(@path).mode
     end
 
     # Public: Read file contents.
diff --git a/lib/linguist/language.rb b/lib/linguist/language.rb
index a8e7a33c..3a354463 100644
--- a/lib/linguist/language.rb
+++ b/lib/linguist/language.rb
@@ -92,18 +92,14 @@ module Linguist
 
     # Public: Detects the Language of the blob.
     #
-    # name - String filename
-    # data - String blob data. A block also maybe passed in for lazy
-    #        loading. This behavior is deprecated and you should always
-    #        pass in a String.
-    # mode - Optional String mode (defaults to nil)
-    #
     # Returns Language or nil.
-    def self.detect(name, data, mode = nil)
+    def self.detect(blob)
+      name = blob.name
+
       # A bit of an elegant hack. If the file is executable but extensionless,
       # append a "magic" extension so it can be classified with other
       # languages that have shebang scripts.
-      if File.extname(name).empty? && mode && (mode.to_i(8) & 05) == 05
+      if File.extname(name).empty? && blob.mode && (blob.mode & 05) == 05
         name += ".script!"
       end
 
@@ -114,7 +110,7 @@ module Linguist
       # extension at all, in the case of extensionless scripts), we need to continue
       # our detection work
       if possible_languages.length > 1
-        data = data.call() if data.respond_to?(:call)
+        data = blob.data
         possible_language_names = possible_languages.map(&:name)
 
         # Don't bother with emptiness
diff --git a/lib/linguist/repository.rb b/lib/linguist/repository.rb
index a62bf281..bbf32171 100644
--- a/lib/linguist/repository.rb
+++ b/lib/linguist/repository.rb
@@ -1,4 +1,6 @@
 require 'linguist/file_blob'
+require 'linguist/lazy_blob'
+require 'rugged'
 
 module Linguist
   # A Repository is an abstraction of a Grit::Repo or a basic file
@@ -7,29 +9,15 @@ module Linguist
   # Its primary purpose is for gathering language statistics across
   # the entire project.
   class Repository
-    # Public: Initialize a new Repository from a File directory
-    #
-    # base_path - A path String
-    #
-    # Returns a Repository
-    def self.from_directory(base_path)
-      new Dir["#{base_path}/**/*"].
-        select { |f| File.file?(f) }.
-        map { |path| FileBlob.new(path, base_path) }
-    end
+    attr_reader :repository
 
     # Public: Initialize a new Repository
     #
-    # enum - Enumerator that responds to `each` and
-    #        yields Blob objects
-    #
     # Returns a Repository
-    def initialize(enum)
-      @enum = enum
-      @computed_stats = false
-      @language = @size = nil
-      @sizes = Hash.new { 0 }
-      @file_breakdown = Hash.new { |h,k| h[k] = Array.new }
+    def initialize(repo, sha1, existing_stats = nil)
+      @repository = repo
+      @current_sha1 = sha1
+      @old_sha1, @old_stats = existing_stats if existing_stats
     end
 
     # Public: Returns a breakdown of language stats.
@@ -41,66 +29,90 @@ module Linguist
     #
     # Returns a Hash of Language keys and Integer size values.
     def languages
-      compute_stats
-      @sizes
+      @sizes ||= begin
+        sizes = Hash.new { 0 }
+        file_map.each do |_, (language, size)|
+          sizes[language] += size
+        end
+        sizes
+      end
     end
 
     # Public: Get primary Language of repository.
     #
     # Returns a Language
     def language
-      compute_stats
-      @language
+      @language ||= begin
+        primary = languages.max_by { |(_, size)| size }
+        primary && primary[0]
+      end
     end
 
     # Public: Get the total size of the repository.
     #
     # Returns a byte size Integer
     def size
-      compute_stats
-      @size
+      @size ||= languages.inject(0) { |s,(_,v)| s + v }
     end
 
     # Public: Return the language breakdown of this repository by file
     def breakdown_by_file
-      compute_stats
-      @file_breakdown
+      @file_breakdown ||= begin
+        breakdown = Hash.new { |h,k| h[k] = Array.new }
+        file_map.each do |filename, (language, _)|
+          breakdown[language.name] << filename
+        end
+        breakdown
+      end
+    end
+
+    def incremental_stats(old_sha1, new_sha1, file_map = nil)
+      file_map = file_map ? file_map.dup : {}
+      old_commit = old_sha1 && Rugged::Commit.lookup(repository, old_sha1)
+      new_commit = Rugged::Commit.lookup(repository, new_sha1)
+
+      diff = Rugged::Tree.diff(repository, old_commit, new_commit)
+
+      diff.each_delta do |delta|
+        old = delta.old_file[:path]
+        new = delta.new_file[:path]
+
+        file_map.delete(old)
+        next if delta.binary
+
+        if [:added, :modified].include? delta.status
+          blob = Linguist::LazyBlob.new(repository, delta.new_file[:oid], new, delta.new_file[:mode])
+
+          # Skip vendored or generated blobs
+          next if blob.vendored? || blob.generated? || blob.language.nil?
+
+          # Only include programming languages and acceptable markup languages
+          if blob.language.type == :programming || Language.detectable_markup.include?(blob.language.name)
+            file_map[new] = [blob.language.group, blob.size]
+          end
+        end
+      end
+
+      file_map
+    end
+
+    def load_stats(file)
+      @old_sha1, @old_stats = JSON.load(file)
+    end
+
+    def dump_stats(file)
+      JSON.dump([@current_sha1, file_map], file)
     end
 
     # Internal: Compute language breakdown for each blob in the Repository.
     #
     # Returns nothing
-    def compute_stats
-      return if @computed_stats
-
-      @enum.each do |blob|
-        # Skip files that are likely binary
-        next if blob.likely_binary?
-
-        # Skip vendored or generated blobs
-        next if blob.vendored? || blob.generated? || blob.language.nil?
-
-        # Only include programming languages and acceptable markup languages
-        if blob.language.type == :programming || Language.detectable_markup.include?(blob.language.name)
-
-          # Build up the per-file breakdown stats
-          @file_breakdown[blob.language.group.name] << blob.name
-
-          @sizes[blob.language.group] += blob.size
-        end
-      end
-
-      # Compute total size
-      @size = @sizes.inject(0) { |s,(_,v)| s + v }
-
-      # Get primary language
-      if primary = @sizes.max_by { |(_, size)| size }
-        @language = primary[0]
-      end
-
-      @computed_stats = true
-
-      nil
+    def file_map
+      @file_map ||= if @old_sha1 == @current_sha1
+                      @old_stats
+                    else
+                      incremental_stats(@old_sha1, @current_sha1, @old_stats)
+                    end
     end
   end
 end
diff --git a/test/test_heuristics.rb b/test/test_heuristics.rb
index 0c1a07ff..33f8a087 100644
--- a/test/test_heuristics.rb
+++ b/test/test_heuristics.rb
@@ -34,11 +34,6 @@ class TestHeuristcs < Test::Unit::TestCase
     assert_equal Language["C++"], results.first
   end
 
-  def test_detect_still_works_if_nothing_matches
-    match = Language.detect("Hello.m", fixture("Objective-C/hello.m"))
-    assert_equal Language["Objective-C"], match
-  end
-  
   def test_pl_prolog_by_heuristics
     languages = ["Perl", "Prolog"]
     results = Heuristics.disambiguate_pl(fixture("Prolog/turing.pl"), languages)
diff --git a/test/test_repository.rb b/test/test_repository.rb
index 832489d3..13dbdaef 100644
--- a/test/test_repository.rb
+++ b/test/test_repository.rb
@@ -5,12 +5,9 @@ require 'test/unit'
 class TestRepository < Test::Unit::TestCase
   include Linguist
 
-  def repo(base_path)
-    Repository.from_directory(base_path)
-  end
-
   def linguist_repo
-    repo(File.expand_path("../..", __FILE__))
+    r = Rugged::Repository.new(File.expand_path("../../.git", __FILE__))
+    Linguist::Repository.new(r, '31921838cdc252536ec07668f73d4b64d8022750')
   end
 
   def test_linguist_language
@@ -30,8 +27,4 @@ class TestRepository < Test::Unit::TestCase
     assert linguist_repo.breakdown_by_file["Ruby"].include?("bin/linguist")
     assert linguist_repo.breakdown_by_file["Ruby"].include?("lib/linguist/language.rb")
   end
-
-  def test_binary_override
-    assert_equal repo(File.expand_path("../../samples/Nimrod", __FILE__)).language, Language["Nimrod"]
-  end
 end

From cd58a30c7c29fb1a0cd4331eb7d1f36ce445bbc7 Mon Sep 17 00:00:00 2001
From: Vicent Marti <tanoku@gmail.com>
Date: Mon, 16 Jun 2014 16:41:58 +0200
Subject: [PATCH 02/20] Only cache strings, thanks

---
 lib/linguist/repository.rb | 37 ++++++++++++++-----------------------
 lib/linguist/version.rb    |  2 +-
 2 files changed, 15 insertions(+), 24 deletions(-)

diff --git a/lib/linguist/repository.rb b/lib/linguist/repository.rb
index bbf32171..960ae3df 100644
--- a/lib/linguist/repository.rb
+++ b/lib/linguist/repository.rb
@@ -31,7 +31,7 @@ module Linguist
     def languages
       @sizes ||= begin
         sizes = Hash.new { 0 }
-        file_map.each do |_, (language, size)|
+        cache.each do |_, (language, size)|
           sizes[language] += size
         end
         sizes
@@ -59,15 +59,15 @@ module Linguist
     def breakdown_by_file
       @file_breakdown ||= begin
         breakdown = Hash.new { |h,k| h[k] = Array.new }
-        file_map.each do |filename, (language, _)|
-          breakdown[language.name] << filename
+        cache.each do |filename, (language, _)|
+          breakdown[language] << filename
         end
         breakdown
       end
     end
 
-    def incremental_stats(old_sha1, new_sha1, file_map = nil)
-      file_map = file_map ? file_map.dup : {}
+    def incremental_stats(old_sha1, new_sha1, cache = nil)
+      file_map = cache ? cache.dup : {}
       old_commit = old_sha1 && Rugged::Commit.lookup(repository, old_sha1)
       new_commit = Rugged::Commit.lookup(repository, new_sha1)
 
@@ -88,7 +88,7 @@ module Linguist
 
           # Only include programming languages and acceptable markup languages
           if blob.language.type == :programming || Language.detectable_markup.include?(blob.language.name)
-            file_map[new] = [blob.language.group, blob.size]
+            file_map[new] = [blob.language.group.name, blob.size]
           end
         end
       end
@@ -96,23 +96,14 @@ module Linguist
       file_map
     end
 
-    def load_stats(file)
-      @old_sha1, @old_stats = JSON.load(file)
-    end
-
-    def dump_stats(file)
-      JSON.dump([@current_sha1, file_map], file)
-    end
-
-    # Internal: Compute language breakdown for each blob in the Repository.
-    #
-    # Returns nothing
-    def file_map
-      @file_map ||= if @old_sha1 == @current_sha1
-                      @old_stats
-                    else
-                      incremental_stats(@old_sha1, @current_sha1, @old_stats)
-                    end
+    def cache
+      @cache ||= begin
+        if @old_sha1 == @current_sha1
+          @old_stats
+        else
+          incremental_stats(@old_sha1, @current_sha1, @old_stats)
+        end
+      end
     end
   end
 end
diff --git a/lib/linguist/version.rb b/lib/linguist/version.rb
index 6704b3fb..ed7d57dc 100644
--- a/lib/linguist/version.rb
+++ b/lib/linguist/version.rb
@@ -1,3 +1,3 @@
 module Linguist
-  VERSION = "2.12.0"
+  VERSION = "2.13.0.github2"
 end

From 463f48f04f4f2137222044ba42d5dc7175f5dbf2 Mon Sep 17 00:00:00 2001
From: Vicent Marti <tanoku@gmail.com>
Date: Mon, 16 Jun 2014 17:31:44 +0200
Subject: [PATCH 03/20] Mode must always be a String

---
 lib/linguist/file_blob.rb  | 4 ++--
 lib/linguist/language.rb   | 2 +-
 lib/linguist/repository.rb | 3 ++-
 lib/linguist/version.rb    | 2 +-
 4 files changed, 6 insertions(+), 5 deletions(-)

diff --git a/lib/linguist/file_blob.rb b/lib/linguist/file_blob.rb
index e84e9b61..9123922f 100644
--- a/lib/linguist/file_blob.rb
+++ b/lib/linguist/file_blob.rb
@@ -34,9 +34,9 @@ module Linguist
 
     # Public: Read file permissions
     #
-    # Returns an Int like 0100644
+    # Returns a String like '0100644'
     def mode
-      File.stat(@path).mode
+      File.stat(@path).mode.to_s(8)
     end
 
     # Public: Read file contents.
diff --git a/lib/linguist/language.rb b/lib/linguist/language.rb
index 3a354463..bdaa91e8 100644
--- a/lib/linguist/language.rb
+++ b/lib/linguist/language.rb
@@ -99,7 +99,7 @@ module Linguist
       # A bit of an elegant hack. If the file is executable but extensionless,
       # append a "magic" extension so it can be classified with other
       # languages that have shebang scripts.
-      if File.extname(name).empty? && blob.mode && (blob.mode & 05) == 05
+      if File.extname(name).empty? && blob.mode && (blob.mode.to_i(8) & 05) == 05
         name += ".script!"
       end
 
diff --git a/lib/linguist/repository.rb b/lib/linguist/repository.rb
index 960ae3df..9fb6ebc7 100644
--- a/lib/linguist/repository.rb
+++ b/lib/linguist/repository.rb
@@ -81,7 +81,8 @@ module Linguist
         next if delta.binary
 
         if [:added, :modified].include? delta.status
-          blob = Linguist::LazyBlob.new(repository, delta.new_file[:oid], new, delta.new_file[:mode])
+          mode = delta.new_file[:mode].to_s(8)
+          blob = Linguist::LazyBlob.new(repository, delta.new_file[:oid], new, mode)
 
           # Skip vendored or generated blobs
           next if blob.vendored? || blob.generated? || blob.language.nil?
diff --git a/lib/linguist/version.rb b/lib/linguist/version.rb
index ed7d57dc..c47a92f3 100644
--- a/lib/linguist/version.rb
+++ b/lib/linguist/version.rb
@@ -1,3 +1,3 @@
 module Linguist
-  VERSION = "2.13.0.github2"
+  VERSION = "2.13.0.github3"
 end

From ea1fc90cf5610acf6dd987af324ceab7ed78dbc0 Mon Sep 17 00:00:00 2001
From: Vicent Marti <tanoku@gmail.com>
Date: Mon, 16 Jun 2014 18:41:51 +0200
Subject: [PATCH 04/20] Handle `nil` blob names

---
 lib/linguist/file_blob.rb | 2 +-
 lib/linguist/language.rb  | 2 +-
 lib/linguist/version.rb   | 2 +-
 3 files changed, 3 insertions(+), 3 deletions(-)

diff --git a/lib/linguist/file_blob.rb b/lib/linguist/file_blob.rb
index 9123922f..7e7f1acd 100644
--- a/lib/linguist/file_blob.rb
+++ b/lib/linguist/file_blob.rb
@@ -34,7 +34,7 @@ module Linguist
 
     # Public: Read file permissions
     #
-    # Returns a String like '0100644'
+    # Returns a String like '100644'
     def mode
       File.stat(@path).mode.to_s(8)
     end
diff --git a/lib/linguist/language.rb b/lib/linguist/language.rb
index bdaa91e8..ffbecb7b 100644
--- a/lib/linguist/language.rb
+++ b/lib/linguist/language.rb
@@ -94,7 +94,7 @@ module Linguist
     #
     # Returns Language or nil.
     def self.detect(blob)
-      name = blob.name
+      name = blob.name.to_s
 
       # A bit of an elegant hack. If the file is executable but extensionless,
       # append a "magic" extension so it can be classified with other
diff --git a/lib/linguist/version.rb b/lib/linguist/version.rb
index c47a92f3..232278b4 100644
--- a/lib/linguist/version.rb
+++ b/lib/linguist/version.rb
@@ -1,3 +1,3 @@
 module Linguist
-  VERSION = "2.13.0.github3"
+  VERSION = "2.13.0.github4"
 end

From 5896bb8fa35d813902dff6a7495f3da3ec2f32f2 Mon Sep 17 00:00:00 2001
From: Vicent Marti <tanoku@gmail.com>
Date: Tue, 24 Jun 2014 17:52:43 +0200
Subject: [PATCH 05/20] Missing file. Duh.

---
 lib/linguist/lazy_blob.rb | 37 +++++++++++++++++++++++++++++++++++++
 1 file changed, 37 insertions(+)
 create mode 100644 lib/linguist/lazy_blob.rb

diff --git a/lib/linguist/lazy_blob.rb b/lib/linguist/lazy_blob.rb
new file mode 100644
index 00000000..bb262241
--- /dev/null
+++ b/lib/linguist/lazy_blob.rb
@@ -0,0 +1,37 @@
+require 'linguist/blob_helper'
+require 'rugged'
+
+module Linguist
+  class LazyBlob
+    include BlobHelper
+
+    MAX_SIZE = 128 * 1024
+
+    attr_reader :repository
+    attr_reader :oid
+    attr_reader :name
+    attr_reader :mode
+
+    def initialize(repo, oid, name, mode = nil)
+      @repository = repo
+      @oid = oid
+      @name = name
+      @mode = mode
+    end
+
+    def data
+      load_blob!
+      @data
+    end
+
+    def size
+      load_blob!
+      @size
+    end
+
+    protected
+    def load_blob!
+      @data, @size = Rugged::Blob.to_buffer(repository, oid, MAX_SIZE) if @data.nil?
+    end
+  end
+end

From 1fd59361b5c4ce323d633261377a9175d3f8f1f9 Mon Sep 17 00:00:00 2001
From: Vicent Marti <tanoku@gmail.com>
Date: Wed, 25 Jun 2014 20:26:44 +0200
Subject: [PATCH 06/20] Proper incremental diffing

---
 lib/linguist/repository.rb | 18 +++++++++---------
 test/test_repository.rb    | 33 +++++++++++++++++++++++++++------
 2 files changed, 36 insertions(+), 15 deletions(-)

diff --git a/lib/linguist/repository.rb b/lib/linguist/repository.rb
index 9fb6ebc7..e11d2b23 100644
--- a/lib/linguist/repository.rb
+++ b/lib/linguist/repository.rb
@@ -14,10 +14,10 @@ module Linguist
     # Public: Initialize a new Repository
     #
     # Returns a Repository
-    def initialize(repo, sha1, existing_stats = nil)
+    def initialize(repo, commit_oid, existing_stats = nil)
       @repository = repo
-      @current_sha1 = sha1
-      @old_sha1, @old_stats = existing_stats if existing_stats
+      @commit_oid = commit_oid
+      @old_commit_oid, @old_stats = existing_stats if existing_stats
     end
 
     # Public: Returns a breakdown of language stats.
@@ -66,12 +66,12 @@ module Linguist
       end
     end
 
-    def incremental_stats(old_sha1, new_sha1, cache = nil)
+    def compute_stats(old_commit_oid, commit_oid, cache = nil)
       file_map = cache ? cache.dup : {}
-      old_commit = old_sha1 && Rugged::Commit.lookup(repository, old_sha1)
-      new_commit = Rugged::Commit.lookup(repository, new_sha1)
+      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_commit, new_commit)
+      diff = Rugged::Tree.diff(repository, old_tree, new_tree)
 
       diff.each_delta do |delta|
         old = delta.old_file[:path]
@@ -99,10 +99,10 @@ module Linguist
 
     def cache
       @cache ||= begin
-        if @old_sha1 == @current_sha1
+        if @old_commit_oid == @commit_oid
           @old_stats
         else
-          incremental_stats(@old_sha1, @current_sha1, @old_stats)
+          compute_stats(@old_commit_oid, @commit_oid, @old_stats)
         end
       end
     end
diff --git a/test/test_repository.rb b/test/test_repository.rb
index 13dbdaef..bd96e66d 100644
--- a/test/test_repository.rb
+++ b/test/test_repository.rb
@@ -3,19 +3,24 @@ require 'linguist/repository'
 require 'test/unit'
 
 class TestRepository < Test::Unit::TestCase
-  include Linguist
+  def rugged_repository
+    @rugged ||= Rugged::Repository.new(File.expand_path("../../.git", __FILE__))
+  end
 
-  def linguist_repo
-    r = Rugged::Repository.new(File.expand_path("../../.git", __FILE__))
-    Linguist::Repository.new(r, '31921838cdc252536ec07668f73d4b64d8022750')
+  def master_oid
+    @master_oid ||= Rugged::Object.rev_parse_oid(rugged_repository, 'master')
+  end
+
+  def linguist_repo(oid = master_oid)
+    Linguist::Repository.new(rugged_repository, oid)
   end
 
   def test_linguist_language
-    # assert_equal Language['Ruby'], linguist_repo.language
+    assert_equal 'Ruby', linguist_repo.language
   end
 
   def test_linguist_languages
-    # assert linguist_repo.languages[Language['Ruby']] > 10_000
+    assert linguist_repo.languages['Ruby'] > 10_000
   end
 
   def test_linguist_size
@@ -27,4 +32,20 @@ class TestRepository < Test::Unit::TestCase
     assert linguist_repo.breakdown_by_file["Ruby"].include?("bin/linguist")
     assert linguist_repo.breakdown_by_file["Ruby"].include?("lib/linguist/language.rb")
   end
+
+  def test_incremental_stats
+    old_commit = Rugged::Object.rev_parse_oid(rugged_repository, 'v2.0.0')
+    old_repo = linguist_repo(old_commit)
+
+    assert old_repo.languages['Ruby'] > 10_000
+    assert old_repo.size > 30_000
+
+    old_cache = [old_commit, old_repo.cache]
+    new_repo = Linguist::Repository.new(rugged_repository, master_oid, old_cache)
+
+    assert new_repo.languages['Ruby'] > old_repo.languages['Ruby']
+    assert new_repo.size > old_repo.size
+
+    assert_equal linguist_repo.cache, new_repo.cache
+  end
 end

From 29072d6eae88b31c56cf7604e5c2aeb39afbcbda Mon Sep 17 00:00:00 2001
From: Vicent Marti <tanoku@gmail.com>
Date: Thu, 26 Jun 2014 12:27:02 +0200
Subject: [PATCH 07/20] Fix travis build

---
 .travis.yml | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/.travis.yml b/.travis.yml
index 83880550..3a5791da 100644
--- a/.travis.yml
+++ b/.travis.yml
@@ -1,4 +1,6 @@
 before_install: 
+  - git fetch origin master:master
+  - git fetch origin v2.0.0:v2.0.0
   - sudo apt-get install libicu-dev -y
   - gem update --system 2.1.11
 rvm:

From 659d27cae589c9cb5a085239dd3833a8cede9462 Mon Sep 17 00:00:00 2001
From: Vicent Marti <tanoku@gmail.com>
Date: Thu, 26 Jun 2014 12:54:08 +0200
Subject: [PATCH 08/20] DOCS

---
 lib/linguist/repository.rb | 74 ++++++++++++++++++++++++++++----------
 test/test_repository.rb    |  3 +-
 2 files changed, 57 insertions(+), 20 deletions(-)

diff --git a/lib/linguist/repository.rb b/lib/linguist/repository.rb
index e11d2b23..82e9ef47 100644
--- a/lib/linguist/repository.rb
+++ b/lib/linguist/repository.rb
@@ -1,4 +1,3 @@
-require 'linguist/file_blob'
 require 'linguist/lazy_blob'
 require 'rugged'
 
@@ -11,23 +10,53 @@ module Linguist
   class Repository
     attr_reader :repository
 
-    # Public: Initialize a new Repository
+    # Public: Create a new Repository based on the stats of
+    # an existing one
+    def self.incremental(repo, commit_oid, old_commit_oid, old_stats)
+      repo = self.new(repo, commit_oid)
+      repo.load_existing_stats(old_commit_oid, old_stats)
+      repo
+    end
+
+    # Public: Initialize a new Repository to be analyzed for language
+    # data
+    #
+    # repo - a Rugged::Repository object
+    # commit_oid - the sha1 of the commit that will be analyzed;
+    #              this is usually the master branch
     #
     # Returns a Repository
-    def initialize(repo, commit_oid, existing_stats = nil)
+    def initialize(repo, commit_oid)
       @repository = repo
       @commit_oid = commit_oid
-      @old_commit_oid, @old_stats = existing_stats if existing_stats
+    end
+
+    # Public: Load the results of a previous analysis on this repository
+    # to speed up the new scan.
+    #
+    # The new analysis will be performed incrementally as to only take
+    # into account the file changes since the last time the repository
+    # was scanned
+    #
+    # old_commit_oid - the sha1 of the commit that was previously analyzed
+    # old_stats - the result of the previous analysis, obtained by calling
+    #             Repository#cache on the old repository
+    #
+    # Returns nothing
+    def load_existing_stats(old_commit_oid, old_stats)
+      @old_commit_oid = old_commit_oid
+      @old_stats = old_stats
+      nil
     end
 
     # Public: Returns a breakdown of language stats.
     #
     # Examples
     #
-    #   # => { Language['Ruby'] => 46319,
-    #          Language['JavaScript'] => 258 }
+    #   # => { 'Ruby' => 46319,
+    #          'JavaScript' => 258 }
     #
-    # Returns a Hash of Language keys and Integer size values.
+    # Returns a Hash of language names and Integer size values.
     def languages
       @sizes ||= begin
         sizes = Hash.new { 0 }
@@ -40,7 +69,7 @@ module Linguist
 
     # Public: Get primary Language of repository.
     #
-    # Returns a Language
+    # Returns a language name
     def language
       @language ||= begin
         primary = languages.max_by { |(_, size)| size }
@@ -56,6 +85,8 @@ module Linguist
     end
 
     # Public: Return the language breakdown of this repository by file
+    #
+    # Returns a map of language names => [filenames...]
     def breakdown_by_file
       @file_breakdown ||= begin
         breakdown = Hash.new { |h,k| h[k] = Array.new }
@@ -66,6 +97,23 @@ module Linguist
       end
     end
 
+    # Public: Return the cached results of the analysis
+    #
+    # This is a per-file breakdown that can be passed to other instances
+    # of Linguist::Repository to perform incremental scans
+    #
+    # Returns a map of filename => [language, size]
+    def cache
+      @cache ||= begin
+        if @old_commit_oid == @commit_oid
+          @old_stats
+        else
+          compute_stats(@old_commit_oid, @commit_oid, @old_stats)
+        end
+      end
+    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
@@ -96,15 +144,5 @@ module Linguist
 
       file_map
     end
-
-    def cache
-      @cache ||= begin
-        if @old_commit_oid == @commit_oid
-          @old_stats
-        else
-          compute_stats(@old_commit_oid, @commit_oid, @old_stats)
-        end
-      end
-    end
   end
 end
diff --git a/test/test_repository.rb b/test/test_repository.rb
index bd96e66d..6637f51f 100644
--- a/test/test_repository.rb
+++ b/test/test_repository.rb
@@ -40,8 +40,7 @@ class TestRepository < Test::Unit::TestCase
     assert old_repo.languages['Ruby'] > 10_000
     assert old_repo.size > 30_000
 
-    old_cache = [old_commit, old_repo.cache]
-    new_repo = Linguist::Repository.new(rugged_repository, master_oid, old_cache)
+    new_repo = Linguist::Repository.incremental(rugged_repository, master_oid, old_commit, old_repo.cache)
 
     assert new_repo.languages['Ruby'] > old_repo.languages['Ruby']
     assert new_repo.size > old_repo.size

From bc34345a56f977ac070675b57ad01fd76cf5778b Mon Sep 17 00:00:00 2001
From: Vicent Marti <tanoku@gmail.com>
Date: Thu, 26 Jun 2014 13:03:30 +0200
Subject: [PATCH 09/20] Fix the linguist binary

---
 bin/linguist               | 4 +++-
 lib/linguist/repository.rb | 2 ++
 2 files changed, 5 insertions(+), 1 deletion(-)

diff --git a/bin/linguist b/bin/linguist
index 2cfa8064..e086dcea 100755
--- a/bin/linguist
+++ b/bin/linguist
@@ -5,6 +5,7 @@
 
 require 'linguist/file_blob'
 require 'linguist/repository'
+require 'rugged'
 
 path = ARGV[0] || Dir.pwd
 
@@ -18,7 +19,8 @@ ARGV.shift
 breakdown = true if ARGV[0] == "--breakdown"
 
 if File.directory?(path)
-  repo = Linguist::Repository.from_directory(path)
+  rugged = Rugged::Repository.new(path)
+  repo = Linguist::Repository.new(rugged, rugged.head.target_id)
   repo.languages.sort_by { |_, size| size }.reverse.each do |language, size|
     percentage = ((size / repo.size.to_f) * 100)
     percentage = sprintf '%.2f' % percentage
diff --git a/lib/linguist/repository.rb b/lib/linguist/repository.rb
index 82e9ef47..e3d2d707 100644
--- a/lib/linguist/repository.rb
+++ b/lib/linguist/repository.rb
@@ -29,6 +29,8 @@ module Linguist
     def initialize(repo, commit_oid)
       @repository = repo
       @commit_oid = commit_oid
+
+      raise TypeError, 'commit_oid must be a commit SHA1' unless commit_oid.is_a?(String)
     end
 
     # Public: Load the results of a previous analysis on this repository

From 00a873dcc7ed88201b2a502f703e04784ddeccb8 Mon Sep 17 00:00:00 2001
From: Vicent Marti <tanoku@gmail.com>
Date: Thu, 26 Jun 2014 13:03:41 +0200
Subject: [PATCH 10/20] Bump 3.0.0b0

---
 lib/linguist/version.rb | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/lib/linguist/version.rb b/lib/linguist/version.rb
index 232278b4..42e0eb95 100644
--- a/lib/linguist/version.rb
+++ b/lib/linguist/version.rb
@@ -1,3 +1,3 @@
 module Linguist
-  VERSION = "2.13.0.github4"
+  VERSION = "3.0.0b0"
 end

From 324ac834891a31b4738cd7a8f17da2d2db93075d Mon Sep 17 00:00:00 2001
From: Vicent Marti <tanoku@gmail.com>
Date: Thu, 26 Jun 2014 14:12:00 +0200
Subject: [PATCH 11/20] Use the new Rugged release

---
 Gemfile                 | 1 -
 github-linguist.gemspec | 1 +
 2 files changed, 1 insertion(+), 1 deletion(-)

diff --git a/Gemfile b/Gemfile
index 3ac3ffa4..fa75df15 100644
--- a/Gemfile
+++ b/Gemfile
@@ -1,4 +1,3 @@
 source 'https://rubygems.org'
 
 gemspec
-gem 'rugged', :git => 'https://github.com/libgit2/rugged.git', branch: 'development', submodules: true
diff --git a/github-linguist.gemspec b/github-linguist.gemspec
index d4c2337a..936550f3 100644
--- a/github-linguist.gemspec
+++ b/github-linguist.gemspec
@@ -17,6 +17,7 @@ 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_development_dependency 'json'
   s.add_development_dependency 'mocha'

From 907d3c5a3676daee59e46cfbf43a716822c47199 Mon Sep 17 00:00:00 2001
From: Vicent Marti <tanoku@gmail.com>
Date: Thu, 26 Jun 2014 18:17:51 +0200
Subject: [PATCH 12/20] b1

---
 lib/linguist/version.rb | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/lib/linguist/version.rb b/lib/linguist/version.rb
index 42e0eb95..c5c3085d 100644
--- a/lib/linguist/version.rb
+++ b/lib/linguist/version.rb
@@ -1,3 +1,3 @@
 module Linguist
-  VERSION = "3.0.0b0"
+  VERSION = "3.0.0b1"
 end

From 621042e63936fbdecf1e526e26ef3ada91e3b7e0 Mon Sep 17 00:00:00 2001
From: Vicent Marti <tanoku@gmail.com>
Date: Thu, 26 Jun 2014 18:42:43 +0200
Subject: [PATCH 13/20] Remove whitespace

---
 Gemfile | 1 -
 1 file changed, 1 deletion(-)

diff --git a/Gemfile b/Gemfile
index fa75df15..851fabc2 100644
--- a/Gemfile
+++ b/Gemfile
@@ -1,3 +1,2 @@
 source 'https://rubygems.org'
-
 gemspec

From 12429b90fe02171cc035d16164ca723bb83f2213 Mon Sep 17 00:00:00 2001
From: Vicent Marti <tanoku@gmail.com>
Date: Thu, 26 Jun 2014 21:24:30 +0200
Subject: [PATCH 14/20] Bring back missing test

---
 test/test_heuristics.rb | 7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/test/test_heuristics.rb b/test/test_heuristics.rb
index 33f8a087..8bbf0695 100644
--- a/test/test_heuristics.rb
+++ b/test/test_heuristics.rb
@@ -1,6 +1,7 @@
 require 'linguist/heuristics'
 require 'linguist/language'
 require 'linguist/samples'
+require 'linguist/file_blob'
 
 require 'test/unit'
 
@@ -34,6 +35,12 @@ class TestHeuristcs < Test::Unit::TestCase
     assert_equal Language["C++"], results.first
   end
 
+  def test_detect_still_works_if_nothing_matches
+    blob = Linguist::FileBlob.new(File.join(samples_path, "Objective-C/hello.m"))
+    match = Language.detect(blob)
+    assert_equal Language["Objective-C"], match
+  end
+  
   def test_pl_prolog_by_heuristics
     languages = ["Perl", "Prolog"]
     results = Heuristics.disambiguate_pl(fixture("Prolog/turing.pl"), languages)

From 65eaf98d0badbb5f6012f9d5bf947e418e28f9a0 Mon Sep 17 00:00:00 2001
From: Vicent Marti <tanoku@gmail.com>
Date: Thu, 26 Jun 2014 21:26:26 +0200
Subject: [PATCH 15/20] docs

---
 lib/linguist/language.rb | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/lib/linguist/language.rb b/lib/linguist/language.rb
index ffbecb7b..ed9b68e9 100644
--- a/lib/linguist/language.rb
+++ b/lib/linguist/language.rb
@@ -92,6 +92,9 @@ module Linguist
 
     # Public: Detects the Language of the blob.
     #
+    # blob - an object that implements the Linguist `Blob` interface;
+    #       see Linguist::LazyBlob and Linguist::FileBlob for examples
+    #
     # Returns Language or nil.
     def self.detect(blob)
       name = blob.name.to_s

From d206131df0df57c9a77b88b4a6709ae65a8075b1 Mon Sep 17 00:00:00 2001
From: Vicent Marti <tanoku@gmail.com>
Date: Fri, 27 Jun 2014 13:51:37 +0200
Subject: [PATCH 16/20] Hardcode OIDs for test

---
 test/test_repository.rb | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/test/test_repository.rb b/test/test_repository.rb
index 6637f51f..f4c5ad70 100644
--- a/test/test_repository.rb
+++ b/test/test_repository.rb
@@ -8,7 +8,7 @@ class TestRepository < Test::Unit::TestCase
   end
 
   def master_oid
-    @master_oid ||= Rugged::Object.rev_parse_oid(rugged_repository, 'master')
+    'd40b4a33deba710e2f494db357c654fbe5d4b419'
   end
 
   def linguist_repo(oid = master_oid)
@@ -34,7 +34,7 @@ class TestRepository < Test::Unit::TestCase
   end
 
   def test_incremental_stats
-    old_commit = Rugged::Object.rev_parse_oid(rugged_repository, 'v2.0.0')
+    old_commit = '3d7364877d6794f6cc2a86b493e893968a597332'
     old_repo = linguist_repo(old_commit)
 
     assert old_repo.languages['Ruby'] > 10_000

From 32828a9af56d748225609e55e99e5c5df4c041cc Mon Sep 17 00:00:00 2001
From: Vicent Marti <tanoku@gmail.com>
Date: Fri, 27 Jun 2014 13:51:56 +0200
Subject: [PATCH 17/20] b2

---
 lib/linguist/version.rb | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/lib/linguist/version.rb b/lib/linguist/version.rb
index c5c3085d..eb7bbeb7 100644
--- a/lib/linguist/version.rb
+++ b/lib/linguist/version.rb
@@ -1,3 +1,3 @@
 module Linguist
-  VERSION = "3.0.0b1"
+  VERSION = "3.0.0b2"
 end

From d9be472ccb22ef2ccb5b131a80757bb9ab6c6b0c Mon Sep 17 00:00:00 2001
From: Vicent Marti <tanoku@gmail.com>
Date: Fri, 27 Jun 2014 16:41:23 +0200
Subject: [PATCH 18/20] Skip submodules when diffing

---
 lib/linguist/repository.rb | 7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)

diff --git a/lib/linguist/repository.rb b/lib/linguist/repository.rb
index e3d2d707..a89c81e6 100644
--- a/lib/linguist/repository.rb
+++ b/lib/linguist/repository.rb
@@ -131,8 +131,11 @@ module Linguist
         next if delta.binary
 
         if [:added, :modified].include? delta.status
-          mode = delta.new_file[:mode].to_s(8)
-          blob = Linguist::LazyBlob.new(repository, delta.new_file[:oid], new, mode)
+          # Skip submodules
+          mode = delta.new_file[:mode]
+          next if (mode & 040000) != 0
+
+          blob = Linguist::LazyBlob.new(repository, delta.new_file[:oid], new, mode.to_s(8))
 
           # Skip vendored or generated blobs
           next if blob.vendored? || blob.generated? || blob.language.nil?

From df09a746a0d76cbe2ebe8858b69cd0e3b61141c4 Mon Sep 17 00:00:00 2001
From: Vicent Marti <tanoku@gmail.com>
Date: Fri, 27 Jun 2014 16:57:58 +0200
Subject: [PATCH 19/20] b3

---
 lib/linguist/version.rb | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/lib/linguist/version.rb b/lib/linguist/version.rb
index eb7bbeb7..ab1a401e 100644
--- a/lib/linguist/version.rb
+++ b/lib/linguist/version.rb
@@ -1,3 +1,3 @@
 module Linguist
-  VERSION = "3.0.0b2"
+  VERSION = "3.0.0b3"
 end

From 9281bd043aec088e656fece3bc5b689fd1af5dc7 Mon Sep 17 00:00:00 2001
From: Arfon Smith <arfon@github.com>
Date: Tue, 1 Jul 2014 11:19:05 -0500
Subject: [PATCH 20/20] Version

---
 lib/linguist/version.rb | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/lib/linguist/version.rb b/lib/linguist/version.rb
index ab1a401e..c77f7734 100644
--- a/lib/linguist/version.rb
+++ b/lib/linguist/version.rb
@@ -1,3 +1,3 @@
 module Linguist
-  VERSION = "3.0.0b3"
+  VERSION = "3.0.0"
 end