Import the gitlab-elasticsearch-git gem unmodified

Gemfile

@@ -113,7 +113,6 @@ gem 'seed-fu', '~> 2.3.5'
 gem 'elasticsearch-model', '~> 0.1.9'
 gem 'elasticsearch-rails', '~> 0.1.9'
 gem 'elasticsearch-api',   '5.0.3'
-gem 'gitlab-elasticsearch-git', '1.2.0', require: "elasticsearch/git"
 gem 'aws-sdk'
 gem 'faraday_middleware-aws-signers-v4'
 

Gemfile.lock

@@ -287,14 +287,6 @@ GEM
       mime-types (>= 1.19)
       rugged (>= 0.23.0b)
     github-markup (1.4.0)
-    gitlab-elasticsearch-git (1.2.0)
-      activemodel (~> 4.2)
-      activesupport (~> 4.2)
-      charlock_holmes (~> 0.7)
-      elasticsearch-api
-      elasticsearch-model (~> 0.1.9)
-      github-linguist (~> 4.7)
-      rugged (~> 0.24)
     gitlab-flowdock-git-hook (1.0.1)
       flowdock (~> 0.7)
       gitlab-grit (>= 2.4.1)
@@ -949,7 +941,6 @@ DEPENDENCIES
   gemojione (~> 3.0)
   gitaly (~> 0.5.0)
   github-linguist (~> 4.7.0)
-  gitlab-elasticsearch-git (= 1.2.0)
   gitlab-flowdock-git-hook (~> 1.0.1)
   gitlab-license (~> 1.0)
   gitlab-markup (~> 1.5.1)

lib/elasticsearch/git.rb

+require "elasticsearch/git/version"
+require "elasticsearch/git/model"
+require "elasticsearch/git/repository"
+
+module Elasticsearch
+  module Git
+  end
+end
+

lib/elasticsearch/git/encoder_helper.rb

+require 'active_support/concern'
+require 'charlock_holmes'
+
+module Elasticsearch
+  module Git
+    module EncoderHelper
+      extend ActiveSupport::Concern
+
+      included do
+        def encode!(message)
+          return nil unless message.respond_to? :force_encoding
+
+          # if message is utf-8 encoding, just return it
+          message.force_encoding("UTF-8")
+          return message if message.valid_encoding?
+
+          # return message if message type is binary
+          detect = CharlockHolmes::EncodingDetector.detect(message)
+          return message.force_encoding("BINARY") if detect && detect[:type] == :binary
+
+          # encoding message to detect encoding
+          if detect && detect[:encoding]
+            message.force_encoding(detect[:encoding])
+          end
+
+          # encode and clean the bad chars
+          message.replace clean(message)
+        rescue
+          encoding = detect ? detect[:encoding] : "unknown"
+          "--broken encoding: #{encoding}"
+        end
+
+        private
+
+        def clean(message)
+          message.encode("UTF-16BE", undef: :replace, invalid: :replace, replace: "")
+          .encode("UTF-8")
+          .gsub("\0".encode("UTF-8"), "")
+        end
+      end
+    end
+  end
+end

lib/elasticsearch/git/lite_blob.rb

+require 'linguist'
+require 'elasticsearch/git/encoder_helper'
+
+module Elasticsearch
+  module Git
+    class LiteBlob
+      include Linguist::BlobHelper
+      include Elasticsearch::Git::EncoderHelper
+
+      attr_accessor :id, :name, :path, :size, :mode, :commit_id
+      attr_writer :data
+
+      def initialize(repo, raw_blob_hash)
+        @id   = raw_blob_hash[:oid]
+        @blob  = repo.lookup(@id)
+
+        @mode = raw_blob_hash[:mode].to_s(8)
+        @size = @blob.size
+        @path = encode!(raw_blob_hash[:path])
+        @name = @path.split('/').last
+      end
+
+      def data
+        @data ||= encode!(@blob.content)
+      end
+    end
+  end
+end

lib/elasticsearch/git/model.rb

+require 'active_support/concern'
+require 'active_model'
+require 'elasticsearch/model'
+
+module Elasticsearch
+  module Git
+    module Model
+      extend ActiveSupport::Concern
+
+      included do
+        extend ActiveModel::Naming
+        include ActiveModel::Model
+        include Elasticsearch::Model
+
+        env = if defined?(::Rails)
+                ::Rails.env.to_s
+              else
+                nil
+              end
+
+        index_name [self.name.downcase, 'index', env].compact.join('-')
+
+        settings \
+          index: {
+          analysis: {
+            analyzer: {
+              path_analyzer: {
+                type: 'custom',
+                tokenizer: 'path_tokenizer',
+                filter: %w(lowercase asciifolding)
+              },
+              sha_analyzer: {
+                type: 'custom',
+                tokenizer: 'sha_tokenizer',
+                filter: %w(lowercase asciifolding)
+              },
+              code_analyzer: {
+                type: 'custom',
+                tokenizer: 'standard',
+                filter: %w(code lowercase asciifolding),
+                char_filter: ["code_mapping"]
+              },
+              code_search_analyzer: {
+                type: 'custom',
+                tokenizer: 'standard',
+                filter: %w(lowercase asciifolding),
+                char_filter: ["code_mapping"]
+              }
+            },
+            tokenizer: {
+              sha_tokenizer: {
+                type: "edgeNGram",
+                min_gram: 5,
+                max_gram: 40,
+                token_chars: %w(letter digit)
+              },
+              path_tokenizer: {
+                type: 'path_hierarchy',
+                reverse: true
+              },
+            },
+            filter: {
+              code: {
+                type: "pattern_capture",
+                preserve_original: 1,
+                patterns: [
+                  "(\\p{Ll}+|\\p{Lu}\\p{Ll}+|\\p{Lu}+)",
+                  "(\\d+)"
+                ]
+              }
+            },
+            char_filter: {
+              code_mapping: {
+                type: "mapping",
+                mappings: [
+                  ". => ' '"
+                ]
+              }
+            },
+          }
+        }
+      end
+    end
+  end
+end