Remove dependencies on Linguist

This saves about 128 MB of baseline RAM usage per Unicorn and
Sidekiq process (!).

Linguist wasn't detecting languages anymore from CE/EE since
9ae8b57467ac8b38f1fa9020a466d94a93cbb9dd. However, Linguist::BlobHelper
was still being depended on by BlobLike and others.

This removes the Linguist gem, given it isn't required anymore.
EscapeUtils were pulled in as dependency, but given Banzai depends on
it, it is now added explicitly.

Previously, Linguist was used to detect the best ACE mode. Instead,
we rely on ACE to guess the best mode based on the file extension.

4 files changed, 147 insertions, 24 deletions

diff --git a/spec/lib/gitlab/blob_helper_spec.rb b/spec/lib/gitlab/blob_helper_spec.rb
new file mode 100644
index 00000000000..0b56f8687c3
--- /dev/null
+++ b/spec/lib/gitlab/blob_helper_spec.rb
@@ -0,0 +1,125 @@
+# frozen_string_literal: true
+
+require 'spec_helper'
+
+describe Gitlab::BlobHelper do
+  include FakeBlobHelpers
+
+  let(:project) { create(:project) }
+  let(:blob) { fake_blob(path: 'file.txt') }
+  let(:large_blob) { fake_blob(path: 'test.pdf', size: 2.megabytes, binary: true) }
+
+  describe '#extname' do
+    it 'returns the extension' do
+      expect(blob.extname).to eq('.txt')
+    end
+  end
+
+  describe '#known_extension?' do
+    it 'returns true' do
+      expect(blob.known_extension?).to be_truthy
+    end
+  end
+
+  describe '#viewable' do
+    it 'returns true' do
+      expect(blob.viewable?).to be_truthy
+    end
+
+    it 'returns false' do
+      expect(large_blob.viewable?).to be_falsey
+    end
+  end
+
+  describe '#large?' do
+    it 'returns false' do
+      expect(blob.large?).to be_falsey
+    end
+
+    it 'returns true' do
+      expect(large_blob.large?).to be_truthy
+    end
+  end
+
+  describe '#binary?' do
+    it 'returns true' do
+      expect(large_blob.binary?).to be_truthy
+    end
+
+    it 'returns false' do
+      expect(blob.binary?).to be_falsey
+    end
+  end
+
+  describe '#text?' do
+    it 'returns true' do
+      expect(blob.text?).to be_truthy
+    end
+
+    it 'returns false' do
+      expect(large_blob.text?).to be_falsey
+    end
+  end
+
+  describe '#image?' do
+    it 'returns false' do
+      expect(blob.image?).to be_falsey
+    end
+  end
+
+  describe '#mime_type' do
+    it 'returns text/plain' do
+      expect(blob.mime_type).to eq('text/plain')
+    end
+
+    it 'returns application/pdf' do
+      expect(large_blob.mime_type).to eq('application/pdf')
+    end
+  end
+
+  describe '#binary_mime_type?' do
+    it 'returns false' do
+      expect(blob.binary_mime_type?).to be_falsey
+    end
+  end
+
+  describe '#lines' do
+    it 'returns the payload in an Array' do
+      expect(blob.lines).to eq(['foo'])
+    end
+  end
+
+  describe '#content_type' do
+    it 'returns text/plain' do
+      expect(blob.content_type).to eq('text/plain; charset=utf-8')
+    end
+
+    it 'returns text/plain' do
+      expect(large_blob.content_type).to eq('application/pdf')
+    end
+  end
+
+  describe '#encoded_newlines_re' do
+    it 'returns a regular expression' do
+      expect(blob.encoded_newlines_re).to eq(/\r\n|\r|\n/)
+    end
+  end
+
+  describe '#ruby_encoding' do
+    it 'returns UTF-8' do
+      expect(blob.ruby_encoding).to eq('UTF-8')
+    end
+  end
+
+  describe '#encoding' do
+    it 'returns UTF-8' do
+      expect(blob.ruby_encoding).to eq('UTF-8')
+    end
+  end
+
+  describe '#empty?' do
+    it 'returns false' do
+      expect(blob.empty?).to be_falsey
+    end
+  end
+end
diff --git a/spec/lib/gitlab/conflict/file_spec.rb b/spec/lib/gitlab/conflict/file_spec.rb
index 1bd077ddbdf..a955ce54e85 100644
--- a/spec/lib/gitlab/conflict/file_spec.rb
+++ b/spec/lib/gitlab/conflict/file_spec.rb
@@ -267,11 +267,6 @@ FILE
       it 'includes the full content of the conflict' do
         expect(conflict_file.as_json(full_content: true)).to have_key(:content)
       end
-
-      it 'includes the detected language of the conflict file' do
-        expect(conflict_file.as_json(full_content: true)[:blob_ace_mode])
-          .to eq('ruby')
-      end
     end
   end
 end
diff --git a/spec/lib/gitlab/git/blob_snippet_spec.rb b/spec/lib/gitlab/git/blob_snippet_spec.rb
deleted file mode 100644
index 6effec8295c..00000000000
--- a/spec/lib/gitlab/git/blob_snippet_spec.rb
+++ /dev/null
@@ -1,19 +0,0 @@
-# encoding: UTF-8
-
-require "spec_helper"
-
-describe Gitlab::Git::BlobSnippet, :seed_helper do
-  describe '#data' do
-    context 'empty lines' do
-      let(:snippet) { Gitlab::Git::BlobSnippet.new('master', nil, nil, nil) }
-
-      it { expect(snippet.data).to be_nil }
-    end
-
-    context 'present lines' do
-      let(:snippet) { Gitlab::Git::BlobSnippet.new('master', %w(wow much), 1, 'wow.rb') }
-
-      it { expect(snippet.data).to eq("wow\nmuch") }
-    end
-  end
-end
diff --git a/spec/lib/gitlab/language_data_spec.rb b/spec/lib/gitlab/language_data_spec.rb
new file mode 100644
index 00000000000..b08150855fe
--- /dev/null
+++ b/spec/lib/gitlab/language_data_spec.rb
@@ -0,0 +1,22 @@
+# frozen_string_literal: true
+
+require 'spec_helper'
+
+describe Gitlab::LanguageData do
+  describe '#extensions' do
+    before do
+      described_class.clear_extensions!
+    end
+
+    it 'loads the extensions once' do
+      expect(YAML).to receive(:load_file).once.and_call_original
+
+      2.times do
+        expect(described_class.extensions).to be_a(Set)
+        expect(described_class.extensions.count).to be > 0
+        # Sanity check for known extensions
+        expect(described_class.extensions).to include(*%w(.rb .yml .json))
+      end
+    end
+  end
+end