Handle encoding in non-binary Blob instancesupdate-gitlab-git

gitlab_git 10.6.4 relies on Rugged marking blobs as binary or not,
instead of relying on Linguist. Linguist in turn would mark text blobs
as binary whenever they would contain byte sequences that could not be
encoded using UTF-8.

However, marking such blobs as binary is not correct. If one pushes a
Markdown document with invalid character sequences it's still a text
based Markdown document and not some random binary blob.

This commit overwrites Blob#data so it automatically converts text-based
content to UTF-8 (the encoding we use everywhere else) while taking care
of replacing any invalid sequences with the UTF-8 replacement character.
The data of binary blobs is left as-is.

2 files changed, 32 insertions, 0 deletions

diff --git a/app/models/blob.rb b/app/models/blob.rb
index 12cc5aaafba..ab92e820335 100644
--- a/app/models/blob.rb
+++ b/app/models/blob.rb
@@ -22,6 +22,18 @@ class Blob < SimpleDelegator
     new(blob)
   end
 
+  # Returns the data of the blob.
+  #
+  # If the blob is a text based blob the content is converted to UTF-8 and any
+  # invalid byte sequences are replaced.
+  def data
+    if binary?
+      super
+    else
+      @data ||= super.encode(Encoding::UTF_8, invalid: :replace, undef: :replace)
+    end
+  end
+
   def no_highlighting?
     size && size > 1.megabyte
   end
diff --git a/spec/models/blob_spec.rb b/spec/models/blob_spec.rb
index cee20234e1f..03d02b4d382 100644
--- a/spec/models/blob_spec.rb
+++ b/spec/models/blob_spec.rb
@@ -1,3 +1,4 @@
+# encoding: utf-8
 require 'rails_helper'
 
 describe Blob do
@@ -7,6 +8,25 @@ describe Blob do
     end
   end
 
+  describe '#data' do
+    context 'using a binary blob' do
+      it 'returns the data as-is' do
+        data = "\n\xFF\xB9\xC3"
+        blob = described_class.new(double(binary?: true, data: data))
+
+        expect(blob.data).to eq(data)
+      end
+    end
+
+    context 'using a text blob' do
+      it 'converts the data to UTF-8' do
+        blob = described_class.new(double(binary?: false, data: "\n\xFF\xB9\xC3"))
+
+        expect(blob.data).to eq("\n���")
+      end
+    end
+  end
+
   describe '#svg?' do
     it 'is falsey when not text' do
       git_blob = double(text?: false)