GitlabCi::Encode lib

2 files changed, 38 insertions, 7 deletions

diff --git a/app/models/build.rb b/app/models/build.rb
index e3ba323..813d585 100644
--- a/app/models/build.rb
+++ b/app/models/build.rb
@@ -94,15 +94,15 @@ class Build < ActiveRecord::Base
   end
 
   def sanitize_build_output(output)
-    output.detect_encoding!.encode!('utf-8', invalid: :replace)
+    GitlabCi::Encode.encode!(output)
   end
 
   def read_tmp_file
-    if tmp_file && File.readable?(tmp_file)
-      File.read(tmp_file)
-    else
-      ''
-    end
+    content = if tmp_file && File.readable?(tmp_file)
+                File.read(tmp_file)
+              end
+
+    content ||= ''
   end
 
   def compose_output
@@ -110,7 +110,7 @@ class Build < ActiveRecord::Base
 
     if running?
       sanitized_output = sanitize_build_output(read_tmp_file)
-      output << sanitized_output
+      output << sanitized_output if sanitized_output.present?
     end
 
     output
diff --git a/lib/gitlab_ci/encode.rb b/lib/gitlab_ci/encode.rb
new file mode 100644
index 0000000..6cf84fd
--- /dev/null
+++ b/lib/gitlab_ci/encode.rb
@@ -0,0 +1,31 @@
+module GitlabCi
+  module Encode
+    extend self
+
+    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 if detect[:type] == :binary
+
+      # if message is not utf-8 encoding, convert it
+      if detect[:encoding]
+        message.force_encoding(detect[:encoding])
+        message.encode!("UTF-8", detect[:encoding], undef: :replace, replace: "", invalid: :replace)
+      end
+
+      # ensure message encoding is utf8
+      message.valid_encoding? ? message : raise
+
+      # Prevent app from crash cause of encoding errors
+    rescue
+      encoding = detect ? detect[:encoding] : "unknown"
+      "--broken encoding: #{encoding}"
+    end
+  end
+end