From 9f55273d5cacb6c1fa26a6d218eb40b5ecc8e12b Mon Sep 17 00:00:00 2001
From: Tietew <tietew@gmail.com>
Date: Fri, 14 Feb 2020 13:14:25 +0900
Subject: [PATCH] Convert source file encoding to UTF-8

---
 .rubocop.yml                 |  3 +++
 lib/simplecov/source_file.rb | 28 +++++++++++++++++++++++++-
 spec/fixtures/euc-jp.rb      |  2 ++
 spec/fixtures/utf-8-magic.rb |  2 ++
 spec/source_file_spec.rb     | 38 ++++++++++++++++++++++++++++++++++++
 5 files changed, 72 insertions(+), 1 deletion(-)
 create mode 100644 spec/fixtures/euc-jp.rb
 create mode 100644 spec/fixtures/utf-8-magic.rb

diff --git a/.rubocop.yml b/.rubocop.yml
index ce1bd2aa..9a1b76f9 100644
--- a/.rubocop.yml
+++ b/.rubocop.yml
@@ -1,6 +1,9 @@
 AllCops:
   Exclude:
     - "spec/fixtures/iso-8859.rb"
+    - "spec/fixtures/utf-8.rb"
+    - "spec/fixtures/utf-8-magic.rb"
+    - "spec/fixtures/euc-jp.rb"
     - "tmp/**/*"
     - "vendor/bundle/**/*"
     - "vendor/bundle/**/.*"
diff --git a/lib/simplecov/source_file.rb b/lib/simplecov/source_file.rb
index af0922d9..65897361 100644
--- a/lib/simplecov/source_file.rb
+++ b/lib/simplecov/source_file.rb
@@ -25,7 +25,7 @@ def project_filename
     def src
       # We intentionally read source code lazily to
       # suppress reading unused source code.
-      @src ||= File.open(filename, "rb", &:readlines)
+      @src ||= load_source
     end
     alias source src
 
@@ -175,6 +175,32 @@ def build_no_cov_chunks
       end
     end
 
+    def load_source
+      lines = []
+      # The default encoding is UTF-8
+      File.open(filename, "rb:UTF-8") do |file|
+        line = file.gets
+
+        # Check for shbang
+        if /\A#!/.match?(line)
+          lines << line
+          line = file.gets
+        end
+        return lines unless line
+
+        check_magic_comment(file, line)
+        lines.concat([line], file.readlines)
+      end
+    end
+
+    def check_magic_comment(file, line)
+      # Check for encoding magic comment
+      # Encoding magic comment must be placed at first line except for shbang
+      if (match = /\A#\s*(?:-\*-)?\s*(?:en)?coding:\s*(\S+)\s*(?:-\*-)?\s*\z/.match(line))
+        file.set_encoding(match[1], "UTF-8")
+      end
+    end
+
     def build_lines
       coverage_exceeding_source_warn if coverage_data["lines"].size > src.size
       lines = src.map.with_index(1) do |src, i|
diff --git a/spec/fixtures/euc-jp.rb b/spec/fixtures/euc-jp.rb
new file mode 100644
index 00000000..0eecd26e
--- /dev/null
+++ b/spec/fixtures/euc-jp.rb
@@ -0,0 +1,2 @@
+# encoding: EUC-JP
+puts "135¡ëC"
diff --git a/spec/fixtures/utf-8-magic.rb b/spec/fixtures/utf-8-magic.rb
new file mode 100644
index 00000000..64996ba2
--- /dev/null
+++ b/spec/fixtures/utf-8-magic.rb
@@ -0,0 +1,2 @@
+# encoding: UTF-8
+puts "135°C"
diff --git a/spec/source_file_spec.rb b/spec/source_file_spec.rb
index 48ad3cc2..b7ea394e 100644
--- a/spec/source_file_spec.rb
+++ b/spec/source_file_spec.rb
@@ -695,4 +695,42 @@
       end
     end
   end
+
+  context "a file contains non-ASCII characters" do
+    COVERAGE_FOR_SINGLE_LINE = {"lines" => [nil]}.freeze
+    COVERAGE_FOR_DOUBLE_LINES = {"lines" => [nil]}.freeze
+
+    shared_examples_for "converting to UTF-8" do
+      it "has all source lines of encoding UTF-8" do
+        subject.lines.each do |line|
+          expect(line.source.encoding).to eq(Encoding::UTF_8)
+          expect(line.source).to be_valid_encoding
+        end
+      end
+    end
+
+    describe "UTF-8 without magic comment" do
+      subject do
+        SimpleCov::SourceFile.new(source_fixture("utf-8.rb"), COVERAGE_FOR_SINGLE_LINE)
+      end
+
+      it_behaves_like "converting to UTF-8"
+    end
+
+    describe "UTF-8 with magic comment" do
+      subject do
+        SimpleCov::SourceFile.new(source_fixture("utf-8-magic.rb"), COVERAGE_FOR_DOUBLE_LINES)
+      end
+
+      it_behaves_like "converting to UTF-8"
+    end
+
+    describe "EUC-JP with magic comment" do
+      subject do
+        SimpleCov::SourceFile.new(source_fixture("euc-jp.rb"), COVERAGE_FOR_DOUBLE_LINES)
+      end
+
+      it_behaves_like "converting to UTF-8"
+    end
+  end
 end