diff --git a/README.md b/README.md
index f90951c..0fce7f6 100644
--- a/README.md
+++ b/README.md
@@ -191,7 +191,10 @@ some input and returns a `ParserResult`.
 
 ## Running tests
 
-    $ ruby -Ilib:test test/test_parser.rb
+``` shell
+rake test
+ruby -Ilib:test test/test_parser.rb # per-file testing
+```
 
 # TODO
 Decent syntax error reporting, eg: Which line, which column failed.
diff --git a/Rakefile b/Rakefile
new file mode 100644
index 0000000..c94dfed
--- /dev/null
+++ b/Rakefile
@@ -0,0 +1,5 @@
+require 'rake/testtask'
+
+Rake::TestTask.new do |t|
+  t.test_files = FileList["test/test_*.rb"]
+end
diff --git a/lib/parser_result.rb b/lib/parser_result.rb
index 0d6fcfe..844a469 100644
--- a/lib/parser_result.rb
+++ b/lib/parser_result.rb
@@ -7,7 +7,7 @@ def initialize(success, remaining, matched, output=nil)
     @output = output.nil? ? [matched] : output
   end
 
-  def self.ok(output=nil, matched:, remaining:)
+  def self.ok(output=nil, matched: "", remaining: "")
 #    yield matched if block_given?
     output = [matched] if output.nil?
     ParserResult.new(true, remaining, matched, output)
diff --git a/test/test_base_parsers.rb b/test/test_base_parsers.rb
index 2c9b583..41becf5 100644
--- a/test/test_base_parsers.rb
+++ b/test/test_base_parsers.rb
@@ -132,7 +132,7 @@
       start(:letterOrNumber)
     end
 
-    assert_equal ["w", "8"], parser.run("w8").output
+    assert_equal [["w"], ["8"]], parser.run("w8").output
 
     parser = Grammar.build do
       rule(:letter)         { many1 { anyLetter } }
@@ -140,7 +140,7 @@
       start(:letterOrNumber)
     end
 
-    assert_equal ["w", "8"], parser.run("w8").output
+    assert_equal [["w"], ["8"]], parser.run("w8").output
   end
 
   it "uses regex" do