Multiline ruby

comment_parser/parsers/ruby_parser.py

@@ -21,7 +21,9 @@ def extract_comments(code):
   """
   pattern = r"""
     (?P<literal> ([\"'])((?:\\\2|(?:(?!\2)).)*)(\2)) |
-    (?P<single> \#(?P<single_content>.*?)$)
+    (?P<single> \#(?P<single_content>.*?)$) |
+    (?P<multi> ^=begin\n(?P<multi_content>(.|\n)*?)?\n=end$) |
+    (?P<error> ^=begin$\*(.*)?)
   """
   compiled = re.compile(pattern, re.VERBOSE | re.MULTILINE)
 
@@ -40,5 +42,10 @@ def extract_comments(code):
       comment_content = match.group("single_content")
       comment = common.Comment(comment_content, line_no + 1)
       comments.append(comment)
-
+    elif kind == "multi":
+      comment_content = match.group("multi_content")
+      comment = common.Comment(comment_content, line_no + 1, multiline=True)
+      comments.append(comment)
+    elif kind == "error":
+      raise common.UnterminatedCommentError()
   return comments

comment_parser/parsers/tests/ruby_parser_test.py

@@ -6,7 +6,7 @@
 from comment_parser.parsers import ruby_parser
 
 
-class ShellParserTest(unittest.TestCase):
+class RubyParserTest(unittest.TestCase):
 
   def testComment(self):
     code = '# comment'
@@ -63,3 +63,15 @@ def testDifferentLiteralsSeparatedByComment(self):
     comments = ruby_parser.extract_comments(code)
     expected = [common.Comment(code[11:], 1, multiline=False)]
     self.assertEqual(comments, expected)
+
+  def testMultilineComment(self):
+    code = '=begin\nThis is a multiline comment.' \
+           'It is not terminated by =end \n=end'
+    comments = ruby_parser.extract_comments(code)
+    expected = [
+        common.Comment(
+            "This is a multiline comment.It is not terminated by =end ",
+            1,
+            multiline=True)
+    ]
+    self.assertEqual(comments, expected)