From 56341b55857879542ac2e2a8b0789cd09ad03eb9 Mon Sep 17 00:00:00 2001
From: Yorick Peterse <yorickpeterse@gmail.com>
Date: Sat, 16 Aug 2014 16:03:55 +0200
Subject: [PATCH] Cleaned up lexing of comments/cdata.

Thanks to @whitequark for suggesting the use of the "--" operator.
---
 ext/ragel/base_lexer.rl | 39 +++++++--------------------------------
 1 file changed, 7 insertions(+), 32 deletions(-)

diff --git a/ext/ragel/base_lexer.rl b/ext/ragel/base_lexer.rl
index 850e11f..e38705e 100644
--- a/ext/ragel/base_lexer.rl
+++ b/ext/ragel/base_lexer.rl
@@ -50,25 +50,12 @@
 
     comment_start = '<!--';
     comment_end   = '-->';
+    comment       = comment_start (any* -- comment_end) comment_end;
 
     action start_comment {
-        mark = te;
-
-        fnext comment_body;
+        callback("on_comment", data, encoding, ts + 4, te - 3);
     }
 
-    comment_body := |*
-        comment_end => {
-            callback("on_comment", data, encoding, mark, te - 3);
-
-            mark = 0;
-
-            fnext main;
-        };
-
-        any;
-    *|;
-
     # CDATA
     #
     # http://www.w3.org/TR/html-markup/syntax.html#cdata-sections
@@ -79,25 +66,12 @@
 
     cdata_start = '<![CDATA[';
     cdata_end   = ']]>';
+    cdata       = cdata_start (any* -- cdata_end) cdata_end;
 
     action start_cdata {
-        mark = te;
-
-        fnext cdata_body;
+        callback("on_cdata", data, encoding, ts + 9, te - 3);
     }
 
-    cdata_body := |*
-        cdata_end => {
-            callback("on_cdata", data, encoding, mark, te - 3);
-
-            mark = 0;
-
-            fnext main;
-        };
-
-        any;
-    *|;
-
     # Processing Instructions
     #
     # http://www.w3.org/TR/xpath/#section-Processing-Instruction-Nodes
@@ -289,10 +263,11 @@
     main := |*
         doctype_start  => start_doctype;
         xml_decl_start => start_xml_decl;
-        comment_start  => start_comment;
-        cdata_start    => start_cdata;
+        comment        => start_comment;
+        cdata          => start_cdata;
         proc_ins_start => start_proc_ins;
 
+
         # The start of an element.
         '<' => start_element;