Lex comments in chunks

Similar to this being added for CDATA tags in
8acc7fc743 comments are now also lexed in
chunks.

Related issue: #93

ext/c/lexer.rl

@@ -93,7 +93,9 @@ VALUE oga_xml_lexer_advance(VALUE self, VALUE data_block)
     ID id_on_cdata_start      = rb_intern("on_cdata_start");
     ID id_on_cdata_body       = rb_intern("on_cdata_body");
     ID id_on_cdata_end        = rb_intern("on_cdata_end");
-    ID id_on_comment          = rb_intern("on_comment");
+    ID id_on_comment_start    = rb_intern("on_comment_start");
+    ID id_on_comment_body     = rb_intern("on_comment_body");
+    ID id_on_comment_end      = rb_intern("on_comment_end");
     ID id_on_doctype_end      = rb_intern("on_doctype_end");
     ID id_on_doctype_inline   = rb_intern("on_doctype_inline");
     ID id_on_doctype_name     = rb_intern("on_doctype_name");

ext/java/org/liboga/xml/Lexer.rl

@@ -107,7 +107,9 @@ public class Lexer extends RubyObject
         String id_on_cdata_start      = "on_cdata_start";
         String id_on_cdata_body       = "on_cdata_body";
         String id_on_cdata_end        = "on_cdata_end";
-        String id_on_comment          = "on_comment";
+        String id_on_comment_start    = "on_comment_start";
+        String id_on_comment_body     = "on_comment_body";
+        String id_on_comment_end      = "on_comment_end";
         String id_on_doctype_end      = "on_doctype_end";
         String id_on_doctype_inline   = "on_doctype_inline";
         String id_on_doctype_name     = "on_doctype_name";

ext/ragel/base_lexer.rl

@@ -67,12 +67,35 @@
 
     comment_start = '<!--';
     comment_end   = '-->';
-    comment       = comment_start (any* -- comment_end) comment_end;
+
+    # Everything except "-" OR a single "-"
+    comment_allowed = (^'-'+ | '-') $count_newlines;
 
     action start_comment {
-        callback(id_on_comment, data, encoding, ts + 4, te - 3);
+        callback_simple(id_on_comment_start);
+
+        fnext comment_body;
     }
 
+    comment_body := |*
+        comment_allowed => {
+            callback(id_on_comment_body, data, encoding, ts, te);
+
+            if ( lines > 0 )
+            {
+                advance_line(lines);
+
+                lines = 0;
+            }
+        };
+
+        comment_end => {
+            callback_simple(id_on_comment_end);
+
+            fnext main;
+        };
+    *|;
+
     # CDATA
     #
     # http://www.w3.org/TR/html-markup/syntax.html#cdata-sections
@@ -461,7 +484,7 @@
     main := |*
         doctype_start  => start_doctype;
         xml_decl_start => start_xml_decl;
-        comment        => start_comment;
+        comment_start  => start_comment;
         cdata_start    => start_cdata;
         proc_ins_start => start_proc_ins;
         element_start  => start_element;