parser.rb in regexp-examples-0.2.4

- old
+ new

@@ -52,15 +52,20 @@
         @current_position += ($1.length + 2)
         group = parse_backreference_group($1)
       when BackslashCharMap.keys.include?(regexp_string[@current_position])
         group = CharGroup.new(
           BackslashCharMap[regexp_string[@current_position]])
-      when rest_of_string =~ /\Ac(.)/ # Control character
-        @current_position += 1
-        group = parse_single_char_group( parse_control_character($1) )
-        # TODO: There are also a bunch of multi-char matches to watch out for:
+      when rest_of_string =~ /\A(c|C-)(.)/ # Control character
         # http://en.wikibooks.org/wiki/Ruby_Programming/Syntax/Literals
+        @current_position += $1.length
+        group = parse_single_char_group( parse_control_character($2) )
+      when rest_of_string =~ /\Ax(\h{1,2})/ # Escape sequence
+        @current_position += $1.length
+        group = parse_single_char_group( parse_escape_sequence($1) )
+      when rest_of_string =~ /\Au(\h{4})/ # Unicode sequence
+        @current_position += 4
+        group = parse_single_char_group( parse_unicode_sequence($1) )
       else
         group = parse_single_char_group( regexp_string[@current_position] )
         # TODO: What about cases like \A, \z, \Z ?
       end
       group
@@ -150,11 +155,19 @@
     def parse_backreference_group(match)
       BackReferenceGroup.new(match)
     end
 
     def parse_control_character(char)
-      # TODO: Is there a better way of doing this? I.e. not just brute-force
-      (0..255).detect { |x| x.chr =~ Regexp.new("\\c#{char}") }.chr
+      (char.ord % 32).chr # Black magic!
+      # eval "?\\C-#{char.chr}" # Doesn't work for e.g. char = "?"
+    end
+
+    def parse_escape_sequence(match)
+      eval "?\\x#{match}"
+    end
+
+    def parse_unicode_sequence(match)
+      eval "?\\u#{match}"
     end
 
     def parse_star_repeater(group)
       @current_position += 1
       StarRepeater.new(group)