support for Universal Charcter Names #17

cpp/examples.cpp

@@ -443,4 +443,10 @@ int main() {
     [[noreturn]] void func2(/*syntax highlighting*/); // no syntax highlighting
     struct st { // syntax highlighting works now
     };
-    void func3();
+    void func3();
+
+// Universal character name
+class LineReader{};
+class Line_Reader{};
+class Line\u00b7Reader{};
+class Line\U0002070EReader{};

cpp/generate.rb

@@ -78,11 +78,11 @@ def blockFinderFor( name:"", tag_as:"", start_pattern:nil, needs_semicolon: true
             ]
         )
     end
-# 
-# 
+#
+#
 # Contexts
-# 
-# 
+#
+#
     preprocessor_context = [
         "#preprocessor-rule-enabled",
         "#preprocessor-rule-disabled",
@@ -278,13 +278,13 @@ def blockFinderFor( name:"", tag_as:"", start_pattern:nil, needs_semicolon: true
 #
     universal_character = /\\u[0-9a-fA-F]{4}/.or(/\\U000[0-9a-fA-F]/)
     first_character = /[a-zA-Z_]/.or(universal_character)
-    subsequent_character = /[a-zA-Z0-9]/.or(universal_character)
+    subsequent_character = /[a-zA-Z0-9_]/.or(universal_character)
     identifier = first_character.then(zeroOrMoreOf(subsequent_character))
     # todo: make a better name for this function
     variableBounds = ->(regex_pattern) do
         lookBehindToAvoid(@standard_character).then(regex_pattern).lookAheadToAvoid(@standard_character)
     end
-    variable_name_without_bounds = /[a-zA-Z_]#{@standard_character.without_default_mode_modifiers}*/
+    variable_name_without_bounds = identifier
     # word bounds are inefficient, but they are accurate
     variable_name = variableBounds[variable_name_without_bounds]