Throw exception when user-provided input contains BOM.

include/chaiscript/language/chaiscript_parser.hpp

@@ -114,6 +114,12 @@ namespace chaiscript
           // little SFINAE trick to avoid base class
           return Char_Parser_Helper<std::true_type>::u8str_from_ll(val);
         }
+        
+        static bool has_utf8_bom(const std::string &t_input)
+        {
+          //skip UTF-8 BOM
+          return ((t_input.size() > 3) && (t_input[0] == '\xef') && (t_input[1] == '\xbb' && t_input[2] == '\xbf'));
+        }
       };
     }
 
@@ -2563,19 +2569,16 @@ namespace chaiscript
         m_position = Position(t_input.begin(), t_input.end());
         m_filename = std::make_shared<std::string>(std::move(t_fname));
         
+        if (detail::Char_Parser_Helper<std::string>::has_utf8_bom(t_input)) {
+           throw exception::eval_error("UTF-8 in user provided input!");
+        }
+
         if ((t_input.size() > 1) && (t_input[0] == '#') && (t_input[1] == '!')) {
           while (m_position.has_more() && (!Eol())) {
             ++m_position;
           }
         }
 
-        //skip UTF-8 BOM
-        if ((t_input.size() > 3) && (t_input[0] == '\xef') && (t_input[1] == '\xbb' && t_input[2] == '\xbf')) {
-            while(m_position.has_more() && (m_position.col < 4)) {
-                ++m_position;
-            }
-        }
-
         if (Statements(true)) {
           if (m_position.has_more()) {
             throw exception::eval_error("Unparsed input", File_Position(m_position.line, m_position.col), *m_filename);

unittests/compiled_tests.cpp

@@ -355,12 +355,7 @@ TEST_CASE("Functor cast")
 TEST_CASE("BOM at beginning of string")
 {
   chaiscript::ChaiScript_Basic chai(create_chaiscript_stdlib(),create_chaiscript_parser());
-
+  CHECK_THROWS_AS(chai.eval<std::string>("\xef\xbb\xbfprint \"Hello World\""), chaiscript::exception::eval_error);
-  chai.add(chaiscript::fun(&functor_cast_test_call), "test_call");
-
-  chai.eval("def func() { return \"Hello World\"; };");
-
-  CHECK(chai.eval<std::string>("\xef\xbb\xbf(func())") == "Hello World");
 }
 
 

unittests/eval_file_with_bom.chai

@@ -0,0 +1,2 @@
+eval_file("file_with_bom.inc")
+assert_true(alwaysTrue())

unittests/file_with_bom.inc

@@ -0,0 +1,3 @@
+def alwaysTrue() {
+   return true
+}