IsStringUTF8 unittest and enforcing UTF-8 in JSON deserialization

base/json_reader.cc

 // Copyright (c) 2006-2008 The Chromium Authors. All rights reserved.
 // Use of this source code is governed by a BSD-style license that can be
 // found in the LICENSE file.
 
 #include "base/json_reader.h"
 
 #include "base/float_util.h"
 #include "base/logging.h"
 #include "base/string_util.h"
 #include "base/values.h"
@@ skipping 65 matching lines @@
                       Value** root,
                       bool allow_trailing_comma) {
   return JsonToValue(json, root, true, allow_trailing_comma);
 }
 
 /* static */
 bool JSONReader::JsonToValue(const std::string& json,
                              Value** root,
                              bool check_root,
                              bool allow_trailing_comma) {
-  // Assume input is UTF8.  The conversion from UTF8 to wstring removes null
-  // bytes for us (a good thing).
+  // The input must be in UTF-8.
+  if (!IsStringUTF8(json.c_str()))
+    return false;
+  // The conversion from UTF8 to wstring removes null bytes for us
+  // (a good thing).
   std::wstring json_wide(UTF8ToWide(json));
   const wchar_t* json_cstr = json_wide.c_str();
 
   // When the input JSON string starts with a UTF-8 Byte-Order-Mark
   // (0xEF, 0xBB, 0xBF), the UTF8ToWide() function converts it to a Unicode
   // BOM (U+FEFF). To avoid the JSONReader::BuildValue() function from
   // mis-treating a Unicode BOM as an invalid character and returning false,
   // skip a converted Unicode BOM if it exists.
   if (!json_wide.empty() && json_cstr[0] == 0xFEFF) {
     ++json_cstr;
@@ skipping 470 matching lines @@
         default:
           ++json_pos_;
       }
     }
   } else {
     return false;
   }
   return true;
 }