mcs/class/referencesource/System.Web/Util/Utf16StringValidator.cs

   1 //------------------------------------------------------------------------------
   2 // <copyright file="Utf16StringValidator.cs" company="Microsoft">
   3 //     Copyright (c) Microsoft Corporation.  All rights reserved.
   4 // </copyright>
   5 //------------------------------------------------------------------------------
   6
   7 namespace System.Web.Util {
   8     using System;
   9
  10     // This class contains utility methods for dealing with security contexts when crossing AppDomain boundaries.
  11
  12     internal static class Utf16StringValidator {
  13
  14         private const char UNICODE_NULL_CHAR = '\0';
  15         private const char UNICODE_REPLACEMENT_CHAR = '\uFFFD';
  16
  17         private static readonly bool _skipUtf16Validation = AppSettings.AllowRelaxedUnicodeDecoding;
  18
  19         public static string ValidateString(string input) {
  20             return ValidateString(input, _skipUtf16Validation);
  21         }
  22
  23         // only internal for unit testing
  24         internal static string ValidateString(string input, bool skipUtf16Validation) {
  25             if (skipUtf16Validation || String.IsNullOrEmpty(input)) {
  26                 return input;
  27             }
  28
  29             // locate the first surrogate character
  30             int idxOfFirstSurrogate = -1;
  31             for (int i = 0; i < input.Length; i++) {
  32                 if (Char.IsSurrogate(input[i])) {
  33                     idxOfFirstSurrogate = i;
  34                     break;
  35                 }
  36             }
  37
  38             // fast case: no surrogates = return input string
  39             if (idxOfFirstSurrogate < 0) {
  40                 return input;
  41             }
  42
  43             // slow case: surrogates exist, so we need to validate them
  44             char[] chars = input.ToCharArray();
  45             for (int i = idxOfFirstSurrogate; i < chars.Length; i++) {
  46                 char thisChar = chars[i];
  47
  48                 // If this character is a low surrogate, then it was not preceded by
  49                 // a high surrogate, so we'll replace it.
  50                 if (Char.IsLowSurrogate(thisChar)) {
  51                     chars[i] = UNICODE_REPLACEMENT_CHAR;
  52                     continue;
  53                 }
  54
  55                 if (Char.IsHighSurrogate(thisChar)) {
  56                     // If this character is a high surrogate and it is followed by a
  57                     // low surrogate, allow both to remain.
  58                     if (i + 1 < chars.Length && Char.IsLowSurrogate(chars[i + 1])) {
  59                         i++; // skip the low surrogate also
  60                         continue;
  61                     }
  62
  63                     // If this character is a high surrogate and it is not followed
  64                     // by a low surrogate, replace it.
  65                     chars[i] = UNICODE_REPLACEMENT_CHAR;
  66                     continue;
  67                 }
  68
  69                 // Otherwise, this is a non-surrogate character and just move to the
  70                 // next character.
  71             }
  72             return new String(chars);
  73         }
  74
  75     }
  76 }