mcs/class/corlib/System.Globalization/StringInfo.cs

   1 //
   2 // System.Globalization.StringInfo.cs
   3 //
   4 // Author:
   5 //      Dick Porter (dick@ximian.com)
   6 //
   7 // (C) 2002 Ximian, Inc.
   8 // (C) 2004 Novell, Inc.
   9 //
  10
  11 //
  12 // Copyright (C) 2004 Novell, Inc (http://www.novell.com)
  13 //
  14 // Permission is hereby granted, free of charge, to any person obtaining
  15 // a copy of this software and associated documentation files (the
  16 // "Software"), to deal in the Software without restriction, including
  17 // without limitation the rights to use, copy, modify, merge, publish,
  18 // distribute, sublicense, and/or sell copies of the Software, and to
  19 // permit persons to whom the Software is furnished to do so, subject to
  20 // the following conditions:
  21 //
  22 // The above copyright notice and this permission notice shall be
  23 // included in all copies or substantial portions of the Software.
  24 //
  25 // THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
  26 // EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
  27 // MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
  28 // NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE
  29 // LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION
  30 // OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION
  31 // WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
  32 //
  33
  34 using System.Collections;
  35
  36 namespace System.Globalization {
  37
  38         [Serializable]
  39         public class StringInfo {
  40                 public StringInfo()
  41                 {
  42                 }
  43
  44                 public static string GetNextTextElement(string str)
  45                 {
  46                         if(str == null || str.Length == 0) {
  47                                 throw new ArgumentNullException("string is null");
  48                         }
  49                         return(GetNextTextElement (str, 0));
  50                 }
  51
  52                 public static string GetNextTextElement(string str, int index)
  53                 {
  54                         if(str == null) {
  55                                 throw new ArgumentNullException("string is null");
  56                         }
  57
  58                         if(index < 0 || index >= str.Length) {
  59                                 throw new ArgumentOutOfRangeException ("Index is not valid");
  60                         }
  61
  62                         /* Find the next base character, surrogate
  63                          * pair or combining character sequence
  64                          */
  65
  66                         char ch = str[index];
  67                         UnicodeCategory cat = char.GetUnicodeCategory (ch);
  68
  69                         if (cat == UnicodeCategory.Surrogate) {
  70                                 /* Check that it's a high surrogate
  71                                  * followed by a low surrogate
  72                                  */
  73                                 if (ch >= 0xD800 && ch <= 0xDBFF) {
  74                                         if ((index + 1) < str.Length &&
  75                                             str[index + 1] >= 0xDC00 &&
  76                                             str[index + 1] <= 0xDFFF) {
  77                                                 /* A valid surrogate pair */
  78                                                 return(str.Substring (index, 2));
  79                                         } else {
  80                                                 /* High surrogate on its own */
  81                                                 return(new String (ch, 1));
  82                                         }
  83                                 } else {
  84                                         /* Low surrogate on its own */
  85                                         return(new String (ch, 1));
  86                                 }
  87                         } else {
  88                                 /* Look for a base character, which
  89                                  * may or may not be followed by a
  90                                  * series of combining characters
  91                                  */
  92
  93                                 if (cat == UnicodeCategory.NonSpacingMark ||
  94                                     cat == UnicodeCategory.SpacingCombiningMark ||
  95                                     cat == UnicodeCategory.EnclosingMark) {
  96                                         /* Not a base character */
  97                                         return(new String (ch, 1));
  98                                 }
  99
 100                                 int count = 1;
 101
 102                                 while (index + count < str.Length) {
 103                                         cat = char.GetUnicodeCategory (str[index + count]);
 104                                         if (cat != UnicodeCategory.NonSpacingMark &&
 105                                             cat != UnicodeCategory.SpacingCombiningMark &&
 106                                             cat != UnicodeCategory.EnclosingMark) {
 107                                                 /* Finished the sequence */
 108                                                 break;
 109                                         }
 110                                         count++;
 111                                 }
 112
 113                                 return(str.Substring (index, count));
 114                         }
 115                 }
 116
 117                 public static TextElementEnumerator GetTextElementEnumerator(string str)
 118                 {
 119                         if(str == null || str.Length == 0) {
 120                                 throw new ArgumentNullException("string is null");
 121                         }
 122                         return(new TextElementEnumerator (str, 0));
 123                 }
 124
 125                 public static TextElementEnumerator GetTextElementEnumerator(string str, int index)
 126                 {
 127                         if(str == null) {
 128                                 throw new ArgumentNullException("string is null");
 129                         }
 130
 131                         if(index < 0 || index >= str.Length) {
 132                                 throw new ArgumentOutOfRangeException ("Index is not valid");
 133                         }
 134
 135                         return(new TextElementEnumerator (str, index));
 136                 }
 137
 138                 public static int[] ParseCombiningCharacters(string str)
 139                 {
 140                         if(str == null) {
 141                                 throw new ArgumentNullException("string is null");
 142                         }
 143
 144                         ArrayList indices = new ArrayList (str.Length);
 145                         TextElementEnumerator tee = GetTextElementEnumerator (str);
 146
 147                         tee.Reset ();
 148                         while(tee.MoveNext ()) {
 149                                 indices.Add (tee.ElementIndex);
 150                         }
 151
 152                         return((int[])indices.ToArray (typeof (int)));
 153                 }
 154         }
 155 }