/// <summary>
/// Transforms input into a phonetic transform that is similar to analyzed data
/// </summary>
/// <param name="source">source term</param>
/// <returns>Analyzed transformation</returns>
public string Transform(string source)
{
int maxSizeNeeded = 4 * source.Length;
char[] poutput = new char[ArrayUtil.GetNextSize(maxSizeNeeded)];
int len = PhoneticTransform(poutput, source.ToCharArray(), source.Length);
char[] aoutput = new char[ArrayUtil.GetNextSize(maxSizeNeeded)];
len = Accent(aoutput, poutput, len);
return(new string(aoutput, 0, len).ToLower());
}
/// <summary> Converts characters with tones. For example,
/// accents are removed from accented characters.
/// </summary>
/// <param name="input">The string to fold
/// </param>
/// <param name="length">The number of characters in the input string
/// </param>
public void Transform(char[] input, int length)
{
// Worst-case length required:
int maxSizeNeeded = 4 * length;
if (output.Length < maxSizeNeeded)
{
output = new char[ArrayUtil.GetNextSize(maxSizeNeeded)];
}
outputPos = 0;
for (int pos = 0; pos < length; ++pos)
{
char c = input[pos];
// Quick test: if it's not in range then just keep current character
if (c < 'Ά')
{
output[outputPos++] = c;
}
else
{
switch (c)
{
case 'ά':
output[outputPos++] = 'α';
break;
case 'Ά':
output[outputPos++] = 'Α';
break;
case 'έ':
output[outputPos++] = 'ε';
break;
case 'Έ':
output[outputPos++] = 'Ε';
break;
case 'ή':
output[outputPos++] = 'η';
break;
case 'Ή':
output[outputPos++] = 'Η';
break;
case 'ί':
case 'ΐ':
case 'ϊ':
output[outputPos++] = 'ι';
break;
case 'Ί':
case 'Ϊ':
output[outputPos++] = 'Ι';
break;
case 'ό':
output[outputPos++] = 'ο';
break;
case 'Ό':
output[outputPos++] = 'Ο';
break;
case 'Ύ':
case 'Ϋ':
output[outputPos++] = 'Υ';
break;
case 'ύ':
case 'ΰ':
case 'ϋ':
output[outputPos++] = 'υ';
break;
case 'ώ':
output[outputPos++] = 'ω';
break;
case 'Ώ':
output[outputPos++] = 'Ω';
break;
default:
output[outputPos++] = c;
break;
}
}
}
}
/// <summary> Converts characters above ASCII to their ASCII equivalents. For example,
/// accents are removed from accented characters.
/// </summary>
/// <param name="input">The string to fold
/// </param>
/// <param name="length">The number of characters in the input string
/// </param>
public void Transform(char[] input, int length)
{
// Worst-case length required:
int maxSizeNeeded = 4 * length;
if (output.Length < maxSizeNeeded)
{
output = new char[ArrayUtil.GetNextSize(maxSizeNeeded)];
}
outputPos = 0;
for (int pos = 0; pos < length; ++pos)
{
char c = input[pos];
char next = (char)0;
char fnext = (char)0;
char fnext2 = (char)0;
if (pos + 1 < length)
{
next = input[pos + 1];
}
if (pos + 2 < length)
{
fnext = input[pos + 2];
}
if (pos + 3 < length)
{
fnext2 = input[pos + 3];
}
// Quick test: if it's not in range then just keep current character
if (c < 'Ά')
{
output[outputPos++] = c;
}
else
{
switch (c)
{
/*
* phoenetic similar αι -> α
*/
case 'α':
case 'Α':
if (next == 'υ' || next == 'ύ')
{
output[outputPos++] = c;
pos++;
if (IsLoud(fnext, fnext2))
{
output[outputPos++] = 'β';
}
else
{
output[outputPos++] = 'φ';
}
}
else if (next == 'Y' || next == 'Ύ')
{
output[outputPos++] = c;
pos++;
if (IsLoud(fnext, fnext2))
{
output[outputPos++] = 'Β';
}
else
{
output[outputPos++] = 'Φ';
}
}
else if (next == 'ι')
{
output[outputPos++] = 'ε';
pos++;
}
else if (next == 'ί')
{
output[outputPos++] = 'έ';
pos++;
}
else if (next == 'Ι')
{
output[outputPos++] = 'Ε';
pos++;
}
else if (next == 'Ί')
{
output[outputPos++] = 'Έ';
pos++;
}
else
{
output[outputPos++] = c;
}
break;
case 'ε':
case 'Ε':
if (next == 'υ' || next == 'ύ')
{
output[outputPos++] = c;
pos++;
if (IsLoud(fnext, fnext2))
{
output[outputPos++] = 'β';
}
else
{
output[outputPos++] = 'φ';
}
}
else if (next == 'Y' || next == 'Ύ')
{
output[outputPos++] = c;
pos++;
if (IsLoud(fnext, fnext2))
{
output[outputPos++] = 'Β';
}
else
{
output[outputPos++] = 'Φ';
}
}
else if (next == 'ι')
{
output[outputPos++] = 'ι';
pos++;
}
else if (next == 'ί')
{
output[outputPos++] = 'ί';
pos++;
}
else if (next == 'Ι')
{
output[outputPos++] = 'Ι';
pos++;
}
else if (next == 'Ί')
{
output[outputPos++] = 'Ί';
pos++;
}
else
{
output[outputPos++] = c;
}
break;
case 'η':
output[outputPos++] = 'ι';
break;
case 'ή':
output[outputPos++] = 'ί';
break;
case 'Η':
output[outputPos++] = 'Ι';
break;
case 'Ή':
output[outputPos++] = 'Ί';
break;
case 'ο':
case 'Ο':
if (next == 'ι')
{
output[outputPos++] = 'ι';
pos++;
}
else if (next == 'ί')
{
output[outputPos++] = 'ί';
pos++;
}
else if (next == 'Ι')
{
output[outputPos++] = 'Ι';
pos++;
}
else if (next == 'Ί')
{
output[outputPos++] = 'Ί';
pos++;
}
else
{
output[outputPos++] = c;
}
break;
case 'υ':
if (next == 'ι')
{
output[outputPos++] = 'ι';
pos++;
}
else if (next == 'ί')
{
output[outputPos++] = 'ί';
pos++;
}
else if (next == 'Ι')
{
output[outputPos++] = 'Ι';
pos++;
}
else if (next == 'Ί')
{
output[outputPos++] = 'Ί';
pos++;
}
else
{
output[outputPos++] = 'ι';
}
break;
case 'ύ':
output[outputPos++] = 'ί';
break;
case 'Υ':
if (next == 'ι')
{
output[outputPos++] = 'ι';
pos++;
}
else if (next == 'ί')
{
output[outputPos++] = 'ί';
pos++;
}
else if (next == 'Ι')
{
output[outputPos++] = 'Ι';
pos++;
}
else if (next == 'Ί')
{
output[outputPos++] = 'Ί';
pos++;
}
else
{
output[outputPos++] = 'Ι';
}
break;
case 'Ύ':
output[outputPos++] = 'Ί';
break;
case 'ω':
output[outputPos++] = 'ο';
break;
case 'ώ':
output[outputPos++] = 'ό';
break;
case 'Ω':
output[outputPos++] = 'Ο';
break;
case 'Ώ':
output[outputPos++] = 'Ό';
break;
default:
output[outputPos++] = c;
break;
}
}
}
}