private static void CopyCharStructToString(StringBuilder str, CharStruct s, int level)
{
/* s is a shaped CharStruct; level is the index into the string */
if (s.BaseChar == 0)
{
return;
}
str.Append(s.BaseChar);
s.LigNum--;
if (s.Mark1 != 0)
{
if ((level & ar_novowel) == 0)
{
str.Append(s.Mark1);
}
s.LigNum--;
}
if (s.Vowel != 0)
{
if ((level & ar_novowel) == 0)
{
str.Append(s.Vowel);
}
s.LigNum--;
}
}
public void BadChar()
{
ParseT9 parseT9 = new ParseT9();
CharStruct s = parseT9.GetCharStruct('%');
Assert.That(s.Code, Is.EqualTo(""));
}
private static void CopyCharStructToString(StringBuilder str, CharStruct s, int level)
{
/* s is a shaped CharStruct; level is the index into the string */
if (s.BaseChar == 0)
{
return;
}
str.Append(s.BaseChar);
}
private static void Shape(char[] text, StringBuilder str, int level)
{
/* string is assumed to be empty and big enough.
* text is the original text.
* This routine does the basic arabic reshaping.
* len the number of non-null characters.
*
* Note: We have to unshape each character first!
*/
int join;
int which;
char nextLetter;
int p = 0; /* initialize for output */
CharStruct oldchar = new CharStruct();
CharStruct curchar = new CharStruct();
while (p < text.Length)
{
nextLetter = text[p++];
//nextletter = unshape (nextletter);
join = Ligature(nextLetter, curchar);
if (join == 0)
{
/* shape curchar */
int nc = ShapeCount(nextLetter); // from 1 to 4
//(*len)++;
/* which 0 == isolated 1 == final 2 == initial 3 == medial */
if (nc == 1)
{
which = 0; /* final or isolated */
}
else
{
which = 2; /* medial or initial */
}
if (ConnectsToLeft(oldchar))
{
which++;
}
which = which % (curchar.NumShapes);
curchar.BaseChar = GetCharShape(curchar.BaseChar, which);
/* get rid of oldchar */
CopyCharStructToString(str, oldchar, level);
oldchar = curchar; /* new values in oldchar */
/* init new curchar */
curchar = new CharStruct();
curchar.BaseChar = nextLetter;
curchar.NumShapes = nc;
curchar.LigNum++;
// (*len) += unligature (&curchar, level);
}
else if (join == 1)
{
}
// else
// {
// (*len) += unligature (&curchar, level);
// }
// p = g_utf8_next_char (p);
}
/* handle last char */
which = 0;
if (ConnectsToLeft(oldchar))
{
which++;
}
which = which % (curchar.NumShapes);
curchar.BaseChar = GetCharShape(curchar.BaseChar, which);
/* get rid of oldchar */
CopyCharStructToString(str, oldchar, level);
CopyCharStructToString(str, curchar, level);
}
private static bool ConnectsToLeft(CharStruct a)
{
return(a.NumShapes > 2);
}
private static int Ligature(char newChar, CharStruct oldChar)
{
/* 0 == no ligature possible; 1 == vowel; 2 == two chars; 3 == Lam+Alef */
int retval = 0;
if (oldChar.BaseChar == 0) // base char yet is not defined
{
return(0);
}
if (IsVowel(newChar))
{
retval = 1;
if ((oldChar.Vowel != 0) && (newChar != SHADDA))
{
retval = 2; /* we eliminate the old vowel .. */
}
switch (newChar)
{
case SHADDA:
{
if (oldChar.Mark1 == 0)
{
oldChar.Mark1 = SHADDA;
}
else
{
return(0); /* no ligature possible */
}
break;
}
case HAMZA_BELOW:
{
switch (oldChar.BaseChar)
{
case ALEF: { oldChar.BaseChar = ALEF_HAMZA_BELOW; retval = 2; break; }
case LAM_ALEF: { oldChar.BaseChar = LAM_ALEF_HAMZA_BELOW; retval = 2; break; }
default: { oldChar.Mark1 = HAMZA_BELOW; break; }
}
break;
}
case HAMZA_ABOVE:
{
switch (oldChar.BaseChar)
{
case ALEF: { oldChar.BaseChar = ALEF_HAMZA_ABOVE; retval = 2; break; }
case LAM_ALEF: { oldChar.BaseChar = LAM_ALEF_HAMZA_ABOVE; retval = 2; break; }
case WAW: { oldChar.BaseChar = WAW_HAMZA_ABOVE; retval = 2; break; }
case YEH:
case ALEF_MAKSURA:
case FARSI_YEH: { oldChar.BaseChar = YEH_HAMZA_ABOVE; retval = 2; break; }
default: { oldChar.Mark1 = HAMZA_ABOVE; break; } /* whatever sense this may make .. */
}
break;
}
case MADDA_ABOVE:
{
switch (oldChar.BaseChar)
{
case ALEF: { oldChar.BaseChar = ALEF_MADDA_ABOVE; retval = 2; break; }
}
break;
}
default: { oldChar.Vowel = newChar; break; }
}
if (retval == 1)
{
oldChar.LigNum++;
}
return(retval);
}
if (oldChar.Vowel != 0) /* if we already joined a vowel, we can't join a Hamza */
{
return(0);
}
switch (oldChar.BaseChar)
{
case LAM:
{
switch (newChar)
{
case ALEF: { oldChar.BaseChar = LAM_ALEF; oldChar.NumShapes = 2; retval = 3; break; }
case ALEF_HAMZA_ABOVE: { oldChar.BaseChar = LAM_ALEF_HAMZA_ABOVE; oldChar.NumShapes = 2; retval = 3; break; }
case ALEF_HAMZA_BELOW: { oldChar.BaseChar = LAM_ALEF_HAMZA_BELOW; oldChar.NumShapes = 2; retval = 3; break; }
case ALEF_MADDA_ABOVE: { oldChar.BaseChar = LAM_ALEF_MADDA_ABOVE; oldChar.NumShapes = 2; retval = 3; break; }
}
break;
}
case (char)0:
{
oldChar.BaseChar = newChar; oldChar.NumShapes = ShapeCount(newChar); retval = 1; break;
}
}
return(retval);
}
private static int Ligature(char newChar, CharStruct oldChar)
{
/* 0 == no ligature possible; 2 == two chars; 3 == three chars*/
int retval = 0;
if (oldChar.BaseChar == 0) // base char yet is not defined
{
return(0);
}
switch (newChar)
{
case HIRIQ:
{
switch (oldChar.BaseChar)
{
case YOD: { oldChar.BaseChar = YOD_HIRIQ; retval = 2; break; }
}
break;
}
case PATAH:
{
switch (oldChar.BaseChar)
{
case ALEF: { oldChar.BaseChar = ALEF_PATAH; retval = 2; break; }
}
break;
}
case QAMATS:
{
switch (oldChar.BaseChar)
{
case ALEF: { oldChar.BaseChar = ALEF_QAMATS; retval = 2; break; }
}
break;
}
case HOLAM:
{
switch (oldChar.BaseChar)
{
case VAV: { oldChar.BaseChar = VAV_HOLAM; retval = 2; break; }
}
break;
}
case DAGESH: // == case MAPIQ:
{
switch (oldChar.BaseChar)
{
case ALEF: { oldChar.BaseChar = ALEF_MAPIQ; retval = 2; break; }
case HE: { oldChar.BaseChar = HE_MAPIQ; retval = 2; break; }
case BET: { oldChar.BaseChar = BET_DAGESH; retval = 2; break; }
case GIMEL: { oldChar.BaseChar = GIMEL_DAGESH; retval = 2; break; }
case DALET: { oldChar.BaseChar = DALET_DAGESH; retval = 2; break; }
case VAV: { oldChar.BaseChar = VAV_DAGESH; retval = 2; break; }
case ZAYIN: { oldChar.BaseChar = ZAYIN_DAGESH; retval = 2; break; }
case TET: { oldChar.BaseChar = TET_DAGESH; retval = 2; break; }
case YOD: { oldChar.BaseChar = YOD_DAGESH; retval = 2; break; }
case FINAL_KAF: { oldChar.BaseChar = FINAL_KAF_DAGESH; retval = 2; break; }
case KAF: { oldChar.BaseChar = KAF_DAGESH; retval = 2; break; }
case LAMED: { oldChar.BaseChar = LAMED_DAGESH; retval = 2; break; }
case MEM: { oldChar.BaseChar = MEM_DAGESH; retval = 2; break; }
case NUN: { oldChar.BaseChar = NUN_DAGESH; retval = 2; break; }
case SAMEKH: { oldChar.BaseChar = SAMEKH_DAGESH; retval = 2; break; }
case FINAL_PE: { oldChar.BaseChar = FINAL_PE_DAGESH; retval = 2; break; }
case PE: { oldChar.BaseChar = PE_DAGESH; retval = 2; break; }
case TSADI: { oldChar.BaseChar = TSADI_DAGESH; retval = 2; break; }
case QOF: { oldChar.BaseChar = QOF_DAGESH; retval = 2; break; }
case RESH: { oldChar.BaseChar = RESH_DAGESH; retval = 2; break; }
case SHIN: { oldChar.BaseChar = SHIN_DAGESH; retval = 2; break; }
case TAV: { oldChar.BaseChar = TAV_DAGESH; retval = 2; break; }
}
break;
}
case SHIN_DOT:
{
switch (oldChar.BaseChar)
{
case SHIN: { oldChar.BaseChar = SHIN_SHIN_DOT; retval = 2; break; }
case SHIN_DAGESH: { oldChar.BaseChar = SHIN_DAGESH_SHIN_DOT; retval = 3; break; } // 3 CHARS
}
break;
}
case SIN_DOT:
{
switch (oldChar.BaseChar)
{
case SHIN: { oldChar.BaseChar = SHIN_SIN_DOT; retval = 2; break; }
case SHIN_DAGESH: { oldChar.BaseChar = SHIN_DAGESH_SIN_DOT; retval = 3; break; } // 3 CHARS
}
break;
}
case RAFE:
{
switch (oldChar.BaseChar)
{
case BET: { oldChar.BaseChar = BET_RAFE; retval = 2; break; }
case KAF: { oldChar.BaseChar = KAF_RAFE; retval = 2; break; }
case PE: { oldChar.BaseChar = PE_RAFE; retval = 2; break; }
}
break;
}
case LAMED:
{
switch (oldChar.BaseChar)
{
case ALEF: { oldChar.BaseChar = ALEF_LAMED; retval = 2; break; }
}
break;
}
}
return(retval);
}