static utf_error get_sequence_3(ref string src, int it, int end, ref int code_point)
{
if (it == end)
{
return(utf_error.NOT_ENOUGH_ROOM);
}
code_point = mask8(src[it]);
{
utf_error ret1 = increase_safely(ref src, it, end);
if (ret1 != utf_error.UTF8_OK)
{
return(ret1);
}
}
code_point = ((code_point << 12) & 0xffff) + ((mask8(src[it]) << 6) & 0xfff);
utf_error ret2 = increase_safely(ref src, it, end);
if (ret2 != utf_error.UTF8_OK)
{
return(ret2);
}
code_point += (src[it]) & 0x3f;
return(utf_error.UTF8_OK);
}
public static bool IsStringValidUTF8(string source)
{
int result = 0;
int end = source.Length;
while (result != end)
{
int t = 0;
utf_error err_code = validate_next(ref source, ref result, end, ref t);
if (err_code != utf_error.UTF8_OK)
{
return(result >= end);
}
}
return(result >= end);
}
static utf_error get_sequence_2(ref string src, int it, int end, ref int code_point)
{
if (it == end)
{
return(utf_error.NOT_ENOUGH_ROOM);
}
code_point = mask8(src[it]);
utf_error ret = increase_safely(ref src, it, end);
if (ret != utf_error.UTF8_OK)
{
return(ret);
}
code_point = ((code_point << 6) & 0x7ff) + ((src[it]) & 0x3f);
return(utf_error.UTF8_OK);
}
static utf_error validate_next(ref string src, ref int it, int end, ref int code_point)
{
if (it == end)
{
return(utf_error.NOT_ENOUGH_ROOM);
}
// Save the original value of it so we can go back ref case of failure
// Of course, it does not make much sense with i.e. stream iterators
int original_it = it;
Int32 cp = 0;
// Determine the sequence length based on the lead octet
int length = sequence_length(ref src, it);
// Get trail octets and calculate the code point
utf_error err = utf_error.UTF8_OK;
switch (length)
{
case 0:
return(utf_error.INVALID_LEAD);
case 1:
err = get_sequence_1(ref src, it, end, ref cp);
break;
case 2:
err = get_sequence_2(ref src, it, end, ref cp);
break;
case 3:
err = get_sequence_3(ref src, it, end, ref cp);
break;
case 4:
err = get_sequence_4(ref src, it, end, ref cp);
break;
}
if (err == utf_error.UTF8_OK)
{
// Decoding succeeded. Now, security checks...
if (is_code_point_valid((UInt32)cp))
{
if (!is_overlong_sequence((UInt32)cp, length))
{
// Passed! Return here.
code_point = cp;
++it;
return(utf_error.UTF8_OK);
}
else
{
err = utf_error.OVERLONG_SEQUENCE;
}
}
else
{
err = utf_error.INVALID_CODE_POINT;
}
}
// Failure branch - restore the original value of the iterator
it = original_it;
return(err);
}
