public static IEnumerable <Utf8Bytes> Split(this Utf8Bytes src, byte delemeter)
{
var start = 0;
var p = new Utf8Iterator(src.Bytes);
while (p.MoveNext())
{
if (p.Current == delemeter)
{
if (p.BytePosition - start == 0)
{
yield return(default(Utf8Bytes));
}
else
{
yield return(src.Subbytes(start, p.BytePosition - start));
}
start = p.BytePosition + 1;
}
}
if (start < p.BytePosition)
{
yield return(src.Subbytes(start, p.BytePosition - start));
}
}
public static UInt64 ToUInt64(this Utf8Bytes src)
{
UInt64 value = 0;
var p = new Utf8Iterator(src.Bytes);
while (p.MoveNext())
{
var b = p.Current;
switch (b)
{
case 0x30: value = (UInt64)(value * 10); break;
case 0x31: value = (UInt64)(value * 10 + 1); break;
case 0x32: value = (UInt64)(value * 10 + 2); break;
case 0x33: value = (UInt64)(value * 10 + 3); break;
case 0x34: value = (UInt64)(value * 10 + 4); break;
case 0x35: value = (UInt64)(value * 10 + 5); break;
case 0x36: value = (UInt64)(value * 10 + 6); break;
case 0x37: value = (UInt64)(value * 10 + 7); break;
case 0x38: value = (UInt64)(value * 10 + 8); break;
case 0x39: value = (UInt64)(value * 10 + 9); break;
default: throw new ArgumentOutOfRangeException();
}
}
return(value);
}
public static Int32 ToInt32(this Utf8Bytes src)
{
Int32 value = 0;
var p = new Utf8Iterator(src.Bytes);
while (p.MoveNext())
{
var b = p.Current;
switch (b)
{
case 0x30: value = value * 10; break;
case 0x31: value = value * 10 + 1; break;
case 0x32: value = value * 10 + 2; break;
case 0x33: value = value * 10 + 3; break;
case 0x34: value = value * 10 + 4; break;
case 0x35: value = value * 10 + 5; break;
case 0x36: value = value * 10 + 6; break;
case 0x37: value = value * 10 + 7; break;
case 0x38: value = value * 10 + 8; break;
case 0x39: value = value * 10 + 9; break;
default: throw new ArgumentOutOfRangeException();
}
}
return(value);
}
public static Utf8Bytes From(string src, Byte[] bytes)
{
var required = src.Sum(c => Utf8Iterator.ByteLengthFromChar(c));
if (required > bytes.Length)
{
throw new OverflowException();
}
int pos = 0;
foreach (var c in src)
{
if (c <= Utf8Iterator.Mask7)
{
// 1bit
bytes[pos++] = (byte)c;
}
else if (c <= Utf8Iterator.Mask11)
{
// 2bit
bytes[pos++] = (byte)(Utf8Iterator.Head2 | Utf8Iterator.Mask5 & (c >> 6));
bytes[pos++] = (byte)(Utf8Iterator.Head1 | Utf8Iterator.Mask6 & (c));
}
else
{
// 3bit
bytes[pos++] = (byte)(Utf8Iterator.Head3 | Utf8Iterator.Mask4 & (c >> 12));
bytes[pos++] = (byte)(Utf8Iterator.Head1 | Utf8Iterator.Mask6 & (c >> 6));
bytes[pos++] = (byte)(Utf8Iterator.Head1 | Utf8Iterator.Mask6 & (c));
}
}
return(new Utf8Bytes(new ArraySegment <byte>(bytes, 0, pos)));
}
public static bool TrySearchAscii(this Utf8Bytes src, Byte target, int start, out int pos)
{
var p = new Utf8Iterator(src.Bytes, start);
while (p.MoveNext())
{
var b = p.Current;
if (b <= 0x7F)
{
// ascii
if (b == target /*'\"'*/)
{
// closed
pos = p.BytePosition;
return(true);
}
else if (b == '\\')
{
// escaped
switch ((char)p.Second)
{
case '"': // fall through
case '\\': // fall through
case '/': // fall through
case 'b': // fall through
case 'f': // fall through
case 'n': // fall through
case 'r': // fall through
case 't': // fall through
// skip next
p.MoveNext();
break;
case 'u': // unicode
// skip next 4
p.MoveNext();
p.MoveNext();
p.MoveNext();
p.MoveNext();
break;
default:
// unkonw escape
throw new KeyNotFoundException("unknown escape: " + p.Second);
}
}
}
}
pos = -1;
return(false);
}