static void ReadText(byte[] encoded, string expected)
{
using (var ms = new MemoryStream(encoded))
{
var utf8 = new Utf8Stream(ms, false);
var charbuf = new char[expected.Length];
var read = utf8.ReadCharacters(charbuf, 0, expected.Length);
var actual = new string(charbuf);
Assert.AreEqual(expected.Length, read, "Actual number of characters read differed from expected.");
Assert.AreEqual(expected, actual, "Decoded string differs from source string.");
}
}
public void TestReadMultipleChars()
{
var mathLowercaseEl = "abc𝓁";
using (var ms = new MemoryStream(Encoding.UTF8.GetBytes(mathLowercaseEl)))
{
var charbuf = new char[5];
var utf8 = new Utf8Stream(ms, false);
var read = utf8.ReadCharacters(charbuf, 0, 5);
var str = new string(charbuf);
Assert.AreEqual(mathLowercaseEl, str);
}
}
public void TestReadChar()
{
// euro, feh, thorn, om, respectively
var charBytes = new Dictionary<char, byte[]>
{
{ '€', new byte[] { 0xE2, 0x82, 0xAC } },
{ 'Ֆ', new byte[] { 0xD5, 0x96 } },
{ 'þ', new byte[] { 0xC3, 0xBE } },
{ 'ॐ', new byte[] { 0xE0, 0xA5, 0x90 } }
};
foreach (var kvp in charBytes)
{
using (var ms = new MemoryStream(kvp.Value))
{
var utf8 = new Utf8Stream(ms, false);
var actual = utf8.ReadCharacter();
Assert.AreEqual(kvp.Key, actual);
}
}
}
private string ReadString()
{
var sb = new StringBuilder();
var finalChunk = false;
do
{
int charCount;
var chunkIndicator = input.ReadByte();
if (chunkIndicator >= 0x00 && chunkIndicator <= 0x1F)
{
finalChunk = true;
charCount = chunkIndicator;
}
else if (chunkIndicator == 'S' || chunkIndicator == 's')
{
finalChunk = chunkIndicator == 'S';
charCount = ReadLength();
}
else
{
throw new SerializationException("Invalid Hessian 2.0 string.");
}
if (charCount == 0)
{
// We have no further work to do in this chunk;
// read the next one. If another chunk exists, this one
// is malformed. Most likely, this chunk is the first and
// only chunk of an empty string. Either way, I don't
// think that we care.
continue;
}
using (var stringstream = new Utf8Stream(input))
{
var charbuf = new char[charCount];
var charsRead = stringstream.ReadCharacters(charbuf, 0, charCount);
if (charsRead != charCount)
throw new SerializationException(string.Format("Invalid string chunk - expected {0} characters, found {1} characters.", charCount, charsRead));
sb.Append(charbuf);
}
/*
var converter = new Utf8StreamConverter(input);
for (var i = 0; i < charCount; ++i)
{
sb.Append(converter.ReadCharacter());
}*/
}
while (!finalChunk);
return sb.ToString();
}
