We Keep Coding

convert serialport.readline() non-ascii string into a byte array

I want to use serialport.readline() because it's a blocking call and the data I'm getting has 0x0D 0x0A (CR LF) at the end. However, what I want is an Hex String instead of an Ascii representation.
For example, the device I am communication with is sending byte arrays like {0xff,0xff,0x45,0x0D,0x0A}. I want to simply print out in my program just like this: 0xff, 0xff,0x45. Readline() kindly trims out LF and CR.
I though about using serialport.read(buff[]...) by specifying how many bytes I want to read. But it didn't work very well because if I am reading too fast half of the array will be 0x00 and if I am reading too slow, there will be an overflow for the com port. I don't want to lose any bytes.
I tried to convert what I got from serialport.readline() to a byte array but the hex string I got usually turns into 0x3f. Code is like this:
var line = string.Join(",", mySerialPort.ReadLine().Select(c => ((Byte)c).ToString("X")).ToArray());
I changed the encoding a few times (ASCII, UTF8,UNICODE) but still no go.
Is there any way to convert the non-Ascii String I got from readline() into a byte array?

It sounds like you shouldn't be reading it as text data at all.
You're fundamentally dealing with binary data, so use the overload of Read which takes a byte array rather than a char array. (Or call ReadByte repeatedly.)
Any time you try to treat arbitrary binary data as if it's text, you're going to have a bad experience.
It sounds like you've already tried this, but done it badly:
But it didn't work very well because if I am reading too fast half of the array will be 0x00
That suggests you're ignoring the return value of Read, which says how many bytes have actually been read. You should have something like:
int bytesRead = port.Read(buffer, 0, buffer.Length);
// Now use the portion of buffer which is from 0 (inclusive) to
// bytesRead (exclusive).

How to properly write a UDP packet

I am trying to rewrite some of my code from a C++ program I wrote a while ago, but I am not sure if/how I can write to a byte array properly, or if I should be using something else. The code I am trying to change to C# .NET is below.
unsigned char pData[1400];
bf_write g_ReplyInfo("SVC_ReplyInfo", &pData, 1400);
void PlayerManager::BuildReplyInfo()
{
// Delete the old packet
g_ReplyInfo.Reset();
g_ReplyInfo.WriteLong(-1);
g_ReplyInfo.WriteByte(73);
g_ReplyInfo.WriteByte(g_ProtocolVersion.GetInt());
g_ReplyInfo.WriteString(iserver->GetName());
g_ReplyInfo.WriteString(iserver->GetMapName());
}

BinaryWriter might work, although strings are written with a preceding 7-bit encoded length, which I suspect the client won't be able to handle. You'll probably have to convert strings to bytes and then either add a length word or 0-terminate it.
No need to manually convert numbers to bytes. If you have a long that you want to write as a byte, just cast it. That is, if your BinaryWriter is bw, then you can write bw.Write((byte)longval);. To write -1 as a long: bw.Write((long)(-1)).

How to treat a byte[] as Int16[]?

I am receiving some raw data from audio input via System.Runtime.InteropServices.Marshal.Copy() and this is stored in a byte[]. The actual data in that byte[] I know to be of type Int16. I would like to treat this byte[] as if it was an array of type Int16 and loop over its elements, process it and at the end of this I would like to pass this altered array as byte[] into another function.
For those who wonder what I'm trying to do: I would like to try adding an echo effect to the incoming sound (from mic) by simply adding up the wave data from previous buffer with the wave data from the current buffer, mixing it in at a lower volume.
EDIT: the byte[] array may be storing a int16 number 258 as a pair of bytes [1][2]
0000 0001 0000 0010

You can't, as far as I'm aware.
I suggest you use Buffer.BlockCopy to copy the data into a short[], work on it, and then copy it back. Of course, that assumes that the bytes are arranged in an appropriate endianness... but it's likely that they will be.
Alternatively... use the overload of Marshal.Copy which accepts a short[] to copy the data into to start with.

You can use BitConverter to treat the bytes as Int16 elements using BitConverter.ToInt16. You can also use Buffer.BlockCopy to copy the array of bytes to an array of Int16s. The first is more memory efficient and the second is more speed efficient.

FileStream.Read(...): Will I ever get a nibble rather than byte?

I'm doing a number of bitwise operations on an array of bytes in C#. I'm obtaining the array by calling FileStream.Read. I just realized that I'm not sure what would happen if a file had a bad byte or corrupt byte in it somewhere. For example, maybe a nibble is chopped off of the end or something like that. What would the FileStream do with it? Would the messed up byte by 'rounded' off by the Read method? Would an exception be thrown? Or is this something that will virtually never happen?
Thanks,
brian

If your FileStream.Read call succeeds, there's no such thing as a file having a bad byte or corrupt byte. Each byte that is successfully read, and part of the file, is a value from 0 to 255. How it is interpreted by a program is what matters.
If FileStream.Read returns for example 5 bytes, then you can rely that those 5 bytes are read successfully from the file and all bits of the bytes were put into your buffer successfully.
There is such thing though for example as a bad cluster on your hard disk, in which case your Read would fail with some kind of exception.
For completeness I should also mention that every file type has a file format. I.e. how you should interpret the binary data. It is possible that a byte or several bytes don't follow the file format. And in that way you can view a byte as being corrupt or invalid, but it's not really corrupt or invalid just wrong in terms of what the file format specifies.

problem with encoding.utf8.getbytes in c#

I am working on C#, trying below code
byte[] buffer = new byte[str.Length];
buffer = Encoding.UTF8.GetBytes(str);
In str I've got lengthy data but I've got problem in getting complete encoded bytes.
Please tell me what's going wrong and how can I overcome this problem?

Why are you creating a new byte array and then ignoring it? The value of buffer before the call to GetBytes is being replaced with a reference to a new byte array returned by GetBytes.
However, you shouldn't expect the UTF-8 encoded version of a string to be the same length in bytes as the original string's length in characters, unless it's all ASCII. Any character over U+007F takes up at least 2 bytes.
What's the bigger picture here? What are you trying to achieve, and why does the length of the byte array matter to you?

The proper use is:
byte[] buffer = Encoding.UTF8.GetBytes(str);

In general, you should not make any assumptions about length/size/count when working with encoding, bytes and chars/strings. Let the Encoding objects do their work and then query the resulting objects for that info.
Having said that, I don't believe there is an inherent length restriction for the encoding classes. I have several production apps doing the same work in the opposite direction (bytes encoded to chars) which are processing byte arrays in the 10s of megabytes.