I have a kotlinx io Source. I would like to read c...

# io

[edenman]

I have a kotlinx io Source. I would like to read chunks of size N and only get a smaller chunk if it’s the end of the file. readAtMostTo looks like what I want, but it’s not guaranteed to read the full chunk size

[edenman]

In particular, looks like if you hit the 8192 buffer size and it doesn’t line up with your chunk size exactly, you’ll get a partial chunk even if the Source still has more data

[edenman]

Do I just wrap readAtMostTo in a while loop? Gross

[edenman]

private fun readChunk(from: Source, buffer: Buffer, chunkSize: Long): ByteArray? {
        var numBytesRead = 0L
        while (numBytesRead != -1L && numBytesRead != chunkSize && !from.exhausted()) {
            val read = from.readAtMostTo(sink = buffer, byteCount = chunkSize - numBytesRead)
            if (read == -1L) {
                return null
            }
            numBytesRead += read
        }
        if (from.exhausted() && numBytesRead == 0L) {
            return null
        }
        return buffer.readByteArray()
    }

[edenman]

Probably could clean that up a bit more but it works ¯\_(ツ)_/¯

[Filipp Zhinkin]

Hey! Source has readTo function which reads exactly N bytes into a sink (which could be a Buffer).
And if all you need is an array of bytes, readByteArray(byteCount: Int) should also read exactly byteCount bytes.

You can always combine them with Source.request(byteCount: Long) to make sure that the source has the whole chunk. Something like:

private fun readChunk(from: Source, chunkSize: Long): ByteArray? {
   require(chunkSize in 0..Int.MAX_VALUE)

   return when {
      from.request(chunkSize) -> from.readByteArray(chunkSize.toInt())
      from.exhausted() -> null
      else -> from.readByteArray()
   }
}

[edenman]

readByteArray(int) doesn’t work for the end of file case, sadly (throws an exception)

[edenman]

Will look at the others, thx!