Bug :java.lang.ArrayIndexOutOfBoundsException: arraycopy: last source index 248 out of bounds for byte[234]

[PanagiotisDrakatos]

Bug :java.lang.ArrayIndexOutOfBoundsException: arraycopy: last source index 248 out of bounds for byte[234]

[PanagiotisDrakatos]

@eduard-vasinskyi Brilliant!!!!!!!!!!! working like charm. Solved

[eduard-vasinskyi]

Hi, @PanagiotisDrakatos

The problem seems to be with chosen message encoding. You use ByteBufsDecoder#ofCrlfTerminatedBytes to decode messages. It expects messages to end with \r\n. However, your data may also contain \r\n, so deserializer does not decode the full message but rather part of it. You can verify that this is the case by modifying the loadData(Transaction) method:

private static @NotNull Promise<ByteBuf> loadData(Transaction transaction) {
    byte transaction_hash[] = encode.encode(transaction, 1024);
    for (int i = 0; i < transaction_hash.length - 1; i++) {
        if (transaction_hash[i] == '\r' && transaction_hash[i+1] == '\n'){
            throw new RuntimeException("CR LF conflict");
        }
    ByteBuf appendedBuf = ByteBufPool.append(ByteBuf.wrapForReading(transaction_hash), ByteBuf.wrapForReading("\r\n".getBytes()));
    return Promise.of(appendedBuf);
}

You could try to somehow escape the \r\n sequence in your data. Alternatively, you could use a different ByteBufsDecoder. One option is to use size-prefixed message encoding. For example, you could use ByteBufsDecoder.ofVarIntSizePrefixedBytes() as your DECODER. And then encode your message in loadData(Transaction) like this:

private static @NotNull Promise<ByteBuf> loadData(Transaction transaction) {
    byte transaction_hash[] = encode.encode(transaction, 1024);
    ByteBuf sizeBuf = ByteBufPool.allocate(2); // enough to serialize size 1024
    sizeBuf.writeVarInt(transaction_hash.length);
    ByteBuf appendedBuf = ByteBufPool.append(sizeBuf, ByteBuf.wrapForReading(transaction_hash));
    return Promise.of(appendedBuf);
}