QtEndian

[ThMars]

I receive raw data from a ble sensor in a QByteArray (5 bytes).
Byte 0 is a flag, byte 1 and 2is a quint16 value, so is byte 3 and 4, data is in little endian format.

I create a pointer to a quint8 array, and then extract the values with qFromLittleEndian

```

auto data = reinterpret_cast<const quint8 *>(arrayValue.constData());

quint8 flag = *data;
quint16 valueOne = qFromLittleEndian<quint16>(data[1]);
quint16 valueTwo = qFromLittleEndian<quint16>(data[3]);

What happens is that only the first byte taken into account, see the example code:

#include <QCoreApplication>
#include <QtEndian>

#include <iostream>

int main(int argc, char *argv[])
{
QCoreApplication a(argc, argv);

char byteData[]{'\x02', '\x02', '\01', '\xfc', '\x13'};

QByteArray arrayValue = QByteArray::fromRawData(byteData, sizeof(byteData));
auto data = reinterpret_cast<const quint8 *>(arrayValue.constData());

quint8 flag = *data;
quint16 valueOne = qFromLittleEndian<quint16>(data[1]);
quint16 valueTwo = qFromLittleEndian<quint16>(data[3]);

std::cout << "Receceived raw data in a QByteArray from a ble sensor." << std::endl << "Byte 0 is a flag, byte 1 and 2 is a quint16 value, so is byte 3 and 4" << std::endl;
std::cout << "Data is in littleEndian so the values should read, flag = " << char(2) << ", valueOne = 258 and valueTwo = 5116" << std::endl << std::endl;

std::cout << "flag: " << flag << " // valueOne: " << valueOne << " // valueTwo: " << valueTwo << std::endl;

std::cout << "As  one can see, only the first byte (being the lsb) is used by qFromLittleEndian" << std::endl << std::endl;

std::cout << "Is this expected behaviour from 'T qFromLittleEndian(const void *src)' and should I do: valueOne = data[2] * 256 + data[1];.\r\n"
             "Am I missing something or is this a bug?, SomeOne?" << std::endl << std::endl;
valueOne = data[2] * 256 + data[1];
valueTwo = data[4] * 256 + data[3];


std::cout << "flag: " << flag << " // valueOne: " << valueOne << " // valueTwo: " << valueTwo << std::endl;



return a.exec();

}

Anyone any suggestion?
Thanks in advance.

[sierdzio]

quint16 valueOne = qFromLittleEndian<quint16>(data[1]);
quint16 valueTwo = qFromLittleEndian<quint16>(data[3]);

So what you do is give qFromLittleEndian 8 bits and expect it to read the next 8 bits automatically? That's not how it works. You need to provide the whole 16 bits, for example:

quint16 valueOne = qFromLittleEndian<quint16>(arrayValue.mid(1,2).data());

Or easier, using QDataStream:

QDataStream stream(arrayValue);
stream.setByteOrder(QDataStream::LittleEndian);
quint8 flag;
quint16 valueOne;
quint16 valueTwo;

stream >> flag;
stream >> valueOne;
stream >> valueTwo;

Edit: corrected some small typos.

[ThMars]

Thanks for the solutions.
I just realised that I should have given the pointer with an offset and not just the 8 bits.

[Pablo J. Rogina]

@ThMars said in QtEndian:

Thanks for the solutions.

is your issue solved then? if so please don't forget to mark your post as such! Thanks.