How to convert const QVector<quint32> to vector<int>.

[JohanSolo]

Once I'm sure I need the conversion, I would use std::transform with a lambda.

[VRonin]

Apart from the ususal disclaimer about converting unsigned to siged, you can use

std::vector<int> l_sensorAddressList(
    reinterpret_cast<const int*>(p_sensorAddressList.constData())
    , reinterpret_cast<const int*>(p_sensorAddressList.constData()+p_sensorAddressList.size())
);

[Christian Ehrlicher]

While we're at it:

std::copy(in.begin(), in.end(), std::back_inserter(out));

[SGaist]

Hi,

One even simpler version:

std::vector<int> l_sensorAddressList(std::begin(p_sensorAddressList), std::end(p_sensorAddressList));

or

std::vector<int> l_sensorAddressList(p_sensorAddressList.begin(), p_sensorAddressList.end());

[VRonin]

Just to clarify: @JohanSolo , @Christian-Ehrlicher and @SGaist 's solutions are correct. My solution should be faster as it's a simple memcpy but the other's might be safer (I don't really see how though)

[fcarney]

@Christian-Ehrlicher Are iterators faster than indexing ([index]) ?

[JonB]

@VRonin

My solution should be faster as it's a simple memcpy

Does yours assume 32-bit where the others don't? Is (part of) the question to convert from qint32 to whatever int is, which I assumed is 64-bit now?

[Christian Ehrlicher]

@JonB said in How to convert const QVector<quint32> to vector<int>.:

which I assumed is 64-bit now?

int is 32 bit, on every platform Qt is running on :)

[JonB]

@Christian-Ehrlicher
Wot? Really?? I thought it was 64 on 64....

[JKSH]

@JonB said in How to convert const QVector<quint32> to vector<int>.:

Wot? Really?? I thought it was 64 on 64....

Pointer size is increased to 64 but integer size remains 32 on today's common 64-bit desktop architectures.

sizeof(qintptr) == sizeof(void*) != sizeof(int)

[J.Hilk]

@JKSH said in How to convert const QVector<quint32> to vector<int>.:

sizeof(qintptr) == sizeof(void*) != sizeof(int)

wow, I totally assumed differently,
I should make it habit to use the explicitly size defined qintXX/quintXX in the future...

[JKSH]

@J-Hilk said in How to convert const QVector<quint32> to vector<int>.:

I should make it habit to use the explicitly size defined qintXX/quintXX in the future...

Good idea!

[JonB]

@JKSH , @J-Hilk

Pointer size is increased to 64 but integer size remains 32 on today's common 64-bit desktop architectures.

Yes, now I vaguely remember reading this, I am glad @J-Hilk shared my mis-assumption!

This is what comes of my being forced to use Python for Qt or JavaScript over so many years now. Ignorance of underlying type sizes! It has been a long time since sizeof(int) != sizeof(long) and sizeof(int) != sizeof(void*), probably not since the initial move from 16-bit to 32-bit architecture! :)