Pass vector as pointer

Question

I have a vector of float

vector<float>nFloats;

I would like to pass it to the following function:

/* This is a non-stream oriented interface to just change the speed of a sound sample */
int sonicChangeFloatSpeed(
    float *samples,
    int numSamples,
    float speed,
    float pitch,
    float rate,
    float volume,
    int useChordPitch,
    int sampleRate,
    int numChannels)
{
    sonicStream stream = sonicCreateStream(sampleRate, numChannels);

    sonicSetSpeed(stream, speed);
    sonicSetPitch(stream, pitch);
    sonicSetRate(stream, rate);
    sonicSetVolume(stream, volume);
    sonicSetChordPitch(stream, useChordPitch);
    sonicWriteFloatToStream(stream, samples, numSamples);
    sonicFlushStream(stream);
    numSamples = sonicSamplesAvailable(stream);
    sonicReadFloatFromStream(stream, samples, numSamples);
    sonicDestroyStream(stream);
    return numSamples;
}

Can anybody tell me how to do it correctly?

My attempts

 1) int iRet = sonicChangeFloatSpeed(&nFloats[0], *num, 1, 0.5, 1, 1, 1, 48000, 1);

 2) int iRet = sonicChangeFloatSpeed(nFloats[0], *num, 1, 0.5, 1, 1, 1, 48000, 1);

 3) int iRet = sonicChangeFloatSpeed(*nFloats, *num, 1, 0.5, 1, 1, 1, 48000, 1);

all don't work.

I'm sure such a question has already been asked, but I couldn't find any exact duplicate.

Can anybody help?

Thank you.

It looks like the pointer is used as output of the function. You must resize() the vector before using it to an appropriate buffer size. — πάντα ῥεῖ
– πάντα ῥεῖ, Commented Jan 29, 2017 at 11:26
The semantics of the function are unclear, because it takes numSamples and returns it. — Christian Hackl
– Christian Hackl, Commented Jan 29, 2017 at 11:43
Is the function supposed to add floats to the vector? And what do you mean by all don't work ? Doesn't work in which way? The first method looks fine to me. — 4386427
– 4386427, Commented Jan 29, 2017 at 11:45

Shaana · Accepted Answer · 2017-01-29 11:29:47Z

4

I think your first version should work, because the vectors use contiguous storage locations for their elements.

int iRet = sonicChangeFloatSpeed(&nFloats[0], nFloats.size(), 1, 0.5, 1, 1, 1, 48000, 1);

In c++11 you can use

float* p = nFloats.data();
int iRet = sonicChangeFloatSpeed(p, nFloats.size(), 1, 0.5, 1, 1, 1, 48000, 1);

answered Jan 29, 2017 at 11:29

Shaana

3361 silver badge5 bronze badges

Sign up to request clarification or add additional context in comments.

Comments

4386427 · Accepted Answer · 2017-01-29 12:06:34Z

0

As far as I can tell, your approach is problematic.

Your function seems to do 2 things:

1) Write a number of floats to a stream

2) Read a number of floats from a stream

In both cases you want the floats to be in a vector.

For number 1 it is fine to use a vector but for number 2 it is very problematic.

You can't pass a pointer to the data held by a vector and then change the number of elements. The original vector will keep its original size. If you attempt to read more floats from the stream than currently held by the vector, the behavior is undefined.

answered Jan 29, 2017 at 12:06

4386427

44.7k4 gold badges44 silver badges69 bronze badges

Collectives™ on Stack Overflow

Pass vector as pointer

2 Answers 2

Comments

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

Comments

Comments

Your Answer

Sign up or log in

Post as a guest

Related