Best Way to Extract a Subvector from a Vector

Best way to extract a subvector from a vector?

vector<T>::const_iterator first = myVec.begin() + 100000;
vector<T>::const_iterator last = myVec.begin() + 101000;
vector<T> newVec(first, last);

It's an O(N) operation to construct the new vector, but there isn't really a better way.

Extract a subvector from a vector with iterator

If you want to assign a subrange to an existing std::vector you can use the assign() member:

NewVec.assign(Original.begin(), Original.begin() + 5);

Of course, this assumes that Original has at least 5 elements.

Creating a new C++ subvector?

One of std::vector's constructor accepts a range:

std::vector<int> v;

// Populate v.
for (int i = 1; i <= 10; i++) v.push_back(i);   

// Construct v1 from subrange in v.
std::vector<int> v1(v.begin() + 4, v.end() - 2);

How to extract sub-vectors of different datatypes from vector<char>?

You need to copy the bytes into objects of the right type

std::vector<char> data = /* get from network */;
Eigen::Matrix4d trajectory;
std::vector<uint8_t> rgb_image_data(rgb_image_size);
std::vector<uint8_t> gray_image_data(gray_image_size);

std::memcpy(&trajectory, data.data() + trajectory_offset, sizeof(Eigen::Matrix4d));
std::memcpy(rgb_image_data.data(), data.data() + rgb_image_offset, rgb_image_size);
std::memcpy(gray_image_data.data(), data.data() + gray_image_offset, gray_image_size);

That is assuming that the source copied in a whole Eigen::Matrix4d, rather than the double[4][4] it contains.

Find value in vector and extract subvector from that position

You tried to copy to somewhere which is not allocated.

You can use a constructor of std::vector that takes two InputIterators and creates copy of the range.

Try this:

vector<int> vec = {10, 20, 30, 40, 50, 60};
std::vector<int>::iterator it;
it = find (vec.begin(), vec.end(), 30);
vector<int> newvec(it, vec.end());

Extracting Subvector from Vector Register in LLVM IR

shufflevector will accomplish the same as above (provided you are only interest in %out.8) and LLVM will replace it with a simple register name change (e.g., if %out.1 is ymm0, %out.8 would be xmm0).

Single line to replace eight:

%out.8 = shufflevector <8 x float> %out.0, <8 x float> undef, <4 x i32> <i32 0, i32 1, i32 2, i32 3>

---

Related Topics