The underlying principle of vector

Investigate vector

One 、 principle ： The dynamic array vector Underlying principles and related interview questions - Handsome play with programming

Two 、 Use ：

 

 

 

 

#if 1

int main() {
	vector<int>  iv(3, 9);
	for (auto e : iv) cout << e << endl;

	cout << "============================" << endl;
	iv.push_back(1);
	iv.push_back(2);
	iv.push_back(3);
	iv.push_back(4);
	iv.push_back(7);
	iv.push_back(0);
	for (auto e : iv) cout << e << endl;
		return 0;
	
}
#endif

（2）vector Medium reserve and resize The difference between

reserve Is directly expanded to a certain size , Can reduce the number of times to open up 、 The problem of releasing space （ Optimize push_back）, You can improve efficiency , Secondly, it can reduce the problem of copying data many times .reserve It's just a guarantee that vector The size of the space in （capacity） At least the size specified by the parameter n.reserve() There is only one parameter .

resize() You can change the size of the available space , There is also a function to change the default value .capacity The size of the will also change with .resize() Can have multiple parameters .

（3）vector Medium size and capacity The difference between .

size At present vector How many elements are there in （finish – start）, and capacity Function indicates how many elements it can hold in the allocated memory （end_of_storage – start）.

（4）vector The element type of can be a reference ？

vector The underlying implementation of requires a continuous arrangement of objects , References are not objects , There is no real address , therefore vector Element cannot be of reference type .

（5）vector Iterator failures

When inserting an element into vector in , Due to memory reallocation , So the iterators that point to the original memory all fail .

When you delete an element in a container , The element that the iterator points to has been deleted , Then it also causes the iterator to fail .erase Method returns the next valid iterator , So when we want to delete an element , need it=vec.erase(it);.

（6） Release... Correctly vector Of memory (clear(), swap(), shrink_to_fit())

vec.clear()： Empty content , But don't free memory .

vector().swap(vec)： Empty content , And free memory , Want a brand new vector.

vec.shrink_to_fit()： Request the container to lower its capacity and size matching .

vec.clear();vec.shrink_to_fit();： Empty content , And free memory .

（7）vector Why should we expand the capacity with 1.5 Times or 2 Double expansion ?

According to the information we have looked up , Consider the possible waste of heap space , Multiple growth can't be too big , There are two widely used expansion methods , With 2 Double the way to expand , Or with 1.5 Double the way to expand .

With 2 Double the way to expand , As a result, the memory of the next application must be larger than the total memory allocated before , As a result, the previously allocated memory can no longer be used , So it's better to set the multiplier factor to (1,2) Between ：