How do I use a for loop to erase elements from a vector by its index ? I am getting a vector out of range error. I have a sample code below.
vector<int> to_erase = {0, 1, 2};
vector<int> data = {3, 3, 3, 3};
for(int i = 0; i < to_erase.size(); i++) {
data.erase(data.begin() + to_erase[i]);
}
I think it is because the size of my vector reduces through every iteration therefore it cannot access index 2.
You would normally employ the erase–remove idiom to delete multiple elements from a vector efficiently (erasing them one by one is generally less efficient, and, as you’ve seen, not always trivial). In its most general form, the idiom looks like this:
data.erase(remove_algorithm(begin(data), end(data)), end(data));
In your case, the remove_algorithm is based off indices in another vector, so we need to provide those as well:
data.erase(
remove_indices(begin(data), end(data), begin(to_erase), end(to_erase)),
end(data));
Unfortunately, such an algorithm isn’t contained in the standard library. However, it’s trivial to write yourself1:
template <typename It, typename It2>
auto remove_indices(It begin, It end, It2 idx_b, It2 idx_e) -> It {
using idx_t = typename std::iterator_traits<It2>::value_type;
std::sort(idx_b, idx_e, std::greater<idx_t>{});
for (; idx_b != idx_e; ++idx_b) {
auto pos = begin + *idx_b;
std::move(std::next(pos), end--, pos);
}
return end;
}
Here, we first sort the indices to be removed from largest to smallest. Next, we loop over those indices. We then (maximally efficiently) move all elements between the current position (to be deleted) and the end of the vector forward by one. Subsequently, the end is decreased by one (to account for the fact that an element got deleted).
Live code
1 *Ahem* Once you’ve removed all the silly typos in your code.
I think it is because the size of my vector reduces through every iteration
Yes!
You could do it by keeping an extra variable, which counts the elements that are deleted, like this:
#include <iostream>
#include <vector>
using namespace std;
int main() {
vector<int> to_erase = {0, 1, 2};
vector<int> data = {3, 3, 3, 3};
int count_removed = 0;
for(unsigned int i = 0; i < to_erase.size(); i++)
data.erase(data.begin() + to_erase[i] - count_removed++);
for(unsigned int i = 0; i < data.size(); ++i)
cout << data[i] << "\n";
return 0;
}
Output:
3
I had the same problem when I first used std::erase(), good question, +1.
Deleting a collection of elements meanwhile you iterate, is unsafe and probalby expensive. I would suggest that each element that meets your criteria gets swapped with an element at the end. (at the end because will be cheaper to erase from the end. You can keep track of how much back you came from the end of the vector (based on the number of swap), and break early our of the loop. Now based on how many elements you swapped you can do something like:
data.resize(data.size() - reverse_counter);
or
int pos = data.size() - reverse_counter;
data.erease(data.begin()+pos, data.end();
It is sudo code just to explain the idea.
As mentioned in the reference, erase not at the end cause re-allocation, which is expensive. Something worth keep in mind:
http://www.cplusplus.com/reference/vector/vector/erase/
I think this is a bad design, because you will change the for loop invariant and will need a lot of workaround to make this happen. Anyway, if you really want to use a for loop, you MAY flag what you whant to delete and run a stl remove_if, something like:
#include <iostream>
#include <vector>
#include <limits>
#include <algorithm>
using namespace std;
int main() {
vector<int> to_erase = {0, 1, 2};
vector<int> data = {3, 3, 3, 3};
cout << "Before:\n" ;
for(int i=0; i<data.size(); i++)
cout << i << "\t";
cout << endl;
for(int i=0; i<data.size(); i++)
cout << data[i] << "\t";
cout << endl;
for(int i = 0; i < to_erase.size(); i++) {
//data.erase(data.begin() + to_erase[i]);
data[i] = numeric_limits<int>::max();
}
data.erase(remove_if(data.begin(),
data.end(),
[](int i){return i==numeric_limits<int>::max();}), data.end());
cout << "Next:\n" ;
for(int i=0; i<data.size(); i++)
cout << i << "\t";
cout << endl;
for(int i=0; i<data.size(); i++)
cout << data[i] << "\t";
return 0;
}
Related
I want to find the duplicate numbers in this array, I'm using vectors and want to print the answer, what am I doing wrong here?
#include <iostream>
#include <vector>
using namespace std;
int findDuplicate(vector<int> &arr)
{
int ans = 0;
// XOR all elements
for (int i = 0; i < arr.size(); i++)
{
ans = ans ^ arr[i];
}
// XOR [1, n-1]
for (int i = 1; i < arr.size(); i++)
{
ans = ans ^ i;
}
cout << ans;
return ans;
}
int main()
{
vector<int> arr = {5, 2, 5, 2, 7, 6, 6};
findDuplicate(arr);
}
There are a number of things that are wrong. Let's start with the two easy ones.
You have a cout statement that prints (it turns out) 0. But you don't do an endl, so you don't get a newline.
cout << ans << endl;
That will actually print a newline, which makes it easier to read.
Second, your method returns a value, which is ignored in main(). You probably want to do this in main:
int answer = findDuplicate(arr);
cout << "And the answer is " << answer << endl;
Or something like that.
That's all fine and good. That's the easy stuff. But why do you think this XOR code is going to tell you what duplicates their are, especially when there might be multiples of the same value or more than one value with duplicates. Maybe there is some algorithm you know about that none of us do.
But it's printing out 0 when the data clearly has duplicates.
Every duplicate finder I know about that's remotely efficient sorts the data then does a loop through it, keeping track of the last value, and if the next value equals the previous value, you have a duplicate.
#include <algorithm>
std::sort(vec.begin(), vec.end());
int prevValue = vec[0];
for (int index = 1; index < vec.size(); ++index) {
int thisValue = vec[index];
if (thisValue == prevValue) {
... Print it
}
prevValue = thisValue;
}
You can make this smarter if you want to know how many are duplicated, and want to be smart about not printing 6 is duplicated 17 times in a row.
I am trying to implement a small program that iterates over a 2d vector backwards and adds the value of the element.
If that element have already been added then I want to overwrite the value of the vector with 99.
So for example if number of climbs is four then add the of the program points should have the value 5 and the vector should look like this at the end
{1, 1 ,1},
{99, 1, 1},
{99(should start here), 99, 99}
But I keep getting a segmentation fault and I don't know whether I am iterating over the vector backwards incorrectly.
This is my full code
#include <iostream>
#include <vector>
using namespace std;
int main()
{
vector<vector<int>> vect
{ {3}, //finish here
{1, 1 ,1},
{2, 1, 1},
{1, 1, 1} //start here
};
int points = 0;
for (int i = 0; i < vect.size(); i++)
{
for (int j = 0; j < vect[i].size(); j++)
{
cout << vect[i][j] << " ";
}
cout << endl;
}
int visited = 99;
int number_of_climbs = 4;
for(int i = 4; i >= 0; i--)
for (int j = 0; j < number_of_climbs; j++)
{
if(vect[i][j] != 99)
{
points += vect[i][j];
vect[i][j] = 99;;
continue;
}
}
return 0;
}
Looping over things backwards always trips me up too, especially when I accidentally declare my loop using an unsigned int. Your backwards loop is close to correct, but can be simplified into:
for (int i = vect.size() - 1; i >= 0; --i)
{
for (int j = vect[i].size() - 1; j >= 0; --j)
{
}
}
You need to start at size() - 1 because the only indices available are [0, size) (that is EXCLUSIVE of size), so in a vector of 4 elements, that's 0, 1, 2 and 3 (and you need to start at 3), which is likely the cause of your segmentation fault since you start your initial loop at 4 so it's immediately out-of-bounds.
Here in the inner loop you can see we account for the vectors being different sizes by just using the size of each vector.
It's also possible to use reverse iterators, but iterators can be confusing to people:
for (auto outer = vect.rbegin(); outer != vect.rend(); ++outer)
{
for (auto inner = outer->rbegin(); inner != (*outer).rend(); ++inner)
{
}
}
Note I've shown both ways to access the iterator, using it-> or (*it). Within the inner loop, you will simply use *inner to get the actual value held by the vector, so if you simply had std::cout << *inner << " "; that would print the 2d vector on one line in reverse order.
Even though we're looping in reverse, we still use ++
i am currently trying to learn some C++ and now i got stuck in an exercise with vectors. So the task is to read ints from a text file and store them in the vector which should be dynamic.
I guess there is something wrong with the while-loop?
If I start this, the program fails and if I set the vector size to 6, I get
6 0 0 0 0 0 as output.
Thanks for any hints.
int main()
{
const string filename = "test.txt";
int s = 0;
fstream f;
f.open(filename, ios::in);
vector<int> v;
if (f){
while(f >> s){
int i = 0;
v[i] = s;
i = i+1;
}
f.close();
}
for(int i = 0; i < 6; i++){
cout << v[i] << "\n";
}
}
You don't grow the vector. It is empty and cannot hold any ints. You'll need to either resize it every time you want to add another int or you use push_back which automatically enlarges the vector.
You set i = 0 for every iteration so you would change the first value of the vector every iteration instead of the next one.
Go for:
v.push_back(s);
in your loop and
for(int i = 0; i < v.size(); i++) { // ...
Remark:
You normally don't hardcode vector sizes/bounds. One major point about using std::vector is its ability to behave dynamically with respect to its size. Thus, the code dealing with vectors should not impose any restrictions about the size of the vector onto the respective object.
Example:
for(int i = 0; i < 6; i++){ cout << v[i] << "\n"; }
requires the vector to have at least 6 elements, otherwise (less than 6 ints) you access values out of bounds (and you potentially miss elements if v contains more than 6 values).
Use either
for(int i = 0; i < v.size(); i++){ cout << v[i] << "\n"; }
or
for(std::vector<int>::const_iterator i = v.begin(); i != v.end(); ++i)
{
cout << *i << "\n";
}
or
for(auto i = v.begin(); i != v.end(); ++i)
{
cout << *i << "\n";
}
or
for(int x : v){ cout << x << "\n"; }
or
for(auto && x : v){ cout << x << "\n"; }
or
std::for_each(v.begin(), v.end(), [](int x){ std::cout << x << "\n"; });
or variants of the above which possibly pre-store v.size() or v.end()
or whatever you like as long as you don't impose any restriction on the dynamic size of your vector.
The issue is in the line i= 0. Fixing that will give an issue in the line v[i] = s.
You always initialise i to 0 in the while loop, and that is responsible for the current output. You should shift it out of the while loop.
After fixing that, you have not allocated memory to that vector, and so v[i] doesn't make sense as it would access memory beyond bounds. This will give a segmentation fault. Instead, it should be v.push_back(i), as that adds elements to the end of a vector, and also allocates memory if needed.
If you are using std::vector you can use v.push_back(i) to fill this vector
Error is this line int i = 0;
because you declare i=0 every time in while-loop.
To correct this move this line outside from loop.
Note: this will work, if you declare v like normal array for example int v[101]
When you use std vectors you can just push element at the end of vector with v.push_back(element);
v[i] = s; //error,you dont malloc room for vector
change into : v.push_back(s);
I'm trying to remove the same integer numbers in a vector. My aim is to have only one copy them. Well I wrote a simple code, but it doesn't work properly. Can anyone help? Thanks in advance.
#include <iostream>
#include <vector>
using namespace std;
int main()
{
int a = 10, b = 10 , c = 8, d = 8, e = 10 , f = 6;
vector<int> vec;
vec.push_back(a);
vec.push_back(b);
vec.push_back(c);
vec.push_back(d);
vec.push_back(e);
vec.push_back(f);
for (int i=vec.size()-1; i>=0; i--)
{
for(int j=vec.size()-1; j>=0; j--)
{
if(vec[j] == vec[i-1])
vec.erase(vec.begin() + j);
}
}
for(int i=0; i<vec.size(); i++)
{
cout<< "vec: "<< vec[i]<<endl;
}
return 0;
}
Don't use a list for this. Use a set:
#include <set>
...
set<int> vec;
This will ensure you will have no duplicates by not adding an element if it already exists.
To remove duplicates it's easier if you sort the array first. The code below uses two different methods for removing the duplicates: one using the built-in C++ algorithms and the other using a loop.
#include <iostream>
#include <vector>
#include <iterator>
#include <algorithm>
using namespace std;
int main() {
int a = 10, b = 10 , c = 8, d = 8, e = 10 , f = 6;
vector<int> vec;
vec.push_back(a);
vec.push_back(b);
vec.push_back(c);
vec.push_back(d);
vec.push_back(e);
vec.push_back(f);
// Sort the vector
std::sort(vec.begin(), vec.end());
// Remove duplicates (v1)
std::vector<int> result;
std::unique_copy(vec.begin(), vec.end(), std::back_inserter(result));
// Print results
std::cout << "Result v1: ";
std::copy(result.begin(), result.end(), std::ostream_iterator<int>(cout, " "));
std::cout << std::endl;
// Remove duplicates (v2)
std::vector<int> result2;
for (int i = 0; i < vec.size(); i++) {
if (i > 0 && vec[i] == vec[i - 1])
continue;
result2.push_back(vec[i]);
}
// Print results (v2)
std::cout << "Result v2: ";
std::copy(result2.begin(), result2.end(), std::ostream_iterator<int>(cout, " "));
std::cout << std::endl;
return 0;
}
If you need to save initial order of numbers you can make a function that will remove duplicates using helper set<int> structure:
void removeDuplicates( vector<int>& v )
{
set<int> s;
vector<int> res;
for( int i = 0; i < v.size(); i++ ) {
int x = v[i];
if( s.find(x) == s.end() ) {
s.insert(x);
res.push_back(x);
}
}
swap(v, res);
}
The problem with your code is here:
for(int j=vec.size()-1; j>=0; j--)
{
if(vec[j] == vec[i-1])
vec.erase(vec.begin() + j);
}
there's going to be a time when j==i-1 and that's going to kill your algorithms and there will be a time when i-1 < 0 so you will get an out of boundary exception.
What you can do is to change your for loop conditions:
for (int i = vec.size() - 1; i>0; i--){
for(int j = i - 1; j >= 0; j--){
//do stuff
}
}
this way, your the two variables your comparing will never be the same and your indices will always be at least 0.
Others have already pointed to std::set. This is certainly simple and easy--but it can be fairly slow (quite a bit slower than std::vector, largely because (like a linked list) it consists of individually allocated nodes, linked together via pointers to form a balanced tree1.
You can (often) improve on that by using an std::unordered_set instead of a std::set. This uses a hash table2 instead of a tree to store the data, so it normally uses contiguous storage, and gives O(1) expected access time instead of the O(log N) expected for a tree.
An alternative that's often faster is to collect the data in the vector, then sort the data and use std::unique to eliminate duplicates. This tends to be best when you have two distinct phases of operation: first you collect all the data, then you need duplicates removed. If you frequently alternate between adding/deleting data, and needing a duplicate free set, then something like std::set or std::unordered_set that maintain the set without duplicates at all times may be more useful.
All of these also affect the order of the items. An std::set always maintains the items sorted in a defined order. With std::unique you need to explicit sort the data. With std::unordered_set you get the items sorted in an arbitrary order that's neither their original order nor is it sorted.
If you need to maintain the original order, but without duplicates, you normally end up needing to store the data twice. For example when you need to add a new item, you attempt to insert it into an std::unordered_set, then if and only if that succeeds, add it to the vector as well.
Technically, implementation as a tree isn't strictly required, but it's about the only possibility of which I'm aware that can meet the requirements, and all the implementations of which I'm aware are based on trees.
Again, other implementations might be theoretically possible, but all of which I'm aware use hashing--but in this case, enough of the implementation is exposed that avoiding a hash table would probably be even more difficult.
The body of a range for must not change the size of the sequence over which it is iterating..
you can remove duplicates before push_back
void push(std::vector<int> & arr, int n)
{
for(int i = 0; i != arr.size(); ++i)
{
if(arr[i] == n)
{
return;
}
}
arr.push_back(n);
}
... ...
push(vec, a);
push(vec, b);
push(vec, c);
...
I want to reuse a std::vector within a for loop. However, I need the vector to be empty for each iteration step of the for loop.
Question: How can I empty a vector rapidly without changing its capacity in the most efficient way?
What I used so far is
std::vector<int> myVec;
for(int i=0; i<A_BIG_NUMBER; ++i) {
std::vector<T>().swap(myVec);
myVec.reserve(STANDARD_MAXIMUM);
/// .. doing business
}
Cheers!
Solution:
Thanks for the answers, here is how I implemented (checked) it:
#include <vector>
#include <iostream>
int main() {
int n = 10;
std::vector< int > myVec;
myVec.reserve(n);
for(int j=0; j<3; ++j) {
myVec.clear();
for(int i=0; i<n; ++i) {
myVec.push_back(i);
}
for(int i=0; i<myVec.size(); ++i) {
std::cout << i << ": " << myVec[i] << std::endl;
}
}
return 0;
}
EDIT: changed from operator[] to push_back.
Use vector::clear method. It will clear the content without reducing its capacity.
myVec.clear();
This is equivalent to myVec.erase(myVec.begin(), myVec.end()).
To retain the current size of a vector with default values for its content, you can assign default values to the vector. In the case of a vector of ints, you can do the following:
myVec.assign( myVec.size(), 0 );
use clear method as below:
std::vector<int> myVec;
for(int i=0; i<A_BIG_NUMBER; ++i)
{
std::vector<T>().swap(myVec);
myVec.reserve(STANDARD_MAXIMUM);
/// .. doing business
myVec.clear();
}
Answer based on OP's solution:
The normal approach for containers is to start with an empty container and fill it up as needed with an exception for std::vector where you can reserve space eventhough there are still no objects in the container.
If you want a different approach where an "empty container" would be a container of default objects that you can access like an array (only works with std::vector and std::deque), then you need to start with resize() and you can "clean up" with fill:
int n = 10;
std::vector<int> myVec;
myVec.resize(n);
myVec[4] = 5;
std::cout << myVec[4] << "\n";
std::fill(myVec.begin(), myVec.end(), int()); // ofcourse for int, you can use 0 instead of int()
std::cout << myVec[4] << "\n";