Here is a version of your code written in reasonably decent C++. I didn't touch the loop in the middle because I have no clue what it's doing. You're using obscure variable names and no comments and doing all kinds of bizarre things with indexes and mixing them up with user input.
Now, reading indexes from user input and using them isn't bad, though in a real program you'd want to be doing lots of bounds checking on that input to make sure people weren't feeding you bad data. But doing all that stuff with such poorly named variables with no explanation is going to leave anybody looking at it scratching their head. Don't do that.
Also, avoid the use of begin and end as variable names, especially if they hold indexes. In most cases it will confuse things terribly as begin and end are important identifiers in the standard library and always refer to iterators, which are sort of like indexes, but most definitely not indexes, adding greatly to the confusion. beginidx and endidx could be acceptable substitutes in this case.
#include <vector>
#include <algorithm>
#include <iostream>
#include <iterator>
using ::std::vector;
using ::std::sort;
using ::std::copy_n;
using ::std::copy;
using ::std::fill;
using ::std::istream_iterator;
using ::std::ostream_iterator;
using ::std::cin;
using ::std::cout;
int main() {
// your code goes here
using vec_el_t = long long int;
int t;
cin >> t;
while (t--) {
int const n = []() { int n; cin >> n; return n; }();
vector<vec_el_t> c{n}, h{n}, a{n};
copy_n(istream_iterator<vec_el_t>{cin}, n, c.begin());
copy_n(istream_iterator<vec_el_t>{cin}, n, h.begin());
// Suggested debugging code:
// cout << "h before sort: "
// copy(h.begin(), h.end(), ostream_iterator<vec_el_t>{cout, " "});
// cout << '\n';
sort(h.begin(), h.end());
// Suggested debugging code:
// cout << "h after sort: "
// copy(h.begin(), h.end(), ostream_iterator<vec_el_t>{cout, " "});
// cout << '\n';
fill(a.begin(), a.end(), 0);
// Weird, unexplained algorithm begins here
int i = 0;
int begin = (i + 1) - c[i];
int end = (i + 1) + c[i];
int j = begin;
while (i < n) {
a[j - 1]++;
j++;
if (j > end){
i++;
begin = (i + 1) - c[i];
end = (i + 1) + c[i];
j = begin;
}
}
// Weird unexplained algorithm ends here
sort(a.begin(), a.end());
copy(h.begin(), h.end(), ostream_iterator<vec_el_t>{cout, " "});
}
return 0;
}
Changes made... Use vector not a variable length array, which isn't valid C++ anyway and will only work in g++ and clang. Don't use explicit loops if there is an algorithm that will do the job. Try to make as many things const as you can so you can make sure that the compiler catches it if you try to change things you didn't mean to change. Avoid using std; and if you want to import names from ::std import exactly the ones you need. Don't use compiler or library implementation specific header files and use the ones from the standard instead (i.e. no bits/stdc++.h).
As for your problem, I have no idea. I suspect that the index manipulation combined with looping isn't doing what you expect. If you print out the arrays before and after sorting, you will discover that sort only alters order, and not content.
As a general rule, always suspect your own code first and make absolutely sure it's correct. And if you really think it's the library code, prove it beyond a shadow of a doubt before coming here to ask why the library isn't doing what it says it does.
The complicated code I didn't touch looks rife with opportunities for out-of-bounds access, and that results in undefined behavior, which means your program might do absolutely anything in that case. You might change uses of operator [] with calls to the at function (one of the many perks of using vector) instead. That way, attempts at out-of-bounds access will throw an exception.
