Loops · C++ tutorial

The counting for#

for (int i = 0; i < 5; i++) {
    cout << i << " ";       // 0 1 2 3 4
}

Start at 0, keep going while i < 5, add one each round. Those are exactly the valid positions of a collection of five items.

Range-based for#

When you want each item and not its position, C++ has a shorter loop:

vector<int> nums = {4, 8, 15, 16, 23, 42};
long long total = 0;
for (int x : nums) {
    total += x;
}

for (char c : string("hey")) {
    cout << c << "\n";
}

Read for (int x : nums) as "for each x in nums". Each x is a copy. To change the items, take a reference:

for (int &x : nums) x *= 2;          // doubles every element in place

For large items such as strings, const auto & avoids copying without allowing changes:

for (const auto &word : words) cout << word << "\n";

while and do while#

int n = 4071, digits = 0;
while (n > 0) {
    n /= 10;
    digits++;
}

int x;
do {
    cin >> x;
} while (x <= 0);       // runs at least once

Something inside a while must change so it eventually stops.

break and continue#

break leaves the loop; continue skips to the next round:

for (int x : nums) {
    if (x < 0) break;
    if (x % 2 == 1) continue;
    cout << x << "\n";
}

Comparing sizes#

v.size() is unsigned, so v.size() - 1 on an empty vector is not -1 but a huge number, and a loop like for (int i = 0; i < v.size() - 1; i++) misbehaves. Convert first:

int n = v.size();
for (int i = 0; i + 1 < n; i++) {
    // compare v[i] with v[i + 1]
}

Cost#

Two nested loops over n items take n × n steps. Fine for a thousand items, far too slow for a hundred thousand. Most of algorithms is turning the second loop into something smarter.

Key points#

for (int i = 0; i < n; i++) counts; for (auto x : v) visits each item.
Use & in a range-for to change items, and const auto & to avoid copies.
size() is unsigned: store it in an int before subtracting.
break leaves; continue skips ahead.

Next: functions, and passing by reference.