Kickstart 2021 D: Cutting Intervals, scanf vs cin

So I checking out the “Analysis” for the Cutting Intervals problem and wrote this

void solve(int testcase) {
ll n, c; scanf("%lld %lld", &n, &c);

map<ll, ll> m;
fo(_, 0, n) {
    ll tl, tr; scanf("%lld %lld", &tl, &tr);
    ++m[tl+1], --m[tr];

// iter(x, m) printf("{%lld: %lld}, ", x.first, x.second);
// printf("\n");

map<ll, ll> a;
ll j = 0;
auto mprev = m.begin(), mcurr = ++m.begin();

while (mcurr != m.end()) {
    ll kprev = mprev->first, kcurr = mcurr->first;
    // printf("%lld %lld\n", kprev, kcurr);

    j += mprev->second;
    a[j] += kcurr - kprev;

    ++mprev, ++mcurr;

// iter(x, a) printf("{%lld: %lld}, ", x.first, x.second);
// printf("\n");

ll res = n;
for (auto it = a.rbegin(); it != a.rend(); ++it) {
    res += it->first * min(it->second, c);
    c -= min(it->second, c);

    if (c == 0) break;

printf("Case #%d: %lld\n", testcase, res);


I believe the macros are self-explanatory. This gave “Sample Failed”. However, it worked on my machine. So I sent a manual “Run Test” and the output was Case #1: 1144119736199078559
instead of Case #1: 7. But then when I switched to cin over scanf it gave AC.

Any idea why?

@ssjgz, tagging you because I have seen you answering such questions before.
(Apologies if you don’t want to be tagged like this and I’ll keep this in mind next time)

Were you using Fast input output ?

Yes! I thought it just caused cin and and scanf to be out of sync, so I made sure to only use `scanf.