#include <bits/stdc++.h>
using namespace std;
#pragma GCC optimize "Ofast"
#define pb push_back
inline constexpr int lg2(int x){return 32 - __builtin_clz(x);}
const int MDIGITS = 7, PW10[MDIGITS] = {0, 10, 100, 1000, 10000, 100000, 1000000};
const int IO_SZ = 6 << 20;
char _buf[IO_SZ], *_ii = _buf, *_oi = _buf;
template <typename T> inline void scan(T &x){for(x=*_ii++-'0'; *_ii++>='0'; x=x*10+_ii[-1]-'0');}
template <typename T> inline void print(T x){int dg=0; while(dg<MDIGITS && PW10[dg]<=x) dg++; for(char *i=_oi+dg-1; i>=_oi; --i, x/=10) *i=x%10+'0'; _oi+=dg;}
// type 1 - increase [l, n-r) by v
// type 2 - query [l, n-r), v is extra
struct Event {
int type, l, r, v;
};
vector<Event> events;
const int MN = 3e5 + 5;
int n, q;
char *arr;
int bit1[MN];
int lvl[MN];
int xArr[MN*2], yArr[MN*2], vArr[MN*2], ai;
int bit1Lb(int x){
if(x <= 0) return -1;
int i = 0;
for(int pw = 1 << lg2(n); pw > 0; pw >>= 1) if(i + pw <= n && bit1[i + pw] < x) x -= bit1[i += pw];
return i;
}
int main(){
fread(_buf, 1, IO_SZ, stdin);
scan(n); scan(q);
arr = _ii; _ii += n+1;
for(int i = 0; i < n; ++i) if(!(arr[i] -= '0')) for(int j = i+1; j <= n; j += j & -j) bit1[j]++;
events.reserve(n*2);
for (int i = 1; i <= q; ++i) {
bool type = *_ii == 't';
_ii += type ? 7 : 6;
if(type){
int a; scan(a); a--;
int cur = 0;
for(int j = a+1; j; j -= j & -j) cur += bit1[j];
int l = bit1Lb(cur - !arr[a]) + 1;
int r = bit1Lb(cur+1);
if(arr[a]){
int prv = lvl[r];
events.pb({1, l, n-r, i-prv});
arr[a] = 0;
lvl[a] = lvl[r] = i;
for(int j = a+1; j <= n; j += j & -j) bit1[j]++;
}
else {
if(l < a){
int prv = lvl[a];
events.pb({1, l, n-a, i-prv});
}
if(r > a+1){
int prv = lvl[r];
events.pb({1, a+1, n-r, i-prv});
}
arr[a] = 1;
lvl[r] = i;
for(int j = a+1; j <= n; j += j & -j) bit1[j]--;
}
}
else {
int l, r;
scan(l); scan(r);
l--; r--;
int qRes = 0;
for(int j = l+1; j; j -= j & -j) qRes += bit1[j];
int mr = bit1Lb(qRes - !arr[l] + 1);
events.pb({2, l, n-r, mr >= r ? i-lvl[mr] : 0});
}
}
for(Event event : events){
if(event.type == 1){
xArr[ai] = event.l;
yArr[ai] = event.r;
vArr[ai] = event.v;
ai++;
}
else {
int x = event.l, y = event.r, res = event.v;
for (int i = 0; i < ai; i++) {
if(xArr[i] <= x && yArr[i] <= y) res += vArr[i];
}
print(res); *_oi++ = '\n';
}
}
fwrite(_buf, 1, _oi-_buf, stdout);
}
