#pragma GCC optimize("02,unroll-loops")
#include <bits/stdc++.h>
#define FOR(i, l, r) for(int i = l; i <= r; i++)
#define FOD(i, r, l) for(int i = r; i >= l; i--)
#define rep(i, n) for(int i = 0; i < n; i++)
#define pb push_back
#define ep emplace_back
#define fi first
#define se second
#define sz(A) (int) A.size()
#define bit(mask, i) ((mask >> i) & 1)
#define lon(x) ((1ll << x))
#define all(A) A.begin(), A.end()
#define el cout << '\n';
#define IOI_26_Gold_Medalist int main()
using namespace std;
const int maxn = 1e6 + 5;
const int mod = 1e9 + 7;
using ii = pair<int, int>;
using ll = long long;
const ll inf = 3e18;
using str = string;
using vi = vector<int>;
template <class X, class Y>
bool minimize(X &x, const Y &y) {
if(x > y) {
x = y;
return true;
}
else return false;
}
template <class X, class Y>
bool maximize(X &x, const Y &y) {
if(x < y) {
x = y;
return true;
}
else return false;
}
int n;
vector<ii> pts;
namespace sub1 {
int ans = 0;
void solve() {
unordered_map<int, int> cntx, cnty;
rep(mask, lon(n)) {
bool valid = true;
cntx.clear(); cnty.clear();
rep(i, n) if(bit(mask, i)) {
cntx[pts[i].fi]++;
cnty[pts[i].se]++;
if(cntx[pts[i].fi] > 2) valid = false;
if(cnty[pts[i].se] > 2) valid = false;
if(!valid) break;
}
if(!valid) continue;
rep(i, n) if(!bit(mask, i)) {
bool okL = false, okR = false;
rep(j, n) if(bit(mask, j)) {
if(pts[j].fi == pts[i].fi) {
if(pts[j].se < pts[i].se) okL = true;
else okR = true;
}
}
if(okL && okR) continue;
bool okT = false, okB = false;
rep(j, n) if(bit(mask, j)) {
if(pts[j].se == pts[i].se) {
if(pts[j].fi < pts[i].fi) okT = true;
else okB = true;
}
}
if(okT && okB) continue;
valid = false; break;
}
if(valid) ans = mask;
}
rep(i, n) cout << bit(ans, i);
}
}
namespace sub2 {
unordered_map<int, int> cntx;
bool check() {
rep(i, n) {
cntx[pts[i].fi]++;
if(cntx[pts[i].fi] > 2) return false;
}
return true;
}
void solve() {
vi comp;
for(auto &pr : pts) {
int x = pr.se;
comp.pb(x);
}
sort(all(comp));
comp.erase(unique(all(comp)), comp.end());
int K = sz(comp);
vi maxR(K + 1, 0), minR(K + 1, 1e9), idmaxR(K + 1, 0), idminR(K + 1, 0);
rep(i, n) {
int x = lower_bound(all(comp), pts[i].se) - comp.begin() + 1;
if(maximize(maxR[x], pts[i].fi)) {
idmaxR[x] = i;
}
if(minimize(minR[x], pts[i].fi)) {
idminR[x] = i;
}
}
vi dd(n, 0);
FOR(i, 1, K) {
dd[idmaxR[i]] = 1;
dd[idminR[i]] = 1;
// cerr << maxR[i] << " " << minR[i] << endl;
}
rep(i, n) cout << dd[i];
}
}
namespace sub3 {
int ans[maxn];
vector<pair<int,int>> col[maxn], row[maxn];
int cnt[maxn];
void solve() {
int maxX = 0, maxY = 0;
rep(i, n) {
maximize(maxX, pts[i].fi);
maximize(maxY, pts[i].se);
}
for (int x = 0; x <= maxX; ++x) {
col[x].clear();
cnt[x] = 0;
}
for (int y = 0; y <= maxY; ++y) row[y].clear();
rep(i, n) ans[i] = 0;
rep(i, n) {
int x = pts[i].fi, y = pts[i].se;
col[x].ep(y, i);
row[y].ep(x, i);
}
for (int x = 0; x <= maxX; ++x) {
if (col[x].empty()) continue;
sort(all(col[x]));
int idBot = col[x].front().second;
int idTop = col[x].back().second;
if (!ans[idBot]) {
ans[idBot] = 1; cnt[x]++;
}
if (idTop != idBot && !ans[idTop]) {
if (cnt[x] < 2) {
ans[idTop] = 1; cnt[x]++;
}
}
}
for (int y = 0; y <= maxY; ++y) {
if (row[y].empty()) continue;
sort(all(row[y]));
int xL = row[y].front().first, idL = row[y].front().second;
int xR = row[y].back().first, idR = row[y].back().second;
if (!ans[idL] && cnt[xL] < 2) {
ans[idL] = 1; cnt[xL]++;
}
if (!ans[idR] && cnt[xR] < 2) {
ans[idR] = 1; cnt[xR]++;
}
}
rep(i, n) cout << ans[i];
}
}
IOI_26_Gold_Medalist {
ios_base::sync_with_stdio(false);
cin.tie(NULL); cout.tie(NULL);
#define TASK "SLAMP"
if(fopen(TASK".INP", "r")) {
freopen(TASK".INP", "r", stdin);
freopen(TASK".OUT", "w", stdout);
}
cin >> n;
rep(i, n) {
int x, y; cin >> x >> y;
pts.ep(x, y);
}
if(n <= 16) return sub1 :: solve(), 0;
if(sub2 :: check()) return sub2 :: solve(), 0;
return sub3 :: solve(), 0;
return 0;
}
/*
3
1 1
1 5
1 4
9
1 1
1 2
5 3
2 1
4 2
2 3
3 1
3 2
4 3
111011100
111001011
8
1 2
1 4
1 5
2 2
2 5
4 2
4 4
4 5
10100101
10111101
*/
