Problem Statement

You are looking for a restaurant for a year-end party, but there are many restaurants in the world, and you do not have enough time to look at all of them.
For each restaurant $i (1 \le i \le N)$, you have data on three axes: the price rating $P_i$, the allowed length of stay rating $T_i$, and the review rating $R_i$. You decided to use this data to reduce the number of restaurants you need to check.

Restaurant $j$ is said to be inferior to restaurant $i$ when at least one of 1, 2, 3 holds.
1. $ P_j \lt P_i$ and $T_j \le T_i$ and $R_j \le R_i$
2. $ P_j \le P_i$ and $T_j \lt T_i$ and $R_j \le R_i$
3. $ P_j \le P_i$ and $T_j \le T_i$ and $R_j \lt R_i$

List the restaurants that are not inferior to any restaurant, in ascending order.

* It has been confirmed that AC is achievable in C++ and Rust. The writer has not achieved AC in PyPy.

Input

$N$
$P_1\ T_1\ R_1$
$P_i\ T_i\ R_i$
$P_N\ T_N\  R_N$

The first line contains the number of restaurants $N$.
Line $i+1$ contains the price rating $P_i$, the allowed length of stay rating $T_i$, and the review rating $R_i$ of restaurant $i (1 \le i \le N)$, separated by single-byte spaces.

$ 1 \le N \le 10^5$
$ 0 \le P_i,T_i, R_i \le 10^4$
$ i \ne j \Leftrightarrow (P_i, T_i, R_i) \ne (P_j, T_j, R_j)$

Output

Print a newline at the end.

Sample

5
5 2 1
4 4 4
3 3 1
0 0 1
5 8 0

1
2
5

3
1 0 0
0 1 0
0 0 1

1
2
3

3
1 1 1
2 3 4
5 5 5

3