Rating changes for last rounds are temporarily rolled back. They will be returned soon. ×

E. Data Structures Fan
time limit per test
2 seconds
memory limit per test
256 megabytes
input
standard input
output
standard output

You are given an array of integers $$$a_1, a_2, \ldots, a_n$$$, as well as a binary string$$$^{\dagger}$$$ $$$s$$$ consisting of $$$n$$$ characters.

Augustin is a big fan of data structures. Therefore, he asked you to implement a data structure that can answer $$$q$$$ queries. There are two types of queries:

  • "1 $$$l$$$ $$$r$$$" ($$$1\le l \le r \le n$$$) — replace each character $$$s_i$$$ for $$$l \le i \le r$$$ with its opposite. That is, replace all $$$\texttt{0}$$$ with $$$\texttt{1}$$$ and all $$$\texttt{1}$$$ with $$$\texttt{0}$$$.
  • "2 $$$g$$$" ($$$g \in \{0, 1\}$$$) — calculate the value of the bitwise XOR of the numbers $$$a_i$$$ for all indices $$$i$$$ such that $$$s_i = g$$$. Note that the $$$\operatorname{XOR}$$$ of an empty set of numbers is considered to be equal to $$$0$$$.

Please help Augustin to answer all the queries!

For example, if $$$n = 4$$$, $$$a = [1, 2, 3, 6]$$$, $$$s = \texttt{1001}$$$, consider the following series of queries:

  1. "2 $$$0$$$" — we are interested in the indices $$$i$$$ for which $$$s_i = \tt{0}$$$, since $$$s = \tt{1001}$$$, these are the indices $$$2$$$ and $$$3$$$, so the answer to the query will be $$$a_2 \oplus a_3 = 2 \oplus 3 = 1$$$.
  2. "1 $$$1$$$ $$$3$$$" — we need to replace the characters $$$s_1, s_2, s_3$$$ with their opposites, so before the query $$$s = \tt{1001}$$$, and after the query: $$$s = \tt{0111}$$$.
  3. "2 $$$1$$$" — we are interested in the indices $$$i$$$ for which $$$s_i = \tt{1}$$$, since $$$s = \tt{0111}$$$, these are the indices $$$2$$$, $$$3$$$, and $$$4$$$, so the answer to the query will be $$$a_2 \oplus a_3 \oplus a_4 = 2 \oplus 3 \oplus 6 = 7$$$.
  4. "1 $$$2$$$ $$$4$$$" — $$$s = \tt{0111}$$$ $$$\to$$$ $$$s = \tt{0000}$$$.
  5. "2 $$$1$$$" — $$$s = \tt{0000}$$$, there are no indices with $$$s_i = \tt{1}$$$, so since the $$$\operatorname{XOR}$$$ of an empty set of numbers is considered to be equal to $$$0$$$, the answer to this query is $$$0$$$.

$$$^{\dagger}$$$ A binary string is a string containing only characters $$$\texttt{0}$$$ or $$$\texttt{1}$$$.

Input

The first line of the input contains one integer $$$t$$$ ($$$1 \le t \le 10^4$$$) — the number of test cases in the test.

The descriptions of the test cases follow.

The first line of each test case description contains an integer $$$n$$$ ($$$1 \le n \le 10^5$$$) — the length of the array.

The second line of the test case contains $$$n$$$ integers $$$a_1, a_2, \dots, a_n$$$ ($$$1 \le a_i \le 10^9$$$).

The third line of the test case contains the binary string $$$s$$$ of length $$$n$$$.

The fourth line of the test case contains one integer $$$q$$$ ($$$1 \le q \le 10^5$$$) — the number of queries.

The subsequent $$$q$$$ lines of the test case describe the queries. The first number of each query, $$$tp \in \{1, 2\}$$$, characterizes the type of the query: if $$$tp = 1$$$, then $$$2$$$ integers $$$1 \le l \le r \le n$$$ follow, meaning that the operation of type $$$1$$$ should be performed with parameters $$$l, r$$$, and if $$$tp = 2$$$, then one integer $$$g \in \{0, 1\}$$$ follows, meaning that the operation of type $$$2$$$ should be performed with parameter $$$g$$$.

It is guaranteed that the sum of $$$n$$$ over all test cases does not exceed $$$10^5$$$, and also that the sum of $$$q$$$ over all test cases does not exceed $$$10^5$$$.

Output

For each test case, and for each query of type $$$2$$$ in it, output the answer to the corresponding query.

Example
Input
5
5
1 2 3 4 5
01000
7
2 0
2 1
1 2 4
2 0
2 1
1 1 3
2 1
6
12 12 14 14 5 5
001001
3
2 1
1 2 4
2 1
4
7 7 7 777
1111
3
2 0
1 2 3
2 0
2
1000000000 996179179
11
1
2 1
5
1 42 20 47 7
00011
5
1 3 4
1 1 1
1 3 4
1 2 4
2 0
Output
3 2 6 7 7 
11 7 
0 0 
16430827 
47
Note

Let's analyze the first test case:

  1. "2 $$$0$$$" — we are interested in the indices $$$i$$$ for which $$$s_i = \tt{0}$$$, since $$$s = \tt{01000}$$$, these are the indices $$$1, 3, 4$$$, and $$$5$$$, so the answer to the query will be $$$a_1 \oplus a_3 \oplus a_4 \oplus a_5 = 1 \oplus 3 \oplus 4 \oplus 5 = 3$$$.
  2. "2 $$$1$$$" — we are interested in the indices $$$i$$$ for which $$$s_i = \tt{1}$$$, since $$$s = \tt{01000}$$$, the only suitable index is $$$2$$$, so the answer to the query will be $$$a_2 = 2$$$.
  3. "1 $$$2$$$ $$$4$$$" — we need to replace the characters $$$s_2, s_3, s_4$$$ with their opposites, so before the query $$$s = \tt{01000}$$$, and after the query: $$$s = \tt{00110}$$$.
  4. "2 $$$0$$$" — we are interested in the indices $$$i$$$ for which $$$s_i = \tt{0}$$$, since $$$s = \tt{00110}$$$, these are the indices $$$1, 2$$$, and $$$5$$$, so the answer to the query will be $$$a_1 \oplus a_2 \oplus a_5 = 1 \oplus 2 \oplus 5 = 6$$$.
  5. "2 $$$1$$$" — we are interested in the indices $$$i$$$ for which $$$s_i = \tt{1}$$$, since $$$s = \tt{00110}$$$, these are the indices $$$3$$$ and $$$4$$$, so the answer to the query will be $$$a_3 \oplus a_4 = 3 \oplus 4 = 7$$$.
  6. "1 $$$1$$$ $$$3$$$" — $$$s = \tt{00110}$$$ $$$\to$$$ $$$s = \tt{11010}$$$.
  7. "2 $$$1$$$" — we are interested in the indices $$$i$$$ for which $$$s_i = \tt{1}$$$, since $$$s = \tt{11010}$$$, these are the indices $$$1, 2$$$, and $$$4$$$, so the answer to the query will be $$$a_1 \oplus a_2 \oplus a_4 = 1 \oplus 2 \oplus 4 = 7$$$.