A. Red Versus Blue

time limit per test

1 second

memory limit per test

256 megabytes

Team Red and Team Blue competed in a competitive FPS. Their match was streamed around the world. They played a series of n matches.

In the end, it turned out Team Red won r times and Team Blue won b times. Team Blue was less skilled than Team Red, so b was strictly less than r.

You missed the stream since you overslept, but you think that the match must have been neck and neck since so many people watched it. So you imagine a string of length n where the i-th character denotes who won the i-th match --- it is R if Team Red won or B if Team Blue won. You imagine the string was such that the maximum number of times a team won in a row was as small as possible. For example, in the series of matches RBBRRRB, Team Red won 3 times in a row, which is the maximum.

You must find a string satisfying the above conditions. If there are multiple answers, print any.

Input

The first line contains a single integer t (1≤t≤1000) --- the number of test cases.

Each test case has a single line containing three integers n, r, and b (3≤n≤100; 1≤b<r≤n, r+b=n).

Output

For each test case, output a single line containing a string satisfying the given conditions. If there are multiple answers, print any.

Examples

Input

Copy

复制代码

3

7 4 3

6 5 1

19 13 6

Output

Copy

复制代码
RBRBRBR
RRRBRR
RRBRRBRRBRRBRRBRRBR

Input

Copy

复制代码

6

3 2 1

10 6 4

11 6 5

10 9 1

10 8 2

11 9 2

Output

Copy

复制代码
RBR
RRBRBRBRBR
RBRBRBRBRBR
RRRRRBRRRR
RRRBRRRBRR
RRRBRRRBRRR

Note

The first test case of the first example gives the optimal answer for the example in the statement. The maximum number of times a team wins in a row in RBRBRBR is 1. We cannot minimize it any further.

The answer for the second test case of the second example is RRBRBRBRBR. The maximum number of times a team wins in a row is 2, given by RR at the beginning. We cannot minimize the answer any further.

解题说明:此题是一道模拟题,为了确保连续R出现的次数最少,需要尽可能将R和B平均分配到队列中。首先统计出R和B的比例情况,然后按照比例输出R和B。

cpp 复制代码
#include<stdio.h>
int main()
{
    int t, i, n, r, b;
    scanf("%d", &t);
    for (i = 0; i < t; i++) 
    {
        scanf("%d %d %d", &n, &r, &b);
        char win[101];
        int k = 0;
        while (r > 0) 
        {
            int z = r / (b + 1);
            for (int j = 0; j < z; j++)
            {
                win[k] = 'R';
                k++;
            }
            if (b > 0) 
            {
                win[k] = 'B';
                k++;
            }
            r = r - z;
            b = b - 1;
        }
        win[n] = '\0';
        printf("%s\n", win);
    }
    return 0;
}
相关推荐
UIU1142 小时前
递归算法与汉诺塔
c++·学习·算法·c#·递归
用户979984807187 小时前
200 行代码写一个能跑的 AI Agent:不依赖任何框架,只靠 tool-calling 原理
算法
threerocks7 小时前
【Muse实战】X 流量作战室搭建保姆级教程 - 拥有你自己的运营团队
算法
GreenTea7 小时前
深度拆解 ScienceBuddy:如何用“双层递归自进化”构建高可靠科研 Agent Harness
前端·后端·算法
花椒技术7 小时前
2.46 秒生成 5 秒视频:拆解 H3 Max 的模型、推理栈与硬件协同
算法·音视频开发·视频编码
vivo互联网技术7 小时前
ART:妆容迁移框架,重新定义高保真妆容迁移 | ECCV 2026
人工智能·算法·图像识别
用户0441440924497 小时前
卫星信号模拟器里的四个坐标转换公式
算法
橘和柠8 小时前
阿里 open-code-review (AI代码审查工具)实战:安装、四层规则链、自定义规则格式与实测避坑
算法·面试
得物技术8 小时前
别再只卷向量检索了,得物交易搜索如何用“生成式”实现召回范式跃迁?
人工智能·算法·llm
罗西的思考8 小时前
[Agent Memory / 强化学习] MemPO源码学习笔记 ---(4)--- Rollout实现细节
人工智能·算法