C++实现游戏角色按时长依次长按W/A/S/D按键功能求助

刚接触C++，想实现游戏角色在任意游戏中按指定时长依次长按按键：先长按W键2秒，接着长按A键2秒，再长按S键2秒，最后长按D键2秒。但自己写的代码有卡顿问题，而且对多数代码内容不理解，附上代码求助：

int sleepBetween = 50;

if (GetAsyncKeyState(VK_NUMPAD1))
{
    std::cout << "Working..." << endl;

    SHORT key;
    UINT mappedkey;

    // 循环
    while (true)
    {
        INPUT input = {};
        key = VkKeyScan('w');

        mappedkey = MapVirtualKey(LOBYTE(key), 0);
        input.type = INPUT_KEYBOARD;
        input.ki.dwFlags = KEYEVENTF_SCANCODE;
        input.ki.wScan = mappedkey;

        SendInput(1, &input, sizeof(INPUT));
        Sleep(10);

        input.ki.dwFlags = KEYEVENTF_SCANCODE | KEYEVENTF_KEYUP;
        SendInput(1, &input, sizeof(INPUT));

        Sleep(sleepBetween);

        key = VkKeyScan('a');

        mappedkey = MapVirtualKey(LOBYTE(key), 0);
        input.type = INPUT_KEYBOARD;
        input.ki.dwFlags = KEYEVENTF_SCANCODE;
        input.ki.wScan = mappedkey;

        SendInput(1, &input, sizeof(INPUT));
        Sleep(10);

        input.ki.dwFlags = KEYEVENTF_SCANCODE | KEYEVENTF_KEYUP;
        SendInput(1, &input, sizeof(INPUT));

        Sleep(sleepBetween);

        key = VkKeyScan('s');

        mappedkey = MapVirtualKey(LOBYTE(key), 0);
        input.type = INPUT_KEYBOARD;
        input.ki.dwFlags = KEYEVENTF_SCANCODE;
        input.ki.wScan = mappedkey;

        SendInput(1, &input, sizeof(INPUT));
        Sleep(10);

        input.ki.dwFlags = KEYEVENTF_SCANCODE | KEYEVENTF_KEYUP;
        SendInput(1, &input, sizeof(INPUT));

        Sleep(sleepBetween);

        Sleep(sleepBetween);

        key = VkKeyScan('d');

        mappedkey = MapVirtualKey(LOBYTE(key), 0);
        input.type = INPUT_KEYBOARD;
        input.ki.dwFlags = KEYEVENTF_SCANCODE;
        input.ki.wScan = mappedkey;

        SendInput(1, &input, sizeof(INPUT));
        Sleep(10);

        input.ki.dwFlags = KEYEVENTF_SCANCODE | KEYEVENTF_KEYUP;
        SendInput(1, &input, sizeof(INPUT));

        Sleep(sleepBetween);

        // 按小键盘2停止
        if (GetAsyncKeyState(VK_NUMPAD2))
        {
            std::cout << "Stopped!" << endl;
            break;
        }
    }
}

1. 原代码核心问题 #

• 长按逻辑错误：原代码按下按键后仅Sleep(10ms)就松开，实际是点按而非长按，完全不符合2秒长按的需求
• 代码冗余：重复编写4次几乎一致的按键操作，维护难度高
• 卡顿原因：主循环中的Sleep会阻塞线程，且未处理停止信号的实时检测；同时原代码的时间间隔设置和需求完全不符
• 按键检测不严谨：GetAsyncKeyState未判断最高位（0x8000），可能导致误触发

2. 改进后的代码实现 #

#include <windows.h>
#include <iostream>

// 通用长按按键函数：传入扫描码和长按时长（毫秒）
void HoldKey(UINT scancode, DWORD duration_ms) {
    INPUT input = {0};
    input.type = INPUT_KEYBOARD;
    input.ki.dwFlags = KEYEVENTF_SCANCODE;
    input.ki.wScan = scancode;

    // 发送按下按键指令
    SendInput(1, &input, sizeof(INPUT));

    // 长按过程中实时检测停止键，避免完全阻塞
    DWORD start_time = GetTickCount();
    while (GetTickCount() - start_time < duration_ms) {
        if (GetAsyncKeyState(VK_NUMPAD2) & 0x8000) {
            // 先松开当前按键再退出
            input.ki.dwFlags = KEYEVENTF_SCANCODE | KEYEVENTF_KEYUP;
            SendInput(1, &input, sizeof(INPUT));
            std::cout << "Stopped during hold!" << std::endl;
            exit(0);
        }
        Sleep(10); // 降低CPU占用
    }

    // 发送松开按键指令
    input.ki.dwFlags = KEYEVENTF_SCANCODE | KEYEVENTF_KEYUP;
    SendInput(1, &input, sizeof(INPUT));
}

int main() {
    std::cout << "按小键盘1开始，小键盘2停止" << std::endl;

    while (true) {
        // 检测开始键：判断最高位确认按键当前按下
        if (GetAsyncKeyState(VK_NUMPAD1) & 0x8000) {
            std::cout << "Working..." << std::endl;

            // 依次长按W/A/S/D各2秒（扫描码为硬件固定值，游戏兼容性更好）
            HoldKey(0x11, 2000); // W键
            HoldKey(0x1E, 2000); // A键
            HoldKey(0x1F, 2000); // S键
            HoldKey(0x20, 2000); // D键

            std::cout << "一轮完成，等待再次触发..." << std::endl;
            // 等待松开开始键，防止重复触发
            while (GetAsyncKeyState(VK_NUMPAD1) & 0x8000) {
                Sleep(10);
            }
        }

        // 全局检测停止键
        if (GetAsyncKeyState(VK_NUMPAD2) & 0x8000) {
            std::cout << "Stopped!" << std::endl;
            break;
        }

        Sleep(10); // 降低CPU占用
    }

    return 0;
}

3. 关键代码解释 #

• 扫描码（scancode）：游戏通常优先响应硬件扫描码，比虚拟键兼容性更好，W/A/S/D的扫描码为固定值（0x11/0x1E/0x1F/0x20）
• HoldKey函数：封装了按键按下、保持、松开的完整逻辑，同时在长按过程中实时检测停止信号，避免程序完全阻塞
• GetAsyncKeyState正确用法：通过按位与0x8000判断按键当前是否处于按下状态，避免误触发
• 卡顿优化：用小间隔Sleep+循环检测的方式，既保证长按时长，又能实时响应停止指令，同时降低CPU占用

内容的提问来源于stack exchange，提问作者Martin