原型模式：從手寫簡歷到克隆對象的藝術

場景：大量相似對象的創建困境

小菜準備面試，打印了多份一模一樣的簡歷。大鳥看代碼發現：main() 中三次 new Resume 並逐字段 SetPersonalInfo、SetWorkExperience。若需 20 份，不僅代碼冗長，一處錯誤還需改 20 次。

初版代碼：重複初始化

class Resume {
public:
    Resume(char* _name) : name(_name), sex(0), age(0), timeArea(0), company(0) {}
    void SetPersonalInfo(char* _sex, char* _age) { sex = _sex; age = _age; }
    void SetWorkExperience(char* _timeArea, char* _company) { timeArea = _timeArea; company = _company; }
    void Display() { printf("%s %s %s 工作經歷:%s %s", name, sex, age, timeArea, company); }
private:
    char *name, *sex, *age, *timeArea, *company;
};

int main() {
    Resume a("大鳥"); a.SetPersonalInfo("男","29"); a.SetWorkExperience("1998-2000","XX公司");
    Resume b("大鳥"); b.SetPersonalInfo("男","29"); b.SetWorkExperience("1998-2000","XX公司");
    Resume c("大鳥"); c.SetPersonalInfo("男","29"); c.SetWorkExperience("1998-2000","XX公司");
    a.Display(); b.Display(); c.Display();
}

原型模式核心思想

用一個原型實例指定創建對象的種類，並通過複製這個原型來創建新對象。避免了繁瑣的初始化，支援動態增減產品類，無需固定繼承體系。

缺點：每個類需實現克隆方法。

UML 類圖

克隆介面與實現

class ICloneable {
public:
    virtual ICloneable* Clone() = 0;
};

class WorkExperience : public ICloneable {
public:
    char*& WorkDate() { return workDate; }
    char*& Company() { return company; }
    WorkExperience* Clone() override {
        auto* result = new WorkExperience();
        result->WorkDate() = workDate;
        result->Company() = company;
        return result;
    }
private:
    char *workDate, *company;
};

class Resume : public ICloneable {
public:
    Resume(char* _name) : name(_name), work(new WorkExperience()) {}
    ~Resume() { delete work; }
    void SetPersonalInfo(char* _sex, char* _age) { sex = _sex; age = _age; }
    void SetWorkExperience(char* _workDate, char* _company) {
        work->WorkDate() = _workDate;
        work->Company() = _company;
    }
    void Display() { printf("%s %s %s 工作經歷:%s %s", name, sex, age, work->WorkDate(), work->Company()); }
    Resume* Clone() override {
        auto* result = new Resume(name); // 注意：此處原碼有誤，應傳 name 而非 work
        result->name = name;
        result->sex = sex;
        result->age = age;
        result->work = work->Clone(); // 深拷貝工作經歷
        return result;
    }
private:
    char *name, *sex, *age;
    WorkExperience* work;
};

客戶端調用：一行代碼複製多份

int main() {
    Resume* proto = new Resume("大鳥");
    proto->SetPersonalInfo("男", "29");
    proto->SetWorkExperience("1998-2000", "XX公司");

    Resume* a = proto->Clone();
    Resume* b = proto->Clone();
    Resume* c = proto->Clone();

    a->Display(); b->Display(); c->Display();
    // 清理內存...
}

關鍵點總結

• 深拷貝 vs 淺拷貝：WorkExperience 指標必須深拷貝，否則多份簡歷共享同一工作經歷對象，修改其一會影響其它。
• 原型管理器：實際專案常引入 PrototypeManager 註冊/查找原型，進一步解耦客戶端與具體類。
• 適用場景：對象初始化昂貴、狀態組合較少、需動態增減產品類時。

從「手寫 N 份簡歷」到「克隆 1 份原型」，原型模式讓對象創建變得優雅高效。