Strategy Design Pattern in Modern C++ vishal chovatiya
Reading Time: 4 minutes

In software engineering, Behavioural Design Patterns deal with the assignment of responsibilities between objects which in turn make the interaction between the objects easy & loosely coupled. In this article of the Behavioural Design Pattern series, we’re going to take a look at Strategy Design Pattern in Modern C++. It allows you to partially specify the behaviour of the class and then augment it later on. This pattern is also known as policy in many programming languages including especially in the C++ language.

By the way, If you haven’t check out my other articles on Behavioural Design Patterns, then here is the list:

  1. Chain of responsibility
  2. Command
  3. Interpreter
  4. Iterator
  5. Mediator
  6. Memento
  7. Observer
  8. State
  9. Strategy
  10. Template Method
  11. Visitor

The code snippets you see throughout this series of articles are simplified not sophisticated. So you often see me not using keywords like override, final, public(while inheritance) just to make code compact & consumable(most of the time) in single standard screen size. I also prefer struct instead of class just to save line by not writing “public:” sometimes and also miss virtual destructor, constructor, copy constructor, prefix std::, deleting dynamic memory, intentionally. I also consider myself a pragmatic person who wants to convey an idea in the simplest way possible rather than the standard way or using Jargons.

Note:

  • If you stumbled here directly, then I would suggest you go through What is design pattern? first, even if it is trivial. I believe it will encourage you to explore more on this topic.
  • All of this code you encounter in this series of articles are compiled using C++20(though I have used Modern C++ features up to C++17 in most cases). So if you don’t have access to the latest compiler you can use https://wandbox.org/ which has preinstalled boost library as well.

Intent

To choose particular algorithm from a family of algorithms depending upon need.

  • So many algorithms can actually decomposed into what I would call the higher & lower-level parts. For example, let’s consider the process of making tea. So the process of making tea can decomposed into something higher level like the process of making a hot beverage.
  • Because, whether you’re making tea or coffee or any other hot drink you have to boil the water. Then, you have to pour the water into the cup. So that is the high-level part can reuse.
  • And the specific things are to be specific but everything else can be reused for making something else like coffee or hot chocolate for example.
  • And this supports the beverage-specific strategies. So that is where the Strategy Design Pattern actually comes in. Strategy Design Pattern essentially enables the exact behaviour of a system to become selective either at compile time or run time.

Strategy Design Pattern Examples in C++

  • We’re going to implement a strategy for printing a list of items in different formats like Markdown & HTML. And I will show you how you can implement Dynamic & Static Strategy with two separate examples.

Dynamic Strategy

enum class Format { Markdown, Html };

struct ListStrategy {
    virtual ~ListStrategy() = default;
    virtual void add_list_item(ostringstream& oss, string& item) {};
    virtual void start(ostringstream& oss) {};
    virtual void end(ostringstream& oss) {};
};

struct MarkdownListStrategy: ListStrategy {
    void add_list_item(ostringstream& oss, string& item) override { oss << " - " << item << endl; }
};

struct HtmlListStrategy: ListStrategy {
    void start(ostringstream& oss) override { oss << "<ul>" << endl; }
    void end(ostringstream& oss) override { oss << "</ul>" << endl; }
    void add_list_item(ostringstream& oss, string& item) override { oss << "\t<li>" << item << "</li>" << endl; }
};

struct TextProcessor {
    void clear() {
        m_oss.str("");
        m_oss.clear();
    }

    void append_list(vector<string>& items) {
        m_list_strategy->start(m_oss);
        for (auto& item: items)
            m_list_strategy->add_list_item(m_oss, item);
        m_list_strategy->end(m_oss);
    }

    void set_output_format(Format& format) {
        switch (format) {
            case Format::Markdown: m_list_strategy = make_unique<MarkdownListStrategy>(); break;
            case Format::Html: m_list_strategy = make_unique<HtmlListStrategy>(); break;
        }
    }

    string str() { return m_oss.str(); }
private:
    ostringstream               m_oss;
    unique_ptr<ListStrategy>    m_list_strategy;
};

int main() {
    // markdown
    TextProcessor tp;
    tp.set_output_format(Format::Markdown);
    tp.append_list({ "foo", "bar", "baz" });
    cout << tp.str() << endl;

    // html
    tp.clear();
    tp.set_output_format(Format::Html);
    tp.append_list({ "foo", "bar", "baz" });
    cout << tp.str() << endl;

    return EXIT_SUCCESS;
}
/*  
 - foo
 - bar
 - baz

<ul>
	<li>foo</li>
	<li>bar</li>
	<li>baz</li>
</ul>
*/
  • Above example is self-explainable, but the thing here to take into account is a smart pointer to TextProcessor::m_list_strategy which we can use to change the strategy on the fly.

Static Strategy

  • I would not consider a static approach to the Strategy Design Pattern as flexible as the dynamic one but it’s still there if you need it.
template<typename LS>
struct TextProcessor {
    void append_list(const vector<string> &items) {
        m_list_strategy.start(m_oss);
        for (auto & item: items)
            m_list_strategy.add_list_item(m_oss, item);
        m_list_strategy.end(m_oss);
    }

    string str() const { return m_oss.str(); }
private:
    ostringstream       m_oss;
    LS                  m_list_strategy;
};

int main() {
    // markdown
    TextProcessor<MarkdownListStrategy> tp1;
    tp1.append_list({ "foo", "bar", "baz" });
    cout << tp1.str() << endl;

    // html
    TextProcessor<HtmlListStrategy> tp2;
    tp2.append_list({ "foo", "bar", "baz" });
    cout << tp2.str() << endl;

    return EXIT_SUCCESS;
}
  • When you looked at the Decorator Design Pattern we saw that the decorator can be implemented as both dynamic as well as static and it just so happens that the Strategy Design Pattern is exactly like this.
  • There is not anything special about static strategy except we are not referring algorithm(i.e. add_list_item()) through virtual table(rather passing it as a template parameter) which means you do not have the ability to change your algorithm/strategy at runtime.

Benefits of Strategy Design Pattern

  1. It’s easy to switch between different algorithms(strategies) in runtime as we’re using polymorphism in the interfaces.
  2. Clean & readable code because we avoid conditional code for algorithms(strategies).
  3. More clean code because you separate the concerns into classes (a class to each strategy) so automatically adhering Single Responsibility Principle.
  4. Preserving Open-Closed Principle also as I described in the example above, Strategy allows you to extend a logic in some parts of your code (“open for extension”) without rewriting those parts (“closed for modification”).

Summary by FAQs

When should the Strategy Design Pattern be used?

– When you need to use several algorithms with different variations.
– While most of your classes have related behaviours.
– When there are conditional statements around several related algorithms.

Why do we use the Strategy Design Pattern?

– For clean & readable code
– To adhere the Single Responsibility Principle & Open-Closed Principle.

What is the difference between the Strategy & State Design Pattern?

– Strategy is only an algorithm that you can change it in different circumstances upon your need.
– State can change whole object behaviour.

Do you like it☝️? Get such articles directly into the inbox…!📥