I've created my own wrapper for GetComponent() so that I could use it wherever I wanted without perfomance concerns.

Why does not Unity do that in the first place? Could there be problems with such simple trick?

My code (intended to be used as a base class instead of MonoBehaviour):

public abstract class MonoBase : MonoBehaviour
{
    private readonly Dictionary<Type, Component> _components = new();

    public new T GetComponent<T>() where T : Component
    {
        var type = typeof(T);
        
        if (!_components.ContainsKey(type))
        {
            var component = gameObject.GetComponent<T>();
            
            if (component != null)
                _components.Add(type, component);
            else
                _components.Add(type, gameObject.AddComponent<T>());
        }
        
        return (T)_components[type];
    }
}

CodePudding user response：

• It adds to the memory overhead of your behaviour and also adds at least one heap allocation for each of its instances.
• It fails if components get removed and maybe re-added at runtime.
• Directly caching the component e.g. in Awake() into its own variable is still more performant.

Of course these points may be negligible in many of your use-cases, and you can surely use it where you see fit, but it is not a good idea in every situation, which is why it should not be put into the general implementation.