Ce diaporama a bien été signalé.
Nous utilisons votre profil LinkedIn et vos données d’activité pour vous proposer des publicités personnalisées et pertinentes. Vous pouvez changer vos préférences de publicités à tout moment.

SOLID Deconstruction

1 001 vues

Publié le

Presented at GOTO Nights (20th September 2016)

The SOLID principles are often presented as being core to good code design practice. Each of S, O, L, I and D do not, however, necessarily mean what programmers expect they mean or are taught. By understanding this range of beliefs we can learn more about practices for objects, components and interfaces than just S, O, L, I and D.

This talk reviews the SOLID principles and reveals contradictions and different interpretations. It is through paradoxes and surprises we often gain insights. We will leave SOLID somewhat more fluid, but having learnt from them more than expected.

Publié dans : Logiciels
  • Soyez le premier à commenter

SOLID Deconstruction

  1. 1. SOLIDDeconstruction @KevlinHenney
  2. 2. S O L I D
  3. 3. Single Responsibility Open-Closed Liskov Substitution Interface Segregation Dependency Inversion
  4. 4. Single Responsibility Open-Closed Liskov Substitution Interface Segregation Dependency Inversion
  5. 5. In object-oriented programming, the single responsibility principle states that every object should have a single responsibility, and that responsibility should be entirely encapsulated by the class. All its services should be narrowly aligned with that responsibility. http://en.wikipedia.org/wiki/Single_responsibility_principle
  6. 6. The term was introduced by Robert C. Martin [...]. Martin described it as being based on the principle of cohesion, as described by Tom DeMarco in his book Structured Analysis and Systems Specification. http://en.wikipedia.org/wiki/Single_responsibility_principle
  7. 7. We refer to a sound line of reasoning, for example, as coherent. The thoughts fit, they go together, they relate to each other. This is exactly the characteristic of a class that makes it coherent: the pieces all seem to be related, they seem to belong together, and it would feel somewhat unnatural to pull them apart. Such a class exhibits cohesion.
  8. 8. utility  the state of being useful, profitable or beneficial  useful, especially through having several functions  functional rather than attractive Concise Oxford English Dictionary
  9. 9. #include <stdlib.h>
  10. 10. Every class should embody only about 3–5 distinct responsibilities. Grady Booch, Object Solutions
  11. 11. One of the most foundational principles of good design is: Gather together those things that change for the same reason, and separate those things that change for different reasons. This principle is often known as the single responsibility principle, or SRP. In short, it says that a subsystem, module, class, or even a function, should not have more than one reason to change.
  12. 12. Single Responsibility Open-Closed Liskov Substitution Interface Segregation Dependency Inversion
  13. 13. Interface inheritance (subtyping) is used whenever one can imagine that client code should depend on less functionality than the full interface. Services are often partitioned into several unrelated interfaces when it is possible to partition the clients into different roles. For example, an administrative interface is often unrelated and distinct in the type system from the interface used by “normal” clients. "General Design Principles" CORBAservices
  14. 14. The dependency should be on the interface, the whole interface, and nothing but the interface.
  15. 15. We refer to a sound line of reasoning, for example, as coherent. The thoughts fit, they go together, they relate to each other. This is exactly the characteristic of a class that makes it coherent: the pieces all seem to be related, they seem to belong together, and it would feel somewhat unnatural to pull them apart. Such a class exhibits cohesion.
  16. 16. We refer to a sound line of reasoning, for example, as coherent. The thoughts fit, they go together, they relate to each other. This is exactly the characteristic of an interface that makes it coherent: the pieces all seem to be related, they seem to belong together, and it would feel somewhat unnatural to pull them apart. Such an interface exhibits cohesion.
  17. 17. public interface LineIO { String read(); void write(String toWrite); }
  18. 18. public interface LineReader { String read(); } public interface LineWriter { void write(String toWrite); }
  19. 19. Single Responsibility Open-Closed Liskov Substitution Interface Segregation Dependency Inversion
  20. 20. In a purist view of object-oriented methodology, dynamic dispatch is the only mechanism for taking advantage of attributes that have been forgotten by subsumption. This position is often taken on abstraction grounds: no knowledge should be obtainable about objects except by invoking their methods. In the purist approach, subsumption provides a simple and effective mechanism for hiding private attributes.
  21. 21. A type hierarchy is composed of subtypes and supertypes. The intuitive idea of a subtype is one whose objects provide all the behavior of objects of another type (the supertype) plus something extra. What is wanted here is something like the following substitution property: If for each object o1 of type S there is an object o2 of type T such that for all programs P defined in terms of T, the behavior of P is unchanged when o1 is substituted for o2, then S is a subtype of T. Barbara Liskov "Data Abstraction and Hierarchy"
  22. 22. generalisation specialisation commonality variation
  23. 23. E.g., Python's dict class, which is a hashed mapping container E.g., Python's OrderedDict class, which preserves order of insertion
  24. 24. public class RecentlyUsedList { ... public int Count { get ... } public string this[int index] { get ... } public void Add(string newItem) ... ... }
  25. 25. public class RecentlyUsedList { private IList<string> items = new List<string>(); public int Count { get { return items.Count; } } public string this[int index] { get { return items[index]; } } public void Add(string newItem) { if(newItem == null) throw new ArgumentNullException(); items.Remove(newItem); items.Insert(0, newItem); } ... }
  26. 26. public class RecentlyUsedList : List<string> { public override void Add(string newItem) { if(newItem == null) throw new ArgumentNullException(); items.Remove(newItem); items.Insert(0, newItem); } ... }
  27. 27. namespace List_spec { ... [TestFixture] public class Addition { private List<string> list; [Setup] public void List_is_initially_empty() { list = ... } ... [Test] public void Addition_of_non_null_item_is_appended() ... [Test] public void Addition_of_null_is_permitted() ... [Test] public void Addition_of_duplicate_item_is_appended() ... ... } ... }
  28. 28. namespace List_spec { ... [TestFixture] public class Addition { private List<string> list; [Setup] public void List_is_initially_empty() { list = new List<string>(); } ... [Test] public void Addition_of_non_null_item_is_appended() ... [Test] public void Addition_of_null_is_permitted() ... [Test] public void Addition_of_duplicate_item_is_appended() ... ... } ... }
  29. 29. namespace List_spec { ... [TestFixture] public class Addition { private List<string> list; [Setup] public void List_is_initially_empty() { list = new RecentlyUsedList(); } ... [Test] public void Addition_of_non_null_item_is_appended() ... [Test] public void Addition_of_null_is_permitted() ... [Test] public void Addition_of_duplicate_item_is_appended() ... ... } ... }
  30. 30. A type hierarchy is composed of subtypes and supertypes. The intuitive idea of a subtype is one whose objects provide all the behavior of objects of another type (the supertype) plus something extra. What is wanted here is something like the following substitution property: If for each object o1 of type S there is an object o2 of type T such that for all programs P defined in terms of T, the behavior of P is unchanged when o1 is substituted for o2, then S is a subtype of T. Barbara Liskov "Data Abstraction and Hierarchy"
  31. 31. A type hierarchy is composed of subtypes and supertypes. The intuitive idea of a subtype is one whose objects provide all the behavior of objects of another type (the supertype) plus something extra. What is wanted here is something like the following substitution property: If for each object o1 of type S there is an object o2 of type T such that for all programs P defined in terms of T, the behavior of P is unchanged when o1 is substituted for o2, then S is a subtype of T. Barbara Liskov "Data Abstraction and Hierarchy"
  32. 32. A type hierarchy is composed of subtypes and supertypes. The intuitive idea of a subtype is one whose objects provide all the behavior of objects of another type (the supertype) plus something extra. What is wanted here is something like the following substitution property: If for each object o1 of type S there is an object o2 of type T such that for all programs P defined in terms of T, the behavior of P is unchanged when o1 is substituted for o2, then S is a subtype of T. Barbara Liskov "Data Abstraction and Hierarchy"
  33. 33. A type hierarchy is composed of subtypes and supertypes. The intuitive idea of a subtype is one whose objects provide all the behavior of objects of another type (the supertype) plus something extra. What is wanted here is something like the following substitution property: If for each object o1 of type S there is an object o2 of type T such that for all programs P defined in terms of T, the behavior of P is unchanged when o1 is substituted for o2, then S is a subtype of T. Barbara Liskov "Data Abstraction and Hierarchy"
  34. 34. Single Responsibility Open-Closed Liskov Substitution Interface Segregation Dependency Inversion
  35. 35. Bertrand Meyer gave us guidance as long ago as 1988 when he coined the now famous open-closed principle. To paraphrase him: Software entites (classes, modules, functions, etc.) should be open for extension, but closed for modification. Robert C Martin "The Open-Closed Principle" C++ Report, January 1996
  36. 36. Open for extension Closed for extension Closed for modification Open for modification Extend existing code without changing it, e.g., using a framework Extend or change existing code, e.g., open-source and unpublished code Non-extensible and unchangeable code, e.g., composable code or legacy code Non-extensible code that can be maintained, e.g., composable code in a maintained library
  37. 37. The principle stated that a good module structure should be both open and closed:  Closed, because clients need the module's services to proceed with their own development, and once they have settled on a version of the module should not be affected by the introduction of new services they do not need.  Open, because there is no guarantee that we will include right from the start every service potentially useful to some client.
  38. 38. [...] A good module structure should be [...] closed [...] because clients need the module's services to proceed with their own development, and once they have settled on a version of the module should not be affected by the introduction of new services they do not need.
  39. 39. Published Interface is a term I used (first in Refactoring) to refer to a class interface that's used outside the code base that it's defined in. Martin Fowler http://martinfowler.com/bliki/PublishedInterface.html
  40. 40. There is no problem changing a method name if you have access to all the code that calls that method. Even if the method is public, as long as you can reach and change all the callers, you can rename the method.
  41. 41. There is a problem only if the interface is being used by code that you cannot find and change. When this happens, I say that the interface becomes a published interface (a step beyond a public interface).
  42. 42. Published Interface is a term I used (first in Refactoring) to refer to a class interface that's used outside the code base that it's defined in. The distinction between published and public is actually more important than that between public and private. The reason is that with a non-published interface you can change it and update the calling code since it is all within a single code base. [...] But anything published so you can't reach the calling code needs more complicated treatment. Martin Fowler http://martinfowler.com/bliki/PublishedInterface.html
  43. 43. [...] A good module structure should be [...] open [...] because there is no guarantee that we will include right from the start every service potentially useful to some client.
  44. 44. Speculative Generality Brian Foote suggested this name for a smell to which we are very sensitive. You get it when people say, "Oh, I think we need the ability to do this kind of thing someday" and thus want all sorts of hooks and special cases to handle things that aren't required.
  45. 45. extends
  46. 46. A myth in the object-oriented design community goes something like this: If you use object-oriented technology, you can take any class someone else wrote, and, by using it as a base class, refine it to do a similar task. Robert B Murray C++ Strategies and Tactics
  47. 47. Design and implement for inheritance or else prohibit it By now, it should be apparent that designing a class for inheritance places substantial limitations on the class.
  48. 48. http://blog.8thlight.com/uncle-bob/2014/05/12/TheOpenClosedPrinciple.html
  49. 49. http://blog.8thlight.com/uncle-bob/2014/05/12/TheOpenClosedPrinciple.html I've heard it said that the OCP is wrong, unworkable, impractical, and not for real programmers with real work to do. The rise of plugin architectures makes it plain that these views are utter nonsense. On the contrary, a strong plugin architecture is likely to be the most important aspect of future software systems.
  50. 50. OCP ≡ Plug-ins?
  51. 51. OCP ≡ Plug-ins/
  52. 52. public abstract class Shape { ... } public class Square : Shape { ... public void DrawSquare() ... } public class Circle : Shape { ... public void DrawCircle() ... }
  53. 53. public abstract class Shape ... public class Square : Shape ... public class Circle : Shape ... static void DrawAllShapes(Shape[] list) { foreach (Shape s in list) if (s is Square) (s as Square).DrawSquare(); else if (s is Circle) (s as Circle).DrawCircle(); }
  54. 54. public abstract class Shape ... public class Square : Shape ... public class Circle : Shape ... static void DrawAllShapes(Shape[] list) { foreach (Shape s in list) if (s is Square) (s as Square).DrawSquare(); else if (s is Circle) (s as Circle).DrawCircle(); }
  55. 55. public abstract class Shape { ... public abstract void Draw(); } public class Square : Shape { ... public override void Draw() ... } public class Circle : Shape { ... public override void Draw() ... }
  56. 56. public abstract class Shape ... public class Square : Shape ... public class Circle : Shape ... static void DrawAllShapes(Shape[] list) { foreach (Shape s in list) s.Draw(); }
  57. 57. public abstract class Shape ... public class Square : Shape ... public class Circle : Shape ... static void DrawAllShapes(Shape[] list) { foreach (Shape s in list) s.Draw(); }
  58. 58. public abstract class Shape ... public class Square : Shape ... public class Circle : Shape ... static void DrawAllShapes(Shape[] list) { foreach (Shape s in list) s.Draw(); }
  59. 59. Bertrand Meyer gave us guidance as long ago as 1988 when he coined the now famous open-closed principle. To paraphrase him: Software entites (classes, modules, functions, etc.) should be open for extension, but closed for modification. Robert C Martin "The Open-Closed Principle" C++ Report, January 1996
  60. 60. Bertrand Meyer gave us guidance as long ago as 1988 when he coined the now famous open-closed principle. To paraphrase him: Software entites (classes, modules, functions, etc.) should be open for extension, but closed for modification. Robert C Martin "The Open-Closed Principle" C++ Report, January 1996
  61. 61. This double requirement looks like a dilemma, and classical module structures offer no clue. This double requirement looks like a dilemma, and classical module structures offer no clue. But inheritance solves it. This double requirement looks like a dilemma, and classical module structures offer no clue. But inheritance solves it. A class is closed, since it may be compiled, stored in a library, baselined, and used by client classes. But it is also open, since any new class may use it as parent, adding new features.
  62. 62. public abstract class Shape ... public class Square : Shape ... public class Circle : Shape ... static void DrawAllShapes(Shape[] list) { foreach (Shape s in list) if (s is Square) (s as Square).DrawSquare(); else if (s is Circle) (s as Circle).DrawCircle(); }
  63. 63. public abstract class Shape ... public class Square : Shape ... public class Circle : Shape ... static void DrawAllShapes(Shape[] list) { foreach (Shape s in list) if (s is Square) (s as Square).DrawSquare(); else if (s is Circle) (s as Circle).DrawCircle(); }
  64. 64. public abstract class Shape ... public class Square : Shape ... public class Circle : Shape ... static void DrawAllShapes(Shape[] list) { foreach (Shape s in list) s.Draw(); }
  65. 65. public abstract class Shape ... public class Square : Shape ... public class Circle : Shape ... static void DrawAllShapes(Shape[] list) { foreach (Shape s in list) s.Draw(); }
  66. 66. public sealed class Shape { public enum Type { Square, Circle } ... public void Draw() ... } static void DrawAllShapes(Shape[] list) { foreach (Shape s in list) s.Draw(); }
  67. 67. public sealed class Shape { public enum Type { Square, Circle } ... public void Draw() ... } static void DrawAllShapes(Shape[] list) { foreach (Shape s in list) s.Draw(); }
  68. 68. Don't publish interfaces prematurely. Modify your code ownership policies to smooth refactoring.
  69. 69. Single Responsibility Open-Closed Liskov Substitution Interface Segregation Dependency Inversion
  70. 70. In object-oriented programming, the dependency inversion principle refers to a specific form of decoupling where conventional dependency relationships established from high- level, policy-setting modules to low-level, dependency modules are inverted (i.e. reversed) for the purpose of rendering high-level modules independent of the low-level module implementation details. http://en.wikipedia.org/wiki/Dependency_inversion_principle
  71. 71. The principle states: A. High-level modules should not depend on low-level modules. Both should depend on abstractions. B. Abstractions should not depend upon details. Details should depend upon abstractions. http://en.wikipedia.org/wiki/Dependency_inversion_principle
  72. 72. inversion, noun  the action of inverting or the state of being inverted  reversal of the normal order of words, normally for rhetorical effect  an inverted interval, chord, or phrase  a reversal of the normal decrease of air temperature with altitude, or of water temperature with depth Concise Oxford English Dictionary
  73. 73. Parameterize from Above Hardwire from Below
  74. 74. Stewart Brand, How Buildings Learn See also http://www.laputan.org/mud/
  75. 75. package com.sun...;
  76. 76.        Scenario buffering by dot-voting possible changes and invert dependencies as needed
  77. 77. One of the most foundational principles of good design is: Gather together those things that change for the same reason, and separate those things that change for different reasons. This principle is often known as the single responsibility principle, or SRP. In short, it says that a subsystem, module, class, or even a function, should not have more than one reason to change.
  78. 78. At some level the style becomes the substance.

×