A question for OOP experts. - page 51
You are missing trading opportunities:
- Free trading apps
- Over 8,000 signals for copying
- Economic news for exploring financial markets
Registration
Log in
You agree to website policy and terms of use
If you do not have an account, please register
"New OOP concept" - not clear on the purpose. What are you doing it for?
I want to understand a few things I've been interested in all my life.
1. Is it possible to create a self-developing system?
2. Is it possible to create a process in which multiple systems interact, change and evolve?
3) Can anything emerge without an initial concept in place?
4. What happens if I combine my representation of objects in the kernel and the standard OOP, with its inheritance and encapsulation of objects? Would it be possible to simplify the creation of complex systems?
...and operate with instances in the bucket:)
Why write something in a bucket, especially something related to a specific object? The object itself stores information about itself, and the bucket only contains pointers to objects.
Have you ever wondered why standard OOP has such big "capsules" (classes) of Objects? After all, it's easier to turn an Object into a node of pointers, and store the contents of Objects outside the classes. Then, the Object is easy to model. Just change the pointers to the material, which it shall combine, and the Object will "turn" into something else. The content of the Object will depend on the pointers, not on the "ingredients" of its Class (capsule). So, I store exactly such bundles, in the Kernel as Objects. And the material itself is outside the Core. Thus, connections between Objects, as well as their contents, can be easily and quickly changed, and systems consisting of such "nodal" Objects can be easily modified.
.
Well cool, does the approach have a name? Maybe JELOP - Nuclear Object Oriented Programming. Sounds solid, associated with weapons, nuclear power plants. A slogan of some sort, "I've taken the best and put it in the kernel!" Well done, Peter, keep up the good work, I'm pleasantly surprised.
Look at the brain from the inside. It is filled with connection centres (neurons), not individual closed complexes. The brain encapsulates connections, not the entire content of each object.
Such links are called "composition/aggregation of objects", but you keep cycling to your core. Add any links you like, there's nothing new about it.
ZS: I watched the video diagonally, maybe not so much.Such links are called "composition/aggregation of objects", but you keep cycling to your core. Add any links you want, it's nothing new.
ZS: I watched the video diagonally, maybe not so much.https://habr.com/ru/post/354046/
This is much clearer.
//-----------------------------------------------
Composition links Objects that are in a common associative chain. Room-Walls-Furniture. The problem is that associative chains are too unstable and undefined, so class-level connections are a mistake. A room can be associated with a child, a cage, a hyper-cube or an underworld, and depending on that, put content in it. A standard OOP class does not provide for easy change of relationships within a composition, while the kernel does.
Such links are called "composition/aggregation of objects", but you keep cycling to your core. Add any links you want, there's nothing new about it.
ZS: I watched the video diagonally, maybe not so much.send him 5-10 roubles to get a haircut ;)
Why does the kernel provide for easy change of compositions (object content)? - Because in the kernel, entities are memory locations with variables. And a class isa description of an object at the editor level. At the code level.
The code changes through the editor. Memory is changed through the interface or by the program itself. That is, if the class (object capsule) is put from the code into an array (kernel), its processing and changing will be 100 times faster and more flexible.