For example, one of the methods is:public abstract class ViewModelImpl<T> : ViewModel where T : IModel { protected void selectData(string query) { // Логика получения reader IModel item = (T)Activator.CreateInstance(typeof(T), reader); this.Data.Add(item); } } Well, inherit specific classes from ViewModelImpl<T> with T♪(If the designer T without parameters, you could use a reflexi. )For better readability, the method may be: private T CreateModelItem(SqlDataReader reader) { return (T)Activator.CreateInstance(typeof(T), reader); } protected void selectData(string query) { // Логика получения reader IModel item = CreateModelItem(reader); this.Data.Add(item); } Well, you already have this auxiliary method, it's called createObject♪

Decision working code:public class MenuIndexException extends ArrayIndexOutOfBoundsException { private static final long serialVersionUID = -505311792750028191L; public MenuIndexException(String massage) { super(massage); } public void answer() { } public static void main(String[] args) { final MenuIndexException ex = new MenuIndexException("s"); ex.answer(); } }

This may be done by means of an index or reference to a reference to a base class initiated by a production class.For example,Circle c( 10 ); Shape *ps = &c; ps->R(); ps->S(); Shape &rs = c; rs.R(); rs.S(); Keep in mind that the primary-class disruptor should also be declared virtual. For example,virtual ~Shape() = default; orvirtual ~Shape(){} Pay attention to the discrepancy.double Circle::P() { return 2 * 3.12 * r; ^^^^^ } double Circle::S() { return 3.14 * r * r; ^^^^ } In order to prevent this kind of error, it is better for the constants to make names. You could identify the relevant member in your class (subject to the compiler supporting the design), for example,class Circle : public Shpes { private: double r; void SetCheck(double R); constexpr static double PI = 3.14; public: Circle(double R = 0); virtual double P(); virtual double S(); void Print(); }; And write down the definitions of functions as follows:double Circle::P() { return 2 * PI * r; ^^^^^ } double Circle::S() { return PI * r * r; ^^^^ } If the compiler does not support such an announcement as a member of the class, you can just writeclass Circle : public Shpes { private: double r; void SetCheck(double R); const static double PI; public: Circle(double R = 0); virtual double P(); virtual double S(); void Print(); }; And then define this statistical variable. double Circle::PI = 3.14;

Let's start dealing with your problem from the end.The same point shall not necessarily be number. It could be a date, time or even IP. There are a lot of combinations and complexity in configuration. Anything ready?Comparison operations exist in all programming languages. It always works for the numbers, but with other types of data, there may be problems.If you're lucky and your programming language supports the redefinition of comparison operations, you can introduce your type of data (e.g.,) IPAddressand implement correct operator behaviour for him >♪ <♪ == (all other comparison operators - derivatives from these three).If you are unlucky and your programming language does not support the redefinition of comparison operators, a special interface will be available to assist with the necessary methods in a clear form:interface IPPointInterface { public function isEqualTo(IPPointInterface $ip): bool; public function isGraterThan(IPPointInterface $ip): bool; public function isLessThan(IPPointInterface $ip): bool; } By analogy, an interface can be introduced for any type of data.It is clear that the interval is defined by two points and many points on the number of direct points between them. If we have a time interval, it makes sense that we can check if there's a point at a given interval.You've rightly caught the main thrust of the interval - he can check the point in it or not. That's what you're supposed to do in his interface. In addition, the customer code should be able to obtain the starting and final points:interface IPIntervalInterface { public function containsPoint(IPPointInterface $ip): bool; public function getStartPoint(IPPointInterface $ip): IPPoint; public function getEndPoint(IPPointInterface $ip): IPPoint; } From personal experience, I would add that in 99% of cases there are several other methods that should maintain interval interfaces, such as checking intervals.And there's a problem, because intervals are four types-- [a, b]♪ (a, b)♪ [a, b)♪ (a, b]♪This is the details of implementation. You can either implement four classes of intervals or shirk and implement the base class for cut (closed intervals) [a, b]class IPClosedInterval implements IPIntervalInterface { private $start; private $end; public function __construct(IPPointInterface $start, IPPointInterface $end) { $this-&gt;start = $start; $this-&gt;end = $end; } public function getStartPoint() { return $this-&gt;start; } public function getEndPoint() { return $this-&gt;end; } public function containsPoint(IPPointInterface $ip) { return $ip-&gt;isLessThan($this-&gt;getEndPoint()) &amp;&amp; $ip-&gt;isGreaterThan($this-&gt;getStartEndPoint()); } } And two border decorators.class IPIntervalLeftOpenDecorator implements IPIntervalInterface { private $interval; public function __construct(IPIntervalInterface $interval) { $this-&gt;interval = $interval; } public function containsPoint(IPPointInterface $ip) { return $this-&gt;interval-&gt;containsPoint($ip) || $ip-&gt;isEqualTo($this-&gt;getStartPoint()); } public function getStartPoint() { return $this-&gt;interval-&gt;getStartPoint(); } public function getEndPoint() { return $this-&gt;interval-&gt;getEndPoint(); } } If desired, the designer and the code returning the boundaries can be removed to the abstract decorator.Here's an example of how this can be used:$a = new IPPoint('...'); $b = new IPPoint('...'); $closed_int = new IPIntervalClosed($a, $b); // [a, b] $left_opened_int = new IPIntervalLeftOpenDecorator($closed_int); // (a, b] $right_opened_int = new IPIntervalRightOpenDecorator($closed_int); // [a, b) $opened_int = new IPIntervalLeftOpenDecorator($right_opened_int); // (a, b) Programming has been in existence for decades, and this task has to be repeated and, accordingly, there must be a general solution that I do not seem to know.It depends very much on your programming language. For example, in C++, you can define the reference type for the point and parameterize the interval class with that type.With regard to PHP, you can also identify a base type for a point. IPPointInterface Except for the types of arguments in the methods, but then you'll have to manually control the matching of the same types in the interval methods. It's not as complicated as it seems, but I still prefer automatic control of types.If there is still a need for automatic control of PHP types, the automatic generation of interval classes for different types of points may be performed, but this topic requires separate discussion.In fact, the question is not exactly how, but the question is, are there solutions?Personally, I only got a class to work at intervals: https://github.com/thephpleague/period ♪ At the same time, I never looked for a solution to work with other types of intervals.

Although the final decision is working. I'd still prefer to do it through a composite.Plus, the combination is that you can combine different data cashing techniques with different data collection techniques.Moreover, data can be downloaded at several locations.interface IProvider{ public function getCategory(); } class DBProvider implements IProvider{ public function getCategory(){ //non cached code } } class MemcachedProvider implements IProvider{ private $provider; public function __construct( IProvider $real_provider ){ $this->provider = $real_provider; } public function getCategory(){ // if in cache return result from memcache // in not in cache then use $this-&gt;provider and save into cache } } class SessionCachedProvider implements IProvider{ // just example private $provider; public function __construct( IProvider $real_provider ){ $this->provider = $real_provider; } public function getCategory(){ // if value in session then use is // otherwise use $this-&gt;provider and save into session } } // обычный provider который ничего не кэшрует $provider = new DBProvider(); // тотже провайдер но эширует в memcached $cached_provider = new MemcachedProvider( $provider ); // провайдер кэширует в сессии $session_cached_provider = new SessionCachedProvider( $provider ); //кэшируем везде где только можно. $super_cache = new SessionCachedProvider( $cached_provider );

This error arises because Image1 It's not a global object, it's a class field. TForm1♪ The processor is the method of this class and sees the field, and it is inaccessible to the outside function.You can do that.function f():byte; begin Form1.Image1.Canvas.Brush.Color:=clBlack; Form1.Image1.Canvas.Rectangle(1, 1, 50, 50); end; But it is. CHEERING Not right. There is an unnecessary recourse to the global variable.Could be transferred to a class object TForm1function f(AForm: TForm1):byte; begin AForm.Image1.Canvas.Brush.Color:=clBlack; AForm.Image1.Canvas.Rectangle(1, 1, 50, 50); end; procedure TForm1.Button2Click(Sender: TObject); begin f(Self); end; But it's stupid. The same success can be achieved f by class TForm♪ I understand you want to write a function for drawing a rectangle on an arbitrary canva. Then all you have to do is give it to me as an index to the right canva.procedure f(ACanvas: TCanvas) begin ACanvas.Brush.Color:=clBlack; ACanvas.Rectangle(1, 1, 50, 50); end; procedure TForm1.Button2Click(Sender: TObject); begin f(Image1.Canvas); end; Starting with Delphi 2006, you can write a chopper for class. TCanvas and to do so f pseudonym of this classtype TCanvasHelper = class helper for TCanvas public procedure f; end; procedure TCanvasHelper.f begin Brush.Color:=clBlack; Rectangle(1, 1, 50, 50); end; procedure TForm1.Button2Click(Sender: TObject); begin Image1.Canvas.f; end;

Having ghetters and grids, you provide an open interface to the users of your class who will not depend on how the ghetters and the grid are implemented. This provides flexibility in design. You can change the ghetter and grid implementation, but the user code will not change.Of course, one thing has to be chosen: either ghetters and grids or open fields. Otherwise, the point is lost.

First, Designer initiated all fields, including protected and closed, which are not accessible to class users. Second, Designer enables the establishment of an object in a given location. corrective condition. Thirdly, you can create so-called temporary Objects that are not tied to a specific identifier, and therefore, it is virtually impossible to cause several grids for such objects. Fourth, One Designer challenge Replacement There's a few calls for the netters.

Use MVC. Besides, you'll figure out what SRP is.From this point of view, break your packets and directories. If necessary, new directories could be added accordingly. Any mvc directory. And resources in a separate director of resources. About that. [Proj dir] ├[src] │ ├[model] │ ├[view] │ └[controller] └[resources]

About why you're not compiling, as you've already written, your mistakes are lynchkin' errors, so it's an unconnected file. Student.cppand not found WinMain Says you're trying to create a final app, not a console.But you have the second part of the question - how to write. You don't have any mistakes, in the sense that they'll cause the program to fail. But like you said, it's not worth writing on C++. ♪I'd start with you having a structure rather than a class. In the sense that your class does not contain a specific internal condition that should be necessarily maintained in an agreed state or some other functionality other than the installation and return of fields. So, personally, I don't see a big sin just to declare your class a structure (i.e. class with all the open members) and work directly with them:struct Student{ string name; string surname; int age; int course; string group; string* marks; }; ... Student test; test.name = "Name"; If, however, some more serious work is planned or the assignment requires that this be a class, it is better for such small single-strong functions to be inline (instrued), i.e. to define them directly in the class announcement. This will allow the compiler not to call them, but simply to install the code, which is beneficial to productivity. And I would have done all the retrieval functions so that they could be called for for stable objects. I'd pass the lines as constant references to copy less.class Student{ public: Student(); void setName(const string& s) { name = s; } string getName() const { return name; } void setSurname(const string& s) { surname = s; } string getSurname() const { return surname; } ... private: string name; string surname; int age; int course; string group; string* marks; }; Also, your grades are clearly extravagant, so I wouldn't be in town with a set of lines, but I just used a line, which, in general, is a set of symbols. evaluation - i.e. joining this line of another symbol. Besides, you'd get rid of what's called magic constants-- you wouldn't have to remember that grades 6:for(char c : mark){ cout << c <<" "; } And you're all in class. copy.what's right is that the object is self-sufficient, "all of its own." And here's the body you're just handing over as an index. External in relation to the class, the essence, and if it's changed, it works, and the state of the class. I mean, your student's score isn't a setup, it's a cover with a paper that says, "and my grades are recorded over there on a class board" where any bully can fix them... I mean, even if you're working with a set of lines, it's better to bury it in, for the propulsion.And, why don't you create another designer that would create a first-named student? I don't think the name will change.

The question is really very vague. I'm suggesting his own combing on the terms and things used. They say "the right question is 50 per cent of the right answer."Well, what's in it is in the form of assumptions...Once the question is marked by the "s+++" the logical assumption is to use the "s++" possibilities itself (as many C) and/or STL.If we need to implement a Segmented Squad based on STL, we read a description http://ru.cppreference.com/w/cpp/container/deque/deque and http://en.cppreference.com/w/cpp/memory/allocator ♪ We're trying to implement our allocator.If we need to implement "S segmented-band" without using STL, we're looking for and reading on " http://en.cppreference.com/w/cpp/language/new "The question of &quot; policy of accommodation &quot; of the elements by means of the above-mentioned allocators or the placement of new ones remains unclear. ♪Turns out the blocks start to be set up in the middle of the directory to leave room for new blocks.It's like programme. Nevertheless, it may be necessary to:The size of "this" where we're going to set asideList of undistributed clusters "tam"A mechanism to deal with the question, "What do you do if the undistributed blocks are over?"The implementation of the list of undistributed blocks depends directly on, say, the use forecast. In STL, different containers operate according to different speeds when access, supply, removal, addition. There should be a better view of the developer in terms of the real challenge - what type of container is to use.Like this. What's the question is the answer:

Ah. I'm Arkhangelski, S++ Builder 6, c.519, M., 2002. I've built objects in pursuit of TChart, TEdit, TPanel, TSeries. It's done, I'm doing it this day.

Reduce beyond:TFile=class(TComponent) private FPath : String; FSize : Int64; FParent: TFile; published property Parent: TFile read FParent write FParent; property Path : String read FPath write FPath; property Size : Int64 read FSize write Fsize; end; procedure FindFiles(const DirPath: string; parent: TFile; Owner: TComponent); var SR: TSearchRec; mfile: TFile; begin if FindFirst(DirPath + '*.*', faAnyFile, SR) = 0 then try repeat if not ((SR.Name = '.') or (SR.Name = '..')) then begin mfile:=TFile.Create(Owner); mfile.Path:=DirPath+ '' + SR.Name; mfile.Size:=SR.Size; mfile.Parent:=parent;//Это новое свойство в TFile if SR.Attr = faDirectory then FindFiles(DirPath + '' + SR.Name, mfile, Owner); end; until FindNext(SR) <> 0; finally FindClose(SR); end; end; procedure TForm1.MagicButtonClick(Sender: TObject); begin FindFiles('D:',nil,Self); end; So the files turned into components. Why do you need that?

The answer to &quot; why &quot; is simple: because https://docs.oracle.com/javase/specs/jls/se8/html/jls-5.html#jls-5.1.5 ♪Each class may be inherited from other classes (and interfaces) and the left part of the assignment may be a pre-class or an interface.More interesting, why. Look, this is it. polymorial♪ It's often that a few different things have a lot in common, similar to each other, and you want to treat them the same way. For example, you have classes that are a plane and a helicopter, and you want to check for each of them if you need repair. The hold is that you announce the general parentage class of the &quot; flying machine &quot; , make a general functionality there. Now you can.To declare the listList<>) Your aircraft, and contain both aircraft and helicoptersTo declare a function that takes place at the entrance of the aircraft and conducts the same actions for the aircraft and the helicopter (e.g. inventory).Why, right after the designer's call, declare a narrower type of variable? Options may also be several:You may wish to emphasize that you treat the object as one of the groups of similar objects. For example, you can set up a plane, but you're using it as an aircraft, and you're not interested in a plane or a helicopter.You can get access to the data you've got. http://ideone.com/F2tTRl for such classesclass Base { public String name = "Base"; } class Derived extends Base { public String name = "Derived"; } access Base.name type reference Derived You won't get it.

You misunderstand the work mechanism of your page, from here, I think it's a problem.I'll take the points.Every download of the page in the browser, every update of the page of the lane uploads your violin. I mean, the php of the laundry processs your code every time.The first download is the file that is listed on the address line of the browser. It may be main.php, news.php, index.php or any other, but it will be downloaded first.Further, as the basic violin has been performed, other violators shall be connected if instructions for their connection - include or requireLet's see your file delete_news.inc.php:You're checking the variable $news and there's gonna be an object in it, but only if it's been installed somewhere before. That is why the main file in the variable will be downloaded by reference to the facility. But if you download the file delete_news.inc.php directly, the variable will be empty and you will get a mistake. So if you want to use this file separately, you need to comment on the NewsBD connection and the $news variable.Let's go back to your main file, code:if ($_SERVER['REQUEST_METHOD']=='POST') { require_once ("save_news.inc.php");} require_once ("delete_news.inc.php"); ?> You don't have a closing } after you switch off the delete_news.inc.php. And the only way to connect this file is when it comes to a post-request, it doesn't make sense, because the news is being removed on the GET parameter and the POST doesn't depend.In fact, your code clearly shows the enormous problem arising from this website approach. By handwriting up the required files, you'll sooner or later come to the fact that for any change, you're gonna have to overload the ear files, and that's gonna be the job.

In order to answer this question, the memory management debrief and the presentation of variables should be addressed. Happiness adds that the presentation has changed quite significantly in PHP7 and the comprehensive response requires both models to be described. The minimum description of this kitchen has already required articles in two parts: http://nikic.github.io/2015/05/05/Internal-value-representation-in-PHP-7-part-1.html ♪ http://nikic.github.io/2015/06/19/Internal-value-representation-in-PHP-7-part-2.html ♪ Translation to Russian: https://habrahabr.ru/company/mailru/blog/257999/ ♪ https://habrahabr.ru/company/mailru/blog/261131/ ♪ Then the articles are the objects. http://jpauli.github.io/2015/03/24/zoom-on-php-objects.html (sighs) https://habrahabr.ru/company/mailru/blog/255237/ change of device http://jpauli.github.io/2016/01/14/php-7-objects.html (sighs) https://habrahabr.ru/company/mailru/blog/275497/ )The answers to so are not recommended, but this is a huge plate of material, and even a basis in a clear form for a reasonable volume of text, I will not be able to state, but also to suggest that the question should be clarified, I do not believe it is.

C#/Java has no multiple class inheritance, but there is a multiple internship. Therefore, in these languages, the separation of interfaces and abstract classes is a non-voluntary option.There is no restriction in C++, so there is no particular need to separate the concept of the interface and to separate it from the abstract class. Any abstract class may serve an interface, but it does not even have to have public virtual functions (except for the destructor if polymorphous disposal is required). Example of such an interface - Pattern https://en.m.wikibooks.org/wiki/More_C%2B%2B_Idioms/Non-Virtual_Interface ♪If the PLO wishes to be more consistent with the C#/Java language paradigm, the names starting from the letter may be given to interfaith classes Ior even add #define interface struct In order to distinguish the Interface Class in the declaration/definition. Such approaches are used in MS COM.But in general, this is not necessary, and the problem of the rumboid hierarchy can be solved through virtual succession.

Example with addition of classes of operationsWhat you described in the example is almost a design template called Visitor. NextOperation ♪ PrevOperation ♪ RandomOperation - class visitors Playback In this case, this template is hardly justified because it is intended to add a hierarchy of classes, and it really plays a hierarchy when working with hierarchy. OCP (i.e. the need to add the behaviour in the tree of heirs by adding the method to the base class as a breach of the OCP). You class one. Playback ♪ OCP No violation in the first case Playback open to extension (if of course change) private ♪ protectedand closed to change (this means you don't need to rewrite it in your heirs). You can't see a minus of a second approach - you're seriously complicating the system by blurting a simple logic for many classes. http://sergeyteplyakov.blogspot.ru/2014/10/about-design-principles.html ) So I'd advise you to leave the first option, at least when the system is simple, it can't be so complicated, https://ru.stackoverflow.com/questions/556137/%D0%9D%D0%B0%D1%80%D1%83%D1%88%D0%B0%D0%B5%D1%82-%D0%BB%D0%B8-ocp-%D0%B8-dip-%D0%B8%D0%B7-solid-%D0%BF%D1%80%D0%B8%D0%BD%D1%86%D0%B8%D0%BF-yagni ♪ But if Playback becomes a basic hierarchy class - the use of visitors will be justified and the method should be added visit(PlayOperation operation)where possible Playback - doesn't mean the method. next()e.g., it is necessary to be delivered to the Visitor &apos; s class: it is better for visitors to take additional operations and for the main to remain in the classroom.

Only the other day. http://php.net/manual/ru/oop4.newref.php ♪If briefly, new returns the copy, not the reference, and if the design is used $a = &new Foo()the reference to it in the global mass $globalref It won't point to the same object as it is. $a♪In addition, based on documentation, a more resource-intensive process has been established.Conduct xdebug_debug_zval I would explain that:$obj = &new A; new A creates a copy of the class, transmits it to the operator. &which generates a reference to it and further transmits to the assignee operator.I mean, $obj We've got a reference to a copy of the installed class. A♪ From here on, it appears that, in fact, the variable containing this class does not exist. There's only a reference to the memory area where it's located.UPDThat's what happens.Establishment and reference of the facility - refcount=1♪ I mean, this one. &obj is a reference. From here. is_ref=1♪Reference to the reference $alterthe reference enumerator shall be incremented - refcount=2♪ She's a link, too, from here. is_ref=1♪You remove the reference to an object that has two references. The reference counter is reduced to the unit I(!!) the last variable that refers to this facility automatically ceases to be a reference because it is the last one. If you know *nix, it's practically analog hardlink.Try this code.$a = 'a'; $b = &$a; $с = &$a; xdebug_debug_zval('a'); xdebug_debug_zval('b'); xdebug_debug_zval('c'); unset($a); xdebug_debug_zval('b'); xdebug_debug_zval('c'); unset($b); xdebug_debug_zval('c'); END UPDPHP 7 Parse error: syntax error, unexpected 'new' in [. ♪ _ ♪ ♪ ] on line 6PHP 5.3-5.6 Deprecated: Assigning the return value of new by reference is deprecated in [. ♪ _ ♪ ♪ ] on line 6

First of all, in a good design, the base class should know nothing about its heirs, otherwise one of the PLO &apos; s contractual programming framework is being violated.Do another abstract method and redesign it to the heirs in such a way that they return new copies of their type:public abstract class AbstractClass { public AbstractClass foo() { AbstractClass a = createInstance(); a.bar(); return a; } public abstract void bar(); protected abstract AbstractClass createInstance(); } public class ConcreteClass extends AbstractClass { protected override AbstractClass createInstance() { return ConcreteClass(); } } If you do not want to touch the heirs and go dirty, reflexia can be used (but in this case all heirs must have a designer without parameters):public Abstract foo() { AbstractClass a = this.getClass().newInstance(); a.bar(); return a; }