, functions, structures or constants) of which one computer program can interact with another program. Typically included in the description of some Internet protocol (for example, RFC), software framework (framework), or operating system function call standard. Often implemented by a separate software library or operating system service. Used by programmers when writing all kinds of applications.

API as a means of application integration

The API defines the functionality that the program (module, library) provides, while the API allows you to abstract from how exactly this functionality is implemented.

If a program (module, library) is considered as a black box, then the API is a set of “handles” that are available to the user of this box and which he can twirl and pull.

Software components interact with each other through APIs. In this case, components usually form a hierarchy - high-level components use the API of low-level components, and they, in turn, use the API of even lower-level components.

Data transfer protocols over the Internet are built on this principle. Standard stack protocols (OSI network model) contains 7 levels (from physical level passing bits down to application protocols like HTTP and IMAP). Each layer uses the functionality of the previous (“lower”) data transfer layer and, in turn, provides the necessary functionality to the next (“higher”) level.

It is important to note that the concept of a protocol is close in meaning to the concept of an API. Both are abstractions of functionality, only in the first case we are talking about data transfer, and in the second we are talking about the interaction of applications.

The function and class library API includes a description signatures And semantics of functions.

Function signature

Sometimes they distinguish call signature And implementation signature functions. A call signature is usually compiled from the syntactic structure of a function call, taking into account the signature of the scope of the given function, the name of the function, the sequence of actual types of arguments in the call, and the type of the result. The implementation signature usually includes some elements from the syntactic structure of the function declaration: a function scope specifier, its name, and a sequence of formal argument types.

For example, in the C++ programming language simple function is uniquely recognized by the compiler by its name and the sequence of types of its arguments, which constitutes the signature of the function in this language. If a function is a method of a certain class, then the class name will also be included in the signature.

In the industry software Common standard APIs for standard functionality are important because they ensure that all programs using the common API will work equally well, or at least in a typically familiar way. In the case of GUI APIs, this means that programs will have similar user interface, which facilitates the process of mastering new software products.

On the other hand, differences in the APIs of different operating systems significantly complicate the transfer of applications between platforms. There are various methods workaround this complexity - writing "intermediate" APIs (APIs of graphical interfaces wxWidgets, GTK, etc.), writing libraries that map system calls of one OS to system calls of another OS (runtimes such as Wine, cygwin, etc. .), introduction of coding standards in programming languages (for example, standard library C language), writing interpreted languages implemented on different platforms (python, perl, php, tcl, Java, etc.).

It should also be noted that the programmer often has several different APIs at his disposal to achieve the same result. Moreover, each API is usually implemented using API software components lower level of abstraction.

For example: in order to see the line “Hello, world!” in the browser. ", you just need to create an HTML document with a minimal title and a simple body containing this line. When the browser opens this document, the browser program will pass the file name (or an already open file descriptor) to the library that processes HTML documents, which, in turn, using the operating system API, will read this file and understand its structure, then sequentially call through API library of standard graphic primitives operations like “clear the window”, “write “Hello, world!” in the selected font”. While performing these operations, the graphics primitives library will contact the window interface library with the appropriate requests, and this library will call the operating system API to write data to the video card buffer.

Moreover, at almost each level there are actually several possible alternative APIs. For example: we could write the source document not in HTML, but in LaTeX, and we could use any browser for display. In addition, different browsers use different HTML libraries, and, in addition, all this can be compiled using different primitive libraries and on different operating systems.

The main challenges of existing multi-level API systems are therefore:

Difficulty in porting program code from one API system to another (for example, when changing the OS);
Loss of functionality when moving from a lower level to a higher one. Roughly speaking, each API “layer” is created to facilitate the execution of some standard set of operations. But at the same time, it becomes really difficult or becomes fundamentally impossible to perform some other operations that are provided by a lower level of API.

Most famous APIs

Operating systems

This short term is well-known to everyone who has at least some experience with development. But not everyone understands what exactly it means and why it is needed. Developer Peter Gazarov talked about the API in simple words on your blog.

The abbreviation API stands for "Application Programming Interface" software interface applications). Majority large companies at a certain stage, APIs are developed for clients or for internal use. To understand how and how APIs are used in development and business, you first need to understand how the World Wide Web works.

World Wide Web and remote servers

The WWW can be thought of as a huge network of interconnected servers on which every page is stored. An ordinary laptop can be turned into a server capable of serving an entire website on the network, and local servers developers use it to create websites before making them available to the public.

When entered into the address bar of the browser www.facebook.com A corresponding request is sent to the remote Facebook server. Once the browser receives the response, it interprets the code and displays the page.

Every time a user visits any page on the web, he interacts with the API remote server. The API is the component part of the server that receives requests and sends responses.

API as a way to serve clients

Many companies offer APIs as a ready-made product. For example, Weather Underground sells access to its weather data API.

Usage scenario: On the website of a small company there is a form for making appointments for clients. The company wants to integrate Google Calendar into it to give customers the ability to automatically create an event and enter details about an upcoming meeting.

API Application: The goal is for the site server to directly contact the Google server with a request to create an event with the specified details, receive Google's response, process it, and send the appropriate information to the browser, for example, a message asking for confirmation to the user.

Alternatively, the browser can make a request to Google's server API without going through the company's server.

Than API Google Calendar different from the API of any other remote server on the network?

Technically, the difference is in the format of the request and response. To generate a complete web page, the browser expects a response in the language HTML markup, while the Google Calendar API will simply return data in a format like JSON.

If a request to the API is made by the server of a company's website, then it is the client (just as the browser is the client when the user opens the website).

Thanks to the API, the user gets the opportunity to perform an action without leaving the company’s website.

Most modern websites use at least a few third-party APIs. Many tasks already have ready-made solutions offered by third-party developers, be it a library or a service. It is often easier and more reliable to resort to a ready-made solution.

Many developers distribute the application to several servers, which interact with each other using the API. Servers that perform a supporting function to the main application server are called microservices.

So, when a company offers an API to its users, it simply means that it has created a series of special URLs that return only data as a response.

Such requests can often be sent through a browser. Since HTTP data transfer occurs in text form, the browser will always be able to display the response. For example, through a browser you can directly access the GitHub API (https://api.github.com/users/petrgazarov), without an access token, and receive this response in JSON format:

The browser perfectly displays the JSON response, which can be inserted into the code. It is easy enough to extract data from such text to use it at your discretion.

Some more API examples

The word "application" can be used in different meanings. In the context of the API, this means:

a piece of software with a specific function,
the entire server, the entire application, or just a separate part of the application.

Any piece of software that can be clearly distinguished from the environment can replace the letter “A” in an English abbreviation, and can also have some kind of API. For example, when a developer implements a third-party library into the code, it becomes part of the entire application. As a standalone piece of software, the library will have some kind of API that allows it to interact with the rest of the application code.

In object-oriented design, code is represented as a collection of objects. In an application, there can be hundreds of such objects interacting with each other. Each of them has its own API - a set public properties and methods for interacting with other objects in the application. Objects may also have private, internal logic that is hidden from the environment and is not an API.

Sooner or later, any programmer is faced with such a concept as API. However, when such a meeting occurs, not everyone knows what it is, why it is needed and how to use it. And in this article I am going to fill this gap in the knowledge of some of you, and also give an example from my practice.

API (application programming interface) - This application programming interface. More to the point in simple language, then this is a set of various functions, constants, classes, query formats that can be used in other programs.

It can be considered that API- this is a certain object, the implementation of which we do not know, however, we can use it. For example, a computer is an object whose implementation very few people know, however, almost everyone can use it to perform some actions: watching videos, surfing the Internet, printing text, etc. Few people know how it all works, but almost everyone can do it.

Example API is Windows API, OpenGL API, Direct3D API and so on.

For example, not long ago I also encountered directly API. I registered for the mailing list service" SmartResponder.ru" and started a newsletter, which people began to subscribe to. The task was as follows: within 24 hours after subscribing, a person can purchase my paid video course at a discount. Since all information about subscribers is stored on the server " SmartResponder.ru", then normal access (for example, through DB) I did not have access to this data, but it was necessary to implement it. Thankfully, " SmartResponder.ru"have your own API, which I used.

I found in them API request format to get the subscription date as a result. Further through cURL I sent the corresponding request and received the required subscription date for a specific e-mail addresses . Next is standard processing and output of the result.

The API defines the functionality that a program (module, library) provides, while the API allows you to abstract from how exactly this functionality is implemented.

If a program (module, library) is considered as a black box, then the API is a set of “knobs” that are available to the user of this box, which he can twirl and pull.

Data transfer protocols are built on this principle. The standard Internet protocol (OSI network model) contains 7 layers (from the physical layer of transmitting bit packets to the layer of application protocols like HTTP and IMAP protocols). Each layer uses the functionality of the previous data transfer layer and, in turn, provides the necessary functionality to the next layer.

The function and class library API includes a description signatures And semantics of functions.

Application Programming Interface (API) is a software interface for interaction between systems, allowing:

Gain access to enterprise business services
Exchange information between systems and applications
Simplify interactions between companies, partners, developers and clients

Open API strategy

API strategy includes:

Development of business products based on existing APIs
Providing internal services to developers
API monetization models for building multi-channel interaction and increasing profits

The implementation of the Open API concept helps transform business, integrate it into a flexible project ecosystem of market players, create conditions for the constant generation of new ideas and the creation of additional value when managing arrays of corporate data.

The market for integration solutions is developing in the context of the evolution of APIs - from EDI and SOAP to Web 2.0, which began the era of public APIs. The number of such interfaces in the next 3 years may grow by more than 50 times and reach 1 million. This is due to omnichannel: the channels of interaction with customers must change along with them. The continuous growth in the number of consumers and the volume of data has led to the emergence of an API economy that helps open interfaces create innovative business models for using corporate assets and services.

Function signature

Function signature- Part general announcement functions that allow translation tools to identify a function among others. Different programming languages have different ideas about the signature of a function, which is also closely related to the capabilities of function overloading in these languages.

Sometimes they distinguish call signature And implementation signature functions. A call signature is usually compiled from the syntactic structure of a function call, taking into account the signature of the scope of the given function, the name of the function, the sequence of actual types of arguments in the call, and the type of the result. The implementation signature usually includes some elements from the syntactic structure of the function declaration: a function scope specifier, its name, and a sequence of formal argument types.

For example, in the C++ programming language, a simple function is uniquely identified by the compiler by its name and the sequence of types of its arguments, which constitutes the function signature in this language. If a function is a method of a certain class, then the class name will also be included in the signature.

For example: in order to see the line “Hello, world!” in the browser. All you need to do is create an HTML document with a minimal title and a simple body containing this line. What happens when the browser opens this document? The browser program will pass the file name (or an already open file descriptor) to the library that processes HTML documents, which, in turn, using the operating system API, will read this file and understand its structure, calling operations like “clear the window”, “write Hello, world in the selected font!”, during these operations the library of graphic primitives will contact the window interface library with the corresponding requests, and this library will contact the operating system API with requests like “put it in my video card buffer this."

Moreover, at almost each level there are actually several possible alternative APIs. For example: we could write the source document not in HTML, but in LaTeX, and we could use any browser for display. Various browsers, generally speaking, use different HTML libraries, and, furthermore, the whole thing can (generally speaking) be compiled using different primitive libraries and on different operating systems.

The main challenges of existing multi-level API systems are therefore:

Difficulty in porting program code from one API system to another (for example, when changing the OS);
Loss of functionality when moving from a lower level to a higher one. Roughly speaking, each API “layer” is created to facilitate the execution of some standard set of operations. But at the same time, it becomes really difficult or becomes fundamentally impossible to perform some other operations that are provided by a lower level of API.