Why does a video file from VK search results refer to a user who does not have more than one video?

Question

In contact with

Disclaimer: The
question I'm asking here is part of a more serious study as part of my thesis. The links will be to porn videos, the question asked came up absolutely by accident, but it’s not appropriate to freak out about the fact that people here are interested in porn, I’m looking for porn specifically and consciously, but the goal is not to watch porn, but to write a diploma consciously, but to reveal my maximum The topic can be precisely on the examples of porn and social networks.
So let's get started.
Kind all the time, how can you explain the following to me, there is a video file https://vk.com/video-118966358_456241274, it was uploaded by this user https://vk.com/id365751935 however, the user does not have more than one video. Can someone explain to me how this is possible?
ps if in your own words, then the topic of the diploma is about Internet fraud and methods of protection, it’s really too lazy to type the correct name from the phone.

Answer 1

bro,
until you become a programmer and you learn about database synchronization, and you don’t try - you still won’t understand anything. If you
want to read about "real time" (difficult), or shove the Bookvarix database in Postgre to evaluate this time

Answer 2

Well, the tests are different: both green and red. ))
It all depends on the level of abstraction that you want to test. Therefore, tests are divided into all sorts of classifications. For example: unit (unit) tests, integration tests, etc. etc.
I, as an ardent supporter of BDD, use two types of tests: feature (read integration tests) and spec (read unit tests)
So, features. We take Behat and hack a bunch of tests by type: I
went to the main page,
I poked something in the registration form I
went to my profile and saw my name and photo on the
Profit page!
In general, we got: "Direct registration with entry into the database", only in this situation it does not matter to us at all what is in the database. It is important for us that the user registers and gets into the closed section of the site.
For such tests, it is good to have environment support (production, development, test) in the code, so that, for example, you can disable captcha or do some other complicated crap.
If the system is terribly sophisticated, you will have to raise the entire environment here for tests. And it’s better to give CI this to do, let a friend work.
Plus, such tests, for example, when we write a complex front - a complex back-back and an even more complex back-back, then you can test the operation of the entire service in one go.
The next level of abstraction spec:
If we didn’t give a damn about the database in integration tests. She seems to write, but what kind of structure is absolutely do not care. Then in the case of the sintered ones, we DO NOT care.
We take our class (function) and check what results it gives. Instead of objects with which the test subject (an object of the class under test), we give him a rubber woman (mocks, etc.), and look at the results of our work.
The main thing with both approaches is that it doesn’t matter to us that we always test the public properties of the system in our guts. In the first case, this is a reaction to the user, in the second, the public API of the class.
Something like that!
PS tests should be written before the code.

Answer 3

Who are the tests for? For man or machine?
I mean, tests are different: manual and automatic. This is not mentioned in your question!
What kind of tests are needed? Modular? Integration? Installations? Load? About this, too, not a word in your question!
Tests are written like this:
1. They first identify working scenarios, i.e. when all is well and the result appears. There are a huge number of situations when a user is ready to pay for a strange product, if at least 1 time and 50 launches he will do what he did for 3 days! Scenarios are sorted by:
1.1. components
1.2. priority and importance.
After sorting create test sets
2. Next, they find out scenarios when something is missing "the network has fallen, what should I do?". In other words, check the operation of positive scenarios with possible negative cases, which may be, but not the fault of the user. Examples: the network has crashed, the flash drive has stopped being seen, access to the database is suddenly not allowed, etc. They especially look at the possible damage to the original data. My friend had a case when they picked up the password to the database and accidentally erased a couple of bytes. ;)
3. Only after that they check the "error for a fool." Examples: instead of a text file, they gave an exe file. Or a string instead of a number.
Let's assume that you want to write unit tests. I will say the following that these types of tests should not do:
1. Check the work with the database, disk, network
2. Time dependent code
The main indicator of a unit test is the speed of work. If a test of this type runs for half a second, then you wrote anything, but this is not a unit test ;)