1 00:00:04,390 --> 00:00:11,230 Another common stage that you may encounter when building S.I pipelines is the unit stage or the unit's 2 00:00:11,230 --> 00:00:11,650 job? 3 00:00:12,530 --> 00:00:13,840 First, what does the unit test? 4 00:00:14,560 --> 00:00:20,460 Unit tests are responsible for testing only single units of code, typically only one class. 5 00:00:20,920 --> 00:00:24,610 The execution time is very fast and they give instant feedback. 6 00:00:25,420 --> 00:00:30,280 A common concept that you'll hear when discussing testing is the so-called testing pyramid. 7 00:00:30,520 --> 00:00:37,090 It takes different forms and shapes, but always at the bottom of the test pyramid you will find the 8 00:00:37,090 --> 00:00:38,050 unit test. 9 00:00:38,680 --> 00:00:41,920 The reason why they are there is the reason I have specified before. 10 00:00:42,100 --> 00:00:43,510 They run very, very fast. 11 00:00:43,630 --> 00:00:45,280 They don't have a lot of preconditions. 12 00:00:45,280 --> 00:00:48,580 They don't need the entire application and they are quite cheap. 13 00:00:48,580 --> 00:00:51,100 They are easy to write, easy to execute. 14 00:00:51,460 --> 00:00:55,840 So overall, the cost of having unit tests is very, very good. 15 00:00:56,880 --> 00:01:01,640 Now, they are not the only kind of test that we can have in our application, just because the unit 16 00:01:01,640 --> 00:01:04,500 tests to run doesn't mean that the entire application works. 17 00:01:05,069 --> 00:01:11,070 But they are a very important level and they are present in practically any continuous integration pipeline. 18 00:01:12,640 --> 00:01:14,530 Let me give you an example from the application. 19 00:01:15,650 --> 00:01:22,190 Kareem Posman, if I go to the statistics folder, I'll be able to get to the average fleet age. 20 00:01:23,390 --> 00:01:30,020 Now, this is something that is computed inside a code, which means that there is a Java class which 21 00:01:30,020 --> 00:01:35,810 will go over all the vehicles that are in the database and you can view them here. 22 00:01:37,060 --> 00:01:41,590 Will it take to build here and will do an average with this building? 23 00:01:43,380 --> 00:01:51,990 Looking at a code, this can be found in source main Java Services Fleet Statistics Service. 24 00:01:52,910 --> 00:01:59,060 And even if you're not too much into Java, you can easily see how this works, it goes over all the 25 00:01:59,060 --> 00:02:05,150 cars and it all use the average function to get an average from the builder and will give this information 26 00:02:05,150 --> 00:02:05,450 back. 27 00:02:06,110 --> 00:02:11,930 Now, this code may or may not work, so we're not really sure if we need to add additional things to 28 00:02:11,930 --> 00:02:15,630 the code, then it may break out of things that previously used to work. 29 00:02:15,830 --> 00:02:22,760 So for that reason, we always write unit tests which test at this particular class or this particular 30 00:02:22,970 --> 00:02:25,790 method within the class still works the same. 31 00:02:27,110 --> 00:02:29,070 This class here has to test. 32 00:02:29,210 --> 00:02:31,520 One test is simply testing for one car. 33 00:02:31,550 --> 00:02:33,320 What happens if there's only one car? 34 00:02:33,590 --> 00:02:40,150 It adds it to the test and then it can check the average, but then it can also add two cars and then 35 00:02:40,170 --> 00:02:42,020 will check the average from the two cars. 36 00:02:42,770 --> 00:02:45,900 So this is the main idea when it comes to unit testing. 37 00:02:45,920 --> 00:02:51,550 You have a class which performs a specific task, in this case calculating the average. 38 00:02:51,830 --> 00:02:57,890 Then we want to include additional things here just to test that this class actually works how it should 39 00:02:57,890 --> 00:02:58,210 work. 40 00:02:59,440 --> 00:03:05,860 When it comes to executing tests that can be done in different ways, one ways is simple to use cradle 41 00:03:06,190 --> 00:03:08,860 verification and to run test. 42 00:03:09,990 --> 00:03:16,590 But this is like a graphical way on how we can start to test the command that we can run locally, but 43 00:03:16,590 --> 00:03:20,300 just as well in Lipsky is using the greatest rapper. 44 00:03:20,650 --> 00:03:24,450 So the great rapper where it's like this we call the great rapper. 45 00:03:25,630 --> 00:03:26,950 And we called test. 46 00:03:33,150 --> 00:03:37,350 Now, let's go ahead and make one of these tests fail, how will do this? 47 00:03:37,350 --> 00:03:39,710 It's a simply modify here a number. 48 00:03:39,720 --> 00:03:41,670 So actually the code will be OK. 49 00:03:41,760 --> 00:03:45,690 The test will not be OK itself because the test is wrong. 50 00:03:45,890 --> 00:03:48,750 It will show you how a failed test looks like. 51 00:03:56,890 --> 00:04:02,380 Now, you will be able to see that this command has failed and again, a report will be generated. 52 00:04:02,380 --> 00:04:09,000 So you can click on the report, however, here on the corner to open the browser and inside the browser, 53 00:04:09,010 --> 00:04:10,810 you'll be able to see a test summary. 54 00:04:10,810 --> 00:04:12,360 We'll see which are the failed test. 55 00:04:13,890 --> 00:04:19,230 You will see here that there is an insertion error that has occurred, expected was 15, but we got 56 00:04:19,230 --> 00:04:21,290 16 and this is why the tests fail. 57 00:04:21,779 --> 00:04:23,900 Of course, this is not a real fail. 58 00:04:24,180 --> 00:04:27,290 I have just made a test fail to demonstrate how it is. 59 00:04:27,300 --> 00:04:29,490 So I'm going to move it back to 15.