1
00:00:00,270 --> 00:00:03,080
Now that we saw this overview of white-box testing, I'd like

2
00:00:03,080 --> 00:00:06,270
to start talking about specific coverage criterion. And I'm going to start

3
00:00:06,270 --> 00:00:09,010
with the first one, which is Statement Coverage. This criterion is

4
00:00:09,010 --> 00:00:12,355
going to be characterized by two aspects, the first one is which

5
00:00:12,355 --> 00:00:15,580
are the Test requirements for the criteria and the second one

6
00:00:15,580 --> 00:00:18,830
is how we measure Coverage for that criteria. In the case

7
00:00:18,830 --> 00:00:22,200
of statement coverage, these test requirements are all the statements in

8
00:00:22,200 --> 00:00:25,804
the program. So this is the basic, the first the, the simplest

9
00:00:25,804 --> 00:00:29,860
coverage criteria in the white-box arena. Let me remind you the assumption

10
00:00:29,860 --> 00:00:33,280
that we made at the beginning. White-box testing is based on the assumption

11
00:00:33,280 --> 00:00:35,990
that if there isn't a faulty element in the code, we need

12
00:00:35,990 --> 00:00:38,390
to exercise it. We need to execute it, in order to find the

13
00:00:38,390 --> 00:00:41,260
fault. And that's exactly what statement coverage does. If there is a

14
00:00:41,260 --> 00:00:44,450
statement that is faulty in the code, we need to exercise it, in

15
00:00:44,450 --> 00:00:47,490
order to find the fault. And therefore, a good measure of how well

16
00:00:47,490 --> 00:00:51,480
we exercise the code, is the ratio of the number of executed statements.

17
00:00:51,480 --> 00:00:54,950
So all the statements that my test cases executed, to the total

18
00:00:54,950 --> 00:00:58,570
number of statements in the program. The higher this number, the better

19
00:00:58,570 --> 00:01:01,870
I exercise my code. And we can also look at coverage criterion

20
00:01:01,870 --> 00:01:04,290
in terms of questions. So what is the questions they were trying

21
00:01:04,290 --> 00:01:06,940
to answer when we look at a specific set of test cases

22
00:01:06,940 --> 00:01:09,870
and we assess the statement coverage that they achieved. And the question

23
00:01:09,870 --> 00:01:13,440
is whether each statement in the program has been executed. So, statement

24
00:01:13,440 --> 00:01:16,920
coverage is satisfied when all the statements in the program have been executed.

25
00:01:16,920 --> 00:01:19,640
And we can satisfy to different degrees and the degrees to which it's

26
00:01:19,640 --> 00:01:23,320
satisfied is measured by this value. So now let's go ahead and measure

27
00:01:23,320 --> 00:01:27,250
statement coverage on our printSum example. What I'm going to show down here is

28
00:01:27,250 --> 00:01:30,500
this progress bar in which we show the amount of coverage, the percentage of

29
00:01:30,500 --> 00:01:33,060
coverage achieved. So what this means is that the, if I get to

30
00:01:33,060 --> 00:01:36,680
this point I've covered 25% of the statements in the code. And my goal

31
00:01:36,680 --> 00:01:39,160
is to get up here to cover all the statements in the code.

32
00:01:39,160 --> 00:01:42,000
We have two test cases for this code. The first one that we just

33
00:01:42,000 --> 00:01:45,700
saw, consists of the inputs a equal to 3 and b equal

34
00:01:45,700 --> 00:01:48,200
to 9, and the second one has the inputs a is equal to

35
00:01:48,200 --> 00:01:50,840
minus 5 and b is equal to minus 8. So now let's see

36
00:01:50,840 --> 00:01:53,450
what happens when we run this test case. When we run this test

37
00:01:53,450 --> 00:01:56,920
case, I'm going to show you by highlighting in the code the parts that

38
00:01:56,920 --> 00:02:00,430
we cover when we start executing the code. We cover the first statement,

39
00:02:00,430 --> 00:02:01,910
then we always execute the second

40
00:02:01,910 --> 00:02:04,180
statement, which computes the result, we continue

41
00:02:04,180 --> 00:02:07,070
the execution, we get to the if statement. If the result is greater

42
00:02:07,070 --> 00:02:10,038
than zero, in this case our result is 12 because we

43
00:02:10,038 --> 00:02:12,360
are working with the inputs 3 and 9, and therefore we

44
00:02:12,360 --> 00:02:15,710
execute the true part of the if, we execute the statement.

45
00:02:15,710 --> 00:02:18,850
And at this point, we just jump to the end. Because we

46
00:02:18,850 --> 00:02:21,530
do not execute the else part of the statement, since we

47
00:02:21,530 --> 00:02:24,240
have executed a true one, and therefore, we cover this final

48
00:02:24,240 --> 00:02:27,534
statement. So at the end of the execution of this test

49
00:02:27,534 --> 00:02:32,313
case, we cover one, two, three, four, five statement out of seven

50
00:02:32,313 --> 00:02:36,299
which is roughly speaking 71%. So we can mark in here

51
00:02:36,299 --> 00:02:39,779
that we more or less got to 71% of coverage for

52
00:02:39,779 --> 00:02:42,719
this code. Now let's look at what happens when we execute

53
00:02:42,719 --> 00:02:45,700
test case number two. In this case again, we execute the

54
00:02:45,700 --> 00:02:48,660
first statement, the second statement, the third statement. In this case

55
00:02:48,660 --> 00:02:52,010
though, the first statement, when it evaluates the value of result,

56
00:02:52,010 --> 00:02:54,250
it sees that the result is not greater than zero because

57
00:02:54,250 --> 00:02:57,590
our inputs are minus five and minus eight. Therefore, you will execute

58
00:02:57,590 --> 00:03:00,090
line number five. And because the result is less than

59
00:03:00,090 --> 00:03:02,810
zero, you will also execute line number six. So, at

60
00:03:02,810 --> 00:03:05,230
this point, all of the statements in our code are

61
00:03:05,230 --> 00:03:09,360
executed and therefore, we achieved a 100% statement coverage, which

62
00:03:09,360 --> 00:03:13,090
was our goal. Before looking at other kinds of coverage,

63
00:03:13,090 --> 00:03:16,060
let's see how our statement coverage is used in practice.

64
00:03:16,060 --> 00:03:19,170
First of all, statement coverage is the most used kind

65
00:03:19,170 --> 00:03:22,830
of coverage criterion in industry. Normally for company that uses

66
00:03:22,830 --> 00:03:27,200
statement coverage, the typical coverage target is 80-90%, which

67
00:03:27,200 --> 00:03:29,690
mean the outcome of the test should be such

68
00:03:29,690 --> 00:03:32,770
that 80-90% of the statements are exercised at the

69
00:03:32,770 --> 00:03:34,880
end of testing. So at this point, you might

70
00:03:34,880 --> 00:03:36,850
be wondering, why don't we just shoot for 100%?

71
00:03:36,850 --> 00:03:38,580
Why don't we try to cover all of the

72
00:03:38,580 --> 00:03:40,050
code? We just saw that we could do it.

73
00:03:40,050 --> 00:03:41,730
And so I'm going to ask you the same question.