In
42 Why We Need Model-Based Analysis
this test, we do not just check a result at the end, we check that the entire run ??“
the whole sequence of events and handler invocations ??“ is what we intend. It is not
clear how to use NUnit assertions to achieve this, so we code the test method to
query the tester (the human user). The tester must watch the test output and make
the judgment whether the test succeeded or failed, so this test is not fully automated.
Executing this test in nunit-console produces this output. The lines that begin
with a question mark ? are written by ReceiveEvent; they report the type of event
that occured, the type of event that the controller expected (the waitfor variable),
and whether the sensor is considered to be erroneous (the sensor variable). The
indented lines are written by the handler methods; they indicate which handler ran,
and the data the handler used. This test succeeds. The sensor status changes from
Error to OK after a valid message is received. After that, the controller carries out a
command to compute the calibration factor.
> nunit-console /include=Typical Tests.dll
...
Included categories: Brief
.?Timeout, Timeout, Error
Reset
?Message ??™99.9??™, Message, Error
CheckMessage ??™99.9??™, compare to 99.9, OK
?Timeout, Timeout, OK
Poll
?Message ??™100.0??™, Message, OK
CheckMessage ??™100.0??™, compare to 99.
Pages:
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80