Real World Instrumentation with Python

Functional Testing

We now have some requirements and hopefully a design document, but before the first line of code is written we need to think about testing. Here I am referring to functional testing, as opposed to unit testing; they are very different things, with different objectives and methods. We’ll get to unit testing in just a bit, but for now, let’s look at functional testing and see why it’s an essential (and arguably the only) method for verifying compliance with the requirements.

Testing to the Requirements

A functional test, as the name implies, tests the functionality of a system against the requirements that defined the system in the first place. Creating functional test cases from the requirements is actually very straightforward. A well-written requirement should contain all the information necessary for the test case. For example, assume we have a requirement that reads:

The data acquisition subsystem software shall be capable of acquiring and converting analog data as 16-bit signed data samples at a sustained rate of 1,000 samples per second.

From this we can see that we’ll need an input that will result in the system generating values between –32,768 and 32,767. We will also need to be able to count the samples generated in a one-second window. If the software has the ability to determine its actual sample rate in real time and display it (or at least record it), we’re basically done. If it doesn’t, maybe this is something that should go into a derived requirement ...