Testing and Debugging

A calculation can return a plausible number and still be wrong. Tests compare the result with a stated rule. Debugging then locates the cause of a failed comparison. This lesson checks a return function with ordinary, boundary and invalid inputs. It also explains why numerical comparisons need a tolerance and why a fix should be recorded with its test in Git.

Learning objectives. By the end of this note you should be able to select normal, boundary and invalid cases; write small assertions; and isolate a failure before changing code.