We encourage you to adopt an interactive style of debugging and programming. Use the Python shell a lot. Write short pieces of code and check them.  It is much  simpler to test the individual parts as you go, rather than  to take hours writing a large program, and then search it does not work,  and must sift through all your program, trying  to search the bugs.

But, if you ?nd  yourself in the (inevitable) position of having a big program with  a error in it, do not despair. Debugging a program does not require brilliance or creativity or much  in the way of inside.  What it needs is a systematic and persistence approach.

First of all, have  a test case (a set of inputs to the procedure you are trying  to debug) and  know what  the answer is considered to be.  To check a program, you might  start  with  some  special  cases: what  if the argument is 0 or the empty function? Those cases may  be simpler to sort through ?rst (and are also cases that can be easy to get worse). Then try more specific cases.

Now,  if your  program gets your  test case incorrect, what  could you do?  Resist the temptation to start  changing your  program around, just to watch if that  will ?x the solution. Do not modify  any code until  you know  what  is wrong with  what  you are doing  now, and therefore seems that the modify  you make is going to correct the problem.

Ultimately, for debugging large programs, it is most useful  to use a software development environment  with  a serious debugger. But these  methods may sometimes have  a steep  explaining curve,  so in this class we will study to debug systematically using  "print" statements.