Note: Some people find this section useful, and some find it confusing. If you find it confusing, feel free to skip it and continue with the next section.
To show more advanced ways of using functions, we'll now do a walk through for the following program:
defmult(a,b):ifb==0:return0rest=mult(a,b-1)value=a+restreturnvalueresult=mult(3,2)print("3 * 2 = ",result)
Basically this program creates a positive integer multiplication function(that is far slower than the built in multiplication function) and then demonstrates this function with a use of the function. This program demonstrates the use of recursion, that is a form of iteration (repetition) in which there is a function that repeatedly calls itself until an exit condition is satisfied. It uses repeated additions to give the same result as multiplication: e.g. 3 + 3 (addition) gives the same result as 3 * 2 (multiplication).
mult
is defined with the lines:defmult(a,b):ifb==0:return0rest=mult(a,b-1)value=a+restreturnvalue
result = mult(3, 2)
is run.mult(3, 2)
to the variableresult
.mult(3, 2)
return?mult
function to find out.a
gets the value 3 assigned to it and the variableb
gets the value 2 assigned to it.if b == 0:
is run. Sinceb
has the value 2 this is false so the linereturn 0
is skipped.rest = mult(a, b - 1)
is run. This line sets the local variablerest
to the value ofmult(a, b - 1)
. The value of a
is 3 and the value of b
is 2 so the function call ismult(3,1)
mult(3, 1)
? mult
with the parameters 3 and 1.a
has the value 3 and b
has the value 1. Since these are local values these do not affect the previous values ofa
andb
. b
has the value 1 the if statement is false, so the next line becomesrest = mult(a, b - 1)
.mult(3, 0)
to rest.a
has the value 3 andb
has the value 0.if b == 0:
. b
has the value 0 so the next line to run isreturn 0
return 0
do?mult(3, 0)
has the value 0. Now we know what the linerest = mult(a, b - 1)
did since we have run the functionmult
with the parameters 3 and 0. We have finished runningmult(3, 0)
and are now back to running mult(3, 1)
. The variablerest
gets assigned the value 0.value = a + rest
is run next. In this run of the function,a = 3
andrest = 0
so nowvalue = 3
.return value
is run. This returns 3 from the function. This also exits from the run of the function mult(3, 1)
. Afterreturn
is called, we go back to running mult(3, 2)
. mult(3, 2)
?a = 3
andb = 2
and were examining the linerest = mult(a, b - 1)
.rest
get 3 assigned to it. The next linevalue = a + rest
setsvalue
to3 + 3
or 6.result = mult(3, 2)
which can now assign the value 6 to the variableresult
.print("3 * 2 = ", result)
is run.3 * 2 =
and the value ofresult
which is 6. The complete line printed is3 * 2 = 6
.x * 0 = 0
). The second is that a number times another number is equal to the first number plus the first number times one less than the second number (x * y = x + x * (y - 1)
). So what happens is3 * 2
is first converted into 3 + 3 * 1
. Then3 * 1
is converted into3 + 3 * 0
. Then we know that any number times 0 is 0 so3 * 0
is 0. Then we can calculate that3 + 3 * 0
is3 + 0
which is3
. Now we know what3 * 1
is so we can calculate that3 + 3 * 1
is3 + 3
which is6
.This is how the whole thing works:
mult(3, 2)3 + mult(3, 1)3 + 3 + mult(3, 0)3 + 3 + 03 + 36
Programming constructs solving a problem by solving a smaller version of the same problem are calledrecursive. In the examples in this chapter, recursion is realized by defining a function calling itself. This facilitates implementing solutions to programming tasks as it may be sufficient to consider the next step of a problem instead of the whole problem at once. It is also useful as it allows to express some mathematical concepts with straightforward, easy to read code.
Any problem that can be solved with recursion could be re-implemented with loops. Using the latter usually results in better performance. However equivalent implementations using loops are usually harder to get done correctly.
Probably the most intuitive definition ofrecursion is:
Try walking through the factorial example if the multiplication example did not make sense.
factorial.py
#defines a function that calculates the factorialdeffactorial(n):ifn==0:return1ifn<0:return"Error, negative numbers do not have factorial values!!"returnn*factorial(n-1)print("2! =",factorial(2))print("3! =",factorial(3))print("4! =",factorial(4))print("5! =",factorial(5))print("-3! =",factorial(-3))
Output:
2! = 23! = 64! = 245! = 120-3! = Error, negative values do not have factorial values!!
countdown.py
defcount_down(n):print(n)ifn>0:returncount_down(n-1)count_down(5)
Output:
543210
Non-Programmer's Tutorial for Python 3 | ||
← Defining Functions | Advanced Functions Example | Lists → |