Functional Programming Facilities in Python

1
Functional Programming Facilities in Python

• Supplementary Lecture
• K.A.Hawick, April 2008

2

• !/usr/bin/python
• print "A Python Example showing lists and map"
• mylist 1, 2, 3, 4, 5
• print mylist
• def func(arg)
• return arg arg
• print "first in imperative style using
  'foreach'"
• for e in mylist
• print func(e)
• print "and again using functional style"
• print map( func, mylist )
• and now show off use of filter
• def test( arg )

• python interpreter is in /usr/bin (Unix path)
• denotes a comment
• python is dynamic so can make lists on the fly
• note the two space indents for eg function
  definitions
• foreach for in syntax
• Output
• A Python Example showing lists and map
• 1, 2, 3, 4, 5
• first in imperative style using 'foreach'
• 1
• 4
• 9
• 16
• 25
• and again using functional style
• 1, 4, 9, 16, 25
• 1, 4, 9

3

• functions are first class entities
• We can have lists of functions
• Note lambda function (a way to make an anonymous
  function)
• Output

• !/usr/bin/python
• def func1( arg )
• return arg 1
• def func2( arg )
• return arg 2
• def func3( arg )
• return arg arg arg
• funclist func1, func2, func3, lambda xxxxx
  
• list 1, 2
• for f in funclist
• print map( f, list )

2, 3 2, 4 1, 8 1, 16

4

• !/usr/bin/python
• print range(1,10,1) start, stop, step
• print range(4,0,-1)
• print reduce( lambda n,mnm, range(4,0,-1) )
• print reduce( lambda n,mnm, range(4,0,-1) )
• print "-------------------------------"
• mylist 1, 2, 3 4, 5 list
  concatenation
• mylist.sort()
• mylist.reverse()
• mylist.pop()
• mylist.append( 42)
• print mylist
• print mylist.index(3)

• range generates a numerical list
• Note it does not include the stop value
• reduce function does cumulative reduction
• ( ( (4 3) 2 ) 1 )
• Also various list methods available
• Output
• 1, 2, 3, 4, 5, 6, 7, 8, 9
• 4, 3, 2, 1
• 24
• 10
• -------------------------------
• 5, 4, 3, 2, 42
• 2
• 4

5

• Can map functions to functions
• Need the lambda expression here to make a
  function call from the identifier argument
• Output
• calling funcA
• calling funcB
• calling funcC

• !/usr/bin/python
• def funcA()
• print "calling funcA"
• def funcB()
• print "calling funcB"
• def funcC()
• print "calling funcC"
• map( lambda f f(), funcA, funcB, funcC )

6
So What is Python?

• python is (mostly) an imperative/OO language
• but it has some useful functional facilities
• hybrid languages that let you use the best of
  both worlds are likely to become common
• (incidentally, in python lists are mutable,
  tuples are almost like lists but are immutable)