Introduction to Python

davidbpython.com

Set Operations and List Comprehensions

container processing: set comparisons

Set comparisons make it easy to compare 2 sets for membership.

set_a = {1, 2, 3, 4}
set_b = {3, 4, 5, 6}

print(set_a.union(set_b))           # {1, 2, 3, 4, 5, 6}  (set_a + set_b)

print(set_a.difference(set_b))      # {1, 2}              (set_a - set_b)

print(set_a.intersection(set_b))    # {3, 4}     (what is common between them?)

• set comparisons aren't essential, as we could accomplish them with a simple algorithm using 'for' loops
• however, set comparisons are very convenient to use
• they are also more efficient to run

Ex. 9.15 - 9.17

"transforming" list comprehension

List comprehensions build a new list based on an existing list.

This list comprehension doubles each value in the nums list

nums = [1, 2, 3, 4, 5]

dblnums = [      val * 2      for val in nums     ]
   #            transform       'for' loop

print(dblnums)            # [2, 4, 6, 8, 10]

• for val in nums is the 'for' loop: this specifies what we are looping through, and the name of the control variable
• val * 2 is the transform: each item will be passed through this change in value and passed to the target list (dblnums).
• (Note that the spacing in the example above is not required but added for clarity and commenting.)

Ex. 9.18

"filtering" list comprehension

List comprehensions can also select values to place in the new list.

This list comprehension selects only those values above 35 degrees Celsius:

daily_temps = [26.1, 31.0, 38.4, 36.1, 38.3, 34.1, 32.7, 33.3]

hitemps = [       t          for t in daily_temps        if t > 35     ]
    #          transform          'for' loop               filter

print(hitemps)          # [37.4, 36.1, 38.3]

• the first t in t for t is the transform: in this case, the value will not be changed
• the if expression is the filter: only value that pass this test will be included in the resulting list

Ex. 9.19

combining a transforming with a filtering list comprehension

We can choose to filter or transform or both.

This list comprehension selects values above 35C and converts them to Fahrenheit:

daily_temps = [26.1, 31.0, 38.4, 36.1, 38.3, 34.1, 32.7, 33.3]

f_hitemps = [ round((t * 9/5) + 32, 1)     for t in daily_temps     if t > 35 ]
     #              transform                'for' loop               filter

print(f_hitemps)          # [37, 36, 37]

Ex. 9.19

list comprehensions: examples

List comprehensions are a powerful convenience, but not ever required.

Some common operations can be accomplished in a single line. In this example, we produce a list of lines from a file, stripped of whitespace.

stripped_lines = [ i.rstrip() for i in open('pyku.txt').readlines() ]

We can even combine expressions for some fancy footwork

totals = [  float(i.split(',')[2])
            for i in open('revenue.csv')
            if i.split(',')[1] == 'NY'    ]

This last example borders on the overcomplicated -- if we are trying to do too much with a list comprehension, we might be better off with a conventional 'for' loop.

list comprehensions: why?

A list comprehension is a single statement.

• since a list comprehension is one statement, it can go in places where a 'for' loop algorithm cannot
• it can be used as a return value
• it can be assigned to a variable
• it can be the value in a dictionary
• it can be made the argument to a function

sidebar: list comprehensions with dictionaries

Since dicts can be converted to and from 2-item tuples, we can manipulate them using list comprehensions.

Recall that dict .items() returns a list of 2-item tuples, and that the dict() constructor uses the same 2-item tuples to build a dict.

mydict =  {'a': 5, 'b': 1, 'c': -3}

# dict -> list of tuples
my_items = list(mydict.items())      # list, [('a', 5), ('b', 1), ('c', -3)]

# list of tuples -> dict
mydict2 = dict(my_items)       # dict, {'a':5,   'b':1,   'c':-3}

Here's an example: filtering a dictionary by value - accepting only those pairs whose value is larger than 0:

mydict = {'a': 5, 'b': 1, 'c': -3}

filtered_dict = dict([ (i, j)
                       for (i, j) in mydict.items()
                       if j > 0 ])

           # {'a': 5, 'b': 1}