Class 9 - Notes

Upcoming Schedule

If you did not earn a Yellow Belt (by receiving a Gold Star on Project 1), see Yellow Belt Promotion for information on how to advance. There is no prescribed deadline on when you need to complete this, but additional requirements for advancing will be added for anyone who has not completed it by Sunday (14 Feburary).

Project 2 is posted now, and is due at the beginning of class on Monday, 15 February. Everyone should have at least completed Problem 3 of Project 2 by today. If you have not gotten that far, it is time (or past time!) to get cracking!

Dave will not be able to hold his usually schedule office hours tomorrow (Thursday morning); if you need to meet with me, send email (that includes a list of all the times you have available) to arrange a time. Yuchi has office hours today (4-5pm, Rice 514) and Friday (immediately after class).

Slides

[Download (PPTX)]

Lists

The Python standard library provides several useful datatypes for managing collections of data. The list type is a mutable, ordered collection of elements. Mutable means we can modify the value of the list after it is created (for example, but adding or deleting elements). Ordered means the elements are in a well-defined order, and we can access them by their index. The elements in a list can be any type, including (of course!) other lists.

Square brackets construct lists:

>>> p = [3, 4]
>>> p
[3, 4]

We can access elements of a list using list[index]: (indices start counting from 0)

>>> p[0]
3
>>> p[1]
4
>>> p[2]
IndexError: list index out of range

Growing Data

The simplest compound data structure is a pair. One way to implement a pair is to use the standard list type (but later we will also see that it is possible to build pairs without using any collection type!).

def make_pair(a, b):
    return [a, b]

def pair_first(pair):
    return pair[0]

def pair_last(pair):
    return pair[1]

How can we make a triple using a pair?

# A list is a pair where the second part is a list,
#          or None

def make_list(first, second): return make_pair(first, second)

def list_first(lst): return pair_first(lst)

def list_rest(lst):  return pair_last(lst)

# Any resemblence to actual politicians is purely pythonic.
democritoots = make_list('Berny', make_list('Hillarie', None))
republicrats = make_list('Thump', make_list('Kasitch', ..., make_list('Carsoon', None))))))))

Is it better to solve problems by thinking about what we need to do to solve the problem (procedures) or by thinking about what we need to represent to solve the problem (data)?