ADVANCED PYTHON CONCEPTS

What we'll cover

List Comprehensions and Generator Expressions
Generators and Iterators
Context Managers (with)
Decorators
Magic Methods and Attributes
Functional Style
Metaprogramming

other stuff

There will be optional assignments for the end of each segments. I recommend that you do them however, they'll really help solidify the things we cover.
If you don't understand something I've said, feel free to ask away.
I'll upload model answers for the previous lecture after each lecture. If that sentence even makes sense.
It might be a good idea to use something like project euler to exercise any skills you learn a little more.
Sometimes code on the slides might not be copied verbatim. But it does all work. As far as I know. Please tell me if it doesn't.

List comprehensions

Syntactic sugar for building lists
Build and modify lists quickly
Can act like map & filter rolled into one

General form:

 [item for item in iterable]

FOR those not familiar with map and filter

 >>> my_list = [1, 2, 3] >>> def add_one(n): ...     return n + 1 >>> map(add_one, my_list) [2, 3, 4]
 >>> def is_even(n):         return True if n%2 == 0 else False >>> filter(is_even, my_list) [2]
 >>> filter(is_even, map(add_one, my_list)) [2, 4]

also, a little aside

I'm going to be using lambdas quite a lot for simple functions. They generally look like this:

 >>> f = lambda arguments: returnvalue

Left side of colon is comma-delimited arguments, right side is an expression that is evaluated and then returned.

I can rewrite the add_one function from the last page in this form:

 >>> add_one = lambda n: n + 1 >>> add_one(1) 2

more lambdas

And remember you don't actually need to assign them to anything either. they're just a quick way of creating a function object.

 >>> (lambda n: n*2)(4) 8
 >>> map(lambda n: n%2, [1,2,3,4]) [1,0,1,0]

anyway, back to list comprehensions

The most basic use is generating flat lists

 >>> [i for i in range(5)] [0, 1, 2, 3, 4]

But we can also apply some operation to the items of a list

 >>> [i**2 for i in range(5)] [0, 1, 4, 9, 16]

Or even remove ones we don't want

 >>> [i for i in range(5) if i%2==0] [0, 2, 4]

And any combination of these

 >>> [i**2 for i in range(5) if i%2==0] [0, 4, 16]

here's something a bit stranger

What will this return?

 >>> [(i,j) for i in (0,1) for j in (0,1)]

DICTCOMPS are pretty similar

 >>> square_lookups = {thing:thing**2 for thing in range(3)} {0:0, 1:1, 2:4}
 >>> square_lookups[2] 4

And the same for sets too

 >>> [i*j for i in range(3) for j in range(3)]
 [0, 0, 0, 0, 1, 2, 0, 2, 4]
 >>> {i*j for i in range(3) for j in range(3)}
 set([0, 1, 2, 4])

pitfall: tuples

This:

 >>> tuple(i for i in range(3)) (0, 1, 2)

Not this:

 >>> (i for i in range(3)) <generator object <genexpr> at 0x...>