{
 "metadata": {
  "name": "list-comprehensions"
 },
 "nbformat": 3,
 "nbformat_minor": 0,
 "worksheets": [
  {
   "cells": [
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "List comprehensions\n",
      "===================\n",
      "\n",
      "Basics\n",
      "------\n",
      "\n",
      "List comprehensions are a Python feature that let us quickly write\n",
      "simple loops that and return the results to a list. This is a common\n",
      "pattern in programming, so it can save us a lot of keystrokes, but \n",
      "they are also often simpler to read for basic problems.\n",
      "\n",
      "For example, let's say we have a list of numbers and we need to do\n",
      "some math (let's square each value) on each of those numbers and store\n",
      "the results in a list. Using a for loop the answer would look something\n",
      "like this."
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "numbers = [2.2, 7.3, 1, 0, 9.5, 6.2, 42]\n",
      "results = []\n",
      "for value in numbers:\n",
      "    results.append(value ** 2)\n",
      "results"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "pyout",
       "prompt_number": 1,
       "text": [
        "[4.840000000000001, 53.29, 1, 0, 90.25, 38.440000000000005, 1764]"
       ]
      }
     ],
     "prompt_number": 1
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "Using a list comprehension we can right the middle 3 lines as a single line:"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "numbers = [2.2, 7.3, 1, 0, 9.5, 6.2, 42]\n",
      "results = [value ** 2 for value in numbers]\n",
      "results"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "pyout",
       "prompt_number": 2,
       "text": [
        "[4.840000000000001, 53.29, 1, 0, 90.25, 38.440000000000005, 1764]"
       ]
      }
     ],
     "prompt_number": 2
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "Nested loops\n",
      "------------\n",
      "List comprehensions can also handle nested for loops. So if we wanted to\n",
      "get the products of all of the possible combinations of numbers from 1 to 4\n",
      "we could write a list comprehension like this:"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "multiplication_table = [a * b for a in range(1, 5) for b in range(1, 5)]\n",
      "multiplication_table"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "pyout",
       "prompt_number": 3,
       "text": [
        "[1, 2, 3, 4, 2, 4, 6, 8, 3, 6, 9, 12, 4, 8, 12, 16]"
       ]
      }
     ],
     "prompt_number": 3
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "The loops can interact just like in a normal nested for loop, so if we\n",
      "only wanted to calculate each unique combination once (i.e., only calculate\n",
      "1 * 3, not also 3 * 1) we can modify the second for loop."
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "multiplication_table = [a * b for a in range(1, 5) for b in range(a, 5)]\n",
      "multiplication_table"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "pyout",
       "prompt_number": 4,
       "text": [
        "[1, 2, 3, 4, 4, 6, 8, 9, 12, 16]"
       ]
      }
     ],
     "prompt_number": 4
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "Conditionals\n",
      "------------\n",
      "They can also handle conditions. So, if in the first example we only wanted,\n",
      "to square the number if it was positive we could check that inside the list\n",
      "comprehension."
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "numbers = [2.2, 7.3, -5, -2.2, 1, 0, 9.5, -7, 6.2, 42]\n",
      "results = [value ** 2 for value in numbers if value > 0]\n",
      "results"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "pyout",
       "prompt_number": 5,
       "text": [
        "[4.840000000000001, 53.29, 1, 90.25, 38.440000000000005, 1764]"
       ]
      }
     ],
     "prompt_number": 5
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "Limits to readability\n",
      "---------------------\n",
      "Simple list comprehensions are easier to read than their equivalent for loops,\n",
      "but this is only true up to a point. A general [rule of thumb](http://python.net/~goodger/projects/pycon/2007/idiomatic/handout.html#list-comprehensions)\n",
      "is to use a regular for loop if you have more than total for loops and conditions,\n",
      "or the if the conditions are complex."
     ]
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "Want to learn more?\n",
      "-------------------\n",
      "Check out the [Python documentation on list comprehensions](http://docs.python.org/2/tutorial/datastructures.html#list-comprehensions)."
     ]
    }
   ],
   "metadata": {}
  }
 ]
}