Add section on attrgetter and itemgetter

Author: gjbex

source-code/functional-programming/functional_programming_style.ipynb

@@ -9,12 +9,14 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 5,
+   "execution_count": 30,
    "metadata": {},
    "outputs": [],
    "source": [
     "from collections import Counter\n",
+    "from dataclasses import dataclass\n",
     "from functools import reduce, partial\n",
+    "from operator import attrgetter, itemgetter\n",
     "import random"
    ]
   },
@@ -225,6 +227,114 @@
     "    counter[x] += 1"
    ]
   },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Convenience functions"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "When using the `map` function, a common tasks is to do a computation involving a specific object attribute.  Consider the data class `Person`.  You would like to compute the average age of people in a list."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 28,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "@dataclass\n",
+    "class Person:\n",
+    "    firstname: str\n",
+    "    lastname: str\n",
+    "    age: int"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 29,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "people = [\n",
+    "    Person('Alice', 'Wonderland', 12),\n",
+    "    Person('Bob', 'Ludlum', 68),\n",
+    "    Person('Carol', 'Christmas', 38),\n",
+    "]"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "For each person in the list, the age has to be extracted.  This can easily be done using a lambda function."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 31,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "39.333333333333336"
+      ]
+     },
+     "execution_count": 31,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "sum(map(lambda p: p.age, people))/len(people)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "However, this is such a common pattern that the Python standard library has a function for it.  This function also has the advantage that the attribute to be extracted is specified as a string, and hcnce can be computed."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 32,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "39.333333333333336"
+      ]
+     },
+     "execution_count": 32,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "sum(map(attrgetter('age'), people))/len(people)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Similar to `attrgetter`, the function `itemgetter` can be used to extract an element of a tuple or a list by index."
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Note that `attrgetter` and `itemgetter` return a callable, so they are examples of higher-order functions."
+   ]
+  },
   {
    "cell_type": "markdown",
    "metadata": {},
@@ -497,7 +607,8 @@
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
    "version": "3.9.5"
-  }
+  },
+  "toc-autonumbering": true
  },
  "nbformat": 4,
  "nbformat_minor": 4