Learn Functional Programming with Elixir

Learn Functional Programming with Elixir

Table of contents generated with markdown-toc

Running Code

Get Elixir version
$ elixir -v
Open Elixir's interactive shell
$ iex
(Press Ctrl + C twice to exit)
Execute script file
$ elixir hello_world.exs
File Extensions
.ex for compiled files
.exs for script files

Chapter 1 - Thinking Functionally

The rules that govern typical everyday programming are changing. That doesn't happen often. When it does, something important is going on.

Why Functional?

We need to write code that takes advantage of concurrency and parallelism.

The limitations of Imperative Languages

Imperative Languages have shared mutating values, which can be dangerous for concorrency and easily introduce bugs.

Instead of libraries that offer mechanisms to help lock and sync the changes, functional programming offers a better alternative.

Moving to Functional Programming

In FP, functions are the basic building blocks, values are immutable, and the code is declarative.

Working with Immutable Data

In FP, all values you create are immutable, no matter the operation we apply to it:

list = [1, 2, 3, 4]
# => [1, 2, 3, 4]

List.delete_at(list, -1)
# => [4]

list ++ [1]
# => [1, 2, 3, 4, 1]

IO.inspect list
# => [1, 2, 3, 4]

Building Programs with Functions

Functions are the primary tools for building a program
They are usually short and expressive
We combine multiple little functions to create a larger program
Values are passed explicitly between functions
Functions can be arguments and results

Pure functions

Values are immutable
The result is affected only by the function's arguments
No side effects beyond the value returned

add2 = fn (n) -> n + 2 end

add2.(2)
# => 4

Impure functions

More complex, unpredictable results and with side effects (Chapter 7)

Transforming Values

Elixir's focus is on the data transformation flow, and the pipe ( |> ) operator can be used to combine multiples functions' calls and results, where the result of each expression will be passed as a value to the next function.

title 
|> String.split
|> capitalize_all
|> join_with_whitespace

Declaring Code

Generates less and simpler code
Less code = fewer bugs
what's necessary to solve a problem instead of how to solve a problem( imperative)

Chapter 2 - Working with Variables and Functions

Values are anything that can represent data: Numbers, Strings, Fuctions, Maps and etc..
Operators compute values and generate a result: +, *, -...
Literals represent values that humans can easily understand.
Elixir generates result for any expression:

iex> a = 2
2

It's not possible to add text and a number:

iex> "2" + 2
(ArithmeticError) ...

Common Types:

Type	Useful for	Examples
string	Text	"Hello, World"
integer	Numbers	10 20 -2
float	Real numbers	10.8 -1.45
boolean	Logical operators	true, false
atom	Identifiers	:ok, :error, :exit
tuple	Building collections of known sizes	{:ok, "Hello"} {1,2,3}
list	Building collections of unknown sizes	[1,2] ["a", "b"]
map	Looking up a value in a dictionary by key	%{id: 123, name: "Anna"} %{12 => "User"}
nil	NULL	nil

Common Operators

Operator	Useful for	Examples
+	Adding numbers	1 + 2, 1.5 + 2
-	Substracting numbers	1-2, 1.5-2
/	Dividing numbers	10/2
*	Multipling numbers	10*2.5
==	Checking if equal	1 == 1.0
!=	Checking if not equal	"1"!=1
<	Less than	1<2
>	Greater than	2>1
++	Concatenating two lists	[1,2] ++ [3,4]
<>	Concatenating tow strings or binaries	"Hello, " <> "World!"
"#{}"	String interpolation	"Hello, #{name}!"

Creating Logical Expressions

and, or, not are made to work with Boolean values,

iex> true and true
true

iex> true or false
true

iex> not true
false

iex> 1 and true
(BadBooleanError)

iex> true and 1
true
(right side part won't be computed)

&&, ||, ! accept falsy(nil and false) or truthy(anything else) values and return a value based on the operator we use

iex> nil && 1
nil

iex> true && "Hello, World!"
"Hello, World!"

iex> nil || 1
1

iex> !true
false

Variables

Explicit names
snake_case format

iex> quantity = 10
10

iex> product_price = 11.5
11.5

iex> total_cost = product_price * quantity
115.0

Anonymous Functions

Functions receive input, computes, and outputs the last expression
Good practice to keep param number below 5
Must be bound to a variable to be reused
Also known as lambdas
func = fn param_1, p_2 -> body end
func.("param", 2)
func_2 = fn -> body end
func_2.()
func_3 = &(&1 * &2)
func_3.(10, 2)
func_4 = & &1 * &2
func_4.(10, 2)

iex> hello = fn name -> "Hello, " <> name <> "!" end
iex> hello.("Ana")
"Hello, Ana!"

Functions as First-Class Citizens

Functions are values of type function

iex> flat_fee = fn _ -> 5 end
iex> full_price = fn price, fee -> price + fee.(price) end
iex> full_price.(10, flat_fee)
15

Sharing Values Without Using Arguments

Use closures, as they have access to var values both inside and outside of the code block
Closures "remember" all the free vars from when they were created
Scope is a part of a program (code block)
Lexical scope is related to the visibility of the vars where they were defined

iex> answer = 42
iex> make_answer = fn -> other_answer = 88 + answer end 
iex> make_answer.()
50
iex> other_answer
*** (Error)
iex> answer = 0
iex> make_answer.()
50

Naming Functions

named functions are defined inside modules
atom or aliases can be used to name a function
Alias is any word that starts with a capital letter ( String )
During compile time, aliases become atoms (String == :"Elixir.String" )
To call a function inside a module, use Module.function()
It's possible to omit the parenthesis when calling a function ( IO.puts "Hello" )

Elixir's Named Functions

Full Docs
Kernel module doesn't need the module name when called div(1,2)
All other modules do String.capitalize("HellO THerE")

Creating Modules and Functions

use snake_case.ex for module file
use lib/namespace/module.ex for module path
use CamelCase, Namespace.Module for module name
only one module per file
use snake_case for function name
one line functions can be defined as def func_name(), do: expression

defmodule Ecommerce.Checkout do
    def total_cost(price, tax_rate) do
        price * (tax_rate + 1)
    end
    def total_cost_single(price, tax_rate), do: price * tax_rate + 1
end

to compile and use a file inside iex, use:

iex> c("lib/ecommerce/checkout.ex")
iex> Ecommerce.Checkout.total_cost(100, 0.2)
120.0

Importing Named Functions

modules attributes can be defined with @ and are used as annotations, temporary storage, or constants @file_name "task_list.md"
use import directive to import modules' functions import File, only: [write: 3, read: 1]
import directive should only be used when function names can't cause confusion

Using Named Functions as Values

use the &function/arity operator to bind a named function to a variable

iex> upcase = &String.upcase/1
iex> upcase.("oi")
OI

Chapter 3 - Using Pattern Matching to Control the Program Flow

Elixir's pattern match shapes everything you program.

Making Two things Match

iex> 1 = 1
1

iex> 2 = 1
** (MatchError)

iex> x = 1 (bind and match)
1
iex> 1 = x
1
iex> 2 = x
** (MatchError)
iex> x = 2 (rebind and match)
2
iex> ^x = 2 (pin operator avoids rebind and uses variable value)
2
iex> ^x = 1 (avoids rebind and uses variable value)
** (MatchError)

You can use the _ wildcard to ignore parts of a Pattern Matching

Unpacking Values from Various Data Types

Pattern matching is useful for extracting parts of values to variables in a process called destructing.

Matching Parts of a String

iex> "Authentication: " <> credentials = "Authentication: Basic dxYz123"
"Authentication: Basic dxYz123"

iex> credentials
"Basic dxYz123"

It's not possible to use a variable on the left side of the <> operator

iex> first_name <> "Doe" = "John Doe"
** (CompileError)

Matching Tuples

Tuples are often used to pass a signal with values

We can match and bind multiple items of a tuple

iex> {a, b, c} = {4, 5, 6}
{4, 5, 6}

iex> b
5

Tuples are useful for signaling successess and failures in a function's return

iex> my_function = fn -> {:ok, 42} end
iex> {:ok, answer} = my_function.()
iex> answer
42
iex> {:error, reason} = my_function.()
** (MatchError)

Matching Lists

In Elixir, lists are linked lists

iex> [a, a, a] = [1, 1, 1]
[1, 1, 1]
iex> [a, a, a] = [1, 2, 1]
** (MatchError)
iex> [a, b, a] = [1, 2, 1]
[1, 2, 1]
iex> [a, b, 1] = [1, 2, 1]
[1, 2, 1]
iex> [_, b, _] = [1, 2, 1]
[1, 2, 1]

-Using the | operator to split arrays

iex> [head | tail] = [1,2,3,4]
iex> head
1
iex> tail
[2,3,4]

iex> [a, b | tail] = [1,2,3,4]
iex> a
1
iex> b
1
iex> tail
[3,4]

iex> [head | tail] = [1]
iex> head
1
iex> tail
[]

iex> [head | tail] = []
** MatchError

Matching Maps

Maps are data types structured in key/value pairs

iex> user_signup = %{email: "johndoe@email.com", password: "1234"}

The key is a atom, if you need to use other values as key you should use =>

iex> sales = %{"2017/01"=> 200, "2017/02"=> 250}

It's possible to create nested structures

%{
  name: "John Doe",
  programming_languages: ["Ruby", "Python", "Java"],
  location: %{city: "New York", country:"US"}
}

Pattern Match and bind to variable

iex> abilities = %{strength: 16, intelligence: 10}

iex> %{strength: s_value} = abilities
iex> s_value
16

iex> %{wisdom: w_value} = abilities
** MatchError

iex> %{strength: 16, intelligence: i_value} = abilities
iex> i_value
10

iex> %{strength: s_value = 16} = abilities (check and bind)
iex> s_value
16

iex> strength_value = 16
iex> %{strength: ^strength_value} = abilities (use value of ^strength_value)

Maps vs. Keyword Lists

Keyword List is a list of two-element tuples
It allows duplicate dkeys
Keys must be atoms

iex> [b, c] = [a: 1, a: 12]
iex> b
{:a, 1}

x = %{a: 1, a:12} # result in {a: 12}
x = [a: 1, a: 12] # OK
x = [{:a, 1}, {:a, 12}] # same from above
x = %{1=> :a, 2 => :b} # OK
x = [1 => :a, 2 => :b] # Syntax error

The syntax is similar, but their limitations make them handy for different cases

Matching Structs

Structs are extensions of mapping structures
Useful for representing consistent structures
Similar to a class with props
Pattern Matching works like it does with Maps
~D is a date sigil
The name of the struct can be used to match only the struct

iex> date = ~D[2018-01-01]
iex> %{year: year} = date
iex> year
2018

iex> date = ~D[2018-01-01]
iex> %Date{year: year} = date
iex> year
2018

iex> date = %{year: 2018}
iex> %Date{year: year} = date
** MatchError

Sigils

Sigils are shortcuts to create values with a simplified text representation

iex> ~D[2018-01-01]
~D[2018-01-01]

iex> ~w(chocoalte jelly mint)
["chocolate", "jelly", "mint"]

Control Flow with Functions

Pattern Matching and Functions are the fundamental tools we use to control the program flow see number_compare.ex
Arguments from functions are pattern-matching expressions
To define a private function inside a module, use defp

Applying Default Values for Functions

Use \\ to apply a default value to a named function
Elixir will create multiple functions automatically on compile time
In Elixir, functions have fixed arity
Arity is part of the function's unique name

defmodule Checkout do
    # &total_cost/1 and &total_cost/2 are created
    def total_cost(price, quantity \\ 10), do: price * quantity
end

iex> Checkout.total_cost(12)
120
iex> Checkout.total_cost(12, 5)
60

Expanding Control with Guard Clauses

They permit us to add Boolean expression in our functions
Help create better function signatures
Can be created by using the when keyword after params
Possible to use guards on named and annonymous functions

def greater(number, other_number) when number >= other_number, do: number
def greater(_, other_number), do: other_number

var_greater = fn
    number, other_number when number >= other_number -> number
    _, other_number -> other_number
end

Macro functions can be easily created with the defguard directive
When macro is used, we need to use the directive require

Elixir Control-Flow Structures

Function clauses are usually used to control the flow of the program
Elixir built-in control-flow structures like case, cond, if and unless can also be used

Case: Control with Pattern Matching

Useful when we want to check an expression with multiple pattern matching clauses
Useful with functions that may have unexpected effect
It's possible to use guards on the clauses
When a clause is matched, it'll execute the code and return the last expression
if no clause is matched, an error is raised

case Boolean expression do
    {:ok, value} -> value
    :error -> IO.puts("Error")
end

Cond: Control with Logical Expressions

Useful we want to check variables in logical expressions
Structure similar to case
When a condition is truthy, execute related code
if no condition is truthy, raise error

result = cond do
    age < 12 -> "kid"
    age <= 18 -> "Teen"

If: Taking a look at our old friend

Useful when you want to execute a command based on a truthy expression
unless can be used as if not
else block is optional, if not given, return nil

def greater_if(number, other_number) do
    if number >= other_number do
        number
    else
        other_number
    end
end

def greater_unless(number, other_number) do
    unless number < other_number do
        number
    else
        other_number
    end
end

result = if(number >= other_number, do: number, else: other_number)

Chapter 4 - Diving into Recursion

While we use for and while loops on imperative languages, we have recursion in FP

Surrounded by Boundaries

A bounded recursion have an end
Most common type
Number of iterations is directly associated with the arguments
Good practice: Declare the boundary clause first

defmodule Sum do
    def up_to(0), do: 0
    def up_to(n), do: n + up_to(n -1)
end

iex> Sum.up_to(3) #will call the function n+1 times
6

Navigating Through Lists

Use syntax [head | tail] to navigate through lists using recursion
Can iterate through any data structure

defmodule Math do
    def sum([]), do: 0
    def sum([head | tail]), do: head + sum(tail)
end

iex> Math.sum([5,2,3])
10

iex> Math.sum([])
0

Transforming Lists

Transforming lists requires repetitive steps
[head | tail] is useful for destructuring and constructing new lists
[|] to prepend is faster than ++

iex> [:a | [:b, :c]]
[:a, :b, :c]

iex> [[:b, :c] | :a ]
[:b, :c, :a]

iex> [:a, :b, :c]
[:a, :b, :c]

The Key-based Accessors

iex> item = %{a: 1, b: 2}
iex> item[:a]
1
iex> item.a
1
iex> item[:c]
nil
iex> item.c
** KeyError

Conquering Recursion

Decrease and Conquer

Reduce a problem to its simplest form and start solving it incrementally
E.g: Factorial

defmodule Factorial do
    def of(0), do: 1
    def of(n) when n > 0, do: n * of(n-1)
end

Divide and Conquer

Separate the problem into two or more parts
Process independly and combine in the end
E.g: Sorting functions

defmodule Sort do
    def asc([]), do: []
    def asc([a]), do: [a]
    def asc(list) do
        half_size = div(Enum.count(list), 2)
        {list_a, list_b} = Enum.split(list, half_size)
        merge(
            asc(list_a),
            asc(list_b)
        )
    end

    defp merge([], list_b), do: list_b
    defp merge(list_a, []), do: list_a
    defp merge([head_a | tail_a], list_b = [head_b | _]) when head_a <= head_b do
        [head_a, merge(tail_a, list_b)]
    end
    defp merge(list_a = [head_a | _], [head_b | tail_b]) when head_a > head_b do
        [head_b, merge(list_a, tail_b)]
    end
end

Tail-Call Optimization

Compiler reduces functions in memory without allocating more memory
Last expression of a function must be a function call
Common approach is to replace the use of the function result with an extra argument
Extra argument accumulates the results of each iteration
body-recursive functions are simpler to read and maintain, so they should be used when a low number of recursion is expected
tail-recursive functions should be used when a large number of iterations is expected

Compare:

defmodule Factorial do
    def of(0), do: 1
    def of(n) when n > 0, do: n * of(n -1) # will create stack
end

defmodule TailCallFactorial do
    def of(n), do: fac_of(n, 1)
    defp fac_of(0, acc), do: acc
    defp fac_of(n, acc) when > 0, do: fac_of(n - 1, n * acc) # will use tail-recursive optmization
end

Functions without borders

Unbounded recursion is when we can't predict the number of repetition for a recursive function
For example: web crawler, directory navigator

Making functions more predictable

Adding Boundaries
Avoiding Infinite Loops

Using Recursion with Anonymous Functions

Recursion with anonymous functions isn't straightforward and should avoided
It's necessary to wrap the function in another function to avoid nonexistent error

iex> fact_gen = fn me ->
    fn
        0 -> 1
        x when x > 0 -> x * me.(me).(x - 1) 
    end
end

iex> factorial = fact_gen.(fact_gen)
iex> factorial.(5)

Chapter 5 - Using Higher-Order Functions

Higher-Order functions are those that take functions as arguments and/or return them.
Useful for hiding complexity and laborius functions
Receive functions as arguments as your receive any other arg def each([], function), do: nil
Use functions as you use function values function.()
It's possible to send annonymous functions masMyList.each([], fn item -> IO.puts item end)
Value functions MyList.each([], my_function)
Named functions MyList.each([], &String.capitalize/1)
Operators as named functions MyList.reduce([], 0, &+/2)
Operators as named functions Enum.sort([], &<=/2)

Using the Enum Module

Works with maps, lists, tuples and any type that implements the Enumerable protocol
Enum.each
Enum.map
Enum.reduce
Enum.filter
Enum.count
Enum.sort
Enum.sum
Enum.uniq
Enum.sort
Enum.join

Using Comprehensions

using for, enumerables can be easily iterated, mapped and filtered

iex> for a <- ["dogs", "cats", "flowers"], do: String.upcase(a)
["DOGS", "CATS", "FLOWERS"]

each item of the list will be assigned to a
it's possible to have more than one generator

iex> for a <- ["Willy", "Anna"], b <- ["Math", "English"], do: {a, b}
[{"Willy", "Math"}, {"Willy", "English"}, {"Anna", "Math"}, {"Anna", "English"}]

we can filter by using pattern matching

iex> parseds = for i <- ["10", "hot dogs", "20"], do: Integer.parse(i)
[{10, ""}, :error, {20, ""}]
iex> for {n, _} <- parseds, do: n
[10, 20]

we can also filter with an expression for truthy values

iex> for n <- [1,2,3,4,5], n > 3, do: n
[4,5]

Pipelining Your Functions

Used to execute many functions in sequence
Elixir doesn't have a built-in compose function like other functional languages, the alternative is to use pipes

iex> first_letter_and_upcase = &(&1 |> String.first |> String.upcase)
iex> first_letter_and_upcase.("works")
"W"

The result of each pipe expression will be passed as the first argument of the next pipe
The result of the pipeline is the result of the last pipe
Pipes should be used for 2+ expressions for better readability

Be Lazy

Lazy operations provide alternative techniques of programming and creating efficient programs

Delay the Function Call

Other FP languages have currying, Elixir has partial application
Partial application is used to postopone a function's execution
The function is wrapped in another function with a fixed value of any of the arguments
It's common to use the function-capturing-syntax

defmodule WordBuilder do
    def build(alphabet, positions) do
        letters = Enum.map(positions, &(String.at(alphabet),&1))
        Enum.join(letters)
    end
end

Working with the Infinite

In programming, infinite is something that's always expanding
E.g. Web Server waiting for calls, messaging broker handling events
streams type represents a flow of data that may not have an end
Stream module contains many functions to create and operate streams

Range

Range literal is a stream iex> range = 1..10
It's a lazy collection, which are evaluated only when necessary iex> Enum.each(range, &IO.puts/1)

Stream

We can represent a collection that is always expanding using Stream.iterate/2
The items of the stream will be evaluated dynamically, every time something asks for an item
Needs a starting value and an increment function

    iex> i = Stream.iterate(1, fn prev -> prev + 1 end)`
    iex> Enum.take(i, 5)
    [1,2,3,4,5]

defmodule Factorial do
    def of(0), do: 1
    def of(n) when n > 0 do
        Stream.iterate(1, &(&1+1))
        Enum.take(n)
        Enum.reduce(&(&1*&2))
    end
end

Stream.cycle lets you create a list that will keep iterating through its items forever

defmodule Halloween do
    def give_candy(kids) do
        ~w(chocolate jelly mint)
        |> Stream.cycle
        |> Enum.zip(kids)
    end
end

iex> Halloween.give_candy(~w(Raff Jorge Anna Elza))
[{"chocolate", "Raff"}, {"jelly", "Jorge"}, {"jelly", "Anna"}, {"chocolate", "Elza"} ....]

Pipelining Data Streams

Combine Elixir pipe and strems
Can be done in two ways: eager or lazy
Eager: each computation processes all items and send to next computation
Lazy: each computation processes a chunk of items and send partial results

Chapter 6 - Designing Your Elixir Applications

Starting your project with Mix

Command-line interface (CLI)
Provides essentials for developing any Elixir application
Helps creates and maintain projects (compile, debug, test, manage dependencies)
built-in Elixir

Running the new Task

mix new creates the initial structure to code your application

mix new project_name

mix test will run all the tests on test folder
iex -S mix innitiates IEx with project compiled

Create the Start Task

Mix tasks are mix commands like mix new and mix test
It's possible to create our own tasks
Must be inside lib/mix/tasks

Designing Entities with Structs

Structs are used to express our application domains
We should define them on lib\namespace\module.ex
Module name should be Namespace.Module
Define struct using defstruct directive
Use attribute_name: default_value for each prop

Alias

alias directive can be used to create module shortcuts alias Mix.Shell.IO, as: Shell

Using Protocols to Create Polymorphic Functions

protocol lets you create an interface that many data types can implement
interfaces leads to better codebase design
define protocol with defprotocol

defprotocol Protocol do
    def function(param)
end

implement with defimpl

defimpl Protocol, for: ModuleThatImplements do
    def function(param), do: param
end

if inside module, for is not needed

def Module do
    defimpl Protocol do
        def fuction(param), do: param
    end
end

Conventions

if you own the struct, put the implementation in the same file as the struct
if you own the protocol and not the struct, put the implementation with the protocol
if you don't own anything, create a file with the protocol name and put the implementation there

Creating Module Behaviours

behavior is a contract between a module and the client code
Mix.Task is a behaviour
@callback directive to define a function rule

    @callback run(param1 :: any, param2 :: any) :: any

@behaviour directive to say module follows a behaviour

    @behaviour Namespace.CallbackModule

Adding Type Specifications

Notations that say what your functions expect and return
Good for static check and documentation
use @type t directive inside the module

@type t :: %ModuleStruct{
    prop: Type.t
}

Use Dialyzer to statically analyze the code

Chapter 7 - Handling Impure Functions

Impure Functions are those that can return different values from the same input.

The main strategy for creating a healthy codebase is to identify and isolate the parts that can have unexpected results, and make them predictable.

We'll discuss 4 strategies for this:

Conditional structures: case, if, etc...
Exception handling: try
Error monads
Elixir's with: pattern maching + conditional

Pure vs Impure Functions

You shouldn't think pure is good and impure is evil, both are important to write an application
It's good practice to build more pure functions and isolating impure parts with proper handling

Pure

Always return consistent output when the same input is given, and no side effects are produced
They can produce errors, but those are predictable

Impure

May not return consistent output with the same input
May produce side effects
References values that aren't in the function's args
Interact with content outside of the program context (File IO, database, API, logs, user input, etc...)
If a functions calls an impure function, it becomes impure

Trying, Rescuing, and Catching

Throwing values and/or raising errors is unusual in FP
Names from functions that raise errors end with !

Try, Raise and Rescue

use defexception to define a exception struct
use try rescue to catch exceptions
throw exceptions with raise Exception

def checkout do
    try do
        {qty, _} = ask_number("Quantity?")
        {price, _} = ask_number("Price?")
        qty * price
    rescue
        MatchError -> "It's not a number"
    end
end

Try, Throw and Catch

Similar to raise and rescue
Doesn't mean error, it's possible to throw variables

throw {:error, "Invalid option"}

def ask_for_option(options) do
    try do 
        options
        |> display_options 
        |> parse_answer 
    catch {:error, message} ->
        display_error(message)
    end
end

Handling Impure Functions with the Error Monad

Error Monad helps combining functions that can result in an error
Clear sequence
Error handling at a unique point
A monad wraps a value with properties that give more information about the value (context)
Makes it possible to combine functions with values to make automatic decisions
Are not Elixir native, you must download a package from the internet

Using with

Makes it possible to combine multiple matching clauses
Should be used when there's a function pipeline that can result in an error

def checkout() do
    result = 
        with {qty, _} <- ask_number("Qty?")
             {price, _} <- ask_number("Price?"),
            do: qty * price
    
    if result == :error, do: IO.puts("It's not a number"), else: result
end

Name		Name	Last commit message	Last commit date
Latest commit History 29 Commits
chapter_02		chapter_02
chapter_03		chapter_03
chapter_04/lib		chapter_04/lib
chapter_05/lib		chapter_05/lib
chapter_06/dungeon_crawl		chapter_06/dungeon_crawl
chapter_07/dungeon_crawl		chapter_07/dungeon_crawl
.gitignore		.gitignore
README.md		README.md

yangknet/LearnFunctionalProgrammingWithElixir-2

Folders and files

Latest commit

History

Repository files navigation