2016-04-21 23:52:12 +03:00
|
|
|
|
---
|
2023-04-27 23:09:38 +03:00
|
|
|
|
language: Elixir
|
2016-04-21 23:52:12 +03:00
|
|
|
|
contributors:
|
|
|
|
|
- ["Joao Marques", "http://github.com/mrshankly"]
|
|
|
|
|
- ["Dzianis Dashkevich", "https://github.com/dskecse"]
|
|
|
|
|
translators:
|
|
|
|
|
- ["Tai An Su", "https://github.com/taiansu"]
|
|
|
|
|
filename: learnelixir-tw.ex
|
|
|
|
|
lang: zh-tw
|
|
|
|
|
---
|
|
|
|
|
|
|
|
|
|
Elixir 是一門建構在 Erlang 虛擬機上的現代函數式語言。它完全與 Erlang 相容,但
|
|
|
|
|
採行了比較常見的語法,並提供更多的功能。
|
|
|
|
|
|
|
|
|
|
```elixir
|
|
|
|
|
# 單行註解以井字號開頭
|
|
|
|
|
|
|
|
|
|
# 沒有多行註解的功能
|
|
|
|
|
# 但你可以連續使用多個單行
|
|
|
|
|
|
2023-04-27 23:09:38 +03:00
|
|
|
|
# 用 `iex` 來進入 Elixir shell
|
2016-04-21 23:52:12 +03:00
|
|
|
|
# 用 `elixirc` 來編譯你的模組
|
|
|
|
|
|
2023-04-27 23:09:38 +03:00
|
|
|
|
# 如果你已成功安裝 Elixir 的話,這兩個命令應已在你的 path 下。
|
2016-04-21 23:52:12 +03:00
|
|
|
|
|
|
|
|
|
## ---------------------------
|
|
|
|
|
## -- 基本型別
|
|
|
|
|
## ---------------------------
|
|
|
|
|
|
|
|
|
|
# 數字
|
|
|
|
|
3 # 整數
|
|
|
|
|
0x1F # 整數
|
|
|
|
|
3.0 # 浮點數
|
|
|
|
|
|
|
|
|
|
# 原子 (Atoms) 是不可變的字面常數,以 `:` 開頭。
|
|
|
|
|
:hello # atom
|
|
|
|
|
|
|
|
|
|
# 元組(Tuples) 會存在記憶體連續的區段裡。
|
|
|
|
|
{1,2,3} # tuple
|
|
|
|
|
|
|
|
|
|
# 我們可以用 `elem` 函式來取得 tuple 中的元素。
|
|
|
|
|
elem({1, 2, 3}, 0) #=> 1
|
|
|
|
|
|
|
|
|
|
# 串列 (List) 是用連結串列實作的。
|
|
|
|
|
[1,2,3] # list
|
|
|
|
|
|
|
|
|
|
# 我們可以這樣取得串列的頭尾元素:
|
|
|
|
|
[head | tail] = [1,2,3]
|
|
|
|
|
head #=> 1
|
|
|
|
|
tail #=> [2,3]
|
|
|
|
|
|
2023-04-27 23:09:38 +03:00
|
|
|
|
# 在 Elixir 中,就如同 Erlang 裡一樣,`=` 代表的是模式比對,而非指派。
|
2016-04-21 23:52:12 +03:00
|
|
|
|
#
|
|
|
|
|
# 這代表將使用左手邊的模式 (pattern) 去與右手邊的值進行比對。
|
|
|
|
|
#
|
|
|
|
|
# 這也是先前取得串列的頭尾元素的運作原理
|
|
|
|
|
|
|
|
|
|
# 當模式比對無法找到合適的配對時,將會回報錯誤,如下例中兩個 tuple 的大小不一致。
|
|
|
|
|
# {a, b, c} = {1, 2} #=> ** (MatchError) no match of right hand side value: {1,2}
|
|
|
|
|
|
|
|
|
|
# 還有二進位的型別
|
|
|
|
|
<<1,2,3>> # binary
|
|
|
|
|
|
|
|
|
|
# 字串與字母串列
|
|
|
|
|
"hello" # string
|
|
|
|
|
'hello' # char list
|
|
|
|
|
|
|
|
|
|
# 多行字串
|
|
|
|
|
"""
|
|
|
|
|
I'm a multi-line
|
|
|
|
|
string.
|
|
|
|
|
"""
|
|
|
|
|
#=> "I'm a multi-line\nstring.\n"
|
|
|
|
|
|
|
|
|
|
# 字串皆使用 UTF-8 編碼
|
|
|
|
|
"héllò" #=> "héllò"
|
|
|
|
|
|
|
|
|
|
# 字串其實是以二進位實作,而字母串列就只是單純的串列。
|
|
|
|
|
<<?a, ?b, ?c>> #=> "abc"
|
|
|
|
|
[?a, ?b, ?c] #=> 'abc'
|
|
|
|
|
|
2023-04-27 23:09:38 +03:00
|
|
|
|
# `?a` 在 Elixir 中會回傳字母 `a` 的 ASCII 整數
|
2016-04-21 23:52:12 +03:00
|
|
|
|
?a #=> 97
|
|
|
|
|
|
|
|
|
|
# 用 `++` 來合併串列,而合併二進位則要用 `<>`
|
|
|
|
|
[1,2,3] ++ [4,5] #=> [1,2,3,4,5]
|
|
|
|
|
'hello ' ++ 'world' #=> 'hello world'
|
|
|
|
|
|
|
|
|
|
<<1,2,3>> <> <<4,5>> #=> <<1,2,3,4,5>>
|
|
|
|
|
"hello " <> "world" #=> "hello world"
|
|
|
|
|
|
|
|
|
|
# 範圍 (Ranges) 則是以 `開頭..結尾`來宣告 (頭尾都包含在內)
|
|
|
|
|
1..10 #=> 1..10
|
|
|
|
|
lower..upper = 1..10 # 可以對 range 進行模式比對
|
|
|
|
|
[lower, upper] #=> [1, 10]
|
|
|
|
|
|
|
|
|
|
## ---------------------------
|
|
|
|
|
## -- 運算元
|
|
|
|
|
## ---------------------------
|
|
|
|
|
|
|
|
|
|
# 簡單算數
|
|
|
|
|
1 + 1 #=> 2
|
|
|
|
|
10 - 5 #=> 5
|
|
|
|
|
5 * 2 #=> 10
|
|
|
|
|
10 / 2 #=> 5.0
|
|
|
|
|
|
2023-04-27 23:09:38 +03:00
|
|
|
|
# 在 Elixir 中, `/` 運算元永遠回傳浮點數。
|
2016-04-21 23:52:12 +03:00
|
|
|
|
|
|
|
|
|
# 若需要回傳整數的除法,用 `div`
|
|
|
|
|
div(10, 2) #=> 5
|
|
|
|
|
|
|
|
|
|
# 要得到除法的餘數時,用 `rem`
|
|
|
|
|
rem(10, 3) #=> 1
|
|
|
|
|
|
|
|
|
|
# 還有布林運算元: `or`, `and` and `not`.
|
|
|
|
|
# 這些運算元要求第一個參數必需為布林值。
|
|
|
|
|
true and true #=> true
|
|
|
|
|
false or true #=> true
|
|
|
|
|
# 1 and true #=> ** (ArgumentError) argument error
|
|
|
|
|
|
|
|
|
|
# Elixir 也提供了 `||`, `&&` 及 `!`,它們接受任何型別的參數。
|
|
|
|
|
# 除了 `false` 與 `nil` 之外的值都會被當做 true。
|
|
|
|
|
1 || true #=> 1
|
|
|
|
|
false && 1 #=> false
|
|
|
|
|
nil && 20 #=> nil
|
|
|
|
|
!true #=> false
|
|
|
|
|
|
|
|
|
|
# 用來比較的運算元有:`==`, `!=`, `===`, `!==`, `<=`, `>=`, `<` and `>`
|
|
|
|
|
1 == 1 #=> true
|
|
|
|
|
1 != 1 #=> false
|
|
|
|
|
1 < 2 #=> true
|
|
|
|
|
|
|
|
|
|
# `===` 和 `!==` 會嚴格比較整數與浮點數
|
|
|
|
|
1 == 1.0 #=> true
|
|
|
|
|
1 === 1.0 #=> false
|
|
|
|
|
|
|
|
|
|
# 兩個不同的型別也可以比較
|
|
|
|
|
1 < :hello #=> true
|
|
|
|
|
|
|
|
|
|
# 所有型別的排序如下:
|
|
|
|
|
# number < atom < reference < functions < port < pid < tuple < list < bit string
|
|
|
|
|
|
|
|
|
|
# 引用 Joe Armstrong 的話: "實際排序的先後並不重要, 但有明確排出全體順序的定
|
|
|
|
|
# 義才是重要的。"
|
|
|
|
|
|
|
|
|
|
## ---------------------------
|
|
|
|
|
## -- 控制流程
|
|
|
|
|
## ---------------------------
|
|
|
|
|
|
|
|
|
|
# `if` 表達式
|
|
|
|
|
if false do
|
|
|
|
|
"This will never be seen"
|
|
|
|
|
else
|
|
|
|
|
"This will"
|
|
|
|
|
end
|
|
|
|
|
|
|
|
|
|
# 也有 `unless`
|
|
|
|
|
unless true do
|
|
|
|
|
"This will never be seen"
|
|
|
|
|
else
|
|
|
|
|
"This will"
|
|
|
|
|
end
|
|
|
|
|
|
|
|
|
|
# 還記得模式比對嗎?Elixir 中許多控制流程的結構都依賴模式比對來運作。
|
|
|
|
|
|
|
|
|
|
# `case` 讓我們可以將一個值與許多模式進行比對:
|
|
|
|
|
case {:one, :two} do
|
|
|
|
|
{:four, :five} ->
|
|
|
|
|
"This won't match"
|
|
|
|
|
{:one, x} ->
|
|
|
|
|
"This will match and bind `x` to `:two` in this clause"
|
|
|
|
|
_ ->
|
|
|
|
|
"This will match any value"
|
|
|
|
|
end
|
|
|
|
|
|
|
|
|
|
# 當我們不需要某個值的時候,通常會將它比對成 `_`。
|
|
|
|
|
# 例如我們只關心串列的第一個值的情況時:
|
|
|
|
|
[head | _] = [1,2,3]
|
|
|
|
|
head #=> 1
|
|
|
|
|
|
|
|
|
|
# 若希望程式更好懂時,我們會這樣處理:
|
|
|
|
|
[head | _tail] = [:a, :b, :c]
|
|
|
|
|
head #=> :a
|
|
|
|
|
|
|
|
|
|
# `cond` 讓我們可以同時檢測多個不同的值。
|
|
|
|
|
# 用 `cond` 來代替巢狀的 `if` 表達式
|
|
|
|
|
cond do
|
|
|
|
|
1 + 1 == 3 ->
|
|
|
|
|
"I will never be seen"
|
|
|
|
|
2 * 5 == 12 ->
|
|
|
|
|
"Me neither"
|
|
|
|
|
1 + 2 == 3 ->
|
|
|
|
|
"But I will"
|
|
|
|
|
end
|
|
|
|
|
|
|
|
|
|
# 把最後一個條件設為 `true` 來捕捉剩下的所有情況是很常見的作法。
|
|
|
|
|
cond do
|
|
|
|
|
1 + 1 == 3 ->
|
|
|
|
|
"I will never be seen"
|
|
|
|
|
2 * 5 == 12 ->
|
|
|
|
|
"Me neither"
|
|
|
|
|
true ->
|
|
|
|
|
"But I will (this is essentially an else)"
|
|
|
|
|
end
|
|
|
|
|
|
|
|
|
|
# `try/catch` 用來捕捉拋出的值,它也提供 `after` 子句,無論是否有接到拋出的值,
|
|
|
|
|
# 最後都會調用其下的程式。
|
|
|
|
|
try do
|
|
|
|
|
throw(:hello)
|
|
|
|
|
catch
|
|
|
|
|
message -> "Got #{message}."
|
|
|
|
|
after
|
|
|
|
|
IO.puts("I'm the after clause.")
|
|
|
|
|
end
|
|
|
|
|
#=> I'm the after clause
|
|
|
|
|
# "Got :hello"
|
|
|
|
|
|
|
|
|
|
## ---------------------------
|
|
|
|
|
## -- 模組與函式
|
|
|
|
|
## ---------------------------
|
|
|
|
|
|
|
|
|
|
# 匿名函式 (注意那個句點)
|
|
|
|
|
square = fn(x) -> x * x end
|
|
|
|
|
square.(5) #=> 25
|
|
|
|
|
|
|
|
|
|
# 匿名函式也接受多個子句及防衛(guards)
|
|
|
|
|
# Guards 可以進行模式比對
|
|
|
|
|
# 用 `when` 來描述 guards
|
|
|
|
|
f = fn
|
|
|
|
|
x, y when x > 0 -> x + y
|
|
|
|
|
x, y -> x * y
|
|
|
|
|
end
|
|
|
|
|
|
|
|
|
|
f.(1, 3) #=> 4
|
|
|
|
|
f.(-1, 3) #=> -3
|
|
|
|
|
|
|
|
|
|
# Elixir 也提供許多內建的函式
|
|
|
|
|
# 這些在預設的作用域下都可以使用
|
|
|
|
|
is_number(10) #=> true
|
|
|
|
|
is_list("hello") #=> false
|
|
|
|
|
elem({1,2,3}, 0) #=> 1
|
|
|
|
|
|
|
|
|
|
# 你可以用模組將多個的函式集合在一起。在模組裡,用 `def` 來定義函式。
|
|
|
|
|
defmodule Math do
|
|
|
|
|
def sum(a, b) do
|
|
|
|
|
a + b
|
|
|
|
|
end
|
|
|
|
|
|
|
|
|
|
def square(x) do
|
|
|
|
|
x * x
|
|
|
|
|
end
|
|
|
|
|
end
|
|
|
|
|
|
|
|
|
|
Math.sum(1, 2) #=> 3
|
|
|
|
|
Math.square(3) #=> 9
|
|
|
|
|
|
|
|
|
|
# 要編譯我們的 Math 模組時,先將它存成 `math.ex`,再用 `elixirc` 進行編譯。
|
|
|
|
|
# 在終端機輸入: elixirc math.ex
|
|
|
|
|
|
|
|
|
|
# 在模組中我們可以用 `def` 宣告函式,及用 `defp` 宣告私有 (private) 函式。
|
|
|
|
|
# 使用 `def` 定義的函式可以在其它的模組中被調用。
|
|
|
|
|
# 私有的函式只能在這個模組內部調用。
|
|
|
|
|
defmodule PrivateMath do
|
|
|
|
|
def sum(a, b) do
|
|
|
|
|
do_sum(a, b)
|
|
|
|
|
end
|
|
|
|
|
|
|
|
|
|
defp do_sum(a, b) do
|
|
|
|
|
a + b
|
|
|
|
|
end
|
|
|
|
|
end
|
|
|
|
|
|
|
|
|
|
PrivateMath.sum(1, 2) #=> 3
|
|
|
|
|
# PrivateMath.do_sum(1, 2) #=> ** (UndefinedFunctionError)
|
|
|
|
|
|
|
|
|
|
# 函式宣告也支援用防衛條件及多個條件子句
|
|
|
|
|
defmodule Geometry do
|
|
|
|
|
def area({:rectangle, w, h}) do
|
|
|
|
|
w * h
|
|
|
|
|
end
|
|
|
|
|
|
|
|
|
|
def area({:circle, r}) when is_number(r) do
|
|
|
|
|
3.14 * r * r
|
|
|
|
|
end
|
|
|
|
|
end
|
|
|
|
|
|
|
|
|
|
Geometry.area({:rectangle, 2, 3}) #=> 6
|
|
|
|
|
Geometry.area({:circle, 3}) #=> 28.25999999999999801048
|
|
|
|
|
# Geometry.area({:circle, "not_a_number"})
|
|
|
|
|
#=> ** (FunctionClauseError) no function clause matching in Geometry.area/1
|
|
|
|
|
|
2023-04-27 23:09:38 +03:00
|
|
|
|
# 由於不可變特性 (immutability),遞迴在 Elixir 中扮演重要的角色。
|
2016-04-21 23:52:12 +03:00
|
|
|
|
defmodule Recursion do
|
|
|
|
|
def sum_list([head | tail], acc) do
|
|
|
|
|
sum_list(tail, acc + head)
|
|
|
|
|
end
|
|
|
|
|
|
|
|
|
|
def sum_list([], acc) do
|
|
|
|
|
acc
|
|
|
|
|
end
|
|
|
|
|
end
|
|
|
|
|
|
|
|
|
|
Recursion.sum_list([1,2,3], 0) #=> 6
|
|
|
|
|
|
|
|
|
|
# Elixir 模組也支援屬性,模組有內建一些屬性,而你也可以定義自己的屬性。
|
|
|
|
|
defmodule MyMod do
|
|
|
|
|
@moduledoc """
|
|
|
|
|
這是內建的屬性,模組文件
|
|
|
|
|
"""
|
|
|
|
|
|
|
|
|
|
@my_data 100 # 這是自訂的屬性
|
|
|
|
|
IO.inspect(@my_data) #=> 100
|
|
|
|
|
end
|
|
|
|
|
|
|
|
|
|
## ---------------------------
|
|
|
|
|
## -- 結構與例外 (Structs and Exceptions)
|
|
|
|
|
## ---------------------------
|
|
|
|
|
|
|
|
|
|
# 結構 (structs) 是 maps 的擴展。是 Elixir 裡可以有預設值,編譯期檢查及
|
|
|
|
|
# 多形 (polymorphism) 的資料結構。
|
|
|
|
|
defmodule Person do
|
|
|
|
|
defstruct name: nil, age: 0, height: 0
|
|
|
|
|
end
|
|
|
|
|
|
|
|
|
|
joe_info = %Person{ name: "Joe", age: 30, height: 180 }
|
|
|
|
|
#=> %Person{age: 30, height: 180, name: "Joe"}
|
|
|
|
|
|
|
|
|
|
# 取得 name 的值
|
|
|
|
|
joe_info.name #=> "Joe"
|
|
|
|
|
|
|
|
|
|
# 更新 age 的值
|
|
|
|
|
older_joe_info = %{ joe_info | age: 31 }
|
|
|
|
|
#=> %Person{age: 31, height: 180, name: "Joe"}
|
|
|
|
|
|
|
|
|
|
# The `try` block with the `rescue` keyword is used to handle exceptions
|
|
|
|
|
# 帶有 `rescue` 關鍵字的 `try` 區塊是用來進行例外處理的。
|
|
|
|
|
try do
|
|
|
|
|
raise "some error"
|
|
|
|
|
rescue
|
|
|
|
|
RuntimeError -> "rescued a runtime error"
|
|
|
|
|
_error -> "this will rescue any error"
|
|
|
|
|
end
|
|
|
|
|
#=> "rescued a runtime error"
|
|
|
|
|
|
|
|
|
|
# 所有的異常都有帶著一個訊息
|
|
|
|
|
try do
|
|
|
|
|
raise "some error"
|
|
|
|
|
rescue
|
|
|
|
|
x in [RuntimeError] ->
|
|
|
|
|
x.message
|
|
|
|
|
end
|
|
|
|
|
#=> "some error"
|
|
|
|
|
|
|
|
|
|
## ---------------------------
|
|
|
|
|
## -- 平行處理
|
|
|
|
|
## ---------------------------
|
|
|
|
|
|
2023-04-27 23:09:38 +03:00
|
|
|
|
# Elixir 依靠 actor 模式來進行平行處理。在 Elixir 中要寫出平行處理程式,
|
2016-04-21 23:52:12 +03:00
|
|
|
|
# 只需要三個基本要素:建立行程,發送訊息及接收訊息。
|
|
|
|
|
|
|
|
|
|
# 我們用 `spawn` 函式來建立行程,它接收一個函式當參數。
|
|
|
|
|
f = fn -> 2 * 2 end #=> #Function<erl_eval.20.80484245>
|
|
|
|
|
spawn(f) #=> #PID<0.40.0>
|
|
|
|
|
|
|
|
|
|
# `spawn` 會回傳一個 pid (行程識別碼),你可以利用這個 pid 來對該行程傳送訊息。
|
|
|
|
|
# 我們會使用 `send` 運算元來傳送訊息。但首先我們要讓該行程可以接收訊息。這要用
|
|
|
|
|
# 到 `receive` 機制來達成。
|
|
|
|
|
|
|
|
|
|
# `receive` 區塊能讓行程監聽接收到的訊息。每個 `receive do` 區塊只能接收一條
|
|
|
|
|
# 訊息。若要接收多條訊息時,含有 `receive do` 的函式必須要在接到訊息後,遞迴呼
|
|
|
|
|
# 叫自己以再次進入 `receive do` 區塊。
|
|
|
|
|
|
|
|
|
|
defmodule Geometry do
|
|
|
|
|
def area_loop do
|
|
|
|
|
receive do
|
|
|
|
|
{:rectangle, w, h} ->
|
|
|
|
|
IO.puts("Area = #{w * h}")
|
|
|
|
|
area_loop()
|
|
|
|
|
{:circle, r} ->
|
|
|
|
|
IO.puts("Area = #{3.14 * r * r}")
|
|
|
|
|
area_loop()
|
|
|
|
|
end
|
|
|
|
|
end
|
|
|
|
|
end
|
|
|
|
|
|
|
|
|
|
# 編譯模組,並在 shell 中創造一個行程來執行 `area_loop`。
|
|
|
|
|
pid = spawn(fn -> Geometry.area_loop() end) #=> #PID<0.40.0>
|
|
|
|
|
# 更簡潔的替代寫法
|
|
|
|
|
pid = spawn(Geometry, :area_loop, [])
|
|
|
|
|
|
|
|
|
|
# 對 `pid` 傳送訊息,則會與接收區塊進行樣式比對。
|
|
|
|
|
send pid, {:rectangle, 2, 3}
|
|
|
|
|
#=> Area = 6
|
|
|
|
|
# {:rectangle,2,3}
|
|
|
|
|
|
|
|
|
|
send pid, {:circle, 2}
|
|
|
|
|
#=> Area = 12.56000000000000049738
|
|
|
|
|
# {:circle,2}
|
|
|
|
|
|
|
|
|
|
# The shell is also a process, you can use `self` to get the current pid
|
|
|
|
|
# shell 也是一個行程 (process),你可以用 `self` 拿到目前的 pid
|
|
|
|
|
self() #=> #PID<0.27.0>
|
|
|
|
|
```
|
|
|
|
|
|
|
|
|
|
## 參考資料
|
|
|
|
|
|
|
|
|
|
* [Getting started guide](http://elixir-lang.org/getting-started/introduction.html) from the [Elixir website](http://elixir-lang.org)
|
|
|
|
|
* [Elixir Documentation](http://elixir-lang.org/docs/master/)
|
|
|
|
|
* ["Programming Elixir"](https://pragprog.com/book/elixir/programming-elixir) by Dave Thomas
|
|
|
|
|
* [Elixir Cheat Sheet](http://media.pragprog.com/titles/elixir/ElixirCheat.pdf)
|
|
|
|
|
* ["Learn You Some Erlang for Great Good!"](http://learnyousomeerlang.com/) by Fred Hebert
|
|
|
|
|
* ["Programming Erlang: Software for a Concurrent World"](https://pragprog.com/book/jaerlang2/programming-erlang) by Joe Armstrong
|