Five Languages in a Moment

Five Languages
in a Moment

Sergio Gil

Sergio Gil Pérez de la Manga

@porras

“Excusatio non petita,
accussatio manifesta”

“Learn a new language each year”

2006 Ruby

2007 -

2008

2006 Ruby

2007 -

2008 -

2006 Ruby

2007 -

2008 -

2009

2006 Ruby

2007 -

2008 -

2009 -

2006 Ruby

2007 -

2008 -

2009 -

2010

2006 Ruby

2007 -

2008 -

2009 -

2010 -

Highest level

Lowest level

Highest level

Machine
Lowest level

Highest level

Assembler

Machine
Lowest level

Highest level

C

Assembler

Machine
Lowest level

Highest level

Java

C

Assembler

Machine
Lowest level

Highest level

Ruby

Java

C

Assembler

Machine
Lowest level

Abstractions are tools for the mind

Created by Joe Armstrong at Ericsson in 1986

Functional Language

Concurrency Oriented

Functional Language


Distribution Oriented

Functional Language


Distribution Oriented

Error Tolerant

http://www.defmacro.org/ramblings/fp.html

> X = 1.
> X.
1
> X = 2.
** exception error

> Point = {point, 15, 20}.
> {point, X, Y} = Point.

> Point = {point, 15, 20}.
> X.
15

> Point = {point, 15, 20}.
> X.
15
> Y.
20

Lists

Head and Tail

[H|T]

> L = [1, 2, 3, 4].
> [H|T] = L.

> L = [1, 2, 3, 4].
> [H|T] = L.
> H.
1

> L = [1, 2, 3, 4].
> [H|T] = L.
> H.
1
> T.
[2, 3, 4]

fact(0) -> 1;
fact(X) -> X * fact(X - 1).

sum([]) -> 0;
sum([H|T]) -> H + sum(T).

sum([]) -> 0;

map(_, []) -> [];
map(F, [H|T]) -> [F(H) | map(F, T)].

sum([]) -> 0;

map(_, []) -> [];
map(F, [H|T]) -> [F(H) | map(F, T)].

...

area({rectangle, Width, Height}) -> Width * Height;
area({square, X}) -> X * X;
area({circle, R}) -> 3.14159 * R.

area({rectangle, Width, Height}) -> Width * Height;
area({square, X}) -> X * X;
area({circle, R}) -> 3.14159 * R.

> area({square, 3}).
9

class Rectangle
def initialize(width, height)
@width = width
@height = height
end

def area
@width * height
end
end

class Square
def initialize(x)
@x = x
end

def area
@x * @x
end
end

class Circle
def initialize(r)
@r = r
end

def area
3.14159 * @r
end
end

def area(*args)
case args[0]
when :rectangle
args[1] * args[2]
when :square
args[1] * args[1]
when :circle
3.141519 * args[1]
end
end

receive
Pattern -> Expression;
Pattern -> Expression
end

Remember map?

map(_, []) -> [];
map(F, [H|T]) -> [F(H) | map(F, T)].

pmap(F, L) ->
Parent = self(),
Pids = map(fun(I) ->
spawn(fun() ->
Parent ! {self(), F(I)}
end)
end, L),
gather(Pids).

pmap(F, L) ->
Parent = self(),
Pids = map(fun(I) ->
spawn(fun() ->
Parent ! {self(), F(I)}
end)
end, L),
gather(Pids).

gather([]) -> [];
gather([Pid|T]) ->
receive
{Pid, Result} -> [Result|gather(T)]
end.

More interesting things about Erlang


Error control, even distributed


OTP


OTP
Web development with ChicagoBoss

...you want to play with functional programming

...you're interested in writing parallel programs

Created by Simon Peyton Jones in 1990

Pure functional language

Lazy evaluation

Pure functional language

Lazy evaluation

Strict but powerful type system

mymap _ [] = []
mymap f (x:xs) = f x : map f xs

Haskell Type System

Strong

Haskell Type System

Strong
Static

Haskell Type System

Strong
Static
Inferred

Char
[Char]
String
Integer
([Char], Integer)

data BookInfo = Book Int String [String]

data BookInfo = Book Int String [String]

myBook = Book 987987 "About Wadus" ["Fulano", "Mengano"]

data Color = Red
| Yellow
| Green
| RGB Int Int Int

data Color = Red
| Yellow
| Green
| RGB Int Int Int

yellow = Yellow
black = RGB 0 0 0

Type System + Pattern Matching

data Shape = Circle Float
| Rectangle Float Float
| Square Float

| Square Float

area (Circle r) = r * 3.14
area (Rectangle w h) = w * h
area (Square x) = x * x

| Square Float


circle = Circle 5
rectangle = Rectangle 2 3
square = Square 1.5

| Square Float


circle = Circle 5
square = Square 1.5

-- > area circle
-- 15.700001

| Square Float


circle = Circle 5
square = Square 1.5

-- > area circle
-- 15.700001

-- > :type area
-- area :: Shape -> Float

square_all list = map (n -> n^2) list

> square_all [1, 2, 5]
[1,4,25]

square_all list = map (^2) list

[1,4,25]

square_all list = map (^2) list
square_all = map (^2)

[1,4,25]

@articles = Article.where(:status => 'published')


<% @articles.each do |article| %>
<h2><%= article.title %></h2>
<p><%= article.body %></p>
<% end %>


<% @articles.limit(5).each do |article| %>
<h2><%= article.title %></h2>
<p><%= article.body %></p>
<% end %>

iforelse cond a b = if cond
then a
else b

then a
else b

> iforelse True (1 + 2) (3 + 4)
3

then a
else b

> iforelse True (1 + 2) (3 + 4)
3
> iforelse False (1 + 2) (3 + 4)
7

...you tried the functional paradigm
and want to go further

...you want to see your belief
“types are useless” challenged

...you want to write extremelly
efficient programs in a high level
language

Invented by John McCarthy in 1958

Lots of
Irritating
Superﬂuou
s
Parenthese
s

“property of some
programming languages,
in which the primary
representation of
programs is also a data
structure in a primitive
type of the language
itself”

(1 2 3 ("wadus" 5) (1.3 1.7))

[1, 2, 3, ["wadus", 5], [1.3, 1.7]]

(+ (* 3 5) (/ 10 2))

20

(1 2 3 ("wadus" 5) (1.3 1.7))

(+ (* 3 5) (/ 10 2))

(1 2 3 ("wadus" 5) (1.3 1.7))

(+ (* 3 5) (/ 10 2))

CODE = DATA

Valid Lisp code is composed of
lists whose ﬁrst element is the
name of a function and the rest are
the parameters passed

Metaprogramming is creating and manipulating
code

Metaprogramming is creating and manipulating
code lists

(defmacro backwards (code)
(reverse code))

(reverse code))

(backwards
("wadus" print))

(reverse code))

(backwards
("wadus" print))

(print "wadus")

You should try Common Lisp if...

...you want to take a ﬂight
in a retrofuturist jetpack

...you want to try one of the
most abstraction-capable
and still efficient languages

...you want to understand
why Lisp-lovers love Lisp so
much

Created by Rick Hickey in 2007

Lisp meets the JVM

Lisp meets purelly functional programming

Lisp meets the JVM


Lisp meets laziness

Lisp meets the JVM


Lisp meets laziness

Lisp meets concurrency

(def whole-numbers (iterate inc 1))


(first whole-numbers)
; 1


; 1

(take 5 whole-numbers)
; (1 2 3 4 5)


; 1

(take 5 whole-numbers)
; (1 2 3 4 5)

(last whole-numbers) ; => CATACROKER!

(defn fibonacci []
(map first (iterate (fn [[a b]] [b (+ a b)]) [0 1])))

(defn fibonacci []
(map first (iterate (fn [[a b]] [b (+ a b)]) [0 1])))

(take 20 (fibonacci))
; (0 1 1 2 3 5 8 13 21 34 55 89 144 233 377 610 987 1597
2584 4181)

(def fulano-balance (ref 100))
(def mengano-balance (ref 80))


(dosync
(ref-set fulano-balance (+ @fulano-balance 20))
(ref-set mengano-balance (- @mengano-balance 20)))


(dosync

@fulano-balance
; 120


(dosync

@fulano-balance
; 120
@mengano-balance
; 60

(dosync
(alter fulano-balance + 20)
(alter mengano-balance - 20))

(dosync

@fulano-balance
; 140

(dosync

@fulano-balance
; 140
@mengano-balance
; 40

More interesting things about Clojure

More interesting things about Clojure

Web development with Compojure

...you like both Lisp and the functional approach

...you need/want Java/JVM interoperability

http://github.com/porras/
ujfalusi

You should create your own language if...

...you want to learn about how
languages internally work

...you want to have fun
doing something very geek

...your favorite player
leaves your team and
breaks your heart

Five Languages in a Moment

Recommended

Recommended

More Related Content

What's hot

What's hot (20)

Similar to Five Languages in a Moment

Similar to Five Languages in a Moment (20)

More from Sergio Gil

More from Sergio Gil (9)

Recently uploaded

Recently uploaded (20)

Five Languages in a Moment

Editor's Notes