Booting into functional programming

Booting into FP
Dhaval Dalal
@softwareartisan
https://dhavaldalal.wordpress.com
14th Nov 2019

Our Tour
• Functions, everywhere!

• Understanding Referential Transparency

• What is a Lambda?

• Modularization

• Building Abstractions with Higher-Order Functions

• Lazy Evaluation

• Tackling Complexity using Function Composition

• Streams (Lazy Lists)

• De-structuring and Pattern Matching

• Currying

• Sequence/Chain Operations together using Monads - Dealing with absence of Values

• Overview of Upcoming Sessions

Touted Advantages
• Less number of lines of code as compared to imperative code.
• Hence, Functional Programmers are more “productive”

Touted Advantages
• Declarative
• Describes “what to do”, rather than imperative - “how to do”

Touted Advantages
• Declarative
• Immutability
• No assignment statements, a variable value never changes.

Touted Advantages
• Declarative
• Immutability
• No side-eﬀects
• Eliminates major source of bugs, order of evaluation does not matter.

Touted Advantages
• Declarative
• Immutability
• No side-eﬀects
• Eliminates major source of bugs, order of evaluation does not matter.
• Referential Transparency
• Enables equational reasoning
Source: Why Functional Programming Matters - John Hughes

Why FP ^ matters?
• Its about Modularity
• “Our ability to decompose a problem into parts depends directly on our ability
to glue solutions” - John Hughes
really

Why FP ^ matters?
• Beneﬁts
• Small can be coded quickly and easily.
• Reusability of general-purpose modules.
• Independent testing of modules.
really

Why FP ^ matters?
• Beneﬁts
• Small can be coded quickly and easily.
• Reusability of general-purpose modules.
• Independent testing of modules.
• The Glue
• Lazy Evaluation.
• Higher-Order Functions.
Source: Why Functional Programming Matters - John Hughes
really

• Function is a smallest unit of computation.
Functions Everywhere!

f : x ↦ x2OR
f(x) = x2

f : x ↦ x2OR
f(x) = x2
• Domain, Co-Domain and Range of a function.
• Domain {-3, -2, -1, 0, 1, 2, 3}
• Co-Domain {0, 1, 2, 3, 4, 5, 6, 7, 8, 9}

f : x ↦ x2OR
f(x) = x2
• Domain {-3, -2, -1, 0, 1, 2, 3}
• Co-Domain {0, 1, 2, 3, 4, 5, 6, 7, 8, 9}
• Range {0, 1, 4, 9}

f : x ↦ x2OR
f(x) = x2
• Domain {-3, -2, -1, 0, 1, 2, 3}
• Co-Domain {0, 1, 2, 3, 4, 5, 6, 7, 8, 9}
• Range {0, 1, 4, 9}
• Function is a relation that maps the domain into co-domain such
that there is exactly one input maps to a output (1:1, can be m:1
but not 1:n).

f : x ↦ x2OR
f(x) = x2
• Domain {-3, -2, -1, 0, 1, 2, 3}
• Co-Domain {0, 1, 2, 3, 4, 5, 6, 7, 8, 9}
• Range {0, 1, 4, 9}
• Function is a relation that maps the domain into co-domain such
that there is exactly one input maps to a output (1:1, can be m:1
but not 1:n).
• Type of the function is int ↦ int

• Uses nothing other than i/p parameters (and its
definition) to produce o/p

• Neither modifies input arguments nor reads/modifies
external state

• Call is substitutable by its body. To understand the
code, you don’t have to look elsewhere.
class Calculator {
// (int, int) -> int
public int add(final int x, final int y) {
return x + y;
}
}
A Pure Function
f : T ↦ R

Black-Hole like
Function
• A typical Setter!
class Calculator {
private int memory = 0;
public Calculator(final int memory) {
this.memory = memory;
}
// int -> ()
public void setMemory(final int memory) {
}
}
f : T ↦ ()

Mother’s Love like
Function
• A typical Getter!
class Calculator {
}
// () -> int
public int recallMemory() {
return memory;
}
}
f : () ↦ R

Politician like
Function
• a Printer!

• Neither consume anything, nor I return anything,
because I do everything under the table.
class Calculator {
}
public void printMemory() {
System.out.println(memory);
}
}
f : () ↦ ()

• Modiﬁes or interacts with things outside of its scope
(interaction with outside world), and may also return a value.

• Reading/Writing to disk, DB, Socket etc… anything that modiﬁes
external world.
A Side-effecting Function
class Calculator {
}
public Integer add(final int x, final int y) { … }
public Integer memoryPlus(final int n) {
memory = add(memory, n);
return memory;
}
}
Outside of
its scope

• Write Side-eﬀect…a typical Setter!
Black-Hole like
Side-effecting Function
class Calculator {
}
// int -> ()
public void setMemory(final int memory) {
}
}
f : T ↦ ()

Mother’s Love like
Side-Effecting Function
• Read Side-eﬀect!
class Calculator {
private Random rnd = new Random();
// () -> int
public int random() {
return rnd.nextInt();
}
}
f : () ↦ R

Politician like
Side-effecting Function
• a Printer!

class Calculator {
}
public void printMemory() {
System.out.println(memory);
}
}
f : () ↦ ()

Side-effects or Useful-effects?
Source: https://channel9.msdn.com/Blogs/Charles/Simon-Peyton-Jones-Towards-a-Programming-Language-Nirvana

Understanding
Referential Transparency
class Calculator {
private int memory;
}
public Integer add(final int x, final int y) {
return x + y;
}
public Integer memoryPlus(final int n) {
memory = add(memory, n);
return memory;
}
}

Understanding
Referential Transparency
c.add(2, 3); // 5
c.add(2, 4); // 6
c.add(2, 3); // 5
c.add(2, 4); // 6
Calculator c = new Calculator();
Referentially Opaque memoryPlus :
1.Cannot replace it with resulting value.
2.Returns different results as time
progresses, as behaviour depends on history.
c.memoryPlus(2); // 2
Time Time
Referentially Transparent add :
1.Substitute any expression with its
resulting value.
2. Returns same results all the time, as
behaviour does not depend on history.

Pure Functions,
Side-Effects and
Evaluation Order
c.add(2, 3); // 5
c.add(2, 4); // 6
c.add(2, 3); // 5
c.add(2, 4); // 6
Calculator c = new Calculator();
Out-of-order execution not
Possible
Time Time
Out-of-order execution possible

How can we make
memoryPlus
Referentially
Transparent?

Ref. Transparent
memoryPlus
class Calculator {
private final int memory;
}
public int add { … }
public Calculator memoryPlus(final int n) {
return new Calculator(add(memory, n));
}
}
Make memory
Immutable
Return new instance from
operation.

Reﬂections
• Referential Transparency is about replacing any expression
(or function) with its resulting value.

• To be referentially transparent, function will require to work
with immutable data.

• To be referentially transparent, a function will require to be
pure.

• To be referentially transparent, a function will require to be
deterministic.

• Referentially transparent functions are context-free and the
context is time.
https://github.com/CodeJugalbandi/FunctionalProgramming/tree/master/melodies/referentialTransparency

3 Pillars of FP
• Referential Transparency - enforces invariant expressions or functions
are pure.

• Substitution Model - RT enables simple and natural mode of
reasoning about program evaluation.

• Equational Reasoning - Computation proceeds like solving an
algebraic equation, where we substitute and reduce to its simplest
form.
Source: Functional and Reactive Domain Modelling, Debashish Ghosh
Ref.Trans
Substitution
Eq.Reasoning

Good Deeds Happen Anonymously
• Naming is one of the hardest problem in software.

• Alleviate using an anonymous function - Lambda 𝝺.
return x + y;
}
return x + y;
}
(final int x, final int y) -> { x + y; }
(x, y) -> x + y;
Drop all
inessentials
what remains
are essentials,
parameters and
body

SAMs Become…
new Button().addActionListener(new ActionListener() {
public void actionPerformed(ActionEvent e) {
System.out.print("essence");
}
});
new Button().addActionListener(new ActionListener() {
public void actionPerformed(ActionEvent e) {
System.out.println("essence");
}
});
new Button().addActionListener(e -> System.out.print("essence"));
Drop all inessentials
what remains
are essentials,
parameters and
body new Button().addActionListener(System.out::print);
OR

Lambda
Function<Integer, Integer> twice = x -> 2 * x;
twice.apply(3); // 6
It would be have been nice if
some more syntactic sugar
was added by Java8 to make
this happentwice(3);Instead of
• Compiled as interface implementation with synthesised
method having signature of the abstract method in that
interface.

• An interface with single abstract method (SAM).

• @FunctionalInterface - Though optional annotation, its better to
have it so that it ensures that it stays as a lambda and not
become something more.

@FunctionalInterface
public interface Function<T, R> {
R apply(T t);
…
}
Function<Float, Float> twice = x -> 2 * x;
twice.apply(3); // 6
// A function returns its argument unchanged
Function<Float, Float> identity = x -> x;
public interface BiFunction<T, U, R> {
R apply(T t, U u);
…
}
BiFunction<Float, Float, Float> add = (x, y) -> x + y;
add.apply(2, 3); // 5
Functional Interfaces

A Black-Hole like
Functional Interface
public interface Consumer<T> {
void accept(T t);
…
}
corrupt.accept(100.34);
public interface BiConsumer<T, U> {
void accept(T t, U u);
…
}
BiConsumer<Celestial, Celestial> blackHole =
(planet, asteroid) -> { }
blackHole.accept(planet, asteriod);

A Mother’s Love like
public interface Supplier<T> {
T get();
}
Supplier<String> mother = () -> “Love";
mother.get(); // Love
Things Emerge
without
consuming anything

public interface Predicate<T> {
boolean test(T t);
…
}
Predicate<T> affirmative = x -> true;
affirmative.test(1); // true
Predicate<T> negative = x -> false;
negative.test(1); // false
Predicate<Integer> isEven = x -> x % 2 == 0;
isEven.test(2); // true
isEven.test(3); // false
Mr. Spock like

Politician like
public interface Runnable {
void run();
}

Every“thing” is a
Lambda
• Like Object, a Function is also a thing, so

• Pass a function to a function

• Return a function from within a function

• A function that produces or consumes a function is
called as Higher Order Function - HOF

• Either pass existing method reference (static or instance) or
write an in-place lambda where a method expects a function
parameter.

• Modularization - Build abstractions with HOFs.

Pass a function
to a function
• Subsumed ‘for’ loop.

• Repetitive behaviour using Recursion.
void iterate(int times, Runnable body) {
if (times <= 0) {
return;
}
body.run();
iterate(times - 1, body);
}
iterate(2, () -> System.out.println("Hello"));
// Hello
// Hello
No need for
explicit looping
constructs!
Just a function!

Return a function
from a function
• Subsumed Factory Method.
Function<Double, Double> power(double raiseTo) {
return x -> Math.pow(x, raiseTo);
}
Function<Double, Double> square = power(2.0);
square.apply(2.0); // 4.0
Function<Double, Double> cube = power(3.0);
cube.apply(2.0); // 8.0

Subsumed Aspect
public<T, R> Function<T, R> time(Function<T, R> fn) {
return t -> {
long startTime = System.currentTimeMillis();
R result = fn.apply(t);
long timeTaken = System.currentTimeMillis() - startTime;
System.out.println("Time: " + timeTaken + " ms");
return result;
}
}
• AOP - Around style

Composition Pipeline
• Compose a function from other functions by aligning
types.
• Function composition is not commutative.

• f ∘ g ≠ g ∘ f
// T -> U
Function<T, U> f;
// U -> R
Function<U, R> g;
// T -> U . U -> R = T -> R
Function<T, R> h = f . g;

Why Composition?
• Tackle complexity by composing behaviors.

• Compose larger functions from smaller ones.

• Smaller functions can be tested independently.

• If the parts are correct, we can then trust the
correctness of the whole.

• Every part of the larger function can be reasoned about
easily.

Composing Functions
• Using compose & andThen
Function<Integer, Integer> square = x -> x * x;
Function<Integer, Integer> twice = x -> 2 * x;
square.compose(twice).apply(2); // 16
square.andThen(twice).apply(2); // 8
Read Left to Right
Think Right to Left
Read Left to Right
Think Left to Right
• For languages that support function composition, look for a way
to go with the grain of thought. In Java, prefer using andThen

Circle of Purity
• Compose behaviours using
pure Functions.

• Data ﬂows through
composed pipes.

• Interact with outside world
using Side-eﬀecting
functions.
Outside
World
Side
Effecting
Functions
Pure
Functions
Courtesy: Venkat
Consumer Supplier

Reﬂections
• Enforce order of evaluation.

• In imperative programming, statements enforce order of evaluation.

• In FP, the order of composition determines the order of evaluation.

• Function composition (and not function application) is the
default way to build sub-routines in Concatenative
Programming Style, a.k.a Point Free Style.

• Functions neither contain argument types nor names, they
are just laid out as computation pipeline.

• Lot of our domain code is just trying to do this!

• Makes code more succinct and readable.
https://github.com/CodeJugalbandi/FunctionalProgramming/tree/master/melodies/sequencing

Eager Evaluation
• Java uses eager evaluation for method arguments.

• Args evaluated before passing to the method.
public int first(int x, int y) { 
System.out.println("Inside First..."); 
return x; 
}
public int loop() { 
System.out.println("Inside Loop..."); 
return loop(); 
} 
System.out.println(first(10, 20));
// Inside First…
// 10 
System.out.println(first(10, loop())); // Stackoverflow 
System.out.println(first(loop(), 20)); // Stackoverflow

Lazy Evaluation
• Lazy means don’t compute until there exists a demand.

• Haskell OTOH is lazy by default.

• Args not evaluated before passing to the method,
instead evaluated upon use.
loop = loop 
 
first x y = x 
 
main :: IO () 
main = do 
print $ first 10 20 -- 10 
print $ first 10 loop -- 10 
print $ first loop 10 -- Infinite Loop

Lazy Evaluation
• Scala is eager by default like Java, and provides a
“lazy” switch.
def loop: Int = loop 
 
def first(x: Int, y: Int) = x 
 
first(10, 20) // 10 
first(loop, 20) // Non-termination 
first(10, loop) // Non-termination
scala> lazy val x = loop 
x: Int = <lazy> 
scala> x // Non-termination

public static int first(Supplier<Integer> x, Supplier<Integer> y) { 
System.out.println("Inside First..."); 
return x.get(); 
} 
public static int loop() { 
System.out.println("Inside Loop..."); 
return loop(); 
} 
System.out.println(first(10, 20)); // 10 
System.out.println(first(() -> 10, () -> loop())); // 10 
System.out.println(first(() -> loop(), () -> 20)); // Stackoverflow
Simulating Lazy Evaluation
Representation of
computation and
not the
computation itself.
• To delay the evaluation of args (lazy), wrap them in lambda.

• Call the lambda when we need to evaluate.

• Immutability and Purity makes lazy evaluation possible.

Imperative Filtering and
Collecting
String sentence = "all mimsy were the borogoves and the momeraths";
String [] words = sentence.split(" ");
StringBuilder caps = new StringBuilder();
for (var word : words) {
if (word.length() < 4) {
caps.append(word.toUpperCase());
caps.append(" ");
}
}
String capitalized = caps.toString().trim();
System.out.println(capitalized); // ALL THE AND THE
Enumeration
and
Filtering
interleaved
Collector
• One of the motivations of Streams is to make
parallelism more accessible to developers.

Streams (Lazy List)
String sentence = "all mimsy were the borogoves and the momeraths"; 
String capitalized = Stream.of(sentence.split(" ")) 
.filter(word -> word.length() < 4) 
.map(String::toUpperCase) 
.collect(Collectors.joining(" ")); 
 
System.out.println(capitalized);
.peek(word -> System.out.println("Filter = " + word)) 
.peek(word -> System.out.println("Map = " + word)) 
.map(String::toUpperCase) 
System.out.println(capitalized);
• One Element at a time passes through the pipeline

public static String capitalize(String s) { 
try { 
Thread.sleep(2000); 
} catch (Exception e) { 
} 
return s.toUpperCase(); 
} 
long startTime = System.currentTimeMillis(); 
.map(word -> capitalize(word)) 
long timeTaken = System.currentTimeMillis() - startTime; 
System.out.println(String.format("%s, Took %d(ms)", capitalized, timeTaken));
// ALL THE AND THE, Took 8043(ms)
How can we improve this?

Parallelizing
public static String capitalize(String s) { 
try { 
Thread.sleep(2000); 
} catch (Exception e) { 
} 
return s.toUpperCase(); 
} 
String capitalized = Stream.of(sentence.split(" “))
.parallel()
.map(word -> capitalize(word)) 
long timeTaken = System.currentTimeMillis() - startTime; 
System.out.println(String.format("%s, Took %d(ms)", capitalized, timeTaken));
// ALL THE AND THE, Took 2027(ms)
• One of the motivations of Streams is to make
parallelism more accessible to developers.

Don’t Parallelize everything
• It takes time to create threads, scatter input and
gather results.
long sum = Stream.iterate(0, x -> x + 1) 
.limit(10000) 
.filter(n -> n % 2 == 0) 
.reduce(0, (acc, elem) -> acc + elem); 
long diff = System.currentTimeMillis() - startTime; 
System.out.println(sum + " Took " + diff + “(ms)");
// 24995000 Took 36(ms)
long sum = Stream.iterate(0, x -> x + 1) 
.limit(10000) 
.parallel() 
.filter(n -> n % 2 == 0) 
.reduce(0, (acc, elem) -> acc + elem); 
long diff = System.currentTimeMillis() - startTime; 
System.out.println(sum + " Took " + diff + "(ms)"); 
// 24995000 Took 56(ms)

Streams - Infinitely Lazy
• Streams are pull-based, consumer decides the pace
of pull.

• With Streams, only essential space is allocated upon
materialization, the rest is in ether :)

• This reduces memory footprint (as you don’t bring every item
in memory).
Stream.iterate(22, x -> x + 1) 
.filter(n -> n % 5 == 0) 
.findFirst() 
.ifPresent(System.out::println); // 25
• Because of laziness findFirst() only runs till it find the
first matching element.
Stream.iterate(22, x -> x + 1);

De-structuring
• It is about extracting data from the innards of the
Data-Structure.

• Gives us a short-hand way of naming parts of the
data-structure.
val list = 1 :: 2 :: 3 :: 4 :: Nil 
val first :: second :: rest = list 
println(first) // 1 
println(second) // 2 
println(rest) // List(3, 4) 
 
case class Name(first: String, last: String, salutation: String) 
 
val dd = Name("Dhaval", "Dalal", "Mr.") 
val Name(fName, lName, s) = dd 
println(fName) // Dhaval 
println(lName) // Dalal

Pattern Matching
• In Pattern Matching, we are not only matching on relative
positions of elements in the data-structure, but also trying to
dispatch to different definitions based on the value inside the
data structure.

• This dispatch is a kind of conditional construct. So, it’s about
associating a definition with a particular set of i/ps.
val dd = Name("Dhaval", "Dalal", "Mr.") 
val nancy = Name("Nancy", "Drew", "Ms.") 
def greet(name: Name) = name match { 
case Name("Nancy", "Drew", _) => s"Hello Ms. Drew" 
case Name(f@"Dhaval", l, s) => s"Hey $f" 
case name => s"Hi ${name.first.toUpperCase}" 
} 
 
println(greet(dd)) // Hey Dhaval 
println(greet(nancy)) // Hello Ms. Drew 
println(greet(Name("Peter", "Shmo", "Mr."))) // Hi PETER

Implementing Recursion
Using Pattern Matching
In Erlang
sum(Acc, []) -> Acc; 
sum(Acc, [X|XS]) -> sum(Acc + X, XS). 
 
sum(List) -> sum(0, List). 
 
main(_) -> 
io:format("~p~n", [sum([])]), 
io:format("~p~n", [sum([1, 2, 3, 4])]).

// (String, Integer) -> String 
String merge(String x, Integer y) { 
return x + y.toString(); 
} 
// String -> (Integer -> String) 
Function<Integer, String> merge(String x) { 
return y -> x + y.toString(); 
} 
// () -> (String -> (Integer -> String)) 
Function<String, Function<Integer, String>> merge() { 
return x -> y -> x + y.toString(); 
}
// (String -> (Integer -> String)) 
Function<String, Function<Integer, String>> merge = x -> y -> x + y.toString(); 
System.out.println(merge("Test#", 1)); // Test#1 
System.out.println(merge("Test#").apply(2)); // Test#2 
System.out.println(merge().apply("Test#").apply(3)); // Test#3
System.out.println(merge.apply("Test#").apply(4)); // Test#4
• Refers to the phenomena of rewriting a N-arg function to a
nest of functions, each taking only 1-arg at a time
Currying

Currying
For each arg, there is
another nested
function, that takes
a arg and returns a
function taking the
subsequent arg, until
all the args are
exhausted.
f : (T, U, V) ↦ R
f′ : T ↦ U ↦ V ↦ R
f = f′

Make Absence Explicit
Optional<T>
• Stop null abuse!

• Don’t return null, use Optional<T> so that it explicitly tells that
in the type.

• A container or view it as a collection containing single
value or is empty.
Presence
Absence
Input(s)
Value
null

Before
String frequentFlyerId = "123456789"; 
UserRepository userRepository = new UserRepository(); 
User user = userRepository.findBy(frequentFlyerId); 
String targetPage = null; 
int miles = 7000; 
if (user != null) { 
int newPoints = user.addPointsUsing(miles); 
targetPage = "http://tierPage"; 
} else { 
targetPage = "http://loginPage"; 
} 
System.out.println(targetPage);
String frequentFlyerId = "123456789"; 
int miles = 7000; 
UserRepository userRepository = new UserRepository(); 
String targetPage = userRepository.findBy(frequentFlyerId) 
.map(user -> { 
int newPoints = user.addPointsUsing(miles); 
return "http://tierPage"; 
}) 
.orElse("http://loginPage"); 
 
System.out.println(targetPage);
After

Thank - You!
https://github.com/DhavalDalal/FunctionalConference-2019-Booting-Into-FP-Demo

Booting into functional programming

More Related Content

What's hot (20)

Similar to Booting into functional programming (20)

More from Dhaval Dalal (20)

Recently uploaded (20)

Booting into functional programming