Overview of Java 8 Streams (Part 1) - Vanderbilt Universityschmidt/cs891f/2018-PDFs/... · •Java 8 streams are a framework in the Java class library that provides several key benefits

Overview of Java 8 Streams (Part 1)

Douglas C. [email protected]

www.dre.vanderbilt.edu/~schmidt

Professor of Computer Science

Institute for Software

Integrated Systems

Vanderbilt University

Nashville, Tennessee, USA

mailto:[email protected]

2

Learning Objectives in this Part of the Lesson• Understand the structure & functionality of Java 8 streams

Aggregate operation (behavior f)

Input x

Output f(x)

Output g(f(x))

Output h(g(f(x)))

Aggregate operation (behavior g)

Aggregate operation (behavior h)

3

Learning Objectives in this Part of the Lesson• Understand the structure & functionality of Java 8 streams, e.g.,

• Fundamentals of streams


Input x

Output f(x)

Output g(f(x))

Output h(g(f(x)))



4

Learning Objectives in this Part of the Lesson• Understand the structure & functionality of Java 8 streams, e.g.,

• Fundamentals of streams

• We’ll use an example program to illustrate key concepts Aggregate operation (behavior f)

Input x

Output f(x)

Output g(f(x))

Output h(g(f(x)))



See github.com/douglascraigschmidt/LiveLessons/tree/master/Java8/ex12

Stream

.of("horatio",

"laertes",

"Hamlet", ...)

.filter(s -> toLowerCase

(s.charAt(0)) == 'h')

.map(this::capitalize)

.sorted()

.forEach(System.out::println);

https://github.com/douglascraigschmidt/LiveLessons/tree/master/Java8/ex12

5

Overview of Java 8 Streams

6

• Java 8 streams are a framework inthe Java class library that provides several key benefits to programs


See docs.oracle.com/javase/tutorial/collections/streams

https://docs.oracle.com/javase/tutorial/collections/streams

7

• Java 8 streams are a framework inthe Java class library that provides several key benefits to programs, e.g.

• Manipulate flows of data declaratively via function composition

Overview of Java 8 Streams…

This stream expresses whatoperations to perform, not

how to perform them

See github.com/douglascraigschmidt/LiveLessons/tree/master/ImageStreamGang

filter(not(this::urlCached))

collect(toList())

map(this::downloadImage)

flatMap(this::applyFilters)

stream()

https://github.com/douglascraigschmidt/LiveLessons/tree/master/ImageStreamGang

8

…

filter(not(this::urlCached))

collect(toList())

map(this::downloadImage)

flatMap(this::applyFilters)

parallelStream()

• Java 8 streams are a framework inthe Java class library that provides several key benefits to programs, e.g.

• Manipulate flows of data declaratively via function composition

• Enable transparent parallelization without the need to write any multi-threaded code


The data elements in this stream are automatically

mapped to processor cores

See docs.oracle.com/javase/tutorial/collections/streams/parallelism.html

http://docs.oracle.com/javase/tutorial/collections/streams/parallelism.html

9

• A stream is a pipeline of aggregate operations that process a sequence of elements (aka, “values” or “data”)




See www.tutorialspoint.com/java8/java8_streams.htm

Input x

Output f(x)

Output g(f(x))


http://www.tutorialspoint.com/java8/java8_streams.htm

10






Input x

Output f(x)

Output g(f(x))


An aggregate operation is a higher-order function that applies a “behavior” parameter to every element in a stream

https://docs.oracle.com/javase/tutorial/collections/streams

11





Input x

Output f(x)

Output g(f(x))

A stream is conceptually unbounded, though it’s often bounded by practical constraints


12






Input x

Output f(x)

Output g(f(x))

Stream

.of("horatio",

"laertes",

"Hamlet", ...)


(s.charAt(0)) == 'h')


.sorted()




13

Stream source• A stream is created via a factory method

See en.wikipedia.org/wiki/Factory_method_pattern




Input x

Output f(x)

Output g(f(x))

Stream

.of("horatio",

"laertes",

"Hamlet", ...)

...


https://en.wikipedia.org/wiki/Factory_method_pattern

14





Input x

Output f(x)

Output g(f(x))

Stream

.of("horatio",

"laertes",

"Hamlet", ...) ...

Stream<String>

Array<String>

“horatio” …

…

“laertes” “Hamlet”

“horatio” “laertes” “Hamlet”

The of() factory method converts an array of T into a stream of T


See docs.oracle.com/javase/8/docs/api/java/util/stream/Stream.html#of

https://docs.oracle.com/javase/8/docs/api/java/util/stream/Stream.html#of-T...-

15





Input x

Output f(x)

Output g(f(x))

List<String> l1 = ...;



Stream

.of(l1, l2, l3)

.flatMap(List::stream)

...


of() is flexible, especially when combined with other aggregate operations




16

• A stream is created via a factory method




Input x

Output f(x)

Output g(f(x))

There are many other factory methods that create streams

collection.stream()

collection.parallelStream()

Pattern.compile(…).splitAsStream()

Stream.of(value1,… ,valueN)

Arrays.stream(array)

Arrays.stream(array, start, end)

Files.lines(file_path)

"string".chars()

Stream.builder().add(...)....build()

Stream.generate(generate_expression)

Files.list(file_path)

Files.find(file_path, max_depth, mathcher)

Stream.generate(iterator::next)

Stream.iterate(init_value, generate_expression)

StreamSupport.stream(iterable.spliterator(), false)

...

Overview of Java 8 StreamsStream source

17

• An aggregate operation performs a behavior on each element in a stream

A behavior is implemented by a lambda expression or method reference


Input x


18



Input x


Stream

.of("horatio",

"laertes",

"Hamlet", ...)


(s.charAt(0)) == 'h')


.sorted()



Stream<String>

Stream<String>

“horatio” “Hamlet”

“Horatio” “Hamlet”Method reference


19


• Ideally, a behavior’s output in a stream depends only on its input arguments


See en.wikipedia.org/wiki/Side_effect_(computer_science)

Input x

Output f(x)


https://en.wikipedia.org/wiki/Side_effect_(computer_science)

20




Input x

Output f(x)String capitalize(String s) {

if (s.length() == 0)

return s;

return s.substring(0, 1)

.toUpperCase()

+ s.substring(1)

.toLowerCase();




21



• Behaviors with side-effects likely incur race conditions in parallel streams


Input x

Output f(x)


See en.wikipedia.org/wiki/Race_condition#Software

Race conditions arise in software when an application depends on the sequence or

timing of threads for it to operate properly

https://en.wikipedia.org/wiki/Race_condition#Software

22In Java you must avoid race conditions, i.e., the compiler & JVM won’t save you..



• Behaviors with side-effects likely incur race conditions in parallel streams


Only you can prevent

race conditions!


Input x

Output f(x)

23

• Streams enhance flexibility by forming a “processing pipeline” that chains multiple aggregate operations together



See en.wikipedia.org/wiki/Pipeline_(software)

Input x

Output f(x)

Output g(f(x))


Output h(g(f(x)))


https://en.wikipedia.org/wiki/Pipeline_(software)

24

• Streams enhance flexibility by forming a “processing pipeline” that chains multiple aggregate operations together

of(“horatio”, “laertes”, “Hamlet”, …)

filter(s->toLowerCase(s.charAt(0)…)

Each aggregate operation in the pipeline can filter and/or transform the stream

Stream<String>

Stream<String>

Stream<String>

Stream<String>

Array<String>

“horatio” …

…

“laertes” “Hamlet”

“horatio” “laertes” “Hamlet”

“horatio” “Hamlet”

“Horatio” “Hamlet”

“Horatio”“Hamlet”

sorted()


map(this::capitalize)

List of names

Stream of names

Stream of names starting with ‘h’

Stream of capitalized names

Stream of sorted names

25

• A stream holds no non-transient storage



Input x

Output f(x)

Output g(f(x))


Output h(g(f(x)))


26

• Every stream works very similarly


27


• Starts with a source of data

e.g., a Java array, collection, generator function, or input channel

Stream

.of("horatio",

"laertes",

"Hamlet", ...)

...

…


28



e.g., a Java array, collection, generator function, or input channel

List<String> characters =

Arrays.asList("horatio",

"laertes",

"Hamlet", ...);

characters

.stream()

...

…


29



• Processes data through a pipeline of intermediate operations that each map an input stream to an output stream



Examples of intermediate operations include filter(), map(), & flatMap()

Input x

Output f(x)

Output g(f(x))

Stream

.of("horatio", "laertes",

"Hamlet", ...)


(s.charAt(0)) == 'h')


.sorted()

...

…


30




• Finishes with a terminal operation that yields a non-stream result




Output f(x)

Output g(f(x))...


(s.charAt(0)) == 'h')


.sorted()


…

Input x


31




Output f(x)

Output g(f(x))

…

Input x


A terminal operation triggers processing of intermediate operations in a stream





...


(s.charAt(0)) == 'h')


.sorted()


32






See www.youtube.com/watch?v=oWlWEKNM5AwEach stream must have one (& only one) terminal operation

http://www.youtube.com/watch?v=oWlWEKNM5Aw

33See radar.oreilly.com/2015/02/java-8-streams-api-and-parallelism.html

• Each aggregate operation in a stream runs its behavior sequentially by default




Output f(x)

Output g(f(x))

Output h(g(f(x)))

…

Input x


http://radar.oreilly.com/2015/02/java-8-streams-api-and-parallelism.html

34

• Each aggregate operation in a stream runs its behavior sequentially by default

• i.e., one at a time in a single thread




Output f(x)

Output g(f(x))

Output h(g(f(x)))

…

Input x



We’ll cover sequential streams first

http://docs.oracle.com/javase/tutorial/collections/streams

35

• A Java 8 parallel stream splits its elements into multiple chunks & uses a common fork-join pool to process the chunks independently


Output f(x)

Output g(f(x))



Output h(g(f(x)))

See docs.oracle.com/javase/tutorial/collections/streams/parallelism.html

…

Input x


We’ll cover parallel streams later

https://docs.oracle.com/javase/tutorial/collections/streams/parallelism.html

36

End of Overview of Java 8 Streams (Part 1)

Overview of Java 8 Streams (Part 1) - Vanderbilt Universityschmidt/cs891f/2018-PDFs/... · •Java 8 streams are a framework in the Java class library that provides several key benefits

Documents