1 /*
2 * Copyright (c) 2012, 2013, Oracle and/or its affiliates. All rights reserved.
3 * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
4 *
5 * This code is free software; you can redistribute it and/or modify it
6 * under the terms of the GNU General Public License version 2 only, as
7 * published by the Free Software Foundation. Oracle designates this
8 * particular file as subject to the "Classpath" exception as provided
9 * by Oracle in the LICENSE file that accompanied this code.
10 *
11 * This code is distributed in the hope that it will be useful, but WITHOUT
12 * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
13 * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License
14 * version 2 for more details (a copy is included in the LICENSE file that
15 * accompanied this code).
16 *
17 * You should have received a copy of the GNU General Public License version
18 * 2 along with this work; if not, write to the Free Software Foundation,
19 * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA.
20 *
21 * Please contact Oracle, 500 Oracle Parkway, Redwood Shores, CA 94065 USA
22 * or visit www.oracle.com if you need additional information or have any
23 * questions.
24 */
25 package java.util.stream;
26
27 import java.nio.charset.Charset;
28 import java.nio.file.Files;
29 import java.nio.file.Path;
30 import java.util.Arrays;
31 import java.util.Collection;
32 import java.util.Comparator;
33 import java.util.Iterator;
34 import java.util.Objects;
35 import java.util.Optional;
36 import java.util.Spliterator;
37 import java.util.Spliterators;
38 import java.util.concurrent.ConcurrentHashMap;
39 import java.util.function.BiConsumer;
40 import java.util.function.BiFunction;
41 import java.util.function.BinaryOperator;
42 import java.util.function.Consumer;
43 import java.util.function.Function;
44 import java.util.function.IntFunction;
45 import java.util.function.Predicate;
46 import java.util.function.Supplier;
47 import java.util.function.ToDoubleFunction;
48 import java.util.function.ToIntFunction;
49 import java.util.function.ToLongFunction;
50 import java.util.function.UnaryOperator;
51
52 /**
53 * A sequence of elements supporting sequential and parallel aggregate
54 * operations. The following example illustrates an aggregate operation using
55 * {@link Stream} and {@link IntStream}:
56 *
57 * <pre>{@code
58 * int sum = widgets.stream()
59 * .filter(w -> w.getColor() == RED)
60 * .mapToInt(w -> w.getWeight())
61 * .sum();
62 * }</pre>
63 *
64 * In this example, {@code widgets} is a {@code Collection<Widget>}. We create
65 * a stream of {@code Widget} objects via {@link Collection#stream Collection.stream()},
66 * filter it to produce a stream containing only the red widgets, and then
67 * transform it into a stream of {@code int} values representing the weight of
68 * each red widget. Then this stream is summed to produce a total weight.
69 *
70 * <p>In addition to {@code Stream}, which is a stream of object references,
71 * there are primitive specializations for {@link IntStream}, {@link LongStream},
72 * and {@link DoubleStream}, all of which are referred to as "streams" and
73 * conform to the characteristics and restrictions described here.
74 *
75 * <p>To perform a computation, stream
76 * <a href="package-summary.html#StreamOps">operations</a> are composed into a
77 * <em>stream pipeline</em>. A stream pipeline consists of a source (which
78 * might be an array, a collection, a generator function, an I/O channel,
79 * etc), zero or more <em>intermediate operations</em> (which transform a
80 * stream into another stream, such as {@link Stream#filter(Predicate)}), and a
81 * <em>terminal operation</em> (which produces a result or side-effect, such
82 * as {@link Stream#count()} or {@link Stream#forEach(Consumer)}).
83 * Streams are lazy; computation on the source data is only performed when the
84 * terminal operation is initiated, and source elements are consumed only
85 * as needed.
86 *
87 * <p>Collections and streams, while bearing some superficial similarities,
88 * have different goals. Collections are primarily concerned with the efficient
89 * management of, and access to, their elements. By contrast, streams do not
90 * provide a means to directly access or manipulate their elements, and are
91 * instead concerned with declaratively describing their source and the
92 * computational operations which will be performed in aggregate on that source.
93 * However, if the provided stream operations do not offer the desired
94 * functionality, the {@link #iterator()} and {@link #spliterator()} operations
95 * can be used to perform a controlled traversal.
96 *
97 * <p>A stream pipeline, like the "widgets" example above, can be viewed as
98 * a <em>query</em> on the stream source. Unless the source was explicitly
99 * designed for concurrent modification (such as a {@link ConcurrentHashMap}),
100 * unpredictable or erroneous behavior may result from modifying the stream
101 * source while it is being queried.
102 *
103 * <p>Most stream operations accept parameters that describe user-specified
104 * behavior, such as the lambda expression {@code w -> w.getWeight()} passed to
105 * {@code mapToInt} in the example above. To preserve correct behavior,
106 * these <em>behavioral parameters</em>:
107 * <ul>
108 * <li>must be <a href="package-summary.html#NonInterference">non-interfering</a>
109 * (they do not modify the stream source); and</li>
110 * <li>in most cases must be <a href="package-summary.html#Statelessness">stateless</a>
111 * (their result should not depend on any state that might change during execution
112 * of the stream pipeline).</li>
113 * </ul>
114 *
115 * <p>Such parameters are always instances of a
116 * <a href="../function/package-summary.html">functional interface</a> such
117 * as {@link java.util.function.Function}, and are often lambda expressions or
118 * method references. Unless otherwise specified these parameters must be
119 * <em>non-null</em>.
120 *
121 * <p>A stream should be operated on (invoking an intermediate or terminal stream
122 * operation) only once. This rules out, for example, "forked" streams, where
123 * the same source feeds two or more pipelines, or multiple traversals of the
124 * same stream. A stream implementation may throw {@link IllegalStateException}
125 * if it detects that the stream is being reused. However, since some stream
126 * operations may return their receiver rather than a new stream object, it may
127 * not be possible to detect reuse in all cases.
128 *
129 * <p>Streams have a {@link #close()} method and implement {@link AutoCloseable},
130 * but nearly all stream instances do not actually need to be closed after use.
131 * Generally, only streams whose source is an IO channel (such as those returned
132 * by {@link Files#lines(Path, Charset)}) will require closing. Most streams
133 * are backed by collections, arrays, or generating functions, which require no
134 * special resource management. (If a stream does require closing, it can be
135 * declared as a resource in a {@code try}-with-resources statement.)
136 *
137 * <p>Stream pipelines may execute either sequentially or in
138 * <a href="package-summary.html#Parallelism">parallel</a>. This
139 * execution mode is a property of the stream. Streams are created
140 * with an initial choice of sequential or parallel execution. (For example,
141 * {@link Collection#stream() Collection.stream()} creates a sequential stream,
142 * and {@link Collection#parallelStream() Collection.parallelStream()} creates
143 * a parallel one.) This choice of execution mode may be modified by the
144 * {@link #sequential()} or {@link #parallel()} methods, and may be queried with
145 * the {@link #isParallel()} method.
146 *
147 * @param <T> the type of the stream elements
148 * @since 1.8
149 * @see IntStream
150 * @see LongStream
151 * @see DoubleStream
152 * @see <a href="package-summary.html">java.util.stream</a>
153 */
154 public interface Stream<T> extends BaseStream<T, Stream<T>> {
155
156 /**
157 * Returns a stream consisting of the elements of this stream that match
158 * the given predicate.
159 *
160 * <p>This is an <a href="package-summary.html#StreamOps">intermediate
161 * operation</a>.
162 *
163 * @param predicate a <a href="package-summary.html#NonInterference">non-interfering</a>,
164 * <a href="package-summary.html#Statelessness">stateless</a>
165 * predicate to apply to each element to determine if it
166 * should be included
167 * @return the new stream
168 */
169 Stream<T> filter(Predicate<? super T> predicate);
170
171 /**
172 * Returns a stream consisting of the results of applying the given
173 * function to the elements of this stream.
174 *
175 * <p>This is an <a href="package-summary.html#StreamOps">intermediate
176 * operation</a>.
177 *
178 * @param <R> The element type of the new stream
179 * @param mapper a <a href="package-summary.html#NonInterference">non-interfering</a>,
180 * <a href="package-summary.html#Statelessness">stateless</a>
181 * function to apply to each element
182 * @return the new stream
183 */
184 <R> Stream<R> map(Function<? super T, ? extends R> mapper);
185
186 /**
187 * Returns an {@code IntStream} consisting of the results of applying the
188 * given function to the elements of this stream.
189 *
190 * <p>This is an <a href="package-summary.html#StreamOps">
191 * intermediate operation</a>.
192 *
193 * @param mapper a <a href="package-summary.html#NonInterference">non-interfering</a>,
194 * <a href="package-summary.html#Statelessness">stateless</a>
195 * function to apply to each element
196 * @return the new stream
197 */
198 IntStream mapToInt(ToIntFunction<? super T> mapper);
199
200 /**
201 * Returns a {@code LongStream} consisting of the results of applying the
202 * given function to the elements of this stream.
203 *
204 * <p>This is an <a href="package-summary.html#StreamOps">intermediate
205 * operation</a>.
206 *
207 * @param mapper a <a href="package-summary.html#NonInterference">non-interfering</a>,
208 * <a href="package-summary.html#Statelessness">stateless</a>
209 * function to apply to each element
210 * @return the new stream
211 */
212 LongStream mapToLong(ToLongFunction<? super T> mapper);
213
214 /**
215 * Returns a {@code DoubleStream} consisting of the results of applying the
216 * given function to the elements of this stream.
217 *
218 * <p>This is an <a href="package-summary.html#StreamOps">intermediate
219 * operation</a>.
220 *
221 * @param mapper a <a href="package-summary.html#NonInterference">non-interfering</a>,
222 * <a href="package-summary.html#Statelessness">stateless</a>
223 * function to apply to each element
224 * @return the new stream
225 */
226 DoubleStream mapToDouble(ToDoubleFunction<? super T> mapper);
227
228 /**
229 * Returns a stream consisting of the results of replacing each element of
230 * this stream with the contents of a mapped stream produced by applying
231 * the provided mapping function to each element. Each mapped stream is
232 * {@link java.util.stream.BaseStream#close() closed} after its contents
233 * have been placed into this stream. (If a mapped stream is {@code null}
234 * an empty stream is used, instead.)
235 *
236 * <p>This is an <a href="package-summary.html#StreamOps">intermediate
237 * operation</a>.
238 *
239 * @apiNote
240 * The {@code flatMap()} operation has the effect of applying a one-to-many
241 * transformation to the elements of the stream, and then flattening the
242 * resulting elements into a new stream.
243 *
244 * <p><b>Examples.</b>
245 *
246 * <p>If {@code orders} is a stream of purchase orders, and each purchase
247 * order contains a collection of line items, then the following produces a
248 * stream containing all the line items in all the orders:
249 * <pre>{@code
250 * orders.flatMap(order -> order.getLineItems().stream())...
251 * }</pre>
252 *
253 * <p>If {@code path} is the path to a file, then the following produces a
254 * stream of the {@code words} contained in that file:
255 * <pre>{@code
256 * Stream<String> lines = Files.lines(path, StandardCharsets.UTF_8);
257 * Stream<String> words = lines.flatMap(line -> Stream.of(line.split(" +")));
258 * }</pre>
259 * The {@code mapper} function passed to {@code flatMap} splits a line,
260 * using a simple regular expression, into an array of words, and then
261 * creates a stream of words from that array.
262 *
263 * @param <R> The element type of the new stream
264 * @param mapper a <a href="package-summary.html#NonInterference">non-interfering</a>,
265 * <a href="package-summary.html#Statelessness">stateless</a>
266 * function to apply to each element which produces a stream
267 * of new values
268 * @return the new stream
269 */
270 <R> Stream<R> flatMap(Function<? super T, ? extends Stream<? extends R>> mapper);
271
272 /**
273 * Returns an {@code IntStream} consisting of the results of replacing each
274 * element of this stream with the contents of a mapped stream produced by
275 * applying the provided mapping function to each element. Each mapped
276 * stream is {@link java.util.stream.BaseStream#close() closed} after its
277 * contents have been placed into this stream. (If a mapped stream is
278 * {@code null} an empty stream is used, instead.)
279 *
280 * <p>This is an <a href="package-summary.html#StreamOps">intermediate
281 * operation</a>.
282 *
283 * @param mapper a <a href="package-summary.html#NonInterference">non-interfering</a>,
284 * <a href="package-summary.html#Statelessness">stateless</a>
285 * function to apply to each element which produces a stream
286 * of new values
287 * @return the new stream
288 * @see #flatMap(Function)
289 */
290 IntStream flatMapToInt(Function<? super T, ? extends IntStream> mapper);
291
292 /**
293 * Returns an {@code LongStream} consisting of the results of replacing each
294 * element of this stream with the contents of a mapped stream produced by
295 * applying the provided mapping function to each element. Each mapped
296 * stream is {@link java.util.stream.BaseStream#close() closed} after its
297 * contents have been placed into this stream. (If a mapped stream is
298 * {@code null} an empty stream is used, instead.)
299 *
300 * <p>This is an <a href="package-summary.html#StreamOps">intermediate
301 * operation</a>.
302 *
303 * @param mapper a <a href="package-summary.html#NonInterference">non-interfering</a>,
304 * <a href="package-summary.html#Statelessness">stateless</a>
305 * function to apply to each element which produces a stream
306 * of new values
307 * @return the new stream
308 * @see #flatMap(Function)
309 */
310 LongStream flatMapToLong(Function<? super T, ? extends LongStream> mapper);
311
312 /**
313 * Returns an {@code DoubleStream} consisting of the results of replacing
314 * each element of this stream with the contents of a mapped stream produced
315 * by applying the provided mapping function to each element. Each mapped
316 * stream is {@link java.util.stream.BaseStream#close() closed} after its
317 * contents have placed been into this stream. (If a mapped stream is
318 * {@code null} an empty stream is used, instead.)
319 *
320 * <p>This is an <a href="package-summary.html#StreamOps">intermediate
321 * operation</a>.
322 *
323 * @param mapper a <a href="package-summary.html#NonInterference">non-interfering</a>,
324 * <a href="package-summary.html#Statelessness">stateless</a>
325 * function to apply to each element which produces a stream
326 * of new values
327 * @return the new stream
328 * @see #flatMap(Function)
329 */
330 DoubleStream flatMapToDouble(Function<? super T, ? extends DoubleStream> mapper);
331
332 /**
333 * Returns a stream consisting of the distinct elements (according to
334 * {@link Object#equals(Object)}) of this stream.
335 *
336 * <p>For ordered streams, the selection of distinct elements is stable
337 * (for duplicated elements, the element appearing first in the encounter
338 * order is preserved.) For unordered streams, no stability guarantees
339 * are made.
340 *
341 * <p>This is a <a href="package-summary.html#StreamOps">stateful
342 * intermediate operation</a>.
343 *
344 * @apiNote
345 * Preserving stability for {@code distinct()} in parallel pipelines is
346 * relatively expensive (requires that the operation act as a full barrier,
347 * with substantial buffering overhead), and stability is often not needed.
348 * Using an unordered stream source (such as {@link #generate(Supplier)})
349 * or removing the ordering constraint with {@link #unordered()} may result
350 * in significantly more efficient execution for {@code distinct()} in parallel
351 * pipelines, if the semantics of your situation permit. If consistency
352 * with encounter order is required, and you are experiencing poor performance
353 * or memory utilization with {@code distinct()} in parallel pipelines,
354 * switching to sequential execution with {@link #sequential()} may improve
355 * performance.
356 *
357 * @return the new stream
358 */
359 Stream<T> distinct();
360
361 /**
362 * Returns a stream consisting of the elements of this stream, sorted
363 * according to natural order. If the elements of this stream are not
364 * {@code Comparable}, a {@code java.lang.ClassCastException} may be thrown
365 * when the terminal operation is executed.
366 *
367 * <p>For ordered streams, the sort is stable. For unordered streams, no
368 * stability guarantees are made.
369 *
370 * <p>This is a <a href="package-summary.html#StreamOps">stateful
371 * intermediate operation</a>.
372 *
373 * @return the new stream
374 */
375 Stream<T> sorted();
376
377 /**
378 * Returns a stream consisting of the elements of this stream, sorted
379 * according to the provided {@code Comparator}.
380 *
381 * <p>For ordered streams, the sort is stable. For unordered streams, no
382 * stability guarantees are made.
383 *
384 * <p>This is a <a href="package-summary.html#StreamOps">stateful
385 * intermediate operation</a>.
386 *
387 * @param comparator a <a href="package-summary.html#NonInterference">non-interfering</a>,
388 * <a href="package-summary.html#Statelessness">stateless</a>
389 * {@code Comparator} to be used to compare stream elements
390 * @return the new stream
391 */
392 Stream<T> sorted(Comparator<? super T> comparator);
393
394 /**
395 * Returns a stream consisting of the elements of this stream, additionally
396 * performing the provided action on each element as elements are consumed
397 * from the resulting stream.
398 *
399 * <p>This is an <a href="package-summary.html#StreamOps">intermediate
400 * operation</a>.
401 *
402 * <p>For parallel stream pipelines, the action may be called at
403 * whatever time and in whatever thread the element is made available by the
404 * upstream operation. If the action modifies shared state,
405 * it is responsible for providing the required synchronization.
406 *
407 * @apiNote This method exists mainly to support debugging, where you want
408 * to see the elements as they flow past a certain point in a pipeline:
409 * <pre>{@code
410 * Stream.of("one", "two", "three", "four")
411 * .filter(e -> e.length() > 3)
412 * .peek(e -> System.out.println("Filtered value: " + e))
413 * .map(String::toUpperCase)
414 * .peek(e -> System.out.println("Mapped value: " + e))
415 * .collect(Collectors.toList());
416 * }</pre>
417 *
418 * @param action a <a href="package-summary.html#NonInterference">
419 * non-interfering</a> action to perform on the elements as
420 * they are consumed from the stream
421 * @return the new stream
422 */
423 Stream<T> peek(Consumer<? super T> action);
424
425 /**
426 * Returns a stream consisting of the elements of this stream, truncated
427 * to be no longer than {@code maxSize} in length.
428 *
429 * <p>This is a <a href="package-summary.html#StreamOps">short-circuiting
430 * stateful intermediate operation</a>.
431 *
432 * @apiNote
433 * While {@code limit()} is generally a cheap operation on sequential
434 * stream pipelines, it can be quite expensive on ordered parallel pipelines,
435 * especially for large values of {@code maxSize}, since {@code limit(n)}
436 * is constrained to return not just any <em>n</em> elements, but the
437 * <em>first n</em> elements in the encounter order. Using an unordered
438 * stream source (such as {@link #generate(Supplier)}) or removing the
439 * ordering constraint with {@link #unordered()} may result in significant
440 * speedups of {@code limit()} in parallel pipelines, if the semantics of
441 * your situation permit. If consistency with encounter order is required,
442 * and you are experiencing poor performance or memory utilization with
443 * {@code limit()} in parallel pipelines, switching to sequential execution
444 * with {@link #sequential()} may improve performance.
445 *
446 * @param maxSize the number of elements the stream should be limited to
447 * @return the new stream
448 * @throws IllegalArgumentException if {@code maxSize} is negative
449 */
450 Stream<T> limit(long maxSize);
451
452 /**
453 * Returns a stream consisting of the remaining elements of this stream
454 * after discarding the first {@code n} elements of the stream.
455 * If this stream contains fewer than {@code n} elements then an
456 * empty stream will be returned.
457 *
458 * <p>This is a <a href="package-summary.html#StreamOps">stateful
459 * intermediate operation</a>.
460 *
461 * @apiNote
462 * While {@code skip()} is generally a cheap operation on sequential
463 * stream pipelines, it can be quite expensive on ordered parallel pipelines,
464 * especially for large values of {@code n}, since {@code skip(n)}
465 * is constrained to skip not just any <em>n</em> elements, but the
466 * <em>first n</em> elements in the encounter order. Using an unordered
467 * stream source (such as {@link #generate(Supplier)}) or removing the
468 * ordering constraint with {@link #unordered()} may result in significant
469 * speedups of {@code skip()} in parallel pipelines, if the semantics of
470 * your situation permit. If consistency with encounter order is required,
471 * and you are experiencing poor performance or memory utilization with
472 * {@code skip()} in parallel pipelines, switching to sequential execution
473 * with {@link #sequential()} may improve performance.
474 *
475 * @param n the number of leading elements to skip
476 * @return the new stream
477 * @throws IllegalArgumentException if {@code n} is negative
478 */
479 Stream<T> skip(long n);
480
481 /**
482 * Performs an action for each element of this stream.
483 *
484 * <p>This is a <a href="package-summary.html#StreamOps">terminal
485 * operation</a>.
486 *
487 * <p>The behavior of this operation is explicitly nondeterministic.
488 * For parallel stream pipelines, this operation does <em>not</em>
489 * guarantee to respect the encounter order of the stream, as doing so
490 * would sacrifice the benefit of parallelism. For any given element, the
491 * action may be performed at whatever time and in whatever thread the
492 * library chooses. If the action accesses shared state, it is
493 * responsible for providing the required synchronization.
494 *
495 * @param action a <a href="package-summary.html#NonInterference">
496 * non-interfering</a> action to perform on the elements
497 */
498 void forEach(Consumer<? super T> action);
499
500 /**
501 * Performs an action for each element of this stream, in the encounter
502 * order of the stream if the stream has a defined encounter order.
503 *
504 * <p>This is a <a href="package-summary.html#StreamOps">terminal
505 * operation</a>.
506 *
507 * <p>This operation processes the elements one at a time, in encounter
508 * order if one exists. Performing the action for one element
509 * <a href="../concurrent/package-summary.html#MemoryVisibility"><i>happens-before</i></a>
510 * performing the action for subsequent elements, but for any given element,
511 * the action may be performed in whatever thread the library chooses.
512 *
513 * @param action a <a href="package-summary.html#NonInterference">
514 * non-interfering</a> action to perform on the elements
515 * @see #forEach(Consumer)
516 */
517 void forEachOrdered(Consumer<? super T> action);
518
519 /**
520 * Returns an array containing the elements of this stream.
521 *
522 * <p>This is a <a href="package-summary.html#StreamOps">terminal
523 * operation</a>.
524 *
525 * @return an array containing the elements of this stream
526 */
527 Object[] toArray();
528
529 /**
530 * Returns an array containing the elements of this stream, using the
531 * provided {@code generator} function to allocate the returned array, as
532 * well as any additional arrays that might be required for a partitioned
533 * execution or for resizing.
534 *
535 * <p>This is a <a href="package-summary.html#StreamOps">terminal
536 * operation</a>.
537 *
538 * @apiNote
539 * The generator function takes an integer, which is the size of the
540 * desired array, and produces an array of the desired size. This can be
541 * concisely expressed with an array constructor reference:
542 * <pre>{@code
543 * Person[] men = people.stream()
544 * .filter(p -> p.getGender() == MALE)
545 * .toArray(Person[]::new);
546 * }</pre>
547 *
548 * @param <A> the element type of the resulting array
549 * @param generator a function which produces a new array of the desired
550 * type and the provided length
551 * @return an array containing the elements in this stream
552 * @throws ArrayStoreException if the runtime type of the array returned
553 * from the array generator is not a supertype of the runtime type of every
554 * element in this stream
555 */
556 <A> A[] toArray(IntFunction<A[]> generator);
557
558 /**
559 * Performs a <a href="package-summary.html#Reduction">reduction</a> on the
560 * elements of this stream, using the provided identity value and an
561 * <a href="package-summary.html#Associativity">associative</a>
562 * accumulation function, and returns the reduced value. This is equivalent
563 * to:
564 * <pre>{@code
565 * T result = identity;
566 * for (T element : this stream)
567 * result = accumulator.apply(result, element)
568 * return result;
569 * }</pre>
570 *
571 * but is not constrained to execute sequentially.
572 *
573 * <p>The {@code identity} value must be an identity for the accumulator
574 * function. This means that for all {@code t},
575 * {@code accumulator.apply(identity, t)} is equal to {@code t}.
576 * The {@code accumulator} function must be an
577 * <a href="package-summary.html#Associativity">associative</a> function.
578 *
579 * <p>This is a <a href="package-summary.html#StreamOps">terminal
580 * operation</a>.
581 *
582 * @apiNote Sum, min, max, average, and string concatenation are all special
583 * cases of reduction. Summing a stream of numbers can be expressed as:
584 *
585 * <pre>{@code
586 * Integer sum = integers.reduce(0, (a, b) -> a+b);
587 * }</pre>
588 *
589 * or:
590 *
591 * <pre>{@code
592 * Integer sum = integers.reduce(0, Integer::sum);
593 * }</pre>
594 *
595 * <p>While this may seem a more roundabout way to perform an aggregation
596 * compared to simply mutating a running total in a loop, reduction
597 * operations parallelize more gracefully, without needing additional
598 * synchronization and with greatly reduced risk of data races.
599 *
600 * @param identity the identity value for the accumulating function
601 * @param accumulator an <a href="package-summary.html#Associativity">associative</a>,
602 * <a href="package-summary.html#NonInterference">non-interfering</a>,
603 * <a href="package-summary.html#Statelessness">stateless</a>
604 * function for combining two values
605 * @return the result of the reduction
606 */
607 T reduce(T identity, BinaryOperator<T> accumulator);
608
609 /**
610 * Performs a <a href="package-summary.html#Reduction">reduction</a> on the
611 * elements of this stream, using an
612 * <a href="package-summary.html#Associativity">associative</a> accumulation
613 * function, and returns an {@code Optional} describing the reduced value,
614 * if any. This is equivalent to:
615 * <pre>{@code
616 * boolean foundAny = false;
617 * T result = null;
618 * for (T element : this stream) {
619 * if (!foundAny) {
620 * foundAny = true;
621 * result = element;
622 * }
623 * else
624 * result = accumulator.apply(result, element);
625 * }
626 * return foundAny ? Optional.of(result) : Optional.empty();
627 * }</pre>
628 *
629 * but is not constrained to execute sequentially.
630 *
631 * <p>The {@code accumulator} function must be an
632 * <a href="package-summary.html#Associativity">associative</a> function.
633 *
634 * <p>This is a <a href="package-summary.html#StreamOps">terminal
635 * operation</a>.
636 *
637 * @param accumulator an <a href="package-summary.html#Associativity">associative</a>,
638 * <a href="package-summary.html#NonInterference">non-interfering</a>,
639 * <a href="package-summary.html#Statelessness">stateless</a>
640 * function for combining two values
641 * @return an {@link Optional} describing the result of the reduction
642 * @throws NullPointerException if the result of the reduction is null
643 * @see #reduce(Object, BinaryOperator)
644 * @see #min(Comparator)
645 * @see #max(Comparator)
646 */
647 Optional<T> reduce(BinaryOperator<T> accumulator);
648
649 /**
650 * Performs a <a href="package-summary.html#Reduction">reduction</a> on the
651 * elements of this stream, using the provided identity, accumulation and
652 * combining functions. This is equivalent to:
653 * <pre>{@code
654 * U result = identity;
655 * for (T element : this stream)
656 * result = accumulator.apply(result, element)
657 * return result;
658 * }</pre>
659 *
660 * but is not constrained to execute sequentially.
661 *
662 * <p>The {@code identity} value must be an identity for the combiner
663 * function. This means that for all {@code u}, {@code combiner(identity, u)}
664 * is equal to {@code u}. Additionally, the {@code combiner} function
665 * must be compatible with the {@code accumulator} function; for all
666 * {@code u} and {@code t}, the following must hold:
667 * <pre>{@code
668 * combiner.apply(u, accumulator.apply(identity, t)) == accumulator.apply(u, t)
669 * }</pre>
670 *
671 * <p>This is a <a href="package-summary.html#StreamOps">terminal
672 * operation</a>.
673 *
674 * @apiNote Many reductions using this form can be represented more simply
675 * by an explicit combination of {@code map} and {@code reduce} operations.
676 * The {@code accumulator} function acts as a fused mapper and accumulator,
677 * which can sometimes be more efficient than separate mapping and reduction,
678 * such as when knowing the previously reduced value allows you to avoid
679 * some computation.
680 *
681 * @param <U> The type of the result
682 * @param identity the identity value for the combiner function
683 * @param accumulator an <a href="package-summary.html#Associativity">associative</a>,
684 * <a href="package-summary.html#NonInterference">non-interfering</a>,
685 * <a href="package-summary.html#Statelessness">stateless</a>
686 * function for incorporating an additional element into a result
687 * @param combiner an <a href="package-summary.html#Associativity">associative</a>,
688 * <a href="package-summary.html#NonInterference">non-interfering</a>,
689 * <a href="package-summary.html#Statelessness">stateless</a>
690 * function for combining two values, which must be
691 * compatible with the accumulator function
692 * @return the result of the reduction
693 * @see #reduce(BinaryOperator)
694 * @see #reduce(Object, BinaryOperator)
695 */
696 <U> U reduce(U identity,
697 BiFunction<U, ? super T, U> accumulator,
698 BinaryOperator<U> combiner);
699
700 /**
701 * Performs a <a href="package-summary.html#MutableReduction">mutable
702 * reduction</a> operation on the elements of this stream. A mutable
703 * reduction is one in which the reduced value is a mutable result container,
704 * such as an {@code ArrayList}, and elements are incorporated by updating
705 * the state of the result rather than by replacing the result. This
706 * produces a result equivalent to:
707 * <pre>{@code
708 * R result = supplier.get();
709 * for (T element : this stream)
710 * accumulator.accept(result, element);
711 * return result;
712 * }</pre>
713 *
714 * <p>Like {@link #reduce(Object, BinaryOperator)}, {@code collect} operations
715 * can be parallelized without requiring additional synchronization.
716 *
717 * <p>This is a <a href="package-summary.html#StreamOps">terminal
718 * operation</a>.
719 *
720 * @apiNote There are many existing classes in the JDK whose signatures are
721 * well-suited for use with method references as arguments to {@code collect()}.
722 * For example, the following will accumulate strings into an {@code ArrayList}:
723 * <pre>{@code
724 * List<String> asList = stringStream.collect(ArrayList::new, ArrayList::add,
725 * ArrayList::addAll);
726 * }</pre>
727 *
728 * <p>The following will take a stream of strings and concatenates them into a
729 * single string:
730 * <pre>{@code
731 * String concat = stringStream.collect(StringBuilder::new, StringBuilder::append,
732 * StringBuilder::append)
733 * .toString();
734 * }</pre>
735 *
736 * @param <R> type of the result
737 * @param supplier a function that creates a new result container. For a
738 * parallel execution, this function may be called
739 * multiple times and must return a fresh value each time.
740 * @param accumulator an <a href="package-summary.html#Associativity">associative</a>,
741 * <a href="package-summary.html#NonInterference">non-interfering</a>,
742 * <a href="package-summary.html#Statelessness">stateless</a>
743 * function for incorporating an additional element into a result
744 * @param combiner an <a href="package-summary.html#Associativity">associative</a>,
745 * <a href="package-summary.html#NonInterference">non-interfering</a>,
746 * <a href="package-summary.html#Statelessness">stateless</a>
747 * function for combining two values, which must be
748 * compatible with the accumulator function
749 * @return the result of the reduction
750 */
751 <R> R collect(Supplier<R> supplier,
752 BiConsumer<R, ? super T> accumulator,
753 BiConsumer<R, R> combiner);
754
755 /**
756 * Performs a <a href="package-summary.html#MutableReduction">mutable
757 * reduction</a> operation on the elements of this stream using a
758 * {@code Collector}. A {@code Collector}
759 * encapsulates the functions used as arguments to
760 * {@link #collect(Supplier, BiConsumer, BiConsumer)}, allowing for reuse of
761 * collection strategies and composition of collect operations such as
762 * multiple-level grouping or partitioning.
763 *
764 * <p>If the stream is parallel, and the {@code Collector}
765 * is {@link Collector.Characteristics#CONCURRENT concurrent}, and
766 * either the stream is unordered or the collector is
767 * {@link Collector.Characteristics#UNORDERED unordered},
768 * then a concurrent reduction will be performed (see {@link Collector} for
769 * details on concurrent reduction.)
770 *
771 * <p>This is a <a href="package-summary.html#StreamOps">terminal
772 * operation</a>.
773 *
774 * <p>When executed in parallel, multiple intermediate results may be
775 * instantiated, populated, and merged so as to maintain isolation of
776 * mutable data structures. Therefore, even when executed in parallel
777 * with non-thread-safe data structures (such as {@code ArrayList}), no
778 * additional synchronization is needed for a parallel reduction.
779 *
780 * @apiNote
781 * The following will accumulate strings into an ArrayList:
782 * <pre>{@code
783 * List<String> asList = stringStream.collect(Collectors.toList());
784 * }</pre>
785 *
786 * <p>The following will classify {@code Person} objects by city:
787 * <pre>{@code
788 * Map<String, List<Person>> peopleByCity
789 * = personStream.collect(Collectors.groupingBy(Person::getCity));
790 * }</pre>
791 *
792 * <p>The following will classify {@code Person} objects by state and city,
793 * cascading two {@code Collector}s together:
794 * <pre>{@code
795 * Map<String, Map<String, List<Person>>> peopleByStateAndCity
796 * = personStream.collect(Collectors.groupingBy(Person::getState,
797 * Collectors.groupingBy(Person::getCity)));
798 * }</pre>
799 *
800 * @param <R> the type of the result
801 * @param <A> the intermediate accumulation type of the {@code Collector}
802 * @param collector the {@code Collector} describing the reduction
803 * @return the result of the reduction
804 * @see #collect(Supplier, BiConsumer, BiConsumer)
805 * @see Collectors
806 */
807 <R, A> R collect(Collector<? super T, A, R> collector);
808
809 /**
810 * Returns the minimum element of this stream according to the provided
811 * {@code Comparator}. This is a special case of a
812 * <a href="package-summary.html#Reduction">reduction</a>.
813 *
814 * <p>This is a <a href="package-summary.html#StreamOps">terminal operation</a>.
815 *
816 * @param comparator a <a href="package-summary.html#NonInterference">non-interfering</a>,
817 * <a href="package-summary.html#Statelessness">stateless</a>
818 * {@code Comparator} to compare elements of this stream
819 * @return an {@code Optional} describing the minimum element of this stream,
820 * or an empty {@code Optional} if the stream is empty
821 * @throws NullPointerException if the minimum element is null
822 */
823 Optional<T> min(Comparator<? super T> comparator);
824
825 /**
826 * Returns the maximum element of this stream according to the provided
827 * {@code Comparator}. This is a special case of a
828 * <a href="package-summary.html#Reduction">reduction</a>.
829 *
830 * <p>This is a <a href="package-summary.html#StreamOps">terminal
831 * operation</a>.
832 *
833 * @param comparator a <a href="package-summary.html#NonInterference">non-interfering</a>,
834 * <a href="package-summary.html#Statelessness">stateless</a>
835 * {@code Comparator} to compare elements of this stream
836 * @return an {@code Optional} describing the maximum element of this stream,
837 * or an empty {@code Optional} if the stream is empty
838 * @throws NullPointerException if the maximum element is null
839 */
840 Optional<T> max(Comparator<? super T> comparator);
841
842 /**
843 * Returns the count of elements in this stream. This is a special case of
844 * a <a href="package-summary.html#Reduction">reduction</a> and is
845 * equivalent to:
846 * <pre>{@code
847 * return mapToLong(e -> 1L).sum();
848 * }</pre>
849 *
850 * <p>This is a <a href="package-summary.html#StreamOps">terminal operation</a>.
851 *
852 * @return the count of elements in this stream
853 */
854 long count();
855
856 /**
857 * Returns whether any elements of this stream match the provided
858 * predicate. May not evaluate the predicate on all elements if not
859 * necessary for determining the result. If the stream is empty then
860 * {@code false} is returned and the predicate is not evaluated.
861 *
862 * <p>This is a <a href="package-summary.html#StreamOps">short-circuiting
863 * terminal operation</a>.
864 *
865 * @apiNote
866 * This method evaluates the <em>existential quantification</em> of the
867 * predicate over the elements of the stream (for some x P(x)).
868 *
869 * @param predicate a <a href="package-summary.html#NonInterference">non-interfering</a>,
870 * <a href="package-summary.html#Statelessness">stateless</a>
871 * predicate to apply to elements of this stream
872 * @return {@code true} if any elements of the stream match the provided
873 * predicate, otherwise {@code false}
874 */
875 boolean anyMatch(Predicate<? super T> predicate);
876
877 /**
878 * Returns whether all elements of this stream match the provided predicate.
879 * May not evaluate the predicate on all elements if not necessary for
880 * determining the result. If the stream is empty then {@code true} is
881 * returned and the predicate is not evaluated.
882 *
883 * <p>This is a <a href="package-summary.html#StreamOps">short-circuiting
884 * terminal operation</a>.
885 *
886 * @apiNote
887 * This method evaluates the <em>universal quantification</em> of the
888 * predicate over the elements of the stream (for all x P(x)). If the
889 * stream is empty, the quantification is said to be <em>vacuously
890 * satisfied</em> and is always {@code true} (regardless of P(x)).
891 *
892 * @param predicate a <a href="package-summary.html#NonInterference">non-interfering</a>,
893 * <a href="package-summary.html#Statelessness">stateless</a>
894 * predicate to apply to elements of this stream
895 * @return {@code true} if either all elements of the stream match the
896 * provided predicate or the stream is empty, otherwise {@code false}
897 */
898 boolean allMatch(Predicate<? super T> predicate);
899
900 /**
901 * Returns whether no elements of this stream match the provided predicate.
902 * May not evaluate the predicate on all elements if not necessary for
903 * determining the result. If the stream is empty then {@code true} is
904 * returned and the predicate is not evaluated.
905 *
906 * <p>This is a <a href="package-summary.html#StreamOps">short-circuiting
907 * terminal operation</a>.
908 *
909 * @apiNote
910 * This method evaluates the <em>universal quantification</em> of the
911 * negated predicate over the elements of the stream (for all x ~P(x)). If
912 * the stream is empty, the quantification is said to be vacuously satisfied
913 * and is always {@code true}, regardless of P(x).
914 *
915 * @param predicate a <a href="package-summary.html#NonInterference">non-interfering</a>,
916 * <a href="package-summary.html#Statelessness">stateless</a>
917 * predicate to apply to elements of this stream
918 * @return {@code true} if either no elements of the stream match the
919 * provided predicate or the stream is empty, otherwise {@code false}
920 */
921 boolean noneMatch(Predicate<? super T> predicate);
922
923 /**
924 * Returns an {@link Optional} describing the first element of this stream,
925 * or an empty {@code Optional} if the stream is empty. If the stream has
926 * no encounter order, then any element may be returned.
927 *
928 * <p>This is a <a href="package-summary.html#StreamOps">short-circuiting
929 * terminal operation</a>.
930 *
931 * @return an {@code Optional} describing the first element of this stream,
932 * or an empty {@code Optional} if the stream is empty
933 * @throws NullPointerException if the element selected is null
934 */
935 Optional<T> findFirst();
936
937 /**
938 * Returns an {@link Optional} describing some element of the stream, or an
939 * empty {@code Optional} if the stream is empty.
940 *
941 * <p>This is a <a href="package-summary.html#StreamOps">short-circuiting
942 * terminal operation</a>.
943 *
944 * <p>The behavior of this operation is explicitly nondeterministic; it is
945 * free to select any element in the stream. This is to allow for maximal
946 * performance in parallel operations; the cost is that multiple invocations
947 * on the same source may not return the same result. (If a stable result
948 * is desired, use {@link #findFirst()} instead.)
949 *
950 * @return an {@code Optional} describing some element of this stream, or an
951 * empty {@code Optional} if the stream is empty
952 * @throws NullPointerException if the element selected is null
953 * @see #findFirst()
954 */
955 Optional<T> findAny();
956
957 // Static factories
958
959 /**
960 * Returns a builder for a {@code Stream}.
961 *
962 * @param <T> type of elements
963 * @return a stream builder
964 */
965 public static<T> Builder<T> builder() {
966 return new Streams.StreamBuilderImpl<>();
967 }
968
969 /**
970 * Returns an empty sequential {@code Stream}.
971 *
972 * @param <T> the type of stream elements
973 * @return an empty sequential stream
974 */
975 public static<T> Stream<T> empty() {
976 return StreamSupport.stream(Spliterators.<T>emptySpliterator(), false);
977 }
978
979 /**
980 * Returns a sequential {@code Stream} containing a single element.
981 *
982 * @param t the single element
983 * @param <T> the type of stream elements
984 * @return a singleton sequential stream
985 */
986 public static<T> Stream<T> of(T t) {
987 return StreamSupport.stream(new Streams.StreamBuilderImpl<>(t), false);
988 }
989
990 /**
991 * Returns a sequential {@code Stream} containing a single element, if
992 * non-null, otherwise returns an empty {@code Stream}.
993 *
994 * @param t the single element
995 * @param <T> the type of stream elements
996 * @return a stream with a single element if the specified element
997 * is non-null, otherwise an empty stream
998 * @since 1.9
999 */
1000 public static<T> Stream<T> ofNullable(T t) {
1001 return t == null ? Stream.empty()
1002 : StreamSupport.stream(new Streams.StreamBuilderImpl<>(t), false);
1003 }
1004
1005 /**
1006 * Returns a sequential ordered stream whose elements are the specified values.
1007 *
1008 * @param <T> the type of stream elements
1009 * @param values the elements of the new stream
1010 * @return the new stream
1011 */
1012 @SafeVarargs
1013 @SuppressWarnings("varargs") // Creating a stream from an array is safe
1014 public static<T> Stream<T> of(T... values) {
1015 return Arrays.stream(values);
1016 }
1017
1018 /**
1019 * Returns an infinite sequential ordered {@code Stream} produced by iterative
1020 * application of a function {@code f} to an initial element {@code seed},
1021 * producing a {@code Stream} consisting of {@code seed}, {@code f(seed)},
1022 * {@code f(f(seed))}, etc.
1023 *
1024 * <p>The first element (position {@code 0}) in the {@code Stream} will be
1025 * the provided {@code seed}. For {@code n > 0}, the element at position
1026 * {@code n}, will be the result of applying the function {@code f} to the
1027 * element at position {@code n - 1}.
1028 *
1029 * @param <T> the type of stream elements
1030 * @param seed the initial element
1031 * @param f a function to be applied to the previous element to produce
1032 * a new element
1033 * @return a new sequential {@code Stream}
1034 */
1035 public static<T> Stream<T> iterate(final T seed, final UnaryOperator<T> f) {
1036 Objects.requireNonNull(f);
1037 final Iterator<T> iterator = new Iterator<T>() {
1038 @SuppressWarnings("unchecked")
1039 T t = (T) Streams.NONE;
1040
1041 @Override
1042 public boolean hasNext() {
1043 return true;
1044 }
1045
1046 @Override
1047 public T next() {
1048 return t = (t == Streams.NONE) ? seed : f.apply(t);
1049 }
1050 };
1051 return StreamSupport.stream(Spliterators.spliteratorUnknownSize(
1052 iterator,
1053 Spliterator.ORDERED | Spliterator.IMMUTABLE), false);
1054 }
1055
1056 /**
1057 * Returns an infinite sequential unordered stream where each element is
1058 * generated by the provided {@code Supplier}. This is suitable for
1059 * generating constant streams, streams of random elements, etc.
1060 *
1061 * @param <T> the type of stream elements
1062 * @param s the {@code Supplier} of generated elements
1063 * @return a new infinite sequential unordered {@code Stream}
1064 */
1065 public static<T> Stream<T> generate(Supplier<T> s) {
1066 Objects.requireNonNull(s);
1067 return StreamSupport.stream(
1068 new StreamSpliterators.InfiniteSupplyingSpliterator.OfRef<>(Long.MAX_VALUE, s), false);
1069 }
1070
1071 /**
1072 * Creates a lazily concatenated stream whose elements are all the
1073 * elements of the first stream followed by all the elements of the
1074 * second stream. The resulting stream is ordered if both
1075 * of the input streams are ordered, and parallel if either of the input
1076 * streams is parallel. When the resulting stream is closed, the close
1077 * handlers for both input streams are invoked.
1078 *
1079 * @implNote
1080 * Use caution when constructing streams from repeated concatenation.
1081 * Accessing an element of a deeply concatenated stream can result in deep
1082 * call chains, or even {@code StackOverflowError}.
1083 *
1084 * @param <T> The type of stream elements
1085 * @param a the first stream
1086 * @param b the second stream
1087 * @return the concatenation of the two input streams
1088 */
1089 public static <T> Stream<T> concat(Stream<? extends T> a, Stream<? extends T> b) {
1090 Objects.requireNonNull(a);
1091 Objects.requireNonNull(b);
1092
1093 @SuppressWarnings("unchecked")
1094 Spliterator<T> split = new Streams.ConcatSpliterator.OfRef<>(
1095 (Spliterator<T>) a.spliterator(), (Spliterator<T>) b.spliterator());
1096 Stream<T> stream = StreamSupport.stream(split, a.isParallel() || b.isParallel());
1097 return stream.onClose(Streams.composedClose(a, b));
1098 }
1099
1100 /**
1101 * A mutable builder for a {@code Stream}. This allows the creation of a
1102 * {@code Stream} by generating elements individually and adding them to the
1103 * {@code Builder} (without the copying overhead that comes from using
1104 * an {@code ArrayList} as a temporary buffer.)
1105 *
1106 * <p>A stream builder has a lifecycle, which starts in a building
1107 * phase, during which elements can be added, and then transitions to a built
1108 * phase, after which elements may not be added. The built phase begins
1109 * when the {@link #build()} method is called, which creates an ordered
1110 * {@code Stream} whose elements are the elements that were added to the stream
1111 * builder, in the order they were added.
1112 *
1113 * @param <T> the type of stream elements
1114 * @see Stream#builder()
1115 * @since 1.8
1116 */
1117 public interface Builder<T> extends Consumer<T> {
1118
1119 /**
1120 * Adds an element to the stream being built.
1121 *
1122 * @throws IllegalStateException if the builder has already transitioned to
1123 * the built state
1124 */
1125 @Override
1126 void accept(T t);
1127
1128 /**
1129 * Adds an element to the stream being built.
1130 *
1131 * @implSpec
1132 * The default implementation behaves as if:
1133 * <pre>{@code
1134 * accept(t)
1135 * return this;
1136 * }</pre>
1137 *
1138 * @param t the element to add
1139 * @return {@code this} builder
1140 * @throws IllegalStateException if the builder has already transitioned to
1141 * the built state
1142 */
1143 default Builder<T> add(T t) {
1144 accept(t);
1145 return this;
1146 }
1147
1148 /**
1149 * Builds the stream, transitioning this builder to the built state.
1150 * An {@code IllegalStateException} is thrown if there are further attempts
1151 * to operate on the builder after it has entered the built state.
1152 *
1153 * @return the built stream
1154 * @throws IllegalStateException if the builder has already transitioned to
1155 * the built state
1156 */
1157 Stream<T> build();
1158
1159 }
1160 }