001/*
002 * Licensed to the Apache Software Foundation (ASF) under one or more
003 * contributor license agreements.  See the NOTICE file distributed with
004 * this work for additional information regarding copyright ownership.
005 * The ASF licenses this file to You under the Apache License, Version 2.0
006 * (the "License"); you may not use this file except in compliance with
007 * the License.  You may obtain a copy of the License at
008 *
009 *      http://www.apache.org/licenses/LICENSE-2.0
010 *
011 * Unless required by applicable law or agreed to in writing, software
012 * distributed under the License is distributed on an "AS IS" BASIS,
013 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
014 * See the License for the specific language governing permissions and
015 * limitations under the License.
016 */
017package org.apache.commons.text;
018
019import java.util.ArrayList;
020import java.util.Collections;
021import java.util.HashSet;
022import java.util.List;
023import java.util.Set;
024import java.util.concurrent.ThreadLocalRandom;
025
026import org.apache.commons.lang3.ArrayUtils;
027import org.apache.commons.lang3.StringUtils;
028import org.apache.commons.lang3.Validate;
029
030/**
031 * Generates random Unicode strings containing the specified number of code points.
032 * Instances are created using a builder class, which allows the
033 * callers to define the properties of the generator. See the documentation for the
034 * {@link Builder} class to see available properties.
035 *
036 * <pre>
037 * // Generates a 20 code point string, using only the letters a-z
038 * RandomStringGenerator generator = RandomStringGenerator.builder()
039 *     .withinRange('a', 'z').build();
040 * String randomLetters = generator.generate(20);
041 * </pre>
042 * <pre>
043 * // Using Apache Commons RNG for randomness
044 * UniformRandomProvider rng = RandomSource.create(...);
045 * // Generates a 20 code point string, using only the letters a-z
046 * RandomStringGenerator generator = RandomStringGenerator.builder()
047 *     .withinRange('a', 'z')
048 *     .usingRandom(rng::nextInt) // uses Java 8 syntax
049 *     .build();
050 * String randomLetters = generator.generate(20);
051 * </pre>
052 * <p>
053 * {@code RandomStringGenerator} instances are thread-safe when using the
054 * default random number generator (RNG). If a custom RNG is set by calling the method
055 * {@link Builder#usingRandom(TextRandomProvider) Builder.usingRandom(TextRandomProvider)}, thread-safety
056 * must be ensured externally.
057 * </p>
058 * @since 1.1
059 */
060public final class RandomStringGenerator {
061
062    /**
063     * A builder for generating {@code RandomStringGenerator} instances.
064     *
065     * <p>The behavior of a generator is controlled by properties set by this
066     * builder. Each property has a default value, which can be overridden by
067     * calling the methods defined in this class, prior to calling {@link #build()}.</p>
068     *
069     * <p>All the property setting methods return the {@code Builder} instance to allow for method chaining.</p>
070     *
071     * <p>The minimum and maximum code point values are defined using {@link #withinRange(int, int)}. The
072     * default values are {@code 0} and {@link Character#MAX_CODE_POINT} respectively.</p>
073     *
074     * <p>The source of randomness can be set using {@link #usingRandom(TextRandomProvider)},
075     * otherwise {@link ThreadLocalRandom} is used.</p>
076     *
077     * <p>The type of code points returned can be filtered using {@link #filteredBy(CharacterPredicate...)},
078     * which defines a collection of tests that are applied to the randomly generated code points.
079     * The code points will only be included in the result if they pass at least one of the tests.
080     * Some commonly used predicates are provided by the {@link CharacterPredicates} enum.</p>
081     *
082     * <p>This class is not thread safe.</p>
083     * @since 1.1
084     */
085    public static class Builder implements org.apache.commons.text.Builder<RandomStringGenerator> {
086
087        /**
088         * The default maximum code point allowed: {@link Character#MAX_CODE_POINT}
089         * ({@value}).
090         */
091        public static final int DEFAULT_MAXIMUM_CODE_POINT = Character.MAX_CODE_POINT;
092
093        /**
094         * The default string length produced by this builder: {@value}.
095         */
096        public static final int DEFAULT_LENGTH = 0;
097
098        /**
099         * The default minimum code point allowed: {@value}.
100         */
101        public static final int DEFAULT_MINIMUM_CODE_POINT = 0;
102
103        /**
104         * The minimum code point allowed.
105         */
106        private int minimumCodePoint = DEFAULT_MINIMUM_CODE_POINT;
107
108        /**
109         * The maximum code point allowed.
110         */
111        private int maximumCodePoint = DEFAULT_MAXIMUM_CODE_POINT;
112
113        /**
114         * Filters for code points.
115         */
116        private Set<CharacterPredicate> inclusivePredicates;
117
118        /**
119         * The source of randomness.
120         */
121        private TextRandomProvider random;
122
123        /**
124         * The source of provided characters.
125         */
126        private List<Character> characterList;
127
128        /**
129         * Builds a new {@code RandomStringGenerator}.
130         *
131         * @return A new {@code RandomStringGenerator}
132         * @deprecated Use {@link #get()}.
133         */
134        @Deprecated
135        @Override
136        public RandomStringGenerator build() {
137            return get();
138        }
139
140        /**
141         * Limits the characters in the generated string to those that match at
142         * least one of the predicates supplied.
143         *
144         * <p>
145         * Passing {@code null} or an empty array to this method will revert to the
146         * default behavior of allowing any character. Multiple calls to this
147         * method will replace the previously stored predicates.
148         * </p>
149         *
150         * @param predicates
151         *            the predicates, may be {@code null} or empty
152         * @return {@code this}, to allow method chaining
153         */
154        public Builder filteredBy(final CharacterPredicate... predicates) {
155            if (ArrayUtils.isEmpty(predicates)) {
156                inclusivePredicates = null;
157                return this;
158            }
159            if (inclusivePredicates == null) {
160                inclusivePredicates = new HashSet<>();
161            } else {
162                inclusivePredicates.clear();
163            }
164            Collections.addAll(inclusivePredicates, predicates);
165            return this;
166        }
167
168        /**
169         * Builds a new {@code RandomStringGenerator}.
170         *
171         * @return A new {@code RandomStringGenerator}
172         * @since 1.12.0
173         */
174        @Override
175        public RandomStringGenerator get() {
176            return new RandomStringGenerator(minimumCodePoint, maximumCodePoint, inclusivePredicates,
177                    random, characterList);
178        }
179
180        /**
181         * Limits the characters in the generated string to those who match at
182         * supplied list of Character.
183         *
184         * <p>
185         * Passing {@code null} or an empty array to this method will revert to the
186         * default behavior of allowing any character. Multiple calls to this
187         * method will replace the previously stored Character.
188         * </p>
189         *
190         * @param chars set of predefined Characters for random string generation
191         *            the Character can be, may be {@code null} or empty
192         * @return {@code this}, to allow method chaining
193         * @since 1.2
194         */
195        public Builder selectFrom(final char... chars) {
196            characterList = new ArrayList<>();
197            if (chars != null) {
198                for (final char c : chars) {
199                    characterList.add(c);
200                }
201            }
202            return this;
203        }
204
205        /**
206         * Overrides the default source of randomness.  It is highly
207         * recommended that a random number generator library like
208         * <a href="https://commons.apache.org/proper/commons-rng/">Apache Commons RNG</a>
209         * be used to provide the random number generation.
210         *
211         * <p>
212         * When using Java 8 or later, {@link TextRandomProvider} is a
213         * functional interface and need not be explicitly implemented:
214         * </p>
215         * <pre>
216         * {@code
217         *     UniformRandomProvider rng = RandomSource.create(...);
218         *     RandomStringGenerator gen = RandomStringGenerator.builder()
219         *         .usingRandom(rng::nextInt)
220         *         // additional builder calls as needed
221         *         .build();
222         * }
223         * </pre>
224         *
225         * <p>
226         * Passing {@code null} to this method will revert to the default source of
227         * randomness.
228         * </p>
229         *
230         * @param random
231         *            the source of randomness, may be {@code null}
232         * @return {@code this}, to allow method chaining
233         */
234        public Builder usingRandom(final TextRandomProvider random) {
235            this.random = random;
236            return this;
237        }
238
239        /**
240         * Sets the array of minimum and maximum char allowed in the
241         * generated string.
242         *
243         * For example:
244         * <pre>
245         * {@code
246         *     char [][] pairs = {{'0','9'}};
247         *     char [][] pairs = {{'a','z'}};
248         *     char [][] pairs = {{'a','z'},{'0','9'}};
249         * }
250         * </pre>
251         *
252         * @param pairs array of characters array, expected is to pass min, max pairs through this arg.
253         * @return {@code this}, to allow method chaining.
254         */
255        public Builder withinRange(final char[]... pairs) {
256            characterList = new ArrayList<>();
257            if (pairs != null) {
258                for (final char[] pair : pairs) {
259                    Validate.isTrue(pair.length == 2, "Each pair must contain minimum and maximum code point");
260                    final int minimumCodePoint = pair[0];
261                    final int maximumCodePoint = pair[1];
262                    Validate.isTrue(minimumCodePoint <= maximumCodePoint, "Minimum code point %d is larger than maximum code point %d", minimumCodePoint,
263                            maximumCodePoint);
264
265                    for (int index = minimumCodePoint; index <= maximumCodePoint; index++) {
266                        characterList.add((char) index);
267                    }
268                }
269            }
270            return this;
271
272        }
273
274        /**
275         * Sets the minimum and maximum code points allowed in the
276         * generated string.
277         *
278         * @param minimumCodePoint
279         *            the smallest code point allowed (inclusive)
280         * @param maximumCodePoint
281         *            the largest code point allowed (inclusive)
282         * @return {@code this}, to allow method chaining
283         * @throws IllegalArgumentException
284         *             if {@code maximumCodePoint >}
285         *             {@link Character#MAX_CODE_POINT}
286         * @throws IllegalArgumentException
287         *             if {@code minimumCodePoint < 0}
288         * @throws IllegalArgumentException
289         *             if {@code minimumCodePoint > maximumCodePoint}
290         */
291        public Builder withinRange(final int minimumCodePoint, final int maximumCodePoint) {
292            Validate.isTrue(minimumCodePoint <= maximumCodePoint,
293                    "Minimum code point %d is larger than maximum code point %d", minimumCodePoint, maximumCodePoint);
294            Validate.isTrue(minimumCodePoint >= 0, "Minimum code point %d is negative", minimumCodePoint);
295            Validate.isTrue(maximumCodePoint <= Character.MAX_CODE_POINT,
296                    "Value %d is larger than Character.MAX_CODE_POINT.", maximumCodePoint);
297            this.minimumCodePoint = minimumCodePoint;
298            this.maximumCodePoint = maximumCodePoint;
299            return this;
300        }
301    }
302
303    /**
304     * Constructs a new builder.
305     * @return a new builder.
306     * @since 1.11.0
307     */
308    public static Builder builder() {
309        return new Builder();
310    }
311
312    /**
313     * The smallest allowed code point (inclusive).
314     */
315    private final int minimumCodePoint;
316
317    /**
318     * The largest allowed code point (inclusive).
319     */
320    private final int maximumCodePoint;
321
322    /**
323     * Filters for code points.
324     */
325    private final Set<CharacterPredicate> inclusivePredicates;
326
327    /**
328     * The source of randomness for this generator.
329     */
330    private final TextRandomProvider random;
331
332    /**
333     * The source of provided characters.
334     */
335    private final List<Character> characterList;
336
337    /**
338     * Constructs the generator.
339     *
340     * @param minimumCodePoint
341     *            smallest allowed code point (inclusive)
342     * @param maximumCodePoint
343     *            largest allowed code point (inclusive)
344     * @param inclusivePredicates
345     *            filters for code points
346     * @param random
347     *            source of randomness
348     * @param characterList list of predefined set of characters.
349     */
350    private RandomStringGenerator(final int minimumCodePoint, final int maximumCodePoint,
351                                  final Set<CharacterPredicate> inclusivePredicates, final TextRandomProvider random,
352                                  final List<Character> characterList) {
353        this.minimumCodePoint = minimumCodePoint;
354        this.maximumCodePoint = maximumCodePoint;
355        this.inclusivePredicates = inclusivePredicates;
356        this.random = random;
357        this.characterList = characterList;
358    }
359
360    /**
361     * Generates a random string, containing the specified number of code points.
362     *
363     * <p>
364     * Code points are randomly selected between the minimum and maximum values defined
365     * in the generator.
366     * Surrogate and private use characters are not returned, although the
367     * resulting string may contain pairs of surrogates that together encode a
368     * supplementary character.
369     * </p>
370     * <p>
371     * Note: the number of {@code char} code units generated will exceed
372     * {@code length} if the string contains supplementary characters. See the
373     * {@link Character} documentation to understand how Java stores Unicode
374     * values.
375     * </p>
376     *
377     * @param length
378     *            the number of code points to generate
379     * @return The generated string
380     * @throws IllegalArgumentException
381     *             if {@code length < 0}
382     */
383    public String generate(final int length) {
384        if (length == 0) {
385            return StringUtils.EMPTY;
386        }
387        Validate.isTrue(length > 0, "Length %d is smaller than zero.", length);
388        final StringBuilder builder = new StringBuilder(length);
389        long remaining = length;
390        do {
391            final int codePoint;
392            if (characterList != null && !characterList.isEmpty()) {
393                codePoint = generateRandomNumber(characterList);
394            } else {
395                codePoint = generateRandomNumber(minimumCodePoint, maximumCodePoint);
396            }
397            switch (Character.getType(codePoint)) {
398            case Character.UNASSIGNED:
399            case Character.PRIVATE_USE:
400            case Character.SURROGATE:
401                continue;
402            default:
403            }
404            if (inclusivePredicates != null) {
405                boolean matchedFilter = false;
406                for (final CharacterPredicate predicate : inclusivePredicates) {
407                    if (predicate.test(codePoint)) {
408                        matchedFilter = true;
409                        break;
410                    }
411                }
412                if (!matchedFilter) {
413                    continue;
414                }
415            }
416            builder.appendCodePoint(codePoint);
417            remaining--;
418        } while (remaining != 0);
419        return builder.toString();
420    }
421
422    /**
423     * Generates a random string, containing between the minimum (inclusive) and the maximum (inclusive)
424     * number of code points.
425     *
426     * @param minLengthInclusive
427     *            the minimum (inclusive) number of code points to generate
428     * @param maxLengthInclusive
429     *            the maximum (inclusive) number of code points to generate
430     * @return The generated string
431     * @throws IllegalArgumentException
432     *             if {@code minLengthInclusive < 0}, or {@code maxLengthInclusive < minLengthInclusive}
433     * @see RandomStringGenerator#generate(int)
434     * @since 1.2
435     */
436    public String generate(final int minLengthInclusive, final int maxLengthInclusive) {
437        Validate.isTrue(minLengthInclusive >= 0, "Minimum length %d is smaller than zero.", minLengthInclusive);
438        Validate.isTrue(minLengthInclusive <= maxLengthInclusive,
439                "Maximum length %d is smaller than minimum length %d.", maxLengthInclusive, minLengthInclusive);
440        return generate(generateRandomNumber(minLengthInclusive, maxLengthInclusive));
441    }
442
443    /**
444     * Generates a random number within a range, using a {@link ThreadLocalRandom} instance
445     * or the user-supplied source of randomness.
446     *
447     * @param minInclusive
448     *            the minimum value allowed
449     * @param maxInclusive
450     *            the maximum value allowed
451     * @return The random number.
452     */
453    private int generateRandomNumber(final int minInclusive, final int maxInclusive) {
454        if (random != null) {
455            return random.nextInt(maxInclusive - minInclusive + 1) + minInclusive;
456        }
457        return ThreadLocalRandom.current().nextInt(minInclusive, maxInclusive + 1);
458    }
459
460    /**
461     * Generates a random number within a range, using a {@link ThreadLocalRandom} instance
462     * or the user-supplied source of randomness.
463     *
464     * @param characterList predefined char list.
465     * @return The random number.
466     */
467    private int generateRandomNumber(final List<Character> characterList) {
468        final int listSize = characterList.size();
469        if (random != null) {
470            return String.valueOf(characterList.get(random.nextInt(listSize))).codePointAt(0);
471        }
472        return String.valueOf(characterList.get(ThreadLocalRandom.current().nextInt(0, listSize))).codePointAt(0);
473    }
474}