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 the {@code RandomStringGenerator} using the properties specified.
130         *
131         * @return The configured {@code RandomStringGenerator}
132         */
133        @Override
134        public RandomStringGenerator build() {
135            return new RandomStringGenerator(minimumCodePoint, maximumCodePoint, inclusivePredicates,
136                    random, characterList);
137        }
138
139        /**
140         * Limits the characters in the generated string to those that match at
141         * least one of the predicates supplied.
142         *
143         * <p>
144         * Passing {@code null} or an empty array to this method will revert to the
145         * default behavior of allowing any character. Multiple calls to this
146         * method will replace the previously stored predicates.
147         * </p>
148         *
149         * @param predicates
150         *            the predicates, may be {@code null} or empty
151         * @return {@code this}, to allow method chaining
152         */
153        public Builder filteredBy(final CharacterPredicate... predicates) {
154            if (ArrayUtils.isEmpty(predicates)) {
155                inclusivePredicates = null;
156                return this;
157            }
158
159            if (inclusivePredicates == null) {
160                inclusivePredicates = new HashSet<>();
161            } else {
162                inclusivePredicates.clear();
163            }
164
165            Collections.addAll(inclusivePredicates, predicates);
166
167            return this;
168        }
169
170        /**
171         * Limits the characters in the generated string to those who match at
172         * supplied list of Character.
173         *
174         * <p>
175         * Passing {@code null} or an empty array to this method will revert to the
176         * default behavior of allowing any character. Multiple calls to this
177         * method will replace the previously stored Character.
178         * </p>
179         *
180         * @param chars set of predefined Characters for random string generation
181         *            the Character can be, may be {@code null} or empty
182         * @return {@code this}, to allow method chaining
183         * @since 1.2
184         */
185        public Builder selectFrom(final char... chars) {
186            characterList = new ArrayList<>();
187            if (chars != null) {
188                for (final char c : chars) {
189                    characterList.add(c);
190                }
191            }
192            return this;
193        }
194
195        /**
196         * Overrides the default source of randomness.  It is highly
197         * recommended that a random number generator library like
198         * <a href="https://commons.apache.org/proper/commons-rng/">Apache Commons RNG</a>
199         * be used to provide the random number generation.
200         *
201         * <p>
202         * When using Java 8 or later, {@link TextRandomProvider} is a
203         * functional interface and need not be explicitly implemented:
204         * </p>
205         * <pre>
206         * {@code
207         *     UniformRandomProvider rng = RandomSource.create(...);
208         *     RandomStringGenerator gen = RandomStringGenerator.builder()
209         *         .usingRandom(rng::nextInt)
210         *         // additional builder calls as needed
211         *         .build();
212         * }
213         * </pre>
214         *
215         * <p>
216         * Passing {@code null} to this method will revert to the default source of
217         * randomness.
218         * </p>
219         *
220         * @param random
221         *            the source of randomness, may be {@code null}
222         * @return {@code this}, to allow method chaining
223         */
224        public Builder usingRandom(final TextRandomProvider random) {
225            this.random = random;
226            return this;
227        }
228
229        /**
230         * Sets the array of minimum and maximum char allowed in the
231         * generated string.
232         *
233         * For example:
234         * <pre>
235         * {@code
236         *     char [][] pairs = {{'0','9'}};
237         *     char [][] pairs = {{'a','z'}};
238         *     char [][] pairs = {{'a','z'},{'0','9'}};
239         * }
240         * </pre>
241         *
242         * @param pairs array of characters array, expected is to pass min, max pairs through this arg.
243         * @return {@code this}, to allow method chaining.
244         */
245        public Builder withinRange(final char[]... pairs) {
246            characterList = new ArrayList<>();
247            if (pairs != null) {
248                for (final char[] pair : pairs) {
249                    Validate.isTrue(pair.length == 2, "Each pair must contain minimum and maximum code point");
250                    final int minimumCodePoint = pair[0];
251                    final int maximumCodePoint = pair[1];
252                    Validate.isTrue(minimumCodePoint <= maximumCodePoint, "Minimum code point %d is larger than maximum code point %d", minimumCodePoint,
253                            maximumCodePoint);
254
255                    for (int index = minimumCodePoint; index <= maximumCodePoint; index++) {
256                        characterList.add((char) index);
257                    }
258                }
259            }
260            return this;
261
262        }
263
264        /**
265         * Sets the minimum and maximum code points allowed in the
266         * generated string.
267         *
268         * @param minimumCodePoint
269         *            the smallest code point allowed (inclusive)
270         * @param maximumCodePoint
271         *            the largest code point allowed (inclusive)
272         * @return {@code this}, to allow method chaining
273         * @throws IllegalArgumentException
274         *             if {@code maximumCodePoint >}
275         *             {@link Character#MAX_CODE_POINT}
276         * @throws IllegalArgumentException
277         *             if {@code minimumCodePoint < 0}
278         * @throws IllegalArgumentException
279         *             if {@code minimumCodePoint > maximumCodePoint}
280         */
281        public Builder withinRange(final int minimumCodePoint, final int maximumCodePoint) {
282            Validate.isTrue(minimumCodePoint <= maximumCodePoint,
283                    "Minimum code point %d is larger than maximum code point %d", minimumCodePoint, maximumCodePoint);
284            Validate.isTrue(minimumCodePoint >= 0, "Minimum code point %d is negative", minimumCodePoint);
285            Validate.isTrue(maximumCodePoint <= Character.MAX_CODE_POINT,
286                    "Value %d is larger than Character.MAX_CODE_POINT.", maximumCodePoint);
287
288            this.minimumCodePoint = minimumCodePoint;
289            this.maximumCodePoint = maximumCodePoint;
290            return this;
291        }
292    }
293
294    /**
295     * Constructs a new builder.
296     * @return a new builder.
297     *
298     * @since 1.11.0
299     */
300    public static Builder builder() {
301        return new Builder();
302    }
303
304    /**
305     * The smallest allowed code point (inclusive).
306     */
307    private final int minimumCodePoint;
308
309    /**
310     * The largest allowed code point (inclusive).
311     */
312    private final int maximumCodePoint;
313
314    /**
315     * Filters for code points.
316     */
317    private final Set<CharacterPredicate> inclusivePredicates;
318
319    /**
320     * The source of randomness for this generator.
321     */
322    private final TextRandomProvider random;
323
324    /**
325     * The source of provided characters.
326     */
327    private final List<Character> characterList;
328
329    /**
330     * Constructs the generator.
331     *
332     * @param minimumCodePoint
333     *            smallest allowed code point (inclusive)
334     * @param maximumCodePoint
335     *            largest allowed code point (inclusive)
336     * @param inclusivePredicates
337     *            filters for code points
338     * @param random
339     *            source of randomness
340     * @param characterList list of predefined set of characters.
341     */
342    private RandomStringGenerator(final int minimumCodePoint, final int maximumCodePoint,
343                                  final Set<CharacterPredicate> inclusivePredicates, final TextRandomProvider random,
344                                  final List<Character> characterList) {
345        this.minimumCodePoint = minimumCodePoint;
346        this.maximumCodePoint = maximumCodePoint;
347        this.inclusivePredicates = inclusivePredicates;
348        this.random = random;
349        this.characterList = characterList;
350    }
351
352    /**
353     * Generates a random string, containing the specified number of code points.
354     *
355     * <p>
356     * Code points are randomly selected between the minimum and maximum values defined
357     * in the generator.
358     * Surrogate and private use characters are not returned, although the
359     * resulting string may contain pairs of surrogates that together encode a
360     * supplementary character.
361     * </p>
362     * <p>
363     * Note: the number of {@code char} code units generated will exceed
364     * {@code length} if the string contains supplementary characters. See the
365     * {@link Character} documentation to understand how Java stores Unicode
366     * values.
367     * </p>
368     *
369     * @param length
370     *            the number of code points to generate
371     * @return The generated string
372     * @throws IllegalArgumentException
373     *             if {@code length < 0}
374     */
375    public String generate(final int length) {
376        if (length == 0) {
377            return StringUtils.EMPTY;
378        }
379        Validate.isTrue(length > 0, "Length %d is smaller than zero.", length);
380
381        final StringBuilder builder = new StringBuilder(length);
382        long remaining = length;
383
384        do {
385            final int codePoint;
386            if (characterList != null && !characterList.isEmpty()) {
387                codePoint = generateRandomNumber(characterList);
388            } else {
389                codePoint = generateRandomNumber(minimumCodePoint, maximumCodePoint);
390            }
391            switch (Character.getType(codePoint)) {
392            case Character.UNASSIGNED:
393            case Character.PRIVATE_USE:
394            case Character.SURROGATE:
395                continue;
396            default:
397            }
398
399            if (inclusivePredicates != null) {
400                boolean matchedFilter = false;
401                for (final CharacterPredicate predicate : inclusivePredicates) {
402                    if (predicate.test(codePoint)) {
403                        matchedFilter = true;
404                        break;
405                    }
406                }
407                if (!matchedFilter) {
408                    continue;
409                }
410            }
411
412            builder.appendCodePoint(codePoint);
413            remaining--;
414
415        } while (remaining != 0);
416
417        return builder.toString();
418    }
419
420    /**
421     * Generates a random string, containing between the minimum (inclusive) and the maximum (inclusive)
422     * number of code points.
423     *
424     * @param minLengthInclusive
425     *            the minimum (inclusive) number of code points to generate
426     * @param maxLengthInclusive
427     *            the maximum (inclusive) number of code points to generate
428     * @return The generated string
429     * @throws IllegalArgumentException
430     *             if {@code minLengthInclusive < 0}, or {@code maxLengthInclusive < minLengthInclusive}
431     * @see RandomStringGenerator#generate(int)
432     * @since 1.2
433     */
434    public String generate(final int minLengthInclusive, final int maxLengthInclusive) {
435        Validate.isTrue(minLengthInclusive >= 0, "Minimum length %d is smaller than zero.", minLengthInclusive);
436        Validate.isTrue(minLengthInclusive <= maxLengthInclusive,
437                "Maximum length %d is smaller than minimum length %d.", maxLengthInclusive, minLengthInclusive);
438        return generate(generateRandomNumber(minLengthInclusive, maxLengthInclusive));
439    }
440
441    /**
442     * Generates a random number within a range, using a {@link ThreadLocalRandom} instance
443     * or the user-supplied source of randomness.
444     *
445     * @param minInclusive
446     *            the minimum value allowed
447     * @param maxInclusive
448     *            the maximum value allowed
449     * @return The random number.
450     */
451    private int generateRandomNumber(final int minInclusive, final int maxInclusive) {
452        if (random != null) {
453            return random.nextInt(maxInclusive - minInclusive + 1) + minInclusive;
454        }
455        return ThreadLocalRandom.current().nextInt(minInclusive, maxInclusive + 1);
456    }
457
458    /**
459     * Generates a random number within a range, using a {@link ThreadLocalRandom} instance
460     * or the user-supplied source of randomness.
461     *
462     * @param characterList predefined char list.
463     * @return The random number.
464     */
465    private int generateRandomNumber(final List<Character> characterList) {
466        final int listSize = characterList.size();
467        if (random != null) {
468            return String.valueOf(characterList.get(random.nextInt(listSize))).codePointAt(0);
469        }
470        return String.valueOf(characterList.get(ThreadLocalRandom.current().nextInt(0, listSize))).codePointAt(0);
471    }
472}