001/* 002 * Copyright (C) 2007 The Guava Authors 003 * 004 * Licensed under the Apache License, Version 2.0 (the "License"); 005 * you may not use this file except in compliance with the License. 006 * You may obtain a copy of the License at 007 * 008 * http://www.apache.org/licenses/LICENSE-2.0 009 * 010 * Unless required by applicable law or agreed to in writing, software 011 * distributed under the License is distributed on an "AS IS" BASIS, 012 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. 013 * See the License for the specific language governing permissions and 014 * limitations under the License. 015 */ 016 017package com.google.common.collect; 018 019import com.google.common.annotations.GwtCompatible; 020import com.google.errorprone.annotations.CanIgnoreReturnValue; 021import com.google.errorprone.annotations.CompatibleWith; 022import java.util.Collection; 023import java.util.Collections; 024import java.util.Iterator; 025import java.util.List; 026import java.util.Set; 027import org.checkerframework.checker.nullness.compatqual.NullableDecl; 028 029/** 030 * A collection that supports order-independent equality, like {@link Set}, but may have duplicate 031 * elements. A multiset is also sometimes called a <i>bag</i>. 032 * 033 * <p>Elements of a multiset that are equal to one another are referred to as <i>occurrences</i> of 034 * the same single element. The total number of occurrences of an element in a multiset is called 035 * the <i>count</i> of that element (the terms "frequency" and "multiplicity" are equivalent, but 036 * not used in this API). Since the count of an element is represented as an {@code int}, a multiset 037 * may never contain more than {@link Integer#MAX_VALUE} occurrences of any one element. 038 * 039 * <p>{@code Multiset} refines the specifications of several methods from {@code Collection}. It 040 * also defines an additional query operation, {@link #count}, which returns the count of an 041 * element. There are five new bulk-modification operations, for example {@link #add(Object, int)}, 042 * to add or remove multiple occurrences of an element at once, or to set the count of an element to 043 * a specific value. These modification operations are optional, but implementations which support 044 * the standard collection operations {@link #add(Object)} or {@link #remove(Object)} are encouraged 045 * to implement the related methods as well. Finally, two collection views are provided: {@link 046 * #elementSet} contains the distinct elements of the multiset "with duplicates collapsed", and 047 * {@link #entrySet} is similar but contains {@link Entry Multiset.Entry} instances, each providing 048 * both a distinct element and the count of that element. 049 * 050 * <p>In addition to these required methods, implementations of {@code Multiset} are expected to 051 * provide two {@code static} creation methods: {@code create()}, returning an empty multiset, and 052 * {@code create(Iterable<? extends E>)}, returning a multiset containing the given initial 053 * elements. This is simply a refinement of {@code Collection}'s constructor recommendations, 054 * reflecting the new developments of Java 5. 055 * 056 * <p>As with other collection types, the modification operations are optional, and should throw 057 * {@link UnsupportedOperationException} when they are not implemented. Most implementations should 058 * support either all add operations or none of them, all removal operations or none of them, and if 059 * and only if all of these are supported, the {@code setCount} methods as well. 060 * 061 * <p>A multiset uses {@link Object#equals} to determine whether two instances should be considered 062 * "the same," <i>unless specified otherwise</i> by the implementation. 063 * 064 * <p>Common implementations include {@link ImmutableMultiset}, {@link HashMultiset}, and {@link 065 * ConcurrentHashMultiset}. 066 * 067 * <p>If your values may be zero, negative, or outside the range of an int, you may wish to use 068 * {@link com.google.common.util.concurrent.AtomicLongMap} instead. Note, however, that unlike 069 * {@code Multiset}, {@code AtomicLongMap} does not automatically remove zeros. 070 * 071 * <p>See the Guava User Guide article on <a href= 072 * "https://github.com/google/guava/wiki/NewCollectionTypesExplained#multiset"> {@code 073 * Multiset}</a>. 074 * 075 * @author Kevin Bourrillion 076 * @since 2.0 077 */ 078@GwtCompatible 079public interface Multiset<E> extends Collection<E> { 080 // Query Operations 081 082 /** 083 * Returns the total number of all occurrences of all elements in this multiset. 084 * 085 * <p><b>Note:</b> this method does not return the number of <i>distinct elements</i> in the 086 * multiset, which is given by {@code entrySet().size()}. 087 */ 088 @Override 089 int size(); 090 091 /** 092 * Returns the number of occurrences of an element in this multiset (the <i>count</i> of the 093 * element). Note that for an {@link Object#equals}-based multiset, this gives the same result as 094 * {@link Collections#frequency} (which would presumably perform more poorly). 095 * 096 * <p><b>Note:</b> the utility method {@link Iterables#frequency} generalizes this operation; it 097 * correctly delegates to this method when dealing with a multiset, but it can also accept any 098 * other iterable type. 099 * 100 * @param element the element to count occurrences of 101 * @return the number of occurrences of the element in this multiset; possibly zero but never 102 * negative 103 */ 104 int count(@NullableDecl @CompatibleWith("E") Object element); 105 106 // Bulk Operations 107 108 /** 109 * Adds a number of occurrences of an element to this multiset. Note that if {@code occurrences == 110 * 1}, this method has the identical effect to {@link #add(Object)}. This method is functionally 111 * equivalent (except in the case of overflow) to the call {@code 112 * addAll(Collections.nCopies(element, occurrences))}, which would presumably perform much more 113 * poorly. 114 * 115 * @param element the element to add occurrences of; may be null only if explicitly allowed by the 116 * implementation 117 * @param occurrences the number of occurrences of the element to add. May be zero, in which case 118 * no change will be made. 119 * @return the count of the element before the operation; possibly zero 120 * @throws IllegalArgumentException if {@code occurrences} is negative, or if this operation would 121 * result in more than {@link Integer#MAX_VALUE} occurrences of the element 122 * @throws NullPointerException if {@code element} is null and this implementation does not permit 123 * null elements. Note that if {@code occurrences} is zero, the implementation may opt to 124 * return normally. 125 */ 126 @CanIgnoreReturnValue 127 int add(@NullableDecl E element, int occurrences); 128 129 /** 130 * Removes a number of occurrences of the specified element from this multiset. If the multiset 131 * contains fewer than this number of occurrences to begin with, all occurrences will be removed. 132 * Note that if {@code occurrences == 1}, this is functionally equivalent to the call {@code 133 * remove(element)}. 134 * 135 * @param element the element to conditionally remove occurrences of 136 * @param occurrences the number of occurrences of the element to remove. May be zero, in which 137 * case no change will be made. 138 * @return the count of the element before the operation; possibly zero 139 * @throws IllegalArgumentException if {@code occurrences} is negative 140 */ 141 @CanIgnoreReturnValue 142 int remove(@NullableDecl @CompatibleWith("E") Object element, int occurrences); 143 144 /** 145 * Adds or removes the necessary occurrences of an element such that the element attains the 146 * desired count. 147 * 148 * @param element the element to add or remove occurrences of; may be null only if explicitly 149 * allowed by the implementation 150 * @param count the desired count of the element in this multiset 151 * @return the count of the element before the operation; possibly zero 152 * @throws IllegalArgumentException if {@code count} is negative 153 * @throws NullPointerException if {@code element} is null and this implementation does not permit 154 * null elements. Note that if {@code count} is zero, the implementor may optionally return 155 * zero instead. 156 */ 157 @CanIgnoreReturnValue 158 int setCount(E element, int count); 159 160 /** 161 * Conditionally sets the count of an element to a new value, as described in {@link 162 * #setCount(Object, int)}, provided that the element has the expected current count. If the 163 * current count is not {@code oldCount}, no change is made. 164 * 165 * @param element the element to conditionally set the count of; may be null only if explicitly 166 * allowed by the implementation 167 * @param oldCount the expected present count of the element in this multiset 168 * @param newCount the desired count of the element in this multiset 169 * @return {@code true} if the condition for modification was met. This implies that the multiset 170 * was indeed modified, unless {@code oldCount == newCount}. 171 * @throws IllegalArgumentException if {@code oldCount} or {@code newCount} is negative 172 * @throws NullPointerException if {@code element} is null and the implementation does not permit 173 * null elements. Note that if {@code oldCount} and {@code newCount} are both zero, the 174 * implementor may optionally return {@code true} instead. 175 */ 176 @CanIgnoreReturnValue 177 boolean setCount(E element, int oldCount, int newCount); 178 179 // Views 180 181 /** 182 * Returns the set of distinct elements contained in this multiset. The element set is backed by 183 * the same data as the multiset, so any change to either is immediately reflected in the other. 184 * The order of the elements in the element set is unspecified. 185 * 186 * <p>If the element set supports any removal operations, these necessarily cause <b>all</b> 187 * occurrences of the removed element(s) to be removed from the multiset. Implementations are not 188 * expected to support the add operations, although this is possible. 189 * 190 * <p>A common use for the element set is to find the number of distinct elements in the multiset: 191 * {@code elementSet().size()}. 192 * 193 * @return a view of the set of distinct elements in this multiset 194 */ 195 Set<E> elementSet(); 196 197 /** 198 * Returns a view of the contents of this multiset, grouped into {@code Multiset.Entry} instances, 199 * each providing an element of the multiset and the count of that element. This set contains 200 * exactly one entry for each distinct element in the multiset (thus it always has the same size 201 * as the {@link #elementSet}). The order of the elements in the element set is unspecified. 202 * 203 * <p>The entry set is backed by the same data as the multiset, so any change to either is 204 * immediately reflected in the other. However, multiset changes may or may not be reflected in 205 * any {@code Entry} instances already retrieved from the entry set (this is 206 * implementation-dependent). Furthermore, implementations are not required to support 207 * modifications to the entry set at all, and the {@code Entry} instances themselves don't even 208 * have methods for modification. See the specific implementation class for more details on how 209 * its entry set handles modifications. 210 * 211 * @return a set of entries representing the data of this multiset 212 */ 213 Set<Entry<E>> entrySet(); 214 215 /** 216 * An unmodifiable element-count pair for a multiset. The {@link Multiset#entrySet} method returns 217 * a view of the multiset whose elements are of this class. A multiset implementation may return 218 * Entry instances that are either live "read-through" views to the Multiset, or immutable 219 * snapshots. Note that this type is unrelated to the similarly-named type {@code Map.Entry}. 220 * 221 * @since 2.0 222 */ 223 interface Entry<E> { 224 225 /** 226 * Returns the multiset element corresponding to this entry. Multiple calls to this method 227 * always return the same instance. 228 * 229 * @return the element corresponding to this entry 230 */ 231 E getElement(); 232 233 /** 234 * Returns the count of the associated element in the underlying multiset. This count may either 235 * be an unchanging snapshot of the count at the time the entry was retrieved, or a live view of 236 * the current count of the element in the multiset, depending on the implementation. Note that 237 * in the former case, this method can never return zero, while in the latter, it will return 238 * zero if all occurrences of the element were since removed from the multiset. 239 * 240 * @return the count of the element; never negative 241 */ 242 int getCount(); 243 244 /** 245 * {@inheritDoc} 246 * 247 * <p>Returns {@code true} if the given object is also a multiset entry and the two entries 248 * represent the same element and count. That is, two entries {@code a} and {@code b} are equal 249 * if: 250 * 251 * <pre>{@code 252 * Objects.equal(a.getElement(), b.getElement()) 253 * && a.getCount() == b.getCount() 254 * }</pre> 255 */ 256 @Override 257 // TODO(kevinb): check this wrt TreeMultiset? 258 boolean equals(Object o); 259 260 /** 261 * {@inheritDoc} 262 * 263 * <p>The hash code of a multiset entry for element {@code element} and count {@code count} is 264 * defined as: 265 * 266 * <pre>{@code 267 * ((element == null) ? 0 : element.hashCode()) ^ count 268 * }</pre> 269 */ 270 @Override 271 int hashCode(); 272 273 /** 274 * Returns the canonical string representation of this entry, defined as follows. If the count 275 * for this entry is one, this is simply the string representation of the corresponding element. 276 * Otherwise, it is the string representation of the element, followed by the three characters 277 * {@code " x "} (space, letter x, space), followed by the count. 278 */ 279 @Override 280 String toString(); 281 } 282 283 // Comparison and hashing 284 285 /** 286 * Compares the specified object with this multiset for equality. Returns {@code true} if the 287 * given object is also a multiset and contains equal elements with equal counts, regardless of 288 * order. 289 */ 290 @Override 291 // TODO(kevinb): caveats about equivalence-relation? 292 boolean equals(@NullableDecl Object object); 293 294 /** 295 * Returns the hash code for this multiset. This is defined as the sum of 296 * 297 * <pre>{@code 298 * ((element == null) ? 0 : element.hashCode()) ^ count(element) 299 * }</pre> 300 * 301 * <p>over all distinct elements in the multiset. It follows that a multiset and its entry set 302 * always have the same hash code. 303 */ 304 @Override 305 int hashCode(); 306 307 /** 308 * {@inheritDoc} 309 * 310 * <p>It is recommended, though not mandatory, that this method return the result of invoking 311 * {@link #toString} on the {@link #entrySet}, yielding a result such as {@code [a x 3, c, d x 2, 312 * e]}. 313 */ 314 @Override 315 String toString(); 316 317 // Refined Collection Methods 318 319 /** 320 * {@inheritDoc} 321 * 322 * <p>Elements that occur multiple times in the multiset will appear multiple times in this 323 * iterator, though not necessarily sequentially. 324 */ 325 @Override 326 Iterator<E> iterator(); 327 328 /** 329 * Determines whether this multiset contains the specified element. 330 * 331 * <p>This method refines {@link Collection#contains} to further specify that it <b>may not</b> 332 * throw an exception in response to {@code element} being null or of the wrong type. 333 * 334 * @param element the element to check for 335 * @return {@code true} if this multiset contains at least one occurrence of the element 336 */ 337 @Override 338 boolean contains(@NullableDecl Object element); 339 340 /** 341 * Returns {@code true} if this multiset contains at least one occurrence of each element in the 342 * specified collection. 343 * 344 * <p>This method refines {@link Collection#containsAll} to further specify that it <b>may not</b> 345 * throw an exception in response to any of {@code elements} being null or of the wrong type. 346 * 347 * <p><b>Note:</b> this method does not take into account the occurrence count of an element in 348 * the two collections; it may still return {@code true} even if {@code elements} contains several 349 * occurrences of an element and this multiset contains only one. This is no different than any 350 * other collection type like {@link List}, but it may be unexpected to the user of a multiset. 351 * 352 * @param elements the collection of elements to be checked for containment in this multiset 353 * @return {@code true} if this multiset contains at least one occurrence of each element 354 * contained in {@code elements} 355 * @throws NullPointerException if {@code elements} is null 356 */ 357 @Override 358 boolean containsAll(Collection<?> elements); 359 360 /** 361 * Adds a single occurrence of the specified element to this multiset. 362 * 363 * <p>This method refines {@link Collection#add}, which only <i>ensures</i> the presence of the 364 * element, to further specify that a successful call must always increment the count of the 365 * element, and the overall size of the collection, by one. 366 * 367 * <p>To both add the element and obtain the previous count of that element, use {@link 368 * #add(Object, int) add}{@code (element, 1)} instead. 369 * 370 * @param element the element to add one occurrence of; may be null only if explicitly allowed by 371 * the implementation 372 * @return {@code true} always, since this call is required to modify the multiset, unlike other 373 * {@link Collection} types 374 * @throws NullPointerException if {@code element} is null and this implementation does not permit 375 * null elements 376 * @throws IllegalArgumentException if {@link Integer#MAX_VALUE} occurrences of {@code element} 377 * are already contained in this multiset 378 */ 379 @CanIgnoreReturnValue 380 @Override 381 boolean add(E element); 382 383 /** 384 * Removes a <i>single</i> occurrence of the specified element from this multiset, if present. 385 * 386 * <p>This method refines {@link Collection#remove} to further specify that it <b>may not</b> 387 * throw an exception in response to {@code element} being null or of the wrong type. 388 * 389 * <p>To both remove the element and obtain the previous count of that element, use {@link 390 * #remove(Object, int) remove}{@code (element, 1)} instead. 391 * 392 * @param element the element to remove one occurrence of 393 * @return {@code true} if an occurrence was found and removed 394 */ 395 @CanIgnoreReturnValue 396 @Override 397 boolean remove(@NullableDecl Object element); 398 399 /** 400 * {@inheritDoc} 401 * 402 * <p><b>Note:</b> This method ignores how often any element might appear in {@code c}, and only 403 * cares whether or not an element appears at all. If you wish to remove one occurrence in this 404 * multiset for every occurrence in {@code c}, see {@link Multisets#removeOccurrences(Multiset, 405 * Multiset)}. 406 * 407 * <p>This method refines {@link Collection#removeAll} to further specify that it <b>may not</b> 408 * throw an exception in response to any of {@code elements} being null or of the wrong type. 409 */ 410 @CanIgnoreReturnValue 411 @Override 412 boolean removeAll(Collection<?> c); 413 414 /** 415 * {@inheritDoc} 416 * 417 * <p><b>Note:</b> This method ignores how often any element might appear in {@code c}, and only 418 * cares whether or not an element appears at all. If you wish to remove one occurrence in this 419 * multiset for every occurrence in {@code c}, see {@link Multisets#retainOccurrences(Multiset, 420 * Multiset)}. 421 * 422 * <p>This method refines {@link Collection#retainAll} to further specify that it <b>may not</b> 423 * throw an exception in response to any of {@code elements} being null or of the wrong type. 424 * 425 * @see Multisets#retainOccurrences(Multiset, Multiset) 426 */ 427 @CanIgnoreReturnValue 428 @Override 429 boolean retainAll(Collection<?> c); 430}