Package com.google.common.collect

Class Maps

  • @GwtCompatible(emulated=true)
public final class Maps
extends Object
    Static utility methods pertaining to Map instances (including instances of SortedMap, BiMap, etc.). Also see this class's counterparts Lists, Sets and Queues.

    See the Guava User Guide article on Maps.

    Since:
    2.0
    Author:
    Kevin Bourrillion, Mike Bostock, Isaac Shum, Louis Wasserman

    • Method Detail

      • immutableEnumMap

        @GwtCompatible(serializable=true)
public static <K extends Enum<K>,​V> ImmutableMap<K,​V> immutableEnumMap​(Map<K,​? extends V> map)
        Returns an immutable map instance containing the given entries. Internally, the returned map will be backed by an EnumMap.

        The iteration order of the returned map follows the enum's iteration order, not the order in which the elements appear in the given map.

        Parameters:
        map - the map to make an immutable copy of
        Returns:
        an immutable map containing those entries
        Since:
        14.0

      • toImmutableEnumMap

        public static <T extends @Nullable Object,​K extends Enum<K>,​V> Collector<T,​?,​ImmutableMap<K,​V>> toImmutableEnumMap​(Function<? super T,​? extends K> keyFunction,
                                                                                                                                                 Function<? super T,​? extends V> valueFunction)
        Returns a Collector that accumulates elements into an ImmutableMap whose keys and values are the result of applying the provided mapping functions to the input elements. The resulting implementation is specialized for enum key types. The returned map and its views will iterate over keys in their enum definition order, not encounter order.

        If the mapped keys contain duplicates, an IllegalArgumentException is thrown when the collection operation is performed. (This differs from the Collector returned by Collectors.toMap(Function, Function), which throws an IllegalStateException.)

        Since:
        21.0

      • toImmutableEnumMap

        public static <T extends @Nullable Object,​K extends Enum<K>,​V> Collector<T,​?,​ImmutableMap<K,​V>> toImmutableEnumMap​(Function<? super T,​? extends K> keyFunction,
                                                                                                                                                 Function<? super T,​? extends V> valueFunction,
                                                                                                                                                 BinaryOperator<V> mergeFunction)
        Returns a Collector that accumulates elements into an ImmutableMap whose keys and values are the result of applying the provided mapping functions to the input elements. The resulting implementation is specialized for enum key types. The returned map and its views will iterate over keys in their enum definition order, not encounter order.

        If the mapped keys contain duplicates, the values are merged using the specified merging function.

        Since:
        21.0

      • newHashMap

        public static <K extends @Nullable Object,​V extends @Nullable Object> HashMap<K,​V> newHashMap​(Map<? extends K,​? extends V> map)
        Creates a mutable HashMap instance with the same mappings as the specified map.

        Note: if mutability is not required, use ImmutableMap.copyOf(Map) instead.

        Note: if K is an Enum type, use newEnumMap(java.lang.Class<K>) instead.

        Note: this method is now unnecessary and should be treated as deprecated. Instead, use the HashMap constructor directly, taking advantage of "diamond" syntax.

        Parameters:
        map - the mappings to be placed in the new map
        Returns:
        a new HashMap initialized with the mappings from map

      • newHashMapWithExpectedSize

        public static <K extends @Nullable Object,​V extends @Nullable Object> HashMap<K,​V> newHashMapWithExpectedSize​(int expectedSize)
        Creates a HashMap instance, with a high enough "initial capacity" that it should hold expectedSize elements without growth. This behavior cannot be broadly guaranteed, but it is observed to be true for OpenJDK 1.7. It also can't be guaranteed that the method isn't inadvertently oversizing the returned map.
        Parameters:
        expectedSize - the number of entries you expect to add to the returned map
        Returns:
        a new, empty HashMap with enough capacity to hold expectedSize entries without resizing
        Throws:
        IllegalArgumentException - if expectedSize is negative

      • newLinkedHashMap

        public static <K extends @Nullable Object,​V extends @Nullable Object> LinkedHashMap<K,​V> newLinkedHashMap()
        Creates a mutable, empty, insertion-ordered LinkedHashMap instance.

        Note: if mutability is not required, use ImmutableMap.of() instead.

        Note: this method is now unnecessary and should be treated as deprecated. Instead, use the LinkedHashMap constructor directly, taking advantage of "diamond" syntax.

        Returns:
        a new, empty LinkedHashMap

      • newLinkedHashMap

        public static <K extends @Nullable Object,​V extends @Nullable Object> LinkedHashMap<K,​V> newLinkedHashMap​(Map<? extends K,​? extends V> map)
        Creates a mutable, insertion-ordered LinkedHashMap instance with the same mappings as the specified map.

        Note: if mutability is not required, use ImmutableMap.copyOf(Map) instead.

        Note: this method is now unnecessary and should be treated as deprecated. Instead, use the LinkedHashMap constructor directly, taking advantage of "diamond" syntax.

        Parameters:
        map - the mappings to be placed in the new map
        Returns:
        a new, LinkedHashMap initialized with the mappings from map

      • newLinkedHashMapWithExpectedSize

        public static <K extends @Nullable Object,​V extends @Nullable Object> LinkedHashMap<K,​V> newLinkedHashMapWithExpectedSize​(int expectedSize)
        Creates a LinkedHashMap instance, with a high enough "initial capacity" that it should hold expectedSize elements without growth. This behavior cannot be broadly guaranteed, but it is observed to be true for OpenJDK 1.7. It also can't be guaranteed that the method isn't inadvertently oversizing the returned map.
        Parameters:
        expectedSize - the number of entries you expect to add to the returned map
        Returns:
        a new, empty LinkedHashMap with enough capacity to hold expectedSize entries without resizing
        Throws:
        IllegalArgumentException - if expectedSize is negative
        Since:
        19.0

      • newTreeMap

        public static <K extends Comparable,​V extends @Nullable Object> TreeMap<K,​V> newTreeMap()
        Creates a mutable, empty TreeMap instance using the natural ordering of its elements.

        Note: if mutability is not required, use ImmutableSortedMap.of() instead.

        Note: this method is now unnecessary and should be treated as deprecated. Instead, use the TreeMap constructor directly, taking advantage of "diamond" syntax.

        Returns:
        a new, empty TreeMap

      • newTreeMap

        public static <K extends @Nullable Object,​V extends @Nullable Object> TreeMap<K,​V> newTreeMap​(SortedMap<K,​? extends V> map)
        Creates a mutable TreeMap instance with the same mappings as the specified map and using the same ordering as the specified map.

        Note: if mutability is not required, use ImmutableSortedMap.copyOfSorted(SortedMap) instead.

        Note: this method is now unnecessary and should be treated as deprecated. Instead, use the TreeMap constructor directly, taking advantage of "diamond" syntax.

        Parameters:
        map - the sorted map whose mappings are to be placed in the new map and whose comparator is to be used to sort the new map
        Returns:
        a new TreeMap initialized with the mappings from map and using the comparator of map

      • newTreeMap

        public static <C extends @Nullable Object,​K extends C,​V extends @Nullable Object> TreeMap<K,​V> newTreeMap​(@CheckForNull
                                                                                                                            Comparator<C> comparator)
        Creates a mutable, empty TreeMap instance using the given comparator.

        Note: if mutability is not required, use ImmutableSortedMap.orderedBy(comparator).build() instead.

        Note: this method is now unnecessary and should be treated as deprecated. Instead, use the TreeMap constructor directly, taking advantage of "diamond" syntax.

        Parameters:
        comparator - the comparator to sort the keys with
        Returns:
        a new, empty TreeMap

      • newEnumMap

        public static <K extends Enum<K>,​V extends @Nullable Object> EnumMap<K,​V> newEnumMap​(Class<K> type)
        Creates an EnumMap instance.
        Parameters:
        type - the key type for this map
        Returns:
        a new, empty EnumMap

      • newEnumMap

        public static <K extends Enum<K>,​V extends @Nullable Object> EnumMap<K,​V> newEnumMap​(Map<K,​? extends V> map)
        Creates an EnumMap with the same mappings as the specified map.

        Note: this method is now unnecessary and should be treated as deprecated. Instead, use the EnumMap constructor directly, taking advantage of "diamond" syntax.

        Parameters:
        map - the map from which to initialize this EnumMap
        Returns:
        a new EnumMap initialized with the mappings from map
        Throws:
        IllegalArgumentException - if m is not an EnumMap instance and contains no mappings

      • difference

        public static <K extends @Nullable Object,​V extends @Nullable Object> MapDifference<K,​V> difference​(Map<? extends K,​? extends V> left,
                                                                                                                Map<? extends K,​? extends V> right)
        Computes the difference between two maps. This difference is an immutable snapshot of the state of the maps at the time this method is called. It will never change, even if the maps change at a later time.

        Since this method uses HashMap instances internally, the keys of the supplied maps must be well-behaved with respect to Object.equals(java.lang.Object) and Object.hashCode().

        Note:If you only need to know whether two maps have the same mappings, call left.equals(right) instead of this method.

        Parameters:
        left - the map to treat as the "left" map for purposes of comparison
        right - the map to treat as the "right" map for purposes of comparison
        Returns:
        the difference between the two maps

      • difference

        public static <K extends @Nullable Object,​V> MapDifference<K,​V> difference​(Map<? extends K,​? extends V> left,
                                                                                       Map<? extends K,​? extends V> right,
                                                                                       Equivalence<? super V> valueEquivalence)
        Computes the difference between two maps. This difference is an immutable snapshot of the state of the maps at the time this method is called. It will never change, even if the maps change at a later time.

        Since this method uses HashMap instances internally, the keys of the supplied maps must be well-behaved with respect to Object.equals(java.lang.Object) and Object.hashCode().

        Parameters:
        left - the map to treat as the "left" map for purposes of comparison
        right - the map to treat as the "right" map for purposes of comparison
        valueEquivalence - the equivalence relationship to use to compare values
        Returns:
        the difference between the two maps
        Since:
        10.0

      • difference

        public static <K extends @Nullable Object,​V extends @Nullable Object> SortedMapDifference<K,​V> difference​(SortedMap<K,​? extends V> left,
                                                                                                                      Map<? extends K,​? extends V> right)
        Computes the difference between two sorted maps, using the comparator of the left map, or Ordering.natural() if the left map uses the natural ordering of its elements. This difference is an immutable snapshot of the state of the maps at the time this method is called. It will never change, even if the maps change at a later time.

        Since this method uses TreeMap instances internally, the keys of the right map must all compare as distinct according to the comparator of the left map.

        Note:If you only need to know whether two sorted maps have the same mappings, call left.equals(right) instead of this method.

        Parameters:
        left - the map to treat as the "left" map for purposes of comparison
        right - the map to treat as the "right" map for purposes of comparison
        Returns:
        the difference between the two maps
        Since:
        11.0

      • asMap

        public static <K extends @Nullable Object,​V extends @Nullable Object> Map<K,​V> asMap​(Set<K> set,
                                                                                                 Function<? super K,​V> function)
        Returns a live Map view whose keys are the contents of set and whose values are computed on demand using function. To get an immutable copy instead, use toMap(Iterable, Function).

        Specifically, for each k in the backing set, the returned map has an entry mapping k to function.apply(k). The keySet, values, and entrySet views of the returned map iterate in the same order as the backing set.

        Modifications to the backing set are read through to the returned map. The returned map supports removal operations if the backing set does. Removal operations write through to the backing set. The returned map does not support put operations.

        Warning: If the function rejects null, caution is required to make sure the set does not contain null, because the view cannot stop null from being added to the set.

        Warning: This method assumes that for any instance k of key type K, k.equals(k2) implies that k2 is also of type K. Using a key type for which this may not hold, such as ArrayList, may risk a ClassCastException when calling methods on the resulting map view.

        Since:
        14.0

      • asMap

        public static <K extends @Nullable Object,​V extends @Nullable Object> SortedMap<K,​V> asMap​(SortedSet<K> set,
                                                                                                       Function<? super K,​V> function)
        Returns a view of the sorted set as a map, mapping keys from the set according to the specified function.

        Specifically, for each k in the backing set, the returned map has an entry mapping k to function.apply(k). The keySet, values, and entrySet views of the returned map iterate in the same order as the backing set.

        Modifications to the backing set are read through to the returned map. The returned map supports removal operations if the backing set does. Removal operations write through to the backing set. The returned map does not support put operations.

        Warning: If the function rejects null, caution is required to make sure the set does not contain null, because the view cannot stop null from being added to the set.

        Warning: This method assumes that for any instance k of key type K, k.equals(k2) implies that k2 is also of type K. Using a key type for which this may not hold, such as ArrayList, may risk a ClassCastException when calling methods on the resulting map view.

        Since:
        14.0

      • asMap

        @GwtIncompatible
public static <K extends @Nullable Object,​V extends @Nullable Object> NavigableMap<K,​V> asMap​(NavigableSet<K> set,
                                                                                                          Function<? super K,​V> function)
        Returns a view of the navigable set as a map, mapping keys from the set according to the specified function.

        Specifically, for each k in the backing set, the returned map has an entry mapping k to function.apply(k). The keySet, values, and entrySet views of the returned map iterate in the same order as the backing set.

        Modifications to the backing set are read through to the returned map. The returned map supports removal operations if the backing set does. Removal operations write through to the backing set. The returned map does not support put operations.

        Warning: If the function rejects null, caution is required to make sure the set does not contain null, because the view cannot stop null from being added to the set.

        Warning: This method assumes that for any instance k of key type K, k.equals(k2) implies that k2 is also of type K. Using a key type for which this may not hold, such as ArrayList, may risk a ClassCastException when calling methods on the resulting map view.

        Since:
        14.0

      • toMap

        public static <K,​V> ImmutableMap<K,​V> toMap​(Iterable<K> keys,
                                                        Function<? super K,​V> valueFunction)
        Returns an immutable map whose keys are the distinct elements of keys and whose value for each key was computed by valueFunction. The map's iteration order is the order of the first appearance of each key in keys.

        When there are multiple instances of a key in keys, it is unspecified whether valueFunction will be applied to more than one instance of that key and, if it is, which result will be mapped to that key in the returned map.

        If keys is a Set, a live view can be obtained instead of a copy using asMap(Set, Function).

        Throws:
        NullPointerException - if any element of keys is null, or if valueFunction produces null for any key
        Since:
        14.0

      • toMap

        public static <K,​V> ImmutableMap<K,​V> toMap​(Iterator<K> keys,
                                                        Function<? super K,​V> valueFunction)
        Returns an immutable map whose keys are the distinct elements of keys and whose value for each key was computed by valueFunction. The map's iteration order is the order of the first appearance of each key in keys.

        When there are multiple instances of a key in keys, it is unspecified whether valueFunction will be applied to more than one instance of that key and, if it is, which result will be mapped to that key in the returned map.

        Throws:
        NullPointerException - if any element of keys is null, or if valueFunction produces null for any key
        Since:
        14.0

      • uniqueIndex

        @CanIgnoreReturnValue
public static <K,​V> ImmutableMap<K,​V> uniqueIndex​(Iterable<V> values,
                                                              Function<? super V,​K> keyFunction)
        Returns a map with the given values, indexed by keys derived from those values. In other words, each input value produces an entry in the map whose key is the result of applying keyFunction to that value. These entries appear in the same order as the input values. Example usage: 
        
 Color red = new Color("red", 255, 0, 0);
 ...
 ImmutableSet<Color> allColors = ImmutableSet.of(red, green, blue);

 Map<String, Color> colorForName =
     uniqueIndex(allColors, toStringFunction());
 assertThat(colorForName).containsEntry("red", red);

        If your index may associate multiple values with each key, use Multimaps.index.

        Parameters:
        values - the values to use when constructing the Map
        keyFunction - the function used to produce the key for each value
        Returns:
        a map mapping the result of evaluating the function keyFunction on each value in the input collection to that value
        Throws:
        IllegalArgumentException - if keyFunction produces the same key for more than one value in the input collection
        NullPointerException - if any element of values is null, or if keyFunction produces null for any value

      • uniqueIndex

        @CanIgnoreReturnValue
public static <K,​V> ImmutableMap<K,​V> uniqueIndex​(Iterator<V> values,
                                                              Function<? super V,​K> keyFunction)
        Returns a map with the given values, indexed by keys derived from those values. In other words, each input value produces an entry in the map whose key is the result of applying keyFunction to that value. These entries appear in the same order as the input values. Example usage: 
        
 Color red = new Color("red", 255, 0, 0);
 ...
 Iterator<Color> allColors = ImmutableSet.of(red, green, blue).iterator();

 Map<String, Color> colorForName =
     uniqueIndex(allColors, toStringFunction());
 assertThat(colorForName).containsEntry("red", red);

        If your index may associate multiple values with each key, use Multimaps.index.

        Parameters:
        values - the values to use when constructing the Map
        keyFunction - the function used to produce the key for each value
        Returns:
        a map mapping the result of evaluating the function keyFunction on each value in the input collection to that value
        Throws:
        IllegalArgumentException - if keyFunction produces the same key for more than one value in the input collection
        NullPointerException - if any element of values is null, or if keyFunction produces null for any value
        Since:
        10.0

      • fromProperties

        @GwtIncompatible
public static ImmutableMap<String,​String> fromProperties​(Properties properties)
        Creates an ImmutableMap<String, String> from a Properties instance. Properties normally derive from Map<Object, Object>, but they typically contain strings, which is awkward. This method lets you get a plain-old-Map out of a Properties.
        Parameters:
        properties - a Properties object to be converted
        Returns:
        an immutable map containing all the entries in properties
        Throws:
        ClassCastException - if any key in properties is not a String
        NullPointerException - if any key or value in properties is null

      • synchronizedBiMap

        public static <K extends @Nullable Object,​V extends @Nullable Object> BiMap<K,​V> synchronizedBiMap​(BiMap<K,​V> bimap)
        Returns a synchronized (thread-safe) bimap backed by the specified bimap. In order to guarantee serial access, it is critical that all access to the backing bimap is accomplished through the returned bimap.

        It is imperative that the user manually synchronize on the returned map when accessing any of its collection views: 

        
 BiMap<Long, String> map = Maps.synchronizedBiMap(
     HashBiMap.<Long, String>create());
 ...
 Set<Long> set = map.keySet();  // Needn't be in synchronized block
 ...
 synchronized (map) {  // Synchronizing on map, not set!
   Iterator<Long> it = set.iterator(); // Must be in synchronized block
   while (it.hasNext()) {
     foo(it.next());
   }
 }

        Failure to follow this advice may result in non-deterministic behavior.

        The returned bimap will be serializable if the specified bimap is serializable.

        Parameters:
        bimap - the bimap to be wrapped in a synchronized view
        Returns:
        a synchronized view of the specified bimap

      • unmodifiableBiMap

        public static <K extends @Nullable Object,​V extends @Nullable Object> BiMap<K,​V> unmodifiableBiMap​(BiMap<? extends K,​? extends V> bimap)
        Returns an unmodifiable view of the specified bimap. This method allows modules to provide users with "read-only" access to internal bimaps. Query operations on the returned bimap "read through" to the specified bimap, and attempts to modify the returned map, whether direct or via its collection views, result in an UnsupportedOperationException.

        The returned bimap will be serializable if the specified bimap is serializable.

        Parameters:
        bimap - the bimap for which an unmodifiable view is to be returned
        Returns:
        an unmodifiable view of the specified bimap

      • transformValues

        public static <K extends @Nullable Object,​V1 extends @Nullable Object,​V2 extends @Nullable Object> Map<K,​V2> transformValues​(Map<K,​V1> fromMap,
                                                                                                                                               Function<? super V1,​V2> function)
        Returns a view of a map where each value is transformed by a function. All other properties of the map, such as iteration order, are left intact. For example, the code: 
        
 Map<String, Integer> map = ImmutableMap.of("a", 4, "b", 9);
 Function<Integer, Double> sqrt =
     new Function<Integer, Double>() {
       public Double apply(Integer in) {
         return Math.sqrt((int) in);
       }
     };
 Map<String, Double> transformed = Maps.transformValues(map, sqrt);
 System.out.println(transformed);
        ... prints {a=2.0, b=3.0}.

        Changes in the underlying map are reflected in this view. Conversely, this view supports removal operations, and these are reflected in the underlying map.

        It's acceptable for the underlying map to contain null keys, and even null values provided that the function is capable of accepting null input. The transformed map might contain null values, if the function sometimes gives a null result.

        The returned map is not thread-safe or serializable, even if the underlying map is.

        The function is applied lazily, invoked when needed. This is necessary for the returned map to be a view, but it means that the function will be applied many times for bulk operations like Map.containsValue(java.lang.Object) and Map.toString(). For this to perform well, function should be fast. To avoid lazy evaluation when the returned map doesn't need to be a view, copy the returned map into a new map of your choosing.

      • transformValues

        public static <K extends @Nullable Object,​V1 extends @Nullable Object,​V2 extends @Nullable Object> SortedMap<K,​V2> transformValues​(SortedMap<K,​V1> fromMap,
                                                                                                                                                     Function<? super V1,​V2> function)
        Returns a view of a sorted map where each value is transformed by a function. All other properties of the map, such as iteration order, are left intact. For example, the code: 
        
 SortedMap<String, Integer> map = ImmutableSortedMap.of("a", 4, "b", 9);
 Function<Integer, Double> sqrt =
     new Function<Integer, Double>() {
       public Double apply(Integer in) {
         return Math.sqrt((int) in);
       }
     };
 SortedMap<String, Double> transformed =
      Maps.transformValues(map, sqrt);
 System.out.println(transformed);
        ... prints {a=2.0, b=3.0}.

        Changes in the underlying map are reflected in this view. Conversely, this view supports removal operations, and these are reflected in the underlying map.

        It's acceptable for the underlying map to contain null keys, and even null values provided that the function is capable of accepting null input. The transformed map might contain null values, if the function sometimes gives a null result.

        The returned map is not thread-safe or serializable, even if the underlying map is.

        The function is applied lazily, invoked when needed. This is necessary for the returned map to be a view, but it means that the function will be applied many times for bulk operations like Map.containsValue(java.lang.Object) and Map.toString(). For this to perform well, function should be fast. To avoid lazy evaluation when the returned map doesn't need to be a view, copy the returned map into a new map of your choosing.

        Since:
        11.0

      • transformValues

        @GwtIncompatible
public static <K extends @Nullable Object,​V1 extends @Nullable Object,​V2 extends @Nullable Object> NavigableMap<K,​V2> transformValues​(NavigableMap<K,​V1> fromMap,
                                                                                                                                                        Function<? super V1,​V2> function)
        Returns a view of a navigable map where each value is transformed by a function. All other properties of the map, such as iteration order, are left intact. For example, the code: 
        
 NavigableMap<String, Integer> map = Maps.newTreeMap();
 map.put("a", 4);
 map.put("b", 9);
 Function<Integer, Double> sqrt =
     new Function<Integer, Double>() {
       public Double apply(Integer in) {
         return Math.sqrt((int) in);
       }
     };
 NavigableMap<String, Double> transformed =
      Maps.transformNavigableValues(map, sqrt);
 System.out.println(transformed);
        ... prints {a=2.0, b=3.0}.

        Changes in the underlying map are reflected in this view. Conversely, this view supports removal operations, and these are reflected in the underlying map.

        It's acceptable for the underlying map to contain null keys, and even null values provided that the function is capable of accepting null input. The transformed map might contain null values, if the function sometimes gives a null result.

        The returned map is not thread-safe or serializable, even if the underlying map is.

        The function is applied lazily, invoked when needed. This is necessary for the returned map to be a view, but it means that the function will be applied many times for bulk operations like Map.containsValue(java.lang.Object) and Map.toString(). For this to perform well, function should be fast. To avoid lazy evaluation when the returned map doesn't need to be a view, copy the returned map into a new map of your choosing.

        Since:
        13.0

      • transformEntries

        public static <K extends @Nullable Object,​V1 extends @Nullable Object,​V2 extends @Nullable Object> Map<K,​V2> transformEntries​(Map<K,​V1> fromMap,
                                                                                                                                                Maps.EntryTransformer<? super K,​? super V1,​V2> transformer)
        Returns a view of a map whose values are derived from the original map's entries. In contrast to transformValues(java.util.Map<K, V1>, com.google.common.base.Function<? super V1, V2>), this method's entry-transformation logic may depend on the key as well as the value.

        All other properties of the transformed map, such as iteration order, are left intact. For example, the code: 

        
 Map<String, Boolean> options =
     ImmutableMap.of("verbose", true, "sort", false);
 EntryTransformer<String, Boolean, String> flagPrefixer =
     new EntryTransformer<String, Boolean, String>() {
       public String transformEntry(String key, Boolean value) {
         return value ? key : "no" + key;
       }
     };
 Map<String, String> transformed =
     Maps.transformEntries(options, flagPrefixer);
 System.out.println(transformed);
        ... prints {verbose=verbose, sort=nosort}.

        Changes in the underlying map are reflected in this view. Conversely, this view supports removal operations, and these are reflected in the underlying map.

        It's acceptable for the underlying map to contain null keys and null values provided that the transformer is capable of accepting null inputs. The transformed map might contain null values if the transformer sometimes gives a null result.

        The returned map is not thread-safe or serializable, even if the underlying map is.

        The transformer is applied lazily, invoked when needed. This is necessary for the returned map to be a view, but it means that the transformer will be applied many times for bulk operations like Map.containsValue(java.lang.Object) and Object.toString(). For this to perform well, transformer should be fast. To avoid lazy evaluation when the returned map doesn't need to be a view, copy the returned map into a new map of your choosing.

        Warning: This method assumes that for any instance k of EntryTransformer key type K, k.equals(k2) implies that k2 is also of type K. Using an EntryTransformer key type for which this may not hold, such as ArrayList, may risk a ClassCastException when calling methods on the transformed map.

        Since:
        7.0

      • transformEntries

        public static <K extends @Nullable Object,​V1 extends @Nullable Object,​V2 extends @Nullable Object> SortedMap<K,​V2> transformEntries​(SortedMap<K,​V1> fromMap,
                                                                                                                                                      Maps.EntryTransformer<? super K,​? super V1,​V2> transformer)
        Returns a view of a sorted map whose values are derived from the original sorted map's entries. In contrast to transformValues(java.util.Map<K, V1>, com.google.common.base.Function<? super V1, V2>), this method's entry-transformation logic may depend on the key as well as the value.

        All other properties of the transformed map, such as iteration order, are left intact. For example, the code: 

        
 Map<String, Boolean> options =
     ImmutableSortedMap.of("verbose", true, "sort", false);
 EntryTransformer<String, Boolean, String> flagPrefixer =
     new EntryTransformer<String, Boolean, String>() {
       public String transformEntry(String key, Boolean value) {
         return value ? key : "yes" + key;
       }
     };
 SortedMap<String, String> transformed =
     Maps.transformEntries(options, flagPrefixer);
 System.out.println(transformed);
        ... prints {sort=yessort, verbose=verbose}.

        Changes in the underlying map are reflected in this view. Conversely, this view supports removal operations, and these are reflected in the underlying map.

        It's acceptable for the underlying map to contain null keys and null values provided that the transformer is capable of accepting null inputs. The transformed map might contain null values if the transformer sometimes gives a null result.

        The returned map is not thread-safe or serializable, even if the underlying map is.

        The transformer is applied lazily, invoked when needed. This is necessary for the returned map to be a view, but it means that the transformer will be applied many times for bulk operations like Map.containsValue(java.lang.Object) and Object.toString(). For this to perform well, transformer should be fast. To avoid lazy evaluation when the returned map doesn't need to be a view, copy the returned map into a new map of your choosing.

        Warning: This method assumes that for any instance k of EntryTransformer key type K, k.equals(k2) implies that k2 is also of type K. Using an EntryTransformer key type for which this may not hold, such as ArrayList, may risk a ClassCastException when calling methods on the transformed map.

        Since:
        11.0

      • transformEntries

        @GwtIncompatible
public static <K extends @Nullable Object,​V1 extends @Nullable Object,​V2 extends @Nullable Object> NavigableMap<K,​V2> transformEntries​(NavigableMap<K,​V1> fromMap,
                                                                                                                                                         Maps.EntryTransformer<? super K,​? super V1,​V2> transformer)
        Returns a view of a navigable map whose values are derived from the original navigable map's entries. In contrast to transformValues(java.util.Map<K, V1>, com.google.common.base.Function<? super V1, V2>), this method's entry-transformation logic may depend on the key as well as the value.

        All other properties of the transformed map, such as iteration order, are left intact. For example, the code: 

        
 NavigableMap<String, Boolean> options = Maps.newTreeMap();
 options.put("verbose", false);
 options.put("sort", true);
 EntryTransformer<String, Boolean, String> flagPrefixer =
     new EntryTransformer<String, Boolean, String>() {
       public String transformEntry(String key, Boolean value) {
         return value ? key : ("yes" + key);
       }
     };
 NavigableMap<String, String> transformed =
     LabsMaps.transformNavigableEntries(options, flagPrefixer);
 System.out.println(transformed);
        ... prints {sort=yessort, verbose=verbose}.

        Changes in the underlying map are reflected in this view. Conversely, this view supports removal operations, and these are reflected in the underlying map.

        It's acceptable for the underlying map to contain null keys and null values provided that the transformer is capable of accepting null inputs. The transformed map might contain null values if the transformer sometimes gives a null result.

        The returned map is not thread-safe or serializable, even if the underlying map is.

        The transformer is applied lazily, invoked when needed. This is necessary for the returned map to be a view, but it means that the transformer will be applied many times for bulk operations like Map.containsValue(java.lang.Object) and Object.toString(). For this to perform well, transformer should be fast. To avoid lazy evaluation when the returned map doesn't need to be a view, copy the returned map into a new map of your choosing.

        Warning: This method assumes that for any instance k of EntryTransformer key type K, k.equals(k2) implies that k2 is also of type K. Using an EntryTransformer key type for which this may not hold, such as ArrayList, may risk a ClassCastException when calling methods on the transformed map.

        Since:
        13.0

      • filterKeys

        public static <K extends @Nullable Object,​V extends @Nullable Object> Map<K,​V> filterKeys​(Map<K,​V> unfiltered,
                                                                                                      Predicate<? super K> keyPredicate)
        Returns a map containing the mappings in unfiltered whose keys satisfy a predicate. The returned map is a live view of unfiltered; changes to one affect the other.

        The resulting map's keySet(), entrySet(), and values() views have iterators that don't support remove(), but all other methods are supported by the map and its views. When given a key that doesn't satisfy the predicate, the map's put() and putAll() methods throw an IllegalArgumentException.

        When methods such as removeAll() and clear() are called on the filtered map or its views, only mappings whose keys satisfy the filter will be removed from the underlying map.

        The returned map isn't threadsafe or serializable, even if unfiltered is.

        Many of the filtered map's methods, such as size(), iterate across every key/value mapping in the underlying map and determine which satisfy the filter. When a live view is not needed, it may be faster to copy the filtered map and use the copy.

        Warning: keyPredicate must be consistent with equals, as documented at Predicate.apply(T). Do not provide a predicate such as Predicates.instanceOf(ArrayList.class), which is inconsistent with equals.

      • filterKeys

        public static <K extends @Nullable Object,​V extends @Nullable Object> SortedMap<K,​V> filterKeys​(SortedMap<K,​V> unfiltered,
                                                                                                            Predicate<? super K> keyPredicate)
        Returns a sorted map containing the mappings in unfiltered whose keys satisfy a predicate. The returned map is a live view of unfiltered; changes to one affect the other.

        The resulting map's keySet(), entrySet(), and values() views have iterators that don't support remove(), but all other methods are supported by the map and its views. When given a key that doesn't satisfy the predicate, the map's put() and putAll() methods throw an IllegalArgumentException.

        When methods such as removeAll() and clear() are called on the filtered map or its views, only mappings whose keys satisfy the filter will be removed from the underlying map.

        The returned map isn't threadsafe or serializable, even if unfiltered is.

        Many of the filtered map's methods, such as size(), iterate across every key/value mapping in the underlying map and determine which satisfy the filter. When a live view is not needed, it may be faster to copy the filtered map and use the copy.

        Warning: keyPredicate must be consistent with equals, as documented at Predicate.apply(T). Do not provide a predicate such as Predicates.instanceOf(ArrayList.class), which is inconsistent with equals.

        Since:
        11.0

      • filterKeys

        @GwtIncompatible
public static <K extends @Nullable Object,​V extends @Nullable Object> NavigableMap<K,​V> filterKeys​(NavigableMap<K,​V> unfiltered,
                                                                                                               Predicate<? super K> keyPredicate)
        Returns a navigable map containing the mappings in unfiltered whose keys satisfy a predicate. The returned map is a live view of unfiltered; changes to one affect the other.

        The resulting map's keySet(), entrySet(), and values() views have iterators that don't support remove(), but all other methods are supported by the map and its views. When given a key that doesn't satisfy the predicate, the map's put() and putAll() methods throw an IllegalArgumentException.

        When methods such as removeAll() and clear() are called on the filtered map or its views, only mappings whose keys satisfy the filter will be removed from the underlying map.

        The returned map isn't threadsafe or serializable, even if unfiltered is.

        Many of the filtered map's methods, such as size(), iterate across every key/value mapping in the underlying map and determine which satisfy the filter. When a live view is not needed, it may be faster to copy the filtered map and use the copy.

        Warning: keyPredicate must be consistent with equals, as documented at Predicate.apply(T). Do not provide a predicate such as Predicates.instanceOf(ArrayList.class), which is inconsistent with equals.

        Since:
        14.0

      • filterKeys

        public static <K extends @Nullable Object,​V extends @Nullable Object> BiMap<K,​V> filterKeys​(BiMap<K,​V> unfiltered,
                                                                                                        Predicate<? super K> keyPredicate)
        Returns a bimap containing the mappings in unfiltered whose keys satisfy a predicate. The returned bimap is a live view of unfiltered; changes to one affect the other.

        The resulting bimap's keySet(), entrySet(), and values() views have iterators that don't support remove(), but all other methods are supported by the bimap and its views. When given a key that doesn't satisfy the predicate, the bimap's put(), forcePut() and putAll() methods throw an IllegalArgumentException.

        When methods such as removeAll() and clear() are called on the filtered bimap or its views, only mappings that satisfy the filter will be removed from the underlying bimap.

        The returned bimap isn't threadsafe or serializable, even if unfiltered is.

        Many of the filtered bimap's methods, such as size(), iterate across every key in the underlying bimap and determine which satisfy the filter. When a live view is not needed, it may be faster to copy the filtered bimap and use the copy.

        Warning: entryPredicate must be consistent with equals , as documented at Predicate.apply(T).

        Since:
        14.0

      • filterValues

        public static <K extends @Nullable Object,​V extends @Nullable Object> Map<K,​V> filterValues​(Map<K,​V> unfiltered,
                                                                                                        Predicate<? super V> valuePredicate)
        Returns a map containing the mappings in unfiltered whose values satisfy a predicate. The returned map is a live view of unfiltered; changes to one affect the other.

        The resulting map's keySet(), entrySet(), and values() views have iterators that don't support remove(), but all other methods are supported by the map and its views. When given a value that doesn't satisfy the predicate, the map's put(), putAll(), and Map.Entry.setValue(V) methods throw an IllegalArgumentException.

        When methods such as removeAll() and clear() are called on the filtered map or its views, only mappings whose values satisfy the filter will be removed from the underlying map.

        The returned map isn't threadsafe or serializable, even if unfiltered is.

        Many of the filtered map's methods, such as size(), iterate across every key/value mapping in the underlying map and determine which satisfy the filter. When a live view is not needed, it may be faster to copy the filtered map and use the copy.

        Warning: valuePredicate must be consistent with equals, as documented at Predicate.apply(T). Do not provide a predicate such as Predicates.instanceOf(ArrayList.class), which is inconsistent with equals.

      • filterValues

        public static <K extends @Nullable Object,​V extends @Nullable Object> SortedMap<K,​V> filterValues​(SortedMap<K,​V> unfiltered,
                                                                                                              Predicate<? super V> valuePredicate)
        Returns a sorted map containing the mappings in unfiltered whose values satisfy a predicate. The returned map is a live view of unfiltered; changes to one affect the other.

        The resulting map's keySet(), entrySet(), and values() views have iterators that don't support remove(), but all other methods are supported by the map and its views. When given a value that doesn't satisfy the predicate, the map's put(), putAll(), and Map.Entry.setValue(V) methods throw an IllegalArgumentException.

        When methods such as removeAll() and clear() are called on the filtered map or its views, only mappings whose values satisfy the filter will be removed from the underlying map.

        The returned map isn't threadsafe or serializable, even if unfiltered is.

        Many of the filtered map's methods, such as size(), iterate across every key/value mapping in the underlying map and determine which satisfy the filter. When a live view is not needed, it may be faster to copy the filtered map and use the copy.

        Warning: valuePredicate must be consistent with equals, as documented at Predicate.apply(T). Do not provide a predicate such as Predicates.instanceOf(ArrayList.class), which is inconsistent with equals.

        Since:
        11.0

      • filterValues

        @GwtIncompatible
public static <K extends @Nullable Object,​V extends @Nullable Object> NavigableMap<K,​V> filterValues​(NavigableMap<K,​V> unfiltered,
                                                                                                                 Predicate<? super V> valuePredicate)
        Returns a navigable map containing the mappings in unfiltered whose values satisfy a predicate. The returned map is a live view of unfiltered; changes to one affect the other.

        The resulting map's keySet(), entrySet(), and values() views have iterators that don't support remove(), but all other methods are supported by the map and its views. When given a value that doesn't satisfy the predicate, the map's put(), putAll(), and Map.Entry.setValue(V) methods throw an IllegalArgumentException.

        When methods such as removeAll() and clear() are called on the filtered map or its views, only mappings whose values satisfy the filter will be removed from the underlying map.

        The returned map isn't threadsafe or serializable, even if unfiltered is.

        Many of the filtered map's methods, such as size(), iterate across every key/value mapping in the underlying map and determine which satisfy the filter. When a live view is not needed, it may be faster to copy the filtered map and use the copy.

        Warning: valuePredicate must be consistent with equals, as documented at Predicate.apply(T). Do not provide a predicate such as Predicates.instanceOf(ArrayList.class), which is inconsistent with equals.

        Since:
        14.0

      • filterValues

        public static <K extends @Nullable Object,​V extends @Nullable Object> BiMap<K,​V> filterValues​(BiMap<K,​V> unfiltered,
                                                                                                          Predicate<? super V> valuePredicate)
        Returns a bimap containing the mappings in unfiltered whose values satisfy a predicate. The returned bimap is a live view of unfiltered; changes to one affect the other.

        The resulting bimap's keySet(), entrySet(), and values() views have iterators that don't support remove(), but all other methods are supported by the bimap and its views. When given a value that doesn't satisfy the predicate, the bimap's put(), forcePut() and putAll() methods throw an IllegalArgumentException. Similarly, the map's entries have a Map.Entry.setValue(V) method that throws an IllegalArgumentException when the provided value doesn't satisfy the predicate.

        When methods such as removeAll() and clear() are called on the filtered bimap or its views, only mappings that satisfy the filter will be removed from the underlying bimap.

        The returned bimap isn't threadsafe or serializable, even if unfiltered is.

        Many of the filtered bimap's methods, such as size(), iterate across every value in the underlying bimap and determine which satisfy the filter. When a live view is not needed, it may be faster to copy the filtered bimap and use the copy.

        Warning: entryPredicate must be consistent with equals , as documented at Predicate.apply(T).

        Since:
        14.0

      • filterEntries

        public static <K extends @Nullable Object,​V extends @Nullable Object> Map<K,​V> filterEntries​(Map<K,​V> unfiltered,
                                                                                                         Predicate<? super Map.Entry<K,​V>> entryPredicate)
        Returns a map containing the mappings in unfiltered that satisfy a predicate. The returned map is a live view of unfiltered; changes to one affect the other.

        The resulting map's keySet(), entrySet(), and values() views have iterators that don't support remove(), but all other methods are supported by the map and its views. When given a key/value pair that doesn't satisfy the predicate, the map's put() and putAll() methods throw an IllegalArgumentException. Similarly, the map's entries have a Map.Entry.setValue(V) method that throws an IllegalArgumentException when the existing key and the provided value don't satisfy the predicate.

        When methods such as removeAll() and clear() are called on the filtered map or its views, only mappings that satisfy the filter will be removed from the underlying map.

        The returned map isn't threadsafe or serializable, even if unfiltered is.

        Many of the filtered map's methods, such as size(), iterate across every key/value mapping in the underlying map and determine which satisfy the filter. When a live view is not needed, it may be faster to copy the filtered map and use the copy.

        Warning: entryPredicate must be consistent with equals, as documented at Predicate.apply(T).

      • filterEntries

        public static <K extends @Nullable Object,​V extends @Nullable Object> SortedMap<K,​V> filterEntries​(SortedMap<K,​V> unfiltered,
                                                                                                               Predicate<? super Map.Entry<K,​V>> entryPredicate)
        Returns a sorted map containing the mappings in unfiltered that satisfy a predicate. The returned map is a live view of unfiltered; changes to one affect the other.

        The resulting map's keySet(), entrySet(), and values() views have iterators that don't support remove(), but all other methods are supported by the map and its views. When given a key/value pair that doesn't satisfy the predicate, the map's put() and putAll() methods throw an IllegalArgumentException. Similarly, the map's entries have a Map.Entry.setValue(V) method that throws an IllegalArgumentException when the existing key and the provided value don't satisfy the predicate.

        When methods such as removeAll() and clear() are called on the filtered map or its views, only mappings that satisfy the filter will be removed from the underlying map.

        The returned map isn't threadsafe or serializable, even if unfiltered is.

        Many of the filtered map's methods, such as size(), iterate across every key/value mapping in the underlying map and determine which satisfy the filter. When a live view is not needed, it may be faster to copy the filtered map and use the copy.

        Warning: entryPredicate must be consistent with equals, as documented at Predicate.apply(T).

        Since:
        11.0

      • filterEntries

        @GwtIncompatible
public static <K extends @Nullable Object,​V extends @Nullable Object> NavigableMap<K,​V> filterEntries​(NavigableMap<K,​V> unfiltered,
                                                                                                                  Predicate<? super Map.Entry<K,​V>> entryPredicate)
        Returns a sorted map containing the mappings in unfiltered that satisfy a predicate. The returned map is a live view of unfiltered; changes to one affect the other.

        The resulting map's keySet(), entrySet(), and values() views have iterators that don't support remove(), but all other methods are supported by the map and its views. When given a key/value pair that doesn't satisfy the predicate, the map's put() and putAll() methods throw an IllegalArgumentException. Similarly, the map's entries have a Map.Entry.setValue(V) method that throws an IllegalArgumentException when the existing key and the provided value don't satisfy the predicate.

        When methods such as removeAll() and clear() are called on the filtered map or its views, only mappings that satisfy the filter will be removed from the underlying map.

        The returned map isn't threadsafe or serializable, even if unfiltered is.

        Many of the filtered map's methods, such as size(), iterate across every key/value mapping in the underlying map and determine which satisfy the filter. When a live view is not needed, it may be faster to copy the filtered map and use the copy.

        Warning: entryPredicate must be consistent with equals, as documented at Predicate.apply(T).

        Since:
        14.0

      • filterEntries

        public static <K extends @Nullable Object,​V extends @Nullable Object> BiMap<K,​V> filterEntries​(BiMap<K,​V> unfiltered,
                                                                                                           Predicate<? super Map.Entry<K,​V>> entryPredicate)
        Returns a bimap containing the mappings in unfiltered that satisfy a predicate. The returned bimap is a live view of unfiltered; changes to one affect the other.

        The resulting bimap's keySet(), entrySet(), and values() views have iterators that don't support remove(), but all other methods are supported by the bimap and its views. When given a key/value pair that doesn't satisfy the predicate, the bimap's put(), forcePut() and putAll() methods throw an IllegalArgumentException. Similarly, the map's entries have an Map.Entry.setValue(V) method that throws an IllegalArgumentException when the existing key and the provided value don't satisfy the predicate.

        When methods such as removeAll() and clear() are called on the filtered bimap or its views, only mappings that satisfy the filter will be removed from the underlying bimap.

        The returned bimap isn't threadsafe or serializable, even if unfiltered is.

        Many of the filtered bimap's methods, such as size(), iterate across every key/value mapping in the underlying bimap and determine which satisfy the filter. When a live view is not needed, it may be faster to copy the filtered bimap and use the copy.

        Warning: entryPredicate must be consistent with equals , as documented at Predicate.apply(T).

        Since:
        14.0

      • unmodifiableNavigableMap

        @GwtIncompatible
public static <K extends @Nullable Object,​V extends @Nullable Object> NavigableMap<K,​V> unmodifiableNavigableMap​(NavigableMap<K,​? extends V> map)
        Returns an unmodifiable view of the specified navigable map. Query operations on the returned map read through to the specified map, and attempts to modify the returned map, whether direct or via its views, result in an UnsupportedOperationException.

        The returned navigable map will be serializable if the specified navigable map is serializable.

        This method's signature will not permit you to convert a NavigableMap<? extends K, V> to a NavigableMap<K, V>. If it permitted this, the returned map's comparator() method might return a Comparator<? extends K>, which works only on a particular subtype of K, but promise that it's a Comparator<? super K>, which must work on any type of K.

        Parameters:
        map - the navigable map for which an unmodifiable view is to be returned
        Returns:
        an unmodifiable view of the specified navigable map
        Since:
        12.0

      • synchronizedNavigableMap

        @GwtIncompatible
public static <K extends @Nullable Object,​V extends @Nullable Object> NavigableMap<K,​V> synchronizedNavigableMap​(NavigableMap<K,​V> navigableMap)
        Returns a synchronized (thread-safe) navigable map backed by the specified navigable map. In order to guarantee serial access, it is critical that all access to the backing navigable map is accomplished through the returned navigable map (or its views).

        It is imperative that the user manually synchronize on the returned navigable map when iterating over any of its collection views, or the collections views of any of its descendingMap, subMap, headMap or tailMap views. 

        
 NavigableMap<K, V> map = synchronizedNavigableMap(new TreeMap<K, V>());

 // Needn't be in synchronized block
 NavigableSet<K> set = map.navigableKeySet();

 synchronized (map) { // Synchronizing on map, not set!
   Iterator<K> it = set.iterator(); // Must be in synchronized block
   while (it.hasNext()) {
     foo(it.next());
   }
 }

        or: 

        
 NavigableMap<K, V> map = synchronizedNavigableMap(new TreeMap<K, V>());
 NavigableMap<K, V> map2 = map.subMap(foo, false, bar, true);

 // Needn't be in synchronized block
 NavigableSet<K> set2 = map2.descendingKeySet();

 synchronized (map) { // Synchronizing on map, not map2 or set2!
   Iterator<K> it = set2.iterator(); // Must be in synchronized block
   while (it.hasNext()) {
     foo(it.next());
   }
 }

        Failure to follow this advice may result in non-deterministic behavior.

        The returned navigable map will be serializable if the specified navigable map is serializable.

        Parameters:
        navigableMap - the navigable map to be "wrapped" in a synchronized navigable map.
        Returns:
        a synchronized view of the specified navigable map.
        Since:
        13.0

      • subMap

        @Beta
@GwtIncompatible
public static <K extends Comparable<? super K>,​V extends @Nullable Object> NavigableMap<K,​V> subMap​(NavigableMap<K,​V> map,
                                                                                                                Range<K> range)
        Returns a view of the portion of map whose keys are contained by range.

        This method delegates to the appropriate methods of NavigableMap (namely subMap(), tailMap(), and headMap()) to actually construct the view. Consult these methods for a full description of the returned view's behavior.

        Warning: Ranges always represent a range of values using the values' natural ordering. NavigableMap on the other hand can specify a custom ordering via a Comparator, which can violate the natural ordering. Using this method (or in general using Range) with unnaturally-ordered maps can lead to unexpected and undefined behavior.

        Since:
        20.0