Class Caffeine<K,V>

java.lang.Object
com.github.benmanes.caffeine.cache.Caffeine<K,V>
Type Parameters:
K - the base key type for all caches created by this builder
V - the base value type for all caches created by this builder

public final class Caffeine<K,V> extends Object
A builder of AsyncLoadingCache, LoadingCache, and Cache instances having a combination of the following features:
  • automatic loading of entries into the cache, optionally asynchronously
  • size-based eviction when a maximum is exceeded based on frequency and recency
  • time-based expiration of entries, measured since last access or last write
  • asynchronously refresh when the first stale request for an entry occurs
  • keys automatically wrapped in weak references
  • values automatically wrapped in weak or soft references
  • writes propagated to an external resource
  • notification of evicted (or otherwise removed) entries
  • accumulation of cache access statistics

These features are all optional; caches can be created using all or none of them. By default cache instances created by Caffeine will not perform any type of eviction.

Usage example:


   LoadingCache<Key, Graph> graphs = Caffeine.newBuilder()
       .maximumSize(10_000)
       .expireAfterWrite(10, TimeUnit.MINUTES)
       .removalListener((Key key, Graph graph, RemovalCause cause) ->
           System.out.printf("Key %s was removed (%s)%n", key, cause))
       .build(key -> createExpensiveGraph(key));
 

The returned cache is implemented as a hash table with similar performance characteristics to ConcurrentHashMap. The asMap view (and its collection views) have weakly consistent iterators. This means that they are safe for concurrent use, but if other threads modify the cache after the iterator is created, it is undefined which of these changes, if any, are reflected in that iterator. These iterators never throw ConcurrentModificationException.

Note: by default, the returned cache uses equality comparisons (the equals method) to determine equality for keys or values. However, if weakKeys() was specified, the cache uses identity (==) comparisons instead for keys. Likewise, if weakValues() or softValues() was specified, the cache uses identity comparisons for values.

Entries are automatically evicted from the cache when any of maximumSize, maximumWeight, expireAfterWrite, expireAfterAccess, weakKeys, weakValues, or softValues are requested.

If maximumSize or maximumWeight is requested entries may be evicted on each cache modification.

If expireAfterWrite or expireAfterAccess is requested entries may be evicted on each cache modification, on occasional cache accesses, or on calls to Cache.cleanUp(). Expired entries may be counted by Cache.estimatedSize(), but will never be visible to read or write operations.

If weakKeys, weakValues, or softValues are requested, it is possible for a key or value present in the cache to be reclaimed by the garbage collector. Entries with reclaimed keys or values may be removed from the cache on each cache modification, on occasional cache accesses, or on calls to Cache.cleanUp(); such entries may be counted in Cache.estimatedSize(), but will never be visible to read or write operations.

Certain cache configurations will result in the accrual of periodic maintenance tasks which will be performed during write operations, or during occasional read operations in the absence of writes. The Cache.cleanUp() method of the returned cache will also perform maintenance, but calling it should not be necessary with a high throughput cache. Only caches built with maximumSize, maximumWeight, expireAfterWrite, expireAfterAccess, weakKeys, weakValues, or softValues perform periodic maintenance.

The caches produced by Caffeine are serializable, and the deserialized caches retain all the configuration properties of the original cache. Note that the serialized form does not include cache contents, but only configuration.

  • Field Details

    • logger

      static final Logger logger
    • ENABLED_STATS_COUNTER_SUPPLIER

      static final Supplier<StatsCounter> ENABLED_STATS_COUNTER_SUPPLIER
    • UNSET_INT

      static final int UNSET_INT
      See Also:
    • DEFAULT_INITIAL_CAPACITY

      static final int DEFAULT_INITIAL_CAPACITY
      See Also:
    • DEFAULT_EXPIRATION_NANOS

      static final int DEFAULT_EXPIRATION_NANOS
      See Also:
    • DEFAULT_REFRESH_NANOS

      static final int DEFAULT_REFRESH_NANOS
      See Also:
    • strictParsing

      boolean strictParsing
    • maximumSize

      long maximumSize
    • maximumWeight

      long maximumWeight
    • initialCapacity

      int initialCapacity
    • refreshNanos

      long refreshNanos
    • expireAfterWriteNanos

      long expireAfterWriteNanos
    • expireAfterAccessNanos

      long expireAfterAccessNanos
    • removalListener

      RemovalListener<? super K,? super V> removalListener
    • statsCounterSupplier

      Supplier<StatsCounter> statsCounterSupplier
    • writer

      CacheWriter<? super K,? super V> writer
    • weigher

      Weigher<? super K,? super V> weigher
    • executor

      Executor executor
    • ticker

      Ticker ticker
    • keyStrength

      Caffeine.Strength keyStrength
    • valueStrength

      Caffeine.Strength valueStrength
  • Constructor Details

    • Caffeine

      private Caffeine()
  • Method Details

    • requireArgument

      static void requireArgument(boolean expression, String template, Object... args)
      Ensures that the argument expression is true.
    • requireArgument

      static void requireArgument(boolean expression)
      Ensures that the argument expression is true.
    • requireState

      static void requireState(boolean expression)
      Ensures that the state expression is true.
    • requireState

      static void requireState(boolean expression, String template, Object... args)
      Ensures that the state expression is true.
    • newBuilder

      @Nonnull public static Caffeine<Object,Object> newBuilder()
      Constructs a new Caffeine instance with default settings, including strong keys, strong values, and no automatic eviction of any kind.
      Returns:
      a new instance with default settings
    • from

      @Nonnull public static Caffeine<Object,Object> from(CaffeineSpec spec)
      Constructs a new Caffeine instance with the settings specified in spec.
      Parameters:
      spec - the specification to build from
      Returns:
      a new instance with the specification's settings
    • from

      @Nonnull public static Caffeine<Object,Object> from(String spec)
      Constructs a new Caffeine instance with the settings specified in spec.
      Parameters:
      spec - a String in the format specified by CaffeineSpec
      Returns:
      a new instance with the specification's settings
    • initialCapacity

      @Nonnull public Caffeine<K,V> initialCapacity(@Nonnegative int initialCapacity)
      Sets the minimum total size for the internal data structures. Providing a large enough estimate at construction time avoids the need for expensive resizing operations later, but setting this value unnecessarily high wastes memory.
      Parameters:
      initialCapacity - minimum total size for the internal data structures
      Returns:
      this builder instance
      Throws:
      IllegalArgumentException - if initialCapacity is negative
      IllegalStateException - if an initial capacity was already set
    • hasInitialCapacity

      boolean hasInitialCapacity()
    • getInitialCapacity

      int getInitialCapacity()
    • executor

      @Nonnull public Caffeine<K,V> executor(@Nonnull Executor executor)
      Specifies the executor to use when running asynchronous tasks. The executor is delegated to when sending removal notifications, when asynchronous computations are performed by AsyncLoadingCache or LoadingCache.refresh(K) or refreshAfterWrite(long, java.util.concurrent.TimeUnit), or when performing periodic maintenance. By default, ForkJoinPool.commonPool() is used.

      The primary intent of this method is to facilitate testing of caches which have been configured with removalListener or utilize asynchronous computations. A test may instead prefer to configure the cache to execute tasks directly on the same thread.

      Beware that configuring a cache with an executor that throws RejectedExecutionException may experience non-deterministic behavior.

      Parameters:
      executor - the executor to use for asynchronous execution
      Returns:
      this builder instance
      Throws:
      NullPointerException - if the specified executor is null
    • getExecutor

      @Nonnull Executor getExecutor()
    • maximumSize

      @Nonnull public Caffeine<K,V> maximumSize(@Nonnegative long maximumSize)
      Specifies the maximum number of entries the cache may contain. Note that the cache may evict an entry before this limit is exceeded or temporarily exceed the threshold while evicting. As the cache size grows close to the maximum, the cache evicts entries that are less likely to be used again. For example, the cache may evict an entry because it hasn't been used recently or very often.

      When size is zero, elements will be evicted immediately after being loaded into the cache. This can be useful in testing, or to disable caching temporarily without a code change.

      This feature cannot be used in conjunction with maximumWeight.

      Parameters:
      maximumSize - the maximum size of the cache
      Returns:
      this builder instance
      Throws:
      IllegalArgumentException - if size is negative
      IllegalStateException - if a maximum size or weight was already set
    • maximumWeight

      @Nonnull public Caffeine<K,V> maximumWeight(@Nonnegative long maximumWeight)
      Specifies the maximum weight of entries the cache may contain. Weight is determined using the Weigher specified with weigher, and use of this method requires a corresponding call to weigher prior to calling build().

      Note that the cache may evict an entry before this limit is exceeded or temporarily exceed the threshold while evicting. As the cache size grows close to the maximum, the cache evicts entries that are less likely to be used again. For example, the cache may evict an entry because it hasn't been used recently or very often.

      When maximumWeight is zero, elements will be evicted immediately after being loaded into cache. This can be useful in testing, or to disable caching temporarily without a code change.

      Note that weight is only used to determine whether the cache is over capacity; it has no effect on selecting which entry should be evicted next.

      This feature cannot be used in conjunction with maximumSize.

      Parameters:
      maximumWeight - the maximum total weight of entries the cache may contain
      Returns:
      this builder instance
      Throws:
      IllegalArgumentException - if maximumWeight is negative
      IllegalStateException - if a maximum weight or size was already set
    • weigher

      @Nonnull public <K1 extends K, V1 extends V> Caffeine<K1,V1> weigher(@Nonnull Weigher<? super K1,? super V1> weigher)
      Specifies the weigher to use in determining the weight of entries. Entry weight is taken into consideration by maximumWeight(long) when determining which entries to evict, and use of this method requires a corresponding call to maximumWeight(long) prior to calling build(). Weights are measured and recorded when entries are inserted into the cache, and are thus effectively static during the lifetime of a cache entry.

      When the weight of an entry is zero it will not be considered for size-based eviction (though it still may be evicted by other means).

      Important note: Instead of returning this as a Caffeine instance, this method returns Caffeine<K1, V1>. From this point on, either the original reference or the returned reference may be used to complete configuration and build the cache, but only the "generic" one is type-safe. That is, it will properly prevent you from building caches whose key or value types are incompatible with the types accepted by the weigher already provided; the Caffeine type cannot do this. For best results, simply use the standard method-chaining idiom, as illustrated in the documentation at top, configuring a Caffeine and building your Cache all in a single statement.

      Warning: if you ignore the above advice, and use this Caffeine to build a cache whose key or value type is incompatible with the weigher, you will likely experience a ClassCastException at some undefined point in the future.

      Type Parameters:
      K1 - key type of the weigher
      V1 - value type of the weigher
      Parameters:
      weigher - the weigher to use in calculating the weight of cache entries
      Returns:
      the cache builder reference that should be used instead of this for any remaining configuration and cache building
      Throws:
      IllegalArgumentException - if size is negative
      IllegalStateException - if a maximum size was already set
    • evicts

      boolean evicts()
    • isWeighted

      boolean isWeighted()
    • getMaximum

      @Nonnegative long getMaximum()
    • getWeigher

      @Nonnull <K1 extends K, V1 extends V> Weigher<K1,V1> getWeigher(boolean isAsync)
    • weakKeys

      @Nonnull public Caffeine<K,V> weakKeys()
      Specifies that each key (not value) stored in the cache should be wrapped in a WeakReference (by default, strong references are used).

      Warning: when this method is used, the resulting cache will use identity (==) comparison to determine equality of keys.

      Entries with keys that have been garbage collected may be counted in Cache.estimatedSize(), but will never be visible to read or write operations; such entries are cleaned up as part of the routine maintenance described in the class javadoc.

      This feature cannot be used in conjunction with writer.

      Returns:
      this builder instance
      Throws:
      IllegalStateException - if the key strength was already set or the writer was set
    • isStrongKeys

      boolean isStrongKeys()
    • isWeakKeys

      boolean isWeakKeys()
    • weakValues

      @Nonnull public Caffeine<K,V> weakValues()
      Specifies that each value (not key) stored in the cache should be wrapped in a WeakReference (by default, strong references are used).

      Weak values will be garbage collected once they are weakly reachable. This makes them a poor candidate for caching; consider softValues() instead.

      Note: when this method is used, the resulting cache will use identity (==) comparison to determine equality of values.

      Entries with values that have been garbage collected may be counted in Cache.estimatedSize(), but will never be visible to read or write operations; such entries are cleaned up as part of the routine maintenance described in the class javadoc.

      This feature cannot be used in conjunction with buildAsync(com.github.benmanes.caffeine.cache.CacheLoader<? super K1, V1>).

      Returns:
      this builder instance
      Throws:
      IllegalStateException - if the value strength was already set
    • isStrongValues

      boolean isStrongValues()
    • isWeakValues

      boolean isWeakValues()
    • isSoftValues

      boolean isSoftValues()
    • softValues

      @Nonnull public Caffeine<K,V> softValues()
      Specifies that each value (not key) stored in the cache should be wrapped in a SoftReference (by default, strong references are used). Softly-referenced objects will be garbage-collected in a globally least-recently-used manner, in response to memory demand.

      Warning: in most circumstances it is better to set a per-cache maximum size instead of using soft references. You should only use this method if you are very familiar with the practical consequences of soft references.

      Note: when this method is used, the resulting cache will use identity (==) comparison to determine equality of values.

      Entries with values that have been garbage collected may be counted in Cache.estimatedSize(), but will never be visible to read or write operations; such entries are cleaned up as part of the routine maintenance described in the class javadoc.

      This feature cannot be used in conjunction with buildAsync(com.github.benmanes.caffeine.cache.CacheLoader<? super K1, V1>).

      Returns:
      this builder instance
      Throws:
      IllegalStateException - if the value strength was already set
    • expireAfterWrite

      @Nonnull public Caffeine<K,V> expireAfterWrite(@Nonnegative long duration, @Nonnull TimeUnit unit)
      Specifies that each entry should be automatically removed from the cache once a fixed duration has elapsed after the entry's creation, or the most recent replacement of its value.

      Expired entries may be counted in Cache.estimatedSize(), but will never be visible to read or write operations. Expired entries are cleaned up as part of the routine maintenance described in the class javadoc.

      Parameters:
      duration - the length of time after an entry is created that it should be automatically removed
      unit - the unit that duration is expressed in
      Returns:
      this builder instance
      Throws:
      IllegalArgumentException - if duration is negative
      IllegalStateException - if the time to live or time to idle was already set
    • getExpiresAfterWriteNanos

      @Nonnegative long getExpiresAfterWriteNanos()
    • expiresAfterWrite

      boolean expiresAfterWrite()
    • expireAfterAccess

      @Nonnull public Caffeine<K,V> expireAfterAccess(@Nonnegative long duration, @Nonnull TimeUnit unit)
      Specifies that each entry should be automatically removed from the cache once a fixed duration has elapsed after the entry's creation, the most recent replacement of its value, or its last read. Access time is reset by all cache read and write operations (including Cache.asMap().get(Object) and Cache.asMap().put(K, V)), but not by operations on the collection-views of Cache.asMap().

      Expired entries may be counted in Cache.estimatedSize(), but will never be visible to read or write operations. Expired entries are cleaned up as part of the routine maintenance described in the class javadoc.

      Parameters:
      duration - the length of time after an entry is last accessed that it should be automatically removed
      unit - the unit that duration is expressed in
      Returns:
      this builder instance
      Throws:
      IllegalArgumentException - if duration is negative
      IllegalStateException - if the time to idle or time to live was already set
    • getExpiresAfterAccessNanos

      @Nonnegative long getExpiresAfterAccessNanos()
    • expiresAfterAccess

      boolean expiresAfterAccess()
    • refreshAfterWrite

      @Nonnull public Caffeine<K,V> refreshAfterWrite(@Nonnegative long duration, @Nonnull TimeUnit unit)
      Specifies that active entries are eligible for automatic refresh once a fixed duration has elapsed after the entry's creation, or the most recent replacement of its value. The semantics of refreshes are specified in LoadingCache.refresh(K), and are performed by calling CacheLoader.reload(K, V).

      Automatic refreshes are performed when the first stale request for an entry occurs. The request triggering refresh will make an asynchronous call to CacheLoader.reload(K, V) and immediately return the old value.

      Note: all exceptions thrown during refresh will be logged and then swallowed.

      Parameters:
      duration - the length of time after an entry is created that it should be considered stale, and thus eligible for refresh
      unit - the unit that duration is expressed in
      Returns:
      this builder instance
      Throws:
      IllegalArgumentException - if duration is negative
      IllegalStateException - if the refresh interval was already set
    • getRefreshAfterWriteNanos

      @Nonnegative long getRefreshAfterWriteNanos()
    • refreshes

      boolean refreshes()
    • ticker

      @Nonnull public Caffeine<K,V> ticker(@Nonnull Ticker ticker)
      Specifies a nanosecond-precision time source for use in determining when entries should be expired or refreshed. By default, System.nanoTime() is used.

      The primary intent of this method is to facilitate testing of caches which have been configured with expireAfterWrite(long, java.util.concurrent.TimeUnit), expireAfterAccess(long, java.util.concurrent.TimeUnit), or refreshAfterWrite(long, java.util.concurrent.TimeUnit).

      Parameters:
      ticker - a nanosecond-precision time source
      Returns:
      this builder instance
      Throws:
      IllegalStateException - if a ticker was already set
      NullPointerException - if the specified ticker is null
    • getTicker

      @Nonnull Ticker getTicker()
    • removalListener

      @Nonnull public <K1 extends K, V1 extends V> Caffeine<K1,V1> removalListener(@Nonnull RemovalListener<? super K1,? super V1> removalListener)
      Specifies a listener instance that caches should notify each time an entry is removed for any reason. Each cache created by this builder will invoke this listener as part of the routine maintenance described in the class documentation above.

      Warning: after invoking this method, do not continue to use this cache builder reference; instead use the reference this method returns. At runtime, these point to the same instance, but only the returned reference has the correct generic type information so as to ensure type safety. For best results, use the standard method-chaining idiom illustrated in the class documentation above, configuring a builder and building your cache in a single statement. Failure to heed this advice can result in a ClassCastException being thrown by a cache operation at some undefined point in the future.

      Warning: any exception thrown by listener will not be propagated to the Cache user, only logged via a Logger.

      Type Parameters:
      K1 - the key type of the listener
      V1 - the value type of the listener
      Parameters:
      removalListener - a listener instance that caches should notify each time an entry is removed
      Returns:
      the cache builder reference that should be used instead of this for any remaining configuration and cache building
      Throws:
      IllegalStateException - if a removal listener was already set
      NullPointerException - if the specified removal listener is null
    • getRemovalListener

      <K1 extends K, V1 extends V> RemovalListener<K1,V1> getRemovalListener(boolean async)
    • writer

      @Nonnull public <K1 extends K, V1 extends V> Caffeine<K1,V1> writer(@Nonnull CacheWriter<? super K1,? super V1> writer)
      Specifies a writer instance that caches should notify each time an entry is explicitly created or modified, or removed for any reason. The writer is not notified when an entry is loaded or computed. Each cache created by this builder will invoke this writer as part of the atomic operation that modifies the cache.

      Warning: after invoking this method, do not continue to use this cache builder reference; instead use the reference this method returns. At runtime, these point to the same instance, but only the returned reference has the correct generic type information so as to ensure type safety. For best results, use the standard method-chaining idiom illustrated in the class documentation above, configuring a builder and building your cache in a single statement. Failure to heed this advice can result in a ClassCastException being thrown by a cache operation at some undefined point in the future.

      Warning: any exception thrown by writer will be propagated to the Cache user.

      This feature cannot be used in conjunction with weakKeys() or buildAsync(com.github.benmanes.caffeine.cache.CacheLoader<? super K1, V1>).

      Type Parameters:
      K1 - the key type of the writer
      V1 - the value type of the writer
      Parameters:
      writer - a writer instance that caches should notify each time an entry is explicitly created or modified, or removed for any reason
      Returns:
      the cache builder reference that should be used instead of this for any remaining configuration and cache building
      Throws:
      IllegalStateException - if a writer was already set or if the key strength is weak
      NullPointerException - if the specified writer is null
    • getCacheWriter

      <K1 extends K, V1 extends V> CacheWriter<K1,V1> getCacheWriter()
    • recordStats

      @Nonnull public Caffeine<K,V> recordStats()
      Enables the accumulation of CacheStats during the operation of the cache. Without this Cache.stats() will return zero for all statistics. Note that recording statistics requires bookkeeping to be performed with each operation, and thus imposes a performance penalty on cache operation.
      Returns:
      this builder instance
    • recordStats

      @Nonnull public Caffeine<K,V> recordStats(@Nonnull Supplier<? extends StatsCounter> statsCounterSupplier)
      Enables the accumulation of CacheStats during the operation of the cache. Without this Cache.stats() will return zero for all statistics. Note that recording statistics requires bookkeeping to be performed with each operation, and thus imposes a performance penalty on cache operation. Any exception thrown by the supplied StatsCounter will be suppressed and logged.
      Parameters:
      statsCounterSupplier - a supplier instance that returns a new StatsCounter
      Returns:
      this builder instance
    • isRecordingStats

      boolean isRecordingStats()
    • getStatsCounterSupplier

      @Nonnull Supplier<? extends StatsCounter> getStatsCounterSupplier()
    • isBounded

      boolean isBounded()
    • build

      @Nonnull public <K1 extends K, V1 extends V> Cache<K1,V1> build()
      Builds a cache which does not automatically load values when keys are requested.

      Consider build(CacheLoader) instead, if it is feasible to implement a CacheLoader.

      This method does not alter the state of this Caffeine instance, so it can be invoked again to create multiple independent caches.

      Type Parameters:
      K1 - the key type of the cache
      V1 - the value type of the cache
      Returns:
      a cache having the requested features
    • build

      @Nonnull public <K1 extends K, V1 extends V> LoadingCache<K1,V1> build(@Nonnull CacheLoader<? super K1,V1> loader)
      Builds a cache, which either returns an already-loaded value for a given key or atomically computes or retrieves it using the supplied CacheLoader. If another thread is currently loading the value for this key, simply waits for that thread to finish and returns its loaded value. Note that multiple threads can concurrently load values for distinct keys.

      This method does not alter the state of this Caffeine instance, so it can be invoked again to create multiple independent caches.

      Type Parameters:
      K1 - the key type of the loader
      V1 - the value type of the loader
      Parameters:
      loader - the cache loader used to obtain new values
      Returns:
      a cache having the requested features
      Throws:
      NullPointerException - if the specified cache loader is null
    • buildAsync

      @Nonnull public <K1 extends K, V1 extends V> AsyncLoadingCache<K1,V1> buildAsync(@Nonnull CacheLoader<? super K1,V1> loader)
      Builds a cache, which either returns a CompletableFuture already loaded or currently computing the value for a given key, or atomically computes the value asynchronously through a supplied mapping function or the supplied CacheLoader. If the asynchronous computation fails or computes a null value then the entry will be automatically removed. Note that multiple threads can concurrently load values for distinct keys.

      This method does not alter the state of this Caffeine instance, so it can be invoked again to create multiple independent caches.

      Type Parameters:
      K1 - the key type of the loader
      V1 - the value type of the loader
      Parameters:
      loader - the cache loader used to obtain new values
      Returns:
      a cache having the requested features
      Throws:
      IllegalStateException - if the value strength is weak or soft
      NullPointerException - if the specified cache loader is null
    • buildAsync

      @Nonnull public <K1 extends K, V1 extends V> AsyncLoadingCache<K1,V1> buildAsync(@Nonnull AsyncCacheLoader<? super K1,V1> loader)
      Builds a cache, which either returns a CompletableFuture already loaded or currently computing the value for a given key, or atomically computes the value asynchronously through a supplied mapping function or the supplied AsyncCacheLoader. If the asynchronous computation fails or computes a null value then the entry will be automatically removed. Note that multiple threads can concurrently load values for distinct keys.

      This method does not alter the state of this Caffeine instance, so it can be invoked again to create multiple independent caches.

      Type Parameters:
      K1 - the key type of the loader
      V1 - the value type of the loader
      Parameters:
      loader - the cache loader used to obtain new values
      Returns:
      a cache having the requested features
      Throws:
      IllegalStateException - if the value strength is weak or soft
      NullPointerException - if the specified cache loader is null
    • requireNonLoadingCache

      void requireNonLoadingCache()
    • requireWeightWithWeigher

      void requireWeightWithWeigher()
    • toString

      public String toString()
      Returns a string representation for this Caffeine instance. The exact form of the returned string is not specified.
      Overrides:
      toString in class Object