com.tangosol.net.NamedMap Maven / Gradle / Ivy
Show all versions of coherence Show documentation
/*
* Copyright (c) 2020 Oracle and/or its affiliates.
*
* Licensed under the Universal Permissive License v 1.0 as shown at
* http://oss.oracle.com/licenses/upl.
*/
package com.tangosol.net;
import com.tangosol.internal.util.DefaultAsyncNamedCache;
import com.tangosol.util.AsynchronousAgent;
import com.tangosol.util.ConcurrentMap;
import com.tangosol.util.InvocableMap;
import com.tangosol.util.ObservableMap;
import com.tangosol.util.QueryMap;
import java.util.Collection;
import java.util.Map;
import java.util.Objects;
import java.util.function.BiConsumer;
/**
* A Map-based data-structure that manages entries across one or more processes.
* Entries are typically managed in memory, and are often comprised of data
* that is also stored persistently, on disk.
*
* @param the type of the map entry keys
* @param the type of the map entry values
*
* @author Aleks Seovic 2020.06.06
*
* @since 20.06
*/
public interface NamedMap
extends NamedCollection, ObservableMap, ConcurrentMap,
QueryMap, InvocableMap
{
@Override
default String getName()
{
return ((NamedCache) this).getCacheName();
}
@Override
default CacheService getService()
{
return ((NamedCache) this).getCacheService();
}
/**
* Get all the specified keys, if they are in the map. For each key
* that is in the map, that key and its corresponding value will be
* placed in the map that is returned by this method. The absence of
* a key in the returned map indicates that it was not in the map,
* which may imply (for maps that can load behind the scenes) that
* the requested data could not be loaded.
*
* The result of this method is defined to be semantically the same as
* the following implementation, without regards to threading issues:
*
*
* Map map = new AnyMap(); // could be a HashMap (but does not have to)
* for (Iterator iter = colKeys.iterator(); iter.hasNext(); )
* {
* Object oKey = iter.next();
* Object oVal = get(oKey);
* if (oVal != null || containsKey(oKey))
* {
* map.put(oKey, oVal);
* }
* }
* return map;
*
*
* @param colKeys a collection of keys that may be in the named map
*
* @return a Map of keys to values for the specified keys passed in
* colKeys
*/
Map getAll(Collection colKeys);
/**
* Removes all mappings from this map.
*
* Note: invoking the {@code clear()} operation against a distributed map
* can be both a memory and CPU intensive task and therefore is generally
* not recommended. Either {@link #truncate()} or {@link #destroy()} operations
* may be suitable alternatives.
*/
void clear();
/**
* Removes all mappings from this map.
*
* Note: the removal of entries caused by this truncate operation will
* not be observable. This includes any registered {@link com.tangosol.util.MapListener
* listeners}, {@link com.tangosol.util.MapTrigger triggers}, or {@link
* com.tangosol.net.events.EventInterceptor interceptors}. However, a
* {@link com.tangosol.net.events.partition.cache.CacheLifecycleEvent CacheLifecycleEvent}
* is raised to notify subscribers of the execution of this operation.
*
* @throws UnsupportedOperationException if the server does not support the truncate operation
*/
default void truncate()
{
throw new UnsupportedOperationException();
}
/**
* Perform the given action for each entry selected by the specified key set
* until all entries have been processed or the action throws an exception.
*
* Exceptions thrown by the action are relayed to the caller.
*
* The implementation processes each entry on the client and should only be
* used for read-only client-side operations (such as adding map entries to
* a UI widget, for example).
*
* Any entry mutation caused by the specified action will not be propagated
* to the server when this method is called on a distributed map, so it
* should be avoided. The mutating operations on a subset of entries
* should be implemented using one of {@link InvocableMap#invokeAll},
* {@link #replaceAll}, {@link #compute}, or {@link #merge} methods instead.
*
* @param collKeys the keys to process; these keys are not required to
* exist within the Map
* @param action the action to be performed for each entry
*
* @since 12.2.1
*/
public default void forEach(Collection collKeys, BiConsumer action)
{
Objects.requireNonNull(action);
getAll(collKeys).forEach(action);
}
/**
* Return an asynchronous wrapper for this NamedMap.
*
* By default, the order of execution of asynchronous operation invoked on
* the returned AsyncNamedMap will be preserved by ensuring that all
* operations invoked from the same client thread are executed on the server
* sequentially, using the same {@link AsynchronousAgent#getUnitOfOrderId
* unit-of-order}. This tends to provide best performance for fast,
* non-blocking operations.
*
* However, when invoking CPU-intensive or blocking operations, such as
* read- or write-through operations that access remote database or web
* service, for example, it may be very beneficial to allow the server to
* parallelize execution by passing {@link AsyncNamedMap.OrderBy#none()}
* configuration option to this method. Note, that in that case there are
* no guarantees for the order of execution.
*
* @param options the configuration options
*
* @return asynchronous wrapper for this NamedMap
*/
default AsyncNamedMap async(AsyncNamedMap.Option... options)
{
return new DefaultAsyncNamedCache<>((NamedCache) this, options);
}
/**
* Construct a {@code view} of this {@link NamedMap}.
*
* @param the type of the entry values in this {@code view}, which
* will be the same as {@code V_BACK}, unless a {@code transformer} is specified
* when creating this {@code view}
*
* @return a local {@code view} for this {@link NamedMap}
*
* @see ViewBuilder
*
* @since 12.2.1.4
*/
public default MapViewBuilder view()
{
return new MapViewBuilder<>(this);
}
// ----- NamedMap.Option interface --------------------------------------
/**
* An immutable option for requesting and configuring {@link NamedMap}s.
*/
interface Option
{
}
}