A {@link Watchable} object is registered with a watch service by invokingits {@link Watchable#register register} method, returning a {@link WatchKey}to represent the registration. When an event for an object is detected the key is signalled, and if not currently signalled, it is queued to the watch service so that it can be retrieved by consumers that invoke the {@link #poll() poll} or {@link #take() take} methods to retrieve keysand process events. Once the events have been processed the consumer invokes the key's {@link WatchKey#reset reset} method to reset the key whichallows the key to be signalled and re-queued with further events.
Registration with a watch service is cancelled by invoking the key's {@link WatchKey#cancel cancel} method. A key that is queued at the time thatit is cancelled remains in the queue until it is retrieved. Depending on the object, a key may be cancelled automatically. For example, suppose a directory is watched and the watch service detects that it has been deleted or its file system is no longer accessible. When a key is cancelled in this manner it is signalled and queued, if not currently signalled. To ensure that the consumer is notified the return value from the {@code reset}method indicates if the key is valid.
A watch service is safe for use by multiple concurrent consumers. To ensure that only one consumer processes the events for a particular object at any time then care should be taken to ensure that the key's {@code reset}method is only invoked after its events have been processed. The {@link #close close} method may be invoked at any time to close the service causingany threads waiting to retrieve keys, to throw {@code ClosedWatchServiceException}.
File systems may report events faster than they can be retrieved or processed and an implementation may impose an unspecified limit on the number of events that it may accumulate. Where an implementation knowingly discards events then it arranges for the key's {@link WatchKey#pollEvents pollEvents} method to return an element with an event type of {@link StandardWatchEventKinds#OVERFLOW OVERFLOW}. This event can be used by the consumer as a trigger to re-examine the state of the object.
When an event is reported to indicate that a file in a watched directory has been modified then there is no guarantee that the program (or programs) that have modified the file have completed. Care should be taken to coordinate access with other programs that may be updating the file. The {@link java.nio.channels.FileChannel FileChannel} class defines methodsto lock regions of a file against access by other programs.
The implementation that observes events from the file system is intended to map directly on to the native file event notification facility where available, or to use a primitive mechanism, such as polling, when a native facility is not available. Consequently, many of the details on how events are detected, their timeliness, and whether their ordering is preserved are highly implementation specific. For example, when a file in a watched directory is modified then it may result in a single {@link StandardWatchEventKinds#ENTRY_MODIFY ENTRY_MODIFY} event in someimplementations but several events in other implementations. Short-lived files (meaning files that are deleted very quickly after they are created) may not be detected by primitive implementations that periodically poll the file system to detect changes.
If a watched file is not located on a local storage device then it is implementation specific if changes to the file can be detected. In particular, it is not required that changes to files carried out on remote systems be detected. @since 1.7 @see FileSystem#newWatchService
| |
| |