public abstract class PipeReaderThread
extends java.lang.Thread
doReading(java.io.InputStream)
method to process the data; this method runs
in the new thread.
Here is an example of using the class to count the bytes written down a stream:
PipeReaderThread reader = new PipeReaderThread() { protected void doReading( InputStream dataIn ) throws IOException { int i; while ( dataIn.read() >= 0 ) i++; System.out.println( i ); } }; reader.start(); OutputStream dataOut = reader.getOutputStream(); // write bytes down dataOut ... dataOut.close(); reader.finishReading();Other uses will look pretty similar, but just override
doReading
in different ways. Note that any exceptions thrown by doReading
are caught and eventually thrown in the reader thread by
finishReading
. The same exception may also be thrown by
the write
method of the writer thread.
This class serves two purposes. Firstly it copes with IOExceptions
encountered during the read, and makes sure they get thrown at the
writing end of the pipe (doReading
is declared to
throw IOException
).
Secondly it shields the user from the implementation of the piped
connection.
Performance of the
PipedInputStream
/PipedOutputStream
is dismal - this class may be able to do better.
The current implementation uses a couple of drop-in Piped*Stream
replacements, but performance still isn't great - it may be possible
to do better in future. You can provide your own paired pipes
by overriding both getInputStream()
and getOutputStream()
.
Constructor and Description |
---|
PipeReaderThread()
Constructs a new reader thread.
|
Modifier and Type | Method and Description |
---|---|
protected abstract void |
doReading(java.io.InputStream dataIn)
This method should be implemented to consume all the bytes in
the given input stream.
|
void |
finishReading()
Waits until the
doReading method has finished reading
the bytes written down the output stream, closes the input stream,
and returns. |
protected java.io.InputStream |
getInputStream()
Returns the stream at the input end of the pipe.
|
java.io.OutputStream |
getOutputStream()
Returns the stream at the output end of the pipe.
|
void |
run()
Implements the thread's
run method to invoke doReading,
catching and saving IOExceptions. |
activeCount, checkAccess, clone, countStackFrames, currentThread, destroy, dumpStack, enumerate, getAllStackTraces, getContextClassLoader, getDefaultUncaughtExceptionHandler, getId, getName, getPriority, getStackTrace, getState, getThreadGroup, getUncaughtExceptionHandler, holdsLock, interrupt, interrupted, isAlive, isDaemon, isInterrupted, join, join, join, resume, setContextClassLoader, setDaemon, setDefaultUncaughtExceptionHandler, setName, setPriority, setUncaughtExceptionHandler, sleep, sleep, start, stop, stop, suspend, toString, yield
public PipeReaderThread() throws java.io.IOException
java.io.IOException
protected java.io.InputStream getInputStream()
public java.io.OutputStream getOutputStream()
public void run()
run
method to invoke doReading,
catching and saving IOExceptions.run
in interface java.lang.Runnable
run
in class java.lang.Thread
protected abstract void doReading(java.io.InputStream dataIn) throws java.io.IOException
dataIn
to the end of the stream (either closing it early or
just stopping reading) may cause an IOException to be thrown in
the thread which is writing to the PipedOutputStream.
Implementations should not close the supplied input stream.dataIn
- stream which will supply bytesjava.io.IOException
- if any I/O error occurs; this exception will
be saved and thrown later by the finishReading
methodpublic void finishReading() throws java.io.IOException
doReading
method has finished reading
the bytes written down the output stream, closes the input stream,
and returns.
Any IOException which has occurred during the read will be thrown
by this method.java.io.InterruptedIOException
- if failure was caused by interruption;
the bytesTransferred
field of this exception
is not set to a useful valuejava.io.IOException
- in case of some other IO failure