Module java.base
Package java.lang

Class Process

java.lang.Object
java.lang.Process

public abstract class Process extends Object
Process provides control of native processes started by ProcessBuilder.start and Runtime.exec. The class provides methods for performing input from the process, performing output to the process, waiting for the process to complete, checking the exit status of the process, and destroying (killing) the process. The ProcessBuilder.start() and Runtime.exec methods create a native process and return an instance of a subclass of Process that can be used to control the process and obtain information about it.

The methods that create processes may not work well for special processes on certain native platforms, such as native windowing processes, daemon processes, Win16/DOS processes on Microsoft Windows, or shell scripts.

By default, the created process does not have its own terminal or console. All its standard I/O (i.e. stdin, stdout, stderr) operations will be redirected to the parent process, where they can be accessed via the streams obtained using the methods getOutputStream(), getInputStream(), and getErrorStream(). The I/O streams of characters and lines can be written and read using the methods outputWriter(), outputWriter(Charset)}, inputReader(), inputReader(Charset), errorReader(), and errorReader(Charset). The parent process uses these streams to feed input to and get output from the process. Because some native platforms only provide limited buffer size for standard input and output streams, failure to promptly write the input stream or read the output stream of the process may cause the process to block, or even deadlock.

Where desired, process I/O can also be redirected using methods of the ProcessBuilder class.

The process is not killed when there are no more references to the Process object, but rather the process continues executing asynchronously.

There is no requirement that the process represented by a Process object execute asynchronously or concurrently with respect to the Java process that owns the Process object.

As of 1.5, ProcessBuilder.start() is the preferred way to create a Process.

Subclasses of Process should override the onExit() and toHandle() methods to provide a fully functional Process including the process id, information about the process, direct children, and direct children plus descendants of those children of the process. Delegating to the underlying Process or ProcessHandle is typically easiest and most efficient.

Since:
1.0
  • Constructor Details

    • Process

      public Process()
      Default constructor for Process.
  • Method Details

    • getOutputStream

      public abstract OutputStream getOutputStream()
      Returns the output stream connected to the normal input of the process. Output to the stream is piped into the standard input of the process represented by this Process object.

      If the standard input of the process has been redirected using ProcessBuilder.redirectInput then this method will return a null output stream.

      API Note:
      When writing to both getOutputStream() and either outputWriter() or outputWriter(Charset), BufferedWriter.flush should be called before writes to the OutputStream.
      Implementation Note:
      Implementation note: It is a good idea for the returned output stream to be buffered.
      Returns:
      the output stream connected to the normal input of the process
    • getInputStream

      public abstract InputStream getInputStream()
      Returns the input stream connected to the normal output of the process. The stream obtains data piped from the standard output of the process represented by this Process object.

      If the standard output of the process has been redirected using ProcessBuilder.redirectOutput then this method will return a null input stream.

      Otherwise, if the standard error of the process has been redirected using ProcessBuilder.redirectErrorStream then the input stream returned by this method will receive the merged standard output and the standard error of the process.

      API Note:
      Use getInputStream() and inputReader() with extreme care. The BufferedReader may have buffered input from the input stream.
      Implementation Note:
      Implementation note: It is a good idea for the returned input stream to be buffered.
      Returns:
      the input stream connected to the normal output of the process
    • getErrorStream

      public abstract InputStream getErrorStream()
      Returns the input stream connected to the error output of the process. The stream obtains data piped from the error output of the process represented by this Process object.

      If the standard error of the process has been redirected using ProcessBuilder.redirectError or ProcessBuilder.redirectErrorStream then this method will return a null input stream.

      API Note:
      Use getErrorStream() and errorReader() with extreme care. The BufferedReader may have buffered input from the error stream.
      Implementation Note:
      Implementation note: It is a good idea for the returned input stream to be buffered.
      Returns:
      the input stream connected to the error output of the process
    • inputReader

      public final BufferedReader inputReader()
      Returns a BufferedReader connected to the standard output of the process. The Charset for the native encoding is used to read characters, lines, or stream lines from standard output.

      This method delegates to inputReader(Charset) using the Charset named by the native.encoding system property. If the native.encoding is not a valid charset name or not supported the Charset.defaultCharset() is used.

      Returns:
      a BufferedReader using the native.encoding if supported, otherwise, the Charset.defaultCharset()
      Since:
      17
    • inputReader

      public final BufferedReader inputReader(Charset charset)
      Returns a BufferedReader connected to the standard output of this process using a Charset. The BufferedReader can be used to read characters, lines, or stream lines of the standard output.

      Characters are read by an InputStreamReader that reads and decodes bytes from this process getInputStream(). Bytes are decoded to characters using the charset; malformed-input and unmappable-character sequences are replaced with the charset's default replacement. The BufferedReader reads and buffers characters from the InputStreamReader.

      The first call to this method creates the BufferedReader, if called again with the same charset the same BufferedReader is returned. It is an error to call this method again with a different charset.

      If the standard output of the process has been redirected using ProcessBuilder.redirectOutput then the InputStreamReader will be reading from a null input stream.

      Otherwise, if the standard error of the process has been redirected using ProcessBuilder.redirectErrorStream then the input reader returned by this method will receive the merged standard output and the standard error of the process.

      API Note:
      Using both getInputStream() and inputReader(Charset) has unpredictable behavior since the buffered reader reads ahead from the input stream.

      When the process has terminated, and the standard input has not been redirected, reading of the bytes available from the underlying stream is on a best effort basis and may be unpredictable.

      Parameters:
      charset - the Charset used to decode bytes to characters
      Returns:
      a BufferedReader for the standard output of the process using the charset
      Throws:
      NullPointerException - if the charset is null
      IllegalStateException - if called more than once with different charset arguments
      Since:
      17
    • errorReader

      public final BufferedReader errorReader()
      Returns a BufferedReader connected to the standard error of the process. The Charset for the native encoding is used to read characters, lines, or stream lines from standard error.

      This method delegates to errorReader(Charset) using the Charset named by the native.encoding system property. If the native.encoding is not a valid charset name or not supported the Charset.defaultCharset() is used.

      Returns:
      a BufferedReader using the native.encoding if supported, otherwise, the Charset.defaultCharset()
      Since:
      17
    • errorReader

      public final BufferedReader errorReader(Charset charset)
      Returns a BufferedReader connected to the standard error of this process using a Charset. The BufferedReader can be used to read characters, lines, or stream lines of the standard error.

      Characters are read by an InputStreamReader that reads and decodes bytes from this process getErrorStream(). Bytes are decoded to characters using the charset; malformed-input and unmappable-character sequences are replaced with the charset's default replacement. The BufferedReader reads and buffers characters from the InputStreamReader.

      The first call to this method creates the BufferedReader, if called again with the same charset the same BufferedReader is returned. It is an error to call this method again with a different charset.

      If the standard error of the process has been redirected using ProcessBuilder.redirectError or ProcessBuilder.redirectErrorStream then the InputStreamReader will be reading from a null input stream.

      API Note:
      Using both getErrorStream() and errorReader(Charset) has unpredictable behavior since the buffered reader reads ahead from the error stream.

      When the process has terminated, and the standard error has not been redirected, reading of the bytes available from the underlying stream is on a best effort basis and may be unpredictable.

      Parameters:
      charset - the Charset used to decode bytes to characters
      Returns:
      a BufferedReader for the standard error of the process using the charset
      Throws:
      NullPointerException - if the charset is null
      IllegalStateException - if called more than once with different charset arguments
      Since:
      17
    • outputWriter

      public final BufferedWriter outputWriter()
      Returns a BufferedWriter connected to the normal input of the process using the native encoding. Writes text to a character-output stream, buffering characters so as to provide for the efficient writing of single characters, arrays, and strings.

      This method delegates to outputWriter(Charset) using the Charset named by the native.encoding system property. If the native.encoding is not a valid charset name or not supported the Charset.defaultCharset() is used.

      Returns:
      a BufferedWriter to the standard input of the process using the charset for the native.encoding system property
      Since:
      17
    • outputWriter

      public final BufferedWriter outputWriter(Charset charset)
      Returns a BufferedWriter connected to the normal input of the process using a Charset. Writes text to a character-output stream, buffering characters so as to provide for the efficient writing of single characters, arrays, and strings.

      Characters written by the writer are encoded to bytes using OutputStreamWriter and the Charset are written to the standard input of the process represented by this Process. Malformed-input and unmappable-character sequences are replaced with the charset's default replacement.

      The first call to this method creates the BufferedWriter, if called again with the same charset the same BufferedWriter is returned. It is an error to call this method again with a different charset.

      If the standard input of the process has been redirected using ProcessBuilder.redirectInput then the OutputStreamWriter writes to a null output stream.

      API Note:
      A BufferedWriter writes characters, arrays of characters, and strings. Wrapping the BufferedWriter with a PrintWriter provides efficient buffering and formatting of primitives and objects as well as support for auto-flush on line endings. Call the BufferedWriter.flush() method to flush buffered output to the process.

      When writing to both getOutputStream() and either outputWriter() or outputWriter(Charset), BufferedWriter.flush should be called before writes to the OutputStream.

      Parameters:
      charset - the Charset to encode characters to bytes
      Returns:
      a BufferedWriter to the standard input of the process using the charset
      Throws:
      NullPointerException - if the charset is null
      IllegalStateException - if called more than once with different charset arguments
      Since:
      17
    • waitFor

      public abstract int waitFor() throws InterruptedException
      Causes the current thread to wait, if necessary, until the process represented by this Process object has terminated. This method returns immediately if the process has already terminated. If the process has not yet terminated, the calling thread will be blocked until the process exits.
      Returns:
      the exit value of the process represented by this Process object. By convention, the value 0 indicates normal termination.
      Throws:
      InterruptedException - if the current thread is interrupted by another thread while it is waiting, then the wait is ended and an InterruptedException is thrown.
    • waitFor

      public boolean waitFor(long timeout, TimeUnit unit) throws InterruptedException
      Causes the current thread to wait, if necessary, until the process represented by this Process object has terminated, or the specified waiting time elapses.

      If the process has already terminated then this method returns immediately with the value true. If the process has not terminated and the timeout value is less than, or equal to, zero, then this method returns immediately with the value false.

      The default implementation of this method polls the exitValue to check if the process has terminated. Concrete implementations of this class are strongly encouraged to override this method with a more efficient implementation.

      Parameters:
      timeout - the maximum time to wait
      unit - the time unit of the timeout argument
      Returns:
      true if the process has exited and false if the waiting time elapsed before the process has exited.
      Throws:
      InterruptedException - if the current thread is interrupted while waiting.
      NullPointerException - if unit is null
      Since:
      1.8
    • exitValue

      public abstract int exitValue()
      Returns the exit value for the process.
      Returns:
      the exit value of the process represented by this Process object. By convention, the value 0 indicates normal termination.
      Throws:
      IllegalThreadStateException - if the process represented by this Process object has not yet terminated
    • destroy

      public abstract void destroy()
      Kills the process. Whether the process represented by this Process object is normally terminated or not is implementation dependent. Forcible process destruction is defined as the immediate termination of a process, whereas normal termination allows the process to shut down cleanly. If the process is not alive, no action is taken.

      The CompletableFuture from onExit() is completed when the process has terminated.

    • destroyForcibly

      public Process destroyForcibly()
      Kills the process forcibly. The process represented by this Process object is forcibly terminated. Forcible process destruction is defined as the immediate termination of a process, whereas normal termination allows the process to shut down cleanly. If the process is not alive, no action is taken.

      The CompletableFuture from onExit() is completed when the process has terminated.

      Invoking this method on Process objects returned by ProcessBuilder.start() and Runtime.exec(java.lang.String) forcibly terminate the process.

      API Note:
      The process may not terminate immediately. i.e. isAlive() may return true for a brief period after destroyForcibly() is called. This method may be chained to waitFor() if needed.
      Implementation Requirements:
      The default implementation of this method invokes destroy() and so may not forcibly terminate the process.
      Implementation Note:
      Concrete implementations of this class are strongly encouraged to override this method with a compliant implementation.
      Returns:
      the Process object representing the process forcibly destroyed
      Since:
      1.8
    • supportsNormalTermination

      public boolean supportsNormalTermination()
      Returns true if the implementation of destroy() is to normally terminate the process, Returns false if the implementation of destroy forcibly and immediately terminates the process.

      Invoking this method on Process objects returned by ProcessBuilder.start() and Runtime.exec(java.lang.String) return true or false depending on the platform implementation.

      Implementation Requirements:
      This implementation throws an instance of UnsupportedOperationException and performs no other action.
      Returns:
      true if the implementation of destroy() is to normally terminate the process; otherwise, destroy() forcibly terminates the process
      Throws:
      UnsupportedOperationException - if the Process implementation does not support this operation
      Since:
      9
    • isAlive

      public boolean isAlive()
      Tests whether the process represented by this Process is alive.
      Returns:
      true if the process represented by this Process object has not yet terminated.
      Since:
      1.8
    • pid

      public long pid()
      Returns the native process ID of the process. The native process ID is an identification number that the operating system assigns to the process.
      Implementation Requirements:
      The implementation of this method returns the process id as: toHandle().pid().
      Returns:
      the native process id of the process
      Throws:
      UnsupportedOperationException - if the Process implementation does not support this operation
      Since:
      9
    • onExit

      public CompletableFuture<Process> onExit()
      Returns a CompletableFuture<Process> for the termination of the Process. The CompletableFuture provides the ability to trigger dependent functions or actions that may be run synchronously or asynchronously upon process termination. When the process has terminated the CompletableFuture is completed regardless of the exit status of the process.

      Calling onExit().get() waits for the process to terminate and returns the Process. The future can be used to check if the process is done or to wait for it to terminate. Cancelling the CompletableFuture does not affect the Process.

      Processes returned from ProcessBuilder.start() override the default implementation to provide an efficient mechanism to wait for process exit.

      API Note:
      Using onExit is an alternative to waitFor that enables both additional concurrency and convenient access to the result of the Process. Lambda expressions can be used to evaluate the result of the Process execution. If there is other processing to be done before the value is used then onExit is a convenient mechanism to free the current thread and block only if and when the value is needed.
      For example, launching a process to compare two files and get a boolean if they are identical:
         Process p = new ProcessBuilder("cmp", "f1", "f2").start();
          Future<Boolean> identical = p.onExit().thenApply(p1 -> p1.exitValue() == 0);
          ...
          if (identical.get()) { ... }
       
      , The process may be observed to have terminated with isAlive() before the ComputableFuture is completed and dependent actions are invoked.
      Implementation Requirements:
      This implementation executes waitFor() in a separate thread repeatedly until it returns successfully. If the execution of waitFor is interrupted, the thread's interrupt status is preserved.

      When waitFor() returns successfully the CompletableFuture is completed regardless of the exit status of the process. This implementation may consume a lot of memory for thread stacks if a large number of processes are waited for concurrently.

      External implementations should override this method and provide a more efficient implementation. For example, to delegate to the underlying process, it can do the following:

      
          public CompletableFuture<Process> onExit() {
             return delegate.onExit().thenApply(p -> this);
          }
       
      Returns:
      a new CompletableFuture<Process> for the Process
      Since:
      9
    • toHandle

      public ProcessHandle toHandle()
      Returns a ProcessHandle for the Process. Process objects returned by ProcessBuilder.start() and Runtime.exec(java.lang.String) implement toHandle as the equivalent of ProcessHandle.of(pid) including the check for a SecurityManager and RuntimePermission("manageProcess").
      Implementation Requirements:
      This implementation throws an instance of UnsupportedOperationException and performs no other action. Subclasses should override this method to provide a ProcessHandle for the process. The methods pid(), info(), children(), and descendants(), unless overridden, operate on the ProcessHandle.
      Returns:
      Returns a ProcessHandle for the Process
      Throws:
      UnsupportedOperationException - if the Process implementation does not support this operation
      SecurityException - if a security manager has been installed and it denies RuntimePermission("manageProcess")
      Since:
      9
    • info

      public ProcessHandle.Info info()
      Returns a snapshot of information about the process.

      A ProcessHandle.Info instance has accessor methods that return information about the process if it is available.

      Implementation Requirements:
      This implementation returns information about the process as: toHandle().info().
      Returns:
      a snapshot of information about the process, always non-null
      Throws:
      UnsupportedOperationException - if the Process implementation does not support this operation
      Since:
      9
    • children

      public Stream<ProcessHandle> children()
      Returns a snapshot of the direct children of the process. The parent of a direct child process is the process. Typically, a process that is not alive has no children.

      Note that processes are created and terminate asynchronously. There is no guarantee that a process is alive.

      Implementation Requirements:
      This implementation returns the direct children as: toHandle().children().
      Returns:
      a sequential Stream of ProcessHandles for processes that are direct children of the process
      Throws:
      UnsupportedOperationException - if the Process implementation does not support this operation
      SecurityException - if a security manager has been installed and it denies RuntimePermission("manageProcess")
      Since:
      9
    • descendants

      public Stream<ProcessHandle> descendants()
      Returns a snapshot of the descendants of the process. The descendants of a process are the children of the process plus the descendants of those children, recursively. Typically, a process that is not alive has no children.

      Note that processes are created and terminate asynchronously. There is no guarantee that a process is alive.

      Implementation Requirements:
      This implementation returns all children as: toHandle().descendants().
      Returns:
      a sequential Stream of ProcessHandles for processes that are descendants of the process
      Throws:
      UnsupportedOperationException - if the Process implementation does not support this operation
      SecurityException - if a security manager has been installed and it denies RuntimePermission("manageProcess")
      Since:
      9