Guido van Rossum | 5fdeeea | 1994-01-02 01:22:07 +0000 | [diff] [blame] | 1 | \section{Built-in Module \sectcode{thread}} |
Guido van Rossum | e47da0a | 1997-07-17 16:34:52 +0000 | [diff] [blame] | 2 | \label{module-thread} |
Guido van Rossum | 5fdeeea | 1994-01-02 01:22:07 +0000 | [diff] [blame] | 3 | \bimodindex{thread} |
| 4 | |
| 5 | This module provides low-level primitives for working with multiple |
Guido van Rossum | 6bb1adc | 1995-03-13 10:03:32 +0000 | [diff] [blame] | 6 | threads (a.k.a.\ \dfn{light-weight processes} or \dfn{tasks}) --- multiple |
Guido van Rossum | 5fdeeea | 1994-01-02 01:22:07 +0000 | [diff] [blame] | 7 | threads of control sharing their global data space. For |
Guido van Rossum | 6bb1adc | 1995-03-13 10:03:32 +0000 | [diff] [blame] | 8 | synchronization, simple locks (a.k.a.\ \dfn{mutexes} or \dfn{binary |
Guido van Rossum | 5fdeeea | 1994-01-02 01:22:07 +0000 | [diff] [blame] | 9 | semaphores}) are provided. |
Fred Drake | 61b0452 | 1998-01-20 05:52:23 +0000 | [diff] [blame] | 10 | \index{light-weight processes} |
| 11 | \index{processes, light-weight} |
| 12 | \index{binary semaphores} |
| 13 | \index{semaphores, binary} |
Guido van Rossum | 5fdeeea | 1994-01-02 01:22:07 +0000 | [diff] [blame] | 14 | |
Guido van Rossum | 3572d37 | 1997-08-14 19:51:26 +0000 | [diff] [blame] | 15 | The module is optional. It is supported on Windows NT and '95, SGI |
Fred Drake | 65b32f7 | 1998-02-09 20:27:12 +0000 | [diff] [blame] | 16 | IRIX, Solaris 2.x, as well as on systems that have a \POSIX{} thread |
Guido van Rossum | 3572d37 | 1997-08-14 19:51:26 +0000 | [diff] [blame] | 17 | (a.k.a. ``pthread'') implementation. |
Fred Drake | 61b0452 | 1998-01-20 05:52:23 +0000 | [diff] [blame] | 18 | \index{pthreads} |
Fred Drake | 1624a50 | 1998-02-09 22:12:28 +0000 | [diff] [blame] | 19 | \indexii{threads}{\POSIX{}} |
Guido van Rossum | 5fdeeea | 1994-01-02 01:22:07 +0000 | [diff] [blame] | 20 | |
| 21 | It defines the following constant and functions: |
| 22 | |
Guido van Rossum | 5fdeeea | 1994-01-02 01:22:07 +0000 | [diff] [blame] | 23 | \begin{excdesc}{error} |
| 24 | Raised on thread-specific errors. |
| 25 | \end{excdesc} |
| 26 | |
Fred Drake | cce1090 | 1998-03-17 06:33:25 +0000 | [diff] [blame] | 27 | \begin{funcdesc}{start_new_thread}{func, arg} |
Guido van Rossum | 5fdeeea | 1994-01-02 01:22:07 +0000 | [diff] [blame] | 28 | Start a new thread. The thread executes the function \var{func} |
| 29 | with the argument list \var{arg} (which must be a tuple). When the |
Guido van Rossum | 6bb1adc | 1995-03-13 10:03:32 +0000 | [diff] [blame] | 30 | function returns, the thread silently exits. When the function |
Guido van Rossum | 5fdeeea | 1994-01-02 01:22:07 +0000 | [diff] [blame] | 31 | terminates with an unhandled exception, a stack trace is printed and |
| 32 | then the thread exits (but other threads continue to run). |
| 33 | \end{funcdesc} |
| 34 | |
Guido van Rossum | 470be14 | 1995-03-17 16:07:09 +0000 | [diff] [blame] | 35 | \begin{funcdesc}{exit}{} |
Fred Drake | d678cb7 | 1998-04-03 06:35:54 +0000 | [diff] [blame] | 36 | This is a shorthand for \function{exit_thread()}. |
Guido van Rossum | 5fdeeea | 1994-01-02 01:22:07 +0000 | [diff] [blame] | 37 | \end{funcdesc} |
| 38 | |
Guido van Rossum | 470be14 | 1995-03-17 16:07:09 +0000 | [diff] [blame] | 39 | \begin{funcdesc}{exit_thread}{} |
Fred Drake | d678cb7 | 1998-04-03 06:35:54 +0000 | [diff] [blame] | 40 | Raise the \exception{SystemExit} exception. When not caught, this |
| 41 | will cause the thread to exit silently. |
Guido van Rossum | 5fdeeea | 1994-01-02 01:22:07 +0000 | [diff] [blame] | 42 | \end{funcdesc} |
| 43 | |
Guido van Rossum | 470be14 | 1995-03-17 16:07:09 +0000 | [diff] [blame] | 44 | %\begin{funcdesc}{exit_prog}{status} |
| 45 | %Exit all threads and report the value of the integer argument |
| 46 | %\var{status} as the exit status of the entire program. |
| 47 | %\strong{Caveat:} code in pending \code{finally} clauses, in this thread |
| 48 | %or in other threads, is not executed. |
| 49 | %\end{funcdesc} |
| 50 | |
Guido van Rossum | 5fdeeea | 1994-01-02 01:22:07 +0000 | [diff] [blame] | 51 | \begin{funcdesc}{allocate_lock}{} |
| 52 | Return a new lock object. Methods of locks are described below. The |
| 53 | lock is initially unlocked. |
| 54 | \end{funcdesc} |
| 55 | |
Guido van Rossum | b8b264b | 1994-08-12 13:13:50 +0000 | [diff] [blame] | 56 | \begin{funcdesc}{get_ident}{} |
| 57 | Return the `thread identifier' of the current thread. This is a |
| 58 | nonzero integer. Its value has no direct meaning; it is intended as a |
| 59 | magic cookie to be used e.g. to index a dictionary of thread-specific |
| 60 | data. Thread identifiers may be recycled when a thread exits and |
| 61 | another thread is created. |
| 62 | \end{funcdesc} |
| 63 | |
Fred Drake | d678cb7 | 1998-04-03 06:35:54 +0000 | [diff] [blame] | 64 | |
Guido van Rossum | 5fdeeea | 1994-01-02 01:22:07 +0000 | [diff] [blame] | 65 | Lock objects have the following methods: |
| 66 | |
Fred Drake | d678cb7 | 1998-04-03 06:35:54 +0000 | [diff] [blame] | 67 | \begin{methoddesc}[lock]{acquire}{\optional{waitflag}} |
Guido van Rossum | 5fdeeea | 1994-01-02 01:22:07 +0000 | [diff] [blame] | 68 | Without the optional argument, this method acquires the lock |
| 69 | unconditionally, if necessary waiting until it is released by another |
| 70 | thread (only one thread at a time can acquire a lock --- that's their |
| 71 | reason for existence), and returns \code{None}. If the integer |
Guido van Rossum | 6bb1adc | 1995-03-13 10:03:32 +0000 | [diff] [blame] | 72 | \var{waitflag} argument is present, the action depends on its value:\ |
Guido van Rossum | 5fdeeea | 1994-01-02 01:22:07 +0000 | [diff] [blame] | 73 | if it is zero, the lock is only acquired if it can be acquired |
| 74 | immediately without waiting, while if it is nonzero, the lock is |
| 75 | acquired unconditionally as before. If an argument is present, the |
Fred Drake | d678cb7 | 1998-04-03 06:35:54 +0000 | [diff] [blame] | 76 | return value is \code{1} if the lock is acquired successfully, |
| 77 | \code{0} if not. |
| 78 | \end{methoddesc} |
Guido van Rossum | 5fdeeea | 1994-01-02 01:22:07 +0000 | [diff] [blame] | 79 | |
Fred Drake | d678cb7 | 1998-04-03 06:35:54 +0000 | [diff] [blame] | 80 | \begin{methoddesc}[lock]{release}{} |
Guido van Rossum | 5fdeeea | 1994-01-02 01:22:07 +0000 | [diff] [blame] | 81 | Releases the lock. The lock must have been acquired earlier, but not |
| 82 | necessarily by the same thread. |
Fred Drake | d678cb7 | 1998-04-03 06:35:54 +0000 | [diff] [blame] | 83 | \end{methoddesc} |
Guido van Rossum | 5fdeeea | 1994-01-02 01:22:07 +0000 | [diff] [blame] | 84 | |
Fred Drake | d678cb7 | 1998-04-03 06:35:54 +0000 | [diff] [blame] | 85 | \begin{methoddesc}[lock]{locked}{} |
| 86 | Return the status of the lock:\ \code{1} if it has been acquired by |
| 87 | some thread, \code{0} if not. |
| 88 | \end{methoddesc} |
Guido van Rossum | 5fdeeea | 1994-01-02 01:22:07 +0000 | [diff] [blame] | 89 | |
Fred Drake | af8a015 | 1998-01-14 14:51:31 +0000 | [diff] [blame] | 90 | \strong{Caveats:} |
Guido van Rossum | 5fdeeea | 1994-01-02 01:22:07 +0000 | [diff] [blame] | 91 | |
| 92 | \begin{itemize} |
| 93 | \item |
| 94 | Threads interact strangely with interrupts: the |
Fred Drake | d678cb7 | 1998-04-03 06:35:54 +0000 | [diff] [blame] | 95 | \exception{KeyboardInterrupt} exception will be received by an |
| 96 | arbitrary thread. (When the \module{signal}\refbimodindex{signal} |
| 97 | module is available, interrupts always go to the main thread.) |
Guido van Rossum | 5fdeeea | 1994-01-02 01:22:07 +0000 | [diff] [blame] | 98 | |
| 99 | \item |
Fred Drake | d678cb7 | 1998-04-03 06:35:54 +0000 | [diff] [blame] | 100 | Calling \function{sys.exit()} or raising the \exception{SystemExit} |
| 101 | exception is equivalent to calling \function{exit_thread()}. |
Guido van Rossum | 5fdeeea | 1994-01-02 01:22:07 +0000 | [diff] [blame] | 102 | |
| 103 | \item |
| 104 | Not all built-in functions that may block waiting for I/O allow other |
Fred Drake | d678cb7 | 1998-04-03 06:35:54 +0000 | [diff] [blame] | 105 | threads to run. (The most popular ones (\function{time.sleep()}, |
| 106 | \method{\var{file}.read()}, \function{select.select()}) work as |
| 107 | expected.) |
Guido van Rossum | 5fdeeea | 1994-01-02 01:22:07 +0000 | [diff] [blame] | 108 | |
Guido van Rossum | 3572d37 | 1997-08-14 19:51:26 +0000 | [diff] [blame] | 109 | \item |
Fred Drake | d678cb7 | 1998-04-03 06:35:54 +0000 | [diff] [blame] | 110 | It is not possible to interrupt the \method{acquire()} method on a lock |
| 111 | --- the \exception{KeyboardInterrupt} exception will happen after the |
| 112 | lock has been acquired. |
Guido van Rossum | 3572d37 | 1997-08-14 19:51:26 +0000 | [diff] [blame] | 113 | |
| 114 | \item |
| 115 | When the main thread exits, it is system defined whether the other |
| 116 | threads survive. On SGI IRIX using the native thread implementation, |
| 117 | they survive. On most other systems, they are killed without |
Fred Drake | d678cb7 | 1998-04-03 06:35:54 +0000 | [diff] [blame] | 118 | executing \keyword{try} ... \keyword{finally} clauses or executing |
| 119 | object destructors. |
Fred Drake | 61b0452 | 1998-01-20 05:52:23 +0000 | [diff] [blame] | 120 | \indexii{threads}{IRIX} |
Guido van Rossum | 3572d37 | 1997-08-14 19:51:26 +0000 | [diff] [blame] | 121 | |
| 122 | \item |
Fred Drake | d678cb7 | 1998-04-03 06:35:54 +0000 | [diff] [blame] | 123 | When the main thread exits, it does not do any of its usual cleanup |
| 124 | (except that \keyword{try} ... \keyword{finally} clauses are honored), |
| 125 | and the standard I/O files are not flushed. |
Guido van Rossum | 3572d37 | 1997-08-14 19:51:26 +0000 | [diff] [blame] | 126 | |
Guido van Rossum | 5fdeeea | 1994-01-02 01:22:07 +0000 | [diff] [blame] | 127 | \end{itemize} |