———
This module provides access to the select() and poll() functions available in most operating systems, devpoll() available on Solaris and derivatives, epoll() available on Linux 2.5+ and kqueue() available on most BSD. Note that on Windows, it only works for sockets; on other operating systems, it also works for other file types (in particular, on Unix, it works on pipes). It cannot be used on regular files to determine whether a file has grown since it was last read.
Note
The
module allows high-level and efficient I/O multiplexing, built upon the select module primitives. Users are encouraged to use the selectors module instead, unless they want precise control over the OS-level primitives used.
The module defines the following:
exceptionselect.error
A deprecated alias of
.
Changed in version 3.3: Following
, this class was made an alias of
.
select.devpoll()
Returns a /dev/poll polling object; see section
below for the methods supported by devpoll objects.
devpoll() objects are linked to the number of file descriptors allowed at the time of instantiation. If your program reduces this value, devpoll() will fail. If your program increases this value, devpoll() may return an incomplete list of active file descriptors.
The new file descriptor is
.
Added in version 3.3.
Changed in version 3.4: The new file descriptor is now non-inheritable.
select.epoll(sizehint=-1, flags=0)
Return an edge polling object, which can be used as Edge or Level Triggered interface for I/O events.
sizehint informs epoll about the expected number of events to be registered. It must be positive, or -1 to use the default. It is only used on older systems where
is not available; otherwise it has no effect (though its value is still checked).
flags is deprecated and completely ignored. However, when supplied, its value must be 0 or select.EPOLL_CLOEXEC, otherwise OSError is raised.
See the
Edge and level trigger polling (epoll) objects
section below for the methods supported by epolling objects.
epoll objects support the context management protocol: when used in a
statement, the new file descriptor is automatically closed at the end of the block.
The new file descriptor is
.
Changed in version 3.3: Added the flags parameter.
Changed in version 3.4: Support for the
statement was added. The new file descriptor is now non-inheritable.
Deprecated since version 3.4: The flags parameter. select.EPOLL_CLOEXEC is used by default now. Use
to make the file descriptor inheritable.
Changed in version 3.15: When CPython is built, this function may be disabled using
.
select.poll()
Returns a polling object, which supports registering and unregistering file descriptors, and then polling them for I/O events; see section
below for the methods supported by polling objects.
select.kqueue()
Returns a kernel queue object; see section
below for the methods supported by kqueue objects.
The new file descriptor is
.
Changed in version 3.4: The new file descriptor is now non-inheritable.
select.kevent(ident, filter=KQ_FILTER_READ, flags=KQ_EV_ADD, fflags=0, data=0, udata=0)
Returns a kernel event object; see section
below for the methods supported by kevent objects.
select.select(rlist, wlist, xlist, timeout=None)
This is a straightforward interface to the Unix select() system call. The first three arguments are iterables of ‘waitable objects’: either integers representing file descriptors or objects with a parameterless method named
returning such an integer:
rlist: wait until ready for reading
wlist: wait until ready for writing
xlist: wait for an “exceptional condition” (see the manual page for what your system considers such a condition)
Empty iterables are allowed, but acceptance of three empty iterables is platform-dependent. (It is known to work on Unix but not on Windows.) The optional timeout argument specifies a time-out in seconds; it may be a non-integer to specify fractions of seconds. When the timeout argument is omitted or None, the function blocks until at least one file descriptor is ready. A time-out value of zero specifies a poll and never blocks.
The return value is a triple of lists of objects that are ready: subsets of the first three arguments. When the time-out is reached without a file descriptor becoming ready, three empty lists are returned.
Among the acceptable object types in the iterables are Python
(e.g. sys.stdin, or objects returned by
or
), socket objects returned by
. You may also define a wrapper class yourself, as long as it has an appropriate
method (that really returns a file descriptor, not just a random integer).
Note
File objects on Windows are not acceptable, but sockets are. On Windows, the underlying select() function is provided by the WinSock library, and does not handle file descriptors that don’t originate from WinSock.
Changed in version 3.5: The function is now retried with a recomputed timeout when interrupted by a signal, except if the signal handler raises an exception (see
for the rationale), instead of raising
.
Changed in version 3.15: Accepts any real number as timeout, not only integer or float.
select.PIPE_BUF
The minimum number of bytes which can be written without blocking to a pipe when the pipe has been reported as ready for writing by
, poll() or another interface in this module. This doesn’t apply to other kinds of file-like objects such as sockets.
This value is guaranteed by POSIX to be at least 512.
Added in version 3.2.
/dev/poll polling objects
Solaris and derivatives have /dev/poll. While select() is O(highest file descriptor) and poll() is O(number of file descriptors), /dev/poll is O(active file descriptors).
/dev/poll behaviour is very close to the standard poll() object.
devpoll.close()
Close the file descriptor of the polling object.
Added in version 3.4.
devpoll.closed
True if the polling object is closed.
Added in version 3.4.
devpoll.fileno()
Return the file descriptor number of the polling object.
Added in version 3.4.
devpoll.register(fd[, eventmask])
Register a file descriptor with the polling object. Future calls to the
method will then check whether the file descriptor has any pending I/O events. fd can be either an integer, or an object with a
method that returns an integer. File objects implement fileno(), so they can also be used as the argument.
eventmask is an optional bitmask describing the type of events you want to check for. The constants are the same as with poll() object. The default value is a combination of the constants POLLIN, POLLPRI, and POLLOUT.
Warning
Registering a file descriptor that’s already registered is not an error, but the result is undefined. The appropriate action is to unregister or modify it first. This is an important difference compared with poll().
devpoll.modify(fd[, eventmask])
This method does an
followed by a
. It is (a bit) more efficient than doing the same explicitly.
devpoll.unregister(fd)
Remove a file descriptor being tracked by a polling object. Just like the
method, fd can be an integer or an object with a
method that returns an integer.
Attempting to remove a file descriptor that was never registered is safely ignored.
devpoll.poll([timeout])
Polls the set of registered file descriptors, and returns a possibly empty list containing (fd,event) 2-tuples for the descriptors that have events or errors to report. fd is the file descriptor, and event is a bitmask with bits set for the reported events for that descriptor — POLLIN for waiting input, POLLOUT to indicate that the descriptor can be written to, and so forth. An empty list indicates that the call timed out and no file descriptors had any events to report. If timeout is given, it specifies the length of time in milliseconds which the system will wait for events before returning. If timeout is omitted, -1, or
, the call will block until there is an event for this poll object.
Changed in version 3.5: The function is now retried with a recomputed timeout when interrupted by a signal, except if the signal handler raises an exception (see
for the rationale), instead of raising
.
Changed in version 3.15: Accepts any real number as timeout, not only integer or float.
Edge and level trigger polling (epoll) objects
https://linux.die.net/man/4/epoll
The eventmask is a bit mask using the following constants:
Constant
Meaning
EPOLLIN
Available for read.
EPOLLOUT
Available for write.
EPOLLPRI
Urgent data for read.
EPOLLERR
Error condition happened on the associated fd.
EPOLLHUP
Hang up happened on the associated fd.
EPOLLET
Set Edge Trigger behavior, the default is Level Trigger behavior.
EPOLLONESHOT
Set one-shot behavior. After one event is pulled out, the fd is internally disabled.
EPOLLEXCLUSIVE
Wake only one epoll object when the associated fd has an event. The default (if this flag is not set) is to wake all epoll objects polling on an fd.
EPOLLRDHUP
Stream socket peer closed connection or shut down writing half of connection.
EPOLLRDNORM
Equivalent to EPOLLIN
EPOLLRDBAND
Priority data band can be read.
EPOLLWRNORM
Equivalent to EPOLLOUT.
EPOLLWRBAND
Priority data may be written.
EPOLLMSG
Ignored.
EPOLLWAKEUP
Prevents sleep during event waiting.
Added in version 3.6: EPOLLEXCLUSIVE was added. It’s only supported by Linux Kernel 4.5 or later.
Added in version 3.14: EPOLLWAKEUP was added. It’s only supported by Linux Kernel 3.5 or later.
epoll.close()
Close the control file descriptor of the epoll object.
epoll.closed
True if the epoll object is closed.
epoll.fileno()
Return the file descriptor number of the control fd.
epoll.fromfd(fd)
Create an epoll object from a given file descriptor.
epoll.register(fd[, eventmask])
Register a file descriptor fd with the epoll object.
epoll.modify(fd, eventmask)
Modify a registered file descriptor fd.
epoll.unregister(fd)
Remove a registered file descriptor from the epoll object.
Changed in version 3.9: The method no longer ignores the
error.
epoll.poll(timeout=None, maxevents=-1)
Wait for events. If timeout is given, it specifies the length of time in seconds (may be non-integer) which the system will wait for events before returning.
Changed in version 3.5: The function is now retried with a recomputed timeout when interrupted by a signal, except if the signal handler raises an exception (see
for the rationale), instead of raising
.
Changed in version 3.15: Accepts any real number as timeout, not only integer or float.
Polling objects
The poll() system call, supported on most Unix systems, provides better scalability for network servers that service many, many clients at the same time. poll() scales better because the system call only requires listing the file descriptors of interest, while select() builds a bitmap, turns on bits for the fds of interest, and then afterward the whole bitmap has to be linearly scanned again. select() is O(highest file descriptor), while poll() is O(number of file descriptors).
poll.register(fd[, eventmask])
Register a file descriptor with the polling object. Future calls to the
method will then check whether the file descriptor has any pending I/O events. fd can be either an integer, or an object with a
method that returns an integer. File objects implement fileno(), so they can also be used as the argument.
eventmask is an optional bitmask describing the type of events you want to check for, and can be a combination of the constants POLLIN, POLLPRI, and POLLOUT, described in the table below. If not specified, the default value used will check for all 3 types of events.
Constant
Meaning
POLLIN
There is data to read.
POLLPRI
There is urgent data to read.
POLLOUT
Ready for output: writing will not block.
POLLERR
Error condition of some sort.
POLLHUP
Hung up.
POLLRDHUP
Stream socket peer closed connection, or shut down writing half of connection.
POLLNVAL
Invalid request: descriptor not open.
Registering a file descriptor that’s already registered is not an error, and has the same effect as registering the descriptor exactly once.
poll.modify(fd, eventmask)
Modifies an already registered fd. This has the same effect as register(fd,eventmask). Attempting to modify a file descriptor that was never registered causes an
exception with errno ENOENT to be raised.
poll.unregister(fd)
Remove a file descriptor being tracked by a polling object. Just like the
method, fd can be an integer or an object with a
method that returns an integer.
Attempting to remove a file descriptor that was never registered causes a
exception to be raised.
poll.poll([timeout])
Polls the set of registered file descriptors, and returns a possibly empty list containing (fd,event) 2-tuples for the descriptors that have events or errors to report. fd is the file descriptor, and event is a bitmask with bits set for the reported events for that descriptor — POLLIN for waiting input, POLLOUT to indicate that the descriptor can be written to, and so forth. An empty list indicates that the call timed out and no file descriptors had any events to report. If timeout is given, it specifies the length of time in milliseconds which the system will wait for events before returning. If timeout is omitted, negative, or
, the call will block until there is an event for this poll object.
Changed in version 3.5: The function is now retried with a recomputed timeout when interrupted by a signal, except if the signal handler raises an exception (see
for the rationale), instead of raising
.
Changed in version 3.15: Accepts any real number as timeout, not only integer or float. If ppoll() function is available, timeout has a resolution of 1 ns (1e-6 ms) instead of 1 ms.
Kqueue objects
kqueue.close()
Close the control file descriptor of the kqueue object.
kqueue.closed
True if the kqueue object is closed.
kqueue.fileno()
Return the file descriptor number of the control fd.
kqueue.fromfd(fd)
Create a kqueue object from a given file descriptor.
kqueue.control(changelist, max_events[, timeout])→eventlist
Low level interface to kevent
changelist must be an iterable of kevent objects or None
max_events must be 0 or a positive integer
timeout in seconds (non-integers are possible); the default is None, to wait forever
Changed in version 3.5: The function is now retried with a recomputed timeout when interrupted by a signal, except if the signal handler raises an exception (see
for the rationale), instead of raising
.
Changed in version 3.15: Accepts any real number as timeout, not only integer or float.
Kevent objects
https://man.freebsd.org/cgi/man.cgi?query=kqueue&sektion=2
kevent.ident
Value used to identify the event. The interpretation depends on the filter but it’s usually the file descriptor. In the constructor ident can either be an int or an object with a
method. kevent stores the integer internally.
kevent.filter
Name of the kernel filter.
Constant
Meaning
KQ_FILTER_READ
Takes a descriptor and returns whenever there is data available to read.
KQ_FILTER_WRITE
Takes a descriptor and returns whenever there is data available to write.
KQ_FILTER_AIO
AIO requests.
KQ_FILTER_VNODE
Returns when one or more of the requested events watched in fflag occurs.
KQ_FILTER_PROC
Watch for events on a process ID.
KQ_FILTER_NETDEV
Watch for events on a network device (not available on macOS).
KQ_FILTER_SIGNAL
Returns whenever the watched signal is delivered to the process.
KQ_FILTER_TIMER
Establishes an arbitrary timer.
kevent.flags
Filter action.
Constant
Meaning
KQ_EV_ADD
Adds or modifies an event.
KQ_EV_DELETE
Removes an event from the queue.
KQ_EV_ENABLE
Permits control() to return the event.
KQ_EV_DISABLE
Disables event.
KQ_EV_ONESHOT
Removes event after first occurrence.
KQ_EV_CLEAR
Reset the state after an event is retrieved.
KQ_EV_SYSFLAGS
Internal event.
KQ_EV_FLAG1
Internal event.
KQ_EV_EOF
Filter-specific EOF condition.
KQ_EV_ERROR
See return values.
kevent.fflags
Filter-specific flags.
KQ_FILTER_READ and KQ_FILTER_WRITE filter flags:
Constant
Meaning
KQ_NOTE_LOWAT
Low water mark of a socket buffer.
KQ_FILTER_VNODE filter flags:
Constant
Meaning
KQ_NOTE_DELETE
unlink() was called.
KQ_NOTE_WRITE
A write occurred.
KQ_NOTE_EXTEND
The file was extended.
KQ_NOTE_ATTRIB
An attribute was changed.
KQ_NOTE_LINK
The link count has changed.
KQ_NOTE_RENAME
The file was renamed.
KQ_NOTE_REVOKE
Access to the file was revoked.
KQ_FILTER_PROC filter flags:
Constant
Meaning
KQ_NOTE_EXIT
The process has exited.
KQ_NOTE_FORK
The process has called fork().
KQ_NOTE_EXEC
The process has executed a new process.
KQ_NOTE_PCTRLMASK
Internal filter flag.
KQ_NOTE_PDATAMASK
Internal filter flag.
KQ_NOTE_TRACK
Follow a process across fork().
KQ_NOTE_CHILD
Returned on the child process for NOTE_TRACK.
KQ_NOTE_TRACKERR
Unable to attach to a child.
KQ_FILTER_NETDEV filter flags (not available on macOS):
Constant
Meaning
KQ_NOTE_LINKUP
Link is up.
KQ_NOTE_LINKDOWN
Link is down.
KQ_NOTE_LINKINV
Link state is invalid.
kevent.data
Filter-specific data.
kevent.udata
User-defined value.