PageRenderTime 50ms CodeModel.GetById 16ms RepoModel.GetById 0ms app.codeStats 0ms

/Doc/library/socket.rst

https://bitbucket.org/python_mirrors/sandbox-regex-integration
ReStructuredText | 1375 lines | 968 code | 407 blank | 0 comment | 0 complexity | d2d54dd0a4d81ceced8def068e22aeaa MD5 | raw file
Possible License(s): BSD-3-Clause, 0BSD

Large files files are truncated, but you can click here to view the full file

  1. :mod:`socket` --- Low-level networking interface
  2. ================================================
  3. .. module:: socket
  4. :synopsis: Low-level networking interface.
  5. This module provides access to the BSD *socket* interface. It is available on
  6. all modern Unix systems, Windows, MacOS, OS/2, and probably additional
  7. platforms.
  8. .. note::
  9. Some behavior may be platform dependent, since calls are made to the operating
  10. system socket APIs.
  11. .. index:: object: socket
  12. The Python interface is a straightforward transliteration of the Unix system
  13. call and library interface for sockets to Python's object-oriented style: the
  14. :func:`socket` function returns a :dfn:`socket object` whose methods implement
  15. the various socket system calls. Parameter types are somewhat higher-level than
  16. in the C interface: as with :meth:`read` and :meth:`write` operations on Python
  17. files, buffer allocation on receive operations is automatic, and buffer length
  18. is implicit on send operations.
  19. .. seealso::
  20. Module :mod:`socketserver`
  21. Classes that simplify writing network servers.
  22. Module :mod:`ssl`
  23. A TLS/SSL wrapper for socket objects.
  24. Socket families
  25. ---------------
  26. Depending on the system and the build options, various socket families
  27. are supported by this module.
  28. The address format required by a particular socket object is automatically
  29. selected based on the address family specified when the socket object was
  30. created. Socket addresses are represented as follows:
  31. - The address of an :const:`AF_UNIX` socket bound to a file system node
  32. is represented as a string, using the file system encoding and the
  33. ``'surrogateescape'`` error handler (see :pep:`383`). An address in
  34. Linux's abstract namespace is returned as a :class:`bytes` object with
  35. an initial null byte; note that sockets in this namespace can
  36. communicate with normal file system sockets, so programs intended to
  37. run on Linux may need to deal with both types of address. A string or
  38. :class:`bytes` object can be used for either type of address when
  39. passing it as an argument.
  40. .. versionchanged:: 3.3
  41. Previously, :const:`AF_UNIX` socket paths were assumed to use UTF-8
  42. encoding.
  43. - A pair ``(host, port)`` is used for the :const:`AF_INET` address family,
  44. where *host* is a string representing either a hostname in Internet domain
  45. notation like ``'daring.cwi.nl'`` or an IPv4 address like ``'100.50.200.5'``,
  46. and *port* is an integral port number.
  47. - For :const:`AF_INET6` address family, a four-tuple ``(host, port, flowinfo,
  48. scopeid)`` is used, where *flowinfo* and *scopeid* represent the ``sin6_flowinfo``
  49. and ``sin6_scope_id`` members in :const:`struct sockaddr_in6` in C. For
  50. :mod:`socket` module methods, *flowinfo* and *scopeid* can be omitted just for
  51. backward compatibility. Note, however, omission of *scopeid* can cause problems
  52. in manipulating scoped IPv6 addresses.
  53. - :const:`AF_NETLINK` sockets are represented as pairs ``(pid, groups)``.
  54. - Linux-only support for TIPC is available using the :const:`AF_TIPC`
  55. address family. TIPC is an open, non-IP based networked protocol designed
  56. for use in clustered computer environments. Addresses are represented by a
  57. tuple, and the fields depend on the address type. The general tuple form is
  58. ``(addr_type, v1, v2, v3 [, scope])``, where:
  59. - *addr_type* is one of :const:`TIPC_ADDR_NAMESEQ`, :const:`TIPC_ADDR_NAME`,
  60. or :const:`TIPC_ADDR_ID`.
  61. - *scope* is one of :const:`TIPC_ZONE_SCOPE`, :const:`TIPC_CLUSTER_SCOPE`, and
  62. :const:`TIPC_NODE_SCOPE`.
  63. - If *addr_type* is :const:`TIPC_ADDR_NAME`, then *v1* is the server type, *v2* is
  64. the port identifier, and *v3* should be 0.
  65. If *addr_type* is :const:`TIPC_ADDR_NAMESEQ`, then *v1* is the server type, *v2*
  66. is the lower port number, and *v3* is the upper port number.
  67. If *addr_type* is :const:`TIPC_ADDR_ID`, then *v1* is the node, *v2* is the
  68. reference, and *v3* should be set to 0.
  69. If *addr_type* is :const:`TIPC_ADDR_ID`, then *v1* is the node, *v2* is the
  70. reference, and *v3* should be set to 0.
  71. - A tuple ``(interface, )`` is used for the :const:`AF_CAN` address family,
  72. where *interface* is a string representing a network interface name like
  73. ``'can0'``. The network interface name ``''`` can be used to receive packets
  74. from all network interfaces of this family.
  75. - Certain other address families (:const:`AF_BLUETOOTH`, :const:`AF_PACKET`)
  76. support specific representations.
  77. .. XXX document them!
  78. For IPv4 addresses, two special forms are accepted instead of a host address:
  79. the empty string represents :const:`INADDR_ANY`, and the string
  80. ``'<broadcast>'`` represents :const:`INADDR_BROADCAST`. This behavior is not
  81. compatible with IPv6, therefore, you may want to avoid these if you intend
  82. to support IPv6 with your Python programs.
  83. If you use a hostname in the *host* portion of IPv4/v6 socket address, the
  84. program may show a nondeterministic behavior, as Python uses the first address
  85. returned from the DNS resolution. The socket address will be resolved
  86. differently into an actual IPv4/v6 address, depending on the results from DNS
  87. resolution and/or the host configuration. For deterministic behavior use a
  88. numeric address in *host* portion.
  89. All errors raise exceptions. The normal exceptions for invalid argument types
  90. and out-of-memory conditions can be raised; starting from Python 3.3, errors
  91. related to socket or address semantics raise :exc:`OSError` or one of its
  92. subclasses (they used to raise :exc:`socket.error`).
  93. Non-blocking mode is supported through :meth:`~socket.setblocking`. A
  94. generalization of this based on timeouts is supported through
  95. :meth:`~socket.settimeout`.
  96. Module contents
  97. ---------------
  98. The module :mod:`socket` exports the following constants and functions:
  99. .. exception:: error
  100. A deprecated alias of :exc:`OSError`.
  101. .. versionchanged:: 3.3
  102. Following :pep:`3151`, this class was made an alias of :exc:`OSError`.
  103. .. exception:: herror
  104. A subclass of :exc:`OSError`, this exception is raised for
  105. address-related errors, i.e. for functions that use *h_errno* in the POSIX
  106. C API, including :func:`gethostbyname_ex` and :func:`gethostbyaddr`.
  107. The accompanying value is a pair ``(h_errno, string)`` representing an
  108. error returned by a library call. *h_errno* is a numeric value, while
  109. *string* represents the description of *h_errno*, as returned by the
  110. :c:func:`hstrerror` C function.
  111. .. versionchanged:: 3.3
  112. This class was made a subclass of :exc:`OSError`.
  113. .. exception:: gaierror
  114. A subclass of :exc:`OSError`, this exception is raised for
  115. address-related errors by :func:`getaddrinfo` and :func:`getnameinfo`.
  116. The accompanying value is a pair ``(error, string)`` representing an error
  117. returned by a library call. *string* represents the description of
  118. *error*, as returned by the :c:func:`gai_strerror` C function. The
  119. numeric *error* value will match one of the :const:`EAI_\*` constants
  120. defined in this module.
  121. .. versionchanged:: 3.3
  122. This class was made a subclass of :exc:`OSError`.
  123. .. exception:: timeout
  124. A subclass of :exc:`OSError`, this exception is raised when a timeout
  125. occurs on a socket which has had timeouts enabled via a prior call to
  126. :meth:`~socket.settimeout` (or implicitly through
  127. :func:`~socket.setdefaulttimeout`). The accompanying value is a string
  128. whose value is currently always "timed out".
  129. .. versionchanged:: 3.3
  130. This class was made a subclass of :exc:`OSError`.
  131. .. data:: AF_UNIX
  132. AF_INET
  133. AF_INET6
  134. These constants represent the address (and protocol) families, used for the
  135. first argument to :func:`socket`. If the :const:`AF_UNIX` constant is not
  136. defined then this protocol is unsupported. More constants may be available
  137. depending on the system.
  138. .. data:: SOCK_STREAM
  139. SOCK_DGRAM
  140. SOCK_RAW
  141. SOCK_RDM
  142. SOCK_SEQPACKET
  143. These constants represent the socket types, used for the second argument to
  144. :func:`socket`. More constants may be available depending on the system.
  145. (Only :const:`SOCK_STREAM` and :const:`SOCK_DGRAM` appear to be generally
  146. useful.)
  147. .. data:: SOCK_CLOEXEC
  148. SOCK_NONBLOCK
  149. These two constants, if defined, can be combined with the socket types and
  150. allow you to set some flags atomically (thus avoiding possible race
  151. conditions and the need for separate calls).
  152. .. seealso::
  153. `Secure File Descriptor Handling <http://udrepper.livejournal.com/20407.html>`_
  154. for a more thorough explanation.
  155. Availability: Linux >= 2.6.27.
  156. .. versionadded:: 3.2
  157. .. data:: SO_*
  158. SOMAXCONN
  159. MSG_*
  160. SOL_*
  161. SCM_*
  162. IPPROTO_*
  163. IPPORT_*
  164. INADDR_*
  165. IP_*
  166. IPV6_*
  167. EAI_*
  168. AI_*
  169. NI_*
  170. TCP_*
  171. Many constants of these forms, documented in the Unix documentation on sockets
  172. and/or the IP protocol, are also defined in the socket module. They are
  173. generally used in arguments to the :meth:`setsockopt` and :meth:`getsockopt`
  174. methods of socket objects. In most cases, only those symbols that are defined
  175. in the Unix header files are defined; for a few symbols, default values are
  176. provided.
  177. .. data:: AF_CAN
  178. PF_CAN
  179. SOL_CAN_*
  180. CAN_*
  181. Many constants of these forms, documented in the Linux documentation, are
  182. also defined in the socket module.
  183. Availability: Linux >= 2.6.25.
  184. .. versionadded:: 3.3
  185. .. data:: AF_RDS
  186. PF_RDS
  187. SOL_RDS
  188. RDS_*
  189. Many constants of these forms, documented in the Linux documentation, are
  190. also defined in the socket module.
  191. Availability: Linux >= 2.6.30.
  192. .. versionadded:: 3.3
  193. .. data:: SIO_*
  194. RCVALL_*
  195. Constants for Windows' WSAIoctl(). The constants are used as arguments to the
  196. :meth:`ioctl` method of socket objects.
  197. .. data:: TIPC_*
  198. TIPC related constants, matching the ones exported by the C socket API. See
  199. the TIPC documentation for more information.
  200. .. data:: has_ipv6
  201. This constant contains a boolean value which indicates if IPv6 is supported on
  202. this platform.
  203. .. function:: create_connection(address[, timeout[, source_address]])
  204. Connect to a TCP service listening on the Internet *address* (a 2-tuple
  205. ``(host, port)``), and return the socket object. This is a higher-level
  206. function than :meth:`socket.connect`: if *host* is a non-numeric hostname,
  207. it will try to resolve it for both :data:`AF_INET` and :data:`AF_INET6`,
  208. and then try to connect to all possible addresses in turn until a
  209. connection succeeds. This makes it easy to write clients that are
  210. compatible to both IPv4 and IPv6.
  211. Passing the optional *timeout* parameter will set the timeout on the
  212. socket instance before attempting to connect. If no *timeout* is
  213. supplied, the global default timeout setting returned by
  214. :func:`getdefaulttimeout` is used.
  215. If supplied, *source_address* must be a 2-tuple ``(host, port)`` for the
  216. socket to bind to as its source address before connecting. If host or port
  217. are '' or 0 respectively the OS default behavior will be used.
  218. .. versionchanged:: 3.2
  219. *source_address* was added.
  220. .. versionchanged:: 3.2
  221. support for the :keyword:`with` statement was added.
  222. .. function:: getaddrinfo(host, port, family=0, type=0, proto=0, flags=0)
  223. Translate the *host*/*port* argument into a sequence of 5-tuples that contain
  224. all the necessary arguments for creating a socket connected to that service.
  225. *host* is a domain name, a string representation of an IPv4/v6 address
  226. or ``None``. *port* is a string service name such as ``'http'``, a numeric
  227. port number or ``None``. By passing ``None`` as the value of *host*
  228. and *port*, you can pass ``NULL`` to the underlying C API.
  229. The *family*, *type* and *proto* arguments can be optionally specified
  230. in order to narrow the list of addresses returned. Passing zero as a
  231. value for each of these arguments selects the full range of results.
  232. The *flags* argument can be one or several of the ``AI_*`` constants,
  233. and will influence how results are computed and returned.
  234. For example, :const:`AI_NUMERICHOST` will disable domain name resolution
  235. and will raise an error if *host* is a domain name.
  236. The function returns a list of 5-tuples with the following structure:
  237. ``(family, type, proto, canonname, sockaddr)``
  238. In these tuples, *family*, *type*, *proto* are all integers and are
  239. meant to be passed to the :func:`socket` function. *canonname* will be
  240. a string representing the canonical name of the *host* if
  241. :const:`AI_CANONNAME` is part of the *flags* argument; else *canonname*
  242. will be empty. *sockaddr* is a tuple describing a socket address, whose
  243. format depends on the returned *family* (a ``(address, port)`` 2-tuple for
  244. :const:`AF_INET`, a ``(address, port, flow info, scope id)`` 4-tuple for
  245. :const:`AF_INET6`), and is meant to be passed to the :meth:`socket.connect`
  246. method.
  247. The following example fetches address information for a hypothetical TCP
  248. connection to ``www.python.org`` on port 80 (results may differ on your
  249. system if IPv6 isn't enabled)::
  250. >>> socket.getaddrinfo("www.python.org", 80, proto=socket.SOL_TCP)
  251. [(2, 1, 6, '', ('82.94.164.162', 80)),
  252. (10, 1, 6, '', ('2001:888:2000:d::a2', 80, 0, 0))]
  253. .. versionchanged:: 3.2
  254. parameters can now be passed as single keyword arguments.
  255. .. function:: getfqdn([name])
  256. Return a fully qualified domain name for *name*. If *name* is omitted or empty,
  257. it is interpreted as the local host. To find the fully qualified name, the
  258. hostname returned by :func:`gethostbyaddr` is checked, followed by aliases for the
  259. host, if available. The first name which includes a period is selected. In
  260. case no fully qualified domain name is available, the hostname as returned by
  261. :func:`gethostname` is returned.
  262. .. function:: gethostbyname(hostname)
  263. Translate a host name to IPv4 address format. The IPv4 address is returned as a
  264. string, such as ``'100.50.200.5'``. If the host name is an IPv4 address itself
  265. it is returned unchanged. See :func:`gethostbyname_ex` for a more complete
  266. interface. :func:`gethostbyname` does not support IPv6 name resolution, and
  267. :func:`getaddrinfo` should be used instead for IPv4/v6 dual stack support.
  268. .. function:: gethostbyname_ex(hostname)
  269. Translate a host name to IPv4 address format, extended interface. Return a
  270. triple ``(hostname, aliaslist, ipaddrlist)`` where *hostname* is the primary
  271. host name responding to the given *ip_address*, *aliaslist* is a (possibly
  272. empty) list of alternative host names for the same address, and *ipaddrlist* is
  273. a list of IPv4 addresses for the same interface on the same host (often but not
  274. always a single address). :func:`gethostbyname_ex` does not support IPv6 name
  275. resolution, and :func:`getaddrinfo` should be used instead for IPv4/v6 dual
  276. stack support.
  277. .. function:: gethostname()
  278. Return a string containing the hostname of the machine where the Python
  279. interpreter is currently executing.
  280. If you want to know the current machine's IP address, you may want to use
  281. ``gethostbyname(gethostname())``. This operation assumes that there is a
  282. valid address-to-host mapping for the host, and the assumption does not
  283. always hold.
  284. Note: :func:`gethostname` doesn't always return the fully qualified domain
  285. name; use ``getfqdn()`` (see above).
  286. .. function:: gethostbyaddr(ip_address)
  287. Return a triple ``(hostname, aliaslist, ipaddrlist)`` where *hostname* is the
  288. primary host name responding to the given *ip_address*, *aliaslist* is a
  289. (possibly empty) list of alternative host names for the same address, and
  290. *ipaddrlist* is a list of IPv4/v6 addresses for the same interface on the same
  291. host (most likely containing only a single address). To find the fully qualified
  292. domain name, use the function :func:`getfqdn`. :func:`gethostbyaddr` supports
  293. both IPv4 and IPv6.
  294. .. function:: getnameinfo(sockaddr, flags)
  295. Translate a socket address *sockaddr* into a 2-tuple ``(host, port)``. Depending
  296. on the settings of *flags*, the result can contain a fully-qualified domain name
  297. or numeric address representation in *host*. Similarly, *port* can contain a
  298. string port name or a numeric port number.
  299. .. function:: getprotobyname(protocolname)
  300. Translate an Internet protocol name (for example, ``'icmp'``) to a constant
  301. suitable for passing as the (optional) third argument to the :func:`socket`
  302. function. This is usually only needed for sockets opened in "raw" mode
  303. (:const:`SOCK_RAW`); for the normal socket modes, the correct protocol is chosen
  304. automatically if the protocol is omitted or zero.
  305. .. function:: getservbyname(servicename[, protocolname])
  306. Translate an Internet service name and protocol name to a port number for that
  307. service. The optional protocol name, if given, should be ``'tcp'`` or
  308. ``'udp'``, otherwise any protocol will match.
  309. .. function:: getservbyport(port[, protocolname])
  310. Translate an Internet port number and protocol name to a service name for that
  311. service. The optional protocol name, if given, should be ``'tcp'`` or
  312. ``'udp'``, otherwise any protocol will match.
  313. .. function:: socket([family[, type[, proto]]])
  314. Create a new socket using the given address family, socket type and protocol
  315. number. The address family should be :const:`AF_INET` (the default),
  316. :const:`AF_INET6`, :const:`AF_UNIX`, :const:`AF_CAN` or :const:`AF_RDS`. The
  317. socket type should be :const:`SOCK_STREAM` (the default),
  318. :const:`SOCK_DGRAM`, :const:`SOCK_RAW` or perhaps one of the other ``SOCK_``
  319. constants. The protocol number is usually zero and may be omitted in that
  320. case or :const:`CAN_RAW` in case the address family is :const:`AF_CAN`.
  321. .. versionchanged:: 3.3
  322. The AF_CAN family was added.
  323. The AF_RDS family was added.
  324. .. function:: socketpair([family[, type[, proto]]])
  325. Build a pair of connected socket objects using the given address family, socket
  326. type, and protocol number. Address family, socket type, and protocol number are
  327. as for the :func:`socket` function above. The default family is :const:`AF_UNIX`
  328. if defined on the platform; otherwise, the default is :const:`AF_INET`.
  329. Availability: Unix.
  330. .. versionchanged:: 3.2
  331. The returned socket objects now support the whole socket API, rather
  332. than a subset.
  333. .. function:: fromfd(fd, family, type[, proto])
  334. Duplicate the file descriptor *fd* (an integer as returned by a file object's
  335. :meth:`fileno` method) and build a socket object from the result. Address
  336. family, socket type and protocol number are as for the :func:`socket` function
  337. above. The file descriptor should refer to a socket, but this is not checked ---
  338. subsequent operations on the object may fail if the file descriptor is invalid.
  339. This function is rarely needed, but can be used to get or set socket options on
  340. a socket passed to a program as standard input or output (such as a server
  341. started by the Unix inet daemon). The socket is assumed to be in blocking mode.
  342. .. function:: ntohl(x)
  343. Convert 32-bit positive integers from network to host byte order. On machines
  344. where the host byte order is the same as network byte order, this is a no-op;
  345. otherwise, it performs a 4-byte swap operation.
  346. .. function:: ntohs(x)
  347. Convert 16-bit positive integers from network to host byte order. On machines
  348. where the host byte order is the same as network byte order, this is a no-op;
  349. otherwise, it performs a 2-byte swap operation.
  350. .. function:: htonl(x)
  351. Convert 32-bit positive integers from host to network byte order. On machines
  352. where the host byte order is the same as network byte order, this is a no-op;
  353. otherwise, it performs a 4-byte swap operation.
  354. .. function:: htons(x)
  355. Convert 16-bit positive integers from host to network byte order. On machines
  356. where the host byte order is the same as network byte order, this is a no-op;
  357. otherwise, it performs a 2-byte swap operation.
  358. .. function:: inet_aton(ip_string)
  359. Convert an IPv4 address from dotted-quad string format (for example,
  360. '123.45.67.89') to 32-bit packed binary format, as a bytes object four characters in
  361. length. This is useful when conversing with a program that uses the standard C
  362. library and needs objects of type :c:type:`struct in_addr`, which is the C type
  363. for the 32-bit packed binary this function returns.
  364. :func:`inet_aton` also accepts strings with less than three dots; see the
  365. Unix manual page :manpage:`inet(3)` for details.
  366. If the IPv4 address string passed to this function is invalid,
  367. :exc:`OSError` will be raised. Note that exactly what is valid depends on
  368. the underlying C implementation of :c:func:`inet_aton`.
  369. :func:`inet_aton` does not support IPv6, and :func:`inet_pton` should be used
  370. instead for IPv4/v6 dual stack support.
  371. .. function:: inet_ntoa(packed_ip)
  372. Convert a 32-bit packed IPv4 address (a bytes object four characters in
  373. length) to its standard dotted-quad string representation (for example,
  374. '123.45.67.89'). This is useful when conversing with a program that uses the
  375. standard C library and needs objects of type :c:type:`struct in_addr`, which
  376. is the C type for the 32-bit packed binary data this function takes as an
  377. argument.
  378. If the byte sequence passed to this function is not exactly 4 bytes in
  379. length, :exc:`OSError` will be raised. :func:`inet_ntoa` does not
  380. support IPv6, and :func:`inet_ntop` should be used instead for IPv4/v6 dual
  381. stack support.
  382. .. function:: inet_pton(address_family, ip_string)
  383. Convert an IP address from its family-specific string format to a packed,
  384. binary format. :func:`inet_pton` is useful when a library or network protocol
  385. calls for an object of type :c:type:`struct in_addr` (similar to
  386. :func:`inet_aton`) or :c:type:`struct in6_addr`.
  387. Supported values for *address_family* are currently :const:`AF_INET` and
  388. :const:`AF_INET6`. If the IP address string *ip_string* is invalid,
  389. :exc:`OSError` will be raised. Note that exactly what is valid depends on
  390. both the value of *address_family* and the underlying implementation of
  391. :c:func:`inet_pton`.
  392. Availability: Unix (maybe not all platforms).
  393. .. function:: inet_ntop(address_family, packed_ip)
  394. Convert a packed IP address (a bytes object of some number of characters) to its
  395. standard, family-specific string representation (for example, ``'7.10.0.5'`` or
  396. ``'5aef:2b::8'``). :func:`inet_ntop` is useful when a library or network protocol
  397. returns an object of type :c:type:`struct in_addr` (similar to :func:`inet_ntoa`)
  398. or :c:type:`struct in6_addr`.
  399. Supported values for *address_family* are currently :const:`AF_INET` and
  400. :const:`AF_INET6`. If the string *packed_ip* is not the correct length for the
  401. specified address family, :exc:`ValueError` will be raised. A
  402. :exc:`OSError` is raised for errors from the call to :func:`inet_ntop`.
  403. Availability: Unix (maybe not all platforms).
  404. ..
  405. XXX: Are sendmsg(), recvmsg() and CMSG_*() available on any
  406. non-Unix platforms? The old (obsolete?) 4.2BSD form of the
  407. interface, in which struct msghdr has no msg_control or
  408. msg_controllen members, is not currently supported.
  409. .. function:: CMSG_LEN(length)
  410. Return the total length, without trailing padding, of an ancillary
  411. data item with associated data of the given *length*. This value
  412. can often be used as the buffer size for :meth:`~socket.recvmsg` to
  413. receive a single item of ancillary data, but :rfc:`3542` requires
  414. portable applications to use :func:`CMSG_SPACE` and thus include
  415. space for padding, even when the item will be the last in the
  416. buffer. Raises :exc:`OverflowError` if *length* is outside the
  417. permissible range of values.
  418. Availability: most Unix platforms, possibly others.
  419. .. versionadded:: 3.3
  420. .. function:: CMSG_SPACE(length)
  421. Return the buffer size needed for :meth:`~socket.recvmsg` to
  422. receive an ancillary data item with associated data of the given
  423. *length*, along with any trailing padding. The buffer space needed
  424. to receive multiple items is the sum of the :func:`CMSG_SPACE`
  425. values for their associated data lengths. Raises
  426. :exc:`OverflowError` if *length* is outside the permissible range
  427. of values.
  428. Note that some systems might support ancillary data without
  429. providing this function. Also note that setting the buffer size
  430. using the results of this function may not precisely limit the
  431. amount of ancillary data that can be received, since additional
  432. data may be able to fit into the padding area.
  433. Availability: most Unix platforms, possibly others.
  434. .. versionadded:: 3.3
  435. .. function:: getdefaulttimeout()
  436. Return the default timeout in seconds (float) for new socket objects. A value
  437. of ``None`` indicates that new socket objects have no timeout. When the socket
  438. module is first imported, the default is ``None``.
  439. .. function:: setdefaulttimeout(timeout)
  440. Set the default timeout in seconds (float) for new socket objects. When
  441. the socket module is first imported, the default is ``None``. See
  442. :meth:`~socket.settimeout` for possible values and their respective
  443. meanings.
  444. .. function:: sethostname(name)
  445. Set the machine's hostname to *name*. This will raise a
  446. :exc:`OSError` if you don't have enough rights.
  447. Availability: Unix.
  448. .. versionadded:: 3.3
  449. .. function:: if_nameindex()
  450. Return a list of network interface information
  451. (index int, name string) tuples.
  452. :exc:`OSError` if the system call fails.
  453. Availability: Unix.
  454. .. versionadded:: 3.3
  455. .. function:: if_nametoindex(if_name)
  456. Return a network interface index number corresponding to an
  457. interface name.
  458. :exc:`OSError` if no interface with the given name exists.
  459. Availability: Unix.
  460. .. versionadded:: 3.3
  461. .. function:: if_indextoname(if_index)
  462. Return a network interface name corresponding to a
  463. interface index number.
  464. :exc:`OSError` if no interface with the given index exists.
  465. Availability: Unix.
  466. .. versionadded:: 3.3
  467. .. data:: SocketType
  468. This is a Python type object that represents the socket object type. It is the
  469. same as ``type(socket(...))``.
  470. .. _socket-objects:
  471. Socket Objects
  472. --------------
  473. Socket objects have the following methods. Except for :meth:`makefile` these
  474. correspond to Unix system calls applicable to sockets.
  475. .. method:: socket.accept()
  476. Accept a connection. The socket must be bound to an address and listening for
  477. connections. The return value is a pair ``(conn, address)`` where *conn* is a
  478. *new* socket object usable to send and receive data on the connection, and
  479. *address* is the address bound to the socket on the other end of the connection.
  480. .. method:: socket.bind(address)
  481. Bind the socket to *address*. The socket must not already be bound. (The format
  482. of *address* depends on the address family --- see above.)
  483. .. method:: socket.close()
  484. Close the socket. All future operations on the socket object will fail. The
  485. remote end will receive no more data (after queued data is flushed). Sockets are
  486. automatically closed when they are garbage-collected.
  487. .. note::
  488. :meth:`close()` releases the resource associated with a connection but
  489. does not necessarily close the connection immediately. If you want
  490. to close the connection in a timely fashion, call :meth:`shutdown()`
  491. before :meth:`close()`.
  492. .. method:: socket.connect(address)
  493. Connect to a remote socket at *address*. (The format of *address* depends on the
  494. address family --- see above.)
  495. .. method:: socket.connect_ex(address)
  496. Like ``connect(address)``, but return an error indicator instead of raising an
  497. exception for errors returned by the C-level :c:func:`connect` call (other
  498. problems, such as "host not found," can still raise exceptions). The error
  499. indicator is ``0`` if the operation succeeded, otherwise the value of the
  500. :c:data:`errno` variable. This is useful to support, for example, asynchronous
  501. connects.
  502. .. method:: socket.detach()
  503. Put the socket object into closed state without actually closing the
  504. underlying file descriptor. The file descriptor is returned, and can
  505. be reused for other purposes.
  506. .. versionadded:: 3.2
  507. .. method:: socket.fileno()
  508. Return the socket's file descriptor (a small integer). This is useful with
  509. :func:`select.select`.
  510. Under Windows the small integer returned by this method cannot be used where a
  511. file descriptor can be used (such as :func:`os.fdopen`). Unix does not have
  512. this limitation.
  513. .. method:: socket.getpeername()
  514. Return the remote address to which the socket is connected. This is useful to
  515. find out the port number of a remote IPv4/v6 socket, for instance. (The format
  516. of the address returned depends on the address family --- see above.) On some
  517. systems this function is not supported.
  518. .. method:: socket.getsockname()
  519. Return the socket's own address. This is useful to find out the port number of
  520. an IPv4/v6 socket, for instance. (The format of the address returned depends on
  521. the address family --- see above.)
  522. .. method:: socket.getsockopt(level, optname[, buflen])
  523. Return the value of the given socket option (see the Unix man page
  524. :manpage:`getsockopt(2)`). The needed symbolic constants (:const:`SO_\*` etc.)
  525. are defined in this module. If *buflen* is absent, an integer option is assumed
  526. and its integer value is returned by the function. If *buflen* is present, it
  527. specifies the maximum length of the buffer used to receive the option in, and
  528. this buffer is returned as a bytes object. It is up to the caller to decode the
  529. contents of the buffer (see the optional built-in module :mod:`struct` for a way
  530. to decode C structures encoded as byte strings).
  531. .. method:: socket.gettimeout()
  532. Return the timeout in seconds (float) associated with socket operations,
  533. or ``None`` if no timeout is set. This reflects the last call to
  534. :meth:`setblocking` or :meth:`settimeout`.
  535. .. method:: socket.ioctl(control, option)
  536. :platform: Windows
  537. The :meth:`ioctl` method is a limited interface to the WSAIoctl system
  538. interface. Please refer to the `Win32 documentation
  539. <http://msdn.microsoft.com/en-us/library/ms741621%28VS.85%29.aspx>`_ for more
  540. information.
  541. On other platforms, the generic :func:`fcntl.fcntl` and :func:`fcntl.ioctl`
  542. functions may be used; they accept a socket object as their first argument.
  543. .. method:: socket.listen(backlog)
  544. Listen for connections made to the socket. The *backlog* argument specifies the
  545. maximum number of queued connections and should be at least 0; the maximum value
  546. is system-dependent (usually 5), the minimum value is forced to 0.
  547. .. method:: socket.makefile(mode='r', buffering=None, *, encoding=None, \
  548. errors=None, newline=None)
  549. .. index:: single: I/O control; buffering
  550. Return a :term:`file object` associated with the socket. The exact returned
  551. type depends on the arguments given to :meth:`makefile`. These arguments are
  552. interpreted the same way as by the built-in :func:`open` function.
  553. Closing the file object won't close the socket unless there are no remaining
  554. references to the socket. The socket must be in blocking mode; it can have
  555. a timeout, but the file object's internal buffer may end up in a inconsistent
  556. state if a timeout occurs.
  557. .. note::
  558. On Windows, the file-like object created by :meth:`makefile` cannot be
  559. used where a file object with a file descriptor is expected, such as the
  560. stream arguments of :meth:`subprocess.Popen`.
  561. .. method:: socket.recv(bufsize[, flags])
  562. Receive data from the socket. The return value is a bytes object representing the
  563. data received. The maximum amount of data to be received at once is specified
  564. by *bufsize*. See the Unix manual page :manpage:`recv(2)` for the meaning of
  565. the optional argument *flags*; it defaults to zero.
  566. .. note::
  567. For best match with hardware and network realities, the value of *bufsize*
  568. should be a relatively small power of 2, for example, 4096.
  569. .. method:: socket.recvfrom(bufsize[, flags])
  570. Receive data from the socket. The return value is a pair ``(bytes, address)``
  571. where *bytes* is a bytes object representing the data received and *address* is the
  572. address of the socket sending the data. See the Unix manual page
  573. :manpage:`recv(2)` for the meaning of the optional argument *flags*; it defaults
  574. to zero. (The format of *address* depends on the address family --- see above.)
  575. .. method:: socket.recvmsg(bufsize[, ancbufsize[, flags]])
  576. Receive normal data (up to *bufsize* bytes) and ancillary data from
  577. the socket. The *ancbufsize* argument sets the size in bytes of
  578. the internal buffer used to receive the ancillary data; it defaults
  579. to 0, meaning that no ancillary data will be received. Appropriate
  580. buffer sizes for ancillary data can be calculated using
  581. :func:`CMSG_SPACE` or :func:`CMSG_LEN`, and items which do not fit
  582. into the buffer might be truncated or discarded. The *flags*
  583. argument defaults to 0 and has the same meaning as for
  584. :meth:`recv`.
  585. The return value is a 4-tuple: ``(data, ancdata, msg_flags,
  586. address)``. The *data* item is a :class:`bytes` object holding the
  587. non-ancillary data received. The *ancdata* item is a list of zero
  588. or more tuples ``(cmsg_level, cmsg_type, cmsg_data)`` representing
  589. the ancillary data (control messages) received: *cmsg_level* and
  590. *cmsg_type* are integers specifying the protocol level and
  591. protocol-specific type respectively, and *cmsg_data* is a
  592. :class:`bytes` object holding the associated data. The *msg_flags*
  593. item is the bitwise OR of various flags indicating conditions on
  594. the received message; see your system documentation for details.
  595. If the receiving socket is unconnected, *address* is the address of
  596. the sending socket, if available; otherwise, its value is
  597. unspecified.
  598. On some systems, :meth:`sendmsg` and :meth:`recvmsg` can be used to
  599. pass file descriptors between processes over an :const:`AF_UNIX`
  600. socket. When this facility is used (it is often restricted to
  601. :const:`SOCK_STREAM` sockets), :meth:`recvmsg` will return, in its
  602. ancillary data, items of the form ``(socket.SOL_SOCKET,
  603. socket.SCM_RIGHTS, fds)``, where *fds* is a :class:`bytes` object
  604. representing the new file descriptors as a binary array of the
  605. native C :c:type:`int` type. If :meth:`recvmsg` raises an
  606. exception after the system call returns, it will first attempt to
  607. close any file descriptors received via this mechanism.
  608. Some systems do not indicate the truncated length of ancillary data
  609. items which have been only partially received. If an item appears
  610. to extend beyond the end of the buffer, :meth:`recvmsg` will issue
  611. a :exc:`RuntimeWarning`, and will return the part of it which is
  612. inside the buffer provided it has not been truncated before the
  613. start of its associated data.
  614. On systems which support the :const:`SCM_RIGHTS` mechanism, the
  615. following function will receive up to *maxfds* file descriptors,
  616. returning the message data and a list containing the descriptors
  617. (while ignoring unexpected conditions such as unrelated control
  618. messages being received). See also :meth:`sendmsg`. ::
  619. import socket, array
  620. def recv_fds(sock, msglen, maxfds):
  621. fds = array.array("i") # Array of ints
  622. msg, ancdata, flags, addr = sock.recvmsg(msglen, socket.CMSG_LEN(maxfds * fds.itemsize))
  623. for cmsg_level, cmsg_type, cmsg_data in ancdata:
  624. if (cmsg_level == socket.SOL_SOCKET and cmsg_type == socket.SCM_RIGHTS):
  625. # Append data, ignoring any truncated integers at the end.
  626. fds.fromstring(cmsg_data[:len(cmsg_data) - (len(cmsg_data) % fds.itemsize)])
  627. return msg, list(fds)
  628. Availability: most Unix platforms, possibly others.
  629. .. versionadded:: 3.3
  630. .. method:: socket.recvmsg_into(buffers[, ancbufsize[, flags]])
  631. Receive normal data and ancillary data from the socket, behaving as
  632. :meth:`recvmsg` would, but scatter the non-ancillary data into a
  633. series of buffers instead of returning a new bytes object. The
  634. *buffers* argument must be an iterable of objects that export
  635. writable buffers (e.g. :class:`bytearray` objects); these will be
  636. filled with successive chunks of the non-ancillary data until it
  637. has all been written or there are no more buffers. The operating
  638. system may set a limit (:func:`~os.sysconf` value ``SC_IOV_MAX``)
  639. on the number of buffers that can be used. The *ancbufsize* and
  640. *flags* arguments have the same meaning as for :meth:`recvmsg`.
  641. The return value is a 4-tuple: ``(nbytes, ancdata, msg_flags,
  642. address)``, where *nbytes* is the total number of bytes of
  643. non-ancillary data written into the buffers, and *ancdata*,
  644. *msg_flags* and *address* are the same as for :meth:`recvmsg`.
  645. Example::
  646. >>> import socket
  647. >>> s1, s2 = socket.socketpair()
  648. >>> b1 = bytearray(b'----')
  649. >>> b2 = bytearray(b'0123456789')
  650. >>> b3 = bytearray(b'--------------')
  651. >>> s1.send(b'Mary had a little lamb')
  652. 22
  653. >>> s2.recvmsg_into([b1, memoryview(b2)[2:9], b3])
  654. (22, [], 0, None)
  655. >>> [b1, b2, b3]
  656. [bytearray(b'Mary'), bytearray(b'01 had a 9'), bytearray(b'little lamb---')]
  657. Availability: most Unix platforms, possibly others.
  658. .. versionadded:: 3.3
  659. .. method:: socket.recvfrom_into(buffer[, nbytes[, flags]])
  660. Receive data from the socket, writing it into *buffer* instead of creating a
  661. new bytestring. The return value is a pair ``(nbytes, address)`` where *nbytes* is
  662. the number of bytes received and *address* is the address of the socket sending
  663. the data. See the Unix manual page :manpage:`recv(2)` for the meaning of the
  664. optional argument *flags*; it defaults to zero. (The format of *address*
  665. depends on the address family --- see above.)
  666. .. method:: socket.recv_into(buffer[, nbytes[, flags]])
  667. Receive up to *nbytes* bytes from the socket, storing the data into a buffer
  668. rather than creating a new bytestring. If *nbytes* is not specified (or 0),
  669. receive up to the size available in the given buffer. Returns the number of
  670. bytes received. See the Unix manual page :manpage:`recv(2)` for the meaning
  671. of the optional argument *flags*; it defaults to zero.
  672. .. method:: socket.send(bytes[, flags])
  673. Send data to the socket. The socket must be connected to a remote socket. The
  674. optional *flags* argument has the same meaning as for :meth:`recv` above.
  675. Returns the number of bytes sent. Applications are responsible for checking that
  676. all data has been sent; if only some of the data was transmitted, the
  677. application needs to attempt delivery of the remaining data.
  678. .. method:: socket.sendall(bytes[, flags])
  679. Send data to the socket. The socket must be connected to a remote socket. The
  680. optional *flags* argument has the same meaning as for :meth:`recv` above.
  681. Unlike :meth:`send`, this method continues to send data from *bytes* until
  682. either all data has been sent or an error occurs. ``None`` is returned on
  683. success. On error, an exception is raised, and there is no way to determine how
  684. much data, if any, was successfully sent.
  685. .. method:: socket.sendto(bytes[, flags], address)
  686. Send data to the socket. The socket should not be connected to a remote socket,
  687. since the destination socket is specified by *address*. The optional *flags*
  688. argument has the same meaning as for :meth:`recv` above. Return the number of
  689. bytes sent. (The format of *address* depends on the address family --- see
  690. above.)
  691. .. method:: socket.sendmsg(buffers[, ancdata[, flags[, address]]])
  692. Send normal and ancillary data to the socket, gathering the
  693. non-ancillary data from a series of buffers and concatenating it
  694. into a single message. The *buffers* argument specifies the
  695. non-ancillary data as an iterable of buffer-compatible objects
  696. (e.g. :class:`bytes` objects); the operating system may set a limit
  697. (:func:`~os.sysconf` value ``SC_IOV_MAX``) on the number of buffers
  698. that can be used. The *ancdata* argument specifies the ancillary
  699. data (control messages) as an iterable of zero or more tuples
  700. ``(cmsg_level, cmsg_type, cmsg_data)``, where *cmsg_level* and
  701. *cmsg_type* are integers specifying the protocol level and
  702. protocol-specific type respectively, and *cmsg_data* is a
  703. buffer-compatible object holding the associated data. Note that
  704. some systems (in particular, systems without :func:`CMSG_SPACE`)
  705. might support sending only one control message per call. The
  706. *flags* argument defaults to 0 and has the same meaning as for
  707. :meth:`send`. If *address* is supplied and not ``None``, it sets a
  708. destination address for the message. The return value is the
  709. number of bytes of non-ancillary data sent.
  710. The following function sends the list of file descriptors *fds*
  711. over an :const:`AF_UNIX` socket, on systems which support the
  712. :const:`SCM_RIGHTS` mechanism. See also :meth:`recvmsg`. ::
  713. import socket, array
  714. def send_fds(sock, msg, fds):
  715. return sock.sendmsg([msg], [(socket.SOL_SOCKET, socket.SCM_RIGHTS, array.array("i", fds))])
  716. Availability: most Unix platforms, possibly others.
  717. .. versionadded:: 3.3
  718. .. method:: socket.setblocking(flag)
  719. Set blocking or non-blocking mode of the socket: if *flag* is false, the
  720. socket is set to non-blocking, else to blocking mode.
  721. This method is a shorthand for certain :meth:`~socket.settimeout` calls:
  722. * ``sock.setblocking(True)`` is equivalent to ``sock.settimeout(None)``
  723. * ``sock.setblocking(False)`` is equivalent to ``sock.settimeout(0.0)``
  724. .. method:: socket.settimeout(value)
  725. Set a timeout on blocking socket operations. The *value* argument can be a
  726. nonnegative floating point number expressing seconds, or ``None``.
  727. If a non-zero value is given, subsequent socket operations will raise a
  728. :exc:`timeout` exception if the timeout period *value* has elapsed before
  729. the operation has completed. If zero is given, the socket is put in
  730. non-blocking mode. If ``None`` is given, the socket is put in blocking mode.
  731. For further information, please consult the :ref:`notes on socket timeouts <socket-timeouts>`.
  732. .. method:: socket.setsockopt(level, optname, value)
  733. .. index:: module: struct
  734. Set the value of the given socket option (see the Unix manual page
  735. :manpage:`setsockopt(2)`). The needed symbolic constants are defined in the
  736. :mod:`socket` module (:const:`SO_\*` etc.). The value can be an integer or a
  737. bytes object representing a buffer. In the latter case it is up to the caller to
  738. ensure that the bytestring contains the proper bits (see the optional built-in
  739. module :mod:`struct` for a way to encode C structures as bytestrings).
  740. .. method:: socket.shutdown(how)
  741. Shut down one or both halves of the connection. If *how* is :const:`SHUT_RD`,
  742. further receives are disallowed. If *how* is :const:`SHUT_WR`, further sends
  743. are disallowed. If *how* is :const:`SHUT_RDWR`, further sends and receives are
  744. disallowed. Depending on the platform, shutting down one half of the connection
  745. can also close the opposite half (e.g. on Mac OS X, ``shutdown(SHUT_WR)`` does
  746. not allow further reads on the other end of the connection).
  747. Note that there are no methods :meth:`read` or :meth:`write`; use
  748. :meth:`~socket.recv` and :meth:`~socket.send` without *flags* argument instead.
  749. Socket objects also have these (read-only) attributes that correspond to the
  750. values given to the :class:`socket` constructor.
  751. .. attribute:: socket.family
  752. The socket family.
  753. .. attribute:: socket.type
  754. The socket type.
  755. .. attribute:: socket.proto
  756. The socket protocol.
  757. .. _socket-timeouts:
  758. Notes on socket timeouts
  759. ------------------------
  760. A socket object can be in one of three modes: blocking, non-blocking, or
  761. timeout. Sockets are by default always created in blocking mode, but this
  762. can be changed by calling :func:`setdefaulttimeout`.
  763. * In *blocking mode*, operations block until complete or the system returns
  764. an error (such as connection timed out).
  765. * In *non-blocking mode*, operations fail (with an error that is unfortunately
  766. system-dependent) if they cannot be completed immediately: functions from the
  767. :mod:`select` can be used to know when and whether a socket is available for
  768. reading or writing.
  769. * In *timeout mode*, operations fail if they cannot be completed within the
  770. timeout specified for the socket (they raise a :exc:`timeout` exception)
  771. or if the system returns an error.
  772. .. note::
  773. At the operating system level, sockets in *timeout mode* are internally set
  774. in non-blocking mode. Also, the blocking and timeout modes are shared between
  775. file descriptors and socket objects that refer to the same network endpoint.
  776. This implementation detail can have visible consequences if e.g. you decide
  777. to use the :meth:`~socket.fileno()` of a socket.
  778. Timeouts and the ``connect`` method
  779. ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  780. The :meth:`~socket.connect` operation is also subject to the timeout
  781. setting, and in general it is recommended to call :meth:`~socket.settimeout`
  782. before calling :meth:`~socket.connect` or pass a timeout parameter to
  783. :meth:`create_connection`. However, the system network stack may also
  784. return a connection timeout error of its own regardless of any Python socket
  785. timeout setting.
  786. Timeouts and the ``accept`` method
  787. ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  788. If :func:`getdefaulttimeout` is not :const:`None`, sockets returned by
  789. the :meth:`~socket.accept` method inherit that timeout. Otherwise, the
  790. behaviour depends on settings of the listening socket:
  791. * if the listening socket is in *blocking mode* or in *timeout mode*,
  792. the socket returned by :meth:`~socket.accept` is in *blocking mode*;
  793. * if the listening socket is in *non-blocking mode*, whether the socket
  794. returned by :meth:`~socket.accept` is in blocking or non-blocking mode
  795. is operating system-dependent. If you want to ensure cross-platform
  796. behaviour, it is recommended you manually override this setting.
  797. .. _socket-example:
  798. Example
  799. -------
  800. Here are four minimal example programs using the TCP/IP protocol: a server that
  801. echoes all data that it receives back (servicing only one client), and a client
  802. using it. Note that a server must perform the sequence :func:`socket`,
  803. :meth:`~socket.bind`, :meth:`~socket.listen`, :meth:`~socket.accept` (possibly
  804. repeating the :meth:`~socket.accept` to service more than one client), while a
  805. client only needs the sequence :func:`socket`, :meth:`~socket.connect`. Also
  806. note that the server does not :meth:`~socket.send`/:meth:`~socket.recv` on the
  807. socket it is listening on but on the new socket returned by
  808. :meth:`~socket.accept`.
  809. The first two examples support IPv4 only. ::
  810. # Echo server program
  811. import socket
  812. HOST = '' # Symbolic name meaning all available interfaces
  813. PORT = 50007 # Arbitrary non-privileged port
  814. s = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
  815. s.bind((HOST, PORT))
  816. s.listen(1)
  817. conn, addr = s.accept()
  818. print('Connected by', addr)
  819. while True:
  820. data = conn.recv(1024)
  821. if not data: break
  822. conn.send(data)
  823. conn.close()
  824. ::
  825. # Echo client program
  826. import socket
  827. HOST = 'daring.cwi.nl' # The remote host
  828. PORT = 50007 # The same port as used by the server
  829. s = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
  830. s.connect((HOST, PORT))
  831. s.send(b'Hello, world')
  832. data = s.recv(1024)
  833. s.close()
  834. print('Received', repr(data))
  835. The next two examples are identical to the above two, but support both IPv4 and
  836. IPv6. The server side will listen to the first address family available (it
  837. should listen to both instead). On most of IPv6-ready systems, IPv6 will take
  838. precedence and the server may not accept IPv4 traffic. The client side will try
  839. to connect to the all addresses returned as a result of the name resolution, and
  840. sends traffic to the first one connected successfully. ::
  841. # Echo server program
  842. import socket
  843. import sys
  844. HOST = None # Symbolic name meaning all available interfaces
  845. PORT = 50007 # Arbitrary non-privileged port
  846. s = None
  847. for res in socket.getaddrinfo(HOST, PORT, socket.AF_UNSPEC,
  848. socket.SOCK_STREAM, 0, socket.AI_PASSIVE):
  849. af, socktype, proto, canonname, sa = res
  850. try:
  851. s = socket.socket(af, socktype, proto)
  852. except OSError as msg:
  853. s = None
  854. continue
  855. try:
  856. s.bind(sa)
  857. s.listen(1)
  858. except OSError as msg:
  859. s.close()
  860. s = None
  861. continue
  862. break
  863. if s is None:
  864. print('could not open socket')
  865. sys.exit(1)
  866. conn, addr = s.accept()
  867. print('Connected by', addr)
  868. while True:
  869. data = conn.recv(1024)
  870. if not data: break
  871. conn.send(data)
  872. conn.close()
  873. ::
  874. # Echo client program
  875. import socket
  876. import sys
  877. HOST = 'daring.cwi.nl' # The remote host
  878. PORT = 50007 # The same port as used …

Large files files are truncated, but you can click here to view the full file