Mercurial > repos > shellac > guppy_basecaller

comparison env/lib/python3.7/site-packages/future/backports/socketserver.py @ 2:6af9afd405e9 draft

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
"planemo upload commit 0a63dd5f4d38a1f6944587f52a8cd79874177fc1"
author shellac
date Thu, 14 May 2020 14:56:58 -0400
parents 26e78fe6e8c4
children
comparison
equal deleted inserted replaced
1:75ca89e9b81c 2:6af9afd405e9
1 """Generic socket server classes.
2
3 This module tries to capture the various aspects of defining a server:
4
5 For socket-based servers:
6
7 - address family:
8 - AF_INET{,6}: IP (Internet Protocol) sockets (default)
9 - AF_UNIX: Unix domain sockets
10 - others, e.g. AF_DECNET are conceivable (see <socket.h>
11 - socket type:
12 - SOCK_STREAM (reliable stream, e.g. TCP)
13 - SOCK_DGRAM (datagrams, e.g. UDP)
14
15 For request-based servers (including socket-based):
16
17 - client address verification before further looking at the request
18 (This is actually a hook for any processing that needs to look
19 at the request before anything else, e.g. logging)
20 - how to handle multiple requests:
21 - synchronous (one request is handled at a time)
22 - forking (each request is handled by a new process)
23 - threading (each request is handled by a new thread)
24
25 The classes in this module favor the server type that is simplest to
26 write: a synchronous TCP/IP server. This is bad class design, but
27 save some typing. (There's also the issue that a deep class hierarchy
28 slows down method lookups.)
29
30 There are five classes in an inheritance diagram, four of which represent
31 synchronous servers of four types:
32
33 +------------+
34 | BaseServer |
35 +------------+
36 |
37 v
38 +-----------+ +------------------+
39 | TCPServer |------->| UnixStreamServer |
40 +-----------+ +------------------+
41 |
42 v
43 +-----------+ +--------------------+
44 | UDPServer |------->| UnixDatagramServer |
45 +-----------+ +--------------------+
46
47 Note that UnixDatagramServer derives from UDPServer, not from
48 UnixStreamServer -- the only difference between an IP and a Unix
49 stream server is the address family, which is simply repeated in both
50 unix server classes.
51
52 Forking and threading versions of each type of server can be created
53 using the ForkingMixIn and ThreadingMixIn mix-in classes. For
54 instance, a threading UDP server class is created as follows:
55
56 class ThreadingUDPServer(ThreadingMixIn, UDPServer): pass
57
58 The Mix-in class must come first, since it overrides a method defined
59 in UDPServer! Setting the various member variables also changes
60 the behavior of the underlying server mechanism.
61
62 To implement a service, you must derive a class from
63 BaseRequestHandler and redefine its handle() method. You can then run
64 various versions of the service by combining one of the server classes
65 with your request handler class.
66
67 The request handler class must be different for datagram or stream
68 services. This can be hidden by using the request handler
69 subclasses StreamRequestHandler or DatagramRequestHandler.
70
71 Of course, you still have to use your head!
72
73 For instance, it makes no sense to use a forking server if the service
74 contains state in memory that can be modified by requests (since the
75 modifications in the child process would never reach the initial state
76 kept in the parent process and passed to each child). In this case,
77 you can use a threading server, but you will probably have to use
78 locks to avoid two requests that come in nearly simultaneous to apply
79 conflicting changes to the server state.
80
81 On the other hand, if you are building e.g. an HTTP server, where all
82 data is stored externally (e.g. in the file system), a synchronous
83 class will essentially render the service "deaf" while one request is
84 being handled -- which may be for a very long time if a client is slow
85 to read all the data it has requested. Here a threading or forking
86 server is appropriate.
87
88 In some cases, it may be appropriate to process part of a request
89 synchronously, but to finish processing in a forked child depending on
90 the request data. This can be implemented by using a synchronous
91 server and doing an explicit fork in the request handler class
92 handle() method.
93
94 Another approach to handling multiple simultaneous requests in an
95 environment that supports neither threads nor fork (or where these are
96 too expensive or inappropriate for the service) is to maintain an
97 explicit table of partially finished requests and to use select() to
98 decide which request to work on next (or whether to handle a new
99 incoming request). This is particularly important for stream services
100 where each client can potentially be connected for a long time (if
101 threads or subprocesses cannot be used).
102
103 Future work:
104 - Standard classes for Sun RPC (which uses either UDP or TCP)
105 - Standard mix-in classes to implement various authentication
106 and encryption schemes
107 - Standard framework for select-based multiplexing
108
109 XXX Open problems:
110 - What to do with out-of-band data?
111
112 BaseServer:
113 - split generic "request" functionality out into BaseServer class.
114 Copyright (C) 2000 Luke Kenneth Casson Leighton <lkcl@samba.org>
115
116 example: read entries from a SQL database (requires overriding
117 get_request() to return a table entry from the database).
118 entry is processed by a RequestHandlerClass.
119
120 """
121
122 # Author of the BaseServer patch: Luke Kenneth Casson Leighton
123
124 # XXX Warning!
125 # There is a test suite for this module, but it cannot be run by the
126 # standard regression test.
127 # To run it manually, run Lib/test/test_socketserver.py.
128
129 from __future__ import (absolute_import, print_function)
130
131 __version__ = "0.4"
132
133
134 import socket
135 import select
136 import sys
137 import os
138 import errno
139 try:
140 import threading
141 except ImportError:
142 import dummy_threading as threading
143
144 __all__ = ["TCPServer","UDPServer","ForkingUDPServer","ForkingTCPServer",
145 "ThreadingUDPServer","ThreadingTCPServer","BaseRequestHandler",
146 "StreamRequestHandler","DatagramRequestHandler",
147 "ThreadingMixIn", "ForkingMixIn"]
148 if hasattr(socket, "AF_UNIX"):
149 __all__.extend(["UnixStreamServer","UnixDatagramServer",
150 "ThreadingUnixStreamServer",
151 "ThreadingUnixDatagramServer"])
152
153 def _eintr_retry(func, *args):
154 """restart a system call interrupted by EINTR"""
155 while True:
156 try:
157 return func(*args)
158 except OSError as e:
159 if e.errno != errno.EINTR:
160 raise
161
162 class BaseServer(object):
163
164 """Base class for server classes.
165
166 Methods for the caller:
167
168 - __init__(server_address, RequestHandlerClass)
169 - serve_forever(poll_interval=0.5)
170 - shutdown()
171 - handle_request() # if you do not use serve_forever()
172 - fileno() -> int # for select()
173
174 Methods that may be overridden:
175
176 - server_bind()
177 - server_activate()
178 - get_request() -> request, client_address
179 - handle_timeout()
180 - verify_request(request, client_address)
181 - server_close()
182 - process_request(request, client_address)
183 - shutdown_request(request)
184 - close_request(request)
185 - service_actions()
186 - handle_error()
187
188 Methods for derived classes:
189
190 - finish_request(request, client_address)
191
192 Class variables that may be overridden by derived classes or
193 instances:
194
195 - timeout
196 - address_family
197 - socket_type
198 - allow_reuse_address
199
200 Instance variables:
201
202 - RequestHandlerClass
203 - socket
204
205 """
206
207 timeout = None
208
209 def __init__(self, server_address, RequestHandlerClass):
210 """Constructor. May be extended, do not override."""
211 self.server_address = server_address
212 self.RequestHandlerClass = RequestHandlerClass
213 self.__is_shut_down = threading.Event()
214 self.__shutdown_request = False
215
216 def server_activate(self):
217 """Called by constructor to activate the server.
218
219 May be overridden.
220
221 """
222 pass
223
224 def serve_forever(self, poll_interval=0.5):
225 """Handle one request at a time until shutdown.
226
227 Polls for shutdown every poll_interval seconds. Ignores
228 self.timeout. If you need to do periodic tasks, do them in
229 another thread.
230 """
231 self.__is_shut_down.clear()
232 try:
233 while not self.__shutdown_request:
234 # XXX: Consider using another file descriptor or
235 # connecting to the socket to wake this up instead of
236 # polling. Polling reduces our responsiveness to a
237 # shutdown request and wastes cpu at all other times.
238 r, w, e = _eintr_retry(select.select, [self], [], [],
239 poll_interval)
240 if self in r:
241 self._handle_request_noblock()
242
243 self.service_actions()
244 finally:
245 self.__shutdown_request = False
246 self.__is_shut_down.set()
247
248 def shutdown(self):
249 """Stops the serve_forever loop.
250
251 Blocks until the loop has finished. This must be called while
252 serve_forever() is running in another thread, or it will
253 deadlock.
254 """
255 self.__shutdown_request = True
256 self.__is_shut_down.wait()
257
258 def service_actions(self):
259 """Called by the serve_forever() loop.
260
261 May be overridden by a subclass / Mixin to implement any code that
262 needs to be run during the loop.
263 """
264 pass
265
266 # The distinction between handling, getting, processing and
267 # finishing a request is fairly arbitrary. Remember:
268 #
269 # - handle_request() is the top-level call. It calls
270 # select, get_request(), verify_request() and process_request()
271 # - get_request() is different for stream or datagram sockets
272 # - process_request() is the place that may fork a new process
273 # or create a new thread to finish the request
274 # - finish_request() instantiates the request handler class;
275 # this constructor will handle the request all by itself
276
277 def handle_request(self):
278 """Handle one request, possibly blocking.
279
280 Respects self.timeout.
281 """
282 # Support people who used socket.settimeout() to escape
283 # handle_request before self.timeout was available.
284 timeout = self.socket.gettimeout()
285 if timeout is None:
286 timeout = self.timeout
287 elif self.timeout is not None:
288 timeout = min(timeout, self.timeout)
289 fd_sets = _eintr_retry(select.select, [self], [], [], timeout)
290 if not fd_sets[0]:
291 self.handle_timeout()
292 return
293 self._handle_request_noblock()
294
295 def _handle_request_noblock(self):
296 """Handle one request, without blocking.
297
298 I assume that select.select has returned that the socket is
299 readable before this function was called, so there should be
300 no risk of blocking in get_request().
301 """
302 try:
303 request, client_address = self.get_request()
304 except socket.error:
305 return
306 if self.verify_request(request, client_address):
307 try:
308 self.process_request(request, client_address)
309 except:
310 self.handle_error(request, client_address)
311 self.shutdown_request(request)
312
313 def handle_timeout(self):
314 """Called if no new request arrives within self.timeout.
315
316 Overridden by ForkingMixIn.
317 """
318 pass
319
320 def verify_request(self, request, client_address):
321 """Verify the request. May be overridden.
322
323 Return True if we should proceed with this request.
324
325 """
326 return True
327
328 def process_request(self, request, client_address):
329 """Call finish_request.
330
331 Overridden by ForkingMixIn and ThreadingMixIn.
332
333 """
334 self.finish_request(request, client_address)
335 self.shutdown_request(request)
336
337 def server_close(self):
338 """Called to clean-up the server.
339
340 May be overridden.
341
342 """
343 pass
344
345 def finish_request(self, request, client_address):
346 """Finish one request by instantiating RequestHandlerClass."""
347 self.RequestHandlerClass(request, client_address, self)
348
349 def shutdown_request(self, request):
350 """Called to shutdown and close an individual request."""
351 self.close_request(request)
352
353 def close_request(self, request):
354 """Called to clean up an individual request."""
355 pass
356
357 def handle_error(self, request, client_address):
358 """Handle an error gracefully. May be overridden.
359
360 The default is to print a traceback and continue.
361
362 """
363 print('-'*40)
364 print('Exception happened during processing of request from', end=' ')
365 print(client_address)
366 import traceback
367 traceback.print_exc() # XXX But this goes to stderr!
368 print('-'*40)
369
370
371 class TCPServer(BaseServer):
372
373 """Base class for various socket-based server classes.
374
375 Defaults to synchronous IP stream (i.e., TCP).
376
377 Methods for the caller:
378
379 - __init__(server_address, RequestHandlerClass, bind_and_activate=True)
380 - serve_forever(poll_interval=0.5)
381 - shutdown()
382 - handle_request() # if you don't use serve_forever()
383 - fileno() -> int # for select()
384
385 Methods that may be overridden:
386
387 - server_bind()
388 - server_activate()
389 - get_request() -> request, client_address
390 - handle_timeout()
391 - verify_request(request, client_address)
392 - process_request(request, client_address)
393 - shutdown_request(request)
394 - close_request(request)
395 - handle_error()
396
397 Methods for derived classes:
398
399 - finish_request(request, client_address)
400
401 Class variables that may be overridden by derived classes or
402 instances:
403
404 - timeout
405 - address_family
406 - socket_type
407 - request_queue_size (only for stream sockets)
408 - allow_reuse_address
409
410 Instance variables:
411
412 - server_address
413 - RequestHandlerClass
414 - socket
415
416 """
417
418 address_family = socket.AF_INET
419
420 socket_type = socket.SOCK_STREAM
421
422 request_queue_size = 5
423
424 allow_reuse_address = False
425
426 def __init__(self, server_address, RequestHandlerClass, bind_and_activate=True):
427 """Constructor. May be extended, do not override."""
428 BaseServer.__init__(self, server_address, RequestHandlerClass)
429 self.socket = socket.socket(self.address_family,
430 self.socket_type)
431 if bind_and_activate:
432 self.server_bind()
433 self.server_activate()
434
435 def server_bind(self):
436 """Called by constructor to bind the socket.
437
438 May be overridden.
439
440 """
441 if self.allow_reuse_address:
442 self.socket.setsockopt(socket.SOL_SOCKET, socket.SO_REUSEADDR, 1)
443 self.socket.bind(self.server_address)
444 self.server_address = self.socket.getsockname()
445
446 def server_activate(self):
447 """Called by constructor to activate the server.
448
449 May be overridden.
450
451 """
452 self.socket.listen(self.request_queue_size)
453
454 def server_close(self):
455 """Called to clean-up the server.
456
457 May be overridden.
458
459 """
460 self.socket.close()
461
462 def fileno(self):
463 """Return socket file number.
464
465 Interface required by select().
466
467 """
468 return self.socket.fileno()
469
470 def get_request(self):
471 """Get the request and client address from the socket.
472
473 May be overridden.
474
475 """
476 return self.socket.accept()
477
478 def shutdown_request(self, request):
479 """Called to shutdown and close an individual request."""
480 try:
481 #explicitly shutdown. socket.close() merely releases
482 #the socket and waits for GC to perform the actual close.
483 request.shutdown(socket.SHUT_WR)
484 except socket.error:
485 pass #some platforms may raise ENOTCONN here
486 self.close_request(request)
487
488 def close_request(self, request):
489 """Called to clean up an individual request."""
490 request.close()
491
492
493 class UDPServer(TCPServer):
494
495 """UDP server class."""
496
497 allow_reuse_address = False
498
499 socket_type = socket.SOCK_DGRAM
500
501 max_packet_size = 8192
502
503 def get_request(self):
504 data, client_addr = self.socket.recvfrom(self.max_packet_size)
505 return (data, self.socket), client_addr
506
507 def server_activate(self):
508 # No need to call listen() for UDP.
509 pass
510
511 def shutdown_request(self, request):
512 # No need to shutdown anything.
513 self.close_request(request)
514
515 def close_request(self, request):
516 # No need to close anything.
517 pass
518
519 class ForkingMixIn(object):
520
521 """Mix-in class to handle each request in a new process."""
522
523 timeout = 300
524 active_children = None
525 max_children = 40
526
527 def collect_children(self):
528 """Internal routine to wait for children that have exited."""
529 if self.active_children is None: return
530 while len(self.active_children) >= self.max_children:
531 # XXX: This will wait for any child process, not just ones
532 # spawned by this library. This could confuse other
533 # libraries that expect to be able to wait for their own
534 # children.
535 try:
536 pid, status = os.waitpid(0, 0)
537 except os.error:
538 pid = None
539 if pid not in self.active_children: continue
540 self.active_children.remove(pid)
541
542 # XXX: This loop runs more system calls than it ought
543 # to. There should be a way to put the active_children into a
544 # process group and then use os.waitpid(-pgid) to wait for any
545 # of that set, but I couldn't find a way to allocate pgids
546 # that couldn't collide.
547 for child in self.active_children:
548 try:
549 pid, status = os.waitpid(child, os.WNOHANG)
550 except os.error:
551 pid = None
552 if not pid: continue
553 try:
554 self.active_children.remove(pid)
555 except ValueError as e:
556 raise ValueError('%s. x=%d and list=%r' % (e.message, pid,
557 self.active_children))
558
559 def handle_timeout(self):
560 """Wait for zombies after self.timeout seconds of inactivity.
561
562 May be extended, do not override.
563 """
564 self.collect_children()
565
566 def service_actions(self):
567 """Collect the zombie child processes regularly in the ForkingMixIn.
568
569 service_actions is called in the BaseServer's serve_forver loop.
570 """
571 self.collect_children()
572
573 def process_request(self, request, client_address):
574 """Fork a new subprocess to process the request."""
575 pid = os.fork()
576 if pid:
577 # Parent process
578 if self.active_children is None:
579 self.active_children = []
580 self.active_children.append(pid)
581 self.close_request(request)
582 return
583 else:
584 # Child process.
585 # This must never return, hence os._exit()!
586 try:
587 self.finish_request(request, client_address)
588 self.shutdown_request(request)
589 os._exit(0)
590 except:
591 try:
592 self.handle_error(request, client_address)
593 self.shutdown_request(request)
594 finally:
595 os._exit(1)
596
597
598 class ThreadingMixIn(object):
599 """Mix-in class to handle each request in a new thread."""
600
601 # Decides how threads will act upon termination of the
602 # main process
603 daemon_threads = False
604
605 def process_request_thread(self, request, client_address):
606 """Same as in BaseServer but as a thread.
607
608 In addition, exception handling is done here.
609
610 """
611 try:
612 self.finish_request(request, client_address)
613 self.shutdown_request(request)
614 except:
615 self.handle_error(request, client_address)
616 self.shutdown_request(request)
617
618 def process_request(self, request, client_address):
619 """Start a new thread to process the request."""
620 t = threading.Thread(target = self.process_request_thread,
621 args = (request, client_address))
622 t.daemon = self.daemon_threads
623 t.start()
624
625
626 class ForkingUDPServer(ForkingMixIn, UDPServer): pass
627 class ForkingTCPServer(ForkingMixIn, TCPServer): pass
628
629 class ThreadingUDPServer(ThreadingMixIn, UDPServer): pass
630 class ThreadingTCPServer(ThreadingMixIn, TCPServer): pass
631
632 if hasattr(socket, 'AF_UNIX'):
633
634 class UnixStreamServer(TCPServer):
635 address_family = socket.AF_UNIX
636
637 class UnixDatagramServer(UDPServer):
638 address_family = socket.AF_UNIX
639
640 class ThreadingUnixStreamServer(ThreadingMixIn, UnixStreamServer): pass
641
642 class ThreadingUnixDatagramServer(ThreadingMixIn, UnixDatagramServer): pass
643
644 class BaseRequestHandler(object):
645
646 """Base class for request handler classes.
647
648 This class is instantiated for each request to be handled. The
649 constructor sets the instance variables request, client_address
650 and server, and then calls the handle() method. To implement a
651 specific service, all you need to do is to derive a class which
652 defines a handle() method.
653
654 The handle() method can find the request as self.request, the
655 client address as self.client_address, and the server (in case it
656 needs access to per-server information) as self.server. Since a
657 separate instance is created for each request, the handle() method
658 can define arbitrary other instance variariables.
659
660 """
661
662 def __init__(self, request, client_address, server):
663 self.request = request
664 self.client_address = client_address
665 self.server = server
666 self.setup()
667 try:
668 self.handle()
669 finally:
670 self.finish()
671
672 def setup(self):
673 pass
674
675 def handle(self):
676 pass
677
678 def finish(self):
679 pass
680
681
682 # The following two classes make it possible to use the same service
683 # class for stream or datagram servers.
684 # Each class sets up these instance variables:
685 # - rfile: a file object from which receives the request is read
686 # - wfile: a file object to which the reply is written
687 # When the handle() method returns, wfile is flushed properly
688
689
690 class StreamRequestHandler(BaseRequestHandler):
691
692 """Define self.rfile and self.wfile for stream sockets."""
693
694 # Default buffer sizes for rfile, wfile.
695 # We default rfile to buffered because otherwise it could be
696 # really slow for large data (a getc() call per byte); we make
697 # wfile unbuffered because (a) often after a write() we want to
698 # read and we need to flush the line; (b) big writes to unbuffered
699 # files are typically optimized by stdio even when big reads
700 # aren't.
701 rbufsize = -1
702 wbufsize = 0
703
704 # A timeout to apply to the request socket, if not None.
705 timeout = None
706
707 # Disable nagle algorithm for this socket, if True.
708 # Use only when wbufsize != 0, to avoid small packets.
709 disable_nagle_algorithm = False
710
711 def setup(self):
712 self.connection = self.request
713 if self.timeout is not None:
714 self.connection.settimeout(self.timeout)
715 if self.disable_nagle_algorithm:
716 self.connection.setsockopt(socket.IPPROTO_TCP,
717 socket.TCP_NODELAY, True)
718 self.rfile = self.connection.makefile('rb', self.rbufsize)
719 self.wfile = self.connection.makefile('wb', self.wbufsize)
720
721 def finish(self):
722 if not self.wfile.closed:
723 try:
724 self.wfile.flush()
725 except socket.error:
726 # An final socket error may have occurred here, such as
727 # the local error ECONNABORTED.
728 pass
729 self.wfile.close()
730 self.rfile.close()
731
732
733 class DatagramRequestHandler(BaseRequestHandler):
734
735 # XXX Regrettably, I cannot get this working on Linux;
736 # s.recvfrom() doesn't return a meaningful client address.
737
738 """Define self.rfile and self.wfile for datagram sockets."""
739
740 def setup(self):
741 from io import BytesIO
742 self.packet, self.socket = self.request
743 self.rfile = BytesIO(self.packet)
744 self.wfile = BytesIO()
745
746 def finish(self):
747 self.socket.sendto(self.wfile.getvalue(), self.client_address)