1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
1001
1002
1003
1004
1005
1006
1007
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
1018
1019
1020
1021
1022
1023
1024
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
1035
1036
1037
1038
1039
1040
1041
1042
1043
1044
1045
1046
1047
1048
1049
1050
1051
1052
1053
1054
1055
1056
1057
1058
1059
1060
1061
1062
1063
1064
1065
1066
1067
1068
1069
1070
1071
1072
1073
1074
1075
1076
1077
1078
1079
1080
1081
1082
1083
1084
1085
1086
1087
1088
1089
1090
1091
1092
1093
1094
1095
1096
1097
1098
1099
1100
1101
1102
1103
1104
1105
1106
1107
1108
1109
1110
1111
1112
1113
1114
1115
1116
1117
1118
1119
1120
1121
1122
1123
1124
1125
1126
1127
1128
1129
1130
1131
1132
1133
1134
1135
1136
1137
1138
1139
1140
1141
1142
1143
1144
1145
1146
1147
1148
1149
1150
1151
1152
1153
1154
1155
1156
1157
1158
1159
1160
1161
1162
1163
1164
1165
1166
1167
1168
1169
1170
1171
1172
1173
1174
1175
1176
1177
1178
1179
1180
1181
1182
1183
1184
1185
1186
1187
1188
1189
1190
1191
1192
1193
1194
1195
1196
1197
1198
1199
1200
1201
1202
1203
1204
1205
1206
1207
1208
1209
1210
1211
1212
1213
1214
1215
1216
1217
1218
1219
1220
1221
1222
1223
1224
1225
1226
1227
1228
1229
1230
1231
1232
1233
1234
1235
1236
1237
1238
1239
1240
1241
1242
1243
1244
1245
1246
1247
1248
1249
1250
1251
1252
1253
1254
1255
1256
1257
1258
1259
1260
1261
1262
1263
1264
1265
1266
1267
1268
1269
1270
1271
1272
1273
1274
1275
1276
1277
1278
1279
1280
1281
1282
1283
1284
1285
1286
1287
1288
1289
1290
1291
1292
1293
1294
1295
1296
1297
1298
1299
1300
1301
1302
1303
1304
1305
1306
1307
1308
1309
1310
1311
1312
1313
1314
1315
1316
1317
1318
1319
1320
1321
1322
1323
1324
1325
1326
1327
1328
1329
1330
1331
1332
1333
1334
1335
1336
1337
1338
1339
1340
1341
1342
1343
1344
1345
1346
1347
1348
1349
1350
1351
1352
1353
1354
1355
1356
1357
1358
1359
1360
1361
1362
1363
1364
1365
1366
1367
1368
1369
1370
1371
1372
1373
1374
1375
1376
1377
1378
1379
1380
1381
1382
1383
1384
1385
1386
1387
1388
1389
1390
1391
1392
1393
1394
1395
1396
1397
1398
1399
1400
1401
1402
1403
1404
1405
1406
1407
1408
1409
1410
1411
1412
1413
1414
1415
1416
1417
1418
1419
1420
1421
1422
1423
1424
1425
1426
1427
1428
1429
1430
1431
1432
1433
1434
1435
1436
1437
1438
1439
1440
1441
1442
1443
1444
1445
1446
1447
1448
1449
1450
1451
1452
1453
1454
1455
1456
1457
1458
1459
1460
1461
1462
1463
1464
1465
1466
1467
1468
1469
1470
1471
1472
1473
1474
1475
1476
1477
1478
1479
1480
1481
1482
1483
1484
1485
1486
1487
1488
1489
1490
1491
1492
1493
1494
1495
1496
1497
1498
1499
1500
1501
1502
1503
1504
1505
1506
1507
1508
1509
1510
1511
1512
1513
1514
1515
1516
1517
1518
1519
1520
1521
1522
1523
1524
1525
1526
1527
1528
1529
1530
1531
1532
1533
1534
1535
1536
1537
1538
1539
1540
1541
1542
1543
1544
1545
1546
1547
1548
1549
1550
1551
1552
1553
1554
1555
1556
1557
1558
1559
1560
1561
1562
1563
1564
1565
1566
1567
1568
1569
1570
1571
1572
1573
1574
1575
1576
1577
1578
1579
1580
1581
1582
1583
1584
1585
1586
1587
1588
1589
1590
1591
1592
1593
1594
1595
1596
1597
1598
1599
1600
|
# -*- test-case-name: twisted.web.test.test_webclient -*-
# Copyright (c) Twisted Matrix Laboratories.
# See LICENSE for details.
"""
HTTP client.
"""
import os, types
from urlparse import urlunparse
from urllib import splithost, splittype
import zlib
from zope.interface import implements
from twisted.python import log
from twisted.python.failure import Failure
from twisted.web import http
from twisted.internet import defer, protocol, task, reactor
from twisted.internet.interfaces import IProtocol
from twisted.internet.endpoints import TCP4ClientEndpoint, SSL4ClientEndpoint
from twisted.python import failure
from twisted.python.util import InsensitiveDict
from twisted.python.components import proxyForInterface
from twisted.web import error
from twisted.web.iweb import UNKNOWN_LENGTH, IBodyProducer, IResponse
from twisted.web.http_headers import Headers
from twisted.python.compat import set
class PartialDownloadError(error.Error):
"""
Page was only partially downloaded, we got disconnected in middle.
@ivar response: All of the response body which was downloaded.
"""
class HTTPPageGetter(http.HTTPClient):
"""
Gets a resource via HTTP, then quits.
Typically used with L{HTTPClientFactory}. Note that this class does not, by
itself, do anything with the response. If you want to download a resource
into a file, use L{HTTPPageDownloader} instead.
@ivar _completelyDone: A boolean indicating whether any further requests are
necessary after this one completes in order to provide a result to
C{self.factory.deferred}. If it is C{False}, then a redirect is going
to be followed. Otherwise, this protocol's connection is the last one
before firing the result Deferred. This is used to make sure the result
Deferred is only fired after the connection is cleaned up.
"""
quietLoss = 0
followRedirect = True
failed = 0
_completelyDone = True
_specialHeaders = set(('host', 'user-agent', 'cookie', 'content-length'))
def connectionMade(self):
method = getattr(self.factory, 'method', 'GET')
self.sendCommand(method, self.factory.path)
if self.factory.scheme == 'http' and self.factory.port != 80:
host = '%s:%s' % (self.factory.host, self.factory.port)
elif self.factory.scheme == 'https' and self.factory.port != 443:
host = '%s:%s' % (self.factory.host, self.factory.port)
else:
host = self.factory.host
self.sendHeader('Host', self.factory.headers.get("host", host))
self.sendHeader('User-Agent', self.factory.agent)
data = getattr(self.factory, 'postdata', None)
if data is not None:
self.sendHeader("Content-Length", str(len(data)))
cookieData = []
for (key, value) in self.factory.headers.items():
if key.lower() not in self._specialHeaders:
# we calculated it on our own
self.sendHeader(key, value)
if key.lower() == 'cookie':
cookieData.append(value)
for cookie, cookval in self.factory.cookies.items():
cookieData.append('%s=%s' % (cookie, cookval))
if cookieData:
self.sendHeader('Cookie', '; '.join(cookieData))
self.endHeaders()
self.headers = {}
if data is not None:
self.transport.write(data)
def handleHeader(self, key, value):
"""
Called every time a header is received. Stores the header information
as key-value pairs in the C{headers} attribute.
@type key: C{str}
@param key: An HTTP header field name.
@type value: C{str}
@param value: An HTTP header field value.
"""
key = key.lower()
l = self.headers.setdefault(key, [])
l.append(value)
def handleStatus(self, version, status, message):
self.version, self.status, self.message = version, status, message
self.factory.gotStatus(version, status, message)
def handleEndHeaders(self):
self.factory.gotHeaders(self.headers)
m = getattr(self, 'handleStatus_'+self.status, self.handleStatusDefault)
m()
def handleStatus_200(self):
pass
handleStatus_201 = lambda self: self.handleStatus_200()
handleStatus_202 = lambda self: self.handleStatus_200()
def handleStatusDefault(self):
self.failed = 1
def handleStatus_301(self):
l = self.headers.get('location')
if not l:
self.handleStatusDefault()
return
url = l[0]
if self.followRedirect:
scheme, host, port, path = \
_parse(url, defaultPort=self.transport.getPeer().port)
self.factory._redirectCount += 1
if self.factory._redirectCount >= self.factory.redirectLimit:
err = error.InfiniteRedirection(
self.status,
'Infinite redirection detected',
location=url)
self.factory.noPage(failure.Failure(err))
self.quietLoss = True
self.transport.loseConnection()
return
self._completelyDone = False
self.factory.setURL(url)
if self.factory.scheme == 'https':
from twisted.internet import ssl
contextFactory = ssl.ClientContextFactory()
reactor.connectSSL(self.factory.host, self.factory.port,
self.factory, contextFactory)
else:
reactor.connectTCP(self.factory.host, self.factory.port,
self.factory)
else:
self.handleStatusDefault()
self.factory.noPage(
failure.Failure(
error.PageRedirect(
self.status, self.message, location = url)))
self.quietLoss = True
self.transport.loseConnection()
def handleStatus_302(self):
if self.afterFoundGet:
self.handleStatus_303()
else:
self.handleStatus_301()
def handleStatus_303(self):
self.factory.method = 'GET'
self.handleStatus_301()
def connectionLost(self, reason):
"""
When the connection used to issue the HTTP request is closed, notify the
factory if we have not already, so it can produce a result.
"""
if not self.quietLoss:
http.HTTPClient.connectionLost(self, reason)
self.factory.noPage(reason)
if self._completelyDone:
# Only if we think we're completely done do we tell the factory that
# we're "disconnected". This way when we're following redirects,
# only the last protocol used will fire the _disconnectedDeferred.
self.factory._disconnectedDeferred.callback(None)
def handleResponse(self, response):
if self.quietLoss:
return
if self.failed:
self.factory.noPage(
failure.Failure(
error.Error(
self.status, self.message, response)))
if self.factory.method == 'HEAD':
# Callback with empty string, since there is never a response
# body for HEAD requests.
self.factory.page('')
elif self.length != None and self.length != 0:
self.factory.noPage(failure.Failure(
PartialDownloadError(self.status, self.message, response)))
else:
self.factory.page(response)
# server might be stupid and not close connection. admittedly
# the fact we do only one request per connection is also
# stupid...
self.transport.loseConnection()
def timeout(self):
self.quietLoss = True
self.transport.loseConnection()
self.factory.noPage(defer.TimeoutError("Getting %s took longer than %s seconds." % (self.factory.url, self.factory.timeout)))
class HTTPPageDownloader(HTTPPageGetter):
transmittingPage = 0
def handleStatus_200(self, partialContent=0):
HTTPPageGetter.handleStatus_200(self)
self.transmittingPage = 1
self.factory.pageStart(partialContent)
def handleStatus_206(self):
self.handleStatus_200(partialContent=1)
def handleResponsePart(self, data):
if self.transmittingPage:
self.factory.pagePart(data)
def handleResponseEnd(self):
if self.length:
self.transmittingPage = 0
self.factory.noPage(
failure.Failure(
PartialDownloadError(self.status)))
if self.transmittingPage:
self.factory.pageEnd()
self.transmittingPage = 0
if self.failed:
self.factory.noPage(
failure.Failure(
error.Error(
self.status, self.message, None)))
self.transport.loseConnection()
class HTTPClientFactory(protocol.ClientFactory):
"""Download a given URL.
@type deferred: Deferred
@ivar deferred: A Deferred that will fire when the content has
been retrieved. Once this is fired, the ivars `status', `version',
and `message' will be set.
@type status: str
@ivar status: The status of the response.
@type version: str
@ivar version: The version of the response.
@type message: str
@ivar message: The text message returned with the status.
@type response_headers: dict
@ivar response_headers: The headers that were specified in the
response from the server.
@type method: str
@ivar method: The HTTP method to use in the request. This should be one of
OPTIONS, GET, HEAD, POST, PUT, DELETE, TRACE, or CONNECT (case
matters). Other values may be specified if the server being contacted
supports them.
@type redirectLimit: int
@ivar redirectLimit: The maximum number of HTTP redirects that can occur
before it is assumed that the redirection is endless.
@type afterFoundGet: C{bool}
@ivar afterFoundGet: Deviate from the HTTP 1.1 RFC by handling redirects
the same way as most web browsers; if the request method is POST and a
302 status is encountered, the redirect is followed with a GET method
@type _redirectCount: int
@ivar _redirectCount: The current number of HTTP redirects encountered.
@ivar _disconnectedDeferred: A L{Deferred} which only fires after the last
connection associated with the request (redirects may cause multiple
connections to be required) has closed. The result Deferred will only
fire after this Deferred, so that callers can be assured that there are
no more event sources in the reactor once they get the result.
"""
protocol = HTTPPageGetter
url = None
scheme = None
host = ''
port = None
path = None
def __init__(self, url, method='GET', postdata=None, headers=None,
agent="Twisted PageGetter", timeout=0, cookies=None,
followRedirect=True, redirectLimit=20,
afterFoundGet=False):
self.followRedirect = followRedirect
self.redirectLimit = redirectLimit
self._redirectCount = 0
self.timeout = timeout
self.agent = agent
self.afterFoundGet = afterFoundGet
if cookies is None:
cookies = {}
self.cookies = cookies
if headers is not None:
self.headers = InsensitiveDict(headers)
else:
self.headers = InsensitiveDict()
if postdata is not None:
self.headers.setdefault('Content-Length', len(postdata))
# just in case a broken http/1.1 decides to keep connection alive
self.headers.setdefault("connection", "close")
self.postdata = postdata
self.method = method
self.setURL(url)
self.waiting = 1
self._disconnectedDeferred = defer.Deferred()
self.deferred = defer.Deferred()
# Make sure the first callback on the result Deferred pauses the
# callback chain until the request connection is closed.
self.deferred.addBoth(self._waitForDisconnect)
self.response_headers = None
def _waitForDisconnect(self, passthrough):
"""
Chain onto the _disconnectedDeferred, preserving C{passthrough}, so that
the result is only available after the associated connection has been
closed.
"""
self._disconnectedDeferred.addCallback(lambda ignored: passthrough)
return self._disconnectedDeferred
def __repr__(self):
return "<%s: %s>" % (self.__class__.__name__, self.url)
def setURL(self, url):
self.url = url
scheme, host, port, path = _parse(url)
if scheme and host:
self.scheme = scheme
self.host = host
self.port = port
self.path = path
def buildProtocol(self, addr):
p = protocol.ClientFactory.buildProtocol(self, addr)
p.followRedirect = self.followRedirect
p.afterFoundGet = self.afterFoundGet
if self.timeout:
timeoutCall = reactor.callLater(self.timeout, p.timeout)
self.deferred.addBoth(self._cancelTimeout, timeoutCall)
return p
def _cancelTimeout(self, result, timeoutCall):
if timeoutCall.active():
timeoutCall.cancel()
return result
def gotHeaders(self, headers):
self.response_headers = headers
if 'set-cookie' in headers:
for cookie in headers['set-cookie']:
cookparts = cookie.split(';')
cook = cookparts[0]
cook.lstrip()
k, v = cook.split('=', 1)
self.cookies[k.lstrip()] = v.lstrip()
def gotStatus(self, version, status, message):
self.version, self.status, self.message = version, status, message
def page(self, page):
if self.waiting:
self.waiting = 0
self.deferred.callback(page)
def noPage(self, reason):
if self.waiting:
self.waiting = 0
self.deferred.errback(reason)
def clientConnectionFailed(self, _, reason):
"""
When a connection attempt fails, the request cannot be issued. If no
result has yet been provided to the result Deferred, provide the
connection failure reason as an error result.
"""
if self.waiting:
self.waiting = 0
# If the connection attempt failed, there is nothing more to
# disconnect, so just fire that Deferred now.
self._disconnectedDeferred.callback(None)
self.deferred.errback(reason)
class HTTPDownloader(HTTPClientFactory):
"""Download to a file."""
protocol = HTTPPageDownloader
value = None
def __init__(self, url, fileOrName,
method='GET', postdata=None, headers=None,
agent="Twisted client", supportPartial=0,
timeout=0, cookies=None, followRedirect=1,
redirectLimit=20, afterFoundGet=False):
self.requestedPartial = 0
if isinstance(fileOrName, types.StringTypes):
self.fileName = fileOrName
self.file = None
if supportPartial and os.path.exists(self.fileName):
fileLength = os.path.getsize(self.fileName)
if fileLength:
self.requestedPartial = fileLength
if headers == None:
headers = {}
headers["range"] = "bytes=%d-" % fileLength
else:
self.file = fileOrName
HTTPClientFactory.__init__(
self, url, method=method, postdata=postdata, headers=headers,
agent=agent, timeout=timeout, cookies=cookies,
followRedirect=followRedirect, redirectLimit=redirectLimit,
afterFoundGet=afterFoundGet)
def gotHeaders(self, headers):
HTTPClientFactory.gotHeaders(self, headers)
if self.requestedPartial:
contentRange = headers.get("content-range", None)
if not contentRange:
# server doesn't support partial requests, oh well
self.requestedPartial = 0
return
start, end, realLength = http.parseContentRange(contentRange[0])
if start != self.requestedPartial:
# server is acting wierdly
self.requestedPartial = 0
def openFile(self, partialContent):
if partialContent:
file = open(self.fileName, 'rb+')
file.seek(0, 2)
else:
file = open(self.fileName, 'wb')
return file
def pageStart(self, partialContent):
"""Called on page download start.
@param partialContent: tells us if the download is partial download we requested.
"""
if partialContent and not self.requestedPartial:
raise ValueError, "we shouldn't get partial content response if we didn't want it!"
if self.waiting:
try:
if not self.file:
self.file = self.openFile(partialContent)
except IOError:
#raise
self.deferred.errback(failure.Failure())
def pagePart(self, data):
if not self.file:
return
try:
self.file.write(data)
except IOError:
#raise
self.file = None
self.deferred.errback(failure.Failure())
def noPage(self, reason):
"""
Close the storage file and errback the waiting L{Deferred} with the
given reason.
"""
if self.waiting:
self.waiting = 0
if self.file:
try:
self.file.close()
except:
log.err(None, "Error closing HTTPDownloader file")
self.deferred.errback(reason)
def pageEnd(self):
self.waiting = 0
if not self.file:
return
try:
self.file.close()
except IOError:
self.deferred.errback(failure.Failure())
return
self.deferred.callback(self.value)
class _URL(tuple):
"""
A parsed URL.
At some point this should be replaced with a better URL implementation.
"""
def __new__(self, scheme, host, port, path):
return tuple.__new__(_URL, (scheme, host, port, path))
def __init__(self, scheme, host, port, path):
self.scheme = scheme
self.host = host
self.port = port
self.path = path
def _parse(url, defaultPort=None):
"""
Split the given URL into the scheme, host, port, and path.
@type url: C{str}
@param url: An URL to parse.
@type defaultPort: C{int} or C{None}
@param defaultPort: An alternate value to use as the port if the URL does
not include one.
@return: A four-tuple of the scheme, host, port, and path of the URL. All
of these are C{str} instances except for port, which is an C{int}.
"""
url = url.strip()
parsed = http.urlparse(url)
scheme = parsed[0]
path = urlunparse(('', '') + parsed[2:])
if defaultPort is None:
if scheme == 'https':
defaultPort = 443
else:
defaultPort = 80
host, port = parsed[1], defaultPort
if ':' in host:
host, port = host.split(':')
try:
port = int(port)
except ValueError:
port = defaultPort
if path == '':
path = '/'
return _URL(scheme, host, port, path)
def _makeGetterFactory(url, factoryFactory, contextFactory=None,
*args, **kwargs):
"""
Create and connect an HTTP page getting factory.
Any additional positional or keyword arguments are used when calling
C{factoryFactory}.
@param factoryFactory: Factory factory that is called with C{url}, C{args}
and C{kwargs} to produce the getter
@param contextFactory: Context factory to use when creating a secure
connection, defaulting to C{None}
@return: The factory created by C{factoryFactory}
"""
scheme, host, port, path = _parse(url)
factory = factoryFactory(url, *args, **kwargs)
if scheme == 'https':
from twisted.internet import ssl
if contextFactory is None:
contextFactory = ssl.ClientContextFactory()
reactor.connectSSL(host, port, factory, contextFactory)
else:
reactor.connectTCP(host, port, factory)
return factory
def getPage(url, contextFactory=None, *args, **kwargs):
"""
Download a web page as a string.
Download a page. Return a deferred, which will callback with a
page (as a string) or errback with a description of the error.
See L{HTTPClientFactory} to see what extra arguments can be passed.
"""
return _makeGetterFactory(
url,
HTTPClientFactory,
contextFactory=contextFactory,
*args, **kwargs).deferred
def downloadPage(url, file, contextFactory=None, *args, **kwargs):
"""
Download a web page to a file.
@param file: path to file on filesystem, or file-like object.
See HTTPDownloader to see what extra args can be passed.
"""
factoryFactory = lambda url, *a, **kw: HTTPDownloader(url, file, *a, **kw)
return _makeGetterFactory(
url,
factoryFactory,
contextFactory=contextFactory,
*args, **kwargs).deferred
# The code which follows is based on the new HTTP client implementation. It
# should be significantly better than anything above, though it is not yet
# feature equivalent.
from twisted.web.error import SchemeNotSupported
from twisted.web._newclient import Request, Response, HTTP11ClientProtocol
from twisted.web._newclient import ResponseDone, ResponseFailed
from twisted.web._newclient import RequestNotSent, RequestTransmissionFailed
from twisted.web._newclient import ResponseNeverReceived
try:
from twisted.internet.ssl import ClientContextFactory
except ImportError:
class WebClientContextFactory(object):
"""
A web context factory which doesn't work because the necessary SSL
support is missing.
"""
def getContext(self, hostname, port):
raise NotImplementedError("SSL support unavailable")
else:
class WebClientContextFactory(ClientContextFactory):
"""
A web context factory which ignores the hostname and port and does no
certificate verification.
"""
def getContext(self, hostname, port):
return ClientContextFactory.getContext(self)
class _WebToNormalContextFactory(object):
"""
Adapt a web context factory to a normal context factory.
@ivar _webContext: A web context factory which accepts a hostname and port
number to its C{getContext} method.
@ivar _hostname: The hostname which will be passed to
C{_webContext.getContext}.
@ivar _port: The port number which will be passed to
C{_webContext.getContext}.
"""
def __init__(self, webContext, hostname, port):
self._webContext = webContext
self._hostname = hostname
self._port = port
def getContext(self):
"""
Called the wrapped web context factory's C{getContext} method with a
hostname and port number and return the resulting context object.
"""
return self._webContext.getContext(self._hostname, self._port)
class FileBodyProducer(object):
"""
L{FileBodyProducer} produces bytes from an input file object incrementally
and writes them to a consumer.
Since file-like objects cannot be read from in an event-driven manner,
L{FileBodyProducer} uses a L{Cooperator} instance to schedule reads from
the file. This process is also paused and resumed based on notifications
from the L{IConsumer} provider being written to.
The file is closed after it has been read, or if the producer is stopped
early.
@ivar _inputFile: Any file-like object, bytes read from which will be
written to a consumer.
@ivar _cooperate: A method like L{Cooperator.cooperate} which is used to
schedule all reads.
@ivar _readSize: The number of bytes to read from C{_inputFile} at a time.
"""
implements(IBodyProducer)
# Python 2.4 doesn't have these symbolic constants
_SEEK_SET = getattr(os, 'SEEK_SET', 0)
_SEEK_END = getattr(os, 'SEEK_END', 2)
def __init__(self, inputFile, cooperator=task, readSize=2 ** 16):
self._inputFile = inputFile
self._cooperate = cooperator.cooperate
self._readSize = readSize
self.length = self._determineLength(inputFile)
def _determineLength(self, fObj):
"""
Determine how many bytes can be read out of C{fObj} (assuming it is not
modified from this point on). If the determination cannot be made,
return C{UNKNOWN_LENGTH}.
"""
try:
seek = fObj.seek
tell = fObj.tell
except AttributeError:
return UNKNOWN_LENGTH
originalPosition = tell()
seek(0, self._SEEK_END)
end = tell()
seek(originalPosition, self._SEEK_SET)
return end - originalPosition
def stopProducing(self):
"""
Permanently stop writing bytes from the file to the consumer by
stopping the underlying L{CooperativeTask}.
"""
self._inputFile.close()
self._task.stop()
def startProducing(self, consumer):
"""
Start a cooperative task which will read bytes from the input file and
write them to C{consumer}. Return a L{Deferred} which fires after all
bytes have been written.
@param consumer: Any L{IConsumer} provider
"""
self._task = self._cooperate(self._writeloop(consumer))
d = self._task.whenDone()
def maybeStopped(reason):
# IBodyProducer.startProducing's Deferred isn't support to fire if
# stopProducing is called.
reason.trap(task.TaskStopped)
return defer.Deferred()
d.addCallbacks(lambda ignored: None, maybeStopped)
return d
def _writeloop(self, consumer):
"""
Return an iterator which reads one chunk of bytes from the input file
and writes them to the consumer for each time it is iterated.
"""
while True:
bytes = self._inputFile.read(self._readSize)
if not bytes:
self._inputFile.close()
break
consumer.write(bytes)
yield None
def pauseProducing(self):
"""
Temporarily suspend copying bytes from the input file to the consumer
by pausing the L{CooperativeTask} which drives that activity.
"""
self._task.pause()
def resumeProducing(self):
"""
Undo the effects of a previous C{pauseProducing} and resume copying
bytes to the consumer by resuming the L{CooperativeTask} which drives
the write activity.
"""
self._task.resume()
class _HTTP11ClientFactory(protocol.Factory):
"""
A factory for L{HTTP11ClientProtocol}, used by L{HTTPConnectionPool}.
@ivar _quiescentCallback: The quiescent callback to be passed to protocol
instances, used to return them to the connection pool.
@since: 11.1
"""
def __init__(self, quiescentCallback):
self._quiescentCallback = quiescentCallback
def buildProtocol(self, addr):
return HTTP11ClientProtocol(self._quiescentCallback)
class _RetryingHTTP11ClientProtocol(object):
"""
A wrapper for L{HTTP11ClientProtocol} that automatically retries requests.
@ivar _clientProtocol: The underlying L{HTTP11ClientProtocol}.
@ivar _newConnection: A callable that creates a new connection for a
retry.
"""
def __init__(self, clientProtocol, newConnection):
self._clientProtocol = clientProtocol
self._newConnection = newConnection
def _shouldRetry(self, method, exception, bodyProducer):
"""
Indicate whether request should be retried.
Only returns C{True} if method is idempotent, no response was
received, and no body was sent. The latter requirement may be relaxed
in the future, and PUT added to approved method list.
"""
if method not in ("GET", "HEAD", "OPTIONS", "DELETE", "TRACE"):
return False
if not isinstance(exception, (RequestNotSent, RequestTransmissionFailed,
ResponseNeverReceived)):
return False
if bodyProducer is not None:
return False
return True
def request(self, request):
"""
Do a request, and retry once (with a new connection) it it fails in
a retryable manner.
@param request: A L{Request} instance that will be requested using the
wrapped protocol.
"""
d = self._clientProtocol.request(request)
def failed(reason):
if self._shouldRetry(request.method, reason.value,
request.bodyProducer):
return self._newConnection().addCallback(
lambda connection: connection.request(request))
else:
return reason
d.addErrback(failed)
return d
class HTTPConnectionPool(object):
"""
A pool of persistent HTTP connections.
Features:
- Cached connections will eventually time out.
- Limits on maximum number of persistent connections.
Connections are stored using keys, which should be chosen such that any
connections stored under a given key can be used interchangeably.
Failed requests done using previously cached connections will be retried
once if they use an idempotent method (e.g. GET), in case the HTTP server
timed them out.
@ivar persistent: Boolean indicating whether connections should be
persistent. Connections are persistent by default.
@ivar maxPersistentPerHost: The maximum number of cached persistent
connections for a C{host:port} destination.
@type maxPersistentPerHost: C{int}
@ivar cachedConnectionTimeout: Number of seconds a cached persistent
connection will stay open before disconnecting.
@ivar retryAutomatically: C{boolean} indicating whether idempotent
requests should be retried once if no response was received.
@ivar _factory: The factory used to connect to the proxy.
@ivar _connections: Map (scheme, host, port) to lists of
L{HTTP11ClientProtocol} instances.
@ivar _timeouts: Map L{HTTP11ClientProtocol} instances to a
C{IDelayedCall} instance of their timeout.
@since: 12.1
"""
_factory = _HTTP11ClientFactory
maxPersistentPerHost = 2
cachedConnectionTimeout = 240
retryAutomatically = True
def __init__(self, reactor, persistent=True):
self._reactor = reactor
self.persistent = persistent
self._connections = {}
self._timeouts = {}
def getConnection(self, key, endpoint):
"""
Retrieve a connection, either new or cached, to be used for a HTTP
request.
If a cached connection is returned, it will not be used for other
requests until it is put back (which will happen automatically), since
we do not support pipelined requests. If no cached connection is
available, the passed in endpoint is used to create the connection.
If the connection doesn't disconnect at the end of its request, it
will be returned to this pool automatically. As such, only a single
request should be sent using the returned connection.
@param key: A unique key identifying connections that can be used
interchangeably.
@param endpoint: An endpoint that can be used to open a new connection
if no cached connection is available.
@return: A C{Deferred} that will fire with a L{HTTP11ClientProtocol}
(or a wrapper) that can be used to send a single HTTP request.
"""
# Try to get cached version:
connections = self._connections.get(key)
while connections:
connection = connections.pop(0)
# Cancel timeout:
self._timeouts[connection].cancel()
del self._timeouts[connection]
if connection.state == "QUIESCENT":
if self.retryAutomatically:
newConnection = lambda: self._newConnection(key, endpoint)
connection = _RetryingHTTP11ClientProtocol(
connection, newConnection)
return defer.succeed(connection)
return self._newConnection(key, endpoint)
def _newConnection(self, key, endpoint):
"""
Create a new connection.
This implements the new connection code path for L{getConnection}.
"""
def quiescentCallback(protocol):
self._putConnection(key, protocol)
factory = self._factory(quiescentCallback)
return endpoint.connect(factory)
def _removeConnection(self, key, connection):
"""
Remove a connection from the cache and disconnect it.
"""
connection.transport.loseConnection()
self._connections[key].remove(connection)
del self._timeouts[connection]
def _putConnection(self, key, connection):
"""
Return a persistent connection to the pool. This will be called by
L{HTTP11ClientProtocol} when the connection becomes quiescent.
"""
if connection.state != "QUIESCENT":
# Log with traceback for debugging purposes:
try:
raise RuntimeError(
"BUG: Non-quiescent protocol added to connection pool.")
except:
log.err()
return
connections = self._connections.setdefault(key, [])
if len(connections) == self.maxPersistentPerHost:
dropped = connections.pop(0)
dropped.transport.loseConnection()
self._timeouts[dropped].cancel()
del self._timeouts[dropped]
connections.append(connection)
cid = self._reactor.callLater(self.cachedConnectionTimeout,
self._removeConnection,
key, connection)
self._timeouts[connection] = cid
def closeCachedConnections(self):
"""
Close all persistent connections and remove them from the pool.
@return: L{defer.Deferred} that fires when all connections have been
closed.
"""
results = []
for protocols in self._connections.itervalues():
for p in protocols:
results.append(p.abort())
self._connections = {}
for dc in self._timeouts.values():
dc.cancel()
self._timeouts = {}
return defer.gatherResults(results).addCallback(lambda ign: None)
class _AgentBase(object):
"""
Base class offering common facilities for L{Agent}-type classes.
@ivar _reactor: The C{IReactorTime} implementation which will be used by
the pool, and perhaps by subclasses as well.
@ivar _pool: The L{HTTPConnectionPool} used to manage HTTP connections.
"""
def __init__(self, reactor, pool):
if pool is None:
pool = HTTPConnectionPool(reactor, False)
self._reactor = reactor
self._pool = pool
def _computeHostValue(self, scheme, host, port):
"""
Compute the string to use for the value of the I{Host} header, based on
the given scheme, host name, and port number.
"""
if (scheme, port) in (('http', 80), ('https', 443)):
return host
return '%s:%d' % (host, port)
def _requestWithEndpoint(self, key, endpoint, method, parsedURI,
headers, bodyProducer, requestPath):
"""
Issue a new request, given the endpoint and the path sent as part of
the request.
"""
# Create minimal headers, if necessary:
if headers is None:
headers = Headers()
if not headers.hasHeader('host'):
headers = headers.copy()
headers.addRawHeader(
'host', self._computeHostValue(parsedURI.scheme, parsedURI.host,
parsedURI.port))
d = self._pool.getConnection(key, endpoint)
def cbConnected(proto):
return proto.request(
Request(method, requestPath, headers, bodyProducer,
persistent=self._pool.persistent))
d.addCallback(cbConnected)
return d
class Agent(_AgentBase):
"""
L{Agent} is a very basic HTTP client. It supports I{HTTP} and I{HTTPS}
scheme URIs (but performs no certificate checking by default).
@param pool: A L{HTTPConnectionPool} instance, or C{None}, in which case a
non-persistent L{HTTPConnectionPool} instance will be created.
@ivar _contextFactory: A web context factory which will be used to create
SSL context objects for any SSL connections the agent needs to make.
@ivar _connectTimeout: If not C{None}, the timeout passed to C{connectTCP}
or C{connectSSL} for specifying the connection timeout.
@ivar _bindAddress: If not C{None}, the address passed to C{connectTCP} or
C{connectSSL} for specifying the local address to bind to.
@since: 9.0
"""
def __init__(self, reactor, contextFactory=WebClientContextFactory(),
connectTimeout=None, bindAddress=None,
pool=None):
_AgentBase.__init__(self, reactor, pool)
self._contextFactory = contextFactory
self._connectTimeout = connectTimeout
self._bindAddress = bindAddress
def _wrapContextFactory(self, host, port):
"""
Create and return a normal context factory wrapped around
C{self._contextFactory} in such a way that C{self._contextFactory} will
have the host and port information passed to it.
@param host: A C{str} giving the hostname which will be connected to in
order to issue a request.
@param port: An C{int} giving the port number the connection will be
on.
@return: A context factory suitable to be passed to
C{reactor.connectSSL}.
"""
return _WebToNormalContextFactory(self._contextFactory, host, port)
def _getEndpoint(self, scheme, host, port):
"""
Get an endpoint for the given host and port, using a transport
selected based on scheme.
@param scheme: A string like C{'http'} or C{'https'} (the only two
supported values) to use to determine how to establish the
connection.
@param host: A C{str} giving the hostname which will be connected to in
order to issue a request.
@param port: An C{int} giving the port number the connection will be
on.
@return: An endpoint which can be used to connect to given address.
"""
kwargs = {}
if self._connectTimeout is not None:
kwargs['timeout'] = self._connectTimeout
kwargs['bindAddress'] = self._bindAddress
if scheme == 'http':
return TCP4ClientEndpoint(self._reactor, host, port, **kwargs)
elif scheme == 'https':
return SSL4ClientEndpoint(self._reactor, host, port,
self._wrapContextFactory(host, port),
**kwargs)
else:
raise SchemeNotSupported("Unsupported scheme: %r" % (scheme,))
def request(self, method, uri, headers=None, bodyProducer=None):
"""
Issue a new request.
@param method: The request method to send.
@type method: C{str}
@param uri: The request URI send.
@type uri: C{str}
@param headers: The request headers to send. If no I{Host} header is
included, one will be added based on the request URI.
@type headers: L{Headers}
@param bodyProducer: An object which will produce the request body or,
if the request body is to be empty, L{None}.
@type bodyProducer: L{IBodyProducer} provider
@return: A L{Deferred} which fires with the result of the request (a
L{twisted.web.iweb.IResponse} provider), or fails if there is a
problem setting up a connection over which to issue the request.
It may also fail with L{SchemeNotSupported} if the scheme of the
given URI is not supported.
@rtype: L{Deferred}
"""
parsedURI = _parse(uri)
try:
endpoint = self._getEndpoint(parsedURI.scheme, parsedURI.host,
parsedURI.port)
except SchemeNotSupported:
return defer.fail(Failure())
key = (parsedURI.scheme, parsedURI.host, parsedURI.port)
return self._requestWithEndpoint(key, endpoint, method, parsedURI,
headers, bodyProducer, parsedURI.path)
class ProxyAgent(_AgentBase):
"""
An HTTP agent able to cross HTTP proxies.
@ivar _proxyEndpoint: The endpoint used to connect to the proxy.
@since: 11.1
"""
def __init__(self, endpoint, reactor=None, pool=None):
if reactor is None:
from twisted.internet import reactor
_AgentBase.__init__(self, reactor, pool)
self._proxyEndpoint = endpoint
def request(self, method, uri, headers=None, bodyProducer=None):
"""
Issue a new request via the configured proxy.
"""
# Cache *all* connections under the same key, since we are only
# connecting to a single destination, the proxy:
key = ("http-proxy", self._proxyEndpoint)
# To support proxying HTTPS via CONNECT, we will use key
# ("http-proxy-CONNECT", scheme, host, port), and an endpoint that
# wraps _proxyEndpoint with an additional callback to do the CONNECT.
return self._requestWithEndpoint(key, self._proxyEndpoint, method,
_parse(uri), headers, bodyProducer,
uri)
class _FakeUrllib2Request(object):
"""
A fake C{urllib2.Request} object for C{cookielib} to work with.
@see: U{http://docs.python.org/library/urllib2.html#request-objects}
@type uri: C{str}
@ivar uri: Request URI.
@type headers: L{twisted.web.http_headers.Headers}
@ivar headers: Request headers.
@type type: C{str}
@ivar type: The scheme of the URI.
@type host: C{str}
@ivar host: The host[:port] of the URI.
@since: 11.1
"""
def __init__(self, uri):
self.uri = uri
self.headers = Headers()
self.type, rest = splittype(self.uri)
self.host, rest = splithost(rest)
def has_header(self, header):
return self.headers.hasHeader(header)
def add_unredirected_header(self, name, value):
self.headers.addRawHeader(name, value)
def get_full_url(self):
return self.uri
def get_header(self, name, default=None):
headers = self.headers.getRawHeaders(name, default)
if headers is not None:
return headers[0]
return None
def get_host(self):
return self.host
def get_type(self):
return self.type
def is_unverifiable(self):
# In theory this shouldn't be hardcoded.
return False
class _FakeUrllib2Response(object):
"""
A fake C{urllib2.Response} object for C{cookielib} to work with.
@type response: C{twisted.web.iweb.IResponse}
@ivar response: Underlying Twisted Web response.
@since: 11.1
"""
def __init__(self, response):
self.response = response
def info(self):
class _Meta(object):
def getheaders(zelf, name):
return self.response.headers.getRawHeaders(name, [])
return _Meta()
class CookieAgent(object):
"""
L{CookieAgent} extends the basic L{Agent} to add RFC-compliant
handling of HTTP cookies. Cookies are written to and extracted
from a C{cookielib.CookieJar} instance.
The same cookie jar instance will be used for any requests through this
agent, mutating it whenever a I{Set-Cookie} header appears in a response.
@type _agent: L{twisted.web.client.Agent}
@ivar _agent: Underlying Twisted Web agent to issue requests through.
@type cookieJar: C{cookielib.CookieJar}
@ivar cookieJar: Initialized cookie jar to read cookies from and store
cookies to.
@since: 11.1
"""
def __init__(self, agent, cookieJar):
self._agent = agent
self.cookieJar = cookieJar
def request(self, method, uri, headers=None, bodyProducer=None):
"""
Issue a new request to the wrapped L{Agent}.
Send a I{Cookie} header if a cookie for C{uri} is stored in
L{CookieAgent.cookieJar}. Cookies are automatically extracted and
stored from requests.
If a C{'cookie'} header appears in C{headers} it will override the
automatic cookie header obtained from the cookie jar.
@see: L{Agent.request}
"""
if headers is None:
headers = Headers()
lastRequest = _FakeUrllib2Request(uri)
# Setting a cookie header explicitly will disable automatic request
# cookies.
if not headers.hasHeader('cookie'):
self.cookieJar.add_cookie_header(lastRequest)
cookieHeader = lastRequest.get_header('Cookie', None)
if cookieHeader is not None:
headers = headers.copy()
headers.addRawHeader('cookie', cookieHeader)
d = self._agent.request(method, uri, headers, bodyProducer)
d.addCallback(self._extractCookies, lastRequest)
return d
def _extractCookies(self, response, request):
"""
Extract response cookies and store them in the cookie jar.
@type response: L{twisted.web.iweb.IResponse}
@param response: Twisted Web response.
@param request: A urllib2 compatible request object.
"""
resp = _FakeUrllib2Response(response)
self.cookieJar.extract_cookies(resp, request)
return response
class GzipDecoder(proxyForInterface(IResponse)):
"""
A wrapper for a L{Response} instance which handles gzip'ed body.
@ivar original: The original L{Response} object.
@since: 11.1
"""
def __init__(self, response):
self.original = response
self.length = UNKNOWN_LENGTH
def deliverBody(self, protocol):
"""
Override C{deliverBody} to wrap the given C{protocol} with
L{_GzipProtocol}.
"""
self.original.deliverBody(_GzipProtocol(protocol, self.original))
class _GzipProtocol(proxyForInterface(IProtocol)):
"""
A L{Protocol} implementation which wraps another one, transparently
decompressing received data.
@ivar _zlibDecompress: A zlib decompress object used to decompress the data
stream.
@ivar _response: A reference to the original response, in case of errors.
@since: 11.1
"""
def __init__(self, protocol, response):
self.original = protocol
self._response = response
self._zlibDecompress = zlib.decompressobj(16 + zlib.MAX_WBITS)
def dataReceived(self, data):
"""
Decompress C{data} with the zlib decompressor, forwarding the raw data
to the original protocol.
"""
try:
rawData = self._zlibDecompress.decompress(data)
except zlib.error:
raise ResponseFailed([failure.Failure()], self._response)
if rawData:
self.original.dataReceived(rawData)
def connectionLost(self, reason):
"""
Forward the connection lost event, flushing remaining data from the
decompressor if any.
"""
try:
rawData = self._zlibDecompress.flush()
except zlib.error:
raise ResponseFailed([reason, failure.Failure()], self._response)
if rawData:
self.original.dataReceived(rawData)
self.original.connectionLost(reason)
class ContentDecoderAgent(object):
"""
An L{Agent} wrapper to handle encoded content.
It takes care of declaring the support for content in the
I{Accept-Encoding} header, and automatically decompresses the received data
if it's effectively using compression.
@param decoders: A list or tuple of (name, decoder) objects. The name
declares which decoding the decoder supports, and the decoder must
return a response object when called/instantiated. For example,
C{(('gzip', GzipDecoder))}. The order determines how the decoders are
going to be advertized to the server.
@since: 11.1
"""
def __init__(self, agent, decoders):
self._agent = agent
self._decoders = dict(decoders)
self._supported = ','.join([decoder[0] for decoder in decoders])
def request(self, method, uri, headers=None, bodyProducer=None):
"""
Send a client request which declares supporting compressed content.
@see: L{Agent.request}.
"""
if headers is None:
headers = Headers()
else:
headers = headers.copy()
headers.addRawHeader('accept-encoding', self._supported)
deferred = self._agent.request(method, uri, headers, bodyProducer)
return deferred.addCallback(self._handleResponse)
def _handleResponse(self, response):
"""
Check if the response is encoded, and wrap it to handle decompression.
"""
contentEncodingHeaders = response.headers.getRawHeaders(
'content-encoding', [])
contentEncodingHeaders = ','.join(contentEncodingHeaders).split(',')
while contentEncodingHeaders:
name = contentEncodingHeaders.pop().strip()
decoder = self._decoders.get(name)
if decoder is not None:
response = decoder(response)
else:
# Add it back
contentEncodingHeaders.append(name)
break
if contentEncodingHeaders:
response.headers.setRawHeaders(
'content-encoding', [','.join(contentEncodingHeaders)])
else:
response.headers.removeHeader('content-encoding')
return response
class RedirectAgent(object):
"""
An L{Agent} wrapper which handles HTTP redirects.
The implementation is rather strict: 301 and 302 behaves like 307, not
redirecting automatically on methods different from C{GET} and C{HEAD}.
@param redirectLimit: The maximum number of times the agent is allowed to
follow redirects before failing with a L{error.InfiniteRedirection}.
@since: 11.1
"""
def __init__(self, agent, redirectLimit=20):
self._agent = agent
self._redirectLimit = redirectLimit
def request(self, method, uri, headers=None, bodyProducer=None):
"""
Send a client request following HTTP redirects.
@see: L{Agent.request}.
"""
deferred = self._agent.request(method, uri, headers, bodyProducer)
return deferred.addCallback(
self._handleResponse, method, uri, headers, 0)
def _handleRedirect(self, response, method, uri, headers, redirectCount):
"""
Handle a redirect response, checking the number of redirects already
followed, and extracting the location header fields.
"""
if redirectCount >= self._redirectLimit:
err = error.InfiniteRedirection(
response.code,
'Infinite redirection detected',
location=uri)
raise ResponseFailed([failure.Failure(err)], response)
locationHeaders = response.headers.getRawHeaders('location', [])
if not locationHeaders:
err = error.RedirectWithNoLocation(
response.code, 'No location header field', uri)
raise ResponseFailed([failure.Failure(err)], response)
location = locationHeaders[0]
deferred = self._agent.request(method, location, headers)
return deferred.addCallback(
self._handleResponse, method, uri, headers, redirectCount + 1)
def _handleResponse(self, response, method, uri, headers, redirectCount):
"""
Handle the response, making another request if it indicates a redirect.
"""
if response.code in (http.MOVED_PERMANENTLY, http.FOUND,
http.TEMPORARY_REDIRECT):
if method not in ('GET', 'HEAD'):
err = error.PageRedirect(response.code, location=uri)
raise ResponseFailed([failure.Failure(err)], response)
return self._handleRedirect(response, method, uri, headers,
redirectCount)
elif response.code == http.SEE_OTHER:
return self._handleRedirect(response, 'GET', uri, headers,
redirectCount)
return response
__all__ = [
'PartialDownloadError', 'HTTPPageGetter', 'HTTPPageDownloader',
'HTTPClientFactory', 'HTTPDownloader', 'getPage', 'downloadPage',
'ResponseDone', 'Response', 'ResponseFailed', 'Agent', 'CookieAgent',
'ProxyAgent', 'ContentDecoderAgent', 'GzipDecoder', 'RedirectAgent',
'HTTPConnectionPool']
|