This repository was archived by the owner on Aug 3, 2022. It is now read-only.
-
Notifications
You must be signed in to change notification settings - Fork 69
/
Copy pathhttpclient.py
512 lines (431 loc) · 16.5 KB
/
httpclient.py
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
#!/usr/bin/env python
# Copyright 2012 Google Inc. All Rights Reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
"""Retrieve web resources over http."""
import copy
import datetime
import httplib
import logging
import random
import ssl
import StringIO
import httparchive
import platformsettings
import script_injector
# PIL isn't always available, but we still want to be able to run without
# the image scrambling functionality in this case.
try:
import Image
except ImportError:
Image = None
TIMER = platformsettings.timer
class HttpClientException(Exception):
"""Base class for all exceptions in httpclient."""
pass
def _InjectScripts(response, injector):
"""Injects script generated by |injector| immediately after <head> or <html>.
Copies |response| if it is modified.
Args:
response: an ArchivedHttpResponse
injector: function which generates JavaScript string
based on recording time (e.g. "Math.random = function(){...}")
Returns:
an ArchivedHttpResponse
"""
if type(response) == tuple:
logging.warn('tuple response: %s', response)
content_type = response.get_header('content-type')
if content_type and content_type.startswith('text/html'):
text_chunks = response.get_data_as_chunks()
text_chunks, just_injected = script_injector.InjectScript(
text_chunks, 'text/html', injector(response.request_time))
if just_injected:
response = copy.deepcopy(response)
response.set_data_from_chunks(text_chunks)
return response
def _ScrambleImages(response):
"""If the |response| is an image, attempt to scramble it.
Copies |response| if it is modified.
Args:
response: an ArchivedHttpResponse
Returns:
an ArchivedHttpResponse
"""
assert Image, '--scramble_images requires the PIL module to be installed.'
content_type = response.get_header('content-type')
if content_type and content_type.startswith('image/'):
try:
image_data = response.response_data[0]
image_data.decode(encoding='base64')
im = Image.open(StringIO.StringIO(image_data))
pixel_data = list(im.getdata())
random.shuffle(pixel_data)
scrambled_image = im.copy()
scrambled_image.putdata(pixel_data)
output_image_io = StringIO.StringIO()
scrambled_image.save(output_image_io, im.format)
output_image_data = output_image_io.getvalue()
output_image_data.encode(encoding='base64')
response = copy.deepcopy(response)
response.set_data(output_image_data)
except Exception:
pass
return response
class DetailedHTTPResponse(httplib.HTTPResponse):
"""Preserve details relevant to replaying responses.
WARNING: This code uses attributes and methods of HTTPResponse
that are not part of the public interface.
"""
def read_chunks(self):
"""Return the response body content and timing data.
The returned chunks have the chunk size and CRLFs stripped off.
If the response was compressed, the returned data is still compressed.
Returns:
(chunks, delays)
chunks:
[response_body] # non-chunked responses
[chunk_1, chunk_2, ...] # chunked responses
delays:
[0] # non-chunked responses
[chunk_1_first_byte_delay, ...] # chunked responses
The delay for the first body item should be recorded by the caller.
"""
buf = []
chunks = []
delays = []
if not self.chunked:
chunks.append(self.read())
delays.append(0)
else:
start = TIMER()
try:
while True:
line = self.fp.readline()
chunk_size = self._read_chunk_size(line)
if chunk_size is None:
raise httplib.IncompleteRead(''.join(chunks))
if chunk_size == 0:
break
delays.append(TIMER() - start)
chunks.append(self._safe_read(chunk_size))
self._safe_read(2) # skip the CRLF at the end of the chunk
start = TIMER()
# Ignore any trailers.
while True:
line = self.fp.readline()
if not line or line == '\r\n':
break
finally:
self.close()
return chunks, delays
@classmethod
def _read_chunk_size(cls, line):
chunk_extensions_pos = line.find(';')
if chunk_extensions_pos != -1:
line = line[:chunk_extensions_pos] # strip chunk-extensions
try:
chunk_size = int(line, 16)
except ValueError:
return None
return chunk_size
class DetailedHTTPConnection(httplib.HTTPConnection):
"""Preserve details relevant to replaying connections."""
response_class = DetailedHTTPResponse
class DetailedHTTPSResponse(DetailedHTTPResponse):
"""Preserve details relevant to replaying SSL responses."""
pass
class DetailedHTTPSConnection(httplib.HTTPSConnection):
"""Preserve details relevant to replaying SSL connections."""
response_class = DetailedHTTPSResponse
def __init__(self, host, port):
# https://www.python.org/dev/peps/pep-0476/#opting-out
if hasattr(ssl, '_create_unverified_context'):
httplib.HTTPSConnection.__init__(
self, host=host, port=port, context=ssl._create_unverified_context())
else:
httplib.HTTPSConnection.__init__(self, host=host, port=port)
class RealHttpFetch(object):
def __init__(self, real_dns_lookup):
"""Initialize RealHttpFetch.
Args:
real_dns_lookup: a function that resolves a host to an IP. RealHttpFetch
will resolve host name to the IP before making fetching request if this
is not None.
"""
self._real_dns_lookup = real_dns_lookup
@staticmethod
def _GetHeaderNameValue(header):
"""Parse the header line and return a name/value tuple.
Args:
header: a string for a header such as "Content-Length: 314".
Returns:
A tuple (header_name, header_value) on success or None if the header
is not in expected format. header_name is in lowercase.
"""
i = header.find(':')
if i > 0:
return (header[:i].lower(), header[i+1:].strip())
return None
@staticmethod
def _ToTuples(headers):
"""Parse headers and save them to a list of tuples.
This method takes HttpResponse.msg.headers as input and convert it
to a list of (header_name, header_value) tuples.
HttpResponse.msg.headers is a list of strings where each string
represents either a header or a continuation line of a header.
1. a normal header consists of two parts which are separated by colon :
"header_name:header_value..."
2. a continuation line is a string starting with whitespace
"[whitespace]continued_header_value..."
If a header is not in good shape or an unexpected continuation line is
seen, it will be ignored.
Should avoid using response.getheaders() directly
because response.getheaders() can't handle multiple headers
with the same name properly. Instead, parse the
response.msg.headers using this method to get all headers.
Args:
headers: an instance of HttpResponse.msg.headers.
Returns:
A list of tuples which looks like:
[(header_name, header_value), (header_name2, header_value2)...]
"""
all_headers = []
for line in headers:
if line[0] in '\t ':
if not all_headers:
logging.warning(
'Unexpected response header continuation line [%s]', line)
continue
name, value = all_headers.pop()
value += '\n ' + line.strip()
else:
name_value = RealHttpFetch._GetHeaderNameValue(line)
if not name_value:
logging.warning(
'Response header in wrong format [%s]', line)
continue
name, value = name_value # pylint: disable=unpacking-non-sequence
all_headers.append((name, value))
return all_headers
@staticmethod
def _get_request_host_port(request):
host_parts = request.host.split(':')
host = host_parts[0]
port = int(host_parts[1]) if len(host_parts) == 2 else None
return host, port
@staticmethod
def _get_system_proxy(is_ssl):
return platformsettings.get_system_proxy(is_ssl)
def _get_connection(self, request_host, request_port, is_ssl):
"""Return a detailed connection object for host/port pair.
If a system proxy is defined (see platformsettings.py), it will be used.
Args:
request_host: a host string (e.g. "www.example.com").
request_port: a port integer (e.g. 8080) or None (for the default port).
is_ssl: True if HTTPS connection is needed.
Returns:
A DetailedHTTPSConnection or DetailedHTTPConnection instance.
"""
connection_host = request_host
connection_port = request_port
system_proxy = self._get_system_proxy(is_ssl)
if system_proxy:
connection_host = system_proxy.host
connection_port = system_proxy.port
# Use an IP address because WPR may override DNS settings.
if self._real_dns_lookup:
connection_ip = self._real_dns_lookup(connection_host)
if not connection_ip:
logging.critical(
'Unable to find IP for host name: %s', connection_host)
return None
connection_host = connection_ip
if is_ssl:
connection = DetailedHTTPSConnection(connection_host, connection_port)
if system_proxy:
connection.set_tunnel(request_host, request_port)
else:
connection = DetailedHTTPConnection(connection_host, connection_port)
return connection
def __call__(self, request):
"""Fetch an HTTP request.
Args:
request: an ArchivedHttpRequest
Returns:
an ArchivedHttpResponse
"""
logging.debug('RealHttpFetch: %s %s', request.host, request.full_path)
request_host, request_port = self._get_request_host_port(request)
retries = 3
while True:
try:
request_time = datetime.datetime.utcnow()
connection = self._get_connection(
request_host, request_port, request.is_ssl)
connect_start = TIMER()
connection.connect()
connect_delay = int((TIMER() - connect_start) * 1000)
start = TIMER()
connection.request(
request.command,
request.full_path,
request.request_body,
request.headers)
response = connection.getresponse()
headers_delay = int((TIMER() - start) * 1000)
chunks, chunk_delays = response.read_chunks()
delays = {
'connect': connect_delay,
'headers': headers_delay,
'data': chunk_delays
}
archived_http_response = httparchive.ArchivedHttpResponse(
response.version,
response.status,
response.reason,
RealHttpFetch._ToTuples(response.msg.headers),
chunks,
delays,
request_time)
return archived_http_response
except Exception, e:
if retries:
retries -= 1
logging.warning('Retrying fetch %s: %s', request, repr(e))
continue
logging.critical('Could not fetch %s: %s', request, repr(e))
return None
class RecordHttpArchiveFetch(object):
"""Make real HTTP fetches and save responses in the given HttpArchive."""
def __init__(self, http_archive, injector):
"""Initialize RecordHttpArchiveFetch.
Args:
http_archive: an instance of a HttpArchive
injector: script injector to inject scripts in all pages
"""
self.http_archive = http_archive
# Do not resolve host name to IP when recording to avoid SSL3 handshake
# failure.
# See https://github.com/chromium/web-page-replay/issues/73 for details.
self.real_http_fetch = RealHttpFetch(real_dns_lookup=None)
self.injector = injector
def __call__(self, request):
"""Fetch the request and return the response.
Args:
request: an ArchivedHttpRequest.
Returns:
an ArchivedHttpResponse
"""
# If request is already in the archive, return the archived response.
if request in self.http_archive:
logging.debug('Repeated request found: %s', request)
response = self.http_archive[request]
else:
response = self.real_http_fetch(request)
if response is None:
return None
self.http_archive[request] = response
if self.injector:
response = _InjectScripts(response, self.injector)
logging.debug('Recorded: %s', request)
return response
class ReplayHttpArchiveFetch(object):
"""Serve responses from the given HttpArchive."""
def __init__(self, http_archive, real_dns_lookup, injector,
use_diff_on_unknown_requests=False,
use_closest_match=False, scramble_images=False):
"""Initialize ReplayHttpArchiveFetch.
Args:
http_archive: an instance of a HttpArchive
real_dns_lookup: a function that resolves a host to an IP.
injector: script injector to inject scripts in all pages
use_diff_on_unknown_requests: If True, log unknown requests
with a diff to requests that look similar.
use_closest_match: If True, on replay mode, serve the closest match
in the archive instead of giving a 404.
"""
self.http_archive = http_archive
self.injector = injector
self.use_diff_on_unknown_requests = use_diff_on_unknown_requests
self.use_closest_match = use_closest_match
self.scramble_images = scramble_images
self.real_http_fetch = RealHttpFetch(real_dns_lookup)
def __call__(self, request):
"""Fetch the request and return the response.
Args:
request: an instance of an ArchivedHttpRequest.
Returns:
Instance of ArchivedHttpResponse (if found) or None
"""
if request.host.startswith('127.0.0.1:'):
return self.real_http_fetch(request)
response = self.http_archive.get(request)
if self.use_closest_match and not response:
closest_request = self.http_archive.find_closest_request(
request, use_path=True)
if closest_request:
response = self.http_archive.get(closest_request)
if response:
logging.info('Request not found: %s\nUsing closest match: %s',
request, closest_request)
if not response:
reason = str(request)
if self.use_diff_on_unknown_requests:
diff = self.http_archive.diff(request)
if diff:
reason += (
"\nNearest request diff "
"('-' for archived request, '+' for current request):\n%s" % diff)
logging.warning('Could not replay: %s', reason)
else:
if self.injector:
response = _InjectScripts(response, self.injector)
if self.scramble_images:
response = _ScrambleImages(response)
return response
class ControllableHttpArchiveFetch(object):
"""Controllable fetch function that can swap between record and replay."""
def __init__(self, http_archive, real_dns_lookup,
injector, use_diff_on_unknown_requests,
use_record_mode, use_closest_match, scramble_images):
"""Initialize HttpArchiveFetch.
Args:
http_archive: an instance of a HttpArchive
real_dns_lookup: a function that resolves a host to an IP.
injector: function to inject scripts in all pages.
takes recording time as datetime.datetime object.
use_diff_on_unknown_requests: If True, log unknown requests
with a diff to requests that look similar.
use_record_mode: If True, start in server in record mode.
use_closest_match: If True, on replay mode, serve the closest match
in the archive instead of giving a 404.
"""
self.http_archive = http_archive
self.record_fetch = RecordHttpArchiveFetch(http_archive, injector)
self.replay_fetch = ReplayHttpArchiveFetch(
http_archive, real_dns_lookup, injector,
use_diff_on_unknown_requests, use_closest_match, scramble_images)
if use_record_mode:
self.SetRecordMode()
else:
self.SetReplayMode()
def SetRecordMode(self):
self.fetch = self.record_fetch
self.is_record_mode = True
def SetReplayMode(self):
self.fetch = self.replay_fetch
self.is_record_mode = False
def __call__(self, *args, **kwargs):
"""Forward calls to Replay/Record fetch functions depending on mode."""
return self.fetch(*args, **kwargs)