-
Notifications
You must be signed in to change notification settings - Fork 514
/
Copy pathbrowser_tab.py
1138 lines (961 loc) · 43.1 KB
/
browser_tab.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
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
# -*- coding: utf-8 -*-
from __future__ import absolute_import
import base64
import functools
import os
import weakref
import uuid
from PyQt5.QtCore import QObject, QSize, Qt, QTimer, pyqtSlot
from PyQt5.QtNetwork import QNetworkRequest
from PyQt5.QtWebKitWidgets import QWebPage
from PyQt5.QtWebKit import QWebSettings
from twisted.internet import defer
from twisted.python import log
import six
from splash import defaults
from splash.har.qt import cookies2har
from splash.qtrender_image import QtImageRenderer
from splash.qtutils import (OPERATION_QT_CONSTANTS, WrappedSignal, qt2py,
qurl2ascii, to_qurl)
from splash.render_options import validate_size_str
from splash.qwebpage import SplashQWebPage, SplashQWebView, RenderErrorInfo
from splash.exceptions import JsError, OneShotCallbackError, ScriptError
from splash.utils import to_bytes
from splash.jsutils import (
get_sanitized_result_js,
SANITIZE_FUNC_JS,
get_process_errors_js,
escape_js,
)
def skip_if_closing(meth):
@functools.wraps(meth)
def wrapped(self, *args, **kwargs):
if self._closing:
self.logger.log("%s is not called because BrowserTab is closing" % meth.__name__, min_level=2)
return
return meth(self, *args, **kwargs)
return wrapped
class BrowserTab(QObject):
"""
An object for controlling a single browser tab (QWebView).
It is created by splash.pool.Pool. Pool attaches to tab's deferred
and waits until either a callback or an errback is called, then destroys
a BrowserTab.
XXX: are cookies shared between "browser tabs"? In real browsers they are,
but maybe this is not what we want.
"""
def __init__(self, network_manager, splash_proxy_factory, verbosity,
render_options, visible=False):
""" Create a new browser tab. """
QObject.__init__(self)
self.deferred = defer.Deferred()
self.network_manager = network_manager
self.verbosity = verbosity
self.visible = visible
self._uid = render_options.get_uid()
self._closing = False
self._closing_normally = False
self._active_timers = set()
self._timers_to_cancel_on_redirect = weakref.WeakKeyDictionary() # timer: callback
self._timers_to_cancel_on_error = weakref.WeakKeyDictionary() # timer: callback
self._callback_proxies_to_cancel = weakref.WeakSet()
self._js_console = None
self._autoload_scripts = []
self._is_unsupported_content = False
self._unsupported_content_reply = None
self._load_finished_after_unsupported_content_ready = False
self.logger = _BrowserTabLogger(uid=self._uid, verbosity=verbosity)
self._init_webpage(verbosity, network_manager, splash_proxy_factory,
render_options)
self.http_client = _SplashHttpClient(self.web_page)
def _init_webpage(self, verbosity, network_manager, splash_proxy_factory, render_options):
""" Create and initialize QWebPage and QWebView """
self.web_page = SplashQWebPage(verbosity)
self.web_page.setNetworkAccessManager(network_manager)
self.web_page.splash_proxy_factory = splash_proxy_factory
self.web_page.render_options = render_options
self._set_default_webpage_options(self.web_page)
self._setup_webpage_events()
self.web_view = SplashQWebView()
self.web_view.setPage(self.web_page)
self.web_view.setAttribute(Qt.WA_DeleteOnClose, True)
self.web_view.onBeforeClose = self._on_before_close
if self.visible:
self.web_view.move(0, 0)
self.web_view.show()
self.set_viewport(defaults.VIEWPORT_SIZE)
# XXX: hack to ensure that default window size is not 640x480.
self.web_view.resize(
QSize(*map(int, defaults.VIEWPORT_SIZE.split('x'))))
def set_js_enabled(self, val):
settings = self.web_page.settings()
settings.setAttribute(QWebSettings.JavascriptEnabled, val)
def get_js_enabled(self):
settings = self.web_page.settings()
return settings.testAttribute(QWebSettings.JavascriptEnabled)
def set_private_mode_enabled(self, val):
settings = self.web_page.settings()
settings.setAttribute(QWebSettings.PrivateBrowsingEnabled, bool(val))
settings.setAttribute(QWebSettings.LocalStorageEnabled, not bool(val))
def get_private_mode_enabled(self):
settings = self.web_page.settings()
return settings.testAttribute(QWebSettings.PrivateBrowsingEnabled)
def _set_default_webpage_options(self, web_page):
"""
Set QWebPage options.
TODO: allow to customize them.
"""
settings = web_page.settings()
settings.setAttribute(QWebSettings.JavascriptEnabled, True)
settings.setAttribute(QWebSettings.PluginsEnabled, False)
settings.setAttribute(QWebSettings.LocalContentCanAccessRemoteUrls, True)
scroll_bars = Qt.ScrollBarAsNeeded if self.visible else Qt.ScrollBarAlwaysOff
web_page.mainFrame().setScrollBarPolicy(Qt.Vertical, scroll_bars)
web_page.mainFrame().setScrollBarPolicy(Qt.Horizontal, scroll_bars)
if self.visible:
web_page.settings().setAttribute(QWebSettings.DeveloperExtrasEnabled, True)
def _setup_webpage_events(self):
self._load_finished = WrappedSignal(self.web_page.mainFrame().loadFinished)
self.web_page.mainFrame().loadFinished.connect(self._on_load_finished)
self.web_page.mainFrame().urlChanged.connect(self._on_url_changed)
self.web_page.mainFrame().javaScriptWindowObjectCleared.connect(self._on_javascript_window_object_cleared)
self.web_page.setForwardUnsupportedContent(True)
self.web_page.unsupportedContent.connect(self._on_unsupported_content)
self.logger.add_web_page(self.web_page)
def return_result(self, result):
""" Return a result to the Pool. """
if self._result_already_returned():
self.logger.log("error: result is already returned", min_level=1)
self.deferred.callback(result)
# self.deferred = None
def return_error(self, error):
""" Return an error to the Pool. """
if self._result_already_returned():
self.logger.log("error: result is already returned", min_level=1)
self.deferred.errback(error)
# self.deferred = None
def _result_already_returned(self):
""" Return True if an error or a result is already returned to Pool """
return self.deferred.called
def set_custom_headers(self, headers):
"""
Set custom HTTP headers to be sent with each request. Passed headers
are merged with QWebKit default headers, overwriting QWebKit values
in case of conflicts.
"""
self.web_page.custom_headers = headers
def set_resource_timeout(self, timeout):
""" Set a default timeout for HTTP requests, in seconds. """
self.web_page.resource_timeout = timeout
def get_resource_timeout(self):
""" Get a default timeout for HTTP requests, in seconds. """
return self.web_page.resource_timeout
def set_images_enabled(self, enabled):
self.web_page.settings().setAttribute(QWebSettings.AutoLoadImages,
enabled)
def get_images_enabled(self):
settings = self.web_page.settings()
return settings.testAttribute(QWebSettings.AutoLoadImages)
def set_viewport(self, size, raise_if_empty=False):
"""
Set viewport size.
If size is "full" viewport size is detected automatically.
If can also be "<width>x<height>".
.. note::
This will update all JS geometry variables, but window resize event
is delivered asynchronously and so ``window.resize`` will not be
invoked until control is yielded to the event loop.
"""
if size == 'full':
size = self.web_page.mainFrame().contentsSize()
self.logger.log("Contents size: %s" % size, min_level=2)
if size.isEmpty():
if raise_if_empty:
raise RuntimeError("Cannot detect viewport size")
else:
size = defaults.VIEWPORT_SIZE
self.logger.log("Viewport is empty, falling back to: %s" %
size)
if not isinstance(size, QSize):
validate_size_str(size)
w, h = map(int, size.split('x'))
size = QSize(w, h)
self.web_page.setViewportSize(size)
self._force_relayout()
w, h = int(size.width()), int(size.height())
self.logger.log("viewport size is set to %sx%s" % (w, h), min_level=2)
return w, h
def _force_relayout(self):
"""Force a relayout of the web page contents."""
# setPreferredContentsSize may be used to force a certain size for
# layout purposes. Passing an invalid size resets the override and
# tells the QWebPage to use the size as requested by the document.
# This is in fact the default behavior, so we don't change anything.
#
# The side-effect of this operation is a forced synchronous relayout of
# the page.
self.web_page.setPreferredContentsSize(QSize())
def lock_navigation(self):
self.web_page.navigation_locked = True
def unlock_navigation(self):
self.web_page.navigation_locked = False
def set_content(self, data, callback, errback, mime_type=None, baseurl=None):
"""
Set page contents to ``data``, then wait until page loads.
Invoke a callback if load was successful or errback if it wasn't.
"""
if mime_type is None:
mime_type = "text/html; charset=utf-8"
if baseurl is None:
baseurl = ''
callback_id = self._load_finished.connect(
self._on_content_ready,
callback=callback,
errback=errback,
)
self.logger.log("callback %s is connected to loadFinished" % callback_id, min_level=3)
self.web_page.mainFrame().setContent(data, mime_type, to_qurl(baseurl))
def set_user_agent(self, value):
""" Set User-Agent header for future requests """
self.http_client.set_user_agent(value)
def get_cookies(self):
""" Return a list of all cookies in the current cookiejar """
return cookies2har(self.network_manager.cookiejar.allCookies())
def init_cookies(self, cookies):
""" Replace all current cookies with ``cookies`` """
self.network_manager.cookiejar.init(cookies)
def clear_cookies(self):
""" Remove all cookies. Return a number of cookies deleted. """
return self.network_manager.cookiejar.clear()
def delete_cookies(self, name=None, url=None):
"""
Delete cookies with name == ``name``.
If ``url`` is not None then only those cookies are deleted wihch
are to be added when a request is sent to ``url``.
Return a number of cookies deleted.
"""
return self.network_manager.cookiejar.delete(name, url)
def add_cookie(self, cookie):
return self.network_manager.cookiejar.add(cookie)
@property
def url(self):
""" Current URL """
return six.text_type(self.web_page.mainFrame().url().toString())
def go(self, url, callback, errback, baseurl=None, http_method='GET',
body=None, headers=None):
"""
Go to an URL. This is similar to entering an URL in
address tab and pressing Enter.
"""
self.store_har_timing("_onStarted")
if body is not None:
body = to_bytes(body)
if baseurl:
# If baseurl is used, we download the page manually,
# then set its contents to the QWebPage and let it
# download related resources and render the result.
cb = functools.partial(
self._on_baseurl_request_finished,
callback=callback,
errback=errback,
baseurl=baseurl,
url=url,
)
self.http_client.request(url,
callback=cb,
method=http_method,
body=body,
headers=headers,
follow_redirects=True,
)
else:
# if not self._goto_callbacks.isempty():
# self.logger.log("Only a single concurrent 'go' request is supported. "
# "Previous go requests will be cancelled.", min_level=1)
# # When a new URL is loaded to mainFrame an errback will
# # be called, so we're not cancelling this callback manually.
callback_id = self._load_finished.connect(
self._on_content_ready,
callback=callback,
errback=errback,
)
self.logger.log("callback %s is connected to loadFinished" % callback_id, min_level=3)
self._load_url_to_mainframe(url, http_method, body, headers=headers)
def stop_loading(self):
"""
Stop loading of the current page and all pending page
refresh/redirect requests.
"""
self.logger.log("stop_loading", min_level=2)
self.web_view.pageAction(QWebPage.StopScheduledPageRefresh)
self.web_view.stop()
def register_callback(self, event, callback):
""" Register a callback for an event """
self.web_page.callbacks[event].append(callback)
def clear_callbacks(self, event):
""" Unregister all callbacks for an event """
del self.web_page.callbacks[event][:]
# def remove_callback(self, event, callback):
# """ Unregister a callback for an event """
# self.web_page.callbacks[event].remove(callback)
@skip_if_closing
def close(self):
""" Destroy this tab """
self.logger.log("close is requested by a script", min_level=2)
self._closing = True
self._closing_normally = True
self.web_view.pageAction(QWebPage.StopScheduledPageRefresh)
self.web_view.stop()
self.web_view.close()
self.web_page.deleteLater()
self.web_view.deleteLater()
self._cancel_all_timers()
def _on_before_close(self):
# self._closing = True
# self._cancel_all_timers()
# if not self._closing_normally:
# self.return_error(Exception("Window is closed by user"))
return True # don't close the window
@skip_if_closing
def _on_load_finished(self, ok):
"""
This callback is called for all web_page.mainFrame()
loadFinished events.
"""
if self._is_unsupported_content:
if self._unsupported_content_reply.isRunning():
# XXX: We'll come back later when download finishes
self.logger.log(
'Still receving unsupported content', min_level=3)
return
else:
self._load_finished_after_unsupported_content_ready = True
self.logger.log('Unsupported content received', min_level=3)
if self.web_page.maybe_redirect(ok):
self.logger.log("Redirect or other non-fatal error detected", min_level=2)
return
if self.web_page.is_ok(ok): # or maybe_redirect:
self.logger.log("loadFinished: ok", min_level=2)
else:
self._cancel_timers(self._timers_to_cancel_on_error)
if self.web_page.error_loading(ok):
self.logger.log("loadFinished: %s" % (str(self.web_page.error_info)), min_level=1)
else:
self.logger.log("loadFinished: unknown error", min_level=1)
def _on_baseurl_request_finished(self, callback, errback, baseurl, url):
"""
This method is called when ``baseurl`` is used and a
reply for the first request is received.
"""
self.logger.log("baseurl_request_finished", min_level=2)
reply = self.sender()
mime_type = reply.header(QNetworkRequest.ContentTypeHeader)
data = reply.readAll()
self.set_content(
data=data,
callback=callback,
errback=errback,
mime_type=mime_type,
baseurl=baseurl,
)
if reply.error():
self.logger.log("Error loading %s: %s" % (url, reply.errorString()), min_level=1)
def _load_url_to_mainframe(self, url, http_method, body=None, headers=None):
request = self.http_client.request_obj(url, headers=headers, body=body)
meth = OPERATION_QT_CONSTANTS[http_method]
if body is None: # PyQT doesn't support body=None
self.web_page.mainFrame().load(request, meth)
else:
assert isinstance(body, bytes)
self.web_page.mainFrame().load(request, meth, body)
@skip_if_closing
def _on_content_ready(self, ok, callback, errback, callback_id):
"""
This method is called when a QWebPage finishes loading its contents.
"""
if self._is_unsupported_content:
if self._unsupported_content_reply.isRunning():
# XXX: We'll come back later when download finishes
return
elif self.web_page.maybe_redirect(ok):
# XXX: It assumes loadFinished will be called again because
# redirect happens. If redirect is detected improperly,
# loadFinished won't be called again, and Splash will return
# the result only after a timeout.
return
self.logger.log("loadFinished: disconnecting callback %s" % callback_id, min_level=3)
self._load_finished.disconnect(callback_id)
if self.web_page.is_ok(ok):
callback()
elif self._is_unsupported_content:
# XXX: Error downloading unsupported content.
# `self.web_page.error_info` shall be `None` now
error_info = RenderErrorInfo(
'Network',
int(self._unsupported_content_reply.error()),
six.text_type(self._unsupported_content_reply.errorString()),
six.text_type(self._unsupported_content_reply.url().url())
)
errback(error_info)
elif self.web_page.error_loading(ok):
# XXX: maybe return a meaningful error page instead of generic
# error message?
errback(self.web_page.error_info)
else:
# XXX: it means ok=False. When does it happen?
errback(self.web_page.error_info)
def wait(self, time_ms, callback, onredirect=None, onerror=None):
"""
Wait for time_ms, then run callback.
If onredirect is True then the timer is cancelled if redirect happens.
If onredirect is callable then in case of redirect the timer is
cancelled and this callable is called.
If onerror is True then the timer is cancelled if a render error
happens. If onerror is callable then in case of a render error the
timer is cancelled and this callable is called.
"""
timer = QTimer()
timer.setSingleShot(True)
timer_callback = functools.partial(self._on_wait_timeout,
timer=timer,
callback=callback,
)
timer.timeout.connect(timer_callback)
self.logger.log("waiting %sms; timer %s" % (time_ms, id(timer)), min_level=2)
timer.start(time_ms)
self._active_timers.add(timer)
if onredirect:
self._timers_to_cancel_on_redirect[timer] = onredirect
if onerror:
self._timers_to_cancel_on_error[timer] = onerror
def _on_wait_timeout(self, timer, callback):
self.logger.log("wait timeout for %s" % id(timer), min_level=2)
if timer in self._active_timers:
self._active_timers.remove(timer)
self._timers_to_cancel_on_redirect.pop(timer, None)
self._timers_to_cancel_on_error.pop(timer, None)
callback()
def _cancel_timer(self, timer, errback=None):
self.logger.log("cancelling timer %s" % id(timer), min_level=2)
if timer in self._active_timers:
self._active_timers.remove(timer)
try:
timer.stop()
if callable(errback):
self.logger.log("calling timer errback", min_level=2)
errback(self.web_page.error_info)
finally:
timer.deleteLater()
def _cancel_timers(self, timers):
for timer, oncancel in list(timers.items()):
self._cancel_timer(timer, oncancel)
timers.pop(timer, None)
def _cancel_all_timers(self):
total_len = len(self._active_timers) + len(self._callback_proxies_to_cancel)
self.logger.log("cancelling %d remaining timers" % total_len, min_level=2)
for timer in list(self._active_timers):
self._cancel_timer(timer)
for callback_proxy in self._callback_proxies_to_cancel:
callback_proxy.use_up()
def _on_url_changed(self, url):
self.web_page.har.store_redirect(six.text_type(url.toString()))
self._cancel_timers(self._timers_to_cancel_on_redirect)
def _on_unsupported_content_finished(self):
self.logger.log('Unsupported content finished', min_level=3)
if not self._load_finished_after_unsupported_content_ready:
# XXX: The unsupported content reply might have finished before the
# original loadFinished signal emits. In such cases we do not want
# the same signal twice.
if not self._unsupported_content_reply.error():
self.web_page.mainFrame().loadFinished.emit(True)
else:
self.web_page.mainFrame().loadFinished.emit(False)
def _on_unsupported_content(self, reply):
self.logger.log('Unsupported content detected', min_level=3)
self._is_unsupported_content = True
self._unsupported_content_reply = reply
if reply.isFinished():
# Already finished. The content might be very short.
self.logger.log('Unsupported content already finished', min_level=3)
self._on_unsupported_content_finished()
else:
reply.finished.connect(self._on_unsupported_content_finished)
def run_js_file(self, filename, handle_errors=True):
"""
Load JS library from file ``filename`` to the current frame.
"""
with open(filename, 'rb') as f:
script = f.read().decode('utf-8')
self.runjs(script, handle_errors=handle_errors)
def run_js_files(self, folder, handle_errors=True):
"""
Load all JS libraries from ``folder`` folder to the current frame.
"""
for jsfile in os.listdir(folder):
if jsfile.endswith('.js'):
filename = os.path.join(folder, jsfile)
self.run_js_file(filename, handle_errors=handle_errors)
def autoload(self, js_source):
""" Execute JS code before each page load """
self._autoload_scripts.append(js_source)
def autoload_reset(self):
""" Remove all scripts scheduled for auto-loading """
self._autoload_scripts = []
def _on_javascript_window_object_cleared(self):
for script in self._autoload_scripts:
# XXX: handle_errors=False is used to execute autoload scripts
# in a global context (not inside a closure).
# One difference is how are `function foo(){}` statements handled:
# if executed globally, `foo` becomes an attribute of window;
# if executed in a closure, `foo` is a name local to this closure.
self.runjs(script, handle_errors=False)
def http_get(self, url, callback, headers=None, follow_redirects=True):
""" Send a GET request; call a callback with the reply as an argument. """
self.http_client.get(url,
callback=callback,
headers=headers,
follow_redirects=follow_redirects
)
def http_post(self, url, callback, headers=None, follow_redirects=True, body=None):
if body is not None:
body = to_bytes(body)
self.http_client.post(url,
callback=callback,
headers=headers,
follow_redirects=follow_redirects,
body=body)
def evaljs(self, js_source, handle_errors=True, result_protection=True):
"""
Run JS code in page context and return the result.
If JavaScript exception or an syntax error happens
and `handle_errors` is True then Python JsError
exception is raised.
When `result_protection` is True (default) protection against
badly written or malicious scripts is activated. Disable it
when the script result is known to be good, i.e. it only
contains objects/arrays/primitives without circular references.
"""
frame = self.web_page.mainFrame()
eval_expr = "eval({})".format(escape_js(js_source))
if result_protection:
eval_expr = get_sanitized_result_js(eval_expr)
if not handle_errors:
return qt2py(frame.evaluateJavaScript(eval_expr))
res = frame.evaluateJavaScript(get_process_errors_js(eval_expr))
if not isinstance(res, dict):
raise JsError({
'type': ScriptError.UNKNOWN_ERROR,
'js_error_message': res,
'message': "unknown JS error: {!r}".format(res)
})
if res.get("error", False):
err_message = res.get('errorMessage')
err_type = res.get('errorType', '<custom JS error>')
err_repr = res.get('errorRepr', '<unknown JS error>')
if err_message is None:
err_message = err_repr
raise JsError({
'type': ScriptError.JS_ERROR,
'js_error_type': err_type,
'js_error_message': err_message,
'js_error': err_repr,
'message': "JS error: {!r}".format(err_repr)
})
return res.get("result", None)
def runjs(self, js_source, handle_errors=True):
""" Run JS code in page context and discard the result. """
# If JS code returns something, and we just discard
# the result of frame.evaluateJavaScript, then Qt still needs to build
# a result - it could be costly. So the original JS code
# is adjusted to make sure it doesn't return anything.
self.evaljs(
js_source="%s;undefined" % js_source,
handle_errors=handle_errors,
result_protection=False
)
def wait_for_resume(self, js_source, callback, errback, timeout):
"""
Run some Javascript asynchronously.
The JavaScript must contain a method called `main()` that accepts
one argument. The first argument will be an object with `resume()`
and `error()` methods. The code _must_ call one of these functions
before the timeout or else it will be canceled.
"""
frame = self.web_page.mainFrame()
callback_proxy = OneShotCallbackProxy(self, callback, errback, timeout)
self._callback_proxies_to_cancel.add(callback_proxy)
frame.addToJavaScriptWindowObject(callback_proxy.name, callback_proxy)
wrapped = """
(function () {
try {
eval(%(script_text)s);
} catch (err) {
var main = function (splash) {
throw err;
}
}
(function () {
var sanitize = %(sanitize_func)s;
var _result = {};
var _splash = window["%(callback_name)s"];
var splash = {
'error': function (message) {
_splash.error(message, false);
},
'resume': function (value) {
_result['value'] = value;
try {
_splash.resume(sanitize(_result));
} catch (err) {
_splash.error(err, true);
}
},
'set': function (key, value) {
_result[key] = value;
}
};
delete window["%(callback_name)s"];
try {
if (typeof main === 'undefined') {
throw "wait_for_resume(): no main() function defined";
}
main(splash);
} catch (err) {
_splash.error(err, true);
}
})();
})();undefined
""" % dict(
sanitize_func=SANITIZE_FUNC_JS,
script_text=escape_js(js_source),
callback_name=callback_proxy.name
)
def cancel_callback():
callback_proxy.cancel(reason='javascript window object cleared')
self.logger.log("wait_for_resume wrapped script:\n%s" % wrapped,
min_level=3)
frame.javaScriptWindowObjectCleared.connect(cancel_callback)
frame.evaluateJavaScript(wrapped)
def store_har_timing(self, name):
self.logger.log("HAR event: %s" % name, min_level=3)
self.web_page.har.store_timing(name)
def _jsconsole_enable(self):
# TODO: add public interface or make console available by default
if self._js_console is not None:
return
self._js_console = _JavascriptConsole()
frame = self.web_page.mainFrame()
frame.addToJavaScriptWindowObject('console', self._js_console)
def _jsconsole_messages(self):
# TODO: add public interface or make console available by default
if self._js_console is None:
return []
return self._js_console.messages[:]
def html(self):
""" Return HTML of the current main frame """
self.logger.log("getting HTML", min_level=2)
frame = self.web_page.mainFrame()
result = frame.toHtml()
self.store_har_timing("_onHtmlRendered")
return result
def _get_image(self, image_format, width, height, render_all, scale_method):
old_size = self.web_page.viewportSize()
try:
if render_all:
self.logger.log("Rendering whole page contents (RENDER_ALL)",
min_level=2)
self.set_viewport('full')
renderer = QtImageRenderer(
self.web_page, self.logger, image_format,
width=width, height=height, scale_method=scale_method)
image = renderer.render_qwebpage()
finally:
if old_size != self.web_page.viewportSize():
# Let's not generate extra "set size" messages in the log.
self.web_page.setViewportSize(old_size)
self.store_har_timing("_onScreenshotPrepared")
return image
def png(self, width=None, height=None, b64=False, render_all=False,
scale_method=None):
""" Return screenshot in PNG format """
self.logger.log(
"Getting PNG: width=%s, height=%s, "
"render_all=%s, scale_method=%s" %
(width, height, render_all, scale_method), min_level=2)
image = self._get_image('PNG', width, height, render_all, scale_method)
result = image.to_png()
if b64:
result = base64.b64encode(result).decode('utf-8')
self.store_har_timing("_onPngRendered")
return result
def jpeg(self, width=None, height=None, b64=False, render_all=False,
scale_method=None, quality=None):
"""Return screenshot in JPEG format."""
self.logger.log(
"Getting JPEG: width=%s, height=%s, "
"render_all=%s, scale_method=%s, quality=%s" %
(width, height, render_all, scale_method, quality), min_level=2)
image = self._get_image('JPEG', width, height, render_all, scale_method)
result = image.to_jpeg(quality=quality)
if b64:
result = base64.b64encode(result).decode('utf-8')
self.store_har_timing("_onJpegRendered")
return result
def iframes_info(self, children=True, html=True):
""" Return information about all iframes """
self.logger.log("getting iframes", min_level=3)
frame = self.web_page.mainFrame()
result = self._frame_to_dict(frame, children, html)
self.store_har_timing("_onIframesRendered")
return result
def har(self, reset=False):
""" Return HAR information """
self.logger.log("getting HAR", min_level=3)
res = self.web_page.har.todict()
if reset:
self.har_reset()
return res
def har_reset(self):
""" Drop current HAR information """
self.logger.log("HAR information is reset", min_level=3)
return self.web_page.reset_har()
def history(self):
""" Return history of 'main' HTTP requests """
self.logger.log("getting history", min_level=3)
return self.web_page.har.get_history()
def last_http_status(self):
"""
Return HTTP status code of the currently loaded webpage
or None if it is not available.
"""
return self.web_page.har.get_last_http_status()
def _frame_to_dict(self, frame, children=True, html=True):
g = frame.geometry()
res = {
"url": six.text_type(frame.url().toString()),
"requestedUrl": six.text_type(frame.requestedUrl().toString()),
"geometry": (g.x(), g.y(), g.width(), g.height()),
"title": six.text_type(frame.title())
}
if html:
res["html"] = six.text_type(frame.toHtml())
if children:
res["childFrames"] = [
self._frame_to_dict(f, True, html)
for f in frame.childFrames()
]
res["frameName"] = six.text_type(frame.frameName())
return res
class _SplashHttpClient(QObject):
""" Wrapper class for making HTTP requests on behalf of a SplashQWebPage """
def __init__(self, web_page):
super(_SplashHttpClient, self).__init__()
self._replies = set()
self.web_page = web_page
self.network_manager = web_page.networkAccessManager()
def set_user_agent(self, value):
""" Set User-Agent header for future requests """
self.web_page.custom_user_agent = value
def request_obj(self, url, headers=None, body=None):
""" Return a QNetworkRequest object """
request = QNetworkRequest()
request.setUrl(to_qurl(url))
request.setOriginatingObject(self.web_page.mainFrame())
if headers is not None:
self.web_page.skip_custom_headers = True
self._set_request_headers(request, headers)
if body and not request.hasRawHeader(b"content-type"):
# there is POST body but no content-type
# QT will set this header, but it will complain so better to do this here
request.setRawHeader(b"content-type", b"application/x-www-form-urlencoded")
return request
def request(self, url, callback, method='GET', body=None,
headers=None, follow_redirects=True, max_redirects=5):
"""
Create a request and return a QNetworkReply object with callback
connected.
"""
cb = functools.partial(
self._on_request_finished,
callback=callback,
method=method,
body=body,
headers=headers,
follow_redirects=follow_redirects,
redirects_remaining=max_redirects,
)
return self._send_request(url, cb, method=method, body=body, headers=headers)
def get(self, url, callback, headers=None, follow_redirects=True):
""" Send a GET HTTP request; call the callback with the reply. """
cb = functools.partial(
self._return_reply,
callback=callback,
url=url,
)
self.request(url, cb, headers=headers, follow_redirects=follow_redirects)
def post(self, url, callback, headers=None, follow_redirects=True, body=None):
""" Send HTTP POST request;
"""
cb = functools.partial(self._return_reply, callback=callback, url=url)
self.request(url, cb, headers=headers, follow_redirects=follow_redirects, body=body,
method="POST")
def _send_request(self, url, callback, method='GET', body=None,
headers=None):
# XXX: The caller must ensure self._delete_reply is called in a callback.
if method.upper() not in ["POST", "GET"]:
raise NotImplementedError()
if body is not None:
assert isinstance(body, bytes)
request = self.request_obj(url, headers=headers, body=body)
if method.upper() == "POST":
reply = self.network_manager.post(request, body)
else:
reply = self.network_manager.get(request)
reply.finished.connect(callback)
self._replies.add(reply)
return reply
def _on_request_finished(self, callback, method, body, headers,
follow_redirects, redirects_remaining):
""" Handle redirects and call the callback. """
reply = self.sender()
try:
if not follow_redirects:
callback()
return
if not redirects_remaining:
callback() # XXX: should it be an error?
return
redirect_url = reply.attribute(QNetworkRequest.RedirectionTargetAttribute)
if redirect_url is None: # no redirect
callback()
return
# handle redirects after POST request
if method.upper() == "POST":
method = "GET"
body = None
redirect_url = reply.url().resolved(redirect_url)
self.request(
url=redirect_url,
callback=callback,
method=method,
body=body,
headers=headers,
follow_redirects=follow_redirects,
max_redirects=redirects_remaining-1,
)
finally:
self._delete_reply(reply)
def _return_reply(self, callback, url):
reply = self.sender()
callback(reply)
def _set_request_headers(self, request, headers):
""" Set HTTP headers for the request. """
if isinstance(headers, dict):
headers = headers.items()
for name, value in headers or []:
request.setRawHeader(to_bytes(name), to_bytes(value))
if name.lower() == 'user-agent':