summaryrefslogtreecommitdiff
path: root/crocoite/browser.py
blob: e58ebcf2b22d72cf639893d33e9bf05c997c19f0 (plain)
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
# Copyright (c) 2017 crocoite contributors
# 
# Permission is hereby granted, free of charge, to any person obtaining a copy
# of this software and associated documentation files (the "Software"), to deal
# in the Software without restriction, including without limitation the rights
# to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
# copies of the Software, and to permit persons to whom the Software is
# furnished to do so, subject to the following conditions:
# 
# The above copyright notice and this permission notice shall be included in
# all copies or substantial portions of the Software.
# 
# THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
# IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
# FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
# AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
# LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
# OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
# THE SOFTWARE.

"""
Chrome browser interactions.
"""

import logging
from urllib.parse import urlsplit
from base64 import b64decode
import pychrome

class Item:
    """
    Simple wrapper containing Chrome request and response
    """

    def __init__ (self, tab):
        self.tab = tab
        self.chromeRequest = None
        self.chromeResponse = None
        self.chromeFinished = None

    def __repr__ (self):
        return '<Item {}>'.format (self.request['url'])

    @property
    def request (self):
        return self.chromeRequest['request']

    @property
    def response (self):
        return self.chromeResponse['response']

    @property
    def initiator (self):
        return self.chromeRequest['initiator']

    @property
    def id (self):
        return self.chromeRequest['requestId']

    @property
    def encodedDataLength (self):
        return self.chromeFinished['encodedDataLength']

    @property
    def url (self):
        return self.response['url']

    @property
    def parsedUrl (self):
        return urlsplit (self.url)

    @property
    def body (self):
        """ Return response body or None """
        try:
            body = self.tab.Network.getResponseBody (requestId=self.id, _timeout=60)
            rawBody = body['body']
            base64Encoded = body['base64Encoded']
            if base64Encoded:
                rawBody = b64decode (rawBody)
            else:
                rawBody = rawBody.encode ('utf8')
            return rawBody, base64Encoded
        except (pychrome.exceptions.CallMethodException, pychrome.exceptions.TimeoutException):
            raise ValueError ('Cannot fetch response body')

    @property
    def requestBody (self):
        """ Get request/POST body """
        req = self.request
        postData = req.get ('postData')
        if postData:
            return postData.encode ('utf8'), False
        elif req.get ('hasPostData', False):
            try:
                return b64decode (self.tab.Network.getRequestPostData (requestId=self.id, _timeout=60)['postData']), True
            except (pychrome.exceptions.CallMethodException, pychrome.exceptions.TimeoutException):
                raise ValueError ('Cannot fetch request body')
        return None, False

    @property
    def requestHeaders (self):
        # the response object may contain refined headers, which were
        # *actually* sent over the wire
        return self._unfoldHeaders (self.response.get ('requestHeaders', self.request['headers']))

    @property
    def responseHeaders (self):
        return self._unfoldHeaders (self.response['headers'])

    @staticmethod
    def _unfoldHeaders (headers):
        """
        A host may send multiple headers using the same key, which Chrome folds
        into the same item. Separate those.
        """
        items = []
        for k in headers.keys ():
            for v in headers[k].split ('\n'):
                items.append ((k, v))
        return items

    def setRequest (self, req):
        self.chromeRequest = req

    def setResponse (self, resp):
        self.chromeResponse = resp

    def setFinished (self, finished):
        self.chromeFinished = finished

class SiteLoader:
    """
    Load site in Chrome and monitor network requests

    XXX: track popup windows/new tabs and close them
    """

    allowedSchemes = {'http', 'https'}

    def __init__ (self, browser, url, logger=logging.getLogger(__name__)):
        self.requests = {}
        self.browser = browser
        self.url = url
        self.logger = logger

        self.tab = browser.new_tab()

    def __enter__ (self):
        tab = self.tab
        # setup callbacks
        tab.Network.requestWillBeSent = self._requestWillBeSent
        tab.Network.responseReceived = self._responseReceived
        tab.Network.loadingFinished = self._loadingFinished
        tab.Network.loadingFailed = self._loadingFailed
        tab.Log.entryAdded = self._entryAdded
        #tab.Page.loadEventFired = loadEventFired
        tab.Page.javascriptDialogOpening = self._javascriptDialogOpening

        # start the tab
        tab.start()

        # enable events
        tab.Log.enable ()
        tab.Network.enable()
        tab.Page.enable ()
        tab.Network.clearBrowserCache ()
        if tab.Network.canClearBrowserCookies ()['result']:
            tab.Network.clearBrowserCookies ()

        return self

    def __len__ (self):
        return len (self.requests)

    def start (self):
        self.tab.Page.navigate(url=self.url)

    def wait (self, timeout=1):
        self.tab.wait (timeout)

    def waitIdle (self, idleTimeout=1, maxTimeout=60):
        step = 0
        for i in range (0, maxTimeout):
            self.wait (1)
            if len (self) == 0:
                step += 1
                if step > idleTimeout:
                    break
            else:
                step = 0

    def stop (self):
        """
        Stop loading site

        XXX: stop executing scripts
        """

        tab = self.tab

        tab.Page.stopLoading ()
        tab.Network.disable ()
        tab.Page.disable ()
        tab.Log.disable ()
        # XXX: we can’t drain the event queue directly, so insert (yet another) wait
        tab.wait (1)
        tab.Network.requestWillBeSent = None
        tab.Network.responseReceived = None
        tab.Network.loadingFinished = None
        tab.Network.loadingFailed = None
        tab.Page.loadEventFired = None
        tab.Page.javascriptDialogOpening = None
        tab.Log.entryAdded = None

    def __exit__ (self, exc_type, exc_value, traceback):
        self.tab.stop ()
        self.browser.close_tab(self.tab)
        return False

    # overrideable callbacks
    def loadingFinished (self, item, redirect=False):
        pass

    def loadingFailed (self, item):
        pass

    # internal chrome callbacks
    def _requestWillBeSent (self, **kwargs):
        reqId = kwargs['requestId']
        req = kwargs['request']

        url = urlsplit (req['url'])
        if url.scheme not in self.allowedSchemes:
            return

        item = self.requests.get (reqId)
        if item:
            # redirects never “finish” loading, but yield another requestWillBeSent with this key set
            redirectResp = kwargs.get ('redirectResponse')
            if redirectResp:
                # create fake responses
                resp = {'requestId': reqId, 'response': redirectResp, 'timestamp': kwargs['timestamp']}
                item.setResponse (resp)
                resp = {'requestId': reqId, 'encodedDataLength': 0, 'timestamp': kwargs['timestamp']}
                item.setFinished (resp)
                self.loadingFinished (item, redirect=True)
                self.logger.info ('redirected request {} has url {}'.format (reqId, req['url']))
            else:
                self.logger.warning ('request {} already exists, overwriting.'.format (reqId))

        item = Item (self.tab)
        item.setRequest (kwargs)
        self.requests[reqId] = item

    def _responseReceived (self, **kwargs):
        reqId = kwargs['requestId']
        item = self.requests.get (reqId)
        if item is None:
            return

        resp = kwargs['response']
        url = urlsplit (resp['url'])
        if url.scheme in self.allowedSchemes:
            self.logger.info ('response {} {}'.format (reqId, resp['url']))
            item.setResponse (kwargs)
        else:
            self.logger.warning ('response: ignoring scheme {}'.format (url.scheme))

    def _loadingFinished (self, **kwargs):
        """
        Item was fully loaded. For some items the request body is not available
        when responseReceived is fired, thus move everything here.
        """
        reqId = kwargs['requestId']
        item = self.requests.pop (reqId, None)
        if item is None:
            # we never recorded this request (blacklisted scheme, for example)
            return
        req = item.request
        resp = item.response
        assert req['url'] == resp['url'], 'req and resp urls are not the same {} vs {}'.format (req['url'], resp['url'])
        url = urlsplit (resp['url'])
        if url.scheme in self.allowedSchemes:
            self.logger.info ('finished {} {}'.format (reqId, req['url']))
            item.setFinished (kwargs)
            self.loadingFinished (item)

    def _loadingFailed (self, **kwargs):
        reqId = kwargs['requestId']
        self.logger.warning ('failed {} {}'.format (reqId, kwargs['errorText'], kwargs.get ('blockedReason')))
        item = self.requests.pop (reqId, None)
        self.loadingFailed (item)

    def _entryAdded (self, **kwargs):
        """ Log entry added """
        entry = kwargs['entry']
        level = {'verbose': logging.DEBUG, 'info': logging.INFO,
                'warning': logging.WARNING,
                'error': logging.ERROR}[entry['level']]
        self.logger.log (level, 'console: {}: {}'.format (entry['source'], entry['text']), extra={'raw': entry})

    def _javascriptDialogOpening (self, **kwargs):
        t = kwargs.get ('type')
        if t in {'alert', 'confirm', 'prompt'}:
            self.logger.info ('javascript opened a dialog: {}, {}, canceling'.format (t, kwargs.get ('message')))
            self.tab.Page.handleJavaScriptDialog (accept=False)
        elif t == 'beforeunload':
            # we must accept this one, otherwise the page will not unload/close
            self.logger.info ('javascript opened a dialog: {}, {}, procceeding'.format (t, kwargs.get ('message')))
            self.tab.Page.handleJavaScriptDialog (accept=True)
        else:
            self.logger.warning ('unknown javascript dialog type {}'.format (t))

class AccountingSiteLoader (SiteLoader):
    """
    SiteLoader that keeps basic statistics about retrieved pages.
    """

    def __init__ (self, browser, url, logger=logging.getLogger(__name__)):
        super ().__init__ (browser, url, logger)

        self.stats = {'requests': 0, 'finished': 0, 'failed': 0, 'bytesRcv': 0}

    def loadingFinished (self, item, redirect=False):
        super ().loadingFinished (item, redirect)

        self.stats['finished'] += 1
        self.stats['bytesRcv'] += item.encodedDataLength

    def loadingFailed (self, item):
        super ().loadingFailed (item)

        self.stats['failed'] += 1

    def _requestWillBeSent (self, **kwargs):
        super ()._requestWillBeSent (**kwargs)

        self.stats['requests'] += 1

import subprocess
from tempfile import mkdtemp
import socket, shutil

class ChromeService:
    """
    Start Chrome with socket activation (i.e. pass listening socket). Polling
    is not required with this method, since reads will block until Chrome is
    ready.
    """

    def __init__ (self, binary='google-chrome-stable', host='localhost', port=9222, windowSize=(1920, 1080)):
        self.binary = binary
        self.host = host
        self.port = port
        self.windowSize = windowSize
        self.p = None

    def __enter__ (self):
        assert self.p is None

        port = self.port
        while True:
            s = socket.socket ()
            s.setsockopt (socket.SOL_SOCKET, socket.SO_REUSEADDR, 1)
            try:
                s.bind ((self.host, port))
                break
            except OSError:
                # try different port
                if port < 65000:
                    port += 1
                else:
                    raise
        s.listen (10)
        self.userDataDir = mkdtemp ()
        args = [self.binary,
                '--window-size={},{}'.format (*self.windowSize),
                '--user-data-dir={}'.format (self.userDataDir), # use temporory user dir
                '--no-default-browser-check',
                '--no-first-run', # don’t show first run screen
                '--disable-breakpad', # no error reports
                '--disable-extensions',
                '--disable-infobars',
                '--disable-notifications', # no libnotify
                '--headless',
                '--disable-gpu',
                '--hide-scrollbars', # hide scrollbars on screenshots
                '--mute-audio', # don’t play any audio
                '--remote-debugging-socket-fd={}'.format (s.fileno ()),
                '--homepage=about:blank',
                'about:blank']
        # start new session, so ^C does not affect subprocess
        self.p = subprocess.Popen (args, pass_fds=[s.fileno()], start_new_session=True,
                stdin=subprocess.DEVNULL, stdout=subprocess.DEVNULL,
                stderr=subprocess.DEVNULL)
        s.close ()

        return 'http://{}:{}'.format (self.host, port)

    def __exit__ (self, *exc):
        self.p.terminate ()
        self.p.wait ()
        shutil.rmtree (self.userDataDir)
        self.p = None

class NullService:
    def __init__ (self, url):
        self.url = url

    def __enter__ (self):
        return self.url

    def __exit__ (self, *exc):
        pass

### tests ###

import unittest, time
from http.server import BaseHTTPRequestHandler

class TestHTTPRequestHandler (BaseHTTPRequestHandler):
    encodingTestString = {
        'latin1': 'äöü',
        'utf-8': 'äöü',
        'ISO-8859-1': 'äöü',
        }
    binaryTestData = b'\x00\x01\x02'
    # 1×1 pixel PNG
    imageTestData = b'\x89PNG\r\n\x1a\n\x00\x00\x00\rIHDR\x00\x00\x00\x01\x00\x00\x00\x01\x08\x00\x00\x00\x00:~\x9bU\x00\x00\x00\nIDAT\x08\x1dc\xf8\x0f\x00\x01\x01\x01\x006_g\x80\x00\x00\x00\x00IEND\xaeB`\x82'
    htmlTestData = '<html><body><img src="/image"><img src="/nonexistent"></body></html>'
    alertData = '<html><body><script>window.addEventListener("beforeunload", function (e) { e.returnValue = "bye?"; return e.returnValue; }); alert("stopping here"); if (confirm("are you sure?") || prompt ("42?")) { window.location = "/nonexistent"; }</script><img src="/image"></body></html>'

    def do_GET(self):
        path = self.path
        if path.startswith ('/redirect/301'):
            self.send_response(301)
            self.send_header ('Location', path[13:])
            self.end_headers()
        elif path == '/empty':
            self.send_response (200)
            self.end_headers ()
        elif path.startswith ('/encoding'):
            # send text data with different encodings
            _, _, encoding = path.split ('/', 3)
            self.send_response (200)
            self.send_header ('Content-Type', 'text/plain; charset={}'.format (encoding))
            self.end_headers ()
            self.wfile.write (self.encodingTestString[encoding].encode (encoding))
        elif path == '/binary':
            # send binary data
            self.send_response (200)
            self.send_header ('Content-Type', 'application/octet-stream')
            self.send_header ('Content-Length', len (self.binaryTestData))
            self.end_headers ()
            self.wfile.write (self.binaryTestData)
        elif path == '/image':
            # send binary data
            self.send_response (200)
            self.send_header ('Content-Type', 'image/png')
            self.end_headers ()
            self.wfile.write (self.imageTestData)
        elif path == '/attachment':
            self.send_response (200)
            self.send_header ('Content-Type', 'text/plain; charset=utf-8')
            self.send_header ('Content-Disposition', 'attachment; filename="attachment.txt"')
            self.end_headers ()
            self.wfile.write (self.encodingTestString['utf-8'].encode ('utf-8'))
        elif path == '/html':
            self.send_response (200)
            self.send_header ('Content-Type', 'text/html; charset=utf-8')
            self.end_headers ()
            self.wfile.write (self.htmlTestData.encode ('utf-8'))
        elif path == '/alert':
            self.send_response (200)
            self.send_header ('Content-Type', 'text/html; charset=utf-8')
            self.end_headers ()
            self.wfile.write (self.alertData.encode ('utf-8'))
        else:
            self.send_response (404)
            self.end_headers ()

    def log_message (self, format, *args):
        pass

def startServer ():
    import http.server
    PORT = 8000
    httpd = http.server.HTTPServer (("localhost", PORT), TestHTTPRequestHandler)
    httpd.serve_forever()

class TestSiteLoaderAdapter (SiteLoader):
    def __init__ (self, browser, url):
        SiteLoader.__init__ (self, browser, url)
        self.finished = []

    def loadingFinished (self, item, redirect=False):
        self.finished.append (item)

class TestSiteLoader (unittest.TestCase):
    def setUp (self):
        from multiprocessing import Process
        self.server = Process (target=startServer)
        self.server.start ()
        self.baseurl = 'http://localhost:8000/'
        self.service = ChromeService ()
        browserUrl = self.service.__enter__ ()
        self.browser = pychrome.Browser(url=browserUrl)

    def buildAdapter (self, path):
        return TestSiteLoaderAdapter (self.browser, '{}{}'.format (self.baseurl, path))

    def assertUrls (self, l, expect):
        urls = set (map (lambda x: x.parsedUrl.path, l.finished))
        expect = set (expect)
        self.assertEqual (urls, expect)
        
    def test_wait (self):
        waittime = 2
        with self.buildAdapter ('empty') as l:
            l.start ()
            before = time.time ()
            l.wait (waittime)
            after = time.time ()
            self.assertTrue ((after-before) >= waittime)

    def test_empty (self):
        with self.buildAdapter ('empty') as l:
            l.start ()
            l.waitIdle ()
            self.assertEqual (len (l.finished), 1)

    def test_redirect301 (self):
        with self.buildAdapter ('redirect/301/empty') as l:
            l.start ()
            l.waitIdle ()
            self.assertEqual (len (l.finished), 2)
            self.assertUrls (l, ['/redirect/301/empty', '/empty'])
            for item in l.finished:
                if item.parsedUrl.path == '/empty':
                    self.assertEqual (item.response['status'], 200)
                    self.assertEqual (item.body[0], b'')
                elif item.parsedUrl.path == '/redirect/301/empty':
                    self.assertEqual (item.response['status'], 301)
                else:
                    self.fail ('unknown url')

    def test_redirect301multi (self):
        with self.buildAdapter ('redirect/301/redirect/301/empty') as l:
            l.start ()
            l.waitIdle ()
            self.assertEqual (len (l.finished), 3)
            self.assertUrls (l, ['/redirect/301/redirect/301/empty', '/redirect/301/empty', '/empty'])
            for item in l.finished:
                if item.parsedUrl.path == '/empty':
                    self.assertEqual (item.response['status'], 200)
                    self.assertEqual (item.body[0], b'')
                elif item.parsedUrl.path in {'/redirect/301/empty', \
                        '/redirect/301/redirect/301/empty'}:
                    self.assertEqual (item.response['status'], 301)
                else:
                    self.fail ('unknown url')

    def test_encoding (self):
        """ Text responses are transformed to UTF-8. Make sure this works
        correctly. """
        for encoding, expected in TestHTTPRequestHandler.encodingTestString.items ():
            with self.buildAdapter ('encoding/{}'.format (encoding)) as l:
                l.start ()
                l.waitIdle ()
                self.assertEqual (len (l.finished), 1)
                self.assertUrls (l, ['/encoding/{}'.format (encoding)])
                self.assertEqual (l.finished[0].body[0], expected.encode ('utf8'))

    def test_binary (self):
        """ Browser should ignore content it cannot display (i.e. octet-stream) """
        with self.buildAdapter ('binary') as l:
            l.start ()
            l.waitIdle ()
            self.assertEqual (len (l.finished), 0)

    def test_image (self):
        """ Images should be displayed inline """
        with self.buildAdapter ('image') as l:
            l.start ()
            l.waitIdle ()
            self.assertEqual (len (l.finished), 1)
            self.assertUrls (l, ['/image'])
            self.assertEqual (l.finished[0].body[0], TestHTTPRequestHandler.imageTestData)

    def test_attachment (self):
        """ And downloads won’t work in headless mode """
        with self.buildAdapter ('attachment') as l:
            l.start ()
            l.waitIdle ()
            self.assertEqual (len (l.finished), 0)

    def test_html (self):
        with self.buildAdapter ('html') as l:
            l.start ()
            l.waitIdle ()
            self.assertEqual (len (l.finished), 3)
            self.assertUrls (l, ['/html', '/image', '/nonexistent'])
            for item in l.finished:
                if item.parsedUrl.path == '/html':
                    self.assertEqual (item.response['status'], 200)
                    self.assertEqual (item.body[0], TestHTTPRequestHandler.htmlTestData.encode ('utf-8'))
                elif item.parsedUrl.path == '/image':
                    self.assertEqual (item.response['status'], 200)
                    self.assertEqual (item.body[0], TestHTTPRequestHandler.imageTestData)
                elif item.parsedUrl.path == '/nonexistent':
                    self.assertEqual (item.response['status'], 404)
                else:
                    self.fail ('unknown url')

    def test_alert (self):
        with self.buildAdapter ('alert') as l:
            l.start ()
            l.waitIdle ()
            self.assertUrls (l, ['/alert', '/image'])
            for item in l.finished:
                if item.parsedUrl.path == '/alert':
                    self.assertEqual (item.response['status'], 200)
                    self.assertEqual (item.body[0], TestHTTPRequestHandler.alertData.encode ('utf-8'))
                elif item.parsedUrl.path == '/image':
                    self.assertEqual (item.response['status'], 200)
                    self.assertEqual (item.body[0], TestHTTPRequestHandler.imageTestData)
                else:
                    self.fail ('unknown url')

    def tearDown (self):
        self.service.__exit__ (None, None, None)
        self.server.terminate ()
        self.server.join ()

if __name__ == '__main__':
    import sys
    if sys.argv[1] == 'server':
        startServer ()