File: test_pipeline_media.py

package info (click to toggle)
python-scrapy 2.13.3-1
  • links: PTS, VCS
  • area: main
  • in suites: forky, sid
  • size: 5,664 kB
  • sloc: python: 52,028; xml: 199; makefile: 25; sh: 7
file content (533 lines) | stat: -rw-r--r-- 20,165 bytes parent folder | download
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
from __future__ import annotations

import warnings

import pytest
from testfixtures import LogCapture
from twisted.internet import reactor
from twisted.internet.defer import Deferred, inlineCallbacks
from twisted.python.failure import Failure
from twisted.trial import unittest

from scrapy import signals
from scrapy.http import Request, Response
from scrapy.http.request import NO_CALLBACK
from scrapy.pipelines.files import FileException
from scrapy.pipelines.media import MediaPipeline
from scrapy.spiders import Spider
from scrapy.utils.log import failure_to_exc_info
from scrapy.utils.signal import disconnect_all
from scrapy.utils.test import get_crawler


def _mocked_download_func(request, info):
    assert request.callback is NO_CALLBACK
    response = request.meta.get("response")
    return response() if callable(response) else response


class UserDefinedPipeline(MediaPipeline):
    def media_to_download(self, request, info, *, item=None):
        pass

    def get_media_requests(self, item, info):
        pass

    def media_downloaded(self, response, request, info, *, item=None):
        return {}

    def media_failed(self, failure, request, info):
        return failure

    def file_path(self, request, response=None, info=None, *, item=None):
        return ""


class TestBaseMediaPipeline(unittest.TestCase):
    pipeline_class = UserDefinedPipeline
    settings = None

    def setUp(self):
        spider_cls = Spider
        self.spider = spider_cls("media.com")
        crawler = get_crawler(spider_cls, self.settings)
        self.pipe = self.pipeline_class.from_crawler(crawler)
        self.pipe.download_func = _mocked_download_func
        self.pipe.open_spider(self.spider)
        self.info = self.pipe.spiderinfo
        self.fingerprint = crawler.request_fingerprinter.fingerprint

    def tearDown(self):
        for name, signal in vars(signals).items():
            if not name.startswith("_"):
                disconnect_all(signal)

    def test_modify_media_request(self):
        request = Request("http://url")
        self.pipe._modify_media_request(request)
        assert request.meta == {"handle_httpstatus_all": True}

    def test_should_remove_req_res_references_before_caching_the_results(self):
        """Regression test case to prevent a memory leak in the Media Pipeline.

        The memory leak is triggered when an exception is raised when a Response
        scheduled by the Media Pipeline is being returned. For example, when a
        FileException('download-error') is raised because the Response status
        code is not 200 OK.

        It happens because we are keeping a reference to the Response object
        inside the FileException context. This is caused by the way Twisted
        return values from inline callbacks. It raises a custom exception
        encapsulating the original return value.

        The solution is to remove the exception context when this context is a
        _DefGen_Return instance, the BaseException used by Twisted to pass the
        returned value from those inline callbacks.

        Maybe there's a better and more reliable way to test the case described
        here, but it would be more complicated and involve running - or at least
        mocking - some async steps from the Media Pipeline. The current test
        case is simple and detects the problem very fast. On the other hand, it
        would not detect another kind of leak happening due to old object
        references being kept inside the Media Pipeline cache.

        This problem does not occur in Python 2.7 since we don't have Exception
        Chaining (https://www.python.org/dev/peps/pep-3134/).
        """
        # Create sample pair of Request and Response objects
        request = Request("http://url")
        response = Response("http://url", body=b"", request=request)

        # Simulate the Media Pipeline behavior to produce a Twisted Failure
        try:
            # Simulate a Twisted inline callback returning a Response
            raise StopIteration(response)
        except StopIteration as exc:
            def_gen_return_exc = exc
            try:
                # Simulate the media_downloaded callback raising a FileException
                # This usually happens when the status code is not 200 OK
                raise FileException("download-error")
            except Exception as exc:
                file_exc = exc
                # Simulate Twisted capturing the FileException
                # It encapsulates the exception inside a Twisted Failure
                failure = Failure(file_exc)

        # The Failure should encapsulate a FileException ...
        assert failure.value == file_exc
        # ... and it should have the StopIteration exception set as its context
        assert failure.value.__context__ == def_gen_return_exc

        # Let's calculate the request fingerprint and fake some runtime data...
        fp = self.fingerprint(request)
        info = self.pipe.spiderinfo
        info.downloading.add(fp)
        info.waiting[fp] = []

        # When calling the method that caches the Request's result ...
        self.pipe._cache_result_and_execute_waiters(failure, fp, info)
        # ... it should store the Twisted Failure ...
        assert info.downloaded[fp] == failure
        # ... encapsulating the original FileException ...
        assert info.downloaded[fp].value == file_exc
        # ... but it should not store the StopIteration exception on its context
        context = getattr(info.downloaded[fp].value, "__context__", None)
        assert context is None

    def test_default_item_completed(self):
        item = {"name": "name"}
        assert self.pipe.item_completed([], item, self.info) is item

        # Check that failures are logged by default
        fail = Failure(Exception())
        results = [(True, 1), (False, fail)]

        with LogCapture() as log:
            new_item = self.pipe.item_completed(results, item, self.info)

        assert new_item is item
        assert len(log.records) == 1
        record = log.records[0]
        assert record.levelname == "ERROR"
        assert record.exc_info == failure_to_exc_info(fail)

        # disable failure logging and check again
        self.pipe.LOG_FAILED_RESULTS = False
        with LogCapture() as log:
            new_item = self.pipe.item_completed(results, item, self.info)
        assert new_item is item
        assert len(log.records) == 0

    @inlineCallbacks
    def test_default_process_item(self):
        item = {"name": "name"}
        new_item = yield self.pipe.process_item(item, self.spider)
        assert new_item is item


class MockedMediaPipeline(UserDefinedPipeline):
    def __init__(self, *args, crawler=None, **kwargs):
        super().__init__(*args, crawler=crawler, **kwargs)
        self._mockcalled = []

    def download(self, request, info):
        self._mockcalled.append("download")
        return super().download(request, info)

    def media_to_download(self, request, info, *, item=None):
        self._mockcalled.append("media_to_download")
        if "result" in request.meta:
            return request.meta.get("result")
        return super().media_to_download(request, info)

    def get_media_requests(self, item, info):
        self._mockcalled.append("get_media_requests")
        return item.get("requests")

    def media_downloaded(self, response, request, info, *, item=None):
        self._mockcalled.append("media_downloaded")
        return super().media_downloaded(response, request, info)

    def media_failed(self, failure, request, info):
        self._mockcalled.append("media_failed")
        return super().media_failed(failure, request, info)

    def item_completed(self, results, item, info):
        self._mockcalled.append("item_completed")
        item = super().item_completed(results, item, info)
        item["results"] = results
        return item


class TestMediaPipeline(TestBaseMediaPipeline):
    pipeline_class = MockedMediaPipeline

    def _errback(self, result):
        self.pipe._mockcalled.append("request_errback")
        return result

    @inlineCallbacks
    def test_result_succeed(self):
        rsp = Response("http://url1")
        req = Request(
            "http://url1",
            meta={"response": rsp},
            errback=self._errback,
        )
        item = {"requests": req}
        new_item = yield self.pipe.process_item(item, self.spider)
        assert new_item["results"] == [(True, {})]
        assert self.pipe._mockcalled == [
            "get_media_requests",
            "media_to_download",
            "media_downloaded",
            "item_completed",
        ]

    @inlineCallbacks
    def test_result_failure(self):
        self.pipe.LOG_FAILED_RESULTS = False
        fail = Failure(Exception())
        req = Request(
            "http://url1",
            meta={"response": fail},
            errback=self._errback,
        )
        item = {"requests": req}
        new_item = yield self.pipe.process_item(item, self.spider)
        assert new_item["results"] == [(False, fail)]
        assert self.pipe._mockcalled == [
            "get_media_requests",
            "media_to_download",
            "media_failed",
            "request_errback",
            "item_completed",
        ]

    @inlineCallbacks
    def test_mix_of_success_and_failure(self):
        self.pipe.LOG_FAILED_RESULTS = False
        rsp1 = Response("http://url1")
        req1 = Request("http://url1", meta={"response": rsp1})
        fail = Failure(Exception())
        req2 = Request("http://url2", meta={"response": fail})
        item = {"requests": [req1, req2]}
        new_item = yield self.pipe.process_item(item, self.spider)
        assert new_item["results"] == [(True, {}), (False, fail)]
        m = self.pipe._mockcalled
        # only once
        assert m[0] == "get_media_requests"  # first hook called
        assert m.count("get_media_requests") == 1
        assert m.count("item_completed") == 1
        assert m[-1] == "item_completed"  # last hook called
        # twice, one per request
        assert m.count("media_to_download") == 2
        # one to handle success and other for failure
        assert m.count("media_downloaded") == 1
        assert m.count("media_failed") == 1

    @inlineCallbacks
    def test_get_media_requests(self):
        # returns single Request (without callback)
        req = Request("http://url")
        item = {"requests": req}  # pass a single item
        new_item = yield self.pipe.process_item(item, self.spider)
        assert new_item is item
        assert self.fingerprint(req) in self.info.downloaded

        # returns iterable of Requests
        req1 = Request("http://url1")
        req2 = Request("http://url2")
        item = {"requests": iter([req1, req2])}
        new_item = yield self.pipe.process_item(item, self.spider)
        assert new_item is item
        assert self.fingerprint(req1) in self.info.downloaded
        assert self.fingerprint(req2) in self.info.downloaded

    @inlineCallbacks
    def test_results_are_cached_across_multiple_items(self):
        rsp1 = Response("http://url1")
        req1 = Request("http://url1", meta={"response": rsp1})
        item = {"requests": req1}
        new_item = yield self.pipe.process_item(item, self.spider)
        assert new_item is item
        assert new_item["results"] == [(True, {})]

        # rsp2 is ignored, rsp1 must be in results because request fingerprints are the same
        req2 = Request(
            req1.url, meta={"response": Response("http://donot.download.me")}
        )
        item = {"requests": req2}
        new_item = yield self.pipe.process_item(item, self.spider)
        assert new_item is item
        assert self.fingerprint(req1) == self.fingerprint(req2)
        assert new_item["results"] == [(True, {})]

    @inlineCallbacks
    def test_results_are_cached_for_requests_of_single_item(self):
        rsp1 = Response("http://url1")
        req1 = Request("http://url1", meta={"response": rsp1})
        req2 = Request(
            req1.url, meta={"response": Response("http://donot.download.me")}
        )
        item = {"requests": [req1, req2]}
        new_item = yield self.pipe.process_item(item, self.spider)
        assert new_item is item
        assert new_item["results"] == [(True, {}), (True, {})]

    @inlineCallbacks
    def test_wait_if_request_is_downloading(self):
        def _check_downloading(response):
            fp = self.fingerprint(req1)
            assert fp in self.info.downloading
            assert fp in self.info.waiting
            assert fp not in self.info.downloaded
            assert len(self.info.waiting[fp]) == 2
            return response

        rsp1 = Response("http://url")

        def rsp1_func():
            dfd = Deferred().addCallback(_check_downloading)
            reactor.callLater(0.1, dfd.callback, rsp1)
            return dfd

        def rsp2_func():
            pytest.fail("it must cache rsp1 result and must not try to redownload")

        req1 = Request("http://url", meta={"response": rsp1_func})
        req2 = Request(req1.url, meta={"response": rsp2_func})
        item = {"requests": [req1, req2]}
        new_item = yield self.pipe.process_item(item, self.spider)
        assert new_item["results"] == [(True, {}), (True, {})]

    @inlineCallbacks
    def test_use_media_to_download_result(self):
        req = Request("http://url", meta={"result": "ITSME", "response": self.fail})
        item = {"requests": req}
        new_item = yield self.pipe.process_item(item, self.spider)
        assert new_item["results"] == [(True, "ITSME")]
        assert self.pipe._mockcalled == [
            "get_media_requests",
            "media_to_download",
            "item_completed",
        ]

    def test_key_for_pipe(self):
        assert (
            self.pipe._key_for_pipe("IMAGES", base_class_name="MediaPipeline")
            == "MOCKEDMEDIAPIPELINE_IMAGES"
        )


class TestMediaPipelineAllowRedirectSettings:
    def _assert_request_no3xx(self, pipeline_class, settings):
        pipe = pipeline_class(crawler=get_crawler(None, settings))
        request = Request("http://url")
        pipe._modify_media_request(request)

        assert "handle_httpstatus_list" in request.meta
        for status, check in [
            (200, True),
            # These are the status codes we want
            # the downloader to handle itself
            (301, False),
            (302, False),
            (302, False),
            (307, False),
            (308, False),
            # we still want to get 4xx and 5xx
            (400, True),
            (404, True),
            (500, True),
        ]:
            if check:
                assert status in request.meta["handle_httpstatus_list"]
            else:
                assert status not in request.meta["handle_httpstatus_list"]

    def test_subclass_standard_setting(self):
        self._assert_request_no3xx(UserDefinedPipeline, {"MEDIA_ALLOW_REDIRECTS": True})

    def test_subclass_specific_setting(self):
        self._assert_request_no3xx(
            UserDefinedPipeline, {"USERDEFINEDPIPELINE_MEDIA_ALLOW_REDIRECTS": True}
        )


class TestBuildFromCrawler:
    def setup_method(self):
        self.crawler = get_crawler(None, {"FILES_STORE": "/foo"})

    def test_simple(self):
        class Pipeline(UserDefinedPipeline):
            pass

        with warnings.catch_warnings(record=True) as w:
            pipe = Pipeline.from_crawler(self.crawler)
            assert pipe.crawler == self.crawler
            assert pipe._fingerprinter
            assert len(w) == 0

    def test_has_old_init(self):
        class Pipeline(UserDefinedPipeline):
            def __init__(self):
                super().__init__()
                self._init_called = True

        with warnings.catch_warnings(record=True) as w:
            pipe = Pipeline.from_crawler(self.crawler)
            assert pipe.crawler == self.crawler
            assert pipe._fingerprinter
            assert len(w) == 2
            assert pipe._init_called

    def test_has_from_settings(self):
        class Pipeline(UserDefinedPipeline):
            _from_settings_called = False

            @classmethod
            def from_settings(cls, settings):
                o = cls()
                o._from_settings_called = True
                return o

        with warnings.catch_warnings(record=True) as w:
            pipe = Pipeline.from_crawler(self.crawler)
            assert pipe.crawler == self.crawler
            assert pipe._fingerprinter
            assert len(w) == 2
            assert pipe._from_settings_called

    def test_has_from_settings_and_from_crawler(self):
        class Pipeline(UserDefinedPipeline):
            _from_settings_called = False
            _from_crawler_called = False

            @classmethod
            def from_settings(cls, settings):
                o = cls()
                o._from_settings_called = True
                return o

            @classmethod
            def from_crawler(cls, crawler):
                o = super().from_crawler(crawler)
                o._from_crawler_called = True
                return o

        with warnings.catch_warnings(record=True) as w:
            pipe = Pipeline.from_crawler(self.crawler)
            assert pipe.crawler == self.crawler
            assert pipe._fingerprinter
            assert len(w) == 2
            assert pipe._from_settings_called
            assert pipe._from_crawler_called

    def test_has_from_settings_and_init(self):
        class Pipeline(UserDefinedPipeline):
            _from_settings_called = False

            def __init__(self, store_uri, settings):
                super().__init__()
                self._init_called = True

            @classmethod
            def from_settings(cls, settings):
                store_uri = settings["FILES_STORE"]
                o = cls(store_uri, settings=settings)
                o._from_settings_called = True
                return o

        with warnings.catch_warnings(record=True) as w:
            pipe = Pipeline.from_crawler(self.crawler)
            assert pipe.crawler == self.crawler
            assert pipe._fingerprinter
            assert len(w) == 2
            assert pipe._from_settings_called
            assert pipe._init_called

    def test_has_from_crawler_and_init(self):
        class Pipeline(UserDefinedPipeline):
            _from_crawler_called = False

            def __init__(self, store_uri, settings, *, crawler):
                super().__init__(crawler=crawler)
                self._init_called = True

            @classmethod
            def from_crawler(cls, crawler):
                settings = crawler.settings
                store_uri = settings["FILES_STORE"]
                o = cls(store_uri, settings=settings, crawler=crawler)
                o._from_crawler_called = True
                return o

        with warnings.catch_warnings(record=True) as w:
            pipe = Pipeline.from_crawler(self.crawler)
            assert pipe.crawler == self.crawler
            assert pipe._fingerprinter
            assert len(w) == 0
            assert pipe._from_crawler_called
            assert pipe._init_called

    def test_has_from_crawler(self):
        class Pipeline(UserDefinedPipeline):
            _from_crawler_called = False

            @classmethod
            def from_crawler(cls, crawler):
                settings = crawler.settings
                o = super().from_crawler(crawler)
                o._from_crawler_called = True
                o.store_uri = settings["FILES_STORE"]
                return o

        with warnings.catch_warnings(record=True) as w:
            pipe = Pipeline.from_crawler(self.crawler)
            # this and the next assert will fail as MediaPipeline.from_crawler() wasn't called
            assert pipe.crawler == self.crawler
            assert pipe._fingerprinter
            assert len(w) == 0
            assert pipe._from_crawler_called