]>
Commit | Line | Data |
---|---|---|
cc52de43 | 1 | #!/usr/bin/env python3 |
54007a45 | 2 | |
44a5f171 PH |
3 | # Allow direct execution |
4 | import os | |
5 | import sys | |
6 | import unittest | |
f8271158 | 7 | |
44a5f171 PH |
8 | sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__)))) |
9 | ||
ac668111 | 10 | |
f2e8dbcc | 11 | import collections |
ac668111 | 12 | import hashlib |
54007a45 | 13 | import http.client |
ac668111 | 14 | import json |
15 | import socket | |
16 | import urllib.error | |
ac668111 | 17 | |
dd508b7c | 18 | from test.helper import ( |
0990305d | 19 | assertGreaterEqual, |
060ac762 | 20 | expect_info_dict, |
70b7e3fb | 21 | expect_warnings, |
dd508b7c | 22 | get_params, |
ff14fc49 | 23 | gettestcases, |
f2e8dbcc | 24 | getwebpagetestcases, |
060ac762 | 25 | is_download_test, |
257cfebf | 26 | report_warning, |
060ac762 | 27 | try_rm, |
dd508b7c | 28 | ) |
54007a45 | 29 | |
ac668111 | 30 | import yt_dlp.YoutubeDL # isort: split |
f8271158 | 31 | from yt_dlp.extractor import get_info_extractor |
7a5c1cfe | 32 | from yt_dlp.utils import ( |
44a5f171 PH |
33 | DownloadError, |
34 | ExtractorError, | |
35 | UnavailableVideoError, | |
f8271158 | 36 | format_bytes, |
f2e8dbcc | 37 | join_nonempty, |
44a5f171 | 38 | ) |
fd5ff020 | 39 | |
8cc83b8d FV |
40 | RETRIES = 3 |
41 | ||
5f6a1245 | 42 | |
7a5c1cfe | 43 | class YoutubeDL(yt_dlp.YoutubeDL): |
fd5ff020 | 44 | def __init__(self, *args, **kwargs): |
fd5ff020 | 45 | self.to_stderr = self.to_screen |
0eaf520d | 46 | self.processed_info_dicts = [] |
86e5f3ed | 47 | super().__init__(*args, **kwargs) |
5f6a1245 | 48 | |
f0500bd1 | 49 | def report_warning(self, message, *args, **kwargs): |
be95cac1 FV |
50 | # Don't accept warnings during tests |
51 | raise ExtractorError(message) | |
5f6a1245 | 52 | |
0eaf520d | 53 | def process_info(self, info_dict): |
f46e2f9d | 54 | self.processed_info_dicts.append(info_dict.copy()) |
86e5f3ed | 55 | return super().process_info(info_dict) |
1535ac2a | 56 | |
5f6a1245 | 57 | |
fd5ff020 FV |
58 | def _file_md5(fn): |
59 | with open(fn, 'rb') as f: | |
60 | return hashlib.md5(f.read()).hexdigest() | |
61 | ||
582be358 | 62 | |
f2e8dbcc | 63 | normal_test_cases = gettestcases() |
64 | webpage_test_cases = getwebpagetestcases() | |
65 | tests_counter = collections.defaultdict(collections.Counter) | |
6b47c7f2 | 66 | |
0eaf520d | 67 | |
060ac762 | 68 | @is_download_test |
1535ac2a | 69 | class TestDownload(unittest.TestCase): |
8936f68a YCH |
70 | # Parallel testing in nosetests. See |
71 | # http://nose.readthedocs.org/en/latest/doc_tests/test_multiprocess/multiprocess.html | |
72 | _multiprocess_shared_ = True | |
73 | ||
744435f2 | 74 | maxDiff = None |
5f6a1245 | 75 | |
243c57cf | 76 | COMPLETED_TESTS = {} |
77 | ||
c6c22e98 JH |
78 | def __str__(self): |
79 | """Identify each test with the `add_ie` attribute, if available.""" | |
f2e8dbcc | 80 | cls, add_ie = type(self), getattr(self, self._testMethodName).add_ie |
81 | return f'{self._testMethodName} ({cls.__module__}.{cls.__name__}){f" [{add_ie}]" if add_ie else ""}:' | |
c6c22e98 | 82 | |
fd5ff020 | 83 | |
5f6a1245 JW |
84 | # Dynamically generate tests |
85 | ||
8936f68a | 86 | def generator(test_case, tname): |
1535ac2a | 87 | def test_template(self): |
243c57cf | 88 | if self.COMPLETED_TESTS.get(tname): |
89 | return | |
90 | self.COMPLETED_TESTS[tname] = True | |
7a5c1cfe | 91 | ie = yt_dlp.extractor.get_info_extractor(test_case['name'])() |
655c4100 | 92 | other_ies = [get_info_extractor(ie_key)() for ie_key in test_case.get('add_ie', [])] |
e8ee972c PH |
93 | is_playlist = any(k.startswith('playlist') for k in test_case) |
94 | test_cases = test_case.get( | |
95 | 'playlist', [] if is_playlist else [test_case]) | |
96 | ||
bc2884af JMF |
97 | def print_skipping(reason): |
98 | print('Skipping %s: %s' % (test_case['name'], reason)) | |
6d1b3489 | 99 | self.skipTest(reason) |
100 | ||
9ee2b5f6 | 101 | if not ie.working(): |
bc2884af | 102 | print_skipping('IE marked as not _WORKING') |
e8ee972c PH |
103 | |
104 | for tc in test_cases: | |
105 | info_dict = tc.get('info_dict', {}) | |
0855702f | 106 | params = tc.get('params', {}) |
107 | if not info_dict.get('id'): | |
d7118397 | 108 | raise Exception(f'Test {tname} definition incorrect - "id" key is not present') |
0855702f | 109 | elif not info_dict.get('ext'): |
110 | if params.get('skip_download') and params.get('ignore_no_formats_error'): | |
111 | continue | |
d7118397 | 112 | raise Exception(f'Test {tname} definition incorrect - "ext" key must be present to define the output file') |
e8ee972c | 113 | |
fd5ff020 | 114 | if 'skip' in test_case: |
bc2884af | 115 | print_skipping(test_case['skip']) |
6d1b3489 | 116 | |
9ee2b5f6 JMF |
117 | for other_ie in other_ies: |
118 | if not other_ie.working(): | |
e075a44a | 119 | print_skipping('test depends on %sIE, marked as not WORKING' % other_ie.ie_key()) |
0eaf520d | 120 | |
44a5f171 | 121 | params = get_params(test_case.get('params', {})) |
8936f68a | 122 | params['outtmpl'] = tname + '_' + params['outtmpl'] |
e8ee972c | 123 | if is_playlist and 'playlist' not in test_case: |
65d49afa | 124 | params.setdefault('extract_flat', 'in_playlist') |
6911e11e | 125 | params.setdefault('playlistend', test_case.get('playlist_mincount')) |
e8ee972c | 126 | params.setdefault('skip_download', True) |
0eaf520d | 127 | |
ac35c266 | 128 | ydl = YoutubeDL(params, auto_init=False) |
023fa8c4 | 129 | ydl.add_default_info_extractors() |
bffbd5f0 | 130 | finished_hook_called = set() |
5f6a1245 | 131 | |
bffbd5f0 PH |
132 | def _hook(status): |
133 | if status['status'] == 'finished': | |
134 | finished_hook_called.add(status['filename']) | |
933605d7 | 135 | ydl.add_progress_hook(_hook) |
70b7e3fb | 136 | expect_warnings(ydl, test_case.get('expected_warnings', [])) |
5c892b0b | 137 | |
702665c0 | 138 | def get_tc_filename(tc): |
ad3dc496 | 139 | return ydl.prepare_filename(dict(tc.get('info_dict', {}))) |
702665c0 | 140 | |
28570840 | 141 | res_dict = None |
5f6a1245 | 142 | |
28570840 PH |
143 | def try_rm_tcs_files(tcs=None): |
144 | if tcs is None: | |
145 | tcs = test_cases | |
146 | for tc in tcs: | |
702665c0 JMF |
147 | tc_filename = get_tc_filename(tc) |
148 | try_rm(tc_filename) | |
149 | try_rm(tc_filename + '.part') | |
4eb92208 | 150 | try_rm(os.path.splitext(tc_filename)[0] + '.info.json') |
702665c0 | 151 | try_rm_tcs_files() |
5c892b0b | 152 | try: |
dd508b7c FV |
153 | try_num = 1 |
154 | while True: | |
8cc83b8d | 155 | try: |
3bef10a5 | 156 | # We're not using .download here since that is just a shim |
e8ee972c PH |
157 | # for outside error handling, and returns the exit code |
158 | # instead of the result dict. | |
308cfe0a S |
159 | res_dict = ydl.extract_info( |
160 | test_case['url'], | |
161 | force_generic_extractor=params.get('force_generic_extractor', False)) | |
8cc83b8d | 162 | except (DownloadError, ExtractorError) as err: |
8cc83b8d | 163 | # Check if the exception is not a network related one |
d7118397 | 164 | if (err.exc_info[0] not in (urllib.error.URLError, socket.timeout, UnavailableVideoError, http.client.BadStatusLine) |
165 | or (err.exc_info[0] == urllib.error.HTTPError and err.exc_info[1].code == 503)): | |
166 | err.msg = f'{getattr(err, "msg", err)} ({tname})' | |
8cc83b8d FV |
167 | raise |
168 | ||
dd508b7c | 169 | if try_num == RETRIES: |
8936f68a | 170 | report_warning('%s failed due to network errors, skipping...' % tname) |
dd508b7c FV |
171 | return |
172 | ||
86e5f3ed | 173 | print(f'Retrying: {try_num} failed tries\n\n##########\n\n') |
dd508b7c FV |
174 | |
175 | try_num += 1 | |
8cc83b8d FV |
176 | else: |
177 | break | |
5c892b0b | 178 | |
e8ee972c | 179 | if is_playlist: |
880ee801 | 180 | self.assertTrue(res_dict['_type'] in ['playlist', 'multi_video']) |
d6e6a422 | 181 | self.assertTrue('entries' in res_dict) |
f74b341d | 182 | expect_info_dict(self, res_dict, test_case.get('info_dict', {})) |
d6e6a422 | 183 | |
e8ee972c | 184 | if 'playlist_mincount' in test_case: |
0990305d PH |
185 | assertGreaterEqual( |
186 | self, | |
e8ee972c PH |
187 | len(res_dict['entries']), |
188 | test_case['playlist_mincount'], | |
189 | 'Expected at least %d in playlist %s, but got only %d' % ( | |
190 | test_case['playlist_mincount'], test_case['url'], | |
191 | len(res_dict['entries']))) | |
829476b8 PH |
192 | if 'playlist_count' in test_case: |
193 | self.assertEqual( | |
194 | len(res_dict['entries']), | |
195 | test_case['playlist_count'], | |
28570840 | 196 | 'Expected %d entries in playlist %s, but got %d.' % ( |
22a6f150 | 197 | test_case['playlist_count'], |
28570840 | 198 | test_case['url'], |
22a6f150 PH |
199 | len(res_dict['entries']), |
200 | )) | |
28570840 PH |
201 | if 'playlist_duration_sum' in test_case: |
202 | got_duration = sum(e['duration'] for e in res_dict['entries']) | |
203 | self.assertEqual( | |
204 | test_case['playlist_duration_sum'], got_duration) | |
e8ee972c | 205 | |
364a69e8 S |
206 | # Generalize both playlists and single videos to unified format for |
207 | # simplicity | |
208 | if 'entries' not in res_dict: | |
209 | res_dict['entries'] = [res_dict] | |
210 | ||
80b2fdf9 | 211 | for tc_num, tc in enumerate(test_cases): |
364a69e8 S |
212 | tc_res_dict = res_dict['entries'][tc_num] |
213 | # First, check test cases' data against extracted data alone | |
80b2fdf9 | 214 | expect_info_dict(self, tc_res_dict, tc.get('info_dict', {})) |
364a69e8 | 215 | # Now, check downloaded file consistency |
702665c0 | 216 | tc_filename = get_tc_filename(tc) |
511eda8e | 217 | if not test_case.get('params', {}).get('skip_download', False): |
702665c0 JMF |
218 | self.assertTrue(os.path.exists(tc_filename), msg='Missing file ' + tc_filename) |
219 | self.assertTrue(tc_filename in finished_hook_called) | |
08a36c35 S |
220 | expected_minsize = tc.get('file_minsize', 10000) |
221 | if expected_minsize is not None: | |
222 | if params.get('test'): | |
223 | expected_minsize = max(expected_minsize, 10000) | |
224 | got_fsize = os.path.getsize(tc_filename) | |
225 | assertGreaterEqual( | |
226 | self, got_fsize, expected_minsize, | |
227 | 'Expected %s to be at least %s, but it\'s only %s ' % | |
228 | (tc_filename, format_bytes(expected_minsize), | |
229 | format_bytes(got_fsize))) | |
230 | if 'md5' in tc: | |
231 | md5_for_file = _file_md5(tc_filename) | |
374560f0 | 232 | self.assertEqual(tc['md5'], md5_for_file) |
364a69e8 S |
233 | # Finally, check test cases' data again but this time against |
234 | # extracted data from info JSON file written during processing | |
4eb92208 | 235 | info_json_fn = os.path.splitext(tc_filename)[0] + '.info.json' |
f744c0f3 PH |
236 | self.assertTrue( |
237 | os.path.exists(info_json_fn), | |
238 | 'Missing info file %s' % info_json_fn) | |
86e5f3ed | 239 | with open(info_json_fn, encoding='utf-8') as infof: |
5c892b0b | 240 | info_dict = json.load(infof) |
f74b341d | 241 | expect_info_dict(self, info_dict, tc.get('info_dict', {})) |
5c892b0b | 242 | finally: |
702665c0 | 243 | try_rm_tcs_files() |
d6e6a422 | 244 | if is_playlist and res_dict is not None and res_dict.get('entries'): |
28570840 PH |
245 | # Remove all other files that may have been extracted if the |
246 | # extractor returns full results even with extract_flat | |
247 | res_tcs = [{'info_dict': e} for e in res_dict['entries']] | |
248 | try_rm_tcs_files(res_tcs) | |
fd5ff020 | 249 | |
1535ac2a | 250 | return test_template |
fd5ff020 | 251 | |
582be358 | 252 | |
5f6a1245 | 253 | # And add them to TestDownload |
f2e8dbcc | 254 | def inject_tests(test_cases, label=''): |
255 | for test_case in test_cases: | |
256 | name = test_case['name'] | |
257 | tname = join_nonempty('test', name, label, tests_counter[name][label], delim='_') | |
258 | tests_counter[name][label] += 1 | |
cdab8aa3 | 259 | |
f2e8dbcc | 260 | test_method = generator(test_case, tname) |
261 | test_method.__name__ = tname | |
262 | test_method.add_ie = ','.join(test_case.get('add_ie', [])) | |
263 | setattr(TestDownload, test_method.__name__, test_method) | |
cdab8aa3 | 264 | |
243c57cf | 265 | |
f2e8dbcc | 266 | inject_tests(normal_test_cases) |
267 | ||
268 | # TODO: disable redirection to the IE to ensure we are actually testing the webpage extraction | |
269 | inject_tests(webpage_test_cases, 'webpage') | |
270 | ||
271 | ||
272 | def batch_generator(name): | |
243c57cf | 273 | def test_template(self): |
f2e8dbcc | 274 | for label, num_tests in tests_counter[name].items(): |
275 | for i in range(num_tests): | |
276 | test_name = join_nonempty('test', name, label, i, delim='_') | |
277 | try: | |
278 | getattr(self, test_name)() | |
279 | except unittest.SkipTest: | |
280 | print(f'Skipped {test_name}') | |
243c57cf | 281 | |
282 | return test_template | |
283 | ||
284 | ||
f2e8dbcc | 285 | for name in tests_counter: |
286 | test_method = batch_generator(name) | |
243c57cf | 287 | test_method.__name__ = f'test_{name}_all' |
288 | test_method.add_ie = '' | |
289 | setattr(TestDownload, test_method.__name__, test_method) | |
f2e8dbcc | 290 | del test_method |
243c57cf | 291 | |
292 | ||
cdab8aa3 PH |
293 | if __name__ == '__main__': |
294 | unittest.main() |