]>
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 | |
11 | import hashlib | |
54007a45 | 12 | import http.client |
ac668111 | 13 | import json |
14 | import socket | |
15 | import urllib.error | |
ac668111 | 16 | |
dd508b7c | 17 | from test.helper import ( |
0990305d | 18 | assertGreaterEqual, |
060ac762 | 19 | expect_info_dict, |
70b7e3fb | 20 | expect_warnings, |
dd508b7c | 21 | get_params, |
ff14fc49 | 22 | gettestcases, |
060ac762 | 23 | is_download_test, |
257cfebf | 24 | report_warning, |
060ac762 | 25 | try_rm, |
dd508b7c | 26 | ) |
54007a45 | 27 | |
ac668111 | 28 | import yt_dlp.YoutubeDL # isort: split |
29 | from yt_dlp.compat import compat_HTTPError | |
f8271158 | 30 | from yt_dlp.extractor import get_info_extractor |
7a5c1cfe | 31 | from yt_dlp.utils import ( |
44a5f171 PH |
32 | DownloadError, |
33 | ExtractorError, | |
34 | UnavailableVideoError, | |
f8271158 | 35 | format_bytes, |
44a5f171 | 36 | ) |
fd5ff020 | 37 | |
8cc83b8d FV |
38 | RETRIES = 3 |
39 | ||
5f6a1245 | 40 | |
7a5c1cfe | 41 | class YoutubeDL(yt_dlp.YoutubeDL): |
fd5ff020 | 42 | def __init__(self, *args, **kwargs): |
fd5ff020 | 43 | self.to_stderr = self.to_screen |
0eaf520d | 44 | self.processed_info_dicts = [] |
86e5f3ed | 45 | super().__init__(*args, **kwargs) |
5f6a1245 | 46 | |
f0500bd1 | 47 | def report_warning(self, message, *args, **kwargs): |
be95cac1 FV |
48 | # Don't accept warnings during tests |
49 | raise ExtractorError(message) | |
5f6a1245 | 50 | |
0eaf520d | 51 | def process_info(self, info_dict): |
f46e2f9d | 52 | self.processed_info_dicts.append(info_dict.copy()) |
86e5f3ed | 53 | return super().process_info(info_dict) |
1535ac2a | 54 | |
5f6a1245 | 55 | |
fd5ff020 FV |
56 | def _file_md5(fn): |
57 | with open(fn, 'rb') as f: | |
58 | return hashlib.md5(f.read()).hexdigest() | |
59 | ||
582be358 | 60 | |
ff14fc49 | 61 | defs = gettestcases() |
6b47c7f2 | 62 | |
0eaf520d | 63 | |
060ac762 | 64 | @is_download_test |
1535ac2a | 65 | class TestDownload(unittest.TestCase): |
8936f68a YCH |
66 | # Parallel testing in nosetests. See |
67 | # http://nose.readthedocs.org/en/latest/doc_tests/test_multiprocess/multiprocess.html | |
68 | _multiprocess_shared_ = True | |
69 | ||
744435f2 | 70 | maxDiff = None |
5f6a1245 | 71 | |
243c57cf | 72 | COMPLETED_TESTS = {} |
73 | ||
c6c22e98 JH |
74 | def __str__(self): |
75 | """Identify each test with the `add_ie` attribute, if available.""" | |
76 | ||
77 | def strclass(cls): | |
78 | """From 2.7's unittest; 2.6 had _strclass so we can't import it.""" | |
86e5f3ed | 79 | return f'{cls.__module__}.{cls.__name__}' |
c6c22e98 JH |
80 | |
81 | add_ie = getattr(self, self._testMethodName).add_ie | |
82 | return '%s (%s)%s:' % (self._testMethodName, | |
83 | strclass(self.__class__), | |
84 | ' [%s]' % add_ie if add_ie else '') | |
85 | ||
fd5ff020 | 86 | def setUp(self): |
fd5ff020 FV |
87 | self.defs = defs |
88 | ||
5f6a1245 JW |
89 | # Dynamically generate tests |
90 | ||
91 | ||
8936f68a | 92 | def generator(test_case, tname): |
5d01a647 | 93 | |
1535ac2a | 94 | def test_template(self): |
243c57cf | 95 | if self.COMPLETED_TESTS.get(tname): |
96 | return | |
97 | self.COMPLETED_TESTS[tname] = True | |
7a5c1cfe | 98 | ie = yt_dlp.extractor.get_info_extractor(test_case['name'])() |
655c4100 | 99 | other_ies = [get_info_extractor(ie_key)() for ie_key in test_case.get('add_ie', [])] |
e8ee972c PH |
100 | is_playlist = any(k.startswith('playlist') for k in test_case) |
101 | test_cases = test_case.get( | |
102 | 'playlist', [] if is_playlist else [test_case]) | |
103 | ||
bc2884af JMF |
104 | def print_skipping(reason): |
105 | print('Skipping %s: %s' % (test_case['name'], reason)) | |
6d1b3489 | 106 | self.skipTest(reason) |
107 | ||
9ee2b5f6 | 108 | if not ie.working(): |
bc2884af | 109 | print_skipping('IE marked as not _WORKING') |
e8ee972c PH |
110 | |
111 | for tc in test_cases: | |
112 | info_dict = tc.get('info_dict', {}) | |
0855702f | 113 | params = tc.get('params', {}) |
114 | if not info_dict.get('id'): | |
115 | raise Exception('Test definition incorrect. \'id\' key is not present') | |
116 | elif not info_dict.get('ext'): | |
117 | if params.get('skip_download') and params.get('ignore_no_formats_error'): | |
118 | continue | |
119 | raise Exception('Test definition incorrect. The output file cannot be known. \'ext\' key is not present') | |
e8ee972c | 120 | |
fd5ff020 | 121 | if 'skip' in test_case: |
bc2884af | 122 | print_skipping(test_case['skip']) |
6d1b3489 | 123 | |
9ee2b5f6 JMF |
124 | for other_ie in other_ies: |
125 | if not other_ie.working(): | |
e075a44a | 126 | print_skipping('test depends on %sIE, marked as not WORKING' % other_ie.ie_key()) |
0eaf520d | 127 | |
44a5f171 | 128 | params = get_params(test_case.get('params', {})) |
8936f68a | 129 | params['outtmpl'] = tname + '_' + params['outtmpl'] |
e8ee972c | 130 | if is_playlist and 'playlist' not in test_case: |
65d49afa | 131 | params.setdefault('extract_flat', 'in_playlist') |
6911e11e | 132 | params.setdefault('playlistend', test_case.get('playlist_mincount')) |
e8ee972c | 133 | params.setdefault('skip_download', True) |
0eaf520d | 134 | |
ac35c266 | 135 | ydl = YoutubeDL(params, auto_init=False) |
023fa8c4 | 136 | ydl.add_default_info_extractors() |
bffbd5f0 | 137 | finished_hook_called = set() |
5f6a1245 | 138 | |
bffbd5f0 PH |
139 | def _hook(status): |
140 | if status['status'] == 'finished': | |
141 | finished_hook_called.add(status['filename']) | |
933605d7 | 142 | ydl.add_progress_hook(_hook) |
70b7e3fb | 143 | expect_warnings(ydl, test_case.get('expected_warnings', [])) |
5c892b0b | 144 | |
702665c0 | 145 | def get_tc_filename(tc): |
ad3dc496 | 146 | return ydl.prepare_filename(dict(tc.get('info_dict', {}))) |
702665c0 | 147 | |
28570840 | 148 | res_dict = None |
5f6a1245 | 149 | |
28570840 PH |
150 | def try_rm_tcs_files(tcs=None): |
151 | if tcs is None: | |
152 | tcs = test_cases | |
153 | for tc in tcs: | |
702665c0 JMF |
154 | tc_filename = get_tc_filename(tc) |
155 | try_rm(tc_filename) | |
156 | try_rm(tc_filename + '.part') | |
4eb92208 | 157 | try_rm(os.path.splitext(tc_filename)[0] + '.info.json') |
702665c0 | 158 | try_rm_tcs_files() |
5c892b0b | 159 | try: |
dd508b7c FV |
160 | try_num = 1 |
161 | while True: | |
8cc83b8d | 162 | try: |
3bef10a5 | 163 | # We're not using .download here since that is just a shim |
e8ee972c PH |
164 | # for outside error handling, and returns the exit code |
165 | # instead of the result dict. | |
308cfe0a S |
166 | res_dict = ydl.extract_info( |
167 | test_case['url'], | |
168 | force_generic_extractor=params.get('force_generic_extractor', False)) | |
8cc83b8d | 169 | except (DownloadError, ExtractorError) as err: |
8cc83b8d | 170 | # Check if the exception is not a network related one |
ac668111 | 171 | if not err.exc_info[0] in (urllib.error.URLError, socket.timeout, UnavailableVideoError, http.client.BadStatusLine) or (err.exc_info[0] == compat_HTTPError and err.exc_info[1].code == 503): |
8cc83b8d FV |
172 | raise |
173 | ||
dd508b7c | 174 | if try_num == RETRIES: |
8936f68a | 175 | report_warning('%s failed due to network errors, skipping...' % tname) |
dd508b7c FV |
176 | return |
177 | ||
86e5f3ed | 178 | print(f'Retrying: {try_num} failed tries\n\n##########\n\n') |
dd508b7c FV |
179 | |
180 | try_num += 1 | |
8cc83b8d FV |
181 | else: |
182 | break | |
5c892b0b | 183 | |
e8ee972c | 184 | if is_playlist: |
880ee801 | 185 | self.assertTrue(res_dict['_type'] in ['playlist', 'multi_video']) |
d6e6a422 | 186 | self.assertTrue('entries' in res_dict) |
f74b341d | 187 | expect_info_dict(self, res_dict, test_case.get('info_dict', {})) |
d6e6a422 | 188 | |
e8ee972c | 189 | if 'playlist_mincount' in test_case: |
0990305d PH |
190 | assertGreaterEqual( |
191 | self, | |
e8ee972c PH |
192 | len(res_dict['entries']), |
193 | test_case['playlist_mincount'], | |
194 | 'Expected at least %d in playlist %s, but got only %d' % ( | |
195 | test_case['playlist_mincount'], test_case['url'], | |
196 | len(res_dict['entries']))) | |
829476b8 PH |
197 | if 'playlist_count' in test_case: |
198 | self.assertEqual( | |
199 | len(res_dict['entries']), | |
200 | test_case['playlist_count'], | |
28570840 | 201 | 'Expected %d entries in playlist %s, but got %d.' % ( |
22a6f150 | 202 | test_case['playlist_count'], |
28570840 | 203 | test_case['url'], |
22a6f150 PH |
204 | len(res_dict['entries']), |
205 | )) | |
28570840 PH |
206 | if 'playlist_duration_sum' in test_case: |
207 | got_duration = sum(e['duration'] for e in res_dict['entries']) | |
208 | self.assertEqual( | |
209 | test_case['playlist_duration_sum'], got_duration) | |
e8ee972c | 210 | |
364a69e8 S |
211 | # Generalize both playlists and single videos to unified format for |
212 | # simplicity | |
213 | if 'entries' not in res_dict: | |
214 | res_dict['entries'] = [res_dict] | |
215 | ||
80b2fdf9 | 216 | for tc_num, tc in enumerate(test_cases): |
364a69e8 S |
217 | tc_res_dict = res_dict['entries'][tc_num] |
218 | # First, check test cases' data against extracted data alone | |
80b2fdf9 | 219 | expect_info_dict(self, tc_res_dict, tc.get('info_dict', {})) |
364a69e8 | 220 | # Now, check downloaded file consistency |
702665c0 | 221 | tc_filename = get_tc_filename(tc) |
511eda8e | 222 | if not test_case.get('params', {}).get('skip_download', False): |
702665c0 JMF |
223 | self.assertTrue(os.path.exists(tc_filename), msg='Missing file ' + tc_filename) |
224 | self.assertTrue(tc_filename in finished_hook_called) | |
08a36c35 S |
225 | expected_minsize = tc.get('file_minsize', 10000) |
226 | if expected_minsize is not None: | |
227 | if params.get('test'): | |
228 | expected_minsize = max(expected_minsize, 10000) | |
229 | got_fsize = os.path.getsize(tc_filename) | |
230 | assertGreaterEqual( | |
231 | self, got_fsize, expected_minsize, | |
232 | 'Expected %s to be at least %s, but it\'s only %s ' % | |
233 | (tc_filename, format_bytes(expected_minsize), | |
234 | format_bytes(got_fsize))) | |
235 | if 'md5' in tc: | |
236 | md5_for_file = _file_md5(tc_filename) | |
374560f0 | 237 | self.assertEqual(tc['md5'], md5_for_file) |
364a69e8 S |
238 | # Finally, check test cases' data again but this time against |
239 | # extracted data from info JSON file written during processing | |
4eb92208 | 240 | info_json_fn = os.path.splitext(tc_filename)[0] + '.info.json' |
f744c0f3 PH |
241 | self.assertTrue( |
242 | os.path.exists(info_json_fn), | |
243 | 'Missing info file %s' % info_json_fn) | |
86e5f3ed | 244 | with open(info_json_fn, encoding='utf-8') as infof: |
5c892b0b | 245 | info_dict = json.load(infof) |
f74b341d | 246 | expect_info_dict(self, info_dict, tc.get('info_dict', {})) |
5c892b0b | 247 | finally: |
702665c0 | 248 | try_rm_tcs_files() |
d6e6a422 | 249 | if is_playlist and res_dict is not None and res_dict.get('entries'): |
28570840 PH |
250 | # Remove all other files that may have been extracted if the |
251 | # extractor returns full results even with extract_flat | |
252 | res_tcs = [{'info_dict': e} for e in res_dict['entries']] | |
253 | try_rm_tcs_files(res_tcs) | |
fd5ff020 | 254 | |
1535ac2a | 255 | return test_template |
fd5ff020 | 256 | |
582be358 | 257 | |
5f6a1245 | 258 | # And add them to TestDownload |
243c57cf | 259 | tests_counter = {} |
260 | for test_case in defs: | |
261 | name = test_case['name'] | |
262 | i = tests_counter.get(name, 0) | |
263 | tests_counter[name] = i + 1 | |
264 | tname = f'test_{name}_{i}' if i else f'test_{name}' | |
8936f68a | 265 | test_method = generator(test_case, tname) |
a0f59cdc | 266 | test_method.__name__ = str(tname) |
c6c22e98 JH |
267 | ie_list = test_case.get('add_ie') |
268 | test_method.add_ie = ie_list and ','.join(ie_list) | |
fd5ff020 | 269 | setattr(TestDownload, test_method.__name__, test_method) |
5d01a647 | 270 | del test_method |
cdab8aa3 PH |
271 | |
272 | ||
243c57cf | 273 | def batch_generator(name, num_tests): |
274 | ||
275 | def test_template(self): | |
276 | for i in range(num_tests): | |
277 | getattr(self, f'test_{name}_{i}' if i else f'test_{name}')() | |
278 | ||
279 | return test_template | |
280 | ||
281 | ||
282 | for name, num_tests in tests_counter.items(): | |
283 | test_method = batch_generator(name, num_tests) | |
284 | test_method.__name__ = f'test_{name}_all' | |
285 | test_method.add_ie = '' | |
286 | setattr(TestDownload, test_method.__name__, test_method) | |
287 | del test_method | |
288 | ||
289 | ||
cdab8aa3 PH |
290 | if __name__ == '__main__': |
291 | unittest.main() |