]> jfr.im git - yt-dlp.git/blame - test/test_download.py
[CBS] Add ParamountPlusSeriesIE (#603)
[yt-dlp.git] / test / test_download.py
CommitLineData
cc52de43 1#!/usr/bin/env python3
fd5ff020 2
a0f59cdc
PH
3from __future__ import unicode_literals
4
44a5f171
PH
5# Allow direct execution
6import os
7import sys
8import unittest
9sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
10
dd508b7c 11from test.helper import (
0990305d 12 assertGreaterEqual,
060ac762 13 expect_info_dict,
70b7e3fb 14 expect_warnings,
dd508b7c 15 get_params,
ff14fc49 16 gettestcases,
060ac762 17 is_download_test,
257cfebf 18 report_warning,
060ac762 19 try_rm,
dd508b7c 20)
44a5f171
PH
21
22
efe8902f 23import hashlib
fd5ff020 24import io
7f60b5aa 25import json
6b3aef80 26import socket
cdab8aa3 27
7a5c1cfe
P
28import yt_dlp.YoutubeDL
29from yt_dlp.compat import (
dcf3eec4 30 compat_http_client,
44a5f171 31 compat_urllib_error,
f6cc16f5 32 compat_HTTPError,
42f7d2f5 33)
7a5c1cfe 34from yt_dlp.utils import (
44a5f171
PH
35 DownloadError,
36 ExtractorError,
753727cd 37 format_bytes,
44a5f171
PH
38 UnavailableVideoError,
39)
7a5c1cfe 40from yt_dlp.extractor import get_info_extractor
fd5ff020 41
8cc83b8d
FV
42RETRIES = 3
43
5f6a1245 44
7a5c1cfe 45class YoutubeDL(yt_dlp.YoutubeDL):
fd5ff020 46 def __init__(self, *args, **kwargs):
fd5ff020 47 self.to_stderr = self.to_screen
0eaf520d 48 self.processed_info_dicts = []
8222d8de 49 super(YoutubeDL, self).__init__(*args, **kwargs)
5f6a1245 50
476203d0 51 def report_warning(self, message):
be95cac1
FV
52 # Don't accept warnings during tests
53 raise ExtractorError(message)
5f6a1245 54
0eaf520d
FV
55 def process_info(self, info_dict):
56 self.processed_info_dicts.append(info_dict)
8222d8de 57 return super(YoutubeDL, self).process_info(info_dict)
1535ac2a 58
5f6a1245 59
fd5ff020
FV
60def _file_md5(fn):
61 with open(fn, 'rb') as f:
62 return hashlib.md5(f.read()).hexdigest()
63
582be358 64
ff14fc49 65defs = gettestcases()
6b47c7f2 66
0eaf520d 67
060ac762 68@is_download_test
1535ac2a 69class 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
c6c22e98
JH
76 def __str__(self):
77 """Identify each test with the `add_ie` attribute, if available."""
78
79 def strclass(cls):
80 """From 2.7's unittest; 2.6 had _strclass so we can't import it."""
81 return '%s.%s' % (cls.__module__, cls.__name__)
82
83 add_ie = getattr(self, self._testMethodName).add_ie
84 return '%s (%s)%s:' % (self._testMethodName,
85 strclass(self.__class__),
86 ' [%s]' % add_ie if add_ie else '')
87
fd5ff020 88 def setUp(self):
fd5ff020
FV
89 self.defs = defs
90
5f6a1245
JW
91# Dynamically generate tests
92
93
8936f68a 94def generator(test_case, tname):
5d01a647 95
1535ac2a 96 def test_template(self):
7a5c1cfe 97 ie = yt_dlp.extractor.get_info_extractor(test_case['name'])()
655c4100 98 other_ies = [get_info_extractor(ie_key)() for ie_key in test_case.get('add_ie', [])]
e8ee972c
PH
99 is_playlist = any(k.startswith('playlist') for k in test_case)
100 test_cases = test_case.get(
101 'playlist', [] if is_playlist else [test_case])
102
bc2884af
JMF
103 def print_skipping(reason):
104 print('Skipping %s: %s' % (test_case['name'], reason))
9ee2b5f6 105 if not ie.working():
bc2884af 106 print_skipping('IE marked as not _WORKING')
fd5ff020 107 return
e8ee972c
PH
108
109 for tc in test_cases:
110 info_dict = tc.get('info_dict', {})
4e980275 111 if not (info_dict.get('id') and info_dict.get('ext')):
2437fbca 112 raise Exception('Test definition incorrect. The output file cannot be known. Are both \'id\' and \'ext\' keys present?')
e8ee972c 113
fd5ff020 114 if 'skip' in test_case:
bc2884af 115 print_skipping(test_case['skip'])
fd5ff020 116 return
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())
9ee2b5f6 120 return
0eaf520d 121
44a5f171 122 params = get_params(test_case.get('params', {}))
8936f68a 123 params['outtmpl'] = tname + '_' + params['outtmpl']
e8ee972c 124 if is_playlist and 'playlist' not in test_case:
65d49afa 125 params.setdefault('extract_flat', 'in_playlist')
6911e11e 126 params.setdefault('playlistend', test_case.get('playlist_mincount'))
e8ee972c 127 params.setdefault('skip_download', True)
0eaf520d 128
ac35c266 129 ydl = YoutubeDL(params, auto_init=False)
023fa8c4 130 ydl.add_default_info_extractors()
bffbd5f0 131 finished_hook_called = set()
5f6a1245 132
bffbd5f0
PH
133 def _hook(status):
134 if status['status'] == 'finished':
135 finished_hook_called.add(status['filename'])
933605d7 136 ydl.add_progress_hook(_hook)
70b7e3fb 137 expect_warnings(ydl, test_case.get('expected_warnings', []))
5c892b0b 138
702665c0 139 def get_tc_filename(tc):
4e980275 140 return ydl.prepare_filename(tc.get('info_dict', {}))
702665c0 141
28570840 142 res_dict = None
5f6a1245 143
28570840
PH
144 def try_rm_tcs_files(tcs=None):
145 if tcs is None:
146 tcs = test_cases
147 for tc in tcs:
702665c0
JMF
148 tc_filename = get_tc_filename(tc)
149 try_rm(tc_filename)
150 try_rm(tc_filename + '.part')
4eb92208 151 try_rm(os.path.splitext(tc_filename)[0] + '.info.json')
702665c0 152 try_rm_tcs_files()
5c892b0b 153 try:
dd508b7c
FV
154 try_num = 1
155 while True:
8cc83b8d 156 try:
3bef10a5 157 # We're not using .download here since that is just a shim
e8ee972c
PH
158 # for outside error handling, and returns the exit code
159 # instead of the result dict.
308cfe0a
S
160 res_dict = ydl.extract_info(
161 test_case['url'],
162 force_generic_extractor=params.get('force_generic_extractor', False))
8cc83b8d 163 except (DownloadError, ExtractorError) as err:
8cc83b8d 164 # Check if the exception is not a network related one
dcf3eec4 165 if not err.exc_info[0] in (compat_urllib_error.URLError, socket.timeout, UnavailableVideoError, compat_http_client.BadStatusLine) or (err.exc_info[0] == compat_HTTPError and err.exc_info[1].code == 503):
8cc83b8d
FV
166 raise
167
dd508b7c 168 if try_num == RETRIES:
8936f68a 169 report_warning('%s failed due to network errors, skipping...' % tname)
dd508b7c
FV
170 return
171
172 print('Retrying: {0} failed tries\n\n##########\n\n'.format(try_num))
173
174 try_num += 1
8cc83b8d
FV
175 else:
176 break
5c892b0b 177
e8ee972c 178 if is_playlist:
880ee801 179 self.assertTrue(res_dict['_type'] in ['playlist', 'multi_video'])
d6e6a422 180 self.assertTrue('entries' in res_dict)
f74b341d 181 expect_info_dict(self, res_dict, test_case.get('info_dict', {}))
d6e6a422 182
e8ee972c 183 if 'playlist_mincount' in test_case:
0990305d
PH
184 assertGreaterEqual(
185 self,
e8ee972c
PH
186 len(res_dict['entries']),
187 test_case['playlist_mincount'],
188 'Expected at least %d in playlist %s, but got only %d' % (
189 test_case['playlist_mincount'], test_case['url'],
190 len(res_dict['entries'])))
829476b8
PH
191 if 'playlist_count' in test_case:
192 self.assertEqual(
193 len(res_dict['entries']),
194 test_case['playlist_count'],
28570840 195 'Expected %d entries in playlist %s, but got %d.' % (
22a6f150 196 test_case['playlist_count'],
28570840 197 test_case['url'],
22a6f150
PH
198 len(res_dict['entries']),
199 ))
28570840
PH
200 if 'playlist_duration_sum' in test_case:
201 got_duration = sum(e['duration'] for e in res_dict['entries'])
202 self.assertEqual(
203 test_case['playlist_duration_sum'], got_duration)
e8ee972c 204
364a69e8
S
205 # Generalize both playlists and single videos to unified format for
206 # simplicity
207 if 'entries' not in res_dict:
208 res_dict['entries'] = [res_dict]
209
80b2fdf9 210 for tc_num, tc in enumerate(test_cases):
364a69e8
S
211 tc_res_dict = res_dict['entries'][tc_num]
212 # First, check test cases' data against extracted data alone
80b2fdf9 213 expect_info_dict(self, tc_res_dict, tc.get('info_dict', {}))
364a69e8 214 # Now, check downloaded file consistency
702665c0 215 tc_filename = get_tc_filename(tc)
511eda8e 216 if not test_case.get('params', {}).get('skip_download', False):
702665c0
JMF
217 self.assertTrue(os.path.exists(tc_filename), msg='Missing file ' + tc_filename)
218 self.assertTrue(tc_filename in finished_hook_called)
08a36c35
S
219 expected_minsize = tc.get('file_minsize', 10000)
220 if expected_minsize is not None:
221 if params.get('test'):
222 expected_minsize = max(expected_minsize, 10000)
223 got_fsize = os.path.getsize(tc_filename)
224 assertGreaterEqual(
225 self, got_fsize, expected_minsize,
226 'Expected %s to be at least %s, but it\'s only %s ' %
227 (tc_filename, format_bytes(expected_minsize),
228 format_bytes(got_fsize)))
229 if 'md5' in tc:
230 md5_for_file = _file_md5(tc_filename)
374560f0 231 self.assertEqual(tc['md5'], md5_for_file)
364a69e8
S
232 # Finally, check test cases' data again but this time against
233 # extracted data from info JSON file written during processing
4eb92208 234 info_json_fn = os.path.splitext(tc_filename)[0] + '.info.json'
f744c0f3
PH
235 self.assertTrue(
236 os.path.exists(info_json_fn),
237 'Missing info file %s' % info_json_fn)
4eb92208 238 with io.open(info_json_fn, encoding='utf-8') as infof:
5c892b0b 239 info_dict = json.load(infof)
f74b341d 240 expect_info_dict(self, info_dict, tc.get('info_dict', {}))
5c892b0b 241 finally:
702665c0 242 try_rm_tcs_files()
d6e6a422 243 if is_playlist and res_dict is not None and res_dict.get('entries'):
28570840
PH
244 # Remove all other files that may have been extracted if the
245 # extractor returns full results even with extract_flat
246 res_tcs = [{'info_dict': e} for e in res_dict['entries']]
247 try_rm_tcs_files(res_tcs)
fd5ff020 248
1535ac2a 249 return test_template
fd5ff020 250
582be358 251
5f6a1245 252# And add them to TestDownload
f7ab6cbe 253for n, test_case in enumerate(defs):
2eb88d95
PH
254 tname = 'test_' + str(test_case['name'])
255 i = 1
256 while hasattr(TestDownload, tname):
a0f59cdc 257 tname = 'test_%s_%d' % (test_case['name'], i)
2eb88d95 258 i += 1
8936f68a 259 test_method = generator(test_case, tname)
a0f59cdc 260 test_method.__name__ = str(tname)
c6c22e98
JH
261 ie_list = test_case.get('add_ie')
262 test_method.add_ie = ie_list and ','.join(ie_list)
fd5ff020 263 setattr(TestDownload, test_method.__name__, test_method)
5d01a647 264 del test_method
cdab8aa3
PH
265
266
267if __name__ == '__main__':
268 unittest.main()