]>
Commit | Line | Data |
---|---|---|
fd5ff020 FV |
1 | #!/usr/bin/env python |
2 | ||
a0f59cdc PH |
3 | from __future__ import unicode_literals |
4 | ||
44a5f171 PH |
5 | # Allow direct execution |
6 | import os | |
7 | import sys | |
8 | import unittest | |
9 | sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__)))) | |
10 | ||
dd508b7c | 11 | from test.helper import ( |
0990305d | 12 | assertGreaterEqual, |
70b7e3fb | 13 | expect_warnings, |
dd508b7c | 14 | get_params, |
ff14fc49 | 15 | gettestcases, |
257cfebf | 16 | expect_info_dict, |
257cfebf PH |
17 | try_rm, |
18 | report_warning, | |
dd508b7c | 19 | ) |
44a5f171 PH |
20 | |
21 | ||
efe8902f | 22 | import hashlib |
fd5ff020 | 23 | import io |
7f60b5aa | 24 | import json |
6b3aef80 | 25 | import socket |
cdab8aa3 | 26 | |
cefecac1 U |
27 | import youtube_dlc.YoutubeDL |
28 | from youtube_dlc.compat import ( | |
dcf3eec4 | 29 | compat_http_client, |
44a5f171 | 30 | compat_urllib_error, |
f6cc16f5 | 31 | compat_HTTPError, |
42f7d2f5 | 32 | ) |
cefecac1 | 33 | from youtube_dlc.utils import ( |
44a5f171 PH |
34 | DownloadError, |
35 | ExtractorError, | |
753727cd | 36 | format_bytes, |
44a5f171 PH |
37 | UnavailableVideoError, |
38 | ) | |
cefecac1 | 39 | from youtube_dlc.extractor import get_info_extractor |
fd5ff020 | 40 | |
8cc83b8d FV |
41 | RETRIES = 3 |
42 | ||
5f6a1245 | 43 | |
cefecac1 | 44 | class YoutubeDL(youtube_dlc.YoutubeDL): |
fd5ff020 | 45 | def __init__(self, *args, **kwargs): |
fd5ff020 | 46 | self.to_stderr = self.to_screen |
0eaf520d | 47 | self.processed_info_dicts = [] |
8222d8de | 48 | super(YoutubeDL, self).__init__(*args, **kwargs) |
5f6a1245 | 49 | |
476203d0 | 50 | def report_warning(self, message): |
be95cac1 FV |
51 | # Don't accept warnings during tests |
52 | raise ExtractorError(message) | |
5f6a1245 | 53 | |
0eaf520d FV |
54 | def process_info(self, info_dict): |
55 | self.processed_info_dicts.append(info_dict) | |
8222d8de | 56 | return super(YoutubeDL, self).process_info(info_dict) |
1535ac2a | 57 | |
5f6a1245 | 58 | |
fd5ff020 FV |
59 | def _file_md5(fn): |
60 | with open(fn, 'rb') as f: | |
61 | return hashlib.md5(f.read()).hexdigest() | |
62 | ||
582be358 | 63 | |
ff14fc49 | 64 | defs = gettestcases() |
6b47c7f2 | 65 | |
0eaf520d | 66 | |
1535ac2a | 67 | class TestDownload(unittest.TestCase): |
8936f68a YCH |
68 | # Parallel testing in nosetests. See |
69 | # http://nose.readthedocs.org/en/latest/doc_tests/test_multiprocess/multiprocess.html | |
70 | _multiprocess_shared_ = True | |
71 | ||
744435f2 | 72 | maxDiff = None |
5f6a1245 | 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.""" | |
79 | return '%s.%s' % (cls.__module__, cls.__name__) | |
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): |
cefecac1 | 95 | ie = youtube_dlc.extractor.get_info_extractor(test_case['name'])() |
655c4100 | 96 | other_ies = [get_info_extractor(ie_key)() for ie_key in test_case.get('add_ie', [])] |
e8ee972c PH |
97 | is_playlist = any(k.startswith('playlist') for k in test_case) |
98 | test_cases = test_case.get( | |
99 | 'playlist', [] if is_playlist else [test_case]) | |
100 | ||
bc2884af JMF |
101 | def print_skipping(reason): |
102 | print('Skipping %s: %s' % (test_case['name'], reason)) | |
9ee2b5f6 | 103 | if not ie.working(): |
bc2884af | 104 | print_skipping('IE marked as not _WORKING') |
fd5ff020 | 105 | return |
e8ee972c PH |
106 | |
107 | for tc in test_cases: | |
108 | info_dict = tc.get('info_dict', {}) | |
4e980275 | 109 | if not (info_dict.get('id') and info_dict.get('ext')): |
2437fbca | 110 | raise Exception('Test definition incorrect. The output file cannot be known. Are both \'id\' and \'ext\' keys present?') |
e8ee972c | 111 | |
fd5ff020 | 112 | if 'skip' in test_case: |
bc2884af | 113 | print_skipping(test_case['skip']) |
fd5ff020 | 114 | return |
9ee2b5f6 JMF |
115 | for other_ie in other_ies: |
116 | if not other_ie.working(): | |
e075a44a | 117 | print_skipping('test depends on %sIE, marked as not WORKING' % other_ie.ie_key()) |
9ee2b5f6 | 118 | return |
0eaf520d | 119 | |
44a5f171 | 120 | params = get_params(test_case.get('params', {})) |
8936f68a | 121 | params['outtmpl'] = tname + '_' + params['outtmpl'] |
e8ee972c | 122 | if is_playlist and 'playlist' not in test_case: |
65d49afa | 123 | params.setdefault('extract_flat', 'in_playlist') |
e8ee972c | 124 | params.setdefault('skip_download', True) |
0eaf520d | 125 | |
ac35c266 | 126 | ydl = YoutubeDL(params, auto_init=False) |
023fa8c4 | 127 | ydl.add_default_info_extractors() |
bffbd5f0 | 128 | finished_hook_called = set() |
5f6a1245 | 129 | |
bffbd5f0 PH |
130 | def _hook(status): |
131 | if status['status'] == 'finished': | |
132 | finished_hook_called.add(status['filename']) | |
933605d7 | 133 | ydl.add_progress_hook(_hook) |
70b7e3fb | 134 | expect_warnings(ydl, test_case.get('expected_warnings', [])) |
5c892b0b | 135 | |
702665c0 | 136 | def get_tc_filename(tc): |
4e980275 | 137 | return ydl.prepare_filename(tc.get('info_dict', {})) |
702665c0 | 138 | |
28570840 | 139 | res_dict = None |
5f6a1245 | 140 | |
28570840 PH |
141 | def try_rm_tcs_files(tcs=None): |
142 | if tcs is None: | |
143 | tcs = test_cases | |
144 | for tc in tcs: | |
702665c0 JMF |
145 | tc_filename = get_tc_filename(tc) |
146 | try_rm(tc_filename) | |
147 | try_rm(tc_filename + '.part') | |
4eb92208 | 148 | try_rm(os.path.splitext(tc_filename)[0] + '.info.json') |
702665c0 | 149 | try_rm_tcs_files() |
5c892b0b | 150 | try: |
dd508b7c FV |
151 | try_num = 1 |
152 | while True: | |
8cc83b8d | 153 | try: |
3bef10a5 | 154 | # We're not using .download here since that is just a shim |
e8ee972c PH |
155 | # for outside error handling, and returns the exit code |
156 | # instead of the result dict. | |
308cfe0a S |
157 | res_dict = ydl.extract_info( |
158 | test_case['url'], | |
159 | force_generic_extractor=params.get('force_generic_extractor', False)) | |
8cc83b8d | 160 | except (DownloadError, ExtractorError) as err: |
8cc83b8d | 161 | # Check if the exception is not a network related one |
dcf3eec4 | 162 | 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 |
163 | raise |
164 | ||
dd508b7c | 165 | if try_num == RETRIES: |
8936f68a | 166 | report_warning('%s failed due to network errors, skipping...' % tname) |
dd508b7c FV |
167 | return |
168 | ||
169 | print('Retrying: {0} failed tries\n\n##########\n\n'.format(try_num)) | |
170 | ||
171 | try_num += 1 | |
8cc83b8d FV |
172 | else: |
173 | break | |
5c892b0b | 174 | |
e8ee972c | 175 | if is_playlist: |
880ee801 | 176 | self.assertTrue(res_dict['_type'] in ['playlist', 'multi_video']) |
d6e6a422 | 177 | self.assertTrue('entries' in res_dict) |
f74b341d | 178 | expect_info_dict(self, res_dict, test_case.get('info_dict', {})) |
d6e6a422 | 179 | |
e8ee972c | 180 | if 'playlist_mincount' in test_case: |
0990305d PH |
181 | assertGreaterEqual( |
182 | self, | |
e8ee972c PH |
183 | len(res_dict['entries']), |
184 | test_case['playlist_mincount'], | |
185 | 'Expected at least %d in playlist %s, but got only %d' % ( | |
186 | test_case['playlist_mincount'], test_case['url'], | |
187 | len(res_dict['entries']))) | |
829476b8 PH |
188 | if 'playlist_count' in test_case: |
189 | self.assertEqual( | |
190 | len(res_dict['entries']), | |
191 | test_case['playlist_count'], | |
28570840 | 192 | 'Expected %d entries in playlist %s, but got %d.' % ( |
22a6f150 | 193 | test_case['playlist_count'], |
28570840 | 194 | test_case['url'], |
22a6f150 PH |
195 | len(res_dict['entries']), |
196 | )) | |
28570840 PH |
197 | if 'playlist_duration_sum' in test_case: |
198 | got_duration = sum(e['duration'] for e in res_dict['entries']) | |
199 | self.assertEqual( | |
200 | test_case['playlist_duration_sum'], got_duration) | |
e8ee972c | 201 | |
364a69e8 S |
202 | # Generalize both playlists and single videos to unified format for |
203 | # simplicity | |
204 | if 'entries' not in res_dict: | |
205 | res_dict['entries'] = [res_dict] | |
206 | ||
80b2fdf9 | 207 | for tc_num, tc in enumerate(test_cases): |
364a69e8 S |
208 | tc_res_dict = res_dict['entries'][tc_num] |
209 | # First, check test cases' data against extracted data alone | |
80b2fdf9 | 210 | expect_info_dict(self, tc_res_dict, tc.get('info_dict', {})) |
364a69e8 | 211 | # Now, check downloaded file consistency |
702665c0 | 212 | tc_filename = get_tc_filename(tc) |
511eda8e | 213 | if not test_case.get('params', {}).get('skip_download', False): |
702665c0 JMF |
214 | self.assertTrue(os.path.exists(tc_filename), msg='Missing file ' + tc_filename) |
215 | self.assertTrue(tc_filename in finished_hook_called) | |
08a36c35 S |
216 | expected_minsize = tc.get('file_minsize', 10000) |
217 | if expected_minsize is not None: | |
218 | if params.get('test'): | |
219 | expected_minsize = max(expected_minsize, 10000) | |
220 | got_fsize = os.path.getsize(tc_filename) | |
221 | assertGreaterEqual( | |
222 | self, got_fsize, expected_minsize, | |
223 | 'Expected %s to be at least %s, but it\'s only %s ' % | |
224 | (tc_filename, format_bytes(expected_minsize), | |
225 | format_bytes(got_fsize))) | |
226 | if 'md5' in tc: | |
227 | md5_for_file = _file_md5(tc_filename) | |
374560f0 | 228 | self.assertEqual(tc['md5'], md5_for_file) |
364a69e8 S |
229 | # Finally, check test cases' data again but this time against |
230 | # extracted data from info JSON file written during processing | |
4eb92208 | 231 | info_json_fn = os.path.splitext(tc_filename)[0] + '.info.json' |
f744c0f3 PH |
232 | self.assertTrue( |
233 | os.path.exists(info_json_fn), | |
234 | 'Missing info file %s' % info_json_fn) | |
4eb92208 | 235 | with io.open(info_json_fn, encoding='utf-8') as infof: |
5c892b0b | 236 | info_dict = json.load(infof) |
f74b341d | 237 | expect_info_dict(self, info_dict, tc.get('info_dict', {})) |
5c892b0b | 238 | finally: |
702665c0 | 239 | try_rm_tcs_files() |
d6e6a422 | 240 | if is_playlist and res_dict is not None and res_dict.get('entries'): |
28570840 PH |
241 | # Remove all other files that may have been extracted if the |
242 | # extractor returns full results even with extract_flat | |
243 | res_tcs = [{'info_dict': e} for e in res_dict['entries']] | |
244 | try_rm_tcs_files(res_tcs) | |
fd5ff020 | 245 | |
1535ac2a | 246 | return test_template |
fd5ff020 | 247 | |
582be358 | 248 | |
5f6a1245 | 249 | # And add them to TestDownload |
f7ab6cbe | 250 | for n, test_case in enumerate(defs): |
2eb88d95 PH |
251 | tname = 'test_' + str(test_case['name']) |
252 | i = 1 | |
253 | while hasattr(TestDownload, tname): | |
a0f59cdc | 254 | tname = 'test_%s_%d' % (test_case['name'], i) |
2eb88d95 | 255 | i += 1 |
8936f68a | 256 | test_method = generator(test_case, tname) |
a0f59cdc | 257 | test_method.__name__ = str(tname) |
c6c22e98 JH |
258 | ie_list = test_case.get('add_ie') |
259 | test_method.add_ie = ie_list and ','.join(ie_list) | |
fd5ff020 | 260 | setattr(TestDownload, test_method.__name__, test_method) |
5d01a647 | 261 | del test_method |
cdab8aa3 PH |
262 | |
263 | ||
264 | if __name__ == '__main__': | |
265 | unittest.main() |