]> jfr.im git - yt-dlp.git/blame - test/test_YoutubeDL.py
[youtube:tab] Reject webpage data if redirected to home page
[yt-dlp.git] / test / test_YoutubeDL.py
CommitLineData
cc52de43 1#!/usr/bin/env python3
db13c16e 2# coding: utf-8
e028d0d1 3
89087418
PH
4from __future__ import unicode_literals
5
5d254f77
PH
6# Allow direct execution
7import os
e028d0d1
JMF
8import sys
9import unittest
5d254f77 10sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
e028d0d1 11
0217c783 12import copy
7d1eb38a 13import json
0217c783 14
c57f7757 15from test.helper import FakeYDL, assertRegexpMatches
7a5c1cfe 16from yt_dlp import YoutubeDL
901130bb 17from yt_dlp.compat import compat_os_name, compat_setenv, compat_str, compat_urllib_error
7a5c1cfe
P
18from yt_dlp.extractor import YoutubeIE
19from yt_dlp.extractor.common import InfoExtractor
20from yt_dlp.postprocessor.common import PostProcessor
6e84b215 21from yt_dlp.utils import ExtractorError, int_or_none, match_filter_func, LazyList
e028d0d1 22
8508557e
JMF
23TEST_URL = 'http://localhost/sample.mp4'
24
e028d0d1
JMF
25
26class YDL(FakeYDL):
f4d96df0
PH
27 def __init__(self, *args, **kwargs):
28 super(YDL, self).__init__(*args, **kwargs)
e028d0d1 29 self.downloaded_info_dicts = []
f4d96df0 30 self.msgs = []
5d254f77 31
e028d0d1 32 def process_info(self, info_dict):
f46e2f9d 33 info_dict = info_dict.copy()
af32f40b 34 info_dict.pop('__original_infodict', None)
e028d0d1
JMF
35 self.downloaded_info_dicts.append(info_dict)
36
f4d96df0
PH
37 def to_screen(self, msg):
38 self.msgs.append(msg)
39
b5ac45b1 40 def dl(self, *args, **kwargs):
41 assert False, 'Downloader must not be invoked for test_YoutubeDL'
42
5d254f77 43
3537b93d
PH
44def _make_result(formats, **kwargs):
45 res = {
46 'formats': formats,
47 'id': 'testid',
48 'title': 'testttitle',
49 'extractor': 'testex',
0396806f 50 'extractor_key': 'TestEx',
732044af 51 'webpage_url': 'http://example.com/watch?v=shenanigans',
3537b93d
PH
52 }
53 res.update(**kwargs)
54 return res
55
56
e028d0d1
JMF
57class TestFormatSelection(unittest.TestCase):
58 def test_prefer_free_formats(self):
59 # Same resolution => download webm
60 ydl = YDL()
61 ydl.params['prefer_free_formats'] = True
5d254f77 62 formats = [
8508557e
JMF
63 {'ext': 'webm', 'height': 460, 'url': TEST_URL},
64 {'ext': 'mp4', 'height': 460, 'url': TEST_URL},
5d254f77 65 ]
3537b93d 66 info_dict = _make_result(formats)
3d4a70b8
PH
67 yie = YoutubeIE(ydl)
68 yie._sort_formats(info_dict['formats'])
e028d0d1
JMF
69 ydl.process_ie_result(info_dict)
70 downloaded = ydl.downloaded_info_dicts[0]
89087418 71 self.assertEqual(downloaded['ext'], 'webm')
e028d0d1
JMF
72
73 # Different resolution => download best quality (mp4)
74 ydl = YDL()
75 ydl.params['prefer_free_formats'] = True
5d254f77 76 formats = [
8508557e
JMF
77 {'ext': 'webm', 'height': 720, 'url': TEST_URL},
78 {'ext': 'mp4', 'height': 1080, 'url': TEST_URL},
5d254f77 79 ]
89087418 80 info_dict['formats'] = formats
3d4a70b8
PH
81 yie = YoutubeIE(ydl)
82 yie._sort_formats(info_dict['formats'])
e028d0d1
JMF
83 ydl.process_ie_result(info_dict)
84 downloaded = ydl.downloaded_info_dicts[0]
89087418 85 self.assertEqual(downloaded['ext'], 'mp4')
e028d0d1 86
5d0c5371 87 # No prefer_free_formats => prefer mp4 and webm
e028d0d1
JMF
88 ydl = YDL()
89 ydl.params['prefer_free_formats'] = False
5d254f77 90 formats = [
8508557e
JMF
91 {'ext': 'webm', 'height': 720, 'url': TEST_URL},
92 {'ext': 'mp4', 'height': 720, 'url': TEST_URL},
93 {'ext': 'flv', 'height': 720, 'url': TEST_URL},
5d254f77 94 ]
89087418 95 info_dict['formats'] = formats
3d4a70b8
PH
96 yie = YoutubeIE(ydl)
97 yie._sort_formats(info_dict['formats'])
98 ydl.process_ie_result(info_dict)
99 downloaded = ydl.downloaded_info_dicts[0]
89087418 100 self.assertEqual(downloaded['ext'], 'mp4')
3d4a70b8
PH
101
102 ydl = YDL()
103 ydl.params['prefer_free_formats'] = False
104 formats = [
8508557e
JMF
105 {'ext': 'flv', 'height': 720, 'url': TEST_URL},
106 {'ext': 'webm', 'height': 720, 'url': TEST_URL},
3d4a70b8 107 ]
89087418 108 info_dict['formats'] = formats
3d4a70b8
PH
109 yie = YoutubeIE(ydl)
110 yie._sort_formats(info_dict['formats'])
e028d0d1
JMF
111 ydl.process_ie_result(info_dict)
112 downloaded = ydl.downloaded_info_dicts[0]
5d0c5371 113 self.assertEqual(downloaded['ext'], 'webm')
e028d0d1 114
a9c58ad9
JMF
115 def test_format_selection(self):
116 formats = [
8508557e 117 {'format_id': '35', 'ext': 'mp4', 'preference': 1, 'url': TEST_URL},
232541df 118 {'format_id': 'example-with-dashes', 'ext': 'webm', 'preference': 1, 'url': TEST_URL},
8508557e
JMF
119 {'format_id': '45', 'ext': 'webm', 'preference': 2, 'url': TEST_URL},
120 {'format_id': '47', 'ext': 'webm', 'preference': 3, 'url': TEST_URL},
121 {'format_id': '2', 'ext': 'flv', 'preference': 4, 'url': TEST_URL},
a9c58ad9 122 ]
3537b93d 123 info_dict = _make_result(formats)
a9c58ad9 124
b5ac45b1 125 def test(inp, *expected, multi=False):
126 ydl = YDL({
127 'format': inp,
128 'allow_multiple_video_streams': multi,
129 'allow_multiple_audio_streams': multi,
130 })
131 ydl.process_ie_result(info_dict.copy())
132 downloaded = map(lambda x: x['format_id'], ydl.downloaded_info_dicts)
133 self.assertEqual(list(downloaded), list(expected))
134
135 test('20/47', '47')
136 test('20/71/worst', '35')
137 test(None, '2')
138 test('webm/mp4', '47')
139 test('3gp/40/mp4', '35')
140 test('example-with-dashes', 'example-with-dashes')
dd2a987d 141 test('all', '2', '47', '45', 'example-with-dashes', '35')
b5ac45b1 142 test('mergeall', '2+47+45+example-with-dashes+35', multi=True)
232541df 143
ba7678f9
PH
144 def test_format_selection_audio(self):
145 formats = [
8508557e
JMF
146 {'format_id': 'audio-low', 'ext': 'webm', 'preference': 1, 'vcodec': 'none', 'url': TEST_URL},
147 {'format_id': 'audio-mid', 'ext': 'webm', 'preference': 2, 'vcodec': 'none', 'url': TEST_URL},
148 {'format_id': 'audio-high', 'ext': 'flv', 'preference': 3, 'vcodec': 'none', 'url': TEST_URL},
149 {'format_id': 'vid', 'ext': 'mp4', 'preference': 4, 'url': TEST_URL},
ba7678f9 150 ]
3537b93d 151 info_dict = _make_result(formats)
ba7678f9 152
89087418 153 ydl = YDL({'format': 'bestaudio'})
ba7678f9
PH
154 ydl.process_ie_result(info_dict.copy())
155 downloaded = ydl.downloaded_info_dicts[0]
89087418 156 self.assertEqual(downloaded['format_id'], 'audio-high')
ba7678f9 157
89087418 158 ydl = YDL({'format': 'worstaudio'})
ba7678f9
PH
159 ydl.process_ie_result(info_dict.copy())
160 downloaded = ydl.downloaded_info_dicts[0]
89087418 161 self.assertEqual(downloaded['format_id'], 'audio-low')
ba7678f9
PH
162
163 formats = [
8508557e
JMF
164 {'format_id': 'vid-low', 'ext': 'mp4', 'preference': 1, 'url': TEST_URL},
165 {'format_id': 'vid-high', 'ext': 'mp4', 'preference': 2, 'url': TEST_URL},
ba7678f9 166 ]
3537b93d 167 info_dict = _make_result(formats)
ba7678f9 168
89087418 169 ydl = YDL({'format': 'bestaudio/worstaudio/best'})
ba7678f9
PH
170 ydl.process_ie_result(info_dict.copy())
171 downloaded = ydl.downloaded_info_dicts[0]
89087418 172 self.assertEqual(downloaded['format_id'], 'vid-high')
ba7678f9 173
0217c783
PH
174 def test_format_selection_audio_exts(self):
175 formats = [
176 {'format_id': 'mp3-64', 'ext': 'mp3', 'abr': 64, 'url': 'http://_', 'vcodec': 'none'},
177 {'format_id': 'ogg-64', 'ext': 'ogg', 'abr': 64, 'url': 'http://_', 'vcodec': 'none'},
178 {'format_id': 'aac-64', 'ext': 'aac', 'abr': 64, 'url': 'http://_', 'vcodec': 'none'},
179 {'format_id': 'mp3-32', 'ext': 'mp3', 'abr': 32, 'url': 'http://_', 'vcodec': 'none'},
180 {'format_id': 'aac-32', 'ext': 'aac', 'abr': 32, 'url': 'http://_', 'vcodec': 'none'},
181 ]
182
183 info_dict = _make_result(formats)
184 ydl = YDL({'format': 'best'})
185 ie = YoutubeIE(ydl)
186 ie._sort_formats(info_dict['formats'])
187 ydl.process_ie_result(copy.deepcopy(info_dict))
188 downloaded = ydl.downloaded_info_dicts[0]
189 self.assertEqual(downloaded['format_id'], 'aac-64')
190
191 ydl = YDL({'format': 'mp3'})
192 ie = YoutubeIE(ydl)
193 ie._sort_formats(info_dict['formats'])
194 ydl.process_ie_result(copy.deepcopy(info_dict))
195 downloaded = ydl.downloaded_info_dicts[0]
196 self.assertEqual(downloaded['format_id'], 'mp3-64')
197
198 ydl = YDL({'prefer_free_formats': True})
199 ie = YoutubeIE(ydl)
200 ie._sort_formats(info_dict['formats'])
201 ydl.process_ie_result(copy.deepcopy(info_dict))
202 downloaded = ydl.downloaded_info_dicts[0]
203 self.assertEqual(downloaded['format_id'], 'ogg-64')
204
bc6d5978
JMF
205 def test_format_selection_video(self):
206 formats = [
8508557e
JMF
207 {'format_id': 'dash-video-low', 'ext': 'mp4', 'preference': 1, 'acodec': 'none', 'url': TEST_URL},
208 {'format_id': 'dash-video-high', 'ext': 'mp4', 'preference': 2, 'acodec': 'none', 'url': TEST_URL},
209 {'format_id': 'vid', 'ext': 'mp4', 'preference': 3, 'url': TEST_URL},
bc6d5978 210 ]
3537b93d 211 info_dict = _make_result(formats)
bc6d5978
JMF
212
213 ydl = YDL({'format': 'bestvideo'})
214 ydl.process_ie_result(info_dict.copy())
215 downloaded = ydl.downloaded_info_dicts[0]
216 self.assertEqual(downloaded['format_id'], 'dash-video-high')
217
218 ydl = YDL({'format': 'worstvideo'})
219 ydl.process_ie_result(info_dict.copy())
220 downloaded = ydl.downloaded_info_dicts[0]
221 self.assertEqual(downloaded['format_id'], 'dash-video-low')
222
4c3b16d5
S
223 ydl = YDL({'format': 'bestvideo[format_id^=dash][format_id$=low]'})
224 ydl.process_ie_result(info_dict.copy())
225 downloaded = ydl.downloaded_info_dicts[0]
226 self.assertEqual(downloaded['format_id'], 'dash-video-low')
227
b913348d 228 formats = [
229 {'format_id': 'vid-vcodec-dot', 'ext': 'mp4', 'preference': 1, 'vcodec': 'avc1.123456', 'acodec': 'none', 'url': TEST_URL},
230 ]
231 info_dict = _make_result(formats)
232
233 ydl = YDL({'format': 'bestvideo[vcodec=avc1.123456]'})
234 ydl.process_ie_result(info_dict.copy())
235 downloaded = ydl.downloaded_info_dicts[0]
236 self.assertEqual(downloaded['format_id'], 'vid-vcodec-dot')
237
2cc779f4
S
238 def test_format_selection_string_ops(self):
239 formats = [
240 {'format_id': 'abc-cba', 'ext': 'mp4', 'url': TEST_URL},
e118a879 241 {'format_id': 'zxc-cxz', 'ext': 'webm', 'url': TEST_URL},
2cc779f4
S
242 ]
243 info_dict = _make_result(formats)
244
245 # equals (=)
246 ydl = YDL({'format': '[format_id=abc-cba]'})
247 ydl.process_ie_result(info_dict.copy())
248 downloaded = ydl.downloaded_info_dicts[0]
249 self.assertEqual(downloaded['format_id'], 'abc-cba')
250
251 # does not equal (!=)
252 ydl = YDL({'format': '[format_id!=abc-cba]'})
e118a879
S
253 ydl.process_ie_result(info_dict.copy())
254 downloaded = ydl.downloaded_info_dicts[0]
255 self.assertEqual(downloaded['format_id'], 'zxc-cxz')
256
257 ydl = YDL({'format': '[format_id!=abc-cba][format_id!=zxc-cxz]'})
2cc779f4
S
258 self.assertRaises(ExtractorError, ydl.process_ie_result, info_dict.copy())
259
260 # starts with (^=)
261 ydl = YDL({'format': '[format_id^=abc]'})
262 ydl.process_ie_result(info_dict.copy())
263 downloaded = ydl.downloaded_info_dicts[0]
264 self.assertEqual(downloaded['format_id'], 'abc-cba')
265
266 # does not start with (!^=)
e118a879
S
267 ydl = YDL({'format': '[format_id!^=abc]'})
268 ydl.process_ie_result(info_dict.copy())
269 downloaded = ydl.downloaded_info_dicts[0]
270 self.assertEqual(downloaded['format_id'], 'zxc-cxz')
271
272 ydl = YDL({'format': '[format_id!^=abc][format_id!^=zxc]'})
2cc779f4
S
273 self.assertRaises(ExtractorError, ydl.process_ie_result, info_dict.copy())
274
275 # ends with ($=)
276 ydl = YDL({'format': '[format_id$=cba]'})
277 ydl.process_ie_result(info_dict.copy())
278 downloaded = ydl.downloaded_info_dicts[0]
279 self.assertEqual(downloaded['format_id'], 'abc-cba')
280
281 # does not end with (!$=)
e118a879
S
282 ydl = YDL({'format': '[format_id!$=cba]'})
283 ydl.process_ie_result(info_dict.copy())
284 downloaded = ydl.downloaded_info_dicts[0]
285 self.assertEqual(downloaded['format_id'], 'zxc-cxz')
286
287 ydl = YDL({'format': '[format_id!$=cba][format_id!$=cxz]'})
2cc779f4
S
288 self.assertRaises(ExtractorError, ydl.process_ie_result, info_dict.copy())
289
290 # contains (*=)
e118a879 291 ydl = YDL({'format': '[format_id*=bc-cb]'})
2cc779f4
S
292 ydl.process_ie_result(info_dict.copy())
293 downloaded = ydl.downloaded_info_dicts[0]
294 self.assertEqual(downloaded['format_id'], 'abc-cba')
295
296 # does not contain (!*=)
e118a879
S
297 ydl = YDL({'format': '[format_id!*=bc-cb]'})
298 ydl.process_ie_result(info_dict.copy())
299 downloaded = ydl.downloaded_info_dicts[0]
300 self.assertEqual(downloaded['format_id'], 'zxc-cxz')
301
302 ydl = YDL({'format': '[format_id!*=abc][format_id!*=zxc]'})
303 self.assertRaises(ExtractorError, ydl.process_ie_result, info_dict.copy())
304
2cc779f4
S
305 ydl = YDL({'format': '[format_id!*=-]'})
306 self.assertRaises(ExtractorError, ydl.process_ie_result, info_dict.copy())
307
3d4a70b8 308 def test_youtube_format_selection(self):
a7191c6f 309 # FIXME: Rewrite in accordance with the new format sorting options
5d0c5371 310 return
5d0c5371 311
3d4a70b8 312 order = [
86bf2905 313 '38', '37', '46', '22', '45', '35', '44', '18', '34', '43', '6', '5', '17', '36', '13',
3d4a70b8
PH
314 # Apple HTTP Live Streaming
315 '96', '95', '94', '93', '92', '132', '151',
316 # 3D
317 '85', '84', '102', '83', '101', '82', '100',
318 # Dash video
c11125f9 319 '137', '248', '136', '247', '135', '246',
3d4a70b8
PH
320 '245', '244', '134', '243', '133', '242', '160',
321 # Dash audio
a053c349 322 '141', '172', '140', '171', '139',
3d4a70b8
PH
323 ]
324
67134eab
JMF
325 def format_info(f_id):
326 info = YoutubeIE._formats[f_id].copy()
1df41411 327
91cb6b50 328 # XXX: In real cases InfoExtractor._parse_mpd_formats() fills up 'acodec'
1df41411
YCH
329 # and 'vcodec', while in tests such information is incomplete since
330 # commit a6c2c24479e5f4827ceb06f64d855329c0a6f593
331 # test_YoutubeDL.test_youtube_format_selection is broken without
332 # this fix
333 if 'acodec' in info and 'vcodec' not in info:
334 info['vcodec'] = 'none'
335 elif 'vcodec' in info and 'acodec' not in info:
336 info['acodec'] = 'none'
337
67134eab
JMF
338 info['format_id'] = f_id
339 info['url'] = 'url:' + f_id
340 return info
341 formats_order = [format_info(f_id) for f_id in order]
342
343 info_dict = _make_result(list(formats_order), extractor='youtube')
344 ydl = YDL({'format': 'bestvideo+bestaudio'})
345 yie = YoutubeIE(ydl)
346 yie._sort_formats(info_dict['formats'])
347 ydl.process_ie_result(info_dict)
348 downloaded = ydl.downloaded_info_dicts[0]
5d0c5371 349 self.assertEqual(downloaded['format_id'], '248+172')
67134eab 350 self.assertEqual(downloaded['ext'], 'mp4')
3d4a70b8 351
cf2ac6df
JMF
352 info_dict = _make_result(list(formats_order), extractor='youtube')
353 ydl = YDL({'format': 'bestvideo[height>=999999]+bestaudio/best'})
354 yie = YoutubeIE(ydl)
355 yie._sort_formats(info_dict['formats'])
356 ydl.process_ie_result(info_dict)
357 downloaded = ydl.downloaded_info_dicts[0]
358 self.assertEqual(downloaded['format_id'], '38')
359
f5f4a27a
JMF
360 info_dict = _make_result(list(formats_order), extractor='youtube')
361 ydl = YDL({'format': 'bestvideo/best,bestaudio'})
362 yie = YoutubeIE(ydl)
363 yie._sort_formats(info_dict['formats'])
364 ydl.process_ie_result(info_dict)
365 downloaded_ids = [info['format_id'] for info in ydl.downloaded_info_dicts]
366 self.assertEqual(downloaded_ids, ['137', '141'])
367
0130afb7
JMF
368 info_dict = _make_result(list(formats_order), extractor='youtube')
369 ydl = YDL({'format': '(bestvideo[ext=mp4],bestvideo[ext=webm])+bestaudio'})
370 yie = YoutubeIE(ydl)
371 yie._sort_formats(info_dict['formats'])
372 ydl.process_ie_result(info_dict)
373 downloaded_ids = [info['format_id'] for info in ydl.downloaded_info_dicts]
374 self.assertEqual(downloaded_ids, ['137+141', '248+141'])
375
376 info_dict = _make_result(list(formats_order), extractor='youtube')
377 ydl = YDL({'format': '(bestvideo[ext=mp4],bestvideo[ext=webm])[height<=720]+bestaudio'})
378 yie = YoutubeIE(ydl)
379 yie._sort_formats(info_dict['formats'])
380 ydl.process_ie_result(info_dict)
381 downloaded_ids = [info['format_id'] for info in ydl.downloaded_info_dicts]
382 self.assertEqual(downloaded_ids, ['136+141', '247+141'])
383
384 info_dict = _make_result(list(formats_order), extractor='youtube')
385 ydl = YDL({'format': '(bestvideo[ext=none]/bestvideo[ext=webm])+bestaudio'})
386 yie = YoutubeIE(ydl)
387 yie._sort_formats(info_dict['formats'])
388 ydl.process_ie_result(info_dict)
389 downloaded_ids = [info['format_id'] for info in ydl.downloaded_info_dicts]
390 self.assertEqual(downloaded_ids, ['248+141'])
391
67134eab 392 for f1, f2 in zip(formats_order, formats_order[1:]):
3537b93d 393 info_dict = _make_result([f1, f2], extractor='youtube')
8dd54188 394 ydl = YDL({'format': 'best/bestvideo'})
3d4a70b8
PH
395 yie = YoutubeIE(ydl)
396 yie._sort_formats(info_dict['formats'])
397 ydl.process_ie_result(info_dict)
398 downloaded = ydl.downloaded_info_dicts[0]
67134eab 399 self.assertEqual(downloaded['format_id'], f1['format_id'])
3d4a70b8 400
3537b93d 401 info_dict = _make_result([f2, f1], extractor='youtube')
8dd54188 402 ydl = YDL({'format': 'best/bestvideo'})
3d4a70b8
PH
403 yie = YoutubeIE(ydl)
404 yie._sort_formats(info_dict['formats'])
405 ydl.process_ie_result(info_dict)
406 downloaded = ydl.downloaded_info_dicts[0]
67134eab 407 self.assertEqual(downloaded['format_id'], f1['format_id'])
3d4a70b8 408
317f7ab6
S
409 def test_audio_only_extractor_format_selection(self):
410 # For extractors with incomplete formats (all formats are audio-only or
411 # video-only) best and worst should fallback to corresponding best/worst
412 # video-only or audio-only formats (as per
067aa17e 413 # https://github.com/ytdl-org/youtube-dl/pull/5556)
317f7ab6
S
414 formats = [
415 {'format_id': 'low', 'ext': 'mp3', 'preference': 1, 'vcodec': 'none', 'url': TEST_URL},
416 {'format_id': 'high', 'ext': 'mp3', 'preference': 2, 'vcodec': 'none', 'url': TEST_URL},
417 ]
418 info_dict = _make_result(formats)
419
420 ydl = YDL({'format': 'best'})
421 ydl.process_ie_result(info_dict.copy())
422 downloaded = ydl.downloaded_info_dicts[0]
423 self.assertEqual(downloaded['format_id'], 'high')
424
425 ydl = YDL({'format': 'worst'})
426 ydl.process_ie_result(info_dict.copy())
427 downloaded = ydl.downloaded_info_dicts[0]
428 self.assertEqual(downloaded['format_id'], 'low')
429
430 def test_format_not_available(self):
431 formats = [
432 {'format_id': 'regular', 'ext': 'mp4', 'height': 360, 'url': TEST_URL},
433 {'format_id': 'video', 'ext': 'mp4', 'height': 720, 'acodec': 'none', 'url': TEST_URL},
434 ]
435 info_dict = _make_result(formats)
436
437 # This must fail since complete video-audio format does not match filter
438 # and extractor does not provide incomplete only formats (i.e. only
439 # video-only or audio-only).
440 ydl = YDL({'format': 'best[height>360]'})
441 self.assertRaises(ExtractorError, ydl.process_ie_result, info_dict.copy())
442
8cda78ef 443 def test_format_selection_issue_10083(self):
067aa17e 444 # See https://github.com/ytdl-org/youtube-dl/issues/10083
8cda78ef
S
445 formats = [
446 {'format_id': 'regular', 'height': 360, 'url': TEST_URL},
447 {'format_id': 'video', 'height': 720, 'acodec': 'none', 'url': TEST_URL},
448 {'format_id': 'audio', 'vcodec': 'none', 'url': TEST_URL},
449 ]
450 info_dict = _make_result(formats)
451
452 ydl = YDL({'format': 'best[height>360]/bestvideo[height>360]+bestaudio'})
453 ydl.process_ie_result(info_dict.copy())
454 self.assertEqual(ydl.downloaded_info_dicts[0]['format_id'], 'video+audio')
455
0a31a350
JMF
456 def test_invalid_format_specs(self):
457 def assert_syntax_error(format_spec):
187986a8 458 self.assertRaises(SyntaxError, YDL, {'format': format_spec})
0a31a350
JMF
459
460 assert_syntax_error('bestvideo,,best')
461 assert_syntax_error('+bestaudio')
462 assert_syntax_error('bestvideo+')
d96d604e 463 assert_syntax_error('/')
187986a8 464 assert_syntax_error('[720<height]')
0a31a350 465
083c9df9
PH
466 def test_format_filtering(self):
467 formats = [
468 {'format_id': 'A', 'filesize': 500, 'width': 1000},
469 {'format_id': 'B', 'filesize': 1000, 'width': 500},
470 {'format_id': 'C', 'filesize': 1000, 'width': 400},
471 {'format_id': 'D', 'filesize': 2000, 'width': 600},
472 {'format_id': 'E', 'filesize': 3000},
473 {'format_id': 'F'},
474 {'format_id': 'G', 'filesize': 1000000},
475 ]
476 for f in formats:
477 f['url'] = 'http://_/'
478 f['ext'] = 'unknown'
479 info_dict = _make_result(formats)
480
481 ydl = YDL({'format': 'best[filesize<3000]'})
482 ydl.process_ie_result(info_dict)
483 downloaded = ydl.downloaded_info_dicts[0]
484 self.assertEqual(downloaded['format_id'], 'D')
485
486 ydl = YDL({'format': 'best[filesize<=3000]'})
487 ydl.process_ie_result(info_dict)
488 downloaded = ydl.downloaded_info_dicts[0]
489 self.assertEqual(downloaded['format_id'], 'E')
490
491 ydl = YDL({'format': 'best[filesize <= ? 3000]'})
492 ydl.process_ie_result(info_dict)
493 downloaded = ydl.downloaded_info_dicts[0]
494 self.assertEqual(downloaded['format_id'], 'F')
495
496 ydl = YDL({'format': 'best [filesize = 1000] [width>450]'})
497 ydl.process_ie_result(info_dict)
498 downloaded = ydl.downloaded_info_dicts[0]
499 self.assertEqual(downloaded['format_id'], 'B')
500
501 ydl = YDL({'format': 'best [filesize = 1000] [width!=450]'})
502 ydl.process_ie_result(info_dict)
503 downloaded = ydl.downloaded_info_dicts[0]
504 self.assertEqual(downloaded['format_id'], 'C')
505
506 ydl = YDL({'format': '[filesize>?1]'})
507 ydl.process_ie_result(info_dict)
508 downloaded = ydl.downloaded_info_dicts[0]
509 self.assertEqual(downloaded['format_id'], 'G')
510
511 ydl = YDL({'format': '[filesize<1M]'})
512 ydl.process_ie_result(info_dict)
513 downloaded = ydl.downloaded_info_dicts[0]
514 self.assertEqual(downloaded['format_id'], 'E')
515
516 ydl = YDL({'format': '[filesize<1MiB]'})
517 ydl.process_ie_result(info_dict)
518 downloaded = ydl.downloaded_info_dicts[0]
519 self.assertEqual(downloaded['format_id'], 'G')
520
5acfa126
JMF
521 ydl = YDL({'format': 'all[width>=400][width<=600]'})
522 ydl.process_ie_result(info_dict)
523 downloaded_ids = [info['format_id'] for info in ydl.downloaded_info_dicts]
dd2a987d 524 self.assertEqual(downloaded_ids, ['D', 'C', 'B'])
5acfa126 525
bb8e5536
JMF
526 ydl = YDL({'format': 'best[height<40]'})
527 try:
528 ydl.process_ie_result(info_dict)
529 except ExtractorError:
530 pass
531 self.assertEqual(ydl.downloaded_info_dicts, [])
532
0017d9ad
S
533 def test_default_format_spec(self):
534 ydl = YDL({'simulate': True})
5d0c5371 535 self.assertEqual(ydl._default_format_spec({}), 'bestvideo*+bestaudio/best')
0017d9ad 536
d08dcd2d
S
537 ydl = YDL({})
538 self.assertEqual(ydl._default_format_spec({'is_live': True}), 'best/bestvideo+bestaudio')
af0f7428 539
d08dcd2d 540 ydl = YDL({'simulate': True})
5d0c5371 541 self.assertEqual(ydl._default_format_spec({'is_live': True}), 'bestvideo*+bestaudio/best')
af0f7428 542
0017d9ad 543 ydl = YDL({'outtmpl': '-'})
af0f7428 544 self.assertEqual(ydl._default_format_spec({}), 'best/bestvideo+bestaudio')
0017d9ad
S
545
546 ydl = YDL({})
5d0c5371 547 self.assertEqual(ydl._default_format_spec({}, download=False), 'bestvideo*+bestaudio/best')
af0f7428 548 self.assertEqual(ydl._default_format_spec({'is_live': True}), 'best/bestvideo+bestaudio')
0017d9ad 549
f20bf146
JMF
550
551class TestYoutubeDL(unittest.TestCase):
ab84349b
JMF
552 def test_subtitles(self):
553 def s_formats(lang, autocaption=False):
554 return [{
555 'ext': ext,
556 'url': 'http://localhost/video.%s.%s' % (lang, ext),
557 '_auto': autocaption,
558 } for ext in ['vtt', 'srt', 'ass']]
559 subtitles = dict((l, s_formats(l)) for l in ['en', 'fr', 'es'])
560 auto_captions = dict((l, s_formats(l, True)) for l in ['it', 'pt', 'es'])
561 info_dict = {
562 'id': 'test',
563 'title': 'Test',
564 'url': 'http://localhost/video.mp4',
565 'subtitles': subtitles,
566 'automatic_captions': auto_captions,
567 'extractor': 'TEST',
732044af 568 'webpage_url': 'http://example.com/watch?v=shenanigans',
ab84349b
JMF
569 }
570
571 def get_info(params={}):
572 params.setdefault('simulate', True)
573 ydl = YDL(params)
574 ydl.report_warning = lambda *args, **kargs: None
575 return ydl.process_video_result(info_dict, download=False)
576
577 result = get_info()
578 self.assertFalse(result.get('requested_subtitles'))
579 self.assertEqual(result['subtitles'], subtitles)
580 self.assertEqual(result['automatic_captions'], auto_captions)
581
582 result = get_info({'writesubtitles': True})
583 subs = result['requested_subtitles']
584 self.assertTrue(subs)
585 self.assertEqual(set(subs.keys()), set(['en']))
586 self.assertTrue(subs['en'].get('data') is None)
587 self.assertEqual(subs['en']['ext'], 'ass')
588
589 result = get_info({'writesubtitles': True, 'subtitlesformat': 'foo/srt'})
590 subs = result['requested_subtitles']
591 self.assertEqual(subs['en']['ext'], 'srt')
592
593 result = get_info({'writesubtitles': True, 'subtitleslangs': ['es', 'fr', 'it']})
594 subs = result['requested_subtitles']
595 self.assertTrue(subs)
596 self.assertEqual(set(subs.keys()), set(['es', 'fr']))
597
c32b0aab 598 result = get_info({'writesubtitles': True, 'subtitleslangs': ['all', '-en']})
599 subs = result['requested_subtitles']
600 self.assertTrue(subs)
601 self.assertEqual(set(subs.keys()), set(['es', 'fr']))
602
603 result = get_info({'writesubtitles': True, 'subtitleslangs': ['en', 'fr', '-en']})
604 subs = result['requested_subtitles']
605 self.assertTrue(subs)
606 self.assertEqual(set(subs.keys()), set(['fr']))
607
608 result = get_info({'writesubtitles': True, 'subtitleslangs': ['-en', 'en']})
609 subs = result['requested_subtitles']
610 self.assertTrue(subs)
611 self.assertEqual(set(subs.keys()), set(['en']))
612
613 result = get_info({'writesubtitles': True, 'subtitleslangs': ['e.+']})
614 subs = result['requested_subtitles']
615 self.assertTrue(subs)
616 self.assertEqual(set(subs.keys()), set(['es', 'en']))
617
ab84349b
JMF
618 result = get_info({'writesubtitles': True, 'writeautomaticsub': True, 'subtitleslangs': ['es', 'pt']})
619 subs = result['requested_subtitles']
620 self.assertTrue(subs)
621 self.assertEqual(set(subs.keys()), set(['es', 'pt']))
622 self.assertFalse(subs['es']['_auto'])
623 self.assertTrue(subs['pt']['_auto'])
624
98c70d6f
JMF
625 result = get_info({'writeautomaticsub': True, 'subtitleslangs': ['es', 'pt']})
626 subs = result['requested_subtitles']
627 self.assertTrue(subs)
628 self.assertEqual(set(subs.keys()), set(['es', 'pt']))
629 self.assertTrue(subs['es']['_auto'])
630 self.assertTrue(subs['pt']['_auto'])
631
b6c45014
JMF
632 def test_add_extra_info(self):
633 test_dict = {
634 'extractor': 'Foo',
635 }
636 extra_info = {
637 'extractor': 'Bar',
638 'playlist': 'funny videos',
639 }
640 YDL.add_extra_info(test_dict, extra_info)
641 self.assertEqual(test_dict['extractor'], 'Foo')
642 self.assertEqual(test_dict['playlist'], 'funny videos')
643
752cda38 644 outtmpl_info = {
645 'id': '1234',
646 'ext': 'mp4',
647 'width': None,
648 'height': 1080,
abbeeebc 649 'filesize': 1024,
752cda38 650 'title1': '$PATH',
651 'title2': '%PATH%',
87ea7dfc 652 'title3': 'foo/bar\\test',
7d1eb38a 653 'title4': 'foo "bar" test',
524e2e4f 654 'title5': 'áéí 𝐀',
752cda38 655 'timestamp': 1618488000,
656 'duration': 100000,
657 'playlist_index': 1,
e6f21b3d 658 'playlist_autonumber': 2,
752cda38 659 '_last_playlist_index': 100,
660 'n_entries': 10,
4476d2c7 661 'formats': [{'id': 'id 1'}, {'id': 'id 2'}, {'id': 'id 3'}]
752cda38 662 }
663
5c6542ce 664 def test_prepare_outtmpl_and_filename(self):
9fea350f 665 def test(tmpl, expected, *, info=None, **params):
752cda38 666 params['outtmpl'] = tmpl
667 ydl = YoutubeDL(params)
668 ydl._num_downloads = 1
5c6542ce 669 self.assertEqual(ydl.validate_outtmpl(tmpl), None)
752cda38 670
819e0531 671 out = ydl.evaluate_outtmpl(tmpl, info or self.outtmpl_info)
9fea350f 672 fname = ydl.prepare_filename(info or self.outtmpl_info)
5c6542ce 673
2b8a2973 674 if not isinstance(expected, (list, tuple)):
675 expected = (expected, expected)
676 for (name, got), expect in zip((('outtmpl', out), ('filename', fname)), expected):
677 if callable(expect):
678 self.assertTrue(expect(got), f'Wrong {name} from {tmpl}')
679 else:
680 self.assertEqual(got, expect, f'Wrong {name} from {tmpl}')
5c6542ce 681
6e84b215 682 # Side-effects
683 original_infodict = dict(self.outtmpl_info)
684 test('foo.bar', 'foo.bar')
685 original_infodict['epoch'] = self.outtmpl_info.get('epoch')
686 self.assertTrue(isinstance(original_infodict['epoch'], int))
687 test('%(epoch)d', int_or_none)
688 self.assertEqual(original_infodict, self.outtmpl_info)
689
5c6542ce 690 # Auto-generated fields
691 test('%(id)s.%(ext)s', '1234.mp4')
692 test('%(duration_string)s', ('27:46:40', '27-46-40'))
5c6542ce 693 test('%(resolution)s', '1080p')
694 test('%(playlist_index)s', '001')
e6f21b3d 695 test('%(playlist_autonumber)s', '02')
5c6542ce 696 test('%(autonumber)s', '00001')
697 test('%(autonumber+2)03d', '005', autonumber_start=3)
698 test('%(autonumber)s', '001', autonumber_size=3)
699
700 # Escaping %
901130bb 701 test('%', '%')
5c6542ce 702 test('%%', '%')
703 test('%%%%', '%%')
901130bb 704 test('%s', '%s')
705 test('%%%s', '%%s')
706 test('%d', '%d')
707 test('%abc%', '%abc%')
5c6542ce 708 test('%%(width)06d.%(ext)s', '%(width)06d.mp4')
901130bb 709 test('%%%(height)s', '%1080')
5c6542ce 710 test('%(width)06d.%(ext)s', 'NA.mp4')
711 test('%(width)06d.%%(ext)s', 'NA.%(ext)s')
712 test('%%(width)06d.%(ext)s', '%(width)06d.mp4')
713
9fea350f 714 # ID sanitization
715 test('%(id)s', '_abcd', info={'id': '_abcd'})
716 test('%(some_id)s', '_abcd', info={'some_id': '_abcd'})
717 test('%(formats.0.id)s', '_abcd', info={'formats': [{'id': '_abcd'}]})
718 test('%(id)s', '-abcd', info={'id': '-abcd'})
719 test('%(id)s', '.abcd', info={'id': '.abcd'})
720 test('%(id)s', 'ab__cd', info={'id': 'ab__cd'})
721 test('%(id)s', ('ab:cd', 'ab -cd'), info={'id': 'ab:cd'})
e0fd9573 722 test('%(id.0)s', '-', info={'id': '--'})
9fea350f 723
5c6542ce 724 # Invalid templates
5c6542ce 725 self.assertTrue(isinstance(YoutubeDL.validate_outtmpl('%(title)'), ValueError))
726 test('%(invalid@tmpl|def)s', 'none', outtmpl_na_placeholder='none')
2b8a2973 727 test('%(..)s', 'NA')
728
729 # Entire info_dict
730 def expect_same_infodict(out):
731 got_dict = json.loads(out)
732 for info_field, expected in self.outtmpl_info.items():
733 self.assertEqual(got_dict.get(info_field), expected, info_field)
734 return True
735
736 test('%()j', (expect_same_infodict, str))
5c6542ce 737
738 # NA placeholder
752cda38 739 NA_TEST_OUTTMPL = '%(uploader_date)s-%(width)d-%(x|def)s-%(id)s.%(ext)s'
5c6542ce 740 test(NA_TEST_OUTTMPL, 'NA-NA-def-1234.mp4')
741 test(NA_TEST_OUTTMPL, 'none-none-def-1234.mp4', outtmpl_na_placeholder='none')
742 test(NA_TEST_OUTTMPL, '--def-1234.mp4', outtmpl_na_placeholder='')
582fad70 743 test('%(non_existent.0)s', 'NA')
752cda38 744
5c6542ce 745 # String formatting
752cda38 746 FMT_TEST_OUTTMPL = '%%(height)%s.%%(ext)s'
5c6542ce 747 test(FMT_TEST_OUTTMPL % 's', '1080.mp4')
748 test(FMT_TEST_OUTTMPL % 'd', '1080.mp4')
749 test(FMT_TEST_OUTTMPL % '6d', ' 1080.mp4')
750 test(FMT_TEST_OUTTMPL % '-6d', '1080 .mp4')
751 test(FMT_TEST_OUTTMPL % '06d', '001080.mp4')
752 test(FMT_TEST_OUTTMPL % ' 06d', ' 01080.mp4')
753 test(FMT_TEST_OUTTMPL % ' 06d', ' 01080.mp4')
754 test(FMT_TEST_OUTTMPL % '0 6d', ' 01080.mp4')
755 test(FMT_TEST_OUTTMPL % '0 6d', ' 01080.mp4')
756 test(FMT_TEST_OUTTMPL % ' 0 6d', ' 01080.mp4')
757
758 # Type casting
759 test('%(id)d', '1234')
760 test('%(height)c', '1')
761 test('%(ext)c', 'm')
762 test('%(id)d %(id)r', "1234 '1234'")
763 test('%(id)r %(height)r', "'1234' 1080")
764 test('%(ext)s-%(ext|def)d', 'mp4-def')
765 test('%(width|0)04d', '0000')
766 test('a%(width|)d', 'a', outtmpl_na_placeholder='none')
767
752cda38 768 FORMATS = self.outtmpl_info['formats']
4476d2c7 769 sanitize = lambda x: x.replace(':', ' -').replace('"', "'").replace('\n', ' ')
7d1eb38a 770
771 # Custom type casting
4476d2c7 772 test('%(formats.:.id)l', 'id 1, id 2, id 3')
773 test('%(formats.:.id)#l', ('id 1\nid 2\nid 3', 'id 1 id 2 id 3'))
7d1eb38a 774 test('%(ext)l', 'mp4')
4476d2c7 775 test('%(formats.:.id) 18l', ' id 1, id 2, id 3')
7d1eb38a 776 test('%(formats)j', (json.dumps(FORMATS), sanitize(json.dumps(FORMATS))))
4476d2c7 777 test('%(formats)#j', (json.dumps(FORMATS, indent=4), sanitize(json.dumps(FORMATS, indent=4))))
f5aa5cfb 778 test('%(title5).3B', 'á')
524e2e4f 779 test('%(title5)U', 'áéí 𝐀')
780 test('%(title5)#U', 'a\u0301e\u0301i\u0301 𝐀')
781 test('%(title5)+U', 'áéí A')
782 test('%(title5)+#U', 'a\u0301e\u0301i\u0301 A')
abbeeebc 783 test('%(height)D', '1k')
784 test('%(filesize)#D', '1Ki')
785 test('%(height)5.2D', ' 1.08k')
37893bb0 786 test('%(title4)#S', 'foo_bar_test')
787 test('%(title4).10S', ('foo \'bar\' ', 'foo \'bar\'' + ('#' if compat_os_name == 'nt' else ' ')))
7d1eb38a 788 if compat_os_name == 'nt':
789 test('%(title4)q', ('"foo \\"bar\\" test"', "'foo _'bar_' test'"))
4476d2c7 790 test('%(formats.:.id)#q', ('"id 1" "id 2" "id 3"', "'id 1' 'id 2' 'id 3'"))
791 test('%(formats.0.id)#q', ('"id 1"', "'id 1'"))
7d1eb38a 792 else:
793 test('%(title4)q', ('\'foo "bar" test\'', "'foo 'bar' test'"))
4476d2c7 794 test('%(formats.:.id)#q', "'id 1' 'id 2' 'id 3'")
795 test('%(formats.0.id)#q', "'id 1'")
7d1eb38a 796
797 # Internal formatting
5c6542ce 798 test('%(timestamp-1000>%H-%M-%S)s', '11-43-20')
901130bb 799 test('%(title|%)s %(title|%%)s', '% %%')
5c6542ce 800 test('%(id+1-height+3)05d', '00158')
801 test('%(width+100)05d', 'NA')
7d1eb38a 802 test('%(formats.0) 15s', ('% 15s' % FORMATS[0], '% 15s' % sanitize(str(FORMATS[0]))))
803 test('%(formats.0)r', (repr(FORMATS[0]), sanitize(repr(FORMATS[0]))))
5c6542ce 804 test('%(height.0)03d', '001')
805 test('%(-height.0)04d', '-001')
806 test('%(formats.-1.id)s', FORMATS[-1]['id'])
807 test('%(formats.0.id.-1)d', FORMATS[0]['id'][-1])
808 test('%(formats.3)s', 'NA')
809 test('%(formats.:2:-1)r', repr(FORMATS[:2:-1]))
810 test('%(formats.0.id.-1+id)f', '1235.000000')
385a27fa 811 test('%(formats.0.id.-1+formats.1.id.-1)d', '3')
5c6542ce 812
7c37ff97 813 # Alternates
814 test('%(title,id)s', '1234')
815 test('%(width-100,height+20|def)d', '1100')
816 test('%(width-100,height+width|def)s', 'def')
817 test('%(timestamp-x>%H\\,%M\\,%S,timestamp>%H\\,%M\\,%S)s', '12,00,00')
818
e0fd9573 819 # Replacement
820 test('%(id&foo)s.bar', 'foo.bar')
821 test('%(title&foo)s.bar', 'NA.bar')
822 test('%(title&foo|baz)s.bar', 'baz.bar')
823
6e84b215 824 # Laziness
825 def gen():
826 yield from range(5)
827 raise self.assertTrue(False, 'LazyList should not be evaluated till here')
828 test('%(key.4)s', '4', info={'key': LazyList(gen())})
829
5c6542ce 830 # Empty filename
831 test('%(foo|)s-%(bar|)s.%(ext)s', '-.mp4')
832 # test('%(foo|)s.%(ext)s', ('.mp4', '_.mp4')) # fixme
833 # test('%(foo|)s', ('', '_')) # fixme
834
901130bb 835 # Environment variable expansion for prepare_filename
836 compat_setenv('__yt_dlp_var', 'expanded')
837 envvar = '%__yt_dlp_var%' if compat_os_name == 'nt' else '$__yt_dlp_var'
838 test(envvar, (envvar, 'expanded'))
b836dc94 839 if compat_os_name == 'nt':
840 test('%s%', ('%s%', '%s%'))
841 compat_setenv('s', 'expanded')
842 test('%s%', ('%s%', 'expanded')) # %s% should be expanded before escaping %s
843 compat_setenv('(test)s', 'expanded')
844 test('%(test)s%', ('NA%', 'expanded')) # Environment should take priority over template
901130bb 845
5c6542ce 846 # Path expansion and escaping
847 test('Hello %(title1)s', 'Hello $PATH')
848 test('Hello %(title2)s', 'Hello %PATH%')
849 test('%(title3)s', ('foo/bar\\test', 'foo_bar_test'))
850 test('folder/%(title3)s', ('folder/foo/bar\\test', 'folder%sfoo_bar_test' % os.path.sep))
26e63931 851
c57f7757
PH
852 def test_format_note(self):
853 ydl = YoutubeDL()
854 self.assertEqual(ydl._format_note({}), '')
855 assertRegexpMatches(self, ydl._format_note({
856 'vbr': 10,
398dea32 857 }), r'^\s*10k$')
5d583bdf
S
858 assertRegexpMatches(self, ydl._format_note({
859 'fps': 30,
398dea32 860 }), r'^30fps$')
5d583bdf 861
2b4ecde2
JMF
862 def test_postprocessors(self):
863 filename = 'post-processor-testfile.mp4'
864 audiofile = filename + '.mp3'
865
866 class SimplePP(PostProcessor):
867 def run(self, info):
2b4ecde2
JMF
868 with open(audiofile, 'wt') as f:
869 f.write('EXAMPLE')
592e97e8 870 return [info['filepath']], info
2b4ecde2 871
592e97e8 872 def run_pp(params, PP):
2b4ecde2
JMF
873 with open(filename, 'wt') as f:
874 f.write('EXAMPLE')
875 ydl = YoutubeDL(params)
592e97e8 876 ydl.add_post_processor(PP())
2b4ecde2
JMF
877 ydl.post_process(filename, {'filepath': filename})
878
592e97e8 879 run_pp({'keepvideo': True}, SimplePP)
2b4ecde2
JMF
880 self.assertTrue(os.path.exists(filename), '%s doesn\'t exist' % filename)
881 self.assertTrue(os.path.exists(audiofile), '%s doesn\'t exist' % audiofile)
882 os.unlink(filename)
883 os.unlink(audiofile)
884
592e97e8 885 run_pp({'keepvideo': False}, SimplePP)
2b4ecde2
JMF
886 self.assertFalse(os.path.exists(filename), '%s exists' % filename)
887 self.assertTrue(os.path.exists(audiofile), '%s doesn\'t exist' % audiofile)
888 os.unlink(audiofile)
889
592e97e8
JMF
890 class ModifierPP(PostProcessor):
891 def run(self, info):
892 with open(info['filepath'], 'wt') as f:
893 f.write('MODIFIED')
894 return [], info
895
896 run_pp({'keepvideo': False}, ModifierPP)
897 self.assertTrue(os.path.exists(filename), '%s doesn\'t exist' % filename)
898 os.unlink(filename)
899
531980d8
JMF
900 def test_match_filter(self):
901 class FilterYDL(YDL):
902 def __init__(self, *args, **kwargs):
903 super(FilterYDL, self).__init__(*args, **kwargs)
904 self.params['simulate'] = True
905
906 def process_info(self, info_dict):
907 super(YDL, self).process_info(info_dict)
908
c77495e3 909 def _match_entry(self, info_dict, incomplete=False):
531980d8
JMF
910 res = super(FilterYDL, self)._match_entry(info_dict, incomplete)
911 if res is None:
f46e2f9d 912 self.downloaded_info_dicts.append(info_dict.copy())
531980d8
JMF
913 return res
914
915 first = {
916 'id': '1',
917 'url': TEST_URL,
918 'title': 'one',
919 'extractor': 'TEST',
920 'duration': 30,
921 'filesize': 10 * 1024,
e5a088dc 922 'playlist_id': '42',
db13c16e
S
923 'uploader': "變態妍字幕版 太妍 тест",
924 'creator': "тест ' 123 ' тест--",
732044af 925 'webpage_url': 'http://example.com/watch?v=shenanigans',
531980d8
JMF
926 }
927 second = {
928 'id': '2',
929 'url': TEST_URL,
930 'title': 'two',
931 'extractor': 'TEST',
932 'duration': 10,
933 'description': 'foo',
934 'filesize': 5 * 1024,
e5a088dc 935 'playlist_id': '43',
db13c16e 936 'uploader': "тест 123",
732044af 937 'webpage_url': 'http://example.com/watch?v=SHENANIGANS',
531980d8
JMF
938 }
939 videos = [first, second]
940
941 def get_videos(filter_=None):
942 ydl = FilterYDL({'match_filter': filter_})
943 for v in videos:
944 ydl.process_ie_result(v, download=True)
945 return [v['id'] for v in ydl.downloaded_info_dicts]
946
947 res = get_videos()
948 self.assertEqual(res, ['1', '2'])
949
950 def f(v):
951 if v['id'] == '1':
952 return None
953 else:
954 return 'Video id is not 1'
955 res = get_videos(f)
956 self.assertEqual(res, ['1'])
957
958 f = match_filter_func('duration < 30')
959 res = get_videos(f)
960 self.assertEqual(res, ['2'])
961
962 f = match_filter_func('description = foo')
963 res = get_videos(f)
964 self.assertEqual(res, ['2'])
965
966 f = match_filter_func('description =? foo')
967 res = get_videos(f)
968 self.assertEqual(res, ['1', '2'])
969
970 f = match_filter_func('filesize > 5KiB')
971 res = get_videos(f)
972 self.assertEqual(res, ['1'])
973
e5a088dc
S
974 f = match_filter_func('playlist_id = 42')
975 res = get_videos(f)
976 self.assertEqual(res, ['1'])
977
db13c16e
S
978 f = match_filter_func('uploader = "變態妍字幕版 太妍 тест"')
979 res = get_videos(f)
980 self.assertEqual(res, ['1'])
981
982 f = match_filter_func('uploader != "變態妍字幕版 太妍 тест"')
983 res = get_videos(f)
984 self.assertEqual(res, ['2'])
985
986 f = match_filter_func('creator = "тест \' 123 \' тест--"')
987 res = get_videos(f)
988 self.assertEqual(res, ['1'])
989
990 f = match_filter_func("creator = 'тест \\' 123 \\' тест--'")
991 res = get_videos(f)
992 self.assertEqual(res, ['1'])
993
994 f = match_filter_func(r"creator = 'тест \' 123 \' тест--' & duration > 30")
995 res = get_videos(f)
996 self.assertEqual(res, [])
997
e9eaf3fb
JMF
998 def test_playlist_items_selection(self):
999 entries = [{
1000 'id': compat_str(i),
1001 'title': compat_str(i),
1002 'url': TEST_URL,
1003 } for i in range(1, 5)]
1004 playlist = {
1005 '_type': 'playlist',
1006 'id': 'test',
1007 'entries': entries,
1008 'extractor': 'test:playlist',
1009 'extractor_key': 'test:playlist',
1010 'webpage_url': 'http://example.com',
1011 }
1012
4e9e1e24 1013 def get_downloaded_info_dicts(params):
e9eaf3fb 1014 ydl = YDL(params)
e0abaab2
S
1015 # make a deep copy because the dictionary and nested entries
1016 # can be modified
1017 ydl.process_ie_result(copy.deepcopy(playlist))
4e9e1e24
S
1018 return ydl.downloaded_info_dicts
1019
9e598870 1020 def test_selection(params, expected_ids):
1021 results = [
1022 (v['playlist_autonumber'] - 1, (int(v['id']), v['playlist_index']))
1023 for v in get_downloaded_info_dicts(params)]
1024 self.assertEqual(results, list(enumerate(zip(expected_ids, expected_ids))))
1025
1026 test_selection({}, [1, 2, 3, 4])
1027 test_selection({'playlistend': 10}, [1, 2, 3, 4])
1028 test_selection({'playlistend': 2}, [1, 2])
1029 test_selection({'playliststart': 10}, [])
1030 test_selection({'playliststart': 2}, [2, 3, 4])
1031 test_selection({'playlist_items': '2-4'}, [2, 3, 4])
1032 test_selection({'playlist_items': '2,4'}, [2, 4])
1033 test_selection({'playlist_items': '10'}, [])
ff1c7fc9 1034 test_selection({'playlist_items': '0'}, [])
cd6fc19e 1035
4e9e1e24 1036 # Tests for https://github.com/ytdl-org/youtube-dl/issues/10591
9e598870 1037 test_selection({'playlist_items': '2-4,3-4,3'}, [2, 3, 4])
1038 test_selection({'playlist_items': '4,2'}, [4, 2])
1039
1040 # Tests for https://github.com/yt-dlp/yt-dlp/issues/720
1041 # https://github.com/yt-dlp/yt-dlp/issues/302
1042 test_selection({'playlistreverse': True}, [4, 3, 2, 1])
1043 test_selection({'playliststart': 2, 'playlistreverse': True}, [4, 3, 2])
1044 test_selection({'playlist_items': '2,4', 'playlistreverse': True}, [4, 2])
1045 test_selection({'playlist_items': '4,2'}, [4, 2])
4e9e1e24 1046
e37afbe0 1047 def test_urlopen_no_file_protocol(self):
067aa17e 1048 # see https://github.com/ytdl-org/youtube-dl/issues/8227
e37afbe0
JMF
1049 ydl = YDL()
1050 self.assertRaises(compat_urllib_error.URLError, ydl.urlopen, 'file:///etc/passwd')
1051
b286f201
YCH
1052 def test_do_not_override_ie_key_in_url_transparent(self):
1053 ydl = YDL()
1054
1055 class Foo1IE(InfoExtractor):
1056 _VALID_URL = r'foo1:'
1057
1058 def _real_extract(self, url):
1059 return {
1060 '_type': 'url_transparent',
1061 'url': 'foo2:',
1062 'ie_key': 'Foo2',
0396806f
S
1063 'title': 'foo1 title',
1064 'id': 'foo1_id',
b286f201
YCH
1065 }
1066
1067 class Foo2IE(InfoExtractor):
1068 _VALID_URL = r'foo2:'
1069
1070 def _real_extract(self, url):
1071 return {
1072 '_type': 'url',
1073 'url': 'foo3:',
1074 'ie_key': 'Foo3',
1075 }
1076
1077 class Foo3IE(InfoExtractor):
1078 _VALID_URL = r'foo3:'
1079
1080 def _real_extract(self, url):
51350db5 1081 return _make_result([{'url': TEST_URL}], title='foo3 title')
b286f201
YCH
1082
1083 ydl.add_info_extractor(Foo1IE(ydl))
1084 ydl.add_info_extractor(Foo2IE(ydl))
1085 ydl.add_info_extractor(Foo3IE(ydl))
1086 ydl.extract_info('foo1:')
1087 downloaded = ydl.downloaded_info_dicts[0]
1088 self.assertEqual(downloaded['url'], TEST_URL)
51350db5 1089 self.assertEqual(downloaded['title'], 'foo1 title')
0396806f
S
1090 self.assertEqual(downloaded['id'], 'testid')
1091 self.assertEqual(downloaded['extractor'], 'testex')
1092 self.assertEqual(downloaded['extractor_key'], 'TestEx')
b286f201 1093
a0566bbf 1094 # Test case for https://github.com/ytdl-org/youtube-dl/issues/27064
1095 def test_ignoreerrors_for_playlist_with_url_transparent_iterable_entries(self):
1096
1097 class _YDL(YDL):
1098 def __init__(self, *args, **kwargs):
1099 super(_YDL, self).__init__(*args, **kwargs)
1100
1101 def trouble(self, s, tb=None):
1102 pass
1103
1104 ydl = _YDL({
1105 'format': 'extra',
1106 'ignoreerrors': True,
1107 })
1108
1109 class VideoIE(InfoExtractor):
1110 _VALID_URL = r'video:(?P<id>\d+)'
1111
1112 def _real_extract(self, url):
1113 video_id = self._match_id(url)
1114 formats = [{
1115 'format_id': 'default',
1116 'url': 'url:',
1117 }]
1118 if video_id == '0':
1119 raise ExtractorError('foo')
1120 if video_id == '2':
1121 formats.append({
1122 'format_id': 'extra',
1123 'url': TEST_URL,
1124 })
1125 return {
1126 'id': video_id,
1127 'title': 'Video %s' % video_id,
1128 'formats': formats,
1129 }
1130
1131 class PlaylistIE(InfoExtractor):
1132 _VALID_URL = r'playlist:'
1133
1134 def _entries(self):
1135 for n in range(3):
1136 video_id = compat_str(n)
1137 yield {
1138 '_type': 'url_transparent',
1139 'ie_key': VideoIE.ie_key(),
1140 'id': video_id,
1141 'url': 'video:%s' % video_id,
1142 'title': 'Video Transparent %s' % video_id,
1143 }
1144
1145 def _real_extract(self, url):
1146 return self.playlist_result(self._entries())
1147
1148 ydl.add_info_extractor(VideoIE(ydl))
1149 ydl.add_info_extractor(PlaylistIE(ydl))
1150 info = ydl.extract_info('playlist:')
1151 entries = info['entries']
1152 self.assertEqual(len(entries), 3)
1153 self.assertTrue(entries[0] is None)
1154 self.assertTrue(entries[1] is None)
1155 self.assertEqual(len(ydl.downloaded_info_dicts), 1)
1156 downloaded = ydl.downloaded_info_dicts[0]
9e907ebd 1157 entries[2].pop('requested_downloads', None)
a0566bbf 1158 self.assertEqual(entries[2], downloaded)
1159 self.assertEqual(downloaded['url'], TEST_URL)
1160 self.assertEqual(downloaded['title'], 'Video Transparent 2')
1161 self.assertEqual(downloaded['id'], '2')
1162 self.assertEqual(downloaded['extractor'], 'Video')
1163 self.assertEqual(downloaded['extractor_key'], 'Video')
1164
2b4ecde2 1165
e028d0d1
JMF
1166if __name__ == '__main__':
1167 unittest.main()