]> jfr.im git - yt-dlp.git/blame - test/test_YoutubeDL.py
Release 2024.04.09
[yt-dlp.git] / test / test_YoutubeDL.py
CommitLineData
cc52de43 1#!/usr/bin/env python3
54007a45 2
5d254f77
PH
3# Allow direct execution
4import os
e028d0d1
JMF
5import sys
6import unittest
f8271158 7
5d254f77 8sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
e028d0d1 9
54007a45 10
0217c783 11import copy
7d1eb38a 12import json
f8271158 13
6c5211ce 14from test.helper import FakeYDL, assertRegexpMatches, try_rm
7a5c1cfe 15from yt_dlp import YoutubeDL
14f25df2 16from yt_dlp.compat import compat_os_name
7a5c1cfe
P
17from yt_dlp.extractor import YoutubeIE
18from yt_dlp.extractor.common import InfoExtractor
19from yt_dlp.postprocessor.common import PostProcessor
f8271158 20from yt_dlp.utils import (
21 ExtractorError,
22 LazyList,
7e88d7d7 23 OnDemandPagedList,
f8271158 24 int_or_none,
25 match_filter_func,
26)
6c5211ce 27from yt_dlp.utils.traversal import traverse_obj
28
8508557e
JMF
29TEST_URL = 'http://localhost/sample.mp4'
30
e028d0d1
JMF
31
32class YDL(FakeYDL):
f4d96df0 33 def __init__(self, *args, **kwargs):
86e5f3ed 34 super().__init__(*args, **kwargs)
e028d0d1 35 self.downloaded_info_dicts = []
f4d96df0 36 self.msgs = []
5d254f77 37
e028d0d1 38 def process_info(self, info_dict):
09b49e1f 39 self.downloaded_info_dicts.append(info_dict.copy())
e028d0d1 40
f0500bd1 41 def to_screen(self, msg, *args, **kwargs):
f4d96df0
PH
42 self.msgs.append(msg)
43
b5ac45b1 44 def dl(self, *args, **kwargs):
45 assert False, 'Downloader must not be invoked for test_YoutubeDL'
46
5d254f77 47
3537b93d
PH
48def _make_result(formats, **kwargs):
49 res = {
50 'formats': formats,
51 'id': 'testid',
52 'title': 'testttitle',
53 'extractor': 'testex',
0396806f 54 'extractor_key': 'TestEx',
732044af 55 'webpage_url': 'http://example.com/watch?v=shenanigans',
3537b93d
PH
56 }
57 res.update(**kwargs)
58 return res
59
60
e028d0d1
JMF
61class TestFormatSelection(unittest.TestCase):
62 def test_prefer_free_formats(self):
63 # Same resolution => download webm
64 ydl = YDL()
65 ydl.params['prefer_free_formats'] = True
5d254f77 66 formats = [
8508557e
JMF
67 {'ext': 'webm', 'height': 460, 'url': TEST_URL},
68 {'ext': 'mp4', 'height': 460, 'url': TEST_URL},
5d254f77 69 ]
3537b93d 70 info_dict = _make_result(formats)
9f14daf2 71 ydl.sort_formats(info_dict)
e028d0d1
JMF
72 ydl.process_ie_result(info_dict)
73 downloaded = ydl.downloaded_info_dicts[0]
89087418 74 self.assertEqual(downloaded['ext'], 'webm')
e028d0d1
JMF
75
76 # Different resolution => download best quality (mp4)
77 ydl = YDL()
78 ydl.params['prefer_free_formats'] = True
5d254f77 79 formats = [
8508557e
JMF
80 {'ext': 'webm', 'height': 720, 'url': TEST_URL},
81 {'ext': 'mp4', 'height': 1080, 'url': TEST_URL},
5d254f77 82 ]
89087418 83 info_dict['formats'] = formats
9f14daf2 84 ydl.sort_formats(info_dict)
e028d0d1
JMF
85 ydl.process_ie_result(info_dict)
86 downloaded = ydl.downloaded_info_dicts[0]
89087418 87 self.assertEqual(downloaded['ext'], 'mp4')
e028d0d1 88
5d0c5371 89 # No prefer_free_formats => prefer mp4 and webm
e028d0d1
JMF
90 ydl = YDL()
91 ydl.params['prefer_free_formats'] = False
5d254f77 92 formats = [
8508557e
JMF
93 {'ext': 'webm', 'height': 720, 'url': TEST_URL},
94 {'ext': 'mp4', 'height': 720, 'url': TEST_URL},
95 {'ext': 'flv', 'height': 720, 'url': TEST_URL},
5d254f77 96 ]
89087418 97 info_dict['formats'] = formats
9f14daf2 98 ydl.sort_formats(info_dict)
3d4a70b8
PH
99 ydl.process_ie_result(info_dict)
100 downloaded = ydl.downloaded_info_dicts[0]
89087418 101 self.assertEqual(downloaded['ext'], 'mp4')
3d4a70b8
PH
102
103 ydl = YDL()
104 ydl.params['prefer_free_formats'] = False
105 formats = [
8508557e
JMF
106 {'ext': 'flv', 'height': 720, 'url': TEST_URL},
107 {'ext': 'webm', 'height': 720, 'url': TEST_URL},
3d4a70b8 108 ]
89087418 109 info_dict['formats'] = formats
9f14daf2 110 ydl.sort_formats(info_dict)
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 = [
9f14daf2 117 {'format_id': '35', 'ext': 'mp4', 'preference': 0, '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)
00cdda4f
SS
143 # See: https://github.com/yt-dlp/yt-dlp/pulls/8797
144 test('7_a/worst', '35')
232541df 145
ba7678f9
PH
146 def test_format_selection_audio(self):
147 formats = [
8508557e
JMF
148 {'format_id': 'audio-low', 'ext': 'webm', 'preference': 1, 'vcodec': 'none', 'url': TEST_URL},
149 {'format_id': 'audio-mid', 'ext': 'webm', 'preference': 2, 'vcodec': 'none', 'url': TEST_URL},
150 {'format_id': 'audio-high', 'ext': 'flv', 'preference': 3, 'vcodec': 'none', 'url': TEST_URL},
151 {'format_id': 'vid', 'ext': 'mp4', 'preference': 4, 'url': TEST_URL},
ba7678f9 152 ]
3537b93d 153 info_dict = _make_result(formats)
ba7678f9 154
89087418 155 ydl = YDL({'format': 'bestaudio'})
ba7678f9
PH
156 ydl.process_ie_result(info_dict.copy())
157 downloaded = ydl.downloaded_info_dicts[0]
89087418 158 self.assertEqual(downloaded['format_id'], 'audio-high')
ba7678f9 159
89087418 160 ydl = YDL({'format': 'worstaudio'})
ba7678f9
PH
161 ydl.process_ie_result(info_dict.copy())
162 downloaded = ydl.downloaded_info_dicts[0]
89087418 163 self.assertEqual(downloaded['format_id'], 'audio-low')
ba7678f9
PH
164
165 formats = [
8508557e
JMF
166 {'format_id': 'vid-low', 'ext': 'mp4', 'preference': 1, 'url': TEST_URL},
167 {'format_id': 'vid-high', 'ext': 'mp4', 'preference': 2, 'url': TEST_URL},
ba7678f9 168 ]
3537b93d 169 info_dict = _make_result(formats)
ba7678f9 170
89087418 171 ydl = YDL({'format': 'bestaudio/worstaudio/best'})
ba7678f9
PH
172 ydl.process_ie_result(info_dict.copy())
173 downloaded = ydl.downloaded_info_dicts[0]
89087418 174 self.assertEqual(downloaded['format_id'], 'vid-high')
ba7678f9 175
0217c783
PH
176 def test_format_selection_audio_exts(self):
177 formats = [
178 {'format_id': 'mp3-64', 'ext': 'mp3', 'abr': 64, 'url': 'http://_', 'vcodec': 'none'},
179 {'format_id': 'ogg-64', 'ext': 'ogg', 'abr': 64, 'url': 'http://_', 'vcodec': 'none'},
180 {'format_id': 'aac-64', 'ext': 'aac', 'abr': 64, 'url': 'http://_', 'vcodec': 'none'},
181 {'format_id': 'mp3-32', 'ext': 'mp3', 'abr': 32, 'url': 'http://_', 'vcodec': 'none'},
182 {'format_id': 'aac-32', 'ext': 'aac', 'abr': 32, 'url': 'http://_', 'vcodec': 'none'},
183 ]
184
185 info_dict = _make_result(formats)
86a97203 186 ydl = YDL({'format': 'best', 'format_sort': ['abr', 'ext']})
9f14daf2 187 ydl.sort_formats(info_dict)
0217c783
PH
188 ydl.process_ie_result(copy.deepcopy(info_dict))
189 downloaded = ydl.downloaded_info_dicts[0]
190 self.assertEqual(downloaded['format_id'], 'aac-64')
191
192 ydl = YDL({'format': 'mp3'})
9f14daf2 193 ydl.sort_formats(info_dict)
0217c783
PH
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
86a97203 198 ydl = YDL({'prefer_free_formats': True, 'format_sort': ['abr', 'ext']})
9f14daf2 199 ydl.sort_formats(info_dict)
0217c783
PH
200 ydl.process_ie_result(copy.deepcopy(info_dict))
201 downloaded = ydl.downloaded_info_dicts[0]
202 self.assertEqual(downloaded['format_id'], 'ogg-64')
203
bc6d5978
JMF
204 def test_format_selection_video(self):
205 formats = [
8508557e
JMF
206 {'format_id': 'dash-video-low', 'ext': 'mp4', 'preference': 1, 'acodec': 'none', 'url': TEST_URL},
207 {'format_id': 'dash-video-high', 'ext': 'mp4', 'preference': 2, 'acodec': 'none', 'url': TEST_URL},
208 {'format_id': 'vid', 'ext': 'mp4', 'preference': 3, 'url': TEST_URL},
bc6d5978 209 ]
3537b93d 210 info_dict = _make_result(formats)
bc6d5978
JMF
211
212 ydl = YDL({'format': 'bestvideo'})
213 ydl.process_ie_result(info_dict.copy())
214 downloaded = ydl.downloaded_info_dicts[0]
215 self.assertEqual(downloaded['format_id'], 'dash-video-high')
216
217 ydl = YDL({'format': 'worstvideo'})
218 ydl.process_ie_result(info_dict.copy())
219 downloaded = ydl.downloaded_info_dicts[0]
220 self.assertEqual(downloaded['format_id'], 'dash-video-low')
221
4c3b16d5
S
222 ydl = YDL({'format': 'bestvideo[format_id^=dash][format_id$=low]'})
223 ydl.process_ie_result(info_dict.copy())
224 downloaded = ydl.downloaded_info_dicts[0]
225 self.assertEqual(downloaded['format_id'], 'dash-video-low')
226
b913348d 227 formats = [
228 {'format_id': 'vid-vcodec-dot', 'ext': 'mp4', 'preference': 1, 'vcodec': 'avc1.123456', 'acodec': 'none', 'url': TEST_URL},
229 ]
230 info_dict = _make_result(formats)
231
232 ydl = YDL({'format': 'bestvideo[vcodec=avc1.123456]'})
233 ydl.process_ie_result(info_dict.copy())
234 downloaded = ydl.downloaded_info_dicts[0]
235 self.assertEqual(downloaded['format_id'], 'vid-vcodec-dot')
236
2cc779f4
S
237 def test_format_selection_string_ops(self):
238 formats = [
239 {'format_id': 'abc-cba', 'ext': 'mp4', 'url': TEST_URL},
e118a879 240 {'format_id': 'zxc-cxz', 'ext': 'webm', 'url': TEST_URL},
2cc779f4
S
241 ]
242 info_dict = _make_result(formats)
243
244 # equals (=)
245 ydl = YDL({'format': '[format_id=abc-cba]'})
246 ydl.process_ie_result(info_dict.copy())
247 downloaded = ydl.downloaded_info_dicts[0]
248 self.assertEqual(downloaded['format_id'], 'abc-cba')
249
250 # does not equal (!=)
251 ydl = YDL({'format': '[format_id!=abc-cba]'})
e118a879
S
252 ydl.process_ie_result(info_dict.copy())
253 downloaded = ydl.downloaded_info_dicts[0]
254 self.assertEqual(downloaded['format_id'], 'zxc-cxz')
255
256 ydl = YDL({'format': '[format_id!=abc-cba][format_id!=zxc-cxz]'})
2cc779f4
S
257 self.assertRaises(ExtractorError, ydl.process_ie_result, info_dict.copy())
258
259 # starts with (^=)
260 ydl = YDL({'format': '[format_id^=abc]'})
261 ydl.process_ie_result(info_dict.copy())
262 downloaded = ydl.downloaded_info_dicts[0]
263 self.assertEqual(downloaded['format_id'], 'abc-cba')
264
265 # does not start with (!^=)
e118a879
S
266 ydl = YDL({'format': '[format_id!^=abc]'})
267 ydl.process_ie_result(info_dict.copy())
268 downloaded = ydl.downloaded_info_dicts[0]
269 self.assertEqual(downloaded['format_id'], 'zxc-cxz')
270
271 ydl = YDL({'format': '[format_id!^=abc][format_id!^=zxc]'})
2cc779f4
S
272 self.assertRaises(ExtractorError, ydl.process_ie_result, info_dict.copy())
273
274 # ends with ($=)
275 ydl = YDL({'format': '[format_id$=cba]'})
276 ydl.process_ie_result(info_dict.copy())
277 downloaded = ydl.downloaded_info_dicts[0]
278 self.assertEqual(downloaded['format_id'], 'abc-cba')
279
280 # does not end with (!$=)
e118a879
S
281 ydl = YDL({'format': '[format_id!$=cba]'})
282 ydl.process_ie_result(info_dict.copy())
283 downloaded = ydl.downloaded_info_dicts[0]
284 self.assertEqual(downloaded['format_id'], 'zxc-cxz')
285
286 ydl = YDL({'format': '[format_id!$=cba][format_id!$=cxz]'})
2cc779f4
S
287 self.assertRaises(ExtractorError, ydl.process_ie_result, info_dict.copy())
288
289 # contains (*=)
e118a879 290 ydl = YDL({'format': '[format_id*=bc-cb]'})
2cc779f4
S
291 ydl.process_ie_result(info_dict.copy())
292 downloaded = ydl.downloaded_info_dicts[0]
293 self.assertEqual(downloaded['format_id'], 'abc-cba')
294
295 # does not contain (!*=)
e118a879
S
296 ydl = YDL({'format': '[format_id!*=bc-cb]'})
297 ydl.process_ie_result(info_dict.copy())
298 downloaded = ydl.downloaded_info_dicts[0]
299 self.assertEqual(downloaded['format_id'], 'zxc-cxz')
300
301 ydl = YDL({'format': '[format_id!*=abc][format_id!*=zxc]'})
302 self.assertRaises(ExtractorError, ydl.process_ie_result, info_dict.copy())
303
2cc779f4
S
304 ydl = YDL({'format': '[format_id!*=-]'})
305 self.assertRaises(ExtractorError, ydl.process_ie_result, info_dict.copy())
306
3d4a70b8 307 def test_youtube_format_selection(self):
a7191c6f 308 # FIXME: Rewrite in accordance with the new format sorting options
5d0c5371 309 return
5d0c5371 310
3d4a70b8 311 order = [
86bf2905 312 '38', '37', '46', '22', '45', '35', '44', '18', '34', '43', '6', '5', '17', '36', '13',
3d4a70b8
PH
313 # Apple HTTP Live Streaming
314 '96', '95', '94', '93', '92', '132', '151',
315 # 3D
316 '85', '84', '102', '83', '101', '82', '100',
317 # Dash video
c11125f9 318 '137', '248', '136', '247', '135', '246',
3d4a70b8
PH
319 '245', '244', '134', '243', '133', '242', '160',
320 # Dash audio
a053c349 321 '141', '172', '140', '171', '139',
3d4a70b8
PH
322 ]
323
67134eab
JMF
324 def format_info(f_id):
325 info = YoutubeIE._formats[f_id].copy()
1df41411 326
91cb6b50 327 # XXX: In real cases InfoExtractor._parse_mpd_formats() fills up 'acodec'
1df41411
YCH
328 # and 'vcodec', while in tests such information is incomplete since
329 # commit a6c2c24479e5f4827ceb06f64d855329c0a6f593
330 # test_YoutubeDL.test_youtube_format_selection is broken without
331 # this fix
332 if 'acodec' in info and 'vcodec' not in info:
333 info['vcodec'] = 'none'
334 elif 'vcodec' in info and 'acodec' not in info:
335 info['acodec'] = 'none'
336
67134eab
JMF
337 info['format_id'] = f_id
338 info['url'] = 'url:' + f_id
339 return info
340 formats_order = [format_info(f_id) for f_id in order]
341
342 info_dict = _make_result(list(formats_order), extractor='youtube')
343 ydl = YDL({'format': 'bestvideo+bestaudio'})
9f14daf2 344 ydl.sort_formats(info_dict)
67134eab
JMF
345 ydl.process_ie_result(info_dict)
346 downloaded = ydl.downloaded_info_dicts[0]
5d0c5371 347 self.assertEqual(downloaded['format_id'], '248+172')
67134eab 348 self.assertEqual(downloaded['ext'], 'mp4')
3d4a70b8 349
cf2ac6df
JMF
350 info_dict = _make_result(list(formats_order), extractor='youtube')
351 ydl = YDL({'format': 'bestvideo[height>=999999]+bestaudio/best'})
9f14daf2 352 ydl.sort_formats(info_dict)
cf2ac6df
JMF
353 ydl.process_ie_result(info_dict)
354 downloaded = ydl.downloaded_info_dicts[0]
355 self.assertEqual(downloaded['format_id'], '38')
356
f5f4a27a
JMF
357 info_dict = _make_result(list(formats_order), extractor='youtube')
358 ydl = YDL({'format': 'bestvideo/best,bestaudio'})
9f14daf2 359 ydl.sort_formats(info_dict)
f5f4a27a
JMF
360 ydl.process_ie_result(info_dict)
361 downloaded_ids = [info['format_id'] for info in ydl.downloaded_info_dicts]
362 self.assertEqual(downloaded_ids, ['137', '141'])
363
0130afb7
JMF
364 info_dict = _make_result(list(formats_order), extractor='youtube')
365 ydl = YDL({'format': '(bestvideo[ext=mp4],bestvideo[ext=webm])+bestaudio'})
9f14daf2 366 ydl.sort_formats(info_dict)
0130afb7
JMF
367 ydl.process_ie_result(info_dict)
368 downloaded_ids = [info['format_id'] for info in ydl.downloaded_info_dicts]
369 self.assertEqual(downloaded_ids, ['137+141', '248+141'])
370
371 info_dict = _make_result(list(formats_order), extractor='youtube')
372 ydl = YDL({'format': '(bestvideo[ext=mp4],bestvideo[ext=webm])[height<=720]+bestaudio'})
9f14daf2 373 ydl.sort_formats(info_dict)
0130afb7
JMF
374 ydl.process_ie_result(info_dict)
375 downloaded_ids = [info['format_id'] for info in ydl.downloaded_info_dicts]
376 self.assertEqual(downloaded_ids, ['136+141', '247+141'])
377
378 info_dict = _make_result(list(formats_order), extractor='youtube')
379 ydl = YDL({'format': '(bestvideo[ext=none]/bestvideo[ext=webm])+bestaudio'})
9f14daf2 380 ydl.sort_formats(info_dict)
0130afb7
JMF
381 ydl.process_ie_result(info_dict)
382 downloaded_ids = [info['format_id'] for info in ydl.downloaded_info_dicts]
383 self.assertEqual(downloaded_ids, ['248+141'])
384
67134eab 385 for f1, f2 in zip(formats_order, formats_order[1:]):
3537b93d 386 info_dict = _make_result([f1, f2], extractor='youtube')
8dd54188 387 ydl = YDL({'format': 'best/bestvideo'})
9f14daf2 388 ydl.sort_formats(info_dict)
3d4a70b8
PH
389 ydl.process_ie_result(info_dict)
390 downloaded = ydl.downloaded_info_dicts[0]
67134eab 391 self.assertEqual(downloaded['format_id'], f1['format_id'])
3d4a70b8 392
3537b93d 393 info_dict = _make_result([f2, f1], extractor='youtube')
8dd54188 394 ydl = YDL({'format': 'best/bestvideo'})
9f14daf2 395 ydl.sort_formats(info_dict)
3d4a70b8
PH
396 ydl.process_ie_result(info_dict)
397 downloaded = ydl.downloaded_info_dicts[0]
67134eab 398 self.assertEqual(downloaded['format_id'], f1['format_id'])
3d4a70b8 399
317f7ab6
S
400 def test_audio_only_extractor_format_selection(self):
401 # For extractors with incomplete formats (all formats are audio-only or
402 # video-only) best and worst should fallback to corresponding best/worst
403 # video-only or audio-only formats (as per
067aa17e 404 # https://github.com/ytdl-org/youtube-dl/pull/5556)
317f7ab6
S
405 formats = [
406 {'format_id': 'low', 'ext': 'mp3', 'preference': 1, 'vcodec': 'none', 'url': TEST_URL},
407 {'format_id': 'high', 'ext': 'mp3', 'preference': 2, 'vcodec': 'none', 'url': TEST_URL},
408 ]
409 info_dict = _make_result(formats)
410
411 ydl = YDL({'format': 'best'})
412 ydl.process_ie_result(info_dict.copy())
413 downloaded = ydl.downloaded_info_dicts[0]
414 self.assertEqual(downloaded['format_id'], 'high')
415
416 ydl = YDL({'format': 'worst'})
417 ydl.process_ie_result(info_dict.copy())
418 downloaded = ydl.downloaded_info_dicts[0]
419 self.assertEqual(downloaded['format_id'], 'low')
420
421 def test_format_not_available(self):
422 formats = [
423 {'format_id': 'regular', 'ext': 'mp4', 'height': 360, 'url': TEST_URL},
424 {'format_id': 'video', 'ext': 'mp4', 'height': 720, 'acodec': 'none', 'url': TEST_URL},
425 ]
426 info_dict = _make_result(formats)
427
428 # This must fail since complete video-audio format does not match filter
429 # and extractor does not provide incomplete only formats (i.e. only
430 # video-only or audio-only).
431 ydl = YDL({'format': 'best[height>360]'})
432 self.assertRaises(ExtractorError, ydl.process_ie_result, info_dict.copy())
433
8cda78ef 434 def test_format_selection_issue_10083(self):
067aa17e 435 # See https://github.com/ytdl-org/youtube-dl/issues/10083
8cda78ef
S
436 formats = [
437 {'format_id': 'regular', 'height': 360, 'url': TEST_URL},
438 {'format_id': 'video', 'height': 720, 'acodec': 'none', 'url': TEST_URL},
439 {'format_id': 'audio', 'vcodec': 'none', 'url': TEST_URL},
440 ]
441 info_dict = _make_result(formats)
442
443 ydl = YDL({'format': 'best[height>360]/bestvideo[height>360]+bestaudio'})
444 ydl.process_ie_result(info_dict.copy())
445 self.assertEqual(ydl.downloaded_info_dicts[0]['format_id'], 'video+audio')
446
0a31a350
JMF
447 def test_invalid_format_specs(self):
448 def assert_syntax_error(format_spec):
187986a8 449 self.assertRaises(SyntaxError, YDL, {'format': format_spec})
0a31a350
JMF
450
451 assert_syntax_error('bestvideo,,best')
452 assert_syntax_error('+bestaudio')
453 assert_syntax_error('bestvideo+')
d96d604e 454 assert_syntax_error('/')
187986a8 455 assert_syntax_error('[720<height]')
0a31a350 456
083c9df9
PH
457 def test_format_filtering(self):
458 formats = [
459 {'format_id': 'A', 'filesize': 500, 'width': 1000},
460 {'format_id': 'B', 'filesize': 1000, 'width': 500},
461 {'format_id': 'C', 'filesize': 1000, 'width': 400},
462 {'format_id': 'D', 'filesize': 2000, 'width': 600},
463 {'format_id': 'E', 'filesize': 3000},
464 {'format_id': 'F'},
465 {'format_id': 'G', 'filesize': 1000000},
466 ]
467 for f in formats:
468 f['url'] = 'http://_/'
469 f['ext'] = 'unknown'
9f14daf2 470 info_dict = _make_result(formats, _format_sort_fields=('id', ))
083c9df9
PH
471
472 ydl = YDL({'format': 'best[filesize<3000]'})
473 ydl.process_ie_result(info_dict)
474 downloaded = ydl.downloaded_info_dicts[0]
475 self.assertEqual(downloaded['format_id'], 'D')
476
477 ydl = YDL({'format': 'best[filesize<=3000]'})
478 ydl.process_ie_result(info_dict)
479 downloaded = ydl.downloaded_info_dicts[0]
480 self.assertEqual(downloaded['format_id'], 'E')
481
482 ydl = YDL({'format': 'best[filesize <= ? 3000]'})
483 ydl.process_ie_result(info_dict)
484 downloaded = ydl.downloaded_info_dicts[0]
485 self.assertEqual(downloaded['format_id'], 'F')
486
487 ydl = YDL({'format': 'best [filesize = 1000] [width>450]'})
488 ydl.process_ie_result(info_dict)
489 downloaded = ydl.downloaded_info_dicts[0]
490 self.assertEqual(downloaded['format_id'], 'B')
491
492 ydl = YDL({'format': 'best [filesize = 1000] [width!=450]'})
493 ydl.process_ie_result(info_dict)
494 downloaded = ydl.downloaded_info_dicts[0]
495 self.assertEqual(downloaded['format_id'], 'C')
496
497 ydl = YDL({'format': '[filesize>?1]'})
498 ydl.process_ie_result(info_dict)
499 downloaded = ydl.downloaded_info_dicts[0]
500 self.assertEqual(downloaded['format_id'], 'G')
501
502 ydl = YDL({'format': '[filesize<1M]'})
503 ydl.process_ie_result(info_dict)
504 downloaded = ydl.downloaded_info_dicts[0]
505 self.assertEqual(downloaded['format_id'], 'E')
506
507 ydl = YDL({'format': '[filesize<1MiB]'})
508 ydl.process_ie_result(info_dict)
509 downloaded = ydl.downloaded_info_dicts[0]
510 self.assertEqual(downloaded['format_id'], 'G')
511
5acfa126
JMF
512 ydl = YDL({'format': 'all[width>=400][width<=600]'})
513 ydl.process_ie_result(info_dict)
514 downloaded_ids = [info['format_id'] for info in ydl.downloaded_info_dicts]
dd2a987d 515 self.assertEqual(downloaded_ids, ['D', 'C', 'B'])
5acfa126 516
bb8e5536
JMF
517 ydl = YDL({'format': 'best[height<40]'})
518 try:
519 ydl.process_ie_result(info_dict)
520 except ExtractorError:
521 pass
522 self.assertEqual(ydl.downloaded_info_dicts, [])
523
0017d9ad
S
524 def test_default_format_spec(self):
525 ydl = YDL({'simulate': True})
5d0c5371 526 self.assertEqual(ydl._default_format_spec({}), 'bestvideo*+bestaudio/best')
0017d9ad 527
d08dcd2d
S
528 ydl = YDL({})
529 self.assertEqual(ydl._default_format_spec({'is_live': True}), 'best/bestvideo+bestaudio')
af0f7428 530
d08dcd2d 531 ydl = YDL({'simulate': True})
5d0c5371 532 self.assertEqual(ydl._default_format_spec({'is_live': True}), 'bestvideo*+bestaudio/best')
af0f7428 533
0017d9ad 534 ydl = YDL({'outtmpl': '-'})
af0f7428 535 self.assertEqual(ydl._default_format_spec({}), 'best/bestvideo+bestaudio')
0017d9ad
S
536
537 ydl = YDL({})
5d0c5371 538 self.assertEqual(ydl._default_format_spec({}, download=False), 'bestvideo*+bestaudio/best')
af0f7428 539 self.assertEqual(ydl._default_format_spec({'is_live': True}), 'best/bestvideo+bestaudio')
0017d9ad 540
f20bf146
JMF
541
542class TestYoutubeDL(unittest.TestCase):
ab84349b
JMF
543 def test_subtitles(self):
544 def s_formats(lang, autocaption=False):
545 return [{
546 'ext': ext,
86e5f3ed 547 'url': f'http://localhost/video.{lang}.{ext}',
ab84349b
JMF
548 '_auto': autocaption,
549 } for ext in ['vtt', 'srt', 'ass']]
86e5f3ed 550 subtitles = {l: s_formats(l) for l in ['en', 'fr', 'es']}
551 auto_captions = {l: s_formats(l, True) for l in ['it', 'pt', 'es']}
ab84349b
JMF
552 info_dict = {
553 'id': 'test',
554 'title': 'Test',
555 'url': 'http://localhost/video.mp4',
556 'subtitles': subtitles,
557 'automatic_captions': auto_captions,
558 'extractor': 'TEST',
732044af 559 'webpage_url': 'http://example.com/watch?v=shenanigans',
ab84349b
JMF
560 }
561
562 def get_info(params={}):
563 params.setdefault('simulate', True)
564 ydl = YDL(params)
565 ydl.report_warning = lambda *args, **kargs: None
566 return ydl.process_video_result(info_dict, download=False)
567
568 result = get_info()
569 self.assertFalse(result.get('requested_subtitles'))
570 self.assertEqual(result['subtitles'], subtitles)
571 self.assertEqual(result['automatic_captions'], auto_captions)
572
573 result = get_info({'writesubtitles': True})
574 subs = result['requested_subtitles']
575 self.assertTrue(subs)
86e5f3ed 576 self.assertEqual(set(subs.keys()), {'en'})
ab84349b
JMF
577 self.assertTrue(subs['en'].get('data') is None)
578 self.assertEqual(subs['en']['ext'], 'ass')
579
580 result = get_info({'writesubtitles': True, 'subtitlesformat': 'foo/srt'})
581 subs = result['requested_subtitles']
582 self.assertEqual(subs['en']['ext'], 'srt')
583
584 result = get_info({'writesubtitles': True, 'subtitleslangs': ['es', 'fr', 'it']})
585 subs = result['requested_subtitles']
586 self.assertTrue(subs)
86e5f3ed 587 self.assertEqual(set(subs.keys()), {'es', 'fr'})
ab84349b 588
c32b0aab 589 result = get_info({'writesubtitles': True, 'subtitleslangs': ['all', '-en']})
590 subs = result['requested_subtitles']
591 self.assertTrue(subs)
86e5f3ed 592 self.assertEqual(set(subs.keys()), {'es', 'fr'})
c32b0aab 593
594 result = get_info({'writesubtitles': True, 'subtitleslangs': ['en', 'fr', '-en']})
595 subs = result['requested_subtitles']
596 self.assertTrue(subs)
86e5f3ed 597 self.assertEqual(set(subs.keys()), {'fr'})
c32b0aab 598
599 result = get_info({'writesubtitles': True, 'subtitleslangs': ['-en', 'en']})
600 subs = result['requested_subtitles']
601 self.assertTrue(subs)
86e5f3ed 602 self.assertEqual(set(subs.keys()), {'en'})
c32b0aab 603
604 result = get_info({'writesubtitles': True, 'subtitleslangs': ['e.+']})
605 subs = result['requested_subtitles']
606 self.assertTrue(subs)
86e5f3ed 607 self.assertEqual(set(subs.keys()), {'es', 'en'})
c32b0aab 608
ab84349b
JMF
609 result = get_info({'writesubtitles': True, 'writeautomaticsub': True, 'subtitleslangs': ['es', 'pt']})
610 subs = result['requested_subtitles']
611 self.assertTrue(subs)
86e5f3ed 612 self.assertEqual(set(subs.keys()), {'es', 'pt'})
ab84349b
JMF
613 self.assertFalse(subs['es']['_auto'])
614 self.assertTrue(subs['pt']['_auto'])
615
98c70d6f
JMF
616 result = get_info({'writeautomaticsub': True, 'subtitleslangs': ['es', 'pt']})
617 subs = result['requested_subtitles']
618 self.assertTrue(subs)
86e5f3ed 619 self.assertEqual(set(subs.keys()), {'es', 'pt'})
98c70d6f
JMF
620 self.assertTrue(subs['es']['_auto'])
621 self.assertTrue(subs['pt']['_auto'])
622
b6c45014
JMF
623 def test_add_extra_info(self):
624 test_dict = {
625 'extractor': 'Foo',
626 }
627 extra_info = {
628 'extractor': 'Bar',
629 'playlist': 'funny videos',
630 }
631 YDL.add_extra_info(test_dict, extra_info)
632 self.assertEqual(test_dict['extractor'], 'Foo')
633 self.assertEqual(test_dict['playlist'], 'funny videos')
634
752cda38 635 outtmpl_info = {
636 'id': '1234',
637 'ext': 'mp4',
638 'width': None,
639 'height': 1080,
abbeeebc 640 'filesize': 1024,
752cda38 641 'title1': '$PATH',
642 'title2': '%PATH%',
87ea7dfc 643 'title3': 'foo/bar\\test',
7d1eb38a 644 'title4': 'foo "bar" test',
524e2e4f 645 'title5': 'áéí 𝐀',
752cda38 646 'timestamp': 1618488000,
647 'duration': 100000,
648 'playlist_index': 1,
e6f21b3d 649 'playlist_autonumber': 2,
0a5a191a 650 '__last_playlist_index': 100,
752cda38 651 'n_entries': 10,
07a1250e 652 'formats': [
653 {'id': 'id 1', 'height': 1080, 'width': 1920},
654 {'id': 'id 2', 'height': 720},
655 {'id': 'id 3'}
656 ]
752cda38 657 }
658
5c6542ce 659 def test_prepare_outtmpl_and_filename(self):
9fea350f 660 def test(tmpl, expected, *, info=None, **params):
752cda38 661 params['outtmpl'] = tmpl
d2c8aadf 662 ydl = FakeYDL(params)
752cda38 663 ydl._num_downloads = 1
5c6542ce 664 self.assertEqual(ydl.validate_outtmpl(tmpl), None)
752cda38 665
819e0531 666 out = ydl.evaluate_outtmpl(tmpl, info or self.outtmpl_info)
9fea350f 667 fname = ydl.prepare_filename(info or self.outtmpl_info)
5c6542ce 668
2b8a2973 669 if not isinstance(expected, (list, tuple)):
670 expected = (expected, expected)
671 for (name, got), expect in zip((('outtmpl', out), ('filename', fname)), expected):
672 if callable(expect):
673 self.assertTrue(expect(got), f'Wrong {name} from {tmpl}')
ad54c913 674 elif expect is not None:
2b8a2973 675 self.assertEqual(got, expect, f'Wrong {name} from {tmpl}')
5c6542ce 676
6e84b215 677 # Side-effects
678 original_infodict = dict(self.outtmpl_info)
679 test('foo.bar', 'foo.bar')
680 original_infodict['epoch'] = self.outtmpl_info.get('epoch')
681 self.assertTrue(isinstance(original_infodict['epoch'], int))
682 test('%(epoch)d', int_or_none)
683 self.assertEqual(original_infodict, self.outtmpl_info)
684
5c6542ce 685 # Auto-generated fields
686 test('%(id)s.%(ext)s', '1234.mp4')
687 test('%(duration_string)s', ('27:46:40', '27-46-40'))
5c6542ce 688 test('%(resolution)s', '1080p')
47bcd437 689 test('%(playlist_index|)s', '001')
62b5c94c 690 test('%(playlist_index&{}!)s', '1!')
e6f21b3d 691 test('%(playlist_autonumber)s', '02')
5c6542ce 692 test('%(autonumber)s', '00001')
693 test('%(autonumber+2)03d', '005', autonumber_start=3)
694 test('%(autonumber)s', '001', autonumber_size=3)
695
696 # Escaping %
901130bb 697 test('%', '%')
5c6542ce 698 test('%%', '%')
699 test('%%%%', '%%')
901130bb 700 test('%s', '%s')
701 test('%%%s', '%%s')
702 test('%d', '%d')
703 test('%abc%', '%abc%')
5c6542ce 704 test('%%(width)06d.%(ext)s', '%(width)06d.mp4')
901130bb 705 test('%%%(height)s', '%1080')
5c6542ce 706 test('%(width)06d.%(ext)s', 'NA.mp4')
707 test('%(width)06d.%%(ext)s', 'NA.%(ext)s')
708 test('%%(width)06d.%(ext)s', '%(width)06d.mp4')
709
9fea350f 710 # ID sanitization
711 test('%(id)s', '_abcd', info={'id': '_abcd'})
712 test('%(some_id)s', '_abcd', info={'some_id': '_abcd'})
713 test('%(formats.0.id)s', '_abcd', info={'formats': [{'id': '_abcd'}]})
714 test('%(id)s', '-abcd', info={'id': '-abcd'})
715 test('%(id)s', '.abcd', info={'id': '.abcd'})
716 test('%(id)s', 'ab__cd', info={'id': 'ab__cd'})
97d9c79e 717 test('%(id)s', ('ab:cd', 'ab:cd'), info={'id': 'ab:cd'})
e0fd9573 718 test('%(id.0)s', '-', info={'id': '--'})
9fea350f 719
5c6542ce 720 # Invalid templates
5c6542ce 721 self.assertTrue(isinstance(YoutubeDL.validate_outtmpl('%(title)'), ValueError))
722 test('%(invalid@tmpl|def)s', 'none', outtmpl_na_placeholder='none')
2b8a2973 723 test('%(..)s', 'NA')
07a1250e 724 test('%(formats.{id)s', 'NA')
2b8a2973 725
726 # Entire info_dict
727 def expect_same_infodict(out):
728 got_dict = json.loads(out)
729 for info_field, expected in self.outtmpl_info.items():
730 self.assertEqual(got_dict.get(info_field), expected, info_field)
731 return True
732
f9fb3ce8 733 test('%()j', (expect_same_infodict, None))
5c6542ce 734
735 # NA placeholder
752cda38 736 NA_TEST_OUTTMPL = '%(uploader_date)s-%(width)d-%(x|def)s-%(id)s.%(ext)s'
5c6542ce 737 test(NA_TEST_OUTTMPL, 'NA-NA-def-1234.mp4')
738 test(NA_TEST_OUTTMPL, 'none-none-def-1234.mp4', outtmpl_na_placeholder='none')
739 test(NA_TEST_OUTTMPL, '--def-1234.mp4', outtmpl_na_placeholder='')
582fad70 740 test('%(non_existent.0)s', 'NA')
752cda38 741
5c6542ce 742 # String formatting
752cda38 743 FMT_TEST_OUTTMPL = '%%(height)%s.%%(ext)s'
5c6542ce 744 test(FMT_TEST_OUTTMPL % 's', '1080.mp4')
745 test(FMT_TEST_OUTTMPL % 'd', '1080.mp4')
746 test(FMT_TEST_OUTTMPL % '6d', ' 1080.mp4')
747 test(FMT_TEST_OUTTMPL % '-6d', '1080 .mp4')
748 test(FMT_TEST_OUTTMPL % '06d', '001080.mp4')
749 test(FMT_TEST_OUTTMPL % ' 06d', ' 01080.mp4')
750 test(FMT_TEST_OUTTMPL % ' 06d', ' 01080.mp4')
751 test(FMT_TEST_OUTTMPL % '0 6d', ' 01080.mp4')
752 test(FMT_TEST_OUTTMPL % '0 6d', ' 01080.mp4')
753 test(FMT_TEST_OUTTMPL % ' 0 6d', ' 01080.mp4')
754
755 # Type casting
756 test('%(id)d', '1234')
757 test('%(height)c', '1')
758 test('%(ext)c', 'm')
759 test('%(id)d %(id)r', "1234 '1234'")
760 test('%(id)r %(height)r', "'1234' 1080")
1619ab3e 761 test('%(title5)a %(height)a', (R"'\xe1\xe9\xed \U0001d400' 1080", None))
5c6542ce 762 test('%(ext)s-%(ext|def)d', 'mp4-def')
ebe1b4e3 763 test('%(width|0)04d', '0')
6f2287cb 764 test('a%(width|b)d', 'ab', outtmpl_na_placeholder='none')
5c6542ce 765
752cda38 766 FORMATS = self.outtmpl_info['formats']
7d1eb38a 767
768 # Custom type casting
4476d2c7 769 test('%(formats.:.id)l', 'id 1, id 2, id 3')
770 test('%(formats.:.id)#l', ('id 1\nid 2\nid 3', 'id 1 id 2 id 3'))
7d1eb38a 771 test('%(ext)l', 'mp4')
4476d2c7 772 test('%(formats.:.id) 18l', ' id 1, id 2, id 3')
ad54c913 773 test('%(formats)j', (json.dumps(FORMATS), None))
774 test('%(formats)#j', (
775 json.dumps(FORMATS, indent=4),
776 json.dumps(FORMATS, indent=4).replace(':', ':').replace('"', """).replace('\n', ' ')
777 ))
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')
97d9c79e 787 test('%(title4).10S', ('foo "bar" ', 'foo "bar"' + ('#' if compat_os_name == 'nt' else ' ')))
7d1eb38a 788 if compat_os_name == 'nt':
de015e93
SS
789 test('%(title4)q', ('"foo ""bar"" test"', None))
790 test('%(formats.:.id)#q', ('"id 1" "id 2" "id 3"', None))
791 test('%(formats.0.id)#q', ('"id 1"', None))
7d1eb38a 792 else:
97d9c79e 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')
993edd3f 802 test('%(filesize*8)d', '8192')
ad54c913 803 test('%(formats.0) 15s', ('% 15s' % FORMATS[0], None))
804 test('%(formats.0)r', (repr(FORMATS[0]), None))
5c6542ce 805 test('%(height.0)03d', '001')
806 test('%(-height.0)04d', '-001')
807 test('%(formats.-1.id)s', FORMATS[-1]['id'])
808 test('%(formats.0.id.-1)d', FORMATS[0]['id'][-1])
809 test('%(formats.3)s', 'NA')
810 test('%(formats.:2:-1)r', repr(FORMATS[:2:-1]))
811 test('%(formats.0.id.-1+id)f', '1235.000000')
385a27fa 812 test('%(formats.0.id.-1+formats.1.id.-1)d', '3')
07a1250e 813 out = json.dumps([{'id': f['id'], 'height.:2': str(f['height'])[:2]}
814 if 'height' in f else {'id': f['id']}
815 for f in FORMATS])
ad54c913 816 test('%(formats.:.{id,height.:2})j', (out, None))
07a1250e 817 test('%(formats.:.{id,height}.id)l', ', '.join(f['id'] for f in FORMATS))
818 test('%(.{id,title})j', ('{"id": "1234"}', '{"id": "1234"}'))
5c6542ce 819
7c37ff97 820 # Alternates
821 test('%(title,id)s', '1234')
822 test('%(width-100,height+20|def)d', '1100')
823 test('%(width-100,height+width|def)s', 'def')
824 test('%(timestamp-x>%H\\,%M\\,%S,timestamp>%H\\,%M\\,%S)s', '12,00,00')
825
e0fd9573 826 # Replacement
827 test('%(id&foo)s.bar', 'foo.bar')
828 test('%(title&foo)s.bar', 'NA.bar')
829 test('%(title&foo|baz)s.bar', 'baz.bar')
34baa9fd 830 test('%(x,id&foo|baz)s.bar', 'foo.bar')
831 test('%(x,title&foo|baz)s.bar', 'baz.bar')
ec9311c4 832 test('%(id&a\nb|)s', ('a\nb', 'a b'))
833 test('%(id&hi {:>10} {}|)s', 'hi 1234 1234')
834 test(R'%(id&{0} {}|)s', 'NA')
835 test(R'%(id&{0.1}|)s', 'NA')
6148833f 836 test('%(height&{:,d})S', '1,080')
e0fd9573 837
6e84b215 838 # Laziness
839 def gen():
840 yield from range(5)
841 raise self.assertTrue(False, 'LazyList should not be evaluated till here')
842 test('%(key.4)s', '4', info={'key': LazyList(gen())})
843
5c6542ce 844 # Empty filename
845 test('%(foo|)s-%(bar|)s.%(ext)s', '-.mp4')
846 # test('%(foo|)s.%(ext)s', ('.mp4', '_.mp4')) # fixme
847 # test('%(foo|)s', ('', '_')) # fixme
848
901130bb 849 # Environment variable expansion for prepare_filename
ac668111 850 os.environ['__yt_dlp_var'] = 'expanded'
901130bb 851 envvar = '%__yt_dlp_var%' if compat_os_name == 'nt' else '$__yt_dlp_var'
852 test(envvar, (envvar, 'expanded'))
b836dc94 853 if compat_os_name == 'nt':
854 test('%s%', ('%s%', '%s%'))
ac668111 855 os.environ['s'] = 'expanded'
b836dc94 856 test('%s%', ('%s%', 'expanded')) # %s% should be expanded before escaping %s
ac668111 857 os.environ['(test)s'] = 'expanded'
b836dc94 858 test('%(test)s%', ('NA%', 'expanded')) # Environment should take priority over template
901130bb 859
5c6542ce 860 # Path expansion and escaping
861 test('Hello %(title1)s', 'Hello $PATH')
862 test('Hello %(title2)s', 'Hello %PATH%')
97d9c79e 863 test('%(title3)s', ('foo/bar\\test', 'foo⧸bar⧹test'))
864 test('folder/%(title3)s', ('folder/foo/bar\\test', 'folder%sfoo⧸bar⧹test' % os.path.sep))
26e63931 865
c57f7757
PH
866 def test_format_note(self):
867 ydl = YoutubeDL()
868 self.assertEqual(ydl._format_note({}), '')
869 assertRegexpMatches(self, ydl._format_note({
870 'vbr': 10,
398dea32 871 }), r'^\s*10k$')
5d583bdf
S
872 assertRegexpMatches(self, ydl._format_note({
873 'fps': 30,
398dea32 874 }), r'^30fps$')
5d583bdf 875
2b4ecde2
JMF
876 def test_postprocessors(self):
877 filename = 'post-processor-testfile.mp4'
878 audiofile = filename + '.mp3'
879
880 class SimplePP(PostProcessor):
881 def run(self, info):
6f2287cb 882 with open(audiofile, 'w') as f:
2b4ecde2 883 f.write('EXAMPLE')
592e97e8 884 return [info['filepath']], info
2b4ecde2 885
592e97e8 886 def run_pp(params, PP):
6f2287cb 887 with open(filename, 'w') as f:
2b4ecde2
JMF
888 f.write('EXAMPLE')
889 ydl = YoutubeDL(params)
592e97e8 890 ydl.add_post_processor(PP())
2b4ecde2
JMF
891 ydl.post_process(filename, {'filepath': filename})
892
592e97e8 893 run_pp({'keepvideo': True}, SimplePP)
2b4ecde2
JMF
894 self.assertTrue(os.path.exists(filename), '%s doesn\'t exist' % filename)
895 self.assertTrue(os.path.exists(audiofile), '%s doesn\'t exist' % audiofile)
896 os.unlink(filename)
897 os.unlink(audiofile)
898
592e97e8 899 run_pp({'keepvideo': False}, SimplePP)
2b4ecde2
JMF
900 self.assertFalse(os.path.exists(filename), '%s exists' % filename)
901 self.assertTrue(os.path.exists(audiofile), '%s doesn\'t exist' % audiofile)
902 os.unlink(audiofile)
903
592e97e8
JMF
904 class ModifierPP(PostProcessor):
905 def run(self, info):
6f2287cb 906 with open(info['filepath'], 'w') as f:
592e97e8
JMF
907 f.write('MODIFIED')
908 return [], info
909
910 run_pp({'keepvideo': False}, ModifierPP)
911 self.assertTrue(os.path.exists(filename), '%s doesn\'t exist' % filename)
912 os.unlink(filename)
913
531980d8 914 def test_match_filter(self):
531980d8
JMF
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):
09b49e1f 942 ydl = YDL({'match_filter': filter_, 'simulate': True})
531980d8 943 for v in videos:
104a7b5a 944 ydl.process_ie_result(v.copy(), download=True)
531980d8
JMF
945 return [v['id'] for v in ydl.downloaded_info_dicts]
946
947 res = get_videos()
948 self.assertEqual(res, ['1', '2'])
949
6db9c4d5 950 def f(v, incomplete):
531980d8
JMF
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 998 def test_playlist_items_selection(self):
7e88d7d7 999 INDICES, PAGE_SIZE = list(range(1, 11)), 3
1000
1001 def entry(i, evaluated):
1002 evaluated.append(i)
1003 return {
1004 'id': str(i),
1005 'title': str(i),
1006 'url': TEST_URL,
1007 }
1008
1009 def pagedlist_entries(evaluated):
1010 def page_func(n):
1011 start = PAGE_SIZE * n
1012 for i in INDICES[start: start + PAGE_SIZE]:
1013 yield entry(i, evaluated)
1014 return OnDemandPagedList(page_func, PAGE_SIZE)
1015
1016 def page_num(i):
1017 return (i + PAGE_SIZE - 1) // PAGE_SIZE
e9eaf3fb 1018
7e88d7d7 1019 def generator_entries(evaluated):
1020 for i in INDICES:
1021 yield entry(i, evaluated)
1022
1023 def list_entries(evaluated):
1024 return list(generator_entries(evaluated))
1025
1026 def lazylist_entries(evaluated):
1027 return LazyList(generator_entries(evaluated))
1028
1029 def get_downloaded_info_dicts(params, entries):
e9eaf3fb 1030 ydl = YDL(params)
7e88d7d7 1031 ydl.process_ie_result({
1032 '_type': 'playlist',
1033 'id': 'test',
1034 'extractor': 'test:playlist',
1035 'extractor_key': 'test:playlist',
1036 'webpage_url': 'http://example.com',
1037 'entries': entries,
1038 })
4e9e1e24
S
1039 return ydl.downloaded_info_dicts
1040
7e88d7d7 1041 def test_selection(params, expected_ids, evaluate_all=False):
1042 expected_ids = list(expected_ids)
1043 if evaluate_all:
1044 generator_eval = pagedlist_eval = INDICES
1045 elif not expected_ids:
1046 generator_eval = pagedlist_eval = []
1047 else:
1048 generator_eval = INDICES[0: max(expected_ids)]
1049 pagedlist_eval = INDICES[PAGE_SIZE * page_num(min(expected_ids)) - PAGE_SIZE:
1050 PAGE_SIZE * page_num(max(expected_ids))]
1051
1052 for name, func, expected_eval in (
1053 ('list', list_entries, INDICES),
1054 ('Generator', generator_entries, generator_eval),
7e9a6125 1055 # ('LazyList', lazylist_entries, generator_eval), # Generator and LazyList follow the exact same code path
7e88d7d7 1056 ('PagedList', pagedlist_entries, pagedlist_eval),
1057 ):
1058 evaluated = []
1059 entries = func(evaluated)
1060 results = [(v['playlist_autonumber'] - 1, (int(v['id']), v['playlist_index']))
1061 for v in get_downloaded_info_dicts(params, entries)]
1062 self.assertEqual(results, list(enumerate(zip(expected_ids, expected_ids))), f'Entries of {name} for {params}')
1063 self.assertEqual(sorted(evaluated), expected_eval, f'Evaluation of {name} for {params}')
f2df4071 1064
7e88d7d7 1065 test_selection({}, INDICES)
1066 test_selection({'playlistend': 20}, INDICES, True)
1067 test_selection({'playlistend': 2}, INDICES[:2])
1068 test_selection({'playliststart': 11}, [], True)
1069 test_selection({'playliststart': 2}, INDICES[1:])
1070 test_selection({'playlist_items': '2-4'}, INDICES[1:4])
9e598870 1071 test_selection({'playlist_items': '2,4'}, [2, 4])
7e88d7d7 1072 test_selection({'playlist_items': '20'}, [], True)
ff1c7fc9 1073 test_selection({'playlist_items': '0'}, [])
cd6fc19e 1074
4e9e1e24 1075 # Tests for https://github.com/ytdl-org/youtube-dl/issues/10591
9e598870 1076 test_selection({'playlist_items': '2-4,3-4,3'}, [2, 3, 4])
1077 test_selection({'playlist_items': '4,2'}, [4, 2])
1078
1079 # Tests for https://github.com/yt-dlp/yt-dlp/issues/720
1080 # https://github.com/yt-dlp/yt-dlp/issues/302
7e88d7d7 1081 test_selection({'playlistreverse': True}, INDICES[::-1])
1082 test_selection({'playliststart': 2, 'playlistreverse': True}, INDICES[:0:-1])
9e598870 1083 test_selection({'playlist_items': '2,4', 'playlistreverse': True}, [4, 2])
1084 test_selection({'playlist_items': '4,2'}, [4, 2])
4e9e1e24 1085
7e88d7d7 1086 # Tests for --playlist-items start:end:step
1087 test_selection({'playlist_items': ':'}, INDICES, True)
1088 test_selection({'playlist_items': '::1'}, INDICES, True)
1089 test_selection({'playlist_items': '::-1'}, INDICES[::-1], True)
1090 test_selection({'playlist_items': ':6'}, INDICES[:6])
1091 test_selection({'playlist_items': ':-6'}, INDICES[:-5], True)
1092 test_selection({'playlist_items': '-1:6:-2'}, INDICES[:4:-2], True)
1093 test_selection({'playlist_items': '9:-6:-2'}, INDICES[8:3:-2], True)
1094
1095 test_selection({'playlist_items': '1:inf:2'}, INDICES[::2], True)
1096 test_selection({'playlist_items': '-2:inf'}, INDICES[-2:], True)
1097 test_selection({'playlist_items': ':inf:-1'}, [], True)
1098 test_selection({'playlist_items': '0-2:2'}, [2])
1099 test_selection({'playlist_items': '1-:2'}, INDICES[::2], True)
1100 test_selection({'playlist_items': '0--2:2'}, INDICES[1:-1:2], True)
1101
1102 test_selection({'playlist_items': '10::3'}, [10], True)
1103 test_selection({'playlist_items': '-1::3'}, [10], True)
1104 test_selection({'playlist_items': '11::3'}, [], True)
1105 test_selection({'playlist_items': '-15::2'}, INDICES[1::2], True)
1106 test_selection({'playlist_items': '-15::15'}, [], True)
1107
b286f201
YCH
1108 def test_do_not_override_ie_key_in_url_transparent(self):
1109 ydl = YDL()
1110
1111 class Foo1IE(InfoExtractor):
1112 _VALID_URL = r'foo1:'
1113
1114 def _real_extract(self, url):
1115 return {
1116 '_type': 'url_transparent',
1117 'url': 'foo2:',
1118 'ie_key': 'Foo2',
0396806f
S
1119 'title': 'foo1 title',
1120 'id': 'foo1_id',
b286f201
YCH
1121 }
1122
1123 class Foo2IE(InfoExtractor):
1124 _VALID_URL = r'foo2:'
1125
1126 def _real_extract(self, url):
1127 return {
1128 '_type': 'url',
1129 'url': 'foo3:',
1130 'ie_key': 'Foo3',
1131 }
1132
1133 class Foo3IE(InfoExtractor):
1134 _VALID_URL = r'foo3:'
1135
1136 def _real_extract(self, url):
51350db5 1137 return _make_result([{'url': TEST_URL}], title='foo3 title')
b286f201
YCH
1138
1139 ydl.add_info_extractor(Foo1IE(ydl))
1140 ydl.add_info_extractor(Foo2IE(ydl))
1141 ydl.add_info_extractor(Foo3IE(ydl))
1142 ydl.extract_info('foo1:')
1143 downloaded = ydl.downloaded_info_dicts[0]
1144 self.assertEqual(downloaded['url'], TEST_URL)
51350db5 1145 self.assertEqual(downloaded['title'], 'foo1 title')
0396806f
S
1146 self.assertEqual(downloaded['id'], 'testid')
1147 self.assertEqual(downloaded['extractor'], 'testex')
1148 self.assertEqual(downloaded['extractor_key'], 'TestEx')
b286f201 1149
a0566bbf 1150 # Test case for https://github.com/ytdl-org/youtube-dl/issues/27064
1151 def test_ignoreerrors_for_playlist_with_url_transparent_iterable_entries(self):
1152
1153 class _YDL(YDL):
1154 def __init__(self, *args, **kwargs):
86e5f3ed 1155 super().__init__(*args, **kwargs)
a0566bbf 1156
1157 def trouble(self, s, tb=None):
1158 pass
1159
1160 ydl = _YDL({
1161 'format': 'extra',
1162 'ignoreerrors': True,
1163 })
1164
1165 class VideoIE(InfoExtractor):
1166 _VALID_URL = r'video:(?P<id>\d+)'
1167
1168 def _real_extract(self, url):
1169 video_id = self._match_id(url)
1170 formats = [{
1171 'format_id': 'default',
1172 'url': 'url:',
1173 }]
1174 if video_id == '0':
1175 raise ExtractorError('foo')
1176 if video_id == '2':
1177 formats.append({
1178 'format_id': 'extra',
1179 'url': TEST_URL,
1180 })
1181 return {
1182 'id': video_id,
1183 'title': 'Video %s' % video_id,
1184 'formats': formats,
1185 }
1186
1187 class PlaylistIE(InfoExtractor):
1188 _VALID_URL = r'playlist:'
1189
1190 def _entries(self):
1191 for n in range(3):
14f25df2 1192 video_id = str(n)
a0566bbf 1193 yield {
1194 '_type': 'url_transparent',
1195 'ie_key': VideoIE.ie_key(),
1196 'id': video_id,
1197 'url': 'video:%s' % video_id,
1198 'title': 'Video Transparent %s' % video_id,
1199 }
1200
1201 def _real_extract(self, url):
1202 return self.playlist_result(self._entries())
1203
1204 ydl.add_info_extractor(VideoIE(ydl))
1205 ydl.add_info_extractor(PlaylistIE(ydl))
1206 info = ydl.extract_info('playlist:')
1207 entries = info['entries']
1208 self.assertEqual(len(entries), 3)
1209 self.assertTrue(entries[0] is None)
1210 self.assertTrue(entries[1] is None)
1211 self.assertEqual(len(ydl.downloaded_info_dicts), 1)
1212 downloaded = ydl.downloaded_info_dicts[0]
9e907ebd 1213 entries[2].pop('requested_downloads', None)
a0566bbf 1214 self.assertEqual(entries[2], downloaded)
1215 self.assertEqual(downloaded['url'], TEST_URL)
1216 self.assertEqual(downloaded['title'], 'Video Transparent 2')
1217 self.assertEqual(downloaded['id'], '2')
1218 self.assertEqual(downloaded['extractor'], 'Video')
1219 self.assertEqual(downloaded['extractor_key'], 'Video')
1220
31215122
SS
1221 def test_header_cookies(self):
1222 from http.cookiejar import Cookie
1223
1224 ydl = FakeYDL()
1225 ydl.report_warning = lambda *_, **__: None
1226
1227 def cookie(name, value, version=None, domain='', path='', secure=False, expires=None):
1228 return Cookie(
1229 version or 0, name, value, None, False,
1230 domain, bool(domain), bool(domain), path, bool(path),
1231 secure, expires, False, None, None, rest={})
1232
1233 _test_url = 'https://yt.dlp/test'
1234
6c5211ce 1235 def test(encoded_cookies, cookies, *, headers=False, round_trip=None, error_re=None):
31215122
SS
1236 def _test():
1237 ydl.cookiejar.clear()
6c5211ce 1238 ydl._load_cookies(encoded_cookies, autoscope=headers)
31215122
SS
1239 if headers:
1240 ydl._apply_header_cookies(_test_url)
1241 data = {'url': _test_url}
1242 ydl._calc_headers(data)
1243 self.assertCountEqual(
1244 map(vars, ydl.cookiejar), map(vars, cookies),
1245 'Extracted cookiejar.Cookie is not the same')
1246 if not headers:
1247 self.assertEqual(
1248 data.get('cookies'), round_trip or encoded_cookies,
1249 'Cookie is not the same as round trip')
1250 ydl.__dict__['_YoutubeDL__header_cookies'] = []
1251
1252 with self.subTest(msg=encoded_cookies):
6c5211ce 1253 if not error_re:
31215122
SS
1254 _test()
1255 return
6c5211ce 1256 with self.assertRaisesRegex(Exception, error_re):
31215122
SS
1257 _test()
1258
1259 test('test=value; Domain=.yt.dlp', [cookie('test', 'value', domain='.yt.dlp')])
6c5211ce 1260 test('test=value', [cookie('test', 'value')], error_re=r'Unscoped cookies are not allowed')
31215122
SS
1261 test('cookie1=value1; Domain=.yt.dlp; Path=/test; cookie2=value2; Domain=.yt.dlp; Path=/', [
1262 cookie('cookie1', 'value1', domain='.yt.dlp', path='/test'),
1263 cookie('cookie2', 'value2', domain='.yt.dlp', path='/')])
1264 test('test=value; Domain=.yt.dlp; Path=/test; Secure; Expires=9999999999', [
1265 cookie('test', 'value', domain='.yt.dlp', path='/test', secure=True, expires=9999999999)])
1266 test('test="value; "; path=/test; domain=.yt.dlp', [
1267 cookie('test', 'value; ', domain='.yt.dlp', path='/test')],
1268 round_trip='test="value\\073 "; Domain=.yt.dlp; Path=/test')
1269 test('name=; Domain=.yt.dlp', [cookie('name', '', domain='.yt.dlp')],
1270 round_trip='name=""; Domain=.yt.dlp')
1271
1272 test('test=value', [cookie('test', 'value', domain='.yt.dlp')], headers=True)
6c5211ce 1273 test('cookie1=value; Domain=.yt.dlp; cookie2=value', [], headers=True, error_re=r'Invalid syntax')
31215122 1274 ydl.deprecated_feature = ydl.report_error
6c5211ce 1275 test('test=value', [], headers=True, error_re=r'Passing cookies as a header is a potential security risk')
1276
1277 def test_infojson_cookies(self):
1278 TEST_FILE = 'test_infojson_cookies.info.json'
1279 TEST_URL = 'https://example.com/example.mp4'
1280 COOKIES = 'a=b; Domain=.example.com; c=d; Domain=.example.com'
1281 COOKIE_HEADER = {'Cookie': 'a=b; c=d'}
1282
1283 ydl = FakeYDL()
1284 ydl.process_info = lambda x: ydl._write_info_json('test', x, TEST_FILE)
1285
1286 def make_info(info_header_cookies=False, fmts_header_cookies=False, cookies_field=False):
1287 fmt = {'url': TEST_URL}
1288 if fmts_header_cookies:
1289 fmt['http_headers'] = COOKIE_HEADER
1290 if cookies_field:
1291 fmt['cookies'] = COOKIES
1292 return _make_result([fmt], http_headers=COOKIE_HEADER if info_header_cookies else None)
1293
1294 def test(initial_info, note):
1295 result = {}
1296 result['processed'] = ydl.process_ie_result(initial_info)
1297 self.assertTrue(ydl.cookiejar.get_cookies_for_url(TEST_URL),
1298 msg=f'No cookies set in cookiejar after initial process when {note}')
1299 ydl.cookiejar.clear()
1300 with open(TEST_FILE) as infojson:
1301 result['loaded'] = ydl.sanitize_info(json.load(infojson), True)
1302 result['final'] = ydl.process_ie_result(result['loaded'].copy(), download=False)
1303 self.assertTrue(ydl.cookiejar.get_cookies_for_url(TEST_URL),
1304 msg=f'No cookies set in cookiejar after final process when {note}')
1305 ydl.cookiejar.clear()
1306 for key in ('processed', 'loaded', 'final'):
1307 info = result[key]
1308 self.assertIsNone(
1309 traverse_obj(info, ((None, ('formats', 0)), 'http_headers', 'Cookie'), casesense=False, get_all=False),
1310 msg=f'Cookie header not removed in {key} result when {note}')
1311 self.assertEqual(
1312 traverse_obj(info, ((None, ('formats', 0)), 'cookies'), get_all=False), COOKIES,
1313 msg=f'No cookies field found in {key} result when {note}')
1314
1315 test({'url': TEST_URL, 'http_headers': COOKIE_HEADER, 'id': '1', 'title': 'x'}, 'no formats field')
1316 test(make_info(info_header_cookies=True), 'info_dict header cokies')
1317 test(make_info(fmts_header_cookies=True), 'format header cookies')
1318 test(make_info(info_header_cookies=True, fmts_header_cookies=True), 'info_dict and format header cookies')
1319 test(make_info(info_header_cookies=True, fmts_header_cookies=True, cookies_field=True), 'all cookies fields')
1320 test(make_info(cookies_field=True), 'cookies format field')
1321 test({'url': TEST_URL, 'cookies': COOKIES, 'id': '1', 'title': 'x'}, 'info_dict cookies field only')
1322
1323 try_rm(TEST_FILE)
1324
1325 def test_add_headers_cookie(self):
1326 def check_for_cookie_header(result):
1327 return traverse_obj(result, ((None, ('formats', 0)), 'http_headers', 'Cookie'), casesense=False, get_all=False)
1328
1329 ydl = FakeYDL({'http_headers': {'Cookie': 'a=b'}})
1330 ydl._apply_header_cookies(_make_result([])['webpage_url']) # Scope to input webpage URL: .example.com
1331
1332 fmt = {'url': 'https://example.com/video.mp4'}
1333 result = ydl.process_ie_result(_make_result([fmt]), download=False)
1334 self.assertIsNone(check_for_cookie_header(result), msg='http_headers cookies in result info_dict')
1335 self.assertEqual(result.get('cookies'), 'a=b; Domain=.example.com', msg='No cookies were set in cookies field')
1336 self.assertIn('a=b', ydl.cookiejar.get_cookie_header(fmt['url']), msg='No cookies were set in cookiejar')
1337
1338 fmt = {'url': 'https://wrong.com/video.mp4'}
1339 result = ydl.process_ie_result(_make_result([fmt]), download=False)
1340 self.assertIsNone(check_for_cookie_header(result), msg='http_headers cookies for wrong domain')
1341 self.assertFalse(result.get('cookies'), msg='Cookies set in cookies field for wrong domain')
1342 self.assertFalse(ydl.cookiejar.get_cookie_header(fmt['url']), msg='Cookies set in cookiejar for wrong domain')
31215122 1343
2b4ecde2 1344
e028d0d1
JMF
1345if __name__ == '__main__':
1346 unittest.main()