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