+ }), r'^\s*10k$')
+ assertRegexpMatches(self, ydl._format_note({
+ 'fps': 30,
+ }), r'^30fps$')
+
+ def test_postprocessors(self):
+ filename = 'post-processor-testfile.mp4'
+ audiofile = filename + '.mp3'
+
+ class SimplePP(PostProcessor):
+ def run(self, info):
+ with open(audiofile, 'wt') as f:
+ f.write('EXAMPLE')
+ return [info['filepath']], info
+
+ def run_pp(params, PP):
+ with open(filename, 'wt') as f:
+ f.write('EXAMPLE')
+ ydl = YoutubeDL(params)
+ ydl.add_post_processor(PP())
+ ydl.post_process(filename, {'filepath': filename})
+
+ run_pp({'keepvideo': True}, SimplePP)
+ self.assertTrue(os.path.exists(filename), '%s doesn\'t exist' % filename)
+ self.assertTrue(os.path.exists(audiofile), '%s doesn\'t exist' % audiofile)
+ os.unlink(filename)
+ os.unlink(audiofile)
+
+ run_pp({'keepvideo': False}, SimplePP)
+ self.assertFalse(os.path.exists(filename), '%s exists' % filename)
+ self.assertTrue(os.path.exists(audiofile), '%s doesn\'t exist' % audiofile)
+ os.unlink(audiofile)
+
+ class ModifierPP(PostProcessor):
+ def run(self, info):
+ with open(info['filepath'], 'wt') as f:
+ f.write('MODIFIED')
+ return [], info
+
+ run_pp({'keepvideo': False}, ModifierPP)
+ self.assertTrue(os.path.exists(filename), '%s doesn\'t exist' % filename)
+ os.unlink(filename)
+
+ def test_match_filter(self):
+ class FilterYDL(YDL):
+ def __init__(self, *args, **kwargs):
+ super(FilterYDL, self).__init__(*args, **kwargs)
+ self.params['simulate'] = True
+
+ def process_info(self, info_dict):
+ super(YDL, self).process_info(info_dict)
+
+ def _match_entry(self, info_dict, incomplete):
+ res = super(FilterYDL, self)._match_entry(info_dict, incomplete)
+ if res is None:
+ self.downloaded_info_dicts.append(info_dict)
+ return res
+
+ first = {
+ 'id': '1',
+ 'url': TEST_URL,
+ 'title': 'one',
+ 'extractor': 'TEST',
+ 'duration': 30,
+ 'filesize': 10 * 1024,
+ 'playlist_id': '42',
+ 'uploader': "變態妍字幕版 太妍 тест",
+ 'creator': "тест ' 123 ' тест--",
+ }
+ second = {
+ 'id': '2',
+ 'url': TEST_URL,
+ 'title': 'two',
+ 'extractor': 'TEST',
+ 'duration': 10,
+ 'description': 'foo',
+ 'filesize': 5 * 1024,
+ 'playlist_id': '43',
+ 'uploader': "тест 123",
+ }
+ videos = [first, second]
+
+ def get_videos(filter_=None):
+ ydl = FilterYDL({'match_filter': filter_})
+ for v in videos:
+ ydl.process_ie_result(v, download=True)
+ return [v['id'] for v in ydl.downloaded_info_dicts]
+
+ res = get_videos()
+ self.assertEqual(res, ['1', '2'])
+
+ def f(v):
+ if v['id'] == '1':
+ return None
+ else:
+ return 'Video id is not 1'
+ res = get_videos(f)
+ self.assertEqual(res, ['1'])
+
+ f = match_filter_func('duration < 30')
+ res = get_videos(f)
+ self.assertEqual(res, ['2'])
+
+ f = match_filter_func('description = foo')
+ res = get_videos(f)
+ self.assertEqual(res, ['2'])
+
+ f = match_filter_func('description =? foo')
+ res = get_videos(f)
+ self.assertEqual(res, ['1', '2'])
+
+ f = match_filter_func('filesize > 5KiB')
+ res = get_videos(f)
+ self.assertEqual(res, ['1'])
+
+ f = match_filter_func('playlist_id = 42')
+ res = get_videos(f)
+ self.assertEqual(res, ['1'])
+
+ f = match_filter_func('uploader = "變態妍字幕版 太妍 тест"')
+ res = get_videos(f)
+ self.assertEqual(res, ['1'])
+
+ f = match_filter_func('uploader != "變態妍字幕版 太妍 тест"')
+ res = get_videos(f)
+ self.assertEqual(res, ['2'])
+
+ f = match_filter_func('creator = "тест \' 123 \' тест--"')
+ res = get_videos(f)
+ self.assertEqual(res, ['1'])
+
+ f = match_filter_func("creator = 'тест \\' 123 \\' тест--'")
+ res = get_videos(f)
+ self.assertEqual(res, ['1'])
+
+ f = match_filter_func(r"creator = 'тест \' 123 \' тест--' & duration > 30")
+ res = get_videos(f)
+ self.assertEqual(res, [])
+
+ def test_playlist_items_selection(self):
+ entries = [{
+ 'id': compat_str(i),
+ 'title': compat_str(i),
+ 'url': TEST_URL,
+ } for i in range(1, 5)]
+ playlist = {
+ '_type': 'playlist',
+ 'id': 'test',
+ 'entries': entries,
+ 'extractor': 'test:playlist',
+ 'extractor_key': 'test:playlist',
+ 'webpage_url': 'http://example.com',
+ }
+
+ def get_downloaded_info_dicts(params):
+ ydl = YDL(params)
+ # make a deep copy because the dictionary and nested entries
+ # can be modified
+ ydl.process_ie_result(copy.deepcopy(playlist))
+ return ydl.downloaded_info_dicts
+
+ def get_ids(params):
+ return [int(v['id']) for v in get_downloaded_info_dicts(params)]
+
+ result = get_ids({})
+ self.assertEqual(result, [1, 2, 3, 4])
+
+ result = get_ids({'playlistend': 10})
+ self.assertEqual(result, [1, 2, 3, 4])
+
+ result = get_ids({'playlistend': 2})
+ self.assertEqual(result, [1, 2])
+
+ result = get_ids({'playliststart': 10})
+ self.assertEqual(result, [])
+
+ result = get_ids({'playliststart': 2})
+ self.assertEqual(result, [2, 3, 4])
+
+ result = get_ids({'playlist_items': '2-4'})
+ self.assertEqual(result, [2, 3, 4])
+
+ result = get_ids({'playlist_items': '2,4'})
+ self.assertEqual(result, [2, 4])
+
+ result = get_ids({'playlist_items': '10'})
+ self.assertEqual(result, [])
+
+ result = get_ids({'playlist_items': '3-10'})
+ self.assertEqual(result, [3, 4])
+
+ result = get_ids({'playlist_items': '2-4,3-4,3'})
+ self.assertEqual(result, [2, 3, 4])
+
+ # Tests for https://github.com/ytdl-org/youtube-dl/issues/10591
+ # @{
+ result = get_downloaded_info_dicts({'playlist_items': '2-4,3-4,3'})
+ self.assertEqual(result[0]['playlist_index'], 2)
+ self.assertEqual(result[1]['playlist_index'], 3)
+
+ result = get_downloaded_info_dicts({'playlist_items': '2-4,3-4,3'})
+ self.assertEqual(result[0]['playlist_index'], 2)
+ self.assertEqual(result[1]['playlist_index'], 3)
+ self.assertEqual(result[2]['playlist_index'], 4)
+
+ result = get_downloaded_info_dicts({'playlist_items': '4,2'})
+ self.assertEqual(result[0]['playlist_index'], 4)
+ self.assertEqual(result[1]['playlist_index'], 2)
+ # @}
+
+ def test_urlopen_no_file_protocol(self):
+ # see https://github.com/ytdl-org/youtube-dl/issues/8227
+ ydl = YDL()
+ self.assertRaises(compat_urllib_error.URLError, ydl.urlopen, 'file:///etc/passwd')
+
+ def test_do_not_override_ie_key_in_url_transparent(self):
+ ydl = YDL()
+
+ class Foo1IE(InfoExtractor):
+ _VALID_URL = r'foo1:'
+
+ def _real_extract(self, url):
+ return {
+ '_type': 'url_transparent',
+ 'url': 'foo2:',
+ 'ie_key': 'Foo2',
+ 'title': 'foo1 title',
+ 'id': 'foo1_id',
+ }
+
+ class Foo2IE(InfoExtractor):
+ _VALID_URL = r'foo2:'
+
+ def _real_extract(self, url):
+ return {
+ '_type': 'url',
+ 'url': 'foo3:',
+ 'ie_key': 'Foo3',
+ }
+
+ class Foo3IE(InfoExtractor):
+ _VALID_URL = r'foo3:'
+
+ def _real_extract(self, url):
+ return _make_result([{'url': TEST_URL}], title='foo3 title')
+
+ ydl.add_info_extractor(Foo1IE(ydl))
+ ydl.add_info_extractor(Foo2IE(ydl))
+ ydl.add_info_extractor(Foo3IE(ydl))
+ ydl.extract_info('foo1:')
+ downloaded = ydl.downloaded_info_dicts[0]
+ self.assertEqual(downloaded['url'], TEST_URL)
+ self.assertEqual(downloaded['title'], 'foo1 title')
+ self.assertEqual(downloaded['id'], 'testid')
+ self.assertEqual(downloaded['extractor'], 'testex')
+ self.assertEqual(downloaded['extractor_key'], 'TestEx')
+