+ def test_get_elements_html_by_attribute(self):
+ html = self.GET_ELEMENTS_BY_CLASS_TEST_STRING
+
+ self.assertEqual(get_elements_html_by_attribute('class', 'foo bar', html), self.GET_ELEMENTS_BY_CLASS_RES)
+ self.assertEqual(get_elements_html_by_attribute('class', 'foo', html), [])
+ self.assertEqual(get_elements_html_by_attribute('class', 'no-such-foo', html), [])
+
+ def test_get_elements_text_and_html_by_attribute(self):
+ html = self.GET_ELEMENTS_BY_CLASS_TEST_STRING
+
+ self.assertEqual(
+ list(get_elements_text_and_html_by_attribute('class', 'foo bar', html)),
+ list(zip(['nice', 'also nice'], self.GET_ELEMENTS_BY_CLASS_RES)))
+ self.assertEqual(list(get_elements_text_and_html_by_attribute('class', 'foo', html)), [])
+ self.assertEqual(list(get_elements_text_and_html_by_attribute('class', 'no-such-foo', html)), [])
+
+ self.assertEqual(list(get_elements_text_and_html_by_attribute(
+ 'class', 'foo', '<a class="foo">nice</a><span class="foo">nice</span>', tag='a')), [('nice', '<a class="foo">nice</a>')])
+
+ GET_ELEMENT_BY_TAG_TEST_STRING = '''
+ random text lorem ipsum</p>
+ <div>
+ this should be returned
+ <span>this should also be returned</span>
+ <div>
+ this should also be returned
+ </div>
+ closing tag above should not trick, so this should also be returned
+ </div>
+ but this text should not be returned
+ '''
+ GET_ELEMENT_BY_TAG_RES_OUTERDIV_HTML = GET_ELEMENT_BY_TAG_TEST_STRING.strip()[32:276]
+ GET_ELEMENT_BY_TAG_RES_OUTERDIV_TEXT = GET_ELEMENT_BY_TAG_RES_OUTERDIV_HTML[5:-6]
+ GET_ELEMENT_BY_TAG_RES_INNERSPAN_HTML = GET_ELEMENT_BY_TAG_TEST_STRING.strip()[78:119]
+ GET_ELEMENT_BY_TAG_RES_INNERSPAN_TEXT = GET_ELEMENT_BY_TAG_RES_INNERSPAN_HTML[6:-7]
+
+ def test_get_element_text_and_html_by_tag(self):
+ html = self.GET_ELEMENT_BY_TAG_TEST_STRING
+
+ self.assertEqual(
+ get_element_text_and_html_by_tag('div', html),
+ (self.GET_ELEMENT_BY_TAG_RES_OUTERDIV_TEXT, self.GET_ELEMENT_BY_TAG_RES_OUTERDIV_HTML))
+ self.assertEqual(
+ get_element_text_and_html_by_tag('span', html),
+ (self.GET_ELEMENT_BY_TAG_RES_INNERSPAN_TEXT, self.GET_ELEMENT_BY_TAG_RES_INNERSPAN_HTML))
+ self.assertRaises(compat_HTMLParseError, get_element_text_and_html_by_tag, 'article', html)
+
+ def test_iri_to_uri(self):
+ self.assertEqual(
+ iri_to_uri('https://www.google.com/search?q=foo&ie=utf-8&oe=utf-8&client=firefox-b'),
+ 'https://www.google.com/search?q=foo&ie=utf-8&oe=utf-8&client=firefox-b') # Same
+ self.assertEqual(
+ iri_to_uri('https://www.google.com/search?q=Käsesoßenrührlöffel'), # German for cheese sauce stirring spoon
+ 'https://www.google.com/search?q=K%C3%A4seso%C3%9Fenr%C3%BChrl%C3%B6ffel')
+ self.assertEqual(
+ iri_to_uri('https://www.google.com/search?q=lt<+gt>+eq%3D+amp%26+percent%25+hash%23+colon%3A+tilde~#trash=?&garbage=#'),
+ 'https://www.google.com/search?q=lt%3C+gt%3E+eq%3D+amp%26+percent%25+hash%23+colon%3A+tilde~#trash=?&garbage=#')
+ self.assertEqual(
+ iri_to_uri('http://правозащита38.рф/category/news/'),
+ 'http://xn--38-6kcaak9aj5chl4a3g.xn--p1ai/category/news/')
+ self.assertEqual(
+ iri_to_uri('http://www.правозащита38.рф/category/news/'),
+ 'http://www.xn--38-6kcaak9aj5chl4a3g.xn--p1ai/category/news/')
+ self.assertEqual(
+ iri_to_uri('https://i❤.ws/emojidomain/👍👏🤝💪'),
+ 'https://xn--i-7iq.ws/emojidomain/%F0%9F%91%8D%F0%9F%91%8F%F0%9F%A4%9D%F0%9F%92%AA')
+ self.assertEqual(
+ iri_to_uri('http://日本語.jp/'),
+ 'http://xn--wgv71a119e.jp/')
+ self.assertEqual(
+ iri_to_uri('http://导航.中国/'),
+ 'http://xn--fet810g.xn--fiqs8s/')
+
+ def test_clean_podcast_url(self):
+ self.assertEqual(clean_podcast_url('https://www.podtrac.com/pts/redirect.mp3/chtbl.com/track/5899E/traffic.megaphone.fm/HSW7835899191.mp3'), 'https://traffic.megaphone.fm/HSW7835899191.mp3')
+ self.assertEqual(clean_podcast_url('https://play.podtrac.com/npr-344098539/edge1.pod.npr.org/anon.npr-podcasts/podcast/npr/waitwait/2020/10/20201003_waitwait_wwdtmpodcast201003-015621a5-f035-4eca-a9a1-7c118d90bc3c.mp3'), 'https://edge1.pod.npr.org/anon.npr-podcasts/podcast/npr/waitwait/2020/10/20201003_waitwait_wwdtmpodcast201003-015621a5-f035-4eca-a9a1-7c118d90bc3c.mp3')
+ self.assertEqual(clean_podcast_url('https://pdst.fm/e/2.gum.fm/chtbl.com/track/chrt.fm/track/34D33/pscrb.fm/rss/p/traffic.megaphone.fm/ITLLC7765286967.mp3?updated=1687282661'), 'https://traffic.megaphone.fm/ITLLC7765286967.mp3?updated=1687282661')
+ self.assertEqual(clean_podcast_url('https://pdst.fm/e/https://mgln.ai/e/441/www.buzzsprout.com/1121972/13019085-ep-252-the-deep-life-stack.mp3'), 'https://www.buzzsprout.com/1121972/13019085-ep-252-the-deep-life-stack.mp3')
+
+ def test_LazyList(self):
+ it = list(range(10))
+
+ self.assertEqual(list(LazyList(it)), it)
+ self.assertEqual(LazyList(it).exhaust(), it)
+ self.assertEqual(LazyList(it)[5], it[5])
+
+ self.assertEqual(LazyList(it)[5:], it[5:])
+ self.assertEqual(LazyList(it)[:5], it[:5])
+ self.assertEqual(LazyList(it)[::2], it[::2])
+ self.assertEqual(LazyList(it)[1::2], it[1::2])
+ self.assertEqual(LazyList(it)[5::-1], it[5::-1])
+ self.assertEqual(LazyList(it)[6:2:-2], it[6:2:-2])
+ self.assertEqual(LazyList(it)[::-1], it[::-1])
+
+ self.assertTrue(LazyList(it))
+ self.assertFalse(LazyList(range(0)))
+ self.assertEqual(len(LazyList(it)), len(it))
+ self.assertEqual(repr(LazyList(it)), repr(it))
+ self.assertEqual(str(LazyList(it)), str(it))
+
+ self.assertEqual(list(LazyList(it, reverse=True)), it[::-1])
+ self.assertEqual(list(reversed(LazyList(it))[::-1]), it)
+ self.assertEqual(list(reversed(LazyList(it))[1:3:7]), it[::-1][1:3:7])
+
+ def test_LazyList_laziness(self):
+
+ def test(ll, idx, val, cache):
+ self.assertEqual(ll[idx], val)
+ self.assertEqual(ll._cache, list(cache))
+
+ ll = LazyList(range(10))
+ test(ll, 0, 0, range(1))
+ test(ll, 5, 5, range(6))
+ test(ll, -3, 7, range(10))
+
+ ll = LazyList(range(10), reverse=True)
+ test(ll, -1, 0, range(1))
+ test(ll, 3, 6, range(10))
+
+ ll = LazyList(itertools.count())
+ test(ll, 10, 10, range(11))
+ ll = reversed(ll)
+ test(ll, -15, 14, range(15))
+
+ def test_format_bytes(self):
+ self.assertEqual(format_bytes(0), '0.00B')
+ self.assertEqual(format_bytes(1000), '1000.00B')
+ self.assertEqual(format_bytes(1024), '1.00KiB')
+ self.assertEqual(format_bytes(1024**2), '1.00MiB')
+ self.assertEqual(format_bytes(1024**3), '1.00GiB')
+ self.assertEqual(format_bytes(1024**4), '1.00TiB')
+ self.assertEqual(format_bytes(1024**5), '1.00PiB')
+ self.assertEqual(format_bytes(1024**6), '1.00EiB')
+ self.assertEqual(format_bytes(1024**7), '1.00ZiB')
+ self.assertEqual(format_bytes(1024**8), '1.00YiB')
+ self.assertEqual(format_bytes(1024**9), '1024.00YiB')
+
+ def test_hide_login_info(self):
+ self.assertEqual(Config.hide_login_info(['-u', 'foo', '-p', 'bar']),
+ ['-u', 'PRIVATE', '-p', 'PRIVATE'])
+ self.assertEqual(Config.hide_login_info(['-u']), ['-u'])
+ self.assertEqual(Config.hide_login_info(['-u', 'foo', '-u', 'bar']),
+ ['-u', 'PRIVATE', '-u', 'PRIVATE'])
+ self.assertEqual(Config.hide_login_info(['--username=foo']),
+ ['--username=PRIVATE'])
+
+ def test_locked_file(self):
+ TEXT = 'test_locked_file\n'
+ FILE = 'test_locked_file.ytdl'
+ MODES = 'war' # Order is important
+
+ try:
+ for lock_mode in MODES:
+ with locked_file(FILE, lock_mode, False) as f:
+ if lock_mode == 'r':
+ self.assertEqual(f.read(), TEXT * 2, 'Wrong file content')
+ else:
+ f.write(TEXT)
+ for test_mode in MODES:
+ testing_write = test_mode != 'r'
+ try:
+ with locked_file(FILE, test_mode, False):
+ pass
+ except (BlockingIOError, PermissionError):
+ if not testing_write: # FIXME
+ print(f'Known issue: Exclusive lock ({lock_mode}) blocks read access ({test_mode})')
+ continue
+ self.assertTrue(testing_write, f'{test_mode} is blocked by {lock_mode}')
+ else:
+ self.assertFalse(testing_write, f'{test_mode} is not blocked by {lock_mode}')
+ finally:
+ with contextlib.suppress(OSError):
+ os.remove(FILE)
+
+ def test_determine_file_encoding(self):
+ self.assertEqual(determine_file_encoding(b''), (None, 0))
+ self.assertEqual(determine_file_encoding(b'--verbose -x --audio-format mkv\n'), (None, 0))
+
+ self.assertEqual(determine_file_encoding(b'\xef\xbb\xbf'), ('utf-8', 3))
+ self.assertEqual(determine_file_encoding(b'\x00\x00\xfe\xff'), ('utf-32-be', 4))
+ self.assertEqual(determine_file_encoding(b'\xff\xfe'), ('utf-16-le', 2))
+
+ self.assertEqual(determine_file_encoding(b'\xff\xfe# coding: utf-8\n--verbose'), ('utf-16-le', 2))
+
+ self.assertEqual(determine_file_encoding(b'# coding: utf-8\n--verbose'), ('utf-8', 0))
+ self.assertEqual(determine_file_encoding(b'# coding: someencodinghere-12345\n--verbose'), ('someencodinghere-12345', 0))
+
+ self.assertEqual(determine_file_encoding(b'#coding:utf-8\n--verbose'), ('utf-8', 0))
+ self.assertEqual(determine_file_encoding(b'# coding: utf-8 \r\n--verbose'), ('utf-8', 0))
+
+ self.assertEqual(determine_file_encoding('# coding: utf-32-be'.encode('utf-32-be')), ('utf-32-be', 0))
+ self.assertEqual(determine_file_encoding('# coding: utf-16-le'.encode('utf-16-le')), ('utf-16-le', 0))
+
+ def test_get_compatible_ext(self):
+ self.assertEqual(get_compatible_ext(
+ vcodecs=[None], acodecs=[None, None], vexts=['mp4'], aexts=['m4a', 'm4a']), 'mkv')
+ self.assertEqual(get_compatible_ext(
+ vcodecs=[None], acodecs=[None], vexts=['flv'], aexts=['flv']), 'flv')
+
+ self.assertEqual(get_compatible_ext(
+ vcodecs=[None], acodecs=[None], vexts=['mp4'], aexts=['m4a']), 'mp4')
+ self.assertEqual(get_compatible_ext(
+ vcodecs=[None], acodecs=[None], vexts=['mp4'], aexts=['webm']), 'mkv')
+ self.assertEqual(get_compatible_ext(
+ vcodecs=[None], acodecs=[None], vexts=['webm'], aexts=['m4a']), 'mkv')
+ self.assertEqual(get_compatible_ext(
+ vcodecs=[None], acodecs=[None], vexts=['webm'], aexts=['webm']), 'webm')
+ self.assertEqual(get_compatible_ext(
+ vcodecs=[None], acodecs=[None], vexts=['webm'], aexts=['weba']), 'webm')
+
+ self.assertEqual(get_compatible_ext(
+ vcodecs=['h264'], acodecs=['mp4a'], vexts=['mov'], aexts=['m4a']), 'mp4')
+ self.assertEqual(get_compatible_ext(
+ vcodecs=['av01.0.12M.08'], acodecs=['opus'], vexts=['mp4'], aexts=['webm']), 'webm')
+
+ self.assertEqual(get_compatible_ext(
+ vcodecs=['vp9'], acodecs=['opus'], vexts=['webm'], aexts=['webm'], preferences=['flv', 'mp4']), 'mp4')
+ self.assertEqual(get_compatible_ext(
+ vcodecs=['av1'], acodecs=['mp4a'], vexts=['webm'], aexts=['m4a'], preferences=('webm', 'mkv')), 'mkv')
+
+ def test_try_call(self):
+ def total(*x, **kwargs):
+ return sum(x) + sum(kwargs.values())
+
+ self.assertEqual(try_call(None), None,
+ msg='not a fn should give None')
+ self.assertEqual(try_call(lambda: 1), 1,
+ msg='int fn with no expected_type should give int')
+ self.assertEqual(try_call(lambda: 1, expected_type=int), 1,
+ msg='int fn with expected_type int should give int')
+ self.assertEqual(try_call(lambda: 1, expected_type=dict), None,
+ msg='int fn with wrong expected_type should give None')
+ self.assertEqual(try_call(total, args=(0, 1, 0, ), expected_type=int), 1,
+ msg='fn should accept arglist')
+ self.assertEqual(try_call(total, kwargs={'a': 0, 'b': 1, 'c': 0}, expected_type=int), 1,
+ msg='fn should accept kwargs')
+ self.assertEqual(try_call(lambda: 1, expected_type=dict), None,
+ msg='int fn with no expected_type should give None')
+ self.assertEqual(try_call(lambda x: {}, total, args=(42, ), expected_type=int), 42,
+ msg='expect first int result with expected_type int')
+
+ def test_variadic(self):
+ self.assertEqual(variadic(None), (None, ))
+ self.assertEqual(variadic('spam'), ('spam', ))
+ self.assertEqual(variadic('spam', allowed_types=dict), 'spam')
+ with warnings.catch_warnings():
+ warnings.simplefilter('ignore')
+ self.assertEqual(variadic('spam', allowed_types=[dict]), 'spam')
+
+ def test_http_header_dict(self):
+ headers = HTTPHeaderDict()
+ headers['ytdl-test'] = b'0'
+ self.assertEqual(list(headers.items()), [('Ytdl-Test', '0')])
+ headers['ytdl-test'] = 1
+ self.assertEqual(list(headers.items()), [('Ytdl-Test', '1')])
+ headers['Ytdl-test'] = '2'
+ self.assertEqual(list(headers.items()), [('Ytdl-Test', '2')])
+ self.assertTrue('ytDl-Test' in headers)
+ self.assertEqual(str(headers), str(dict(headers)))
+ self.assertEqual(repr(headers), str(dict(headers)))
+
+ headers.update({'X-dlp': 'data'})
+ self.assertEqual(set(headers.items()), {('Ytdl-Test', '2'), ('X-Dlp', 'data')})
+ self.assertEqual(dict(headers), {'Ytdl-Test': '2', 'X-Dlp': 'data'})
+ self.assertEqual(len(headers), 2)
+ self.assertEqual(headers.copy(), headers)
+ headers2 = HTTPHeaderDict({'X-dlp': 'data3'}, **headers, **{'X-dlp': 'data2'})
+ self.assertEqual(set(headers2.items()), {('Ytdl-Test', '2'), ('X-Dlp', 'data2')})
+ self.assertEqual(len(headers2), 2)
+ headers2.clear()
+ self.assertEqual(len(headers2), 0)
+
+ # ensure we prefer latter headers
+ headers3 = HTTPHeaderDict({'Ytdl-TeSt': 1}, {'Ytdl-test': 2})
+ self.assertEqual(set(headers3.items()), {('Ytdl-Test', '2')})
+ del headers3['ytdl-tesT']
+ self.assertEqual(dict(headers3), {})
+
+ headers4 = HTTPHeaderDict({'ytdl-test': 'data;'})
+ self.assertEqual(set(headers4.items()), {('Ytdl-Test', 'data;')})
+
+ # common mistake: strip whitespace from values
+ # https://github.com/yt-dlp/yt-dlp/issues/8729
+ headers5 = HTTPHeaderDict({'ytdl-test': ' data; '})
+ self.assertEqual(set(headers5.items()), {('Ytdl-Test', 'data;')})
+
+ def test_extract_basic_auth(self):
+ assert extract_basic_auth('http://:foo.bar') == ('http://:foo.bar', None)
+ assert extract_basic_auth('http://foo.bar') == ('http://foo.bar', None)
+ assert extract_basic_auth('http://@foo.bar') == ('http://foo.bar', 'Basic Og==')
+ assert extract_basic_auth('http://:pass@foo.bar') == ('http://foo.bar', 'Basic OnBhc3M=')
+ assert extract_basic_auth('http://user:@foo.bar') == ('http://foo.bar', 'Basic dXNlcjo=')
+ assert extract_basic_auth('http://user:pass@foo.bar') == ('http://foo.bar', 'Basic dXNlcjpwYXNz')
+
+ @unittest.skipUnless(compat_os_name == 'nt', 'Only relevant on Windows')
+ def test_windows_escaping(self):
+ tests = [
+ 'test"&',
+ '%CMDCMDLINE:~-1%&',
+ 'a\nb',
+ '"',
+ '\\',
+ '!',
+ '^!',
+ 'a \\ b',
+ 'a \\" b',
+ 'a \\ b\\',
+ # We replace \r with \n
+ ('a\r\ra', 'a\n\na'),
+ ]
+
+ def run_shell(args):
+ stdout, stderr, error = Popen.run(
+ args, text=True, shell=True, stdout=subprocess.PIPE, stderr=subprocess.PIPE)
+ assert not stderr
+ assert not error
+ return stdout
+
+ for argument in tests:
+ if isinstance(argument, str):
+ expected = argument
+ else:
+ argument, expected = argument
+
+ args = [sys.executable, '-c', 'import sys; print(end=sys.argv[1])', argument, 'end']
+ assert run_shell(args) == expected
+ assert run_shell(shell_quote(args, shell=True)) == expected
+