Compare commits

...

11 Commits

Author SHA1 Message Date
dirkf
03b354a729
Merge a631e79b3344dd6dfbca3571940870be97cc6afc into 2b4fbfce25902d557b86b003cf48f738129efce4 2025-03-26 07:21:53 +00:00
dirkf
2b4fbfce25 [YouTube] Support player 4fcd6e4a
thx seproDev, bashonly: yt-dlp/yt-dlp#12748
2025-03-26 02:27:25 +00:00
dirkf
1bc45b8b6c [JSInterp] Use , for join() with null/undefined argument
Eg: [1,2,3].join(null) -> '1,2,3'
2025-03-25 22:35:06 +00:00
dirkf
b982d77d0b [YouTube] Align signature tests with yt-dlp
thx bashonly, yt-dlp/yt-dlp#12725
2025-03-25 22:35:06 +00:00
dirkf
c55dbf4838 [YouTube] Update signature extraction for players 643afba4, 363db69b 2025-03-25 22:35:06 +00:00
dirkf
087d865230 [YouTube] Support new player URL patterns 2025-03-25 22:35:06 +00:00
dirkf
a4fc1151f1 [JSInterp] Improve indexing
* catch invalid list index with `ValueError` (eg [1, 2]['ab'] -> undefined)
* allow assignment outside existing list (eg var l = [1,2]; l[9] = 0;)
2025-03-25 22:35:05 +00:00
dirkf
a464c159e6 [YouTube] Make _extract_player_info() use _search_regex() 2025-03-25 22:35:05 +00:00
dirkf
7dca08eff0 [YouTube] Also get original of translated automatic captions 2025-03-25 22:35:05 +00:00
dirkf
2239ee7965 [YouTube] Get subtitles/automatic captions from both web and API responses 2025-03-25 22:35:05 +00:00
dirkf
a631e79b33 [core] Implement a simple version of yt-dlp's --print option 2022-03-05 13:58:02 +00:00
7 changed files with 165 additions and 58 deletions

View File

@ -310,6 +310,9 @@ Alternatively, refer to the [developer instructions](#developer-instructions) fo
--get-filename Simulate, quiet but print output --get-filename Simulate, quiet but print output
filename filename
--get-format Simulate, quiet but print output format --get-format Simulate, quiet but print output format
-O, --print TEMPLATE Simulate, quiet but print the given fields.
Either a field name or similar formatting
as the output template can be used
-j, --dump-json Simulate, quiet but print JSON -j, --dump-json Simulate, quiet but print JSON
information. See the "OUTPUT TEMPLATE" information. See the "OUTPUT TEMPLATE"
for a description of available keys. for a description of available keys.
@ -620,6 +623,12 @@ Available for the media that is a track or a part of a music album:
- `disc_number` (numeric): Number of the disc or other physical medium the track belongs to - `disc_number` (numeric): Number of the disc or other physical medium the track belongs to
- `release_year` (numeric): Year (YYYY) when the album was released - `release_year` (numeric): Year (YYYY) when the album was released
Available only when used in `--print`:
- `urls` (string): The URLs of all requested formats, one in each line
- `duration_string` (string): Length of the video (HH:mm:ss)
- `filename` (string): Name of the video file. Note that the actual filename may be different due to post-processing. Use `--exec echo` to get the name after all postprocessing is complete
Each aforementioned sequence when referenced in an output template will be replaced by the actual value corresponding to the sequence name. Note that some of the sequences are not guaranteed to be present since they depend on the metadata obtained by a particular extractor. Such sequences will be replaced with placeholder value provided with `--output-na-placeholder` (`NA` by default). Each aforementioned sequence when referenced in an output template will be replaced by the actual value corresponding to the sequence name. Note that some of the sequences are not guaranteed to be present since they depend on the metadata obtained by a particular extractor. Such sequences will be replaced with placeholder value provided with `--output-na-placeholder` (`NA` by default).
For example for `-o %(title)s-%(id)s.%(ext)s` and an mp4 video with title `youtube-dl test video` and id `BaW_jenozKcj`, this will result in a `youtube-dl test video-BaW_jenozKcj.mp4` file created in the current directory. For example for `-o %(title)s-%(id)s.%(ext)s` and an mp4 video with title `youtube-dl test video` and id `BaW_jenozKcj`, this will result in a `youtube-dl test video-BaW_jenozKcj.mp4` file created in the current directory.

View File

@ -84,6 +84,21 @@ _SIG_TESTS = [
'2aq0aqSyOoJXtK73m-uME_jv7-pT15gOFC02RFkGMqWpzEICs69VdbwQ0LDp1v7j8xx92efCJlYFYb1sUkkBSPOlPmXgIARw8JQ0qOAOAA', '2aq0aqSyOoJXtK73m-uME_jv7-pT15gOFC02RFkGMqWpzEICs69VdbwQ0LDp1v7j8xx92efCJlYFYb1sUkkBSPOlPmXgIARw8JQ0qOAOAA',
'0QJ8wRAIgXmPlOPSBkkUs1bYFYlJCfe29xxAj7v1pDL0QwbdV96sCIEzpWqMGkFR20CFOg51Tp-7vj_EMu-m37KtXJ2OySqa0q', '0QJ8wRAIgXmPlOPSBkkUs1bYFYlJCfe29xxAj7v1pDL0QwbdV96sCIEzpWqMGkFR20CFOg51Tp-7vj_EMu-m37KtXJ2OySqa0q',
), ),
(
'https://www.youtube.com/s/player/643afba4/tv-player-ias.vflset/tv-player-ias.js',
'2aq0aqSyOoJXtK73m-uME_jv7-pT15gOFC02RFkGMqWpzEICs69VdbwQ0LDp1v7j8xx92efCJlYFYb1sUkkBSPOlPmXgIARw8JQ0qOAOAA',
'AAOAOq0QJ8wRAIgXmPlOPSBkkUs1bYFYlJCfe29xx8j7vgpDL0QwbdV06sCIEzpWqMGkFR20CFOS21Tp-7vj_EMu-m37KtXJoOy1',
),
(
'https://www.youtube.com/s/player/363db69b/player_ias.vflset/en_US/base.js',
'2aq0aqSyOoJXtK73m-uME_jv7-pT15gOFC02RFkGMqWpzEICs69VdbwQ0LDp1v7j8xx92efCJlYFYb1sUkkBSPOlPmXgIARw8JQ0qOAOAA',
'0aqSyOoJXtK73m-uME_jv7-pT15gOFC02RFkGMqWpz2ICs6EVdbwQ0LDp1v7j8xx92efCJlYFYb1sUkkBSPOlPmXgIARw8JQ0qOAOAA',
),
(
'https://www.youtube.com/s/player/4fcd6e4a/player_ias.vflset/en_US/base.js',
'2aq0aqSyOoJXtK73m-uME_jv7-pT15gOFC02RFkGMqWpzEICs69VdbwQ0LDp1v7j8xx92efCJlYFYb1sUkkBSPOlPmXgIARw8JQ0qOAOAA',
'wAOAOq0QJ8ARAIgXmPlOPSBkkUs1bYFYlJCfe29xx8q7v1pDL0QwbdV96sCIEzpWqMGkFR20CFOg51Tp-7vj_EMu-m37KtXJoOySqa0',
),
] ]
_NSIG_TESTS = [ _NSIG_TESTS = [
@ -153,7 +168,7 @@ _NSIG_TESTS = [
), ),
( (
'https://www.youtube.com/s/player/c57c113c/player_ias.vflset/en_US/base.js', 'https://www.youtube.com/s/player/c57c113c/player_ias.vflset/en_US/base.js',
'-Txvy6bT5R6LqgnQNx', 'dcklJCnRUHbgSg', 'M92UUMHa8PdvPd3wyM', '3hPqLJsiNZx7yA',
), ),
( (
'https://www.youtube.com/s/player/5a3b6271/player_ias.vflset/en_US/base.js', 'https://www.youtube.com/s/player/5a3b6271/player_ias.vflset/en_US/base.js',
@ -173,7 +188,7 @@ _NSIG_TESTS = [
), ),
( (
'https://www.youtube.com/s/player/cfa9e7cb/player_ias.vflset/en_US/base.js', 'https://www.youtube.com/s/player/cfa9e7cb/player_ias.vflset/en_US/base.js',
'qO0NiMtYQ7TeJnfFG2', 'k9cuJDHNS5O7kQ', 'aCi3iElgd2kq0bxVbQ', 'QX1y8jGb2IbZ0w',
), ),
( (
'https://www.youtube.com/s/player/8c7583ff/player_ias.vflset/en_US/base.js', 'https://www.youtube.com/s/player/8c7583ff/player_ias.vflset/en_US/base.js',
@ -231,10 +246,6 @@ _NSIG_TESTS = [
'https://www.youtube.com/s/player/f6e09c70/player_ias_tce.vflset/en_US/base.js', 'https://www.youtube.com/s/player/f6e09c70/player_ias_tce.vflset/en_US/base.js',
'W9HJZKktxuYoDTqW', 'jHbbkcaxm54', 'W9HJZKktxuYoDTqW', 'jHbbkcaxm54',
), ),
(
'https://www.youtube.com/s/player/643afba4/player_ias.vflset/en_US/base.js',
'W9HJZKktxuYoDTqW', 'larxUlagTRAcSw',
),
( (
'https://www.youtube.com/s/player/e7567ecf/player_ias_tce.vflset/en_US/base.js', 'https://www.youtube.com/s/player/e7567ecf/player_ias_tce.vflset/en_US/base.js',
'Sy4aDGc0VpYRR9ew_', '5UPOT1VhoZxNLQ', 'Sy4aDGc0VpYRR9ew_', '5UPOT1VhoZxNLQ',
@ -259,6 +270,22 @@ _NSIG_TESTS = [
'https://www.youtube.com/s/player/643afba4/tv-player-ias.vflset/tv-player-ias.js', 'https://www.youtube.com/s/player/643afba4/tv-player-ias.vflset/tv-player-ias.js',
'ir9-V6cdbCiyKxhr', '2PL7ZDYAALMfmA', 'ir9-V6cdbCiyKxhr', '2PL7ZDYAALMfmA',
), ),
(
'https://www.youtube.com/s/player/643afba4/player_ias.vflset/en_US/base.js',
'W9HJZKktxuYoDTqW', 'larxUlagTRAcSw',
),
(
'https://www.youtube.com/s/player/363db69b/player_ias.vflset/en_US/base.js',
'eWYu5d5YeY_4LyEDc', 'XJQqf-N7Xra3gg',
),
(
'https://www.youtube.com/s/player/4fcd6e4a/player_ias.vflset/en_US/base.js',
'o_L251jm8yhZkWtBW', 'lXoxI3XvToqn6A',
),
(
'https://www.youtube.com/s/player/4fcd6e4a/tv-player-ias.vflset/tv-player-ias.js',
'o_L251jm8yhZkWtBW', 'lXoxI3XvToqn6A',
),
] ]
@ -271,6 +298,8 @@ class TestPlayerInfo(unittest.TestCase):
('https://www.youtube.com/s/player/64dddad9/player-plasma-ias-phone-en_US.vflset/base.js', '64dddad9'), ('https://www.youtube.com/s/player/64dddad9/player-plasma-ias-phone-en_US.vflset/base.js', '64dddad9'),
('https://www.youtube.com/s/player/64dddad9/player-plasma-ias-phone-de_DE.vflset/base.js', '64dddad9'), ('https://www.youtube.com/s/player/64dddad9/player-plasma-ias-phone-de_DE.vflset/base.js', '64dddad9'),
('https://www.youtube.com/s/player/64dddad9/player-plasma-ias-tablet-en_US.vflset/base.js', '64dddad9'), ('https://www.youtube.com/s/player/64dddad9/player-plasma-ias-tablet-en_US.vflset/base.js', '64dddad9'),
('https://www.youtube.com/s/player/e7567ecf/player_ias_tce.vflset/en_US/base.js', 'e7567ecf'),
('https://www.youtube.com/s/player/643afba4/tv-player-ias.vflset/tv-player-ias.js', '643afba4'),
# obsolete # obsolete
('https://www.youtube.com/yts/jsbin/player_ias-vfle4-e03/en_US/base.js', 'vfle4-e03'), ('https://www.youtube.com/yts/jsbin/player_ias-vfle4-e03/en_US/base.js', 'vfle4-e03'),
('https://www.youtube.com/yts/jsbin/player_ias-vfl49f_g4/en_US/base.js', 'vfl49f_g4'), ('https://www.youtube.com/yts/jsbin/player_ias-vfl49f_g4/en_US/base.js', 'vfl49f_g4'),
@ -280,8 +309,9 @@ class TestPlayerInfo(unittest.TestCase):
('https://s.ytimg.com/yts/jsbin/html5player-en_US-vflXGBaUN.js', 'vflXGBaUN'), ('https://s.ytimg.com/yts/jsbin/html5player-en_US-vflXGBaUN.js', 'vflXGBaUN'),
('https://s.ytimg.com/yts/jsbin/html5player-en_US-vflKjOTVq/html5player.js', 'vflKjOTVq'), ('https://s.ytimg.com/yts/jsbin/html5player-en_US-vflKjOTVq/html5player.js', 'vflKjOTVq'),
) )
ie = YoutubeIE(FakeYDL({'cachedir': False}))
for player_url, expected_player_id in PLAYER_URLS: for player_url, expected_player_id in PLAYER_URLS:
player_id = YoutubeIE._extract_player_info(player_url) player_id = ie._extract_player_info(player_url)
self.assertEqual(player_id, expected_player_id) self.assertEqual(player_id, expected_player_id)
@ -301,8 +331,8 @@ class TestSignature(unittest.TestCase):
def t_factory(name, sig_func, url_pattern): def t_factory(name, sig_func, url_pattern):
def make_tfunc(url, sig_input, expected_sig): def make_tfunc(url, sig_input, expected_sig):
m = url_pattern.match(url) m = url_pattern.match(url)
assert m, '%r should follow URL format' % url assert m, '{0!r} should follow URL format'.format(url)
test_id = m.group('id') test_id = re.sub(r'[/.-]', '_', m.group('id') or m.group('compat_id'))
def test_func(self): def test_func(self):
basename = 'player-{0}-{1}.js'.format(name, test_id) basename = 'player-{0}-{1}.js'.format(name, test_id)
@ -335,12 +365,16 @@ def n_sig(jscode, sig_input):
make_sig_test = t_factory( make_sig_test = t_factory(
'signature', signature, re.compile(r'.*(?:-|/player/)(?P<id>[a-zA-Z0-9_-]+)(?:/.+\.js|(?:/watch_as3|/html5player)?\.[a-z]+)$')) 'signature', signature,
re.compile(r'''(?x)
.+/(?P<h5>html5)?player(?(h5)(?:-en_US)?-|/)(?P<id>[a-zA-Z0-9/._-]+)
(?(h5)/(?:watch_as3|html5player))?\.js$
'''))
for test_spec in _SIG_TESTS: for test_spec in _SIG_TESTS:
make_sig_test(*test_spec) make_sig_test(*test_spec)
make_nsig_test = t_factory( make_nsig_test = t_factory(
'nsig', n_sig, re.compile(r'.+/player/(?P<id>[a-zA-Z0-9_-]+)/.+.js$')) 'nsig', n_sig, re.compile(r'.+/player/(?P<id>[a-zA-Z0-9_/.-]+)\.js$'))
for test_spec in _NSIG_TESTS: for test_spec in _NSIG_TESTS:
make_nsig_test(*test_spec) make_nsig_test(*test_spec)

View File

@ -1912,25 +1912,49 @@ class YoutubeDL(object):
return subs return subs
def __forced_printings(self, info_dict, filename, incomplete): def __forced_printings(self, info_dict, filename, incomplete):
FIELD_ALIASES = {}
def print_mandatory(field): def print_mandatory(field):
actual_field = FIELD_ALIASES.get(field, field)
if (self.params.get('force%s' % field, False) if (self.params.get('force%s' % field, False)
and (not incomplete or info_dict.get(field) is not None)): and (not incomplete or info_dict.get(actual_field) is not None)):
self.to_stdout(info_dict[field]) self.to_stdout(info_dict[actual_field])
def print_optional(field): def print_optional(field):
if (self.params.get('force%s' % field, False) if (self.params.get('force%s' % field, False)
and info_dict.get(field) is not None): and info_dict.get(field) is not None):
self.to_stdout(info_dict[field]) self.to_stdout(info_dict[field])
info_dict = info_dict.copy()
info_dict['duration_string'] = ( # %(duration>%H-%M-%S)s is wrong if duration > 24hrs
formatSeconds(info_dict['duration'])
if info_dict.get('duration', None) is not None
else None)
if info_dict.get('resolution') is None:
info_dict['resolution'] = self.format_resolution(info_dict, default=None)
if filename is not None:
info_dict['filename'] = filename
if info_dict.get('requested_formats') is not None:
# For RTMP URLs, also include the playpath
info_dict['urls'] = '\n'.join(f['url'] + f.get('play_path', '') for f in info_dict['requested_formats'])
elif 'url' in info_dict:
info_dict['urls'] = info_dict['url'] + info_dict.get('play_path', '')
if 'urls' in info_dict:
FIELD_ALIASES['url'] = 'urls'
for tmpl in self.params.get('forceprint', []):
if re.match(r'\w+$', tmpl):
tmpl = '%({0})s'.format(tmpl)
try:
out_txt = tmpl % info_dict
except KeyError:
self.report_warning('Skipping invalid print string "%s"' % (tmpl, ))
continue
self.to_stdout(out_txt)
print_mandatory('title') print_mandatory('title')
print_mandatory('id') print_mandatory('id')
if self.params.get('forceurl', False) and not incomplete: print_mandatory('url')
if info_dict.get('requested_formats') is not None:
for f in info_dict['requested_formats']:
self.to_stdout(f['url'] + f.get('play_path', ''))
else:
# For RTMP URLs, also include the playpath
self.to_stdout(info_dict['url'] + info_dict.get('play_path', ''))
print_optional('thumbnail') print_optional('thumbnail')
print_optional('description') print_optional('description')
if self.params.get('forcefilename', False) and filename is not None: if self.params.get('forcefilename', False) and filename is not None:
@ -1938,6 +1962,7 @@ class YoutubeDL(object):
if self.params.get('forceduration', False) and info_dict.get('duration') is not None: if self.params.get('forceduration', False) and info_dict.get('duration') is not None:
self.to_stdout(formatSeconds(info_dict['duration'])) self.to_stdout(formatSeconds(info_dict['duration']))
print_mandatory('format') print_mandatory('format')
if self.params.get('forcejson', False): if self.params.get('forcejson', False):
self.to_stdout(json.dumps(self.sanitize_info(info_dict))) self.to_stdout(json.dumps(self.sanitize_info(info_dict)))

View File

@ -245,7 +245,7 @@ def _real_main(argv=None):
' file! Use "{0}.%(ext)s" instead of "{0}" as the output' ' file! Use "{0}.%(ext)s" instead of "{0}" as the output'
' template'.format(outtmpl)) ' template'.format(outtmpl))
any_getting = opts.geturl or opts.gettitle or opts.getid or opts.getthumbnail or opts.getdescription or opts.getfilename or opts.getformat or opts.getduration or opts.dumpjson or opts.dump_single_json any_getting = opts.print_ or opts.geturl or opts.gettitle or opts.getid or opts.getthumbnail or opts.getdescription or opts.getfilename or opts.getformat or opts.getduration or opts.dumpjson or opts.dump_single_json
any_printing = opts.print_json any_printing = opts.print_json
download_archive_fn = expand_path(opts.download_archive) if opts.download_archive is not None else opts.download_archive download_archive_fn = expand_path(opts.download_archive) if opts.download_archive is not None else opts.download_archive
@ -335,6 +335,7 @@ def _real_main(argv=None):
'forceduration': opts.getduration, 'forceduration': opts.getduration,
'forcefilename': opts.getfilename, 'forcefilename': opts.getfilename,
'forceformat': opts.getformat, 'forceformat': opts.getformat,
'forceprint': opts.print_,
'forcejson': opts.dumpjson or opts.print_json, 'forcejson': opts.dumpjson or opts.print_json,
'dump_single_json': opts.dump_single_json, 'dump_single_json': opts.dump_single_json,
'simulate': opts.simulate or any_getting, 'simulate': opts.simulate or any_getting,

View File

@ -692,9 +692,9 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
'invidious': '|'.join(_INVIDIOUS_SITES), 'invidious': '|'.join(_INVIDIOUS_SITES),
} }
_PLAYER_INFO_RE = ( _PLAYER_INFO_RE = (
r'/s/player/(?P<id>[a-zA-Z0-9_-]{8,})//(?:tv-)?player', r'/s/player/(?P<id>[a-zA-Z0-9_-]{8,})/(?:tv-)?player',
r'/(?P<id>[a-zA-Z0-9_-]{8,})/player(?:_ias\.vflset(?:/[a-zA-Z]{2,3}_[a-zA-Z]{2,3})?|-plasma-ias-(?:phone|tablet)-[a-z]{2}_[A-Z]{2}\.vflset)/base\.js$', r'/(?P<id>[a-zA-Z0-9_-]{8,})/player(?:_ias(?:_tce)?\.vflset(?:/[a-zA-Z]{2,3}_[a-zA-Z]{2,3})?|-plasma-ias-(?:phone|tablet)-[a-z]{2}_[A-Z]{2}\.vflset)/base\.js$',
r'\b(?P<id>vfl[a-zA-Z0-9_-]+)\b.*?\.js$', r'\b(?P<id>vfl[a-zA-Z0-9_-]{6,})\b.*?\.js$',
) )
_SUBTITLE_FORMATS = ('json3', 'srv1', 'srv2', 'srv3', 'ttml', 'vtt') _SUBTITLE_FORMATS = ('json3', 'srv1', 'srv2', 'srv3', 'ttml', 'vtt')
@ -1626,15 +1626,13 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
""" Return a string representation of a signature """ """ Return a string representation of a signature """
return '.'.join(compat_str(len(part)) for part in example_sig.split('.')) return '.'.join(compat_str(len(part)) for part in example_sig.split('.'))
@classmethod def _extract_player_info(self, player_url):
def _extract_player_info(cls, player_url): try:
for player_re in cls._PLAYER_INFO_RE: return self._search_regex(
id_m = re.search(player_re, player_url) self._PLAYER_INFO_RE, player_url, 'player info', group='id')
if id_m: except ExtractorError as e:
break raise ExtractorError(
else: 'Cannot identify player %r' % (player_url,), cause=e)
raise ExtractorError('Cannot identify player %r' % player_url)
return id_m.group('id')
def _load_player(self, video_id, player_url, fatal=True, player_id=None): def _load_player(self, video_id, player_url, fatal=True, player_id=None):
if not player_id: if not player_id:
@ -1711,6 +1709,23 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
' return %s\n') % (signature_id_tuple, expr_code) ' return %s\n') % (signature_id_tuple, expr_code)
self.to_screen('Extracted signature function:\n' + code) self.to_screen('Extracted signature function:\n' + code)
def _extract_sig_fn(self, jsi, funcname):
var_ay = self._search_regex(
r'''(?x)
(?:\*/|\{|\n|^)\s*(?:'[^']+'\s*;\s*)
(var\s*[\w$]+\s*=\s*(?:
('|")(?:\\\2|(?!\2).)+\2\s*\.\s*split\(\s*('|")\W+\3\s*\)|
\[\s*(?:('|")(?:\\\4|(?!\4).)*\4\s*(?:(?=\])|,\s*))+\]
))(?=\s*[,;])
''', jsi.code, 'useful values', default='')
sig_fn = jsi.extract_function_code(funcname)
if var_ay:
sig_fn = (sig_fn[0], ';\n'.join((var_ay, sig_fn[1])))
return sig_fn
def _parse_sig_js(self, jscode): def _parse_sig_js(self, jscode):
# Examples where `sig` is funcname: # Examples where `sig` is funcname:
# sig=function(a){a=a.split(""); ... ;return a.join("")}; # sig=function(a){a=a.split(""); ... ;return a.join("")};
@ -1736,8 +1751,12 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
jscode, 'Initial JS player signature function name', group='sig') jscode, 'Initial JS player signature function name', group='sig')
jsi = JSInterpreter(jscode) jsi = JSInterpreter(jscode)
initial_function = jsi.extract_function(funcname)
return lambda s: initial_function([s]) initial_function = self._extract_sig_fn(jsi, funcname)
func = jsi.extract_function_from_code(*initial_function)
return lambda s: func([s])
def _cached(self, func, *cache_id): def _cached(self, func, *cache_id):
def inner(*args, **kwargs): def inner(*args, **kwargs):
@ -1856,15 +1875,9 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
def _extract_n_function_code_jsi(self, video_id, jsi, player_id=None): def _extract_n_function_code_jsi(self, video_id, jsi, player_id=None):
var_ay = self._search_regex(
r'(?:[;\s]|^)\s*(var\s*[\w$]+\s*=\s*"(?:\\"|[^"])+"\s*\.\s*split\("\W+"\))(?=\s*[,;])',
jsi.code, 'useful values', default='')
func_name = self._extract_n_function_name(jsi.code) func_name = self._extract_n_function_name(jsi.code)
func_code = jsi.extract_function_code(func_name) func_code = self._extract_sig_fn(jsi, func_name)
if var_ay:
func_code = (func_code[0], ';\n'.join((var_ay, func_code[1])))
if player_id: if player_id:
self.cache.store('youtube-nsig', player_id, func_code) self.cache.store('youtube-nsig', player_id, func_code)
@ -2136,7 +2149,8 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
video_details = merge_dicts(*traverse_obj( video_details = merge_dicts(*traverse_obj(
(player_response, api_player_response), (player_response, api_player_response),
(Ellipsis, 'videoDetails', T(dict)))) (Ellipsis, 'videoDetails', T(dict))))
player_response.update(api_player_response or {}) player_response.update(filter_dict(
api_player_response or {}, cndn=lambda k, _: k != 'captions'))
player_response['videoDetails'] = video_details player_response['videoDetails'] = video_details
def is_agegated(playability): def is_agegated(playability):
@ -2566,8 +2580,8 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
} }
pctr = traverse_obj( pctr = traverse_obj(
player_response, (player_response, api_player_response),
('captions', 'playerCaptionsTracklistRenderer', T(dict))) (Ellipsis, 'captions', 'playerCaptionsTracklistRenderer', T(dict)))
if pctr: if pctr:
def process_language(container, base_url, lang_code, query): def process_language(container, base_url, lang_code, query):
lang_subs = [] lang_subs = []
@ -2584,20 +2598,21 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
def process_subtitles(): def process_subtitles():
subtitles = {} subtitles = {}
for caption_track in traverse_obj(pctr, ( for caption_track in traverse_obj(pctr, (
'captionTracks', lambda _, v: v.get('baseUrl'))): Ellipsis, 'captionTracks', lambda _, v: (
v.get('baseUrl') and v.get('languageCode')))):
base_url = self._yt_urljoin(caption_track['baseUrl']) base_url = self._yt_urljoin(caption_track['baseUrl'])
if not base_url: if not base_url:
continue continue
lang_code = caption_track['languageCode']
if caption_track.get('kind') != 'asr': if caption_track.get('kind') != 'asr':
lang_code = caption_track.get('languageCode')
if not lang_code:
continue
process_language( process_language(
subtitles, base_url, lang_code, {}) subtitles, base_url, lang_code, {})
continue continue
automatic_captions = {} automatic_captions = {}
process_language(
automatic_captions, base_url, lang_code, {})
for translation_language in traverse_obj(pctr, ( for translation_language in traverse_obj(pctr, (
'translationLanguages', lambda _, v: v.get('languageCode'))): Ellipsis, 'translationLanguages', lambda _, v: v.get('languageCode'))):
translation_language_code = translation_language['languageCode'] translation_language_code = translation_language['languageCode']
process_language( process_language(
automatic_captions, base_url, translation_language_code, automatic_captions, base_url, translation_language_code,

View File

@ -678,7 +678,7 @@ class JSInterpreter(object):
return len(obj) return len(obj)
try: try:
return obj[int(idx)] if isinstance(obj, list) else obj[compat_str(idx)] return obj[int(idx)] if isinstance(obj, list) else obj[compat_str(idx)]
except (TypeError, KeyError, IndexError) as e: except (TypeError, KeyError, IndexError, ValueError) as e:
# allow_undefined is None gives correct behaviour # allow_undefined is None gives correct behaviour
if allow_undefined or ( if allow_undefined or (
allow_undefined is None and not isinstance(e, TypeError)): allow_undefined is None and not isinstance(e, TypeError)):
@ -1038,6 +1038,10 @@ class JSInterpreter(object):
left_val = self._index(left_val, idx) left_val = self._index(left_val, idx)
if isinstance(idx, float): if isinstance(idx, float):
idx = int(idx) idx = int(idx)
if isinstance(left_val, list) and len(left_val) <= int_or_none(idx, default=-1):
# JS Array is a sparsely assignable list
# TODO: handle extreme sparsity without memory bloat, eg using auxiliary dict
left_val.extend((idx - len(left_val) + 1) * [JS_Undefined])
left_val[idx] = self._operator( left_val[idx] = self._operator(
m.group('op'), self._index(left_val, idx) if m.group('op') else None, m.group('op'), self._index(left_val, idx) if m.group('op') else None,
m.group('expr'), expr, local_vars, allow_recursion) m.group('expr'), expr, local_vars, allow_recursion)
@ -1204,7 +1208,8 @@ class JSInterpreter(object):
elif member == 'join': elif member == 'join':
assertion(isinstance(obj, list), 'must be applied on a list') assertion(isinstance(obj, list), 'must be applied on a list')
assertion(len(argvals) <= 1, 'takes at most one argument') assertion(len(argvals) <= 1, 'takes at most one argument')
return (',' if len(argvals) == 0 else argvals[0]).join( return (',' if len(argvals) == 0 or argvals[0] in (None, JS_Undefined)
else argvals[0]).join(
('' if x in (None, JS_Undefined) else _js_toString(x)) ('' if x in (None, JS_Undefined) else _js_toString(x))
for x in obj) for x in obj)
elif member == 'reverse': elif member == 'reverse':
@ -1364,19 +1369,21 @@ class JSInterpreter(object):
code, _ = self._separate_at_paren(func_m.group('code')) # refine the match code, _ = self._separate_at_paren(func_m.group('code')) # refine the match
return self.build_arglist(func_m.group('args')), code return self.build_arglist(func_m.group('args')), code
def extract_function(self, funcname): def extract_function(self, funcname, *global_stack):
return function_with_repr( return function_with_repr(
self.extract_function_from_code(*self.extract_function_code(funcname)), self.extract_function_from_code(*itertools.chain(
self.extract_function_code(funcname), global_stack)),
'F<%s>' % (funcname,)) 'F<%s>' % (funcname,))
def extract_function_from_code(self, argnames, code, *global_stack): def extract_function_from_code(self, argnames, code, *global_stack):
local_vars = {} local_vars = {}
start = None
while True: while True:
mobj = re.search(r'function\((?P<args>[^)]*)\)\s*{', code) mobj = re.search(r'function\((?P<args>[^)]*)\)\s*{', code[start:])
if mobj is None: if mobj is None:
break break
start, body_start = mobj.span() start, body_start = ((start or 0) + x for x in mobj.span())
body, remaining = self._separate_at_paren(code[body_start - 1:]) body, remaining = self._separate_at_paren(code[body_start - 1:])
name = self._named_object(local_vars, self.extract_function_from_code( name = self._named_object(local_vars, self.extract_function_from_code(
[x.strip() for x in mobj.group('args').split(',')], [x.strip() for x in mobj.group('args').split(',')],

View File

@ -13,6 +13,7 @@ from .compat import (
compat_kwargs, compat_kwargs,
compat_open as open, compat_open as open,
compat_shlex_split, compat_shlex_split,
compat_str,
) )
from .utils import ( from .utils import (
preferredencoding, preferredencoding,
@ -109,6 +110,14 @@ def parseOpts(overrideArguments=None):
def _comma_separated_values_options_callback(option, opt_str, value, parser): def _comma_separated_values_options_callback(option, opt_str, value, parser):
setattr(parser.values, option.dest, value.split(',')) setattr(parser.values, option.dest, value.split(','))
def _list_from_options_callback(option, opt_str, value, parser, append=True, delim=',', process=compat_str.strip):
# append can be True, False or -1 (prepend)
current = list(getattr(parser.values, option.dest)) if append else []
value = list(filter(None, [process(value)] if delim is None else map(process, value.split(delim))))
setattr(
parser.values, option.dest,
current + value if append is True else value + current)
# No need to wrap help messages if we're on a wide console # No need to wrap help messages if we're on a wide console
columns = compat_get_terminal_size().columns columns = compat_get_terminal_size().columns
max_width = columns if columns else 80 max_width = columns if columns else 80
@ -594,6 +603,13 @@ def parseOpts(overrideArguments=None):
'--skip-download', '--skip-download',
action='store_true', dest='skip_download', default=False, action='store_true', dest='skip_download', default=False,
help='Do not download the video') help='Do not download the video')
verbosity.add_option(
'-O', '--print', metavar='TEMPLATE',
action='callback', dest='print_', type='str', default=[],
callback=_list_from_options_callback, callback_kwargs={'delim': None},
help=(
'Simulate, quiet but print the given fields. Either a field name '
'or similar formatting as the output template can be used'))
verbosity.add_option( verbosity.add_option(
'-g', '--get-url', '-g', '--get-url',
action='store_true', dest='geturl', default=False, action='store_true', dest='geturl', default=False,