Compare commits

..

1 Commits

11 changed files with 258 additions and 553 deletions

View File

@ -63,7 +63,7 @@ class TestCache(unittest.TestCase):
obj = {'x': 1, 'y': ['ä', '\\a', True]} obj = {'x': 1, 'y': ['ä', '\\a', True]}
c.store('test_cache', 'k.', obj) c.store('test_cache', 'k.', obj)
self.assertEqual(c.load('test_cache', 'k.', min_ver='1970.01.01'), obj) self.assertEqual(c.load('test_cache', 'k.', min_ver='1970.01.01'), obj)
new_version = '.'.join(('%0.2d' % ((v + 1) if i == 0 else v, )) for i, v in enumerate(version_tuple(__version__))) new_version = '.'.join(('%d' % ((v + 1) if i == 0 else v, )) for i, v in enumerate(version_tuple(__version__)))
self.assertIs(c.load('test_cache', 'k.', min_ver=new_version), None) self.assertIs(c.load('test_cache', 'k.', min_ver=new_version), None)

View File

@ -7,7 +7,6 @@ from __future__ import unicode_literals
import os import os
import sys import sys
import unittest import unittest
sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__)))) sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
import math import math
@ -147,25 +146,6 @@ class TestJSInterpreter(unittest.TestCase):
# https://github.com/ytdl-org/youtube-dl/issues/32815 # https://github.com/ytdl-org/youtube-dl/issues/32815
self._test('function f(){return 0 - 7 * - 6;}', 42) self._test('function f(){return 0 - 7 * - 6;}', 42)
def test_bitwise_operators_typecast(self):
# madness
self._test('function f(){return null << 5}', 0)
self._test('function f(){return undefined >> 5}', 0)
self._test('function f(){return 42 << NaN}', 42)
self._test('function f(){return 42 << Infinity}', 42)
self._test('function f(){return 0.0 << null}', 0)
self._test('function f(){return NaN << 42}', 0)
self._test('function f(){return "21.9" << 1}', 42)
self._test('function f(){return true << "5";}', 32)
self._test('function f(){return true << true;}', 2)
self._test('function f(){return "19" & "21.9";}', 17)
self._test('function f(){return "19" & false;}', 0)
self._test('function f(){return "11.0" >> "2.1";}', 2)
self._test('function f(){return 5 ^ 9;}', 12)
self._test('function f(){return 0.0 << NaN}', 0)
self._test('function f(){return null << undefined}', 0)
self._test('function f(){return 21 << 4294967297}', 42)
def test_array_access(self): def test_array_access(self):
self._test('function f(){var x = [1,2,3]; x[0] = 4; x[0] = 5; x[2.0] = 7; return x;}', [5, 2, 7]) self._test('function f(){var x = [1,2,3]; x[0] = 4; x[0] = 5; x[2.0] = 7; return x;}', [5, 2, 7])
@ -180,7 +160,6 @@ class TestJSInterpreter(unittest.TestCase):
self._test('function f(){var x = 20; x = 30 + 1; return x;}', 31) self._test('function f(){var x = 20; x = 30 + 1; return x;}', 31)
self._test('function f(){var x = 20; x += 30 + 1; return x;}', 51) self._test('function f(){var x = 20; x += 30 + 1; return x;}', 51)
self._test('function f(){var x = 20; x -= 30 + 1; return x;}', -11) self._test('function f(){var x = 20; x -= 30 + 1; return x;}', -11)
self._test('function f(){var x = 2; var y = ["a", "b"]; y[x%y["length"]]="z"; return y}', ['z', 'b'])
def test_comments(self): def test_comments(self):
self._test(''' self._test('''
@ -372,13 +351,6 @@ class TestJSInterpreter(unittest.TestCase):
self._test('function f() { a=5; return (a -= 1, a+=3, a); }', 7) self._test('function f() { a=5; return (a -= 1, a+=3, a); }', 7)
self._test('function f() { return (l=[0,1,2,3], function(a, b){return a+b})((l[1], l[2]), l[3]) }', 5) self._test('function f() { return (l=[0,1,2,3], function(a, b){return a+b})((l[1], l[2]), l[3]) }', 5)
def test_not(self):
self._test('function f() { return ! undefined; }', True)
self._test('function f() { return !0; }', True)
self._test('function f() { return !!0; }', False)
self._test('function f() { return ![]; }', False)
self._test('function f() { return !0 !== false; }', True)
def test_void(self): def test_void(self):
self._test('function f() { return void 42; }', JS_Undefined) self._test('function f() { return void 42; }', JS_Undefined)
@ -463,7 +435,6 @@ class TestJSInterpreter(unittest.TestCase):
def test_regex(self): def test_regex(self):
self._test('function f() { let a=/,,[/,913,/](,)}/; }', None) self._test('function f() { let a=/,,[/,913,/](,)}/; }', None)
self._test('function f() { let a=/,,[/,913,/](,)}/; return a.source; }', ',,[/,913,/](,)}')
jsi = JSInterpreter(''' jsi = JSInterpreter('''
function x() { let a=/,,[/,913,/](,)}/; "".replace(a, ""); return a; } function x() { let a=/,,[/,913,/](,)}/; "".replace(a, ""); return a; }
@ -511,6 +482,25 @@ class TestJSInterpreter(unittest.TestCase):
self._test('function f(){return -524999584 << 5}', 379882496) self._test('function f(){return -524999584 << 5}', 379882496)
self._test('function f(){return 1236566549 << 5}', 915423904) self._test('function f(){return 1236566549 << 5}', 915423904)
def test_bitwise_operators_typecast(self):
# madness
self._test('function f(){return null << 5}', 0)
self._test('function f(){return undefined >> 5}', 0)
self._test('function f(){return 42 << NaN}', 42)
self._test('function f(){return 42 << Infinity}', 42)
self._test('function f(){return 0.0 << null}', 0)
self._test('function f(){return NaN << 42}', 0)
self._test('function f(){return "21.9" << 1}', 42)
self._test('function f(){return 21 << 4294967297}', 42)
self._test('function f(){return true << "5";}', 32)
self._test('function f(){return true << true;}', 2)
self._test('function f(){return "19" & "21.9";}', 17)
self._test('function f(){return "19" & false;}', 0)
self._test('function f(){return "11.0" >> "2.1";}', 2)
self._test('function f(){return 5 ^ 9;}', 12)
self._test('function f(){return 0.0 << NaN}', 0)
self._test('function f(){return null << undefined}', 0)
def test_negative(self): def test_negative(self):
self._test('function f(){return 2 * -2.0 ;}', -4) self._test('function f(){return 2 * -2.0 ;}', -4)
self._test('function f(){return 2 - - -2 ;}', 0) self._test('function f(){return 2 - - -2 ;}', 0)
@ -553,8 +543,6 @@ class TestJSInterpreter(unittest.TestCase):
test_result = list('test') test_result = list('test')
tests = [ tests = [
'function f(a, b){return a.split(b)}', 'function f(a, b){return a.split(b)}',
'function f(a, b){return a["split"](b)}',
'function f(a, b){let x = ["split"]; return a[x[0]](b)}',
'function f(a, b){return String.prototype.split.call(a, b)}', 'function f(a, b){return String.prototype.split.call(a, b)}',
'function f(a, b){return String.prototype.split.apply(a, [b])}', 'function f(a, b){return String.prototype.split.apply(a, [b])}',
] ]
@ -605,9 +593,6 @@ class TestJSInterpreter(unittest.TestCase):
self._test('function f(){return "012345678".slice(-1, 1)}', '') self._test('function f(){return "012345678".slice(-1, 1)}', '')
self._test('function f(){return "012345678".slice(-3, -1)}', '67') self._test('function f(){return "012345678".slice(-3, -1)}', '67')
def test_splice(self):
self._test('function f(){var T = ["0", "1", "2"]; T["splice"](2, 1, "0")[0]; return T }', ['0', '1', '0'])
def test_pop(self): def test_pop(self):
# pop # pop
self._test('function f(){var a = [0, 1, 2, 3, 4, 5, 6, 7, 8]; return [a.pop(), a]}', self._test('function f(){var a = [0, 1, 2, 3, 4, 5, 6, 7, 8]; return [a.pop(), a]}',
@ -642,16 +627,6 @@ class TestJSInterpreter(unittest.TestCase):
'return [ret.length, ret[0][0], ret[1][1], ret[0][2]]}', 'return [ret.length, ret[0][0], ret[1][1], ret[0][2]]}',
[2, 4, 1, [4, 2]]) [2, 4, 1, [4, 2]])
def test_extract_function(self):
jsi = JSInterpreter('function a(b) { return b + 1; }')
func = jsi.extract_function('a')
self.assertEqual(func([2]), 3)
def test_extract_function_with_global_stack(self):
jsi = JSInterpreter('function c(d) { return d + e + f + g; }')
func = jsi.extract_function('c', {'e': 10}, {'f': 100, 'g': 1000})
self.assertEqual(func([1]), 1111)
if __name__ == '__main__': if __name__ == '__main__':
unittest.main() unittest.main()

View File

@ -94,51 +94,11 @@ _SIG_TESTS = [
'2aq0aqSyOoJXtK73m-uME_jv7-pT15gOFC02RFkGMqWpzEICs69VdbwQ0LDp1v7j8xx92efCJlYFYb1sUkkBSPOlPmXgIARw8JQ0qOAOAA', '2aq0aqSyOoJXtK73m-uME_jv7-pT15gOFC02RFkGMqWpzEICs69VdbwQ0LDp1v7j8xx92efCJlYFYb1sUkkBSPOlPmXgIARw8JQ0qOAOAA',
'0aqSyOoJXtK73m-uME_jv7-pT15gOFC02RFkGMqWpz2ICs6EVdbwQ0LDp1v7j8xx92efCJlYFYb1sUkkBSPOlPmXgIARw8JQ0qOAOAA', '0aqSyOoJXtK73m-uME_jv7-pT15gOFC02RFkGMqWpz2ICs6EVdbwQ0LDp1v7j8xx92efCJlYFYb1sUkkBSPOlPmXgIARw8JQ0qOAOAA',
), ),
(
'https://www.youtube.com/s/player/363db69b/player_ias_tce.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', 'https://www.youtube.com/s/player/4fcd6e4a/player_ias.vflset/en_US/base.js',
'2aq0aqSyOoJXtK73m-uME_jv7-pT15gOFC02RFkGMqWpzEICs69VdbwQ0LDp1v7j8xx92efCJlYFYb1sUkkBSPOlPmXgIARw8JQ0qOAOAA', '2aq0aqSyOoJXtK73m-uME_jv7-pT15gOFC02RFkGMqWpzEICs69VdbwQ0LDp1v7j8xx92efCJlYFYb1sUkkBSPOlPmXgIARw8JQ0qOAOAA',
'wAOAOq0QJ8ARAIgXmPlOPSBkkUs1bYFYlJCfe29xx8q7v1pDL0QwbdV96sCIEzpWqMGkFR20CFOg51Tp-7vj_EMu-m37KtXJoOySqa0', 'wAOAOq0QJ8ARAIgXmPlOPSBkkUs1bYFYlJCfe29xx8q7v1pDL0QwbdV96sCIEzpWqMGkFR20CFOg51Tp-7vj_EMu-m37KtXJoOySqa0',
), ),
(
'https://www.youtube.com/s/player/4fcd6e4a/player_ias_tce.vflset/en_US/base.js',
'2aq0aqSyOoJXtK73m-uME_jv7-pT15gOFC02RFkGMqWpzEICs69VdbwQ0LDp1v7j8xx92efCJlYFYb1sUkkBSPOlPmXgIARw8JQ0qOAOAA',
'wAOAOq0QJ8ARAIgXmPlOPSBkkUs1bYFYlJCfe29xx8q7v1pDL0QwbdV96sCIEzpWqMGkFR20CFOg51Tp-7vj_EMu-m37KtXJoOySqa0',
),
(
'https://www.youtube.com/s/player/20830619/player_ias.vflset/en_US/base.js',
'2aq0aqSyOoJXtK73m-uME_jv7-pT15gOFC02RFkGMqWpzEICs69VdbwQ0LDp1v7j8xx92efCJlYFYb1sUkkBSPOlPmXgIARw8JQ0qOAOAA',
'7AOq0QJ8wRAIgXmPlOPSBkkAs1bYFYlJCfe29xx8jOv1pDL0Q2bdV96sCIEzpWqMGkFR20CFOg51Tp-7vj_EMu-m37KtXJoOySqa0qaw',
),
(
'https://www.youtube.com/s/player/20830619/player_ias_tce.vflset/en_US/base.js',
'2aq0aqSyOoJXtK73m-uME_jv7-pT15gOFC02RFkGMqWpzEICs69VdbwQ0LDp1v7j8xx92efCJlYFYb1sUkkBSPOlPmXgIARw8JQ0qOAOAA',
'7AOq0QJ8wRAIgXmPlOPSBkkAs1bYFYlJCfe29xx8jOv1pDL0Q2bdV96sCIEzpWqMGkFR20CFOg51Tp-7vj_EMu-m37KtXJoOySqa0qaw',
),
(
'https://www.youtube.com/s/player/20830619/player-plasma-ias-phone-en_US.vflset/base.js',
'2aq0aqSyOoJXtK73m-uME_jv7-pT15gOFC02RFkGMqWpzEICs69VdbwQ0LDp1v7j8xx92efCJlYFYb1sUkkBSPOlPmXgIARw8JQ0qOAOAA',
'7AOq0QJ8wRAIgXmPlOPSBkkAs1bYFYlJCfe29xx8jOv1pDL0Q2bdV96sCIEzpWqMGkFR20CFOg51Tp-7vj_EMu-m37KtXJoOySqa0qaw',
),
(
'https://www.youtube.com/s/player/20830619/player-plasma-ias-tablet-en_US.vflset/base.js',
'2aq0aqSyOoJXtK73m-uME_jv7-pT15gOFC02RFkGMqWpzEICs69VdbwQ0LDp1v7j8xx92efCJlYFYb1sUkkBSPOlPmXgIARw8JQ0qOAOAA',
'7AOq0QJ8wRAIgXmPlOPSBkkAs1bYFYlJCfe29xx8jOv1pDL0Q2bdV96sCIEzpWqMGkFR20CFOg51Tp-7vj_EMu-m37KtXJoOySqa0qaw',
),
(
'https://www.youtube.com/s/player/8a8ac953/player_ias_tce.vflset/en_US/base.js',
'2aq0aqSyOoJXtK73m-uME_jv7-pT15gOFC02RFkGMqWpzEICs69VdbwQ0LDp1v7j8xx92efCJlYFYb1sUkkBSPOlPmXgIARw8JQ0qOAOAA',
'IAOAOq0QJ8wRAAgXmPlOPSBkkUs1bYFYlJCfe29xx8j7v1pDL0QwbdV96sCIEzpWqMGkFR20CFOg51Tp-7vj_E2u-m37KtXJoOySqa0',
),
(
'https://www.youtube.com/s/player/8a8ac953/tv-player-es6.vflset/tv-player-es6.js',
'2aq0aqSyOoJXtK73m-uME_jv7-pT15gOFC02RFkGMqWpzEICs69VdbwQ0LDp1v7j8xx92efCJlYFYb1sUkkBSPOlPmXgIARw8JQ0qOAOAA',
'IAOAOq0QJ8wRAAgXmPlOPSBkkUs1bYFYlJCfe29xx8j7v1pDL0QwbdV96sCIEzpWqMGkFR20CFOg51Tp-7vj_E2u-m37KtXJoOySqa0',
),
] ]
_NSIG_TESTS = [ _NSIG_TESTS = [
@ -312,7 +272,7 @@ _NSIG_TESTS = [
), ),
( (
'https://www.youtube.com/s/player/643afba4/player_ias.vflset/en_US/base.js', 'https://www.youtube.com/s/player/643afba4/player_ias.vflset/en_US/base.js',
'ir9-V6cdbCiyKxhr', '2PL7ZDYAALMfmA', 'W9HJZKktxuYoDTqW', 'larxUlagTRAcSw',
), ),
( (
'https://www.youtube.com/s/player/363db69b/player_ias.vflset/en_US/base.js', 'https://www.youtube.com/s/player/363db69b/player_ias.vflset/en_US/base.js',
@ -326,26 +286,6 @@ _NSIG_TESTS = [
'https://www.youtube.com/s/player/4fcd6e4a/tv-player-ias.vflset/tv-player-ias.js', 'https://www.youtube.com/s/player/4fcd6e4a/tv-player-ias.vflset/tv-player-ias.js',
'o_L251jm8yhZkWtBW', 'lXoxI3XvToqn6A', 'o_L251jm8yhZkWtBW', 'lXoxI3XvToqn6A',
), ),
(
'https://www.youtube.com/s/player/20830619/tv-player-ias.vflset/tv-player-ias.js',
'ir9-V6cdbCiyKxhr', '9YE85kNjZiS4',
),
(
'https://www.youtube.com/s/player/20830619/player-plasma-ias-phone-en_US.vflset/base.js',
'ir9-V6cdbCiyKxhr', '9YE85kNjZiS4',
),
(
'https://www.youtube.com/s/player/20830619/player-plasma-ias-tablet-en_US.vflset/base.js',
'ir9-V6cdbCiyKxhr', '9YE85kNjZiS4',
),
(
'https://www.youtube.com/s/player/8a8ac953/player_ias_tce.vflset/en_US/base.js',
'MiBYeXx_vRREbiCCmh', 'RtZYMVvmkE0JE',
),
(
'https://www.youtube.com/s/player/8a8ac953/tv-player-es6.vflset/tv-player-es6.js',
'MiBYeXx_vRREbiCCmh', 'RtZYMVvmkE0JE',
),
] ]
@ -395,7 +335,7 @@ def t_factory(name, sig_func, url_pattern):
test_id = re.sub(r'[/.-]', '_', m.group('id') or m.group('compat_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}.js'.format(test_id) basename = 'player-{0}-{1}.js'.format(name, test_id)
fn = os.path.join(self.TESTDATA_DIR, basename) fn = os.path.join(self.TESTDATA_DIR, basename)
if not os.path.exists(fn): if not os.path.exists(fn):

View File

@ -540,14 +540,10 @@ class YoutubeDL(object):
"""Print message to stdout if not in quiet mode.""" """Print message to stdout if not in quiet mode."""
return self.to_stdout(message, skip_eol, check_quiet=True) return self.to_stdout(message, skip_eol, check_quiet=True)
def _write_string(self, s, out=None, only_once=False, _cache=set()): def _write_string(self, s, out=None):
if only_once and s in _cache:
return
write_string(s, out=out, encoding=self.params.get('encoding')) write_string(s, out=out, encoding=self.params.get('encoding'))
if only_once:
_cache.add(s)
def to_stdout(self, message, skip_eol=False, check_quiet=False, only_once=False): def to_stdout(self, message, skip_eol=False, check_quiet=False):
"""Print message to stdout if not in quiet mode.""" """Print message to stdout if not in quiet mode."""
if self.params.get('logger'): if self.params.get('logger'):
self.params['logger'].debug(message) self.params['logger'].debug(message)
@ -556,9 +552,9 @@ class YoutubeDL(object):
terminator = ['\n', ''][skip_eol] terminator = ['\n', ''][skip_eol]
output = message + terminator output = message + terminator
self._write_string(output, self._screen_file, only_once=only_once) self._write_string(output, self._screen_file)
def to_stderr(self, message, only_once=False): def to_stderr(self, message):
"""Print message to stderr.""" """Print message to stderr."""
assert isinstance(message, compat_str) assert isinstance(message, compat_str)
if self.params.get('logger'): if self.params.get('logger'):
@ -566,7 +562,7 @@ class YoutubeDL(object):
else: else:
message = self._bidi_workaround(message) message = self._bidi_workaround(message)
output = message + '\n' output = message + '\n'
self._write_string(output, self._err_file, only_once=only_once) self._write_string(output, self._err_file)
def to_console_title(self, message): def to_console_title(self, message):
if not self.params.get('consoletitle', False): if not self.params.get('consoletitle', False):
@ -645,11 +641,18 @@ class YoutubeDL(object):
raise DownloadError(message, exc_info) raise DownloadError(message, exc_info)
self._download_retcode = 1 self._download_retcode = 1
def report_warning(self, message, only_once=False): def report_warning(self, message, only_once=False, _cache={}):
''' '''
Print the message to stderr, it will be prefixed with 'WARNING:' Print the message to stderr, it will be prefixed with 'WARNING:'
If stderr is a tty file the 'WARNING:' will be colored If stderr is a tty file the 'WARNING:' will be colored
''' '''
if only_once:
m_hash = hash((self, message))
m_cnt = _cache.setdefault(m_hash, 0)
_cache[m_hash] = m_cnt + 1
if m_cnt > 0:
return
if self.params.get('logger') is not None: if self.params.get('logger') is not None:
self.params['logger'].warning(message) self.params['logger'].warning(message)
else: else:
@ -660,7 +663,7 @@ class YoutubeDL(object):
else: else:
_msg_header = 'WARNING:' _msg_header = 'WARNING:'
warning_message = '%s %s' % (_msg_header, message) warning_message = '%s %s' % (_msg_header, message)
self.to_stderr(warning_message, only_once=only_once) self.to_stderr(warning_message)
def report_error(self, message, *args, **kwargs): def report_error(self, message, *args, **kwargs):
''' '''
@ -674,16 +677,6 @@ class YoutubeDL(object):
kwargs['message'] = '%s %s' % (_msg_header, message) kwargs['message'] = '%s %s' % (_msg_header, message)
self.trouble(*args, **kwargs) self.trouble(*args, **kwargs)
def write_debug(self, message, only_once=False):
'''Log debug message or Print message to stderr'''
if not self.params.get('verbose', False):
return
message = '[debug] {0}'.format(message)
if self.params.get('logger'):
self.params['logger'].debug(message)
else:
self.to_stderr(message, only_once)
def report_unscoped_cookies(self, *args, **kwargs): def report_unscoped_cookies(self, *args, **kwargs):
# message=None, tb=False, is_error=False # message=None, tb=False, is_error=False
if len(args) <= 2: if len(args) <= 2:
@ -2521,7 +2514,7 @@ class YoutubeDL(object):
self.get_encoding())) self.get_encoding()))
write_string(encoding_str, encoding=None) write_string(encoding_str, encoding=None)
writeln_debug = lambda *s: self.write_debug(''.join(s)) writeln_debug = lambda *s: self._write_string('[debug] %s\n' % (''.join(s), ))
writeln_debug('youtube-dl version ', __version__) writeln_debug('youtube-dl version ', __version__)
if _LAZY_LOADER: if _LAZY_LOADER:
writeln_debug('Lazy loading extractors enabled') writeln_debug('Lazy loading extractors enabled')

View File

@ -1,6 +1,6 @@
# coding: utf-8
from __future__ import unicode_literals from __future__ import unicode_literals
import errno
import json import json
import os import os
import re import re
@ -8,17 +8,14 @@ import shutil
import traceback import traceback
from .compat import ( from .compat import (
compat_contextlib_suppress,
compat_getenv, compat_getenv,
compat_open as open, compat_open as open,
compat_os_makedirs,
) )
from .utils import ( from .utils import (
error_to_compat_str, error_to_compat_str,
escape_rfc3986,
expand_path, expand_path,
is_outdated_version, is_outdated_version,
traverse_obj, try_get,
write_json_file, write_json_file,
) )
from .version import __version__ from .version import __version__
@ -33,35 +30,23 @@ class Cache(object):
def __init__(self, ydl): def __init__(self, ydl):
self._ydl = ydl self._ydl = ydl
def _write_debug(self, *args, **kwargs):
self._ydl.write_debug(*args, **kwargs)
def _report_warning(self, *args, **kwargs):
self._ydl.report_warning(*args, **kwargs)
def _to_screen(self, *args, **kwargs):
self._ydl.to_screen(*args, **kwargs)
def _get_param(self, k, default=None):
return self._ydl.params.get(k, default)
def _get_root_dir(self): def _get_root_dir(self):
res = self._get_param('cachedir') res = self._ydl.params.get('cachedir')
if res is None: if res is None:
cache_root = compat_getenv('XDG_CACHE_HOME', '~/.cache') cache_root = compat_getenv('XDG_CACHE_HOME', '~/.cache')
res = os.path.join(cache_root, self._YTDL_DIR) res = os.path.join(cache_root, self._YTDL_DIR)
return expand_path(res) return expand_path(res)
def _get_cache_fn(self, section, key, dtype): def _get_cache_fn(self, section, key, dtype):
assert re.match(r'^[\w.-]+$', section), \ assert re.match(r'^[a-zA-Z0-9_.-]+$', section), \
'invalid section %r' % section 'invalid section %r' % section
key = escape_rfc3986(key, safe='').replace('%', ',') # encode non-ascii characters assert re.match(r'^[a-zA-Z0-9_.-]+$', key), 'invalid key %r' % key
return os.path.join( return os.path.join(
self._get_root_dir(), section, '%s.%s' % (key, dtype)) self._get_root_dir(), section, '%s.%s' % (key, dtype))
@property @property
def enabled(self): def enabled(self):
return self._get_param('cachedir') is not False return self._ydl.params.get('cachedir') is not False
def store(self, section, key, data, dtype='json'): def store(self, section, key, data, dtype='json'):
assert dtype in ('json',) assert dtype in ('json',)
@ -71,55 +56,61 @@ class Cache(object):
fn = self._get_cache_fn(section, key, dtype) fn = self._get_cache_fn(section, key, dtype)
try: try:
compat_os_makedirs(os.path.dirname(fn), exist_ok=True) try:
self._write_debug('Saving {section}.{key} to cache'.format(section=section, key=key)) os.makedirs(os.path.dirname(fn))
except OSError as ose:
if ose.errno != errno.EEXIST:
raise
write_json_file({self._VERSION_KEY: __version__, 'data': data}, fn) write_json_file({self._VERSION_KEY: __version__, 'data': data}, fn)
except Exception: except Exception:
tb = traceback.format_exc() tb = traceback.format_exc()
self._report_warning('Writing cache to {fn!r} failed: {tb}'.format(fn=fn, tb=tb)) self._ydl.report_warning(
'Writing cache to %r failed: %s' % (fn, tb))
def _validate(self, data, min_ver): def _validate(self, data, min_ver):
version = traverse_obj(data, self._VERSION_KEY) version = try_get(data, lambda x: x[self._VERSION_KEY])
if not version: # Backward compatibility if not version: # Backward compatibility
data, version = {'data': data}, self._DEFAULT_VERSION data, version = {'data': data}, self._DEFAULT_VERSION
if not is_outdated_version(version, min_ver or '0', assume_new=False): if not is_outdated_version(version, min_ver or '0', assume_new=False):
return data['data'] return data['data']
self._write_debug('Discarding old cache from version {version} (needs {min_ver})'.format(version=version, min_ver=min_ver)) self._ydl.to_screen(
'Discarding old cache from version {version} (needs {min_ver})'.format(**locals()))
def load(self, section, key, dtype='json', default=None, **kw_min_ver): def load(self, section, key, dtype='json', default=None, min_ver=None):
assert dtype in ('json',) assert dtype in ('json',)
min_ver = kw_min_ver.get('min_ver')
if not self.enabled: if not self.enabled:
return default return default
cache_fn = self._get_cache_fn(section, key, dtype) cache_fn = self._get_cache_fn(section, key, dtype)
with compat_contextlib_suppress(IOError): # If no cache available
try: try:
with open(cache_fn, encoding='utf-8') as cachef: try:
self._write_debug('Loading {section}.{key} from cache'.format(section=section, key=key), only_once=True) with open(cache_fn, 'r', encoding='utf-8') as cachef:
return self._validate(json.load(cachef), min_ver) return self._validate(json.load(cachef), min_ver)
except (ValueError, KeyError): except ValueError:
try: try:
file_size = os.path.getsize(cache_fn) file_size = os.path.getsize(cache_fn)
except (OSError, IOError) as oe: except (OSError, IOError) as oe:
file_size = error_to_compat_str(oe) file_size = error_to_compat_str(oe)
self._report_warning('Cache retrieval from %s failed (%s)' % (cache_fn, file_size)) self._ydl.report_warning(
'Cache retrieval from %s failed (%s)' % (cache_fn, file_size))
except IOError:
pass # No cache available
return default return default
def remove(self): def remove(self):
if not self.enabled: if not self.enabled:
self._to_screen('Cache is disabled (Did you combine --no-cache-dir and --rm-cache-dir?)') self._ydl.to_screen('Cache is disabled (Did you combine --no-cache-dir and --rm-cache-dir?)')
return return
cachedir = self._get_root_dir() cachedir = self._get_root_dir()
if not any((term in cachedir) for term in ('cache', 'tmp')): if not any((term in cachedir) for term in ('cache', 'tmp')):
raise Exception('Not removing directory %s - this does not look like a cache dir' % (cachedir,)) raise Exception('Not removing directory %s - this does not look like a cache dir' % cachedir)
self._to_screen( self._ydl.to_screen(
'Removing cache dir %s .' % (cachedir,), skip_eol=True, ), 'Removing cache dir %s .' % cachedir, skip_eol=True)
if os.path.exists(cachedir): if os.path.exists(cachedir):
self._to_screen('.', skip_eol=True) self._ydl.to_screen('.', skip_eol=True)
shutil.rmtree(cachedir) shutil.rmtree(cachedir)
self._to_screen('.') self._ydl.to_screen('.')

View File

@ -2498,7 +2498,8 @@ try:
from urllib.parse import urlencode as compat_urllib_parse_urlencode from urllib.parse import urlencode as compat_urllib_parse_urlencode
from urllib.parse import parse_qs as compat_parse_qs from urllib.parse import parse_qs as compat_parse_qs
except ImportError: # Python 2 except ImportError: # Python 2
_asciire = getattr(compat_urllib_parse, '_asciire', None) or re.compile(r'([\x00-\x7f]+)') _asciire = (compat_urllib_parse._asciire if hasattr(compat_urllib_parse, '_asciire')
else re.compile(r'([\x00-\x7f]+)'))
# HACK: The following are the correct unquote_to_bytes, unquote and unquote_plus # HACK: The following are the correct unquote_to_bytes, unquote and unquote_plus
# implementations from cpython 3.4.3's stdlib. Python 2's version # implementations from cpython 3.4.3's stdlib. Python 2's version
@ -2566,21 +2567,24 @@ except ImportError: # Python 2
# Possible solutions are to either port it from python 3 with all # Possible solutions are to either port it from python 3 with all
# the friends or manually ensure input query contains only byte strings. # the friends or manually ensure input query contains only byte strings.
# We will stick with latter thus recursively encoding the whole query. # We will stick with latter thus recursively encoding the whole query.
def compat_urllib_parse_urlencode(query, doseq=0, safe='', encoding='utf-8', errors='strict'): def compat_urllib_parse_urlencode(query, doseq=0, encoding='utf-8'):
def encode_elem(e): def encode_elem(e):
if isinstance(e, dict): if isinstance(e, dict):
e = encode_dict(e) e = encode_dict(e)
elif isinstance(e, (list, tuple,)): elif isinstance(e, (list, tuple,)):
e = type(e)(encode_elem(el) for el in e) list_e = encode_list(e)
e = tuple(list_e) if isinstance(e, tuple) else list_e
elif isinstance(e, compat_str): elif isinstance(e, compat_str):
e = e.encode(encoding, errors) e = e.encode(encoding)
return e return e
def encode_dict(d): def encode_dict(d):
return tuple((encode_elem(k), encode_elem(v)) for k, v in d.items()) return dict((encode_elem(k), encode_elem(v)) for k, v in d.items())
return compat_urllib_parse._urlencode(encode_elem(query), doseq=doseq).decode('ascii') def encode_list(l):
return [encode_elem(e) for e in l]
return compat_urllib_parse._urlencode(encode_elem(query), doseq=doseq)
# HACK: The following is the correct parse_qs implementation from cpython 3's stdlib. # HACK: The following is the correct parse_qs implementation from cpython 3's stdlib.
# Python 2's version is apparently totally broken # Python 2's version is apparently totally broken
@ -2635,57 +2639,6 @@ except ImportError: # Python 2
('parse_qs', compat_parse_qs)): ('parse_qs', compat_parse_qs)):
setattr(compat_urllib_parse, name, fix) setattr(compat_urllib_parse, name, fix)
try:
all(chr(i) in b'' for i in range(256))
except TypeError:
# not all chr(i) are str: patch Python2 quote
_safemaps = getattr(compat_urllib_parse, '_safemaps', {})
_always_safe = frozenset(compat_urllib_parse.always_safe)
def _quote(s, safe='/'):
"""quote('abc def') -> 'abc%20def'"""
if not s and s is not None: # fast path
return s
safe = frozenset(safe)
cachekey = (safe, _always_safe)
try:
safe_map = _safemaps[cachekey]
except KeyError:
safe = _always_safe | safe
safe_map = {}
for i in range(256):
c = chr(i)
safe_map[c] = (
c if (i < 128 and c in safe)
else b'%{0:02X}'.format(i))
_safemaps[cachekey] = safe_map
if safe.issuperset(s):
return s
return ''.join(safe_map[c] for c in s)
# linked code
def _quote_plus(s, safe=''):
return (
_quote(s, safe + b' ').replace(b' ', b'+') if b' ' in s
else _quote(s, safe))
# linked code
def _urlcleanup():
if compat_urllib_parse._urlopener:
compat_urllib_parse._urlopener.cleanup()
_safemaps.clear()
compat_urllib_parse.ftpcache.clear()
for name, fix in (
('quote', _quote),
('quote_plus', _quote_plus),
('urlcleanup', _urlcleanup)):
setattr(compat_urllib_parse, '_' + name, getattr(compat_urllib_parse, name))
setattr(compat_urllib_parse, name, fix)
compat_urllib_parse_parse_qs = compat_parse_qs compat_urllib_parse_parse_qs = compat_parse_qs
@ -3167,21 +3120,6 @@ else:
compat_os_path_expanduser = compat_expanduser compat_os_path_expanduser = compat_expanduser
# compat_os_makedirs
try:
os.makedirs('.', exist_ok=True)
compat_os_makedirs = os.makedirs
except TypeError: # < Py3.2
from errno import EEXIST as _errno_EEXIST
def compat_os_makedirs(name, mode=0o777, exist_ok=False):
try:
return os.makedirs(name, mode=mode)
except OSError as ose:
if not (exist_ok and ose.errno == _errno_EEXIST):
raise
# compat_os_path_realpath # compat_os_path_realpath
if compat_os_name == 'nt' and sys.version_info < (3, 8): if compat_os_name == 'nt' and sys.version_info < (3, 8):
# os.path.realpath on Windows does not follow symbolic links # os.path.realpath on Windows does not follow symbolic links
@ -3699,7 +3637,6 @@ __all__ = [
'compat_numeric_types', 'compat_numeric_types',
'compat_open', 'compat_open',
'compat_ord', 'compat_ord',
'compat_os_makedirs',
'compat_os_name', 'compat_os_name',
'compat_os_path_expanduser', 'compat_os_path_expanduser',
'compat_os_path_realpath', 'compat_os_path_realpath',

View File

@ -505,7 +505,7 @@ class InfoExtractor(object):
if not self._x_forwarded_for_ip: if not self._x_forwarded_for_ip:
# Geo bypass mechanism is explicitly disabled by user # Geo bypass mechanism is explicitly disabled by user
if not self.get_param('geo_bypass', True): if not self._downloader.params.get('geo_bypass', True):
return return
if not geo_bypass_context: if not geo_bypass_context:
@ -527,7 +527,7 @@ class InfoExtractor(object):
# Explicit IP block specified by user, use it right away # Explicit IP block specified by user, use it right away
# regardless of whether extractor is geo bypassable or not # regardless of whether extractor is geo bypassable or not
ip_block = self.get_param('geo_bypass_ip_block', None) ip_block = self._downloader.params.get('geo_bypass_ip_block', None)
# Otherwise use random IP block from geo bypass context but only # Otherwise use random IP block from geo bypass context but only
# if extractor is known as geo bypassable # if extractor is known as geo bypassable
@ -538,8 +538,8 @@ class InfoExtractor(object):
if ip_block: if ip_block:
self._x_forwarded_for_ip = GeoUtils.random_ipv4(ip_block) self._x_forwarded_for_ip = GeoUtils.random_ipv4(ip_block)
if self.get_param('verbose', False): if self._downloader.params.get('verbose', False):
self.to_screen( self._downloader.to_screen(
'[debug] Using fake IP %s as X-Forwarded-For.' '[debug] Using fake IP %s as X-Forwarded-For.'
% self._x_forwarded_for_ip) % self._x_forwarded_for_ip)
return return
@ -548,7 +548,7 @@ class InfoExtractor(object):
# Explicit country code specified by user, use it right away # Explicit country code specified by user, use it right away
# regardless of whether extractor is geo bypassable or not # regardless of whether extractor is geo bypassable or not
country = self.get_param('geo_bypass_country', None) country = self._downloader.params.get('geo_bypass_country', None)
# Otherwise use random country code from geo bypass context but # Otherwise use random country code from geo bypass context but
# only if extractor is known as geo bypassable # only if extractor is known as geo bypassable
@ -559,8 +559,8 @@ class InfoExtractor(object):
if country: if country:
self._x_forwarded_for_ip = GeoUtils.random_ipv4(country) self._x_forwarded_for_ip = GeoUtils.random_ipv4(country)
if self.get_param('verbose', False): if self._downloader.params.get('verbose', False):
self.to_screen( self._downloader.to_screen(
'[debug] Using fake IP %s (%s) as X-Forwarded-For.' '[debug] Using fake IP %s (%s) as X-Forwarded-For.'
% (self._x_forwarded_for_ip, country.upper())) % (self._x_forwarded_for_ip, country.upper()))
@ -586,9 +586,9 @@ class InfoExtractor(object):
raise ExtractorError('An extractor error has occurred.', cause=e) raise ExtractorError('An extractor error has occurred.', cause=e)
def __maybe_fake_ip_and_retry(self, countries): def __maybe_fake_ip_and_retry(self, countries):
if (not self.get_param('geo_bypass_country', None) if (not self._downloader.params.get('geo_bypass_country', None)
and self._GEO_BYPASS and self._GEO_BYPASS
and self.get_param('geo_bypass', True) and self._downloader.params.get('geo_bypass', True)
and not self._x_forwarded_for_ip and not self._x_forwarded_for_ip
and countries): and countries):
country_code = random.choice(countries) country_code = random.choice(countries)
@ -698,7 +698,7 @@ class InfoExtractor(object):
if fatal: if fatal:
raise ExtractorError(errmsg, sys.exc_info()[2], cause=err) raise ExtractorError(errmsg, sys.exc_info()[2], cause=err)
else: else:
self.report_warning(errmsg) self._downloader.report_warning(errmsg)
return False return False
def _download_webpage_handle(self, url_or_request, video_id, note=None, errnote=None, fatal=True, encoding=None, data=None, headers={}, query={}, expected_status=None): def _download_webpage_handle(self, url_or_request, video_id, note=None, errnote=None, fatal=True, encoding=None, data=None, headers={}, query={}, expected_status=None):
@ -770,11 +770,11 @@ class InfoExtractor(object):
webpage_bytes = prefix + webpage_bytes webpage_bytes = prefix + webpage_bytes
if not encoding: if not encoding:
encoding = self._guess_encoding_from_content(content_type, webpage_bytes) encoding = self._guess_encoding_from_content(content_type, webpage_bytes)
if self.get_param('dump_intermediate_pages', False): if self._downloader.params.get('dump_intermediate_pages', False):
self.to_screen('Dumping request to ' + urlh.geturl()) self.to_screen('Dumping request to ' + urlh.geturl())
dump = base64.b64encode(webpage_bytes).decode('ascii') dump = base64.b64encode(webpage_bytes).decode('ascii')
self.to_screen(dump) self._downloader.to_screen(dump)
if self.get_param('write_pages', False): if self._downloader.params.get('write_pages', False):
basen = '%s_%s' % (video_id, urlh.geturl()) basen = '%s_%s' % (video_id, urlh.geturl())
if len(basen) > 240: if len(basen) > 240:
h = '___' + hashlib.md5(basen.encode('utf-8')).hexdigest() h = '___' + hashlib.md5(basen.encode('utf-8')).hexdigest()
@ -976,9 +976,19 @@ class InfoExtractor(object):
"""Print msg to screen, prefixing it with '[ie_name]'""" """Print msg to screen, prefixing it with '[ie_name]'"""
self._downloader.to_screen(self.__ie_msg(msg)) self._downloader.to_screen(self.__ie_msg(msg))
def write_debug(self, msg, only_once=False): def write_debug(self, msg, only_once=False, _cache=[]):
'''Log debug message or Print message to stderr''' '''Log debug message or Print message to stderr'''
self._downloader.write_debug(self.__ie_msg(msg), only_once=only_once) if not self.get_param('verbose', False):
return
message = '[debug] ' + self.__ie_msg(msg)
logger = self.get_param('logger')
if logger:
logger.debug(message)
else:
if only_once and hash(message) in _cache:
return
self._downloader.to_stderr(message)
_cache.append(hash(message))
# name, default=None, *args, **kwargs # name, default=None, *args, **kwargs
def get_param(self, name, *args, **kwargs): def get_param(self, name, *args, **kwargs):
@ -1074,7 +1084,7 @@ class InfoExtractor(object):
if mobj: if mobj:
break break
if not self.get_param('no_color') and compat_os_name != 'nt' and sys.stderr.isatty(): if not self._downloader.params.get('no_color') and compat_os_name != 'nt' and sys.stderr.isatty():
_name = '\033[0;34m%s\033[0m' % name _name = '\033[0;34m%s\033[0m' % name
else: else:
_name = name _name = name
@ -1092,7 +1102,7 @@ class InfoExtractor(object):
elif fatal: elif fatal:
raise RegexNotFoundError('Unable to extract %s' % _name) raise RegexNotFoundError('Unable to extract %s' % _name)
else: else:
self.report_warning('unable to extract %s' % _name + bug_reports_message()) self._downloader.report_warning('unable to extract %s' % _name + bug_reports_message())
return None return None
def _search_json(self, start_pattern, string, name, video_id, **kwargs): def _search_json(self, start_pattern, string, name, video_id, **kwargs):
@ -1162,7 +1172,7 @@ class InfoExtractor(object):
username = None username = None
password = None password = None
if self.get_param('usenetrc', False): if self._downloader.params.get('usenetrc', False):
try: try:
netrc_machine = netrc_machine or self._NETRC_MACHINE netrc_machine = netrc_machine or self._NETRC_MACHINE
info = netrc.netrc().authenticators(netrc_machine) info = netrc.netrc().authenticators(netrc_machine)
@ -1173,7 +1183,7 @@ class InfoExtractor(object):
raise netrc.NetrcParseError( raise netrc.NetrcParseError(
'No authenticators for %s' % netrc_machine) 'No authenticators for %s' % netrc_machine)
except (AttributeError, IOError, netrc.NetrcParseError) as err: except (AttributeError, IOError, netrc.NetrcParseError) as err:
self.report_warning( self._downloader.report_warning(
'parsing .netrc: %s' % error_to_compat_str(err)) 'parsing .netrc: %s' % error_to_compat_str(err))
return username, password return username, password
@ -1210,10 +1220,10 @@ class InfoExtractor(object):
""" """
if self._downloader is None: if self._downloader is None:
return None return None
downloader_params = self._downloader.params
twofactor = self.get_param('twofactor') if downloader_params.get('twofactor') is not None:
if twofactor is not None: return downloader_params['twofactor']
return twofactor
return compat_getpass('Type %s and press [Return]: ' % note) return compat_getpass('Type %s and press [Return]: ' % note)
@ -1348,7 +1358,7 @@ class InfoExtractor(object):
elif fatal: elif fatal:
raise RegexNotFoundError('Unable to extract JSON-LD') raise RegexNotFoundError('Unable to extract JSON-LD')
else: else:
self.report_warning('unable to extract JSON-LD %s' % bug_reports_message()) self._downloader.report_warning('unable to extract JSON-LD %s' % bug_reports_message())
return {} return {}
def _json_ld(self, json_ld, video_id, fatal=True, expected_type=None): def _json_ld(self, json_ld, video_id, fatal=True, expected_type=None):
@ -1579,7 +1589,7 @@ class InfoExtractor(object):
if f.get('vcodec') == 'none': # audio only if f.get('vcodec') == 'none': # audio only
preference -= 50 preference -= 50
if self.get_param('prefer_free_formats'): if self._downloader.params.get('prefer_free_formats'):
ORDER = ['aac', 'mp3', 'm4a', 'webm', 'ogg', 'opus'] ORDER = ['aac', 'mp3', 'm4a', 'webm', 'ogg', 'opus']
else: else:
ORDER = ['webm', 'opus', 'ogg', 'mp3', 'aac', 'm4a'] ORDER = ['webm', 'opus', 'ogg', 'mp3', 'aac', 'm4a']
@ -1591,7 +1601,7 @@ class InfoExtractor(object):
else: else:
if f.get('acodec') == 'none': # video only if f.get('acodec') == 'none': # video only
preference -= 40 preference -= 40
if self.get_param('prefer_free_formats'): if self._downloader.params.get('prefer_free_formats'):
ORDER = ['flv', 'mp4', 'webm'] ORDER = ['flv', 'mp4', 'webm']
else: else:
ORDER = ['webm', 'flv', 'mp4'] ORDER = ['webm', 'flv', 'mp4']
@ -1657,7 +1667,7 @@ class InfoExtractor(object):
""" Either "http:" or "https:", depending on the user's preferences """ """ Either "http:" or "https:", depending on the user's preferences """
return ( return (
'http:' 'http:'
if self.get_param('prefer_insecure', False) if self._downloader.params.get('prefer_insecure', False)
else 'https:') else 'https:')
def _proto_relative_url(self, url, scheme=None): def _proto_relative_url(self, url, scheme=None):
@ -3189,7 +3199,7 @@ class InfoExtractor(object):
if fatal: if fatal:
raise ExtractorError(msg) raise ExtractorError(msg)
else: else:
self.report_warning(msg) self._downloader.report_warning(msg)
return res return res
def _float(self, v, name, fatal=False, **kwargs): def _float(self, v, name, fatal=False, **kwargs):
@ -3199,7 +3209,7 @@ class InfoExtractor(object):
if fatal: if fatal:
raise ExtractorError(msg) raise ExtractorError(msg)
else: else:
self.report_warning(msg) self._downloader.report_warning(msg)
return res return res
def _set_cookie(self, domain, name, value, expire_time=None, port=None, def _set_cookie(self, domain, name, value, expire_time=None, port=None,
@ -3208,12 +3218,12 @@ class InfoExtractor(object):
0, name, value, port, port is not None, domain, True, 0, name, value, port, port is not None, domain, True,
domain.startswith('.'), path, True, secure, expire_time, domain.startswith('.'), path, True, secure, expire_time,
discard, None, None, rest) discard, None, None, rest)
self.cookiejar.set_cookie(cookie) self._downloader.cookiejar.set_cookie(cookie)
def _get_cookies(self, url): def _get_cookies(self, url):
""" Return a compat_cookies_SimpleCookie with the cookies for the url """ """ Return a compat_cookies_SimpleCookie with the cookies for the url """
req = sanitized_Request(url) req = sanitized_Request(url)
self.cookiejar.add_cookie_header(req) self._downloader.cookiejar.add_cookie_header(req)
return compat_cookies_SimpleCookie(req.get_header('Cookie')) return compat_cookies_SimpleCookie(req.get_header('Cookie'))
def _apply_first_set_cookie_header(self, url_handle, cookie): def _apply_first_set_cookie_header(self, url_handle, cookie):
@ -3273,8 +3283,8 @@ class InfoExtractor(object):
return not any_restricted return not any_restricted
def extract_subtitles(self, *args, **kwargs): def extract_subtitles(self, *args, **kwargs):
if (self.get_param('writesubtitles', False) if (self._downloader.params.get('writesubtitles', False)
or self.get_param('listsubtitles')): or self._downloader.params.get('listsubtitles')):
return self._get_subtitles(*args, **kwargs) return self._get_subtitles(*args, **kwargs)
return {} return {}
@ -3295,11 +3305,7 @@ class InfoExtractor(object):
""" Merge subtitle dictionaries, language by language. """ """ Merge subtitle dictionaries, language by language. """
# ..., * , target=None # ..., * , target=None
target = kwargs.get('target') target = kwargs.get('target') or dict(subtitle_dict1)
if target is None:
target = dict(subtitle_dict1)
else:
subtitle_dicts = (subtitle_dict1,) + subtitle_dicts
for subtitle_dict in subtitle_dicts: for subtitle_dict in subtitle_dicts:
for lang in subtitle_dict: for lang in subtitle_dict:
@ -3307,8 +3313,8 @@ class InfoExtractor(object):
return target return target
def extract_automatic_captions(self, *args, **kwargs): def extract_automatic_captions(self, *args, **kwargs):
if (self.get_param('writeautomaticsub', False) if (self._downloader.params.get('writeautomaticsub', False)
or self.get_param('listsubtitles')): or self._downloader.params.get('listsubtitles')):
return self._get_automatic_captions(*args, **kwargs) return self._get_automatic_captions(*args, **kwargs)
return {} return {}
@ -3316,9 +3322,9 @@ class InfoExtractor(object):
raise NotImplementedError('This method must be implemented by subclasses') raise NotImplementedError('This method must be implemented by subclasses')
def mark_watched(self, *args, **kwargs): def mark_watched(self, *args, **kwargs):
if (self.get_param('mark_watched', False) if (self._downloader.params.get('mark_watched', False)
and (self._get_login_info()[0] is not None and (self._get_login_info()[0] is not None
or self.get_param('cookiefile') is not None)): or self._downloader.params.get('cookiefile') is not None)):
self._mark_watched(*args, **kwargs) self._mark_watched(*args, **kwargs)
def _mark_watched(self, *args, **kwargs): def _mark_watched(self, *args, **kwargs):
@ -3326,7 +3332,7 @@ class InfoExtractor(object):
def geo_verification_headers(self): def geo_verification_headers(self):
headers = {} headers = {}
geo_verification_proxy = self.get_param('geo_verification_proxy') geo_verification_proxy = self._downloader.params.get('geo_verification_proxy')
if geo_verification_proxy: if geo_verification_proxy:
headers['Ytdl-request-proxy'] = geo_verification_proxy headers['Ytdl-request-proxy'] = geo_verification_proxy
return headers return headers

View File

@ -342,7 +342,14 @@ class YoutubeBaseInfoExtractor(InfoExtractor):
if not self._login(): if not self._login():
return return
_DEFAULT_API_DATA = {'context': _INNERTUBE_CLIENTS['web']['INNERTUBE_CONTEXT']} _DEFAULT_API_DATA = {
'context': {
'client': {
'clientName': 'WEB',
'clientVersion': '2.20201021.03.00',
},
},
}
_YT_INITIAL_DATA_RE = r'(?:window\s*\[\s*["\']ytInitialData["\']\s*\]|ytInitialData)\s*=\s*({.+?})\s*;' _YT_INITIAL_DATA_RE = r'(?:window\s*\[\s*["\']ytInitialData["\']\s*\]|ytInitialData)\s*=\s*({.+?})\s*;'
_YT_INITIAL_PLAYER_RESPONSE_RE = r'ytInitialPlayerResponse\s*=\s*({.+?})\s*;' _YT_INITIAL_PLAYER_RESPONSE_RE = r'ytInitialPlayerResponse\s*=\s*({.+?})\s*;'
@ -490,15 +497,11 @@ class YoutubeBaseInfoExtractor(InfoExtractor):
data['params'] = params data['params'] = params
for page_num in itertools.count(1): for page_num in itertools.count(1):
search = self._download_json( search = self._download_json(
'https://www.youtube.com/youtubei/v1/search', 'https://www.youtube.com/youtubei/v1/search?key=AIzaSyAO_FJ2SlqU8Q4STEHLGCilw_Y9_11qcW8',
video_id='query "%s"' % query, video_id='query "%s"' % query,
note='Downloading page %s' % page_num, note='Downloading page %s' % page_num,
errnote='Unable to download API page', fatal=False, errnote='Unable to download API page', fatal=False,
data=json.dumps(data).encode('utf8'), data=json.dumps(data).encode('utf8'),
query={
# 'key': 'AIzaSyAO_FJ2SlqU8Q4STEHLGCilw_Y9_11qcW8',
'prettyPrint': 'false',
},
headers={'content-type': 'application/json'}) headers={'content-type': 'application/json'})
if not search: if not search:
break break
@ -1652,7 +1655,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
assert os.path.basename(func_id) == func_id assert os.path.basename(func_id) == func_id
self.write_debug('Extracting signature function {0}'.format(func_id)) self.write_debug('Extracting signature function {0}'.format(func_id))
cache_spec, code = self.cache.load('youtube-sigfuncs', func_id, min_ver='2025.04.07'), None cache_spec, code = self.cache.load('youtube-sigfuncs', func_id), None
if not cache_spec: if not cache_spec:
code = self._load_player(video_id, player_url, player_id) code = self._load_player(video_id, player_url, player_id)
@ -1813,10 +1816,6 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
return ret return ret
def _extract_n_function_name(self, jscode): def _extract_n_function_name(self, jscode):
func_name, idx = None, None
# these special cases are redundant and probably obsolete (2025-04):
# they make the tests run ~10% faster without fallback warnings
r"""
func_name, idx = self._search_regex( func_name, idx = self._search_regex(
# (y=NuD(),Mw(k),q=k.Z[y]||null)&&(q=narray[idx](q),k.set(y,q),k.V||NuD(''))}}; # (y=NuD(),Mw(k),q=k.Z[y]||null)&&(q=narray[idx](q),k.set(y,q),k.V||NuD(''))}};
# (R="nn"[+J.Z],mW(J),N=J.K[R]||null)&&(N=narray[idx](N),J.set(R,N))}}; # (R="nn"[+J.Z],mW(J),N=J.K[R]||null)&&(N=narray[idx](N),J.set(R,N))}};
@ -1843,28 +1842,9 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
\(\s*[\w$]+\s*\) \(\s*[\w$]+\s*\)
''', jscode, 'Initial JS player n function name', group=('nfunc', 'idx'), ''', jscode, 'Initial JS player n function name', group=('nfunc', 'idx'),
default=(None, None)) default=(None, None))
"""
if not func_name:
# nfunc=function(x){...}|function nfunc(x); ...
# ... var y=[nfunc]|y[idx]=nfunc);
# obvious REs hang, so use a two-stage tactic
for m in re.finditer(r'''(?x)
[\n;]var\s(?:(?:(?!,).)+,|\s)*?(?!\d)[\w$]+(?:\[(?P<idx>\d+)\])?\s*=\s*
(?(idx)|\[\s*)(?P<nfunc>(?!\d)[\w$]+)(?(idx)|\s*\])
\s*?[;\n]
''', jscode):
func_name = self._search_regex(
r'[;,]\s*(function\s+)?({0})(?(1)|\s*=\s*function)\s*\((?!\d)[\w$]+\)\s*\{1}(?!\s*return\s)'.format(
re.escape(m.group('nfunc')), '{'),
jscode, 'Initial JS player n function name (2)', group=2, default=None)
if func_name:
idx = m.group('idx')
break
# thx bashonly: yt-dlp/yt-dlp/pull/10611 # thx bashonly: yt-dlp/yt-dlp/pull/10611
if not func_name: if not func_name:
self.report_warning('Falling back to generic n function search', only_once=True) self.report_warning('Falling back to generic n function search')
return self._search_regex( return self._search_regex(
r'''(?xs) r'''(?xs)
(?:(?<=[^\w$])|^) # instead of \b, which ignores $ (?:(?<=[^\w$])|^) # instead of \b, which ignores $
@ -1878,14 +1858,14 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
return func_name return func_name
return self._search_json( return self._search_json(
r'(?<![\w-])var\s(?:(?:(?!,).)+,|\s)*?{0}\s*='.format(re.escape(func_name)), jscode, r'var\s+{0}\s*='.format(re.escape(func_name)), jscode,
'Initial JS player n function list ({0}.{1})'.format(func_name, idx), 'Initial JS player n function list ({0}.{1})'.format(func_name, idx),
func_name, contains_pattern=r'\[.+\]', end_pattern='[,;]', func_name, contains_pattern=r'\[[\s\S]+\]', end_pattern='[,;]',
transform_source=js_to_json)[int(idx)] transform_source=js_to_json)[int(idx)]
def _extract_n_function_code(self, video_id, player_url): def _extract_n_function_code(self, video_id, player_url):
player_id = self._extract_player_info(player_url) player_id = self._extract_player_info(player_url)
func_code = self.cache.load('youtube-nsig', player_id, min_ver='2025.04.07') func_code = self.cache.load('youtube-nsig', player_id)
jscode = func_code or self._load_player(video_id, player_url) jscode = func_code or self._load_player(video_id, player_url)
jsi = JSInterpreter(jscode) jsi = JSInterpreter(jscode)
@ -3359,20 +3339,6 @@ class YoutubeTabIE(YoutubeBaseInfoExtractor):
'thumbnailViewModel', 'image'), final_key='sources'), 'thumbnailViewModel', 'image'), final_key='sources'),
}) })
def _extract_shorts_lockup_view_model(self, view_model):
content_id = traverse_obj(view_model, (
'onTap', 'innertubeCommand', 'reelWatchEndpoint', 'videoId',
T(lambda v: v if YoutubeIE.suitable(v) else None)))
if not content_id:
return
return merge_dicts(self.url_result(
content_id, ie=YoutubeIE.ie_key(), video_id=content_id), {
'title': traverse_obj(view_model, (
'overlayMetadata', 'primaryText', 'content', T(compat_str))),
'thumbnails': self._extract_thumbnails(
view_model, 'thumbnail', final_key='sources'),
})
def _video_entry(self, video_renderer): def _video_entry(self, video_renderer):
video_id = video_renderer.get('videoId') video_id = video_renderer.get('videoId')
if video_id: if video_id:
@ -3419,9 +3385,10 @@ class YoutubeTabIE(YoutubeBaseInfoExtractor):
yield entry yield entry
def _rich_grid_entries(self, contents): def _rich_grid_entries(self, contents):
for content in traverse_obj( for content in contents:
contents, (Ellipsis, 'richItemRenderer', 'content'), content = traverse_obj(
expected_type=dict): content, ('richItemRenderer', 'content'),
expected_type=dict) or {}
video_renderer = traverse_obj( video_renderer = traverse_obj(
content, 'videoRenderer', 'reelItemRenderer', content, 'videoRenderer', 'reelItemRenderer',
expected_type=dict) expected_type=dict)
@ -3429,12 +3396,6 @@ class YoutubeTabIE(YoutubeBaseInfoExtractor):
entry = self._video_entry(video_renderer) entry = self._video_entry(video_renderer)
if entry: if entry:
yield entry yield entry
# shorts item
shorts_lockup_view_model = content.get('shortsLockupViewModel')
if shorts_lockup_view_model:
entry = self._extract_shorts_lockup_view_model(shorts_lockup_view_model)
if entry:
yield entry
# playlist # playlist
renderer = traverse_obj( renderer = traverse_obj(
content, 'playlistRenderer', expected_type=dict) or {} content, 'playlistRenderer', expected_type=dict) or {}
@ -3473,15 +3434,23 @@ class YoutubeTabIE(YoutubeBaseInfoExtractor):
next_continuation = cls._extract_next_continuation_data(renderer) next_continuation = cls._extract_next_continuation_data(renderer)
if next_continuation: if next_continuation:
return next_continuation return next_continuation
for command in traverse_obj(renderer, ( contents = []
('contents', 'items', 'rows'), Ellipsis, 'continuationItemRenderer', for key in ('contents', 'items'):
('continuationEndpoint', ('button', 'buttonRenderer', 'command')), contents.extend(try_get(renderer, lambda x: x[key], list) or [])
(('commandExecutorCommand', 'commands', Ellipsis), None), T(dict))): for content in contents:
continuation = traverse_obj(command, ('continuationCommand', 'token', T(compat_str))) if not isinstance(content, dict):
continue
continuation_ep = try_get(
content, lambda x: x['continuationItemRenderer']['continuationEndpoint'],
dict)
if not continuation_ep:
continue
continuation = try_get(
continuation_ep, lambda x: x['continuationCommand']['token'], compat_str)
if not continuation: if not continuation:
continue continue
ctp = command.get('clickTrackingParams') ctp = continuation_ep.get('clickTrackingParams')
return cls._build_continuation_query(continuation, ctp) return YoutubeTabIE._build_continuation_query(continuation, ctp)
def _entries(self, tab, item_id, webpage): def _entries(self, tab, item_id, webpage):
tab_content = try_get(tab, lambda x: x['content'], dict) tab_content = try_get(tab, lambda x: x['content'], dict)
@ -3530,13 +3499,6 @@ class YoutubeTabIE(YoutubeBaseInfoExtractor):
entry = self._video_entry(renderer) entry = self._video_entry(renderer)
if entry: if entry:
yield entry yield entry
renderer = isr_content.get('richGridRenderer')
if renderer:
for from_ in self._rich_grid_entries(
traverse_obj(renderer, ('contents', Ellipsis, T(dict)))):
yield from_
continuation = self._extract_continuation(renderer)
continue
if not continuation: if not continuation:
continuation = self._extract_continuation(is_renderer) continuation = self._extract_continuation(is_renderer)
@ -3546,9 +3508,8 @@ class YoutubeTabIE(YoutubeBaseInfoExtractor):
rich_grid_renderer = tab_content.get('richGridRenderer') rich_grid_renderer = tab_content.get('richGridRenderer')
if not rich_grid_renderer: if not rich_grid_renderer:
return return
for from_ in self._rich_grid_entries( for entry in self._rich_grid_entries(rich_grid_renderer.get('contents') or []):
traverse_obj(rich_grid_renderer, ('contents', Ellipsis, T(dict)))): yield entry
yield from_
continuation = self._extract_continuation(rich_grid_renderer) continuation = self._extract_continuation(rich_grid_renderer)
@ -3594,12 +3555,8 @@ class YoutubeTabIE(YoutubeBaseInfoExtractor):
# Downloading page may result in intermittent 5xx HTTP error # Downloading page may result in intermittent 5xx HTTP error
# that is usually worked around with a retry # that is usually worked around with a retry
response = self._download_json( response = self._download_json(
'https://www.youtube.com/youtubei/v1/browse', 'https://www.youtube.com/youtubei/v1/browse?key=AIzaSyAO_FJ2SlqU8Q4STEHLGCilw_Y9_11qcW8',
None, 'Downloading page %d%s' % (page_num, ' (retry #%d)' % count if count else ''), None, 'Downloading page %d%s' % (page_num, ' (retry #%d)' % count if count else ''),
query={
# 'key': 'AIzaSyAO_FJ2SlqU8Q4STEHLGCilw_Y9_11qcW8',
'prettyPrint': 'false',
},
headers=headers, data=json.dumps(data).encode('utf8')) headers=headers, data=json.dumps(data).encode('utf8'))
break break
except ExtractorError as e: except ExtractorError as e:

View File

@ -240,7 +240,7 @@ def _js_ternary(cndn, if_true=True, if_false=False):
def _js_unary_op(op): def _js_unary_op(op):
@wraps_op(op) @wraps_op(op)
def wrapped(a, _): def wrapped(_, a):
return op(a) return op(a)
return wrapped return wrapped
@ -283,6 +283,17 @@ _OPERATORS = (
('**', _js_exp), ('**', _js_exp),
) )
_COMP_OPERATORS = (
('===', _js_id_op(operator.is_)),
('!==', _js_id_op(operator.is_not)),
('==', _js_eq),
('!=', _js_neq),
('<=', _js_comp_op(operator.le)),
('>=', _js_comp_op(operator.ge)),
('<', _js_comp_op(operator.lt)),
('>', _js_comp_op(operator.gt)),
)
_LOG_OPERATORS = ( _LOG_OPERATORS = (
('|', _js_bit_op(operator.or_)), ('|', _js_bit_op(operator.or_)),
('^', _js_bit_op(operator.xor)), ('^', _js_bit_op(operator.xor)),
@ -299,27 +310,13 @@ _SC_OPERATORS = (
_UNARY_OPERATORS_X = ( _UNARY_OPERATORS_X = (
('void', _js_unary_op(lambda _: JS_Undefined)), ('void', _js_unary_op(lambda _: JS_Undefined)),
('typeof', _js_unary_op(_js_typeof)), ('typeof', _js_unary_op(_js_typeof)),
# avoid functools.partial here since Py2 update_wrapper(partial) -> no __module__
('!', _js_unary_op(lambda x: _js_ternary(x, if_true=False, if_false=True))),
) )
_COMP_OPERATORS = ( _OPERATOR_RE = '|'.join(map(lambda x: re.escape(x[0]), _OPERATORS + _LOG_OPERATORS))
('===', _js_id_op(operator.is_)),
('!==', _js_id_op(operator.is_not)),
('==', _js_eq),
('!=', _js_neq),
('<=', _js_comp_op(operator.le)),
('>=', _js_comp_op(operator.ge)),
('<', _js_comp_op(operator.lt)),
('>', _js_comp_op(operator.gt)),
)
_OPERATOR_RE = '|'.join(map(lambda x: re.escape(x[0]), _OPERATORS + _LOG_OPERATORS + _SC_OPERATORS))
_NAME_RE = r'[a-zA-Z_$][\w$]*' _NAME_RE = r'[a-zA-Z_$][\w$]*'
_MATCHING_PARENS = dict(zip(*zip('()', '{}', '[]'))) _MATCHING_PARENS = dict(zip(*zip('()', '{}', '[]')))
_QUOTES = '\'"/' _QUOTES = '\'"/'
_NESTED_BRACKETS = r'[^[\]]+(?:\[[^[\]]+(?:\[[^\]]+\])?\])?'
class JS_Break(ExtractorError): class JS_Break(ExtractorError):
@ -356,7 +353,7 @@ class LocalNameSpace(ChainMap):
raise NotImplementedError('Deleting is not supported') raise NotImplementedError('Deleting is not supported')
def __repr__(self): def __repr__(self):
return 'LocalNameSpace({0!r})'.format(self.maps) return 'LocalNameSpace%s' % (self.maps, )
class Debugger(object): class Debugger(object):
@ -377,9 +374,6 @@ class Debugger(object):
@classmethod @classmethod
def wrap_interpreter(cls, f): def wrap_interpreter(cls, f):
if not cls.ENABLED:
return f
@wraps(f) @wraps(f)
def interpret_statement(self, stmt, local_vars, allow_recursion, *args, **kwargs): def interpret_statement(self, stmt, local_vars, allow_recursion, *args, **kwargs):
if cls.ENABLED and stmt.strip(): if cls.ENABLED and stmt.strip():
@ -420,17 +414,7 @@ class JSInterpreter(object):
msg = '{0} in: {1!r:.100}'.format(msg.rstrip(), expr) msg = '{0} in: {1!r:.100}'.format(msg.rstrip(), expr)
super(JSInterpreter.Exception, self).__init__(msg, *args, **kwargs) super(JSInterpreter.Exception, self).__init__(msg, *args, **kwargs)
class JS_Object(object): class JS_RegExp(object):
def __getitem__(self, key):
if hasattr(self, key):
return getattr(self, key)
raise KeyError(key)
def dump(self):
"""Serialise the instance"""
raise NotImplementedError
class JS_RegExp(JS_Object):
RE_FLAGS = { RE_FLAGS = {
# special knowledge: Python's re flags are bitmask values, current max 128 # special knowledge: Python's re flags are bitmask values, current max 128
# invent new bitmask values well above that for literal parsing # invent new bitmask values well above that for literal parsing
@ -451,24 +435,16 @@ class JSInterpreter(object):
def __init__(self, pattern_txt, flags=0): def __init__(self, pattern_txt, flags=0):
if isinstance(flags, compat_str): if isinstance(flags, compat_str):
flags, _ = self.regex_flags(flags) flags, _ = self.regex_flags(flags)
# First, avoid https://github.com/python/cpython/issues/74534
self.__self = None self.__self = None
pattern_txt = str_or_none(pattern_txt) or '(?:)' pattern_txt = str_or_none(pattern_txt) or '(?:)'
# escape unintended embedded flags self.__pattern_txt = pattern_txt.replace('[[', r'[\[')
pattern_txt = re.sub(
r'(\(\?)([aiLmsux]*)(-[imsx]+:|(?<!\?)\))',
lambda m: ''.join(
(re.escape(m.group(1)), m.group(2), re.escape(m.group(3)))
if m.group(3) == ')'
else ('(?:', m.group(2), m.group(3))),
pattern_txt)
# Avoid https://github.com/python/cpython/issues/74534
self.source = pattern_txt.replace('[[', r'[\[')
self.__flags = flags self.__flags = flags
def __instantiate(self): def __instantiate(self):
if self.__self: if self.__self:
return return
self.__self = re.compile(self.source, self.__flags) self.__self = re.compile(self.__pattern_txt, self.__flags)
# Thx: https://stackoverflow.com/questions/44773522/setattr-on-python2-sre-sre-pattern # Thx: https://stackoverflow.com/questions/44773522/setattr-on-python2-sre-sre-pattern
for name in dir(self.__self): for name in dir(self.__self):
# Only these? Obviously __class__, __init__. # Only these? Obviously __class__, __init__.
@ -476,15 +452,16 @@ class JSInterpreter(object):
# that can't be setattr'd but also can't need to be copied. # that can't be setattr'd but also can't need to be copied.
if name in ('__class__', '__init__', '__weakref__'): if name in ('__class__', '__init__', '__weakref__'):
continue continue
if name == 'flags':
setattr(self, name, getattr(self.__self, name, self.__flags))
else:
setattr(self, name, getattr(self.__self, name)) setattr(self, name, getattr(self.__self, name))
def __getattr__(self, name): def __getattr__(self, name):
self.__instantiate() self.__instantiate()
if name == 'pattern': # make Py 2.6 conform to its lying documentation
self.pattern = self.source if name == 'flags':
self.flags = self.__flags
return self.flags
elif name == 'pattern':
self.pattern = self.__pattern_txt
return self.pattern return self.pattern
elif hasattr(self.__self, name): elif hasattr(self.__self, name):
v = getattr(self.__self, name) v = getattr(self.__self, name)
@ -492,26 +469,6 @@ class JSInterpreter(object):
return v return v
elif name in ('groupindex', 'groups'): elif name in ('groupindex', 'groups'):
return 0 if name == 'groupindex' else {} return 0 if name == 'groupindex' else {}
else:
flag_attrs = ( # order by 2nd elt
('hasIndices', 'd'),
('global', 'g'),
('ignoreCase', 'i'),
('multiline', 'm'),
('dotAll', 's'),
('unicode', 'u'),
('unicodeSets', 'v'),
('sticky', 'y'),
)
for k, c in flag_attrs:
if name == k:
return bool(self.RE_FLAGS[c] & self.__flags)
else:
if name == 'flags':
return ''.join(
(c if self.RE_FLAGS[c] & self.__flags else '')
for _, c in flag_attrs)
raise AttributeError('{0} has no attribute named {1}'.format(self, name)) raise AttributeError('{0} has no attribute named {1}'.format(self, name))
@classmethod @classmethod
@ -525,16 +482,7 @@ class JSInterpreter(object):
flags |= cls.RE_FLAGS[ch] flags |= cls.RE_FLAGS[ch]
return flags, expr[idx + 1:] return flags, expr[idx + 1:]
def dump(self): class JS_Date(object):
return '(/{0}/{1})'.format(
re.sub(r'(?<!\\)/', r'\/', self.source),
self.flags)
@staticmethod
def escape(string_):
return re.escape(string_)
class JS_Date(JS_Object):
_t = None _t = None
@staticmethod @staticmethod
@ -601,9 +549,6 @@ class JSInterpreter(object):
def valueOf(self): def valueOf(self):
return _NaN if self._t is None else self._t return _NaN if self._t is None else self._t
def dump(self):
return '(new Date({0}))'.format(self.toString())
@classmethod @classmethod
def __op_chars(cls): def __op_chars(cls):
op_chars = set(';,[') op_chars = set(';,[')
@ -707,68 +652,6 @@ class JSInterpreter(object):
_SC_OPERATORS, _LOG_OPERATORS, _COMP_OPERATORS, _OPERATORS, _UNARY_OPERATORS_X)) _SC_OPERATORS, _LOG_OPERATORS, _COMP_OPERATORS, _OPERATORS, _UNARY_OPERATORS_X))
return _cached return _cached
def _separate_at_op(self, expr, max_split=None):
for op, _ in self._all_operators():
# hackety: </> have higher priority than <</>>, but don't confuse them
skip_delim = (op + op) if op in '<>*?' else None
if op == '?':
skip_delim = (skip_delim, '?.')
separated = list(self._separate(expr, op, skip_delims=skip_delim))
if len(separated) < 2:
continue
right_expr = separated.pop()
# handle operators that are both unary and binary, minimal BODMAS
if op in ('+', '-'):
# simplify/adjust consecutive instances of these operators
undone = 0
separated = [s.strip() for s in separated]
while len(separated) > 1 and not separated[-1]:
undone += 1
separated.pop()
if op == '-' and undone % 2 != 0:
right_expr = op + right_expr
elif op == '+':
while len(separated) > 1 and set(separated[-1]) <= self.OP_CHARS:
right_expr = separated.pop() + right_expr
if separated[-1][-1:] in self.OP_CHARS:
right_expr = separated.pop() + right_expr
# hanging op at end of left => unary + (strip) or - (push right)
separated.append(right_expr)
dm_ops = ('*', '%', '/', '**')
dm_chars = set(''.join(dm_ops))
def yield_terms(s):
skip = False
for i, term in enumerate(s[:-1]):
if skip:
skip = False
continue
if not (dm_chars & set(term)):
yield term
continue
for dm_op in dm_ops:
bodmas = list(self._separate(term, dm_op, skip_delims=skip_delim))
if len(bodmas) > 1 and not bodmas[-1].strip():
bodmas[-1] = (op if op == '-' else '') + s[i + 1]
yield dm_op.join(bodmas)
skip = True
break
else:
if term:
yield term
if not skip and s[-1]:
yield s[-1]
separated = list(yield_terms(separated))
right_expr = separated.pop() if len(separated) > 1 else None
expr = op.join(separated)
if right_expr is None:
continue
return op, separated, right_expr
def _operator(self, op, left_val, right_expr, expr, local_vars, allow_recursion): def _operator(self, op, left_val, right_expr, expr, local_vars, allow_recursion):
if op in ('||', '&&'): if op in ('||', '&&'):
if (op == '&&') ^ _js_ternary(left_val): if (op == '&&') ^ _js_ternary(left_val):
@ -779,7 +662,7 @@ class JSInterpreter(object):
elif op == '?': elif op == '?':
right_expr = _js_ternary(left_val, *self._separate(right_expr, ':', 1)) right_expr = _js_ternary(left_val, *self._separate(right_expr, ':', 1))
right_val = self.interpret_expression(right_expr, local_vars, allow_recursion) if right_expr else left_val right_val = self.interpret_expression(right_expr, local_vars, allow_recursion)
opfunc = op and next((v for k, v in self._all_operators() if k == op), None) opfunc = op and next((v for k, v in self._all_operators() if k == op), None)
if not opfunc: if not opfunc:
return right_val return right_val
@ -824,9 +707,51 @@ class JSInterpreter(object):
_FINALLY_RE = re.compile(r'finally\s*\{') _FINALLY_RE = re.compile(r'finally\s*\{')
_SWITCH_RE = re.compile(r'switch\s*\(') _SWITCH_RE = re.compile(r'switch\s*\(')
def _eval_operator(self, op, left_expr, right_expr, expr, local_vars, allow_recursion): def handle_operators(self, expr, local_vars, allow_recursion):
left_val = self.interpret_expression(left_expr, local_vars, allow_recursion)
return self._operator(op, left_val, right_expr, expr, local_vars, allow_recursion) for op, _ in self._all_operators():
# hackety: </> have higher priority than <</>>, but don't confuse them
skip_delim = (op + op) if op in '<>*?' else None
if op == '?':
skip_delim = (skip_delim, '?.')
separated = list(self._separate(expr, op, skip_delims=skip_delim))
if len(separated) < 2:
continue
right_expr = separated.pop()
# handle operators that are both unary and binary, minimal BODMAS
if op in ('+', '-'):
# simplify/adjust consecutive instances of these operators
undone = 0
separated = [s.strip() for s in separated]
while len(separated) > 1 and not separated[-1]:
undone += 1
separated.pop()
if op == '-' and undone % 2 != 0:
right_expr = op + right_expr
elif op == '+':
while len(separated) > 1 and set(separated[-1]) <= self.OP_CHARS:
right_expr = separated.pop() + right_expr
if separated[-1][-1:] in self.OP_CHARS:
right_expr = separated.pop() + right_expr
# hanging op at end of left => unary + (strip) or - (push right)
left_val = separated[-1] if separated else ''
for dm_op in ('*', '%', '/', '**'):
bodmas = tuple(self._separate(left_val, dm_op, skip_delims=skip_delim))
if len(bodmas) > 1 and not bodmas[-1].strip():
expr = op.join(separated) + op + right_expr
if len(separated) > 1:
separated.pop()
right_expr = op.join((left_val, right_expr))
else:
separated = [op.join((left_val, right_expr))]
right_expr = None
break
if right_expr is None:
continue
left_val = self.interpret_expression(op.join(separated), local_vars, allow_recursion)
return self._operator(op, left_val, right_expr, expr, local_vars, allow_recursion), True
@Debugger.wrap_interpreter @Debugger.wrap_interpreter
def interpret_statement(self, stmt, local_vars, allow_recursion=100): def interpret_statement(self, stmt, local_vars, allow_recursion=100):
@ -882,19 +807,15 @@ class JSInterpreter(object):
else: else:
raise self.Exception('Unsupported object {obj:.100}'.format(**locals()), expr=expr) raise self.Exception('Unsupported object {obj:.100}'.format(**locals()), expr=expr)
# apply unary operators (see new above)
for op, _ in _UNARY_OPERATORS_X: for op, _ in _UNARY_OPERATORS_X:
if not expr.startswith(op): if not expr.startswith(op):
continue continue
operand = expr[len(op):] operand = expr[len(op):]
if not operand or (op.isalpha() and operand[0] != ' '): if not operand or operand[0] != ' ':
continue continue
separated = self._separate_at_op(operand, max_split=1) op_result = self.handle_operators(expr, local_vars, allow_recursion)
if separated: if op_result:
next_op, separated, right_expr = separated return op_result[0], should_return
separated.append(right_expr)
operand = next_op.join(separated)
return self._eval_operator(op, operand, '', expr, local_vars, allow_recursion), should_return
if expr.startswith('{'): if expr.startswith('{'):
inner, outer = self._separate_at_paren(expr) inner, outer = self._separate_at_paren(expr)
@ -1089,18 +1010,15 @@ class JSInterpreter(object):
m = re.match(r'''(?x) m = re.match(r'''(?x)
(?P<assign> (?P<assign>
(?P<out>{_NAME_RE})(?P<out_idx>(?:\[{_NESTED_BRACKETS}\])+)?\s* (?P<out>{_NAME_RE})(?:\[(?P<out_idx>(?:.+?\]\s*\[)*.+?)\])?\s*
(?P<op>{_OPERATOR_RE})? (?P<op>{_OPERATOR_RE})?
=(?!=)(?P<expr>.*)$ =(?!=)(?P<expr>.*)$
)|(?P<return> )|(?P<return>
(?!if|return|true|false|null|undefined|NaN|Infinity)(?P<name>{_NAME_RE})$ (?!if|return|true|false|null|undefined|NaN|Infinity)(?P<name>{_NAME_RE})$
)|(?P<attribute>
(?P<var>{_NAME_RE})(?:
(?P<nullish>\?)?\.(?P<member>[^(]+)|
\[(?P<member2>{_NESTED_BRACKETS})\]
)\s*
)|(?P<indexing> )|(?P<indexing>
(?P<in>{_NAME_RE})(?P<in_idx>\[.+\])$ (?P<in>{_NAME_RE})\[(?P<in_idx>(?:.+?\]\s*\[)*.+?)\]$
)|(?P<attribute>
(?P<var>{_NAME_RE})(?:(?P<nullish>\?)?\.(?P<member>[^(]+)|\[(?P<member2>[^\]]+)\])\s*
)|(?P<function> )|(?P<function>
(?P<fname>{_NAME_RE})\((?P<args>.*)\)$ (?P<fname>{_NAME_RE})\((?P<args>.*)\)$
)'''.format(**globals()), expr) )'''.format(**globals()), expr)
@ -1115,11 +1033,10 @@ class JSInterpreter(object):
elif left_val in (None, JS_Undefined): elif left_val in (None, JS_Undefined):
raise self.Exception('Cannot index undefined variable ' + m.group('out'), expr=expr) raise self.Exception('Cannot index undefined variable ' + m.group('out'), expr=expr)
indexes = md['out_idx'] indexes = re.split(r'\]\s*\[', m.group('out_idx'))
while indexes: for i, idx in enumerate(indexes, 1):
idx, indexes = self._separate_at_paren(indexes)
idx = self.interpret_expression(idx, local_vars, allow_recursion) idx = self.interpret_expression(idx, local_vars, allow_recursion)
if indexes: if i < len(indexes):
left_val = self._index(left_val, idx) left_val = self._index(left_val, idx)
if isinstance(idx, float): if isinstance(idx, float):
idx = int(idx) idx = int(idx)
@ -1164,17 +1081,14 @@ class JSInterpreter(object):
if md.get('indexing'): if md.get('indexing'):
val = local_vars[m.group('in')] val = local_vars[m.group('in')]
indexes = m.group('in_idx') for idx in re.split(r'\]\s*\[', m.group('in_idx')):
while indexes:
idx, indexes = self._separate_at_paren(indexes)
idx = self.interpret_expression(idx, local_vars, allow_recursion) idx = self.interpret_expression(idx, local_vars, allow_recursion)
val = self._index(val, idx) val = self._index(val, idx)
return val, should_return return val, should_return
separated = self._separate_at_op(expr) op_result = self.handle_operators(expr, local_vars, allow_recursion)
if separated: if op_result:
op, separated, right_expr = separated return op_result[0], should_return
return self._eval_operator(op, op.join(separated), right_expr, expr, local_vars, allow_recursion), should_return
if md.get('attribute'): if md.get('attribute'):
variable, member, nullish = m.group('var', 'member', 'nullish') variable, member, nullish = m.group('var', 'member', 'nullish')
@ -1195,15 +1109,13 @@ class JSInterpreter(object):
def eval_method(variable, member): def eval_method(variable, member):
if (variable, member) == ('console', 'debug'): if (variable, member) == ('console', 'debug'):
if Debugger.ENABLED: if Debugger.ENABLED:
Debugger.write(self.interpret_expression('[{0}]'.format(arg_str), local_vars, allow_recursion)) Debugger.write(self.interpret_expression('[{}]'.format(arg_str), local_vars, allow_recursion))
return return
types = { types = {
'String': compat_str, 'String': compat_str,
'Math': float, 'Math': float,
'Array': list, 'Array': list,
'Date': self.JS_Date, 'Date': self.JS_Date,
'RegExp': self.JS_RegExp,
# 'Error': self.Exception, # has no std static methods
} }
obj = local_vars.get(variable) obj = local_vars.get(variable)
if obj in (JS_Undefined, None): if obj in (JS_Undefined, None):
@ -1211,7 +1123,7 @@ class JSInterpreter(object):
if obj is JS_Undefined: if obj is JS_Undefined:
try: try:
if variable not in self._objects: if variable not in self._objects:
self._objects[variable] = self.extract_object(variable, local_vars) self._objects[variable] = self.extract_object(variable)
obj = self._objects[variable] obj = self._objects[variable]
except self.Exception: except self.Exception:
if not nullish: if not nullish:
@ -1222,7 +1134,7 @@ class JSInterpreter(object):
# Member access # Member access
if arg_str is None: if arg_str is None:
return self._index(obj, member, nullish) return self._index(obj, member)
# Function call # Function call
argvals = [ argvals = [
@ -1365,8 +1277,7 @@ class JSInterpreter(object):
assertion(len(argvals) == 2, 'takes exactly two arguments') assertion(len(argvals) == 2, 'takes exactly two arguments')
# TODO: argvals[1] callable, other Py vs JS edge cases # TODO: argvals[1] callable, other Py vs JS edge cases
if isinstance(argvals[0], self.JS_RegExp): if isinstance(argvals[0], self.JS_RegExp):
# access JS member with Py reserved name count = 0 if argvals[0].flags & self.JS_RegExp.RE_FLAGS['g'] else 1
count = 0 if self._index(argvals[0], 'global') else 1
assertion(member != 'replaceAll' or count == 0, assertion(member != 'replaceAll' or count == 0,
'replaceAll must be called with a global RegExp') 'replaceAll must be called with a global RegExp')
return argvals[0].sub(argvals[1], obj, count=count) return argvals[0].sub(argvals[1], obj, count=count)
@ -1407,7 +1318,7 @@ class JSInterpreter(object):
for v in self._separate(list_txt): for v in self._separate(list_txt):
yield self.interpret_expression(v, local_vars, allow_recursion) yield self.interpret_expression(v, local_vars, allow_recursion)
def extract_object(self, objname, *global_stack): def extract_object(self, objname):
_FUNC_NAME_RE = r'''(?:{n}|"{n}"|'{n}')'''.format(n=_NAME_RE) _FUNC_NAME_RE = r'''(?:{n}|"{n}"|'{n}')'''.format(n=_NAME_RE)
obj = {} obj = {}
fields = next(filter(None, ( fields = next(filter(None, (
@ -1428,8 +1339,7 @@ class JSInterpreter(object):
fields): fields):
argnames = self.build_arglist(f.group('args')) argnames = self.build_arglist(f.group('args'))
name = remove_quotes(f.group('key')) name = remove_quotes(f.group('key'))
obj[name] = function_with_repr( obj[name] = function_with_repr(self.build_function(argnames, f.group('code')), 'F<{0}>'.format(name))
self.build_function(argnames, f.group('code'), *global_stack), 'F<{0}>'.format(name))
return obj return obj

View File

@ -4204,16 +4204,12 @@ def lowercase_escape(s):
s) s)
def escape_rfc3986(s, safe=None): def escape_rfc3986(s):
"""Escape non-ASCII characters as suggested by RFC 3986""" """Escape non-ASCII characters as suggested by RFC 3986"""
if sys.version_info < (3, 0): if sys.version_info < (3, 0):
s = _encode_compat_str(s, 'utf-8') s = _encode_compat_str(s, 'utf-8')
if safe is not None:
safe = _encode_compat_str(safe, 'utf-8')
if safe is None:
safe = b"%/;:@&=+$,!~*'()?#[]"
# ensure unicode: after quoting, it can always be converted # ensure unicode: after quoting, it can always be converted
return compat_str(compat_urllib_parse.quote(s, safe)) return compat_str(compat_urllib_parse.quote(s, b"%/;:@&=+$,!~*'()?#[]"))
def escape_url(url): def escape_url(url):

View File

@ -1,3 +1,3 @@
from __future__ import unicode_literals from __future__ import unicode_literals
__version__ = '2025.04.07' __version__ = '2021.12.17'