Upgrade yt_dlp and download script

This commit is contained in:
2025-05-02 16:11:08 -05:00
parent 3a2e8eeb08
commit d68d9ce4f9
1194 changed files with 60099 additions and 44436 deletions

View File

@@ -25,7 +25,7 @@ def _js_bit_op(op):
with contextlib.suppress(TypeError):
if math.isnan(x): # NB: NaN cannot be checked by membership
return 0
return x
return int(float(x))
def wrapped(a, b):
return op(zeroise(a), zeroise(b)) & 0xffffffff
@@ -95,6 +95,61 @@ def _js_ternary(cndn, if_true=True, if_false=False):
return if_true
# Ref: https://es5.github.io/#x9.8.1
def js_number_to_string(val: float, radix: int = 10):
if radix in (JS_Undefined, None):
radix = 10
assert radix in range(2, 37), 'radix must be an integer at least 2 and no greater than 36'
if math.isnan(val):
return 'NaN'
if val == 0:
return '0'
if math.isinf(val):
return '-Infinity' if val < 0 else 'Infinity'
if radix == 10:
# TODO: implement special cases
...
ALPHABET = b'0123456789abcdefghijklmnopqrstuvwxyz.-'
result = collections.deque()
sign = val < 0
val = abs(val)
fraction, integer = math.modf(val)
delta = max(math.nextafter(.0, math.inf), math.ulp(val) / 2)
if fraction >= delta:
result.append(-2) # `.`
while fraction >= delta:
delta *= radix
fraction, digit = math.modf(fraction * radix)
result.append(int(digit))
# if we need to round, propagate potential carry through fractional part
needs_rounding = fraction > 0.5 or (fraction == 0.5 and int(digit) & 1)
if needs_rounding and fraction + delta > 1:
for index in reversed(range(1, len(result))):
if result[index] + 1 < radix:
result[index] += 1
break
result.pop()
else:
integer += 1
break
integer, digit = divmod(int(integer), radix)
result.appendleft(digit)
while integer > 0:
integer, digit = divmod(integer, radix)
result.appendleft(digit)
if sign:
result.appendleft(-1) # `-`
return bytes(ALPHABET[digit] for digit in result).decode('ascii')
# Ref: https://developer.mozilla.org/en-US/docs/Web/JavaScript/Reference/Operators/Operator_Precedence
_OPERATORS = { # None => Defined in JSInterpreter._operator
'?': None,
@@ -133,6 +188,7 @@ _COMP_OPERATORS = {'===', '!==', '==', '!=', '<=', '>=', '<', '>'}
_NAME_RE = r'[a-zA-Z_$][\w$]*'
_MATCHING_PARENS = dict(zip(*zip('()', '{}', '[]')))
_QUOTES = '\'"/'
_NESTED_BRACKETS = r'[^[\]]+(?:\[[^[\]]+(?:\[[^\]]+\])?\])?'
class JS_Undefined:
@@ -190,7 +246,7 @@ class Debugger:
cls.write('=> Raises:', e, '<-|', stmt, level=allow_recursion)
raise
if cls.ENABLED and stmt.strip():
if should_ret or not repr(ret) == stmt:
if should_ret or repr(ret) != stmt:
cls.write(['->', '=>'][should_ret], repr(ret), '<-|', stmt, level=allow_recursion)
return ret, should_ret
return interpret_statement
@@ -216,7 +272,7 @@ class JSInterpreter:
self.code, self._functions = code, {}
self._objects = {} if objects is None else objects
class Exception(ExtractorError):
class Exception(ExtractorError): # noqa: A001
def __init__(self, msg, expr=None, *args, **kwargs):
if expr is not None:
msg = f'{msg.rstrip()} in: {truncate_string(expr, 50, 50)}'
@@ -235,7 +291,7 @@ class JSInterpreter:
flags = 0
if not expr:
return flags, expr
for idx, ch in enumerate(expr):
for idx, ch in enumerate(expr): # noqa: B007
if ch not in cls._RE_FLAGS:
break
flags |= cls._RE_FLAGS[ch]
@@ -246,7 +302,7 @@ class JSInterpreter:
OP_CHARS = '+-*/%&|^=<>!,;{}:['
if not expr:
return
counters = {k: 0 for k in _MATCHING_PARENS.values()}
counters = dict.fromkeys(_MATCHING_PARENS.values(), 0)
start, splits, pos, delim_len = 0, 0, 0, len(delim) - 1
in_quote, escaping, after_op, in_regex_char_group = None, False, True, False
for idx, char in enumerate(expr):
@@ -474,7 +530,7 @@ class JSInterpreter:
if remaining.startswith('{'):
body, expr = self._separate_at_paren(remaining)
else:
switch_m = re.match(r'switch\s*\(', remaining) # FIXME
switch_m = re.match(r'switch\s*\(', remaining) # FIXME: ?
if switch_m:
switch_val, remaining = self._separate_at_paren(remaining[switch_m.end() - 1:])
body, expr = self._separate_at_paren(remaining, '}')
@@ -551,15 +607,18 @@ class JSInterpreter:
m = re.match(fr'''(?x)
(?P<assign>
(?P<out>{_NAME_RE})(?:\[(?P<index>[^\]]+?)\])?\s*
(?P<out>{_NAME_RE})(?:\[(?P<index>{_NESTED_BRACKETS})\])?\s*
(?P<op>{"|".join(map(re.escape, set(_OPERATORS) - _COMP_OPERATORS))})?
=(?!=)(?P<expr>.*)$
)|(?P<return>
(?!if|return|true|false|null|undefined|NaN)(?P<name>{_NAME_RE})$
)|(?P<attribute>
(?P<var>{_NAME_RE})(?:
(?P<nullish>\?)?\.(?P<member>[^(]+)|
\[(?P<member2>{_NESTED_BRACKETS})\]
)\s*
)|(?P<indexing>
(?P<in>{_NAME_RE})\[(?P<idx>.+)\]$
)|(?P<attribute>
(?P<var>{_NAME_RE})(?:(?P<nullish>\?)?\.(?P<member>[^(]+)|\[(?P<member2>[^\]]+)\])\s*
)|(?P<function>
(?P<fname>{_NAME_RE})\((?P<args>.*)\)$
)''', expr)
@@ -585,9 +644,9 @@ class JSInterpreter:
return int(expr), should_return
elif expr == 'break':
raise JS_Break()
raise JS_Break
elif expr == 'continue':
raise JS_Continue()
raise JS_Continue
elif expr == 'undefined':
return JS_Undefined, should_return
elif expr == 'NaN':
@@ -636,6 +695,8 @@ class JSInterpreter:
raise self.Exception(f'{member} {msg}', expr)
def eval_method():
nonlocal member
if (variable, member) == ('console', 'debug'):
if Debugger.ENABLED:
Debugger.write(self.interpret_expression(f'[{arg_str}]', local_vars, allow_recursion))
@@ -644,12 +705,13 @@ class JSInterpreter:
types = {
'String': str,
'Math': float,
'Array': list,
}
obj = local_vars.get(variable, types.get(variable, NO_DEFAULT))
if obj is NO_DEFAULT:
if variable not in self._objects:
try:
self._objects[variable] = self.extract_object(variable)
self._objects[variable] = self.extract_object(variable, local_vars)
except self.Exception:
if not nullish:
raise
@@ -667,12 +729,27 @@ class JSInterpreter:
self.interpret_expression(v, local_vars, allow_recursion)
for v in self._separate(arg_str)]
if obj == str:
# Fixup prototype call
if isinstance(obj, type) and member.startswith('prototype.'):
new_member, _, func_prototype = member.partition('.')[2].partition('.')
assertion(argvals, 'takes one or more arguments')
assertion(isinstance(argvals[0], obj), f'needs binding to type {obj}')
if func_prototype == 'call':
obj, *argvals = argvals
elif func_prototype == 'apply':
assertion(len(argvals) == 2, 'takes two arguments')
obj, argvals = argvals
assertion(isinstance(argvals, list), 'second argument needs to be a list')
else:
raise self.Exception(f'Unsupported Function method {func_prototype}', expr)
member = new_member
if obj is str:
if member == 'fromCharCode':
assertion(argvals, 'takes one or more arguments')
return ''.join(map(chr, argvals))
raise self.Exception(f'Unsupported String method {member}', expr)
elif obj == float:
elif obj is float:
if member == 'pow':
assertion(len(argvals) == 2, 'takes two arguments')
return argvals[0] ** argvals[1]
@@ -691,18 +768,18 @@ class JSInterpreter:
obj.reverse()
return obj
elif member == 'slice':
assertion(isinstance(obj, list), 'must be applied on a list')
assertion(len(argvals) == 1, 'takes exactly one argument')
return obj[argvals[0]:]
assertion(isinstance(obj, (list, str)), 'must be applied on a list or string')
assertion(len(argvals) <= 2, 'takes between 0 and 2 arguments')
return obj[slice(*argvals, None)]
elif member == 'splice':
assertion(isinstance(obj, list), 'must be applied on a list')
assertion(argvals, 'takes one or more arguments')
index, howMany = map(int, (argvals + [len(obj)])[:2])
index, how_many = map(int, ([*argvals, len(obj)])[:2])
if index < 0:
index += len(obj)
add_items = argvals[2:]
res = []
for i in range(index, min(index + howMany, len(obj))):
for _ in range(index, min(index + how_many, len(obj))):
res.append(obj.pop(index))
for i, item in enumerate(add_items):
obj.insert(index + i, item)
@@ -726,12 +803,12 @@ class JSInterpreter:
elif member == 'forEach':
assertion(argvals, 'takes one or more arguments')
assertion(len(argvals) <= 2, 'takes at-most 2 arguments')
f, this = (argvals + [''])[:2]
f, this = ([*argvals, ''])[:2]
return [f((item, idx, obj), {'this': this}, allow_recursion) for idx, item in enumerate(obj)]
elif member == 'indexOf':
assertion(argvals, 'takes one or more arguments')
assertion(len(argvals) <= 2, 'takes at-most 2 arguments')
idx, start = (argvals + [0])[:2]
idx, start = ([*argvals, 0])[:2]
try:
return obj.index(idx, start)
except ValueError:
@@ -774,7 +851,7 @@ class JSInterpreter:
raise self.Exception('Cannot return from an expression', expr)
return ret
def extract_object(self, objname):
def extract_object(self, objname, *global_stack):
_FUNC_NAME_RE = r'''(?:[a-zA-Z$0-9]+|"[a-zA-Z$0-9]+"|'[a-zA-Z$0-9]+')'''
obj = {}
obj_m = re.search(
@@ -796,7 +873,8 @@ class JSInterpreter:
for f in fields_m:
argnames = f.group('args').split(',')
name = remove_quotes(f.group('key'))
obj[name] = function_with_repr(self.build_function(argnames, f.group('code')), f'F<{name}>')
obj[name] = function_with_repr(
self.build_function(argnames, f.group('code'), *global_stack), f'F<{name}>')
return obj
@@ -817,9 +895,9 @@ class JSInterpreter:
code, _ = self._separate_at_paren(func_m.group('code'))
return [x.strip() for x in func_m.group('args').split(',')], code
def extract_function(self, funcname):
def extract_function(self, funcname, *global_stack):
return function_with_repr(
self.extract_function_from_code(*self.extract_function_code(funcname)),
self.extract_function_from_code(*self.extract_function_code(funcname), *global_stack),
f'F<{funcname}>')
def extract_function_from_code(self, argnames, code, *global_stack):