Compare commits

...

24 Commits

Author SHA1 Message Date
wfaulk
b77b369fda
Merge 92d09b249d into 4d05f84325 2024-06-27 06:36:58 +08:00
dirkf
4d05f84325 [PalcoMP3] Conform to new linter rule
* no space after @ in decorator
2024-06-20 20:03:49 +01:00
dirkf
e0094e63c3 [jsinterp] Various tweaks
* treat Infinity like NaN
* cache operator list
2024-06-20 20:03:49 +01:00
dirkf
fd8242e3ef [jsinterp] Fix and improve expression parsing
* improve BODMAS (fixes https://github.com/ytdl-org/youtube-dl/issues/32815)
* support more weird expressions with multiple unary ops
2024-06-20 20:03:49 +01:00
dirkf
ad01fa6cca [jsinterp] Add Debugger from yt-dlp
* https://github.com/yt-dlp/yt-dlp/commit/8f53dc4
* thx pukkandan
2024-06-20 20:03:49 +01:00
dirkf
2eac0fa379 [utils] Save orig_msg in ExtractorError 2024-06-20 20:03:49 +01:00
wfaulk
92d09b249d
Update youtube_dl/extractor/tubitv.py
Co-authored-by: dirkf <fieldhouse@gmx.net>
2023-05-11 14:13:50 -04:00
wfaulk
0f8938bd1e
Update youtube_dl/extractor/tubitv.py
Co-authored-by: dirkf <fieldhouse@gmx.net>
2023-05-11 14:13:38 -04:00
dirkf
7607aa7589
Merge branch 'master' into tubitv-live 2023-04-06 00:05:56 +01:00
William Faulk
fab9851619 better test for description 2023-02-13 09:07:53 -05:00
William Faulk
83a70160ae Add live stream support for TubiTv 2023-02-11 23:01:57 -05:00
fonkap
70a4a8b752
[StreamsbIE] Add extractor for streamsb.com (viewsb.com) (#31517)
* Add extractor for streamsb.com (viewsb.com)

* make data url using app.js version

---------

Co-authored-by: dirkf <fieldhouse@gmx.net>
2023-02-11 02:54:45 +00:00
fonkap
99c3cf6419
[KommunetvIE] Add extractor for kommunetv.no (#31516)
* Add extractor for kommunetv.no
* Using utils.update_url instead of regex

---------

Co-authored-by: dirkf <fieldhouse@gmx.net>
2023-02-11 02:47:43 +00:00
fonkap
cf5fb7e139
[FileMoonIE] Add extractor for filemoon.sx (#31515)
---------

Co-authored-by: dirkf <fieldhouse@gmx.net>
2023-02-11 02:37:45 +00:00
teddy171
9256b026d0
[feat]: Add support to external downloader aria2p (#31500)
* feat: add class Aria2pFD

* feat: create call_downloader function

* feat: a colorful download interface to aria2pFD

* feat: change value name

* Apply suggestions from code review

Co-authored-by: dirkf <fieldhouse@gmx.net>

* Typo in suggestion

* fix: remove unused value

* fix: add not function to return value(0 is normal); add total_seconds to download.eta(timedelta object); add waiting status when hook progress

* fix: remove unuse method ..utils.format_bytes

* fix: be up to flake8

* fix: be up to flake8

* Apply suggestions from code review

* [feat] test external downloader aria2p

* [feat] test external downloader aria2p

* [fix] test_external_downloader.py

* Apply suggestions from code review

Co-authored-by: dirkf <fieldhouse@gmx.net>

* Apply suggestions from code review

Co-authored-by: dirkf <fieldhouse@gmx.net>

* Update test/test_external_downloader.py

Co-authored-by: dirkf <fieldhouse@gmx.net>

* Update test/test_external_downloader.py

Co-authored-by: dirkf <fieldhouse@gmx.net>

* Update youtube_dl/downloader/external.py

Co-authored-by: dirkf <fieldhouse@gmx.net>

* refactoring code and fix bugs

* Apply suggestions from code review

* Rename test_external_downloader.py to test_downloader_external.py

---------

Co-authored-by: dirkf <fieldhouse@gmx.net>
2023-02-09 20:19:27 +00:00
Valentin Metz
80c0228777
[rbgtum] Add new extractor (#31305)
* [rbgtum] Add new extractor

* Small update, force CI

---------

Co-authored-by: dirkf <fieldhouse@gmx.net>
2023-02-09 11:25:28 +00:00
dirkf
bcf597ea17 [YouTube] Fix tests 2023-02-09 11:01:57 +00:00
dirkf
f640916de1 [YouTube] Refresh compat/utils usage
* import parse_qs()
* import parse_qs in lazy_extractors (clears old TODO)
* clean up old compiled lazy_extractors for Py2
* use update_url()
2023-02-09 11:01:57 +00:00
dirkf
b337af9c62 [compat] Update test_compat
[skip ci]
2023-02-09 11:01:57 +00:00
dirkf
a6f7d10d44 [utils] Add parse_qs, update_url
[skip ci]
2023-02-09 11:01:57 +00:00
dirkf
9ca224b697 [compat] Systematise compat_ naming
[skip ci]
2023-02-09 11:01:57 +00:00
dirkf
ce81ae3846 [test] Fix TestAgeRestriction
* age restriction may cause DownloadError
* update obsolete test URLs
[skip ci]
2023-02-09 11:01:57 +00:00
dirkf
2be0cd2616 [YouTube] Add signatureTimestamp for age-gate bypass 2023-02-09 11:01:57 +00:00
dirkf
b14b33a2e9 [YouTube] Bypass age-gating for certain restricted videos
* Use TVHTML5_SIMPLY_EMBEDDED_PLAYER client

* Also add and fix tests

* Introduce and use new utility function `update_url()`
2023-02-09 11:01:57 +00:00
7 changed files with 101 additions and 49 deletions

View File

@ -577,9 +577,11 @@ class TestJSInterpreter(unittest.TestCase):
def test_unary_operators(self):
jsi = JSInterpreter('function f(){return 2 - - - 2;}')
self.assertEqual(jsi.call_function('f'), 0)
# fails
# jsi = JSInterpreter('function f(){return 2 + - + - - 2;}')
# self.assertEqual(jsi.call_function('f'), 0)
jsi = JSInterpreter('function f(){return 2 + - + - - 2;}')
self.assertEqual(jsi.call_function('f'), 0)
# https://github.com/ytdl-org/youtube-dl/issues/32815
jsi = JSInterpreter('function f(){return 0 - 7 * - 6;}')
self.assertEqual(jsi.call_function('f'), 42)
""" # fails so far
def test_packed(self):

View File

@ -158,6 +158,10 @@ _NSIG_TESTS = [
'https://www.youtube.com/s/player/b7910ca8/player_ias.vflset/en_US/base.js',
'_hXMCwMt9qE310D', 'LoZMgkkofRMCZQ',
),
(
'https://www.youtube.com/s/player/590f65a6/player_ias.vflset/en_US/base.js',
'1tm7-g_A9zsI8_Lay_', 'xI4Vem4Put_rOg',
),
]

View File

@ -3033,7 +3033,6 @@ class InfoExtractor(object):
transform_source=transform_source, default=None)
def _extract_jwplayer_data(self, webpage, video_id, *args, **kwargs):
# allow passing `transform_source` through to _find_jwplayer_data()
transform_source = kwargs.pop('transform_source', None)
kwfind = compat_kwargs({'transform_source': transform_source}) if transform_source else {}

View File

@ -8,7 +8,7 @@ from ..compat import compat_str
from ..utils import (
int_or_none,
str_or_none,
try_get,
traverse_obj,
)
@ -109,7 +109,7 @@ class PalcoMP3ArtistIE(PalcoMP3BaseIE):
}
name'''
@ classmethod
@classmethod
def suitable(cls, url):
return False if re.match(PalcoMP3IE._VALID_URL, url) else super(PalcoMP3ArtistIE, cls).suitable(url)
@ -118,7 +118,8 @@ class PalcoMP3ArtistIE(PalcoMP3BaseIE):
artist = self._call_api(artist_slug, self._ARTIST_FIELDS_TMPL)['artist']
def entries():
for music in (try_get(artist, lambda x: x['musics']['nodes'], list) or []):
for music in traverse_obj(artist, (
'musics', 'nodes', lambda _, m: m['musicID'])):
yield self._parse_music(music)
return self.playlist_result(
@ -137,7 +138,7 @@ class PalcoMP3VideoIE(PalcoMP3BaseIE):
'title': 'Maiara e Maraisa - Você Faz Falta Aqui - DVD Ao Vivo Em Campo Grande',
'description': 'md5:7043342c09a224598e93546e98e49282',
'upload_date': '20161107',
'uploader_id': 'maiaramaraisaoficial',
'uploader_id': '@maiaramaraisaoficial',
'uploader': 'Maiara e Maraisa',
}
}]

View File

@ -4,6 +4,7 @@ from __future__ import unicode_literals
import re
from .common import InfoExtractor
from ..compat import compat_str
from ..utils import (
ExtractorError,
int_or_none,
@ -13,38 +14,29 @@ from ..utils import (
class TubiTvIE(InfoExtractor):
_VALID_URL = r'https?://(?:www\.)?tubitv\.com/(?:video|movies|tv-shows)/(?P<id>[0-9]+)'
_VALID_URL = r'https?://(?:www\.)?tubitv\.com/(?:video|movies|tv-shows|live)/(?P<id>[0-9]+)'
_LOGIN_URL = 'http://tubitv.com/login'
_NETRC_MACHINE = 'tubitv'
_GEO_COUNTRIES = ['US']
_TESTS = [{
'url': 'http://tubitv.com/video/283829/the_comedian_at_the_friday',
'md5': '43ac06be9326f41912dc64ccf7a80320',
'url': 'https://tubitv.com/movies/229877/carnival-of-souls',
'md5': '275a9f78428cfd6428520989408bcfe9',
'info_dict': {
'id': '283829',
'id': '229877',
'ext': 'mp4',
'title': 'The Comedian at The Friday',
'description': 'A stand up comedian is forced to look at the decisions in his life while on a one week trip to the west coast.',
'uploader_id': 'bc168bee0d18dd1cb3b86c68706ab434',
'title': 'Carnival of Souls',
'description': 'After a traumatic accident, a woman becomes drawn to a mysterious abandoned carnival.',
'uploader_id': 'da2522b1c94be4a578da6ba674925159',
},
}, {
'url': 'http://tubitv.com/tv-shows/321886/s01_e01_on_nom_stories',
'only_matching': True,
}, {
'url': 'http://tubitv.com/movies/383676/tracker',
'only_matching': True,
}, {
'url': 'https://tubitv.com/movies/560057/penitentiary?start=true',
'url': 'https://tubitv.com/live/715943/wb-tv-watchlist',
'info_dict': {
'id': '560057',
'id': '715943',
'ext': 'mp4',
'title': 'Penitentiary',
'description': 'md5:8d2fc793a93cc1575ff426fdcb8dd3f9',
'uploader_id': 'd8fed30d4f24fcb22ec294421b9defc2',
'release_year': 1979,
},
'params': {
'skip_download': True,
'title': 'WB TV Watchlist',
'description': compat_str,
'uploader_id': 'a12923a141282c8b62e593bac425db99',
'is_live': True,
},
}]
@ -76,7 +68,7 @@ class TubiTvIE(InfoExtractor):
title = video_data['title']
formats = self._extract_m3u8_formats(
self._proto_relative_url(video_data['url']),
self._proto_relative_url(traverse_obj(video_data, 'url', ('video_resources', 0, 'manifest', 'url'))),
video_id, 'mp4', 'm3u8_native')
self._sort_formats(formats)
@ -97,6 +89,8 @@ class TubiTvIE(InfoExtractor):
'url': self._proto_relative_url(sub_url),
})
is_live = ('live' in video_data.get('tags') or []) or None
return {
'id': video_id,
'title': title,
@ -107,4 +101,5 @@ class TubiTvIE(InfoExtractor):
'duration': int_or_none(video_data.get('duration')),
'uploader_id': video_data.get('publisher_id'),
'release_year': int_or_none(video_data.get('year')),
'is_live': is_live,
}

View File

@ -14,6 +14,7 @@ from .utils import (
remove_quotes,
unified_timestamp,
variadic,
write_string,
)
from .compat import (
compat_basestring,
@ -53,15 +54,16 @@ def wraps_op(op):
# NB In principle NaN cannot be checked by membership.
# Here all NaN values are actually this one, so _NaN is _NaN,
# although _NaN != _NaN.
# although _NaN != _NaN. Ditto Infinity.
_NaN = float('nan')
_Infinity = float('inf')
def _js_bit_op(op):
def zeroise(x):
return 0 if x in (None, JS_Undefined, _NaN) else x
return 0 if x in (None, JS_Undefined, _NaN, _Infinity) else x
@wraps_op(op)
def wrapped(a, b):
@ -84,7 +86,7 @@ def _js_arith_op(op):
def _js_div(a, b):
if JS_Undefined in (a, b) or not (a or b):
return _NaN
return operator.truediv(a or 0, b) if b else float('inf')
return operator.truediv(a or 0, b) if b else _Infinity
def _js_mod(a, b):
@ -220,6 +222,42 @@ class LocalNameSpace(ChainMap):
return 'LocalNameSpace%s' % (self.maps, )
class Debugger(object):
ENABLED = False
@staticmethod
def write(*args, **kwargs):
level = kwargs.get('level', 100)
def truncate_string(s, left, right=0):
if s is None or len(s) <= left + right:
return s
return '...'.join((s[:left - 3], s[-right:] if right else ''))
write_string('[debug] JS: {0}{1}\n'.format(
' ' * (100 - level),
' '.join(truncate_string(compat_str(x), 50, 50) for x in args)))
@classmethod
def wrap_interpreter(cls, f):
def interpret_statement(self, stmt, local_vars, allow_recursion, *args, **kwargs):
if cls.ENABLED and stmt.strip():
cls.write(stmt, level=allow_recursion)
try:
ret, should_ret = f(self, stmt, local_vars, allow_recursion, *args, **kwargs)
except Exception as e:
if cls.ENABLED:
if isinstance(e, ExtractorError):
e = e.orig_msg
cls.write('=> Raises:', e, '<-|', stmt, level=allow_recursion)
raise
if cls.ENABLED and stmt.strip():
if should_ret or not repr(ret) == stmt:
cls.write(['->', '=>'][should_ret], repr(ret), '<-|', stmt, level=allow_recursion)
return ret, should_ret
return interpret_statement
class JSInterpreter(object):
__named_object_counter = 0
@ -307,8 +345,7 @@ class JSInterpreter(object):
def __op_chars(cls):
op_chars = set(';,[')
for op in cls._all_operators():
for c in op[0]:
op_chars.add(c)
op_chars.update(op[0])
return op_chars
def _named_object(self, namespace, obj):
@ -326,9 +363,8 @@ class JSInterpreter(object):
# collections.Counter() is ~10% slower in both 2.7 and 3.9
counters = dict((k, 0) for k in _MATCHING_PARENS.values())
start, splits, pos, delim_len = 0, 0, 0, len(delim) - 1
in_quote, escaping, skipping = None, False, 0
after_op, in_regex_char_group = True, False
in_quote, escaping, after_op, in_regex_char_group = None, False, True, False
skipping = 0
for idx, char in enumerate(expr):
paren_delta = 0
if not in_quote:
@ -382,10 +418,12 @@ class JSInterpreter(object):
return separated[0][1:].strip(), separated[1].strip()
@staticmethod
def _all_operators():
return itertools.chain(
# Ref: https://developer.mozilla.org/en-US/docs/Web/JavaScript/Reference/Operators/Operator_Precedence
_SC_OPERATORS, _LOG_OPERATORS, _COMP_OPERATORS, _OPERATORS)
def _all_operators(_cached=[]):
if not _cached:
_cached.extend(itertools.chain(
# Ref: https://developer.mozilla.org/en-US/docs/Web/JavaScript/Reference/Operators/Operator_Precedence
_SC_OPERATORS, _LOG_OPERATORS, _COMP_OPERATORS, _OPERATORS))
return _cached
def _operator(self, op, left_val, right_expr, expr, local_vars, allow_recursion):
if op in ('||', '&&'):
@ -416,7 +454,7 @@ class JSInterpreter(object):
except Exception as e:
if allow_undefined:
return JS_Undefined
raise self.Exception('Cannot get index {idx:.100}'.format(**locals()), expr=repr(obj), cause=e)
raise self.Exception('Cannot get index {idx!r:.100}'.format(**locals()), expr=repr(obj), cause=e)
def _dump(self, obj, namespace):
try:
@ -438,6 +476,7 @@ class JSInterpreter(object):
_FINALLY_RE = re.compile(r'finally\s*\{')
_SWITCH_RE = re.compile(r'switch\s*\(')
@Debugger.wrap_interpreter
def interpret_statement(self, stmt, local_vars, allow_recursion=100):
if allow_recursion < 0:
raise self.Exception('Recursion limit reached')
@ -511,7 +550,6 @@ class JSInterpreter(object):
expr = self._dump(inner, local_vars) + outer
if expr.startswith('('):
m = re.match(r'\((?P<d>[a-z])%(?P<e>[a-z])\.length\+(?P=e)\.length\)%(?P=e)\.length', expr)
if m:
# short-cut eval of frequently used `(d%e.length+e.length)%e.length`, worth ~6% on `pytest -k test_nsig`
@ -693,7 +731,7 @@ class JSInterpreter(object):
(?P<op>{_OPERATOR_RE})?
=(?!=)(?P<expr>.*)$
)|(?P<return>
(?!if|return|true|false|null|undefined)(?P<name>{_NAME_RE})$
(?!if|return|true|false|null|undefined|NaN|Infinity)(?P<name>{_NAME_RE})$
)|(?P<indexing>
(?P<in>{_NAME_RE})\[(?P<idx>.+)\]$
)|(?P<attribute>
@ -727,11 +765,12 @@ class JSInterpreter(object):
raise JS_Break()
elif expr == 'continue':
raise JS_Continue()
elif expr == 'undefined':
return JS_Undefined, should_return
elif expr == 'NaN':
return _NaN, should_return
elif expr == 'Infinity':
return _Infinity, should_return
elif md.get('return'):
return local_vars[m.group('name')], should_return
@ -760,18 +799,28 @@ class JSInterpreter(object):
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
while len(separated) > 1 and not separated[-1].strip():
undone += 1
separated.pop()
if op == '-' and undone % 2 != 0:
right_expr = op + right_expr
elif op == '+':
while len(separated) > 1 and separated[-1].strip() 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]
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
right_expr = None
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
@ -797,6 +846,8 @@ class JSInterpreter(object):
def eval_method():
if (variable, member) == ('console', 'debug'):
if Debugger.ENABLED:
Debugger.write(self.interpret_expression('[{}]'.format(arg_str), local_vars, allow_recursion))
return
types = {
'String': compat_str,

View File

@ -2406,7 +2406,7 @@ class ExtractorError(YoutubeDLError):
""" tb, if given, is the original traceback (so that it can be printed out).
If expected is set, this is a normal error message and most likely not a bug in youtube-dl.
"""
self.orig_msg = msg
if sys.exc_info()[0] in (compat_urllib_error.URLError, socket.timeout, UnavailableVideoError):
expected = True
if video_id is not None: