Compare commits

...

8 Commits

Author SHA1 Message Date
marieell
3247ee621b
Merge 48d0994b5b into 4d05f84325 2024-06-27 06:36:12 +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
marieell
48d0994b5b [tube8] Fix extractor a bit, replace broken test 2022-02-15 12:22:58 +01:00
marieell
7561727db6 Remove extractors for dead pages
They all redirect to youporn.com.
2022-02-15 09:41:32 +01:00
12 changed files with 190 additions and 334 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

@ -371,7 +371,6 @@ from .esri import EsriVideoIE
from .europa import EuropaIE
from .expotv import ExpoTVIE
from .expressen import ExpressenIE
from .extremetube import ExtremeTubeIE
from .eyedotv import EyedoTVIE
from .facebook import (
FacebookIE,
@ -562,7 +561,6 @@ from .kaltura import KalturaIE
from .kankan import KankanIE
from .karaoketv import KaraoketvIE
from .karrierevideos import KarriereVideosIE
from .keezmovies import KeezMoviesIE
from .ketnet import KetnetIE
from .khanacademy import (
KhanAcademyIE,
@ -713,10 +711,6 @@ from .mlb import (
)
from .mnet import MnetIE
from .moevideo import MoeVideoIE
from .mofosex import (
MofosexIE,
MofosexEmbedIE,
)
from .mojvideo import MojvideoIE
from .morningstar import MorningstarIE
from .motherless import (

View File

@ -1,50 +0,0 @@
from __future__ import unicode_literals
from ..utils import str_to_int
from .keezmovies import KeezMoviesIE
class ExtremeTubeIE(KeezMoviesIE):
_VALID_URL = r'https?://(?:www\.)?extremetube\.com/(?:[^/]+/)?video/(?P<id>[^/#?&]+)'
_TESTS = [{
'url': 'http://www.extremetube.com/video/music-video-14-british-euro-brit-european-cumshots-swallow-652431',
'md5': '92feaafa4b58e82f261e5419f39c60cb',
'info_dict': {
'id': 'music-video-14-british-euro-brit-european-cumshots-swallow-652431',
'ext': 'mp4',
'title': 'Music Video 14 british euro brit european cumshots swallow',
'uploader': 'anonim',
'view_count': int,
'age_limit': 18,
}
}, {
'url': 'http://www.extremetube.com/gay/video/abcde-1234',
'only_matching': True,
}, {
'url': 'http://www.extremetube.com/video/latina-slut-fucked-by-fat-black-dick',
'only_matching': True,
}, {
'url': 'http://www.extremetube.com/video/652431',
'only_matching': True,
}]
def _real_extract(self, url):
webpage, info = self._extract_info(url)
if not info['title']:
info['title'] = self._search_regex(
r'<h1[^>]+title="([^"]+)"[^>]*>', webpage, 'title')
uploader = self._html_search_regex(
r'Uploaded by:\s*</[^>]+>\s*<a[^>]+>(.+?)</a>',
webpage, 'uploader', fatal=False)
view_count = str_to_int(self._search_regex(
r'Views:\s*</[^>]+>\s*<[^>]+>([\d,\.]+)</',
webpage, 'view count', fatal=False))
info.update({
'uploader': uploader,
'view_count': view_count,
})
return info

View File

@ -66,7 +66,6 @@ from .tnaflix import TNAFlixNetworkEmbedIE
from .drtuber import DrTuberIE
from .redtube import RedTubeIE
from .tube8 import Tube8IE
from .mofosex import MofosexEmbedIE
from .spankwire import SpankwireIE
from .youporn import YouPornIE
from .vimeo import (
@ -3044,11 +3043,6 @@ class GenericIE(InfoExtractor):
if tube8_urls:
return self.playlist_from_matches(tube8_urls, video_id, video_title, ie=Tube8IE.ie_key())
# Look for embedded Mofosex player
mofosex_urls = MofosexEmbedIE._extract_urls(webpage)
if mofosex_urls:
return self.playlist_from_matches(mofosex_urls, video_id, video_title, ie=MofosexEmbedIE.ie_key())
# Look for embedded Spankwire player
spankwire_urls = SpankwireIE._extract_urls(webpage)
if spankwire_urls:

View File

@ -1,133 +0,0 @@
from __future__ import unicode_literals
import re
from .common import InfoExtractor
from ..aes import aes_decrypt_text
from ..compat import compat_urllib_parse_unquote
from ..utils import (
determine_ext,
ExtractorError,
int_or_none,
str_to_int,
strip_or_none,
url_or_none,
)
class KeezMoviesIE(InfoExtractor):
_VALID_URL = r'https?://(?:www\.)?keezmovies\.com/video/(?:(?P<display_id>[^/]+)-)?(?P<id>\d+)'
_TESTS = [{
'url': 'https://www.keezmovies.com/video/arab-wife-want-it-so-bad-i-see-she-thirsty-and-has-tiny-money-18070681',
'md5': '2ac69cdb882055f71d82db4311732a1a',
'info_dict': {
'id': '18070681',
'display_id': 'arab-wife-want-it-so-bad-i-see-she-thirsty-and-has-tiny-money',
'ext': 'mp4',
'title': 'Arab wife want it so bad I see she thirsty and has tiny money.',
'thumbnail': None,
'view_count': int,
'age_limit': 18,
}
}, {
'url': 'http://www.keezmovies.com/video/18070681',
'only_matching': True,
}]
def _extract_info(self, url, fatal=True):
mobj = re.match(self._VALID_URL, url)
video_id = mobj.group('id')
display_id = (mobj.group('display_id')
if 'display_id' in mobj.groupdict()
else None) or mobj.group('id')
webpage = self._download_webpage(
url, display_id, headers={'Cookie': 'age_verified=1'})
formats = []
format_urls = set()
title = None
thumbnail = None
duration = None
encrypted = False
def extract_format(format_url, height=None):
format_url = url_or_none(format_url)
if not format_url or not format_url.startswith(('http', '//')):
return
if format_url in format_urls:
return
format_urls.add(format_url)
tbr = int_or_none(self._search_regex(
r'[/_](\d+)[kK][/_]', format_url, 'tbr', default=None))
if not height:
height = int_or_none(self._search_regex(
r'[/_](\d+)[pP][/_]', format_url, 'height', default=None))
if encrypted:
format_url = aes_decrypt_text(
video_url, title, 32).decode('utf-8')
formats.append({
'url': format_url,
'format_id': '%dp' % height if height else None,
'height': height,
'tbr': tbr,
})
flashvars = self._parse_json(
self._search_regex(
r'flashvars\s*=\s*({.+?});', webpage,
'flashvars', default='{}'),
display_id, fatal=False)
if flashvars:
title = flashvars.get('video_title')
thumbnail = flashvars.get('image_url')
duration = int_or_none(flashvars.get('video_duration'))
encrypted = flashvars.get('encrypted') is True
for key, value in flashvars.items():
mobj = re.search(r'quality_(\d+)[pP]', key)
if mobj:
extract_format(value, int(mobj.group(1)))
video_url = flashvars.get('video_url')
if video_url and determine_ext(video_url, None):
extract_format(video_url)
video_url = self._html_search_regex(
r'flashvars\.video_url\s*=\s*(["\'])(?P<url>http.+?)\1',
webpage, 'video url', default=None, group='url')
if video_url:
extract_format(compat_urllib_parse_unquote(video_url))
if not formats:
if 'title="This video is no longer available"' in webpage:
raise ExtractorError(
'Video %s is no longer available' % video_id, expected=True)
try:
self._sort_formats(formats)
except ExtractorError:
if fatal:
raise
if not title:
title = self._html_search_regex(
r'<h1[^>]*>([^<]+)', webpage, 'title')
return webpage, {
'id': video_id,
'display_id': display_id,
'title': strip_or_none(title),
'thumbnail': thumbnail,
'duration': duration,
'age_limit': 18,
'formats': formats,
}
def _real_extract(self, url):
webpage, info = self._extract_info(url, fatal=False)
if not info['formats']:
return self.url_result(url, 'Generic')
info['view_count'] = str_to_int(self._search_regex(
r'<b>([\d,.]+)</b> Views?', webpage, 'view count', fatal=False))
return info

View File

@ -1,79 +0,0 @@
from __future__ import unicode_literals
import re
from .common import InfoExtractor
from ..utils import (
int_or_none,
str_to_int,
unified_strdate,
)
from .keezmovies import KeezMoviesIE
class MofosexIE(KeezMoviesIE):
_VALID_URL = r'https?://(?:www\.)?mofosex\.com/videos/(?P<id>\d+)/(?P<display_id>[^/?#&.]+)\.html'
_TESTS = [{
'url': 'http://www.mofosex.com/videos/318131/amateur-teen-playing-and-masturbating-318131.html',
'md5': '558fcdafbb63a87c019218d6e49daf8a',
'info_dict': {
'id': '318131',
'display_id': 'amateur-teen-playing-and-masturbating-318131',
'ext': 'mp4',
'title': 'amateur teen playing and masturbating',
'thumbnail': r're:^https?://.*\.jpg$',
'upload_date': '20121114',
'view_count': int,
'like_count': int,
'dislike_count': int,
'age_limit': 18,
}
}, {
# This video is no longer available
'url': 'http://www.mofosex.com/videos/5018/japanese-teen-music-video.html',
'only_matching': True,
}]
def _real_extract(self, url):
webpage, info = self._extract_info(url)
view_count = str_to_int(self._search_regex(
r'VIEWS:</span>\s*([\d,.]+)', webpage, 'view count', fatal=False))
like_count = int_or_none(self._search_regex(
r'id=["\']amountLikes["\'][^>]*>(\d+)', webpage,
'like count', fatal=False))
dislike_count = int_or_none(self._search_regex(
r'id=["\']amountDislikes["\'][^>]*>(\d+)', webpage,
'like count', fatal=False))
upload_date = unified_strdate(self._html_search_regex(
r'Added:</span>([^<]+)', webpage, 'upload date', fatal=False))
info.update({
'view_count': view_count,
'like_count': like_count,
'dislike_count': dislike_count,
'upload_date': upload_date,
'thumbnail': self._og_search_thumbnail(webpage),
})
return info
class MofosexEmbedIE(InfoExtractor):
_VALID_URL = r'https?://(?:www\.)?mofosex\.com/embed/?\?.*?\bvideoid=(?P<id>\d+)'
_TESTS = [{
'url': 'https://www.mofosex.com/embed/?videoid=318131&referrer=KM',
'only_matching': True,
}]
@staticmethod
def _extract_urls(webpage):
return re.findall(
r'<iframe[^>]+\bsrc=["\']((?:https?:)?//(?:www\.)?mofosex\.com/embed/?\?.*?\bvideoid=\d+)',
webpage)
def _real_extract(self, url):
video_id = self._match_id(url)
return self.url_result(
'http://www.mofosex.com/videos/{0}/{0}.html'.format(video_id),
ie=MofosexIE.ie_key(), video_id=video_id)

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

@ -3,32 +3,33 @@ from __future__ import unicode_literals
import re
from ..utils import (
determine_ext,
ExtractorError,
int_or_none,
str_to_int,
url_or_none,
)
from .keezmovies import KeezMoviesIE
from .common import InfoExtractor
from ..aes import aes_decrypt_text
from ..compat import compat_urllib_parse_unquote
class Tube8IE(KeezMoviesIE):
class Tube8IE(InfoExtractor):
_VALID_URL = r'https?://(?:www\.)?tube8\.com/(?:[^/]+/)+(?P<display_id>[^/]+)/(?P<id>\d+)'
_TESTS = [{
'url': 'http://www.tube8.com/teen/kasia-music-video/229795/',
'md5': '65e20c48e6abff62ed0c3965fff13a39',
'url': 'https://www.tube8.com/erotic/playtime/81807731/',
'md5': 'fefa69ff76debaa63aa59374bfc51c95',
'info_dict': {
'id': '229795',
'display_id': 'kasia-music-video',
'id': '81807731',
'display_id': 'playtime',
'ext': 'mp4',
'description': 'hot teen Kasia grinding',
'uploader': 'unknown',
'title': 'Kasia music video',
'uploader': 'kikkijay-ph',
'title': 'Playtime',
'age_limit': 18,
'duration': 230,
'categories': ['Teen'],
'tags': ['dancing'],
'duration': 988,
'categories': ['Erotic'],
'tags': ['adult toys', 'big boobs', 'butt', 'masturbate'],
},
}, {
'url': 'http://www.tube8.com/shemale/teen/blonde-cd-gets-kidnapped-by-two-blacks-and-punished-for-being-a-slutty-girl/19569151/',
'only_matching': True,
}]
@staticmethod
@ -38,49 +39,121 @@ class Tube8IE(KeezMoviesIE):
webpage)
def _real_extract(self, url):
webpage, info = self._extract_info(url)
mobj = re.match(self._VALID_URL, url)
video_id = mobj.group('id')
display_id = mobj.group('display_id')
if not info['title']:
info['title'] = self._html_search_regex(
r'videoTitle\s*=\s*"([^"]+)', webpage, 'title')
webpage = self._download_webpage(
url, display_id, headers={'Cookie': 'age_verified=1'})
description = self._html_search_regex(
r'(?s)Description:</dt>\s*<dd>(.+?)</dd>', webpage, 'description', fatal=False)
uploader = self._html_search_regex(
r'<span class="username">\s*(.+?)\s*<',
webpage, 'uploader', fatal=False)
formats = []
format_urls = set()
title = None
thumbnail = None
duration = None
encrypted = False
def extract_format(format_url, height=None):
format_url = url_or_none(format_url)
if not format_url or not format_url.startswith(('http', '//')):
return
if format_url in format_urls:
return
format_urls.add(format_url)
tbr = int_or_none(self._search_regex(
r'[/_](\d+)[kK][/_]', format_url, 'tbr', default=None))
if not height:
height = int_or_none(self._search_regex(
r'[/_](\d+)[pP][/_]', format_url, 'height', default=None))
if encrypted:
format_url = aes_decrypt_text(
video_url, title, 32).decode('utf-8')
formats.append({
'url': format_url,
'format_id': '%dp' % height if height else None,
'height': height,
'tbr': tbr,
})
flashvars = self._parse_json(
self._search_regex(
r'flashvars\s*=\s*({.+?});', webpage,
'flashvars', default='{}'),
display_id, fatal=False)
if flashvars:
title = flashvars.get('video_title')
thumbnail = flashvars.get('image_url')
duration = int_or_none(flashvars.get('video_duration'))
encrypted = flashvars.get('encrypted') is True
uploader = flashvars.get('sponsor')
for key, value in flashvars.items():
mobj = re.search(r'quality_(\d+)[pP]', key)
if mobj:
extract_format(value, int(mobj.group(1)))
video_url = flashvars.get('video_url')
for key, value in flashvars.items():
mobj = re.search(r'quality_(\d+)[pP]', key)
if mobj:
extract_format(value, int(mobj.group(1)))
video_url = flashvars.get('video_url')
if video_url and determine_ext(video_url, None):
extract_format(video_url)
video_url = self._html_search_regex(
r'flashvars\.video_url\s*=\s*(["\'])(?P<url>http.+?)\1',
webpage, 'video url', default=None, group='url')
if video_url:
extract_format(compat_urllib_parse_unquote(video_url))
if not formats:
if 'title="This video is no longer available"' in webpage:
raise ExtractorError(
'Video %s is no longer available' % video_id, expected=True)
self._sort_formats(formats)
if not title:
title = self._html_search_regex([
r'<h1[^>]*>([^<]+)'
r'videoTitle\s*=\s*"([^"]+)'], webpage, 'title', default=display_id.capitalize())
like_count = int_or_none(self._search_regex(
r'rupVar\s*=\s*"(\d+)"', webpage, 'like count', fatal=False))
r'rupVar\s*=\s*(\d+);', webpage, 'like count', fatal=False))
dislike_count = int_or_none(self._search_regex(
r'rdownVar\s*=\s*"(\d+)"', webpage, 'dislike count', fatal=False))
r'rdownVar\s*=\s*(\d+);', webpage, 'dislike count', fatal=False))
view_count = str_to_int(self._search_regex(
r'Views:\s*</dt>\s*<dd>([\d,\.]+)',
webpage, 'view count', fatal=False))
comment_count = str_to_int(self._search_regex(
r'<span id="allCommentsCount">(\d+)</span>',
r'<span id="allCommentsCount">\((\d+)\)</span>',
webpage, 'comment count', fatal=False))
category = self._search_regex(
r'Category:\s*</dt>\s*<dd>\s*<a[^>]+href=[^>]+>([^<]+)',
r'videoCategoryByName\s*=\s*"([^"]+)";',
webpage, 'category', fatal=False)
categories = [category] if category else None
tags_str = self._search_regex(
r'(?s)Tags:\s*</dt>\s*<dd>(.+?)</(?!a)',
r"(?s)<li\s+class\s*=\s*'video-tag'\s+data-esp-node\s*=\s*'tag'>(.*?)</div>",
webpage, 'tags', fatal=False)
tags = [t for t in re.findall(
r'<a[^>]+href=[^>]+>([^<]+)', tags_str)] if tags_str else None
info.update({
'description': description,
return {
'uploader': uploader,
'id': video_id,
'display_id': display_id,
'title': title,
'thumbnail': thumbnail,
'duration': duration,
'age_limit': 18,
'formats': formats,
'view_count': view_count,
'like_count': like_count,
'dislike_count': dislike_count,
'comment_count': comment_count,
'categories': categories,
'tags': tags,
})
return info
}

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: