Compare commits

...

7 Commits

Author SHA1 Message Date
jastrab
4251693ec6
Merge 98fb8efcb8 into c5098961b0 2024-08-21 22:32:54 -04:00
dirkf
c5098961b0 [Youtube] Rework n function extraction pattern
Now also succeeds with player b12cc44b
2024-08-06 20:59:09 +01:00
dirkf
dbc08fba83 [jsinterp] Improve slice implementation for player b12cc44b
Partly taken from yt-dlp/yt-dlp#10664, thx seproDev
        Fixes #32896
2024-08-06 20:51:38 +01:00
Aiur Adept
71223bff39
[Youtube] Fix nsig extraction for player 20dfca59 (#32891)
* dirkf's patch for nsig extraction
* add generic search per  yt-dlp/yt-dlp/pull/10611 - thx bashonly

---------

Co-authored-by: dirkf <fieldhouse@gmx.net>
2024-08-01 19:18:34 +01:00
Jastrab
98fb8efcb8 [markiza] Fixed extractor 2022-01-16 15:40:59 +01:00
Jastrab
4cf557e765 [markiza] Fixed extractor 2022-01-15 21:27:57 +01:00
Jastrab
63af7465cc [rtvs] Fixed extractor for Slovak television and radio 2022-01-15 21:08:06 +01:00
6 changed files with 243 additions and 54 deletions

View File

@ -425,6 +425,34 @@ class TestJSInterpreter(unittest.TestCase):
self._test(jsi, [''], args=['', '-'])
self._test(jsi, [], args=['', ''])
def test_slice(self):
self._test('function f(){return [0, 1, 2, 3, 4, 5, 6, 7, 8].slice()}', [0, 1, 2, 3, 4, 5, 6, 7, 8])
self._test('function f(){return [0, 1, 2, 3, 4, 5, 6, 7, 8].slice(0)}', [0, 1, 2, 3, 4, 5, 6, 7, 8])
self._test('function f(){return [0, 1, 2, 3, 4, 5, 6, 7, 8].slice(5)}', [5, 6, 7, 8])
self._test('function f(){return [0, 1, 2, 3, 4, 5, 6, 7, 8].slice(99)}', [])
self._test('function f(){return [0, 1, 2, 3, 4, 5, 6, 7, 8].slice(-2)}', [7, 8])
self._test('function f(){return [0, 1, 2, 3, 4, 5, 6, 7, 8].slice(-99)}', [0, 1, 2, 3, 4, 5, 6, 7, 8])
self._test('function f(){return [0, 1, 2, 3, 4, 5, 6, 7, 8].slice(0, 0)}', [])
self._test('function f(){return [0, 1, 2, 3, 4, 5, 6, 7, 8].slice(1, 0)}', [])
self._test('function f(){return [0, 1, 2, 3, 4, 5, 6, 7, 8].slice(0, 1)}', [0])
self._test('function f(){return [0, 1, 2, 3, 4, 5, 6, 7, 8].slice(3, 6)}', [3, 4, 5])
self._test('function f(){return [0, 1, 2, 3, 4, 5, 6, 7, 8].slice(1, -1)}', [1, 2, 3, 4, 5, 6, 7])
self._test('function f(){return [0, 1, 2, 3, 4, 5, 6, 7, 8].slice(-1, 1)}', [])
self._test('function f(){return [0, 1, 2, 3, 4, 5, 6, 7, 8].slice(-3, -1)}', [6, 7])
self._test('function f(){return "012345678".slice()}', '012345678')
self._test('function f(){return "012345678".slice(0)}', '012345678')
self._test('function f(){return "012345678".slice(5)}', '5678')
self._test('function f(){return "012345678".slice(99)}', '')
self._test('function f(){return "012345678".slice(-2)}', '78')
self._test('function f(){return "012345678".slice(-99)}', '012345678')
self._test('function f(){return "012345678".slice(0, 0)}', '')
self._test('function f(){return "012345678".slice(1, 0)}', '')
self._test('function f(){return "012345678".slice(0, 1)}', '0')
self._test('function f(){return "012345678".slice(3, 6)}', '345')
self._test('function f(){return "012345678".slice(1, -1)}', '1234567')
self._test('function f(){return "012345678".slice(-1, 1)}', '')
self._test('function f(){return "012345678".slice(-3, -1)}', '67')
if __name__ == '__main__':
unittest.main()

View File

@ -174,6 +174,14 @@ _NSIG_TESTS = [
'https://www.youtube.com/s/player/5604538d/player_ias.vflset/en_US/base.js',
'7X-he4jjvMx7BCX', 'sViSydX8IHtdWA',
),
(
'https://www.youtube.com/s/player/20dfca59/player_ias.vflset/en_US/base.js',
'-fLCxedkAk4LUTK2', 'O8kfRq1y1eyHGw',
),
(
'https://www.youtube.com/s/player/b12cc44b/player_ias.vflset/en_US/base.js',
'keLa5R2U00sR9SQK', 'N1OGyujjEwMnLw',
),
]

View File

@ -4,18 +4,21 @@ from __future__ import unicode_literals
import re
from .common import InfoExtractor
from ..compat import compat_str
from ..utils import (
orderedSet,
parse_duration,
url_or_none,
determine_ext,
try_get,
compat_str
)
class MarkizaIE(InfoExtractor):
_VALID_URL = r'https?://(?:www\.)?videoarchiv\.markiza\.sk/(?:video/(?:[^/]+/)*|embed/)(?P<id>\d+)(?:[_/]|$)'
_VALID_URL = r'https?://(?:www\.)?videoarchiv\.markiza\.sk/(?:video/(?:[^/]+/)*|embed/)\S+/(?P<id>\d+)(?:[_/-]|$)'
_TESTS = [{
'url': 'http://videoarchiv.markiza.sk/video/oteckovia/84723_oteckovia-109',
'url': 'http://videoarchiv.markiza.sk/video/oteckovia/\
84723_oteckovia-109',
'md5': 'ada4e9fad038abeed971843aa028c7b0',
'info_dict': {
'id': '139078',
@ -26,54 +29,104 @@ class MarkizaIE(InfoExtractor):
'duration': 2760,
},
}, {
'url': 'http://videoarchiv.markiza.sk/video/televizne-noviny/televizne-noviny/85430_televizne-noviny',
'url': 'https://videoarchiv.markiza.sk/video/laska-na-prenajom/epizoda/58779-seria-1-epizoda-14',
'info_dict': {
'id': '85430',
'title': 'Televízne noviny',
},
'playlist_count': 23,
}, {
'url': 'http://videoarchiv.markiza.sk/video/oteckovia/84723',
'url': 'https://videoarchiv.markiza.sk/video/oteckovia/84723',
'only_matching': True,
}, {
'url': 'http://videoarchiv.markiza.sk/video/84723',
'url': 'https://videoarchiv.markiza.sk/video/84723',
'only_matching': True,
}, {
'url': 'http://videoarchiv.markiza.sk/video/filmy/85190_kamenak',
'url': 'https://videoarchiv.markiza.sk/video/filmy/85190_kamenak',
'only_matching': True,
}, {
'url': 'http://videoarchiv.markiza.sk/video/reflex/zo-zakulisia/84651_pribeh-alzbetky',
'url': 'https://videoarchiv.markiza.sk/video/reflex/zo-zakulisia/84651_pribeh-alzbetky',
'only_matching': True,
}, {
'url': 'http://videoarchiv.markiza.sk/embed/85295',
'url': 'https://videoarchiv.markiza.sk/embed/85295',
'only_matching': True,
}]
def _real_extract(self, url):
video_id = self._match_id(url)
data = self._download_json(
'http://videoarchiv.markiza.sk/json/video_jwplayer7.json',
video_id, query={'id': video_id})
webpage = self._download_webpage(url, video_id)
info = self._parse_jwplayer_data(data, m3u8_id='hls', mpd_id='dash')
embed = self._search_regex(
r'<iframe src="(https:\/\/media.*?)" style="',
webpage, 'embed', fatal=False)
if info.get('_type') == 'playlist':
info.update({
'id': video_id,
'title': try_get(
data, lambda x: x['details']['name'], compat_str),
})
else:
info['duration'] = parse_duration(
try_get(data, lambda x: x['details']['duration'], compat_str))
return info
webpage = self._download_webpage(embed, video_id)
data = self._search_regex(
r'processAdTagModifier\s*\(\s*(\{.*)\)\s*,\s*\{\s*"video"\s*:',
webpage, 'embed', default='')
data2 = self._search_regex(
r'processAdTagModifier\s*\(\s*\{.*\)\s*,\s*(\{\s*"video"\s*:.*)\s*\);',
webpage, 'embed', default='')
data = re.sub('\\\\/', '/', data)
data2 = re.sub('\\\\/', '/', data2)
info = self._parse_json(data, video_id)
info2 = self._parse_json(data2, video_id)
formats = []
for format_id, format_list in (
try_get(info, lambda x: x['tracks'], dict) or {}).items():
for format_dict in format_list:
if not isinstance(format_dict, dict):
continue
format_url = try_get(
format_dict, lambda x: url_or_none(x['src']))
if not format_url:
continue
format_type = format_dict.get('type')
ext = determine_ext(format_url)
if (format_type == 'application/x-mpegURL'
or format_id == 'HLS' or ext == 'm3u8'):
formats.extend(self._extract_m3u8_formats(
format_url, video_id, 'mp4',
entry_protocol='m3u8_native', m3u8_id='hls',
fatal=False))
elif (format_type == 'application/dash+xml'
or format_id == 'DASH' or ext == 'mpd'):
formats.extend(self._extract_mpd_formats(
format_url, video_id, mpd_id='dash', fatal=False))
else:
formats.append({
'url': format_url,
})
# thumbnail = info.get('plugins').get('thumbnails').get('url')
thumbnail = try_get(
info, lambda x: x['plugins']['thumbnails']['url'], compat_str)
thumbnail = re.sub('$Num$', '001', thumbnail)
duration = info.get('duration')
# title2 = info2.get('video').get('title')
# title = info2.get('video').get('custom').get('show_title')
title2 = try_get(info2, lambda x: x['video']['title'], compat_str)
title = try_get(
info2, lambda x: x['video']['custom']['show_title'], compat_str)
title = ' - '.join(x for x in (title, title2, ) if x)
# if not title:
# continue
return {
'id': video_id,
'title': title,
'thumbnail': thumbnail,
'duration': duration,
'formats': formats
}
class MarkizaPageIE(InfoExtractor):
_VALID_URL = r'https?://(?:www\.)?(?:(?:[^/]+\.)?markiza|tvnoviny)\.sk/(?:[^/]+/)*(?P<id>\d+)_'
_TESTS = [{
'url': 'http://www.markiza.sk/soubiz/zahranicny/1923705_oteckovia-maju-svoj-den-ti-slavni-nie-su-o-nic-menej-rozkosni',
'url': 'https://www.markiza.sk/soubiz/zahranicny/1923705_oteckovia-maju-svoj-den-ti-slavni-nie-su-o-nic-menej-rozkosni',
'md5': 'ada4e9fad038abeed971843aa028c7b0',
'info_dict': {
'id': '139355',
@ -87,29 +140,31 @@ class MarkizaPageIE(InfoExtractor):
'skip_download': True,
},
}, {
'url': 'http://dajto.markiza.sk/filmy-a-serialy/1774695_frajeri-vo-vegas',
'url': 'https://dajto.markiza.sk/filmy-a-serialy/1774695_frajeri-vo-vegas',
'only_matching': True,
}, {
'url': 'http://superstar.markiza.sk/aktualne/1923870_to-je-ale-telo-spevacka-ukazala-sexy-postavicku-v-bikinach',
'url': 'https://superstar.markiza.sk/aktualne/1923870_to-je-ale-telo-spevacka-ukazala-sexy-postavicku-v-bikinach',
'only_matching': True,
}, {
'url': 'http://hybsa.markiza.sk/aktualne/1923790_uzasna-atmosfera-na-hybsa-v-poprade-superstaristi-si-prve-koncerty-pred-davom-ludi-poriadne-uzili',
'url': 'https://hybsa.markiza.sk/aktualne/1923790_uzasna-atmosfera-na-hybsa-v-poprade-superstaristi-si-prve-koncerty-pred-davom-ludi-poriadne-uzili',
'only_matching': True,
}, {
'url': 'http://doma.markiza.sk/filmy/1885250_moja-vysnivana-svadba',
'url': 'https://doma.markiza.sk/filmy/1885250_moja-vysnivana-svadba',
'only_matching': True,
}, {
'url': 'http://www.tvnoviny.sk/domace/1923887_po-smrti-manzela-ju-cakalo-poriadne-prekvapenie',
'url': 'https://www.tvnoviny.sk/domace/1923887_po-smrti-manzela-ju-cakalo-poriadne-prekvapenie',
'only_matching': True,
}]
@classmethod
def suitable(cls, url):
return False if MarkizaIE.suitable(url) else super(MarkizaPageIE, cls).suitable(url)
return (
False if MarkizaIE.suitable(url)
else super(MarkizaPageIE, cls).suitable(url)
)
def _real_extract(self, url):
playlist_id = self._match_id(url)
webpage = self._download_webpage(
# Downloading for some hosts (e.g. dajto, doma) fails with 500
# although everything seems to be OK, so considering 500
@ -117,7 +172,8 @@ class MarkizaPageIE(InfoExtractor):
url, playlist_id, expected_status=500)
entries = [
self.url_result('http://videoarchiv.markiza.sk/video/%s' % video_id)
self.url_result(
'http://videoarchiv.markiza.sk/video/%s' % video_id)
for video_id in orderedSet(re.findall(
r'(?:initPlayer_|data-entity=["\']|id=["\']player_)(\d+)',
webpage))]

View File

@ -1,7 +1,13 @@
# coding: utf-8
from __future__ import unicode_literals
import re
from .common import InfoExtractor
from ..utils import (
url_or_none,
determine_ext
)
class RTVSIE(InfoExtractor):
@ -26,7 +32,8 @@ class RTVSIE(InfoExtractor):
'id': '63118',
'ext': 'mp4',
'title': 'Amaro Džives - Náš deň',
'description': 'Galavečer pri príležitosti Medzinárodného dňa Rómov.'
'description':
'Galavečer pri príležitosti Medzinárodného dňa Rómov.'
},
'params': {
'skip_download': True,
@ -36,12 +43,66 @@ class RTVSIE(InfoExtractor):
def _real_extract(self, url):
video_id = self._match_id(url)
webpage = self._download_webpage(url, video_id)
if url.find('/radio/') != -1:
a2 = url.split('/')[-1]
a1 = url.split('/')[-2]
embed = self._download_webpage(
"https://www.rtvs.sk/embed/radio/archive/%s/%s" % (a1, a2),
video_id)
audio_id = re.search('audio5f.json?id=(?P<id>[^\"]+)', embed)
audio_id = audio_id.group('id')
info = self._download_json(
"https://www.rtvs.sk/json/audio5f.json?id=%s" % audio_id,
audio_id)
playlist_url = self._search_regex(
r'playlist["\']?\s*:\s*(["\'])(?P<url>(?:(?!\1).)+)\1', webpage,
'playlist url', group='url')
formats = []
formats.append({
'url': info['playlist'][0]['sources'][0]['src'],
'format_id': None,
'height': 0
})
info = info['playlist'][0]
return {
'id': audio_id,
'title': info.get('title'),
'thumbnail': info.get('image'),
'formats': formats
}
else:
info = self._download_json(
"https://www.rtvs.sk/json/archive5f.json?id=%s" % video_id,
video_id)
info = info.get('clip')
data = self._download_json(
playlist_url, video_id, 'Downloading playlist')[0]
return self._parse_jwplayer_data(data, video_id=video_id)
formats = []
for format_id, format_list in info.items():
if not isinstance(format_list, list):
format_list = [format_list]
for format_dict in format_list:
if not isinstance(format_dict, dict):
continue
format_url = url_or_none(format_dict.get('src'))
format_type = format_dict.get('type')
ext = determine_ext(format_url)
if (format_type == 'application/x-mpegURL'
or format_id == 'HLS' or ext == 'm3u8'):
formats.extend(self._extract_m3u8_formats(
format_url, video_id, 'mp4',
entry_protocol='m3u8_native', m3u8_id='hls',
fatal=False))
elif (format_type == 'application/dash+xml'
or format_id == 'DASH' or ext == 'mpd'):
pass
else:
formats.append({
'url': format_url,
})
formats = sorted(formats, key=lambda i: i['tbr'])
dt = info.get('datetime_create')
return {
'id': video_id,
'title': info.get('title') + '-' + dt[:10],
'thumbnail': info.get('image'),
'description': info.get('description'),
'formats': formats
}

View File

@ -1659,17 +1659,46 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
def _extract_n_function_name(self, jscode):
func_name, idx = self._search_regex(
# new: (b=String.fromCharCode(110),c=a.get(b))&&c=nfunc[idx](c)
# or: (b="nn"[+a.D],c=a.get(b))&&(c=nfunc[idx](c)s
# old: .get("n"))&&(b=nfunc[idx](b)
# older: .get("n"))&&(b=nfunc(b)
# or: (b="nn"[+a.D],c=a.get(b))&&(c=nfunc[idx](c)
# or: (PL(a),b=a.j.n||null)&&(b=nfunc[idx](b)
# or: (b="nn"[+a.D],vL(a),c=a.j[b]||null)&&(c=narray[idx](c),a.set(b,c),narray.length||nfunc("")
# old: (b=a.get("n"))&&(b=nfunc[idx](b)(?P<c>[a-z])\s*=\s*[a-z]\s*
# older: (b=a.get("n"))&&(b=nfunc(b)
r'''(?x)
(?:\(\s*(?P<b>[a-z])\s*=\s*(?:
String\s*\.\s*fromCharCode\s*\(\s*110\s*\)|
"n+"\[\s*\+?s*[\w$.]+\s*]
)\s*,(?P<c>[a-z])\s*=\s*[a-z]\s*)?
\.\s*get\s*\(\s*(?(b)(?P=b)|"n{1,2}")(?:\s*\)){2}\s*&&\s*\(\s*(?(c)(?P=c)|b)\s*=\s*
(?P<nfunc>[a-zA-Z_$][\w$]*)(?:\s*\[(?P<idx>\d+)\])?\s*\(\s*[\w$]+\s*\)
''', jscode, 'Initial JS player n function name', group=('nfunc', 'idx'))
\((?:[\w$()\s]+,)*?\s* # (
(?P<b>[a-z])\s*=\s* # b=
(?:
(?: # expect ,c=a.get(b) (etc)
String\s*\.\s*fromCharCode\s*\(\s*110\s*\)|
"n+"\[\s*\+?s*[\w$.]+\s*]
)\s*(?:,[\w$()\s]+(?=,))*|
(?P<old>[\w$]+) # a (old[er])
)\s*
(?(old)
# b.get("n")
(?:\.\s*[\w$]+\s*|\[\s*[\w$]+\s*]\s*)*?
(?:\.\s*n|\[\s*"n"\s*]|\.\s*get\s*\(\s*"n"\s*\))
| # ,c=a.get(b)
,\s*(?P<c>[a-z])\s*=\s*[a-z]\s*
(?:\.\s*[\w$]+\s*|\[\s*[\w$]+\s*]\s*)*?
(?:\[\s*(?P=b)\s*]|\.\s*get\s*\(\s*(?P=b)\s*\))
)
# interstitial junk
\s*(?:\|\|\s*null\s*)?(?:\)\s*)?&&\s*(?:\(\s*)?
(?(c)(?P=c)|(?P=b))\s*=\s* # [c|b]=
# nfunc|nfunc[idx]
(?P<nfunc>[a-zA-Z_$][\w$]*)(?:\s*\[(?P<idx>\d+)\])?\s*\(\s*[\w$]+\s*\)
''', jscode, 'Initial JS player n function name', group=('nfunc', 'idx'),
default=(None, None))
# thx bashonly: yt-dlp/yt-dlp/pull/10611
if not func_name:
self.report_warning('Falling back to generic n function search')
return self._search_regex(
r'''(?xs)
(?:(?<=[^\w$])|^) # instead of \b, which ignores $
(?P<name>(?!\d)[a-zA-Z\d_$]+)\s*=\s*function\((?!\d)[a-zA-Z\d_$]+\)
\s*\{(?:(?!};).)+?["']enhanced_except_
''', jscode, 'Initial JS player n function name', group='name')
if not idx:
return func_name

View File

@ -925,9 +925,16 @@ class JSInterpreter(object):
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, compat_str)), 'must be applied on a list or string')
# From [1]:
# .slice() - like [:]
# .slice(n) - like [n:] (not [slice(n)]
# .slice(m, n) - like [m:n] or [slice(m, n)]
# [1] https://developer.mozilla.org/en-US/docs/Web/JavaScript/Reference/Global_Objects/Array/slice
assertion(len(argvals) <= 2, 'takes between 0 and 2 arguments')
if len(argvals) < 2:
argvals += (None,)
return obj[slice(*argvals)]
elif member == 'splice':
assertion(isinstance(obj, list), 'must be applied on a list')
assertion(argvals, 'takes one or more arguments')