Compare commits

...

5 Commits

Author SHA1 Message Date
filip-hejsek
29a828d588
Merge 74bbdc079b into c5098961b0 2024-08-21 22:33:39 -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
filip-hejsek
74bbdc079b
Add an option to discard style information from TTML subtitles 2022-08-11 05:28:03 +02:00
8 changed files with 96 additions and 18 deletions

View File

@ -425,6 +425,34 @@ class TestJSInterpreter(unittest.TestCase):
self._test(jsi, [''], args=['', '-']) self._test(jsi, [''], args=['', '-'])
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__': if __name__ == '__main__':
unittest.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', 'https://www.youtube.com/s/player/5604538d/player_ias.vflset/en_US/base.js',
'7X-he4jjvMx7BCX', 'sViSydX8IHtdWA', '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

@ -282,6 +282,7 @@ def _real_main(argv=None):
postprocessors.append({ postprocessors.append({
'key': 'FFmpegSubtitlesConvertor', 'key': 'FFmpegSubtitlesConvertor',
'format': opts.convertsubtitles, 'format': opts.convertsubtitles,
'ttml_convert_style': opts.ttmlconvertstyle,
}) })
if opts.embedsubtitles: if opts.embedsubtitles:
postprocessors.append({ postprocessors.append({

View File

@ -1659,17 +1659,46 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
def _extract_n_function_name(self, jscode): def _extract_n_function_name(self, jscode):
func_name, idx = self._search_regex( func_name, idx = self._search_regex(
# new: (b=String.fromCharCode(110),c=a.get(b))&&c=nfunc[idx](c) # 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 # or: (b="nn"[+a.D],c=a.get(b))&&(c=nfunc[idx](c)
# old: .get("n"))&&(b=nfunc[idx](b) # or: (PL(a),b=a.j.n||null)&&(b=nfunc[idx](b)
# older: .get("n"))&&(b=nfunc(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) r'''(?x)
(?:\(\s*(?P<b>[a-z])\s*=\s*(?: \((?:[\w$()\s]+,)*?\s* # (
String\s*\.\s*fromCharCode\s*\(\s*110\s*\)| (?P<b>[a-z])\s*=\s* # b=
"n+"\[\s*\+?s*[\w$.]+\s*] (?:
)\s*,(?P<c>[a-z])\s*=\s*[a-z]\s*)? (?: # expect ,c=a.get(b) (etc)
\.\s*get\s*\(\s*(?(b)(?P=b)|"n{1,2}")(?:\s*\)){2}\s*&&\s*\(\s*(?(c)(?P=c)|b)\s*=\s* String\s*\.\s*fromCharCode\s*\(\s*110\s*\)|
(?P<nfunc>[a-zA-Z_$][\w$]*)(?:\s*\[(?P<idx>\d+)\])?\s*\(\s*[\w$]+\s*\) "n+"\[\s*\+?s*[\w$.]+\s*]
''', jscode, 'Initial JS player n function name', group=('nfunc', 'idx')) )\s*(?:,[\w$()\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: if not idx:
return func_name return func_name

View File

@ -925,9 +925,16 @@ class JSInterpreter(object):
obj.reverse() obj.reverse()
return obj return obj
elif member == 'slice': elif member == 'slice':
assertion(isinstance(obj, list), 'must be applied on a list') assertion(isinstance(obj, (list, compat_str)), 'must be applied on a list or string')
assertion(len(argvals) == 1, 'takes exactly one argument') # From [1]:
return obj[argvals[0]:] # .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': elif member == 'splice':
assertion(isinstance(obj, list), 'must be applied on a list') assertion(isinstance(obj, list), 'must be applied on a list')
assertion(argvals, 'takes one or more arguments') assertion(argvals, 'takes one or more arguments')

View File

@ -870,6 +870,10 @@ def parseOpts(overrideArguments=None):
'--convert-subs', '--convert-subtitles', '--convert-subs', '--convert-subtitles',
metavar='FORMAT', dest='convertsubtitles', default=None, metavar='FORMAT', dest='convertsubtitles', default=None,
help='Convert the subtitles to other format (currently supported: srt|ass|vtt|lrc)') help='Convert the subtitles to other format (currently supported: srt|ass|vtt|lrc)')
postproc.add_option(
'--ttml-convert-no-style',
action='store_false', dest='ttmlconvertstyle', default=True,
help='Discard all style information when converting subtitles from ttml/dfxp')
parser.add_option_group(general) parser.add_option_group(general)
parser.add_option_group(network) parser.add_option_group(network)

View File

@ -591,9 +591,10 @@ class FFmpegFixupM3u8PP(FFmpegPostProcessor):
class FFmpegSubtitlesConvertorPP(FFmpegPostProcessor): class FFmpegSubtitlesConvertorPP(FFmpegPostProcessor):
def __init__(self, downloader=None, format=None): def __init__(self, downloader=None, format=None, ttml_convert_style=True):
super(FFmpegSubtitlesConvertorPP, self).__init__(downloader) super(FFmpegSubtitlesConvertorPP, self).__init__(downloader)
self.format = format self.format = format
self.ttml_convert_style = ttml_convert_style
def run(self, info): def run(self, info):
subs = info.get('requested_subtitles') subs = info.get('requested_subtitles')
@ -626,7 +627,7 @@ class FFmpegSubtitlesConvertorPP(FFmpegPostProcessor):
srt_file = subtitles_filename(filename, lang, 'srt', info.get('ext')) srt_file = subtitles_filename(filename, lang, 'srt', info.get('ext'))
with open(dfxp_file, 'rb') as f: with open(dfxp_file, 'rb') as f:
srt_data = dfxp2srt(f.read()) srt_data = dfxp2srt(f.read(), self.ttml_convert_style)
with open(srt_file, 'w', encoding='utf-8') as f: with open(srt_file, 'w', encoding='utf-8') as f:
f.write(srt_data) f.write(srt_data)

View File

@ -4902,7 +4902,7 @@ def srt_subtitles_timecode(seconds):
return '%02d:%02d:%02d,%03d' % (seconds / 3600, (seconds % 3600) / 60, seconds % 60, (seconds % 1) * 1000) return '%02d:%02d:%02d,%03d' % (seconds / 3600, (seconds % 3600) / 60, seconds % 60, (seconds % 1) * 1000)
def dfxp2srt(dfxp_data): def dfxp2srt(dfxp_data, convert_style=True):
''' '''
@param dfxp_data A bytes-like object containing DFXP data @param dfxp_data A bytes-like object containing DFXP data
@returns A unicode object containing converted SRT data @returns A unicode object containing converted SRT data
@ -4944,7 +4944,7 @@ def dfxp2srt(dfxp_data):
def start(self, tag, attrib): def start(self, tag, attrib):
if tag in (_x('ttml:br'), 'br'): if tag in (_x('ttml:br'), 'br'):
self._out += '\n' self._out += '\n'
else: elif convert_style:
unclosed_elements = [] unclosed_elements = []
style = {} style = {}
element_style_id = attrib.get('style') element_style_id = attrib.get('style')
@ -4987,7 +4987,7 @@ def dfxp2srt(dfxp_data):
self._unclosed_elements.append(unclosed_elements) self._unclosed_elements.append(unclosed_elements)
def end(self, tag): def end(self, tag):
if tag not in (_x('ttml:br'), 'br'): if convert_style and tag not in (_x('ttml:br'), 'br'):
unclosed_elements = self._unclosed_elements.pop() unclosed_elements = self._unclosed_elements.pop()
for element in reversed(unclosed_elements): for element in reversed(unclosed_elements):
self._out += '</%s>' % element self._out += '</%s>' % element