Merge 1e4ea76441 into c5098961b0

Now also succeeds with player b12cc44b

test/test_jsinterp.py

@@ -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()

test/test_youtube_signature.py

@@ -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',
+    ),
 ]
 
 

youtube_dl/extractor/extractors.py

@@ -750,6 +750,10 @@ from .nationalgeographic import (
     NationalGeographicVideoIE,
     NationalGeographicTVIE,
 )
+from .nate import (
+    NateIE,
+    NateProgramIE,
+)
 from .naver import NaverIE
 from .nba import (
     NBAWatchEmbedIE,

youtube_dl/extractor/nate.py

@@ -0,0 +1,194 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import itertools
+
+from .common import InfoExtractor
+
+from ..utils import (
+    ExtractorError,
+    int_or_none,
+    merge_dicts,
+    T,
+    traverse_obj,
+    txt_or_none,
+    unified_strdate,
+    url_or_none,
+)
+
+
+class NateBaseIE(InfoExtractor):
+    _API_BASE = 'https://tv.nate.com/api/v1/'
+
+    def _download_webpage_handle(self, url_or_request, video_id, *args, **kwargs):
+        fatal = kwargs.get('fatal', True)
+        kwargs['fatal'] = False
+        res = super(NateBaseIE, self)._download_webpage_handle(
+            url_or_request, video_id, *args, **kwargs)
+        if not res:
+            if fatal:
+                raise ExtractorError('Failed to download webpage')
+            return res
+        status = res[1].getcode()
+        if 200 <= status < 400:
+            new_url = res[1].geturl()
+            if url_or_request != new_url and '/Error.html' in new_url:
+                raise ExtractorError(
+                    'Download redirected to Error.html: expired?',
+                    expected=True)
+        else:
+            msg = 'Failed to download webpage: HTTP code %d' % status
+            if fatal:
+                raise ExtractorError(msg)
+            else:
+                self.report_warning(msg)
+        return res
+
+
+class NateIE(NateBaseIE):
+    _VALID_URL = r'https?://(?:m\.)?tv\.nate\.com/clip/(?P<id>[0-9]+)'
+    _TESTS = [{
+        'url': 'https://tv.nate.com/clip/1848976',
+        'info_dict': {
+            'id': '1848976',
+            'ext': 'mp4',
+            'title': '[결승 오프닝 타이틀] 2018 LCK 서머 스플릿 결승전 kt Rolster VS Griffin',
+            'description': 'md5:e1b79a7dcf0d8d586443f11366f50e6f',
+            'thumbnail': r're:^http?://.*\.jpg$',
+            'upload_date': '20180908',
+            'age_limit': 15,
+            'duration': 73,
+            'uploader': '2018 LCK 서머 스플릿(롤챔스)',
+            'channel': '2018 LCK 서머 스플릿(롤챔스)',
+            'channel_id': '3606',
+            'uploader_id': '3606',
+            'tags': 'count:59',
+        },
+        'skip': 'Redirect to Error.html',
+    }, {
+        'url': 'https://tv.nate.com/clip/4300566',
+        # 'md5': '02D3CAB3907B60C58043761F8B5BF2B3',
+        'info_dict': {
+            'id': '4300566',
+            'ext': 'mp4',
+            'title': '[심쿵엔딩] 이준호x이세영, 서로를 기억하며 끌어안는 두 사람!💕, MBC 211204 방송',
+            'description': 'md5:edf489c54ea2682c7973154b2089aa0e',
+            'thumbnail': r're:^http?://.*\.jpg$',
+            'upload_date': '20211204',
+            'age_limit': 15,
+            'duration': 201,
+            'uploader': '옷소매 붉은 끝동',
+            'channel': '옷소매 붉은 끝동',
+            'channel_id': '27987',
+            'uploader_id': '27987',
+            'tags': 'count:20',
+        },
+        'params': {'skip_download': True},
+    }, {
+        'url': 'https://tv.nate.com/clip/4764792',
+        'info_dict': {
+            'id': '4764792',
+            'ext': 'mp4',
+            'title': '흥을 돋우는 가야금 연주와 트롯의 만남⬈ ‘열두줄’♪ TV CHOSUN 230625 방송',
+            'description': 'md5:85734d3f9daebe4aa4f20cc73bdcc90c',
+            'upload_date': '20230625',
+            'uploader_id': '29116',
+            'uploader': '쇼퀸',
+            'age_limit': 15,
+            'thumbnail': r're:^http?://.*\.jpg$',
+            'duration': 182,
+            'channel': '쇼퀸',
+            'channel_id': '29116',
+            'tags': 'count:25',
+        },
+        'params': {'skip_download': True},
+    }]
+
+    _QUALITY = {
+        '36': 2160,
+        '35': 1080,
+        '34': 720,
+        '33': 480,
+        '32': 360,
+        '31': 270,
+    }
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+
+        video_data, urlh = self._download_json_handle(
+            '{0}clip/{1}'.format(self._API_BASE, video_id), video_id,
+            fatal=False)
+        if not video_data:
+            raise ExtractorError('Empty programme JSON')
+
+        title = video_data['clipTitle']
+        formats = []
+        for f_url in traverse_obj(video_data, ('smcUriList', Ellipsis, T(url_or_none))):
+            fmt_id = f_url[-2:]
+            formats.append({
+                'format_id': fmt_id,
+                'url': f_url,
+                'height': self._QUALITY.get(fmt_id),
+                'quality': int_or_none(fmt_id),
+            })
+        self._sort_formats(formats)
+
+        info = traverse_obj(video_data, {
+            'uploader': ('programTitle', T(txt_or_none)),
+            'uploader_id': ('programSeq', T(txt_or_none)),
+        })
+        for up, ch in (('uploader', 'channel'), ('uploader_id', 'channel_id')):
+            info[ch] = info.get(up)
+
+        return merge_dicts({
+            'id': video_id,
+            'title': title,
+            'formats': formats,
+        }, info, traverse_obj(video_data, {
+            'description': ('synopsis', T(txt_or_none)),
+            'thumbnail': ('contentImg', T(url_or_none)),
+            'upload_date': (('broadDate', 'regDate'), T(unified_strdate)),
+            'age_limit': ('targetAge', T(int_or_none)),
+            'duration': ('playTime', T(int_or_none)),
+            'tags': ('hashTag', T(lambda s: s.split(',') or None)),
+        }, get_all=False))
+
+
+class NateProgramIE(NateBaseIE):
+    _VALID_URL = r'https?://tv\.nate\.com/program/clips/(?P<id>[0-9]+)'
+
+    _TESTS = [{
+        'url': 'https://tv.nate.com/program/clips/27987',
+        'playlist_mincount': 191,
+        'info_dict': {
+            'id': '27987',
+        },
+    }, {
+        'url': 'https://tv.nate.com/program/clips/3606',
+        'playlist_mincount': 15,
+        'info_dict': {
+            'id': '3606',
+        },
+        'skip': 'Redirect to Error.html',
+    }]
+
+    def _entries(self, pl_id):
+        for page_num in itertools.count(1):
+            program_data, urlh = self._download_json_handle(
+                '{0}program/{1}/clip/ranking'.format(self._API_BASE, pl_id),
+                pl_id, query={'size': 20, 'page': page_num},
+                note='Downloading page {0}'.format(page_num), fatal=False)
+
+            empty = True
+            for clip_id in traverse_obj(program_data, ('content', Ellipsis, 'clipSeq', T(txt_or_none))):
+                yield self.url_result(
+                    'https://tv.nate.com/clip/%s' % clip_id,
+                    ie=NateIE.ie_key(), video_id=clip_id)
+                empty = False
+            if traverse_obj(program_data, 'last') or (program_data and empty):
+                break
+
+    def _real_extract(self, url):
+        pl_id = self._match_id(url)
+        return self.playlist_result(self._entries(pl_id), playlist_id=pl_id)

youtube_dl/extractor/youtube.py

@@ -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
 

youtube_dl/jsinterp.py

@@ -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')