Merge 69a40d3eb0 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

@@ -1154,7 +1154,10 @@ from .slideshare import SlideshareIE
 from .slideslive import SlidesLiveIE
 from .slutload import SlutloadIE
 from .snotr import SnotrIE
-from .sohu import SohuIE
+from .sohu import (
+    SohuIE,
+    SohuPlaylistIE,
+)
 from .sonyliv import SonyLIVIE
 from .soundcloud import (
     SoundcloudEmbedIE,

youtube_dl/extractor/sohu.py

@@ -1,6 +1,7 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
+import math
 import re
 
 from .common import InfoExtractor
@@ -11,12 +12,13 @@ from ..compat import (
 from ..utils import (
     ExtractorError,
     int_or_none,
-    try_get,
+    try_get, NO_DEFAULT, HTMLAttributeParser,
 )
 
 
 class SohuIE(InfoExtractor):
-    _VALID_URL = r'https?://(?P<mytv>my\.)?tv\.sohu\.com/.+?/(?(mytv)|n)(?P<id>\d+)\.shtml.*?'
+    _VALID_URL = r'https?://(?:my\.)?tv\.sohu\.com/.+?/.+(?:\.html|\.shtml).*?'
+    _VALID_URL_OG_URL = r'https?://(?P<mytv>my\.)?tv\.sohu\.com/.+?/(?(mytv)|n)(?P<id>\d+)\.shtml.*?'
 
     # Sohu videos give different MD5 sums on Travis CI and my machine
     _TESTS = [{
@@ -82,6 +84,58 @@ class SohuIE(InfoExtractor):
         'params': {
             'skip_download': True
         }
+    }, {
+        'url': 'https://tv.sohu.com/v/dXMvMjMyNzk5ODg5Lzc4NjkzNDY0LnNodG1s.html',
+        'info_dict': {
+            'id': '78693464',
+            'ext': 'mp4',
+            'title': '【爱范品】第31期：MWC见不到的奇葩手机',
+        }
+    }, {
+        'note': 'Video in issue #18542: https://github.com/ytdl-org/youtube-dl/issues/18542',
+        'url': 'https://tv.sohu.com/v/MjAxNzA3MTMvbjYwMDA1MzM4MS5zaHRtbA==.html',
+        'info_dict': {
+            'id': '600053381',
+            'ext': 'mp4',
+            'title': '试听：侯旭《孤鸟》',
+        },
+    }, {
+        'note': 'Video in issue #28944: https://github.com/ytdl-org/youtube-dl/issues/28944',
+        'url': 'https://tv.sohu.com/v/dXMvNTAyMjA5MTMvNjg1NjIyNTYuc2h0bWw=.html?src=pl',
+        'info_dict': {
+            'id': '68562256',
+            'ext': 'mp4',
+            'title': "Cryin'  [HD 1080p]  Chris.Botti(feat. Steven Tyler",
+        },
+    }, {
+        'note': 'Multipart video with new address format',
+        'url': 'https://tv.sohu.com/v/dXMvMjQyNTYyMTYzLzc4OTEwMzM5LnNodG1s.html?src=pl',
+        'info_dict': {
+            'id': '78910339',
+            'title': '【神探苍实战秘籍】第13期 战争之影 赫卡里姆',
+        },
+        'playlist': [{
+            'info_dict': {
+                'id': '78910339_part1',
+                'ext': 'mp4',
+                'duration': 294,
+                'title': '【神探苍实战秘籍】第13期 战争之影 赫卡里姆',
+            }
+        }, {
+            'info_dict': {
+                'id': '78910339_part2',
+                'ext': 'mp4',
+                'duration': 300,
+                'title': '【神探苍实战秘籍】第13期 战争之影 赫卡里姆',
+            }
+        }, {
+            'info_dict': {
+                'id': '78910339_part3',
+                'ext': 'mp4',
+                'duration': 150,
+                'title': '【神探苍实战秘籍】第13期 战争之影 赫卡里姆',
+            }
+        }]
     }]
 
     def _real_extract(self, url):
@@ -97,7 +151,12 @@ class SohuIE(InfoExtractor):
                 'Downloading JSON data for %s' % vid_id,
                 headers=self.geo_verification_headers())
 
-        mobj = re.match(self._VALID_URL, url)
+        mobj = re.match(self._VALID_URL_OG_URL, url)
+        if mobj is None:
+            webpage = self._download_webpage(url, '')
+            url = self._og_search_url(webpage)
+            mobj = re.match(self._VALID_URL_OG_URL, url)
+
         video_id = mobj.group('id')
         mytv = mobj.group('mytv') is not None
 
@@ -200,3 +259,65 @@ class SohuIE(InfoExtractor):
             }
 
         return info
+
+
+class SohuPlaylistIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:my\.)?tv\.sohu\.com/pl/(?P<pl_id>\d+)$'
+    _URL_IN_PLAYLIST = re.compile(r'<strong>.*?</strong>')
+    parser = HTMLAttributeParser()
+    _TESTS = [{
+        'url': 'https://my.tv.sohu.com/pl/9637148',
+        'info_dict': {
+            'title': '【语文大师】初中必背、常考经典古诗词',
+            'id': '9637148',
+        },
+        'playlist_count': 70,
+    }, {
+        'url': 'https://my.tv.sohu.com/pl/9700995',
+        'info_dict': {
+            'title': '牛人游戏',
+            'id': '9700995',
+        },
+        'playlist_count': 198,
+    },
+    ]
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        playlist_id = mobj.group('pl_id')
+
+        webpage = self._download_webpage(url, playlist_id)
+        title = self._get_playlist_title(webpage)
+        all_pages = self._get_all_pages_in_playlist(webpage, url)
+        video_list = self._get_video_list(all_pages, playlist_id)
+        playlist = self.playlist_result(self._entries(video_list), playlist_id, title)
+
+        return playlist
+
+    def _get_playlist_title(self, webpage):
+        title = self._search_regex(r'<title>(.*?)</title>', webpage, 'title')
+        return re.sub(r'(?:^栏目：| -搜狐视频$)', '', title)
+
+    def _entries(self, video_list):
+        entries = []
+        for mobj in re.finditer(self._URL_IN_PLAYLIST, video_list):
+            self.parser.feed(mobj.group(0))
+            url = self.parser.attrs['href']
+            title = self.parser.attrs['title']
+            entry = self.url_result(url, SohuIE.ie_key(), '', title)
+            entries.append(entry)
+        return entries
+
+    def _get_all_pages_in_playlist(self, first_page, url):
+        pgcount = int(self._search_regex(r'var pgcount = \'(\d+)\'', first_page, 'pgcount'))
+        pgsize = int(self._search_regex(r'var pgsize = \'(\d+)\'', first_page, 'pgsize'))
+        return [url + '/index%d.shtml' % (i + 1) for i in range(0, math.ceil(pgcount / pgsize))]
+
+    def _get_video_list(self, all_pages, playlist_id):
+        video_list = ''
+        for i, url in enumerate(all_pages):
+            webpage = self._download_webpage(url, "playlist " + playlist_id + " page: %d" % (1 + i))
+            video_list += self._search_regex(
+                r'<ul class="uList cfix">(.*?)</ul>',
+                webpage, 'video list', NO_DEFAULT, True, re.DOTALL)
+        return video_list

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')