Merge b0f50733a1 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/duboku.py

@@ -0,0 +1,242 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..compat import compat_urlparse
+from ..utils import (
+    clean_html,
+    extract_attributes,
+    ExtractorError,
+    get_elements_by_class,
+    int_or_none,
+    js_to_json,
+    smuggle_url,
+    unescapeHTML,
+)
+
+
+def _get_elements_by_tag_and_attrib(html, tag=None, attribute=None, value=None, escape_value=True):
+    """Return the content of the tag with the specified attribute in the passed HTML document"""
+
+    if tag is None:
+        tag = '[a-zA-Z0-9:._-]+'
+    if attribute is None:
+        attribute = ''
+    else:
+        attribute = r'\s+(?P<attribute>%s)' % re.escape(attribute)
+    if value is None:
+        value = ''
+    else:
+        value = re.escape(value) if escape_value else value
+        value = '=[\'"]?(?P<value>%s)[\'"]?' % value
+
+    retlist = []
+    for m in re.finditer(r'''(?xs)
+        <(?P<tag>%s)
+         (?:\s+[a-zA-Z0-9:._-]+(?:=[a-zA-Z0-9:._-]*|="[^"]*"|='[^']*'|))*?
+         %s%s
+         (?:\s+[a-zA-Z0-9:._-]+(?:=[a-zA-Z0-9:._-]*|="[^"]*"|='[^']*'|))*?
+        \s*>
+        (?P<content>.*?)
+        </\1>
+    ''' % (tag, attribute, value), html):
+        retlist.append(m)
+
+    return retlist
+
+
+def _get_element_by_tag_and_attrib(html, tag=None, attribute=None, value=None, escape_value=True):
+    retval = _get_elements_by_tag_and_attrib(html, tag, attribute, value, escape_value)
+    return retval[0] if retval else None
+
+
+class DubokuIE(InfoExtractor):
+    IE_NAME = 'duboku'
+    IE_DESC = 'www.duboku.co'
+
+    _VALID_URL = r'(?:https?://[^/]+\.duboku\.co/vodplay/)(?P<id>[0-9]+-[0-9-]+)\.html.*'
+    _TESTS = [{
+        'url': 'https://www.duboku.co/vodplay/1575-1-1.html',
+        'info_dict': {
+            'id': '1575-1-1',
+            'ext': 'ts',
+            'series': '白色月光',
+            'title': 'contains:白色月光',
+            'season_number': 1,
+            'episode_number': 1,
+        },
+        'params': {
+            'skip_download': 'm3u8 download',
+        },
+    }, {
+        'url': 'https://www.duboku.co/vodplay/1588-1-1.html',
+        'info_dict': {
+            'id': '1588-1-1',
+            'ext': 'ts',
+            'series': '亲爱的自己',
+            'title': 'contains:预告片',
+            'season_number': 1,
+            'episode_number': 1,
+        },
+        'params': {
+            'skip_download': 'm3u8 download',
+        },
+    }]
+
+    _PLAYER_DATA_PATTERN = r'player_data\s*=\s*(\{\s*(.*)})\s*;?\s*</script'
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+        temp = video_id.split('-')
+        series_id = temp[0]
+        season_id = temp[1]
+        episode_id = temp[2]
+
+        webpage_url = 'https://www.duboku.co/vodplay/%s.html' % video_id
+        webpage_html = self._download_webpage(webpage_url, video_id)
+
+        # extract video url
+
+        player_data = self._search_regex(
+            self._PLAYER_DATA_PATTERN, webpage_html, 'player_data')
+        player_data = self._parse_json(player_data, video_id, js_to_json)
+
+        # extract title
+
+        temp = get_elements_by_class('title', webpage_html)
+        series_title = None
+        title = None
+        for html in temp:
+            mobj = re.search(r'<a\s+.*>(.*)</a>', html)
+            if mobj:
+                href = extract_attributes(mobj.group(0)).get('href')
+                if href:
+                    mobj1 = re.search(r'/(\d+)\.html', href)
+                    if mobj1 and mobj1.group(1) == series_id:
+                        series_title = clean_html(mobj.group(0))
+                        series_title = re.sub(r'[\s\r\n\t]+', ' ', series_title)
+                        title = clean_html(html)
+                        title = re.sub(r'[\s\r\n\t]+', ' ', title)
+                        break
+
+        data_url = player_data.get('url')
+        if not data_url:
+            raise ExtractorError('Cannot find url in player_data')
+        data_from = player_data.get('from')
+
+        # if it is an embedded iframe, maybe it's an external source
+        if data_from == 'iframe':
+            # use _type url_transparent to retain the meaningful details
+            # of the video.
+            return {
+                '_type': 'url_transparent',
+                'url': smuggle_url(data_url, {'http_headers': {'Referer': webpage_url}}),
+                'id': video_id,
+                'title': title,
+                'series': series_title,
+                'season_number': int_or_none(season_id),
+                'season_id': season_id,
+                'episode_number': int_or_none(episode_id),
+                'episode_id': episode_id,
+            }
+
+        formats = self._extract_m3u8_formats(data_url, video_id, 'mp4')
+
+        return {
+            'id': video_id,
+            'title': title,
+            'series': series_title,
+            'season_number': int_or_none(season_id),
+            'season_id': season_id,
+            'episode_number': int_or_none(episode_id),
+            'episode_id': episode_id,
+            'formats': formats,
+            'http_headers': {'Referer': 'https://www.duboku.co/static/player/videojs.html'}
+        }
+
+
+class DubokuPlaylistIE(InfoExtractor):
+    IE_NAME = 'duboku:list'
+    IE_DESC = 'www.duboku.co entire series'
+
+    _VALID_URL = r'(?:https?://[^/]+\.duboku\.co/voddetail/)(?P<id>[0-9]+)\.html.*'
+    _TESTS = [{
+        'url': 'https://www.duboku.co/voddetail/1575.html',
+        'info_dict': {
+            'id': 'startswith:1575',
+            'title': '白色月光',
+        },
+        'playlist_count': 12,
+    }, {
+        'url': 'https://www.duboku.co/voddetail/1554.html',
+        'info_dict': {
+            'id': 'startswith:1554',
+            'title': '以家人之名',
+        },
+        'playlist_mincount': 30,
+    }, {
+        'url': 'https://www.duboku.co/voddetail/1554.html#playlist2',
+        'info_dict': {
+            'id': '1554#playlist2',
+            'title': '以家人之名',
+        },
+        'playlist_mincount': 27,
+    }]
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        if mobj is None:
+            raise ExtractorError('Invalid URL: %s' % url)
+        series_id = mobj.group('id')
+        fragment = compat_urlparse.urlparse(url).fragment
+
+        webpage_url = 'https://www.duboku.co/voddetail/%s.html' % series_id
+        webpage_html = self._download_webpage(webpage_url, series_id)
+
+        # extract title
+
+        title = _get_element_by_tag_and_attrib(webpage_html, 'h1', 'class', 'title')
+        title = unescapeHTML(title.group('content')) if title else None
+        if not title:
+            title = self._html_search_meta('keywords', webpage_html)
+        if not title:
+            title = _get_element_by_tag_and_attrib(webpage_html, 'title')
+            title = unescapeHTML(title.group('content')) if title else None
+
+        # extract playlists
+
+        playlists = {}
+        for div in _get_elements_by_tag_and_attrib(
+                webpage_html, attribute='id', value='playlist\\d+', escape_value=False):
+            playlist_id = div.group('value')
+            playlist = []
+            for a in _get_elements_by_tag_and_attrib(
+                    div.group('content'), 'a', 'href', value='[^\'"]+?', escape_value=False):
+                playlist.append({
+                    'href': unescapeHTML(a.group('value')),
+                    'title': unescapeHTML(a.group('content'))
+                })
+            playlists[playlist_id] = playlist
+
+        # select the specified playlist if url fragment exists
+        playlist = None
+        playlist_id = None
+        if fragment:
+            playlist = playlists.get(fragment)
+            playlist_id = fragment
+        else:
+            first = next(iter(playlists.items()), None)
+            if first:
+                (playlist_id, playlist) = first
+        if not playlist:
+            raise ExtractorError(
+                'Cannot find %s' % fragment if fragment else 'Cannot extract playlist')
+
+        # return url results
+        return self.playlist_result([
+            self.url_result(
+                compat_urlparse.urljoin('https://www.duboku.co', x['href']),
+                ie=DubokuIE.ie_key(), video_title=x.get('title'))
+            for x in playlist], series_id + '#' + playlist_id, title)

youtube_dl/extractor/extractors.py

@@ -323,6 +323,10 @@ from .drtv import (
 )
 from .dtube import DTubeIE
 from .dvtv import DVTVIE
+from .duboku import (
+    DubokuIE,
+    DubokuPlaylistIE
+)
 from .dumpert import DumpertIE
 from .defense import DefenseGouvFrIE
 from .discovery import DiscoveryIE

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*(?:
+                \((?:[\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*,(?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*
+                    )\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'))
+            ''', 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')