Merge b1699d4752 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/newgrounds.py

@@ -8,12 +8,14 @@ from ..utils import (
     int_or_none,
     parse_duration,
     parse_filesize,
+    parse_count,
+    parse_resolution,
     unified_timestamp,
 )
 
 
 class NewgroundsIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?newgrounds\.com/(?:audio/listen|portal/view)/(?P<id>[0-9]+)'
+    _VALID_URL = r'https?://(?:www\.)?newgrounds\.com/(?:audio/listen|portal/view)/(?P<id>[0-9]+)(?:/format/flash)?'
     _TESTS = [{
         'url': 'https://www.newgrounds.com/audio/listen/549479',
         'md5': 'fe6033d297591288fa1c1f780386f07a',
@@ -25,17 +27,18 @@ class NewgroundsIE(InfoExtractor):
             'timestamp': 1378878540,
             'upload_date': '20130911',
             'duration': 143,
+            'description': 'Sonic blahblahblah I\'m late again SEPTEMBER 11TH: Busmode-The story of ants on a log. ',
         },
     }, {
-        'url': 'https://www.newgrounds.com/portal/view/673111',
-        'md5': '3394735822aab2478c31b1004fe5e5bc',
+        'url': 'https://www.newgrounds.com/portal/view/297383',
         'info_dict': {
-            'id': '673111',
+            'id': '297383',
             'ext': 'mp4',
-            'title': 'Dancin',
-            'uploader': 'Squirrelman82',
-            'timestamp': 1460256780,
-            'upload_date': '20160410',
+            'title': 'Metal Gear Awesome',
+            'uploader': 'Egoraptor',
+            'timestamp': 1140663240,
+            'upload_date': '20060223',
+            'description': 'Metal Gear is awesome is so is this movie.',
         },
     }, {
         # source format unavailable, additional mp4 formats
@@ -44,9 +47,10 @@ class NewgroundsIE(InfoExtractor):
             'id': '689400',
             'ext': 'mp4',
             'title': 'ZTV News Episode 8',
-            'uploader': 'BennettTheSage',
+            'uploader': 'ZONE-SAMA',
             'timestamp': 1487965140,
             'upload_date': '20170224',
+            'description': 'ZTV News Episode 8 (February 2017)',
         },
         'params': {
             'skip_download': True,
@@ -55,38 +59,44 @@ class NewgroundsIE(InfoExtractor):
 
     def _real_extract(self, url):
         media_id = self._match_id(url)
+        formats = []
+        uploader = None
 
         webpage = self._download_webpage(url, media_id)
 
         title = self._html_search_regex(
-            r'<title>([^>]+)</title>', webpage, 'title')
+            r'<title>(.+?)</title>', webpage, 'title')
 
-        media_url = self._parse_json(self._search_regex(
-            r'"url"\s*:\s*("[^"]+"),', webpage, ''), media_id)
+        media_url_string = self._search_regex(r'"url"\s*:\s*("[^"]+"),', webpage, 'media url string', fatal=False, default=None)
 
+        if media_url_string:
+            media_url = self._parse_json(media_url_string, media_id)
             formats = [{
                 'url': media_url,
                 'format_id': 'source',
                 'quality': 1,
             }]
-
-        max_resolution = int_or_none(self._search_regex(
-            r'max_resolution["\']\s*:\s*(\d+)', webpage, 'max resolution',
-            default=None))
-        if max_resolution:
-            url_base = media_url.rpartition('.')[0]
-            for resolution in (360, 720, 1080):
-                if resolution > max_resolution:
-                    break
-                formats.append({
-                    'url': '%s.%dp.mp4' % (url_base, resolution),
-                    'format_id': '%dp' % resolution,
-                    'height': resolution,
+        else:
+            json_data = self._download_json('https://www.newgrounds.com/portal/video/' + media_id, media_id, headers={
+                'Accept': 'application/json, text/javascript, */*; q=0.01',
+                'Accept-Encoding': 'gzip, deflate, br',
+                'X-Requested-With': 'XMLHttpRequest',
             })
 
-        self._check_formats(formats, media_id)
-        self._sort_formats(formats)
+            uploader = json_data.get('author')
+            media_formats = json_data['sources']
+            for media_format in media_formats:
+                media_sources = media_formats[media_format]
+                for source in media_sources:
+                    format = {
+                        'format_id': media_format,
+                        'url': source.get('src'),
+                    }
+                    format.update(parse_resolution(media_format))
+                    format['quality'] = format.get('height')
+                    formats.append(format)
 
+        if not uploader:
             uploader = self._html_search_regex(
                 (r'(?s)<h4[^>]*>(.+?)</h4>.*?<em>\s*Author\s*</em>',
                  r'(?:Author|Writer)\s*<a[^>]+>([^<]+)'), webpage, 'uploader',
@@ -96,10 +106,18 @@ class NewgroundsIE(InfoExtractor):
             (r'<dt>\s*Uploaded\s*</dt>\s*<dd>([^<]+</dd>\s*<dd>[^<]+)',
              r'<dt>\s*Uploaded\s*</dt>\s*<dd>([^<]+)'), webpage, 'timestamp',
             default=None))
+
+        thumbnail = self._og_search_thumbnail(webpage)
+
         duration = parse_duration(self._search_regex(
             r'(?s)<dd>\s*Song\s*</dd>\s*<dd>.+?</dd>\s*<dd>([^<]+)', webpage,
             'duration', default=None))
 
+        description = self._og_search_description(webpage)
+
+        view_count = parse_count(self._html_search_regex(r'(?s)<dt>\s*Views\s*</dt>\s*<dd>([\d\.,]+)</dd>', webpage,
+                                                         'view_count', fatal=False, default=None))
+
         filesize_approx = parse_filesize(self._html_search_regex(
             r'(?s)<dd>\s*Song\s*</dd>\s*<dd>(.+?)</dd>', webpage, 'filesize',
             default=None))
@@ -109,6 +127,9 @@ class NewgroundsIE(InfoExtractor):
         if '<dd>Song' in webpage:
             formats[0]['vcodec'] = 'none'
 
+        self._check_formats(formats, media_id)
+        self._sort_formats(formats)
+
         return {
             'id': media_id,
             'title': title,
@@ -116,6 +137,9 @@ class NewgroundsIE(InfoExtractor):
             'timestamp': timestamp,
             'duration': duration,
             'formats': formats,
+            'description': description,
+            'thumbnail': thumbnail,
+            'view_count': view_count,
         }
 
 
@@ -127,14 +151,14 @@ class NewgroundsPlaylistIE(InfoExtractor):
             'id': 'cats',
             'title': 'Cats',
         },
-        'playlist_mincount': 46,
+        'playlist_mincount': 45,
     }, {
-        'url': 'http://www.newgrounds.com/portal/search/author/ZONE-SAMA',
+        'url': 'https://www.newgrounds.com/collection/dogs',
         'info_dict': {
-            'id': 'ZONE-SAMA',
-            'title': 'Portal Search: ZONE-SAMA',
+            'id': 'dogs',
+            'title': 'Dogs',
         },
-        'playlist_mincount': 47,
+        'playlist_mincount': 25,
     }, {
         'url': 'http://www.newgrounds.com/audio/search/title/cats',
         'only_matching': True,
@@ -155,14 +179,14 @@ class NewgroundsPlaylistIE(InfoExtractor):
 
         entries = []
         for a, path, media_id in re.findall(
-                r'(<a[^>]+\bhref=["\']/?((?:portal/view|audio/listen)/(\d+))[^>]+>)',
+                r'(<a[^>]+href="https?://[^/]+/(audio/listen|portal/view)/([0-9]+)"[^>]+>)',
                 webpage):
             a_class = extract_attributes(a).get('class')
             if a_class not in ('item-portalsubmission', 'item-audiosubmission'):
                 continue
             entries.append(
                 self.url_result(
-                    'https://www.newgrounds.com/%s' % path,
+                    'https://www.newgrounds.com/%s/%s' % (path, media_id),
                     ie=NewgroundsIE.ie_key(), video_id=media_id))
 
         return self.playlist_result(entries, playlist_id, title)

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')