Merge e8eb999df4 into c5098961b0

[Youtube] Rework n function extraction pattern
Now also succeeds with player b12cc44b
2024-12-22 07:56:49 +00:00 · 2024-08-21 22:33:29 -04:00 · 2024-08-06 20:59:09 +01:00 · 2024-08-06 20:51:38 +01:00 · 2024-08-01 19:18:34 +01:00 · 2023-10-27 14:42:48 -04:00
7 changed files with 298 additions and 13 deletions
--- a/test/test_jsinterp.py
+++ b/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()
--- a/test/test_youtube_signature.py
+++ b/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',
+    ),
 ]


--- a/youtube_dl/extractor/bandlab.py
+++ b/youtube_dl/extractor/bandlab.py
@ -0,0 +1,169 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+
+from ..compat import (
+    compat_str,
+)
+
+from ..utils import (
+    strip_or_none,
+    try_get,
+    url_or_none,
+)
+
+
+class BandlabIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?bandlab\.com/post/(?P<id>[^/]+)'
+
+    _TESTS = [{
+        'url': 'https://www.bandlab.com/post/f5f04998635a44ea819cacdba7ae2076_f8d8574c3bdaec11b6562818783151a1',
+        'info_dict': {
+            'id': 'f5f04998635a44ea819cacdba7ae2076_f8d8574c3bdaec11b6562818783151a1',
+            'ext': 'm4a',
+            'title': 'ON MY OWN (unreleased)',
+            'artist': 'Michael MacDonald',
+        },
+    }]
+
+    def _real_extract(self, url):
+        track_id = self._match_id(url)
+        config = self._download_json(
+            'http://www.bandlab.com/api/v1.3/posts/%s' % track_id, track_id)
+        track_url = config['track']['sample']['audioUrl']
+
+        return {
+            'id': track_id,
+            'title': config['track']['name'],
+            'url': track_url,
+            'artist': config['creator']['name']
+        }
+
+
+class BandlabAlbumOrPlaylistIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?bandlab\.com/[^/]+/(?P<kind>albums|collections)/(?P<id>[^/]+)'
+    _TRACK_ID_RE = r'.+/(?P<id>[^/]+)\.m4a'
+    _TESTS = [{
+        'url': 'https://www.bandlab.com/sbsdasani/albums/dc26e307-e51f-ed11-95d7-002248452390',
+        'playlist': [
+            {
+                'info_dict': {
+                    'id': '91feeb36-8e10-4c91-ae57-ffac0a98c6b4',
+                    'title': 'How\'d I Lose You? (Intro)',
+                    'ext': 'm4a',
+                },
+            },
+            {
+                'info_dict': {
+                    'id': 'd87c50a2-70cb-4937-9b97-3ae8646ca3aa',
+                    'title': 'Money $$$',
+                    'ext': 'm4a',
+                },
+            },
+            {
+                'info_dict': {
+                    'id': 'ff2909ff-348f-448d-9d2c-7edbf2f0ec5e',
+                    'title': 'You\'ll Be Mine',
+                    'ext': 'm4a',
+                },
+            },
+            {
+                'info_dict': {
+                    'id': '53786c38-1f3c-4921-9271-793a64af7186',
+                    'title': 'Who You Are',
+                    'ext': 'm4a',
+                },
+            },
+            {
+                'info_dict': {
+                    'id': '0394bb27-113f-4d19-b902-f9c1fe6ba8a8',
+                    'title': 'In Your Eyes',
+                    'ext': 'm4a',
+                },
+            },
+            {
+                'info_dict': {
+                    'id': '2aa44689-c7fa-4d0e-b28a-e0d1dd570372',
+                    'title': 'The Same',
+                    'ext': 'm4a',
+                },
+            },
+            {
+                'info_dict': {
+                    'id': '281fe589-a559-4260-802d-78a6c7a973d8',
+                    'title': 'Fall In Love',
+                    'ext': 'm4a',
+                },
+            },
+            {
+                'info_dict': {
+                    'id': '83a9dc0a-702f-40fd-82f6-ab847f6a7b46',
+                    'title': 'Tried So Hard',
+                    'ext': 'm4a',
+                },
+            },
+            {
+                'info_dict': {
+                    'id': '3f6a4a1c-eb73-449f-bd45-f7958d6f2de1',
+                    'title': 'The End Of Everything',
+                    'ext': 'm4a',
+                },
+            }
+        ],
+        'info_dict': {
+            'id': 'dc26e307-e51f-ed11-95d7-002248452390',
+            'album': 'ENDLESS SUMMER',
+            'artist': 'Michael MacDonald'
+        },
+    }, {
+        'url': 'https://www.bandlab.com/hexatetrahedronx/collections/8fb1041c-e865-eb11-9889-0050f28a2802',
+        'playlist': [
+            {
+                'info_dict': {
+                    'id': '8f37e4aa-92c4-eb11-a7ad-0050f280467f',
+                    'title': 'psych ward',
+                    'ext': 'm4a'
+                }
+            }
+        ],
+        'info_dict': {
+            'id': '8fb1041c-e865-eb11-9889-0050f28a2802',
+            'album': 'DOOMTAPE',
+            'artist': '12days'
+        }
+    }]
+
+    def _real_extract(self, url):
+        resource_id, kind = re.match(self._VALID_URL, url).group('id', 'kind')
+        config = self._download_json(
+            'http://www.bandlab.com/api/v1.3/%s/%s' % (kind, resource_id), resource_id)
+        entries = []
+        for track in try_get(config, lambda x: x['posts'], list) or []:
+            url, name = try_get(
+                track,
+                (lambda x: (x['track']['sample']['audioUrl'], x['track']['name']),
+                 lambda x: (x['revision']['mixdown']['file'], x['revision']['song']['name'])),
+                tuple) or (None, '', )
+            url = url_or_none(url)
+            name = strip_or_none(name)
+            if not (url and name):
+                continue
+            track_id = re.match(self._TRACK_ID_RE, url).group('id')
+            if not track_id:
+                continue
+            entries.append({
+                'url': url,
+                'id': track_id,
+                'title': name,
+            })
+
+        return {
+            '_type': 'playlist',
+            'id': resource_id,
+            'entries': entries,
+            'album': try_get(config, lambda x: x['name'].strip(), compat_str),
+            'artist': try_get(config, lambda x: x['creator']['name'].strip(), compat_str)
+        }
--- a/youtube_dl/extractor/extractors.py
+++ b/youtube_dl/extractor/extractors.py
@ -776,8 +776,13 @@ from .ndr import (
    NJoyEmbedIE,
 )
 from .ndtv import NDTVIE
+from .bandlab import (
+    BandlabIE, 
+    BandlabAlbumOrPlaylistIE
+)
 from .netzkino import NetzkinoIE
 from .nerdcubed import NerdCubedFeedIE
+from .nest import NestIE
 from .neteasemusic import (
    NetEaseMusicIE,
    NetEaseMusicAlbumIE,
--- a/youtube_dl/extractor/nest.py
+++ b/youtube_dl/extractor/nest.py
@ -0,0 +1,39 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+from .common import InfoExtractor
+
+
+class NestIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?video.nest\.com/clip/(?P<id>)(.mp4)?'
+    _TEST = {
+        'url': 'https://video.nest.com/clip/73ddb6bd57c4485597a76e154a4429ea.mp4',
+        'md5': '7ab4eb6d4c2480be1740cc014a76ee96',
+        'info_dict': {
+            'id': '73ddb6bd57c4485597a76e154a4429ea',
+            'ext': 'mp4',
+            'title': "\"\"",
+            'description': '#caughtonNestCam',
+        }
+    }
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+        webpage = self._download_webpage(url, video_id)
+        video_id = self._search_regex(
+            r'https:\/\/video.nest.com\/clip\/(.+?)(\.|")', webpage, 'video_id', fatal=False)
+        title = self._html_search_meta(['og:title', 'title'], webpage, 'title')
+        if title == "":
+            title = "\"\""
+        description = self._html_search_meta(['og:description', 'description'], webpage, 'description')
+        ext = self._html_search_meta('og:video:type', webpage, 'ext')
+        if "/" in ext:
+            ext = ext[ext.index("/") + 1:]
+        return {
+            'url': self._html_search_meta(['og:video:url', 'url'], webpage, 'url'),
+            'id': video_id,
+            'title': title,
+            'description': description,
+            'ext': ext,
+            'thumbnail': self._html_search_meta(['og:image', 'image'], webpage, 'thumbnail'),
+        }
--- a/youtube_dl/extractor/youtube.py
+++ b/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

--- a/youtube_dl/jsinterp.py
+++ b/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')
Author	SHA1	Message	Date
Ethan Muir	29c553f9f6	Merge `e8eb999df4` into `c5098961b0`	2024-08-21 22:33:29 -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
Ethan Muir	e8eb999df4	Added extractor for bandlab	2023-10-27 14:42:48 -04:00
Ethan Muir	171b7deda0	Added changes to add nest extractor	2023-10-27 01:52:11 -04:00