Merge 1a57b3c03f into 1036478d13

* WEB URLs are, MWEB not
* resolves #33017

youtube_dl/extractor/soundcloud.py

@@ -3,6 +3,8 @@ from __future__ import unicode_literals
 
 import itertools
 import re
+import json
+import random
 
 from .common import (
     InfoExtractor,
@@ -28,6 +30,7 @@ from ..utils import (
     update_url_query,
     url_or_none,
     urlhandle_detect_ext,
+    sanitized_Request,
 )
 
 
@@ -309,7 +312,81 @@ class SoundcloudIE(InfoExtractor):
                 raise
 
     def _real_initialize(self):
-        self._CLIENT_ID = self._downloader.cache.load('soundcloud', 'client_id') or 'YUKXoArFcqrlQn9tfNHvvyfnDISj04zk'
+        self._CLIENT_ID = self._downloader.cache.load('soundcloud', 'client_id') or "T5R4kgWS2PRf6lzLyIravUMnKlbIxQag"  # 'EXLwg5lHTO2dslU5EePe3xkw0m1h86Cd' # 'YUKXoArFcqrlQn9tfNHvvyfnDISj04zk'
+        self._login()
+
+    _USER_AGENT = "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/84.0.4147.105 Safari/537.36"
+    _API_AUTH_QUERY_TEMPLATE = '?client_id=%s'
+    _API_AUTH_URL_PW = 'https://api-auth.soundcloud.com/web-auth/sign-in/password%s'
+    _access_token = None
+    _HEADERS = {}
+    _NETRC_MACHINE = 'soundcloud'
+
+    def _login(self):
+        username, password = self._get_login_info()
+        if username is None:
+            return
+
+        def genDevId():
+            def genNumBlock():
+                return ''.join([str(random.randrange(10)) for i in range(6)])
+            return '-'.join([genNumBlock() for i in range(4)])
+
+        payload = {
+            'client_id': self._CLIENT_ID,
+            'recaptcha_pubkey': 'null',
+            'recaptcha_response': 'null',
+            'credentials': {
+                'identifier': username,
+                'password': password
+            },
+            'signature': self.sign(username, password, self._CLIENT_ID),
+            'device_id': genDevId(),
+            'user_agent': self._USER_AGENT
+        }
+
+        query = self._API_AUTH_QUERY_TEMPLATE % self._CLIENT_ID
+        login = sanitized_Request(self._API_AUTH_URL_PW % query, json.dumps(payload).encode('utf-8'))
+        response = self._download_json(login, None)
+        self._access_token = response.get('session').get('access_token')
+        if not self._access_token:
+            self.report_warning('Unable to get access token, login may has failed')
+        else:
+            self._HEADERS = {'Authorization': 'OAuth ' + self._access_token}
+
+    # signature generation
+    def sign(self, user, pw, clid):
+        a = 33
+        i = 1
+        s = 440123
+        w = 117
+        u = 1800000
+        l = 1042
+        b = 37
+        k = 37
+        c = 5
+        n = "0763ed7314c69015fd4a0dc16bbf4b90"  # _KEY
+        y = "8"  # _REV
+        r = "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/84.0.4147.105 Safari/537.36"  # _USER_AGENT
+        e = user  # _USERNAME
+        t = clid  # _CLIENT_ID
+
+        d = '-'.join([str(mInt) for mInt in [a, i, s, w, u, l, b, k]])
+        p = n + y + d + r + e + t + d + n
+        h = p
+
+        m = 8011470
+        f = 0
+
+        for f in range(f, len(h)):
+            m = (m >> 1) + ((1 & m) << 23)
+            m += ord(h[f])
+            m &= 16777215
+
+        # c is not even needed
+        out = str(y) + ':' + str(d) + ':' + format(m, 'x') + ':' + str(c)
+
+        return out
 
     @classmethod
     def _resolv_url(cls, url):
@@ -389,7 +466,7 @@ class SoundcloudIE(InfoExtractor):
             if not format_url:
                 continue
             stream = self._download_json(
-                format_url, track_id, query=query, fatal=False)
+                format_url, track_id, query=query, fatal=False, headers=self._HEADERS)
             if not isinstance(stream, dict):
                 continue
             stream_url = url_or_none(stream.get('url'))
@@ -487,7 +564,7 @@ class SoundcloudIE(InfoExtractor):
             info_json_url = self._resolv_url(self._BASE_URL + resolve_title)
 
         info = self._download_json(
-            info_json_url, full_title, 'Downloading info JSON', query=query)
+            info_json_url, full_title, 'Downloading info JSON', query=query, headers=self._HEADERS)
 
         return self._extract_info_dict(info, full_title, token)
 
@@ -503,7 +580,7 @@ class SoundcloudPlaylistBaseIE(SoundcloudIE):
                     'ids': ','.join([compat_str(t['id']) for t in tracks]),
                     'playlistId': playlist_id,
                     'playlistSecretToken': token,
-                })
+                }, headers=self._HEADERS)
         entries = []
         for track in tracks:
             track_id = str_or_none(track.get('id'))
@@ -547,7 +624,7 @@ class SoundcloudSetIE(SoundcloudPlaylistBaseIE):
             full_title += '/' + token
 
         info = self._download_json(self._resolv_url(
-            self._BASE_URL + full_title), full_title)
+            self._BASE_URL + full_title), full_title, headers=self._HEADERS)
 
         if 'errors' in info:
             msgs = (compat_str(err['error_message']) for err in info['errors'])
@@ -574,7 +651,7 @@ class SoundcloudPagedPlaylistBaseIE(SoundcloudIE):
         for i in itertools.count():
             response = self._download_json(
                 next_href, playlist_id,
-                'Downloading track page %s' % (i + 1), query=query)
+                'Downloading track page %s' % (i + 1), query=query, headers=self._HEADERS)
 
             collection = response['collection']
 
@@ -696,7 +773,7 @@ class SoundcloudUserIE(SoundcloudPagedPlaylistBaseIE):
 
         user = self._download_json(
             self._resolv_url(self._BASE_URL + uploader),
-            uploader, 'Downloading user info')
+            uploader, 'Downloading user info', headers=self._HEADERS)
 
         resource = mobj.group('rsrc') or 'all'
 
@@ -721,7 +798,7 @@ class SoundcloudTrackStationIE(SoundcloudPagedPlaylistBaseIE):
     def _real_extract(self, url):
         track_name = self._match_id(url)
 
-        track = self._download_json(self._resolv_url(url), track_name)
+        track = self._download_json(self._resolv_url(url), track_name, headers=self._HEADERS)
         track_id = self._search_regex(
             r'soundcloud:track-stations:(\d+)', track['id'], 'track id')
 
@@ -754,7 +831,7 @@ class SoundcloudPlaylistIE(SoundcloudPlaylistBaseIE):
 
         data = self._download_json(
             self._API_V2_BASE + 'playlists/' + playlist_id,
-            playlist_id, 'Downloading playlist', query=query)
+            playlist_id, 'Downloading playlist', query=query, headers=self._HEADERS)
 
         return self._extract_set(data, token)
 
@@ -791,7 +868,7 @@ class SoundcloudSearchIE(SearchInfoExtractor, SoundcloudIE):
         for i in itertools.count(1):
             response = self._download_json(
                 next_url, collection_id, 'Downloading page {0}'.format(i),
-                'Unable to download API page')
+                'Unable to download API page', headers=self._HEADERS)
 
             collection = response.get('collection', [])
             if not collection:

youtube_dl/extractor/youtube.py

@@ -9,6 +9,7 @@ import json
 import os.path
 import random
 import re
+import string
 import time
 import traceback
 
@@ -67,6 +68,7 @@ from ..utils import (
 
 class YoutubeBaseInfoExtractor(InfoExtractor):
     """Provide base functions for Youtube extractors"""
+
     _LOGIN_URL = 'https://accounts.google.com/ServiceLogin'
     _TWOFACTOR_URL = 'https://accounts.google.com/signin/challenge'
 
@@ -138,7 +140,7 @@ class YoutubeBaseInfoExtractor(InfoExtractor):
                 [2, 1, None, 1,
                  'https://accounts.google.com/ServiceLogin?passive=true&continue=https%3A%2F%2Fwww.youtube.com%2Fsignin%3Fnext%3D%252F%26action_handle_signin%3Dtrue%26hl%3Den%26app%3Ddesktop%26feature%3Dsign_in_button&hl=en&service=youtube&uilel=3&requestPath=%2FServiceLogin&Page=PasswordSeparationSignIn',
                  None, [], 4],
-                1, [None, None, []], None, None, None, True
+                1, [None, None, []], None, None, None, True,
             ],
             username,
         ]
@@ -160,7 +162,7 @@ class YoutubeBaseInfoExtractor(InfoExtractor):
             None, 1, None, [1, None, None, None, [password, None, True]],
             [
                 None, None, [2, 1, None, 1, 'https://accounts.google.com/ServiceLogin?passive=true&continue=https%3A%2F%2Fwww.youtube.com%2Fsignin%3Fnext%3D%252F%26action_handle_signin%3Dtrue%26hl%3Den%26app%3Ddesktop%26feature%3Dsign_in_button&hl=en&service=youtube&uilel=3&requestPath=%2FServiceLogin&Page=PasswordSeparationSignIn', None, [], 4],
-                1, [None, None, []], None, None, None, True
+                1, [None, None, []], None, None, None, True,
             ]]
 
         challenge_results = req(
@@ -213,7 +215,7 @@ class YoutubeBaseInfoExtractor(InfoExtractor):
                     user_hash, None, 2, None,
                     [
                         9, None, None, None, None, None, None, None,
-                        [None, tfa_code, True, 2]
+                        [None, tfa_code, True, 2],
                     ]]
 
                 tfa_results = req(
@@ -284,7 +286,7 @@ class YoutubeBaseInfoExtractor(InfoExtractor):
             'client': {
                 'clientName': 'WEB',
                 'clientVersion': '2.20201021.03.00',
-            }
+            },
         },
     }
 
@@ -385,7 +387,7 @@ class YoutubeBaseInfoExtractor(InfoExtractor):
                 'client': {
                     'clientName': 'WEB',
                     'clientVersion': '2.20201021.03.00',
-                }
+                },
             },
             'query': query,
         }
@@ -462,7 +464,7 @@ class YoutubeBaseInfoExtractor(InfoExtractor):
             #       (HTML, videodetails, metadata, renderers)
             'name': ('content', 'author', (('ownerChannelName', None), 'title'), ['text']),
             'url': ('href', 'ownerProfileUrl', 'vanityChannelUrl',
-                    ['navigationEndpoint', 'browseEndpoint', 'canonicalBaseUrl'])
+                    ['navigationEndpoint', 'browseEndpoint', 'canonicalBaseUrl']),
         }
         if any((videodetails, metadata, renderers)):
             result = (
@@ -671,7 +673,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/8KVIDEO',
                 'description': '',
                 'uploader': '8KVIDEO',
-                'title': 'UHDTV TEST 8K VIDEO.mp4'
+                'title': 'UHDTV TEST 8K VIDEO.mp4',
             },
             'params': {
                 'youtube_include_dash_manifest': True,
@@ -711,7 +713,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                 'uploader_url': r're:https?://(?:www\.)?youtube\.com/@theamazingatheist',
                 'title': 'Burning Everyone\'s Koran',
                 'description': 'SUBSCRIBE: http://www.youtube.com/saturninefilms \r\n\r\nEven Obama has taken a stand against freedom on this issue: http://www.huffingtonpost.com/2010/09/09/obama-gma-interview-quran_n_710282.html',
-            }
+            },
         },
         # Age-gated videos
         {
@@ -839,7 +841,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
             },
             'expected_warnings': [
                 'DASH manifest missing',
-            ]
+            ],
         },
         # Olympics (https://github.com/ytdl-org/youtube-dl/issues/4431)
         {
@@ -1820,8 +1822,8 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
 
         # cpn generation algorithm is reverse engineered from base.js.
         # In fact it works even with dummy cpn.
-        CPN_ALPHABET = 'abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789-_'
-        cpn = ''.join(CPN_ALPHABET[random.randint(0, 256) & 63] for _ in range(0, 16))
+        CPN_ALPHABET = string.ascii_letters + string.digits + '-_'
+        cpn = ''.join(CPN_ALPHABET[random.randint(0, 256) & 63] for _ in range(16))
 
         # more consistent results setting it to right before the end
         qs = parse_qs(playback_url)
@@ -1881,8 +1883,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
         mobj = re.match(cls._VALID_URL, url, re.VERBOSE)
         if mobj is None:
             raise ExtractorError('Invalid URL: %s' % url)
-        video_id = mobj.group(2)
-        return video_id
+        return mobj.group(2)
 
     def _extract_chapters_from_json(self, data, video_id, duration):
         chapters_list = try_get(
@@ -1951,7 +1952,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
             pb_context = {'html5Preference': 'HTML5_PREF_WANTS'}
 
             player_url = self._extract_player_url(webpage)
-            ytcfg = self._extract_ytcfg(video_id, webpage)
+            ytcfg = self._extract_ytcfg(video_id, webpage or '')
             sts = self._extract_signature_timestamp(video_id, player_url, ytcfg)
             if sts:
                 pb_context['signatureTimestamp'] = sts
@@ -2035,7 +2036,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
             headers = {
                 'X-YouTube-Client-Name': '85',
                 'X-YouTube-Client-Version': '2.0',
-                'Origin': 'https://www.youtube.com'
+                'Origin': 'https://www.youtube.com',
             }
 
             video_info = self._call_api('player', query, video_id, fatal=False, headers=headers)
@@ -2064,8 +2065,8 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
             return ''.join([r['text'] for r in runs if isinstance(r.get('text'), compat_str)])
 
         search_meta = (
-            lambda x: self._html_search_meta(x, webpage, default=None)) \
-            if webpage else lambda x: None
+            (lambda x: self._html_search_meta(x, webpage, default=None))
+            if webpage else lambda _: None)
 
         video_details = player_response.get('videoDetails') or {}
         microformat = try_get(
@@ -2137,7 +2138,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
         def build_fragments(f):
             return LazyList({
                 'url': update_url_query(f['url'], {
-                    'range': '{0}-{1}'.format(range_start, min(range_start + CHUNK_SIZE - 1, f['filesize']))
+                    'range': '{0}-{1}'.format(range_start, min(range_start + CHUNK_SIZE - 1, f['filesize'])),
                 })
             } for range_start in range(0, f['filesize'], CHUNK_SIZE))
 
@@ -2236,7 +2237,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                     'protocol': 'http_dash_segments',
                     'fragments': build_fragments(dct),
                 } if dct['filesize'] else {
-                    'downloader_options': {'http_chunk_size': CHUNK_SIZE}  # No longer useful?
+                    'downloader_options': {'http_chunk_size': CHUNK_SIZE},  # No longer useful?
                 })
 
             formats.append(dct)
@@ -2414,9 +2415,9 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
             'is_live': is_live,
         }
 
-        pctr = try_get(
+        pctr = traverse_obj(
             player_response,
-            lambda x: x['captions']['playerCaptionsTracklistRenderer'], dict)
+            ('captions', 'playerCaptionsTracklistRenderer', T(dict)))
         if pctr:
             def process_language(container, base_url, lang_code, query):
                 lang_subs = []
@@ -2430,31 +2431,34 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                     })
                 container[lang_code] = lang_subs
 
-            subtitles = {}
-            for caption_track in (pctr.get('captionTracks') or []):
-                base_url = caption_track.get('baseUrl')
-                if not base_url:
-                    continue
-                if caption_track.get('kind') != 'asr':
-                    lang_code = caption_track.get('languageCode')
-                    if not lang_code:
+            def process_subtitles():
+                subtitles = {}
+                for caption_track in traverse_obj(pctr, (
+                        'captionTracks', lambda _, v: v.get('baseUrl'))):
+                    base_url = self._yt_urljoin(caption_track['baseUrl'])
+                    if not base_url:
                         continue
-                    process_language(
-                        subtitles, base_url, lang_code, {})
-                    continue
-                automatic_captions = {}
-                for translation_language in (pctr.get('translationLanguages') or []):
-                    translation_language_code = translation_language.get('languageCode')
-                    if not translation_language_code:
+                    if caption_track.get('kind') != 'asr':
+                        lang_code = caption_track.get('languageCode')
+                        if not lang_code:
+                            continue
+                        process_language(
+                            subtitles, base_url, lang_code, {})
                         continue
-                    process_language(
-                        automatic_captions, base_url, translation_language_code,
-                        {'tlang': translation_language_code})
-                info['automatic_captions'] = automatic_captions
-            info['subtitles'] = subtitles
+                    automatic_captions = {}
+                    for translation_language in traverse_obj(pctr, (
+                            'translationLanguages', lambda _, v: v.get('languageCode'))):
+                        translation_language_code = translation_language['languageCode']
+                        process_language(
+                            automatic_captions, base_url, translation_language_code,
+                            {'tlang': translation_language_code})
+                    info['automatic_captions'] = automatic_captions
+                info['subtitles'] = subtitles
+
+            process_subtitles()
 
         parsed_url = compat_urllib_parse_urlparse(url)
-        for component in [parsed_url.fragment, parsed_url.query]:
+        for component in (parsed_url.fragment, parsed_url.query):
             query = compat_parse_qs(component)
             for k, v in query.items():
                 for d_k, s_ks in [('start', ('start', 't')), ('end', ('end',))]:
@@ -2684,7 +2688,7 @@ class YoutubeTabIE(YoutubeBaseInfoExtractor):
             'title': 'Super Cooper Shorts - Shorts',
             'uploader': 'Super Cooper Shorts',
             'uploader_id': '@SuperCooperShorts',
-        }
+        },
     }, {
         # Channel that does not have a Shorts tab. Test should just download videos on Home tab instead
         'url': 'https://www.youtube.com/@emergencyawesome/shorts',
@@ -2738,7 +2742,7 @@ class YoutubeTabIE(YoutubeBaseInfoExtractor):
             'description': 'md5:609399d937ea957b0f53cbffb747a14c',
             'uploader': 'ThirstForScience',
             'uploader_id': '@ThirstForScience',
-        }
+        },
     }, {
         'url': 'https://www.youtube.com/c/ChristophLaimer/playlists',
         'only_matching': True,
@@ -3037,7 +3041,7 @@ class YoutubeTabIE(YoutubeBaseInfoExtractor):
             'uploader': '3Blue1Brown',
             'uploader_id': '@3blue1brown',
             'channel_id': 'UCYO_jab_esuFRV4b17AJtAw',
-        }
+        },
     }]
 
     @classmethod
@@ -3335,7 +3339,7 @@ class YoutubeTabIE(YoutubeBaseInfoExtractor):
             'client': {
                 'clientName': 'WEB',
                 'clientVersion': client_version,
-            }
+            },
         }
         visitor_data = try_get(context, lambda x: x['client']['visitorData'], compat_str)
 
@@ -3354,7 +3358,7 @@ class YoutubeTabIE(YoutubeBaseInfoExtractor):
                 headers['x-goog-visitor-id'] = visitor_data
             data['continuation'] = continuation['continuation']
             data['clickTracking'] = {
-                'clickTrackingParams': continuation['itct']
+                'clickTrackingParams': continuation['itct'],
             }
             count = 0
             retries = 3
@@ -3613,7 +3617,7 @@ class YoutubePlaylistIE(InfoExtractor):
             'uploader': 'milan',
             'uploader_id': '@milan5503',
             'channel_id': 'UCEI1-PVPcYXjB73Hfelbmaw',
-        }
+        },
     }, {
         'url': 'http://www.youtube.com/embed/_xDOZElKyNU?list=PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl',
         'playlist_mincount': 455,
@@ -3623,7 +3627,7 @@ class YoutubePlaylistIE(InfoExtractor):
             'uploader': 'LBK',
             'uploader_id': '@music_king',
             'channel_id': 'UC21nz3_MesPLqtDqwdvnoxA',
-        }
+        },
     }, {
         'url': 'TLGGrESM50VT6acwMjAyMjAxNw',
         'only_matching': True,
@@ -3734,7 +3738,7 @@ class YoutubeSearchIE(SearchInfoExtractor, YoutubeBaseInfoExtractor):
         'info_dict': {
             'id': 'youtube-dl test video',
             'title': 'youtube-dl test video',
-        }
+        },
     }]
 
     def _get_n_results(self, query, n):
@@ -3754,7 +3758,7 @@ class YoutubeSearchDateIE(YoutubeSearchIE):
         'info_dict': {
             'id': 'youtube-dl test video',
             'title': 'youtube-dl test video',
-        }
+        },
     }]
 
 
@@ -3769,7 +3773,7 @@ class YoutubeSearchURLIE(YoutubeBaseInfoExtractor):
             'id': 'youtube-dl test video',
             'title': 'youtube-dl test video',
         },
-        'params': {'playlistend': 5}
+        'params': {'playlistend': 5},
     }, {
         'url': 'https://www.youtube.com/results?q=test&sp=EgQIBBgB',
         'only_matching': True,
@@ -3785,6 +3789,7 @@ class YoutubeSearchURLIE(YoutubeBaseInfoExtractor):
 class YoutubeFeedsInfoExtractor(YoutubeTabIE):
     """
     Base class for feed extractors
+
     Subclasses must define the _FEED_NAME property.
     """
     _LOGIN_REQUIRED = True