yt_dlp/extractor/allstar.py

   1 import functools
   2 import json
   3
   4 from .common import InfoExtractor
   5 from ..utils import (
   6     ExtractorError,
   7     OnDemandPagedList,
   8     int_or_none,
   9     join_nonempty,
  10     parse_qs,
  11     urljoin,
  12 )
  13 from ..utils.traversal import traverse_obj
  14
  15
  16 _FIELDS = '''
  17     _id
  18     clipImageSource
  19     clipImageThumb
  20     clipLink
  21     clipTitle
  22     createdDate
  23     shareId
  24     user { _id }
  25     username
  26     views'''
  27
  28 _EXTRA_FIELDS = '''
  29     clipLength
  30     clipSizeBytes'''
  31
  32 _QUERIES = {
  33     'clip': '''query ($id: String!) {
  34         video: getClip(clipIdentifier: $id) {
  35             %s %s
  36         }
  37     }''' % (_FIELDS, _EXTRA_FIELDS),
  38     'montage': '''query ($id: String!) {
  39         video: getMontage(clipIdentifier: $id) {
  40             %s
  41         }
  42     }''' % _FIELDS,
  43     'Clips': '''query ($page: Int!, $user: String!, $game: Int) {
  44         videos: clips(search: createdDate, page: $page, user: $user, mobile: false, game: $game) {
  45             data { %s %s }
  46         }
  47     }''' % (_FIELDS, _EXTRA_FIELDS),
  48     'Montages': '''query ($page: Int!, $user: String!) {
  49         videos: montages(search: createdDate, page: $page, user: $user) {
  50             data { %s }
  51         }
  52     }''' % _FIELDS,
  53     'Mobile Clips': '''query ($page: Int!, $user: String!) {
  54         videos: clips(search: createdDate, page: $page, user: $user, mobile: true) {
  55             data { %s %s }
  56         }
  57     }''' % (_FIELDS, _EXTRA_FIELDS),
  58 }
  59
  60
  61 class AllstarBaseIE(InfoExtractor):
  62     @staticmethod
  63     def _parse_video_data(video_data):
  64         def media_url_or_none(path):
  65             return urljoin('https://media.allstar.gg/', path)
  66
  67         info = traverse_obj(video_data, {
  68             'id': ('_id', {str}),
  69             'display_id': ('shareId', {str}),
  70             'title': ('clipTitle', {str}),
  71             'url': ('clipLink', {media_url_or_none}),
  72             'thumbnails': (('clipImageThumb', 'clipImageSource'), {'url': {media_url_or_none}}),
  73             'duration': ('clipLength', {int_or_none}),
  74             'filesize': ('clipSizeBytes', {int_or_none}),
  75             'timestamp': ('createdDate', {functools.partial(int_or_none, scale=1000)}),
  76             'uploader': ('username', {str}),
  77             'uploader_id': ('user', '_id', {str}),
  78             'view_count': ('views', {int_or_none}),
  79         })
  80
  81         if info.get('id') and info.get('url'):
  82             basename = 'clip' if '/clips/' in info['url'] else 'montage'
  83             info['webpage_url'] = f'https://allstar.gg/{basename}?{basename}={info["id"]}'
  84
  85         info.update({
  86             'extractor_key': AllstarIE.ie_key(),
  87             'extractor': AllstarIE.IE_NAME,
  88             'uploader_url': urljoin('https://allstar.gg/u/', info.get('uploader_id')),
  89         })
  90
  91         return info
  92
  93     def _call_api(self, query, variables, path, video_id=None, note=None):
  94         response = self._download_json(
  95             'https://a1.allstar.gg/graphql', video_id, note=note,
  96             headers={'content-type': 'application/json'},
  97             data=json.dumps({'variables': variables, 'query': query}).encode())
  98
  99         errors = traverse_obj(response, ('errors', ..., 'message', {str}))
 100         if errors:
 101             raise ExtractorError('; '.join(errors))
 102
 103         return traverse_obj(response, path)
 104
 105
 106 class AllstarIE(AllstarBaseIE):
 107     _VALID_URL = r'https?://(?:www\.)?allstar\.gg/(?P<type>(?:clip|montage))\?(?P=type)=(?P<id>[^/?#&]+)'
 108
 109     _TESTS = [{
 110         'url': 'https://allstar.gg/clip?clip=64482c2da9eec30008a67d1b',
 111         'info_dict': {
 112             'id': '64482c2da9eec30008a67d1b',
 113             'title': '4K on Inferno',
 114             'url': 'md5:66befb5381eef0c9456026386c25fa55',
 115             'thumbnail': r're:https://media\.allstar\.gg/.+\.(?:png|jpg)$',
 116             'uploader': 'chrk.',
 117             'ext': 'mp4',
 118             'duration': 20,
 119             'filesize': 21199257,
 120             'timestamp': 1682451501,
 121             'uploader_id': '62b8bdfc9021052f7905882d',
 122             'uploader_url': 'https://allstar.gg/u/62b8bdfc9021052f7905882d',
 123             'upload_date': '20230425',
 124             'view_count': int,
 125         }
 126     }, {
 127         'url': 'https://allstar.gg/clip?clip=8LJLY4JKB',
 128         'info_dict': {
 129             'id': '64a1ec6b887f4c0008dc50b8',
 130             'display_id': '8LJLY4JKB',
 131             'title': 'AK-47 3K on Mirage',
 132             'url': 'md5:dde224fd12f035c0e2529a4ae34c4283',
 133             'ext': 'mp4',
 134             'thumbnail': r're:https://media\.allstar\.gg/.+\.(?:png|jpg)$',
 135             'duration': 16,
 136             'filesize': 30175859,
 137             'timestamp': 1688333419,
 138             'uploader': 'cherokee',
 139             'uploader_id': '62b8bdfc9021052f7905882d',
 140             'uploader_url': 'https://allstar.gg/u/62b8bdfc9021052f7905882d',
 141             'upload_date': '20230702',
 142             'view_count': int,
 143         }
 144     }, {
 145         'url': 'https://allstar.gg/montage?montage=643e64089da7e9363e1fa66c',
 146         'info_dict': {
 147             'id': '643e64089da7e9363e1fa66c',
 148             'display_id': 'APQLGM2IMXW',
 149             'title': 'cherokee Rapid Fire Snipers Montage',
 150             'url': 'md5:a3ee356022115db2b27c81321d195945',
 151             'thumbnail': r're:https://media\.allstar\.gg/.+\.(?:png|jpg)$',
 152             'ext': 'mp4',
 153             'timestamp': 1681810448,
 154             'uploader': 'cherokee',
 155             'uploader_id': '62b8bdfc9021052f7905882d',
 156             'uploader_url': 'https://allstar.gg/u/62b8bdfc9021052f7905882d',
 157             'upload_date': '20230418',
 158             'view_count': int,
 159         }
 160     }, {
 161         'url': 'https://allstar.gg/montage?montage=RILJMH6QOS',
 162         'info_dict': {
 163             'id': '64a2697372ce3703de29e868',
 164             'display_id': 'RILJMH6QOS',
 165             'title': 'cherokee Rapid Fire Snipers Montage',
 166             'url': 'md5:d5672e6f88579730c2310a80fdbc4030',
 167             'thumbnail': r're:https://media\.allstar\.gg/.+\.(?:png|jpg)$',
 168             'ext': 'mp4',
 169             'timestamp': 1688365434,
 170             'uploader': 'cherokee',
 171             'uploader_id': '62b8bdfc9021052f7905882d',
 172             'uploader_url': 'https://allstar.gg/u/62b8bdfc9021052f7905882d',
 173             'upload_date': '20230703',
 174             'view_count': int,
 175         }
 176     }]
 177
 178     def _real_extract(self, url):
 179         query_id, video_id = self._match_valid_url(url).group('type', 'id')
 180
 181         return self._parse_video_data(
 182             self._call_api(
 183                 _QUERIES.get(query_id), {'id': video_id}, ('data', 'video'), video_id))
 184
 185
 186 class AllstarProfileIE(AllstarBaseIE):
 187     _VALID_URL = r'https?://(?:www\.)?allstar\.gg/(?:profile\?user=|u/)(?P<id>[^/?#&]+)'
 188
 189     _TESTS = [{
 190         'url': 'https://allstar.gg/profile?user=62b8bdfc9021052f7905882d',
 191         'info_dict': {
 192             'id': '62b8bdfc9021052f7905882d-clips',
 193             'title': 'cherokee - Clips',
 194         },
 195         'playlist_mincount': 15
 196     }, {
 197         'url': 'https://allstar.gg/u/cherokee?game=730&view=Clips',
 198         'info_dict': {
 199             'id': '62b8bdfc9021052f7905882d-clips-730',
 200             'title': 'cherokee - Clips - 730',
 201         },
 202         'playlist_mincount': 15
 203     }, {
 204         'url': 'https://allstar.gg/u/62b8bdfc9021052f7905882d?view=Montages',
 205         'info_dict': {
 206             'id': '62b8bdfc9021052f7905882d-montages',
 207             'title': 'cherokee - Montages',
 208         },
 209         'playlist_mincount': 4
 210     }, {
 211         'url': 'https://allstar.gg/profile?user=cherokee&view=Mobile Clips',
 212         'info_dict': {
 213             'id': '62b8bdfc9021052f7905882d-mobile',
 214             'title': 'cherokee - Mobile Clips',
 215         },
 216         'playlist_mincount': 1
 217     }]
 218
 219     _PAGE_SIZE = 10
 220
 221     def _get_page(self, user_id, display_id, game, query, page_num):
 222         page_num += 1
 223
 224         for video_data in self._call_api(
 225                 query, {
 226                     'user': user_id,
 227                     'page': page_num,
 228                     'game': game,
 229                 }, ('data', 'videos', 'data'), display_id, f'Downloading page {page_num}'):
 230             yield self._parse_video_data(video_data)
 231
 232     def _real_extract(self, url):
 233         display_id = self._match_id(url)
 234         profile_data = self._download_json(
 235             urljoin('https://api.allstar.gg/v1/users/profile/', display_id), display_id)
 236         user_id = traverse_obj(profile_data, ('data', ('_id'), {str}))
 237         if not user_id:
 238             raise ExtractorError('Unable to extract the user id')
 239
 240         username = traverse_obj(profile_data, ('data', 'profile', ('username'), {str}))
 241         url_query = parse_qs(url)
 242         game = traverse_obj(url_query, ('game', 0, {int_or_none}))
 243         query_id = traverse_obj(url_query, ('view', 0), default='Clips')
 244
 245         if query_id not in ('Clips', 'Montages', 'Mobile Clips'):
 246             raise ExtractorError(f'Unsupported playlist URL type {query_id!r}')
 247
 248         return self.playlist_result(
 249             OnDemandPagedList(
 250                 functools.partial(
 251                     self._get_page, user_id, display_id, game, _QUERIES.get(query_id)), self._PAGE_SIZE),
 252             playlist_id=join_nonempty(user_id, query_id.lower().split()[0], game),
 253             playlist_title=join_nonempty((username or display_id), query_id, game, delim=' - '))