yt_dlp/extractor/naver.py

   1 import base64
   2 import hashlib
   3 import hmac
   4 import itertools
   5 import json
   6 import re
   7 import time
   8 import urllib.parse
   9
  10 from .common import InfoExtractor
  11 from ..utils import (
  12     ExtractorError,
  13     dict_get,
  14     int_or_none,
  15     join_nonempty,
  16     merge_dicts,
  17     parse_iso8601,
  18     traverse_obj,
  19     try_get,
  20     unified_timestamp,
  21     update_url_query,
  22     url_or_none,
  23 )
  24
  25
  26 class NaverBaseIE(InfoExtractor):
  27     _CAPTION_EXT_RE = r'\.(?:ttml|vtt)'
  28
  29     @staticmethod  # NB: Used in WeverseIE
  30     def process_subtitles(vod_data, process_url):
  31         ret = {'subtitles': {}, 'automatic_captions': {}}
  32         for caption in traverse_obj(vod_data, ('captions', 'list', ...)):
  33             caption_url = caption.get('source')
  34             if not caption_url:
  35                 continue
  36             type_ = 'automatic_captions' if caption.get('type') == 'auto' else 'subtitles'
  37             lang = caption.get('locale') or join_nonempty('language', 'country', from_dict=caption) or 'und'
  38             if caption.get('type') == 'fan':
  39                 lang += '_fan{}'.format(next(i for i in itertools.count(1) if f'{lang}_fan{i}' not in ret[type_]))
  40             ret[type_].setdefault(lang, []).extend({
  41                 'url': sub_url,
  42                 'name': join_nonempty('label', 'fanName', from_dict=caption, delim=' - '),
  43             } for sub_url in process_url(caption_url))
  44         return ret
  45
  46     def _extract_video_info(self, video_id, vid, key):
  47         video_data = self._download_json(
  48             'http://play.rmcnmv.naver.com/vod/play/v2.0/' + vid,
  49             video_id, query={
  50                 'key': key,
  51             })
  52         meta = video_data['meta']
  53         title = meta['subject']
  54         formats = []
  55         get_list = lambda x: try_get(video_data, lambda y: y[x + 's']['list'], list) or []
  56
  57         def extract_formats(streams, stream_type, query={}):
  58             for stream in streams:
  59                 stream_url = stream.get('source')
  60                 if not stream_url:
  61                     continue
  62                 stream_url = update_url_query(stream_url, query)
  63                 encoding_option = stream.get('encodingOption', {})
  64                 bitrate = stream.get('bitrate', {})
  65                 formats.append({
  66                     'format_id': '{}_{}'.format(stream.get('type') or stream_type, dict_get(encoding_option, ('name', 'id'))),
  67                     'url': stream_url,
  68                     'ext': 'mp4',
  69                     'width': int_or_none(encoding_option.get('width')),
  70                     'height': int_or_none(encoding_option.get('height')),
  71                     'vbr': int_or_none(bitrate.get('video')),
  72                     'abr': int_or_none(bitrate.get('audio')),
  73                     'filesize': int_or_none(stream.get('size')),
  74                     'protocol': 'm3u8_native' if stream_type == 'HLS' else None,
  75                 })
  76
  77         extract_formats(get_list('video'), 'H264')
  78         for stream_set in video_data.get('streams', []):
  79             query = {}
  80             for param in stream_set.get('keys', []):
  81                 query[param['name']] = param['value']
  82             stream_type = stream_set.get('type')
  83             videos = stream_set.get('videos')
  84             if videos:
  85                 extract_formats(videos, stream_type, query)
  86             elif stream_type == 'HLS':
  87                 stream_url = stream_set.get('source')
  88                 if not stream_url:
  89                     continue
  90                 formats.extend(self._extract_m3u8_formats(
  91                     update_url_query(stream_url, query), video_id,
  92                     'mp4', 'm3u8_native', m3u8_id=stream_type, fatal=False))
  93
  94         replace_ext = lambda x, y: re.sub(self._CAPTION_EXT_RE, '.' + y, x)
  95
  96         def get_subs(caption_url):
  97             if re.search(self._CAPTION_EXT_RE, caption_url):
  98                 return [
  99                     replace_ext(caption_url, 'ttml'),
 100                     replace_ext(caption_url, 'vtt'),
 101                 ]
 102             return [caption_url]
 103
 104         user = meta.get('user', {})
 105
 106         return {
 107             'id': video_id,
 108             'title': title,
 109             'formats': formats,
 110             'thumbnail': try_get(meta, lambda x: x['cover']['source']),
 111             'view_count': int_or_none(meta.get('count')),
 112             'uploader_id': user.get('id'),
 113             'uploader': user.get('name'),
 114             'uploader_url': user.get('url'),
 115             **self.process_subtitles(video_data, get_subs),
 116         }
 117
 118     def _call_api(self, path, video_id):
 119         api_endpoint = f'https://apis.naver.com/now_web2/now_web_api/v1{path}'
 120         key = b'nbxvs5nwNG9QKEWK0ADjYA4JZoujF4gHcIwvoCxFTPAeamq5eemvt5IWAYXxrbYM'
 121         msgpad = int(time.time() * 1000)
 122         md = base64.b64encode(hmac.HMAC(
 123             key, f'{api_endpoint[:255]}{msgpad}'.encode(), digestmod=hashlib.sha1).digest()).decode()
 124
 125         return self._download_json(api_endpoint, video_id=video_id, headers=self.geo_verification_headers(), query={
 126             'msgpad': msgpad,
 127             'md': md,
 128         })['result']
 129
 130
 131 class NaverIE(NaverBaseIE):
 132     _VALID_URL = r'https?://(?:m\.)?tv(?:cast)?\.naver\.com/(?:v|embed)/(?P<id>\d+)'
 133     _GEO_BYPASS = False
 134     _TESTS = [{
 135         'url': 'http://tv.naver.com/v/81652',
 136         'info_dict': {
 137             'id': '81652',
 138             'ext': 'mp4',
 139             'title': '[9월 모의고사 해설강의][수학_김상희] 수학 A형 16~20번',
 140             'description': '메가스터디 수학 김상희 선생님이 9월 모의고사 수학A형 16번에서 20번까지 해설강의를 공개합니다.',
 141             'timestamp': 1378200754,
 142             'upload_date': '20130903',
 143             'uploader': '메가스터디, 합격불변의 법칙',
 144             'uploader_id': 'megastudy',
 145             'uploader_url': 'https://tv.naver.com/megastudy',
 146             'view_count': int,
 147             'like_count': int,
 148             'comment_count': int,
 149             'duration': 2118,
 150             'thumbnail': r're:^https?://.*\.jpg',
 151         },
 152     }, {
 153         'url': 'http://tv.naver.com/v/395837',
 154         'md5': '7791205fa89dbed2f5e3eb16d287ff05',
 155         'info_dict': {
 156             'id': '395837',
 157             'ext': 'mp4',
 158             'title': '9년이 지나도 아픈 기억, 전효성의 아버지',
 159             'description': 'md5:c76be23e21403a6473d8119678cdb5cb',
 160             'timestamp': 1432030253,
 161             'upload_date': '20150519',
 162             'uploader': '4가지쇼',
 163             'uploader_id': '4show',
 164             'uploader_url': 'https://tv.naver.com/4show',
 165             'view_count': int,
 166             'like_count': int,
 167             'comment_count': int,
 168             'duration': 277,
 169             'thumbnail': r're:^https?://.*\.jpg',
 170         },
 171     }, {
 172         'url': 'http://tvcast.naver.com/v/81652',
 173         'only_matching': True,
 174     }]
 175
 176     def _real_extract(self, url):
 177         video_id = self._match_id(url)
 178         data = self._call_api(f'/clips/{video_id}/play-info', video_id)
 179
 180         vid = traverse_obj(data, ('clip', 'videoId', {str}))
 181         in_key = traverse_obj(data, ('play', 'inKey', {str}))
 182
 183         if not vid or not in_key:
 184             raise ExtractorError('Unable to extract video info')
 185
 186         info = self._extract_video_info(video_id, vid, in_key)
 187         info.update(traverse_obj(data, ('clip', {
 188             'title': 'title',
 189             'description': 'description',
 190             'timestamp': ('firstExposureDatetime', {parse_iso8601}),
 191             'duration': ('playTime', {int_or_none}),
 192             'like_count': ('likeItCount', {int_or_none}),
 193             'view_count': ('playCount', {int_or_none}),
 194             'comment_count': ('commentCount', {int_or_none}),
 195             'thumbnail': ('thumbnailImageUrl', {url_or_none}),
 196             'uploader': 'channelName',
 197             'uploader_id': 'channelId',
 198             'uploader_url': ('channelUrl', {url_or_none}),
 199             'age_limit': ('adultVideo', {lambda x: 19 if x else None}),
 200         })))
 201         return info
 202
 203
 204 class NaverLiveIE(NaverBaseIE):
 205     IE_NAME = 'Naver:live'
 206     _VALID_URL = r'https?://(?:m\.)?tv(?:cast)?\.naver\.com/l/(?P<id>\d+)'
 207     _GEO_BYPASS = False
 208     _TESTS = [{
 209         'url': 'https://tv.naver.com/l/127062',
 210         'info_dict': {
 211             'id': '127062',
 212             'ext': 'mp4',
 213             'live_status': 'is_live',
 214             'channel': '뉴스는 YTN',
 215             'channel_id': 'ytnnews24',
 216             'title': 're:^대한민국 24시간 뉴스 채널 [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$',
 217             'description': 'md5:f938b5956711beab6f882314ffadf4d5',
 218             'start_time': 1677752280,
 219             'thumbnail': r're:^https?://.*\.(jpg|jpeg|png)',
 220             'like_count': int,
 221         },
 222     }, {
 223         'url': 'https://tv.naver.com/l/140535',
 224         'info_dict': {
 225             'id': '140535',
 226             'ext': 'mp4',
 227             'live_status': 'is_live',
 228             'channel': 'KBS뉴스',
 229             'channel_id': 'kbsnews',
 230             'start_time': 1696867320,
 231             'title': 're:^언제 어디서나! KBS 뉴스 24 [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$',
 232             'description': 'md5:6ad419c0bf2f332829bda3f79c295284',
 233             'thumbnail': r're:^https?://.*\.(jpg|jpeg|png)',
 234             'like_count': int,
 235         },
 236     }, {
 237         'url': 'https://tv.naver.com/l/54887',
 238         'only_matching': True,
 239     }]
 240
 241     def _real_extract(self, url):
 242         video_id = self._match_id(url)
 243         data = self._call_api(f'/live-end/normal/{video_id}/play-info?renewLastPlayDate=true', video_id)
 244
 245         status = traverse_obj(data, ('live', 'liveStatus'))
 246         if status == 'CLOSED':
 247             raise ExtractorError('Stream is offline.', expected=True)
 248         elif status != 'OPENED':
 249             raise ExtractorError(f'Unknown status {status!r}')
 250
 251         return {
 252             'id': video_id,
 253             'formats': self._extract_m3u8_formats(
 254                 traverse_obj(data, ('playbackBody', {json.loads}, 'media', 0, 'path')), video_id, live=True),
 255             **traverse_obj(data, ('live', {
 256                 'title': 'title',
 257                 'channel': 'channelName',
 258                 'channel_id': 'channelId',
 259                 'description': 'description',
 260                 'like_count': (('likeCount', 'likeItCount'), {int_or_none}),
 261                 'thumbnail': ('thumbnailImageUrl', {url_or_none}),
 262                 'start_time': (('startTime', 'startDateTime', 'startYmdt'), {parse_iso8601}),
 263             }), get_all=False),
 264             'is_live': True,
 265         }
 266
 267
 268 class NaverNowIE(NaverBaseIE):
 269     IE_NAME = 'navernow'
 270     _VALID_URL = r'https?://now\.naver\.com/s/now\.(?P<id>\w+)'
 271     _API_URL = 'https://apis.naver.com/now_web/oldnow_web/v4'
 272     _TESTS = [{
 273         'url': 'https://now.naver.com/s/now.4759?shareReplayId=26331132#replay=',
 274         'md5': 'e05854162c21c221481de16b2944a0bc',
 275         'info_dict': {
 276             'id': '4759-26331132',
 277             'title': '아이키X노제\r\n💖꽁냥꽁냥💖(1)',
 278             'ext': 'mp4',
 279             'thumbnail': r're:^https?://.*\.jpg',
 280             'timestamp': 1650369600,
 281             'upload_date': '20220419',
 282             'uploader_id': 'now',
 283             'view_count': int,
 284             'uploader_url': 'https://now.naver.com/show/4759',
 285             'uploader': '아이키의 떰즈업',
 286         },
 287         'params': {
 288             'noplaylist': True,
 289         },
 290     }, {
 291         'url': 'https://now.naver.com/s/now.4759?shareHightlight=26601461#highlight=',
 292         'md5': '9f6118e398aa0f22b2152f554ea7851b',
 293         'info_dict': {
 294             'id': '4759-26601461',
 295             'title': '아이키: 나 리정한테 흔들렸어,,, 질투 폭발하는 노제 여보😾 [아이키의 떰즈업]ㅣ네이버 NOW.',
 296             'ext': 'mp4',
 297             'thumbnail': r're:^https?://.*\.jpg',
 298             'upload_date': '20220504',
 299             'timestamp': 1651648311,
 300             'uploader_id': 'now',
 301             'view_count': int,
 302             'uploader_url': 'https://now.naver.com/show/4759',
 303             'uploader': '아이키의 떰즈업',
 304         },
 305         'params': {
 306             'noplaylist': True,
 307         },
 308     }, {
 309         'url': 'https://now.naver.com/s/now.4759',
 310         'info_dict': {
 311             'id': '4759',
 312             'title': '아이키의 떰즈업',
 313         },
 314         'playlist_mincount': 101,
 315     }, {
 316         'url': 'https://now.naver.com/s/now.4759?shareReplayId=26331132#replay',
 317         'info_dict': {
 318             'id': '4759',
 319             'title': '아이키의 떰즈업',
 320         },
 321         'playlist_mincount': 101,
 322     }, {
 323         'url': 'https://now.naver.com/s/now.4759?shareHightlight=26601461#highlight=',
 324         'info_dict': {
 325             'id': '4759',
 326             'title': '아이키의 떰즈업',
 327         },
 328         'playlist_mincount': 101,
 329     }, {
 330         'url': 'https://now.naver.com/s/now.kihyunplay?shareReplayId=30573291#replay',
 331         'only_matching': True,
 332     }]
 333
 334     def _extract_replay(self, show_id, replay_id):
 335         vod_info = self._download_json(f'{self._API_URL}/shows/now.{show_id}/vod/{replay_id}', replay_id)
 336         in_key = self._download_json(f'{self._API_URL}/shows/now.{show_id}/vod/{replay_id}/inkey', replay_id)['inKey']
 337         return merge_dicts({
 338             'id': f'{show_id}-{replay_id}',
 339             'title': traverse_obj(vod_info, ('episode', 'title')),
 340             'timestamp': unified_timestamp(traverse_obj(vod_info, ('episode', 'start_time'))),
 341             'thumbnail': vod_info.get('thumbnail_image_url'),
 342         }, self._extract_video_info(replay_id, vod_info['video_id'], in_key))
 343
 344     def _extract_show_replays(self, show_id):
 345         page_size = 15
 346         page = 1
 347         while True:
 348             show_vod_info = self._download_json(
 349                 f'{self._API_URL}/vod-shows/now.{show_id}', show_id,
 350                 query={'page': page, 'page_size': page_size},
 351                 note=f'Downloading JSON vod list for show {show_id} - page {page}',
 352             )['response']['result']
 353             for v in show_vod_info.get('vod_list') or []:
 354                 yield self._extract_replay(show_id, v['id'])
 355
 356             if len(show_vod_info.get('vod_list') or []) < page_size:
 357                 break
 358             page += 1
 359
 360     def _extract_show_highlights(self, show_id, highlight_id=None):
 361         page_size = 10
 362         page = 1
 363         while True:
 364             highlights_videos = self._download_json(
 365                 f'{self._API_URL}/shows/now.{show_id}/highlights/videos/', show_id,
 366                 query={'page': page, 'page_size': page_size},
 367                 note=f'Downloading JSON highlights for show {show_id} - page {page}')
 368
 369             for highlight in highlights_videos.get('results') or []:
 370                 if highlight_id and highlight.get('clip_no') != int(highlight_id):
 371                     continue
 372                 yield merge_dicts({
 373                     'id': f'{show_id}-{highlight["clip_no"]}',
 374                     'title': highlight.get('title'),
 375                     'timestamp': unified_timestamp(highlight.get('regdate')),
 376                     'thumbnail': highlight.get('thumbnail_url'),
 377                 }, self._extract_video_info(highlight['clip_no'], highlight['video_id'], highlight['video_inkey']))
 378
 379             if len(highlights_videos.get('results') or []) < page_size:
 380                 break
 381             page += 1
 382
 383     def _extract_highlight(self, show_id, highlight_id):
 384         try:
 385             return next(self._extract_show_highlights(show_id, highlight_id))
 386         except StopIteration:
 387             raise ExtractorError(f'Unable to find highlight {highlight_id} for show {show_id}')
 388
 389     def _real_extract(self, url):
 390         show_id = self._match_id(url)
 391         qs = urllib.parse.parse_qs(urllib.parse.urlparse(url).query)
 392
 393         if not self._yes_playlist(show_id, qs.get('shareHightlight')):
 394             return self._extract_highlight(show_id, qs['shareHightlight'][0])
 395         elif not self._yes_playlist(show_id, qs.get('shareReplayId')):
 396             return self._extract_replay(show_id, qs['shareReplayId'][0])
 397
 398         show_info = self._download_json(
 399             f'{self._API_URL}/shows/now.{show_id}/', show_id,
 400             note=f'Downloading JSON vod list for show {show_id}')
 401
 402         return self.playlist_result(
 403             itertools.chain(self._extract_show_replays(show_id), self._extract_show_highlights(show_id)),
 404             show_id, show_info.get('title'))