yt_dlp/extractor/litv.py

   1 import json
   2 import uuid
   3
   4 from .common import InfoExtractor
   5 from ..utils import (
   6     ExtractorError,
   7     int_or_none,
   8     join_nonempty,
   9     smuggle_url,
  10     traverse_obj,
  11     try_call,
  12     unsmuggle_url,
  13     urljoin,
  14 )
  15
  16
  17 class LiTVIE(InfoExtractor):
  18     _VALID_URL = r'https?://(?:www\.)?litv\.tv/(?:[^/?#]+/watch/|vod/[^/?#]+/content\.do\?content_id=)(?P<id>[\w-]+)'
  19     _URL_TEMPLATE = 'https://www.litv.tv/%s/watch/%s'
  20     _GEO_COUNTRIES = ['TW']
  21     _TESTS = [{
  22         'url': 'https://www.litv.tv/drama/watch/VOD00041610',
  23         'info_dict': {
  24             'id': 'VOD00041606',
  25             'title': '花千骨',
  26         },
  27         'playlist_count': 51,  # 50 episodes + 1 trailer
  28     }, {
  29         'url': 'https://www.litv.tv/drama/watch/VOD00041610',
  30         'md5': 'b90ff1e9f1d8f5cfcd0a44c3e2b34c7a',
  31         'info_dict': {
  32             'id': 'VOD00041610',
  33             'ext': 'mp4',
  34             'title': '花千骨第1集',
  35             'thumbnail': r're:https?://.*\.jpg$',
  36             'description': '《花千骨》陸劇線上看。十六年前，平靜的村莊內，一名女嬰隨異相出生，途徑此地的蜀山掌門清虛道長算出此女命運非同一般，她體內散發的異香易招惹妖魔。一念慈悲下，他在村莊周邊設下結界阻擋妖魔入侵，讓其年滿十六後去蜀山，並賜名花千骨。',
  37             'categories': ['奇幻', '愛情', '仙俠', '古裝'],
  38             'episode': 'Episode 1',
  39             'episode_number': 1,
  40         },
  41         'params': {
  42             'noplaylist': True,
  43         },
  44     }, {
  45         'url': 'https://www.litv.tv/drama/watch/VOD00044841',
  46         'md5': '88322ea132f848d6e3e18b32a832b918',
  47         'info_dict': {
  48             'id': 'VOD00044841',
  49             'ext': 'mp4',
  50             'title': '芈月傳第1集　霸星芈月降世楚國',
  51             'description': '楚威王二年，太史令唐昧夜觀星象，發現霸星即將現世。王后得知霸星的預言後，想盡辦法不讓孩子順利出生，幸得莒姬相護化解危機。沒想到眾人期待下出生的霸星卻是位公主，楚威王對此失望至極。楚王后命人將女嬰丟棄河中，居然奇蹟似的被少司命像攔下，楚威王認為此女非同凡響，為她取名芈月。',
  52         },
  53         'skip': 'No longer exists',
  54     }]
  55
  56     def _extract_playlist(self, playlist_data, content_type):
  57         all_episodes = [
  58             self.url_result(smuggle_url(
  59                 self._URL_TEMPLATE % (content_type, episode['content_id']),
  60                 {'force_noplaylist': True}))  # To prevent infinite recursion
  61             for episode in traverse_obj(playlist_data, ('seasons', ..., 'episodes', lambda _, v: v['content_id']))]
  62
  63         return self.playlist_result(all_episodes, playlist_data['content_id'], playlist_data.get('title'))
  64
  65     def _real_extract(self, url):
  66         url, smuggled_data = unsmuggle_url(url, {})
  67         video_id = self._match_id(url)
  68         webpage = self._download_webpage(url, video_id)
  69         vod_data = self._search_nextjs_data(webpage, video_id)['props']['pageProps']
  70
  71         program_info = traverse_obj(vod_data, ('programInformation', {dict})) or {}
  72         playlist_data = traverse_obj(vod_data, ('seriesTree'))
  73         if playlist_data and self._yes_playlist(program_info.get('series_id'), video_id, smuggled_data):
  74             return self._extract_playlist(playlist_data, program_info.get('content_type'))
  75
  76         asset_id = traverse_obj(program_info, ('assets', 0, 'asset_id', {str}))
  77         if asset_id:  # This is a VOD
  78             media_type = 'vod'
  79         else:  # This is a live stream
  80             asset_id = program_info['content_id']
  81             media_type = program_info['content_type']
  82         puid = try_call(lambda: self._get_cookies('https://www.litv.tv/')['PUID'].value)
  83         if puid:
  84             endpoint = 'get-urls'
  85         else:
  86             puid = str(uuid.uuid4())
  87             endpoint = 'get-urls-no-auth'
  88         video_data = self._download_json(
  89             f'https://www.litv.tv/api/{endpoint}', video_id,
  90             data=json.dumps({'AssetId': asset_id, 'MediaType': media_type, 'puid': puid}).encode(),
  91             headers={'Content-Type': 'application/json'})
  92
  93         if error := traverse_obj(video_data, ('error', {dict})):
  94             error_msg = traverse_obj(error, ('message', {str}))
  95             if error_msg and 'OutsideRegionError' in error_msg:
  96                 self.raise_geo_restricted('This video is available in Taiwan only')
  97             elif error_msg:
  98                 raise ExtractorError(f'{self.IE_NAME} said: {error_msg}', expected=True)
  99             raise ExtractorError(f'Unexpected error from {self.IE_NAME}')
 100
 101         formats = self._extract_m3u8_formats(
 102             video_data['result']['AssetURLs'][0], video_id, ext='mp4', m3u8_id='hls')
 103         for a_format in formats:
 104             # LiTV HLS segments doesn't like compressions
 105             a_format.setdefault('http_headers', {})['Accept-Encoding'] = 'identity'
 106
 107         return {
 108             'id': video_id,
 109             'formats': formats,
 110             'title': join_nonempty('title', 'secondary_mark', delim='', from_dict=program_info),
 111             **traverse_obj(program_info, {
 112                 'description': ('description', {str}),
 113                 'thumbnail': ('picture', {urljoin('https://p-cdnstatic.svc.litv.tv/')}),
 114                 'categories': ('genres', ..., 'name', {str}),
 115                 'episode_number': ('episode', {int_or_none}),
 116             }),
 117         }