yt_dlp/extractor/litv.py

   1 import json
   2
   3 from .common import InfoExtractor
   4 from ..utils import (
   5     ExtractorError,
   6     int_or_none,
   7     smuggle_url,
   8     traverse_obj,
   9     try_call,
  10     unsmuggle_url,
  11 )
  12
  13
  14 class LiTVIE(InfoExtractor):
  15     _VALID_URL = r'https?://(?:www\.)?litv\.tv/(?:vod|promo)/[^/]+/(?:content\.do)?\?.*?\b(?:content_)?id=(?P<id>[^&]+)'
  16
  17     _URL_TEMPLATE = 'https://www.litv.tv/vod/%s/content.do?content_id=%s'
  18
  19     _TESTS = [{
  20         'url': 'https://www.litv.tv/vod/drama/content.do?brc_id=root&id=VOD00041610&isUHEnabled=true&autoPlay=1',
  21         'info_dict': {
  22             'id': 'VOD00041606',
  23             'title': '花千骨',
  24         },
  25         'playlist_count': 51,  # 50 episodes + 1 trailer
  26     }, {
  27         'url': 'https://www.litv.tv/vod/drama/content.do?brc_id=root&id=VOD00041610&isUHEnabled=true&autoPlay=1',
  28         'md5': 'b90ff1e9f1d8f5cfcd0a44c3e2b34c7a',
  29         'info_dict': {
  30             'id': 'VOD00041610',
  31             'ext': 'mp4',
  32             'title': '花千骨第1集',
  33             'thumbnail': r're:https?://.*\.jpg$',
  34             '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。',
  35             'categories': ['奇幻', '愛情', '中國', '仙俠'],
  36             'episode': 'Episode 1',
  37             'episode_number': 1,
  38         },
  39         'params': {
  40             'noplaylist': True,
  41         },
  42         'skip': 'Georestricted to Taiwan',
  43     }, {
  44         'url': 'https://www.litv.tv/promo/miyuezhuan/?content_id=VOD00044841&',
  45         'md5': '88322ea132f848d6e3e18b32a832b918',
  46         'info_dict': {
  47             'id': 'VOD00044841',
  48             'ext': 'mp4',
  49             'title': '芈月傳第1集　霸星芈月降世楚國',
  50             '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。',
  51         },
  52         'skip': 'No longer exists',
  53     }]
  54
  55     def _extract_playlist(self, playlist_data, content_type):
  56         all_episodes = [
  57             self.url_result(smuggle_url(
  58                 self._URL_TEMPLATE % (content_type, episode['contentId']),
  59                 {'force_noplaylist': True}))  # To prevent infinite recursion
  60             for episode in traverse_obj(playlist_data, ('seasons', ..., 'episode', lambda _, v: v['contentId']))]
  61
  62         return self.playlist_result(all_episodes, playlist_data['contentId'], playlist_data.get('title'))
  63
  64     def _real_extract(self, url):
  65         url, smuggled_data = unsmuggle_url(url, {})
  66
  67         video_id = self._match_id(url)
  68
  69         webpage = self._download_webpage(url, video_id)
  70
  71         if self._search_regex(
  72                 r'(?i)<meta\s[^>]*http-equiv="refresh"\s[^>]*content="[0-9]+;\s*url=https://www\.litv\.tv/"',
  73                 webpage, 'meta refresh redirect', default=False, group=0):
  74             raise ExtractorError('No such content found', expected=True)
  75
  76         program_info = self._parse_json(self._search_regex(
  77             r'var\s+programInfo\s*=\s*([^;]+)', webpage, 'VOD data', default='{}'),
  78             video_id)
  79
  80         # In browsers `getProgramInfo` request is always issued. Usually this
  81         # endpoint gives the same result as the data embedded in the webpage.
  82         # If, for some reason, there are no embedded data, we do an extra request.
  83         if 'assetId' not in program_info:
  84             program_info = self._download_json(
  85                 'https://www.litv.tv/vod/ajax/getProgramInfo', video_id,
  86                 query={'contentId': video_id},
  87                 headers={'Accept': 'application/json'})
  88
  89         series_id = program_info['seriesId']
  90         if self._yes_playlist(series_id, video_id, smuggled_data):
  91             playlist_data = self._download_json(
  92                 'https://www.litv.tv/vod/ajax/getSeriesTree', video_id,
  93                 query={'seriesId': series_id}, headers={'Accept': 'application/json'})
  94             return self._extract_playlist(playlist_data, program_info['contentType'])
  95
  96         video_data = self._parse_json(self._search_regex(
  97             r'uiHlsUrl\s*=\s*testBackendData\(([^;]+)\);',
  98             webpage, 'video data', default='{}'), video_id)
  99         if not video_data:
 100             payload = {'assetId': program_info['assetId']}
 101             puid = try_call(lambda: self._get_cookies('https://www.litv.tv/')['PUID'].value)
 102             if puid:
 103                 payload.update({
 104                     'type': 'auth',
 105                     'puid': puid,
 106                 })
 107                 endpoint = 'getUrl'
 108             else:
 109                 payload.update({
 110                     'watchDevices': program_info['watchDevices'],
 111                     'contentType': program_info['contentType'],
 112                 })
 113                 endpoint = 'getMainUrlNoAuth'
 114             video_data = self._download_json(
 115                 f'https://www.litv.tv/vod/ajax/{endpoint}', video_id,
 116                 data=json.dumps(payload).encode(),
 117                 headers={'Content-Type': 'application/json'})
 118
 119         if not video_data.get('fullpath'):
 120             error_msg = video_data.get('errorMessage')
 121             if error_msg == 'vod.error.outsideregionerror':
 122                 self.raise_geo_restricted('This video is available in Taiwan only')
 123             if error_msg:
 124                 raise ExtractorError(f'{self.IE_NAME} said: {error_msg}', expected=True)
 125             raise ExtractorError(f'Unexpected result from {self.IE_NAME}')
 126
 127         formats = self._extract_m3u8_formats(
 128             video_data['fullpath'], video_id, ext='mp4',
 129             entry_protocol='m3u8_native', m3u8_id='hls')
 130         for a_format in formats:
 131             # LiTV HLS segments doesn't like compressions
 132             a_format.setdefault('http_headers', {})['Accept-Encoding'] = 'identity'
 133
 134         title = program_info['title'] + program_info.get('secondaryMark', '')
 135         description = program_info.get('description')
 136         thumbnail = program_info.get('imageFile')
 137         categories = [item['name'] for item in program_info.get('category', [])]
 138         episode = int_or_none(program_info.get('episode'))
 139
 140         return {
 141             'id': video_id,
 142             'formats': formats,
 143             'title': title,
 144             'description': description,
 145             'thumbnail': thumbnail,
 146             'categories': categories,
 147             'episode_number': episode,
 148         }