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