youtube_dl/extractor/litv.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import json
   5
   6 from .common import InfoExtractor
   7 from ..utils import (
   8     ExtractorError,
   9     int_or_none,
  10     smuggle_url,
  11     unsmuggle_url,
  12 )
  13
  14
  15 class LiTVIE(InfoExtractor):
  16     _VALID_URL = r'https?://(?:www\.)?litv\.tv/(?:vod|promo)/[^/]+/(?:content\.do)?\?.*?\b(?:content_)?id=(?P<id>[^&]+)'
  17
  18     _URL_TEMPLATE = 'https://www.litv.tv/vod/%s/content.do?id=%s'
  19
  20     _TESTS = [{
  21         'url': 'https://www.litv.tv/vod/drama/content.do?brc_id=root&id=VOD00041610&isUHEnabled=true&autoPlay=1',
  22         'info_dict': {
  23             'id': 'VOD00041606',
  24             'title': '花千骨',
  25         },
  26         'playlist_count': 50,
  27     }, {
  28         'url': 'https://www.litv.tv/vod/drama/content.do?brc_id=root&id=VOD00041610&isUHEnabled=true&autoPlay=1',
  29         'md5': '969e343d9244778cb29acec608e53640',
  30         'info_dict': {
  31             'id': 'VOD00041610',
  32             'ext': 'mp4',
  33             'title': '花千骨第1集',
  34             'thumbnail': r're:https?://.*\.jpg$',
  35             'description': 'md5:c7017aa144c87467c4fb2909c4b05d6f',
  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': 'Georestricted to Taiwan',
  52     }]
  53
  54     def _extract_playlist(self, season_list, video_id, program_info, prompt=True):
  55         episode_title = program_info['title']
  56         content_id = season_list['contentId']
  57
  58         if prompt:
  59             self.to_screen('Downloading playlist %s - add --no-playlist to just download video %s' % (content_id, video_id))
  60
  61         all_episodes = [
  62             self.url_result(smuggle_url(
  63                 self._URL_TEMPLATE % (program_info['contentType'], episode['contentId']),
  64                 {'force_noplaylist': True}))  # To prevent infinite recursion
  65             for episode in season_list['episode']]
  66
  67         return self.playlist_result(all_episodes, content_id, episode_title)
  68
  69     def _real_extract(self, url):
  70         url, data = unsmuggle_url(url, {})
  71
  72         video_id = self._match_id(url)
  73
  74         noplaylist = self._downloader.params.get('noplaylist')
  75         noplaylist_prompt = True
  76         if 'force_noplaylist' in data:
  77             noplaylist = data['force_noplaylist']
  78             noplaylist_prompt = False
  79
  80         webpage = self._download_webpage(url, video_id)
  81
  82         program_info = self._parse_json(self._search_regex(
  83             r'var\s+programInfo\s*=\s*([^;]+)', webpage, 'VOD data', default='{}'),
  84             video_id)
  85
  86         season_list = list(program_info.get('seasonList', {}).values())
  87         if season_list:
  88             if not noplaylist:
  89                 return self._extract_playlist(
  90                     season_list[0], video_id, program_info,
  91                     prompt=noplaylist_prompt)
  92
  93             if noplaylist_prompt:
  94                 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
  95
  96         # In browsers `getMainUrl` request is always issued. Usually this
  97         # endpoint gives the same result as the data embedded in the webpage.
  98         # If georestricted, there are no embedded data, so an extra request is
  99         # necessary to get the error code
 100         if 'assetId' not in program_info:
 101             program_info = self._download_json(
 102                 'https://www.litv.tv/vod/ajax/getProgramInfo', video_id,
 103                 query={'contentId': video_id},
 104                 headers={'Accept': 'application/json'})
 105         video_data = self._parse_json(self._search_regex(
 106             r'uiHlsUrl\s*=\s*testBackendData\(([^;]+)\);',
 107             webpage, 'video data', default='{}'), video_id)
 108         if not video_data:
 109             payload = {
 110                 'assetId': program_info['assetId'],
 111                 'watchDevices': program_info['watchDevices'],
 112                 'contentType': program_info['contentType'],
 113             }
 114             video_data = self._download_json(
 115                 'https://www.litv.tv/vod/getMainUrl', video_id,
 116                 data=json.dumps(payload).encode('utf-8'),
 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('%s said: %s' % (self.IE_NAME, error_msg), expected=True)
 125             raise ExtractorError('Unexpected result from %s' % 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', {})['Youtubedl-no-compression'] = True
 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         }