youtube_dl/extractor/litv.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import json
   5 import re
   6
   7 from .common import InfoExtractor
   8 from ..utils import (
   9     ExtractorError,
  10     int_or_none,
  11     smuggle_url,
  12     unsmuggle_url,
  13 )
  14
  15
  16 class LiTVIE(InfoExtractor):
  17     _VALID_URL = r'https?://www\.litv\.tv/(?:vod|promo)/[^/]+/(?:content\.do)?\?.*?\b(?:content_)?id=(?P<id>[^&]+)'
  18
  19     _URL_TEMPLATE = 'https://www.litv.tv/vod/%s/content.do?id=%s'
  20
  21     _TESTS = [{
  22         'url': 'https://www.litv.tv/vod/drama/content.do?brc_id=root&id=VOD00041610&isUHEnabled=true&autoPlay=1',
  23         'info_dict': {
  24             'id': 'VOD00041606',
  25             'title': '花千骨',
  26         },
  27         'playlist_count': 50,
  28     }, {
  29         'url': 'https://www.litv.tv/vod/drama/content.do?brc_id=root&id=VOD00041610&isUHEnabled=true&autoPlay=1',
  30         'md5': '969e343d9244778cb29acec608e53640',
  31         'info_dict': {
  32             'id': 'VOD00041610',
  33             'ext': 'mp4',
  34             'title': '花千骨第1集',
  35             'thumbnail': 're:https?://.*\.jpg$',
  36             'description': 'md5:c7017aa144c87467c4fb2909c4b05d6f',
  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': 'Georestricted to Taiwan',
  53     }]
  54
  55     def _extract_playlist(self, season_list, video_id, vod_data, view_data, prompt=True):
  56         episode_title = view_data['title']
  57         content_id = season_list['contentId']
  58
  59         if prompt:
  60             self.to_screen('Downloading playlist %s - add --no-playlist to just download video %s' % (content_id, video_id))
  61
  62         all_episodes = [
  63             self.url_result(smuggle_url(
  64                 self._URL_TEMPLATE % (view_data['contentType'], episode['contentId']),
  65                 {'force_noplaylist': True}))  # To prevent infinite recursion
  66             for episode in season_list['episode']]
  67
  68         return self.playlist_result(all_episodes, content_id, episode_title)
  69
  70     def _real_extract(self, url):
  71         url, data = unsmuggle_url(url, {})
  72
  73         video_id = self._match_id(url)
  74
  75         noplaylist = self._downloader.params.get('noplaylist')
  76         noplaylist_prompt = True
  77         if 'force_noplaylist' in data:
  78             noplaylist = data['force_noplaylist']
  79             noplaylist_prompt = False
  80
  81         webpage = self._download_webpage(url, video_id)
  82
  83         view_data = dict(map(lambda t: (t[0], t[2]), re.findall(
  84             r'viewData\.([a-zA-Z]+)\s*=\s*(["\'])([^"\']+)\2',
  85             webpage)))
  86
  87         vod_data = self._parse_json(self._search_regex(
  88             'var\s+vod\s*=\s*([^;]+)', webpage, 'VOD data', default='{}'),
  89             video_id)
  90
  91         season_list = list(vod_data.get('seasonList', {}).values())
  92         if season_list:
  93             if not noplaylist:
  94                 return self._extract_playlist(
  95                     season_list[0], video_id, vod_data, view_data,
  96                     prompt=noplaylist_prompt)
  97
  98             if noplaylist_prompt:
  99                 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
 100
 101         # In browsers `getMainUrl` request is always issued. Usually this
 102         # endpoint gives the same result as the data embedded in the webpage.
 103         # If georestricted, there are no embedded data, so an extra request is
 104         # necessary to get the error code
 105         if 'assetId' not in view_data:
 106             view_data = self._download_json(
 107                 'https://www.litv.tv/vod/ajax/getProgramInfo', video_id,
 108                 query={'contentId': video_id},
 109                 headers={'Accept': 'application/json'})
 110         video_data = self._parse_json(self._search_regex(
 111             r'uiHlsUrl\s*=\s*testBackendData\(([^;]+)\);',
 112             webpage, 'video data', default='{}'), video_id)
 113         if not video_data:
 114             payload = {
 115                 'assetId': view_data['assetId'],
 116                 'watchDevices': view_data['watchDevices'],
 117                 'contentType': view_data['contentType'],
 118             }
 119             video_data = self._download_json(
 120                 'https://www.litv.tv/vod/getMainUrl', video_id,
 121                 data=json.dumps(payload).encode('utf-8'),
 122                 headers={'Content-Type': 'application/json'})
 123
 124         if not video_data.get('fullpath'):
 125             error_msg = video_data.get('errorMessage')
 126             if error_msg == 'vod.error.outsideregionerror':
 127                 self.raise_geo_restricted('This video is available in Taiwan only')
 128             if error_msg:
 129                 raise ExtractorError('%s said: %s' % (self.IE_NAME, error_msg), expected=True)
 130             raise ExtractorError('Unexpected result from %s' % self.IE_NAME)
 131
 132         formats = self._extract_m3u8_formats(
 133             video_data['fullpath'], video_id, ext='mp4',
 134             entry_protocol='m3u8_native', m3u8_id='hls')
 135         for a_format in formats:
 136             # LiTV HLS segments doesn't like compressions
 137             a_format.setdefault('http_headers', {})['Youtubedl-no-compression'] = True
 138
 139         title = view_data['title'] + view_data.get('secondaryMark', '')
 140         description = view_data.get('description')
 141         thumbnail = view_data.get('imageFile')
 142         categories = [item['name'] for item in vod_data.get('category', [])]
 143         episode = int_or_none(view_data.get('episode'))
 144
 145         return {
 146             'id': video_id,
 147             'formats': formats,
 148             'title': title,
 149             'description': description,
 150             'thumbnail': thumbnail,
 151             'categories': categories,
 152             'episode_number': episode,
 153         }