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/[^/]+/content\.do\?.*?\bid=(?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         'info_dict': {
  31             'id': 'VOD00041610',
  32             'ext': 'mp4',
  33             'title': '花千骨第1集',
  34             'thumbnail': 're:https?://.*\.jpg$',
  35             'description': 'md5:c7017aa144c87467c4fb2909c4b05d6f',
  36             'episode_number': 1,
  37         },
  38         'params': {
  39             'noplaylist': True,
  40             'skip_download': True,  # m3u8 download
  41         },
  42         'skip': 'Georestricted to Taiwan',
  43     }]
  44
  45     def _extract_playlist(self, season_list, video_id, vod_data, view_data, prompt=True):
  46         episode_title = view_data['title']
  47         content_id = season_list['contentId']
  48
  49         if prompt:
  50             self.to_screen('Downloading playlist %s - add --no-playlist to just download video %s' % (content_id, video_id))
  51
  52         all_episodes = [
  53             self.url_result(smuggle_url(
  54                 self._URL_TEMPLATE % (view_data['contentType'], episode['contentId']),
  55                 {'force_noplaylist': True}))  # To prevent infinite recursion
  56             for episode in season_list['episode']]
  57
  58         return self.playlist_result(all_episodes, content_id, episode_title)
  59
  60     def _real_extract(self, url):
  61         url, data = unsmuggle_url(url, {})
  62
  63         video_id = self._match_id(url)
  64
  65         noplaylist = self._downloader.params.get('noplaylist')
  66         noplaylist_prompt = True
  67         if 'force_noplaylist' in data:
  68             noplaylist = data['force_noplaylist']
  69             noplaylist_prompt = False
  70
  71         webpage = self._download_webpage(url, video_id)
  72
  73         view_data = dict(map(lambda t: (t[0], t[2]), re.findall(
  74             r'viewData\.([a-zA-Z]+)\s*=\s*(["\'])([^"\']+)\2',
  75             webpage)))
  76
  77         vod_data = self._parse_json(self._search_regex(
  78             'var\s+vod\s*=\s*([^;]+)', webpage, 'VOD data', default='{}'),
  79             video_id)
  80
  81         season_list = list(vod_data.get('seasonList', {}).values())
  82         if season_list:
  83             if not noplaylist:
  84                 return self._extract_playlist(
  85                     season_list[0], video_id, vod_data, view_data,
  86                     prompt=noplaylist_prompt)
  87
  88             if noplaylist_prompt:
  89                 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
  90
  91         # In browsers `getMainUrl` request is always issued. Usually this
  92         # endpoint gives the same result as the data embedded in the webpage.
  93         # If georestricted, there are no embedded data, so an extra request is
  94         # necessary to get the error code
  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': view_data['assetId'],
 101                 'watchDevices': vod_data['watchDevices'],
 102                 'contentType': view_data['contentType'],
 103             }
 104             video_data = self._download_json(
 105                 'https://www.litv.tv/vod/getMainUrl', 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', m3u8_id='hls')
 119         for a_format in formats:
 120             # LiTV HLS segments doesn't like compressions
 121             a_format.setdefault('http_headers', {})['Youtubedl-no-compression'] = True
 122
 123         title = view_data['title'] + view_data.get('secondaryMark', '')
 124         description = view_data.get('description')
 125         thumbnail = view_data.get('imageFile')
 126         categories = [item['name'] for item in vod_data.get('category', [])]
 127         episode = int_or_none(view_data.get('episode'))
 128
 129         return {
 130             'id': video_id,
 131             'formats': formats,
 132             'title': title,
 133             'description': description,
 134             'thumbnail': thumbnail,
 135             'categories': categories,
 136             'episode_number': episode,
 137         }