Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/soompi.py

   1 # encoding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5
   6 from .crunchyroll import CrunchyrollIE
   7
   8 from .common import InfoExtractor
   9 from ..compat import compat_HTTPError
  10 from ..utils import (
  11     ExtractorError,
  12     int_or_none,
  13     remove_start,
  14     xpath_text,
  15 )
  16
  17
  18 class SoompiBaseIE(InfoExtractor):
  19     def _get_episodes(self, webpage, episode_filter=None):
  20         episodes = self._parse_json(
  21             self._search_regex(
  22                 r'VIDEOS\s*=\s*(\[.+?\]);', webpage, 'episodes JSON'),
  23             None)
  24         return list(filter(episode_filter, episodes))
  25
  26
  27 class SoompiIE(SoompiBaseIE, CrunchyrollIE):
  28     IE_NAME = 'soompi'
  29     _VALID_URL = r'https?://tv\.soompi\.com/(?:en/)?watch/(?P<id>[0-9]+)'
  30     _TESTS = [{
  31         'url': 'http://tv.soompi.com/en/watch/29235',
  32         'info_dict': {
  33             'id': '29235',
  34             'ext': 'mp4',
  35             'title': 'Episode 1096',
  36             'description': '2015-05-20'
  37         },
  38         'params': {
  39             'skip_download': True,
  40         },
  41     }]
  42
  43     def _get_episode(self, webpage, video_id):
  44         return self._get_episodes(webpage, lambda x: x['id'] == video_id)[0]
  45
  46     def _get_subtitles(self, config, video_id):
  47         sub_langs = {}
  48         for subtitle in config.findall('./{default}preload/subtitles/subtitle'):
  49             sub_langs[subtitle.attrib['id']] = subtitle.attrib['title']
  50
  51         subtitles = {}
  52         for s in config.findall('./{default}preload/subtitle'):
  53             lang_code = sub_langs.get(s.attrib['id'])
  54             if not lang_code:
  55                 continue
  56             sub_id = s.get('id')
  57             data = xpath_text(s, './data', 'data')
  58             iv = xpath_text(s, './iv', 'iv')
  59             if not id or not iv or not data:
  60                 continue
  61             subtitle = self._decrypt_subtitles(data, iv, sub_id).decode('utf-8')
  62             subtitles[lang_code] = self._extract_subtitles(subtitle)
  63         return subtitles
  64
  65     def _real_extract(self, url):
  66         video_id = self._match_id(url)
  67
  68         try:
  69             webpage = self._download_webpage(
  70                 url, video_id, 'Downloading episode page')
  71         except ExtractorError as ee:
  72             if isinstance(ee.cause, compat_HTTPError) and ee.cause.code == 403:
  73                 webpage = ee.cause.read()
  74                 block_message = self._html_search_regex(
  75                     r'(?s)<div class="block-message">(.+?)</div>', webpage,
  76                     'block message', default=None)
  77                 if block_message:
  78                     raise ExtractorError(block_message, expected=True)
  79             raise
  80
  81         formats = []
  82         config = None
  83         for format_id in re.findall(r'\?quality=([0-9a-zA-Z]+)', webpage):
  84             config = self._download_xml(
  85                 'http://tv.soompi.com/en/show/_/%s-config.xml?mode=hls&quality=%s' % (video_id, format_id),
  86                 video_id, 'Downloading %s XML' % format_id)
  87             m3u8_url = xpath_text(
  88                 config, './{default}preload/stream_info/file',
  89                 '%s m3u8 URL' % format_id)
  90             if not m3u8_url:
  91                 continue
  92             formats.extend(self._extract_m3u8_formats(
  93                 m3u8_url, video_id, 'mp4', m3u8_id=format_id))
  94         self._sort_formats(formats)
  95
  96         episode = self._get_episode(webpage, video_id)
  97
  98         title = episode['name']
  99         description = episode.get('description')
 100         duration = int_or_none(episode.get('duration'))
 101
 102         thumbnails = [{
 103             'id': thumbnail_id,
 104             'url': thumbnail_url,
 105         } for thumbnail_id, thumbnail_url in episode.get('img_url', {}).items()]
 106
 107         subtitles = self.extract_subtitles(config, video_id)
 108
 109         return {
 110             'id': video_id,
 111             'title': title,
 112             'description': description,
 113             'thumbnails': thumbnails,
 114             'duration': duration,
 115             'formats': formats,
 116             'subtitles': subtitles
 117         }
 118
 119
 120 class SoompiShowIE(SoompiBaseIE):
 121     IE_NAME = 'soompi:show'
 122     _VALID_URL = r'https?://tv\.soompi\.com/en/shows/(?P<id>[0-9a-zA-Z\-_]+)'
 123     _TESTS = [{
 124         'url': 'http://tv.soompi.com/en/shows/liar-game',
 125         'info_dict': {
 126             'id': 'liar-game',
 127             'title': 'Liar Game',
 128             'description': 'md5:52c02bce0c1a622a95823591d0589b66',
 129         },
 130         'playlist_count': 14,
 131     }]
 132
 133     def _real_extract(self, url):
 134         show_id = self._match_id(url)
 135
 136         webpage = self._download_webpage(
 137             url, show_id, 'Downloading show page')
 138
 139         title = remove_start(self._og_search_title(webpage), 'SoompiTV | ')
 140         description = self._og_search_description(webpage)
 141
 142         entries = [
 143             self.url_result('http://tv.soompi.com/en/watch/%s' % episode['id'], 'Soompi')
 144             for episode in self._get_episodes(webpage)]
 145
 146         return self.playlist_result(entries, show_id, title, description)