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

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import itertools
   5 import json
   6
   7 from .common import InfoExtractor
   8 from ..compat import (
   9     compat_HTTPError,
  10     compat_urlparse,
  11 )
  12 from ..utils import (
  13     clean_html,
  14     ExtractorError,
  15     int_or_none,
  16     parse_age_limit,
  17     parse_duration,
  18     unified_timestamp,
  19     url_or_none,
  20 )
  21
  22
  23 class DramaFeverBaseIE(InfoExtractor):
  24     _NETRC_MACHINE = 'dramafever'
  25
  26     _CONSUMER_SECRET = 'DA59dtVXYLxajktV'
  27
  28     _consumer_secret = None
  29
  30     def _get_consumer_secret(self):
  31         mainjs = self._download_webpage(
  32             'http://www.dramafever.com/static/51afe95/df2014/scripts/main.js',
  33             None, 'Downloading main.js', fatal=False)
  34         if not mainjs:
  35             return self._CONSUMER_SECRET
  36         return self._search_regex(
  37             r"var\s+cs\s*=\s*'([^']+)'", mainjs,
  38             'consumer secret', default=self._CONSUMER_SECRET)
  39
  40     def _real_initialize(self):
  41         self._consumer_secret = self._get_consumer_secret()
  42         self._login()
  43
  44     def _login(self):
  45         username, password = self._get_login_info()
  46         if username is None:
  47             return
  48
  49         login_form = {
  50             'username': username,
  51             'password': password,
  52         }
  53
  54         try:
  55             response = self._download_json(
  56                 'https://www.dramafever.com/api/users/login', None, 'Logging in',
  57                 data=json.dumps(login_form).encode('utf-8'), headers={
  58                     'x-consumer-key': self._consumer_secret,
  59                 })
  60         except ExtractorError as e:
  61             if isinstance(e.cause, compat_HTTPError) and e.cause.code in (403, 404):
  62                 response = self._parse_json(
  63                     e.cause.read().decode('utf-8'), None)
  64             else:
  65                 raise
  66
  67         # Successful login
  68         if response.get('result') or response.get('guid') or response.get('user_guid'):
  69             return
  70
  71         errors = response.get('errors')
  72         if errors and isinstance(errors, list):
  73             error = errors[0]
  74             message = error.get('message') or error['reason']
  75             raise ExtractorError('Unable to login: %s' % message, expected=True)
  76         raise ExtractorError('Unable to log in')
  77
  78
  79 class DramaFeverIE(DramaFeverBaseIE):
  80     IE_NAME = 'dramafever'
  81     _VALID_URL = r'https?://(?:www\.)?dramafever\.com/(?:[^/]+/)?drama/(?P<id>[0-9]+/[0-9]+)(?:/|$)'
  82     _TESTS = [{
  83         'url': 'https://www.dramafever.com/drama/4274/1/Heirs/',
  84         'info_dict': {
  85             'id': '4274.1',
  86             'ext': 'wvm',
  87             'title': 'Heirs - Episode 1',
  88             'description': 'md5:362a24ba18209f6276e032a651c50bc2',
  89             'thumbnail': r're:^https?://.*\.jpg',
  90             'duration': 3783,
  91             'timestamp': 1381354993,
  92             'upload_date': '20131009',
  93             'series': 'Heirs',
  94             'season_number': 1,
  95             'episode': 'Episode 1',
  96             'episode_number': 1,
  97         },
  98         'params': {
  99             # m3u8 download
 100             'skip_download': True,
 101         },
 102     }, {
 103         'url': 'http://www.dramafever.com/drama/4826/4/Mnet_Asian_Music_Awards_2015/?ap=1',
 104         'info_dict': {
 105             'id': '4826.4',
 106             'ext': 'flv',
 107             'title': 'Mnet Asian Music Awards 2015',
 108             'description': 'md5:3ff2ee8fedaef86e076791c909cf2e91',
 109             'episode': 'Mnet Asian Music Awards 2015 - Part 3',
 110             'episode_number': 4,
 111             'thumbnail': r're:^https?://.*\.jpg',
 112             'timestamp': 1450213200,
 113             'upload_date': '20151215',
 114             'duration': 5359,
 115         },
 116         'params': {
 117             # m3u8 download
 118             'skip_download': True,
 119         },
 120     }, {
 121         'url': 'https://www.dramafever.com/zh-cn/drama/4972/15/Doctor_Romantic/',
 122         'only_matching': True,
 123     }]
 124
 125     def _call_api(self, path, video_id, note, fatal=False):
 126         return self._download_json(
 127             'https://www.dramafever.com/api/5/' + path,
 128             video_id, note=note, headers={
 129                 'x-consumer-key': self._consumer_secret,
 130             }, fatal=fatal)
 131
 132     def _get_subtitles(self, video_id):
 133         subtitles = {}
 134         subs = self._call_api(
 135             'video/%s/subtitles/webvtt/' % video_id, video_id,
 136             'Downloading subtitles JSON', fatal=False)
 137         if not subs or not isinstance(subs, list):
 138             return subtitles
 139         for sub in subs:
 140             if not isinstance(sub, dict):
 141                 continue
 142             sub_url = url_or_none(sub.get('url'))
 143             if not sub_url:
 144                 continue
 145             subtitles.setdefault(
 146                 sub.get('code') or sub.get('language') or 'en', []).append({
 147                     'url': sub_url
 148                 })
 149         return subtitles
 150
 151     def _real_extract(self, url):
 152         video_id = self._match_id(url).replace('/', '.')
 153
 154         series_id, episode_number = video_id.split('.')
 155
 156         video = self._call_api(
 157             'series/%s/episodes/%s/' % (series_id, episode_number), video_id,
 158             'Downloading video JSON')
 159
 160         formats = []
 161         download_assets = video.get('download_assets')
 162         if download_assets and isinstance(download_assets, dict):
 163             for format_id, format_dict in download_assets.items():
 164                 if not isinstance(format_dict, dict):
 165                     continue
 166                 format_url = url_or_none(format_dict.get('url'))
 167                 if not format_url:
 168                     continue
 169                 formats.append({
 170                     'url': format_url,
 171                     'format_id': format_id,
 172                     'filesize': int_or_none(video.get('filesize')),
 173                 })
 174
 175         stream = self._call_api(
 176             'video/%s/stream/' % video_id, video_id, 'Downloading stream JSON',
 177             fatal=False)
 178         if stream:
 179             stream_url = stream.get('stream_url')
 180             if stream_url:
 181                 formats.extend(self._extract_m3u8_formats(
 182                     stream_url, video_id, 'mp4', entry_protocol='m3u8_native',
 183                     m3u8_id='hls', fatal=False))
 184         self._sort_formats(formats)
 185
 186         title = video.get('title') or 'Episode %s' % episode_number
 187         description = video.get('description')
 188         thumbnail = video.get('thumbnail')
 189         timestamp = unified_timestamp(video.get('release_date'))
 190         duration = parse_duration(video.get('duration'))
 191         age_limit = parse_age_limit(video.get('tv_rating'))
 192         series = video.get('series_title')
 193         season_number = int_or_none(video.get('season'))
 194
 195         if series:
 196             title = '%s - %s' % (series, title)
 197
 198         subtitles = self.extract_subtitles(video_id)
 199
 200         return {
 201             'id': video_id,
 202             'title': title,
 203             'description': description,
 204             'thumbnail': thumbnail,
 205             'duration': duration,
 206             'timestamp': timestamp,
 207             'age_limit': age_limit,
 208             'series': series,
 209             'season_number': season_number,
 210             'episode_number': int_or_none(episode_number),
 211             'formats': formats,
 212             'subtitles': subtitles,
 213         }
 214
 215
 216 class DramaFeverSeriesIE(DramaFeverBaseIE):
 217     IE_NAME = 'dramafever:series'
 218     _VALID_URL = r'https?://(?:www\.)?dramafever\.com/(?:[^/]+/)?drama/(?P<id>[0-9]+)(?:/(?:(?!\d+(?:/|$)).+)?)?$'
 219     _TESTS = [{
 220         'url': 'http://www.dramafever.com/drama/4512/Cooking_with_Shin/',
 221         'info_dict': {
 222             'id': '4512',
 223             'title': 'Cooking with Shin',
 224             'description': 'md5:84a3f26e3cdc3fb7f500211b3593b5c1',
 225         },
 226         'playlist_count': 4,
 227     }, {
 228         'url': 'http://www.dramafever.com/drama/124/IRIS/',
 229         'info_dict': {
 230             'id': '124',
 231             'title': 'IRIS',
 232             'description': 'md5:b3a30e587cf20c59bd1c01ec0ee1b862',
 233         },
 234         'playlist_count': 20,
 235     }]
 236
 237     _PAGE_SIZE = 60  # max is 60 (see http://api.drama9.com/#get--api-4-episode-series-)
 238
 239     def _real_extract(self, url):
 240         series_id = self._match_id(url)
 241
 242         series = self._download_json(
 243             'http://www.dramafever.com/api/4/series/query/?cs=%s&series_id=%s'
 244             % (self._consumer_secret, series_id),
 245             series_id, 'Downloading series JSON')['series'][series_id]
 246
 247         title = clean_html(series['name'])
 248         description = clean_html(series.get('description') or series.get('description_short'))
 249
 250         entries = []
 251         for page_num in itertools.count(1):
 252             episodes = self._download_json(
 253                 'http://www.dramafever.com/api/4/episode/series/?cs=%s&series_id=%s&page_size=%d&page_number=%d'
 254                 % (self._consumer_secret, series_id, self._PAGE_SIZE, page_num),
 255                 series_id, 'Downloading episodes JSON page #%d' % page_num)
 256             for episode in episodes.get('value', []):
 257                 episode_url = episode.get('episode_url')
 258                 if not episode_url:
 259                     continue
 260                 entries.append(self.url_result(
 261                     compat_urlparse.urljoin(url, episode_url),
 262                     'DramaFever', episode.get('guid')))
 263             if page_num == episodes['num_pages']:
 264                 break
 265
 266         return self.playlist_result(entries, series_id, title, description)