youtube_dl/extractor/smotri.py

   1 # encoding: utf-8
   2
   3 import re
   4 import json
   5 import hashlib
   6 import uuid
   7
   8 from .common import InfoExtractor
   9 from ..utils import (
  10     compat_urllib_parse,
  11     compat_urllib_request,
  12     ExtractorError,
  13 )
  14
  15
  16 class SmotriIE(InfoExtractor):
  17     IE_DESC = u'Smotri.com'
  18     IE_NAME = u'smotri'
  19     _VALID_URL = r'^https?://(?:www\.)?(?P<url>smotri\.com/video/view/\?id=(?P<videoid>v(?P<realvideoid>[0-9]+)[a-z0-9]{4}))'
  20
  21     _TESTS = [
  22         # real video id 2610366
  23         {
  24             u'url': u'http://smotri.com/video/view/?id=v261036632ab',
  25             u'file': u'v261036632ab.mp4',
  26             u'md5': u'2a7b08249e6f5636557579c368040eb9',
  27             u'info_dict': {
  28                 u'title': u'катастрофа с камер видеонаблюдения',
  29                 u'uploader': u'rbc2008',
  30                 u'uploader_id': u'rbc08',
  31                 u'upload_date': u'20131118',
  32                 u'description': u'катастрофа с камер видеонаблюдения, видео катастрофа с камер видеонаблюдения',
  33                 u'thumbnail': u'http://frame6.loadup.ru/8b/a9/2610366.3.3.jpg',
  34             },
  35         },
  36         # real video id 57591
  37         {
  38             u'url': u'http://smotri.com/video/view/?id=v57591cb20',
  39             u'file': u'v57591cb20.flv',
  40             u'md5': u'830266dfc21f077eac5afd1883091bcd',
  41             u'info_dict': {
  42                 u'title': u'test',
  43                 u'uploader': u'Support Photofile@photofile',
  44                 u'uploader_id': u'support-photofile',
  45                 u'upload_date': u'20070704',
  46                 u'description': u'test, видео test',
  47                 u'thumbnail': u'http://frame4.loadup.ru/03/ed/57591.2.3.jpg',
  48             },
  49         },
  50         # video-password
  51         {
  52             u'url': u'http://smotri.com/video/view/?id=v1390466a13c',
  53             u'file': u'v1390466a13c.mp4',
  54             u'md5': u'f6331cef33cad65a0815ee482a54440b',
  55             u'info_dict': {
  56                 u'title': u'TOCCA_A_NOI_-_LE_COSE_NON_VANNO_CAMBIAMOLE_ORA-1',
  57                 u'uploader': u'timoxa40',
  58                 u'uploader_id': u'timoxa40',
  59                 u'upload_date': u'20100404',
  60                 u'thumbnail': u'http://frame7.loadup.ru/af/3f/1390466.3.3.jpg',
  61                 u'description': u'TOCCA_A_NOI_-_LE_COSE_NON_VANNO_CAMBIAMOLE_ORA-1, видео TOCCA_A_NOI_-_LE_COSE_NON_VANNO_CAMBIAMOLE_ORA-1',
  62             },
  63             u'params': {
  64                 u'videopassword': u'qwerty',
  65             },
  66         },
  67         # age limit + video-password
  68         {
  69             u'url': u'http://smotri.com/video/view/?id=v15408898bcf',
  70             u'file': u'v15408898bcf.flv',
  71             u'md5': u'91e909c9f0521adf5ee86fbe073aad70',
  72             u'info_dict': {
  73                 u'title': u'этот ролик не покажут по ТВ',
  74                 u'uploader': u'zzxxx',
  75                 u'uploader_id': u'ueggb',
  76                 u'upload_date': u'20101001',
  77                 u'thumbnail': u'http://frame3.loadup.ru/75/75/1540889.1.3.jpg',
  78                 u'age_limit': 18,
  79                 u'description': u'этот ролик не покажут по ТВ, видео этот ролик не покажут по ТВ',
  80             },
  81             u'params': {
  82                 u'videopassword': u'333'
  83             }
  84         }
  85     ]
  86
  87     _SUCCESS = 0
  88     _PASSWORD_NOT_VERIFIED = 1
  89     _PASSWORD_DETECTED = 2
  90     _VIDEO_NOT_FOUND = 3
  91
  92     def _search_meta(self, name, html, display_name=None):
  93         if display_name is None:
  94             display_name = name
  95         return self._html_search_regex(
  96             r'<meta itemprop="%s" content="([^"]+)" />' % re.escape(name),
  97             html, display_name, fatal=False)
  98         return self._html_search_meta(name, html, display_name)
  99
 100     def _real_extract(self, url):
 101         mobj = re.match(self._VALID_URL, url)
 102         video_id = mobj.group('videoid')
 103         real_video_id = mobj.group('realvideoid')
 104
 105         # Download video JSON data
 106         video_json_url = 'http://smotri.com/vt.php?id=%s' % real_video_id
 107         video_json_page = self._download_webpage(video_json_url, video_id, u'Downloading video JSON')
 108         video_json = json.loads(video_json_page)
 109
 110         status = video_json['status']
 111         if status == self._VIDEO_NOT_FOUND:
 112             raise ExtractorError(u'Video %s does not exist' % video_id, expected=True)
 113         elif status == self._PASSWORD_DETECTED:  # The video is protected by a password, retry with
 114                                                 # video-password set
 115             video_password = self._downloader.params.get('videopassword', None)
 116             if not video_password:
 117                 raise ExtractorError(u'This video is protected by a password, use the --video-password option', expected=True)
 118             video_json_url += '&md5pass=%s' % hashlib.md5(video_password.encode('utf-8')).hexdigest()
 119             video_json_page = self._download_webpage(video_json_url, video_id, u'Downloading video JSON (video-password set)')
 120             video_json = json.loads(video_json_page)
 121             status = video_json['status']
 122             if status == self._PASSWORD_NOT_VERIFIED:
 123                 raise ExtractorError(u'Video password is invalid', expected=True)
 124
 125         if status != self._SUCCESS:
 126             raise ExtractorError(u'Unexpected status value %s' % status)
 127
 128         # Extract the URL of the video
 129         video_url = video_json['file_data']
 130
 131         # Video JSON does not provide enough meta data
 132         # We will extract some from the video web page instead
 133         video_page_url = 'http://' + mobj.group('url')
 134         video_page = self._download_webpage(video_page_url, video_id, u'Downloading video page')
 135
 136         # Adult content
 137         if re.search(u'EroConfirmText">', video_page) is not None:
 138             self.report_age_confirmation()
 139             confirm_string = self._html_search_regex(
 140                 r'<a href="/video/view/\?id=%s&confirm=([^"]+)" title="[^"]+">' % video_id,
 141                 video_page, u'confirm string')
 142             confirm_url = video_page_url + '&confirm=%s' % confirm_string
 143             video_page = self._download_webpage(confirm_url, video_id, u'Downloading video page (age confirmed)')
 144             adult_content = True
 145         else:
 146             adult_content = False
 147
 148         # Extract the rest of meta data
 149         video_title = self._search_meta(u'name', video_page, u'title')
 150         if not video_title:
 151             video_title = video_url.rsplit('/', 1)[-1]
 152
 153         video_description = self._search_meta(u'description', video_page)
 154         END_TEXT = u' на сайте Smotri.com'
 155         if video_description.endswith(END_TEXT):
 156             video_description = video_description[:-len(END_TEXT)]
 157         START_TEXT = u'Смотреть онлайн ролик '
 158         if video_description.startswith(START_TEXT):
 159             video_description = video_description[len(START_TEXT):]
 160         video_thumbnail = self._search_meta(u'thumbnail', video_page)
 161
 162         upload_date_str = self._search_meta(u'uploadDate', video_page, u'upload date')
 163         upload_date_m = re.search(r'(?P<year>\d{4})\.(?P<month>\d{2})\.(?P<day>\d{2})T', upload_date_str)
 164         video_upload_date = (
 165             (
 166                 upload_date_m.group('year') +
 167                 upload_date_m.group('month') +
 168                 upload_date_m.group('day')
 169             )
 170             if upload_date_m else None
 171         )
 172
 173         duration_str = self._search_meta(u'duration', video_page)
 174         duration_m = re.search(r'T(?P<hours>[0-9]{2})H(?P<minutes>[0-9]{2})M(?P<seconds>[0-9]{2})S', duration_str)
 175         video_duration = (
 176             (
 177                 (int(duration_m.group('hours')) * 60 * 60) +
 178                 (int(duration_m.group('minutes')) * 60) +
 179                 int(duration_m.group('seconds'))
 180             )
 181             if duration_m else None
 182         )
 183
 184         video_uploader = self._html_search_regex(
 185             u'<div class="DescrUser"><div>Автор.*?onmouseover="popup_user_info[^"]+">(.*?)</a>',
 186             video_page, u'uploader', fatal=False, flags=re.MULTILINE|re.DOTALL)
 187
 188         video_uploader_id = self._html_search_regex(
 189             u'<div class="DescrUser"><div>Автор.*?onmouseover="popup_user_info\\(.*?\'([^\']+)\'\\);">',
 190             video_page, u'uploader id', fatal=False, flags=re.MULTILINE|re.DOTALL)
 191
 192         video_view_count = self._html_search_regex(
 193             u'Общее количество просмотров.*?<span class="Number">(\\d+)</span>',
 194             video_page, u'view count', fatal=False, flags=re.MULTILINE|re.DOTALL)
 195
 196         return {
 197             'id': video_id,
 198             'url': video_url,
 199             'title': video_title,
 200             'thumbnail': video_thumbnail,
 201             'description': video_description,
 202             'uploader': video_uploader,
 203             'upload_date': video_upload_date,
 204             'uploader_id': video_uploader_id,
 205             'duration': video_duration,
 206             'view_count': video_view_count,
 207             'age_limit': 18 if adult_content else 0,
 208             'video_page_url': video_page_url
 209         }
 210
 211
 212 class SmotriCommunityIE(InfoExtractor):
 213     IE_DESC = u'Smotri.com community videos'
 214     IE_NAME = u'smotri:community'
 215     _VALID_URL = r'^https?://(?:www\.)?smotri\.com/community/video/(?P<communityid>[0-9A-Za-z_\'-]+)'
 216
 217     def _real_extract(self, url):
 218         mobj = re.match(self._VALID_URL, url)
 219         community_id = mobj.group('communityid')
 220
 221         url = 'http://smotri.com/export/rss/video/by/community/-/%s/video.xml' % community_id
 222         rss = self._download_xml(url, community_id, u'Downloading community RSS')
 223
 224         entries = [self.url_result(video_url.text, 'Smotri')
 225                    for video_url in rss.findall('./channel/item/link')]
 226
 227         description_text = rss.find('./channel/description').text
 228         community_title = self._html_search_regex(
 229             u'^Видео сообщества "([^"]+)"$', description_text, u'community title')
 230
 231         return self.playlist_result(entries, community_id, community_title)
 232
 233
 234 class SmotriUserIE(InfoExtractor):
 235     IE_DESC = u'Smotri.com user videos'
 236     IE_NAME = u'smotri:user'
 237     _VALID_URL = r'^https?://(?:www\.)?smotri\.com/user/(?P<userid>[0-9A-Za-z_\'-]+)'
 238
 239     def _real_extract(self, url):
 240         mobj = re.match(self._VALID_URL, url)
 241         user_id = mobj.group('userid')
 242
 243         url = 'http://smotri.com/export/rss/user/video/-/%s/video.xml' % user_id
 244         rss = self._download_xml(url, user_id, u'Downloading user RSS')
 245
 246         entries = [self.url_result(video_url.text, 'Smotri')
 247                    for video_url in rss.findall('./channel/item/link')]
 248
 249         description_text = rss.find('./channel/description').text
 250         user_nickname = self._html_search_regex(
 251             u'^Видео режиссера (.*)$', description_text,
 252             u'user nickname')
 253
 254         return self.playlist_result(entries, user_id, user_nickname)
 255
 256
 257 class SmotriBroadcastIE(InfoExtractor):
 258     IE_DESC = u'Smotri.com broadcasts'
 259     IE_NAME = u'smotri:broadcast'
 260     _VALID_URL = r'^https?://(?:www\.)?(?P<url>smotri\.com/live/(?P<broadcastid>[^/]+))/?.*'
 261
 262     def _real_extract(self, url):
 263         mobj = re.match(self._VALID_URL, url)
 264         broadcast_id = mobj.group('broadcastid')
 265
 266         broadcast_url = 'http://' + mobj.group('url')
 267         broadcast_page = self._download_webpage(broadcast_url, broadcast_id, u'Downloading broadcast page')
 268
 269         if re.search(u'>Режиссер с логином <br/>"%s"<br/> <span>не существует<' % broadcast_id, broadcast_page) is not None:
 270             raise ExtractorError(u'Broadcast %s does not exist' % broadcast_id, expected=True)
 271
 272         # Adult content
 273         if re.search(u'EroConfirmText">', broadcast_page) is not None:
 274
 275             (username, password) = self._get_login_info()
 276             if username is None:
 277                 raise ExtractorError(u'Erotic broadcasts allowed only for registered users, '
 278                     u'use --username and --password options to provide account credentials.', expected=True)
 279
 280             # Log in
 281             login_form_strs = {
 282                 u'login-hint53': '1',
 283                 u'confirm_erotic': '1',
 284                 u'login': username,
 285                 u'password': password,
 286             }
 287             # Convert to UTF-8 *before* urlencode because Python 2.x's urlencode
 288             # chokes on unicode
 289             login_form = dict((k.encode('utf-8'), v.encode('utf-8')) for k,v in login_form_strs.items())
 290             login_data = compat_urllib_parse.urlencode(login_form).encode('utf-8')
 291             login_url = broadcast_url + '/?no_redirect=1'
 292             request = compat_urllib_request.Request(login_url, login_data)
 293             request.add_header('Content-Type', 'application/x-www-form-urlencoded')
 294             broadcast_page = self._download_webpage(
 295                 request, broadcast_id, note=u'Logging in and confirming age')
 296
 297             if re.search(u'>Неверный логин или пароль<', broadcast_page) is not None:
 298                 raise ExtractorError(u'Unable to log in: bad username or password', expected=True)
 299
 300             adult_content = True
 301         else:
 302             adult_content = False
 303
 304         ticket = self._html_search_regex(
 305             u'window\.broadcast_control\.addFlashVar\\(\'file\', \'([^\']+)\'\\);',
 306             broadcast_page, u'broadcast ticket')
 307
 308         url = 'http://smotri.com/broadcast/view/url/?ticket=%s' % ticket
 309
 310         broadcast_password = self._downloader.params.get('videopassword', None)
 311         if broadcast_password:
 312             url += '&pass=%s' % hashlib.md5(broadcast_password.encode('utf-8')).hexdigest()
 313
 314         broadcast_json_page = self._download_webpage(url, broadcast_id, u'Downloading broadcast JSON')
 315
 316         try:
 317             broadcast_json = json.loads(broadcast_json_page)
 318
 319             protected_broadcast = broadcast_json['_pass_protected'] == 1
 320             if protected_broadcast and not broadcast_password:
 321                 raise ExtractorError(u'This broadcast is protected by a password, use the --video-password option', expected=True)
 322
 323             broadcast_offline = broadcast_json['is_play'] == 0
 324             if broadcast_offline:
 325                 raise ExtractorError(u'Broadcast %s is offline' % broadcast_id, expected=True)
 326
 327             rtmp_url = broadcast_json['_server']
 328             if not rtmp_url.startswith('rtmp://'):
 329                 raise ExtractorError(u'Unexpected broadcast rtmp URL')
 330
 331             broadcast_playpath = broadcast_json['_streamName']
 332             broadcast_thumbnail = broadcast_json['_imgURL']
 333             broadcast_title = broadcast_json['title']
 334             broadcast_description = broadcast_json['description']
 335             broadcaster_nick = broadcast_json['nick']
 336             broadcaster_login = broadcast_json['login']
 337             rtmp_conn = 'S:%s' % uuid.uuid4().hex
 338         except KeyError:
 339             if protected_broadcast:
 340                 raise ExtractorError(u'Bad broadcast password', expected=True)
 341             raise ExtractorError(u'Unexpected broadcast JSON')
 342
 343         return {
 344             'id': broadcast_id,
 345             'url': rtmp_url,
 346             'title': broadcast_title,
 347             'thumbnail': broadcast_thumbnail,
 348             'description': broadcast_description,
 349             'uploader': broadcaster_nick,
 350             'uploader_id': broadcaster_login,
 351             'age_limit': 18 if adult_content else 0,
 352             'ext': 'flv',
 353             'play_path': broadcast_playpath,
 354             'rtmp_live': True,
 355             'rtmp_conn': rtmp_conn
 356         }