_ Git - youtube-dl/blob - 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 and video_description.endswith(END_TEXT):
 156             video_description = video_description[:-len(END_TEXT)]
 157         START_TEXT = u'Смотреть онлайн ролик '
 158         if video_description and 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         if upload_date_str:
 164             upload_date_m = re.search(r'(?P<year>\d{4})\.(?P<month>\d{2})\.(?P<day>\d{2})T', upload_date_str)
 165             video_upload_date = (
 166                 (
 167                     upload_date_m.group('year') +
 168                     upload_date_m.group('month') +
 169                     upload_date_m.group('day')
 170                 )
 171                 if upload_date_m else None
 172             )
 173         else:
 174             video_upload_date = None
 175
 176         duration_str = self._search_meta(u'duration', video_page)
 177         if duration_str:
 178             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)
 179             video_duration = (
 180                 (
 181                     (int(duration_m.group('hours')) * 60 * 60) +
 182                     (int(duration_m.group('minutes')) * 60) +
 183                     int(duration_m.group('seconds'))
 184                 )
 185                 if duration_m else None
 186             )
 187         else:
 188             video_duration = None
 189
 190         video_uploader = self._html_search_regex(
 191             u'<div class="DescrUser"><div>Автор.*?onmouseover="popup_user_info[^"]+">(.*?)</a>',
 192             video_page, u'uploader', fatal=False, flags=re.MULTILINE|re.DOTALL)
 193
 194         video_uploader_id = self._html_search_regex(
 195             u'<div class="DescrUser"><div>Автор.*?onmouseover="popup_user_info\\(.*?\'([^\']+)\'\\);">',
 196             video_page, u'uploader id', fatal=False, flags=re.MULTILINE|re.DOTALL)
 197
 198         video_view_count = self._html_search_regex(
 199             u'Общее количество просмотров.*?<span class="Number">(\\d+)</span>',
 200             video_page, u'view count', fatal=False, flags=re.MULTILINE|re.DOTALL)
 201
 202         return {
 203             'id': video_id,
 204             'url': video_url,
 205             'title': video_title,
 206             'thumbnail': video_thumbnail,
 207             'description': video_description,
 208             'uploader': video_uploader,
 209             'upload_date': video_upload_date,
 210             'uploader_id': video_uploader_id,
 211             'duration': video_duration,
 212             'view_count': video_view_count,
 213             'age_limit': 18 if adult_content else 0,
 214             'video_page_url': video_page_url
 215         }
 216
 217
 218 class SmotriCommunityIE(InfoExtractor):
 219     IE_DESC = u'Smotri.com community videos'
 220     IE_NAME = u'smotri:community'
 221     _VALID_URL = r'^https?://(?:www\.)?smotri\.com/community/video/(?P<communityid>[0-9A-Za-z_\'-]+)'
 222
 223     def _real_extract(self, url):
 224         mobj = re.match(self._VALID_URL, url)
 225         community_id = mobj.group('communityid')
 226
 227         url = 'http://smotri.com/export/rss/video/by/community/-/%s/video.xml' % community_id
 228         rss = self._download_xml(url, community_id, u'Downloading community RSS')
 229
 230         entries = [self.url_result(video_url.text, 'Smotri')
 231                    for video_url in rss.findall('./channel/item/link')]
 232
 233         description_text = rss.find('./channel/description').text
 234         community_title = self._html_search_regex(
 235             u'^Видео сообщества "([^"]+)"$', description_text, u'community title')
 236
 237         return self.playlist_result(entries, community_id, community_title)
 238
 239
 240 class SmotriUserIE(InfoExtractor):
 241     IE_DESC = u'Smotri.com user videos'
 242     IE_NAME = u'smotri:user'
 243     _VALID_URL = r'^https?://(?:www\.)?smotri\.com/user/(?P<userid>[0-9A-Za-z_\'-]+)'
 244
 245     def _real_extract(self, url):
 246         mobj = re.match(self._VALID_URL, url)
 247         user_id = mobj.group('userid')
 248
 249         url = 'http://smotri.com/export/rss/user/video/-/%s/video.xml' % user_id
 250         rss = self._download_xml(url, user_id, u'Downloading user RSS')
 251
 252         entries = [self.url_result(video_url.text, 'Smotri')
 253                    for video_url in rss.findall('./channel/item/link')]
 254
 255         description_text = rss.find('./channel/description').text
 256         user_nickname = self._html_search_regex(
 257             u'^Видео режиссера (.*)$', description_text,
 258             u'user nickname')
 259
 260         return self.playlist_result(entries, user_id, user_nickname)
 261
 262
 263 class SmotriBroadcastIE(InfoExtractor):
 264     IE_DESC = u'Smotri.com broadcasts'
 265     IE_NAME = u'smotri:broadcast'
 266     _VALID_URL = r'^https?://(?:www\.)?(?P<url>smotri\.com/live/(?P<broadcastid>[^/]+))/?.*'
 267
 268     def _real_extract(self, url):
 269         mobj = re.match(self._VALID_URL, url)
 270         broadcast_id = mobj.group('broadcastid')
 271
 272         broadcast_url = 'http://' + mobj.group('url')
 273         broadcast_page = self._download_webpage(broadcast_url, broadcast_id, u'Downloading broadcast page')
 274
 275         if re.search(u'>Режиссер с логином <br/>"%s"<br/> <span>не существует<' % broadcast_id, broadcast_page) is not None:
 276             raise ExtractorError(u'Broadcast %s does not exist' % broadcast_id, expected=True)
 277
 278         # Adult content
 279         if re.search(u'EroConfirmText">', broadcast_page) is not None:
 280
 281             (username, password) = self._get_login_info()
 282             if username is None:
 283                 raise ExtractorError(u'Erotic broadcasts allowed only for registered users, '
 284                     u'use --username and --password options to provide account credentials.', expected=True)
 285
 286             # Log in
 287             login_form_strs = {
 288                 u'login-hint53': '1',
 289                 u'confirm_erotic': '1',
 290                 u'login': username,
 291                 u'password': password,
 292             }
 293             # Convert to UTF-8 *before* urlencode because Python 2.x's urlencode
 294             # chokes on unicode
 295             login_form = dict((k.encode('utf-8'), v.encode('utf-8')) for k,v in login_form_strs.items())
 296             login_data = compat_urllib_parse.urlencode(login_form).encode('utf-8')
 297             login_url = broadcast_url + '/?no_redirect=1'
 298             request = compat_urllib_request.Request(login_url, login_data)
 299             request.add_header('Content-Type', 'application/x-www-form-urlencoded')
 300             broadcast_page = self._download_webpage(
 301                 request, broadcast_id, note=u'Logging in and confirming age')
 302
 303             if re.search(u'>Неверный логин или пароль<', broadcast_page) is not None:
 304                 raise ExtractorError(u'Unable to log in: bad username or password', expected=True)
 305
 306             adult_content = True
 307         else:
 308             adult_content = False
 309
 310         ticket = self._html_search_regex(
 311             u'window\.broadcast_control\.addFlashVar\\(\'file\', \'([^\']+)\'\\);',
 312             broadcast_page, u'broadcast ticket')
 313
 314         url = 'http://smotri.com/broadcast/view/url/?ticket=%s' % ticket
 315
 316         broadcast_password = self._downloader.params.get('videopassword', None)
 317         if broadcast_password:
 318             url += '&pass=%s' % hashlib.md5(broadcast_password.encode('utf-8')).hexdigest()
 319
 320         broadcast_json_page = self._download_webpage(url, broadcast_id, u'Downloading broadcast JSON')
 321
 322         try:
 323             broadcast_json = json.loads(broadcast_json_page)
 324
 325             protected_broadcast = broadcast_json['_pass_protected'] == 1
 326             if protected_broadcast and not broadcast_password:
 327                 raise ExtractorError(u'This broadcast is protected by a password, use the --video-password option', expected=True)
 328
 329             broadcast_offline = broadcast_json['is_play'] == 0
 330             if broadcast_offline:
 331                 raise ExtractorError(u'Broadcast %s is offline' % broadcast_id, expected=True)
 332
 333             rtmp_url = broadcast_json['_server']
 334             if not rtmp_url.startswith('rtmp://'):
 335                 raise ExtractorError(u'Unexpected broadcast rtmp URL')
 336
 337             broadcast_playpath = broadcast_json['_streamName']
 338             broadcast_thumbnail = broadcast_json['_imgURL']
 339             broadcast_title = broadcast_json['title']
 340             broadcast_description = broadcast_json['description']
 341             broadcaster_nick = broadcast_json['nick']
 342             broadcaster_login = broadcast_json['login']
 343             rtmp_conn = 'S:%s' % uuid.uuid4().hex
 344         except KeyError:
 345             if protected_broadcast:
 346                 raise ExtractorError(u'Bad broadcast password', expected=True)
 347             raise ExtractorError(u'Unexpected broadcast JSON')
 348
 349         return {
 350             'id': broadcast_id,
 351             'url': rtmp_url,
 352             'title': broadcast_title,
 353             'thumbnail': broadcast_thumbnail,
 354             'description': broadcast_description,
 355             'uploader': broadcaster_nick,
 356             'uploader_id': broadcaster_login,
 357             'age_limit': 18 if adult_content else 0,
 358             'ext': 'flv',
 359             'play_path': broadcast_playpath,
 360             'rtmp_live': True,
 361             'rtmp_conn': rtmp_conn
 362         }