youtube_dl/extractor/smotri.py

   1 # encoding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import os.path
   5 import re
   6 import json
   7 import hashlib
   8 import uuid
   9
  10 from .common import InfoExtractor
  11 from ..utils import (
  12     compat_urllib_parse,
  13     compat_urllib_request,
  14     ExtractorError,
  15     url_basename,
  16 )
  17
  18
  19 class SmotriIE(InfoExtractor):
  20     IE_DESC = 'Smotri.com'
  21     IE_NAME = 'smotri'
  22     _VALID_URL = r'^https?://(?:www\.)?(?P<url>smotri\.com/video/view/\?id=(?P<videoid>v(?P<realvideoid>[0-9]+)[a-z0-9]{4}))'
  23
  24     _TESTS = [
  25         # real video id 2610366
  26         {
  27             'url': 'http://smotri.com/video/view/?id=v261036632ab',
  28             'file': 'v261036632ab.mp4',
  29             'md5': '2a7b08249e6f5636557579c368040eb9',
  30             'info_dict': {
  31                 'title': 'катастрофа с камер видеонаблюдения',
  32                 'uploader': 'rbc2008',
  33                 'uploader_id': 'rbc08',
  34                 'upload_date': '20131118',
  35                 'description': 'катастрофа с камер видеонаблюдения, видео катастрофа с камер видеонаблюдения',
  36                 'thumbnail': 'http://frame6.loadup.ru/8b/a9/2610366.3.3.jpg',
  37             },
  38         },
  39         # real video id 57591
  40         {
  41             'url': 'http://smotri.com/video/view/?id=v57591cb20',
  42             'file': 'v57591cb20.flv',
  43             'md5': '830266dfc21f077eac5afd1883091bcd',
  44             'info_dict': {
  45                 'title': 'test',
  46                 'uploader': 'Support Photofile@photofile',
  47                 'uploader_id': 'support-photofile',
  48                 'upload_date': '20070704',
  49                 'description': 'test, видео test',
  50                 'thumbnail': 'http://frame4.loadup.ru/03/ed/57591.2.3.jpg',
  51             },
  52         },
  53         # video-password
  54         {
  55             'url': 'http://smotri.com/video/view/?id=v1390466a13c',
  56             'file': 'v1390466a13c.mp4',
  57             'md5': 'f6331cef33cad65a0815ee482a54440b',
  58             'info_dict': {
  59                 'title': 'TOCCA_A_NOI_-_LE_COSE_NON_VANNO_CAMBIAMOLE_ORA-1',
  60                 'uploader': 'timoxa40',
  61                 'uploader_id': 'timoxa40',
  62                 'upload_date': '20100404',
  63                 'thumbnail': 'http://frame7.loadup.ru/af/3f/1390466.3.3.jpg',
  64                 'description': 'TOCCA_A_NOI_-_LE_COSE_NON_VANNO_CAMBIAMOLE_ORA-1, видео TOCCA_A_NOI_-_LE_COSE_NON_VANNO_CAMBIAMOLE_ORA-1',
  65             },
  66             'params': {
  67                 'videopassword': 'qwerty',
  68             },
  69         },
  70         # age limit + video-password
  71         {
  72             'url': 'http://smotri.com/video/view/?id=v15408898bcf',
  73             'file': 'v15408898bcf.flv',
  74             'md5': '91e909c9f0521adf5ee86fbe073aad70',
  75             'info_dict': {
  76                 'title': 'этот ролик не покажут по ТВ',
  77                 'uploader': 'zzxxx',
  78                 'uploader_id': 'ueggb',
  79                 'upload_date': '20101001',
  80                 'thumbnail': 'http://frame3.loadup.ru/75/75/1540889.1.3.jpg',
  81                 'age_limit': 18,
  82                 'description': 'этот ролик не покажут по ТВ, видео этот ролик не покажут по ТВ',
  83             },
  84             'params': {
  85                 'videopassword': '333'
  86             }
  87         }
  88     ]
  89
  90     _SUCCESS = 0
  91     _PASSWORD_NOT_VERIFIED = 1
  92     _PASSWORD_DETECTED = 2
  93     _VIDEO_NOT_FOUND = 3
  94
  95     def _search_meta(self, name, html, display_name=None):
  96         if display_name is None:
  97             display_name = name
  98         return self._html_search_regex(
  99             r'<meta itemprop="%s" content="([^"]+)" />' % re.escape(name),
 100             html, display_name, fatal=False)
 101         return self._html_search_meta(name, html, display_name)
 102
 103     def _real_extract(self, url):
 104         mobj = re.match(self._VALID_URL, url)
 105         video_id = mobj.group('videoid')
 106         real_video_id = mobj.group('realvideoid')
 107
 108         # Download video JSON data
 109         video_json_url = 'http://smotri.com/vt.php?id=%s' % real_video_id
 110         video_json_page = self._download_webpage(video_json_url, video_id, 'Downloading video JSON')
 111         video_json = json.loads(video_json_page)
 112
 113         status = video_json['status']
 114         if status == self._VIDEO_NOT_FOUND:
 115             raise ExtractorError('Video %s does not exist' % video_id, expected=True)
 116         elif status == self._PASSWORD_DETECTED: # The video is protected by a password, retry with
 117                                                 # video-password set
 118             video_password = self._downloader.params.get('videopassword', None)
 119             if not video_password:
 120                 raise ExtractorError('This video is protected by a password, use the --video-password option', expected=True)
 121             video_json_url += '&md5pass=%s' % hashlib.md5(video_password.encode('utf-8')).hexdigest()
 122             video_json_page = self._download_webpage(video_json_url, video_id, 'Downloading video JSON (video-password set)')
 123             video_json = json.loads(video_json_page)
 124             status = video_json['status']
 125             if status == self._PASSWORD_NOT_VERIFIED:
 126                 raise ExtractorError('Video password is invalid', expected=True)
 127
 128         if status != self._SUCCESS:
 129             raise ExtractorError('Unexpected status value %s' % status)
 130
 131         # Extract the URL of the video
 132         video_url = video_json['file_data']
 133
 134         # Video JSON does not provide enough meta data
 135         # We will extract some from the video web page instead
 136         video_page_url = 'http://' + mobj.group('url')
 137         video_page = self._download_webpage(video_page_url, video_id, 'Downloading video page')
 138
 139         # Warning if video is unavailable
 140         warning = self._html_search_regex(
 141             r'<div class="videoUnModer">(.*?)</div>', video_page,
 142             'warning message', default=None)
 143         if warning is not None:
 144             self._downloader.report_warning(
 145                 'Video %s may not be available; smotri said: %s ' %
 146                 (video_id, warning))
 147
 148         # Adult content
 149         if re.search('EroConfirmText">', video_page) is not None:
 150             self.report_age_confirmation()
 151             confirm_string = self._html_search_regex(
 152                 r'<a href="/video/view/\?id=%s&confirm=([^"]+)" title="[^"]+">' % video_id,
 153                 video_page, 'confirm string')
 154             confirm_url = video_page_url + '&confirm=%s' % confirm_string
 155             video_page = self._download_webpage(confirm_url, video_id, 'Downloading video page (age confirmed)')
 156             adult_content = True
 157         else:
 158             adult_content = False
 159
 160         # Extract the rest of meta data
 161         video_title = self._search_meta('name', video_page, 'title')
 162         if not video_title:
 163             video_title = os.path.splitext(url_basename(video_url))[0]
 164
 165         video_description = self._search_meta('description', video_page)
 166         END_TEXT = ' на сайте Smotri.com'
 167         if video_description and video_description.endswith(END_TEXT):
 168             video_description = video_description[:-len(END_TEXT)]
 169         START_TEXT = 'Смотреть онлайн ролик '
 170         if video_description and video_description.startswith(START_TEXT):
 171             video_description = video_description[len(START_TEXT):]
 172         video_thumbnail = self._search_meta('thumbnail', video_page)
 173
 174         upload_date_str = self._search_meta('uploadDate', video_page, 'upload date')
 175         if upload_date_str:
 176             upload_date_m = re.search(r'(?P<year>\d{4})\.(?P<month>\d{2})\.(?P<day>\d{2})T', upload_date_str)
 177             video_upload_date = (
 178                 (
 179                     upload_date_m.group('year') +
 180                     upload_date_m.group('month') +
 181                     upload_date_m.group('day')
 182                 )
 183                 if upload_date_m else None
 184             )
 185         else:
 186             video_upload_date = None
 187
 188         duration_str = self._search_meta('duration', video_page)
 189         if duration_str:
 190             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)
 191             video_duration = (
 192                 (
 193                     (int(duration_m.group('hours')) * 60 * 60) +
 194                     (int(duration_m.group('minutes')) * 60) +
 195                     int(duration_m.group('seconds'))
 196                 )
 197                 if duration_m else None
 198             )
 199         else:
 200             video_duration = None
 201
 202         video_uploader = self._html_search_regex(
 203             '<div class="DescrUser"><div>Автор.*?onmouseover="popup_user_info[^"]+">(.*?)</a>',
 204             video_page, 'uploader', fatal=False, flags=re.MULTILINE|re.DOTALL)
 205
 206         video_uploader_id = self._html_search_regex(
 207             '<div class="DescrUser"><div>Автор.*?onmouseover="popup_user_info\\(.*?\'([^\']+)\'\\);">',
 208             video_page, 'uploader id', fatal=False, flags=re.MULTILINE|re.DOTALL)
 209
 210         video_view_count = self._html_search_regex(
 211             'Общее количество просмотров.*?<span class="Number">(\\d+)</span>',
 212             video_page, 'view count', fatal=False, flags=re.MULTILINE|re.DOTALL)
 213
 214         return {
 215             'id': video_id,
 216             'url': video_url,
 217             'title': video_title,
 218             'thumbnail': video_thumbnail,
 219             'description': video_description,
 220             'uploader': video_uploader,
 221             'upload_date': video_upload_date,
 222             'uploader_id': video_uploader_id,
 223             'duration': video_duration,
 224             'view_count': video_view_count,
 225             'age_limit': 18 if adult_content else 0,
 226             'video_page_url': video_page_url
 227         }
 228
 229
 230 class SmotriCommunityIE(InfoExtractor):
 231     IE_DESC = 'Smotri.com community videos'
 232     IE_NAME = 'smotri:community'
 233     _VALID_URL = r'^https?://(?:www\.)?smotri\.com/community/video/(?P<communityid>[0-9A-Za-z_\'-]+)'
 234
 235     def _real_extract(self, url):
 236         mobj = re.match(self._VALID_URL, url)
 237         community_id = mobj.group('communityid')
 238
 239         url = 'http://smotri.com/export/rss/video/by/community/-/%s/video.xml' % community_id
 240         rss = self._download_xml(url, community_id, 'Downloading community RSS')
 241
 242         entries = [self.url_result(video_url.text, 'Smotri')
 243                    for video_url in rss.findall('./channel/item/link')]
 244
 245         description_text = rss.find('./channel/description').text
 246         community_title = self._html_search_regex(
 247             '^Видео сообщества "([^"]+)"$', description_text, 'community title')
 248
 249         return self.playlist_result(entries, community_id, community_title)
 250
 251
 252 class SmotriUserIE(InfoExtractor):
 253     IE_DESC = 'Smotri.com user videos'
 254     IE_NAME = 'smotri:user'
 255     _VALID_URL = r'^https?://(?:www\.)?smotri\.com/user/(?P<userid>[0-9A-Za-z_\'-]+)'
 256
 257     def _real_extract(self, url):
 258         mobj = re.match(self._VALID_URL, url)
 259         user_id = mobj.group('userid')
 260
 261         url = 'http://smotri.com/export/rss/user/video/-/%s/video.xml' % user_id
 262         rss = self._download_xml(url, user_id, 'Downloading user RSS')
 263
 264         entries = [self.url_result(video_url.text, 'Smotri')
 265                    for video_url in rss.findall('./channel/item/link')]
 266
 267         description_text = rss.find('./channel/description').text
 268         user_nickname = self._html_search_regex(
 269             '^Видео режиссера (.*)$', description_text,
 270             'user nickname')
 271
 272         return self.playlist_result(entries, user_id, user_nickname)
 273
 274
 275 class SmotriBroadcastIE(InfoExtractor):
 276     IE_DESC = 'Smotri.com broadcasts'
 277     IE_NAME = 'smotri:broadcast'
 278     _VALID_URL = r'^https?://(?:www\.)?(?P<url>smotri\.com/live/(?P<broadcastid>[^/]+))/?.*'
 279
 280     def _real_extract(self, url):
 281         mobj = re.match(self._VALID_URL, url)
 282         broadcast_id = mobj.group('broadcastid')
 283
 284         broadcast_url = 'http://' + mobj.group('url')
 285         broadcast_page = self._download_webpage(broadcast_url, broadcast_id, 'Downloading broadcast page')
 286
 287         if re.search('>Режиссер с логином <br/>"%s"<br/> <span>не существует<' % broadcast_id, broadcast_page) is not None:
 288             raise ExtractorError('Broadcast %s does not exist' % broadcast_id, expected=True)
 289
 290         # Adult content
 291         if re.search('EroConfirmText">', broadcast_page) is not None:
 292
 293             (username, password) = self._get_login_info()
 294             if username is None:
 295                 raise ExtractorError('Erotic broadcasts allowed only for registered users, '
 296                     'use --username and --password options to provide account credentials.', expected=True)
 297
 298             login_form = {
 299                 'login-hint53': '1',
 300                 'confirm_erotic': '1',
 301                 'login': username,
 302                 'password': password,
 303             }
 304
 305             request = compat_urllib_request.Request(broadcast_url + '/?no_redirect=1', compat_urllib_parse.urlencode(login_form))
 306             request.add_header('Content-Type', 'application/x-www-form-urlencoded')
 307             broadcast_page = self._download_webpage(request, broadcast_id, 'Logging in and confirming age')
 308
 309             if re.search('>Неверный логин или пароль<', broadcast_page) is not None:
 310                 raise ExtractorError('Unable to log in: bad username or password', expected=True)
 311
 312             adult_content = True
 313         else:
 314             adult_content = False
 315
 316         ticket = self._html_search_regex(
 317             'window\.broadcast_control\.addFlashVar\\(\'file\', \'([^\']+)\'\\);',
 318             broadcast_page, 'broadcast ticket')
 319
 320         url = 'http://smotri.com/broadcast/view/url/?ticket=%s' % ticket
 321
 322         broadcast_password = self._downloader.params.get('videopassword', None)
 323         if broadcast_password:
 324             url += '&pass=%s' % hashlib.md5(broadcast_password.encode('utf-8')).hexdigest()
 325
 326         broadcast_json_page = self._download_webpage(url, broadcast_id, 'Downloading broadcast JSON')
 327
 328         try:
 329             broadcast_json = json.loads(broadcast_json_page)
 330
 331             protected_broadcast = broadcast_json['_pass_protected'] == 1
 332             if protected_broadcast and not broadcast_password:
 333                 raise ExtractorError('This broadcast is protected by a password, use the --video-password option', expected=True)
 334
 335             broadcast_offline = broadcast_json['is_play'] == 0
 336             if broadcast_offline:
 337                 raise ExtractorError('Broadcast %s is offline' % broadcast_id, expected=True)
 338
 339             rtmp_url = broadcast_json['_server']
 340             if not rtmp_url.startswith('rtmp://'):
 341                 raise ExtractorError('Unexpected broadcast rtmp URL')
 342
 343             broadcast_playpath = broadcast_json['_streamName']
 344             broadcast_thumbnail = broadcast_json['_imgURL']
 345             broadcast_title = broadcast_json['title']
 346             broadcast_description = broadcast_json['description']
 347             broadcaster_nick = broadcast_json['nick']
 348             broadcaster_login = broadcast_json['login']
 349             rtmp_conn = 'S:%s' % uuid.uuid4().hex
 350         except KeyError:
 351             if protected_broadcast:
 352                 raise ExtractorError('Bad broadcast password', expected=True)
 353             raise ExtractorError('Unexpected broadcast JSON')
 354
 355         return {
 356             'id': broadcast_id,
 357             'url': rtmp_url,
 358             'title': broadcast_title,
 359             'thumbnail': broadcast_thumbnail,
 360             'description': broadcast_description,
 361             'uploader': broadcaster_nick,
 362             'uploader_id': broadcaster_login,
 363             'age_limit': 18 if adult_content else 0,
 364             'ext': 'flv',
 365             'play_path': broadcast_playpath,
 366             'rtmp_live': True,
 367             'rtmp_conn': rtmp_conn
 368         }