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

   1 # encoding: utf-8
   2
   3 import os.path
   4 import re
   5 import json
   6 import hashlib
   7 import uuid
   8
   9 from .common import InfoExtractor
  10 from ..utils import (
  11     compat_urllib_parse,
  12     compat_urllib_request,
  13     ExtractorError,
  14     url_basename,
  15 )
  16
  17
  18 class SmotriIE(InfoExtractor):
  19     IE_DESC = u'Smotri.com'
  20     IE_NAME = u'smotri'
  21     _VALID_URL = r'^https?://(?:www\.)?(?P<url>smotri\.com/video/view/\?id=(?P<videoid>v(?P<realvideoid>[0-9]+)[a-z0-9]{4}))'
  22
  23     _TESTS = [
  24         # real video id 2610366
  25         {
  26             u'url': u'http://smotri.com/video/view/?id=v261036632ab',
  27             u'file': u'v261036632ab.mp4',
  28             u'md5': u'2a7b08249e6f5636557579c368040eb9',
  29             u'info_dict': {
  30                 u'title': u'катастрофа с камер видеонаблюдения',
  31                 u'uploader': u'rbc2008',
  32                 u'uploader_id': u'rbc08',
  33                 u'upload_date': u'20131118',
  34                 u'description': u'катастрофа с камер видеонаблюдения, видео катастрофа с камер видеонаблюдения',
  35                 u'thumbnail': u'http://frame6.loadup.ru/8b/a9/2610366.3.3.jpg',
  36             },
  37         },
  38         # real video id 57591
  39         {
  40             u'url': u'http://smotri.com/video/view/?id=v57591cb20',
  41             u'file': u'v57591cb20.flv',
  42             u'md5': u'830266dfc21f077eac5afd1883091bcd',
  43             u'info_dict': {
  44                 u'title': u'test',
  45                 u'uploader': u'Support Photofile@photofile',
  46                 u'uploader_id': u'support-photofile',
  47                 u'upload_date': u'20070704',
  48                 u'description': u'test, видео test',
  49                 u'thumbnail': u'http://frame4.loadup.ru/03/ed/57591.2.3.jpg',
  50             },
  51         },
  52         # video-password
  53         {
  54             u'url': u'http://smotri.com/video/view/?id=v1390466a13c',
  55             u'file': u'v1390466a13c.mp4',
  56             u'md5': u'f6331cef33cad65a0815ee482a54440b',
  57             u'info_dict': {
  58                 u'title': u'TOCCA_A_NOI_-_LE_COSE_NON_VANNO_CAMBIAMOLE_ORA-1',
  59                 u'uploader': u'timoxa40',
  60                 u'uploader_id': u'timoxa40',
  61                 u'upload_date': u'20100404',
  62                 u'thumbnail': u'http://frame7.loadup.ru/af/3f/1390466.3.3.jpg',
  63                 u'description': u'TOCCA_A_NOI_-_LE_COSE_NON_VANNO_CAMBIAMOLE_ORA-1, видео TOCCA_A_NOI_-_LE_COSE_NON_VANNO_CAMBIAMOLE_ORA-1',
  64             },
  65             u'params': {
  66                 u'videopassword': u'qwerty',
  67             },
  68         },
  69         # age limit + video-password
  70         {
  71             u'url': u'http://smotri.com/video/view/?id=v15408898bcf',
  72             u'file': u'v15408898bcf.flv',
  73             u'md5': u'91e909c9f0521adf5ee86fbe073aad70',
  74             u'info_dict': {
  75                 u'title': u'этот ролик не покажут по ТВ',
  76                 u'uploader': u'zzxxx',
  77                 u'uploader_id': u'ueggb',
  78                 u'upload_date': u'20101001',
  79                 u'thumbnail': u'http://frame3.loadup.ru/75/75/1540889.1.3.jpg',
  80                 u'age_limit': 18,
  81                 u'description': u'этот ролик не покажут по ТВ, видео этот ролик не покажут по ТВ',
  82             },
  83             u'params': {
  84                 u'videopassword': u'333'
  85             }
  86         }
  87     ]
  88
  89     _SUCCESS = 0
  90     _PASSWORD_NOT_VERIFIED = 1
  91     _PASSWORD_DETECTED = 2
  92     _VIDEO_NOT_FOUND = 3
  93
  94     def _search_meta(self, name, html, display_name=None):
  95         if display_name is None:
  96             display_name = name
  97         return self._html_search_regex(
  98             r'<meta itemprop="%s" content="([^"]+)" />' % re.escape(name),
  99             html, display_name, fatal=False)
 100         return self._html_search_meta(name, html, display_name)
 101
 102     def _real_extract(self, url):
 103         mobj = re.match(self._VALID_URL, url)
 104         video_id = mobj.group('videoid')
 105         real_video_id = mobj.group('realvideoid')
 106
 107         # Download video JSON data
 108         video_json_url = 'http://smotri.com/vt.php?id=%s' % real_video_id
 109         video_json_page = self._download_webpage(video_json_url, video_id, u'Downloading video JSON')
 110         video_json = json.loads(video_json_page)
 111
 112         status = video_json['status']
 113         if status == self._VIDEO_NOT_FOUND:
 114             raise ExtractorError(u'Video %s does not exist' % video_id, expected=True)
 115         elif status == self._PASSWORD_DETECTED:  # The video is protected by a password, retry with
 116                                                 # video-password set
 117             video_password = self._downloader.params.get('videopassword', None)
 118             if not video_password:
 119                 raise ExtractorError(u'This video is protected by a password, use the --video-password option', expected=True)
 120             video_json_url += '&md5pass=%s' % hashlib.md5(video_password.encode('utf-8')).hexdigest()
 121             video_json_page = self._download_webpage(video_json_url, video_id, u'Downloading video JSON (video-password set)')
 122             video_json = json.loads(video_json_page)
 123             status = video_json['status']
 124             if status == self._PASSWORD_NOT_VERIFIED:
 125                 raise ExtractorError(u'Video password is invalid', expected=True)
 126
 127         if status != self._SUCCESS:
 128             raise ExtractorError(u'Unexpected status value %s' % status)
 129
 130         # Extract the URL of the video
 131         video_url = video_json['file_data']
 132
 133         # Video JSON does not provide enough meta data
 134         # We will extract some from the video web page instead
 135         video_page_url = 'http://' + mobj.group('url')
 136         video_page = self._download_webpage(video_page_url, video_id, u'Downloading video page')
 137
 138         # Warning if video is unavailable
 139         warning = self._html_search_regex(
 140             r'<div class="videoUnModer">(.*?)</div>', video_page,
 141             u'warning message', default=None)
 142         if warning is not None:
 143             self._downloader.report_warning(
 144                 u'Video %s may not be available; smotri said: %s ' %
 145                 (video_id, warning))
 146
 147         # Adult content
 148         if re.search(u'EroConfirmText">', video_page) is not None:
 149             self.report_age_confirmation()
 150             confirm_string = self._html_search_regex(
 151                 r'<a href="/video/view/\?id=%s&confirm=([^"]+)" title="[^"]+">' % video_id,
 152                 video_page, u'confirm string')
 153             confirm_url = video_page_url + '&confirm=%s' % confirm_string
 154             video_page = self._download_webpage(confirm_url, video_id, u'Downloading video page (age confirmed)')
 155             adult_content = True
 156         else:
 157             adult_content = False
 158
 159         # Extract the rest of meta data
 160         video_title = self._search_meta(u'name', video_page, u'title')
 161         if not video_title:
 162             video_title = os.path.splitext(url_basename(video_url))[0]
 163
 164         video_description = self._search_meta(u'description', video_page)
 165         END_TEXT = u' на сайте Smotri.com'
 166         if video_description and video_description.endswith(END_TEXT):
 167             video_description = video_description[:-len(END_TEXT)]
 168         START_TEXT = u'Смотреть онлайн ролик '
 169         if video_description and video_description.startswith(START_TEXT):
 170             video_description = video_description[len(START_TEXT):]
 171         video_thumbnail = self._search_meta(u'thumbnail', video_page)
 172
 173         upload_date_str = self._search_meta(u'uploadDate', video_page, u'upload date')
 174         if upload_date_str:
 175             upload_date_m = re.search(r'(?P<year>\d{4})\.(?P<month>\d{2})\.(?P<day>\d{2})T', upload_date_str)
 176             video_upload_date = (
 177                 (
 178                     upload_date_m.group('year') +
 179                     upload_date_m.group('month') +
 180                     upload_date_m.group('day')
 181                 )
 182                 if upload_date_m else None
 183             )
 184         else:
 185             video_upload_date = None
 186
 187         duration_str = self._search_meta(u'duration', video_page)
 188         if duration_str:
 189             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)
 190             video_duration = (
 191                 (
 192                     (int(duration_m.group('hours')) * 60 * 60) +
 193                     (int(duration_m.group('minutes')) * 60) +
 194                     int(duration_m.group('seconds'))
 195                 )
 196                 if duration_m else None
 197             )
 198         else:
 199             video_duration = None
 200
 201         video_uploader = self._html_search_regex(
 202             u'<div class="DescrUser"><div>Автор.*?onmouseover="popup_user_info[^"]+">(.*?)</a>',
 203             video_page, u'uploader', fatal=False, flags=re.MULTILINE|re.DOTALL)
 204
 205         video_uploader_id = self._html_search_regex(
 206             u'<div class="DescrUser"><div>Автор.*?onmouseover="popup_user_info\\(.*?\'([^\']+)\'\\);">',
 207             video_page, u'uploader id', fatal=False, flags=re.MULTILINE|re.DOTALL)
 208
 209         video_view_count = self._html_search_regex(
 210             u'Общее количество просмотров.*?<span class="Number">(\\d+)</span>',
 211             video_page, u'view count', fatal=False, flags=re.MULTILINE|re.DOTALL)
 212
 213         return {
 214             'id': video_id,
 215             'url': video_url,
 216             'title': video_title,
 217             'thumbnail': video_thumbnail,
 218             'description': video_description,
 219             'uploader': video_uploader,
 220             'upload_date': video_upload_date,
 221             'uploader_id': video_uploader_id,
 222             'duration': video_duration,
 223             'view_count': video_view_count,
 224             'age_limit': 18 if adult_content else 0,
 225             'video_page_url': video_page_url
 226         }
 227
 228
 229 class SmotriCommunityIE(InfoExtractor):
 230     IE_DESC = u'Smotri.com community videos'
 231     IE_NAME = u'smotri:community'
 232     _VALID_URL = r'^https?://(?:www\.)?smotri\.com/community/video/(?P<communityid>[0-9A-Za-z_\'-]+)'
 233
 234     def _real_extract(self, url):
 235         mobj = re.match(self._VALID_URL, url)
 236         community_id = mobj.group('communityid')
 237
 238         url = 'http://smotri.com/export/rss/video/by/community/-/%s/video.xml' % community_id
 239         rss = self._download_xml(url, community_id, u'Downloading community RSS')
 240
 241         entries = [self.url_result(video_url.text, 'Smotri')
 242                    for video_url in rss.findall('./channel/item/link')]
 243
 244         description_text = rss.find('./channel/description').text
 245         community_title = self._html_search_regex(
 246             u'^Видео сообщества "([^"]+)"$', description_text, u'community title')
 247
 248         return self.playlist_result(entries, community_id, community_title)
 249
 250
 251 class SmotriUserIE(InfoExtractor):
 252     IE_DESC = u'Smotri.com user videos'
 253     IE_NAME = u'smotri:user'
 254     _VALID_URL = r'^https?://(?:www\.)?smotri\.com/user/(?P<userid>[0-9A-Za-z_\'-]+)'
 255
 256     def _real_extract(self, url):
 257         mobj = re.match(self._VALID_URL, url)
 258         user_id = mobj.group('userid')
 259
 260         url = 'http://smotri.com/export/rss/user/video/-/%s/video.xml' % user_id
 261         rss = self._download_xml(url, user_id, u'Downloading user RSS')
 262
 263         entries = [self.url_result(video_url.text, 'Smotri')
 264                    for video_url in rss.findall('./channel/item/link')]
 265
 266         description_text = rss.find('./channel/description').text
 267         user_nickname = self._html_search_regex(
 268             u'^Видео режиссера (.*)$', description_text,
 269             u'user nickname')
 270
 271         return self.playlist_result(entries, user_id, user_nickname)
 272
 273
 274 class SmotriBroadcastIE(InfoExtractor):
 275     IE_DESC = u'Smotri.com broadcasts'
 276     IE_NAME = u'smotri:broadcast'
 277     _VALID_URL = r'^https?://(?:www\.)?(?P<url>smotri\.com/live/(?P<broadcastid>[^/]+))/?.*'
 278
 279     def _real_extract(self, url):
 280         mobj = re.match(self._VALID_URL, url)
 281         broadcast_id = mobj.group('broadcastid')
 282
 283         broadcast_url = 'http://' + mobj.group('url')
 284         broadcast_page = self._download_webpage(broadcast_url, broadcast_id, u'Downloading broadcast page')
 285
 286         if re.search(u'>Режиссер с логином <br/>"%s"<br/> <span>не существует<' % broadcast_id, broadcast_page) is not None:
 287             raise ExtractorError(u'Broadcast %s does not exist' % broadcast_id, expected=True)
 288
 289         # Adult content
 290         if re.search(u'EroConfirmText">', broadcast_page) is not None:
 291
 292             (username, password) = self._get_login_info()
 293             if username is None:
 294                 raise ExtractorError(u'Erotic broadcasts allowed only for registered users, '
 295                     u'use --username and --password options to provide account credentials.', expected=True)
 296
 297             # Log in
 298             login_form_strs = {
 299                 u'login-hint53': '1',
 300                 u'confirm_erotic': '1',
 301                 u'login': username,
 302                 u'password': password,
 303             }
 304             # Convert to UTF-8 *before* urlencode because Python 2.x's urlencode
 305             # chokes on unicode
 306             login_form = dict((k.encode('utf-8'), v.encode('utf-8')) for k,v in login_form_strs.items())
 307             login_data = compat_urllib_parse.urlencode(login_form).encode('utf-8')
 308             login_url = broadcast_url + '/?no_redirect=1'
 309             request = compat_urllib_request.Request(login_url, login_data)
 310             request.add_header('Content-Type', 'application/x-www-form-urlencoded')
 311             broadcast_page = self._download_webpage(
 312                 request, broadcast_id, note=u'Logging in and confirming age')
 313
 314             if re.search(u'>Неверный логин или пароль<', broadcast_page) is not None:
 315                 raise ExtractorError(u'Unable to log in: bad username or password', expected=True)
 316
 317             adult_content = True
 318         else:
 319             adult_content = False
 320
 321         ticket = self._html_search_regex(
 322             u'window\.broadcast_control\.addFlashVar\\(\'file\', \'([^\']+)\'\\);',
 323             broadcast_page, u'broadcast ticket')
 324
 325         url = 'http://smotri.com/broadcast/view/url/?ticket=%s' % ticket
 326
 327         broadcast_password = self._downloader.params.get('videopassword', None)
 328         if broadcast_password:
 329             url += '&pass=%s' % hashlib.md5(broadcast_password.encode('utf-8')).hexdigest()
 330
 331         broadcast_json_page = self._download_webpage(url, broadcast_id, u'Downloading broadcast JSON')
 332
 333         try:
 334             broadcast_json = json.loads(broadcast_json_page)
 335
 336             protected_broadcast = broadcast_json['_pass_protected'] == 1
 337             if protected_broadcast and not broadcast_password:
 338                 raise ExtractorError(u'This broadcast is protected by a password, use the --video-password option', expected=True)
 339
 340             broadcast_offline = broadcast_json['is_play'] == 0
 341             if broadcast_offline:
 342                 raise ExtractorError(u'Broadcast %s is offline' % broadcast_id, expected=True)
 343
 344             rtmp_url = broadcast_json['_server']
 345             if not rtmp_url.startswith('rtmp://'):
 346                 raise ExtractorError(u'Unexpected broadcast rtmp URL')
 347
 348             broadcast_playpath = broadcast_json['_streamName']
 349             broadcast_thumbnail = broadcast_json['_imgURL']
 350             broadcast_title = broadcast_json['title']
 351             broadcast_description = broadcast_json['description']
 352             broadcaster_nick = broadcast_json['nick']
 353             broadcaster_login = broadcast_json['login']
 354             rtmp_conn = 'S:%s' % uuid.uuid4().hex
 355         except KeyError:
 356             if protected_broadcast:
 357                 raise ExtractorError(u'Bad broadcast password', expected=True)
 358             raise ExtractorError(u'Unexpected broadcast JSON')
 359
 360         return {
 361             'id': broadcast_id,
 362             'url': rtmp_url,
 363             'title': broadcast_title,
 364             'thumbnail': broadcast_thumbnail,
 365             'description': broadcast_description,
 366             'uploader': broadcaster_nick,
 367             'uploader_id': broadcaster_login,
 368             'age_limit': 18 if adult_content else 0,
 369             'ext': 'flv',
 370             'play_path': broadcast_playpath,
 371             'rtmp_live': True,
 372             'rtmp_conn': rtmp_conn
 373         }