youtube_dl/extractor/smotri.py

   1 # encoding: utf-8
   2
   3 import re
   4 import json
   5 import hashlib
   6
   7 from .common import InfoExtractor
   8 from ..utils import (
   9     ExtractorError
  10 )
  11
  12
  13 class SmotriIE(InfoExtractor):
  14     IE_DESC = u'Smotri.com'
  15     IE_NAME = u'smotri'
  16     _VALID_URL = r'^https?://(?:www\.)?(?P<url>smotri\.com/video/view/\?id=(?P<videoid>v(?P<realvideoid>[0-9]+)[a-z0-9]{4}))'
  17
  18     _TESTS = [
  19         # real video id 2610366
  20         {
  21             u'url': u'http://smotri.com/video/view/?id=v261036632ab',
  22             u'file': u'v261036632ab.mp4',
  23             u'md5': u'2a7b08249e6f5636557579c368040eb9',
  24             u'info_dict': {
  25                 u'title': u'катастрофа с камер видеонаблюдения',
  26                 u'uploader': u'rbc2008',
  27                 u'uploader_id': u'rbc08',
  28                 u'upload_date': u'20131118',
  29                 u'description': u'катастрофа с камер видеонаблюдения, видео катастрофа с камер видеонаблюдения',
  30                 u'thumbnail': u'http://frame6.loadup.ru/8b/a9/2610366.3.3.jpg',
  31             },
  32         },
  33         # real video id 57591
  34         {
  35             u'url': u'http://smotri.com/video/view/?id=v57591cb20',
  36             u'file': u'v57591cb20.flv',
  37             u'md5': u'830266dfc21f077eac5afd1883091bcd',
  38             u'info_dict': {
  39                 u'title': u'test',
  40                 u'uploader': u'Support Photofile@photofile',
  41                 u'uploader_id': u'support-photofile',
  42                 u'upload_date': u'20070704',
  43                 u'description': u'test, видео test',
  44                 u'thumbnail': u'http://frame4.loadup.ru/03/ed/57591.2.3.jpg',
  45             },
  46         },
  47         # video-password
  48         {
  49             u'url': u'http://smotri.com/video/view/?id=v1390466a13c',
  50             u'file': u'v1390466a13c.mp4',
  51             u'md5': u'f6331cef33cad65a0815ee482a54440b',
  52             u'info_dict': {
  53                 u'title': u'TOCCA_A_NOI_-_LE_COSE_NON_VANNO_CAMBIAMOLE_ORA-1',
  54                 u'uploader': u'timoxa40',
  55                 u'uploader_id': u'timoxa40',
  56                 u'upload_date': u'20100404',
  57                 u'thumbnail': u'http://frame7.loadup.ru/af/3f/1390466.3.3.jpg',
  58                 u'description': u'TOCCA_A_NOI_-_LE_COSE_NON_VANNO_CAMBIAMOLE_ORA-1, видео TOCCA_A_NOI_-_LE_COSE_NON_VANNO_CAMBIAMOLE_ORA-1',
  59             },
  60             u'params': {
  61                 u'videopassword': u'qwerty',
  62             },
  63         },
  64         # age limit + video-password
  65         {
  66             u'url': u'http://smotri.com/video/view/?id=v15408898bcf',
  67             u'file': u'v15408898bcf.flv',
  68             u'md5': u'91e909c9f0521adf5ee86fbe073aad70',
  69             u'info_dict': {
  70                 u'title': u'этот ролик не покажут по ТВ',
  71                 u'uploader': u'zzxxx',
  72                 u'uploader_id': u'ueggb',
  73                 u'upload_date': u'20101001',
  74                 u'thumbnail': u'http://frame3.loadup.ru/75/75/1540889.1.3.jpg',
  75                 u'age_limit': 18,
  76                 u'description': u'этот ролик не покажут по ТВ, видео этот ролик не покажут по ТВ',
  77             },
  78             u'params': {
  79                 u'videopassword': u'333'
  80             }
  81         }
  82     ]
  83
  84     _SUCCESS = 0
  85     _PASSWORD_NOT_VERIFIED = 1
  86     _PASSWORD_DETECTED = 2
  87     _VIDEO_NOT_FOUND = 3
  88
  89     def _search_meta(self, name, html, display_name=None):
  90         if display_name is None:
  91             display_name = name
  92         return self._html_search_regex(
  93             r'<meta itemprop="%s" content="([^"]+)" />' % re.escape(name),
  94             html, display_name, fatal=False)
  95         return self._html_search_meta(name, html, display_name)
  96
  97     def _real_extract(self, url):
  98         mobj = re.match(self._VALID_URL, url)
  99         video_id = mobj.group('videoid')
 100         real_video_id = mobj.group('realvideoid')
 101
 102         # Download video JSON data
 103         video_json_url = 'http://smotri.com/vt.php?id=%s' % real_video_id
 104         video_json_page = self._download_webpage(video_json_url, video_id, u'Downloading video JSON')
 105         video_json = json.loads(video_json_page)
 106
 107         status = video_json['status']
 108         if status == self._VIDEO_NOT_FOUND:
 109             raise ExtractorError(u'Video %s does not exist' % video_id, expected=True)
 110         elif status == self._PASSWORD_DETECTED:  # The video is protected by a password, retry with
 111                                                 # video-password set
 112             video_password = self._downloader.params.get('videopassword', None)
 113             if not video_password:
 114                 raise ExtractorError(u'This video is protected by a password, use the --video-password option', expected=True)
 115             video_json_url += '&md5pass=%s' % hashlib.md5(video_password.encode('utf-8')).hexdigest()
 116             video_json_page = self._download_webpage(video_json_url, video_id, u'Downloading video JSON (video-password set)')
 117             video_json = json.loads(video_json_page)
 118             status = video_json['status']
 119             if status == self._PASSWORD_NOT_VERIFIED:
 120                 raise ExtractorError(u'Video password is invalid', expected=True)
 121
 122         if status != self._SUCCESS:
 123             raise ExtractorError(u'Unexpected status value %s' % status)
 124
 125         # Extract the URL of the video
 126         video_url = video_json['file_data']
 127
 128         # Video JSON does not provide enough meta data
 129         # We will extract some from the video web page instead
 130         video_page_url = 'http://' + mobj.group('url')
 131         video_page = self._download_webpage(video_page_url, video_id, u'Downloading video page')
 132
 133         # Adult content
 134         if re.search(u'EroConfirmText">', video_page) is not None:
 135             self.report_age_confirmation()
 136             confirm_string = self._html_search_regex(
 137                 r'<a href="/video/view/\?id=%s&confirm=([^"]+)" title="[^"]+">' % video_id,
 138                 video_page, u'confirm string')
 139             confirm_url = video_page_url + '&confirm=%s' % confirm_string
 140             video_page = self._download_webpage(confirm_url, video_id, u'Downloading video page (age confirmed)')
 141             adult_content = True
 142         else:
 143             adult_content = False
 144
 145         # Extract the rest of meta data
 146         video_title = self._search_meta(u'name', video_page, u'title')
 147         if not video_title:
 148             video_title = video_url.rsplit('/', 1)[-1]
 149
 150         video_description = self._search_meta(u'description', video_page)
 151         END_TEXT = u' на сайте Smotri.com'
 152         if video_description.endswith(END_TEXT):
 153             video_description = video_description[:-len(END_TEXT)]
 154         START_TEXT = u'Смотреть онлайн ролик '
 155         if video_description.startswith(START_TEXT):
 156             video_description = video_description[len(START_TEXT):]
 157         video_thumbnail = self._search_meta(u'thumbnail', video_page)
 158
 159         upload_date_str = self._search_meta(u'uploadDate', video_page, u'upload date')
 160         upload_date_m = re.search(r'(?P<year>\d{4})\.(?P<month>\d{2})\.(?P<day>\d{2})T', upload_date_str)
 161         video_upload_date = (
 162             (
 163                 upload_date_m.group('year') +
 164                 upload_date_m.group('month') +
 165                 upload_date_m.group('day')
 166             )
 167             if upload_date_m else None
 168         )
 169
 170         duration_str = self._search_meta(u'duration', video_page)
 171         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)
 172         video_duration = (
 173             (
 174                 (int(duration_m.group('hours')) * 60 * 60) +
 175                 (int(duration_m.group('minutes')) * 60) +
 176                 int(duration_m.group('seconds'))
 177             )
 178             if duration_m else None
 179         )
 180
 181         video_uploader = self._html_search_regex(
 182             u'<div class="DescrUser"><div>Автор.*?onmouseover="popup_user_info[^"]+">(.*?)</a>',
 183             video_page, u'uploader', fatal=False, flags=re.MULTILINE|re.DOTALL)
 184
 185         video_uploader_id = self._html_search_regex(
 186             u'<div class="DescrUser"><div>Автор.*?onmouseover="popup_user_info\\(.*?\'([^\']+)\'\\);">',
 187             video_page, u'uploader id', fatal=False, flags=re.MULTILINE|re.DOTALL)
 188
 189         video_view_count = self._html_search_regex(
 190             u'Общее количество просмотров.*?<span class="Number">(\\d+)</span>',
 191             video_page, u'view count', fatal=False, flags=re.MULTILINE|re.DOTALL)
 192
 193         return {
 194             'id': video_id,
 195             'url': video_url,
 196             'title': video_title,
 197             'thumbnail': video_thumbnail,
 198             'description': video_description,
 199             'uploader': video_uploader,
 200             'upload_date': video_upload_date,
 201             'uploader_id': video_uploader_id,
 202             'video_duration': video_duration,
 203             'view_count': video_view_count,
 204             'age_limit': 18 if adult_content else 0,
 205             'video_page_url': video_page_url
 206         }
 207
 208
 209 class SmotriCommunityIE(InfoExtractor):
 210     IE_DESC = u'Smotri.com community videos'
 211     IE_NAME = u'smotri:community'
 212     _VALID_URL = r'^https?://(?:www\.)?smotri\.com/community/video/(?P<communityid>[0-9A-Za-z_\'-]+)'
 213
 214     def _real_extract(self, url):
 215         mobj = re.match(self._VALID_URL, url)
 216         community_id = mobj.group('communityid')
 217
 218         url = 'http://smotri.com/export/rss/video/by/community/-/%s/video.xml' % community_id
 219         rss = self._download_xml(url, community_id, u'Downloading community RSS')
 220
 221         entries = [self.url_result(video_url.text, 'Smotri')
 222                    for video_url in rss.findall('./channel/item/link')]
 223
 224         description_text = rss.find('./channel/description').text
 225         community_title = self._html_search_regex(
 226             u'^Видео сообщества "([^"]+)"$', description_text, u'community title')
 227
 228         return self.playlist_result(entries, community_id, community_title)
 229
 230
 231 class SmotriUserIE(InfoExtractor):
 232     IE_DESC = u'Smotri.com user videos'
 233     IE_NAME = u'smotri:user'
 234     _VALID_URL = r'^https?://(?:www\.)?smotri\.com/user/(?P<userid>[0-9A-Za-z_\'-]+)'
 235
 236     def _real_extract(self, url):
 237         mobj = re.match(self._VALID_URL, url)
 238         user_id = mobj.group('userid')
 239
 240         url = 'http://smotri.com/export/rss/user/video/-/%s/video.xml' % user_id
 241         rss = self._download_xml(url, user_id, u'Downloading user RSS')
 242
 243         entries = [self.url_result(video_url.text, 'Smotri')
 244                    for video_url in rss.findall('./channel/item/link')]
 245
 246         description_text = rss.find('./channel/description').text
 247         user_nickname = self._html_search_regex(
 248             u'^Видео режиссера (.*)$', description_text,
 249             u'user nickname')
 250
 251         return self.playlist_result(entries, user_id, user_nickname)