_ Git - youtube-dl/blob - 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     determine_ext,
  10     ExtractorError
  11 )
  12
  13
  14 class SmotriIE(InfoExtractor):
  15     IE_DESC = u'Smotri.com'
  16     IE_NAME = u'smotri'
  17     _VALID_URL = r'^(?:http://)?(?:www\.)?(?P<url>smotri\.com/video/view/\?id=(?P<videoid>v(?P<realvideoid>[0-9]+)[a-z0-9]{4}))'
  18
  19     _TESTS = [
  20         # real video id 2610366
  21         {
  22             u'url': u'http://smotri.com/video/view/?id=v261036632ab',
  23             u'file': u'v261036632ab.mp4',
  24             u'md5': u'46a72e83a6ad8862b64fa6953fa93f8a',
  25             u'info_dict': {
  26                 u'title': u'катастрофа с камер видеонаблюдения',
  27                 u'uploader': u'rbc2008',
  28                 u'uploader_id': u'rbc08',
  29                 u'upload_date': u'20131118',
  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'9eae59f6dda7087bf39a140e2fff5757',
  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'thumbnail': u'http://frame4.loadup.ru/03/ed/57591.2.3.jpg'
  44             },
  45         },
  46         # video-password
  47         {
  48             u'url': u'http://smotri.com/video/view/?id=v1390466a13c',
  49             u'file': u'v1390466a13c.mp4',
  50             u'md5': u'fe4dd9357558d5ee3c8fc0ef0d39de66',
  51             u'info_dict': {
  52                 u'title': u'TOCCA_A_NOI_-_LE_COSE_NON_VANNO_CAMBIAMOLE_ORA-1',
  53                 u'uploader': u'timoxa40',
  54                 u'uploader_id': u'timoxa40',
  55                 u'upload_date': u'20100404',
  56                 u'thumbnail': u'http://frame7.loadup.ru/af/3f/1390466.3.3.jpg'
  57             },
  58             u'params': {
  59                 u'videopassword': u'qwerty',
  60             },
  61         },
  62         # age limit + video-password
  63         {
  64             u'url': u'http://smotri.com/video/view/?id=v15408898bcf',
  65             u'file': u'v15408898bcf.flv',
  66             u'md5': u'c66a5d61379ac6fde06f07eebe436316',
  67             u'info_dict': {
  68                 u'title': u'этот ролик не покажут по ТВ',
  69                 u'uploader': u'zzxxx',
  70                 u'uploader_id': u'ueggb',
  71                 u'upload_date': u'20101001',
  72                 u'thumbnail': u'http://frame3.loadup.ru/75/75/1540889.1.3.jpg',
  73                 u'age_limit': 18
  74             },
  75             u'params': {
  76                 u'videopassword': u'333'
  77             }
  78         }
  79     ]
  80
  81     _SUCCESS = 0
  82     _PASSWORD_NOT_VERIFIED = 1
  83     _PASSWORD_DETECTED = 2
  84     _VIDEO_NOT_FOUND = 3
  85
  86     def _search_meta(self, name, html, display_name=None):
  87         if display_name is None:
  88             display_name = name
  89         return self._html_search_regex(
  90             r'<meta itemprop="%s" content="([^"]+)" />' % re.escape(name),
  91             html, display_name, fatal=False)
  92
  93     def _real_extract(self, url):
  94         mobj = re.match(self._VALID_URL, url)
  95         video_id = mobj.group('videoid')
  96         real_video_id = mobj.group('realvideoid')
  97
  98         # Download video JSON data
  99         video_json_url = 'http://smotri.com/vt.php?id=%s' % real_video_id
 100         video_json_page = self._download_webpage(video_json_url, video_id, u'Downloading video JSON')
 101         video_json = json.loads(video_json_page)
 102
 103         status = video_json['status']
 104         if status == self._VIDEO_NOT_FOUND:
 105             raise ExtractorError(u'Video %s does not exist' % video_id, expected=True)
 106         elif status == self._PASSWORD_DETECTED: # The video is protected by a password, retry with
 107                                                 # video-password set
 108             video_password = self._downloader.params.get('videopassword', None)
 109             if not video_password:
 110                 raise ExtractorError(u'This video is protected by a password, use the --video-password option', expected=True)
 111             video_json_url += '&md5pass=%s' % hashlib.md5(video_password).hexdigest()
 112             video_json_page = self._download_webpage(video_json_url, video_id, u'Downloading video JSON (video-password set)')
 113             video_json = json.loads(video_json_page)
 114             status = video_json['status']
 115             if status == self._PASSWORD_NOT_VERIFIED:
 116                 raise ExtractorError(u'Video password is invalid', expected=True)
 117
 118         if status != self._SUCCESS:
 119             raise ExtractorError(u'Unexpected status value %s' % status)
 120
 121         # Extract the URL of the video
 122         video_url = video_json['file_data']
 123         video_ext = determine_ext(video_url)
 124
 125         # Video JSON does not provide enough meta data
 126         # We will extract some from the video web page instead
 127         video_page_url = 'http://' + mobj.group('url')
 128         video_page = self._download_webpage(video_page_url, video_id, u'Downloading video page')
 129
 130         # Adult content
 131         if re.search(u'EroConfirmText">', video_page) is not None:
 132             self.report_age_confirmation()
 133             confirm_string = self._html_search_regex(
 134                 ur'<a href="/video/view/\?id=%s&confirm=([^"]+)" title="[^"]+">' % video_id,
 135                 video_page, u'confirm string')
 136             confirm_url = video_page_url + '&confirm=%s' % confirm_string
 137             video_page = self._download_webpage(confirm_url, video_id, u'Downloading video page (age confirmed)')
 138             adult_content = True
 139         else:
 140             adult_content = False
 141
 142         # Extract the rest of meta data
 143         video_title = self._search_meta(u'name', video_page, u'title')
 144         if not video_title:
 145             video_title = video_url.rsplit('/', 1)[-1]
 146
 147         video_description = self._search_meta(u'description', video_page)
 148         video_thumbnail = self._search_meta(u'thumbnail', video_page)
 149
 150         upload_date_str = self._search_meta(u'uploadDate', video_page, u'upload date')
 151         upload_date_m = re.search(r'(?P<year>\d{4})\.(?P<month>\d{2})\.(?P<day>\d{2})T', upload_date_str)
 152         video_upload_date = (
 153             (
 154                 upload_date_m.group('year') +
 155                 upload_date_m.group('month') +
 156                 upload_date_m.group('day')
 157             )
 158             if upload_date_m else None
 159         )
 160
 161         duration_str = self._search_meta(u'duration', video_page)
 162         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)
 163         video_duration = (
 164             (
 165                 (int(duration_m.group('hours')) * 60 * 60) +
 166                 (int(duration_m.group('minutes')) * 60) +
 167                 int(duration_m.group('seconds'))
 168             )
 169             if duration_m else None
 170         )
 171
 172         video_uploader = self._html_search_regex(
 173             ur'<div class="DescrUser"><div>Автор.*?onmouseover="popup_user_info[^"]+">(.*?)</a>',
 174             video_page, u'uploader', fatal=False, flags=re.MULTILINE|re.DOTALL)
 175
 176         video_uploader_id = self._html_search_regex(
 177             ur'<div class="DescrUser"><div>Автор.*?onmouseover="popup_user_info\(.*?\'([^\']+)\'\);">',
 178             video_page, u'uploader id', fatal=False, flags=re.MULTILINE|re.DOTALL)
 179
 180         video_view_count = self._html_search_regex(
 181             ur'Общее количество просмотров.*?<span class="Number">(\d+)</span>',
 182             video_page, u'view count', fatal=False, flags=re.MULTILINE|re.DOTALL)
 183
 184         return {
 185             'id': video_id,
 186             'url': video_url,
 187             'title': video_title,
 188             'ext': video_ext,
 189             'thumbnail': video_thumbnail,
 190             'description': video_description,
 191             'uploader': video_uploader,
 192             'upload_date': video_upload_date,
 193             'uploader_id': video_uploader_id,
 194             'video_duration': video_duration,
 195             'view_count': video_view_count,
 196             'age_limit': 18 if adult_content else 0,
 197             'video_page_url': video_page_url
 198         }
 199
 200 class SmotriCommunityIE(InfoExtractor):
 201     IE_DESC = u'Smotri.com community videos'
 202     IE_NAME = u'smotri:community'
 203     _VALID_URL = r'^(?:http://)?(?:www\.)?smotri\.com/community/video/(?P<communityid>[0-9A-Za-z_\'-]+)'
 204
 205     def _real_extract(self, url):
 206         mobj = re.match(self._VALID_URL, url)
 207         community_id = mobj.group('communityid')
 208
 209         url = 'http://smotri.com/export/rss/video/by/community/-/%s/video.xml' % community_id
 210         rss = self._download_xml(url, community_id, u'Downloading community RSS')
 211
 212         entries = [self.url_result(video_url.text, 'Smotri')
 213                    for video_url in rss.findall('./channel/item/link')]
 214
 215         community_title = self._html_search_regex(
 216             ur'^Видео сообщества "([^"]+)"$', rss.find('./channel/description').text, u'community title')
 217
 218         return self.playlist_result(entries, community_id, community_title)
 219
 220 class SmotriUserIE(InfoExtractor):
 221     IE_DESC = u'Smotri.com user videos'
 222     IE_NAME = u'smotri:user'
 223     _VALID_URL = r'^(?:http://)?(?:www\.)?smotri\.com/user/(?P<userid>[0-9A-Za-z_\'-]+)'
 224
 225     def _real_extract(self, url):
 226         mobj = re.match(self._VALID_URL, url);
 227         user_id = mobj.group('userid')
 228
 229         url = 'http://smotri.com/export/rss/user/video/-/%s/video.xml' % user_id
 230         rss = self._download_xml(url, user_id, u'Downloading user RSS')
 231
 232         entries = [self.url_result(video_url.text, 'Smotri')
 233                    for video_url in rss.findall('./channel/item/link')]
 234
 235         user_nickname = self._html_search_regex(
 236             ur'^Видео режиссера (.*)$', rss.find('./channel/description').text, u'user nickname')
 237
 238         return self.playlist_result(entries, user_id, user_nickname)
 239