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