_ Git - youtube-dl/blob - youtube_dl/extractor/flickr.py

   1 import re
   2
   3 from .common import InfoExtractor
   4 from ..utils import (
   5     ExtractorError,
   6     unescapeHTML,
   7 )
   8
   9
  10 class FlickrIE(InfoExtractor):
  11     """Information Extractor for Flickr videos"""
  12     _VALID_URL = r'(?:https?://)?(?:www\.)?flickr\.com/photos/(?P<uploader_id>[\w\-_@]+)/(?P<id>\d+).*'
  13
  14     def _real_extract(self, url):
  15         mobj = re.match(self._VALID_URL, url)
  16
  17         video_id = mobj.group('id')
  18         video_uploader_id = mobj.group('uploader_id')
  19         webpage_url = 'http://www.flickr.com/photos/' + video_uploader_id + '/' + video_id
  20         webpage = self._download_webpage(webpage_url, video_id)
  21
  22         secret = self._search_regex(r"photo_secret: '(\w+)'", webpage, u'secret')
  23
  24         first_url = 'https://secure.flickr.com/apps/video/video_mtl_xml.gne?v=x&photo_id=' + video_id + '&secret=' + secret + '&bitrate=700&target=_self'
  25         first_xml = self._download_webpage(first_url, video_id, 'Downloading first data webpage')
  26
  27         node_id = self._html_search_regex(r'<Item id="id">(\d+-\d+)</Item>',
  28             first_xml, u'node_id')
  29
  30         second_url = 'https://secure.flickr.com/video_playlist.gne?node_id=' + node_id + '&tech=flash&mode=playlist&bitrate=700&secret=' + secret + '&rd=video.yahoo.com&noad=1'
  31         second_xml = self._download_webpage(second_url, video_id, 'Downloading second data webpage')
  32
  33         self.report_extraction(video_id)
  34
  35         mobj = re.search(r'<STREAM APP="(.+?)" FULLPATH="(.+?)"', second_xml)
  36         if mobj is None:
  37             raise ExtractorError(u'Unable to extract video url')
  38         video_url = mobj.group(1) + unescapeHTML(mobj.group(2))
  39
  40         video_title = self._html_search_regex(r'<meta property="og:title" content=(?:"([^"]+)"|\'([^\']+)\')',
  41             webpage, u'video title')
  42
  43         video_description = self._html_search_regex(r'<meta property="og:description" content=(?:"([^"]+)"|\'([^\']+)\')',
  44             webpage, u'description', fatal=False)
  45
  46         thumbnail = self._html_search_regex(r'<meta property="og:image" content=(?:"([^"]+)"|\'([^\']+)\')',
  47             webpage, u'thumbnail', fatal=False)
  48
  49         return [{
  50             'id':          video_id,
  51             'url':         video_url,
  52             'ext':         'mp4',
  53             'title':       video_title,
  54             'description': video_description,
  55             'thumbnail':   thumbnail,
  56             'uploader_id': video_uploader_id,
  57         }]