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\.|secure\.)?flickr\.com/photos/(?P<uploader_id>[\w\-_@]+)/(?P<id>\d+).*'
  13     _TEST = {
  14         u'url': u'http://www.flickr.com/photos/forestwander-nature-pictures/5645318632/in/photostream/',
  15         u'file': u'5645318632.mp4',
  16         u'md5': u'6fdc01adbc89d72fc9c4f15b4a4ba87b',
  17         u'info_dict': {
  18             u"description": u"Waterfalls in the Springtime at Dark Hollow Waterfalls. These are located just off of Skyline Drive in Virginia. They are only about 6/10 of a mile hike but it is a pretty steep hill and a good climb back up.",
  19             u"uploader_id": u"forestwander-nature-pictures",
  20             u"title": u"Dark Hollow Waterfalls"
  21         }
  22     }
  23
  24     def _real_extract(self, url):
  25         mobj = re.match(self._VALID_URL, url)
  26
  27         video_id = mobj.group('id')
  28         video_uploader_id = mobj.group('uploader_id')
  29         webpage_url = 'http://www.flickr.com/photos/' + video_uploader_id + '/' + video_id
  30         webpage = self._download_webpage(webpage_url, video_id)
  31
  32         secret = self._search_regex(r"photo_secret: '(\w+)'", webpage, u'secret')
  33
  34         first_url = 'https://secure.flickr.com/apps/video/video_mtl_xml.gne?v=x&photo_id=' + video_id + '&secret=' + secret + '&bitrate=700&target=_self'
  35         first_xml = self._download_webpage(first_url, video_id, 'Downloading first data webpage')
  36
  37         node_id = self._html_search_regex(r'<Item id="id">(\d+-\d+)</Item>',
  38             first_xml, u'node_id')
  39
  40         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'
  41         second_xml = self._download_webpage(second_url, video_id, 'Downloading second data webpage')
  42
  43         self.report_extraction(video_id)
  44
  45         mobj = re.search(r'<STREAM APP="(.+?)" FULLPATH="(.+?)"', second_xml)
  46         if mobj is None:
  47             raise ExtractorError(u'Unable to extract video url')
  48         video_url = mobj.group(1) + unescapeHTML(mobj.group(2))
  49
  50         return [{
  51             'id':          video_id,
  52             'url':         video_url,
  53             'ext':         'mp4',
  54             'title':       self._og_search_title(webpage),
  55             'description': self._og_search_description(webpage),
  56             'thumbnail':   self._og_search_thumbnail(webpage),
  57             'uploader_id': video_uploader_id,
  58         }]