youtube_dl/extractor/flickr.py

   1 from __future__ import unicode_literals
   2
   3 import re
   4
   5 from .common import InfoExtractor
   6 from ..utils import (
   7     ExtractorError,
   8     unescapeHTML,
   9 )
  10
  11
  12 class FlickrIE(InfoExtractor):
  13     _VALID_URL = r'https?://(?:www\.|secure\.)?flickr\.com/photos/(?P<uploader_id>[\w\-_@]+)/(?P<id>\d+).*'
  14     _TEST = {
  15         'url': 'http://www.flickr.com/photos/forestwander-nature-pictures/5645318632/in/photostream/',
  16         'md5': '6fdc01adbc89d72fc9c4f15b4a4ba87b',
  17         'info_dict': {
  18             'id': '5645318632',
  19             'ext': 'mp4',
  20             "description": "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.",
  21             "uploader_id": "forestwander-nature-pictures",
  22             "title": "Dark Hollow Waterfalls"
  23         }
  24     }
  25
  26     def _real_extract(self, url):
  27         mobj = re.match(self._VALID_URL, url)
  28
  29         video_id = mobj.group('id')
  30         video_uploader_id = mobj.group('uploader_id')
  31         webpage_url = 'http://www.flickr.com/photos/' + video_uploader_id + '/' + video_id
  32         webpage = self._download_webpage(webpage_url, video_id)
  33
  34         secret = self._search_regex(r"photo_secret: '(\w+)'", webpage, 'secret')
  35
  36         first_url = 'https://secure.flickr.com/apps/video/video_mtl_xml.gne?v=x&photo_id=' + video_id + '&secret=' + secret + '&bitrate=700&target=_self'
  37         first_xml = self._download_webpage(first_url, video_id, 'Downloading first data webpage')
  38
  39         node_id = self._html_search_regex(r'<Item id="id">(\d+-\d+)</Item>',
  40             first_xml, 'node_id')
  41
  42         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'
  43         second_xml = self._download_webpage(second_url, video_id, 'Downloading second data webpage')
  44
  45         self.report_extraction(video_id)
  46
  47         mobj = re.search(r'<STREAM APP="(.+?)" FULLPATH="(.+?)"', second_xml)
  48         if mobj is None:
  49             raise ExtractorError('Unable to extract video url')
  50         video_url = mobj.group(1) + unescapeHTML(mobj.group(2))
  51
  52         return {
  53             'id': video_id,
  54             'url': video_url,
  55             'ext': 'mp4',
  56             'title': self._og_search_title(webpage),
  57             'description': self._og_search_description(webpage),
  58             'thumbnail': self._og_search_thumbnail(webpage),
  59             'uploader_id': video_uploader_id,
  60         }