Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/flickr.py

   1 from __future__ import unicode_literals
   2
   3 import re
   4
   5 from .common import InfoExtractor
   6 from ..compat import compat_urllib_request
   7 from ..utils import (
   8     ExtractorError,
   9     find_xpath_attr,
  10 )
  11
  12
  13 class FlickrIE(InfoExtractor):
  14     _VALID_URL = r'https?://(?:www\.|secure\.)?flickr\.com/photos/(?P<uploader_id>[\w\-_@]+)/(?P<id>\d+).*'
  15     _TEST = {
  16         'url': 'http://www.flickr.com/photos/forestwander-nature-pictures/5645318632/in/photostream/',
  17         'md5': '6fdc01adbc89d72fc9c4f15b4a4ba87b',
  18         'info_dict': {
  19             'id': '5645318632',
  20             'ext': 'mp4',
  21             "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.",
  22             "uploader_id": "forestwander-nature-pictures",
  23             "title": "Dark Hollow Waterfalls"
  24         }
  25     }
  26
  27     def _real_extract(self, url):
  28         mobj = re.match(self._VALID_URL, url)
  29
  30         video_id = mobj.group('id')
  31         video_uploader_id = mobj.group('uploader_id')
  32         webpage_url = 'http://www.flickr.com/photos/' + video_uploader_id + '/' + video_id
  33         req = compat_urllib_request.Request(webpage_url)
  34         req.add_header(
  35             'User-Agent',
  36             # it needs a more recent version
  37             'Mozilla/5.0 (X11; Linux x86_64; rv:38.0) Gecko/20150101 Firefox/38.0 (Chrome)')
  38         webpage = self._download_webpage(req, video_id)
  39
  40         secret = self._search_regex(r'secret"\s*:\s*"(\w+)"', webpage, 'secret')
  41
  42         first_url = 'https://secure.flickr.com/apps/video/video_mtl_xml.gne?v=x&photo_id=' + video_id + '&secret=' + secret + '&bitrate=700&target=_self'
  43         first_xml = self._download_xml(first_url, video_id, 'Downloading first data webpage')
  44
  45         node_id = find_xpath_attr(
  46             first_xml, './/{http://video.yahoo.com/YEP/1.0/}Item', 'id',
  47             'id').text
  48
  49         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'
  50         second_xml = self._download_xml(second_url, video_id, 'Downloading second data webpage')
  51
  52         self.report_extraction(video_id)
  53
  54         stream = second_xml.find('.//STREAM')
  55         if stream is None:
  56             raise ExtractorError('Unable to extract video url')
  57         video_url = stream.attrib['APP'] + stream.attrib['FULLPATH']
  58
  59         return {
  60             'id': video_id,
  61             'url': video_url,
  62             'ext': 'mp4',
  63             'title': self._og_search_title(webpage),
  64             'description': self._og_search_description(webpage),
  65             'thumbnail': self._og_search_thumbnail(webpage),
  66             'uploader_id': video_uploader_id,
  67         }