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

   1 import os.path
   2 import xml.etree.ElementTree
   3
   4 from .common import InfoExtractor
   5 from ..utils import (
   6     compat_urllib_parse_urlparse,
   7
   8     ExtractorError,
   9 )
  10
  11
  12 class MySpassIE(InfoExtractor):
  13     _VALID_URL = r'http://www.myspass.de/.*'
  14
  15     def _real_extract(self, url):
  16         META_DATA_URL_TEMPLATE = 'http://www.myspass.de/myspass/includes/apps/video/getvideometadataxml.php?id=%s'
  17
  18         # video id is the last path element of the URL
  19         # usually there is a trailing slash, so also try the second but last
  20         url_path = compat_urllib_parse_urlparse(url).path
  21         url_parent_path, video_id = os.path.split(url_path)
  22         if not video_id:
  23             _, video_id = os.path.split(url_parent_path)
  24
  25         # get metadata
  26         metadata_url = META_DATA_URL_TEMPLATE % video_id
  27         metadata_text = self._download_webpage(metadata_url, video_id)
  28         metadata = xml.etree.ElementTree.fromstring(metadata_text.encode('utf-8'))
  29
  30         # extract values from metadata
  31         url_flv_el = metadata.find('url_flv')
  32         if url_flv_el is None:
  33             raise ExtractorError(u'Unable to extract download url')
  34         video_url = url_flv_el.text
  35         extension = os.path.splitext(video_url)[1][1:]
  36         title_el = metadata.find('title')
  37         if title_el is None:
  38             raise ExtractorError(u'Unable to extract title')
  39         title = title_el.text
  40         format_id_el = metadata.find('format_id')
  41         if format_id_el is None:
  42             format = 'mp4'
  43         else:
  44             format = format_id_el.text
  45         description_el = metadata.find('description')
  46         if description_el is not None:
  47             description = description_el.text
  48         else:
  49             description = None
  50         imagePreview_el = metadata.find('imagePreview')
  51         if imagePreview_el is not None:
  52             thumbnail = imagePreview_el.text
  53         else:
  54             thumbnail = None
  55         info = {
  56             'id': video_id,
  57             'url': video_url,
  58             'title': title,
  59             'ext': extension,
  60             'format': format,
  61             'thumbnail': thumbnail,
  62             'description': description
  63         }
  64         return [info]