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

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5
   6 from .common import InfoExtractor
   7 from ..utils import (
   8     js_to_json,
   9     determine_ext,
  10 )
  11
  12
  13 class BpbIE(InfoExtractor):
  14     IE_DESC = 'Bundeszentrale für politische Bildung'
  15     _VALID_URL = r'https?://(?:www\.)?bpb\.de/mediathek/(?P<id>[0-9]+)/'
  16
  17     _TEST = {
  18         'url': 'http://www.bpb.de/mediathek/297/joachim-gauck-zu-1989-und-die-erinnerung-an-die-ddr',
  19         # md5 fails in Python 2.6 due to buggy server response and wrong handling of urllib2
  20         'md5': 'c4f84c8a8044ca9ff68bb8441d300b3f',
  21         'info_dict': {
  22             'id': '297',
  23             'ext': 'mp4',
  24             'title': 'Joachim Gauck zu 1989 und die Erinnerung an die DDR',
  25             'description': 'Joachim Gauck, erster Beauftragter für die Stasi-Unterlagen, spricht auf dem Geschichtsforum über die friedliche Revolution 1989 und eine "gewisse Traurigkeit" im Umgang mit der DDR-Vergangenheit.'
  26         }
  27     }
  28
  29     def _real_extract(self, url):
  30         video_id = self._match_id(url)
  31         webpage = self._download_webpage(url, video_id)
  32
  33         title = self._html_search_regex(
  34             r'<h2 class="white">(.*?)</h2>', webpage, 'title')
  35         video_info_dicts = re.findall(
  36             r"({\s*src:\s*'http://film\.bpb\.de/[^}]+})", webpage)
  37
  38         formats = []
  39         for video_info in video_info_dicts:
  40             video_info = self._parse_json(video_info, video_id, transform_source=js_to_json)
  41             quality = video_info['quality']
  42             video_url = video_info['src']
  43             formats.append({
  44                 'url': video_url,
  45                 'preference': 10 if quality == 'high' else 0,
  46                 'format_note': quality,
  47                 'format_id': '%s-%s' % (quality, determine_ext(video_url)),
  48             })
  49
  50         self._sort_formats(formats)
  51
  52         return {
  53             'id': video_id,
  54             'formats': formats,
  55             'title': title,
  56             'description': self._og_search_description(webpage),
  57         }