_ Git - youtube-dl/blob - youtube_dl/extractor/imdb.py

   1 import re
   2 import json
   3
   4 from .common import InfoExtractor
   5 from ..utils import (
   6     compat_urlparse,
   7     get_element_by_attribute,
   8 )
   9
  10
  11 class ImdbIE(InfoExtractor):
  12     IE_NAME = u'imdb'
  13     IE_DESC = u'Internet Movie Database trailers'
  14     _VALID_URL = r'http://www\.imdb\.com/video/imdb/vi(?P<id>\d+)'
  15
  16     _TEST = {
  17         u'url': u'http://www.imdb.com/video/imdb/vi2524815897',
  18         u'md5': u'9f34fa777ade3a6e57a054fdbcb3a068',
  19         u'info_dict': {
  20             u'id': u'2524815897',
  21             u'ext': u'mp4',
  22             u'title': u'Ice Age: Continental Drift Trailer (No. 2) - IMDb',
  23             u'description': u'md5:9061c2219254e5d14e03c25c98e96a81',
  24             u'duration': 151,
  25         }
  26     }
  27
  28     def _real_extract(self, url):
  29         mobj = re.match(self._VALID_URL, url)
  30         video_id = mobj.group('id')
  31         webpage = self._download_webpage(url,video_id)
  32         descr = get_element_by_attribute('itemprop', 'description', webpage)
  33         available_formats = re.findall(
  34             r'case \'(?P<f_id>.*?)\' :$\s+url = \'(?P<path>.*?)\'', webpage,
  35             flags=re.MULTILINE)
  36         formats = []
  37         for f_id, f_path in available_formats:
  38             format_page = self._download_webpage(
  39                 compat_urlparse.urljoin(url, f_path),
  40                 u'Downloading info for %s format' % f_id)
  41             json_data = self._search_regex(
  42                 r'<script[^>]+class="imdb-player-data"[^>]*?>(.*?)</script>',
  43                 format_page, u'json data', flags=re.DOTALL)
  44             info = json.loads(json_data)
  45             format_info = info['videoPlayerObject']['video']
  46             formats.append({
  47                 'format_id': f_id,
  48                 'url': format_info['url'],
  49                 'height': format_info['height'],
  50                 'width': format_info['width'],
  51             })
  52
  53         return {
  54             'id': video_id,
  55             'title': self._og_search_title(webpage),
  56             'formats': formats,
  57             'description': descr,
  58             'thumbnail': format_info['slate'],
  59             'duration': int(info['titleObject']['title']['duration_seconds']),
  60         }