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

   1 import re
   2
   3 from .common import InfoExtractor
   4 from ..utils import month_by_name
   5
   6
   7 class NDTVIE(InfoExtractor):
   8     _VALID_URL = r'^https?://(?:www\.)?ndtv\.com/video/player/[^/]*/[^/]*/(?P<id>[a-z0-9]+)'
   9
  10     _TEST = {
  11         u"url": u"http://www.ndtv.com/video/player/news/ndtv-exclusive-don-t-need-character-certificate-from-rahul-gandhi-says-arvind-kejriwal/300710",
  12         u"file": u"300710.mp4",
  13         u"md5": u"39f992dbe5fb531c395d8bbedb1e5e88",
  14         u"info_dict": {
  15             u"title": u"NDTV exclusive: Don't need character certificate from Rahul Gandhi, says Arvind Kejriwal",
  16             u"description": u"In an exclusive interview to NDTV, Aam Aadmi Party's Arvind Kejriwal says it makes no difference to him that Rahul Gandhi said the Congress needs to learn from his party.",
  17             u"upload_date": u"20131208",
  18             u"duration": 1327,
  19             u"thumbnail": u"http://i.ndtvimg.com/video/images/vod/medium/2013-12/big_300710_1386518307.jpg",
  20         },
  21     }
  22
  23     def _real_extract(self, url):
  24         mobj = re.match(self._VALID_URL, url)
  25         video_id = mobj.group('id')
  26
  27         webpage = self._download_webpage(url, video_id)
  28
  29         filename = self._search_regex(
  30             r"__filename='([^']+)'", webpage, u'video filename')
  31         video_url = (u'http://bitcast-b.bitgravity.com/ndtvod/23372/ndtv/%s' %
  32                      filename)
  33
  34         duration_str = filename = self._search_regex(
  35             r"__duration='([^']+)'", webpage, u'duration', fatal=False)
  36         duration = None if duration_str is None else int(duration_str)
  37
  38         date_m = re.search(r'''(?x)
  39             <p\s+class="vod_dateline">\s*
  40                 Published\s+On:\s*
  41                 (?P<monthname>[A-Za-z]+)\s+(?P<day>[0-9]+),\s*(?P<year>[0-9]+)
  42             ''', webpage)
  43         upload_date = None
  44         assert date_m
  45         if date_m is not None:
  46             month = month_by_name(date_m.group('monthname'))
  47             if month is not None:
  48                 upload_date = '%s%02d%02d' % (
  49                     date_m.group('year'), month, int(date_m.group('day')))
  50
  51         description = self._og_search_description(webpage)
  52         READ_MORE = u' (Read more)'
  53         if description.endswith(READ_MORE):
  54             description = description[:-len(READ_MORE)]
  55
  56         return {
  57             'id': video_id,
  58             'url': video_url,
  59             'title': self._og_search_title(webpage),
  60             'description': description,
  61             'thumbnail': self._og_search_thumbnail(webpage),
  62             'duration': duration,
  63             'upload_date': upload_date,
  64         }