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

   1 import json
   2 import re
   3 import time
   4
   5 from .common import InfoExtractor
   6 from ..utils import month_by_name
   7
   8
   9 class NDTVIE(InfoExtractor):
  10     _VALID_URL = r'^https?://(?:www\.)?ndtv\.com/video/player/[^/]*/[^/]*/(?P<id>[a-z0-9]+)'
  11
  12     _TEST = {
  13         u"url": u"http://www.ndtv.com/video/player/news/ndtv-exclusive-don-t-need-character-certificate-from-rahul-gandhi-says-arvind-kejriwal/300710",
  14         u"file": u"300710.mp4",
  15         u"md5": u"39f992dbe5fb531c395d8bbedb1e5e88",
  16         u"info_dict": {
  17             u"title": u"NDTV exclusive: Don't need character certificate from Rahul Gandhi, says Arvind Kejriwal",
  18             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.",
  19             u"upload_date": u"20131208",
  20             u"duration": 1327,
  21             u"thumbnail": u"http://i.ndtvimg.com/video/images/vod/medium/2013-12/big_300710_1386518307.jpg",
  22         },
  23     }
  24
  25     def _real_extract(self, url):
  26         mobj = re.match(self._VALID_URL, url)
  27         video_id = mobj.group('id')
  28
  29         webpage = self._download_webpage(url, video_id)
  30
  31         filename = self._search_regex(
  32             r"__filename='([^']+)'", webpage, u'video filename')
  33         video_url = (u'http://bitcast-b.bitgravity.com/ndtvod/23372/ndtv/%s' %
  34                      filename)
  35
  36         duration_str = filename = self._search_regex(
  37             r"__duration='([^']+)'", webpage, u'duration', fatal=False)
  38         duration = None if duration_str is None else int(duration_str)
  39
  40         date_m = re.search(r'''(?x)
  41             <p\s+class="vod_dateline">\s*
  42                 Published\s+On:\s*
  43                 (?P<monthname>[A-Za-z]+)\s+(?P<day>[0-9]+),\s*(?P<year>[0-9]+)
  44             ''', webpage)
  45         upload_date = None
  46         assert date_m
  47         if date_m is not None:
  48             month = month_by_name(date_m.group('monthname'))
  49             if month is not None:
  50                 upload_date = '%s%02d%02d' % (
  51                     date_m.group('year'), month, int(date_m.group('day')))
  52
  53         description = self._og_search_description(webpage)
  54         READ_MORE = u' (Read more)'
  55         if description.endswith(READ_MORE):
  56             description = description[:-len(READ_MORE)]
  57
  58         return {
  59             'id': video_id,
  60             'url': video_url,
  61             'title': self._og_search_title(webpage),
  62             'description': description,
  63             'thumbnail': self._og_search_thumbnail(webpage),
  64             'duration': duration,
  65             'upload_date': upload_date,
  66         }