youtube_dl/extractor/thechive.py

   1 from __future__ import unicode_literals
   2
   3 import re
   4
   5 from .common import InfoExtractor
   6 from ..utils import unified_strdate
   7
   8
   9
  10 class TheChiveIE(InfoExtractor):
  11     _VALID_URL = r'http://(www\.)?thechive\.com/[^/]+/[^/]+/[^/]+/(?P<video_id>[A-Za-z\-]+)'
  12     _TEST = {
  13         'url': "http://thechive.com/2015/02/20/so-thats-what-a-set-of-redneck-bagpipes-sound-like-video/",
  14         'md5': "366710dda77cfa727bdef3523ba8466f",
  15         'info_dict': {
  16             'id': "so-thats-what-a-set-of-redneck-bagpipes-sound-like-video",
  17             'title': "So that's what a set of redneck bagpipes sound like... (Video)",
  18             'description': "Okay that was pretty good. Now play Freebird!...",
  19             'thumbnail': "https://thechive.files.wordpress.com/2015/02/0_07dghz0w-thumbnail2.jpg",
  20             'author': "Ben",
  21             'upload_date': "20150220",
  22             'ext': "mp4"
  23         }
  24     }
  25
  26     def _real_extract(self, url):
  27         mobj = re.match(self._VALID_URL, url)
  28         video_id = mobj.group('video_id')
  29         webpage = self._download_webpage(url, video_id)
  30
  31         title = self._og_search_title(webpage)
  32         description = self._html_search_regex(r'(?s)<meta name="description" content="(.*?)" />', webpage, 'description')
  33         thumbnail = self._og_search_thumbnail(webpage)
  34         author = self._html_search_regex(
  35             r'(?s)itemprop="author">(.+?)</span>', webpage, 'author', fatal=False).capitalize()
  36         upload_date = unified_strdate(self._html_search_regex(
  37             r'(?s)itemprop="datePublished" datetime="(.+?)">', webpage, 'upload_date', fatal=False))
  38
  39         # Adapted from extractor/musicvault.py
  40         VIDEO_URL_TEMPLATE = 'http://cdnapi.kaltura.com/p/%(uid)s/sp/%(wid)s/playManifest/entryId/%(entry_id)s/format/url/protocol/http'
  41
  42         kaltura_id = self._search_regex(
  43             r'entry_id=([^"]+)',
  44             webpage, 'kaltura ID')
  45         video_url = VIDEO_URL_TEMPLATE % {
  46             'entry_id': kaltura_id,
  47             'wid': self._search_regex(r'partner_id/([0-9]+)\?', webpage, 'wid'),
  48             'uid': self._search_regex(r'uiconf_id/([0-9]+)/', webpage, 'uid'),
  49         }
  50
  51         return {
  52             'url': video_url,
  53             'id': video_id,
  54             'title': title,
  55             'description': description,
  56             'thumbnail': thumbnail,
  57             'author': author,
  58             'upload_date': upload_date,
  59             'ext': 'mp4'
  60         }