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

   1 from __future__ import unicode_literals
   2
   3 import json
   4 import re
   5
   6 from .common import InfoExtractor
   7 from ..utils import (
   8     strip_jsonp,
   9     parse_duration,
  10     unified_strdate,
  11 )
  12
  13
  14 class MusicVaultIE(InfoExtractor):
  15     _VALID_URL = r'https?://www\.musicvault\.com/(?P<uploader_id>[^/?#]*)/video/(?P<display_id>[^/?#]*)_(?P<id>[0-9]+)\.html'
  16     _TEST = {
  17         'url': 'http://www.musicvault.com/the-allman-brothers-band/video/straight-from-the-heart_1010863.html',
  18         'md5': '2cdbb3ae75f7fb3519821507d2fb3c15',
  19         'info_dict': {
  20             'id': '1010863',
  21             'ext': 'mp4',
  22             'uploader_id': 'the-allman-brothers-band',
  23             'title': 'Straight from the Heart',
  24             'duration': 244,
  25             'uploader': 'The Allman Brothers Band',
  26             'thumbnail': 're:^https?://.*/thumbnail/.*',
  27             'upload_date': '19811216',
  28             'location': 'Capitol Theatre (Passaic, NJ)',
  29             'description': 'Listen to The Allman Brothers Band perform Straight from the Heart at Capitol Theatre (Passaic, NJ) on Dec 16, 1981',
  30         }
  31     }
  32
  33     def _real_extract(self, url):
  34         mobj = re.match(self._VALID_URL, url)
  35         display_id = mobj.group('display_id')
  36         webpage = self._download_webpage(url, display_id)
  37
  38         thumbnail = self._search_regex(
  39             r'<meta itemprop="thumbnail" content="([^"]+)"',
  40             webpage, 'thumbnail', fatal=False)
  41
  42         data_div = self._search_regex(
  43             r'(?s)<div class="data">(.*?)</div>', webpage, 'data fields')
  44         uploader = self._html_search_regex(
  45             r'<h1.*?>(.*?)</h1>', data_div, 'uploader', fatal=False)
  46         title = self._html_search_regex(
  47             r'<h2.*?>(.*?)</h2>', data_div, 'title')
  48         upload_date = unified_strdate(self._html_search_regex(
  49             r'<h3.*?>(.*?)</h3>', data_div, 'uploader', fatal=False))
  50         location = self._html_search_regex(
  51             r'<h4.*?>(.*?)</h4>', data_div, 'location', fatal=False)
  52
  53         duration = parse_duration(self._html_search_meta('duration', webpage))
  54
  55         VIDEO_URL_TEMPLATE = 'http://cdnapi.kaltura.com/p/%(uid)s/sp/%(wid)s/playManifest/entryId/%(entry_id)s/format/url/protocol/http'
  56         kaltura_id = self._search_regex(
  57             r'<div id="video-detail-player" data-kaltura-id="([^"]+)"',
  58             webpage, 'kaltura ID')
  59         video_url = VIDEO_URL_TEMPLATE % {
  60             'entry_id': kaltura_id,
  61             'wid': self._search_regex(r'/wid/_([0-9]+)/', webpage, 'wid'),
  62             'uid': self._search_regex(r'uiconf_id/([0-9]+)/', webpage, 'uid'),
  63         }
  64
  65         return {
  66             'id': mobj.group('id'),
  67             'url': video_url,
  68             'ext': 'mp4',
  69             'display_id': display_id,
  70             'uploader_id': mobj.group('uploader_id'),
  71             'thumbnail': thumbnail,
  72             'description': self._html_search_meta('description', webpage),
  73             'upload_date': upload_date,
  74             'location': location,
  75             'title': title,
  76             'uploader': uploader,
  77             'duration': duration,
  78         }