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

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5
   6 from .common import InfoExtractor
   7 from ..compat import compat_str
   8 from ..utils import (
   9     int_or_none,
  10     float_or_none,
  11     unified_timestamp,
  12     url_or_none,
  13 )
  14
  15
  16 class VzaarIE(InfoExtractor):
  17     _VALID_URL = r'https?://(?:(?:www|view)\.)?vzaar\.com/(?:videos/)?(?P<id>\d+)'
  18     _TESTS = [{
  19         # HTTP and HLS
  20         'url': 'https://vzaar.com/videos/1152805',
  21         'md5': 'bde5ddfeb104a6c56a93a06b04901dbf',
  22         'info_dict': {
  23             'id': '1152805',
  24             'ext': 'mp4',
  25             'title': 'sample video (public)',
  26         },
  27     }, {
  28         'url': 'https://view.vzaar.com/27272/player',
  29         'md5': '3b50012ac9bbce7f445550d54e0508f2',
  30         'info_dict': {
  31             'id': '27272',
  32             'ext': 'mp3',
  33             'title': 'MP3',
  34         },
  35     }, {
  36         # hlsAes = true
  37         'url': 'https://view.vzaar.com/11379930/player',
  38         'info_dict': {
  39             'id': '11379930',
  40             'ext': 'mp4',
  41             'title': 'Videoaula',
  42         },
  43         'params': {
  44             # m3u8 download
  45             'skip_download': True,
  46         },
  47     }, {
  48         # with null videoTitle
  49         'url': 'https://view.vzaar.com/20313539/download',
  50         'only_matching': True,
  51     }]
  52
  53     @staticmethod
  54     def _extract_urls(webpage):
  55         return re.findall(
  56             r'<iframe[^>]+src=["\']((?:https?:)?//(?:view\.vzaar\.com)/[0-9]+)',
  57             webpage)
  58
  59     def _real_extract(self, url):
  60         video_id = self._match_id(url)
  61         video_data = self._download_json(
  62             'http://view.vzaar.com/v2/%s/video' % video_id, video_id)
  63
  64         title = video_data.get('videoTitle') or video_id
  65
  66         formats = []
  67
  68         source_url = url_or_none(video_data.get('sourceUrl'))
  69         if source_url:
  70             f = {
  71                 'url': source_url,
  72                 'format_id': 'http',
  73                 'preference': 1,
  74             }
  75             if 'audio' in source_url:
  76                 f.update({
  77                     'vcodec': 'none',
  78                     'ext': 'mp3',
  79                 })
  80             else:
  81                 f.update({
  82                     'width': int_or_none(video_data.get('width')),
  83                     'height': int_or_none(video_data.get('height')),
  84                     'ext': 'mp4',
  85                     'fps': float_or_none(video_data.get('fps')),
  86                 })
  87             formats.append(f)
  88
  89         video_guid = video_data.get('guid')
  90         usp = video_data.get('usp')
  91         if video_data.get('uspEnabled') and isinstance(video_guid, compat_str) and isinstance(usp, dict):
  92             hls_aes = video_data.get('hlsAes')
  93             qs = '&'.join('%s=%s' % (k, v) for k, v in usp.items())
  94             url_templ = 'http://%%s.vzaar.com/v5/usp%s/%s/%s.ism%%s?' % ('aes' if hls_aes else '', video_guid, video_id)
  95             m3u8_formats = self._extract_m3u8_formats(
  96                 url_templ % ('fable', '/.m3u8') + qs, video_id, 'mp4', 'm3u8_native',
  97                 m3u8_id='hls', fatal=False)
  98             if hls_aes:
  99                 for f in m3u8_formats:
 100                     f['_decryption_key_url'] = url_templ % ('goose', '') + qs
 101             formats.extend(m3u8_formats)
 102
 103         self._sort_formats(formats)
 104
 105         return {
 106             'id': video_id,
 107             'title': title,
 108             'thumbnail': self._proto_relative_url(video_data.get('poster')),
 109             'duration': float_or_none(video_data.get('videoDuration')),
 110             'timestamp': unified_timestamp(video_data.get('ts')),
 111             'formats': formats,
 112         }