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         # with null videoTitle
  37         'url': 'https://view.vzaar.com/20313539/download',
  38         'only_matching': True,
  39     }]
  40
  41     @staticmethod
  42     def _extract_urls(webpage):
  43         return re.findall(
  44             r'<iframe[^>]+src=["\']((?:https?:)?//(?:view\.vzaar\.com)/[0-9]+)',
  45             webpage)
  46
  47     def _real_extract(self, url):
  48         video_id = self._match_id(url)
  49         video_data = self._download_json(
  50             'http://view.vzaar.com/v2/%s/video' % video_id, video_id)
  51
  52         title = video_data.get('videoTitle') or video_id
  53
  54         formats = []
  55
  56         source_url = url_or_none(video_data.get('sourceUrl'))
  57         if source_url:
  58             f = {
  59                 'url': source_url,
  60                 'format_id': 'http',
  61             }
  62             if 'audio' in source_url:
  63                 f.update({
  64                     'vcodec': 'none',
  65                     'ext': 'mp3',
  66                 })
  67             else:
  68                 f.update({
  69                     'width': int_or_none(video_data.get('width')),
  70                     'height': int_or_none(video_data.get('height')),
  71                     'ext': 'mp4',
  72                     'fps': float_or_none(video_data.get('fps')),
  73                 })
  74             formats.append(f)
  75
  76         video_guid = video_data.get('guid')
  77         usp = video_data.get('usp')
  78         if isinstance(video_guid, compat_str) and isinstance(usp, dict):
  79             m3u8_url = ('http://fable.vzaar.com/v4/usp/%s/%s.ism/.m3u8?'
  80                         % (video_guid, video_id)) + '&'.join(
  81                 '%s=%s' % (k, v) for k, v in usp.items())
  82             formats.extend(self._extract_m3u8_formats(
  83                 m3u8_url, video_id, 'mp4', entry_protocol='m3u8_native',
  84                 m3u8_id='hls', fatal=False))
  85
  86         self._sort_formats(formats)
  87
  88         return {
  89             'id': video_id,
  90             'title': title,
  91             'thumbnail': self._proto_relative_url(video_data.get('poster')),
  92             'duration': float_or_none(video_data.get('videoDuration')),
  93             'timestamp': unified_timestamp(video_data.get('ts')),
  94             'formats': formats,
  95         }