youtube_dl/extractor/nuvid.py

   1 from __future__ import unicode_literals
   2
   3 import re
   4
   5 from .common import InfoExtractor
   6 from ..utils import (
   7     parse_duration,
   8 )
   9
  10
  11 class NuvidIE(InfoExtractor):
  12     _VALID_URL = r'https?://(?:www|m)\.nuvid\.com/video/(?P<id>[0-9]+)'
  13     _TEST = {
  14         'url': 'http://m.nuvid.com/video/1310741/',
  15         'md5': 'eab207b7ac4fccfb4e23c86201f11277',
  16         'info_dict': {
  17             'id': '1310741',
  18             'ext': 'mp4',
  19             'title': 'Horny babes show their awesome bodeis and',
  20             'duration': 129,
  21             'age_limit': 18,
  22         }
  23     }
  24
  25     def _real_extract(self, url):
  26         video_id = self._match_id(url)
  27
  28         page_url = 'http://m.nuvid.com/video/%s' % video_id
  29         webpage = self._download_webpage(
  30             page_url, video_id, 'Downloading video page')
  31         # When dwnld_speed exists and has a value larger than the MP4 file's
  32         # bitrate, Nuvid returns the MP4 URL
  33         # It's unit is 100bytes/millisecond, see mobile-nuvid-min.js for the algorithm
  34         self._set_cookie('nuvid.com', 'dwnld_speed', '10.0')
  35         mp4_webpage = self._download_webpage(
  36             page_url, video_id, 'Downloading video page for MP4 format')
  37
  38         html5_video_re = r'(?s)<(?:video|audio)[^<]*(?:>.*?<source[^>]*)?\s+src=["\'](.*?)["\']',
  39         video_url = self._html_search_regex(html5_video_re, webpage, video_id)
  40         mp4_video_url = self._html_search_regex(html5_video_re, mp4_webpage, video_id)
  41         formats = [{
  42             'url': video_url,
  43         }]
  44         if mp4_video_url != video_url:
  45             formats.append({
  46                 'url': mp4_video_url,
  47             })
  48
  49         title = self._html_search_regex(
  50             [r'<span title="([^"]+)">',
  51              r'<div class="thumb-holder video">\s*<h5[^>]*>([^<]+)</h5>',
  52              r'<span[^>]+class="title_thumb">([^<]+)</span>'], webpage, 'title').strip()
  53         thumbnails = [
  54             {
  55                 'url': thumb_url,
  56             } for thumb_url in re.findall(r'<img src="([^"]+)" alt="" />', webpage)
  57         ]
  58         thumbnail = thumbnails[0]['url'] if thumbnails else None
  59         duration = parse_duration(self._html_search_regex(
  60             [r'<i class="fa fa-clock-o"></i>\s*(\d{2}:\d{2})',
  61              r'<span[^>]+class="view_time">([^<]+)</span>'], webpage, 'duration', fatal=False))
  62
  63         return {
  64             'id': video_id,
  65             'title': title,
  66             'thumbnails': thumbnails,
  67             'thumbnail': thumbnail,
  68             'duration': duration,
  69             'age_limit': 18,
  70             'formats': formats,
  71         }