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

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import random
   5 import re
   6
   7 from .common import InfoExtractor
   8 from ..compat import compat_str
   9 from ..utils import (
  10     determine_ext,
  11     float_or_none,
  12     parse_age_limit,
  13     qualities,
  14     try_get,
  15     unified_timestamp,
  16     urljoin,
  17 )
  18
  19
  20 class VideoPressIE(InfoExtractor):
  21     _VALID_URL = r'https?://videopress\.com/embed/(?P<id>[\da-zA-Z]+)'
  22     _TESTS = [{
  23         'url': 'https://videopress.com/embed/kUJmAcSf',
  24         'md5': '706956a6c875873d51010921310e4bc6',
  25         'info_dict': {
  26             'id': 'kUJmAcSf',
  27             'ext': 'mp4',
  28             'title': 'VideoPress Demo',
  29             'thumbnail': r're:^https?://.*\.jpg',
  30             'duration': 634.6,
  31             'timestamp': 1434983935,
  32             'upload_date': '20150622',
  33             'age_limit': 0,
  34         },
  35     }, {
  36         # 17+, requires birth_* params
  37         'url': 'https://videopress.com/embed/iH3gstfZ',
  38         'only_matching': True,
  39     }]
  40
  41     @staticmethod
  42     def _extract_urls(webpage):
  43         return re.findall(
  44             r'<iframe[^>]+src=["\']((?:https?://)?videopress\.com/embed/[\da-zA-Z]+)',
  45             webpage)
  46
  47     def _real_extract(self, url):
  48         video_id = self._match_id(url)
  49
  50         video = self._download_json(
  51             'https://public-api.wordpress.com/rest/v1.1/videos/%s' % video_id,
  52             video_id, query={
  53                 'birth_month': random.randint(1, 12),
  54                 'birth_day': random.randint(1, 31),
  55                 'birth_year': random.randint(1950, 1995),
  56             })
  57
  58         title = video['title']
  59
  60         def base_url(scheme):
  61             return try_get(
  62                 video, lambda x: x['file_url_base'][scheme], compat_str)
  63
  64         base_url = base_url('https') or base_url('http')
  65
  66         QUALITIES = ('std', 'dvd', 'hd')
  67         quality = qualities(QUALITIES)
  68
  69         formats = []
  70         for format_id, f in video['files'].items():
  71             if not isinstance(f, dict):
  72                 continue
  73             for ext, path in f.items():
  74                 if ext in ('mp4', 'ogg'):
  75                     formats.append({
  76                         'url': urljoin(base_url, path),
  77                         'format_id': '%s-%s' % (format_id, ext),
  78                         'ext': determine_ext(path, ext),
  79                         'quality': quality(format_id),
  80                     })
  81         original_url = try_get(video, lambda x: x['original'], compat_str)
  82         if original_url:
  83             formats.append({
  84                 'url': original_url,
  85                 'format_id': 'original',
  86                 'quality': len(QUALITIES),
  87             })
  88         self._sort_formats(formats)
  89
  90         return {
  91             'id': video_id,
  92             'title': title,
  93             'description': video.get('description'),
  94             'thumbnail': video.get('poster'),
  95             'duration': float_or_none(video.get('duration'), 1000),
  96             'timestamp': unified_timestamp(video.get('upload_date')),
  97             'age_limit': parse_age_limit(video.get('rating')),
  98             'formats': formats,
  99         }