yt_dlp/extractor/ora.py

   1 import re
   2 import urllib.parse
   3
   4 from .common import InfoExtractor
   5 from ..utils import (
   6     get_element_by_attribute,
   7     qualities,
   8     unescapeHTML,
   9 )
  10
  11
  12 class OraTVIE(InfoExtractor):
  13     _VALID_URL = r'https?://(?:www\.)?(?:ora\.tv|unsafespeech\.com)/([^/]+/)*(?P<id>[^/\?#]+)'
  14     _TESTS = [{
  15         'url': 'https://www.ora.tv/larrykingnow/2015/12/16/vine-youtube-stars-zach-king-king-bach-on-their-viral-videos-0_36jupg6090pq',
  16         'md5': 'fa33717591c631ec93b04b0e330df786',
  17         'info_dict': {
  18             'id': '50178',
  19             'ext': 'mp4',
  20             'title': 'Vine & YouTube Stars Zach King & King Bach On Their Viral Videos!',
  21             'description': 'md5:ebbc5b1424dd5dba7be7538148287ac1',
  22         },
  23     }, {
  24         'url': 'http://www.unsafespeech.com/video/2016/5/10/student-self-censorship-and-the-thought-police-on-university-campuses-0_6622bnkppw4d',
  25         'only_matching': True,
  26     }]
  27
  28     def _real_extract(self, url):
  29         display_id = self._match_id(url)
  30         webpage = self._download_webpage(url, display_id)
  31
  32         video_data = self._search_regex(
  33             r'"(?:video|current)"\s*:\s*({[^}]+?})', webpage, 'current video')
  34         m3u8_url = self._search_regex(
  35             r'hls_stream"?\s*:\s*"([^"]+)', video_data, 'm3u8 url', None)
  36         if m3u8_url:
  37             formats = self._extract_m3u8_formats(
  38                 m3u8_url, display_id, 'mp4', 'm3u8_native',
  39                 m3u8_id='hls', fatal=False)
  40             # similar to GameSpotIE
  41             m3u8_path = urllib.parse.urlparse(m3u8_url).path
  42             QUALITIES_RE = r'((,[a-z]+\d+)+,?)'
  43             available_qualities = self._search_regex(
  44                 QUALITIES_RE, m3u8_path, 'qualities').strip(',').split(',')
  45             http_path = m3u8_path[1:].split('/', 1)[1]
  46             http_template = re.sub(QUALITIES_RE, r'%s', http_path)
  47             http_template = http_template.replace('.csmil/master.m3u8', '')
  48             http_template = urllib.parse.urljoin(
  49                 'http://videocdn-pmd.ora.tv/', http_template)
  50             preference = qualities(
  51                 ['mobile400', 'basic400', 'basic600', 'sd900', 'sd1200', 'sd1500', 'hd720', 'hd1080'])
  52             for q in available_qualities:
  53                 formats.append({
  54                     'url': http_template % q,
  55                     'format_id': q,
  56                     'quality': preference(q),
  57                 })
  58         else:
  59             return self.url_result(self._search_regex(
  60                 r'"youtube_id"\s*:\s*"([^"]+)', webpage, 'youtube id'), 'Youtube')
  61
  62         return {
  63             'id': self._search_regex(
  64                 r'"id"\s*:\s*(\d+)', video_data, 'video id', default=display_id),
  65             'display_id': display_id,
  66             'title': unescapeHTML(self._og_search_title(webpage)),
  67             'description': get_element_by_attribute(
  68                 'class', 'video_txt_decription', webpage),
  69             'thumbnail': self._proto_relative_url(self._search_regex(
  70                 r'"thumb"\s*:\s*"([^"]+)', video_data, 'thumbnail', None)),
  71             'formats': formats,
  72         }