once.py (2167B)
- # coding: utf-8
- from __future__ import unicode_literals
- import re
- from .common import InfoExtractor
- class OnceIE(InfoExtractor):
- _VALID_URL = r'https?://.+?\.unicornmedia\.com/now/(?:ads/vmap/)?[^/]+/[^/]+/(?P<domain_id>[^/]+)/(?P<application_id>[^/]+)/(?:[^/]+/)?(?P<media_item_id>[^/]+)/content\.(?:once|m3u8|mp4)'
- ADAPTIVE_URL_TEMPLATE = 'http://once.unicornmedia.com/now/master/playlist/%s/%s/%s/content.m3u8'
- PROGRESSIVE_URL_TEMPLATE = 'http://once.unicornmedia.com/now/media/progressive/%s/%s/%s/%s/content.mp4'
- def _extract_once_formats(self, url, http_formats_preference=None):
- domain_id, application_id, media_item_id = re.match(
- OnceIE._VALID_URL, url).groups()
- formats = self._extract_m3u8_formats(
- self.ADAPTIVE_URL_TEMPLATE % (
- domain_id, application_id, media_item_id),
- media_item_id, 'mp4', m3u8_id='hls', fatal=False)
- progressive_formats = []
- for adaptive_format in formats:
- # Prevent advertisement from embedding into m3u8 playlist (see
- # https://github.com/ytdl-org/youtube-dl/issues/8893#issuecomment-199912684)
- adaptive_format['url'] = re.sub(
- r'\badsegmentlength=\d+', r'adsegmentlength=0', adaptive_format['url'])
- rendition_id = self._search_regex(
- r'/now/media/playlist/[^/]+/[^/]+/([^/]+)',
- adaptive_format['url'], 'redition id', default=None)
- if rendition_id:
- progressive_format = adaptive_format.copy()
- progressive_format.update({
- 'url': self.PROGRESSIVE_URL_TEMPLATE % (
- domain_id, application_id, rendition_id, media_item_id),
- 'format_id': adaptive_format['format_id'].replace(
- 'hls', 'http'),
- 'protocol': 'http',
- 'preference': http_formats_preference,
- })
- progressive_formats.append(progressive_format)
- self._check_formats(progressive_formats, media_item_id)
- formats.extend(progressive_formats)
- return formats