1
0
mirror of https://github.com/yt-dlp/yt-dlp.git synced 2024-12-23 23:00:41 +01:00
yt-dlp/youtube_dl/extractor/once.py

44 lines
2.1 KiB
Python

# 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/rg3/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