0
0
Fork 0
mirror of https://github.com/yt-dlp/yt-dlp.git synced 2024-11-30 03:33:02 +00:00
yt-dlp/yt_dlp/extractor/yapfiles.py
pukkandan bfd973ece3 [extractors] Use new framework for existing embeds (#4307)
`Brightcove` is difficult to migrate because it's subclasses may depend
on the signature of the current functions. So it is left as-is for now

Note: Tests have not been migrated
2022-08-02 01:08:16 +05:30

90 lines
2.9 KiB
Python

from .common import InfoExtractor
from ..utils import (
ExtractorError,
int_or_none,
qualities,
url_or_none,
)
class YapFilesIE(InfoExtractor):
_YAPFILES_URL = r'//(?:(?:www|api)\.)?yapfiles\.ru/get_player/*\?.*?\bv=(?P<id>\w+)'
_VALID_URL = r'https?:%s' % _YAPFILES_URL
_EMBED_REGEX = [rf'<iframe\b[^>]+\bsrc=(["\'])(?P<url>(?:https?:)?{_YAPFILES_URL}.*?)\1']
_TESTS = [{
# with hd
'url': 'http://www.yapfiles.ru/get_player/?v=vMDE1NjcyNDUt0413',
'md5': '2db19e2bfa2450568868548a1aa1956c',
'info_dict': {
'id': 'vMDE1NjcyNDUt0413',
'ext': 'mp4',
'title': 'Самый худший пароль WIFI',
'thumbnail': r're:^https?://.*\.jpg$',
'duration': 72,
},
}, {
# without hd
'url': 'https://api.yapfiles.ru/get_player/?uid=video_player_1872528&plroll=1&adv=1&v=vMDE4NzI1Mjgt690b',
'only_matching': True,
}]
def _real_extract(self, url):
video_id = self._match_id(url)
webpage = self._download_webpage(url, video_id, fatal=False)
player_url = None
query = {}
if webpage:
player_url = self._search_regex(
r'player\.init\s*\(\s*(["\'])(?P<url>(?:(?!\1).)+)\1', webpage,
'player url', default=None, group='url')
if not player_url:
player_url = 'http://api.yapfiles.ru/load/%s/' % video_id
query = {
'md5': 'ded5f369be61b8ae5f88e2eeb2f3caff',
'type': 'json',
'ref': url,
}
player = self._download_json(
player_url, video_id, query=query)['player']
playlist_url = player['playlist']
title = player['title']
thumbnail = player.get('poster')
if title == 'Ролик удален' or 'deleted.jpg' in (thumbnail or ''):
raise ExtractorError(
'Video %s has been removed' % video_id, expected=True)
playlist = self._download_json(
playlist_url, video_id)['player']['main']
hd_height = int_or_none(player.get('hd'))
QUALITIES = ('sd', 'hd')
quality_key = qualities(QUALITIES)
formats = []
for format_id in QUALITIES:
is_hd = format_id == 'hd'
format_url = url_or_none(playlist.get(
'file%s' % ('_hd' if is_hd else '')))
if not format_url:
continue
formats.append({
'url': format_url,
'format_id': format_id,
'quality': quality_key(format_id),
'height': hd_height if is_hd else None,
})
self._sort_formats(formats)
return {
'id': video_id,
'title': title,
'thumbnail': thumbnail,
'duration': int_or_none(player.get('length')),
'formats': formats,
}