yt-dlp/youtube_dl/extractor/afreecatv.py

99 lines
3.5 KiB
Python
Raw Normal View History

# coding: utf-8
from __future__ import unicode_literals
import re
from .common import InfoExtractor
from ..compat import (
compat_urllib_parse_urlparse,
compat_urlparse,
)
from ..utils import (
ExtractorError,
int_or_none,
2016-05-05 18:24:02 +02:00
xpath_text,
)
class AfreecaTVIE(InfoExtractor):
IE_DESC = 'afreecatv.com'
_VALID_URL = r'''(?x)^
https?://(?:(live|afbbs|www)\.)?afreeca(?:tv)?\.com(?::\d+)?
(?:
/app/(?:index|read_ucc_bbs)\.cgi|
/player/[Pp]layer\.(?:swf|html))
\?.*?\bnTitleNo=(?P<id>\d+)'''
_TEST = {
'url': 'http://live.afreecatv.com:8079/app/index.cgi?szType=read_ucc_bbs&szBjId=dailyapril&nStationNo=16711924&nBbsNo=18605867&nTitleNo=36164052&szSkin=',
'md5': 'f72c89fe7ecc14c1b5ce506c4996046e',
'info_dict': {
'id': '36164052',
'ext': 'mp4',
'title': '데일리 에이프릴 요정들의 시상식!',
'thumbnail': 're:^https?://videoimg.afreecatv.com/.*$',
'uploader': 'dailyapril',
'uploader_id': 'dailyapril',
}
}
@staticmethod
def parse_video_key(key):
video_key = {'upload_date': None, 'part': '0'}
m = re.match(r'^(?P<upload_date>\d{8})_\w+_(?P<part>\d+)$', key)
if m:
video_key['upload_date'] = m.group('upload_date')
video_key['part'] = m.group('part')
return video_key
def _real_extract(self, url):
video_id = self._match_id(url)
parsed_url = compat_urllib_parse_urlparse(url)
info_url = compat_urlparse.urlunparse(parsed_url._replace(
netloc='afbbs.afreecatv.com:8080',
path='/api/video/get_video_info.php'))
video_xml = self._download_xml(info_url, video_id)
2016-05-05 18:24:02 +02:00
if xpath_text(video_xml, './track/flag', default='FAIL') != 'SUCCEED':
raise ExtractorError('Specified AfreecaTV video does not exist',
expected=True)
2016-05-05 18:24:02 +02:00
title = xpath_text(video_xml, './track/title', 'title')
uploader = xpath_text(video_xml, './track/nickname', 'uploader')
uploader_id = xpath_text(video_xml, './track/bj_id', 'uploader id')
duration = int_or_none(xpath_text(video_xml, './track/duration',
'duration'))
thumbnail = xpath_text(video_xml, './track/titleImage', 'thumbnail')
entries = []
2016-05-05 18:24:02 +02:00
for video_file in video_xml.findall('./track/video/file'):
video_key = self.parse_video_key(video_file.get('key'))
2016-05-05 18:24:02 +02:00
entries.append({
'id': '%s_%s' % (video_id, video_key['part']),
2016-05-05 18:24:02 +02:00
'title': title,
'upload_date': video_key['upload_date'],
2016-05-05 18:24:02 +02:00
'duration': int_or_none(video_file.get('duration')),
'url': video_file.text,
2016-05-05 18:24:02 +02:00
})
info = {
'id': video_id,
'title': title,
'uploader': uploader,
'uploader_id': uploader_id,
'duration': duration,
'thumbnail': thumbnail,
}
if len(entries) > 1:
info['_type'] = 'multi_video'
info['entries'] = entries
elif len(entries) == 1:
info['url'] = entries[0]['url']
info['upload_date'] = entries[0]['upload_date']
else:
raise ExtractorError(
'No files found for the specified AfreecaTV video, either'
' the URL is incorrect or the video has been made private.',
expected=True)
return info