yt-dlc/youtube_dlc/extractor/udn.py

103 lines
3.5 KiB
Python
Raw Normal View History

2015-04-08 09:26:51 +00:00
# coding: utf-8
from __future__ import unicode_literals
import re
2015-04-08 09:26:51 +00:00
from .common import InfoExtractor
from ..utils import (
determine_ext,
int_or_none,
js_to_json,
)
2015-04-08 13:39:34 +00:00
from ..compat import compat_urlparse
2015-04-08 09:26:51 +00:00
class UDNEmbedIE(InfoExtractor):
2015-07-10 16:52:19 +00:00
IE_DESC = '聯合影音'
_PROTOCOL_RELATIVE_VALID_URL = r'//video\.udn\.com/(?:embed|play)/news/(?P<id>\d+)'
_VALID_URL = r'https?:' + _PROTOCOL_RELATIVE_VALID_URL
2015-04-08 09:26:51 +00:00
_TESTS = [{
'url': 'http://video.udn.com/embed/news/300040',
'info_dict': {
'id': '300040',
'ext': 'mp4',
'title': '生物老師男變女 全校挺"做自己"',
'thumbnail': r're:^https?://.*\.jpg$',
},
'params': {
# m3u8 download
'skip_download': True,
},
2017-10-12 16:04:41 +00:00
'expected_warnings': ['Failed to parse JSON Expecting value'],
2015-04-08 09:26:51 +00:00
}, {
'url': 'https://video.udn.com/embed/news/300040',
'only_matching': True,
}, {
# From https://video.udn.com/news/303776
'url': 'https://video.udn.com/play/news/303776',
2015-04-08 09:26:51 +00:00
'only_matching': True,
}]
def _real_extract(self, url):
video_id = self._match_id(url)
page = self._download_webpage(url, video_id)
2017-10-12 16:04:41 +00:00
options_str = self._html_search_regex(
r'var\s+options\s*=\s*([^;]+);', page, 'options')
trans_options_str = js_to_json(options_str)
options = self._parse_json(trans_options_str, 'options', fatal=False) or {}
if options:
video_urls = options['video']
title = options['title']
poster = options.get('poster')
else:
video_urls = self._parse_json(self._html_search_regex(
r'"video"\s*:\s*({.+?})\s*,', trans_options_str, 'video urls'), 'video urls')
title = self._html_search_regex(
r"title\s*:\s*'(.+?)'\s*,", options_str, 'title')
poster = self._html_search_regex(
r"poster\s*:\s*'(.+?)'\s*,", options_str, 'poster', default=None)
2015-04-08 09:26:51 +00:00
if video_urls.get('youtube'):
return self.url_result(video_urls.get('youtube'), 'Youtube')
formats = []
for video_type, api_url in video_urls.items():
if not api_url:
continue
2015-04-08 09:26:51 +00:00
video_url = self._download_webpage(
2015-04-08 13:39:34 +00:00
compat_urlparse.urljoin(url, api_url), video_id,
note='retrieve url for %s video' % video_type)
ext = determine_ext(video_url)
if ext == 'm3u8':
formats.extend(self._extract_m3u8_formats(
video_url, video_id, ext='mp4', m3u8_id='hls'))
elif ext == 'f4m':
formats.extend(self._extract_f4m_formats(
video_url, video_id, f4m_id='hds'))
else:
2017-10-12 16:04:41 +00:00
mobj = re.search(r'_(?P<height>\d+)p_(?P<tbr>\d+)\.mp4', video_url)
a_format = {
'url': video_url,
# video_type may be 'mp4', which confuses YoutubeDL
'format_id': 'http-' + video_type,
}
if mobj:
a_format.update({
'height': int_or_none(mobj.group('height')),
'tbr': int_or_none(mobj.group('tbr')),
})
formats.append(a_format)
2015-04-08 09:26:51 +00:00
self._sort_formats(formats)
return {
'id': video_id,
'formats': formats,
2017-10-12 16:04:41 +00:00
'title': title,
'thumbnail': poster,
2015-04-08 09:26:51 +00:00
}