yt-dlc/youtube_dl/extractor/vzaar.py

92 lines
2.8 KiB
Python
Raw Normal View History

# coding: utf-8
from __future__ import unicode_literals
2017-08-11 15:00:39 +00:00
import re
from .common import InfoExtractor
2018-09-10 19:41:05 +00:00
from ..compat import compat_str
from ..utils import (
int_or_none,
float_or_none,
2018-09-10 19:41:05 +00:00
unified_timestamp,
url_or_none,
)
class VzaarIE(InfoExtractor):
_VALID_URL = r'https?://(?:(?:www|view)\.)?vzaar\.com/(?:videos/)?(?P<id>\d+)'
_TESTS = [{
2018-09-10 19:41:05 +00:00
# HTTP and HLS
'url': 'https://vzaar.com/videos/1152805',
'md5': 'bde5ddfeb104a6c56a93a06b04901dbf',
'info_dict': {
'id': '1152805',
'ext': 'mp4',
'title': 'sample video (public)',
},
}, {
'url': 'https://view.vzaar.com/27272/player',
'md5': '3b50012ac9bbce7f445550d54e0508f2',
'info_dict': {
'id': '27272',
'ext': 'mp3',
'title': 'MP3',
},
}]
2017-08-11 15:00:39 +00:00
@staticmethod
def _extract_urls(webpage):
return re.findall(
r'<iframe[^>]+src=["\']((?:https?:)?//(?:view\.vzaar\.com)/[0-9]+)',
webpage)
def _real_extract(self, url):
video_id = self._match_id(url)
video_data = self._download_json(
'http://view.vzaar.com/v2/%s/video' % video_id, video_id)
2018-09-10 19:41:05 +00:00
title = video_data['videoTitle']
formats = []
source_url = url_or_none(video_data.get('sourceUrl'))
if source_url:
f = {
'url': source_url,
'format_id': 'http',
}
if 'audio' in source_url:
f.update({
'vcodec': 'none',
'ext': 'mp3',
})
else:
f.update({
'width': int_or_none(video_data.get('width')),
'height': int_or_none(video_data.get('height')),
'ext': 'mp4',
'fps': float_or_none(video_data.get('fps')),
})
formats.append(f)
video_guid = video_data.get('guid')
usp = video_data.get('usp')
if isinstance(video_guid, compat_str) and isinstance(usp, dict):
m3u8_url = ('http://fable.vzaar.com/v4/usp/%s/%s.ism/.m3u8?'
% (video_guid, video_id)) + '&'.join(
'%s=%s' % (k, v) for k, v in usp.items())
formats.extend(self._extract_m3u8_formats(
m3u8_url, video_id, 'mp4', entry_protocol='m3u8_native',
m3u8_id='hls', fatal=False))
self._sort_formats(formats)
return {
'id': video_id,
2018-09-10 19:41:05 +00:00
'title': title,
'thumbnail': self._proto_relative_url(video_data.get('poster')),
'duration': float_or_none(video_data.get('videoDuration')),
2018-09-10 19:41:05 +00:00
'timestamp': unified_timestamp(video_data.get('ts')),
'formats': formats,
}