ytdl-nightly/youtube_dl/extractor/canvas.py

66 lines
2.5 KiB
Python
Raw Normal View History

2015-10-12 09:07:18 +02:00
from __future__ import unicode_literals
from .common import InfoExtractor
2016-01-14 18:15:12 +01:00
from ..utils import float_or_none
2015-10-12 09:07:18 +02:00
class CanvasIE(InfoExtractor):
2016-01-14 18:15:12 +01:00
_VALID_URL = r'https?://(?:www\.)?canvas\.be/video/(?:[^/]+/)*(?P<id>[^/?#&]+)'
2015-10-12 09:07:18 +02:00
_TEST = {
'url': 'http://www.canvas.be/video/de-afspraak/najaar-2015/de-afspraak-veilt-voor-de-warmste-week',
'md5': 'ea838375a547ac787d4064d8c7860a6c',
'info_dict': {
2016-01-14 18:15:12 +01:00
'id': 'mz-ast-5e5f90b6-2d72-4c40-82c2-e134f884e93e',
'display_id': 'de-afspraak-veilt-voor-de-warmste-week',
2015-10-12 09:07:18 +02:00
'ext': 'mp4',
2016-01-14 18:15:12 +01:00
'title': 'De afspraak veilt voor de Warmste Week',
'description': 'md5:24cb860c320dc2be7358e0e5aa317ba6',
'thumbnail': 're:^https?://.*\.jpg$',
'duration': 49.02,
2015-10-12 09:07:18 +02:00
}
}
def _real_extract(self, url):
2016-01-14 18:15:12 +01:00
display_id = self._match_id(url)
2015-10-12 09:07:18 +02:00
2016-01-14 18:15:12 +01:00
webpage = self._download_webpage(url, display_id)
2015-10-12 09:07:18 +02:00
title = self._search_regex(
2016-01-14 18:15:12 +01:00
r'<h1[^>]+class="video__body__header__title"[^>]*>(.+?)</h1>',
webpage, 'title', default=None) or self._og_search_title(webpage)
video_id = self._html_search_regex(
r'data-video=(["\'])(?P<id>.+?)\1', webpage, 'video id', group='id')
data = self._download_json(
'https://mediazone.vrt.be/api/v1/canvas/assets/%s' % video_id, display_id)
2015-10-12 09:07:18 +02:00
formats = []
for target in data['targetUrls']:
2016-01-14 18:15:12 +01:00
format_url, format_type = target.get('url'), target.get('type')
if not format_url or not format_type:
continue
if format_type == 'HLS':
formats.extend(self._extract_m3u8_formats(
format_url, display_id, entry_protocol='m3u8_native',
ext='mp4', preference=0, fatal=False, m3u8_id=format_type))
elif format_type == 'HDS':
formats.extend(self._extract_f4m_formats(
format_url, display_id, f4m_id=format_type, fatal=False))
else:
formats.append({
'format_id': format_type,
'url': format_url,
})
2015-10-12 09:07:18 +02:00
self._sort_formats(formats)
2016-01-14 18:15:12 +01:00
2015-10-12 09:07:18 +02:00
return {
'id': video_id,
2016-01-14 18:15:12 +01:00
'display_id': display_id,
2015-10-12 09:07:18 +02:00
'title': title,
2016-01-14 18:15:12 +01:00
'description': self._og_search_description(webpage),
2015-10-12 09:07:18 +02:00
'formats': formats,
2016-01-14 18:15:12 +01:00
'duration': float_or_none(data.get('duration'), 1000),
'thumbnail': data.get('posterImageUrl'),
2015-10-12 09:07:18 +02:00
}