mirror of
https://github.com/yt-dlp/yt-dlp.git
synced 2024-11-16 01:39:38 +01:00
119 lines
4.3 KiB
Python
119 lines
4.3 KiB
Python
from __future__ import unicode_literals
|
|
|
|
from .common import InfoExtractor
|
|
from ..utils import (
|
|
float_or_none,
|
|
int_or_none,
|
|
)
|
|
from ..compat import (
|
|
compat_urllib_request
|
|
)
|
|
|
|
|
|
class ViddlerIE(InfoExtractor):
|
|
_VALID_URL = r'https?://(?:www\.)?viddler\.com/(?:v|embed|player)/(?P<id>[a-z0-9]+)'
|
|
_TESTS = [{
|
|
'url': 'http://www.viddler.com/v/43903784',
|
|
'md5': 'ae43ad7cb59431ce043f0ff7fa13cbf4',
|
|
'info_dict': {
|
|
'id': '43903784',
|
|
'ext': 'mp4',
|
|
'title': 'Video Made Easy',
|
|
'description': 'md5:6a697ebd844ff3093bd2e82c37b409cd',
|
|
'uploader': 'viddler',
|
|
'timestamp': 1335371429,
|
|
'upload_date': '20120425',
|
|
'duration': 100.89,
|
|
'thumbnail': 're:^https?://.*\.jpg$',
|
|
'view_count': int,
|
|
'comment_count': int,
|
|
'categories': ['video content', 'high quality video', 'video made easy', 'how to produce video with limited resources', 'viddler'],
|
|
}
|
|
}, {
|
|
'url': 'http://www.viddler.com/v/4d03aad9/',
|
|
'md5': 'faa71fbf70c0bee7ab93076fd007f4b0',
|
|
'info_dict': {
|
|
'id': '4d03aad9',
|
|
'ext': 'mp4',
|
|
'title': 'WALL-TO-GORTAT',
|
|
'upload_date': '20150126',
|
|
'uploader': 'deadspin',
|
|
'timestamp': 1422285291,
|
|
'view_count': int,
|
|
'comment_count': int,
|
|
}
|
|
}, {
|
|
'url': 'http://www.viddler.com/player/221ebbbd/0/',
|
|
'md5': '0defa2bd0ea613d14a6e9bd1db6be326',
|
|
'info_dict': {
|
|
'id': '221ebbbd',
|
|
'ext': 'mp4',
|
|
'title': 'LETeens-Grammar-snack-third-conditional',
|
|
'description': ' ',
|
|
'upload_date': '20140929',
|
|
'uploader': 'BCLETeens',
|
|
'timestamp': 1411997190,
|
|
'view_count': int,
|
|
'comment_count': int,
|
|
}
|
|
}]
|
|
|
|
def _real_extract(self, url):
|
|
video_id = self._match_id(url)
|
|
|
|
json_url = (
|
|
'http://api.viddler.com/api/v2/viddler.videos.getPlaybackDetails.json?video_id=%s&key=v0vhrt7bg2xq1vyxhkct' %
|
|
video_id)
|
|
headers = {'Referer': 'http://static.cdn-ec.viddler.com/js/arpeggio/v2/embed.html'}
|
|
request = compat_urllib_request.Request(json_url, None, headers)
|
|
data = self._download_json(request, video_id)['video']
|
|
|
|
formats = []
|
|
for filed in data['files']:
|
|
if filed.get('status', 'ready') != 'ready':
|
|
continue
|
|
format_id = filed.get('profile_id') or filed['profile_name']
|
|
f = {
|
|
'format_id': format_id,
|
|
'format_note': filed['profile_name'],
|
|
'url': self._proto_relative_url(filed['url']),
|
|
'width': int_or_none(filed.get('width')),
|
|
'height': int_or_none(filed.get('height')),
|
|
'filesize': int_or_none(filed.get('size')),
|
|
'ext': filed.get('ext'),
|
|
'source_preference': -1,
|
|
}
|
|
formats.append(f)
|
|
|
|
if filed.get('cdn_url'):
|
|
f = f.copy()
|
|
f['url'] = self._proto_relative_url(filed['cdn_url'], 'http:')
|
|
f['format_id'] = format_id + '-cdn'
|
|
f['source_preference'] = 1
|
|
formats.append(f)
|
|
|
|
if filed.get('html5_video_source'):
|
|
f = f.copy()
|
|
f['url'] = self._proto_relative_url(filed['html5_video_source'])
|
|
f['format_id'] = format_id + '-html5'
|
|
f['source_preference'] = 0
|
|
formats.append(f)
|
|
self._sort_formats(formats)
|
|
|
|
categories = [
|
|
t.get('text') for t in data.get('tags', []) if 'text' in t]
|
|
|
|
return {
|
|
'id': video_id,
|
|
'title': data['title'],
|
|
'formats': formats,
|
|
'description': data.get('description'),
|
|
'timestamp': int_or_none(data.get('upload_time')),
|
|
'thumbnail': self._proto_relative_url(data.get('thumbnail_url')),
|
|
'uploader': data.get('author'),
|
|
'duration': float_or_none(data.get('length')),
|
|
'view_count': int_or_none(data.get('view_count')),
|
|
'comment_count': int_or_none(data.get('comment_count')),
|
|
'categories': categories,
|
|
}
|