1
1
mirror of https://github.com/ytdl-org/youtube-dl synced 2024-12-27 01:45:49 +01:00
youtube-dl/youtube_dl/extractor/comcarcoff.py

68 lines
2.5 KiB
Python
Raw Normal View History

2014-12-12 20:32:02 +01:00
# encoding: utf-8
from __future__ import unicode_literals
2014-12-12 20:32:02 +01:00
from .common import InfoExtractor
2015-12-22 20:18:14 +01:00
from ..utils import (
int_or_none,
parse_duration,
parse_iso8601,
)
2014-12-12 20:32:02 +01:00
class ComCarCoffIE(InfoExtractor):
_VALID_URL = r'http://(?:www\.)?comediansincarsgettingcoffee\.com/(?P<id>[a-z0-9\-]*)'
2014-12-12 20:32:02 +01:00
_TESTS = [{
'url': 'http://comediansincarsgettingcoffee.com/miranda-sings-happy-thanksgiving-miranda/',
'info_dict': {
'id': 'miranda-sings-happy-thanksgiving-miranda',
'ext': 'mp4',
'upload_date': '20141127',
'timestamp': 1417107600,
2015-12-22 20:18:14 +01:00
'duration': 1232,
2014-12-12 20:32:02 +01:00
'title': 'Happy Thanksgiving Miranda',
'description': 'Jerry Seinfeld and his special guest Miranda Sings cruise around town in search of coffee, complaining and apologizing along the way.',
'thumbnail': 'http://ccc.crackle.com/images/s5e4_thumb.jpg',
},
'params': {
'skip_download': 'requires ffmpeg',
}
}]
def _real_extract(self, url):
display_id = self._match_id(url)
if not display_id:
display_id = 'comediansincarsgettingcoffee.com'
2014-12-12 20:32:02 +01:00
webpage = self._download_webpage(url, display_id)
full_data = self._parse_json(
self._search_regex(
r'window\.app\s*=\s*({.+?});\n', webpage, 'full data json'),
display_id)['videoData']
2014-12-12 20:32:02 +01:00
video_id = full_data['activeVideo']['video']
video_data = full_data.get('videos', {}).get(video_id) or full_data['singleshots'][video_id]
2014-12-12 20:32:02 +01:00
thumbnails = [{
'url': video_data['images']['thumb'],
}, {
'url': video_data['images']['poster'],
}]
formats = self._extract_m3u8_formats(
video_data['mediaUrl'], video_id, ext='mp4')
2015-12-22 20:18:14 +01:00
timestamp = int_or_none(video_data.get('pubDateTime')) or parse_iso8601(
video_data.get('pubDate'))
duration = int_or_none(video_data.get('durationSeconds')) or parse_duration(
video_data.get('duration'))
2014-12-12 20:32:02 +01:00
return {
'id': video_id,
'display_id': display_id,
'title': video_data['title'],
'description': video_data.get('description'),
2015-12-22 20:18:14 +01:00
'timestamp': timestamp,
'duration': duration,
2014-12-12 20:32:02 +01:00
'thumbnails': thumbnails,
'formats': formats,
2014-12-12 20:49:50 +01:00
'webpage_url': 'http://comediansincarsgettingcoffee.com/%s' % (video_data.get('urlSlug', video_data.get('slug'))),
2014-12-12 20:32:02 +01:00
}