yt-dlp/youtube_dl/extractor/comcarcoff.py

75 lines
2.7 KiB
Python
Raw Normal View History

2014-12-12 14:32:02 -05:00
# encoding: utf-8
from __future__ import unicode_literals
2014-12-12 14:32:02 -05:00
from .common import InfoExtractor
2016-02-10 16:16:21 -05:00
from ..compat import compat_str
2015-12-22 14:18:14 -05:00
from ..utils import (
int_or_none,
parse_duration,
parse_iso8601,
)
2014-12-12 14:32:02 -05:00
class ComCarCoffIE(InfoExtractor):
_VALID_URL = r'https?://(?:www\.)?comediansincarsgettingcoffee\.com/(?P<id>[a-z0-9\-]*)'
2014-12-12 14:32:02 -05:00
_TESTS = [{
'url': 'http://comediansincarsgettingcoffee.com/miranda-sings-happy-thanksgiving-miranda/',
'info_dict': {
2016-02-10 16:16:21 -05:00
'id': '2494164',
2014-12-12 14:32:02 -05:00
'ext': 'mp4',
'upload_date': '20141127',
'timestamp': 1417107600,
2015-12-22 14:18:14 -05:00
'duration': 1232,
2014-12-12 14:32:02 -05:00
'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.',
},
'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 14:32:02 -05:00
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 14:32:02 -05:00
2016-02-10 16:16:21 -05:00
display_id = full_data['activeVideo']['video']
video_data = full_data.get('videos', {}).get(display_id) or full_data['singleshots'][display_id]
2016-02-10 16:16:21 -05:00
video_id = compat_str(video_data['mediaId'])
title = video_data['title']
formats = self._extract_m3u8_formats(
video_data['mediaUrl'], video_id, 'mp4')
self._sort_formats(formats)
2014-12-12 14:32:02 -05:00
thumbnails = [{
'url': video_data['images']['thumb'],
}, {
'url': video_data['images']['poster'],
}]
2015-12-22 14:18:14 -05:00
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 14:32:02 -05:00
return {
'id': video_id,
'display_id': display_id,
'title': title,
2014-12-12 14:32:02 -05:00
'description': video_data.get('description'),
2015-12-22 14:18:14 -05:00
'timestamp': timestamp,
'duration': duration,
2014-12-12 14:32:02 -05:00
'thumbnails': thumbnails,
'formats': formats,
2016-02-10 16:16:21 -05:00
'season_number': int_or_none(video_data.get('season')),
'episode_number': int_or_none(video_data.get('episode')),
2014-12-12 14:49:50 -05:00
'webpage_url': 'http://comediansincarsgettingcoffee.com/%s' % (video_data.get('urlSlug', video_data.get('slug'))),
2014-12-12 14:32:02 -05:00
}