yt-dlp/youtube_dl/extractor/crackle.py

139 lines
4.8 KiB
Python
Raw Normal View History

2016-02-10 22:16:21 +01:00
# coding: utf-8
from __future__ import unicode_literals, division
import re
2016-02-10 22:16:21 +01:00
from .common import InfoExtractor
from ..compat import compat_str
from ..utils import (
determine_ext,
float_or_none,
int_or_none,
parse_age_limit,
parse_duration,
)
2016-02-10 22:16:21 +01:00
class CrackleIE(InfoExtractor):
2017-02-21 19:37:26 +01:00
_GEO_COUNTRIES = ['US']
_VALID_URL = r'(?:crackle:|https?://(?:(?:www|m)\.)?crackle\.com/(?:playlist/\d+/|(?:[^/]+/)+))(?P<id>\d+)'
2016-02-10 22:16:21 +01:00
_TEST = {
'url': 'https://www.crackle.com/andromeda/2502343',
2016-02-10 22:16:21 +01:00
'info_dict': {
'id': '2502343',
2016-02-10 22:16:21 +01:00
'ext': 'mp4',
'title': 'Under The Night',
'description': 'md5:d2b8ca816579ae8a7bf28bfff8cefc8a',
'duration': 2583,
'view_count': int,
'average_rating': 0,
'age_limit': 14,
'genre': 'Action, Sci-Fi',
'creator': 'Allan Kroeker',
'artist': 'Keith Hamilton Cobb, Kevin Sorbo, Lisa Ryder, Lexa Doig, Robert Hewitt Wolfe',
'release_year': 2000,
'series': 'Andromeda',
'episode': 'Under The Night',
'season_number': 1,
'episode_number': 1,
2016-02-10 22:16:21 +01:00
},
'params': {
# m3u8 download
'skip_download': True,
}
}
def _real_extract(self, url):
video_id = self._match_id(url)
media = self._download_json(
'https://web-api-us.crackle.com/Service.svc/details/media/%s/%s'
% (video_id, self._GEO_COUNTRIES[0]), video_id, query={
'disableProtocols': 'true',
'format': 'json'
})
title = media['Title']
formats = []
for e in media['MediaURLs']:
if e.get('UseDRM') is True:
continue
format_url = e.get('Path')
if not format_url or not isinstance(format_url, compat_str):
continue
ext = determine_ext(format_url)
if ext == 'm3u8':
formats.extend(self._extract_m3u8_formats(
format_url, video_id, 'mp4', entry_protocol='m3u8_native',
m3u8_id='hls', fatal=False))
elif ext == 'mpd':
formats.extend(self._extract_mpd_formats(
format_url, video_id, mpd_id='dash', fatal=False))
self._sort_formats(formats)
description = media.get('Description')
duration = int_or_none(media.get(
'DurationInSeconds')) or parse_duration(media.get('Duration'))
view_count = int_or_none(media.get('CountViews'))
average_rating = float_or_none(media.get('UserRating'))
age_limit = parse_age_limit(media.get('Rating'))
genre = media.get('Genre')
release_year = int_or_none(media.get('ReleaseYear'))
creator = media.get('Directors')
artist = media.get('Cast')
if media.get('MediaTypeDisplayValue') == 'Full Episode':
series = media.get('ShowName')
episode = title
season_number = int_or_none(media.get('Season'))
episode_number = int_or_none(media.get('Episode'))
else:
series = episode = season_number = episode_number = None
2016-02-10 22:16:21 +01:00
subtitles = {}
cc_files = media.get('ClosedCaptionFiles')
if isinstance(cc_files, list):
for cc_file in cc_files:
if not isinstance(cc_file, dict):
continue
cc_url = cc_file.get('Path')
if not cc_url or not isinstance(cc_url, compat_str):
continue
lang = cc_file.get('Locale') or 'en'
subtitles.setdefault(lang, []).append({'url': cc_url})
thumbnails = []
images = media.get('Images')
if isinstance(images, list):
for image_key, image_url in images.items():
mobj = re.search(r'Img_(\d+)[xX](\d+)', image_key)
if not mobj:
continue
thumbnails.append({
'url': image_url,
'width': int(mobj.group(1)),
'height': int(mobj.group(2)),
2016-02-10 22:16:21 +01:00
})
return {
'id': video_id,
'title': title,
'description': description,
'duration': duration,
'view_count': view_count,
'average_rating': average_rating,
'age_limit': age_limit,
'genre': genre,
'creator': creator,
'artist': artist,
'release_year': release_year,
'series': series,
'episode': episode,
'season_number': season_number,
'episode_number': episode_number,
'thumbnails': thumbnails,
2016-02-10 22:16:21 +01:00
'subtitles': subtitles,
'formats': formats,
}