yt-dlp/youtube_dl/extractor/crackle.py

122 lines
4.4 KiB
Python
Raw Normal View History

2016-02-10 16:16:21 -05:00
# coding: utf-8
from __future__ import unicode_literals, division
import re
2016-02-10 16:16:21 -05:00
from .common import InfoExtractor
from ..utils import int_or_none
class CrackleIE(InfoExtractor):
_VALID_URL = r'(?:crackle:|https?://(?:www\.)?crackle\.com/(?:playlist/\d+/|(?:[^/]+/)+))(?P<id>\d+)'
_TEST = {
'url': 'http://www.crackle.com/comedians-in-cars-getting-coffee/2498934',
2016-02-10 16:16:21 -05:00
'info_dict': {
'id': '2498934',
2016-02-10 16:16:21 -05:00
'ext': 'mp4',
'title': 'Everybody Respects A Bloody Nose',
'description': 'Jerry is kaffeeklatsching in L.A. with funnyman J.B. Smoove (Saturday Night Live, Real Husbands of Hollywood). Theyre headed for brew at 10 Speed Coffee in a 1964 Studebaker Avanti.',
'thumbnail': 're:^https?://.*\.jpg',
'duration': 906,
'series': 'Comedians In Cars Getting Coffee',
'season_number': 8,
'episode_number': 4,
'subtitles': {
'en-US': [{
'ext': 'ttml',
}]
},
2016-02-10 16:16:21 -05:00
},
'params': {
# m3u8 download
'skip_download': True,
}
}
# extracted from http://legacyweb-us.crackle.com/flash/ReferrerRedirect.ashx
_MEDIA_FILE_SLOTS = {
'c544.flv': {
'width': 544,
'height': 306,
},
'360p.mp4': {
'width': 640,
'height': 360,
},
'480p.mp4': {
'width': 852,
'height': 478,
},
'480p_1mbps.mp4': {
'width': 852,
'height': 478,
},
}
def _real_extract(self, url):
video_id = self._match_id(url)
config_doc = self._download_xml(
'http://legacyweb-us.crackle.com/flash/QueryReferrer.ashx?site=16',
video_id, 'Downloading config')
2016-02-10 16:16:21 -05:00
item = self._download_xml(
2016-02-10 16:23:56 -05:00
'http://legacyweb-us.crackle.com/app/revamp/vidwallcache.aspx?flags=-1&fm=%s' % video_id,
video_id).find('i')
2016-02-10 16:16:21 -05:00
title = item.attrib['t']
subtitles = {}
2016-02-10 16:23:56 -05:00
formats = self._extract_m3u8_formats(
'http://content.uplynk.com/ext/%s/%s.m3u8' % (config_doc.attrib['strUplynkOwnerId'], video_id),
2016-02-10 16:39:33 -05:00
video_id, 'mp4', m3u8_id='hls', fatal=None)
2016-02-10 16:16:21 -05:00
path = item.attrib.get('p')
if path:
http_base_url = 'http://ahttp.crackle.com/' + path
for mfs_path, mfs_info in self._MEDIA_FILE_SLOTS.items():
formats.append({
'url': http_base_url + mfs_path,
2016-02-10 16:39:33 -05:00
'format_id': 'http-' + mfs_path.split('.')[0],
2016-02-10 16:16:21 -05:00
'width': mfs_info['width'],
'height': mfs_info['height'],
})
for cc in item.findall('cc'):
locale = cc.attrib.get('l')
v = cc.attrib.get('v')
if locale and v:
if locale not in subtitles:
subtitles[locale] = []
subtitles[locale] = [{
'url': '%s/%s%s_%s.xml' % (config_doc.attrib['strSubtitleServer'], path, locale, v),
2016-02-10 16:16:21 -05:00
'ext': 'ttml',
}]
2016-02-10 16:23:56 -05:00
self._sort_formats(formats, ('width', 'height', 'tbr', 'format_id'))
2016-02-10 16:16:21 -05:00
media_details = self._download_json(
'https://web-api-us.crackle.com/Service.svc/details/media/%s/TW?format=json' % video_id,
video_id, fatal=False)
thumbnails = []
if media_details:
for key, value in media_details.items():
mobj = re.match('^Thumbnail_(\d+)x(\d+)$', key)
if mobj:
width, height = list(map(int, mobj.groups()))
thumbnails.append({
'id': '%dp' % height,
'url': value,
'width': width,
'height': height,
})
2016-02-10 16:16:21 -05:00
return {
'id': video_id,
'title': title,
'description': item.attrib.get('d'),
'duration': int(item.attrib.get('r'), 16) / 1000 if item.attrib.get('r') else None,
2016-02-10 16:16:21 -05:00
'series': item.attrib.get('sn'),
'season_number': int_or_none(item.attrib.get('se')),
'episode_number': int_or_none(item.attrib.get('ep')),
'thumbnails': thumbnails,
2016-02-10 16:16:21 -05:00
'subtitles': subtitles,
'formats': formats,
}