1
0
mirror of https://github.com/ytdl-org/youtube-dl.git synced 2024-11-14 11:07:50 +01:00
youtube-dl/youtube_dl/extractor/nowness.py
Sergey M? 5c2266df4b Switch codebase to use sanitized_Request instead of
compat_urllib_request.Request

[downloader/dash] Use sanitized_Request

[downloader/http] Use sanitized_Request

[atresplayer] Use sanitized_Request

[bambuser] Use sanitized_Request

[bliptv] Use sanitized_Request

[brightcove] Use sanitized_Request

[cbs] Use sanitized_Request

[ceskatelevize] Use sanitized_Request

[collegerama] Use sanitized_Request

[extractor/common] Use sanitized_Request

[crunchyroll] Use sanitized_Request

[dailymotion] Use sanitized_Request

[dcn] Use sanitized_Request

[dramafever] Use sanitized_Request

[dumpert] Use sanitized_Request

[eitb] Use sanitized_Request

[escapist] Use sanitized_Request

[everyonesmixtape] Use sanitized_Request

[extremetube] Use sanitized_Request

[facebook] Use sanitized_Request

[fc2] Use sanitized_Request

[flickr] Use sanitized_Request

[4tube] Use sanitized_Request

[gdcvault] Use sanitized_Request

[extractor/generic] Use sanitized_Request

[hearthisat] Use sanitized_Request

[hotnewhiphop] Use sanitized_Request

[hypem] Use sanitized_Request

[iprima] Use sanitized_Request

[ivi] Use sanitized_Request

[keezmovies] Use sanitized_Request

[letv] Use sanitized_Request

[lynda] Use sanitized_Request

[metacafe] Use sanitized_Request

[minhateca] Use sanitized_Request

[miomio] Use sanitized_Request

[meovideo] Use sanitized_Request

[mofosex] Use sanitized_Request

[moniker] Use sanitized_Request

[mooshare] Use sanitized_Request

[movieclips] Use sanitized_Request

[mtv] Use sanitized_Request

[myvideo] Use sanitized_Request

[neteasemusic] Use sanitized_Request

[nfb] Use sanitized_Request

[niconico] Use sanitized_Request

[noco] Use sanitized_Request

[nosvideo] Use sanitized_Request

[novamov] Use sanitized_Request

[nowness] Use sanitized_Request

[nuvid] Use sanitized_Request

[played] Use sanitized_Request

[pluralsight] Use sanitized_Request

[pornhub] Use sanitized_Request

[pornotube] Use sanitized_Request

[primesharetv] Use sanitized_Request

[promptfile] Use sanitized_Request

[qqmusic] Use sanitized_Request

[rtve] Use sanitized_Request

[safari] Use sanitized_Request

[sandia] Use sanitized_Request

[shared] Use sanitized_Request

[sharesix] Use sanitized_Request

[sina] Use sanitized_Request

[smotri] Use sanitized_Request

[sohu] Use sanitized_Request

[spankwire] Use sanitized_Request

[sportdeutschland] Use sanitized_Request

[streamcloud] Use sanitized_Request

[streamcz] Use sanitized_Request

[tapely] Use sanitized_Request

[tube8] Use sanitized_Request

[tubitv] Use sanitized_Request

[twitch] Use sanitized_Request

[twitter] Use sanitized_Request

[udemy] Use sanitized_Request

[vbox7] Use sanitized_Request

[veoh] Use sanitized_Request

[vessel] Use sanitized_Request

[vevo] Use sanitized_Request

[viddler] Use sanitized_Request

[videomega] Use sanitized_Request

[viewvster] Use sanitized_Request

[viki] Use sanitized_Request

[vk] Use sanitized_Request

[vodlocker] Use sanitized_Request

[voicerepublic] Use sanitized_Request

[wistia] Use sanitized_Request

[xfileshare] Use sanitized_Request

[xtube] Use sanitized_Request

[xvideos] Use sanitized_Request

[yandexmusic] Use sanitized_Request

[youku] Use sanitized_Request

[youporn] Use sanitized_Request

[youtube] Use sanitized_Request

[patreon] Use sanitized_Request

[extractor/common] Remove unused import

[nfb] PEP 8
2015-11-23 21:56:23 +06:00

135 lines
5.4 KiB
Python

# encoding: utf-8
from __future__ import unicode_literals
from .brightcove import BrightcoveLegacyIE
from .common import InfoExtractor
from ..compat import compat_str
from ..utils import (
ExtractorError,
sanitized_Request,
)
class NownessBaseIE(InfoExtractor):
def _extract_url_result(self, post):
if post['type'] == 'video':
for media in post['media']:
if media['type'] == 'video':
video_id = media['content']
source = media['source']
if source == 'brightcove':
player_code = self._download_webpage(
'http://www.nowness.com/iframe?id=%s' % video_id, video_id,
note='Downloading player JavaScript',
errnote='Unable to download player JavaScript')
bc_url = BrightcoveLegacyIE._extract_brightcove_url(player_code)
if bc_url is None:
raise ExtractorError('Could not find player definition')
return self.url_result(bc_url, 'BrightcoveLegacy')
elif source == 'vimeo':
return self.url_result('http://vimeo.com/%s' % video_id, 'Vimeo')
elif source == 'youtube':
return self.url_result(video_id, 'Youtube')
elif source == 'cinematique':
# youtube-dl currently doesn't support cinematique
# return self.url_result('http://cinematique.com/embed/%s' % video_id, 'Cinematique')
pass
def _api_request(self, url, request_path):
display_id = self._match_id(url)
request = sanitized_Request(
'http://api.nowness.com/api/' + request_path % display_id,
headers={
'X-Nowness-Language': 'zh-cn' if 'cn.nowness.com' in url else 'en-us',
})
return display_id, self._download_json(request, display_id)
class NownessIE(NownessBaseIE):
IE_NAME = 'nowness'
_VALID_URL = r'https?://(?:(?:www|cn)\.)?nowness\.com/(?:story|(?:series|category)/[^/]+)/(?P<id>[^/]+?)(?:$|[?#])'
_TESTS = [{
'url': 'https://www.nowness.com/story/candor-the-art-of-gesticulation',
'md5': '068bc0202558c2e391924cb8cc470676',
'info_dict': {
'id': '2520295746001',
'ext': 'mp4',
'title': 'Candor: The Art of Gesticulation',
'description': 'Candor: The Art of Gesticulation',
'thumbnail': 're:^https?://.*\.jpg',
'uploader': 'Nowness',
},
}, {
'url': 'https://cn.nowness.com/story/kasper-bjorke-ft-jaakko-eino-kalevi-tnr',
'md5': 'e79cf125e387216f86b2e0a5b5c63aa3',
'info_dict': {
'id': '3716354522001',
'ext': 'mp4',
'title': 'Kasper Bjørke ft. Jaakko Eino Kalevi: TNR',
'description': 'Kasper Bjørke ft. Jaakko Eino Kalevi: TNR',
'thumbnail': 're:^https?://.*\.jpg',
'uploader': 'Nowness',
},
}, {
# vimeo
'url': 'https://www.nowness.com/series/nowness-picks/jean-luc-godard-supercut',
'md5': '9a5a6a8edf806407e411296ab6bc2a49',
'info_dict': {
'id': '130020913',
'ext': 'mp4',
'title': 'Bleu, Blanc, Rouge - A Godard Supercut',
'description': 'md5:f0ea5f1857dffca02dbd37875d742cec',
'thumbnail': 're:^https?://.*\.jpg',
'upload_date': '20150607',
'uploader': 'Cinema Sem Lei',
'uploader_id': 'cinemasemlei',
},
}]
def _real_extract(self, url):
_, post = self._api_request(url, 'post/getBySlug/%s')
return self._extract_url_result(post)
class NownessPlaylistIE(NownessBaseIE):
IE_NAME = 'nowness:playlist'
_VALID_URL = r'https?://(?:(?:www|cn)\.)?nowness\.com/playlist/(?P<id>\d+)'
_TEST = {
'url': 'https://www.nowness.com/playlist/3286/i-guess-thats-why-they-call-it-the-blues',
'info_dict': {
'id': '3286',
},
'playlist_mincount': 8,
}
def _real_extract(self, url):
playlist_id, playlist = self._api_request(url, 'post?PlaylistId=%s')
entries = [self._extract_url_result(item) for item in playlist['items']]
return self.playlist_result(entries, playlist_id)
class NownessSeriesIE(NownessBaseIE):
IE_NAME = 'nowness:series'
_VALID_URL = r'https?://(?:(?:www|cn)\.)?nowness\.com/series/(?P<id>[^/]+?)(?:$|[?#])'
_TEST = {
'url': 'https://www.nowness.com/series/60-seconds',
'info_dict': {
'id': '60',
'title': '60 Seconds',
'description': 'One-minute wisdom in a new NOWNESS series',
},
'playlist_mincount': 4,
}
def _real_extract(self, url):
display_id, series = self._api_request(url, 'series/getBySlug/%s')
entries = [self._extract_url_result(post) for post in series['posts']]
series_title = None
series_description = None
translations = series.get('translations', [])
if translations:
series_title = translations[0].get('title') or translations[0]['seoTitle']
series_description = translations[0].get('seoDescription')
return self.playlist_result(
entries, compat_str(series['id']), series_title, series_description)