[vice] add support for uplynk preplay videos(#11101)

This commit is contained in:
Remita Amine 2016-11-03 10:37:07 +01:00
parent f4dfa9a5ed
commit b811b4c93b
2 changed files with 114 additions and 92 deletions

View file

@ -1,12 +1,92 @@
from __future__ import unicode_literals from __future__ import unicode_literals
import re import re
import time
import hashlib
import json
from .adobepass import AdobePassIE
from .common import InfoExtractor from .common import InfoExtractor
from ..utils import ExtractorError from ..compat import compat_HTTPError
from ..utils import (
int_or_none,
parse_age_limit,
str_or_none,
parse_duration,
ExtractorError,
extract_attributes,
)
class ViceIE(InfoExtractor): class ViceBaseIE(AdobePassIE):
def _extract_preplay_video(self, url, webpage):
watch_hub_data = extract_attributes(self._search_regex(
r'(?s)(<watch-hub\s*.+?</watch-hub>)', webpage, 'watch hub'))
video_id = watch_hub_data['vms-id']
title = watch_hub_data['video-title']
query = {}
is_locked = watch_hub_data.get('video-locked') == '1'
if is_locked:
resource = self._get_mvpd_resource(
'VICELAND', title, video_id,
watch_hub_data.get('video-rating'))
query['tvetoken'] = self._extract_mvpd_auth(url, video_id, 'VICELAND', resource)
# signature generation algorithm is reverse engineered from signatureGenerator in
# webpack:///../shared/~/vice-player/dist/js/vice-player.js in
# https://www.viceland.com/assets/common/js/web.vendor.bundle.js
exp = int(time.time()) + 14400
query.update({
'exp': exp,
'sign': hashlib.sha512(('%s:GET:%d' % (video_id, exp)).encode()).hexdigest(),
})
try:
host = 'www.viceland' if is_locked else self._PREPLAY_HOST
preplay = self._download_json('https://%s.com/en_us/preplay/%s' % (host, video_id), video_id, query=query)
except ExtractorError as e:
if isinstance(e.cause, compat_HTTPError) and e.cause.code == 400:
error = json.loads(e.cause.read().decode())
raise ExtractorError('%s said: %s' % (self.IE_NAME, error['details']), expected=True)
raise
video_data = preplay['video']
base = video_data['base']
uplynk_preplay_url = preplay['preplayURL']
episode = video_data.get('episode', {})
channel = video_data.get('channel', {})
subtitles = {}
cc_url = preplay.get('ccURL')
if cc_url:
subtitles['en'] = [{
'url': cc_url,
}]
return {
'_type': 'url_transparent',
'url': uplynk_preplay_url,
'id': video_id,
'title': title,
'description': base.get('body'),
'thumbnail': watch_hub_data.get('cover-image') or watch_hub_data.get('thumbnail'),
'duration': parse_duration(video_data.get('video_duration') or watch_hub_data.get('video-duration')),
'timestamp': int_or_none(video_data.get('created_at')),
'age_limit': parse_age_limit(video_data.get('video_rating')),
'series': video_data.get('show_title') or watch_hub_data.get('show-title'),
'episode_number': int_or_none(episode.get('episode_number') or watch_hub_data.get('episode')),
'episode_id': str_or_none(episode.get('id') or video_data.get('episode_id')),
'season_number': int_or_none(watch_hub_data.get('season')),
'season_id': str_or_none(episode.get('season_id')),
'uploader': channel.get('base', {}).get('title') or watch_hub_data.get('channel-title'),
'uploader_id': str_or_none(channel.get('id')),
'subtitles': subtitles,
'ie_key': 'UplynkPreplay',
}
class ViceIE(ViceBaseIE):
_VALID_URL = r'https?://(?:.+?\.)?vice\.com/(?:[^/]+/)?videos?/(?P<id>[^/?#&]+)' _VALID_URL = r'https?://(?:.+?\.)?vice\.com/(?:[^/]+/)?videos?/(?P<id>[^/?#&]+)'
_TESTS = [{ _TESTS = [{
@ -21,7 +101,7 @@ class ViceIE(InfoExtractor):
'add_ie': ['Ooyala'], 'add_ie': ['Ooyala'],
}, { }, {
'url': 'http://www.vice.com/video/how-to-hack-a-car', 'url': 'http://www.vice.com/video/how-to-hack-a-car',
'md5': '6fb2989a3fed069fb8eab3401fc2d3c9', 'md5': 'a7ecf64ee4fa19b916c16f4b56184ae2',
'info_dict': { 'info_dict': {
'id': '3jstaBeXgAs', 'id': '3jstaBeXgAs',
'ext': 'mp4', 'ext': 'mp4',
@ -32,6 +112,22 @@ class ViceIE(InfoExtractor):
'upload_date': '20140529', 'upload_date': '20140529',
}, },
'add_ie': ['Youtube'], 'add_ie': ['Youtube'],
}, {
'url': 'https://video.vice.com/en_us/video/the-signal-from-tolva/5816510690b70e6c5fd39a56',
'md5': '',
'info_dict': {
'id': '5816510690b70e6c5fd39a56',
'ext': 'mp4',
'uploader': 'Waypoint',
'title': 'The Signal From Tölva',
'uploader_id': '57f7d621e05ca860fa9ccaf9',
'timestamp': 1477941983938,
},
'params': {
# m3u8 download
'skip_download': True,
},
'add_ie': ['UplynkPreplay'],
}, { }, {
'url': 'https://news.vice.com/video/experimenting-on-animals-inside-the-monkey-lab', 'url': 'https://news.vice.com/video/experimenting-on-animals-inside-the-monkey-lab',
'only_matching': True, 'only_matching': True,
@ -42,21 +138,21 @@ class ViceIE(InfoExtractor):
'url': 'https://munchies.vice.com/en/videos/watch-the-trailer-for-our-new-series-the-pizza-show', 'url': 'https://munchies.vice.com/en/videos/watch-the-trailer-for-our-new-series-the-pizza-show',
'only_matching': True, 'only_matching': True,
}] }]
_PREPLAY_HOST = 'video.vice'
def _real_extract(self, url): def _real_extract(self, url):
video_id = self._match_id(url) video_id = self._match_id(url)
webpage = self._download_webpage(url, video_id) webpage, urlh = self._download_webpage_handle(url, video_id)
try: embed_code = self._search_regex(
embed_code = self._search_regex( r'embedCode=([^&\'"]+)', webpage,
r'embedCode=([^&\'"]+)', webpage, 'ooyala embed code', default=None)
'ooyala embed code', default=None) if embed_code:
if embed_code: return self.url_result('ooyala:%s' % embed_code, 'Ooyala')
return self.url_result('ooyala:%s' % embed_code, 'Ooyala') youtube_id = self._search_regex(
youtube_id = self._search_regex( r'data-youtube-id="([^"]+)"', webpage, 'youtube id', default=None)
r'data-youtube-id="([^"]+)"', webpage, 'youtube id') if youtube_id:
return self.url_result(youtube_id, 'Youtube') return self.url_result(youtube_id, 'Youtube')
except ExtractorError: return self._extract_preplay_video(urlh.geturl(), webpage)
raise ExtractorError('The page doesn\'t contain a video', expected=True)
class ViceShowIE(InfoExtractor): class ViceShowIE(InfoExtractor):

View file

@ -1,23 +1,10 @@
# coding: utf-8 # coding: utf-8
from __future__ import unicode_literals from __future__ import unicode_literals
import time from .vice import ViceBaseIE
import hashlib
import json
from .adobepass import AdobePassIE
from ..compat import compat_HTTPError
from ..utils import (
int_or_none,
parse_age_limit,
str_or_none,
parse_duration,
ExtractorError,
extract_attributes,
)
class VicelandIE(AdobePassIE): class VicelandIE(ViceBaseIE):
_VALID_URL = r'https?://(?:www\.)?viceland\.com/[^/]+/video/[^/]+/(?P<id>[a-f0-9]+)' _VALID_URL = r'https?://(?:www\.)?viceland\.com/[^/]+/video/[^/]+/(?P<id>[a-f0-9]+)'
_TEST = { _TEST = {
'url': 'https://www.viceland.com/en_us/video/cyberwar-trailer/57608447973ee7705f6fbd4e', 'url': 'https://www.viceland.com/en_us/video/cyberwar-trailer/57608447973ee7705f6fbd4e',
@ -38,70 +25,9 @@ class VicelandIE(AdobePassIE):
}, },
'add_ie': ['UplynkPreplay'], 'add_ie': ['UplynkPreplay'],
} }
_PREPLAY_HOST = 'www.viceland'
def _real_extract(self, url): def _real_extract(self, url):
video_id = self._match_id(url) video_id = self._match_id(url)
webpage = self._download_webpage(url, video_id) webpage = self._download_webpage(url, video_id)
watch_hub_data = extract_attributes(self._search_regex( return self._extract_preplay_video(url, webpage)
r'(?s)(<watch-hub\s*.+?</watch-hub>)', webpage, 'watch hub'))
video_id = watch_hub_data['vms-id']
title = watch_hub_data['video-title']
query = {}
if watch_hub_data.get('video-locked') == '1':
resource = self._get_mvpd_resource(
'VICELAND', title, video_id,
watch_hub_data.get('video-rating'))
query['tvetoken'] = self._extract_mvpd_auth(url, video_id, 'VICELAND', resource)
# signature generation algorithm is reverse engineered from signatureGenerator in
# webpack:///../shared/~/vice-player/dist/js/vice-player.js in
# https://www.viceland.com/assets/common/js/web.vendor.bundle.js
exp = int(time.time()) + 14400
query.update({
'exp': exp,
'sign': hashlib.sha512(('%s:GET:%d' % (video_id, exp)).encode()).hexdigest(),
})
try:
preplay = self._download_json('https://www.viceland.com/en_us/preplay/%s' % video_id, video_id, query=query)
except ExtractorError as e:
if isinstance(e.cause, compat_HTTPError) and e.cause.code == 400:
error = json.loads(e.cause.read().decode())
raise ExtractorError('%s said: %s' % (self.IE_NAME, error['details']), expected=True)
raise
video_data = preplay['video']
base = video_data['base']
uplynk_preplay_url = preplay['preplayURL']
episode = video_data.get('episode', {})
channel = video_data.get('channel', {})
subtitles = {}
cc_url = preplay.get('ccURL')
if cc_url:
subtitles['en'] = [{
'url': cc_url,
}]
return {
'_type': 'url_transparent',
'url': uplynk_preplay_url,
'id': video_id,
'title': title,
'description': base.get('body'),
'thumbnail': watch_hub_data.get('cover-image') or watch_hub_data.get('thumbnail'),
'duration': parse_duration(video_data.get('video_duration') or watch_hub_data.get('video-duration')),
'timestamp': int_or_none(video_data.get('created_at')),
'age_limit': parse_age_limit(video_data.get('video_rating')),
'series': video_data.get('show_title') or watch_hub_data.get('show-title'),
'episode_number': int_or_none(episode.get('episode_number') or watch_hub_data.get('episode')),
'episode_id': str_or_none(episode.get('id') or video_data.get('episode_id')),
'season_number': int_or_none(watch_hub_data.get('season')),
'season_id': str_or_none(episode.get('season_id')),
'uploader': channel.get('base', {}).get('title') or watch_hub_data.get('channel-title'),
'uploader_id': str_or_none(channel.get('id')),
'subtitles': subtitles,
'ie_key': 'UplynkPreplay',
}