commit b811b4c93bd21c5a053f7deddd3c0a3fe2486184
parent f4dfa9a5ed0756c882a947bd455cc2488d51ffd7
Author: Remita Amine <remitamine@gmail.com>
Date: Thu, 3 Nov 2016 10:37:07 +0100
[vice] add support for uplynk preplay videos(#11101)
Diffstat:
2 files changed, 114 insertions(+), 92 deletions(-)
diff --git a/youtube_dl/extractor/vice.py b/youtube_dl/extractor/vice.py
@@ -1,12 +1,92 @@
from __future__ import unicode_literals
import re
+import time
+import hashlib
+import json
+from .adobepass import AdobePassIE
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>[^/?#&]+)'
_TESTS = [{
@@ -21,7 +101,7 @@ class ViceIE(InfoExtractor):
'add_ie': ['Ooyala'],
}, {
'url': 'http://www.vice.com/video/how-to-hack-a-car',
- 'md5': '6fb2989a3fed069fb8eab3401fc2d3c9',
+ 'md5': 'a7ecf64ee4fa19b916c16f4b56184ae2',
'info_dict': {
'id': '3jstaBeXgAs',
'ext': 'mp4',
@@ -33,6 +113,22 @@ class ViceIE(InfoExtractor):
},
'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',
'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',
'only_matching': True,
}]
+ _PREPLAY_HOST = 'video.vice'
def _real_extract(self, url):
video_id = self._match_id(url)
- webpage = self._download_webpage(url, video_id)
- try:
- embed_code = self._search_regex(
- r'embedCode=([^&\'"]+)', webpage,
- 'ooyala embed code', default=None)
- if embed_code:
- return self.url_result('ooyala:%s' % embed_code, 'Ooyala')
- youtube_id = self._search_regex(
- r'data-youtube-id="([^"]+)"', webpage, 'youtube id')
+ webpage, urlh = self._download_webpage_handle(url, video_id)
+ embed_code = self._search_regex(
+ r'embedCode=([^&\'"]+)', webpage,
+ 'ooyala embed code', default=None)
+ if embed_code:
+ return self.url_result('ooyala:%s' % embed_code, 'Ooyala')
+ youtube_id = self._search_regex(
+ r'data-youtube-id="([^"]+)"', webpage, 'youtube id', default=None)
+ if youtube_id:
return self.url_result(youtube_id, 'Youtube')
- except ExtractorError:
- raise ExtractorError('The page doesn\'t contain a video', expected=True)
+ return self._extract_preplay_video(urlh.geturl(), webpage)
class ViceShowIE(InfoExtractor):
diff --git a/youtube_dl/extractor/viceland.py b/youtube_dl/extractor/viceland.py
@@ -1,23 +1,10 @@
# coding: utf-8
from __future__ import unicode_literals
-import time
-import hashlib
-import json
+from .vice import ViceBaseIE
-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]+)'
_TEST = {
'url': 'https://www.viceland.com/en_us/video/cyberwar-trailer/57608447973ee7705f6fbd4e',
@@ -38,70 +25,9 @@ class VicelandIE(AdobePassIE):
},
'add_ie': ['UplynkPreplay'],
}
+ _PREPLAY_HOST = 'www.viceland'
def _real_extract(self, url):
video_id = self._match_id(url)
-
webpage = self._download_webpage(url, video_id)
- 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 = {}
- 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',
- }
+ return self._extract_preplay_video(url, webpage)
